Metabolomic Biomarkers in Gestational Diabetes Mellitus: A Review of the Evidence

Gestational diabetes mellitus (GDM) is the fastest growing type of diabetes, affecting between 2 to 38% of pregnancies worldwide, varying considerably depending on diagnostic criteria used and sample population studied. Adverse obstetric outcomes include an increased risk of macrosomia, and higher rates of stillbirth, instrumental delivery, and birth trauma. Metabolomics, which is a platform used to analyse and characterise a large number of metabolites, is increasingly used to explore the pathophysiology of cardiometabolic conditions such as GDM. This review aims to summarise metabolomics studies in GDM (from inception to January 2021) in order to highlight prospective biomarkers for diagnosis, and to better understand the dysfunctional metabolic pathways underlying the condition. We found that the most commonly deranged pathways in GDM include amino acids (glutathione, alanine, valine, and serine), carbohydrates (2-hydroxybutyrate and 1,5-anhydroglucitol), and lipids (phosphatidylcholines and lysophosphatidylcholines). We also highlight the possibility of using certain metabolites as predictive markers for developing GDM, with the use of highly stratified modelling techniques. Limitations for metabolomic research are evaluated, and future directions for the field are suggested to aid in the integration of these findings into clinical practice.     

Previous Gestational Diabetes Increases Atherogenic Dyslipidemia in Subsequent Pregnancy and Postpartum

In a cohort of women with previous gestational diabetes mellitus (GDM), we aimed to ascertain whether women with abnormal glucose tolerance 1‐year postdelivery had a more atherogenic lipid profile during and after pregnancy than those with normal glucose tolerance. A prospective cohort study with longitudinal design between January 2004 and March 2016 was conducted. Three hundred and six (56.8%) of 537 women diagnosed with GDM during the studied period attended a control visit during the first year after delivery. Of these, 112 (36.6%) had prediabetes and 16 (5.2%) had type 2 diabetes mellitus. No significant differences during pregnancy were found in total, low‐density lipoprotein, high‐density lipoprotein (HDL) cholesterol, and triacylglycerol (TAG) concentrations among the three groups. Only HDL cholesterol and TAG levels differed significantly among groups at 2 and 12 months after delivery. Logistic regression analysis revealed pregnancy HDL and glucose metabolism status to be associated with the HDL cholesterol concentration 1‐year postdelivery. Furthermore, the only independent factor associated with TAG levels 1 year after delivery was the gestational TAG concentration. In summary, an overweight multiethnic group of women with prior GDM presented a high incidence of postpartum dysglycemia (41.8%). HDL‐cholesterol and TAG levels, both components of the metabolic syndrome, differed significantly among the three study groups in the glucose‐metabolism status at 2 and 12 months after delivery. Women with previous GDM must be followed up in the postpartum period for early detection and management of lipid and glucose disorders.     

First Trimester Urine and Serum Metabolomics for Prediction of Preeclampsia and Gestational Hypertension: A Prospective Screening Study

Hypertensive disorders of pregnancy, including preeclampsia, are major contributors to maternal morbidity. The goal of this study was to evaluate the potential of metabolomics to predict preeclampsia and gestational hypertension from urine and serum samples in early pregnancy, and elucidate the metabolic changes related to the diseases. Metabolic profiles were obtained by nuclear magnetic resonance spectroscopy of serum and urine samples from 599 women at medium to high risk of preeclampsia (nulliparous or previous preeclampsia/gestational hypertension). Preeclampsia developed in 26 (4.3%) and gestational hypertension in 21 (3.5%) women. Multivariate analyses of the metabolic profiles were performed to establish prediction models for the hypertensive disorders individually and combined. Urinary metabolomic profiles predicted preeclampsia and gestational hypertension at 51.3% and 40% sensitivity, respectively, at 10% false positive rate, with hippurate as the most important metabolite for the prediction. Serum metabolomic profiles predicted preeclampsia and gestational hypertension at 15% and 33% sensitivity, respectively, with increased lipid levels and an atherogenic lipid profile as most important for the prediction. Combining maternal characteristics with the urinary hippurate/creatinine level improved the prediction rates of preeclampsia in a logistic regression model. The study indicates a potential future role of clinical importance for metabolomic analysis of urine in prediction of preeclampsia.     

