OBJECTIVE

To determine the association between maternal blood glucose patterns throughout pregnancy and neonatal amino acids and acylcarnitines.

RESEARCH DESIGN AND METHODS

We conducted a prospective cohort study involving 11,457 singleton pregnant women without preexisting diabetes from the Beijing Birth Cohort Study, along with their neonates born between July 2021 and October 2022 in Beijing, China. Distinct maternal glucose trajectories were identified using a latent class model based on blood glucose levels across the three trimesters, and their association with neonatal circulating metabolites, including 11 amino acids and 33 acylcarnitines, was examined, adjusting for potential confounding factors.

RESULTS

Three distinct groups of maternal glucose trajectories were identified: consistent normoglycemia (n = 8,648), mid-to-late gestational hyperglycemia (n = 2,540), and early-onset hyperglycemia (n = 269). Mid-to-late gestational hyperglycemia was associated with decreased levels of amino acids (alanine, arginine, ornithine, and proline) involved in the arginine and proline metabolism and urea cycle pathway, as well as increased levels of C4DC+C5-OH and decreased level of C6DC and C10:1. Early-onset hyperglycemia was associated with elevated levels of free acylcarnitine and C4DC+C5-OH and a decreased level of C10:1, involved in the fatty acid oxidation pathway. However, these associations were primarily observed in male neonates rather than in female neonates.

CONCLUSIONS

Our findings revealed a significant link between maternal glucose trajectories throughout pregnancy and neonatal arginine and proline metabolism, urea cycle pathway, and fatty acid oxidation pathway. These results highlight the importance of maintaining optimal blood glucose levels throughout pregnancy to promote healthy neonatal metabolic outcomes.

The developmental origin of health and disease hypothesis posits that an individual’s lifelong health is significantly impacted by the intrauterine environment (1). Gestational hyperglycemia, which primarily includes gestational diabetes mellitus (GDM), as well as type 1 diabetes mellitus (T1DM) and type 2 diabetes mellitus (T2DM), is estimated to affect ∼21.1 million (16.7%) pregnancies worldwide (2). It is a well-established and modifiable risk factor for both short- and long-term health outcomes in offspring, encompassing preterm birth, macrosomia, and unfavorable long-term health consequences, including neurodevelopmental, cardiovascular, and metabolic disorders (3–6).

Recent research has revealed that maternal hyperglycemia during pregnancy can lead to prenatal changes in the metabolic processes of offspring (7), which thus significantly affects their long-term metabolic health and neurodevelopment (8,9). Several studies, including the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study and Role of Nutrition and Maternal Genetics on the Programming of Development of Fetal Adipose Tissue (PREOBE) cohort study, have identified noticeable alterations in the metabolic composition of fetal cord blood, meconium, and neonatal urine associated with maternal hyperglycemia in the midpregnancy period (7,10–13).

These findings have highlighted the significance of maternal hyperglycemia in influencing fetal metabolism. However, it is important to note that previous research has predominantly focused on women with GDM diagnosed between the 24th and 28th weeks of pregnancy (7,10,11). Nevertheless, recent evidence suggests that elevated blood glucose levels during both early and late pregnancy have significant implications for the health of newborns, including an increased risk of macrosomia and preterm birth (14,15). Given that macrosomia and preterm birth are often associated with metabolic changes (16,17), it is plausible that hyperglycemia during early and late pregnancy may also affect the metabolic functioning of newborns. However, limited evidence is available regarding the specific metabolic changes in newborns resulting from hyperglycemia beyond the 24th to 28th week of pregnancy.

Therefore, this study used a large sample size to examine the potential relationship between hyperglycemia trajectories throughout pregnancy and the neonatal amino acids (AAs) and acylcarnitines (ACs).

Setting, Design, and Participants

The study cohort comprised pregnant Chinese women enrolled in the Beijing Birth Cohort Study (ChiCTR220058395) at Beijing Obstetrics and Gynecology Hospital. Commencing in 2016, the birth cohort recruited pregnant women between 6 and 13 weeks of gestation and followed them until delivery. Beginning in July 2021, the neonates underwent peripheral blood AAs and ACs screening using tandem mass spectrometry (MS/MS).

The present investigation involved singleton pregnant women who received routine prenatal care and delivered at this hospital within the time frame of 1 July 2021 to 30 October 2022, along with their neonates. The inclusion criteria required pregnant women to be between 18 and 50 years old and within 6–13 weeks of gestation at the time of recruitment. Additionally, they were expected to attend routine antenatal examinations and deliver at the study site. They were also required to provide informed written consent. The study excluded twin pregnancies and women with chronic conditions such as heart, kidney, and liver diseases, T1DM, and T2DM. Those without baseline information, maternal blood glucose measurements, or neonatal metabolites information were also excluded. Further information regarding this cohort can be found in a previous study (18).

The screening was conducted on the original group of 14,832 women, with the exclusion of 414 women who had twin pregnancies, 53 with chronic conditions, such as heart, kidney, or liver diseases, and 132 with T1DM or T2DM. Afterward, 1,543 neonates without metabolomic data, 141 women without baseline information data, and 1,092 women without blood glucose data throughout pregnancy were removed from the analysis, resulting in a final data set of 11,457 mother-child pairs (as shown in Supplementary Fig. 1). The study was conducted in accordance with the ethical standards outlined by the Declaration of Helsinki and was approved by the Beijing Obstetrics and Gynecology Hospital Ethics Committee (2018-ky-009–01). Written informed consent was obtained from all participants or their legal guardians.

Clinical Characteristic Measurements

The data on demographics and sociological factors were collected through a questionnaire, whereas the clinical data were derived from the medical records. Trained researchers followed a standardized protocol to take anthropometric measurements of pregnant women and neonates and also collected self-reported prepregnancy maternal body weight at the time of recruitment. In cases where participants were unable to recall their prepregnancy weight, the body weight at recruitment was used instead. The prepregnancy BMI was calculated by dividing the prepregnancy body weight (kg) by the square of the height (m2). Furthermore, the criteria for identifying large for gestational age (LGA) and small for gestational age (SGA) were based on the growth standard curves of Chinese newborns, with LGA and SGA being defined as values >90th percentile and <10th percentile for gestational age, respectively (19). Hypertensive disorders in pregnancy included gestational hypertension, preeclampsia, and eclampsia diagnosed according to ICD-10 code definitions.

Maternal Glucose Measurements Throughout Pregnancy

In this study, fasting blood glucose (FBG) levels during the 7–13 weeks (i.e., the first trimester [T1]) and 32–36 weeks of gestation (i.e., the third trimester [T3]) and 0-h, 1-h, and 2-h glucose levels in oral glucose tolerance test (OGTT) or FBG at 24–28 weeks of gestation (i.e., the second trimester [T2]) were measured using the glucose oxidase method with a DxC800 automatic biochemical analyzer.

In T1, elevated maternal glucose was defined as fasting glucose levels ≥5.3 mmol/L (20). In T2, elevated glucose levels corresponding to the diagnosis of GDM was made using the International Association of Diabetes and Pregnancy Study Groups (IADPSG) criteria, which included FBG levels ≥5.1 mmol/L and/or 1‐h glucose levels ≥10.0 mmol/L and/or 2‐h glucose levels ≥8.5 mmol/L during the 75-g OGTT (21). In T3, a fasting glucose of ≥5.3 mmol/L was categorized as elevated glucose based on previous report criteria (15).

Neonatal Metabolomics Measurements

Neonatal heel blood samples were collected 72 h postpartum and subsequently preserved at 4°C following filtration paper drying. MS/MS was used to analyze 42 metabolites, comprising 11 AAs and 31 ACs. The AAs and ACs were extracted using a methanolic extraction solution containing stable isotope-labeled internal standards according to the expert consensus on MS/MS techniques for neonatal diseases in China (22). The supernatant was quantified via electrospray ionization using a Waters Xevo TQD Mass Spectrometer (Milford, MA) under the following conditions: capillary, 3.20 KV; cone, 50 V; desolvation temperature, 350°C; desolvation, 800 L/h; cone, 50 L/h.

Statistical Analysis

We first explored various patterns of maternal glucose across T1, T2, and T3 using a group-based trajectory model. To distinguish discrete latent groups of maternal glucose levels, we used latent class logistic analysis using PROC TRAJ in SAS 9.4 software. The analysis used binomial maternal glucose levels over the three trimesters. Maternal glucose trajectories were modeled using a quadratic polynomial, and the optimal number of classes was determined based on the lowest Bayesian information criterion. Each class was named according to the most predominant characteristics observed within it.

