Abstract
Obesity in pregnant women increases offspring cardiovascular risk and causes fetal cardiac dysfunction. The underpinning mechanisms remain unclear. We hypothesised that circulating factors in serum from fetuses of women with obesity induce pathological cardiomyocyte hypertrophy. Pregnant women with obesity or healthy weight were recruited at term and provided umbilical cord serum and placentas, which were used for isolation of primary trophoblast cells. Primary cardiomyocytes were isolated from neonatal rats. Compared to cord serum from healthy weight women, cord serum from women with obesity upregulated cardiomyocyte mRNA expression of atrial natriuretic factor (Anf) and brain natriuretic peptide (Bnp) and increased the ratio of beta- to alpha-myosin heavy chain expression (Myh7:Myh6), when it was supplemented into the culture medium. This effect was prevented by treating the cord serum with heat-freeze cycling and DNase or RNase digestion. Separately, conditioned medium from trophoblast cells from women with obesity increased cardiomyocyte Anf expression without altering Bnp or Myh7:Myh6. MicroRNAs miR-142 and miR-17, which are associated with cardiac function, were increased in abundance in extracellular vesicles isolated from cord serum from women with obesity. However, miR-142-3p, miR-142-5p and miR-17-5p did not increase Anf, Bnp or Myh7:Myh6 expression when they were transfected into cardiomyocytes. Neither cord serum nor the upregulated microRNAs from women with obesity altered cardiomyocyte size. The results show that human fetal circulating and placenta-derived factors induce gene expression hallmarks of pathological hypertrophy in cardiomyocytes and may mediate cardiac dysfunction in children of women with obesity.
Keywords: metabolic syndrome, cardiomyopathy, hypertrophic, fetal organ maturity, exosomes, diabetes, gestational
NEW AND NOTEWORTHY
Obesity in pregnant women increases risk for heart problems in their children. This study treated heart cells growing in a dish with blood plasma from the umbilical cords of newborn babies. Plasma from babies of women with obesity activated genes linked to heart failure. This means we could design treatments targeting plasma molecules, like microRNAs, or the way the placenta releases them. This could improve children’s heart health if the mother has obesity.
INTRODUCTION
Obesity in pregnant women is associated with increased risk for cardiometabolic disease in the offspring throughout life (1). This predisposition may be established in utero by adverse alterations in the structure and function of the fetal heart. Maternal obesity increases the thickness of the fetal left ventricular walls and intraventricular septum and reduces cardiac strain rate (2, 3), consistent with cardiac hypertrophy and contractile dysfunction. Cardiac dysfunction persists postnatally in the offspring, at least into childhood (4). Experimentally-induced obesity in pregnant animals similarly causes fetal cardiac dysfunction in association with hypertrophy and impaired contractility of individual cardiomyocytes (5–10). The mechanisms that induce fetal cardiomyocyte hypertrophy in pregnancies complicated by obesity remain unknown.
Humoral factors contribute mechanistically to pathological cardiac hypertrophy in postnatal life and catecholamines, angiotensin II, inflammatory cytokines, growth factors and insulin are all implicated (11). Circulating factors in serum samples from patients with pediatric dilated cardiomyopathy increase cell size and stiffness and induce expression of genes associated with pathology in cardiomyocytes in vitro (12, 13). Maternal obesity and diabetes alter the fetal hormonal and metabolic milieu, and cord plasma circulating factors are strongly associated with cardiac function in the infant (14).
Recently, circulating nucleic acids including microRNAs have been shown to mediate inter-organ crosstalk when they are stabilised in extracellular vesicles or carrier proteins, and cause cardiomyocyte hypertrophy (12, 15). MicroRNAs bind to specific target messenger RNAs to regulate their degradation and/or translation and are known to influence intracellular proteins that are important in cardiomyocyte hypertrophy. Obesity in pregnant women alters the maternal circulating and placental abundance of microRNAs that predict the postnatal growth and metabolic phenotype of the infant (16, 17). Maternal obesity also alters microRNA expression in the fetal heart in non-human primates (18). The contribution of circulating factors and microRNAs to fetal cardiac hypertrophy, as opposed to mechanical or neural inputs, remains to be established.
Fetal cardiac development is linked to placental function. The placental circulation and fetal heart develop in parallel in early gestation (19). Genetic manipulations specifically targeted to placental trophoblast cells alter fetal cardiac phenotype in mice (20). In pregnant mice with diet-induced obesity, maternal adiponectin supplementation normalises both placental function and postnatal cardiac dysfunction (21). Placental cells synthesise hormones, metabolites, signalling molecules and extracellular vesicles that are released into the fetal circulation (22–24). Placenta-secreted factors may therefore play a role in fetal cardiac development in pregnancies complicated by maternal obesity.
We hypothesised that circulating factors in cord serum from fetuses of women with obesity induce pathological cardiomyocyte hypertrophy. We aimed to determine the effect of fetal circulating factors on cultured primary neonatal rat ventricular cardiomyocytes, a well-established model used to investigate mechanisms of pathological hypertrophy. We also tested the cardiomyocyte effects of circulating microRNAs and placenta-derived factors.
MATERIALS AND METHODS
Study participants
Pregnant women (n=62) with obesity (pre-pregnancy body mass index > 30) or healthy weight (pre-pregnancy body mass index 20–24.9) were recruited to the study prior to delivery at University of Colorado Hospital, with informed consent and ethical approval from the Institutional Review Board of the University of Colorado (COMIRB 14–1073). Inclusion criteria for both study groups included singleton pregnancy and maternal age 18–45 years. Exclusion criteria included infection, fetal anomaly, maternal smoking, or alcohol or recreational drug use during pregnancy. None of the study participants had pre-gestational diabetes or chronic hypertension. Umbilical cord blood and/or placenta were collected at delivery and processed within 30 minutes.
Cord blood and serum handling
Cord blood collected from 55 participants was allowed to coagulate then serum was processed by centrifugation and aliquoted and stored at −80°C. Frozen aliquots of 11 of these cord serum specimens were subjected to rapid heating/re-freezing and nuclease digestion as described (25). Briefly, serum was heated to 65°C for 5 minutes then frozen in an ethanol-dry ice bath, in three repeated cycles. This method enhances microRNA detection in human serum samples (Mariner et al, 2016). Subsequently, serum nucleic acids were digested with either DNase (0.1U TURBO DNase per μl serum, AM1907 Thermo Fisher Scientific, Waltham, MA, USA), RNase (750pg RNase A per μl serum, EN0531 Thermo Fisher Scientific, Waltham, MA, USA), or both DNase and RNase consecutively. Enzymes were inactivated after the digest and treated serum was used immediately in subsequent experiments.
Primary human trophoblast cell isolation and culture
Freshly-dissected placental villous tissue from 12 participants was used for isolation of primary human trophoblast cells, as described (26). Briefly, washed and minced tissue was dispersed with trypsin and DNase and cells were subsequently separated by centrifugation over a discontinuous percoll gradient. Trophoblast cells were plated at 5 × 106 cells per dish in 5ml of complete medium (50:50 Ham’s F12:Dulbecco’s Modified Eagle’s Medium, with 10% fetal bovine serum and antibiotics). Culture medium was changed daily thereafter and an aliquot was collected for measurement of human chorionic gonadotrophin, to confirm biochemical differentiation of the cells into syncytiotrophoblast (ELISA from IBL America, Minneapolis, MN, USA). At the final medium change at 66 hours in culture, medium serum content was reduced to 1% and conditioned medium was thereafter collected at 90 hours and stored at −80°C.
Neonatal rat ventricular myocyte isolation and culture
Primary neonatal rat ventricular myocytes (NRVMs) were isolated from 1- to 2-day-old rat pups as described (27, 28). Animal procedures were reviewed and approved by the Institutional Animal Care and Use Committee of the University of Colorado Anschutz Medical Campus (#00235). Briefly, timed-mated pregnant Sprague-Dawley rats (n=19, RRID:RGD_70508) were purchased from Charles River Laboratories (MA, USA) and allowed to deliver at term. Dams were euthanised by CO2 asphyxiation and cervical dislocation. Pups were killed by decapitation and their hearts immediately excised. Ventricles were dissected, minced and digested in trypsin. Resultant cells underwent an initial short, ~45-minute plating to remove non-myocytes. Unattached NRVMs were then plated at 3.2 × 105 cells per well on gelatine-coated 6-well or 12-well plates in Minimum Essential Medium with 5% calf serum and vitamin B12. After 24 hours, NRVMs were transferred to serum-free medium, with insulin, transferrin, bovine serum albumin, vitamin B12 and bromodeoxyuridine (0.1mM), to inhibit proliferation of non-myocardial cells (28). All treatments were subsequently added to the medium and the final outcome measurements were made 72 hours later, without further medium change.
