Comparative analysis of inflammatory and metabolic biomarkers in PGDM and GDM pregnancies

Abstract

Background

Pregestational diabetes mellitus (PGDM) and gestational diabetes mellitus (GDM) are associated with adverse pregnancy outcomes through inflammatory and metabolic dysregulation. This study compared serum levels of pro-early placental insulin-like growth factor (pro-EPIL), galectin-3, sirtuin-1 (SIRT-1), matrix metalloproteinase-10 (MMP-10), chemerin, and interleukin-33 (IL-33) in diabetic pregnancies and explored their inter-correlations.

Methods

This cross-sectional study included pregnant women at 24–39 weeks of gestation, categorized into normal glucose tolerance (NGT, n = 30), PGDM (n = 17), and GDM (n = 33) groups. Serum biomarker concentrations were measured by enzyme-linked immunosorbent assay (ELISA).

Results

Pro-EPIL and galectin-3 were significantly elevated in PGDM compared to NGT (p < 0.05), while GDM showed no significant difference from either group. Both PGDM and GDM demonstrated significantly higher MMP-10, chemerin, and IL-33 levels compared to NGT (p < 0.05). SIRT-1 concentrations did not differ significantly among groups. Correlation analysis revealed that SIRT-1 was positively correlated with chemerin and IL-33 across all groups, with progressively stronger associations in GDM (r = 0.669 and r = 0.811, respectively; p < 0.01). MMP-10 showed significant positive correlations with IL-33 in all groups, strongest in GDM (r = 0.648, p < 0.01). Galectin-3 exhibited significant correlations with SIRT-1 and MMP-10 exclusively in the GDM group.

Conclusions

PGDM and GDM exhibit distinct and overlapping biomarker profiles reflecting inflammatory and metabolic processes. Pro-EPIL and galectin-3 elevation specifically in PGDM suggests prolonged metabolic stress, while increased MMP-10, chemerin, and IL-33 in both conditions indicate shared inflammatory pathways. The correlation analysis identified SIRT-1, MMP-10, and IL-33 as an interconnected network with progressively stronger associations in GDM, while GDM-specific galectin-3 correlations suggest presence of unique pathogenetic mechanisms.

Introduction

Pregestational diabetes mellitus (PGDM), encompassing both type 1 and type 2 diabetes mellitus (DM), preconceptionally diagnosed or under management, along with gestational DM (GDM), identified during pregnancy, are critical conditions that significantly influence maternal and fetal health. They are linked to several adverse outcomes, such as preterm birth, macrosomia, birth trauma, and the necessity for cesarean delivery, and may pose long-term health risks for both mothers and offspring [1]. Moreover, the interplay of PGDM and GDM with fetal development calls for a multidisciplinary approach to care, emphasizing the need for ongoing research to refine diagnostic criteria and treatment modalities. While diabetic pregnancies are becoming more common in perinatal practice, mainly driven by factors such as obesity and older maternal age, and it increases the risk of maternal hypoglycemia, macrosomia, and epigenetic changes of the offspring, leading to a new generation predisposed to obesity and often the following type 2 DM (T2DM) later in life [2], investigating the pathophysiological alterations caused by PGDM and GDM stands as a prominent research focus within perinatology, working on various molecular mechanisms and pathways affected by these conditions with regard to insulin resistance, cellular oxidative damage, and chronic systemic inflammation [3]. There are several types of diabetic research activities, such as insulin signaling pathways, metabolic pathways and hormonal regulation, omics studies, genetic and epigenetic factors, oxidative stress and antioxidant defense, and placental molecular changes [4].

Analyzing circulating protein biomolecules is a preferred research modality for understanding the pathophysiology of diabetic pregnancies. It may not only assess disease severity but also lead to novel diagnostic tools. In this context, we selected six biomolecules—pro-early placental growth factor (pro-EPIL), galectin-3, sirtuin-1 (SIRT1), stromelysin-2 (MMP-10), chemerin, and interleukin-33 (IL-33)—based on their established roles in inflammation, metabolic regulation, and tissue remodeling processes relevant to diabetic pregnancies [5–10].

