IJG-08264; No of Pages 5 International Journal of Gynecology and Obstetrics xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo
CLINICAL ARTICLE
Profiling maternal plasma microRNA expression in early pregnancy to predict gestational diabetes mellitus Yanan Zhu a, Fei Tian a, Hailing Li b, Youxia Zhou a, Jiafeng Lu c, Qinyu Ge a,⁎ a b c
Key Lab for Child Development and Learning Science, Ministry of Education, Research Center for Learning Science, Southeast University, Nanjing, China Department of Gynecology and Obstetrics, Zhongda Hospital, Southeast University, Nanjing, China State Key Lab of Bioelectronics, Southeast University, Nanjing, China
a r t i c l e
i n f o
Article history: Received 4 September 2014 Received in revised form 20 December 2014 Accepted 10 March 2015 Keywords: Gestational diabetes mellitus microRNA Next-generation sequencing Profiling
a b s t r a c t Objective: To profile the differential expression of plasma miRNAs in gestational diabetes mellitus (GDM). Methods: In a pilot study conducted at a tertiary hospital in China between 2010 and 2014, peripheral blood samples were collected from women at 16–19 weeks of pregnancy. Pooled samples from 10 women who were subsequently diagnosed with GDM and from 10 healthy controls were used to construct two small RNA libraries. High-throughput sequencing was performed, and differentially expressed miRNAs were validated by quantitative real-time polymerase chain reaction (qRT-PCR), followed by target prediction, Gene Ontology analysis, and pathway identification. Results: Sequencing revealed 32 miRNAs that were differentially expressed in GDM, including 12 miRNAs that were upregulated and 20 that were downregulated. Differential expression of five upregulated miRNAs (hsa-miR-16-5p, hsa-miR-17-5p, hsa-miR-19a-3p, hsa-miR-19b-3p, hsa-miR-20a-5p) was confirmed by qRT-PCR. Target prediction showed that the major targets of these miRNAs were associated with insulin resistance and abnormal pregnancies. Conclusion: The five miRNAs that were differentially expressed in GDM could serve as noninvasive biomarkers. The results also provide insights into the molecular mechanisms that underlie GDM, thereby contributing to the diagnosis and treatment of this disease. © 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics.
1. Introduction Gestational diabetes mellitus (GDM) is diagnosed when glucose intolerance is identified for the first time during pregnancy. The incidence of this condition varies from region to region, but it affects approximately 3%–8% of all pregnancies. GDM increases maternal and fetal morbidity and mortality: it is associated with maternal hypertension, polyhydramnios, neonatal hypoglycemia, postpartum hemorrhage, and other perinatal complications [1]. It is also a major contributor to macrosomia [2]. Moreover, half of women with GDM will develop type 2 diabetes within 22–28 years of delivery [3]. Although there is no standard method for the screening and diagnosis of GDM, the oral glucose tolerance test (OGTT) is widely used in clinical practice. The test is usually performed at 24–28 weeks of pregnancy, although it is often delayed until 32 weeks [4], which leaves little time for women with an increased risk of GDM to make changes to their diet and exercise regimens, or to start taking medication. Additionally, the International Association of Diabetes and Pregnancy Study Groups determined a new glycemic threshold for the diagnosis of GDM in ⁎ Corresponding author at: Key Lab for Child Development and Learning Science, Ministry of Education, Research Center for Learning Science, Southeast University, Sipailou No.2, Nanjing, 210096, China. Tel./fax: +86 25 83792396. E-mail address: [email protected] (Q. Ge).
2008 [5]. Use of the new diagnostic criteria will triple the number of women who are diagnosed with GDM because the diagnosis can be made on the basis of one blood glucose test result, leading to excessive medical intervention. It is therefore important to investigate novel risk assessment approaches for GDM in early pregnancy. MicroRNAs (miRNAs) are a class of small noncoding RNA molecules 18–22 nucleotides in length [6]. They regulate gene expression and transcription of up to 50% of genes at the post-transcriptional level by interacting with target mRNA [7]. Since the first miRNA was discovered in 1993, the functional characteristics and working patterns of miRNAs have attracted the attention of many researchers. So far, more than 800 miRNAs have been discovered in animal cells and it has become clear that they are involved in many biological processes, including cell proliferation and differentiation, metastasis, and apoptosis [8,9]. Evidence is emerging that abnormal expression of miRNAs is associated with pregnancy complications, such as pre-eclampsia and GDM [10]. Zhao et al. [1] investigated the serum miRNA expression profile in pregnant women who were subsequently diagnosed with GDM. They found that miR-29a, miR-222, and miR-132 were significantly downregulated in these women. Quantitative real-time polymerase chain reaction (qRT-PCR) and hybridization-based microarray platforms have been used to identify miRNA aberrations [1,2]. Yet these technologies only measure relatively abundant and known miRNA sequences, and have limited capacity for
http://dx.doi.org/10.1016/j.ijgo.2015.01.010 0020-7292/© 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics.
Please cite this article as: Zhu Y, et al, Profiling maternal plasma microRNA expression in early pregnancy to predict gestational diabetes mellitus, Int J Gynecol Obstet (2015), http://dx.doi.org/10.1016/j.ijgo.2015.01.010
2
Y. Zhu et al. / International Journal of Gynecology and Obstetrics xxx (2015) xxx–xxx
identifying novel miRNAs with aberrations that are associated with GDM. Next-generation deep sequencing uses massively parallel sequencing, generating millions of small RNA sequence reads from one sample. It not only measures absolute abundance, but also enables the discovery of novel miRNAs [11]. The application of this technique could lead to a more comprehensive identification of miRNAs that are differentially expressed in GDM. The aim of the present study was to profile the expression of plasma miRNAs in women with GDM using next-generation sequencing technology, and to identify differentially expressed miRNAs that could serve as biomarkers for the early diagnosis of GDM. 2. Materials and methods A pilot study was conducted at Zhongda Hospital, Southeast University, Nanjing, China, between June 1, 2010, and January 31, 2014. Written informed consent was obtained from all participants and the study protocol was approved by the Ethics Committee of Zhongda Hospital. Peripheral blood samples (2 mL) were collected from women at 16–19 weeks of pregnancy. At 24–28 weeks of pregnancy, a 50-g glucose challenge test (GCT) was conducted. Women with an abnormal 1-h post-GCT glucose level (≥7.8 mmol/L) were recommended to undergo a 3-hour, 75-g OGTT, during which the blood glucose level would be tested four times after intake of a cola-like drink (at 0, 1, 2, and 3 hours). Women were deemed to have GDM when at least two of the four concentrations measured were above the cutoff values (10.3 mmol/L at 0 hours, 8.6 mmol/L at 1 hour, 6.7 mmol/L at 2 