1H NMR-Based Metabolic Profiling Reveals the Effects of Fluoxetine on Lipid and Amino Acid Metabolism in Astrocytes

Fluoxetine, a selective serotonin reuptake inhibitor (SSRI), is a prescribed and effective antidepressant and generally used for the treatment of depression. Previous studies have revealed that the antidepressant mechanism of fluoxetine was related to astrocytes. However, the therapeutic mechanism underlying its mode of action in astrocytes remains largely unclear. In this study, primary astrocytes were exposed to 10 µM fluoxetine; 24 h post-treatment, a high-resolution proton nuclear magnetic resonance (¹H NMR)-based metabolomic approach coupled with multivariate statistical analysis was used to characterize the metabolic variations of intracellular metabolites. The orthogonal partial least-squares discriminant analysis (OPLS-DA) score plots of the spectra demonstrated that the fluoxetine-treated astrocytes were significantly distinguished from the untreated controls. In total, 17 differential metabolites were identified to discriminate the two groups. These key metabolites were mainly involved in lipids, lipid metabolism-related molecules and amino acids. This is the first study to indicate that fluoxetine may exert antidepressant action by regulating the astrocyte’s lipid and amino acid metabolism. These findings should aid our understanding of the biological mechanisms underlying fluoxetine therapy.     

Parentage-Based Group Composition and Dispersal Pattern Studies of the Yangtze Finless Porpoise Population in Poyang Lake

Social behaviors are poorly known for the critically endangered Yangtze finless porpoise (YFP, Neophocaena asiaeorientalis asiaeorientalis). Here, group composition and dispersal patterns of the YFP population living in the Poyang Lake were studied by parentage-based pedigree analyses using 21 microsatellite loci and a 597 bp segment of the mitochondrial DNA control region. In this study, 21 potential mother-offspring pairs and six potential father-offspring pairs (including two potential parents-offspring pairs) were determined, among which 12 natural mother-offspring groups and a maternal group of three generations were found. No genetically-determined fathers were found associated with their offspring. This study also found that maternally related porpoises at the reproductive state tend to group together. This suggest maternal relationship and reproductive state may be factors for grouping in the YFP population. In natural mother-offspring groups, male offspring were all younger than two years old, which suggest male offspring may leave their mothers at approximately two years of age, or at least they were not in tight association with their mothers as they may have been under two years old. However, female offspring can stay longer with their mothers and can reproduce in the natal group.     

Changes in body composition in lactating adolescent mothers

We assessed changes in body composition in lactating adolescent mothers living under unfavorable socioeconomic and sanitary conditions. A total of 17 healthy primiparous adolescents under 17 years of age attending the Maternity Hospital of the city of La Plata, Argentina, were followed at 4 time points (15 days and 3, 6 and 12 months postpartum) to assess, a) dietary intake, b) practice of physical activity, c) nutritional condition (weight, height, body mass index [BMI] according to age, and body composition by the sum of skinfold measurements), d) characteristics of lactation, and e) growth parameters of the child. The mean age of adolescents was 15.06 +/- 0.66 years (mean menarchal age, 11.59 +/- 0.80 years). All adolescents breastfed up to 12 months postpartum, and maternal milk covered above 80% the baby intake (mean 7.06 +/- 2.54 breast feeds/day). While the daily intake of nutrients by adolescent mothers was constant up to 6 months postpartum, there was a modest decrease in that of proteins, carbohydrates and lipids 15 days and 12 months postpartum. The decrease in energetic intake during the same period was significant (p < 0.05). The practice of physical activity was classified as moderate during the follow-up period. Whereas mean basal percent of fat body mass (FBM) was 29.85 +/- 2.87, and decreased significantly at 6 (27.2% +/- 3.9%; p = 0.02) and 12 (26.1% +/- 3.9%; p = 0.002) months postpartum, changes in lean body mass (LBM) were not significant. In conclusion, lactating adolescents maintained LBM, whereas weight, FBM and BMI decreased markedly from 3 months postpartum.     