Maternal and neonatal clinical characteristics were compared among different groups of maternal glucose trajectories. One-way ANOVA was used for normally distributed continuous variables, the Kruskal-Wallis test was used for nonnormally distributed continuous variables, and the χ2 test was used for categorical variables. The neonatal metabolites (including 11 AAs and 33 ACs) underwent log-transformation prior to the analysis. A one-way ANOVA analysis was used to compare neonatal metabolites across different maternal glucose trajectories. The Benjamini-Hochberg method was used to control for multiple testing, with a false discovery rate (FDR) threshold set at 0.05. The analysis was adjusted for several potential confounders, including maternal age (continuous), parity (multiparity or not), prepregnancy BMI (continuous), hypertensive disorders in pregnancy (yes or no), medical treatment (insulin and/or metformin use or nonuse), and neonatal sex (male or female). Pairwise post hoc tests were conducted using the Tukey-Kramer methods to identify specific differences between each group.

Subsequently, we conducted a linear regression analysis to examine the association between maternal glucose levels at different stages of pregnancy (expressed as continuous variables) and neonatal AAs and ACs. The model was controlled for the aforementioned covariates, and a Benjamini-Hochberg adjusted FDR <0.05 was implemented to address multiple comparisons.

To further explore and visualize the interactions between maternal glucose levels at different stages of pregnancy and neonatal metabolites, the Debiased Sparse Partial Correlation algorithm was used through the online MetaboAnalyst 6.0 tool (https://new.metaboanalyst.ca) (23). Neonatal metabolites demonstrating significant associations with maternal glucose indicators were selected for the partial correlation analysis.

To gain insight into functional pathways affected by different maternal glucose trajectories in neonates, a pathway analysis was conducted using metabolites associated with maternal glucose trajectories. The common metabolite names were converted into corresponding Human Metabolome Database (HMDB) IDs, which were subsequently used for further pathway analysis. The Pathway Analysis module within the MetaboAnalyst tool was used to integrate pathway enrichment data for the calculation of P values and topology analyses to assess pathway impact values, referencing the Small Molecule Pathway Database (SMPDB).

Additionally, our study included an analysis to investigate potential sex disparities regarding the association between maternal glucose trajectories and neonatal metabolites. A subgroup analysis was also conducted, with a specific focus on appropriate-for-age (AGA) neonates, to determine whether the influence of maternal blood glucose levels on neonatal metabolites was evident within this subgroup.

Characteristics of the Study Participants by Maternal Glucose Trajectories

The study included 11,457 women and their offspring, as revealed by the study flowchart (Supplementary Fig. 1). The participants were an average maternal age of 32.90 ± 3.87 years, and 77.04% of their neonates were firstborn. The average birth weight of the neonates was 3,316.15 ± 447.75 g, with 51.53% being male neonates. Among them, 14.74% were classified as LGA and 5.46% as SGA, as indicated in Table 1.

Table 1

Basic characteristics of the participants

Maternal and neonatal characteristicsAll (N = 11,457)Group 1* (n = 8,648)Group 2 (n = 2,540)Group 3 (n = 269)P value§
Maternal characteristics      
 Maternal age, years 32.90 ± 3.87 32.61 ± 3.78 33.72 ± 3.94 34.72 ± 4.13 <0.0001 
 Gravidity, 1st 6,473 (56.50) 5,025 (58.11) 1,334 (52.52) 114 (42.38) <0.0001 
 Parity, 1st 8,826 (77.04) 6,747 (76.44) 1,899 (74.76) 180 (66.91) <0.0001 
 Maternal height, cm 162.81 ± 5.06 162.88 ± 5.05 162.56 ± 5.10 162.85 ± 4.98 0.0178 
 Prepregnancy weight, kg 58.56 ± 9.41 57.59 ± 8.76 60.97 ± 10.36 67.13 ± 11.57 <0.0001 
 Prepregnancy BMI, kg/cm2 22.08 ± 3.28 21.69 ± 3.03 23.05 ± 3.61 25.29 ± 4.01 <0.0001 
 Cesarean section 4,694 (40.97) 3,431 (39.67) 1,140 (44.88) 123 (45.72) <0.0001 
 Hypertensive disorders in pregnancy 1,211 (10.57) 748 (8.65) 406 (15.98) 57 (21.19) <0.0001 
Maternal glucose levels in pregnancy      
 Fasting glucose in T1, mmol/L 4.66 ± 0.38 4.59 ± 0.31 4.79 ± 0.44 5.63 ± 0.45 <0.0001 
 Gestational weeks of measurement in T1 8.43 (7.57–9.86) 8.43 (7.57–9.86) 8.43 (7.57–9.86) 8.29 (7.57–9.71) 0.6808 
 Glucose levels during 75-g OGTT in T2, mmol/L      
  Fasting glucose 4.52 ± 0.38 4.43 ± 0.29 4.76 ± 0.44 5.30 ± 0.58 <0.0001 
  1-h glucose 7.80 ± 1.70 7.30 ± 1.34 9.27 ± 1.75 10.47 ± 1.81 <0.0001 
  2-h glucose 6.70 ± 1.40 6.29 ± 1.02 7.95 ± 1.57 8.67 ± 1.84 <0.0001 
 Gestational weeks of measurement in T2 25.43 (24.57–26.29) 25.14 (24.57–26.00) 26.00 (25.00–26.71) 26.29 (25.14–26.86) <0.0001 
 Fasting glucose in T3, mmol/L 4.45 ± 0.66 4.28 ± 0.38 4.97 ± 1.03 4.97 ± 0.78 <0.0001 
 Gestational weeks of measurement in T3 34.86 (34.00–38.57) 34.71 (34.00–38.43) 35.86 (34.14–39.00) 35.43 (34.14–38.57) <0.0001 
 Medical treatment (use of insulin and/or metformin) 150 (1.31) 0 (0) 97 (3.82) 53 (19.70) <0.0001 
Neonatal characteristics      
 Male sex 5,896 (51.46) 4,456 (51.53) 1,295 (50.98) 145 (53.90) 0.6414 
 Birth weight, g 3,316.15 ± 447.75 3,324.36 ± 429.33 3,283.07 ± 491.63 3,364.57 ± 564.02 <0.0001 
 Gestational age, weeks 39.0 (38.0–40.0) 39.0 (38.0–40.0) 39.0 (38.0–40.0) 39.0 (38.0–39.0) <0.0001 
 Macrosomia 644 (5.62) 465 (5.38) 151 (5.94) 28 (10.41) 0.0014 
 Low birth weight 404 (3.53) 235 (2.72) 152 (5.98) 17 (6.32) <0.0001 
 LGA 1,689 (14.74) 1,191 (13.77) 432 (17.01) 66 (24.54) <0.0001 
 SGA 625 (5.46) 479 (5.54) 140 (5.51) 6 (2.23) 0.0621 
 Preterm birth 547 (4.77) 320 (3.70) 200 (7.87) 27 (10.04) <0.0001 
Maternal and neonatal characteristicsAll (N = 11,457)Group 1* (n = 8,648)Group 2 (n = 2,540)Group 3 (n = 269)P value§
Maternal characteristics      
 Maternal age, years 32.90 ± 3.87 32.61 ± 3.78 33.72 ± 3.94 34.72 ± 4.13 <0.0001 
 Gravidity, 1st 6,473 (56.50) 5,025 (58.11) 1,334 (52.52) 114 (42.38) <0.0001 
 Parity, 1st 8,826 (77.04) 6,747 (76.44) 1,899 (74.76) 180 (66.91) <0.0001 
 Maternal height, cm 162.81 ± 5.06 162.88 ± 5.05 162.56 ± 5.10 162.85 ± 4.98 0.0178 
 Prepregnancy weight, kg 58.56 ± 9.41 57.59 ± 8.76 60.97 ± 10.36 67.13 ± 11.57 <0.0001 
 Prepregnancy BMI, kg/cm2 22.08 ± 3.28 21.69 ± 3.03 23.05 ± 3.61 25.29 ± 4.01 <0.0001 
 Cesarean section 4,694 (40.97) 3,431 (39.67) 1,140 (44.88) 123 (45.72) <0.0001 
 Hypertensive disorders in pregnancy 1,211 (10.57) 748 (8.65) 406 (15.98) 57 (21.19) <0.0001 
Maternal glucose levels in pregnancy      
 Fasting glucose in T1, mmol/L 4.66 ± 0.38 4.59 ± 0.31 4.79 ± 0.44 5.63 ± 0.45 <0.0001 
 Gestational weeks of measurement in T1 8.43 (7.57–9.86) 8.43 (7.57–9.86) 8.43 (7.57–9.86) 8.29 (7.57–9.71) 0.6808 
 Glucose levels during 75-g OGTT in T2, mmol/L      
  Fasting glucose 4.52 ± 0.38 4.43 ± 0.29 4.76 ± 0.44 5.30 ± 0.58 <0.0001 
  1-h glucose 7.80 ± 1.70 7.30 ± 1.34 9.27 ± 1.75 10.47 ± 1.81 <0.0001 
  2-h glucose 6.70 ± 1.40 6.29 ± 1.02 7.95 ± 1.57 8.67 ± 1.84 <0.0001 
 Gestational weeks of measurement in T2 25.43 (24.57–26.29) 25.14 (24.57–26.00) 26.00 (25.00–26.71) 26.29 (25.14–26.86) <0.0001 
 Fasting glucose in T3, mmol/L 4.45 ± 0.66 4.28 ± 0.38 4.97 ± 1.03 4.97 ± 0.78 <0.0001 
 Gestational weeks of measurement in T3 34.86 (34.00–38.57) 34.71 (34.00–38.43) 35.86 (34.14–39.00) 35.43 (34.14–38.57) <0.0001 
 Medical treatment (use of insulin and/or metformin) 150 (1.31) 0 (0) 97 (3.82) 53 (19.70) <0.0001 
Neonatal characteristics      
 Male sex 5,896 (51.46) 4,456 (51.53) 1,295 (50.98) 145 (53.90) 0.6414 
 Birth weight, g 3,316.15 ± 447.75 3,324.36 ± 429.33 3,283.07 ± 491.63 3,364.57 ± 564.02 <0.0001 
 Gestational age, weeks 39.0 (38.0–40.0) 39.0 (38.0–40.0) 39.0 (38.0–40.0) 39.0 (38.0–39.0) <0.0001 
 Macrosomia 644 (5.62) 465 (5.38) 151 (5.94) 28 (10.41) 0.0014 
 Low birth weight 404 (3.53) 235 (2.72) 152 (5.98) 17 (6.32) <0.0001 
 LGA 1,689 (14.74) 1,191 (13.77) 432 (17.01) 66 (24.54) <0.0001 
 SGA 625 (5.46) 479 (5.54) 140 (5.51) 6 (2.23) 0.0621 
 Preterm birth 547 (4.77) 320 (3.70) 200 (7.87) 27 (10.04) <0.0001 