hours, and 5.8 mmol/L at 3 hours [12]). A total of 10 women with GDM and 10 controls—matched for maternal age and length of pregnancy at the time of blood collection—were selected for miRNA profiling. All the selected patients had an uncomplicated singleton pregnancy with no fetal malformations, and no history of diabetes. Women who had a 1-h post-GCT glucose level of less than 7.8 mmol/L were eligible for inclusion as controls. For the miRNA profiling, plasma was separated from the peripheral blood samples obtained at 16–19 weeks of pregnancy by centrifugation at 16 000 g for approximately 10 minutes. The plasma samples of the 10 cases were pooled, as were those of the 10 controls, using 200 μL of each individual plasma sample [1]. The pooled plasma was immediately submitted to RNA extraction or frozen to − 20 °C for short-term storage. The miRNeasy Serum/Plasma Kit (Qiagen, Hilden, Germany) was used to extract miRNA from 500 μL of the pooled plasma samples, following the manufacturer’s instructions with some modifications. The quantity and quality of the obtained miRNAs were evaluated with the ND-1000 spectrophotometer (NanoDrop, Rockland, DE, USA). The concentration of the RNA fraction was determined using the Qubit RNA HS Assay Kit and the Qubit 2.0 Fluorometer (Life Technologies, Grand Island, NY, USA) with strict adherence to the manufacturer’s protocol. Subsequently, an miRNA library was prepared with Ion Total RNASeq Kit v2 (Life Technologies, Grand Island, NY, USA) on the basis of the manufacturer’s protocol with some minor modifications. The procedure involved ligation of the sample RNA to adapters followed by reverse transcription, library pre-amplification, size selection, and finally purification of the small RNA library. Ion Torrent (Life Technologies, Grand Island, NY, USA) high-throughput sequencing technology was employed to sequence the small RNA transcripts in these libraries. To prepare the data for analysis, sequences from impurities were removed, followed by removal of the adapter sequences and inferiorquality reads. Subsequently, the length distribution of the sequences was determined and the sequences were mapped to reference genomic data and Rfam version 11.0 (http://rfam.xfam.org). Clean miRNA sequences that mapped to sequences of multiple reference genes were removed. The remaining miRNA sequences were aligned to the miRNA precursors in miRBase 18 (http://www.mirbase.org). The Short
Oligonucleotide Analysis Package version 2.20 (http://soap.genomics. org.cn/) was used to analyze the unique mappable reads. For matching sequences, information was collected on the structure of the known miRNA precursor as well as on the read length and number of reads in the sample. Because data for miRNA sequences with low read counts are less reliable, only miRNA sequences with more than five copies were retained to compose a reliable expression profile. Digital Gene Expression II (Illumina, San Diego, CA, USA) was used to compare miRNA expression in the GDM and control groups [13]. For each miRNA, the number of unambiguous reads was calculated and this figure was then normalized to the number of transcripts per million clean reads. Three parameters were considered in evaluating the significance of expression differences: the original number of read counts, the P value (based on the Poisson distribution), and the logarithmic fold change (log2 ratio) of reads counts. Only miRNAs with P b 0.01, a fold change of more than 1 (sample/control expression ratio more than two), and an absolute read count of more than 100 were considered to be differentially expressed in the two groups. To validate differentially expressed miRNAs, qRT-PCR was performed on the Applied Biosystems 7500 Real-Time PCR System (Life Technologies, Grand Island, NY, USA) using SYBR Premix Ex Taq II polymerase (TaKaRa, Dalian, China) and stem–loop primers from Invitrogen (Shanghai, China). All miRNAs were measured in triplicate with inclusion of a no-template control. Normalization was performed with miR-221 serving as endogenous control [14], and the relative expression of the target miRNAs was determined with the comparative quantification cycle (Cq) (2–ΔΔCq) method, in which ΔΔCq was calculated by subtracting the mean ΔCq value of the control group from the mean ΔCq value of the GDM group, and ΔCq was calculated by subtracting the Cq value of the endogenous control from the Cq value of the sample. The TargetScan, PicTar, and miRanda algorithms were used in combination to identify the potential target genes of validated miRNAs that were differentially expressed. The resulting data were uploaded into the Database for Annotation, Visualization, and Integrated Discovery version 6.7 (http://david.abcc.ncifcrf.gov/home.jsp) for Gene Ontology analysis and pathway identification using the Kyoto Encyclopedia of Genes and Genomes database. Gene Ontology terms and Kyoto Encyclopedia of Genes and Genomes pathways with a false discovery rate (FDR) below 0.01 were considered as significant for further analysis. FDR control is a statistical method used in multiple hypothesis testing to correct for multiple comparisons. SPSS version 17.0 (SPSS Inc, Chicago, IL, USA) was used for the statistical analysis. Demographic and clinical characteristics were compared using the t test. P values for the significance of differential miRNA expresy ðxþyÞ! sion were calculated using the formula P ðyjxÞ ¼ N2 ðxþyþ1Þ , N1 x!y!ð1þN2 N1Þ where N1 and N2 are the total number of clean reads in the GDM and control groups, respectively, and x and y are the number of reads of a specific miRNA in the GDM and control groups, respectively. P ≤ 0.05 was considered statistically significant. 3. Results Blood samples from 10 women with GDM and 10 controls were assessed. The two groups were matched for age, with a mean age of 30.03 years (range 23–35 years) in the GDM group and a mean age of 26.67 years (range 23–34 years) in the control group. The body mass index in the GDM group was significantly higher than that in the control group (P = 0.01) (Table 1). The percentage of mappable miRNA sequences was 52.42% in the GDM group and 56.86% in the control group. The number of reads (expression level) per unique sequence varied between 1 and 286 821. In total, 187 miRNAs from maternal plasma were sequenced in the GDM group, and 156 miRNAs were sequenced in the control group. Most sequenced small RNAs were 21–23 nucleotides long, which corresponds to the average length of miRNAs after Dicer digestion
Please cite this article as: Zhu Y, et al, Profiling maternal plasma microRNA expression in early pregnancy to predict gestational diabetes mellitus, Int J Gynecol Obstet (2015), http://dx.doi.org/10.1016/j.ijgo.2015.01.010
Y. Zhu et al. / International Journal of Gynecology and Obstetrics xxx (2015) xxx–xxx Table 1 Characteristics of the patients selected for profiling at blood sample collection.a Variable
Age, y Body mass indexb Pregnancy duration, wk a b
Participants with gestational diabetes mellitus (n = 10)
Controls (n = 10)
30.03 ± 3.56 23.94 ± 2.98 17.66 ± 0.85
26.67 ± 4.59 19.24 ± 1.07 18.17 ± 0.93
P value
0.084 0.001 0.213
Values are given as mean ± SD unless indicated otherwise. Calculated as weight in kilograms divided by the square of height in meters.
