Correlation between serum advanced glycation end-products and their receptor-mediated oxidative stress and perinatal outcomes in gestational diabetes mellitus

Abstract

BACKGROUND

Gestational diabetes mellitus (GDM) is one of the most prevalent metabolic disorders of pregnancy. Advanced glycation end-products (AGEs) are a complex and highly heterogeneous group of compounds formed from amino acids and reducing sugars. High-AGE diet exposure during pregnancy may cause adverse effects.

AIM

To investigate the expression levels of AGE and AGE receptor (RAGE) in the serum and placenta of pregnant women with GDM and to assess the association of their mediated oxidative stress response with perinatal outcomes.

METHODS

This study retrospectively analyzed the clinical data of 126 pregnant women with GDM who gave birth in the Obstetrics Department of Obstetrics and Gynecology Hospital of Fudan University from January 2023 to January 2024. A total of 85 pregnant women of similar age without GDM during the same period were selected as the control group. Fasting blood glucose, glycated hemoglobin, AGEs, soluble RAGE (sRAGE), and oxidative stress were compared in both groups. Postpartum placental tissue was collected to identify RAGE protein expression. Participants with GDM were categorized based on perinatal outcomes into normal (n = 89) and adverse perinatal outcome groups (n = 37), and differences in serum AGE–RAGE levels and oxidative stress were analyzed. The influencing factors of adverse perinatal outcomes were analyzed using logistic regression.

RESULTS

The GDM group demonstrated notably higher serum AGE (t = 8.955) and malondialdehyde (MDA) levels (t = 14.14) and lower sRAGE (t = 16.37) and superoxide dismutase (SOD) levels (t = 18.50) than the control group at 24-28 weeks of gestation and before delivery (P < 0.0001). Serum AGE levels were positively correlated with MDA and negatively related to SOD at 24-28 weeks of pregnancy (SOD: r = 0.393, MDA: r = 0.424, P < 0.0001) and before delivery (SOD: r = 0.443, MDA: r = 0.492, P < 0.0001), whereas AGE was inversely associated with sRAGE in the GDM group (r = -0.495, P < 0.0001). Serum AGE levels were significantly higher (t = 9.225, P < 0.0001) and the sRAGE level (r = 3.563, P < 0.0001) was significantly lower in participants with adverse perinatal outcomes than those with normal perinatal outcomes in the GDM group. Logistic regression analysis revealed AGE level as a risk factor (OR = 1.056, P < 0.0001) and sRAGE level (OR = 0.949, P < 0.0001) as a protective factor for adverse perinatal outcomes in GDM.

CONCLUSION

High serum AGE level is a risk factor for adverse perinatal outcomes in GDM, whereas high sRAGE levels are protective. AGEs and RAGE may be associated with oxidative stress in pregnant women with GDM.

Key Words: Gestational diabetes mellitus; Advanced glycation end-products; Hemoglobin; Fasting blood glucose; Perinatal outcomes

Core Tip: Advanced glycation end-products (AGEs) are a complex and highly heterogeneous group of compounds formed from amino acids and reducing sugars, primarily produced during thermal and metabolic processing of food. To date, a few studies have demonstrated the role of AGEs in gestational diabetes mellitus (GDM) pathophysiology. Therefore, in this study, we aimed to understand the association of AGEs with oxidative stress in GDM. We found that serum AGE levels were significantly correlated with patient oxidative stress status at 24-28 weeks of pregnancy and before delivery. Meanwhile, AGE level was a risk factor, and its receptor soluble receptor AGE level was a protective factor for adverse perinatal outcomes in GDM.

INTRODUCTION

Gestational diabetes mellitus (GDM) is a disorder resulting from abnormal glucose metabolism of varying degrees that first occurs during pregnancy and is classified by the World Health Organization as an independent type of diabetes mellitus (DM)[1]. GDM is one of the most prevalent endocrine and metabolic disorders of pregnancy, with an incidence rate of 5%-25% in all pregnancies, depending upon the population and race studied[2]. The International Diabetes Federation reports that approximately 16.7% of pregnant women globally suffer from gestational hyperglycemia, with an estimated 80.3% of those diagnosed with GDM[3].

