This article is mentioned in:

Introduction

Fetal macrosomia has been inconsistently defined as a birth weight of >4,000, 4,500 or 5,000 g, regardless of gestational age (1,2). In China, the most widely used definition of macrosomia is a birth weight ≥4,000 g. Macrosomia can lead to adverse birth outcomes for the newborn, including perinatal asphyxia, neonatal mortality and shoulder dystocia (1,3). In addition, mothers delivering macrosomic fetuses are at an increased risk of experiencing prolonged labor, C-section, abnormal hemorrhage and perineal laceration (4,5). Numerous studies have indicated that high birth weight is associated with long-term health risks for the newborn, including an increased risk of obesity, diabetes and certain types of cancer (6–9). In the past two to three decades, the incidence of macrosomia in developed countries has increased by 15–25% (2). In China, the macrosomia occurrence rate increased from 6.5% in 2006 (10) to 7.3% in 2011 (11). However, the mechanisms responsible for fetal macrosomia remains poorly understood.

A large body of research has indicated that epigenetic alterations in placental tissues are associated with adverse pregnancy outcomes and fetal programming (12–14). MicroRNAs (miRNAs) are short noncoding RNA molecules (~22 nucleotides in length) and are important in post-transcriptional gene regulation through their binding with hundreds of different mRNAs (15). A wide range of biological activities are affected by miRNAs, including cellular proliferation, differentiation, apoptosis and the processes of numerous diseases, including cancer, obesity, insulin resistance and diabetes (16,17). Previous investigations have demonstrated that aberrant expression of a number of miRNAs in the placenta was associated with low birth weight and preeclampsia (18,19). Maccani et al (20) reported that decreased levels of placental miR-16 and -21 were associated with poor fetal growth. In addition, miR-143 was implicated as a crucial regulator of adipogenesis (21). Since evidence suggested that miR-16, -21 and -143 are important in fetal development, the current study sought to further characterize the association of these miRNAs with macrosomia.

The present study examined the expression of miRs-16, -21 and -143 in the placenta of macrosomic and normal pregnancies and demonstrated miRNA dysregulation in macrosomic placenta.

Materials and methods

Study population

Subjects were recruited at Yuying Children's Hospital of Wenzhou Medical University (Wenzhou, China). Macrosomia was defined as a neonate with a birth weight ≥4,000 g. Normal birth weight was defined as 2,500–3,999 g. Samples from macrosomia cases and controls were collected from women between the ages of 18 and 42 years with healthy pregnancies (i.e. pregnancies without hypertension, hepatitis, heart disease, psychological disorders, gestational diabetes or impaired glucose tolerance), whose infants were full-term (≥37 weeks) and viable, with no known genetic disorders. Maternal weight gain during pregnancy was divided into three levels, including low, moderate and high, according to recommendations from the Institute of Medicine (22). A total of 67 macrosomic and 64 control pregnancies were selected. All subjects provided written informed consent to participate in the current study, and the research protocol was approved by the Ethics Committee of Wenzhou Medical University.

Placenta collection

A placental biopsy of ~1 g was excised from the maternal side of each placenta, 2 cm from the umbilical cord insertion site and free of maternal decidua, within 15 min of delivery of the placenta. Biopsies were cut into small sections and immediately placed in RNAlater (Thermo Fisher Scientific, Inc., Waltham, MA, USA). Samples were incubated at 4°C overnight and stored at −80°C until RNA extraction was performed.