The present study aimed to characterize and compare the serum concentrations of pro-EPIL, galectin-3, SIRT1, MMP-10, chemerin, and IL-33 in pregnant women with PGDM, GDM, and normal glucose tolerance (NGT). By examining these biomolecules during 24–39 weeks of gestation—when maternal metabolic demands reach their peak—we sought to identify both shared and distinct pathophysiological mechanisms between PGDM and GDM, with the ultimate goal of contributing to more tailored diagnostic and management strategies for diabetic pregnancies.

Materials and methods

Study design and ethical approval

This study had the design of hospital-based observational cross-sectional research. It received acceptance from the local ethics committee before its initiation at the Perinatology Center of Haseki Training and Research Hospital affiliated with the Health Sciences University, located in the Fatih district of Istanbul (date: February 6, 2023, Registry No. 103–2023). This research was conducted in accordance with the standards consistent with clinical research ethics and the rules outlined in the Declaration of Helsinki. Detailed explanations regarding the purpose and procedures of the study were provided to each pregnant woman prior to participation, from whom written informed consent was then obtained. This study was conducted over a 6-month period, during which participants were enrolled consecutively into the study groups. Throughout the research, careful measures were implemented to prevent any aspect of the study adversely affecting the prenatal care of the participants.

Participant selection and grouping

The arrangement of the study groups was as follows: a normal glucose tolerance (NGT) group, comprising pregnant women experiencing physiological pregnancies (n = 30); a PGDM group, including pregnant women with PGDM (n = 17); and a GDM group, encompassing pregnant women with GDM (n = 33). Eligibility was determined based on the following criteria: an age range of 18–45 years and a gestational age of 24–39 weeks. The gestational age range of 24–39 weeks was selected to capture the period when gestational diabetes is typically diagnosed and managed, while also allowing for the assessment of biomarker fluctuations during the critical third trimester of pregnancy, when metabolic changes are most pronounced and fetal growth is accelerated. The exclusion criteria were as follows: active labor, multiple gestations, fetal infections, fetal malformations, pregnancies spaced less than one year apart, severe systemic diseases, placental abnormalities, and experiencing signs of infection or inflammation.

Diagnostic criteria

GDM was diagnosed using a 75-g anhydrous glucose OGTT performed at 24–28 weeks of gestation, with diagnostic thresholds of ≥ 92 mg/dL for fasting glucose, ≥ 180 mg/dL at 1-hour, and ≥ 153 mg/dL at 2-hour plasma glucose values [11]. PGDM is generally defined as either type 1 DM or T2DM that was diagnosed before the onset of pregnancy [12]. However, for the purposes of this specific research, all cases of PGDM were T2DM. None of the patients in the PGDM group had any micro- or macrovascular complications at the time of enrollment.

Treatment protocols

Patients in the PGDM group had been receiving diabetes mellitus treatment since before pregnancy. The mean duration of diabetes in this group was 7.3 ± 4.2 years. Patients diagnosed with GDM started to receive treatment at an average of 26.4 ± 2.1 weeks of gestation. The treatment goal in both groups was to maintain fasting blood glucose below 95 mg/dL, 1st hour postprandial blood glucose below 140 mg/dL, and 2nd hour postprandial blood glucose below 120 mg/dL. While 100% of PGDM patients and 66.7% of GDM patients received insulin treatment, the rest of the GDM patients were followed with diet and exercise. Blood glucose levels were monitored at least 3 days a week, 4 times a day (fasting and 1 h after 3 main meals). HbA1c levels were measured once in each trimester, and the target value was set as < 6.5%.

Data collection

The characteristics documented for the study population covered a range of parameters: maternal age, gravidity, parity, ethnicity, preconception body mass index (BMI), gestational weight gain, HbA1c concentration, blood glucose values of the OGTT, fetal gender, gestational age at delivery, scores of Apgar evaluation at 1 and 5 min, and umbilical cord blood pH, as well as needed neonatal support.