Age, obesity, parity, and genetic factors are considered independent risk factors for GDM[4-6]. GDM exhibits a wide range of short- and long-term adverse health effects on both mothers and offspring[7]. In the short term, GDM increases pregnancy and peripartum complications, and women with GDM have a higher risk of developing pregnancy-induced hypertension and preeclampsia[8,9]. Further, GDM is considered an independent risk factor for large-for-gestational-age babies and macrosomia[10,11]. Therefore, effective interventions in the first trimester and even before pregnancy is particularly important to prevent GDM.

Among the many risk factors for GDM, dietary factors are an important intervention target and have gained attention from international and domestic health guidelines. Advanced glycation end-products (AGEs) are a complex and highly heterogeneous group of compounds formed from amino acids and reducing sugars, primarily produced during the thermal and metabolic processing of food[12]. AGEs originate from a wide range of sources. Besides in vitro exposure, endogenous AGEs are also produced by the human body with aging; however, exogenous AGEs that are ingested through diet are the primary source[13]. The dietary intake of AGEs is increasing with the popularity of Western dietary patterns and the elevated intake of ultra-processed foods[14,15]. High-AGE diet exposure during pregnancy may cause adverse effects, but currently, the evidence is limited. AGEs function by binding to the AGE receptors (RAGEs)[16]. The mechanism of action of AGEs in the occurrence and development of DM and the associated complications have been extensively studied, and the underlying biological mechanisms of AGEs and GDM may be comparable with those of DM[17]. AGEs are products formed when reducing sugars, such as glucose and fructose in the body, resulting in nonenzymatic glycosylation reactions with the free amino groups of proteins and polypeptides under the hyperglycemic state[18]. RAGEs are predominantly present on the surfaces of macrophages, monocytes, and other cells[19]. Current studies have revealed that AGE and RAGE binding activates a series of intracellular signaling pathways, which increases the production of various cytokines and oxidative stress, thereby regulating the expression of inflammation-related genes[20,21]. AGEs have activated certain signaling cascades that damage endothelial cells and play a crucial role in developing diabetic vascular complications[22]. Further, RAGEs transform short-term proinflammatory reactions into long-term cellular dysfunction and chronic diseases[23]. Furthermore, AGEs induce inflammation, as evidenced by proinflammatory cytokine and inflammatory molecule production, such as tumor necrosis factor (TNF)-α and interleukin (IL)-6[24]. Soluble RAGE (sRAGE), as an alternative splice of RAGE, functions as a “decoy”, which prevents the deleterious effects of the AGE–RAGE axis. Serum sRAGE levels may reflect RAGE expressions at the tissue level and serve as a biomarker for reflecting the function of RAGE[25].

Fetal exposure to maternal diabetes is associated with a high risk of abnormal glucose homeostasis in the offspring[26]. Abnormal endothelial function is a prevalent feature of GDM[27]. However, the cellular mechanisms underlying the altered endothelial function of the umbilical cord in patients with GDM remain unclear. To date, a few studies have demonstrated the role of AGEs in the pathophysiology of GDM. Therefore, in this study, we aim to understand the association of AGEs with oxidative stress in GDM. Further, we discuss the correlation of blood AGE concentration in GDM pregnancies with blood glucose indexes and perinatal outcomes and its value in predicting adverse pregnancy outcomes by studying the mechanisms underlying endothelial inflammation in GDM.