(Supplementary Material S1). miRNAs accounted for most small RNAs in the two samples (Supplementary Material S1). The hierarchical clustering analysis of miRNA expression in the GDM groups compared with the control group is shown in Supplementary Material S2. In total, 32 miRNAs were differentially expressed, including 12 miRNAs that were upregulated among women with GDM and 20 miRNAs that were downregulated (Fig. 1A). The fold change between the two groups was greater than two for 10 miRNAs. Five differentially expressed miRNAs (Table 2) were selected for sequence validation by qRT-PCR, on the basis of the number of reads and the fold change. All five miRNAs were upregulated. The qRT-PCR results demonstrated a high concordance with the miRNA sequencing data (Fig. 1B), indicating that the sequencing results were reliable. The target genes of the five upregulated miRNAs had roles in small GTPase mediated signal transduction, actin cytoskeleton organization, and insulin receptor signaling (FDR b 0.01) (Fig. 2A). Pathway analysis using the integrated results from TargetScan, PicTar, and miRanda resulted in a rank-ordered list of 18 pathways with significant enrichment of the predicted miRNA target genes (Fig. 2B). These pathways included endocytosis, mitogen-activated protein kinase (MAPK) signaling, insulin signaling, mTOR signaling, type 2 diabetes, Wnt signaling, proteoglycans in cancer, and transforming growth factor-β (TGF-β) signaling. On the basis of these findings, a regulatory network involving the five differentially expressed miRNAs was constructed, including seven target genes and five pathways (Fig. 3). 4. Discussion The present study led to the identification of a plasma miRNA signature that predicts GDM in the early second trimester: five miRNAs (miR-16-5p, miR-17-5p, miR-19a-3p, miR-19b-3p, and miR-20a-5p) were aberrantly expressed before serum glucose abnormality. Network analysis revealed that these miRNAs are mainly associated with five
3
Table 2 Differentially expressed miRNAs in pregnant women who subsequently developed gestational diabetes mellitus compared with healthy controls. miRNA
Fold change (log2 sample/control)
P value
hsa-miR-16-5p hsa-miR-17-5p hsa-miR-19a-3p hsa-miR-19b-3p hsa-miR-20a-5p
1.832058 1.415629 1.246886 1.314308 1.203325
5.36E-11 1.10E-10 6.57E-43 1.73E-74 5.27E-37
pathways: MAPK signaling, insulin signaling, type 2 diabetes mellitus, TGF-β signaling, and mTOR signaling. Similar to other forms of hyperglycemia, GDM is characterized by insulin levels that are insufficient to meet insulin demands [15]. The five miRNAs that were upregulated in GDM were all associated with insulin secretion. For example, miR-16 targets genes encoding the insulin receptor substrate (IRS) proteins 1 and 2. These adaptor proteins mediate insulin-like growth factor-I (IGF-I)/insulin signaling; they mainly occur in insulin-sensitive tissues (e.g. fat, bone, and liver) and are involved in various biological processes including diabetes, metastasis, and adipocyte and bone differentiation [16,17]. Following activation by insulin/IGF-I receptors, IRS1 and IRS2 recruit intracellular proteins containing Src Homology 2 (SH2) domains and initiate signal transduction cascades such as the phosphoinositide-3 kinase (PI3K)–AKT and Ras–MAPK pathways [18]. In addition, IRS1 and IRS2 promote Wnt/β-catenin signaling [19], which is critical for cell growth. Abnormal Wnt/β-catenin signaling has been linked with cancer, obesity, and diabetes [19]. These findings indicate that miR-16 might have an important role in the development of insulin insufficiency. Upregulation of miR-16 will result in downregulation of IRS1 and IRS2, which might in turn lead to aberrant Wnt/βcatenin signaling and eventually diabetes. Downregulation of IRS1 and IRS2 will also block insulin signaling, causing insulin resistance. Severe insulin resistance is a sign of GDM [20,21]. The MAPK signaling pathway plays an important role in the development of vascular lesions such as high blood pressure and diabetes [22]. Extracellular signal-regulated kinase (ERK) 1 and ERK2 from the MAPK family have a high sensitivity for insulin. In addition, abnormal MAPK signaling is associated with pregnancy complications. For example, the expression of phosphorylated MAPKs—phosphorylated ERK, phosphorylated Jun N-terminal kinase (JNK), and phosphorylated p38—is increased in fetal membranes close to the cervix in preterm pregnant women [23]. Activation of the p38 and JNK pathways can lead to the activation of various transcription factors, raising the expression of some genes that can lead to cell damage (e.g. those for NF-κB, and JUN and
Fig. 1. Differential miRNA expression in GDM. (A) Log2 fold change of miRNA expression between the GDM group and the control group; (B) Validation of differentially expressed miRNAs by qRT-PCR. Abbreviations: GDM, gestational diabetes mellitus; NGS, next-generation sequencing; qRT-PCR, qualitative real-time polymerase chain reaction.
Please cite this article as: Zhu Y, et al, Profiling maternal plasma microRNA expression in early pregnancy to predict gestational diabetes mellitus, Int J Gynecol Obstet (2015), http://dx.doi.org/10.1016/j.ijgo.2015.01.010
4
Y. Zhu et al. / International Journal of Gynecology and Obstetrics xxx (2015) xxx–xxx
Fig. 2. Gene Ontology and pathway analysis for the predicted targets of five miRNAs that were differentially expressed in gestational diabetes mellitus. (A) Gene Ontology categories enriched with predicted targets; (B) Kyoto Encyclopedia of Genes and Genomes pathways enriched with predicted targets.
FOS) and decreasing expression of some protective genes (e.g. BCL2 and HSPB1). The TGF-β signaling pathway is associated with pre-eclampsia [24], and the mTOR signaling pathway is involved in the control of energy balance and food intake in the hypothalamus [25]. Blocking of these two pathways might also contribute to the development of GDM. Of the five miRNAs that were upregulated in GDM, three (miR146b-5p, miR-20a, and miR-17) are also upregulated in pre-eclampsia [26,27]. Like GDM, pre-eclampsia is a disease of pregnancy and the pathogenesis of these two conditions is related to a certain extent. Both women with pre-eclampsia and those with GDM have a higher degree of insulin resistance than do healthy pregnant women [28]. Four of the five miRNAs (miR-17-5p, miR-19a-3p, miR-19b-3p, and miR-20a-5p) belong to the miRNA-17-92 cluster. This cluster is associated with angiogenesis, which is a key physiological process during pregnancy. Upregulation of these miRNAs might lead to an abnormal pregnancy. The strengths of the present study include the use of a highthroughput sequencing platform and the validation of miRNA sequences
MAPK signaling pathway
by qRT-PCR. However, the study also has a few limitations. Firstly, the sample size was small, which may have caused unreliable results. Secondly, only some of the miRNA sequences were validated by qRTPCR, and the qRT-PCR was performed on pooled samples. Further studies are needed that include validation of a larger number of miRNAs using individual samples. Finally, the target genes and pathways were all predicted by software and the conclusions about the roles of the differentially expressed miRNAs in these pathways were made on the basis of findings from previous studies. More advanced functional studies are needed to better understand the functions of these miRNAs. In summary, the present study demonstrates the use of deep sequencing for the comprehensively profiling of miRNAs in the plasma of women with GDM. The study does not only confirm existing findings, but it also led to the discovery of dysregulated miRNAs that were not known to be associated with GDM, thereby providing a new miRNA signature for the detection of early-stage GDM. The blood samples were collected at 16–19 weeks of pregnancy, before GDM was diagnosed. The identified miRNAs constitute promising biomarkers that might help to improve the early detection and monitoring of GDM. Moreover,
Insulin signaling pathway
IRS-1
MAPK-1
miR-19a
miR-16
SMAD5
Type 2 diabetes mellitus
SOS-1
IRS-2
miR-19b
SMAD4
TGF-β signaling pathway
miR-20a
miR-17
AKT3
mTOR signaling pathway
Fig. 3. Gene regulatory network involving five miRNAs that were differentially expressed in gestational diabetes mellitus.