NMR-Based Metabolomics in Investigation of the Radiation Induced Changes in Blood Serum of Head and Neck Cancer Patients and Its Correlation with the Tissue Volumes Exposed to the Particulate Doses

In the present study, we analyze the nuclear magnetic resonance (NMR) blood serum metabolic profiles of 106 head and neck squamous cell carcinoma (HNSCC) patients during radio (RT) and concurrent radio-chemotherapy (CHRT). Four different fractionation schemes were compared. The blood samples were collected weekly, from the day before the treatment until the last week of CHRT/RT. The NMR spectra were acquired on A Bruker 400 MHz spectrometer at 310 K and analyzed using multivariate methods. Seven metabolites were found significantly to be altered solely by radiotherapy: N-acetyl-glycoprotein (NAG), N-acetylcysteine, glycerol, glycolate and the lipids at 0.9, 1.3 and 3.2 ppm. The NMR results were correlated with the tissue volumes receiving a particular dose of radiation. The influence of the irradiated volume on the metabolic profile is weak and mainly limited to sparse correlations with the inflammatory markers, creatinine and the lymphocyte count in RT and the branched-chain amino-acids in CHRT. This is probably due to the optimal planning and delivery of radiotherapy improving sparing of the surrounding normal tissues and minimizing the differences between the patients (caused by the tumor location and size).     

Identification of Plasma Metabolomic Profiling for Diagnosis of Esophageal Squamous-Cell Carcinoma Using an UPLC/TOF/MS Platform

Epidemiological studies indicated that esophageal squamous-cell carcinoma (ESCC) is still one of the most common causes of cancer incidence in the world. Searching for valuable markers including circulating endogenous metabolites associated with the risk of esophageal cancer, is extremely important A comparative metabolomics study was performed by using ultraperformance liquid chromatography-electrospray ionization-accurate mass time-of-flight mass spectrometry to analyze 53 pairs of plasma samples from ESCC patients and healthy controls recruited in Huaian, China. The result identified a metabolomic profiling of plasma including 25 upregulated metabolites and five downregulated metabolites, for early diagnosis of ESCC. With a database-based verification protocol, 11 molecules were identified, and six upregulated molecules of interest in ESCC were found to belong to phospholipids as follows: phosphatidylserine, phosphatidic acid, phosphatidyl choline, phosphatidylinositol, phosphatidyl ethanolamine, and sphinganine 1-phosphate. Clinical estimation of metabolic biomarkers through hierarchical cluster analysis in plasma samples from 17 ESCC patients and 29 healthy volunteers indicated that the present metabolite profile could distinguish ESCC patients from healthy individuals. The cluster of aberrant expression of these metabolites in ESCC indicates the critical role of phospholipid metabolism in the oncogenesis of ESCC and suggests its potential ability to assess the risk of ESCC development in addition to currently used risk factors.     

Metabolomic profiling can differentiate between bactericidal effects of free and polymer bound halogen

The effects of two halogenated compounds (sodium hypochlorite and N‐halamine polymers) on the Escherichia coli metabolome were investigated. Changes in the intracellular metabolite pools of bacterial cells treated with different formulations of these compounds were analysed using FTIR (Fourier Transform Infra Red) spectroscopy and LC‐MS (Liquid Chromatography‐Mass Spectroscopy). Principal component analysis was used to generate metabolic profiles of the intracellular metabolites to investigate the effect of sublethal concentrations on the metabolome of treated cells. The effect of treatment with sodium hypochlorite was quantitatively dependent on the exposure time. The resulting metabolic profiles supported our previous hypothesis that the mode of action of some halogenated compounds, such as N‐halamine polymers, can be initiated by release of halogen ions into the aqueous environment, in addition to direct contact between the solid polymer material and the bacterial cells. Moreover, the metabolic profiles were able to differentiate between the effect of free and polymer‐bound halogen. Our metabolomic approach was used for hypothesis generation to distinguish apparently different bactericidal effects of free and polymer‐bound halogen. © 2010 Wiley Periodicals, Inc. J Appl Polym Sci, 2011     

