Association of MIF rs1007888 and ARAP1 rs1552224 genetic variants with the risk of gestational diabetes mellitus in a chinese population; case study and meta-analysis

Background: Macrophage migration inhibitory factor (MIF) rs1007888 is significantly associated with pancreatic β-cell function and insulin resistance in patients with gestational diabetes mellitus (GDM). The ArfGAP with RhoGAP domain, ankyrin repeat, and PH domain-containing protein 1 (ARAP1) rs1552224 locus has been identified as a risk locus for type 2 diabetes, and recent reports have linked it to elevated blood glucose levels and reduced insulin release upon glucose stimulation. Few studies have been conducted on these genetic variants and their risk of GDM. This study aimed to investigate the association between these two genetic variants (ARAP1) rs1552224 and (MIF) rs1007888 and the risk of developing GDM.

Methods: A case-control study involving 500 GDM patients and 502 healthy controls was conducted. DNA was extracted, and rs1007888 and rs1552224 were systematically genotyped using the SNPscan™ genotyping kit. Statistical methods assessed genotype and allele differences linked to GDM risk, followed by a meta-analysis to evaluate the impact of regional factors on GDM.

Results: Analyses of (MIF) rs1007888 showed no link to higher GDM risk, but meta-analysis found a significant association (OR>1), indicating a connection to increased GDM risk. ARAP1 rs1552224 was significantly linked to reduced GDM incidence (Allele Model A vs. C: OR = 0.624; 95% CI: 0.425-0.916; p-value = 0.016; Dominant Model AA vs. AC+CC: OR = 0.641; 95% CI: 0.429-0.959; p-value = 0.030), especially in women under 30, rs1552224 Aelle Model (A vs. C: OR = 0.490; 95% CI: 0.281-0.857; p -value = 0.012), Dominant Model (AA vs. AC + CC: OR = 0.523; 95% CI: 0.292-0.938; p -value = 0.030). and those with a BMI≥24, Aelle Model (A vs. C: OR = 0.345; 95% CI: 0.124-0.960; p-value = 0.042). Conversely, a meta-analysis suggested an increased GDM risk with the ARAP1 variant (OR>1).

Conclusion: The meta-analysis results demonstrate that there is an enhanced likelihood of GDM associated with the MIF rs1007888 mutation. Moreover, our findings indicate that the ARAP1 rs1552224 variant, specifically the AC genotype and C allele, confers a decreased risk of developing gestational diabetes mellitus (GDM). The outcomes obtained give GDM testing a theoretical foundation.

1 Introduction

Gestational diabetes mellitus (GDM) is a pregnancy-specific disorder of glucose metabolism, distinct from pre-existing diabetes, and is increasingly prevalent in China. It poses significant risks to maternal health, including preeclampsia, higher rates of cesarean delivery, and a markedly increased likelihood of developing type 2 diabetes mellitus (T2DM) later in life. Offspring of affected mothers are also at risk of adverse outcomes such as macrosomia, neonatal hypoglycemia, and long-term metabolic complications (1–3). Established risk factors for GDM include advanced maternal age, obesity, family history of diabetes, and genetic susceptibility (4).

Although the pathophysiology of GDM is not fully understood, it shares essential features with T2DM, including insulin resistance, impaired glucose tolerance, and β-cell dysfunction (2, 3). Increasing evidence points to the contribution of genetic variation, particularly single-nucleotide polymorphisms (SNPs), in modulating GDM risk. Among candidate loci, the macrophage migration inhibitory factor (MIF) gene and the ARAP1 locus have been implicated in diabetes-related traits. MIF rs1007888 has been associated with insulin resistance and β-cell dysfunction (5, 6), while ARAP1 rs1552224 has been linked to reduced insulin secretion (7). However, their roles in GDM remain insufficiently investigated.

MIF is a pleiotropic cytokine expressed abundantly by placental trophoblasts and upregulated in GDM placental tissue, correlating with fasting glucose levels and insulin resistance (8, 9). Functionally, MIF influences insulin secretion and glucose metabolism and exerts pro-inflammatory effects by regulating cytokines such as TNF-α, IFN-γ, and IL-10, thereby contributing to hyperglycemia (9). Elevated systemic MIF levels are a feature of T2DM (10), but its mechanistic role in pregnancy-related glucose dysregulation remains poorly defined (8).

Meanwhile, genome-wide association studies (GWAS) consistently implicate ARAP1 variants in T2DM susceptibility (11, 12). In particular, rs1552224 has been associated with fasting glucose levels and impaired glucose-stimulated insulin secretion, with evidence suggesting the A allele contributes to β-cell dysfunction (11). Interestingly, preliminary data indicate that this allele may be protective against GDM (13), highlighting potential differences between pregnancy-related and non-pregnancy-related diabetes risk mechanisms.

Despite these insights, data on the association between MIF rs1007888 and ARAP1 rs1552224 with GDM remain sparse, especially in Asian populations, and no systematic synthesis of existing findings has been undertaken. Given the high burden of GDM and the distinct genetic architecture of the Chinese population, further investigation is warranted.

Therefore, in this study, we examined the associations of MIF rs1007888 and ARAP1 rs1552224 with GDM in a Han Chinese cohort and performed a meta-analysis to contextualize our findings with the broader literature. We aimed to clarify the potential contribution of these variants to GDM risk and provide new insights into their role in pregnancy-specific glucose metabolism.

2 Materials and methods

2.1 Study participants

A total of 1002 participants were recruited for the study, including 500 patients with gestational diabetes mellitus (GDM) and 502 pregnant women without GDM who served as controls. This study protocol was approved by the Ethics Committee of Shunde Women’s and Children’s Hospital Affiliated to Guangdong Medical University (approval ID: 2020072). The inclusion criteria were as follows: participants must provide voluntarily signed informed consent; be of Han ethnicity; be at least 18 years old; have undergone a 75g oral glucose tolerance test (OGTT) between 24 and 28 weeks of gestation and be diagnosed with gestational diabetes mellitus (GDM) or have standard glucose tolerance according to the criteria of the International Association of Diabetes and Pregnancy Study Groups (IADPSG). In the present study, the International Association of Diabetes and Pregnancy Investigation Groups (IADPSG) diagnostic guidelines were employed. If one or more points satisfy the following criteria, GDM was diagnosed: fasting blood glucose (FBG) ≥ 5.1 mmol/L, 1-hour postprandial glucose (PG) ≥ 10.0 mmol/L, or 2-hour PG ≥ 8.5 mmol/L. Expectant mothers falling below these specified limits were classified as healthy control subjects.