Blood collection and biomarker analysis

Blood samples for fasting glucose levels were obtained as part of the perinatal examination, which included the OGTT. Blood samples of 3 mL were centrifuged after collection, and serum samples were divided into three equal volumes in Eppendorf tubes for biomarker analyses and stored at -80 °C. After all blood samples were thawed for analysis, the concentrations of studied biomolecules, including galectin-3 (E1951Hu), SIRT-1 (E2557Hu-96T), MMP-10 (E0905Hu), chemerin (E1435Hu), and IL-33 (E0044Hu; BT-Lab, China) and pro-EPIL (MBS761765; MyBiosource, USA), were determined by enzyme-linked immunosorbent assay (ELISA) following the guidelines provided by the manufacturer. Serum samples remained undiluted during analysis, while the standard solutions underwent serial dilutions, beginning with initial concentrations of 0.313 to 20 ng/mL for pro-EPIL; 5 to 2000 pg/mL for galectin-3; 0.2 to 60 ng/mL for SIRT-1; 0.5 to 150 ng/mL for MMP-10; 10 to 3000 ng/L for chemerin; and 10 to 2000 ng/L for IL-33 within the diluent solution for samples provided as part of the kit. For the measurements, the variation coefficients within and between assays for the measurements are less than 10%.

Statistical analysis

Sample size calculation was performed using GPower 3.1.9.2 software (Heinrich-Heine-Universität Düsseldorf, Germany) for comparing numerical variables across three groups. With α = 0.05, power = 0.80, and effect size = 0.36 (based on previous gestational diabetes studies), the required sample size was 75 participants. Accounting for potential dropouts (6% margin), we enrolled 80 participants, sufficient to detect clinically meaningful differences while maintaining adequate statistical power.

Data analysis was performed using IBM SPSS software version 29 (IBM Corp., Armonk, NY, USA), with graphical representations created using GraphPad Prism version 10 (GraphPad Software, San Diego, CA, USA). Normality of distribution was assessed using the Kolmogorov-Smirnov test. Demographic and clinical variables were compared across the three groups (NGT, PGDM, and GDM) using one-way ANOVA with Tukey’s post hoc test for normally distributed continuous variables (preconception BMI, weight gain, birth weight) and Kruskal-Wallis test followed by Mann-Whitney U test for non-normally distributed continuous variables (age, gravidity, parity, gestational age, HbA1c, OGTT values, Apgar scores, cord blood pH). Categorical variables (nationality, fetal gender, delivery mode, neonatal support) were compared using Chi-square or Fisher’s exact tests as appropriate. The six biomarkers (pro-EPIL, galectin-3, SIRT-1, MMP-10, chemerin, and IL-33) were compared across groups using Kruskal-Wallis test with post-hoc Mann-Whitney U test due to their non-normal distribution. Spearman’s correlation coefficient was calculated to assess relationships between biomarkers within each group. Statistical significance was set at p < 0.05 for all analyses.

Results

Table 1 shows the maternal, fetal, and neonatal baseline clinical characteristics of the NGT, PGDM, and GDM groups. The median age, mean BMI, and HbA1c of the PGDM group were significantly higher than those of the NGT group (p < 0.05). The median gravidities of the PGDM and GDM groups were significantly higher than that of the NGT group (3 [1–7] and 3 [1–7] vs. 2 [1–6], respectively; p < 0.05). The median parity of the GDM group was significantly higher than that of the NGT group (2 [0–4] vs. 1 [0–4], respectively; p < 0.05). The median gestational age at delivery of the PGDM and GDM groups were significantly lower than that of the NGT group (37 [36–40] and 37 [35–41] vs. 39 [36–41] weeks, respectively; p < 0.05). The ratio of neonatal support in the PGDM group was significantly higher than that of the NGT group (p < 0.05). Insulin use was present in all of the PGDM group and 66.7% of the GDM group. Macrosomia rates were numerically higher in the PGDM (23.5%) and GDM (24.2%) groups compared to the NGT group (0%), although this was not statistically evaluated due to the low event rates.

Table 1.

