Gene Variants Tied to Gestational Diabetes Risk in Iranian Pregnant Women

Gestational diabetes mellitus, a form of glucose intolerance first recognized during pregnancy, affects an estimated 14 to 17 percent of pregnancies worldwide and raises the risk of perinatal complications as well as later type 2 diabetes in the mother. While obesity, advanced maternal age, and family history are well-established clinical risk factors, genetic predisposition is also a key component of susceptibility. Now, a case-control study of 400 Iranian pregnant women has reported that specific variants in three genes previously linked to gestational diabetes in East Asian populations also show associations in a Middle Eastern cohort, a population in which these variants had not been examined before.

The study, conducted by researchers affiliated with Mashhad University of Medical Sciences, recruited 200 women with gestational diabetes and 200 healthy controls from tertiary care maternity hospitals and affiliated clinics in Mashhad, in Razavi Khorasan Province. Participants were Iranian pregnant women aged 18 to 45 with singleton pregnancies who underwent a 75-gram oral glucose tolerance test between 24 and 48 weeks of gestation. Gestational diabetes was diagnosed according to the criteria of the International Association of the Diabetes and Pregnancy Study Groups. Women with pre-existing diabetes, chronic hypertension, or renal disease were excluded from the study.

The researchers focused on six single nucleotide polymorphisms selected on the basis of prior genome-wide association study evidence in other populations, minor allele frequencies above 0.20 in Asian populations, and their locations in functional genomic regions including intronic, regulatory, 3’UTR, and splice-site regions. Two variants came from each of three genes: rs1545458 and rs905941 in SYNPR, which encodes synaptoporin; rs1344692 and rs11744487 in CDH18, which encodes cadherin 18; and rs6095547 and rs5602 in PTGIS, which encodes prostacyclin synthase. Genotyping was performed using the Amplification-Refractory Mutation System, a polymerase chain reaction method, with quality control monitored through 5 percent duplicate samples.

After adjustment for age, height, smoking status, history of macrosomia, and family history of type 2 diabetes, several of the variants showed significant associations with gestational diabetes risk. In the CDH18 gene, the rs11744487 variant was associated with the condition, with the AC genotype conferring an odds ratio of 0.327 compared with the AA genotype, a 95 percent confidence interval of 0.165 to 0.647, and a p-value of 0.001. In the PTGIS gene, rs6095547 showed a markedly elevated risk with the CT genotype compared with CC, yielding an odds ratio of 3.839 with a 95 percent confidence interval of 1.849 to 7.969 and a p-value below 0.001. The rs5602 variant in the same gene showed the opposite direction of effect, with the CT genotype associated with reduced odds, an odds ratio of 0.154, a 95 percent confidence interval of 0.067 to 0.356, and a p-value below 0.001.

This pattern of both risk and protective alleles within a single gene was among the most striking findings of the study. The authors suggest that PTGIS may harbor multiple functional variants in linkage disequilibrium with the genotyped markers, exerting opposing effects on prostacyclin synthesis and downstream insulin sensitivity, possibly mediated by cis-regulatory variants that influence gene expression in a tissue-specific or context-dependent manner. Notably, the odds ratio observed for rs6095547 is substantially higher than the modest effects of roughly 1.2 to 1.4 reported in some Asian genome-wide studies, a discrepancy the authors attribute to possible differences in allele frequencies between populations, gene-environment interactions specific to Iranian diet or lifestyle, or stochastic variation given the moderate sample size.

Beyond individual variants, haplotype analyses revealed increased gestational diabetes risk for specific haplotype combinations within each gene: AG within SYNPR, GA and AA within CDH18, and TC within PTGIS. Not all variants were associated, however. No significant associations were observed for rs1545458 in SYNPR or rs1344692 in CDH18. All genotyped SNPs were in Hardy-Weinberg equilibrium in the control group, and the Bonferroni correction for multiple comparisons set a significance threshold of p below 0.0083. Associations were assessed using multivariate logistic regression under dominant, recessive, and codominant genetic models, with haplotype estimation and linkage disequilibrium analysis conducted using dedicated software tools.