The exclusion criteria included: presence of pregnancy-related diseases or use of drugs affecting glucose metabolism; history of severe cardiovascular and cerebrovascular diseases, hepatic or renal insufficiency, tumors, or pathogenic infections; and engagement in smoking, alcoholism, drug abuse, or presence of intellectual disability or mental disorders. This study was conducted following the principles outlined in the Declaration of Helsinki.

2.2 Data collection

During the 24–28 gestational weeks, comprehensive data were collected, including parity (primigravida or multigravida), pre-pregnancy weight, ethnicity, age, height, blood pressure, and blood glucose levels. The collected data were subsequently employed to compute the pregestational body mass index (pre-BMI, kg/m²), which was defined as the pregestational weight (in kilograms) divided by the square of height (in meters). To determine the obesity status in line with Chinese standards, the following classification criteria were utilized: underweight (< 18.5 kg/m²), normal weight (18.5–24.9 kg/m²), overweight (25–29.9 kg/m²), and obese (≥ 29 kg/m²). This method of classification was adopted based on the research carried out by (14).

2.3 SNP genotyping and quality control

Based on findings from genome-wide association studies (GWAS) of type 2 diabetes mellitus (T2DM) in Asian populations, we selected two candidate SNPs, MIF rs1007888 and ARAP1 rs1552224, for evaluation of their potential association with GDM. Selection criteria included a minor allele frequency (MAF) > 0.05, supported by evidence from prior studies (8, 15, 16).

For each participant, 2 mL of peripheral blood was collected into EDTA tubes and stored at −80 °C until analysis. Genomic DNA was extracted using the QIAamp DNA Blood Kit (Qiagen, Germany) according to the manufacturer’s protocol. SNP genotyping was performed using the SNPscan™ method (Genesky Biotechnologies, Shanghai, China), a high-throughput and highly accurate technique based on dual ligation probe hybridization and multiplex fluorescent PCR. The procedure involves probe ligation to discriminate wild-type and variant alleles, multiplex PCR amplification with fluorescently labeled primers, and capillary electrophoresis to separate amplified fragments. Genotypes were assigned based on fragment length and fluorescent signal intensity.

To ensure accuracy, rigorous quality control measures were implemented by Genesky Biotechnologies (12, 17). Pre-experiments were conducted to optimize assay performance. In addition, 6% of randomly selected samples were re-genotyped by Sanger sequencing, yielding 100% concordance with the SNPscan results.

2.4 Statistical analyses

The analytical evaluations were performed via SPSS 20.0 (SPSS Inc., Chicago, IL, USA), with a bilateral p-value less than 0.05 considered statistically significant. Standard distribution-aligned variables were recorded as means ± standard deviations, and the non-overlapping samples t-test was used to assess the variances in relevant parameters between the two groups. When the normality assumption was violated, non-parametric testing methods were employed. Descriptive data were evaluated using the chi-square (χ²) test. To verify the representativeness of the control group in the population, the Hardy-Weinberg equilibrium (HWE) test, estimated by the goodness-of-fit χ², was applied. Six genetic models, specifically codominant homozygous, codominant heterozygous, over-dominant, dominant, recessive, and allele models, were used to assess GDM risk through the χ² test and logistic regression analysis. The presentation included basic and adjusted odds ratios (ORs) along with their respective 95% confidence intervals (CIs), considering variables such as age, pre-BMI, blood pressure, parity, and more. A stratified analysis was performed to delve deeper into how age and pre-BMI might affect the outcomes. A one-way ANOVA was used to explore the link between SNPs and blood sugar levels. The least significant difference (LSD) approach was adopted for multiple comparisons. Investigations of specific subgroups were conducted for both GDM and T2DM.

2.5 Meta-analysis

A thorough literature review and meta-analysis were conducted utilizing the Google Scholar, PubMed, and CNKI databases to assess the association between the MIF rs1007888 and ARAP1 rs1552224 polymorphisms and the risk of gestational diabetes mellitus (GDM) and type 2 diabetes mellitus (T2DM) (Supplementary Table 7). A limited number of studies were available for the meta-analysis concerning these genes and their associated polymorphisms. The authors propose further studies with these genes. For the analysis of rs1552224 concerning GDM, one eligible study was included, while three studies were incorporated for rs1552224 concerning T2DM. Additionally, four studies were selected for the association between MIF rs1007888 and GDM (refer to Supplementary Table 7). The analyses were performed using a fixed-effects model.

The search strategy employed combinations of the terms rs1007888, rs1552224, type 2 diabetes mellitus (T2DM), and gestational diabetes mellitus (GDM). Eligible studies were those that focused on case-control or cohort analyses exploring the association between the rs1007888 and rs1552224 polymorphisms and T2DM or GDM, provided they contained sufficient original data. Studies that did not meet the established diagnostic criteria or deviated from the Hardy-Weinberg equilibrium were excluded from the analysis. Data extraction was carried out independently by two authors, with any discrepancies resolved through consultation with a third party. The kappa coefficient was calculated to evaluate inter-researcher agreement, ensuring the objectivity and accuracy of the study selection process.

Meta-analyses were performed across six genetic models, employing fixed-effects or random-effects models depending on the level of heterogeneity observed. Publication bias was assessed using Egger’s and Begg’s tests. All analyses were conducted using STATA version 16.0. This study adheres to the PRISMA guidelines and the Cochrane Handbook. The protocol has been registered with PROSPERO, an internationally recognized platform for the registration of systematic reviews and meta-analyses, under the registration number CRD420251122128.

3 Results

3.1 Overview of the clinical characteristics of the subjects

The study’s foundational analysis of 500 GDM patients and 502 non-diabetic controls (Table 1) revealed several critical baseline differences. Notably, the GDM cohort exhibited significantly higher mean values for key clinical markers, including systolic blood pressure (SBP), diastolic blood pressure (DBP), age, pre-pregnancy body mass index (pre-BMI), fasting plasma glucose (FPG), 1-hour postprandial glucose (1h-PG), and 2-hour postprandial glucose (2h-PG) when compared to the control group (all p < 0.05). This suggests that beyond the defining glucose dysregulation, the GDM group presents with a broader metabolic and physiological profile distinct from non-diabetic individuals, even at baseline. Furthermore, a significant disparity in parity was evident between the GDM and control groups (p < 0.05), indicating that reproductive history may serve as an additional differentiating factor. This finding warrants further exploration into its potential implications for GDM pathogenesis or its role as a risk indicator. Stratified analyses reinforced the robustness of these observations. When stratifying the cohort using age 30 and pre-BMI cut-off points of 18.5 and 24, significant differences between the GDM and control groups persisted (all p < 0.05). This reason suggests that broad population characteristics do not merely drive the observed disparities but remain significant even within specific demographic and anthropometric subgroups, emphasizing the pervasive nature of these differences in the GDM cohort.