Sex-Specific Differences in Lysine, 3-Hydroxybutyric Acid and Acetic Acid in Offspring Exposed to Maternal and Postnatal High Linoleic Acid Diet,Independent of Diet

Background: Linoleic acid (LA) is an essential polyunsaturated fatty acid (PUFA) that is required for foetal growth and development. Excess intake of LA can be detrimental for metabolic health due to its pro-inflammatory properties; however, the effect of a diet high in LA on offspring metabolites is unknown. In this study, we aimed to determine the role of maternal or postnatal high linoleic acid (HLA) diet on plasma metabolites in adult offspring. Methods: Female Wistar Kyoto (WKY) rats were fed with either low LA (LLA) or HLA diet for 10 weeks prior to conception and during gestation/lactation. Offspring were weaned at postnatal day 25 (PN25), treated with either LLA or HLA diets and sacrificed at PN180. Metabolite analysis was performed in plasma samples using Nuclear Magnetic Resonance. Results: Maternal and postnatal HLA diet did not alter plasma metabolites in male and female adult offspring. There was no specific clustering among different treatment groups as demonstrated by principal component analysis. Interestingly, there was clustering among male and female offspring independent of maternal and postnatal dietary intervention. Lysine was higher in female offspring, while 3-hydroxybutyric acid and acetic acid were significantly higher in male offspring. Conclusion: In summary, maternal or postnatal HLA diet did not alter the plasma metabolites in the adult rat offspring; however, differences in metabolites between male and female offspring occurred independently of dietary intervention.     

Untargeted Metabolomics Reveals Predominant Alterations in Lipid Metabolism Following Light Exposure in Broccoli Sprouts

The consumption of vegetables belonging to the family Brassicaceae (e.g., broccoli and cauliflower) is linked to a reduced incidence of cancer and cardiovascular diseases. The molecular composition of such plants is strongly affected by growing conditions. Here we developed an unbiased metabolomics approach to investigate the effect of light and dark exposure on the metabolome of broccoli sprouts and we applied such an approach to provide a bird’s-eye view of the overall metabolic response after light exposure. Broccoli seeds were germinated and grown hydroponically for five days in total darkness or with a light/dark photoperiod (16 h light/8 h dark cycle). We used an ultra-performance liquid-chromatography system coupled to an ion-mobility, time-of-flight mass spectrometer to profile the large array of metabolites present in the sprouts. Differences at the metabolite level between groups were analyzed using multivariate statistical analyses, including principal component analysis and correlation analysis. Altered metabolites were identified by searching publicly available and in-house databases. Metabolite pathway analyses were used to support the identification of subtle but significant changes among groups of related metabolites that may have gone unnoticed with conventional approaches. Besides the chlorophyll pathway, light exposure activated the biosynthesis and metabolism of sterol lipids, prenol lipids, and polyunsaturated lipids, which are essential for the photosynthetic machinery. Our results also revealed that light exposure increased the levels of polyketides, including flavonoids, and oxylipins, which play essential roles in the plant’s developmental processes and defense mechanism against herbivores. This study highlights the significant contribution of light exposure to the ultimate metabolic phenotype, which might affect the cellular physiology and nutritional value of broccoli sprouts. Furthermore, this study highlights the potential of an unbiased omics approach for the comprehensive study of the metabolism.     

Discovery of a Metabolic Signature Predisposing High Risk Patients with Mild Cognitive Impairment to Converting to Alzheimer’s Disease

