Untargeted metabolomics of saliva in pregnant women with and without gestational diabetes mellitus and healthy non-pregnant women

Yueheng Li, Yang Feng, Zhengyan Yang, Zhi Zhou, Dan Jiang, Jun LuoFrontiers in Cellular and Infection Microbiology 2023 · DOI

TL;DR

Researchers used untargeted LC-MS/MS metabolomics to identify distinct chemical signatures in the saliva of women with gestational diabetes mellitus (GDM). By identifying nine high-confidence differential metabolites—including key amino acids and fatty acids—this study demonstrates that saliva could serve as a viable, non-invasive medium for monitoring metabolic health and periodontal inflammation during pregnancy.

What Problem Does This Paper Address?

Gestational diabetes mellitus (GDM) is a significant pregnancy complication that increases risks for both mother and child, including preeclampsia and future type 2 diabetes. Currently, GDM is diagnosed via the Oral Glucose Tolerance Test (OGTT), which is invasive, time-consuming, and often unpleasant for patients.

Furthermore, pregnancy is frequently associated with gingivitis, and there is a known but complex interplay between systemic metabolic disorders like GDM and oral health. This paper seeks to determine if saliva—a complex fluid that reflects systemic health through paracellular pathways—can provide a "molecular snapshot" of GDM. The goal is to find non-invasive biomarkers that could eventually replace or supplement traditional blood-based testing while also exploring the link between GDM and periodontal inflammation.

Methods

The authors employed untargeted metabolomics, an approach that scans for thousands of small molecules simultaneously rather than looking for a pre-defined set of markers.

  1. Sample Collection: Saliva was collected from 27 women (9 with GDM, 9 healthy pregnant women (HPW), and 9 healthy non-pregnant women (HNPW)).
  2. Analytical Platform: They used Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS), specifically a high-resolution Q Exactive mass spectrometer. This allowed them to detect metabolites in both positive and negative ion modes to maximize coverage.
  3. Bioinformatics: Data was processed using Compound Discoverer 3.1 for peak alignment and identification.
  4. Statistical Analysis: The team used a combination of Principal Component Analysis (PCA) and Partial Least Squares Discriminant Analysis (PLS-DA) to visualize group differences. They applied a Variable Importance in Projection (VIP) score ≥ 1 and a p-value < 0.05 to filter for significant differential metabolites.

Datasets

The study utilized a primary clinical dataset of 27 participants recruited from the Maternal and Child Health Hospital of Changshou District, China.

  • GDM Group: n=9, mean age 28.9.
  • HPW Group: n=9, mean age 27.9.
  • HNPW Group: n=9, mean age 27.7.

Metabolite identification was cross-referenced against the ChemSpider, mzCloud, KEGG, and HMDB databases.

Key Insights and Ideas

  • Saliva as a Systemic Mirror: The study confirmed that salivary metabolic shifts in GDM largely mirror those found in blood and urine in previous studies. This validates saliva as a clinically relevant diagnostic fluid.
  • The "Big Nine" Metabolites: The researchers identified nine metabolites with high confidence levels across both ion modes: L-isoleucine, D-glucose 6-phosphate, docosahexaenoic acid (DHA), arachidonic acid (AA), adenosine, AMP, adenosine 5'-monophosphate, xanthine, and hypoxanthine.
  • Energy and Hypoxia: The presence of xanthine and hypoxanthine suggests shifts in purine metabolism. The authors hypothesize that lower levels of these in the GDM group (compared to healthy pregnant women) might be due to a higher degree of hypoxia in GDM patients, which accelerates the conversion of these precursors into uric acid.
  • Fatty Acid Dysregulation: Significant changes in DHA and AA levels highlight the disruption of polyunsaturated fatty acid (PUFA) metabolism in GDM, which is critical for fetal development and inflammatory response regulation.

Key Results

  • Feature Detection: In positive ion mode, 2,529 molecular features were detected; in negative mode, 983 were detected.
  • Differential Metabolites:
    • GDM vs. HPW: 166 differential metabolites in positive mode (89 up, 77 down).
  • GDM vs. HNPW: 823 differential metabolites in positive mode (402 up, 421 down).
  • Pathway Enrichment: The most significantly impacted pathways included ABC transporters, purine metabolism, amino acid biosynthesis (specifically valine, leucine, and isoleucine), and fatty acid biosynthesis.
  • BCAA Shift: L-isoleucine (a branched-chain amino acid) was notably decreased in GDM saliva compared to healthy pregnant women, suggesting a potential reduction in absorption or an increase in metabolic utilization associated with insulin resistance.

Limitations and Future Work

The authors acknowledge several limitations:

  1. Small Sample Size: With only 9 participants per group, the study is a pilot. Larger cohorts are needed to validate these biomarkers across diverse populations.
  2. Single-Point Sampling: The study provides a cross-sectional view. Longitudinal studies following women throughout their entire pregnancy would better capture the progression of GDM.
  3. Gingivitis Correlation: While the study aimed to link metabolites to gingivitis, the authors note that these associations require further validation in subsequent studies specifically designed to isolate periodontal variables.
  4. Database Gaps: Many detected molecular features could not be identified because they are not yet present in existing metabolite databases, highlighting the need for continued expansion of metabolomic libraries.

Summary updated

Abstract

Untargeted metabolomic analysis of saliva samples from pregnant women with GDM, HPW, and HNPW identified nine differential metabolites with high confidence. The results are similar to findings from previous metabolomics studies of serum and urine samples, which offer the possibility of using saliva for regular noninvasive testing in the population of pregnant women with and without GDM. Meanwhile, the associations between these identified differential metabolites and gingivitis need to be further validated by subsequent studies.
BibTeX
@inproceedings{li2023untargeted,
  author = {Yueheng Li and Yang Feng and Zhengyan Yang and Zhi Zhou and Dan Jiang and Jun Luo},
  title = {{Untargeted metabolomics of saliva in pregnant women with and without gestational diabetes mellitus and healthy non-pregnant women}},
  booktitle = {Frontiers in Cellular and Infection Microbiology},
  year = {2023},
  doi = {10.3389/fcimb.2023.1206462},
}
Figure