Secondary Logo

Journal Logo

Original Article

Genetic Variants of Potassium Voltage-gated Channel Subfamily J Member 11 in Gestational Diabetes Mellitus: A Case-control Study

Wei, Li-Jie; Zhou, Xuan; Zhu, Sheng-Lan; Li, Jia-Qi; Zeng, Yu; Yu, Jun; Wang, Shao-Shuai; Feng, Ling

Section Editor(s): Pan, Yang; Shi, Dan-Dan

Author Information
doi: 10.1097/FM9.0000000000000030
  • Open



Gestational diabetes mellitus (GDM) is identified as onset hyperglycemia in the second trimester of pregnancy, which is characterized by pancreatic β cell dysfunction. The incidence of GDM is increasing rapidly. According to the latest statistics, the incidence of GDM has reached 17.5%–18.9% in China.1 Pregnant women who do not control or monitor well enough their sugar level during pregnancy would follow severe pregnancy outcomes, including postpartum diabetes mellitus, macrosomia (defined as the birth weight achieving or exceeding 4 kg), hypoglycemia of neonatal (defined as infantile blood glucose lower than 2.2 mmol/L), diabetes mellitus in offspring, and so on.2,3 Until now, the etiology of GDM has not been fully clarified.

GDM has a sequence of risk factors, such as elder age, pre-pregnancy overweight/obese, family history of diabetes, polycystic ovary syndrome, and so on.4 According to the previous studies, gene susceptibility may also attribute to its pathogenesis.5,6 Recently, genome-wide association study, an organization focusing on genetic loci of diseases and various gene types have identified some genes which may associate with the occurrence of GDM.7 Although numerous studies were conducted in genetic loci and GDM association, however, they have not find the exact relationship between them.

The protein encoded by potassium voltage-gated channel subfamily J member 11 (KCNJ11) is an inward-rectifier type potassium channel and integral membrane protein which is controlled by G-proteins. The opening and closure of adenosine triphosphate-sensitive potassium channel encoded by KCNJ11 can control insulin secretion by coupling β-cell metabolism to calcium entry.8 As reported, KCNJ11 gene was associated with type 2 diabetes mellitus (T2DM), which affected the secretion of insulin in its pathway.9,10 But the relationship of KCNJ11 with GDM were still discordant. Of all the variant loci, rs5219 is the most common loci being studied, followed by rs5215. However, it is worth noting that rs5210 was associated with the occurrence of T2DM in several studies, but the loci variant with GDM was just only one report. It is known that rs5210 is in the 3′-untranslated region of KCNJ11 gene, which is a potential micro-RNA (miRNA) binding site. miRNA is a type of small noncoding RNA molecular, which binds to target messenger RNA (mRNA) to inhibit posttranscriptional gene expression. Therefore, the abnormal interaction between miRNA and its target mRNA due to the variant of rs5210 may affect insulin secretion.

In this study, we focused on the genetic susceptibility of rs5210 in KCNJ11 of GDM. Therefore, exploring the mechanisms of genetic variants in KCNJ11 may lead to a promising research direction.

Material and methods


Six hundred and thirty-two uncorrelated pregnant women included in the study were recruited from obstetrics inpatient department in the Tongji Hospital, Tongji Medical College of Huazhong University of Science and Technology, Wuhan, Hubei, China, from October 2017 to June 2018, in which 241 women were identified as GDM group, while 391 were non-GDM control group. All the participants were of Chinese ethnic origin and visited the hospital during 8 months mentioned above to match case and control participants. Two researchers evaluated the participants independently, and disagreements were resolved by discussion. Their clinical and biochemical characterization were collected and recorded as shown in Table 1. All the participants included in the study were informed and consent. The study protocol was approved by the Ethics Committee of the Tongji Hospital (TJ-IRB20170506), Tongji Medical College of Huazhong University of Science and Technology and performed in accordance with the Declaration of Obstetrics Department. The written consent was obtained from each participant.

Table 1
Table 1:
Clinical and biochemical characterization of GDM and control groups.

Including criteria

All pregnant women were recruited without a previous diagnosis of glucose intolerance. Seventy-five gram oral glucose tolerance test (OGTT) tests were screened in 24–28 weeks of gestation, GDM was diagnosed according to the criteria of the International Association of Diabetes and Pregnancy Study Groups. The measurements of plasma glucose values were as follows: fasting blood glucose ≥5.1 mmol/L, 1-hour blood glucose ≥10.0 mmol/L, 2-hour blood glucose ≥8.5 mmol/L. Any pregnant women having one of the above OGTT higher was identified and grouped as GDM, yet any with the three above plasma glucose level normal were identified and grouped as non-GDM control group.

