The relationship between maternal diet and gestational diabetes mellitus: A metabolomic approach

Abstract

Gestational diabetes mellitus (GDM) is a carbohydrate intolerance first recognised during pregnancy and is associated with serious health consequences to both mother and offspring. The best approach to reduce the adverse consequences of GDM is to prevent its development. Despite evidence suggesting a role of maternal diet in GDM development, there is little agreement on the best approach to prevent its occurrence. The observational nature of dietary research has made it difficult to determine mechanisms underlying the associations. Metabolomics, the study of low-molecular weight compounds, is an analytical approach that could be used to explore how maternal diet is associated with GDM. Aim: The aim of this thesis was to investigate the relationship between maternal diet and GDM, by exploring the effect of both diet and disease on the maternal metabolome, in a Southeast Asian multi-ethnic population (N=1,247) from Singapore (The GUSTO cohort). Method: Maternal dietary pattern analysis was conducted from 24-hr dietary recalls, to explore associations between maternal diet and GDM. The effect of GDM on metabolism was determined by metabolite profiling of maternal plasma and hair, using gas chromatography-mass spectrometry. The dietary pattern data and metabolomic data were combined to look for associations between maternal diet and GDM. A seafood-noodle-based dietary pattern, consisting of higher amounts of soup, fish and seafood products, and noodles; and lower amounts of white rice, curry-based gravy, legumes/pulses, and ethnic bread, was found to be significantly associated with a reduced likelihood of GDM (P < 0.01). Metabolite profiling of maternal hair revealed 277 metabolites, none of which were associated with GDM. Of the 219 metabolites detected in maternal plasma, 26 were associated with GDM. One metabolite, 2-hydroxybutyric acid, was significantly higher in GDM cases (P < 0.001, FDR = 7.04E-07) and could discriminate GDM outcome with higher accuracy than the model of clinical risk factors. Metabolic pathway analysis showed that a combination of the significant GDM metabolites may be linked to dysfunction in mitochondrial oxidative phosphorylation. Two plasma metabolites had a significant linear association with the seafood-noodle-based dietary pattern. However, this significance was removed after adjustment for multiple comparisons. Two significant canonical roots were derived from canonical correlation of metabolites and food groups but explained only 1% of the variation in the food groups and 4% of the variation in the metabolites, deeming them unlikely candidates as dietary biomarkers of GDM development. Conclusion: A seafood-noodle-based dietary pattern was associated with GDM in this multi-ethnic Asian cohort. The pattern was distinct from those found previously, emphasising the importance of applying caution when extrapolating dietary findings outside of the studied population. The plasma metabolome could differentiate women that developed GDM from those that did not, linking GDM to a potential dysregulation in mitochondrial function. Although the diet-metabolite associations were not strong enough to deem them clinically useful as dietary biomarkers of GDM, they highlight some interesting associations. Future work should validate these findings in early pregnancy.