NRVM treatment
Effect of maternal obesity:
Cord serum samples from study participants with healthy weight (n=10) and obesity (n=9) were added to serum-free NRVM medium at 2% (v/v) as described (12). Control NRVMs remained in serum-free medium. Each serum sample was added to at least 2 technical replicate wells of NRVMs, which were analysed separately then averaged before statistical analysis so that the experimental unit was the human study participant.
Effect of serum nucleic acids:
Cord serum samples subjected to heat-freeze treatment and originating from participants with both healthy weight and obesity were added to NRVM medium at 2% (v/v). Liposomal transfection reagent (Lipofectamine RNAiMAX, 7.5 μl) was included in the medium to facilitate cellular uptake of serum nucleic acids after the heat-freeze. To determine the effect of depleting serum nucleic acids, heat-freeze treated cord samples were digested with DNase and/or RNase before addition to the medium.
Effect of microRNA mimics:
NRVMs were transfected with mirVana miRNA mimics for miR-142-3p, miR-142-5p and miR-17-5p (20μM, Table 1) using 7.5μl Lipofectamine RNAiMAX per well. Transfections were performed in two technical replicate NRVM wells, which were averaged prior to statistical analysis. The experiment was repeated in 3–6 independent biological replicate batches of NRVMs, isolated from primary tissue on different days, so that the experimental unit in the final analysis was the primary cell isolation batch.
Table 1.
Primer, probe and microRNA mimic sequences
| Target | Sequence | Supplier ID |
|---|---|---|
| Rat qRT-PCR primers | ||
| Anf forward | GCGAAGGTCAAGCTGCTT | MLS |
| Anf reverse | CTGGGCTCCAATCCTGTCAAT | MLS |
| Bnp forward | GGTGCTGCCCCAGATGATT | MLS |
| Bnp reverse | CTGGAGACTGGCTAGGACTTC | MLS |
| Myh7 forward | CGCTCAGTCATGGCGGAT | MLS |
| Myh7 reverse | GCCCCAAATGCAGCCAT | MLS |
| Myh6 forward | CCTGTCCAGCAGAAAGAGC | MLS |
| Myh6 reverse | CAGGCAAAGTCAAGCATTCATATTTATTGTG | MLS |
| Serca forward | GGCCAGATCGCGCTACA | MLS |
| Serca reverse | GGGCCAATTAGAGAGCAGGTT T | MLS |
| 18s forward | GCCGCTAGAGGTGAAATTCTTG | MLS |
| 18s reverse | CTTTCGCTCTGGTCCGTCTT | MLS |
| Human Taqman Advanced miRNA assay probes | ||
| hsa-miR-186-5p | CAAAGAAUUCUCCUUUUGGGCU | ThermoFisher Scientific, 477940_mir |
| hsa-miR-744-5p | UGCGGGGCUAGGGCUAACAGCA | ThermoFisher Scientific, 478200_ mir |
| hsa-miR-142-3p | UGUAGUGUUUCCUACUUUAUGGA | ThermoFisher Scientific, 477910_mir |
| hsa-miR-142-5p | CAUAAAGUAGAAAGCACUAC U | ThermoFisher Scientific, 477911_mir |
| hsa-miR-17-5p | CAAAGUGCUUACAGUGCAGGUAG | ThermoFisher Scientific, 478447_mir |
| hsa-miR-100-5p | AACCCGUAGAUCCGAACUUGUG | ThermoFisher Scientific, 478224_ mir |
| hsa-miR-483-5p | AAGACGGGAGGAAAGAAGGGAG | ThermoFisher Scientific, 478432_ mir |
| hsa-miR-149-3p | AGGGAGGGACGGGGGCUGUGC | ThermoFisher Scientific, 478720_ mir |
| hsa-miR-5010-5p | AGGGGGAUGGCAGAGCAAAAUU | ThermoFisher Scientific, 480087_ mir |
| Rat miRVana microRNA mimics | ||
| Rno-miR-17-5p | CAAAGUGCUUACAGUGCAGGUAG | ThermoFisher Scientific, MC12412 |
| Rno-miR-142-3p | UGUAGUGUUUCCUACUUUAUGGA | ThermoFisher Scientific, MC10398 |
| Rno-miR-142-5p | CAUAAAGUAGAAAGCACUAC U | ThermoFisher Scientific, MC10979 |
| Negative control | Proprietary validated random sequence | ThermoFisher Scientific, 4464058 |
Effect of primary human trophoblast cell conditioned medium:
Conditioned medium from primary human trophoblast cells isolated from participants with healthy weight (n=7) or obesity (n=5) was also added to NRVM medium at 2%. Controls were supplemented with unconditioned trophoblast cell medium.
Gene expression
NRVMs were lysed and RNA purified 72 hours after treatment, using TRIzol reagent and chloroform extraction. RNA was reverse transcribed (High Capacity cDNA Reverse Trascription Kit, Applied Biosystems) then expression of the rat mRNAs Anf, Bnp, Myh7, Myh6 and Serca was determined using specific primers and SYBR green chemistry. Expression of Anf and Bnp and the ratio Myh7:Myh6 are commonly higher in both fetal hearts and failing adult hearts than in healthy adult hearts so they are collectively called the fetal gene program and used as readouts of pathological cardiomyocyte hypertrophy (29). Expression was determined relative to 18s mRNA, using the ddCt method. Primer sequences are given in Table 1.
Immunostaining and cell size measurement
NRVMs were either visualised live and unstained (microRNA mimic experiments) or fixed in 4% paraformaldehyde then immunostained (serum treatment experiments) using an anti-α-actinin primary antibody (1:2000 overnight, #3134, Cell Signalling Technology) and Alexa-fluor 488-conjugated anti-mouse secondary antibody (1:1000, 45 minutes, RRID:AB_2534069). Two-dimensional images were acquired using an EVOS FL Auto Imaging System (Thermo Fisher Scientific). Cell area was then measured blind to treatment, using Image J (RRID:SCR_003070).
Isolation of RNA from cord serum extracellular vesicles
Total RNA was isolated from extracellular vesicles in 51 cord serum specimens using a commercially available kit (exoRNeasy Serum/Plasma, Qiagen, Germantown, MD, USA). Deep frozen cord serum was thawed by brief incubation at 37°C then filtered using a 0.8μm syringe filter. Extracellular vesicles were isolated from 400μl of filtered serum using membrane-based affinity binding spin columns. Vesicles were lysed and RNA isolated using a phenol/guanidine-based approach with chloroform extraction, ethanol precipitation and silica membrane purification. A synthetic miRNA (5.6 × 108 copies of C. elegans miR-39) was spiked into the vesicle lysate, prior to chloroform extraction, as an internal control for RNA recovery in the isolation procedure.
Small RNASeq
Small RNA libraries were prepared from 6μl total extracellular vesicle RNA (Small RNA-Seq Library Prep Kit, Lexogen, NH, USA). Small RNASeq was then performed with 150 cycles of paired-end reads using the Illumina NovaSEQ 6000, in collaboration with the Genomics and Microarray Core facility of the University of Colorado Anschutz Medical Campus. Resultant reads were trimmed of adapter sequences using BBDuk (https://jgi.doe.gov/data-and-tools/software-tools/bbtools/bb-tools-user-guide/bbduk-guide/) with only reads between 15–31 bases kept. The reads were aligned to miRbase human mature miRNA sequences (30) using Burrows-Wheeler Aligner (31). Counts were compiled and converted to counts per million. Low abundance microRNAs were removed by requiring greater than 20 counts in at least 12 samples. Differentially expressed microRNAs was calculated using EdgeR (RRID:SCR_012802) (32). Gene ontology pathways significantly enriched with mRNA targets of the differentially expressed microRNAs were identified by filtering for experimentally-confirmed interactions using miRPathDB 2.0 (33).
microRNA qPCR
Differentially expressed miRNAs in cord serum extracellular vesicles were validated in a separate set of biological samples from participants with obesity or healthy weight, using primer/probe-based Taqman Advanced miRNA Assays and the Taqman Advanced miRNA cDNA Synthesis Kit (both Life Technologies Corp., CA, USA). Using 2μl of total extracellular vesicle RNA, mature miRNAs were extended with 3’ poly(A) addition and 5’ adaptor ligation then reverse transcribed and uniformly amplified. Relative miRNA quantity was determined by qRT-PCR, using the ddCt method normalised to the geometric mean of endogenous (human miR-186-3p and miR-744-5p) and exogenous (spiked-in) control sequences (C. elegans miR-39). Taqman assay IDs are given in Table 1.