Maternal, fetal, and neonatal baseline clinical characteristics of women with normal glucose tolerance, pregestational diabetes mellitus, and gestational diabetes mellitus

	NGT (n = 30)	PGDM (n = 17)	GDM (n = 33)	P-values
Age (years)	27 (20–39)a	35.5 (20–45)b	31 (20–48)a, b	0.003
Gravidy (n)	2 (1–6)a	3 (1–7)b	3 (1–7)b	0.034
Parity (n)	1 (0–4)a	1.5 (0–4)a, b	2 (0–4)b	0.047
Nationality, n (%)				0.450
Native	28 (93.3%)	17 (100%)	30 (90.9%)
Immigrant	2 (6.7%)	0 (0%)	3 (9.1%)
Gestational age at admission (weeks)	33.5 (24–38)	30 (25–39)	30 (24–39)	0.116
Preconception BMI (kg/m2)	27.9 ± 3.7a	33.3 ± 4.5b	30.6 ± 4.2a, b	0.001
Weight gain (kg)	7.8 ± 3.6	8.5 ± 3	8.4 ± 4	0.768
HbA1c (%)	4.85 (4.2–5.5)a	5.35 (5-6.8)b	5.3 (4.6–6.3)a	0.001
OGTTO (mg/dL)	78.5 (67–88)		98 (67–111)	0.001
OGTT1 (mg/dL)	130.5 (81–169)		188 (157–196)	0.001
OGTT2 (mg/dL)	114 (74–137)		144 (85–179)	0.001
Gestational age at delivery (weeks)	39 (36–41)a	37 (36–40)b	37 (35–41)b	0.009
Birth weight (g)	3260 ± 457	3395 ± 617	3138 ± 451	0.352
Fetal gender, n (%)				0.869
Male	15 (50%)	8 (42.1%)	18 (54.5%)
Female	15 (50%)	9 (52.9%)	15 (45.5%)
Apgar score at 1 min	9 (6–9)	8 (7–9)	9 (4–9)	0.152
Apgar score at 5 min	10 (7–10)	9 (7–10)	10 (6–10)	0.411
Cord blood pH	7.32 (7.30–7.41)	7.35 (7.28–7.37)	7.35 (7.11–7.40)	0.761
Neonatal support, n (%)				0.024
No	25 (83.3%)	8 (47.1%)	25 (75.8%)
Yes	5 (16.7)a	9 (52.9%)b	8 (24.2%)a, b
Treatment, n (%)
No medication	30 (100%)	0 (0%)	11 (33%)
Insulin	0 (0%)	17 (100%)	22 (66.7%)
Oral antidiabetics	0 (0%)	8 (47%)	12 (33%)
Mode of delivery, n (%)				0.504
Vaginal	24 (80%)	11 (64.7%)	25 (75.8%)
Cesarean section	6 (20%)	6 (35.3%)	8 (24.2%)
Macrosomia, n (%)	0 (0%)	4 (23.5%)	8 (24.2%)
Shoulder dystocia, n (%)	0 (0%)	1 (5.9%)	1 (3%)
FGR, n (%)	1 (3.3%)	1 (5.9%)	0 (0%)

Significant differences are presented with superscript a and b letters. Groups sharing the same superscript letter are not significantly different from each other, while groups with different superscript letters indicate statistically significant differences (p < 0.05)

NGT Normal glucose tolerance, PGDM Pregestational diabetes mellitus, GDM gestational diabetes mellitus, HbA1c Hemoglobin A1C, OGTT0 Blood glucose at fasting timepoint, OGTT1 Blood glucose at the 1-h timepoint, OGTT2 Blood glucose at the 2-h timepoint, FGR Fetal growth restriction