Table 1

Table 1. Fundamental and categorized traits of the study subjects.

3.2 The relationship between genetic polymorphisms and GDM risk

3.2.1 Overall analysis results

The control group’s genetic data provides crucial insights into the Hardy-Weinberg Equilibrium (HWE) of two specific SNPs, rs1007888 and rs1552224. These SNPs are located at chromosomal positions 11:72722053 and 22:23898914, respectively. For rs1007888, the major and minor alleles are C and A, with a Minor Allele Frequency (MAF) of 0.486. For rs1552224, the major and minor alleles are T and C, and its MAF is 0.079 (Table 2). Critically, the HWE test yielded high p-values for both SNPs (0.997 for rs1007888 and 0.86 for rs1552224), strongly indicating that both loci are in Hardy-Weinberg equilibrium within the studied population. This finding is fundamental, as it confirms that the control group is representative of an unperturbed genetic population, making it a reliable baseline for further genetic association studies.

Table 2

Table 2. Information on SNPs and Hardy-Weinberg equilibrium (HWE) test among the controls.

3.3 Association between SNPs and the risk of GDM in all of the subjects

In our study, even after accounting for potential confounding factors such as pre-BMI, systolic and diastolic blood pressure, maternal age, and parity, a robust association between the rs1552224 genetic variant and a reduced risk of GDM remained evident. Specifically, analysis within the dominant model (AA genotype compared to AC + CC genotypes) yielded an odds ratio of 0.641 (95% CI: 0.429-0.959; p = 0.030), while the allele model (A vs. C allele) showed an odds ratio of 0.624 (95% CI: 0.425-0.916; p = 0.016). These findings, detailed in Table 3, strongly suggest a protective effect of the variant against GDM. Conversely, our analysis revealed no significant correlation between the rs1007888 locus and GDM risk (Table 3).

Table 3

Table 3. The correlations between SNPs and the risk of GDM in all of the subjects.

3.4 Stratified analysis of the differences between SNPs in the ARAP1 and MIF genes and GDM risk in subjects under 30 years of age.

Our stratified analysis across six genetic models, meticulously accounting for age and pre-BMI, revealed a compelling association between specific SNPs and GDM risk in subjects under 30 years of age. Notably, the ARAP1 rs1552224 allele model demonstrated a significant protective effect, correlating with a substantially lower incidence of GDM in women under 30 years of age. This association remained robust even after comprehensive adjustment for potential confounding factors (A vs. C: OR = 0.490, 95% Confidence Interval: 0.281-0.857; p = 0.012). Further supporting this finding, the dominant model (AA vs. AC + CC: OR = 0.523, 95% Confidence Interval: 0.292-0.938; p = 0.030) also indicated a statistically significant reduction in GDM risk within this younger subgroup (Table 4). These results suggest that ARAP1 rs1552224 warrants further investigation as a potential genetic marker for GDM susceptibility, particularly in younger individuals.

Table 4

Table 4. The connections between SNPs in the ARAP1 and MIF genes and GDM risk in subjects under 30 years of age.

3.5 The association between SNPs and GDM risk in individuals with a pre-BMI of 24 or above

In our investigation, we observed a compelling association between the ARAP1 rs1552224 allele and a reduced risk of GDM specifically within a subset of females. Among those with a pre-BMI of 24 kg/m² or greater, the presence of the C allele of the ARAP1 rs1552224 SNP was significantly associated with 65.5% lower odds of developing GDM compared to carriers of the A allele (OR = 0.345, 95% CI: 0.124-0.960; p = 0.042). This finding, presented in Table 5, suggests a potential protective role of the ARAP1 rs1552224 C allele in GDM susceptibility, particularly in the context of higher pre-BMI. Further research is warranted to elucidate the underlying biological mechanisms by which ARAP1 may influence glucose homeostasis and GDM risk in this specific demographic.

Table 5

Table 5. The correlations between SNPs and GDM risk in individuals with a pre-BMI of 24 or above.

Furthermore, our analysis did not reveal any significant associations between the genetic factors investigated and GDM across the other subcategories (Supplementary Tables 1–3). Furthermore, a one-way ANOVA was conducted to explore potential relationships between polymorphism genotypes and various blood glucose parameters. However, the associations between the genotypes of the SNPs and fasting plasma glucose (FPG), 1-hour post-glucose (1h-PG), and 2-hour post-glucose (2h-PG) levels were not statistically significant, as all p-values consistently exceeded 0.05 (Supplementary Tables 4–6).

3.6 Meta-analysis results

Our meta-analysis findings reveal a significant association between the rs1007888 polymorphism and an elevated risk of GDM across multiple genetic models, including the codominant heterozygote, codominant homozygote, allelic, dominant, and over-dominant models (Figure 1). This consistent association strongly suggests a role for rs1007888 in GDM susceptibility. Furthermore, subgroup analysis highlighted the rs1552224 polymorphism as significantly linked to an increased GDM risk in the codominant heterozygote, allelic, dominant, and over-dominant models. While the codominant homozygote and recessive models for rs1552224 did not show a broad overall effect on T2DM, they did exert a statistically significant influence specifically on GDM, with respective p-values of 0.035 and 0.044. Other genetic models for this polymorphism did not yield substantial correlations (Figure 2).

Figure 1

Figure 1. A meta-analytic method employing a fixed effects strategy to explore the link between MIF rs1007888 and the vulnerability to GDM. (A) Codominant Heterozygous Model, CC versus CT (B). Codominant Homozygous Model, CC versus TT (C) Aelle model, C versus T (D) Dominant Model, CC in contrast to CT+TT (E) Recessive Model, CT+ CC versus TT (F) Over-dominant model, CC+TT versus CT. OR stands for odds ratio, CI for confidence interval, and I-squared represents the measure used to assess the level of diversity in the meta-analytic method.