Data are presented as n (%), mean ± SD, or median (interquartile range).

*Group 1 indicates consistent normoglycemia.

†Group 2 indicates mid-to-late gestational hyperglycemia.

‡Group 3 indicates early-onset hyperglycemia.

§P values were calculated by the one-way ANOVA for normally distributed continuous variables, the Kruskal-Wallis test for nonnormally distributed continuous variables, and the χ2 test for categorical variables.

Among these women, 524 had elevated glucose levels in T1, 2,060 had GDM, and 801 had elevated glucose levels in T3 (Supplementary Fig. 1). Based on these glucose level classifications across pregnancy, we identified three distinct groups of maternal glucose trajectories using a latent class model. Group 1, named consistent normoglycemia, consisted of 8,648 participants who maintained normal glucose levels throughout pregnancy. Group 2, known as mid-to-late gestational hyperglycemia, comprised 2,540 participants primarily diagnosed with GDM and/or those with elevated blood glucose levels in T3. Lastly, group 3, designated as early-onset hyperglycemia, consisted of 269 participants with elevated blood glucose levels from T1 continuing into T2 or T3, or both (Fig. 1A).

Figure 1

A: Maternal glucose trajectories identified by the latent class model. The horizontal axis represents the gestational stage, and the vertical axis represents the probability of elevated blood glucose levels in each group. B: Differential neonatal metabolites between maternal glucose trajectory groups (FDR <0.05).

Figure 1

A: Maternal glucose trajectories identified by the latent class model. The horizontal axis represents the gestational stage, and the vertical axis represents the probability of elevated blood glucose levels in each group. B: Differential neonatal metabolites between maternal glucose trajectory groups (FDR <0.05).

Close modal

The participants of various maternal glucose trajectories showed significant differences in maternal age, gravidity, parity, prepregnancy weight and BMI, presence of hypertensive disorders in pregnancy, the proportion of cesarean section, risks of macrosomia, LGA, low birth weight, and preterm delivery (Table 1).

Maternal Blood Glucose Levels and Neonatal Metabolites

Figure 1B shows the 11 neonatal metabolites associated with maternal blood glucose trajectories, taking into account potential confounders. In the mid-to-late gestational hyperglycemia group, decreased levels of alanine, arginine, glycine, leucine + isoleucine + hydroxyproline, ornithine, proline, valine, C6DC, and C10:1, as well as increased level of C4DC+C5-OH were observed compared with the consistent normoglycemia group. The early-onset hyperglycemia group showed increased levels of free AC (C0) and C4DC+C5-OH and decreased levels of C10:1 compared with the normoglycemia group.

Table 2 presents the coefficient between neonatal metabolites and FBG in T1, T2, and T3, as well as postload blood glucose after 75-g OGTT in T2, after controlling for potential confounding factors. The metabolites linked to different maternal blood glucose indicators exhibit both common and distinctive characteristics.

Table 2

Associations between maternal glucose indicators throughout pregnancy and neonatal metabolites