Statistics
Results are presented as mean ± SD. We did not use a power calculation. Clinical characteristics and cord serum microRNA abundances were compared between participants with healthy weight and obesity by Student’s t-test for continuous variables or Chi-squared test for categorical variables. Effects of serum treatments, microRNA mimics and conditioned medium on cardiomyocytes were determined by Student’s t-test or one-way ANOVA with Sidak post-hoc test. Interacting effects of maternal obesity and serum heat-freezing/digest were determined by mixed effects model. Interacting effects of maternal obesity and fetal sex were determined by two-way ANOVA. Homogeneity of variances was assessed by F-test and Welch’s correction was used if variances differed between groups. Distribution was assessed by Shapiro-Wilk test and data were log transformed or compared by Mann-Whitney test if they were not normally distributed. Effects were considered statistically significant when P<0.05.
RESULTS
Effect of human umbilical cord serum on cardiomyocytes in vitro
Umbilical cord serum from infants of women with healthy weight or obesity was added to the medium of primary neonatal rat ventricular myocytes. By definition, women with obesity had significantly higher pre-pregnancy BMI than healthy weight controls but the two groups had similar maternal ethnicity, age, GDM status, gestational age, delivery mode, birth weight, placenta weight and fetal sex (Table 2). Cord serum from fetuses of women with obesity upregulated NRVM expression of the fetal genes Anf and Bnp, compared to cord serum from fetuses of healthy BMI control women (Fig 1A,B). The ratio of fetal Myh7 to adult Myh6 expression was also greater in NRVMs supplemented with cord serum from fetuses of women with obesity compared to healthy weight women (Fig. 1C). Serca expression was not affected by the BMI of the participant providing cord serum (Fig. 1D). The effect of cord serum from pregnancies complicated by maternal obesity on Bnp expression in NRVMs depended on fetal sex (two-way ANOVA interaction P<0.05, Table 3). Maternal obesity increased NRVM Bnp expression when the fetus was male but not female. Fetal sex did not affect the Anf, Myh7/Myh6 or Serca response (Table 3). Cardiomyocyte size, determined by Image J analysis of α-actinin immunostained cells, was similar in NRVMs treated with cord serum from women with obesity and healthy weight (Fig. 1E).
Table 2.
Clinical characteristics of human study participants providing cord serum for NRVM experiment.
| Healthy weight | n | Obesity | n | P value | |
|---|---|---|---|---|---|
| Pre-pregnancy BMI, kg/m2 (range) | 22.6 ± 1.8 (19.7–25.6) | 10 | 37.5 ± 3.2 (31.5–39.8) | 9 | *P<0.001 |
| Ethnicity, number Hispanic (%) | 1 (10%) | 10 | 4 (44%) | 9 | P=0.089 |
| Maternal age, years | 31.7 ± 4.4 | 10 | 29.2 ± 6.3 | 9 | P=0.327 |
| Gestational diabetes, number (%) | 1 (10%) | 10 | 0 (0%) | 9 | P=0.330 |
| Gestational age at delivery, weeks | 39.2 ± 0.9 | 10 | 39.9 ± 2.2 | 9 | P=0.693 |
| Vaginal delivery, number (%) | 1 (10%) | 10 | 0 (0%) | 9 | P=0.330 |
| Birth weight, kg | 3.42 ± 0.18 | 9 | 3.78 ± 1.21 | 9 | P=0.382 |
| Placenta weight, g | 653 ± 125 | 10 | 774 ± 287 | 9 | P=0.242 |
| Sex, number female (%) | 5 (50%) | 10 | 3 (33%) | 9 | P=0.463 |
Continuous variables compared by Student’s t-test. Categorical variables compared by Chi-squared test. Mean ± SD.
Figure 1. Effect of human umbilical cord serum on cardiomyocytes.
(A-D) mRNA expression of fetal gene program markers of pathological hypertrophy and (E) cell size in NRVMs supplemented with cord serum from women with healthy weight (con, n=7–10) or obesity (n=9). Effect of serum treatment determined by Student’s t-test or Mann-Whitney test (C), P value given in figure. Mean ± SD, points represent means of technical replicate cell wells treated with serum from one study participant.
Table 3.
Interacting effects of fetal sex and maternal obesity on the cardiomyocyte response to cord serum.
| Gene | Healthy weight | Maternal obesity | P value (two-way ANOVA) | ||||
|---|---|---|---|---|---|---|---|
| Female fetus n=5 | Male fetus n=5 | Female fetus n=3 | Male fetus n=6 | Interaction | Maternal obesity | Fetal sex | |
| Anf | 3.13 ± 0.41 | 2.58 ± 0.31 | 3.94 ± 1.08 | 4.16 ± 0.97 | P=0.288 | *P=0.004 | P=0.642 |
| Bnp | 3.73 ± 0.63 | 2.57 ± 0.65 | 4.02 ± 0.63 | 4.46 ± 1.03† | *P=0.050 | *P=0.011 | P=0.353 |
| Myh7/Myh6 | 1.92 ± 0.68 | 1.41 ± 0.38 | 2.22 ± 0.70 | 2.36 ± 0.67 | P=0.282 | P=0.053 | P=0.539 |
| Serca | 1.26 ± 1.68 | 0.50 ± 0.21 | 0.51 ± 0.10 | 0.61 ± 0.23 | P=0.320 | P=0.457 | P=0.447 |
Effects of fetal sex, obesity and interaction determined by two-way ANOVA.
P<0.05 Sidak multiple comparison test vs healthy weight, within fetal sex.
Mean ± SD.
Exposing cord serum to repeated heat-freeze cycles before supplementing it into the NRVM medium altered the effect of maternal obesity (P (interaction) = 0.032, Fig. 2A). After heat-freeze treatment, cord serum from pregnancies complicated by obesity did not upregulate cardiomyocyte Anf, relative to healthy BMI serum (Fig. 2A). Serum heat-freezing also significantly reduced the cardiomyocyte Myh7:Myh6 ratio, irrespective of the BMI group providing the serum (Fig. 2B). The effect of cord serum from pregnancies complicated by obesity on Anf and Myh7:Myh6 was restored when liposomal transfection reagent was included in the medium after heat-freezing (Fig. 2C, D). Treating cord serum with DNase, RNase, or a combination of both before supplementing it into the NRVM medium prevented its effect on Anf, such that expression was similar in cells treated with cord serum from healthy weight women or women with obesity (Fig. 2C). Serum nuclease treatment also tended to prevent the effect of cord serum from women with obesity on Myh7:Myh6 ratio (Fig. 2D).
Figure 2. Effect of depleting nucleic acids from cord serum on cardiomyocytes.
(A-D) mRNA expression of fetal gene program markers of pathological hypertrophy in NRVMs supplemented with cord serum from women with healthy weight (con, n=6) or obesity (ob, n=5). Effects of maternal obesity (P(obesity)), serum heat-freezing or nuclease treatment (P(heat-freeze), P(enzyme)) and interaction (P(interaction) determined by mixed effects model. P values for main effects and Sidak’s post-hoc comparisons given in figure. Mean ± SD, points represent means of technical replicate cell wells treated with serum from one study participant.
Effect of maternal obesity on cord serum microRNA abundance
RNA was purified from extracellular vesicles isolated from cord serum from fetuses of women with healthy weight or obesity. In this larger set of participants, women with obesity had significantly higher pre-pregnancy BMI, heavier babies and greater placental weights than healthy weight controls (Table 4). The women with obesity were also more commonly of Hispanic ethnicity but none of the other clinical characteristics differed in the two groups (Table 4). Small RNASeq identified 27 upregulated miRNAs and 22 downregulated miRNAs with uncorrected P<0.05 in cord serum from women with obesity, compared to women with healthy weight (Fig. 3A, Table 5). The upregulated microRNAs included miR-142-3p, miR-142-5p and miR-17-5p, which are conserved between humans and rodents and have been shown to regulate cardiac function (34, 35). According to miRPathDB 2.0 (33), the experimentally-confirmed downstream mRNA targets of miR-142-3p, miR-142-5p and miR-17-5p, were significantly enriched in transcripts involved in regulating cardiac muscle hypertrophy, metabolic process and growth (Table 6). Following multiplicity correction using false discovery rate, maternal obesity significantly upregulated only miR-142-3p and downregulated miR-5010-5p, miR-100-5p, miR483-5p, and miR-149-3p (FDR<0.05, Fig. 3A, Table 5).