Figure 2

Figure 2. A meta-analytic method employing a fixed effects strategy to explore the link between ARAP1 rs1552224 and the susceptibility to GDM. (A) Codominant Heterozygous Model, AA versus AC (B). Codominant Homozygous Model, AA versus CC (C) Aelle model, A versus C (D). Dominant Model, AA versus AC+CC (E) Recessive Model, AC+AA versus CC (F) Over-dominant model, AA+CC versus AC. OR stands for odds ratio, CI for confidence interval, and I-squared represents the measure used to assess the level of diversity in the meta-analytic method.

Regarding potential publication bias, Begg’s test indicated no substantial bias across the genetic frameworks (all p > 0.05). However, the observed asymmetry in the funnel plot (Supplementary Figures 1, 2) warrants consideration. While this asymmetry could suggest publication bias, it’s also important to acknowledge that it may arise from other factors, such as underlying heterogeneity in study outcomes or variations in study size.

4 Discussion

Baseline characteristics from our cohort showed that women who developed gestational diabetes mellitus (GDM) had significantly higher systolic and diastolic blood pressure, fasting plasma glucose, 1-hour and 2-hour postprandial glucose levels, pre-pregnancy BMI, and mean age compared with controls (all p < 0.05, several p < 0.001). These observations are consistent with previous reports and point to several mechanistic underpinnings (18, 19).

Although none of the GDM participants met the diagnostic threshold for hypertension, their higher blood pressure values suggest early manifestations of insulin resistance–related endothelial dysfunction and subclinical inflammation (20). This pathophysiological background helps to explain the increased risk of preeclampsia and gestational hypertension in GDM pregnancies (4, 21).

The most pronounced group differences were seen in postprandial glucose excursions particularly at 1 and 2 hours supporting the concept that postprandial hyperglycaemia represents the core metabolic abnormality of GDM (4). This likely reflects inadequate first-phase insulin release and reduced peripheral glucose uptake.

Pre-pregnancy BMI emerged as another critical determinant. Even within the “normal” range (18.5–24 kg/m²), women who later developed GDM had higher pre-BMI values, suggesting that mild elevations in adiposity may confer risk. Adipokines such as leptin and TNF-α, secreted from adipose tissue, may further aggravate insulin resistance and impair β-cell function (22, 23).

Maternal age was also independently associated with GDM: the mean age of affected women was 31 years compared with 29 years in controls (p < 0.001), and stratified analysis showed that those aged ≥30 years had more pronounced metabolic impairments. This is consistent with evidence that both insulin sensitivity and β-cell function decline progressively with advancing age (4, 24).

Collectively, these findings underscore that even modest elevations in blood pressure, adiposity, and maternal age contribute to GDM risk. They highlight the need for pre-pregnancy counselling and weight optimisation in women of reproductive age as a preventive strategy to reduce the burden of GDM and its complications.

The human macrophage migration inhibitory factor (MIF) gene, located on chromosome 22q11.2, is constitutively and abundantly expressed in immune cells, particularly T lymphocytes and macrophages (25). MIF exerts diverse biological functions, and altered expression has been reported in several inflammatory and metabolic disorders, including rheumatoid arthritis, atherosclerosis, and diabetes (26). In the placental decidua, macrophages and dendritic cells regulate T-cell activation, initiating immune responses. Subsequent differentiation of T cells into Th1 or Th2 subtypes—driven by both intrinsic and extrinsic factors alters cytokine synthesis and distribution. These shifts directly influence MIF regulation and are essential for immunological adaptation across pregnancy stages (15).

Beyond its role in immunity, MIF functions as a pleiotropic cytokine linking inflammation to glucose metabolism. At the molecular level, it regulates glucose transporter 4 and fructose-bisphosphatase 2 expression (8). Moreover, MIF promotes TNF-α production (27, 28), thereby influencing glycolysis and inflammatory signalling. In adipocytes, MIF-driven TNF-α secretion contributes to insulin resistance (8), while in endothelial cells, MIF disrupts insulin-mediated nitric oxide release, fostering endothelial insulin resistance (8).

These mechanisms align with the lipid overload hypothesis, which posits that ectopic lipid deposition in insulin-sensitive tissues such as skeletal muscle, pancreas, and liver underpins insulin resistance (29–31). Supporting this, both adipose tissue MIF mRNA and circulating MIF concentrations are elevated in obese women with type 2 diabetes (32). Elevated adipose-derived MIF in obesity and diabetes further impairs insulin signalling and amplifies pro-inflammatory cascades, thereby reinforcing insulin resistance (33).

Despite previous findings linking MIF variants to T2D, our study found no significant association between the MIF rs1007888 variant and GDM. This divergence suggests that pregnancy-specific genetic or physiological factors might influence the role of this genetic locus in GDM development (6, 32, 33). Moreover, despite the known involvement of MIF in inflammatory and metabolic disorders (34), its genetic influence via rs1007888 on GDM risk within the context of higher pre-BMI appears to be negligible or absent in our specific cohort. This suggests that the genetic impact of rs1007888 on GDM risk may be either minor or does not significantly interact with the metabolic environment of pre-existing obesity, at least in the population we investigated. It is plausible that other genetic or environmental factors exert a more predominant influence on GDM development within this particular BMI subgroup.

ARAP1, which encodes the protein Arap1 (ARF-GAP, Rho-GAP Anchoring Protein Repeat Sequence, and Pleckstrin Homology Domain Protein 1, also known as centaurin delta 2), functions as a phosphatidylinositol 1,4,5-trisphosphate-regulated Arf GTPase-activating protein. This protein primarily targets the small GTP-binding protein Arf6, which is known to play a critical role in the modulation of insulin exocytosis (6). The rs1552224 SNP is situated within an intronic region of the ARAP1 gene.

Research indicates that ARAP1 is significantly associated with elevated blood sugar levels and reduced insulin secretion in response to glucose stimulation, suggesting that impaired pancreatic β-cell function may be a mechanism through which this genetic locus contributes to the development of diabetes (35). Recent findings propose that the expression of the ARAP1 locus is predominantly influenced by markedly increased levels of STARD10 expression in the pancreas (36).

