Affordable Access

Publisher Website

Insulin Sensitivity Is Reflected by Characteristic Metabolic Fingerprints - A Fourier Transform Mass Spectrometric Non-Targeted Metabolomics Approach

Public Library of Science
Publication Date
DOI: 10.1371/journal.pone.0013317
  • Research Article
  • Biochemistry
  • Diabetes And Endocrinology/Type 2 Diabetes
  • Public Health And Epidemiology/Screening
  • Medicine


Background A decline in body insulin sensitivity in apparently healthy individuals indicates a high risk to develop type 2 diabetes. Investigating the metabolic fingerprints of individuals with different whole body insulin sensitivity according to the formula of Matsuda, et al. (ISIMatsuda) by a non-targeted metabolomics approach we aimed a) to figure out an unsuspicious and altered metabolic pattern, b) to estimate a threshold related to these changes based on the ISI, and c) to identify the metabolic pathways responsible for the discrimination of the two patterns. Methodology and Principal Findings By applying infusion ion cyclotron resonance Fourier transform mass spectrometry, we analyzed plasma of 46 non-diabetic subjects exhibiting high to low insulin sensitivities. The orthogonal partial least square model revealed a cluster of 28 individuals with alterations in their metabolic fingerprints associated with a decline in insulin sensitivity. This group could be separated from 18 subjects with an unsuspicious metabolite pattern. The orthogonal signal correction score scatter plot suggests a threshold of an ISIMatsuda of 15 for the discrimination of these two groups. Of note, a potential subgroup represented by eight individuals (ISIMatsuda value between 8.5 and 15) was identified in different models. This subgroup may indicate a metabolic transition state, since it is already located within the cluster of individuals with declined insulin sensitivity but the metabolic fingerprints still show some similarities with unaffected individuals (ISI >15). Moreover, the highest number of metabolite intensity differences between unsuspicious and altered metabolic fingerprints was detected in lipid metabolic pathways (arachidonic acid metabolism, metabolism of essential fatty acids and biosynthesis of unsaturated fatty acids), steroid hormone biosyntheses and bile acid metabolism, based on data evaluation using the metabolic annotation interface MassTRIX. Conclusions Our results suggest that altered metabolite patterns that reflect changes in insulin sensitivity respectively the ISIMatsuda are dominated by lipid-related pathways. Furthermore, a metabolic transition state reflected by heterogeneous metabolite fingerprints may precede severe alterations of metabolism. Our findings offer future prospects for novel insights in the pathogenesis of the pre-diabetic phase.

There are no comments yet on this publication. Be the first to share your thoughts.