Table 4.
Clinical characteristics of human study participants providing cord serum for RNASeq
| Healthy weight | n | Obesity | n | P value | |
|---|---|---|---|---|---|
| Pre-pregnancy BMI, kg/m2 (range) | 22.6 ± 1.5 (19.7–24.8) | 12 | 37.2 ± 2.9 (31.5–42.0) | 14 | *P<0.001 |
| Ethnicity, number Hispanic (%) | 1 (8%) | 12 | 7 (50%) | 14 | *P=0.022 |
| Maternal age, years | 30.3 ± 5.0 | 12 | 30.1 ± 5.6 | 14 | P=0.928 |
| Gestational age at delivery, weeks | 39.0 ± 1.05 | 12 | 38.9 ± 1.7 | 14 | P=0.911 |
| Gestational diabetes, number (%) | 2 (17%) | 12 | 2 (14%) | 14 | P=0.867 |
| Vaginal delivery, number (%) | 2 (17%) | 12 | 1 (7%) | 14 | P=0.449 |
| Birth weight, kg | 3.22 ± 0.36 | 11 | 3.94 ± 1.06 | 14 | *P=0.031 |
| Placenta weight, g | 624 ± 136 | 11 | 797 ± 249 | 14 | *P=0.050 |
| Sex, number female (%) | 8 (67%) | 12 | 4 (29%) | 14 | P=0.052 |
Continuous variables compared by Student’s t-test. Categorical variables compared by Chi-squared test. Welch’s correction was used if variances differed between groups. Mean ± SD.
Figure 3. Effect of maternal obesity on fetal circulating microRNA abundance.
(A) Volcano plot showing upregulated and downregulated microRNAs by small RNASeq in cord serum from women with obesity (n=14), compared to healthy weight controls (n=12). Points represent individual microRNAs. (B) Relative abundance of selected microRNAs in cord serum from a validation set of women with obesity (n=21) and healthy weight controls (n=18), determined by qT-PCR. Effect of obesity was determined by Student’s t-test (miR-142-3p) or Mann-Whitney test (miR-142-5p, miR-17-5p). Abundance values were log transformed before statistical analysis. Mean ± SD, points represent individual study participants. (C-E) MicroRNA expression, (F-H) mRNA expression of fetal gene program markers of pathological hypertrophy and (I-K) cell size in NRVMs transfected with synthetic mimics for miR-142-3p, miR-142-5p or miR-17-5p. n=3–5 independent transfections. Effect of transfection determined by paired Student’s t-test. Mean ± SD, points represent biological replicate experiments in independently isolated batches of primary cardiomyocytes.
Table 5.
Differentially abundant microRNAs in cord serum from fetuses of women with obesity versus healthy weight, identified by RNASeq
| microRNA | Fold change (maternal obesity / healthy weight) | P-value | False discovery rate |
|---|---|---|---|
| Upregulated | |||
| has-miR-142-3p | 1.588 | 0.001 | 0.045 |
| hsa-miR-20a-5p | 1.474 | 0.006 | 0.130 |
| hsa-let-7g-5p | 1.326 | 0.007 | 0.130 |
| hsa-miR-454-3p | 1.534 | 0.008 | 0.130 |
| hsa-miR-301a-3p | 1.678 | 0.008 | 0.130 |
| hsa-miR-17-5p | 1.360 | 0.012 | 0.152 |
| hsa-let-7i-5p | 1.271 | 0.013 | 0.156 |
| hsa-miR-93-5p | 1.313 | 0.016 | 0.159 |
| hsa-miR-30e-3p | 1.311 | 0.018 | 0.162 |
| hsa-miR-26b-5p | 1.349 | 0.019 | 0.162 |
| hsa-miR-142-5p | 1.381 | 0.021 | 0.162 |
| hsa-miR-625-5p | 1.587 | 0.022 | 0.162 |
| Downregulated | |||
| hsa-miR-149-3p | 0.565 | 0.001 | 0.045 |
| hsa-miR-5010-5p | 0.442 | 0.001 | 0.045 |
| hsa-miR-483-5p | 0.523 | 0.001 | 0.045 |
| hsa-miR-100-5p | 0.487 | 0.001 | 0.045 |
| hsa-miR-99a-5p | 0.530 | 0.002 | 0.077 |
| hsa-miR-483-3p | 0.506 | 0.005 | 0.130 |
| hsa-miR-10400-5p | 0.579 | 0.007 | 0.130 |
| hsa-miR-3168 | 0.514 | 0.008 | 0.130 |
| hsa-miR-4466 | 0.602 | 0.010 | 0.136 |
| hsa-miR-6858-5p | 0.592 | 0.010 | 0.136 |
| hsa-miR-127-3p | 0.733 | 0.014 | 0.156 |
| hsa-miR-516a-5p | 0.462 | 0.015 | 0.156 |
Table 6.
Top biological process GO terms significantly enriched with targets of serum miRNAs upregulated by maternal obesity.
| Pathway | Enrichment | P-value |
|---|---|---|
| miR-142-3p | ||
| cellular protein modification process | 1.43 | 1.0E-02 |
| protein modification process | 1.43 | 1.0E-02 |
| enzyme binding | 1.61 | 1.0E-02 |
| positive regulation of metabolic process | 1.45 | 1.4E-02 |
| regulation of protein metabolic process | 1.53 | 1.5E-02 |
| innate immune response-activating signal transduction | 2.89 | 1.7E-02 |
| activation of innate immune response | 2.68 | 1.9E-02 |
| cardiac muscle hypertrophy | 4.75 | 1.9E-02 |
| positive regulation of cardiac muscle hypertrophy | 8.25 | 1.9E-02 |
| positive regulation of immune response | 2.05 | 1.9E-02 |
| miR-142-5p | ||
| negative regulation of growth | 13.76 | 4.6E-04 |
| positive regulation of cell differentiation | 5.18 | 4.6E-04 |
| regulation of transforming growth factor beta2 production | 103.45 | 4.6E-04 |
| cellular response to transforming growth factor beta stimulus | 10.43 | 6.7E-04 |
| enzyme linked receptor protein signaling pathway | 4.83 | 6.7E-04 |
| epithelial to mesenchymal transition | 14.71 | 6.7E-04 |
| negative regulation of cell differentiation | 5.76 | 6.7E-04 |
| negative regulation of oxidative stress-induced intrinsic apoptotic signaling pathway | 69.77 | 6.7E-04 |
| regulation of cell differentiation | 3.40 | 6.7E-04 |
| regulation of cell migration | 5.16 | 6.7E-04 |
| miR-17-5p | ||
| organelle organization | 1.44 | 2.1E-10 |
| cytosol | 1.33 | 2.4E-10 |
| cytoplasm | 1.15 | 6.4E-09 |
| intracellular organelle | 1.13 | 6.4E-09 |
| regulation of metabolic process | 1.26 | 2.2E-08 |
| intracellular membrane-bounded organelle | 1.15 | 2.7E-08 |
| nucleoplasm | 1.35 | 4.0E-08 |
| enzyme binding | 1.51 | 1.4E-07 |
| protein binding | 1.13 | 1.4E-07 |
| regulation of gene expression | 1.32 | 1.6E-07 |
Based on searching against experimentally-confirmed targets using miRPathDB v2.0.
Selected differentially abundant cord serum miRNAs were validated using qRT-PCR in a different set of biological samples. In the validation sample set, women with obesity again had higher pre-pregnancy BMI and heavier babies and were more frequently of Hispanic ethnicity, but were otherwise similar to women with healthy weight (Table 7). In line with the RNASeq data, miR-142-5p and miR-17-5p were more abundant in cord serum from women with obesity compared to women with healthy weight (Fig. 3B). MiR-142-3p also tended to be higher in women with obesity, albeit the effect was not statistically significant overall (Fig. 3B). Maternal obesity increased cord serum miR-142-3p abundance in female but not male fetuses (Table 8). There was no effect of fetal sex on cord serum miR-142-4p or miR-17-5p (Table 8). We were unable to validate the downregulated miRNAs identified by RNASeq, and abundances of miR-100-5p and miR-483-5p were similar in samples from the two study groups (P>0.05). Neither miR-149-3p nor miR-5010-5p were detectable within 40 cycles by qRT-PCR.