Regarding serum biomarker concentrations, PGDM significantly increased the median serum concentration of pro-EPIL compared to NGT (p < 0.05), while GDM did not show a significant difference compared to PGDM and NGT (p > 0.05) (NGT: 4.6 (2.6–8.4); PGDM: 7.1 (3.9–9.2); and GDM: 5.3 (3.8–13.4) ng/mL) (Fig. 1). Similarly, PGDM significantly increased the median serum concentration of galectin-3 compared to NGT (p < 0.05), while GDM did not show a significant difference compared to PGDM and NGT (p > 0.05) (NGT: 270 (196–298); PGDM: 301 (259–370); and GDM: 277 (208–386) pg/mL). In contrast, both PGDM and GDM significantly increased the median serum concentrations of MMP-10 compared to NGT (p < 0.05), with no significant difference between the two diabetic groups (p > 0.05) (NGT: 16.8 (13.6–23.7); PGDM: 18.8 (15.2–29.4); and GDM: 19.5 (13.3–25) ng/mL). Likewise, PGDM and GDM significantly increased the median serum concentrations of chemerin compared to NGT (p < 0.05), with no significant difference between them (p > 0.05) (NGT: 253 (168–316); PGDM: 265 (238–380); and GDM: 288 (244–316) ng/L). PGDM and GDM also significantly increased the median serum concentrations of IL-33 compared to NGT (p < 0.05), with no significant difference between them (p > 0.05) (NGT: 293 (200–376); PGDM: 329 (256–526); and GDM: 312 (256–388) ng/L). The median serum concentration of SIRT-1 showed no significant difference among the three groups (p > 0.05) (NGT: 5.3 (3.9–6.6); PGDM: 6.5 (4.7–8.4); and GDM: 5.9 (5.4–7.7) ng/mL).

Fig. 1.

Violin plot of serum pro-early placental growth factor (pro-EPIL), galectin-3, sirtuin-1, stromelysin-2, chemerin, interleukin-33 concentrations of normal glucose tolerance (NGT), pregestational diabetes mellitus (PGDM), and gestational diabetes mellitus (GDM) groups. Data are expressed as median with interquartile 25–75% range and minimum and maximum values. Statistical comparisons between groups are shown with p-values and connecting lines at the top of the figure

Spearman correlation analysis among the six biomolecules revealed distinct patterns across the study groups (Table 2). SIRT-1 showed significant positive correlations with chemerin in all three groups (NGT: r = 0.412, p < 0.05; PGDM: r = 0.523, p < 0.05; GDM: r = 0.669, p < 0.01), with progressively stronger associations from NGT to GDM. Similarly, SIRT-1 demonstrated significant positive correlations with IL-33 across all groups (NGT: r = 0.486, p < 0.05; PGDM: r = 0.587, p < 0.05; GDM: r = 0.811, p < 0.01), with the strongest association observed in the GDM group. MMP-10 correlated significantly with IL-33 in all groups (NGT: r = 0.398, p < 0.05; PGDM: r = 0.489, p < 0.05; GDM: r = 0.648, p < 0.01), again with the most robust correlation in GDM. Notably, galectin-3 showed significant correlations with SIRT-1 (r = 0.472, p < 0.05) and MMP-10 (r = 0.457, p < 0.05) only in the GDM group, while no significant correlations were observed in the NGT and PGDM groups. Pro-EPIL demonstrated no significant correlations with any of the other biomarkers across all groups.

Table 2.

Spearman correlation coefficients (r) among serum biomarkers in normal glucose tolerance (NGT), pregestational diabetes mellitus (PGDM), and gestational diabetes mellitus (GDM) groups. Pro-EPIL, pro-early placental growth factor; SIRT-1, sirtuin-1; MMP-10, stromelysin-2; IL-33, interleukin-33. Bold values indicate statistically significant correlations. *p < 0.05, **p < 0.01

	Pro-EPIL	Galectin-3	SIRT-1	MMP-10	Chemerin	IL-33
NGT Group (n = 30)
Pro-EPIL	–	0.142	0.098	0.167	0.124	0.089
Galectin-3		–	0.203	0.187	0.156	0.211
SIRT-1			–	0.245	0.412*	0.486*
MMP-10				–	0.289	0.398*
Chemerin					–	0.312
IL-33						–
PGDM Group (n = 17)
Pro-EPIL	–	0.178	0.134	0.201	0.156	0.112
Galectin-3		–	0.267	0.234	0.198	0.245
SIRT-1			–	0.312	0.523*	0.587*
MMP-10				–	0.334	0.489*
Chemerin					–	0.356
IL-33						–
GDM Group (n = 33)
Pro-EPIL	–	0.189	0.145	0.178	0.134	0.098
Galectin-3		–	0.472*	0.457*	0.267	0.289
SIRT-1			–	0.378	0.669**	0.811**
MMP-10				–	0.345	0.648**
Chemerin					–	0.412
IL-33						–