FBG-T1OGTT-0 h-T2OGTT-1 h-T2OGTT-2 h-T2FBG-T3
β (95% CI)FDR*β (95% CI)FDR*β (95% CI)FDR*β (95% CI)FDR*β (95% CI)FDR*
Alanine 0.005 (−0.002 to 0.011) 0.247 −0.011 (−0.017 to −0.004) 0.003 −0.003 (−0.005 to −0.002) <0.001 −0.003 (−0.004 to −0.001) 0.012 −0.001 (−0.005 to 0.002) 0.651 
Arginine 0.006 (−0.008 to 0.021) 0.559 −0.052 (−0.067 to −0.036) <0.001 −0.008 (−0.011 to −0.004) <0.001 −0.006 (−0.010 to −0.002) 0.013 0.000 (−0.009 to 0.008) 0.952 
Citrulline 0.002 (−0.004 to 0.008) 0.596 −0.008 (−0.014 to −0.001) 0.049 −0.001 (−0.003 to 0.000) 0.265 −0.001 (−0.003 to 0.001) 0.249 −0.002 (−0.006 to 0.002) 0.467 
Glycine −0.003 (−0.010 to 0.003) 0.487 −0.001 (−0.008 to 0.006) 0.877 −0.002 (−0.004 to −0.001) 0.013 −0.003 (−0.005 to −0.001) 0.003 −0.002 (−0.006 to 0.001) 0.373 
Leucine + isoleucine + hydroxyproline 0.009 (0.003 to 0.014) 0.016 −0.003 (−0.009 to 0.003) 0.512 −0.002 (−0.004 to −0.001) 0.002 −0.002 (−0.003 to −0.001) 0.058 0.0006 (−0.003 to 0.004) 0.834 
Methionine 0.008 (0.003 to 0.014) 0.028 −0.008 (−0.014 to −0.002) 0.034 −0.001 (−0.002 to 0.001) 0.4110 −0.001 (−0.003 to 0.000) 0.205 0.002 (−0.001 to 0.006) 0.373 
Ornithine 0.004 (−0.004 to 0.011) 0.538 −0.012 (−0.020 to −0.004) 0.006 −0.004 (−0.006 to −0.002) <0.001 −0.003 (−0.005 to −0.001) 0.012 −0.005 (−0.009 to −0.000) 0.147 
Phenylalanine 0.008 (0.004 to 0.012) 0.008 0.008 (0.003 to 0.012) 0.003 0.001 (0.000 to 0.002) 0.196 0.001 (−0.000 to 0.002) 0.274 0.004 (0.001 to 0.006) 0.033 
Proline 0.002 (−0.003 to 0.008) 0.563 −0.011 (−0.016 to −0.005) 0.002 −0.004 (−0.006 to −0.003) <0.001 −0.005 (−0.007 to −0.004) 0.001 −0.004 (−0.007 to −0.001) 0.081 
Tyrosine 0.007 (−0.000 to 0.015) 0.134 −0.001 (−0.009 to 0.007) 0.885 −0.001 (−0.003 to 0.001) 0.370 −0.001 (−0.003 to 0.001) 0.341 −0.004 (−0.008 to 0.000) 0.257 
Valine 0.009 (0.004 to 0.015) 0.014 −0.002 (−0.008 to 0.004) 0.760 −0.003 (−0.004 to −0.001) <0.001 −0.003 (−0.004 to −0.001) 0.004 −0.001 (−0.004 to 0.003) 0.834 
C0 0.002 (−0.004 to 0.009) 0.580 −0.005 (−0.012 to 0.002) 0.304 0.001 (−0.001 to 0.003) 0.411 0.002 (0.000 to 0.004) 0.049 0.000 (−0.004 to 0.004) 0.949 
C2 −0.009 (−0.017 to −0.002) 0.074 0.007 (−0.002 to 0.015) 0.201 0.001 (−0.001 to 0.003) 0.370 0.000 (−0.002 to 0.002) 0.850 0.003 (−0.001 to 0.008) 0.366 
C3 0.001 (−0.008 to 0.010) 0.910 0.025 (0.016 to 0.034) <0.001 0.002 (0.000 to 0.004) 0.265 0.002 (0.000 to 0.005) 0.155 0.015 (0.010 to 0.020) <0.001 
C3DC+C4-OH −0.018 (−0.028 to −0.007) 0.014 0.011 (−0.001 to 0.022) 0.123 0.000 (−0.002 to 0.003) 0.867 −0.004 (−0.007 to −0.001) 0.036 0.004 (−0.002 to 0.010) 0.366 
C4 −0.007 (−0.013 to −0.000) 0.119 −0.001 (−0.007 to 0.006) 0.877 0.001 (−0.001 to 0.002) 0.501 0.000 (−0.002 to 0.002) 0.850 0.000 (−0.004 to 0.004) 1.000 
C4DC+C5-OH 0.002 (−0.005 to 0.008) 0.642 0.017 (0.010 to 0.024) <0.001 0.004 (0.002 to 0.005) <0.001 0.003 (0.001 to 0.005) 0.006 0.004 (0.001 to 0.008) 0.099 
C5 0.006 (−0.0017 to 0.014) 0.236 −0.002 (−0.010 to 0.006) 0.778 0.001 (−0.001 to 0.003) 0.370 0.002 (0.000 to 0.004) 0.155 0.002 (−0.002 to 0.007) 0.502 
C5:1 −0.001 (−0.005 to 0.002) 0.559 0.000 (−0.004 to 0.003) 0.909 0.000 (−0.001 to 0.001) 0.574 0.000 (−0.001 to 0.001) 0.691 0.003 (0.002 to 0.005) 0.007 
C5DC+C6-OH −0.003 (−0.009 to 0.005) 0.580 0.015 (0.007 to 0.022) <0.001 0.002 (0.000 to 0.004) 0.078 −0.001 (−0.002 to 0.002) 0.691 0.006 (0.002 to 0.010) 0.034 
C6 −0.008 (−0.015 to −0.000) 0.124 −0.003 (−0.011 to 0.004) 0.568 0.000 (−0.001 to 0.002) 0.867 −0.001 (−0.003 to 0.001) 0.301 −0.001 (−0.006 to 0.003) 0.641 
C6DC −0.011 (−0.020 to −0.003) 0.052 −0.023 (−0.032 to −0.013) <0.001 −0.003 (−0.005 to −0.001) 0.005 −0.006 (−0.009 to −0.004) 0.001 0.002 (−0.004 to 0.008) 0.641 
C8 −0.015 (−0.024 to −0.005) 0.028 −0.002 (−0.013 to 0.008) 0.804 0.001 (−0.001 to 0.004) 0.433 0.001 (−0.002 to 0.004) 0.567 0.002 (−0.004 to 0.008) 0.641 
SA 0.004 (−0.001 to 0.009) 0.247 0.001 (−0.005 to 0.007) 0.877 0.000 (−0.001 to 0.001) 0.867 −0.002 (−0.004–0.001) 0.012 0.004 (0.001 to 0.007) 0.081 
C10 −0.013 (−0.022 to −0.004) 0.032 −0.002 (−0.012 to 0.008) 0.877 0.000 (−0.002 to 0.002) 0.871 −0.002 (−0.005 to 0.000) 0.128 0.001 (−0.004 to 0.006) 0.813 
C10:1 −0.013 (−0.021 to −0.005) 0.016 −0.026 (−0.035 to −0.018) <0.001 −0.007 (−0.009 to −0.005) <0.001 −0.010 (−0.012 to −0.008) 0.001 −0.010 (−0.015 to −0.005) <0.001 
C10:2 −0.003 (−0.010 to 0.004) 0.559 −0.001 (−0.017 to −0.002) 0.031 −0.002 (−0.003 to 0.000) 0.102 −0.004 (−0.006 to −0.002) 0.001 −0.002 (−0.006 to 0.002) 0.614 
C12 −0.009 (−0.019 to 0.001) 0.144 0.004 (−0.006 to 0.015) 0.612 0.000 (−0.002 to 0.002) 0.925 −0.004 (−0.007 to −0.002) 0.011 0.002 (−0.003 to 0.008) 0.614 
C12:1 0.000 (−0.015 to 0.014) 0.991 0.001 (−0.015 to 0.017) 0.909 −0.002 (−0.005 to 0.002) 0.532 −0.007 (−0.011 to −0.003) 0.008 −0.001 (−0.010 to 0.007) 0.834 
C14 −0.004 (−0.012 to 0.004) 0.512 0.006 (−0.003 to 0.014) 0.300 0.001 (−0.001 to 0.003) 0.549 −0.002 (−0.004 to 0.000) 0.151 0.002 (−0.003 to 0.007) 0.614 
C14-OH 0.008 (−0.000 to 0.017) 0.134 0.002 (−0.007 to 0.011) 0.804 0.001 (−0.001 to 0.003) 0.656 −0.003 (−0.005 to −0.000) 0.050 0.003 (−0.002 to 0.008) 0.373 
C14:1 −0.010 (−0.020 to 0.000) 0.124 0.003 (−0.007 to 0.014) 0.735 0.001 (−0.002 to 0.003) 0.801 −0.003 (−0.006 to −0.001) 0.040 0.002 (−0.003 to 0.008) 0.614 
C14:2 −0.008 (−0.016 to 0.001) 0.166 −0.016 (−0.025 to −0.007) 0.003 −0.001 (−0.003 to 0.001) 0.501 −0.004 (−0.006 to −0.001) 0.012 0.000 (−0.005 to 0.005) 0.949 
C16 −0.010 (−0.022 to 0.001) 0.144 0.020 (0.009 to 0.032) 0.003 0.002 (−0.001 to 0.004) 0.399 −0.001 (−0.004 to 0.002) 0.631 0.005 (−0.001 to 0.012) 0.314 
C16-OH −0.003 (−0.011 to 0.006) 0.629 0.011 (0.002 to 0.020) 0.037 0.001 (−0.001 to 0.003) 0.741 −0.003 (−0.005 to 0.000) 0.058 0.004 (−0.001 to 0.009) 0.257 
C16:1 −0.016 (−0.030 to −0.003) 0.074 0.025 (0.011 to 0.040) 0.003 0.003 (0.000 to 0.006) 0.220 0.000 (−0.004 to 0.004) 0.962 0.005 (−0.003 to 0.013) 0.427 
C16:1-OH −0.008 (−0.017 to 0.000) 0.134 −0.009 (−0.017 to 0.000) 0.113 0.001 (−0.002 to 0.002) 0.753 −0.001 (−0.004 to 0.001) 0.390 −0.003 (−0.007 to 0.002) 0.502 
C18 −0.004 (−0.013 to 0.004) 0.521 0.014 (0.006 to 0.023) 0.004 0.002 (0.000 to 0.004) 0.074 0.005 (−0.002 to 0.003) 0.691 0.006 (0.001 to 0.011) 0.081 
C18-OH 0.000 (−0.006 to 0.006) 1.000 −0.001 (−0.007 to 0.006) 0.909 0.000 (−0.001 to 0.001) 0.932 −0.003 (−0.004 to −0.001) 0.006 0.002 (−0.002 to 0.005) 0.522 
C18:1 −0.008 (−0.015 to −0.001) 0.114 0.006 (−0.002 to 0.014) 0.204 0.003 (0.001 to 0.004) 0.018 0.002 (−0.001 to 0.004) 0.205 0.003 (−0.001 to 0.008) 0.323 
C18:1-OH −0.003 (−0.012 to 0.005) 0.563 0.015 (0.006 to 0.024) 0.003 0.001 (−0.001 to 0.003) 0.411 0.001 (−0.001 to 0.004) 0.341 0.005 (−0.000 to 0.010) 0.234 
C18:2 0.003 (−0.007 to 0.013) 0.616 −0.020 (−0.030 to −0.010) <0.001 −0.001 (−0.004 to 0.001) 0.411 0.002 (−0.001 to 0.005) 0.249 −0.004 (−0.010 to 0.002) 0.366 