MATERIALS AND METHODS

Research participants

This study retrospectively analyzed the clinical data of pregnant women who visited and gave birth in the Obstetrics Department of Obstetrics and Gynecology Hospital of Fudan University and were diagnosed with GDM at 24-28 weeks of pregnancy from January 2023 to January 2024. The inclusion criteria were: (1) Meeting the diagnostic criteria for GDM following the Recommendations on the Classification and Diagnosis of Hyperglycemia in Pregnancy; (2) Age > 20 years; (3) Singleton pregnancy; (4) Absence of autoimmune diseases, hereditary diseases, or malignancies; (5) Absence of heart, liver, kidney, or other vital organ dysfunction; and (6) Complete clinical data. All pregnant women were tested for oral glucose tolerance at 24-28 gestational weeks. GDM was diagnosed if fasting blood glucose (FBG) was ≥ 5.1 mmol/L, 1-hour postprandial blood glucose was ≥ 10.0 mmol/L, and 2-hour postprandial blood glucose was ≥ 8.5 mmol/L, or any of them significantly exceeded the upper limit of the normal value. Exclusion criteria were: (1) Preexisting DM; (2) Pregnancy-induced hypertension; (3) Congenital heart disease; (4) Twin pregnancies; (5) Pregnant women with serum creatinine levels > 150 μmol/L and transaminase levels > 2.5 times the normal value; (6) Acute and chronic infectious diseases with a history of radiation exposure; (7) Infectious diseases; (8) Long-term medication history; (9) Phenomena such as placenta previa and placental abruption; (10) Pregnant women with acute inflammation, such as prenatal fever; and (11) Incomplete clinical data. This study included 126 pregnant women with GDM after screening. Another 85 pregnant women of comparable age, gestational weeks, and pre-pregnancy body mass index (BMI) without GDM were selected as the control group. Figure 1 illustrates the flow diagram of patient selection.

Figure 1 Flow diagram of the patient selection. GDM: Gestational diabetes mellitus; DM: Diabetes mellitus; BMI: Body mass index; AGE: Advanced glycation end-product; sRAGE: Soluble advanced glycation end-product receptor; RAGE: Advanced glycation end-product receptor.

Patient data collection

General information: Patients’ general information, including age, gestational age, parity, and pre-pregnancy BMI, was collected by reviewing electronic medical records.

Biochemical indicators: Routine blood indicator levels of the participants were collected based on the electronic medical record files of prenatal check-ups. Further, blood samples were collected from pregnant women at 24–28 gestational weeks and before delivery in the GDM group to measure FBG, glycosylated hemoglobin (HbA1c), AGEs, and sRAGE. Immediately after delivery, the placenta was collected and cryopreserved for RAGE protein expression analysis in tissues.

Detection methods

Serum AGE, sRAGE, and oxidative stress index determination: Enzyme-linked immunosorbent assays (ELISAs) were conducted to detect AGE and sRAGE levels in serum and placental tissue lysates of the two groups using ELISA kits (Wuhan Boster Biological Technology., Ltd. and Qingdao Fulin Biological Co., Ltd., respectively). All specimens were measured in the same batch, with an intra-batch coefficient of variation of < 5%. Superoxide dismutase (SOD) activity and malondialdehyde (MDA) levels were identified using the hydroxylamine method and thiobarbituric acid assay, respectively, with kits supplied by Nanjing Jiancheng Bioengineering Institute. The above measurements were performed based on the manufacturers’ instructions.

Determination of RAGE protein expression in placental tissue: Total protein was separated from placental biopsy tissue with a radioimmunoprecipitation assay buffer, and protein quantification was performed using bicinchoninic acid assay. Electrophoresis separation of 30 µg protein in 4%-15% polyacrylamide gel, conventional membrane transfer, sealing, and addition of primary antibodies RAGE (1:1000, Cell Signaling Technology) and β-actin (1:1000, Sigma-Aldrich) for overnight incubation were performed. A secondary antibody (1:2000, Santa Cruz Biotechnology) was added to incubate at room temperature, followed by development with an ECL kit and grayscale analysis of the bands using Image J software.

Newborns of pregnant women in the GDM group were admitted to the neonatal ward as high-risk infants, and their birth weight (consistent or inconsistent with the gestational age), the presence or absence of congenital developmental abnormalities, birth injuries, neonatal asphyxia, or neonatal hypoglycemia, neonatal ultrasound examination or neonatal electrocardiogram, and neonatal neurobehavioral score were assessed. In the GDM group, pregnant women with abnormal findings and maternal-fetal complications during pregnancy were categorized into the adverse perinatal outcome group, whereas those with normal neonatal assessments and examinations were classified into the normal perinatal outcome group. The serum levels of pregnant women were compared between the two groups.

Statistical analysis

Statistical Package for the Social Sciences (SPSS) version 25.0 was used for analysis of data. Measured data were expressed as mean ± SD. Between-group comparisons were conducted using independent sample t-tests. Count data were expressed as rates, and comparisons between groups were performed using χ² tests. Multivariate analysis was conducted using the binary logistic regression model, the percentage of missing values was calculated, and the presence of missing values was resolved using listwise deletion or substitution. Pearson correlations were used for correlation analysis. A statistically significant difference was considered at P < 0.05.