The biological rationale for examining these three genes draws on earlier work. A genome-wide association study in a Chinese population had identified 23 SNPs significantly associated with gestational diabetes, mapping to four genes: SYNPR, PTGIS, CDH18, and CTIF. PTGIS encodes prostacyclin synthase, a key enzyme in the prostaglandin pathway; prostaglandins are lipid mediators that regulate inflammation, vascular tone, and pancreatic beta-cell function, and prostacyclin production has been shown to influence insulin secretion and vascular function. CDH18, a cell adhesion molecule, has been linked to insulin signal transduction via the Wnt signaling pathway and implicated in the regulation of adiponectin levels and systemic inflammation. SYNPR, involved in synaptic vesicle function, may influence central neurohormonal regulation of metabolism, although its role in gestational diabetes remains less clear.

The clinical characteristics of the participants differed between groups in ways consistent with the diagnosis. Women with gestational diabetes had significantly higher fasting plasma glucose and both one-hour and two-hour oral glucose tolerance values compared with controls, all with p-values below 0.001. Significant differences were also observed in maternal height, smoking status, and infant birth weight, and a positive family history of type 2 diabetes was more prevalent in the gestational diabetes group, at 21.2 percent versus 5.0 percent in controls, a difference that was itself statistically significant.

The authors frame their findings as the first evidence for the involvement of the SYNPR, CDH18, and PTGIS loci in gestational diabetes susceptibility in an Iranian population, and as a replication of previous reports from East Asian cohorts in a Middle Eastern setting. They argue that the spectrum of risk variants can differ across ethnic groups, making validation in diverse populations essential, particularly given Iran’s unique ethnic background, dietary patterns, and high consanguinity rates in some regions. The observation of multiple independent associations, including risk and protective alleles and haplotype effects, reinforces the polygenic model of susceptibility, in which no single variant confers a large effect and risk reflects the cumulative burden of many small-effect alleles. A population-specific polygenic risk score incorporating these variants alongside clinical factors such as BMI and family history could, the authors suggest, eventually help stratify pregnant women for earlier screening or lifestyle intervention.

The study’s limitations are considerable and are acknowledged by the authors. The sample of 400 women is modest for a genetic association study, and there was no independent replication cohort. Despite adjustment for several confounders, residual population stratification within the Iranian population cannot be excluded. The candidate gene approach limits the ability to discover novel variants, and no functional outcomes such as PTGIS expression levels or prostaglandin concentrations were measured. The lack of detailed dietary or physical activity data also precludes analysis of gene-environment interactions. The authors call for functional studies, including reporter assays for PTGIS promoter variants and measurement of prostacyclin levels in carriers of risk versus protective alleles, as well as larger multi-ethnic cohorts including other Middle Eastern populations to validate the findings and develop clinically useful risk prediction tools.

Subject of Research: Genetic variants in SYNPR, CDH18, and PTGIS associated with gestational diabetes mellitus susceptibility in an Iranian case-control cohort

Article Title: Genetic links between SYNPR, CDH18, and PTGIS variants and gestational diabetes mellitus in an Iranian cohort

Article References: Jafarpour, S., Pasdar, A., Akhlaghi, F., & Hamzehloei, T. (2026). Genetic links between SYNPR, CDH18, and PTGIS variants and gestational diabetes mellitus in an Iranian cohort. Egyptian Journal of Medical Human Genetics, 27(1), Article 76. https://doi.org/10.1186/s43042-026-00886-w

Image Credits: AI Generated

DOI: 10.1186/s43042-026-00886-w

Keywords: gestational diabetes, genetic association study, SYNPR, CDH18, PTGIS, single nucleotide polymorphisms, Iranian cohort, haplotype analysis, polygenic risk, case-control study, pregnancy complications, insulin resistance