miRNA extraction and quantitative detection

Total RNA was extracted from placental tissue using a MicroRNA Isolation kit (BioTek China, Beijing, China), according to the manufacturer's instructions. Extracted RNA was quantified by measuring absorbance at 260 and 280 nm using the Nano-200 Micro-Spectrophotometer (Allsheng Instruments, Co., Ltd., Hangzhou, China). Reverse transcription (RT) of miRNA (200 ng) was performed using the ReverTra Ace® qPCR RT kit (Toyobo Co., Ltd., Osaka, Japan) with Bulge-Loop™ hsa-miR-16-5p (miRQ0000069-1-2), hsa-miR-21-5p (miRQ0000076-1-2), hsa-miR-143-3p (miRQ0000435-1-2) and U6 small nuclear (sn) RNA (MQP-0202) stem-loop RT-primers (Guangzhou RiboBio Co., Ltd., Guangzhou, China), according to the manufacturer's protocol. U6 snRNA was used as an internal control. Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) assay was performed using Thunderbird SYBR qPCR Mix (Toyobo Co., Ltd.) and the CFX96 Touch Real-Time PCR Detection System (Bio-Rad Laboratories, Inc., Hercules, CA, USA), in accordance with the manufacturer's instructions. Cycling conditions were as follows: 95°C for 3 min, followed by 30 cycles of 95°C for 10 sec, 60°C for 20 sec, and 70°C for 10 sec. To check PCR product homogeneity, the temperature was elevated from 65 to 99°C in 0.1°C/sec increments to obtain a melting curve. miRNA expression levels were calculated using the 2−ΔΔCq method (23).

Bioinformatic prediction of miRNA targets and pathway analysis

miRTarBase (mirtarbase.mbc.nctu.edu.tw, release 4.5) was used to identify target genes (24). DAVID (Database for Annotation, Visualization and Integrated Discovery; david.abcc.ncifcrf.gov) was used to identify pathways involving these target genes (25).

Statistical analysis

Basic characteristics of the macrosomia and control groups were compared using unpaired t-tests for continuous variables and reported as the mean ± standard deviation. Categorical variables were compared using χ2 tests. The miRNA expression data were not normally distributed, therefore, relative expression was presented as the median (interquartile range) and the Mann-Whitney U test was used to analyze differences in miRNA expression levels between the macrosomia and control groups. miRNA expression levels were divided by quartile, using the lowest quartile as the reference group. Logistic regression models were used to determine the risk factors of giving birth to a newborn with macrosomia. Spearman's correlation coefficients were used to examine the relationship between miR-21 and -143 expression. All statistical analyses were conducted using SPSS software, version 14.0 (SPSS, Inc., Chicago, IL, USA). All P-values reported were two-tailed, and P<0.05 was considered to indicate a statistically significant difference.

Results

Subject characteristics

Demographic data are presented in Table I. A total of 131 mothers and their neonates were recruited for the current study. The average birth weight of neonates with macrosomia was 4,303.4±219.5 g, while the average weight of control neonates was 3,398.0±362.2 g. Maternal weight gain during pregnancy was significantly higher in mothers with macrosomic neonates compared with the control group (P<0.001), and neonates with macrosomia were more likely to be male (P=0.004). There were no significant differences in gestational age, maternal weight prior to pregnancy, maternal body mass index (BMI) prior to pregnancy, or maternal age between the macrosomia and control groups.

Table I Characteristics of the study sample.

Table I

Characteristics of the study sample.

Characteristic	Controls (n=64)	Macrosomia cases (n=67)	P-value
Birthweight, g	3,398.0±362.2	4,303.4±219.5	<0.001
Gestational age, weeks	39.1±1.0	39.4±1.0	0.094
Maternal weight prior to pregnancy, kg	52.3±7.0	54.5±7.2	0.110
Height, cm	159.9±3.6	160.5±4.5	0.445
Maternal BMI prior to pregnancy, kg/m2	20.5±2.7	21.2±2.8	0.195
Maternal weight gain during pregnancy, kg	15.4±5.7	20.1±5.7	<0.001
Maternal age, years	27.8±4.4	29.0±3.8	0.097
Tobacco during pregnancy, n (%)
No	64 (100)	67 (100)
Yes	0	0
Infant gender, n (%)			0.004
Male	30 (46.9)	48 (71.6)
Female	34 (53.1)	19 (28.4)
Delivery method, n (%)			<0.001
Vaginal	55 (85.9)	32 (47.8)
C-section	9 (14.1)	35 (52.2)

[i] P-values were obtained using unpaired t-tests for continuous variables and χ² tests for categorical variables. BMI, body mass index.

miRNA expression in macrosomic and control placenta

To determine whether miR-16, -21 or -143 were differentially expressed in macrosomic and control placenta, their expression was measured using RT-qPCR. The Mann-Whitney U-test demonstrated that miR-143 was expressed at lower levels in macrosomic placenta (Fig. 1A; P=0.003), whereas miR-21 expression was significantly higher in macrosomic placenta (Fig. 1B; P= 0.043) compared with the control samples. However, no significant difference in miR-16 expression levels was detected between the two groups (Fig. 1C; P=0.955).