Table 7.
Clinical characteristics of human study participants providing cord serum for qPCR validation of miRNA abundance.
| Healthy weight | n | Obesity | n | P value | |
|---|---|---|---|---|---|
| Pre-pregnancy BMI, kg/m2 (range) | 22.5 ± 1.4 (20.5–24.9) | 18 | 36.4 ± 4.6 (30.1–44.8) | 21 | *P<0.001 |
| Ethnicity, number Hispanic (%) | 0 (0%) | 18 | 11 (52%) | 21 | *P<0.001 |
| Maternal age, years | 32.0 ± 5.0 | 18 | 30.0 ± 4.7 | 21 | P=0.211 |
| Gestational age at delivery, weeks | 39.1 ± 0.7 | 18 | 39.0 ± 1.4 | 21 | P=0.786 |
| Gestational diabetes, number (%) | 0 (0%) | 18 | 0 (0%) | 21 | P>0.999 |
| Vaginal delivery, number (%) | 2 (11%) | 18 | 0 (0%) | 21 | P=0.207 |
| Birth weight, kg | 3.23 ± 0.34 | 17 | 3.72 ± 0.83 | 21 | *P=0.019 |
| Placenta weight, g | 613 ± 144 | 17 | 734 ± 224 | 19 | P=0.076 |
| Sex, number female (%) | 7 (39%) | 18 | 8 (38%) | 21 | P=0.960 |
Continuous variables compared by Student’s t-test. Categorical variables compared by Chi-squared test. Welch’s correction was used if variances differed between groups. Mean ± SD.
Table 8.
Interacting effects of fetal sex and maternal obesity on cord serum miRNA abundance in the validation set of participants.
| miRNA | Healthy weight | Maternal obesity | P value (two-way ANOVA) | ||||
|---|---|---|---|---|---|---|---|
| Female fetus n=7 | Male fetus n=11 | Female fetus n=8 | Male fetus n=13 | Interaction | Maternal obesity | Fetal sex | |
| miR-142-3p | 1.60 ± 0.71 | 2.22 ± 1.02 | 3.48 ± 2.50† | 2.35 ± 1.31 | *P=0.040 | *P=0.030 | P=0.963 |
| miR-142-5p | 3.04 ± 1.71 | 4.42 ± 4.52 | 5.73 ± 2.79 | 5.61 ± 4.12 | P=0.800 | P=0.154 | P=0.641 |
| miR-17-5p | 1.66 ± 0.36 | 1.70 ± 0.62 | 2.34 ± 0.55 | 2.10 ± 1.24 | P=0.427 | P=0.074 | P=0.408 |
Effects of fetal sex, obesity and interaction determined by two-way ANOVA.
P<0.05 Sidak multiple comparison test vs healthy weight, within fetal sex.
Mean ± SD.
Effect of microRNA mimics on cardiomyocytes in vitro
Transfecting NRVMs with synthetic mimics for either miR-142-3p, miR-142-5p or miR-17-5p increased corresponding microRNA expression in the cardiomyocytes (Fig. 3C–E). Transfection with the miR-17-5p mimic also increased NRVM expression of miR-142-5p but there was no other crosstalk between the selected microRNAs (Fig. 3E). The miR-142-3p mimic reduced Myh7:Myh6 expression ratio in the transfected cardiomyocytes without affecting Anf, Bnp or Serca expression (Fig. 3F), whilst the miR-142-5p mimic did not affect expression of any of the markers of hypertrophy (Fig. 3G). MiR-17-5p reduced cardiomyocyte expression of Anf and the Myh7:Myh6 ratio but did not affect Bnp or Serca (Fig. 3H). None of the microRNA mimics altered cardiomyocyte size (Fig. 3I–K).
Effect of primary human trophoblast conditioned medium on cardiomyocytes in vitro.
Conditioned medium was collected from cultured primary human trophoblast cells isolated from placentas of women with healthy weight or obesity and used to supplement the medium of separately-cultured NRVMs. Women with obesity had higher pre-pregnancy BMI than women with healthy weight but all other clinical characteristics were similar in the two groups (Table 9). Trophoblast conditioned medium from placentas of women with obesity increased Anf expression in NRVMs, compared with trophoblast medium from women with healthy weight (Fig. 4A). By contrast, there was no difference in Bnp or Serca expression or in the Myh7:Myh6 ratio in NRVMs given conditioned medium from trophoblasts of healthy weight women and women with obesity (Fig. 4B–D).
Table 9.
Clinical characteristics of human study participants providing primary trophoblast cells
| Healthy weight | n | Obesity | n | P value | |
|---|---|---|---|---|---|
| Pre-pregnancy BMI, kg/m2 (range) | 24.1 ± 2.5 (19.9–26.8) | 7 | 34.7 ± 5.4 (30.1–43.9) | 6 | *P<0.001 |
| Ethnicity, number Hispanic (%) | 2 (25%) | 8 | 3 (50%) | 6 | P=0.334 |
| Maternal age, years | 28.1 ± 6.1 | 8 | 28.0 ± 3.4 | 6 | P=0.965 |
| Gestational age at delivery, weeks | 39.0 ± 0.8 | 8 | 38.9 ± 1.0 | 5 | P=0.855 |
| Gestational diabetes, number (%) | 0 (0%) | 8 | 0 (0%) | 6 | P>0.999 |
| Vaginal delivery, number (%) | 0 (0%) | 8 | 0 (0%) | 6 | P>0.999 |
| Birth weight, kg | 3.28 ± 0.40 | 8 | 3.39 ± 0.42 | 6 | P=0.615 |
| Placenta weight, g | 637 ± 99 | 8 | 611 ± 84 | 6 | P=0.604 |
| Sex, number female (%) | 3 (38%) | 8 | 5 (83%) | 6 | P=0.086 |
Continuous variables compared by Student’s t-test. Categorical variables compared by Chi-squared test. Welch’s correction was used if variances differed between groups. Mean ± SD.
Figure 4. Effect of primary human trophoblast conditioned medium on cardiomyocytes.
(A-D) mRNA expression of fetal gene program markers of pathological hypertrophy in NRVMs supplemented with conditioned medium from primary human trophoblasts isolated from women with healthy weight (con, n=8) or obesity (ob, n=6). Effect of conditioned medium determined by Student’s t-test, P value given in figure. Mean ± SD, points represent individual primary trophoblast cell isolations from different participants.
Discussion
This study shows that circulating factors in human umbilical cord serum from pregnancies complicated by obesity induce hallmarks of pathological hypertrophy in cultured neonatal cardiomyocytes. Cord serum from fetuses of women with obesity specifically increased neonatal rat ventricular myocyte expression of fetal genes Anf and Bnp and the ratio of Myh7:Myh6 expression. Subjecting the cord serum to DNase or RNase digestion prevented the effect of maternal obesity on cardiomyocyte gene expression. Maternal obesity increased cord serum abundance of the mature microRNA strands of miR-142 and miR-17, which are associated with cardiac hypertrophy, but miR-142 and miR-17 mimics did not activate the fetal cardiac gene program in cultured neonatal rat ventricular myocytes. Conditioned medium from cultured primary human trophoblast cells from pregnancies complicated by obesity also induced hallmarks of pathological hypertrophy in cardiomyocytes. The findings indicate that fetal circulating microRNAs or placenta-derived factors may mediate cardiomyocyte hypertrophy in utero in pregnancies complicated by maternal obesity.
Upregulation of cardiomyocyte natriuretic peptide gene expression and an increase in the Myh7:Myh6 expression ratio are consistent with the pattern of cardiac gene expression in adults with cardiomyopathy and therefore represent markers of cardiomyocyte dysfunction (29). Neonatal rat ventricular myocytes also exhibit these hallmarks when supplemented in vitro with human serum from children with dilated cardiomyopathy (12). Inotropic factors, such as β-adrenergic agonists, that are mechanistically implicated in cardiac hypertrophic dysfunction cause similar changes in gene expression in cultured cardiomyocytes, alongside increasing cell size (27). Our findings therefore indicate that there are circulating factors in human fetuses that have a pathological effect on cardiomyocytes and are more abundant or more potent when pregnancy is complicated by maternal obesity. The transcriptional effect in our study was not accompanied by increased cell size so the change in gene expression is unlikely to be a consequence of, for example, altered intracellular diffusion distance or changes in substrate availability.