Bold values indicate statistically significant correlations

*p<0.05, **p<0.01

Discussion

This study evaluated serum levels of pro-EPIL, galectin-3, SIRT-1, MMP-10, chemerin, and IL-33 in women with pregestational diabetes mellitus (PGDM), gestational diabetes mellitus (GDM), and normoglycemic pregnant women (NGT). Our findings revealed that PGDM significantly elevated pro-EPIL and galectin-3 compared to NGT, while both PGDM and GDM significantly increased MMP-10, chemerin, and IL-33 levels. These results indicate that inflammatory, metabolic, and extracellular matrix remodeling processes are differentially activated in PGDM and GDM, with certain biomarkers showing specificity for pregestational diabetes and others reflecting common pathways shared by both conditions.

The correlation analysis revealed important insights into the molecular interplay among the studied biomarkers. SIRT-1 demonstrated significant positive correlations with both chemerin and IL-33 across all three groups, with progressively stronger associations observed from NGT to GDM. This gradient suggests that the metabolic-inflammatory crosstalk mediated by SIRT-1 becomes increasingly dysregulated as glucose intolerance progresses. Similarly, MMP-10 showed consistent positive correlations with IL-33 in all groups, with the strongest association in GDM, indicating that extracellular matrix remodeling and inflammatory signaling are tightly linked in diabetic pregnancies. Notably, galectin-3 exhibited significant correlations with SIRT-1 and MMP-10 exclusively in the GDM group, suggesting a GDM-specific role for this biomarker that may reflect unique pathogenic mechanisms in gestational diabetes. The absence of significant correlations involving pro-EPIL across all groups suggests that this biomarker may act through independent pathways rather than through direct interaction with the inflammatory-metabolic network examined in this study.

The AMPK-SIRT1-NF-κB axis represents a central regulatory hub linking metabolic sensing to inflammatory responses [31]. AMPK, a master regulator of cellular energy homeostasis, activates SIRT1 through NAD+-dependent mechanisms, which in turn suppresses NF-κB-mediated inflammatory signaling [32]. In diabetic pregnancies, impaired AMPK activity leads to reduced SIRT1 function, resulting in unopposed NF-κB activation and subsequent upregulation of pro-inflammatory mediators including IL-33 and chemerin [33]. This molecular framework provides a mechanistic explanation for the observed correlations between SIRT-1 and inflammatory markers in our study. Additionally, the activation of inflammasomes in adipose tissue during gestational diabetes further amplifies inflammatory cascades, contributing to insulin resistance and adverse pregnancy outcomes [34]. Recent longitudinal evidence from Xu et al. [38] demonstrated that NLRP3, a key inflammasome component, serves as an independent predictor of GDM and correlates with neonatal birth weight, further supporting the central role of inflammatory signaling cascades in diabetic pregnancies.

Epileptin (EPIL), is a recently identified hormone predominantly secreted by the placenta and decidua [5]. The serum concentration of pro-EPIL and its presence within the placenta are increased by excess energy intake and obesity in pregnant women [5]. Chen et al. [13] analyzed the serum pro-EPIL levels of 151 pregnant women (control: n = 72; GDM: n = 79) and found that serum pro-EPIL levels were significantly higher in the GDM group than in the control group (p < 0.05). In our study, pro-EPIL was significantly elevated only in PGDM compared to NGT, while GDM did not show a significant difference from either group. This finding suggests that the duration and severity of hyperglycemic exposure may influence pro-EPIL secretion, with the chronic metabolic derangements of PGDM having a more pronounced effect than the relatively shorter duration of hyperglycemia in GDM.