Assessing dementia conversion in patients with mild cognitive impairment (MCI) remains challenging owing to pathological heterogeneity. While many MCI patients ultimately proceed to Alzheimer’s disease (AD), a subset of patients remain stable for various times. Our aim was to characterize the plasma metabolites of nineteen MCI patients proceeding to AD (P-MCI) and twenty-nine stable MCI (S-MCI) patients by untargeted metabolomics profiling. Alterations in the plasma metabolites between the P-MCI and S-MCI groups, as well as between the P-MCI and AD groups, were compared over the observation period. With the help of machine learning-based stratification, a 20-metabolite signature panel was identified that was associated with the presence and progression of AD. Furthermore, when the metabolic signature panel was used for classification of the three patient groups, this gave an accuracy of 73.5% using the panel. Moreover, when specifically classifying the P-MCI and S-MCI subjects, a fivefold cross-validation accuracy of 80.3% was obtained using the random forest model. Importantly, indole-3-propionic acid, a bacteria-generated metabolite from tryptophan, was identified as a predictor of AD progression, suggesting a role for gut microbiota in AD pathophysiology. Our study establishes a metabolite panel to assist in the stratification of MCI patients and to predict conversion to AD.     

Cyclic Fatty Acids Found in Frying Oils are Detoxified via Classical Drug Metabolic Pathway but also by β-Oxidation and Eliminated as Conjugates in Rats

Cyclic fatty acid monomers (CFAM) are mainly formed during heat treatments, such as frying, of edible oils. These fatty acids are mixtures of disubstituted five‐ or six‐carbon‐membered ring structures. Some earlier studies have suggested that some of these molecules could be metabolized and detoxified, but so far, neither the detoxification mechanisms nor the metabolite identifications have been elucidated. The objective of the present study was to identify the metabolites resulting from the metabolism and detoxification of CFAM. A deuterium‐labeled CFAM, [9‐²H]‐10‐(6‐propyl‐2‐cyclohexenyl)‐dodecenoic acid, was synthesized and fed to rats for 3 days, along with a standard chow diet while the control group was fed the same chow diet which did not contain any CFAM. Biological fluids (urine, blood) were collected for both groups of rats and analyzed using an untargeted metabolomic approach by ultra‐performance liquid chromatography coupled with mass spectrometry. Two discriminant metabolites and 18 molecules derived from CFAM were identified or tentatively identified in plasma and urine samples, respectively. The structures of the metabolites suggest that CFAM having a six‐carbon‐membered ring could be detoxified by the classical drug metabolic pathway (phase I and phase II reactions), but our study also indicates that these are substrates for the β‐oxidation pathway and eliminated as glucuronide, sulphate, and/or nitrate conjugates. Urine metabolomics investigations without diet effects have indicated a higher excretion of medium‐chain acylcarnitines in the D‐CFAM diet group, which may indicate an incomplete β‐oxidation.     

Metabolite Analysis of Jerusalem Artichoke (Helianthus tuberosus L.) Seedlings in Response to Polyethylene Glycol-Simulated Drought Stress

Jerusalem artichokes are a perennial crop with high drought tolerance and high value as a raw material to produce biofuels, functional feed, and food. However, there are few comprehensive metabolomic studies on Jerusalem artichokes under drought conditions. Methods: Ultra-performance liquid chromatography and tandem mass spectrometry were used to identify differential metabolites in Jerusalem artichoke seedling leaves under polyethylene glycol (PEG) 6000-simulated drought stress at 0, 18, 24, and 36 h. Results: A total of 661 metabolites and 236 differential metabolites were identified at 0 vs. 18, 18 vs. 24, and 24 vs. 36 h. 146 differential metabolites and 56 common were identified and at 0 vs. 18, 24, and 36 h. Kyoto Encyclopedia of Genes and Genomes enrichment identified 236 differential metabolites involved in the biosynthesis of secondary metabolites and amino acids. Metabolites involved in glycolysis, phenolic metabolism, tricarboxylic cycle, glutamate-mediated proline biosynthesis, urea cycle, amino acid metabolism, unsaturated fatty acid biosynthesis, and the met salvage pathway responded to drought stress. Conclusion: A metabolic network in the leaves of Jerusalem artichokes under drought stress is proposed. These results will improve understanding of the metabolite response to drought stress in Jerusalem artichokes and develop a foundation for breeding drought-resistant varieties.     