Our finding that treating cord serum from pregnancies complicated by obesity with nucleases prevents its effect on cardiomyocytes partly agrees with another study showing that DNase digestion prevents the hypertrophic effect of serum from paediatric dilated cardiomyopathy patients (13). However, in contrast with our results, RNase digestion exacerbated, rather than prevented, the hypertrophic effect of paediatric dilated cardiomyopathy in that study (13). Our results are more in line with studies in mice with obesity, which show that increased levels of circulating microRNAs originating from adipose tissue and other metabolic organs increase cardiac dysfunction and specifically affect cardiomyocytes (36, 37). Circulating microRNAs are stabilised by carrier proteins or packaged in extracellular vesicles. The repeated heat-freeze cycling used to prepare serum samples for our nuclease experiment disrupts these carrier proteins and extracellular vesicles (25). Therefore, our findings may indicate that extracellular vesicle or protein-bound microRNAs are responsible for the pathological cardiomyocyte effects of cord serum from fetuses of women with obesity.
The increased abundance of extracellular vesicle miR-142 in cord serum from fetuses of women with obesity is consistent with previous studies showing that circulating levels of miR-142-3p are elevated in children (38) and adults with obesity (39, 40). Circulating miR-17-5p is increased in abundance in pregnant women with gestational diabetes (41). Maternal circulating microRNA abundance is also altered in pregnant women with obesity, although miR-142 and miR-17 are not amongst the differentially abundant microRNAs identified previously (17). Both miR-142-3p and −5p expression is increased in adipose tissue from mice with diet-induced obesity (42). Separately, increased circulating miR-142 and miR-17 abundance is associated with both paediatric and adult cardiomyopathy (43, 44). MiR-142 and miR-17 control glycolytic and oxidative metabolism in skeletal muscle and tumour cells, respectively (45, 46). Thus, it is plausible that these specific microRNAs modulate the metabolic switch that occurs normally in fetal cardiomyocytes around birth and cause them to dysfunction when they are upregulated by maternal obesity. Indeed, experimentally induced diabetes in pregnant mice upregulates fetal cardiac miR-142 expression (47).
Despite this, transfecting cardiomyocytes with synthetic miR-142-3p, miR-142-5p and miR-17-5p mimics did not induce the gene expression hallmarks of pathological hypertrophy in our study. This appears to be in contrast with previous gain-of-function studies showing that in vivo miR-142 overexpression exacerbates cardiac dysfunction in adult mice (34) and miR-17-3p transfection increases cardiomyocyte size in vitro (35). These discrepant findings may be explained by differences in the duration or magnitude of the gain-of-function, differing conditions in vitro versus in vivo, or strand-specific effects of the mature microRNAs. Alternatively, fetal cardiomyocytes may respond differently to these miRNAs than the adult heart. Endogenously expressed microRNAs are likely to act at much lower concentrations than experimentally transfected microRNAs and their in vivo cardiac effects are likely to be mediated in combination with other humoral and physical factors. Moreover, loss-of-function studies suggest that endogenous miR-142 and miR-17 are required for more complex roles in physiological cardiac growth (34, 35). Thus, whilst our findings do not support the concept that miR-142 or miR-17 directly mediate the pathological effect of cord serum on cardiomyocytes in vitro they may still be consistent with a role in modulating fetal cardiac development in vivo in pregnancies complicated by obesity.
Our data indicating a pathological hypertrophic effect of the primary trophoblast secretome on cultured cardiomyocytes is in line with previous studies showing that trophoblast stem cells secrete factors that regulate cardiomyocyte differentiation, cardiomyocyte apoptosis and myocardial growth and function (48–50). They are also in line with mouse studies showing that placenta-specific gene manipulations lead to alterations in fetal cardiac development (20). Primary human trophoblast cells secrete peptides and extracellular vesicles that contain microRNAs (22, 23, 51). Previous studies indicate that extracellular vesicles mediate communication between trophoblasts and cardiomyocytes (48, 49). Circulating extracellular vesicles from adults with obesity cause pathological changes in cardiomyocyte hypertrophic signalling and function (36, 52). Obesity alters placental microRNA expression and the placenta specific miR profile in the fetal circulation (16, 53). Therefore, our findings may indicate that altered trophoblast secretion of factors such as microRNAs contributes to the adverse effect of maternal obesity on the fetal heart.
We cannot rule out the potential effects of other factors acting in combination with maternal obesity in our study. Women with obesity providing cord serum for RNASeq and qPCR microRNA measurements were more commonly Hispanic and had larger placentas and fetuses than healthy weight women. Ethnicity and population specific genetic variation have been shown to impact circulating microRNA abundance in adults (54, 55), although we are unaware of any studies specifically determining the effect of ethnicity on fetal circulating microRNAs. The abundance of microRNAs and extracellular vesicles in both the fetal and maternal circulation is influenced by fetal and placental weight, independently of maternal obesity (56, 57). Although we controlled maternal BMI and most clinical characteristics were similar across study groups, these covarying factors and the differing subsets of participants providing specimens for each cell and microRNA quantification experiment may explain the varying magnitude of the maternal obesity effect in vitro. They may also explain the lack of hypertrophic effect of the miR-142 and miR-17 mimics in our experiment. We also cannot discount differences in physical activity between study groups and subsets, maternal physical activity level was not recorded in our study. Certainly aerobic exercise training alters circulating microRNA abundance in pregnant women without obesity (58). We acknowledge these limitations of our study and speculate that ethnicity, feto-placental size and maternal physical activity also determine the identity and cardiomyocyte effect of fetal circulating factors.
Overall, the results support our hypothesis by showing that circulating factors in serum from fetuses of women with obesity induce hallmarks of pathological hypertrophy in neonatal cardiomyocytes. They also identify fetal circulating microRNAs that are altered in abundance with maternal obesity and indicate that pro-hypertrophic fetal circulating factors could originate from the placenta. Our findings imply that fetal circulating factors contribute mechanistically to the adverse effects of maternal obesity on fetal development and offspring cardiometabolic disease risk. We speculate that interventions targeting circulating microRNAs or altering placental function could prevent offspring cardiac dysfunction in pregnancies complicated by maternal obesity.
GRANTS
This study was supported by the Ludeman Family Center for Women’s Health Research and National Institutes of Health R01HD065007, P30DK048520 via Colorado Nutrition Obesity Research Center Pilot and Feasibility Program, P30CA046934 via the University of Colorado Cancer Center Bioinformatics and Biostatistics Shared Resource, and by the Medical Research Council UK MR/W028301/1.
Footnotes
DISCLOSURES
The authors declare no relevant conflicts of interest.
DATA AVAILABILITY
Source data for this study are openly available at https://doi.org/10.5522/04/30407794.