The genetic variant STARD10 rs11603334 is functionally linked to the expression of ARAP1 and is in complete linkage disequilibrium with rs1552224. The risk allele (C) of STARD10 rs11603334 for type 2 diabetes is positioned near the ARAP1 promoter. This specific placement interferes with the collaborative complex formation between the transcriptional regulators PAX6 and PAX4. This interference leads to an increased amplification of ARAP1 P1 promoter transcription, resulting in higher ARAP1 transcription levels within the Islets of Langerhans (37).

Conversely, research by Carrat et al (36) suggests that compromised insulin secretion from pancreatic and β-cells is associated with decreased STARD10 levels, rather than elevated ARAP1. Furthermore, the rs1552224C allele might offer a protective effect against type 2 diabetes. This protective role could be mediated by the overexpression of STARD10, which in turn enhances Ca²⁺ dynamics and glucose-triggered insulin release from pancreatic and β-cells (34). Despite these findings, the precise operational mechanism of this process in GDM requires further extensive investigation.

The protective influence of the C allele at rs1552224 warrants particular attention. ARAP1 (ArfGAP with RhoGAP domain, ankyrin repeat, and PH domain 1) plays a critical role in insulin secretion and glucose homeostasis by regulating membrane trafficking and cytoskeletal dynamics within pancreatic β cells (38). While previous research has linked ARAP1 polymorphisms to an elevated risk of type 2 diabetes (39, 40), its specific contribution to GDM, especially in younger populations, remains underexplored. Our findings suggest that the C allele may modify ARAP1’s function, potentially enhancing β-cell compensation or insulin sensitivity to meet the increased metabolic demands of pregnancy, thereby lowering GDM risk. This aligns with existing evidence that genetic variations affecting β-cell function are significant predictors of GDM susceptibility (41).

In the context of the allelic model for the ARAP1 gene, the presence of each C allele is associated with a 38% reduction in the risk of developing GDM. This observation is consistent with the established role of ARAP1 in insulin granule trafficking and the functionality of βcells (37). Notably, the absence of CC homozygotes among GDM cases (0% compared to 0.8% in controls) indicates a potential dosage-dependent protective effect. However, the low minor allele frequency (MAF = 5–8%) (42) may limit the statistical power of analyses based on recessive models.

Furthermore, our results corroborate previous functional studies that demonstrate that ARAP1 knockdown adversely affects glucose-stimulated insulin secretion (GSIS) in β cells Benjamin (31). The protective C allele may enhance the expression of ARAP1, thereby improving the first-phase insulin response, which is crucial for regulating postprandial glucose levels. This is consistent with the elevated 1-hour and 2-hour post-glucose challenge levels presented in Table 1.

The observation that the rs1552224 C allele exhibits a protective association in individuals with elevated pre-BMI is a compelling finding. Given that obesity is a well-established driver of insulin resistance and a significant risk factor for GDM (43), our data imply that the C allele may offer a unique advantage by ameliorating the adverse metabolic consequences of obesity during gestation (38).

The SNP rs1552224 has the potential to enhance the prediction of GDM risk in women with high BMI and nulliparous status. However, its modest effect size, with an odds ratio of approximately 0.6, limits its utility as a standalone predictive marker. We hypothesize that this allele could modulate ARAP1 function in a manner that either bolsters β-cell compensatory mechanisms or enhances insulin signaling pathways. This modulation would be particularly beneficial in counteracting the heightened insulin resistance characteristic of a higher pre-BMI, especially within a population already predisposed to GDM due to their initial weight. This interpretation aligns with the broader concept that genetic variants can influence an individual’s susceptibility to obesity-induced metabolic dysfunction (44).

A comparison of findings in Table 3 with the data in Table 5, which utilized age as a stratification factor, indicates that the protective effect of the rs1552224 C allele is consistently observed across various risk stratifications, including age and BMI. This consistency enhances the potential clinical significance of this genetic variant as a biomarker for assessing the risk of GDM. Our research highlights the critical interaction between genetic factors and environmental influences, such as pre-BMI, in shaping susceptibility to GDM. By identifying genetic variants that may either increase or decrease the risk associated with obesity, we can develop more tailored strategies for the prevention and management of GDM. Future research should investigate the functional implications of rs1552224 and its potential role in targeted interventions for pregnant women with elevated BMI.

This meta-analysis provides a comprehensive assessment of the association between the rs1007888 variant and gestational diabetes mellitus (GDM), synthesising evidence from three independent studies (5, 45) together with data from Zhang et al. (our current study, 2024). The pooled analyses, illustrated in the accompanying forest plots (A–F), consistently demonstrate an increased risk of GDM associated with this locus, thereby offering new insights into the genetic basis of disease susceptibility.

In particular, Figure 1 highlights that Plots A and B yielded statistically significant associations, with pooled odds ratios of 1.48 and 1.45, respectively. Importantly, both models displayed very low between-study heterogeneity (I² = 0.0% for A and 17.6% for B), underscoring the stability and reproducibility of these findings. Such homogeneity suggests that the genetic effect captured in these models—likely reflecting an additive or dominant contribution of the risk allele—represents a robust and biologically meaningful determinant of GDM.

The consistency of associations across both Chinese cohorts (5, 45) and an Iranian cohort (6) further strengthens the evidence, supporting the notion of a shared genetic predisposition that transcends ethnic boundaries. This aligns with prior meta-analyses reporting that common genetic variants contribute significantly to GDM risk across diverse populations (26, 41). The low heterogeneity observed in several key models adds to the reliability of our effect estimates and suggests that rs1007888 may represent a stable genetic marker of GDM susceptibility.

Our study has several significant limitations that should be acknowledged. First, the relatively small number of studies included in the meta-analysis restricts the strength of the conclusions. This underscores the need for larger, multi-ethnic investigations to validate our findings and to explore potential gene-environment interactions. Functional studies will also be essential to elucidate the biological mechanisms by which these variants may contribute to GDM risk. Ultimately, identifying reliable genetic markers could inform personalized risk stratification, early screening, and targeted prevention strategies for GDM.

Discrepancies were observed between the results of our case–control analysis, the meta-analysis, and previous literature. We propose that a limited sample size represents a key contributor to these inconsistencies. Power calculations using G*Power software indicated that approximately 19,788 participants would be required to assess the association with rs1007888 adequately, and 2,273 participants for rs1552224, under the following parameters: test family = Exact; statistical test = Proportions: Inequality, two independent groups (Fisher’s exact test); test direction = two-tailed; power = 0.8; significance level α = 0.05.