RESULTS

Participant characteristics

The two groups demonstrated little difference in age, pre-pregnancy BMI, parity, gestational age at delivery, neonatal weight, alcohol consumption history, smoking history, and family history of DM (P > 0.05). However, GDM pregnancies exhibited higher FPG and HbA1c levels than normal pregnancies at 24-28 weeks of pregnancy and before delivery (P < 0.05). We did not identify any significant differences in total cholesterol, triglyceride, high-density lipoprotein-cholesterol, and low-density lipoprotein-cholesterol (P > 0.05; Table 1).

Table 1 Inter-group comparison of general information.

Parameter	Control group (n = 85)	GDM group (n = 126)	χ2/t	P value
Age (years)	28.33 ± 3.17	28.58 ± 3.02	1.083	0.280
Pre-pregnancy BMI (kg/m2)	22.57 ± 2.19	22.44 ± 2.16	1.428	0.155
Parity (time)			0.148	0.929
0	38	53
1	42	65
≥ 2	5	8
Gestational age at delivery (weeks)	38.37 ± 0.72	38.23 ± 0.73	0.808	0.420
Newborn body mass (g)	3587 ± 438	3675 ± 323	1.614	0.108
Smoking history	7	12	0.103	0.748
Alcohol consumption history	9	18	0.622	0.430
Family history of diabetes mellitus	18	36	1.458	0.227
FPG
At 24-28 weeks of pregnancy	4.11 ± 0.44	5.51 ± 0.30	26.40	< 0.00011
Before delivery	4.08 ± 0.32	4.81 ± 0.44	13.49	< 0.00011
HbA1c (%)
At 24-28 weeks of pregnancy	4.93 ± 0.46	6.05 ± 0.29	9.280	< 0.00011
Before delivery	4.88 ± 0.39	5.56 ± 0.47	25.00	< 0.00011
TC (mmol/L)
At 24-28 weeks of pregnancy	5.46 ± 0.33	5.49 ± 0.36	0.143	0.886
Before delivery	5.89 ± 0.46	5.95 ± 0.37	1.127	0.261
TG (mmol/L)
At 24-28 weeks of pregnancy	2.88 ± 0.22	2.93 ± 0.19	1.862	0.064
Before delivery	3.17 ± 0.31	3.22 ± 0.34	0.228	0.822
HDL-C (mmol/L)
At 24-28 weeks of pregnancy	1.65 ± 0.24	1.69 ± 0.34	1.412	0.159
Before delivery	1.81 ± 0.21	1.87 ± 0.29	1.085	0.279
LDL-C (mmol/L)
At 24-28 weeks of pregnancy	3.55 ± 0.46	3.66 ± 0.50	1.211	0.227
Before delivery	3.76 ± 0.51	3.85 ± 0.48	1.393	0.165

¹Statistically significant difference.

GDM: Gestational diabetes mellitus; BMI: Body mass index; FPG: Fasting plasma glucose; HbA1c: Glycosylated hemoglobin; AGE: Advanced glycation end-product; sRAGE: Soluble advanced glycation end-product receptor; RAGE: Advanced glycation end-product receptor; TC: Total cholesterol; TG: Triglyceride; HDL-C: High density lipoprotein cholesterol; LDL-C: Low density lipoprotein cholesterol.

AGEs/sRAGE and oxidative stress markers

GDM pregnancies were associated with higher serum AGEs and MDA levels and lower sRAGE and SOD levels compared to normal pregnancies (P < 0.05; Table 2).

Table 2 Comparison of serum indexes.

