AUTHOR=Lijing Wang , Sujie Ke , Linxi Wang , Lishan Huang , Liqin Qi , Zhidong Zhan , Kejun Wu , Mengjun Zhang , Xiaoying Liu , Xiaohong Liu , Libin Liu 

TITLE=Altered Caffeine Metabolism Is Associated With Recurrent Hypoglycemia in Type 2 Diabetes Mellitus: A UPLC–MS-Based Untargeted Metabolomics Study

JOURNAL=Frontiers in Endocrinology

VOLUME=Volume 13 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2022.843556

DOI=10.3389/fendo.2022.843556

ISSN=1664-2392

ABSTRACT=Background: Recurrent hypoglycemia (RH) is well known to impair awareness of hypoglycemia and increase the risk of severe hypoglycemia; the underlying mechanism requires further understanding. We aimed to investigate the metabolic characteristic profile for RH in type 2 diabetes mellitus (T2DM) patients and explore the potential metabolic mechanism and prevention strategies.
Methods: We screened 553 community-based T2DM patients. T2DM with RH (DH group, n=40) and T2DM without hypoglycemia (DC group, n=40) were assigned in the case-control study, matched by propensity score matching. Non-targeted, global metabolite profiling was conducted using ultra-high performance liquid chromatography-mass spectrometry. Principal component analysis and supervised projections to latent structures-discriminant analysis  were constructed to evaluate the potential biomarkers. Metabolites with a fold change of >2.0 or <0.5, a t-test p-value <0.05, and variable importance in projection value of >1 were identified as significantly differential metabolites. MetaboAnalyst was performed to analyze the related metabolic pathways.
Results: We identified 12 significantly distinct metabolites as potential biomarkers of RH, which were enriched in five pathways; the caffeine metabolic pathway was the most dominant related one. Caffeine and its main downstream metabolites (theophylline and paraxanthine, all P <0.005) were significantly lower during RH. The combination of these metabolites can serve as a reliable predictor biomarker for RH(area under the curve = 0.88). Regarding lipid metabolism, triglyceride was upregulated (P=0.003)and the O-Acylcarnitine was downregulated (P < 0.001). Besides, RH was accompanied by lower phenylalanine (P=0.003) and higher cortisone (P=0.005) levels. 
Conclusions: RH in T2DM is accompanied by caffeine, lipolysis, phenylalanine, and cortisone metabolism abnormalities. Caffeine might be a reliable candidate biomarker and potential prevention strategy for RH, but further validation studies are needed.