Obesogenic diets alter metabolism in mice

Abstract

Obesity and accompanying metabolic disease is negatively correlated with lung health yet the exact mechanisms by which obesity affects the lung are not well characterized. Since obesity is associated with lung diseases as chronic bronchitis and asthma, we designed a series of experiments to measure changes in lung metabolism in mice fed obesogenic diets. Mice were fed either control or high fat/sugar diet (45%kcal fat/17%kcal sucrose), or very high fat diet (60%kcal fat/7% sucrose) for 150 days. We performed untargeted metabolomics by GC-TOFMS and HILIC-QTOFMS and lipidomics by RPLC-QTOFMS to reveal global changes in lung metabolism resulting from obesity and diet composition. From a total of 447 detected metabolites, we found 91 metabolite and lipid species significantly altered in mouse lung tissues upon dietary treatments. Significantly altered metabolites included complex lipids, free fatty acids, energy metabolites, amino acids and adenosine and NAD pathway members. While some metabolites were altered in both obese groups compared to control, others were different between obesogenic diet groups. Furthermore, a comparison of changes between lung, kidney and liver tissues indicated few metabolic changes were shared across organs, suggesting the lung is an independent metabolic organ. These results indicate obesity and diet composition have direct mechanistic effects on composition of the lung metabolome, which may contribute to disease progression by lung-specific pathways.

Fig 1

Distribution and number of structurally identified or annotated metabolites found to be significantly altered by dietary intervention with raw p<0.05 in all dietary comparisons in all tissues measured (A) and in the lung by obesogenic diet compared to control (B). A: In total 91, 122 and 132 metabolites were significantly altered in the lung, kidney and liver respectively. B: In the lung, 80 metabolites were significantly altered in HFS and VHF compared to control diet, with 35 compounds significantly altered between both obese groups and control group. Significance determined using student’s t-test with p-value <0.05 as significant cut off and n = 8.

Fig 2. Metabolic network maps of lung tissue metabolomic and lipidomic results.

Metabolic modules indicated by colored regions using chemical similarity. A) Metabolic differences in lung metabolism of mice fed High Fat Sugar (HFS) diet compared to samples from mice fed with control chow. B) Metabolic differences in lung metabolism of mice fed Very High Fat (VHF) diet compared to samples from mice fed with control chow. Significant metabolites shown in color with direction of indicated by color. Significance determined using student’s t-test with p-value <0.05 considered significant and n = 8.

Fig 3. Heat-map of significant polar and primary metabolites with fold change and significance levels in the lung by diet compared to control.

Metabolites were measured using GC-TOFMS and HILIC-QTOFMS and raw p-values from student’s t-test <0.05 were used for significance levels. P-values for each metabolite are indicated by asterisks. Averages for each compound by group are shown to indicate fold change magnitude and direction with n = 8.

Fig 4. Heat-map of significant glycerophospholipids, fatty acids and sphingolipids with fold change and significance levels in the lung by diet compared to control.

Lipid species were measured using RPLC-QTOFMS and raw p-values from student’s t-test <0.05 were used to determine significance levels. P-value of each lipid species are indicated by asterisks. Mean direction of change is indicated by color and intensity, with red representing increased values compared to control, and blue decreased values compared to control. Averages for each compound by group are shown with n = 8.

Fig 5. Chemical similarity enrichment analysis of annotated metabolites in both obesogenic diets compared to control chow.

Statistical enrichment analysis utilized chemical similarity and ontology mapping to generate metabolite clusters. The y-axis shows most significantly altered clusters on top, x-axis shows XlogP values of metabolite clusters. Cluster colors give the proportion of increased or decreased compounds (red = increased, blue = decreased) in each cluster. Chemical enrichment statistics is calculated by Kolmogorov–Smirnov test. Only enrichment clusters are shown that are significantly different at p < 0.05. Plots and calculations were done using ChemRICH.