Time Frame	Grouping	AGEs (ng/L)	sRAGE (ng/L)	SOD (U/L)	MDA (μmol/L)
At 24-28 weeks of pregnancy	Control group (n = 85)	385.26 ± 51.25	374.25 ± 54.26	92.25 ± 5.21	3.31 ± 0.64
	GDM group (n = 126)	461.25 ± 69.25	281.25 ± 39.58	83.65 ± 3.25	4.86 ± 0.81
	t value	8.955	16.37	18.50	14.14
	P value	< 0.00011	< 0.00011	< 0.00011	< 0.00011
Before delivery	Control group (n = 85)	401.25 ± 42.65	368.56 ± 51.24	90.36 ± 5.08	3.28 ± 0.61
	GDM group (n = 126)	475.26 ± 62.58	274.15 ± 33.59	80.34 ± 3.41	5.24 ± 0.84
	t value	8.546	18.40	16.99	18.69
	P value	< 0.00011	< 0.00011	< 0.00011	< 0.00011

¹Statistically significant difference.

AGE: Advanced glycation end-product; sRAGE: Soluble advanced glycation end-product receptor; SOD: Superoxide dismutase; MDA: Malondialdehyde; GDM: Gestational diabetes mellitus.

sRAGE expression in placental tissue

Western blotting revealed that sRAGE protein levels in the control and GDM groups were 0.30 ± 0.05 and 0.87 ± 0.06, respectively, with a statistically significant inter-group difference (t = 72.265, P < 0.0001; Figure 2).

Figure 2 Receptors for soluble advanced glycation end-product protein expression in placental tissues of pregnant women in the two groups. GDM: Gestational diabetes mellitus; sRAGE: Soluble advanced glycation end-product receptor; ^aP < 0.001.

Correlation between AGE levels and oxidative stress markers in the GDM group

Serum AGEs and sRAGE levels in pregnant women in the GDM group at 24-28 weeks of gestation and before delivery were negatively correlated, whereas the AGE level positively correlated with the MDA level and negatively correlated with the SOD level (Figure 3).

Figure 3 Correlation of serum indicators in pregnant women in the gestational diabetes mellitus group. A: Correlation between serum advanced glycation end-products (AGEs) and soluble AGE receptor levels at 24-28 weeks of gestation and before delivery; B: Correlation between serum AGEs and superoxide dismutase levels at 24-28 weeks of gestation and before delivery; C: Correlation between serum AGEs and malondialdehyde levels at 24-28 weeks of gestation and before delivery. sRAGE: Soluble advanced glycation end-product receptor; SOD: Superoxide dismutase; MDA: Malondialdehyde; AGEs: Advanced glycation end-products.

Clinical characteristics in patients with different perinatal outcomes

Among GDM pregnancies, 89 cases were in the normal perinatal outcome group and 37 cases were in the adverse perinatal outcome group. We found that mothers with GDM and adverse perinatal outcomes had higher serum AGE and MDA levels compared to those with normal perinatal outcomes at 24-28 weeks of pregnancy and before delivery (P < 0.05). Furthermore, we identified lower sRAGE and SOD levels in mothers with GDM having adverse perinatal outcomes than in those with normal perinatal outcomes (P < 0.05). We did not identify any notable inter-group differences in other general data or biochemical indexes (P > 0.05; Table 3).

Table 3 Comparison of general data of pregnant women with different perinatal outcomes.