Brain Metabolic Profile after Intranasal vs. Intraperitoneal Clomipramine Treatment in Rats with Ultrasound Model of Depression

Background: Molecular mechanisms of depression remain unclear. The brain metabolome after antidepressant therapy is poorly understood and had not been performed for different routes of drug administration before the present study. Rats were exposed to chronic ultrasound stress and treated with intranasal and intraperitoneal clomipramine. We then analyzed 28 metabolites in the frontal cortex and hippocampus. Methods: Rats’ behavior was identified in such tests: social interaction, sucrose preference, forced swim, and Morris water maze. Metabolic analysis was performed with liquid chromatography. Results: After ultrasound stress pronounced depressive-like behavior, clomipramine had an equally antidepressant effect after intranasal and intraperitoneal administration on behavior. Ultrasound stress contributed to changes of the metabolomic pathways associated with pathophysiology of depression. Clomipramine affected global metabolome in frontal cortex and hippocampus in a different way that depended on the route of administration. Intranasal route was associated with more significant changes of metabolites composition in the frontal cortex compared to the control and ultrasound groups while the intraperitoneal route corresponded with more profound changes in hippocampal metabolome compared to other groups. Since far metabolic processes in the brain can change in many ways depending on different routes of administration, the antidepressant therapy should also be evaluated from this point of view.     

The Impact of Soy Isoflavones on MCF-7 and MDA-MB-231 Breast Cancer Cells Using a Global Metabolomic Approach

Despite substantial research, the understanding of the chemopreventive mechanisms of soy isoflavones remains challenging. Promising tools, such as metabolomics, can provide now a deeper insight into their biochemical mechanisms. The purpose of this study was to offer a comprehensive assessment of the metabolic alterations induced by genistein, daidzein and a soy seed extract on estrogen responsive (MCF-7) and estrogen non-responsive breast cancer cells (MDA-MB-231), using a global metabolomic approach. The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay showed that all test compounds induced a biphasic effect on MCF-7 cells and only a dose-dependent inhibitory effect on MDA-MB-231 cells. Proton nuclear magnetic resonance (¹H-NMR) profiling of extracellular metabolites and gas chromatography-mass spectrometry (GC-MS) profiling of intracellular metabolites confirmed that all test compounds shared similar metabolic mechanisms. Exposing MCF-7 cells to stimulatory concentrations of isoflavones led to increased intracellular levels of 6-phosphogluconate and ribose 5-phosphate, suggesting a possible upregulation of the pentose phosphate pathway. After exposure to inhibitory doses of isoflavones, a significant decrease in glucose uptake was observed, especially for MCF-7 cells. In MDA-MB-231 cells, the glutamine uptake was significantly restricted, leading to alterations in protein biosynthesis. Understanding the metabolomic alterations of isoflavones represents a step forward in considering soy and soy derivates as functional foods in breast cancer chemoprevention.     

Dietary patterns during pregnancy and risk of gestational diabetes: a prospective cohort study in Western China

Background

Previous research has demonstrated the association between maternal dietary patterns and gestational diabetes (GDM), but evidence in Asian populations remains limited and inconsistent. This study investigated the association between dietary patterns during early pregnancy and the risk of GDM among pregnant women in Western China.

Methods

A prospective cohort study was conducted among 1337 pregnant women in Western China. Dietary intakes were assessed at 15–20 weeks of gestation using a validated food frequency questionnaire. GDM was diagnosed by oral glucose tolerance tests at 24–28 weeks of gestation. Exploratory factor analysis was performed to derive dietary patterns, and logistic regression models were used to examine the association between dietary patterns and GDM.

Results

A total of 199 women (14.9%) developed GDM. Three dietary patterns were identified, namely, a plant-based pattern, a meat-based pattern and a high protein-low starch pattern. Notwithstanding a lack of association between dietary patterns and GDM risk in the whole cohort, there was a significant reduction in GDM risk among overweight women (BMI ≥24 kg/m²); the odds ratio being 0.29 (95% confidence interval 0.09 to 0.94) when comparing the highest versus the lowest score of the high protein-low starch pattern.

Conclusions

There was no significant association between early pregnancy dietary patterns and GDM risk later in pregnancy for women in Western China, but high protein-low starch diet was associated with lower risk for GDM among women who were overweight at pre-pregnancy.