Triacylglycerol fatty acid composition in diet-induced weight loss in subjects with abnormal glucose metabolism--the GENOBIN study

Abstract

Background: The effect of weight loss on different plasma lipid subclasses at the molecular level is unknown. The aim of this study was to examine whether a diet-induced weight reduction result in changes in the extended plasma lipid profiles (lipidome) in subjects with features of metabolic syndrome in a 33-week intervention.

Methodology/principal findings: Plasma samples of 9 subjects in the weight reduction group and 10 subjects in the control group were analyzed using mass spectrometry based lipidomic and fatty acid analyses. Body weight decreased in the weight reduction group by 7.8+/-2.9% (p<0.01). Most of the serum triacylglycerols and phosphatidylcholines were reduced. The decrease in triacylglycerols affected predominantly the saturated short chain fatty acids. This decrease of saturated short chain fatty acid containing triacylglycerols correlated with the increase of insulin sensitivity. However, levels of several longer chain fatty acids, including arachidonic and docosahexanoic acid, were not affected by weight loss. Levels of other lipids known to be associated with obesity such as sphingolipids and lysophosphatidylcholines were not altered by weight reduction.

Conclusions/significance: Diet-induced weight loss caused significant changes in global lipid profiles in subjects with abnormal glucose metabolism. The observed changes may affect insulin sensitivity and glucose metabolism in these subjects.

Trial registration: ClinicalTrials.gov NCT00621205.

Figure 1. CONSORT chart showing selection of subjects for lipidomic analysis.

Figure 2. Fold changes and t-test p-values between mean concentrations in the weight reduction and control groups at (A) baseline and (B) after the intervention for the 309 peaks, among them180 identified lipid molecular species.

Also shown are fold changes between lipid levels after (33 weeks) vs. before (0 weeks) the intervention and corresponding paired t-test p-values for the (C) control group and (D) weight reduction group. The ‘volcano plot’ arranges lipids along dimensions of biological and statistical significance. The first (horizontal) dimension is the fold change between the two groups (on a log scale, so that up and down regulation appear symmetric), and the second (vertical) axis represents the p-value for a t-test of differences within the group after and before the intervention. The line above which False Discovery Rate q-value <0.05 is marked. No lipid passed the q<0.05 treshold in panels A and C.

Figure 3. Box plots for four selected triacylglycerol (TG) lipids in the weight reduction group before (WR0) and after (WR33) intervention, as well as for the control group before (Ctr0) and after (Ctr33) intervention.

The fatty acid composition of the individual acyls of the respective TG particles are shown in parentheses.

Figure 4. Box plots for selected phospho- and sphingolipids in the weight reduction (WR) and control (Ctr) groups at the beginning (WR0, Ctr0) and at the end (WR33, Ctr33) of the study.

GPCho, phosphatidylcholine; SM, sphingomyelin; Cer, ceramide.

Figure 5. Spearman rank correlation between the fold change of triacylglycerol molecular species after and before the intervention in weight reduction group, and the corresponding (A) total TG fatty acid carbon chain length and (B) degree of TG fatty acid unsaturation.

(C) Changes in esterified fatty acid composition in weight reduction and control groups, as obtained by gas chromatography based fatty acid measurement. p-values were calculated using paired samples t-test (*p<0.05, **p<0.01).

Figure 6. Change in insulin sensitivity index S_I (33 weeks vs. 0 weeks) vs. log-change of triacylglycerol TG(16:0/14:0/14:1) concentration.

The Spearman rank correlation was used to calculate correlation coefficient r. The regression line is based on linear regression.