Parameter	Normal perinatal outcome (n = 89)	Adverse perinatal outcome (n = 37)	χ2/t	P value
Age (years)	28.66 ± 3.26	29.24 ± 3.76	0.869	0.387
Pre-pregnancy BMI (kg/m2)	22.11 ± 2.14	22.21 ± 2.17	0.238	0.812
Parity (time)			3.965	0.138
0	33	20
1	51	14
≥ 2	5	3
Gestational age at delivery (weeks)	38.27 ± 0.75	38.31 ± 0.71	0.277	0.782
Newborn body mass (g)	3677 ± 335	3663 ± 347	0.211	0.833
Adverse outcome			-	-
Newborn birth weight does not correspond to the week of gestation	-	9
Congenital dysplasia	-	6
Birth injury	-	3
Neonatal asphyxia	-	6
Newborns hypoglycemia	-	4
Neonatal ultrasound or neonatal electrocardiogram abnormalities	-	6
Neonatal neurobehavioral score < 35 points	-	3
History of smoking	6	6	2.723	0.099
History of drinking	10	8	2.302	0.129
Family history of diabetes	22	14	2.204	0.138
FPG
At 24-28 weeks of pregnancy	5.51 ± 0.32	5.47 ± 0.33	0.633	0.528
Before delivery	4.77 ± 0.38	4.83 ± 0.48	0.745	0.457
HbA1c (%)
At 24-28 weeks of pregnancy	6.06 ± 0.28	6.04 ± 0.32	0.350	0.727
Before delivery	5.56 ± 0.52	5.50 ± 0.38	0.634	0.527
TC (mmol/L)
At 24-28 weeks of pregnancy	5.42 ± 0.36	5.51 ± 0.38	1.257	0.211
Before delivery	5.98 ± 0.39	5.89 ± 0.34	1.223	0.224
TG (mmol/L)
At 24-28 weeks of pregnancy	2.94 ± 0.21	2.92 ± 0.22	0.480	0.632
Before delivery	3.18 ± 0.37	3.19 ± 0.35	0.140	0.889
HDL-C (mmol/L)
At 24-28 weeks of pregnancy	1.70 ± 0.29	1.72 ± 0.35	0.331	0.741
Before delivery	1.86 ± 0.24	1.81 ± 0.35	0.924	0.357
LDL-C (mmol/L)
At 24-28 weeks of pregnancy	3.64 ± 0.45	3.60 ± 0.54	0.428	0.669
Before delivery	3.84 ± 0.48	3.89 ± 0.46	0.539	0.591
AGE (ng/L)
At 24-28 weeks of pregnancy	436.22 ± 56.73	525.93 ± 38.75	8.795	< 0.00011
Before delivery	445.23 ± 53.78	541.10 ± 51.51	9.225	< 0.00011
sRAGE (ng/L)
At 24-28 weeks of pregnancy	295.75 ± 36.52	255.70 ± 28.03	5.973	< 0.00011
Before delivery	275.72 ± 33.17	254.18 ± 24.51	3.563	< 0.0011
SOD (U/L)
At 24-28 weeks of pregnancy	83.94 ± 3.18	82.16 ± 3.06	2.898	0.0051
Before delivery	80.76 ± 2.81	79.10 ± 3.39	2.284	0.0051
MDA (μmol/L)
At 24-28 weeks of pregnancy	4.63 ± 0.64	5.12 ± 0.75	3.667	< 0.0011
Before delivery	5.09 ± 0.89	5.56 ± 0.70	2.877	0.0051

¹Statistically significant difference.

BMI: Body mass index; FPG: Fasting plasma glucose; HbA1c: Glycosylated hemoglobin; AGE: Advanced glycation end-product; sRAGE: Soluble advanced glycation end-product receptor; TC: Total cholesterol; TG: Triglyceride; HDL-C: High density lipoprotein cholesterol; LDL-C: Low density lipoprotein cholesterol; SOD: Superoxide dismutase; MDA: Malondialdehyde.

Risk factor analysis of adverse perinatal outcomes in the GDM group

We used a logistic regression model for multivariate analysis, and assignment was performed based on continuous variables. We utilized prenatal indicators with significant differences in Table 3 as covariates (AGEs, sRAGE, SOD, and MDA) and pregnancy outcomes (1 = adverse perinatal outcome, 0 = normal perinatal outcome) as dependent variables. After controlling for the mutual influence of each variable, we identified AGE levels in pregnant women with GDM as a risk factor for adverse perinatal outcomes [Odds ratio (OR) = 1.056, P < 0.0001], while sRAGE level was protective against adverse perinatal outcomes in GDM (OR = 0.949, P < 0.0001; Table 4).

Table 4 Logistic regression of risk factors for adverse perinatal outcomes.

Measurement	β	SE	Wald	P value	Odd ratios	Unadjusted 95%CI	Adjusted 95%CI
Constant	-11.302	9.945	1.292	0.256	0.000		-
AGEs	0.054	0.012	21.838	0.0001	1.056	8.562-169.764	1.032-1.080
sRAGE	-0.053	0.014	14.451	0.0001	0.949	0.012-0.150	0.923-0.975
SOD	-0.025	0.102	0.058	0.810	0.976	0.268-1.267	0.799-1.191
MDA	0.113	0.552	0.042	0.838	1.119	1.461-7.553	0.380-3.299

¹Statistically significant difference.