Figure 1 Relative expression of (A) miR-143, (B) miR-21 and (C) miR-16 in placenta. miRNA expression was measured using reverse transcription-quantitative polymerase chain reaction in placenta from macrosomic (n=67) and control (n=64) pregnancies. U6 RNA was used as an internal control. Data are presented as Tukey Box plots. *P<0.05 and **P<0.01, comparison shown by brackets.

Association of miRNAs with macrosomia

The present study examined whether increased miR-21 expression and decreased miR-143 expression were associated with the likelihood of macrosomia using logistic regression models, and adjusted for potential confounders, including maternal age, weight prior to pregnancy, weight gain during pregnancy, gestational age, infant gender and delivery method. miRNA expression levels were divided by quartile (Table II). As presented in Table II, the likelihood of macrosomia increased with miR-21 expression quartile as follows: Q2, OR=6.67 (95% CI, 1.39–32.05); Q3, OR=4.10 (95% CI, 0.88–19.11); and Q4, OR=16.19 (95% CI, 2.46–106.68). Conversely, higher miR-143 quartiles were associated with protection against macrosomia as follows: Q2, OR=0.22 (95% CI, 0.049–0.98); Q3, OR=0.11 (95% CI, 0.024–0.55); and Q4 OR=0.16 (95% CI, 0.032–0.79). High maternal weight gain during pregnancy and male gender were also demonstrated to be risk factors for macrosomia.

Table II Results of logistic regression models investigating the association between individual miRNA expression and macrosomiaa.

Table II

Results of logistic regression models investigating the association between individual miRNA expression and macrosomiaa.

A, miR-21
Effect	n (%)	χ2	OR (95% CI)	P-value
miR-21 relative expression quartiles				0.025
Q1, ≤0.67	28 (25)		1.00
Q2, 0.68–1.10	27 (24)	5.61	6.67 (1.39–32.05)	0.018
Q3, 1.11–1.95	33 (30)	3.22	4.10 (0.88–19.11)	0.073
Q4, >1.95	23 (21)	8.38	16.19 (2.46–106.68)	0.004
Maternal weight gain during pregnancyb				0.025
Low	16 (16)		1.00
Moderate	34 (34)	5.87	3.20 (0.481–21.25)	0.229
High	55 (50)	3.48	9.78 (1.55–61.90)	0.015
Infant gender				0.002
Female	46 (41)		1.00
Male	65 (59)	9.29	6.36 (1.94–20.90)	0.002

B, miR-143
Effect	n (%)	χ2	OR (95% CI)	P-value
miR-143 relative expression quartiles				0.042
Q1, ≤0.52	27 (24)		1.00
Q2, 0.53–0.90	29 (26)	3.92	0.22 (0.049–0.98)	0.048
Q3, 0.91–1.35	29 (26)	7.33	0.11 (0.024–0.55)	0.007
Q4, >1.35	27 (24)	5.08	0.16 (0.032–0.79)	0.024
Maternal weight gain during pregnancyb				0.010
Low	18 (16)		1.00
Moderate	38 (34)	6.60	3.76 (0.488–28.95)	0.204
High	56 (50)	5.07	14.43 (1.88–110.49)	0.010
Infant gender				0.015
Female	46 (41)		1.00
Male	66 (59)	5.91	4.05 (1.31–12.52)	0.015

a Models adjusted for maternal weight prior to pregnancy, gestational age, delivery method and maternal age. Samples lacking values for one or more covariates were excluded from the model.

b Maternal weight gain during pregnancy was divided according to Institute of Medicine recommendations (22). OR, odds ratio; CI, confidence interval.