FBG-T1OGTT-0 h-T2OGTT-1 h-T2OGTT-2 h-T2FBG-T3
β (95% CI)FDR*β (95% CI)FDR*β (95% CI)FDR*β (95% CI)FDR*β (95% CI)FDR*
Alanine 0.005 (−0.002 to 0.011) 0.247 −0.011 (−0.017 to −0.004) 0.003 −0.003 (−0.005 to −0.002) <0.001 −0.003 (−0.004 to −0.001) 0.012 −0.001 (−0.005 to 0.002) 0.651 
Arginine 0.006 (−0.008 to 0.021) 0.559 −0.052 (−0.067 to −0.036) <0.001 −0.008 (−0.011 to −0.004) <0.001 −0.006 (−0.010 to −0.002) 0.013 0.000 (−0.009 to 0.008) 0.952 
Citrulline 0.002 (−0.004 to 0.008) 0.596 −0.008 (−0.014 to −0.001) 0.049 −0.001 (−0.003 to 0.000) 0.265 −0.001 (−0.003 to 0.001) 0.249 −0.002 (−0.006 to 0.002) 0.467 
Glycine −0.003 (−0.010 to 0.003) 0.487 −0.001 (−0.008 to 0.006) 0.877 −0.002 (−0.004 to −0.001) 0.013 −0.003 (−0.005 to −0.001) 0.003 −0.002 (−0.006 to 0.001) 0.373 
Leucine + isoleucine + hydroxyproline 0.009 (0.003 to 0.014) 0.016 −0.003 (−0.009 to 0.003) 0.512 −0.002 (−0.004 to −0.001) 0.002 −0.002 (−0.003 to −0.001) 0.058 0.0006 (−0.003 to 0.004) 0.834 
Methionine 0.008 (0.003 to 0.014) 0.028 −0.008 (−0.014 to −0.002) 0.034 −0.001 (−0.002 to 0.001) 0.4110 −0.001 (−0.003 to 0.000) 0.205 0.002 (−0.001 to 0.006) 0.373 
Ornithine 0.004 (−0.004 to 0.011) 0.538 −0.012 (−0.020 to −0.004) 0.006 −0.004 (−0.006 to −0.002) <0.001 −0.003 (−0.005 to −0.001) 0.012 −0.005 (−0.009 to −0.000) 0.147 
Phenylalanine 0.008 (0.004 to 0.012) 0.008 0.008 (0.003 to 0.012) 0.003 0.001 (0.000 to 0.002) 0.196 0.001 (−0.000 to 0.002) 0.274 0.004 (0.001 to 0.006) 0.033 
Proline 0.002 (−0.003 to 0.008) 0.563 −0.011 (−0.016 to −0.005) 0.002 −0.004 (−0.006 to −0.003) <0.001 −0.005 (−0.007 to −0.004) 0.001 −0.004 (−0.007 to −0.001) 0.081 
Tyrosine 0.007 (−0.000 to 0.015) 0.134 −0.001 (−0.009 to 0.007) 0.885 −0.001 (−0.003 to 0.001) 0.370 −0.001 (−0.003 to 0.001) 0.341 −0.004 (−0.008 to 0.000) 0.257 
Valine 0.009 (0.004 to 0.015) 0.014 −0.002 (−0.008 to 0.004) 0.760 −0.003 (−0.004 to −0.001) <0.001 −0.003 (−0.004 to −0.001) 0.004 −0.001 (−0.004 to 0.003) 0.834 
C0 0.002 (−0.004 to 0.009) 0.580 −0.005 (−0.012 to 0.002) 0.304 0.001 (−0.001 to 0.003) 0.411 0.002 (0.000 to 0.004) 0.049 0.000 (−0.004 to 0.004) 0.949 
C2 −0.009 (−0.017 to −0.002) 0.074 0.007 (−0.002 to 0.015) 0.201 0.001 (−0.001 to 0.003) 0.370 0.000 (−0.002 to 0.002) 0.850 0.003 (−0.001 to 0.008) 0.366 
C3 0.001 (−0.008 to 0.010) 0.910 0.025 (0.016 to 0.034) <0.001 0.002 (0.000 to 0.004) 0.265 0.002 (0.000 to 0.005) 0.155 0.015 (0.010 to 0.020) <0.001 
C3DC+C4-OH −0.018 (−0.028 to −0.007) 0.014 0.011 (−0.001 to 0.022) 0.123 0.000 (−0.002 to 0.003) 0.867 −0.004 (−0.007 to −0.001) 0.036 0.004 (−0.002 to 0.010) 0.366 
C4 −0.007 (−0.013 to −0.000) 0.119 −0.001 (−0.007 to 0.006) 0.877 0.001 (−0.001 to 0.002) 0.501 0.000 (−0.002 to 0.002) 0.850 0.000 (−0.004 to 0.004) 1.000 
C4DC+C5-OH 0.002 (−0.005 to 0.008) 0.642 0.017 (0.010 to 0.024) <0.001 0.004 (0.002 to 0.005) <0.001 0.003 (0.001 to 0.005) 0.006 0.004 (0.001 to 0.008) 0.099 
C5 0.006 (−0.0017 to 0.014) 0.236 −0.002 (−0.010 to 0.006) 0.778 0.001 (−0.001 to 0.003) 0.370 0.002 (0.000 to 0.004) 0.155 0.002 (−0.002 to 0.007) 0.502 
C5:1 −0.001 (−0.005 to 0.002) 0.559 0.000 (−0.004 to 0.003) 0.909 0.000 (−0.001 to 0.001) 0.574 0.000 (−0.001 to 0.001) 0.691 0.003 (0.002 to 0.005) 0.007 
C5DC+C6-OH −0.003 (−0.009 to 0.005) 0.580 0.015 (0.007 to 0.022) <0.001 0.002 (0.000 to 0.004) 0.078 −0.001 (−0.002 to 0.002) 0.691 0.006 (0.002 to 0.010) 0.034 
C6 −0.008 (−0.015 to −0.000) 0.124 −0.003 (−0.011 to 0.004) 0.568 0.000 (−0.001 to 0.002) 0.867 −0.001 (−0.003 to 0.001) 0.301 −0.001 (−0.006 to 0.003) 0.641 
C6DC −0.011 (−0.020 to −0.003) 0.052 −0.023 (−0.032 to −0.013) <0.001 −0.003 (−0.005 to −0.001) 0.005 −0.006 (−0.009 to −0.004) 0.001 0.002 (−0.004 to 0.008) 0.641 
C8 −0.015 (−0.024 to −0.005) 0.028 −0.002 (−0.013 to 0.008) 0.804 0.001 (−0.001 to 0.004) 0.433 0.001 (−0.002 to 0.004) 0.567 0.002 (−0.004 to 0.008) 0.641 
SA 0.004 (−0.001 to 0.009) 0.247 0.001 (−0.005 to 0.007) 0.877 0.000 (−0.001 to 0.001) 0.867 −0.002 (−0.004–0.001) 0.012 0.004 (0.001 to 0.007) 0.081 
C10 −0.013 (−0.022 to −0.004) 0.032 −0.002 (−0.012 to 0.008) 0.877 0.000 (−0.002 to 0.002) 0.871 −0.002 (−0.005 to 0.000) 0.128 0.001 (−0.004 to 0.006) 0.813 
C10:1 −0.013 (−0.021 to −0.005) 0.016 −0.026 (−0.035 to −0.018) <0.001 −0.007 (−0.009 to −0.005) <0.001 −0.010 (−0.012 to −0.008) 0.001 −0.010 (−0.015 to −0.005) <0.001 
C10:2 −0.003 (−0.010 to 0.004) 0.559 −0.001 (−0.017 to −0.002) 0.031 −0.002 (−0.003 to 0.000) 0.102 −0.004 (−0.006 to −0.002) 0.001 −0.002 (−0.006 to 0.002) 0.614 
C12 −0.009 (−0.019 to 0.001) 0.144 0.004 (−0.006 to 0.015) 0.612 0.000 (−0.002 to 0.002) 0.925 −0.004 (−0.007 to −0.002) 0.011 0.002 (−0.003 to 0.008) 0.614 
C12:1 0.000 (−0.015 to 0.014) 0.991 0.001 (−0.015 to 0.017) 0.909 −0.002 (−0.005 to 0.002) 0.532 −0.007 (−0.011 to −0.003) 0.008 −0.001 (−0.010 to 0.007) 0.834 
C14 −0.004 (−0.012 to 0.004) 0.512 0.006 (−0.003 to 0.014) 0.300 0.001 (−0.001 to 0.003) 0.549 −0.002 (−0.004 to 0.000) 0.151 0.002 (−0.003 to 0.007) 0.614 
C14-OH 0.008 (−0.000 to 0.017) 0.134 0.002 (−0.007 to 0.011) 0.804 0.001 (−0.001 to 0.003) 0.656 −0.003 (−0.005 to −0.000) 0.050 0.003 (−0.002 to 0.008) 0.373 
C14:1 −0.010 (−0.020 to 0.000) 0.124 0.003 (−0.007 to 0.014) 0.735 0.001 (−0.002 to 0.003) 0.801 −0.003 (−0.006 to −0.001) 0.040 0.002 (−0.003 to 0.008) 0.614 
C14:2 −0.008 (−0.016 to 0.001) 0.166 −0.016 (−0.025 to −0.007) 0.003 −0.001 (−0.003 to 0.001) 0.501 −0.004 (−0.006 to −0.001) 0.012 0.000 (−0.005 to 0.005) 0.949 
C16 −0.010 (−0.022 to 0.001) 0.144 0.020 (0.009 to 0.032) 0.003 0.002 (−0.001 to 0.004) 0.399 −0.001 (−0.004 to 0.002) 0.631 0.005 (−0.001 to 0.012) 0.314 
C16-OH −0.003 (−0.011 to 0.006) 0.629 0.011 (0.002 to 0.020) 0.037 0.001 (−0.001 to 0.003) 0.741 −0.003 (−0.005 to 0.000) 0.058 0.004 (−0.001 to 0.009) 0.257 
C16:1 −0.016 (−0.030 to −0.003) 0.074 0.025 (0.011 to 0.040) 0.003 0.003 (0.000 to 0.006) 0.220 0.000 (−0.004 to 0.004) 0.962 0.005 (−0.003 to 0.013) 0.427 
C16:1-OH −0.008 (−0.017 to 0.000) 0.134 −0.009 (−0.017 to 0.000) 0.113 0.001 (−0.002 to 0.002) 0.753 −0.001 (−0.004 to 0.001) 0.390 −0.003 (−0.007 to 0.002) 0.502 
C18 −0.004 (−0.013 to 0.004) 0.521 0.014 (0.006 to 0.023) 0.004 0.002 (0.000 to 0.004) 0.074 0.005 (−0.002 to 0.003) 0.691 0.006 (0.001 to 0.011) 0.081 
C18-OH 0.000 (−0.006 to 0.006) 1.000 −0.001 (−0.007 to 0.006) 0.909 0.000 (−0.001 to 0.001) 0.932 −0.003 (−0.004 to −0.001) 0.006 0.002 (−0.002 to 0.005) 0.522 
C18:1 −0.008 (−0.015 to −0.001) 0.114 0.006 (−0.002 to 0.014) 0.204 0.003 (0.001 to 0.004) 0.018 0.002 (−0.001 to 0.004) 0.205 0.003 (−0.001 to 0.008) 0.323 
C18:1-OH −0.003 (−0.012 to 0.005) 0.563 0.015 (0.006 to 0.024) 0.003 0.001 (−0.001 to 0.003) 0.411 0.001 (−0.001 to 0.004) 0.341 0.005 (−0.000 to 0.010) 0.234 
C18:2 0.003 (−0.007 to 0.013) 0.616 −0.020 (−0.030 to −0.010) <0.001 −0.001 (−0.004 to 0.001) 0.411 0.002 (−0.001 to 0.005) 0.249 −0.004 (−0.010 to 0.002) 0.366 