Please cite this article as: Zhu Y, et al, Profiling maternal plasma microRNA expression in early pregnancy to predict gestational diabetes mellitus, Int J Gynecol Obstet (2015), http://dx.doi.org/10.1016/j.ijgo.2015.01.010
Y. Zhu et al. / International Journal of Gynecology and Obstetrics xxx (2015) xxx–xxx
the present insights into the molecular mechanisms that underlie GDM could aid the diagnosis and treatment of the disease. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ijgo.2015.01.010. Acknowledgments This work was supported by project 61271055 of the National Natural Science Foundation of China. Conflict of interest The authors have no conflicts of interest. References [1] Liang Z, Dong M, Cheng Q, Chen D. Gestational diabetes mellitus screening based on the gene chip technique. Diabetes Res Clin Pract 2010;89(2):167–73. [2] HAPO Study Cooperative Research Group, Metzger BE, Lowe LP, Dyer AR, Trimble ER, Chaovarindr U, et al. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 2008;358(19):1991–2002. [3] England LJ, Dietz PM, Njoroge T, Callaghan WM, Bruce C, Buus RM, et al. Preventing type 2 diabetes: public health implications for women with a history of gestational diabetes mellitus. Am J Obstet Gynecol 2009;200(4):365.e1–8. [4] Tieu J, Middleton P, McPhee AJ, Crowther CA. Screening and subsequent management for gestational diabetes for improving maternal and infant health. Cochrane Database Syst Rev 2010;7:CD007222. [5] Holt RI, Coleman MA, McCance DR. The implications of the new International Association of Diabetes and Pregnancy Study Groups (IADPSG) diagnostic criteria for gestational diabetes. Diabet Med 2011;28(4):382–5. [6] Sayed D, Abdellatif M. MicroRNAs in development and disease. Physiol Rev 2011; 91(3):827–87. [7] Coordes A, Lenarz M, Kaufmann AM, Albers AE. Role of miRNA in malignoma of the head and neck. Laryngorhinootologie 2014;93(3):201–9. [8] Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 2004; 116(2):281–97. [9] Ambros V. The functions of animal microRNAs. Nature 2004;431(7006):350–5. [10] Zhao Z, Moley KH, Gronowski AM. Diagnostic potential for miRNAs as biomarkers for pregnancy-specific diseases. Clin Biochem 2013;46(10–11):953–60. [11] van Rooij E. The art of microRNA research. Circ Res 2011;108(2):219–34. [12] American Diabetes Association. Standards of medical care in diabetes–2011. Diabetes Care 2011;34(Suppl. 1):S11–61.
5
[13] Audic S, Claverie JM. The significance of digital gene expression profiles. Genome Res 1997;7(10):986–95. [14] Hu J, Wang Z, Liao BY, Yu L, Gao X, Lu S, et al. Human miR-1228 as a stable endogenous control for the quantification of circulating microRNAs in cancer patients. Int J Cancer 2014;135(5):1187–94. [15] Metzger BE, Buchanan TA, Coustan DR, de Leiva A, Dunger DB, Hadden DR, et al. Summary and recommendations of the Fifth International Workshop-Conference on Gestational Diabetes Mellitus. Diabetes Care 2007;30(Suppl. 2):S251–60. [16] Carew RM, Browne MB, Hickey FB, Brazil DP. Insulin receptor substrate 2 and FoxO3a signalling are involved in E-cadherin expression and transforming growth factor-β1-induced repression in kidney epithelial cells. FEBS J 2011;278(18): 3370–80. [17] Sorokin AV, Chen J. MEMO1, a new IRS1-interacting protein, induces epithelialmesenchymal transition in mammary epithelial cells. Oncogene 2013;32(26): 3130–8. [18] Taniguchi CM, Emanuelli B, Kahn CR. Critical nodes in signalling pathways: insights into insulin action. Nat Rev Mol Cell Biol 2006;7(2):85–96. [19] Geng Y, Ju Y, Ren F, Qiu Y, Tomita Y, Tomoeda M, et al. Insulin receptor substrate 1/2 (IRS1/2) regulates Wnt/β-catenin signaling through blocking autophagic degradation of dishevelled2. J Biol Chem 2014;289(16):11230–41. [20] Rao R, Hao CM, Redha R, Wasserman DH, McGuinness OP, Breyer MD. Glycogen synthase kinase 3 inhibition improves insulin-stimulated glucose metabolism but not hypertension in high-fat-fed C57BL/6 J mice. Diabetologia 2007;50(2):452–60. [21] Tsai YL, Hou CW, Liao YH, Chen CY, Lin FC, Lee WC, et al. Exercise training exacerbates tourniquet ischemia-induced decreases in GLUT4 expression and muscle atrophy in rats. Life Sci 2006;78(25):2953–9. [22] Nishiyama A, Yoshizumi M, Rahman M, Kobori H, Seth DM, Miyatake A, et al. Effects of AT1 receptor blockade on renal injury and mitogen-activated protein activity in Dahl salt-sensitive rats. Kidney Int 2004;65(3):972–81. [23] Lappas M, Riley C, Lim R, Barker G, Rice GE, Menon R, et al. MAPK and AP-1 proteins are increased in term pre-labour fetal membranes overlying the cervix: regulation of enzymes involved in the degradation of fetal membranes. Placenta 2011;32(12): 1016–25. [24] Perucci LO, Gomes KB, Freitas LG, Godoi LC, Alpoim PN, Pinheiro MB, et al. Soluble endoglin, transforming growth factor-Beta 1 and soluble tumor necrosis factor alpha receptors in different clinical manifestations of preeclampsia. PLoS One 2014;9(5):e97632. [25] Kumar A, Lawrence Jr JC, Jung DY, Ko HJ, Keller SR, Kim JK, et al. Fat cell-specific ablation of rictor in mice impairs insulin-regulated fat cell and whole-body glucose and lipid metabolism. Diabetes 2010;59(6):1397–406. [26] Fu G, Brkić J, Hayder H, Peng C. MicroRNAs in Human Placental Development and Pregnancy Complications. Int J Mol Sci 2013;14(3):5519–44. [27] Wang W, Feng L, Zhang H, Hachy S, Satohisa S, Laurent LC, et al. Preeclampsia up-regulates angiogenesis-associated microRNA (i.e., miR-17, − 20a, and -20b) that target ephrin-B2 and EPHB4 in human placenta. J Clin Endocrinol Metab 2012; 97(6):E1051–9. [28] Briana DD, Malamitsi-Puchner A. Reviews: adipocytokines in normal and complicated pregnancies. Reprod Sci 2009;16(10):921–37.