The challenges of sample size requirements differ between variants. For rs1552224, the low minor allele frequency (0.079) limits statistical power, necessitating large cohorts to detect associations. For rs1007888, although the reference allele is more common (p2 = 0.486), the effect size is minimal (OR = 0.923), which again demands a substantial cohort to achieve reliable power. This highlights a broader difficulty in genetic association studies of weak-effect variants, where enormous sample sizes are often required to detect modest associations.

In addition to sample size, the absence of comprehensive adjustment for potential confounders may have contributed to the discrepancies observed. Significant clinical and biochemical variables such as lipid profiles, fasting glucose, HOMA indices, C-peptide concentrations, and comorbidities were not systematically included. Moreover, the lack of multicenter data collection limits generalizability, as regional differences within China may influence allele frequencies and disease risk. Addressing these issues in future studies through larger sample sizes, standardized adjustment factors, and multicenter recruitment will be crucial to resolving these inconsistencies and advancing understanding of the genetic basis of GDM.

5 Conclusions

This study investigated the relationship between the MIF rs1007888 and ARAP1 rs1552224 genetic variants and the risk of GDM. Our findings revealed no significant association between the MIF rs1007888 variant and GDM risk. ARAP1 rs1552224 was significantly linked to reduced GDM incidence. In contrast, meta-analysis results indicated that the MIF rs1007888 mutation was associated with an increased likelihood of developing GDM. Furthermore, the ARAP1 rs1552224 mutation was linked to a heightened risk of GDM. The discrepancies between our findings and those of previous studies may be attributed to limitations in sample size and ethnic diversity. To enhance the understanding of the relationship between these genetic variants and GDM risk, future research should prioritize increasing sample sizes and incorporating a broader range of adjustment factors and diverse ethnicities.

Data availability statement

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

Ethics statement

The studies involving humans were approved by Guangdong Medical University’s Shunde Women and Children’s Hospital (Maternity and Child Healthcare Hospital of Shunde Foshan) gave its approval for the human subjects’ study. The studies were carried out in compliance with institutional norms and local laws. To take part in this study, the subjects gave their written informed consent. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

Author contributions

YZ: Formal analysis, Data curation, Methodology, Writing – original draft. YYW: Writing – original draft, Conceptualization. QZ: Conceptualization, Software, Writing – original draft, Methodology. WG: Writing – review & editing, Investigation. SH: Investigation, Software, Writing – review & editing. XD: Writing – review & editing, Data curation, Formal analysis. JL: Methodology, Investigation, Writing – review & editing. XL: Investigation, Writing – review & editing. YW: Supervision, Validation, Writing – review & editing, Visualization. RG: Supervision, Funding acquisition, Writing – review & editing, Resources.

Funding

The author(s) declare financial support was received for the research and/or publication of this article. Supported by the National Natural Science Foundation of China (81873649); Doctoral scientific research Initiate funding project of Shunde Women and Children s Hospital of Guangdong Medical University (Maternity and Child Healthcare Hospital of Shunde Foshan) (2020BSQD007); Self-financing science and technology project of Foshan (2320001006049).

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declare that no Generative AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fendo.2025.1650782/full#supplementary-material

References

1. McIntyre HD, Catalano P, Zhang C, Desoye G, Mathiesen ER, and Damm P. Gestational diabetes mellitus. Nat Rev Dis Primers. (2019) 5:1–19. doi: 10.1038/s41572-019-0098-8

PubMed Abstract | Crossref Full Text | Google Scholar

2. Yan J, Su R, Ao D, Wang Y, Wang H, and Yang H. Genetic variants and clinical relevance associated with gestational diabetes mellitus in Chinese women: a case-control study. J Maternal-Fetal Neonatal Med. (2018) 31:2115–21. doi: 10.1080/14767058.2017.1336225

PubMed Abstract | Crossref Full Text | Google Scholar

3. Sweeting A, Wong J, Murphy HR, and Ross GP. A clinical update on gestational diabetes mellitus. Endocr Rev. (2022) 43:763–93. doi: 10.1210/endrev/bnac003

PubMed Abstract | Crossref Full Text | Google Scholar

4. Gyan WR, Zhang H, Shao T, Yang T, Wei Y, Li M, et al. Association of CDKAL1 gene polymorphisms variations with gestational diabetes mellitus risk in women: A case-control study and meta-analysis. BMC Endocr Disord. (2025) 25:1–12. doi: 10.1186/s12902-025-01874-8

PubMed Abstract | Crossref Full Text | Google Scholar

5. Zhan Y, Wang Y, Li C, Liu S, and Gao Q. Association between single nucleotide polymorphism of macrophage migration inhibitory factor-rs1007888 and the pathogenesis of gestational diabetes mellitus. Chin J Obstet Gynecol. (2013) 48:5. doi: 10.3760/cam.j.issn.0529-567x.2013.05.002

PubMed Abstract | Crossref Full Text | Google Scholar

6. Aslani S, Hossein-Nezhad A, Maghbooli Z, Mirzaei K, and Karimi F. Genetic variation in macrophage migration inhibitory factor associated with gestational diabetes mellitus and metabolic syndrome. Hormone Metab Res. (2011) 43:557–61. doi: 10.1055/s-0031-1275706

PubMed Abstract | Crossref Full Text | Google Scholar

7. Rutter GA. Cellular and animal models of type 2 diabetes GWAS gene polymorphisms: What can we learn? Drug Discov Today Dis Models. (2013) 10:e59–64. doi: 10.1016/j.ddmod.2013.02.002

Crossref Full Text | Google Scholar

8. Jing L, Yun P, Fu-Kai B, and Ai-Hua L. Correlation between gene polymorphism of macrophage migration inhibitory factor and disease susceptibility: Recognition, target and significance. Chin J Tissue Eng Res. (2018) 22:4574–9. doi: 10.3969/j.issn.2095-4344.0299

Crossref Full Text | Google Scholar

9. Zheng L, Li C, Qi W, Qiao B, Zhao H, Zhou Y, et al. Expression of macrophage migration inhibitory factor gene in placenta tissue and its correlation with gestational diabetes mellitus. Natl Med J China. (2017) 97:43. doi: 10.3760/cma.j.issn.0376-2491.2017.43.006