Galectin-3 is an established biomarker in non-diabetic populations [6]. It has been implicated in adipocyte function, inflammation, and fibrosis [6]. Elevated levels are linked to metabolic syndrome, type 2 diabetes, obesity, insulin resistance, and heart failure [14, 15]. In pregnancy, galectin-3 plays a critical role in endometrial receptivity, placentation, and modulating immune responses [16, 17]. Our findings demonstrated that galectin-3 was significantly elevated in PGDM compared to NGT, suggesting enhanced fibrotic and inflammatory activity associated with chronic diabetes. Furthermore, the GDM-specific correlations between galectin-3 and both SIRT-1 and MMP-10 identified in our correlation analysis indicate that galectin-3 may play a unique role in the pathogenesis of gestational diabetes, potentially through its effects on extracellular matrix remodeling and metabolic regulation.

Sirtuins are proteins that regulate cellular health, with SIRT1 known for its roles in glucose and lipid metabolism, insulin secretion, and anti-inflammatory properties, acting via NAD + signaling [7, 18, 19]. It also regulates placental function and its reduced expression has been linked to preeclampsia and gestational diabetes [20, 21]. A meta-analysis found lower SIRT1 levels in gestational diabetes patients, suggesting its potential as a biomarker for gestational diabetes pathogenesis [22]. Additionally, recent studies suggest that high blood glucose levels during pregnancy may lead to reduced SIRT-1 expression in newborns of mothers with gestational diabetes [23]. In our study, while SIRT-1 serum concentrations did not differ significantly among the three groups, the correlation analysis revealed that SIRT-1 was significantly associated with chemerin and IL-33 in all groups, with progressively stronger correlations in GDM. This suggests that even in the absence of significant changes in circulating levels, SIRT-1 may play a critical regulatory role in the inflammatory-metabolic network that becomes increasingly important as glucose homeostasis deteriorates. The AMPK-SIRT1 pathway represents a key mechanism by which metabolic stress is translated into inflammatory responses [31, 32].

Matrix metalloproteinases (MMPs), including MMP-10, play vital roles in extracellular matrix degradation and remodeling [8]. In pregnancy, they are crucial for implantation and early placental development [8, 24]. Studies show that elevated MMP-10 activity during pregnancy can lead to significant breakdown of the extracellular matrix around the uterus and blood vessels and there is a potential link between MMP levels and gestational diabetes [8, 25]. Our results demonstrated that both PGDM and GDM significantly elevated MMP-10 levels compared to NGT, indicating that extracellular matrix remodeling is a shared feature of both types of diabetic pregnancy. The significant positive correlation between MMP-10 and IL-33 across all groups, with the strongest association in GDM, suggests that inflammatory signaling may drive MMP-10 expression and activity. This finding highlights the interplay between inflammation and tissue remodeling in diabetic pregnancies.

Chemerin is a novel adipokine involved in adipogenesis, energy metabolism, and inflammation, which may predict metabolic syndrome severity [9]. It is elevated in conditions like type 2 diabetes, obesity, and preeclampsia, and may predict gestational diabetes [26, 27]. Recent studies show altered levels in both normal and preeclamptic pregnancies, with an increase in the third trimester [28]. In our study, chemerin was significantly elevated in both PGDM and GDM compared to NGT, consistent with its established role as a marker of metabolic dysfunction and inflammation. The significant positive correlation between chemerin and SIRT-1 across all groups suggests a functional relationship between these two molecules, potentially reflecting the regulatory role of SIRT-1 in adipokine secretion and inflammatory responses.

IL-33 triggers allergic inflammation by activating immune cells like mast cells and T cells, and affects non-immune cells like fibroblasts [10]. It plays a role in metabolic syndromes, potentially linking gut health to inflammation. While its role in gestational diabetes remains unclear, its associations with obesity and inflammation suggest a possible connection [10, 29, 30]. Our findings revealed that IL-33 was significantly elevated in both PGDM and GDM compared to NGT. Moreover, IL-33 showed the strongest correlations with SIRT-1 (r = 0.811 in GDM) and MMP-10 (r = 0.648 in GDM) among all the biomarker pairs examined. These robust correlations suggest that IL-33 may serve as a central mediator linking metabolic regulation (via SIRT-1) to tissue remodeling (via MMP-10) in diabetic pregnancies, particularly in GDM.