REFERENCES
- 1.Reynolds RM, Allan KM, Raja EA, Bhattacharya S, McNeill G, Hannaford PC, Sarwar N, Lee AJ, Bhattacharya S, and Norman JE. Maternal obesity during pregnancy and premature mortality from cardiovascular event in adult offspring: follow-up of 1 323 275 person years. Bmj 347: f4539, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ece İ, Uner A, Balli S, Kibar AE, Oflaz MB, and Kurdoglu M. The Effects of Pre-pregnancy Obesity on Fetal Cardiac Functions. Pediatric Cardiology ? 35: 838–843, 2014. [DOI] [PubMed] [Google Scholar]
- 3.Ingul CB, Loras L, Tegnander E, Eik-Nes SH, and Brantberg A. Maternal obesity affects fetal myocardial function as early as in the first trimester. Ultrasound Obstet Gynecol 47: 433–442, 2016. [DOI] [PubMed] [Google Scholar]
- 4.Toemen L, Gishti O, van Osch-Gevers L, Steegers EA, Helbing WA, Felix JF, Reiss IK, Duijts L, Gaillard R, and Jaddoe VW. Maternal obesity, gestational weight gain and childhood cardiac outcomes: role of childhood body mass index. Int J Obes (Lond) 40: 1070–1078, 2016. [DOI] [PubMed] [Google Scholar]
- 5.Wang J, Ma H, Tong C, Zhang H, Lawlis GB, Li Y, Zang M, Ren J, Nijland MJ, Ford SP, Nathanielsz PW, and Li J. Overnutrition and maternal obesity in sheep pregnancy alter the JNK-IRS-1 signaling cascades and cardiac function in the fetal heart. Faseb j 24: 2066–2076, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wang Q, Zhu C, Sun M, Maimaiti R, Ford SP, Nathanielsz PW, Ren J, and Guo W. Maternal obesity impairs fetal cardiomyocyte contractile function in sheep. Faseb j 33: 2587–2598, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Turdi S, Ge W, Hu N, Bradley KM, Wang X, and Ren J. Interaction between maternal and postnatal high fat diet leads to a greater risk of myocardial dysfunction in offspring via enhanced lipotoxicity, IRS-1 serine phosphorylation and mitochondrial defects. Journal of Molecular and Cellular Cardiology 55: 117–129, 2013. [DOI] [PubMed] [Google Scholar]
- 8.Kandadi MR, Hua Y, Zhu M, Turdi S, Nathanielsz PW, Ford SP, Nair S, and Ren J. Influence of gestational overfeeding on myocardial proinflammatory mediators in fetal sheep heart. J Nutr Biochem 24: 1982–1990, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
- 9.Fan X, Turdi S, Ford SP, Hua Y, Nijland MJ, Zhu M, Nathanielsz PW, and Ren J. Influence of gestational overfeeding on cardiac morphometry and hypertrophic protein markers in fetal sheep. J Nutr Biochem 22: 30–37, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Vaughan OR, Rosario FJ, Chan J, Cox LA, Ferchaud-Roucher V, Zemski-Berry KA, Reusch JEB, Keller AC, Powell TL, and Jansson T. Maternal obesity causes fetal cardiac hypertrophy and alters adult offspring myocardial metabolism in mice. J Physiol 600: 3169–3191, 2022. [DOI] [PubMed] [Google Scholar]
- 11.Shimizu I, and Minamino T. Physiological and pathological cardiac hypertrophy. Journal of Molecular and Cellular Cardiology 97: 245–262, 2016. [DOI] [PubMed] [Google Scholar]
- 12.Jiang X, Sucharov J, Stauffer BL, Miyamoto SD, and Sucharov CC. Exosomes from pediatric dilated cardiomyopathy patients modulate a pathological response in cardiomyocytes. Am J Physiol Heart Circ Physiol 312: H818–h826, 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Jeffrey DA, Pires Da Silva J, Garcia AM, Jiang X, Karimpour-Fard A, Toni LS, Lanzicher T, Peña B, Miyano CA, Nunley K, Korst A, Sbaizero O, Taylor MR, Miyamoto SD, Stauffer BL, and Sucharov CC. Serum circulating proteins from pediatric patients with dilated cardiomyopathy cause pathologic remodeling and cardiomyocyte stiffness. JCI Insight 6: 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Cade WT, Levy PT, Tinius RA, Patel MD, Choudhry S, Holland MR, Singh GK, and Cahill AG. Markers of maternal and infant metabolism are associated with ventricular dysfunction in infants of obese Women with type 2 diabetes. Pediatr Res 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Bang C, Batkai S, Dangwal S, Gupta SK, Foinquinos A, Holzmann A, Just A, Remke J, Zimmer K, Zeug A, Ponimaskin E, Schmiedl A, Yin X, Mayr M, Halder R, Fischer A, Engelhardt S, Wei Y, Schober A, Fiedler J, and Thum T. Cardiac fibroblast-derived microRNA passenger strand-enriched exosomes mediate cardiomyocyte hypertrophy. J Clin Invest 124: 2136–2146, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Carreras-Badosa G, Bonmatí A, Ortega F-J, Mercader J-M, Guindo-Martínez M, Torrents D, Prats-Puig A, Martinez-Calcerrada J-M, de Zegher F, Ibáñez L, Fernandez-Real J-M, Lopez-Bermejo A, and Bassols J. Dysregulation of Placental miRNA in Maternal Obesity Is Associated With Pre- and Postnatal Growth. The Journal of Clinical Endocrinology & Metabolism 102: 2584–2594, 2017. [DOI] [PubMed] [Google Scholar]
- 17.Carreras-Badosa G, Bonmati A, Ortega FJ, Mercader JM, Guindo-Martinez M, Torrents D, Prats-Puig A, Martinez-Calcerrada JM, Platero-Gutierrez E, De Zegher F, Ibanez L, Fernandez-Real JM, Lopez-Bermejo A, and Bassols J. Altered Circulating miRNA Expression Profile in Pregestational and Gestational Obesity. J Clin Endocrinol Metab 100: E1446–1456, 2015. [DOI] [PubMed] [Google Scholar]
- 18.Maloyan A, Muralimanoharan S, Huffman S, Cox LA, Nathanielsz PW, Myatt L, and Nijland MJ. Identification and comparative analyses of myocardial miRNAs involved in the fetal response to maternal obesity. Physiol Genomics 45: 889–900, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Burton GJ, and Jauniaux E. Development of the Human Placenta and Fetal Heart: Synergic or Independent? Front Physiol 9: 373, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Radford BN, Zhao X, Glazer T, Eaton M, Blackwell D, Mohammad S, Lo Vercio LD, Devine J, Shalom-Barak T, Hallgrimsson B, Cross JC, Sucov HM, Barak Y, Dean W, and Hemberger M. Defects in placental syncytiotrophoblast cells are a common cause of developmental heart disease. Nat Commun 14: 1174, 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Vaughan OR, Rosario FJ, Powell TL, and Jansson T. Normalisation of circulating adiponectin levels in obese pregnant mice prevents cardiac dysfunction in adult offspring. Int J Obes (Lond) 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Donker RB, Mouillet JF, Chu T, Hubel CA, Stolz DB, Morelli AE, and Sadovsky Y. The expression profile of C19MC microRNAs in primary human trophoblast cells and exosomes. Molecular Human Reproduction 18: 417–424, 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Rosario FJ, Pardo S, Michelsen TM, Erickson K, Moore L, Powell TL, Weintraub ST, and Jansson T. Characterization of the Primary Human Trophoblast Cell Secretome Using Stable Isotope Labeling With Amino Acids in Cell Culture. Front Cell Dev Biol 9: 704781, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Michelsen TM, Henriksen T, Reinhold D, Powell TL, and Jansson T. The human placental proteome secreted into the maternal and fetal circulations in normal pregnancy based on 4-vessel sampling. Faseb j fj201801193R, 2018. [DOI] [PubMed] [Google Scholar]
- 25.Mariner PD, Korst A, Karimpour-Fard A, Stauffer BL, Miyamoto SD, and Sucharov CC. Improved Detection of Circulating miRNAs in Serum and Plasma Following Rapid Heat/Freeze Cycling. Microrna 7: 138–147, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Roos S, Lagerlof O, Wennergren M, Powell TL, and Jansson T. Regulation of amino acid transporters by glucose and growth factors in cultured primary human trophoblast cells is mediated by mTOR signaling. Am J Physiol Cell Physiol 297: C723–731, 2009. [DOI] [PubMed] [Google Scholar]
- 27.Sucharov CC, Mariner PD, Nunley KR, Long C, Leinwand L, and Bristow MR. A beta1-adrenergic receptor CaM kinase II-dependent pathway mediates cardiac myocyte fetal gene induction. Am J Physiol Heart Circ Physiol 291: H1299–1308, 2006. [DOI] [PubMed] [Google Scholar]
- 28.Simpson P, and Savion S. Differentiation of rat myocytes in single cell cultures with and without proliferating nonmyocardial cells. Cross-striations, ultrastructure, and chronotropic response to isoproterenol. Circ Res 50: 101–116, 1982. [DOI] [PubMed] [Google Scholar]