Interaction between miR-21 and -143

Low miR-143 and high miR-21 expression levels were associated with an increased risk of developing macrosomia. Therefore, the current study analyzed the interaction between the expression of the two miRNAs and macrosomia risk (Table III). Logistic regression analysis demonstrated no significant difference in the risk of developing macrosomia among samples with high miR-143 and low miR-21, high miR-143 and high miR-21 or low miR-143 and low miR-21 levels. However, low placental miR-143 combined with high miR-21 levels were associated with increased likelihood of macrosomia (OR=26.47, 95% CI, 2.91-240.79). This suggested an interaction between miR-21 and -143 expression in macrosomia risk.

Table III Interaction between miR-143 and miR-21 expression and macrosomiaa.

Table III

Interaction between miR-143 and miR-21 expression and macrosomiaa.

Effect	n (%)	χ2	OR (95% CI)	P-value
miR-143				0.005
High	54 (48.6)		1.00
Low	57 (51.4)	7.74	5.87 (1.69–20.44)	0.005
miR-21				0.018
Low	57 (51.4)		1.00
High	54 (48.6)	5.64	4.59 (1.31–16.15)	0.018
miR-143 and miR-21				0.032b
High miR-143, low miR-21	17 (15.3)		1.00
High miR-143, high miR-21	37 (33.4)	1.92	3.57 (0.59–21.63)	0.166
Low miR-143, low miR-21	40 (36.0)	2.65	4.53 (0.74–27.95)	0.103
Low miR-143, high miR-21	17 (15.3)	8.46	26.47(2.91–240.79)	0.004

a Models adjusted for maternal weight prior to pregnancy, maternal weight gain during pregnancy, gestational age, delivery method, infant gender, and maternal age. Samples lacking one or more covariate values were excluded from the model. Expression levels were cut-off by median.

b P-value for trend. OR, odds ratio; CI, confidence interval.

Correlation between miR-143 and miR-21 expression

As low miR-143 and high miR-21 levels were demonstrated to be associated with macrosomia risk, the current study analyzed whether miR-143 and -21 levels were inversely correlated in macrosomic placenta. However, the microRNAs were positively correlated in macrosomia samples (Spearman rs= 0.43; P<0.001; Fig. 2A) Levels of miR-143 and miR-21 were not correlated in the control group (P>0.05; Fig. 2B).

Figure 2 (A) Positive correlation between miR-21 and miR-143 expression in placenta from macrosomic pregnancies (P<0.001). (B) No correlation was observed in the control group (P>0.05).

Pathway analysis

To understand the potential mechanisms by which miR-143 and -21 function in shaping macrosomia risk, the present study identified their potential target genes using miRTarBase, a database that provides experimentally validated miRNA-target interactions. The database suggested 489 target genes for miR-21 and 20 for miR-143 (Table IV). Potential signaling pathways associated with the target genes, generated by DAVID, are presented in Fig. 3. Identified target genes for miR-21 and -143 included those associated with the mitogen-activated protein kinase (MAPK) signaling pathway, an important pathway in cellular proliferation, growth and apoptosis.

Figure 3 Possible pathways regulated by (A) miR-21 and (B) miR-143. The number of enriched genes for each pathway is indicated in parentheses on the vertical axis. The horizontal axis denotes enrichment score log10 (P-value). For each pathway shown, P<0.05. TGF-β, transforming growth factor-β; Jak-STAT, Janus kinase-signal transducer and activator of transcription; MAPK, mitogen-activated protein kinase; GnRH, gonadotropin releasing hormone; VEGF, vascular endothelial growth factor.

Table IV Representative target genes of miR-21 and miR-143.

Table IV

Representative target genes of miR-21 and miR-143.

