Background
Gestational diabetes mellitus (GDM) complicates 1 in 6 Australian births1. Although lifestyle modifications and medications are mainstay of GDM management, it cannot be prevented2. Understanding the preconception microbiome may assist in overcoming this shortcoming.
Aim of Research
To characterise the preconception and pregnancy microbiome of women who developed GDM compared with those who did not.
Methods
Our study was conducted as a substudy of the MothersBabies cohort, with pregnancy complications self-reported during virtual visits3. We collected maternal oral and stool samples at each trimester (T1, T2, T3) and used shotgun metagenomic sequencing to characterise the species profile from these samples at each timepoint, for both groups. Our rigorous approach included differential abundance analysis to filter species, and linear discriminant analysis to evaluate effect size, using microbiomeMarker (1.6.0) within R (v4.3.1) to pinpoint biomarkers (p <0.05 considered significant)4,5.
Results
Demographics: total women recruited (n=42); GDM cohort (n= 21, age, 33.19 years ± 3.97, BMI = 27.38 kg/m2 ± 5.55); and healthy controls (n=21, age, 32.81 years ± 3.89, BMI = 26.05 kg/m2 ± 5.33).
At preconception, GDM oral samples exhibited less enriched species compared to controls (GDM = 2 species, controls = 17 species, p<0.05). Notably, Actinomyces sp000195595 was enriched (p<0.05) and Prevotella jejuni was reduced (p<0.05).
Oral species enrichment was not seen at T1, however, there were more enriched species in the GDM group at T2 (GDM = 14 species, controls = 6 species) and T3 (GDM = 24 species, controls = 20 species, p<0.05). Notably, enriched Phocaeicola vulgatus (T1), Porphyromonas somerae_A (T2) and Veillonella rogosae (T3) (p<0.05).
At preconception, the GDM group stool had higher levels of enriched species at preconception compared to control (GDM = 26 species, controls = 21 species)(p<0.05), including enriched Bacteroides eggerthi (p<0.05) but reduced Bacteroides cellulosilyticus (p<0.05).
GDM stool had less enriched species than controls at T1 (GDM = 14 species, controls = 43 species) and T2 (GDM = 11 species, controls = 35 species)(p<0.05), however, there was no difference at T3 (21 species in GDM and controls)(p<0.05). Remarkably, Phocaeicola dorei was reduced (p<0.05) and Bifidobacterium longum was enriched (p<0.05).
Conclusion
We found enriched microorganisms in GDM mothers which are associated with inflammation, impaired glucose metabolism and SCFA production6-18. Additionally, our results suggest that the oral and stool maternal microbiome during preconception and pregnancy is distinguishable between women with and without GDM. Future functional analysis is warranted to determine the functional and causative role of the microbiome in GDM.