AGE: Advanced glycation end-product; sRAGE: Soluble advanced glycation end-product receptor; SOD: Superoxide dismutase; MDA: Malondialdehyde.

DISCUSSION

The accumulation of AGEs in the body results in an augmented metabolic burden (e.g., hyperglycemia, hyperlipidemia), as well as pathological conditions, including inflammation and oxidative stress[28]. AGE binding to RAGEs activates a series of signaling pathways, thereby increasing adverse reactions, including oxidative stress, inflammatory responses, cellular dysfunction, and apoptosis[29]. Recent research has revealed that AGEs and RAGE are closely associated with the onset and progression of metabolic diseases such as DM and diabetic complications[30]. Both AGEs and sRAGE can be detected in the serum of pregnant women throughout all pregnancy stages. However, serum AGE and sRAGE levels can vary in pathological pregnancies such as preterm birth, preeclampsia, and fetal growth restriction[31].

In this study, we measured serum AGE and sRAGE levels of pregnant women with GDM at 24-28 weeks of pregnancy and before delivery. The results indicated that compared with normal pregnancies, the serum AGE level was statistically increased and serum sRAGE level was decreased in GDM pregnancies; hence, AGE level negatively correlated with the sRAGE level. Concurrently, the GDM group demonstrated markedly higher serum MDA levels and lower SOD levels than the normal control group; thus, AGEs exhibited a positive correlation with MDA and an inverse association with SOD. A study that collected plasma samples from normal pregnancy and GDM women in India demonstrated comparable results, indicating that AGE levels were significantly higher in GDM women than in normal pregnant women[16]. Oxidative stress is present in patients with GDM, and this oxidative stress response is closely associated with postpartum vascular endothelial injury, hypertension, and DM development[32]. Meanwhile, AGE–RAGE interactions induced oxidative stress in vascular endothelial cells[33]. AGEs induced reactive oxygen species (ROS) and reactive nit rogen species (RNS) production by activating the NADP(H) oxidase pathway[34,35]. Therefore, AGEs and sRAGE may be associated with oxidative stress in GDM pregnancies. Current research indicates that by binding to receptors, AGEs activate signal transduction pathways, such as C-Jun N-terminal kinase and NADPH oxidase, and induce inflammatory factor production, such as IL-1 and TNF-α. This results in inflammatory reactions, interferes with insulin signaling, and reduces insulin sensitivity, thereby causing insulin resistance and subsequently leading to the occurrence and development of type 2 DM and GDM[36,37]. As a soluble splice variant of RAGE in the circulatory system, sRAGE can competitively bind to circulating AGEs, thereby attenuating the AGE–RAGE signal and protecting cells from AGE-mediated pathological effects[38]. Conversely, the increased clearance rate of the AGE /sRAGE complex under hyperglycemia may explain the lower sRAGE level[39]. Tang et al[40] used sRAGE to intervene in a GDM mouse model and improved fetal growth, suggesting that sRAGE can alleviate hyperglycemia-induced tissue damage. Yu et al[41] found significantly lower serum sRAGE levels in pregnant patients with type 1 DM who subsequently developed preeclampsia compared to those with normal blood pressure during pregnancy in the early and middle trimesters, demonstrating that sRAGE levels may be a key factor for GDM complicated with preeclampsia; however, the specific mechanism of action remains elusive, warranting further investigation.