- 29.Razeghi P, Young ME, Alcorn JL, Moravec CS, Frazier OH, and Taegtmeyer H. Metabolic gene expression in fetal and failing human heart. Circulation 104: 2923–2931, 2001. [DOI] [PubMed] [Google Scholar]
- 30.Kozomara A, Birgaoanu M, and Griffiths-Jones S. miRBase: from microRNA sequences to function. Nucleic Acids Res 47: D155–d162, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Li H, and Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25: 1754–1760, 2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Robinson MD, McCarthy DJ, and Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26: 139–140, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Kehl T, Kern F, Backes C, Fehlmann T, Stöckel D, Meese E, Lenhof H-P, and Keller A. miRPathDB 2.0: a novel release of the miRNA Pathway Dictionary Database. Nucleic Acids Research 48: D142–D147, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Sharma S, Liu J, Wei J, Yuan H, Zhang T, and Bishopric NH. Repression of miR-142 by p300 and MAPK is required for survival signalling via gp130 during adaptive hypertrophy. EMBO Mol Med 4: 617–632, 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Shi J, Bei Y, Kong X, Liu X, Lei Z, Xu T, Wang H, Xuan Q, Chen P, Xu J, Che L, Liu H, Zhong J, Sluijter JP, Li X, Rosenzweig A, and Xiao J. miR-17-3p Contributes to Exercise-Induced Cardiac Growth and Protects against Myocardial Ischemia-Reperfusion Injury. Theranostics 7: 664–676, 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Gan L, Xie D, Liu J, Bond Lau W, Christopher TA, Lopez B, Zhang L, Gao E, Koch W, Ma XL, and Wang Y. Small Extracellular Microvesicles Mediated Pathological Communications Between Dysfunctional Adipocytes and Cardiomyocytes as a Novel Mechanism Exacerbating Ischemia/Reperfusion Injury in Diabetic Mice. Circulation 141: 968–983, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Li F, Zhang K, Xu T, Du W, Yu B, Liu Y, and Nie H. Exosomal microRNA-29a mediates cardiac dysfunction and mitochondrial inactivity in obesity-related cardiomyopathy. Endocrine 63: 480–488, 2019. [DOI] [PubMed] [Google Scholar]
- 38.Prats-Puig A, Ortega FJ, Mercader JM, Moreno-Navarrete JM, Moreno M, Bonet N, Ricart W, Lopez-Bermejo A, and Fernandez-Real JM. Changes in circulating microRNAs are associated with childhood obesity. J Clin Endocrinol Metab 98: E1655–1660, 2013. [DOI] [PubMed] [Google Scholar]
- 39.Al-Rawaf HA. Circulating microRNAs and adipokines as markers of metabolic syndrome in adolescents with obesity. Clinical Nutrition 38: 2231–2238, 2019. [DOI] [PubMed] [Google Scholar]
- 40.Ortega FJ, Mercader JM, Catalán V, Moreno-Navarrete JM, Pueyo N, Sabater M, Gómez-Ambrosi J, Anglada R, Fernández-Formoso JA, Ricart W, Frühbeck G, and Fernández-Real JM. Targeting the circulating microRNA signature of obesity. Clin Chem 59: 781–792, 2013. [DOI] [PubMed] [Google Scholar]
- 41.Cao YL, Jia YJ, Xing BH, Shi DD, and Dong XJ. Plasma microRNA-16–5p, −17–5p and −20a-5p: Novel diagnostic biomarkers for gestational diabetes mellitus. J Obstet Gynaecol Res 43: 974–981, 2017. [DOI] [PubMed] [Google Scholar]
- 42.Chartoumpekis DV, Zaravinos A, Ziros PG, Iskrenova RP, Psyrogiannis AI, Kyriazopoulou VE, and Habeos IG. Differential expression of microRNAs in adipose tissue after long-term high-fat diet-induced obesity in mice. PLoS One 7: e34872, 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Jiao M, You HZ, Yang XY, Yuan H, Li YL, Liu WX, Jin M, and Du J. Circulating microRNA signature for the diagnosis of childhood dilated cardiomyopathy. Sci Rep 8: 724, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Fang L, Ellims AH, Moore XL, White DA, Taylor AJ, Chin-Dusting J, and Dart AM. Circulating microRNAs as biomarkers for diffuse myocardial fibrosis in patients with hypertrophic cardiomyopathy. J Transl Med 13: 314, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Chemello F, Grespi F, Zulian A, Cancellara P, Hebert-Chatelain E, Martini P, Bean C, Alessio E, Buson L, Bazzega M, Armani A, Sandri M, Ferrazza R, Laveder P, Guella G, Reggiani C, Romualdi C, Bernardi P, Scorrano L, Cagnin S, and Lanfranchi G. Transcriptomic Analysis of Single Isolated Myofibers Identifies miR-27a-3p and miR-142-3p as Regulators of Metabolism in Skeletal Muscle. Cell Rep 26: 3784–3797.e3788, 2019. [DOI] [PubMed] [Google Scholar]
- 46.Izreig S, Samborska B, Johnson Radia M, Sergushichev, Ma Eric H, Lussier C, Loginicheva E, Donayo Ariel O, Poffenberger Maya C, Sagan Selena M, Vincent Emma E, Artyomov Maxim, Duchaine Thomas F, and Jones Russell G. The miR-17~92 microRNA Cluster Is a Global Regulator of Tumor Metabolism. Cell Reports 16: 1915–1928, 2016. [DOI] [PubMed] [Google Scholar]
- 47.Dong D, Zhang Y, Reece EA, Wang L, Harman CR, and Yang P. microRNA expression profiling and functional annotation analysis of their targets modulated by oxidative stress during embryonic heart development in diabetic mice. Reproductive toxicology (Elmsford, NY) 65: 365–374, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Ni J, Liu Y, Wang K, Wu M, Kang L, Sha D, Xu B, and Gu R. Trophoblast Stem-Cell-Derived Exosomes Improve Doxorubicin-Induced Dilated Cardiomyopathy by Modulating the let-7i/YAP Pathway. Molecular Therapy - Nucleic Acids 22: 948–956, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Jeyarajah MJ, Patterson VS, Jaju Bhattad G, Zhao L, Whitehead SN, and Renaud SJ. Placental extracellular vesicles promote cardiomyocyte maturation and fetal heart development. Communications Biology 7: 1254, 2024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Zhao X, Radford BN, Ungrin M, Dean W, and Hemberger M. The Trophoblast Compartment Helps Maintain Embryonic Pluripotency and Delays Differentiation towards Cardiomyocytes. International Journal of Molecular Sciences 24: 12423, 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Ouyang Y, Bayer A, Chu T, Tyurin VA, Kagan VE, Morelli AE, Coyne CB, and Sadovsky Y. Isolation of human trophoblastic extracellular vesicles and characterization of their cargo and antiviral activity. Placenta 47: 86–95, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Fandl HK, Garcia VP, Treuth JW, Brewster LM, Greiner JJ, Davy KP, Stauffer BL, and Desouza CA. Endothelial-derived extracellular vesicles from obese/hypertensive adults increase factors associated with hypertrophy and fibrosis in cardiomyocytes. Am J Physiol Heart Circ Physiol 324: H675–h685, 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Jing J, Wang Y, Quan Y, Wang Z, Liu Y, and Ding Z. Maternal obesity alters C19MC microRNAs expression profile in fetal umbilical cord blood. Nutrition & Metabolism 17: 52, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Rawlings-Goss RA, Campbell MC, and Tishkoff SA. Global population-specific variation in miRNA associated with cancer risk and clinical biomarkers. BMC Medical Genomics 7: 53, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Flowers E, Kanaya AM, Zhang L, and Aouizerat BE. The Role of Racial and Ethnic Factors in MicroRNA Expression and Risk for Type 2 Diabetes. Front Genet 13: 853633, 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Miranda J, Paules C, Nair S, Lai A, Palma C, Scholz-Romero K, Rice GE, Gratacos E, Crispi F, and Salomon C. Placental exosomes profile in maternal and fetal circulation in intrauterine growth restriction - Liquid biopsies to monitoring fetal growth. Placenta 64: 34–43, 2018. [DOI] [PubMed] [Google Scholar]
- 57.Miura K, Morisaki S, Abe S, Higashijima A, Hasegawa Y, Miura S, Tateishi S, Mishima H, Yoshiura K, and Masuzaki H. Circulating levels of maternal plasma cell-free pregnancy-associated placenta-specific microRNAs are associated with placental weight. Placenta 35: 848–851, 2014. [DOI] [PubMed] [Google Scholar]
- 58.Candia AA, Skow RJ, Quon A, Steinback CD, Davidge ST, and Davenport MH. Circulating miRNA levels following a 14-wk aerobic exercise intervention during pregnancy: a secondary exploratory analysis of the PEACH RCT. Journal of Applied Physiology 139: 37–44, 2025. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Source data for this study are openly available at https://doi.org/10.5522/04/30407794.