miR-21 target gene		miR-143 target gene
Gene ID	Official full name	Gene ID	Official full name
PLPP1	Phospholipid phosphatase 1	KRAS	Kirsten rat sarcoma viral oncogene homolog
PLD2	Phospholipase D2	KLF4	Kruppel-like factor 4
BCAT1	Branched chain amino-acid transaminase 1	MYO6	Myosin VI
LARS	Leucyl-tRNA synthetase	DNMT3A	DNA (cytosine-5-)-methyltransferase 3α
PDHA2	Pyruvate dehydrogenase (lipoamide) α2	FNDC3B	Fibronectin type III domain containing 3B
TGFB1	Transforming growth factor β1	MAPK7	Mitogen-activated protein kinase 7
TGFBR2	Transforming growth factor β receptor II	COX2	Cytochrome c oxidase subunit II
SP1	Sp1 transcription factor	COL1A1	Collagen, type I, α1
MYC	v-myc avian myelocytomatosis viral oncogene homolog	HRAS	Harvey rat sarcoma viral oncogene homolog
FAS	Fas cell surface death receptor	FSCN1	Fascin actin-bundling protein 1
IL1B	Interleukin 1β	HK2	Hexokinase 2
BCL2	B-cell CLL/lymphoma 2	SERPINE1	Serpin peptidase inhibitor, clade E, member 1
STAT3	Signal transducer and activator of transcription 3	FHIT	Fragile histidine triad
PIAS3	Protein inhibitor of activated STAT3	MACC1	Metastasis associated in colon cancer 1
AKT2	v-akt murine thymoma viral oncogene homolog 2	PTGS2	Prostaglandin-endoperoxide synthase 2
PTEN	Phosphatase and tensin homolog	JAG1	Jagged 1
CDK6	Cyclin-dependent kinase 6	AKT1	v-akt murine thymoma viral oncogene homolog 1
MAPK3	Mitogen-activated protein kinase 3	MDM2	MDM2 proto-oncogene E3 ubiquitin protein ligase
WNT5A	Wingless-type MMTV integration site family member 5A	BCL2	B-cell CLL/lymphoma 2
EGFR	Epidermal growth factor receptor	MMP13	Matrix metallopeptidase 13

Discussion

As the mediator between mother and fetus, the placenta secretes various hormones, delivers nutrients to the fetus, facilitates gas exchange and serves as a barrier to protect the fetus from adverse environmental conditions, thus, is crucial for fetal development (26). Previous studies have demonstrated that aberrant miRNA expression in the placenta was associated with preeclampsia (19), intrauterine growth retardation and large birth size (18). The results of the current study indicated that dysregulation of miR-21 and -143 expression was associated with macrosomia, and that there was an association between the two miRNAs with regards to risk of macrosomia. Collectively, the evidence suggested that aberrant placental miRNA expression serves a prominent role in abnormal fetal development.

miR-21 is established as an oncogene that is important during tumor progression (27). In previous studies, the upregulation of miR-21 in esophageal squamous cell carcinoma induced cell proliferation and invasion (28), and anti-miR-21 inhibited cell growth in breast tumor tissue (29). In addition, aberrant expression of miR-21 has been demonstrated to affect apoptosis and metastasis in glioma and colorectal cancer cells (30,31). Placental and cancer cells share numerous similar features, including proliferative, invasive and migratory capacity, and shared epigenetic mechanisms for regulating trophoblast invasion (32,33). Thus, the present study hypothesized that miR-21 may affect placental function by influencing cellular processes such as proliferation, thereby affecting fetal development. The data demonstrated that placental miR-21 expression was upregulated in macrosomia cases, which is consistent with a previous investigation (34). Maccani et al (20) reported that low placental miR-21 expression was a risk factor for small for gestational age (SGA). Similarly, a study in mice demonstrated that miR-21 inhibition reduced obesity through the reduction of adipocyte size (35). Other reports have indicated that miR-21 regulated adipogenesis in human primary adipocytes and that its expression was positively correlated with BMI (36). miR-21 was also reported to regulate adipogenic differentiation and adipose proliferation (37,38). Therefore, the present study hypothesized that the upregulation of placental miR-21 leads to macrosomia by promoting cell differentiation and proliferation. However, further experimental evidence is required to evaluate this.