The models were adjusted for maternal age, parity, prepregnancy BMI hypertensive disorders in pregnancy, medications for hyperglycemia, and neonatal sex.

*FDR was calculated to account for multiple comparisons using the Benjamini-Hochberg method.

Notably, there was a negative association between alanine, arginine, ornithine, proline, and OGTT 0-h, 1-h, and 2-h glucose levels. Branched-chain amino acids (BCAAs), including leucine + isoleucine + hydroxyproline and valine, were positively associated with FBG in T1 but negatively associated with postload glucose levels in T2. For ACs, there was a positive association between levels of short-chain ACs C4DC+C5-OH, C5DC+C6-OH, and glucose levels in T2 or T3, whereas a midchain AC (C6DC) and unsaturated-chain ACs (C10:1, C10:2, and C14:2) were negatively associated with glucose indicators. Some long-chain ACs (C16, C16-OH, C16:1, C18, C18:1, and C18:1-OH) were positively associated with glucose levels in T2. Associations between other ACs with glucose indicators varied throughout gestation (Table 2).

Metabolic Pathways Associated With Maternal Blood Glucose Levels

The results of the DSPC network analyses depicted in Fig. 2A demonstrated a significant correlation between maternal glucose levels and the circulating metabolite network in neonates. The network of maternal glucose indicators, as well as maternal age and prepregnancy BMI, were closely associated with the network of AAs and ACs (Fig. 2A).

Figure 2

A: Debiased sparse partial correlation networks of the maternal glucose levels and neonatal metabolites. The nodes are input metabolites, and the edges represent the association measures. B: Maternal glucose trajectory groups associated with neonatal metabolic pathways identified by the pathway enrichment analysis and pathway topology analysis. Group 1: consistent normoglycemia; group 2: mid-to-late gestational hyperglycemia; group 3: early-onset hyperglycemia. C: Maternal glucose trajectory groups associated with neonatal metabolic urea cycle pathway (a), arginine and proline metabolism (b), and β-oxidation of very long-chain fatty acids (c) identified based on the Small Molecule Pathway Database. tRNA, transfer RNA.

Figure 2

A: Debiased sparse partial correlation networks of the maternal glucose levels and neonatal metabolites. The nodes are input metabolites, and the edges represent the association measures. B: Maternal glucose trajectory groups associated with neonatal metabolic pathways identified by the pathway enrichment analysis and pathway topology analysis. Group 1: consistent normoglycemia; group 2: mid-to-late gestational hyperglycemia; group 3: early-onset hyperglycemia. C: Maternal glucose trajectory groups associated with neonatal metabolic urea cycle pathway (a), arginine and proline metabolism (b), and β-oxidation of very long-chain fatty acids (c) identified based on the Small Molecule Pathway Database. tRNA, transfer RNA.

Close modal

Aligned with the observed associations between neonatal metabolites and maternal blood glucose levels, neonatal AA metabolism pathways and fatty acid oxidation pathways were linked to maternal glycemic levels. The pathway enrichment and topology analyses, leveraging the SMPDB database, revealed that arginine and proline metabolism and the urea cycle pathway were associated with mid-to-late gestational hyperglycemia, and the β-oxidation very long-chain fatty acids pathway was associated with early-onset hyperglycemia (Fig. 2B). Crucially, these metabolic pathways intersected with the citric acid cycle (TCA), which is indispensable for mitochondrial aerobic respiration and energy generation. The intricate relationship between the urea cycle and arginine and proline metabolism was also pertinent to the development of interneuronal communication and maintenance of islet function (Fig. 2C).

Association Between Maternal Glucose Trajectories and Neonatal Metabolites in Subgroups

This study delves deeper into the association between maternal glucose trajectories and neonatal metabolites in specific subgroups, particularly stratified by sex and in AGA neonates. Our results, illustrated in Supplementary Fig. 2, indicate that most of the associations between maternal glucose trajectories and neonatal metabolites remained statistically significant in male neonates. Conversely, in female neonates, only a decrease in proline levels was observed in the mid-to-late gestational hyperglycemia group compared with the consistent normoglycemia group. Furthermore, Supplementary Fig. 3 demonstrates that when restricting the analysis to AGA neonates, most of the associations between maternal blood glucose levels and neonatal metabolites persist.

The current study has revealed a significant association between maternal blood glucose levels throughout pregnancy and neonatal circulating metabolites, encompassing a wide array of AAs and ACs. Women with mid-to-late gestational hyperglycemia exhibited associations with metabolic pathways related to arginine and proline metabolism, as well as urea cycle pathways, in contrast to participants with consistent normoglycemia. Those with early-onset hyperglycemia, on the other hand, showed associations with the fatty acid oxidation pathway. All of these pathways are intricately involved in TCA cycles and energy production (24). Interestingly, sex disparities were observed, with maternal glucose trajectories mainly associated with metabolites in male neonates rather than female neonates. That most of these associations remained statistically significant when the analysis was limited to AGA neonates is also noteworthy.

Previous research has established a correlation between maternal hyperglycemia and neonatal metabolomics, specifically in relation to various metabolites, including AAs, lipids, and related metabolites ACs (7,10,11,25). However, the specific neonatal metabolites associated with maternal hyperglycemia have shown some inconsistencies among previous reports, as well as in our own study. These inconsistencies may be attributed to differences in sample types, determination methods, ethnicity, region, and the influence of other factors such as BMI. In addition, the abnormal glucose levels (fasting and postload glucose) and glucose levels at different stages of pregnancy may also contribute to the identified disparities (7).

In this study, we categorized women into three groups based on their blood glucose levels throughout pregnancy (consistent normoglycemia, mid-to-late gestational hyperglycemia, and early-onset hyperglycemia) and explored their associations with neonatal metabolites. We observed decreased levels of alanine, arginine, ornithine, and proline in the urea cycle, as well as in the arginine and proline metabolism pathway in the mid-to-late gestational hyperglycemia group compared with the consistent normoglycemia group. This decrease in arginine level aligns with the findings of Reiss et al. (25), who used neonatal heel blood samples. In contrast, Dani et al. (12) reported elevated levels of alanine and arginine in cord blood samples from neonates born to mothers with GDM compared with those in the control group. Additionally, Peng et al. (13) revealed an increase in argininosuccinic acid levels, a precursor of arginine, and a decrease in uric acid levels in the meconium samples of neonates born to women with GDM, suggesting an alteration in the urea cycle pathway.

Arginine is an essential component in human growth and development. The urea cycle plays a crucial role in converting arginine into urea and reducing nitric oxide production, a key mediator of inflammatory cytotoxicity. Dysfunction in the arginine-nitric oxide pathway may impair β-cell survival and function, as well as interneuronal communication (26,27). Thus, alterations in arginine levels may elucidate the effect of maternal hyperglycemia on offspring metabolic health and neurodevelopment (26,28). Additionally, the urea cycle pathway and arginine and proline metabolism both are associated with the TCA cycle, which regulates glucose homeostasis and lipolysis (29,30), thereby playing an important role in metabolic health.