Please cite this article as: Zhu Y, et al, Profiling maternal plasma microRNA expression in early pregnancy to predict gestational diabetes mellitus, Int J Gynecol Obstet (2015), http://dx.doi.org/10.1016/j.ijgo.2015.01.010
Contents lists available at ScienceDirect
International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo
CLINICAL ARTICLE
Profiling maternal plasma microRNA expression in early pregnancy to predict gestational diabetes mellitus Yanan Zhu a, Fei Tian a, Hailing Li b, Youxia Zhou a, Jiafeng Lu c, Qinyu Ge a,⁎ a b c
Key Lab for Child Development and Learning Science, Ministry of Education, Research Center for Learning Science, Southeast University, Nanjing, China Department of Gynecology and Obstetrics, Zhongda Hospital, Southeast University, Nanjing, China State Key Lab of Bioelectronics, Southeast University, Nanjing, China
a r t i c l e
i n f o
Article history: Received 4 September 2014 Received in revised form 20 December 2014 Accepted 10 March 2015 Keywords: Gestational diabetes mellitus microRNA Next-generation sequencing Profiling
a b s t r a c t Objective: To profile the differential expression of plasma miRNAs in gestational diabetes mellitus (GDM). Methods: In a pilot study conducted at a tertiary hospital in China between 2010 and 2014, peripheral blood samples were collected from women at 16–19 weeks of pregnancy. Pooled samples from 10 women who were subsequently diagnosed with GDM and from 10 healthy controls were used to construct two small RNA libraries. High-throughput sequencing was performed, and differentially expressed miRNAs were validated by quantitative real-time polymerase chain reaction (qRT-PCR), followed by target prediction, Gene Ontology analysis, and pathway identification. Results: Sequencing revealed 32 miRNAs that were differentially expressed in GDM, including 12 miRNAs that were upregulated and 20 that were downregulated. Differential expression of five upregulated miRNAs (hsa-miR-16-5p, hsa-miR-17-5p, hsa-miR-19a-3p, hsa-miR-19b-3p, hsa-miR-20a-5p) was confirmed by qRT-PCR. Target prediction showed that the major targets of these miRNAs were associated with insulin resistance and abnormal pregnancies. Conclusion: The five miRNAs that were differentially expressed in GDM could serve as noninvasive biomarkers. The results also provide insights into the molecular mechanisms that underlie GDM, thereby contributing to the diagnosis and treatment of this disease. © 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics.
1. Introduction Gestational diabetes mellitus (GDM) is diagnosed when glucose intolerance is identified for the first time during pregnancy. The incidence of this condition varies from region to region, but it affects approximately 3%–8% of all pregnancies. GDM increases maternal and fetal morbidity and mortality: it is associated with maternal hypertension, polyhydramnios, neonatal hypoglycemia, postpartum hemorrhage, and other perinatal complications [1]. It is also a major contributor to macrosomia [2]. Moreover, half of women with GDM will develop type 2 diabetes within 22–28 years of delivery [3]. Although there is no standard method for the screening and diagnosis of GDM, the oral glucose tolerance test (OGTT) is widely used in clinical practice. The test is usually performed at 24–28 weeks of pregnancy, although it is often delayed until 32 weeks [4], which leaves little time for women with an increased risk of GDM to make changes to their diet and exercise regimens, or to start taking medication. Additionally, the International Association of Diabetes and Pregnancy Study Groups determined a new glycemic threshold for the diagnosis of GDM in ⁎ Corresponding author at: Key Lab for Child Development and Learning Science, Ministry of Education, Research Center for Learning Science, Southeast University, Sipailou No.2, Nanjing, 210096, China. Tel./fax: +86 25 83792396. E-mail address: [email protected] (Q. Ge).
2008 [5]. Use of the new diagnostic criteria will triple the number of women who are diagnosed with GDM because the diagnosis can be made on the basis of one blood glucose test result, leading to excessive medical intervention. It is therefore important to investigate novel risk assessment approaches for GDM in early pregnancy. MicroRNAs (miRNAs) are a class of small noncoding RNA molecules 18–22 nucleotides in length [6]. They regulate gene expression and transcription of up to 50% of genes at the post-transcriptional level by interacting with target mRNA [7]. Since the first miRNA was discovered in 1993, the functional characteristics and working patterns of miRNAs have attracted the attention of many researchers. So far, more than 800 miRNAs have been discovered in animal cells and it has become clear that they are involved in many biological processes, including cell proliferation and differentiation, metastasis, and apoptosis [8,9]. Evidence is emerging that abnormal expression of miRNAs is associated with pregnancy complications, such as pre-eclampsia and GDM [10]. Zhao et al. [1] investigated the serum miRNA expression profile in pregnant women who were subsequently diagnosed with GDM. They found that miR-29a, miR-222, and miR-132 were significantly downregulated in these women. Quantitative real-time polymerase chain reaction (qRT-PCR) and hybridization-based microarray platforms have been used to identify miRNA aberrations [1,2]. Yet these technologies only measure relatively abundant and known miRNA sequences, and have limited capacity for
http://dx.doi.org/10.1016/j.ijgo.2015.01.010 0020-7292/© 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics.
Please cite this article as: Zhu Y, et al, Profiling maternal plasma microRNA expression in early pregnancy to predict gestational diabetes mellitus, Int J Gynecol Obstet (2015), http://dx.doi.org/10.1016/j.ijgo.2015.01.010
2
Y. Zhu et al. / International Journal of Gynecology and Obstetrics xxx (2015) xxx–xxx
identifying novel miRNAs with aberrations that are associated with GDM. Next-generation deep sequencing uses massively parallel sequencing, generating millions of small RNA sequence reads from one sample. It not only measures absolute abundance, but also enables the discovery of novel miRNAs [11]. The application of this technique could lead to a more comprehensive identification of miRNAs that are differentially expressed in GDM. The aim of the present study was to profile the expression of plasma miRNAs in women with GDM using next-generation sequencing technology, and to identify differentially expressed miRNAs that could serve as biomarkers for the early diagnosis of GDM. 2. Materials and methods A pilot study was conducted at Zhongda Hospital, Southeast University, Nanjing, China, between June 1, 2010, and January 31, 2014. Written informed consent was obtained from all participants and the study protocol was approved by the Ethics Committee of Zhongda Hospital. Peripheral blood samples (2 mL) were collected from women at 16–19 weeks of pregnancy. At 24–28 weeks of pregnancy, a 50-g glucose challenge test (GCT) was conducted. Women with an abnormal 1-h post-GCT glucose level (≥7.8 mmol/L) were recommended to undergo a 3-hour, 75-g OGTT, during which the blood glucose level would be tested four times after intake of a cola-like drink (at 0, 1, 2, and 3 hours). Women were deemed to have GDM when at least two of the four concentrations