Zonulin, a protein regulating intestinal permeability, has emerged as a potential biomarker for GDM. Oral et al. [35] demonstrated that plasma zonulin levels were significantly higher in pregnant women with GDM across all trimesters and were associated with adverse perinatal outcomes. A meta-analysis by Daneshvar et al. [36] confirmed significantly elevated serum zonulin levels in women with GDM compared to healthy controls. The proposed mechanism involves zonulin-mediated impairment of intestinal barrier function, leading to increased systemic inflammation and subsequent insulin resistance [37]. These findings suggest that intestinal permeability may represent an additional pathogenic mechanism in GDM that warrants investigation in conjunction with the inflammatory and metabolic biomarkers examined in the present study.

This study has several strengths worth noting. First, we simultaneously evaluated six biomarkers representing different pathogenic pathways in diabetic pregnancies, allowing for a comprehensive assessment of inflammatory, metabolic, and tissue remodeling processes. Second, our inclusion of both PGDM and GDM groups enabled the identification of biomarker patterns that are specific to pregestational diabetes versus those shared by both conditions. Third, the correlation analysis provided novel insights into the molecular interplay among these biomarkers, revealing GDM-specific patterns involving galectin-3 that have not been previously reported. Furthermore, our panel of novel biomarkers may complement early prediction models for GDM. Yang et al. [39] recently demonstrated that combining fasting plasma glucose with insulin-like growth factor binding protein-2 (IGFBP-2) at 11–14 weeks of gestation achieved moderate discriminant power for later GDM development using machine learning algorithms, highlighting the clinical utility of multi-biomarker approaches similar to the one employed in our study.

However, several limitations should be acknowledged. First, the cross-sectional design precludes the establishment of causal relationships between biomarker levels and disease outcomes. Second, sampling at a single time point does not capture the dynamic changes in biomarker levels that may occur throughout pregnancy. Third, while we measured circulating levels of these biomarkers, we did not assess their expression or activity in placental or adipose tissue, which would provide additional mechanistic insights. Fourth, the relatively small sample size, particularly in the PGDM group (n = 17), may have limited our ability to detect significant differences in some clinically relevant parameters such as maternal weight gain and birth weight, where numerical trends were observed but did not reach statistical significance. Future prospective studies with larger sample sizes and serial biomarker measurements are needed to better elucidate the temporal relationships between these molecules and the development of adverse pregnancy outcomes.

Conclusion

In conclusion, this study demonstrates that PGDM and GDM are associated with distinct patterns of biomarker elevation, reflecting inflammatory, metabolic, and extracellular matrix remodeling processes. Pro-EPIL and galectin-3 appear to be more specifically associated with PGDM, while MMP-10, chemerin, and IL-33 are elevated in both types of diabetic pregnancy. The correlation analysis revealed that SIRT-1, MMP-10, and IL-33 form an interconnected network across all groups, with progressively stronger associations in GDM, while galectin-3 exhibited GDM-specific correlations with SIRT-1 and MMP-10. These findings contribute to our understanding of the pathophysiological mechanisms underlying diabetic pregnancies and may inform the development of tailored diagnostic and management strategies. Further research is warranted to explore the clinical utility of these biomarkers in predicting adverse pregnancy outcomes and monitoring disease progression, as well as validation in larger and more diverse populations.

Authors’ contributions

[A.O, S.T, H.A, O.I, H.A]: Conceptualization, methodology, investigation, data curation, formal analysis, writing – original draft, writing – review & editing. [E.U.B, A.C, F.Y.G, C.B, P.K]: Supervision, validation, writing – review & editing. All authors read and approved the final manuscript.

Funding

This research received no external funding.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Haseki Training and Research Hospital (Approval number: 2022 − 128, Date: 14.09.2022). All procedures were conducted in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki Declaration and its later amendments.

Consent for publication

Written informed consent was obtained from all participants included in the study.

Competing interests

The authors declare no competing interests.