In addition to the effect of maternal glucose levels on neonatal AA levels, our study also revealed significant associations between maternal glucose levels and neonatal AC levels. Specifically, we observed elevated levels of C0 and C4DC+C5-OH, as well as decreased levels of midchain C6DC and C10:1 in neonates from mothers with elevated blood glucose levels. Although a prior study by Shokry et al. (10) reported a substantial decrease in the C0 level and an overall decrease in ACs in cord blood from women with GDM, another study by Reiss et al. (25) indicated that offspring of women with hyperglycemia showed an elevation of short-chain C0, C2, C3, C5, and C5-OH, which was similar to our findings. Our findings were more consistent with those of Reiss et al. (25), as both studies used neonatal blood samples obtained at the time of the neonatal screening test (7,10,11).

ACs are esters formed by combining acyl groups (fatty acids) with C0. They play a significant role in cellular energy metabolism pathways. By combining with fatty acids, carnitines enable fatty acid oxidation and the transfer of energy through the mitochondrial membrane in the form of ACs (31). Disruption of this process can lead to impaired fat and carbohydrate oxidation as well as dysfunction in the TCA cycle (32). Moreover, various short-chain ACs, such as C3, C4/Ci4, C5, and C5-DC, have been reported to be metabolic byproducts of BCAAs (7,33). Elevated levels of short-chain ACs have been associated with disrupted BCAA metabolism, which is believed to play a crucial role in the development of insulin resistance (32,33).

We also identified positive associations between some long-chain ACs (C16, C16-OH, C16:1, C18, C18:1, and C18:1-OH) in the neonates and maternal glucose indicators during the OGTT in T2. As reported by previous studies, elevated long-chain AC levels have been considered as involved in disturbed lipid metabolism and energy production and identified as biomarkers of insulin resistance, diabetes, and metabolic disease (33).

The study also presented some contradictory findings, such as positive associations between neonatal BCAAs (valine and leucine + isoleucine + hydroxyproline) and maternal FBG levels in T1, as well as a negative association between neonatal BCAAs and maternal postload glucose levels during the OGTT in T2. A previous study by Lowe et al. (7) also indicated that fasting and postload glucose during the OGTT were associated with different metabolites in cord blood. These findings imply that AA and fatty acid metabolism and utilization status vary across different conditions, including fasting and following a glycemic load.

This study found associations between early-onset maternal hyperglycemia and specific ACs in neonatal fatty acid oxidation pathways, whereas certain AAs in neonates decreased in response to mid-to-late gestational hyperglycemia. Previous research also identified disparities in neonatal AAs and ACs associated with different types of maternal hyperglycemia, including GDM, T1DM, and T2DM (25), but did not evaluate gestational glucose levels. Meek et al. (34) reported an increase in diverse AC levels in neonates exposed to maternal hyperglycemia across different trimesters in women with T1DM. Research on diabetic mice models demonstrated decreased fetal AAs derived from maternal hyperglycemia in mid-to-late pregnancy, with the effects changing as the pregnancy progressed and the fetus developed anabolic capacity (35). These findings suggest that the fetal growth and development stages may affect neonatal AA and AC metabolism in response to maternal hyperglycemia. Additionally, the smaller sample size in the early-onset hyperglycemia group, compared with the mid-to-late gestational hyperglycemia group, could have hindered detecting subtle differences, potentially explaining the lack of AA changes in this group. Further studies are needed to fully comprehend the impact of maternal hyperglycemia on fetal metabolism throughout pregnancy.

In the subgroup analysis, we observed associations between maternal glucose trajectories and metabolites in male neonates rather than in female neonates. This finding suggests that the AA and AC metabolism of male neonates may be more susceptible to the influence of maternal hyperglycemia. This aligns with the results of the Japan Environment and Children’s Study, which reported neurodevelopmental delays in boys born to women with GDM, whereas no such delays were observed in girls (36). It is widely recognized that male infants are generally more vulnerable to stressful environments (37), emphasizing the need for increased attention to the development of male offspring in the context of maternal hyperglycemia.

Another subgroup analysis was conducted focusing on AGA neonates. Previous studies have shown that the impact of maternal hyperglycemia on neonatal metabolism is usually accompanied by increased newborn size (38,39). Building upon this understanding, our sensitivity analysis revealed within the AGA subgroup, most of the neonatal AAs and ACs remained significantly associated with maternal glucose levels. These findings suggest that maternal hyperglycemia can have enduring effects on neonatal AAs and ACs beyond just weight considerations. Therefore, it is crucial to prioritize the maintenance of offspring metabolic health during the prevention and management of hyperglycemia in pregnancy, rather than solely focusing on achieving full-term and normal birth weight as the ultimate goal.

The current study possesses several notable strengths. Firstly, the examination of maternal blood glucose levels during different trimesters enables us to compare the neonatal AAs and ACs associated with different glycemic indicators, an aspect that has often been overlooked in prior research. Previous research has primarily focused on glucose levels in the middle and late stages of pregnancy. However, this study revealed that different maternal glucose trajectories have been associated with notable neonatal metabolic changes. Secondly, the use of a prospective study design reduced the selection bias. Lastly, the study's results were further reinforced by using a large sample size.

This study has several limitations. Firstly, the single-center design may limit the generalizability of the results. Secondly, the detection of only two classes of metabolites (11 AAs and 33 ACs) may have limited the identification of additional differential metabolites. Thirdly, neonatal characteristics, such as glucose metabolism, may confound the association between maternal glucose levels and neonatal metabolites. Moreover, neonatal blood samples were collected 3 days after birth; feeding patterns might have influenced these metabolic profiles during this period. Unfortunately, we did not collect data on these characteristics. Future research with more comprehensive neonatal data collected from multiple centers will contribute to a deeper understanding of the association between maternal glucose levels during pregnancy and neonatal AAs and ACs.

Conclusion

In summary, we found that maternal blood glucose levels throughout pregnancy were associated with neonatal metabolites. Specifically, the mid-to-late gestation hyperglycemia trajectory was associated with decreased levels of AAs (alanine, arginine, ornithine, and proline) involved in the arginine and proline metabolism and urea cycle pathway. Early-onset hyperglycemia was associated with elevated levels of C0 and C4DC+C5-OH and decreased levels of C10:1, which is involved in the fatty acid oxidation pathway. However, these associations were mainly observed in male neonates, with minimal significance in female neonates, and were also present when restricted to AGA neonates.

These modifications in neonatal AA and fatty acid pathways in infants born to mothers with elevated blood glucose levels in pregnancy indicate early signs of insulin resistance and may contribute to the “metabolic memory” phenomenon, where unfavorable metabolic health is passed down from one generation to the next (40). This highlights the importance of early intervention for maternal hyperglycemia to improve neonatal metabolic health.

See accompanying article, p. 2107.

This article contains supplementary material online at https://doi.org/10.2337/figshare.25848655.

Acknowledgments. The authors express their gratitude to the participants for their cooperation and extend their appreciation to the medical staff for their diligent efforts in gathering information.

Funding. The study was supported by Beijing Hospitals Authority Youth Programme (QML20231401), National Natural Science Foundation of China (82171671, 82301916), and High-level construction project of public health technical personnel in Beijing Municipal Health System (No. Lingjunrencai-02-02).

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. W.Z. conducted the population study, analyzed and interpreted the data, and drafted the manuscript. X.Yu. and J.Z. participated in data analysis, interpretation, and made revisions to the draft. W.H. and J.H. were involved in follow-up, maternal and neonatal data collection, and result interpretation. X.Ya. and L.Z. took part in data collection and results interpretation. L.L. and S.W. provided assistance with neonatal metabolic data determination and analysis. Y.K. and G.L. designed the study and made critical revisions to the manuscript. All of the authors read and approved the final manuscript. G.L. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Handling Editors. The journal editors responsible for overseeing the review of the manuscript were Elizabeth Selvin and Cuilin Zhang.