measured were above the cutoff values (10.3 mmol/L at 0 hours, 8.6 mmol/L at 1 hour, 6.7 mmol/L at 2 hours, and 5.8 mmol/L at 3 hours [12]). A total of 10 women with GDM and 10 controls—matched for maternal age and length of pregnancy at the time of blood collection—were selected for miRNA profiling. All the selected patients had an uncomplicated singleton pregnancy with no fetal malformations, and no history of diabetes. Women who had a 1-h post-GCT glucose level of less than 7.8 mmol/L were eligible for inclusion as controls. For the miRNA profiling, plasma was separated from the peripheral blood samples obtained at 16–19 weeks of pregnancy by centrifugation at 16 000 g for approximately 10 minutes. The plasma samples of the 10 cases were pooled, as were those of the 10 controls, using 200 μL of each individual plasma sample [1]. The pooled plasma was immediately submitted to RNA extraction or frozen to − 20 °C for short-term storage. The miRNeasy Serum/Plasma Kit (Qiagen, Hilden, Germany) was used to extract miRNA from 500 μL of the pooled plasma samples, following the manufacturer’s instructions with some modifications. The quantity and quality of the obtained miRNAs were evaluated with the ND-1000 spectrophotometer (NanoDrop, Rockland, DE, USA). The concentration of the RNA fraction was determined using the Qubit RNA HS Assay Kit and the Qubit 2.0 Fluorometer (Life Technologies, Grand Island, NY, USA) with strict adherence to the manufacturer’s protocol. Subsequently, an miRNA library was prepared with Ion Total RNASeq Kit v2 (Life Technologies, Grand Island, NY, USA) on the basis of the manufacturer’s protocol with some minor modifications. The procedure involved ligation of the sample RNA to adapters followed by reverse transcription, library pre-amplification, size selection, and finally purification of the small RNA library. Ion Torrent (Life Technologies, Grand Island, NY, USA) high-throughput sequencing technology was employed to sequence the small RNA transcripts in these libraries. To prepare the data for analysis, sequences from impurities were removed, followed by removal of the adapter sequences and inferiorquality reads. Subsequently, the length distribution of the sequences was determined and the sequences were mapped to reference genomic data and Rfam version 11.0 (http://rfam.xfam.org). Clean miRNA sequences that mapped to sequences of multiple reference genes were removed. The remaining miRNA sequences were aligned to the miRNA precursors in miRBase 18 (http://www.mirbase.org). The Short
Oligonucleotide Analysis Package version 2.20 (http://soap.genomics. org.cn/) was used to analyze the unique mappable reads. For matching sequences, information was collected on the structure of the known miRNA precursor as well as on the read length and number of reads in the sample. Because data for miRNA sequences with low read counts are less reliable, only miRNA sequences with more than five copies were retained to compose a reliable expression profile. Digital Gene Expression II (Illumina, San Diego, CA, USA) was used to compare miRNA expression in the GDM and control groups [13]. For each miRNA, the number of unambiguous reads was calculated and this figure was then normalized to the number of transcripts per million clean reads. Three parameters were considered in evaluating the significance of expression differences: the original number of read counts, the P value (based on the Poisson distribution), and the logarithmic fold change (log2 ratio) of reads counts. Only miRNAs with P b 0.01, a fold change of more than 1 (sample/control expression ratio more than two), and an absolute read count of more than 100 were considered to be differentially expressed in the two groups. To validate differentially expressed miRNAs, qRT-PCR was performed on the Applied Biosystems 7500 Real-Time PCR System (Life Technologies, Grand Island, NY, USA) using SYBR Premix Ex Taq II polymerase (TaKaRa, Dalian, China) and stem–loop primers from Invitrogen (Shanghai, China). All miRNAs were measured in triplicate with inclusion of a no-template control. Normalization was performed with miR-221 serving as endogenous control [14], and the relative expression of the target miRNAs was determined with the comparative quantification cycle (Cq) (2–ΔΔCq) method, in which ΔΔCq was calculated by subtracting the mean ΔCq value of the control group from the mean ΔCq value of the GDM group, and ΔCq was calculated by subtracting the Cq value of the endogenous control from the Cq value of the sample. The TargetScan, PicTar, and miRanda algorithms were used in combination to identify the potential target genes of validated miRNAs that were differentially expressed. The resulting data were uploaded into the Database for Annotation, Visualization, and Integrated Discovery version 6.7 (http://david.abcc.ncifcrf.gov/home.jsp) for Gene Ontology analysis and pathway identification using the Kyoto Encyclopedia of Genes and Genomes database. Gene Ontology terms and Kyoto Encyclopedia of Genes and Genomes pathways with a false discovery rate (FDR) below 0.01 were considered as significant for further analysis. FDR control is a statistical method used in multiple hypothesis testing to correct for multiple comparisons. SPSS version 17.0 (SPSS Inc, Chicago, IL, USA) was used for the statistical analysis. Demographic and clinical characteristics were compared using the t test. P values for the significance of differential miRNA expresy ðxþyÞ! sion were calculated using the formula P ðyjxÞ ¼ N2 ðxþyþ1Þ , N1 x!y!ð1þN2 N1Þ where N1 and N2 are the total number of clean reads in the GDM and control groups, respectively, and x and y are the number of reads of a specific miRNA in the GDM and control groups, respectively. P ≤ 0.05 was considered statistically significant. 3. Results Blood samples from 10 women with GDM and 10 controls were assessed. The two groups were matched for age, with a mean age of 30.03 years (range 23–35 years) in the GDM group and a mean age of 26.67 years (range 23–34 years) in the control group. The body mass index in the GDM group was significantly higher than that in the control group (P = 0.01) (Table 1). The percentage of mappable miRNA sequences was 52.42% in the GDM group and 56.86% in the control group. The number of reads (expression level) per unique sequence varied between 1 and 286 821. In total, 187 miRNAs from maternal plasma were sequenced in the GDM group, and 156 miRNAs were sequenced in the control group. Most sequenced small RNAs were 21–23 nucleotides long, which corresponds to the average length of miRNAs after Dicer digestion
Please cite this article as: Zhu Y, et al, Profiling maternal plasma microRNA expression in early pregnancy to predict gestational diabetes mellitus, Int J Gynecol Obstet (2015), http://dx.doi.org/10.1016/j.ijgo.2015.01.010
Y. Zhu et al. / International Journal of Gynecology and Obstetrics xxx (2015) xxx–xxx Table 1 Characteristics of the patients selected for profiling at blood sample collection.a Variable
Age, y Body mass indexb Pregnancy duration, wk a b
Participants with gestational diabetes mellitus (n = 10)
Controls (n = 10)
30.03 ± 3.56 23.94 ± 2.98 17.66 ± 0.85
26.67 ± 4.59 19.24 ± 1.07 18.17 ± 0.93
P value
0.084 0.001 0.213
Values are given as mean ± SD unless indicated otherwise. Calculated as weight in kilograms divided by the square of height in meters.