Excluding criteria

(1) Pregnant women who were diagnosed as glucose intolerance before pregnancy; (2) pregnant women suffering from other metabolic disease before or during pregnancy, such as hyperthyroidism, hypothyroidism, hypertension disorder; (3) pregnant women experiencing other complications, such as antiphospholipid syndrome, intrahepatic cholestasis of pregnancy, and infectious disease; (4) pregnant women who with other major organic associated complications, like heart disease, hepatic function damage, renal disease, mental problems; (5) twins pregnancy or more.

DNA extraction of peripheral blood

All the pregnant women recruited in this study their peripheral venous blood of 5 mL were withdrawn, with ethylenediaminetetraacetic acid tubes. All the blood samples were centrifuged as 3 000 rpm in 8 minutes, then the supernate and sediment were separated in 1.5 mL centrifuge tubes and stored at −80°C. DNA was extracted by using TIANamp Blood DNA kit (Beijing, China) according to the manufacturer's instructions. The concentration and purity were measured by ThermoScientificTMNanoDrop Lite Spectrophotometer (Thermo Fisher Scientific, Waltham, Massachusetts, USA) and the A260/A280 ratio was controlled in 1.7–2.0.

Genotype discrimination

All the genomic DNA samples extracted as mentioned above were genotyped by using TaqMan allelic discrimination assay (Applied Biosystems, Foster City, California, USA), real-time polymerase chain reaction was conducted by Bio-rad CFX CONNECT Real-Time System (Bio-rad, Hercules, California, USA). An allelic discrimination assay was performed in 10 μL of 2 × iTaqTM Universal Probes Supermix (Bio-rad, Hercules, California, USA) containing 1 μL of genomic DNA and 0.25 μL of 40× assays-on-Demand single nucleotide polymorphism (SNP) genotyping product (Applied Biosystems, Foster City, California, USA), according to the manufacturer's instructions. Then, plates were placed on CFX CONNECT Real-Time System (Bio-rad) and heated for 2 minutes at 50°C and 10 minutes at 95°C, followed by 45 cycles at 95°C for 15 seconds and 60°C for 1 minute. The fluorescence data files of each plate were analyzed using CFX Manager version 3.1 (Bio-rad). Ten percent of blind samples were genotyped repeatedly, and all of them were confirmed to the first time genotyping.

Statistical analyses

All continuous variables were expressed as mean ± SD. Statistical analyses were conducted using SPSS 23.0 (IBM, Armonk, NY, USA). Two-independent sample t test was used to determine the differences of sample means between two groups if they subjected to normal distribution, otherwise, nonparametric test would be run. Chi-square test was used to determine whether individual polymorphisms were in Hardy-Weinberg equilibrium (HWE). The study could be analyzed further only if the mode of control group inheritance conforms to Mendel's law of heredity (P > 0.050). We compared the allelic and genotype frequencies of the 241 GDM patients with those of the 391 controls. Logistic regression analyses were used for calculating the odds ratio (OR), 95% confidence interval (CI) and the corresponding P with regard to the number of risk alleles using dominant model, recessive model and additive model. Multivariate linear regressions, adjusted for age and body mass index (BMI), were used for comparing neonatal weight and neonatal blood glucose in GDM group with full-term infants according to different allelic models. A P with a two-side <0.050 was considered as statistically significant.


Clinical and biochemical characterization of GDM and control group

As shown in Table 1, we had 632 pregnant women in the study, in which 241 were defined as GDM with an average age as (32.66 ± 4.86) years, while 391 were non-GDM control group with an average age as (30.20 ± 4.13) years. Age, pre-pregnancy BMI of case group were higher than that of control group, which indicated that they were risk factors in GDM (P < 0.050). There were significant differences of second trimester's OGTT and triglyceride between two groups, of which GDM group were significantly higher than of control group (P < 0.001). Interestingly, total cholesterol (TC) level in GDM group were significantly lower than control group (P < 0.001) due to satisfied plasma glucose level. There was no significant difference of blood pressure, liver function, and renal function between GDM and control groups (P > 0.050). In addition, weight-gain during pregnancy, birth weight, and plasma glucose level of full-term delivery infants had no significant difference in the two groups (P > 0.050). None of the participants in our study had a smoking history for at least 6 months before pregnancy.

Association of variant in rs5210 of KCNJ11 and GDM

The genotype distribution of control group was in line with HWE (P = 0.305). And the variant locus was used for the following analyses. As shown in Table 2, heterozygote (AG) had the highest frequency among three genotypes (49% for GDM group; 52% for control group, respectively). Homozygote genotype AA was higher than that of GG in two groups (AA = 27%, GG = 24% for GDM group; AA = 28%, GG = 20% for control group, respectively). Allelic gene G had the minor allele frequency as 48% and 46% in GDM and control groups, respectively. The risk allele of rs5210 in KCNJ11 gene was G allele, which was not significantly higher in GDM group compared to control group. Indicate that G allele of rs5210 did not function in development of GDM. Moreover, logistic analyses were used to adjust age and BMI among the variant genotypes of rs5210 in three genotype models, the results were as follows: P for dominant model was 0.945, (OR: 0.987, 95% CI: 0.681–1.430); P for recessive model was 0.556, (OR: 1.217, 95% CI: 0.633–2.343); P for addictive model was 0.098 (AA vs. GG), (OR: 1.435, 95% CI: 0.936–2.201). However, no significant difference was present in three genotype models as above.

