Fatty fish intake decreases lipids related to inflammation and insulin signaling--a lipidomics approach

Abstract

Background: The evidence of the multiple beneficial health effects of fish consumption is strong, but physiological mechanisms behind these effects are not completely known. Little information is available on the effects of consumption of different type of fish. The aim of this study was to investigate how fatty fish or lean fish in a diet affect serum lipidomic profiles in subjects with coronary heart disease.

Methodology and principal findings: A pilot study was designed which included altogether 33 subjects with myocardial infarction or unstable ischemic attack in an 8-week parallel controlled intervention. The subjects were randomized to either fatty fish (n = 11), lean fish (n = 12) or control (n = 10) groups. Subjects in the fish groups had 4 fish meals per week and subjects in the control group consumed lean beef, pork and chicken. A fish meal was allowed once a week maximum. Lipidomics analyses were performed using ultra performance liquid chromatography coupled to electrospray ionization mass spectrometry and gas chromatography. Multiple bioactive lipid species, including ceramides, lysophosphatidylcholines and diacylglycerols, decreased significantly in the fatty fish group, whereas in the lean fish group cholesterol esters and specific long-chain triacylglycerols increased significantly (False Discovery Rate q-value <0.05).

Conclusions/significance: The 8-week consumption of fatty fish decreased lipids which are potential mediators of lipid-induced insulin resistance and inflammation, and may be related to the protective effects of fatty fish on the progression of atherosclerotic vascular diseases or insulin resistance.

Trial registration: ClinicalTrials.gov NCT00720655.

Figure 1. The CONSORT flow diagram.

Figure 2. False Discovery Rate q-values based on one-way ANOVA test for each of the 307 identified lipids (represented by black dots) from the lipidomics dataset across the three intervention groups at baseline (horizontal axis) and after the 8 week intervention (vertical axis).

None of the lipids reached q<0.05 at baseline.

Figure 3. Bar charts of fold changes for different lipid classes (calculated as a sum of lipid concentrations within a class) in each group.

*, P<0.05; **, P<0.01; ***, P<0.001. P-values were calculated using paired samples t-test (differences inside the groups before and after intervention).