PubMed Abstract | Crossref Full Text | Google Scholar

10. Nielsen T, Sparsø T, Grarup N, Jørgensen T, Pisinger C, Witte DR, et al. Type 2 diabetes risk allele near CENTD2 is associated with decreased glucose-stimulated insulin release. Diabetologia. (2011) 54:1052–6. doi: 10.1007/s00125-011-2054-3

PubMed Abstract | Crossref Full Text | Google Scholar

11. Voight BF, Scott LJ, Steinthorsdottir V, Morris AP, Dina C, Welch RP, et al. Twelve type 2 diabetes susceptibility loci identified through large-scale association analysis. Nat Genet. (2010) 42:579–89. doi: 10.1038/ng.609

PubMed Abstract | Crossref Full Text | Google Scholar

12. Gao L, Xiong YJ, Liang YX, Huang PF, Liu S, Xiao Y, et al. The effects of IL-27 and IL-35 gene variation and expression levels on the susceptibility and clinical manifestations of pulmonary tuberculosis. Front Immunol. (2024) 15:1267624. doi: 10.3389/fimmu.2024.1267624

PubMed Abstract | Crossref Full Text | Google Scholar

13. Arora GP, Almgren P, Brøns C, Thaman RG, Vaag AA, Groop L, et al. Association between genetic risk variants and glucose intolerance during pregnancy in north Indian women. BMC Med Genomics. (2018) 11:1–10. doi: 10.1186/s12920-018-0380-8

PubMed Abstract | Crossref Full Text | Google Scholar

14. Li C, Zhou P, Cai Y, Peng B, Liu Y, Yang T, et al. Associations between gestational diabetes mellitus and the neurodevelopment of offspring from 1 month to 72 months: Study protocol for a cohort study. BMJ Open. (2020) 10:1–6. doi: 10.1136/bmjopen-2020-040305

PubMed Abstract | Crossref Full Text | Google Scholar

15. Jovanović Krivokuća M, Vilotić A, Stefanoska I, Bojić-Trbojević Ž, and Vićovac L. Macrophage migration inhibitory factor in human early pregnancy events and association with placental pathologies. Placenta. (2021) 116:51–7. doi: 10.1016/j.placenta.2021.02.007

PubMed Abstract | Crossref Full Text | Google Scholar

16. Yoo SA, Leng L, Kim BJ, Du X, Tilstam PV, Kim KH, et al. MIF allele-dependent regulation of the MIF coreceptor CD44 and role in rheumatoid arthritis. Proc Natl Acad Sci U.S.A. (2016) 113:E7917–26. doi: 10.1073/pnas.1612717113

PubMed Abstract | Crossref Full Text | Google Scholar

17. Cai XQ, Huang Q, and Zhang TP. The methylation in B7-H4 and BTLA genes are associated with the risk of pulmonary tuberculosis. Immunotargets Ther. (2023) 12:149–63. doi: 10.2147/ITT.S434403

PubMed Abstract | Crossref Full Text | Google Scholar

18. Dorans KS, Bazzano LA, Li X, Bundy JD, Tian L, and He J. Lifestyle behaviors and cardiovascular risk profiles among parous women by gestational diabetes status, 2007–2018. Nutrition Metab Cardiovasc Dis. (2022) 32:1121–30. doi: 10.1016/j.numecd.2022.01.012

PubMed Abstract | Crossref Full Text | Google Scholar

19. Santanapipatkul P, Luewan S, Sittiwangkul R, Krongphaiklang N, Jatavan P, and Tongsong T. Comparison of fetal and neonatal cardiac morphology between the infants of mothers with well-controlled gestational diabetes mellitus and normal controls. Diabetes Vasc Dis Res. (2023) 20:1–6. doi: 10.1177/14791641231190531

PubMed Abstract | Crossref Full Text | Google Scholar

20. Luk C, Haywood NJ, Bridge KI, and Kearney MT. Paracrine role of the endothelium in metabolic homeostasis in health and nutrient excess. Front Cardiovasc Med. (2022) 9:882923. doi: 10.3389/fcvm.2022.882923

PubMed Abstract | Crossref Full Text | Google Scholar

21. Zeng Q, Tan B, Han F, Huang X, Huang J, Wei Y, et al. Association of solute carrier family 30 A8 zinc transporter gene variations with gestational diabetes mellitus risk in a Chinese population. Front Endocrinol (Lausanne). (2023) 14:1159714. doi: 10.3389/fendo.2023.1159714

PubMed Abstract | Crossref Full Text | Google Scholar

22. Moghbeli M, Khedmatgozar H, Yadegari M, Avan A, Ferns GA, and Ghayour Mobarhan M. Cytokines and the immune response in obesity-related disorders. In: Advances in clinical chemistry. Amsterdam, the Netherlands: Academic Press Inc (2021). p. 135–68. doi: 10.1016/bs.acc.2020.06.004

PubMed Abstract | Crossref Full Text | Google Scholar

23. Al-Mansoori L, Al-Jaber H, Prince MS, and Elrayess MA. Role of inflammatory cytokines, growth factors and adipokines in adipogenesis and insulin resistance. Inflammation. (2022) 45:31–44. doi: 10.1007/s10753-021-01559-z

PubMed Abstract | Crossref Full Text | Google Scholar

24. Zeng Q, Zou D, Gu S, Han F, Cao S, Wei Y, et al. Different associations between CDKAL1 variants and type 2 diabetes mellitus susceptibility: A meta-analysis. Front Genet. (2022) 12:783078. doi: 10.3389/fgene.2021.783078

PubMed Abstract | Crossref Full Text | Google Scholar

25. Sun C, Xu J, Zhang B, Huang H, Chen L, Yan H, et al. Whole-genome sequencing suggests a role of MIF in the pathophysiology of TEMPI syndrome. Blood Adv. (2021) 5:2563–8. doi: 10.1182/bloodadvances.2020003783

PubMed Abstract | Crossref Full Text | Google Scholar

26. Sumaiya K, Langford D, Natarajaseenivasan K, and Shanmughapriya S. Macrophage migration inhibitory factor (MIF): A multifaceted cytokine regulated by genetic and physiological strategies. Pharmacol Ther. (2022) 233:1–20. doi: 10.1016/j.pharmthera.2021.108024

PubMed Abstract | Crossref Full Text | Google Scholar

27. Miller EJ, Li J, Leng L, McDonald C, Atsumi T, Bucala R, et al. Macrophage migration inhibitory factor stimulates AMP-activated protein kinase in the ischaemic heart. Nature. (2008) 451:578–82. doi: 10.1038/nature06504