Table 2
Table 2:
Gene variant frequencies of GDM and control groups.

Association of genotypes in rs5210 with relatively metabolic parameters

We conducted the relationship of genotype distribution for rs5210 in KCNJ11 with relatively metabolic parameters of all the participants (control and GDM groups), including weight-gain during pregnancy, second trimester OGTT values and TC values (Table 3). The results showed significant difference of weight-gain during pregnancy in recessive model (P = 0.015) and TC in recessive model (P = 0.022).

Table 3
Table 3:
Association of genotypes in rs5210 with relatively metabolic parameters.


From the clinical and biochemical characteristics of GDM and control group shown in Table 1, age, BMI, and TC might be the risk factors in the onset of GDM. As expected, fasting blood glucose, 1-hour blood glucose and 2-hour blood glucose compared by two groups showed significant difference (P < 0.050). Beyond our expectation, birth weight of infants and the proportion of neonatal hypoglycemia showed no significant difference, which may imply that well-control of blood glucose level in the third trimester will play a role in neonatal outcomes in some ways. If pregnant women are diagnosed earlier with glucose intolerance before 28 weeks, then diet control, exercise, or drugs intervention are advised and prescribed to them on time. So that less neonatal complications may occur. Moreover, numbers of women can less suffer from a series of postpartum complications and that may lead to a high-quality future life.

In our study, we found that GG genotype is linked with the lowest weight gain during pregnancy, compared with AG and AA genotypes, which has not been reported. Furthermore, we observed the significant association of GG genotype with lower TC level. This indicates that GG genotype might be protective to metabolism disease. As reported by Wang et al. (2017) in Table 4, rs5210 was predicted to interact with hsa-miR-219a-2-3p. Variant of GG genotype may create a new binding site of miRNA or affect the interaction between miRNA and target mRNA. However, the mechanism was still unclear.

Until now, although numerous studies have been conducted to investigate the association of KCNJ11 with GDM, there are discordant results of the same variant. As reported, KCNJ11 gene is associated with T2DM, which affects the secretion of insulin in its pathway.

As we have reported above, the variant susceptibility relationship of rs5210 with T2DM and GDM was still discordant (Table 4). Six studies from “NCBI” using the keywords “KCNJ11” and “single nucleotide polymorphism (SNP)” as displayed in Table 4, Koo et al. (2007) and Khan et al. (2015) were excluded after recalculating H-W equilibrium by author (P = 0.026 and P < 0.001, respectively). While, the remained four studies, three of them (Willer et al. (2007), Cruz et al. (2010), Wang et al. (2017)) indicated a minor allelic frequency as “A allele”. To the contrary, one study (Sakamoto et al. (2007)) indicated minor allele of rs5210 “G allele”. Only one study (Wang et al. (2017)) investigated the association of the variant with GDM, the rest of three studies focused on the genetic susceptibility of T2DM. From the above mentioned studies that focused on T2DM, only two showed significant differences, which indicated that gene variant of rs5210 may play a role in T2DM. Although the study for Wang et al. (2017) showed no significant difference in all genetic models of rs5210 of GDM. Various research on rs5210 with more samples are needed to explore whether the variant loci does act or affect on certain unknown metabolism disorders during pregnancy resulting in GDM.

Table 4
Table 4:
Publications about rs5210 in KCNJ11.

The results of our study indicated that the genetic polymorphism of rs5210 in KCNJ11 may associated with some metabolic parameters during pregnancy. GG genotype variant may be a benign factor for the development of GDM. However, some limitations should be addressed. First, the number of the participants in our study was not enough. And more regions of participants in China should be recruited to investigate the difference between various lifestyles. Second, further studies are needed to research the interaction between genetic variant of rs5210 and its miRNA in GDM.


More studies and samples are needed to further investigate the relationship of KCNJ11 and the occurrence of GDM. This also suggests our next study focusing on the KCNJ11 mechanism of its biologic pathways and function, which will provide a powerful evidence for researchers in the prevention and intervention of GDM. Furthermore, gene variant will also facilitate the treatment of GDM and its complications.


The study was funded by the research grants from the National Key Research and Development Program of China (2016YFC1000405, 2018YFC1002900) and the National Natural Science Foundation of China (41671497).