1.
Fleming
TP
,
Watkins
AJ
,
Velazquez
MA
, et al
.
Origins of lifetime health around the time of conception: causes and consequences
.
Lancet
2018
;
391
:
1842
1852
2.
Magliano
DJ
,
Boyko
EJ
; IDF Diabetes Atlas 10th Edition Scientific Committee. Chapter 3, Global picture. IDF Diabetes Atlas. 10th edition,
2021
. Accessed 7 January 2024. Available from https://www.ncbi.nlm.nih.gov/books/NBK581940/
3.
Jiang
L
,
Tang
K
,
Magee
LA
, et al
.
A global view of hypertensive disorders and diabetes mellitus during pregnancy
.
Nat Rev Endocrinol
2022
;
18
:
760
775
4.
Kawasaki
M
,
Arata
N
,
Miyazaki
C
, et al
.
Obesity and abnormal glucose tolerance in offspring of diabetic mothers: a systematic review and meta-analysis
.
PLoS One
2018
;
13
:
e0190676
5.
Abokaf
H
,
Shoham-Vardi
I
,
Sergienko
R
,
Landau
D
,
Sheiner
E.
In utero exposure to gestational diabetes mellitus and long term endocrine morbidity of the offspring
.
Diabetes Res Clin Pract
2018
;
144
:
231
235
6.
Zhang
TN
,
Huang
XM
,
Zhao
XY
,
Wang
W
,
Wen
R
,
Gao
SY.
Risks of specific congenital anomalies in offspring of women with diabetes: a systematic review and meta-analysis of population-based studies including over 80 million births
.
PLoS Med
2022
;
19
:
e1003900
7.
Lowe
WL
Jr
,
Bain
JR
,
Nodzenski
M
, et al.;
HAPO Study Cooperative Research Group
.
Maternal BMI and glycemia impact the fetal metabolome
.
Diabetes Care
2017
;
40
:
902
910
8.
Friedes
BD
,
Molloy
E
,
Strickland
T
, et al
.
Neonatal encephalopathy plasma metabolites are associated with neurodevelopmental outcomes
.
Pediatr Res
2022
;
92
:
466
473
9.
Cao
T
,
Zhao
J
,
Hong
X
, et al
.
Cord blood metabolome and BMI trajectory from birth to adolescence: a prospective birth cohort study on early life biomarkers of persistent obesity
.
Metabolites
2021
;
11
:
739
10.
Shokry
E
,
Marchioro
L
,
Uhl
O
, et al
.
Impact of maternal BMI and gestational diabetes mellitus on maternal and cord blood metabolome: results from the PREOBE cohort study
.
Acta Diabetol
2019
;
56
:
421
430
11.
Huhtala
MS
,
Rönnemaa
T
,
Pellonperä
O
,
Tertti
K.
Cord serum metabolome and birth weight in patients with gestational diabetes treated with metformin, insulin, or diet alone
.
BMJ Open Diabetes Res Care
2021
;
9
:
e002022
12.
Dani
C
,
Bresci
C
,
Berti
E
, et al
.
Metabolomic profile of term infants of gestational diabetic mothers
.
J Matern Fetal Neonatal Med
2014
;
27
:
537
542
13.
Peng
S
,
Zhang
J
,
Liu
L
, et al
.
Newborn meconium and urinary metabolome response to maternal gestational diabetes mellitus: a preliminary case-control study
.
J Proteome Res
2015
;
14
:
1799
1809
14.
Wang
C
,
Wei
Y
,
Yang
Y
, et al
.
Evaluation of the value of fasting plasma glucose in the first trimester for the prediction of adverse pregnancy outcomes
.
Diabetes Res Clin Pract
2021
;
174
:
108736
15.
Cauldwell
M
,
Chmielewska
B
,
Kaur
K
, et al
.
Screening for late-onset gestational diabetes: Are there any clinical benefits
?
BJOG
2022
;
129
:
2176
2183
16.
Sun
H
,
Wang
YC
,
Wang
CC
, et al
.
Metabolic profiling of umbilical cord blood in macrosomia
.
Int J Obes
2018
;
42
:
679
685
17.
Möllers
LS
,
Yousuf
EI
,
Hamatschek
C
, et al
.
Metabolic-endocrine disruption due to preterm birth impacts growth, body composition, and neonatal outcome
.
Pediatr Res
2022
;
91
:
1350
1360
18.
Wang
J
,
Zheng
W
,
Wang
Y
, et al
.
Cohort profile: the Beijing Birth Cohort Study (BBCS)
.
Int J Epidemiol
2024
;53yad155
19.
Capital Institute of Pediatrics
;
Coordinating Study Group of Nine Cities on the Physical Growth and Development of Children
.
Growth standard curves of birth weight, length and head circumference of Chinese newborns of different gestation
.
Zhonghua Er Ke Za Zhi
2020
;
58
:
738
746
[in Chinese]
20.
Simmons
D
,
Immanuel
J
,
Hague
WM
, et al.;
TOBOGM Research Group
.
Treatment of gestational diabetes mellitus diagnosed early in pregnancy
.
N Engl J Med
2023
;
388
:
2132
2144
21.
American Diabetes Association
.
Standards of Medical Care in Diabetes—2014
.
Diabetes Care
2014
;
37
(
Suppl. 1
):
S14
S80
22.
Clinical Laboratory Center, Ministry of Health, Interlaboratory Quality Evaluation Committee for Neonatal Genetic Metabolic Disease Screening
.
Expert consensus on tandem mass spectrometry screening techniques for neonatal diseases
.
Chin J Lab Med
2019
;
42
:
9
23.
Basu
S
,
Duren
W
,
Evans
CR
,
Burant
CF
,
Michailidis
G
,
Karnovsky
A.
Sparse network modeling and metscape-based visualization methods for the analysis of large-scale metabolomics data
.
Bioinformatics
2017
;
33
:
1545
1553
24.
Hara
Y
,
Kume
S
,
Kataoka
Y
,
Watanabe
N.
Changes in TCA cycle and TCA cycle-related metabolites in plasma upon citric acid administration in rats
.
Heliyon
2021
;
7
:
e08501
25.
Reiss
JD
,
Chang
AL
,
Mayo
JA
, et al
.
Newborn screen metabolic panels reflect the impact of common disorders of pregnancy
.
Pediatr Res
2022
;
92
:
490
497
26.
Fu
A
,
Alvarez-Perez
JC
,
Avizonis
D
, et al
.
Glucose-dependent partitioning of arginine to the urea cycle protects β-cells from inflammation
.
Nat Metab
2020
;
2
:
432
446
27.
Tripathi
MK
,
Kartawy
M
,
Amal
H.
The role of nitric oxide in brain disorders: Autism spectrum disorder and other psychiatric, neurological, and neurodegenerative disorders
.
Redox Biol
2020
;
34
:
101567
28.
Rodolaki
K
,
Pergialiotis
V
,
Iakovidou
N
,
Boutsikou
T
,
Iliodromiti
Z
,
Kanaka-Gantenbein
C.
The impact of maternal diabetes on the future health and neurodevelopment of the offspring: a review of the evidence
.
Front Endocrinol (Lausanne)
2023
;
14
:
1125628
29.
Arnold
PK
,
Finley
LWS.
Regulation and function of the mammalian tricarboxylic acid cycle
.
J Biol Chem
2023
;
299
:
102838
30.
Hu
S
,
Han
M
,
Rezaei
A
,
Li
D
,
Wu
G
,
Ma
X.
L-arginine modulates glucose and lipid metabolism in obesity and diabetes
.
Curr Protein Pept Sci
2017
;
18
:
599
608
31.
Adeva-Andany
MM
,
Carneiro-Freire
N
,
Seco-Filgueira
M
,
Fernández-Fernández
C
,
Mouriño-Bayolo
D.
Mitochondrial β-oxidation of saturated fatty acids in humans
.
Mitochondrion
2019
;
46
:
73
90
32.
Aguer
C
,
McCoin
CS
,
Knotts
TA
, et al
.
Acylcarnitines: potential implications for skeletal muscle insulin resistance
.
FASEB J
2015
;
29
:
336
345
33.
Dambrova
M
,
Makrecka-Kuka
M
,
Kuka
J
, et al
.
Acylcarnitines: nomenclature, biomarkers, therapeutic potential, drug targets, and clinical trials
.
Pharmacol Rev
2022
;
74
:
506
551
34.
Meek
CL
,
Stewart
ZA
,
Feig
DS
, et al.;
CONCEPTT collaborative group
.
Metabolomic insights into maternal and neonatal complications in pregnancies affected by type 1 diabetes
.
Diabetologia
2023
;
66
:
2101
2116
35.
Perez-Ramirez
CA
,
Nakano
H
,
Law
RC
, et al
.
Atlas of fetal metabolism during mid-to-late gestation and diabetic pregnancy
.
Cell
2024
;
187
:
204
215.e14
36.
Saito
Y
,
Kobayashi
S
,
Ito
S
, et al.;
Japan Environment and Children’s Study group
.
Neurodevelopmental delay up to the age of 4 years in infants born to women with gestational diabetes mellitus: The Japan Environment and Children’s Study
.
J Diabetes Investig
2022
;
13
:
2054
2062
37.
Wells
JC.
Natural selection and sex differences in morbidity and mortality in early life
.
J Theor Biol
2000
;
202
:
65
76
38.
Kadakia
R
,
Talbot
O
,
Kuang
A
, et al.;
HAPO Study Cooperative Research Group
.
Cord blood metabolomics: association with newborn anthropometrics and C-peptide across ancestries
.
J Clin Endocrinol Metab
2019
;
104
:
4459
4472
39.
Zhai
X
,
Liu
J
,
Yu
M
, et al
.
Nontargeted metabolomics reveals the potential mechanism underlying the association between birthweight and metabolic disturbances
.
BMC Pregnancy Childbirth
2023
;
23
:
14
40.
Ott
R
,
Pawlow
X
,
Weiß
A
, et al
.
Intergenerational metabolomic analysis of mothers with a history of gestational diabetes mellitus and their offspring
.
Int J Mol Sci
2020
;
21
:
9647
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at https://www.diabetesjournals.org/journals/pages/license.