Metabolomics signatures in type 2 diabetes: a systematic review and integrative analysis

Objective Metabolic signatures have emerged as valuable signaling
molecules in the biochemical process of type 2 diabetes (T2D). To
summarize and identify metabolic biomarkers in T2D, we performed a
systematic review and meta-analysis of the associations between
metabolites and T2D using high-throughput metabolomics techniques. Methods
We searched relevant studies from MEDLINE (PubMed), Embase, Web of
Science, and Cochrane Library as well as Chinese databases (Wanfang, Vip,
and CNKI) inception through 31 December 2018. Meta-analysis was conducted
using STATA 14.0 under random effect. Besides, bioinformatic analysis was
performed to explore molecule mechanism by MetaboAnalyst and R 3.5.2.
Results Finally, 46 articles were included in this review on metabolites
involved amino acids, acylcarnitines, lipids, carbohydrates, organic
acids, and others. Results of meta-analysis in prospective studies
indicated that isoleucine, leucine, valine, tyrosine, phenylalanine,
glutamate, alanine, valerylcarnitine (C5), palmitoylcarnitine (C16),
palmitic acid, and linoleic acid were associated with higher T2D risk.
Conversely, serine, glutamine, and lysophosphatidylcholine C18:2 decreased
risk of T2D. Arginine and glycine increased risk of T2D in the Western
countries subgroup, and betaine was negatively correlated with T2D in
nested case-control subgroup. In addition, slight improvements in T2D
prediction beyond traditional risk factors were observed when adding these
metabolites in predictive analysis. Pathway analysis identified 17
metabolic pathways may alter in the process of T2D and metabolite-related
genes were also enriched in functions and pathways associated with T2D.
Conclusions Several metabolites and metabolic pathways associated with T2D
have been identified, which provide valuable biomarkers and novel targets
for prevention and drug therapy.