Author Contributions:

Study concept and design: Li-Jie Wei and Ling Feng. Acquisition of data: Li-Jie Wei, Xuan Zhou, Sheng-Lan Zhu, Jia-Qi Li, Yu Zeng. Analysis and interpretation of data: Li-Jie Wei, Jun Yu, Shao-Shuai Wang. Drafting of the manuscript: Li-Jie Wei, Xuan Zhou.

Conflicts of Interest



[1]. Wei Y, Yang H, Zhu W, et al. International association of diabetes and pregnancy study group criteria is suitable for gestational diabetes mellitus diagnosis: further evidence from China. Chin Med J (Engl) 2014;127(20):3553–3556.
[2]. Bakun OV, Karatieieva SY, Semenenko SB, et al. Obesity during pregnancy: literature review. Wiad Lek 2018;71(4):913–916.
[3]. Plows JF, Stanley JL, Baker PN, et al. The pathophysiology of gestational diabetes mellitus. Int J Mol Sci 2018;19(11):3442. doi:10.3390/ijms19113342.
[4]. Jaffe A, Giveon S, Rubin C, et al. Gestational diabetes risk in a multi-ethnic population. Acta Diabetol 2019;doi:10.1007/s00592-019-01404-8.
[5]. Shaat N, Ekelund M, Lernmark A, et al. Genotypic and phenotypic differences between Arabian and Scandinavian women with gestational diabetes mellitus. Diabetologia 2004;47(5):878–884. doi:10.1007/s00125-004-1388-5.
[6]. Kwak SH, Kim SH, Cho YM, et al. A genome-wide association study of gestational diabetes mellitus in Korean women. Diabetes 2012;61(2):531–541. doi:10.2337/db11-1034.
[7]. Zhang C, Bao W, Rong Y, et al. Genetic variants and the risk of gestational diabetes mellitus: a systematic review. Hum Reprod Update 2013;19(4):376–390. doi:10.1093/humupd/dmt013.
[8]. Vidal-Taboada JM, Pugliese M, Salvadó M, et al. KATP channel expression and genetic polymorphisms associated with progression and survival in amyotrophic lateral sclerosis. Mol Neurobiol 2018;55(10):7962–7972. doi:10.1007/s12035-018-0970-7.
[9]. Wang DD, Chen X, Yang Y, et al. Association of Kir6.2 gene rs5219 variation with type 2 diabetes: a meta-analysis of 21,464 individuals. Prim Care Diabetes 2018;12(4):345–353. doi:10.1016/j.pcd.2018.03.004.
[10]. Qin LJ, Lv Y, Huang QY. Meta-analysis of association of common variants in the KCNJ11-ABCC8 region with type 2 diabetes. Genet Mol Res 2013;12(3):2990–3002. doi:10.4238/2013.August.20.1.
[11]. Koo BK, Cho YM, Park BL, et al. Polymorphisms of KCNJ11 (Kir6.2 gene) are associated with Type 2 diabetes and hypertension in the Korean population. Diabet Med 2007;24(2):178–186. doi:10.1111/j.1464-5491.2006.02050.x.
[12]. Willer CJ, Bonnycastle LL, Conneely KN, et al. Screening of 134 single nucleotide polymorphisms (SNPs) previously associated with type 2 diabetes replicates association with 12 SNPs in nine genes. Diabetes 2007;56(1):256–264. doi:10.2337/db06-0461.
[13]. Sakamoto Y, Inoue H, Keshavarz P, et al. SNPs in the KCNJ11-ABCC8 gene locus are associated with type 2 diabetes and blood pressure levels in the Japanese population. J Hum Genet 2007;52(10):781–793. doi:10.1007/s10038-007-0190-x.
[14]. Cruz M, Valladares-Salgado A, Garcia-Mena J, et al. Candidate gene association study conditioning on individual ancestry in patients with type 2 diabetes and metabolic syndrome from Mexico City. Diabetes Metab Res Rev 2010;26(4):261–270. doi:10.1002/dmrr.1082.
[15]. Khan IA, Vattam KK, Jahan P, et al. Correlation between KCNQ1 and KCNJ11 gene polymorphisms and type 2 and post-transplant diabetes mellitus in the Asian Indian population. Genes Dis 2015;2(3):276–282. doi:10.1016/j.gendis.2015.02.009.
    [16]. Wang X, Li W, Ma L, et al. Investigation of miRNA-binding site variants and risk of gestational diabetes mellitus in Chinese pregnant women. Acta Diabetol 2017;54(3):309–316. doi:10.1007/s00592-017-0969-y.

    Diabetes, gestational; Polymorphism, single nucleotide; Potassium channels, voltage-gated

    Copyright © 2020 The Chinese Medical Association, published by Wolters Kluwer Health, Inc.