Maternal oxidative stress affects the formation of reactive oxygen species in embryos. Further, hyperglycemia is the main driving force of AGE formation, especially in the presence of oxidative stress[42]. A considerable amount of evidence indicates that an improved oxidative stress response during pregnancy is associated with embryonic defects[43]. The main mechanisms of various chronic diseases associated with AGEs involve increased oxidative stress and inflammation[44]. Pregnancy is a period of physiological and physical disorganization (adapted to maintain fetal growth and to prepare for delivery and breastfeeding) to maintain the proper internal balance of the mother[45]. It is characterized by many physiological changes that increase basal oxygen consumption and change the use of energy substrates by various organs, including the fetoplacental unit. Further, pregnancy is associated with an increased susceptibility to oxidative stress due to the systemic inflammatory response, which plays a crucial role during pregnancy, normal labor, and preterm labor onset[46]. The systemic inflammatory response during pregnancy activates peripheral granulocytes, monocytes, and lymphocytes in late pregnancy, generating large amounts of reactive oxygen species (ROS)[47,48]. Further, increased AGE levels in maternal serum may have important effects on the mother in addition to influences on fetal and newborn health[16,49]. The results of this study revealed that mothers with GDM having adverse perinatal outcomes had notably higher serum AGE and MDA levels than those with normal perinatal outcomes at 24-28 weeks of pregnancy and before delivery. Further, mothers with GDM having adverse perinatal outcomes had statistically lower serum sRAGE and SOD levels vs those with GDM with normal perinatal outcomes. Generally, GDM may cause adverse pregnancy outcomes, including miscarriage, premature birth, intrauterine distress, fetal malformations, intrauterine death, intrauterine infection, macrosomia and hypertension during pregnancy, preeclampsia, and polyhydramnios[50,51]. Further, as the pathological hub of DM, the AGE–RAGE axis is strongly associated with the outcome of perinatal infants. Logistic regression results identified AGE level as a risk factor for adverse perinatal outcomes in GDM, whereas the sRAGE level served as a protective factor. The results of several animal and cell studies are consistent with a model that AGE and RAGE activation as well as the initiation of their downstream mechanisms play a pivotal role in the complications of DM. Animal studies have revealed that blocking the effect of AGE or RAGE alleviates DM complications[52]. In addition, Kansu-Celik et al[53] measured serum AGE levels in 53 women in early pregnancy and found a significant association with the risk of preterm labor. Therefore, high serum AGE levels in the first trimester of pregnancy may be a useful biomarker for predicting preterm labor. Meanwhile, pregnancy is a process that lasts more than 37 weeks on average; thus, in clinical practice, rather than a specific threshold, AGE levels are monitored during pregnancy for increasing trends. When one is identified, intervention is required, including dietary interventions such as reduction of intake of foods high in AGEs to reduce the risk of complications.

This study has some limitations. First, the experiments were conducted at a single center (the Obstetrics Department of Obstetrics and Gynecology Hospital of Fudan University); larger studies in multiple centers may make the data more convincing. Second, due to the limitations of retrospective studies, the sample size of the patients is small, and the retrospective design and small sample size limited statistical power; Also, a sample size that is too small will lead to insufficient data volume and thereby cause overfitting of the model. For example, in this study, the confidence interval range of AGE is too large (8.562-169.764); thus, the sample size should be further increased for analysis in future studies. Third, we did not compare AGE/RAGE/oxidative stress markers across the two time points (24-28 weeks vs predelivery) to assess changes over time, which may be impossible to guide pregnant women from pregnancy to pregnancy outcome. Fourth, the short period of this study may affect the generalizability of the results and the long-term effects. The period of the study should be extended in subsequent studies to include more data to observe the long-term effects of AGEs and sRAGE levels DM outcomes. Finally, we provided data on indicators related to oxidative stress and AGE–sRAGE in labor; however, drawing any conclusions about the role or mechanisms of these data on pregnancy outcomes is challenging. Therefore, well-designed, randomized, and controlled trials with prospective data collection and sample size calculation are needed to confirm the findings of our study and to explore the underlying mechanism of the AGE-RAGE axis in GDM.

We have shown that AGE-induced oxidative stress in GDM may contribute to adverse perinatal outcomes. This is a clinical association study, and thus future work focusing on the cause and effect of AGEs on GDM complications may be interesting. However, this study highlights the significance of pursuing AGE as a potential biomarker for predicting GDM complications.

CONCLUSION

Altogether, AGEs and RAGE may be associated with systemic oxidative stress in pregnant women with GDM, and a high serum AGE level is a risk factor for adverse perinatal outcomes in GDM. Our results may provide a new scientific basis for precision nutrition and prevention of GDM. For diseases associated with the AGE-RAGE axis, AGE/sRAGE levels may be a valuable alternative and practical generalized biomarker, and reducing AGEs in common foods through changes in cooking practices is a feasible and safe intervention. In the future, large-scale clinical randomized intervention trials will provide broader evidence of the deleterious effects of dietary AGEs on chronic diseases and to provide a scientific basis for public health policy formulation.