PubMed Abstract | Crossref Full Text | Google Scholar

28. Atsumi T, Cho Y-R, Leng L, Mcdonald C, Yu T, Danton C, et al. The proinflammatory cytokine macrophage migration inhibitory factor regulates glucose metabolism during systemic inflammation 1. J Immunol. (2007) 179:5399–406. doi: 10.4049/jimmunol.179.8.5399

PubMed Abstract | Crossref Full Text | Google Scholar

29. Elbein SC, Gamazon ER, Das SK, Rasouli N, Kern PA, and Cox NJ. Genetic risk factors for type 2 diabetes: A trans-regulatory genetic architecture? Am J Hum Genet. (2012) 91:466–77. doi: 10.1016/j.ajhg.2012.08.002

PubMed Abstract | Crossref Full Text | Google Scholar

30. Toso C, Emamaullee JA, Merani S, and Shapiro AMJ. The role of macrophage migration inhibitory factor on glucose metabolism and diabetes. Diabetologia. (2008) 51:1937–46. doi: 10.1007/s00125-008-1063-3

PubMed Abstract | Crossref Full Text | Google Scholar

31. Tsatsoulis A, Mantzaris MD, Bellou S, and Andrikoula M. Insulin resistance: An adaptive mechanism becomes maladaptive in the current environment - An evolutionary perspective. Metabolism. (2013) 62:622–33. doi: 10.1016/j.metabol.2012.11.004

PubMed Abstract | Crossref Full Text | Google Scholar

32. Herder C, Klopp N, Baumert J, Müller M, Khuseyinova N, Meisinger C, et al. Effect of macrophage migration inhibitory factor (MIF) gene variants and MIF serum concentrations on the risk of type 2 diabetes: Results from the MONICA/KORA Augsburg Case-Cohort Study, 1984-2002. Diabetologia. (2008) 51:276–84. doi: 10.1007/s00125-007-0800-3

PubMed Abstract | Crossref Full Text | Google Scholar

33. Stošić-Grujičić S, Saksida T, Miljković Đ, and Stojanović I. MIF and insulin: Lifetime companions from common genesis to common pathogenesis. Cytokine. (2020) 125:1–10. doi: 10.1016/j.cyto.2019.154792

PubMed Abstract | Crossref Full Text | Google Scholar

34. Li Y, Shen K, Li C, Yang Y, Yang M, Tao W, et al. Identifying the association between single nucleotide polymorphisms in kcnq1, arap1, and kcnj11 and type 2 diabetes mellitus in a chinese population. Int J Med Sci. (2020) 17:2379–86. doi: 10.7150/ijms.48072

PubMed Abstract | Crossref Full Text | Google Scholar

35. Wang L, Wu J, Sramek M, Obayomi SMB, Gao P, Li Y, et al. Heterogeneous enhancer states orchestrate β cell responses to metabolic stress. Nat Commun. (2024) 15:1–19. doi: 10.1038/s41467-024-53717-0

PubMed Abstract | Crossref Full Text | Google Scholar

36. Carrat GR, Hu M, Nguyen-Tu MS, Chabosseau P, Gaulton KJ, van de Bunt M, et al. Decreased STARD10 expression is associated with defective insulin secretion in humans and mice. Am J Hum Genet. (2017) 100:238–56. doi: 10.1016/j.ajhg.2017.01.011

PubMed Abstract | Crossref Full Text | Google Scholar

37. Li L, Xu L, Wen S, Yang Y, Li X, and Fan Q. The effect of lncRNA-ARAP1-AS2/ARAP1 on high glucose-induced cytoskeleton rearrangement and epithelial–mesenchymal transition in human renal tubular epithelial cells. J Cell Physiol. (2020) 235:5787–95. doi: 10.1002/jcp.29512

PubMed Abstract | Crossref Full Text | Google Scholar

38. Zhang Z, Xie W, Gong B, Liang X, Yu H, Yu Y, et al. ARAP1 negatively regulates stress fibers formation and metastasis in lung adenocarcinoma via controlling Rho signaling. Discover Oncol. (2023) 14:1–14. doi: 10.1007/s12672-023-00832-x

PubMed Abstract | Crossref Full Text | Google Scholar

39. Neale BM. It’s ARAP” for type II diabetes. Sci Transl Med. (2014) 6:221ec22–221ec22. doi: 10.1126/scitranslmed.3008474

Crossref Full Text | Google Scholar

40. Horigan G, McGuigan K, Quinn L, and Coates V. Poster: Does gender make a difference to self-management of diabetes in adults ≤ 55 years? A pilot study. Diabetic Med. (2015) 32(S1):140–1. doi: 10.1111/dme.12668_1

Crossref Full Text | Google Scholar

41. Dalfrà MG, Burlina S, Del Vescovo GG, and Lapolla A. Genetics and epigenetics: new insight on gestational diabetes mellitus. Front Endocrinol (Lausanne). (2020) 11:602477. doi: 10.3389/fendo.2020.602477

PubMed Abstract | Crossref Full Text | Google Scholar

42. Gong Y, Zack TI, Morris LGT, Lin K, Hukkelhoven E, Raheja R, et al. Pan-cancer genetic analysis identifies PARK2 as a master regulator of G1/S cyclins. Nat Genet. (2014) 46:588–94. doi: 10.1038/ng.2981

PubMed Abstract | Crossref Full Text | Google Scholar

43. Gao C, Sun X, Lu L, Liu F, and Yuan J. Prevalence of gestational diabetes mellitus in mainland China: A systematic review and meta-analysis. J Diabetes Investig. (2019) 10:154–62. doi: 10.1111/jdi.12854

PubMed Abstract | Crossref Full Text | Google Scholar

44. Schnurr TM, Jakupović H, Carrasquilla GD, Ängquist L, Grarup N, Sørensen TIA, et al. Obesity, unfavourable lifestyle and genetic risk of type 2 diabetes: a case-cohort study. Diabetologia. (2020) 63:1324–32. doi: 10.1007/s00125-020-05140-5

PubMed Abstract | Crossref Full Text | Google Scholar

45. Zhan Y, Li C, Chen J, Yu S, Gao Q, Wang YP, et al. Association between macrophage migration inhibitory factor rs1007888 and GDM. Genet Mol Res. (2015) 14:797–804. doi: 10.4238/2015

PubMed Abstract | Crossref Full Text | Google Scholar