Metabolomic Profiling of Gestational Diabetes Mellitus

A 1H-NMR Based Approach

Authors

  • Caylin Chadwick University of Lethbridge, Department of Chemistry & Biochemistry; University of Lethbridge, Department of Neuroscience
  • Mariya Markovina* University of Lethbridge, Department of Neuroscience
  • Hannah Scott University of Lethbridge, Department of Chemistry & Biochemistry; University of Lethbridge, Department of Neuroscience
  • Brenda Leung University of Lethbridge, Faculty of Health Sciences, Public Health Program
  • Gerlinde Metz University of Lethbridge, Department of Neuroscience
  • Tony Montina University of Lethbridge, Department of Chemistry & Biochemistry

Abstract

Introduction: Gestational diabetes mellitus (GDM) has been shown to increase the occurrence of pregnancy complications, including preeclampsia, preterm births, caesarean sections, intrauterine growth retardation, and neonatal intensive care unit admissions. Current tests for diagnosing GDM and pre-diabetic states in pregnancy are time-consuming and often unreliable. Our main objective was to characterize any metabolomic differences in the urine of women with GDM and healthy pregnant controls to identify diagnostic and predictive biomarkers.

Methods: Urine samples from the second trimester of pregnancy from 33 women with GDM and 34 age-, income-, and education-matched healthy controls were obtained from the Alberta Pregnancy Outcomes and Nutrition (APrON) cohort. Proton nuclear magnetic resonance (1H-NMR) data was acquired using a 700 MHz spectrometer and subsequently binned for data analysis. Group separation was evaluated using principle component analysis (PCA) and partial least square discriminant analysis (PLS-DA). Variable importance analysis based on random variable combination (VIAVC) and Mann-Whitney testing were used to determine which metabolites were significantly altered across the comparison groups. Lastly, pathway analysis was carried out to determine which biological pathways were potentially altered by these metabolites.

Results: The analyses revealed group separation between the metabolomic profiles of pregnant women diagnosed with GDM and health pregnant controls. Several significantly altered metabolites, along with some potentially altered biological pathways, were identified across the comparison groups.

Conclusions: These results demonstrate the potential for 1H-NMR and metabolomics to diagnose pregnant woman with, or assess their risk for developing, GDM. Furthermore, the subset of metabolites identified as significantly altered provides potential biomarkers that can be correlated with pregnancy outcomes in order to identify women at high risk of negative pregnancy outcomes.

*Indicates presenter

URSCA 2018

Published

2018-06-21

Issue

Vol. 4 (2018)

Section

Poster Abstracts

License

Authors retain any and all existing copyright to works contributed to these proceedings.

By submitting work to the URSCA Proceedings, contributors grant non-exclusive rights to MacEwan University and MacEwan University Library to make items accessible online and take any necessary steps to preserve them.