To date, the role of miR-143 in the placenta has remained unknown. To the best of our knowledge, the current study was the first to demonstrate that placental miR-143 was downregulated in macrosomia. Among the miRNAs associated with human obesity, miR-143 was the first reported to regulate adipocyte differentiation (39). Chen et al (21) observed that decreased miR-143 expression promoted adipogenic growth and differentiation of adipose tissue-derived stromal cells by targeting MAPK kinase 5. Compared with control placenta, placenta from macrosomic pregnancies has greater capacity to support growth and differentiation, as it must provide more space and transport more nutrients to a larger fetus (40). Therefore, it can be explained why miR-143 expression was reduced in macrosomic placenta. Other research has reported miR-143 to be a pivotal regulator of cellular activities, targeting multiple mRNAs involved in cell proliferation, survival and apoptosis. For example, overexpression of miR-143 repressed cell proliferation in melanoma and gastric cancer by targeting syndecan-1 (41) and cyclooxygenase-2 (42), which are also expressed in the placenta (43,44). This may partially explain the downregulation of miR-143 in macrosomic placenta demonstrated in the current study. Consistent with the present results, another investigation demonstrated that miR-143 was downregulated in maternal plasma in macrosomic pregnancies (45). However, an association between miR-143 in maternal plasma and placenta remains unclear.

Another notable observation of the current study was the opposing changes to the miR-21 and -143 levels in association with macrosomia. Similarly, Keller et al (36) reported that the expression of miR-21 was positively correlated with BMI. Expression of miR-21 was higher in the adipocyte tissue of adults with BMI>30 than in that of adults with lower BMIs, whereas miR-143 expression was reduced adults with lower BMIs. The same trend was also been demonstrated in colorectal carcinoma and colorectal liver metastasis tissues, which exhibit overexpression of miR-21 and underexpression of miR-143 relative to control tissues (46). These results suggest a functional correlation between miR-21 and -143 expression, as does the relationship identified by the logistic regression analysis in the current study.

According to DAVID analysis (Fig. 3), miR-21 and -143 were associated with MAPK signaling, which is a pivotal pathway in cell proliferation and differentiation. miR-143 has been previously reported to target extracellular signal-regulated kinase 5 (ERK5), a member of the MAPK family that promotes cell proliferation and growth. Overexpression of miR-143 reduced ERK5 expression and activation, thus inhibiting cell proliferation and growth in human colon carcinoma (47). By contrast, miR-21 promoted proliferation of Eca109 cells and inhibited apoptosis by activating the ERK1/2/MAPK pathway (48). Together, these results predict an inverse correlation between levels of miR-143 and -21. Unexpectedly, the current study demonstrated a positive correlation between the two miRNAs in macrosomic but not in control placenta, suggesting that a more complex mechanism co-regulates miR-21 and -143. One such plausible mechanism is proposed as follows: A restrictive correlation between miR-21 and -143 in the regulation of fetal weight. Essentially, the two miRNAs independently and inversely regulate growth in fetuses of normal weight, but in macrosomia the two have a mutually restraining relationship, that is, with the risk factor miR-21 increased, the protective factor miR-143 was also increased, to protect the fetus from overgrowth. This may be a homeostatic mechanism to maintain normal growth in the fetus.

Previous studies have demonstrated that miR-16 targets genes in the pathways regulating numerous cellular processes (49,50). Low levels of placental miR-16 expression are associated with SGA (20). However, the current study did not identify significant differences in placental miR-16 expression between macrosomic and control placenta. In addition, the present study demonstrated that male infants were at a greater risk of developing macrosomia than females. This result is consistent with a number previous reports (11,51). One explanation for this result is that male embryos have a higher rate of cell division and metabolism than female embryos (52).

It is noted that this was a hospital-based case-control study. All subjects were recruited from a large, comprehensive hospital that serves all people in the Wenzhou region. Thus, our data are likely to be representative of the population in the Wenzhou region. Regardless, due to the small sample size, the confidence intervals were relatively large.

In summary, the present study demonstrated that low miR-143 expression and high miR-21 expression are risk factors for macrosomia. Furthermore, there is an association between the two miRNAs and macrosomia. These results provide evidence for the effect of aberrant placental miRNA expression in macrosomia and offer valuable insight into the mechanisms underlying macrosomia. Future studies are required to determine the functional mechanism linking placental miRNA expression to macrosomia and may provide novel information on the potential use of miRNAs as a marker of abnormal fetal growth.

Acknowledgments

The National Natural Science Foundation of China (no. 81072378), and the Natural Science Funds of Zhejiang (no. Y2101185). We gratefully acknowledge the placenta samples provided by the Department of Obstetrics, Yuying Children's Hospital of Wenzhou Medical University. We also thank the residents of the Wenzhou region for their support for our epidemiological study.

References