(Supplementary Material S1). miRNAs accounted for most small RNAs in the two samples (Supplementary Material S1). The hierarchical clustering analysis of miRNA expression in the GDM groups compared with the control group is shown in Supplementary Material S2. In total, 32 miRNAs were differentially expressed, including 12 miRNAs that were upregulated among women with GDM and 20 miRNAs that were downregulated (Fig. 1A). The fold change between the two groups was greater than two for 10 miRNAs. Five differentially expressed miRNAs (Table 2) were selected for sequence validation by qRT-PCR, on the basis of the number of reads and the fold change. All five miRNAs were upregulated. The qRT-PCR results demonstrated a high concordance with the miRNA sequencing data (Fig. 1B), indicating that the sequencing results were reliable. The target genes of the five upregulated miRNAs had roles in small GTPase mediated signal transduction, actin cytoskeleton organization, and insulin receptor signaling (FDR b 0.01) (Fig. 2A). Pathway analysis using the integrated results from TargetScan, PicTar, and miRanda resulted in a rank-ordered list of 18 pathways with significant enrichment of the predicted miRNA target genes (Fig. 2B). These pathways included endocytosis, mitogen-activated protein kinase (MAPK) signaling, insulin signaling, mTOR signaling, type 2 diabetes, Wnt signaling, proteoglycans in cancer, and transforming growth factor-β (TGF-β) signaling. On the basis of these findings, a regulatory network involving the five differentially expressed miRNAs was constructed, including seven target genes and five pathways (Fig. 3). 4. Discussion The present study led to the identification of a plasma miRNA signature that predicts GDM in the early second trimester: five miRNAs (miR-16-5p, miR-17-5p, miR-19a-3p, miR-19b-3p, and miR-20a-5p) were aberrantly expressed before serum glucose abnormality. Network analysis revealed that these miRNAs are mainly associated with five
3
Table 2 Differentially expressed miRNAs in pregnant women who subsequently developed gestational diabetes mellitus compared with healthy controls. miRNA
Fold change (log2 sample/control)
P value
hsa-miR-16-5p hsa-miR-17-5p hsa-miR-19a-3p hsa-miR-19b-3p hsa-miR-20a-5p
1.832058 1.415629 1.246886 1.314308 1.203325
5.36E-11 1.10E-10 6.57E-43 1.73E-74 5.27E-37
pathways: MAPK signaling, insulin signaling, type 2 diabetes mellitus, TGF-β signaling, and mTOR signaling. Similar to other forms of hyperglycemia, GDM is characterized by insulin levels that are insufficient to meet insulin demands [15]. The five miRNAs that were upregulated in GDM were all associated with insulin secretion. For example, miR-16 targets genes encoding the insulin receptor substrate (IRS) proteins 1 and 2. These adaptor proteins mediate insulin-like growth factor-I (IGF-I)/insulin signaling; they mainly occur in insulin-sensitive tissues (e.g. fat, bone, and liver) and are involved in various biological processes including diabetes, metastasis, and adipocyte and bone differentiation [16,17]. Following activation by insulin/IGF-I receptors, IRS1 and IRS2 recruit intracellular proteins containing Src Homology 2 (SH2) domains and initiate signal transduction cascades such as the phosphoinositide-3 kinase (PI3K)–AKT and Ras–MAPK pathways [18]. In addition, IRS1 and IRS2 promote Wnt/β-catenin signaling [19], which is critical for cell growth. Abnormal Wnt/β-catenin signaling has been linked with cancer, obesity, and diabetes [19]. These findings indicate that miR-16 might have an important role in the development of insulin insufficiency. Upregulation of miR-16 will result in downregulation of IRS1 and IRS2, which might in turn lead to aberrant Wnt/βcatenin signaling and eventually diabetes. Downregulation of IRS1 and IRS2 will also block insulin signaling, causing insulin resistance. Severe insulin resistance is a sign of GDM [20,21]. The MAPK signaling pathway plays an important role in the development of vascular lesions such as high blood pressure and diabetes [22]. Extracellular signal-regulated kinase (ERK) 1 and ERK2 from the MAPK family have a high sensitivity for insulin. In addition, abnormal MAPK signaling is associated with pregnancy complications. For example, the expression of phosphorylated MAPKs—phosphorylated ERK, phosphorylated Jun N-terminal kinase (JNK), and phosphorylated p38—is increased in fetal membranes close to the cervix in preterm pregnant women [23]. Activation of the p38 and JNK pathways can lead to the activation of various transcription factors, raising the expression of some genes that can lead to cell damage (e.g. those for NF-κB, and JUN and
Fig. 1. Differential miRNA expression in GDM. (A) Log2 fold change of miRNA expression between the GDM group and the control group; (B) Validation of differentially expressed miRNAs by qRT-PCR. Abbreviations: GDM, gestational diabetes mellitus; NGS, next-generation sequencing; qRT-PCR, qualitative real-time polymerase chain reaction.
Please cite this article as: Zhu Y, et al, Profiling maternal plasma microRNA expression in early pregnancy to predict gestational diabetes mellitus, Int J Gynecol Obstet (2015), http://dx.doi.org/10.1016/j.ijgo.2015.01.010
4
Y. Zhu et al. / International Journal of Gynecology and Obstetrics xxx (2015) xxx–xxx
Fig. 2. Gene Ontology and pathway analysis for the predicted targets of five miRNAs that were differentially expressed in gestational diabetes mellitus. (A) Gene Ontology categories enriched with predicted targets; (B) Kyoto Encyclopedia of Genes and Genomes pathways enriched with predicted targets.
FOS) and decreasing expression of some protective genes (e.g. BCL2 and HSPB1). The TGF-β signaling pathway is associated with pre-eclampsia [24], and the mTOR signaling pathway is involved in the control of energy balance and food intake in the hypothalamus [25]. Blocking of these two pathways might also contribute to the development of GDM. Of the five miRNAs that were upregulated in GDM, three (miR146b-5p, miR-20a, and miR-17) are also upregulated in pre-eclampsia [26,27]. Like GDM, pre-eclampsia is a disease of pregnancy and the pathogenesis of these two conditions is related to a certain extent. Both women with pre-eclampsia and those with GDM have a higher degree of insulin resistance than do healthy pregnant women [28]. Four of the five miRNAs (miR-17-5p, miR-19a-3p, miR-19b-3p, and miR-20a-5p) belong to the miRNA-17-92 cluster. This cluster is associated with angiogenesis, which is a key physiological process during pregnancy. Upregulation of these miRNAs might lead to an abnormal pregnancy. The strengths of the present study include the use of a highthroughput sequencing platform and the validation of miRNA sequences
MAPK signaling pathway
by qRT-PCR. However, the study also has a few limitations. Firstly, the sample size was small, which may have caused unreliable results. Secondly, only some of the miRNA sequences were validated by qRTPCR, and the qRT-PCR was performed on pooled samples. Further studies are needed that include validation of a larger number of miRNAs using individual samples. Finally, the target genes and pathways were all predicted by software and the conclusions about the roles of the differentially expressed miRNAs in these pathways were made on the basis of findings from previous studies. More advanced functional studies are needed to better understand the functions of these miRNAs. In summary, the present study demonstrates the use of deep sequencing for the comprehensively profiling of miRNAs in the plasma of women with GDM. The study does not only confirm existing findings, but it also led to the discovery of dysregulated miRNAs that were not known to be associated with GDM, thereby providing a new miRNA signature for the detection of early-stage GDM. The blood samples were collected at 16–19 weeks of pregnancy, before GDM was diagnosed. The identified miRNAs constitute promising biomarkers that might help to improve the early detection and monitoring of GDM. Moreover,
Insulin signaling pathway
IRS-1
MAPK-1
miR-19a
miR-16
SMAD5
Type 2 diabetes mellitus
SOS-1
IRS-2
miR-19b
SMAD4
TGF-β signaling pathway
miR-20a
miR-17
AKT3
mTOR signaling pathway
Fig. 3. Gene regulatory network involving five miRNAs that were differentially expressed in gestational diabetes mellitus.
Please cite this article as: Zhu Y, et al, Profiling maternal plasma microRNA expression in early pregnancy to predict gestational diabetes mellitus, Int J Gynecol Obstet (2015), http://dx.doi.org/10.1016/j.ijgo.2015.01.010
Y. Zhu et al. / International Journal of Gynecology and Obstetrics xxx (2015) xxx–xxx
the present insights into the molecular mechanisms that underlie GDM could aid the diagnosis and treatment of the disease. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ijgo.2015.01.010. Acknowledgments This work was supported by project 61271055 of the National Natural Science Foundation of China. Conflict of interest The authors have no conflicts of interest. References [1] Liang Z, Dong M, Cheng Q, Chen D. Gestational diabetes mellitus screening based on the gene chip technique. Diabetes Res Clin Pract 2010;89(2):167–73. [2] HAPO Study Cooperative Research Group, Metzger BE, Lowe LP, Dyer AR, Trimble ER, Chaovarindr U, et al. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 2008;358(19):1991–2002. [3] England LJ, Dietz PM, Njoroge T, Callaghan WM, Bruce C, Buus RM, et al. Preventing type 2 diabetes: public health implications for women with a history of gestational diabetes mellitus. Am J Obstet Gynecol 2009;200(4):365.e1–8. [4] Tieu J, Middleton P, McPhee AJ, Crowther CA. Screening and subsequent management for gestational diabetes for improving maternal and infant health. Cochrane Database Syst Rev 2010;7:CD007222. [5] Holt RI, Coleman MA, McCance DR. The implications of the new International Association of Diabetes and Pregnancy Study Groups (IADPSG) diagnostic criteria for gestational diabetes. Diabet Med 2011;28(4):382–5. [6] Sayed D, Abdellatif M. MicroRNAs in development and disease. Physiol Rev 2011; 91(3):827–87. [7] Coordes A, Lenarz M, Kaufmann AM, Albers AE. Role of miRNA in malignoma of the head and neck. Laryngorhinootologie 2014;93(3):201–9. [8] Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 2004; 116(2):281–97. [9] Ambros V. The functions of animal microRNAs. Nature 2004;431(7006):350–5. [10] Zhao Z, Moley KH, Gronowski AM. Diagnostic potential for miRNAs as biomarkers for pregnancy-specific diseases. Clin Biochem 2013;46(10–11):953–60. [11] van Rooij E. The art of microRNA research. Circ Res 2011;108(2):219–34. [12] American Diabetes Association. Standards of medical care in diabetes–2011. Diabetes Care 2011;34(Suppl. 1):S11–61.
5
[13] Audic S, Claverie JM. The significance of digital gene expression profiles. Genome Res 1997;7(10):986–95. [14] Hu J, Wang Z, Liao BY, Yu L, Gao X, Lu S, et al. Human miR-1228 as a stable endogenous control for the quantification of circulating microRNAs in cancer patients. Int J Cancer 2014;135(5):1187–94. [15] Metzger BE, Buchanan TA, Coustan DR, de Leiva A, Dunger DB, Hadden DR, et al. Summary and recommendations of the Fifth International Workshop-Conference on Gestational Diabetes Mellitus. Diabetes Care 2007;30(Suppl. 2):S251–60. [16] Carew RM, Browne MB, Hickey FB, Brazil DP. Insulin receptor substrate 2 and FoxO3a signalling are involved in E-cadherin expression and transforming growth factor-β1-induced repression in kidney epithelial cells. FEBS J 2011;278(18): 3370–80. [17] Sorokin AV, Chen J. MEMO1, a new IRS1-interacting protein, induces epithelialmesenchymal transition in mammary epithelial cells. Oncogene 2013;32(26): 3130–8. [18] Taniguchi CM, Emanuelli B, Kahn CR. Critical nodes in signalling pathways: insights into insulin action. Nat Rev Mol Cell Biol 2006;7(2):85–96. [19] Geng Y, Ju Y, Ren F, Qiu Y, Tomita Y, Tomoeda M, et al. Insulin receptor substrate 1/2 (IRS1/2) regulates Wnt/β-catenin signaling through blocking autophagic degradation of dishevelled2. J Biol Chem 2014;289(16):11230–41. [20] Rao R, Hao CM, Redha R, Wasserman DH, McGuinness OP, Breyer MD. Glycogen synthase kinase 3 inhibition improves insulin-stimulated glucose metabolism but not hypertension in high-fat-fed C57BL/6 J mice. Diabetologia 2007;50(2):452–60. [21] Tsai YL, Hou CW, Liao YH, Chen CY, Lin FC, Lee WC, et al. Exercise training exacerbates tourniquet ischemia-induced decreases in GLUT4 expression and muscle atrophy in rats. Life Sci 2006;78(25):2953–9. [22] Nishiyama A, Yoshizumi M, Rahman M, Kobori H, Seth DM, Miyatake A, et al. Effects of AT1 receptor blockade on renal injury and mitogen-activated protein activity in Dahl salt-sensitive rats. Kidney Int 2004;65(3):972–81. [23] Lappas M, Riley C, Lim R, Barker G, Rice GE, Menon R, et al. MAPK and AP-1 proteins are increased in term pre-labour fetal membranes overlying the cervix: regulation of enzymes involved in the degradation of fetal membranes. Placenta 2011;32(12): 1016–25. [24] Perucci LO, Gomes KB, Freitas LG, Godoi LC, Alpoim PN, Pinheiro MB, et al. Soluble endoglin, transforming growth factor-Beta 1 and soluble tumor necrosis factor alpha receptors in different clinical manifestations of preeclampsia. PLoS One 2014;9(5):e97632. [25] Kumar A, Lawrence Jr JC, Jung DY, Ko HJ, Keller SR, Kim JK, et al. Fat cell-specific ablation of rictor in mice impairs insulin-regulated fat cell and whole-body glucose and lipid metabolism. Diabetes 2010;59(6):1397–406. [26] Fu G, Brkić J, Hayder H, Peng C. MicroRNAs in Human Placental Development and Pregnancy Complications. Int J Mol Sci 2013;14(3):5519–44. [27] Wang W, Feng L, Zhang H, Hachy S, Satohisa S, Laurent LC, et al. Preeclampsia up-regulates angiogenesis-associated microRNA (i.e., miR-17, − 20a, and -20b) that target ephrin-B2 and EPHB4 in human placenta. J Clin Endocrinol Metab 2012; 97(6):E1051–9. [28] Briana DD, Malamitsi-Puchner A. Reviews: adipocytokines in normal and complicated pregnancies. Reprod Sci 2009;16(10):921–37.
Please cite this article as: Zhu Y, et al, Profiling maternal plasma microRNA expression in early pregnancy to predict gestational diabetes mellitus, Int J Gynecol Obstet (2015), http://dx.doi.org/10.1016/j.ijgo.2015.01.010
