Identifying molecular effects of diet through systems biology: influence of herring diet on sterol metabolism and protein turnover in mice

Abstract

Background: Changes in lifestyle have resulted in an epidemic development of obesity-related diseases that challenge the healthcare systems worldwide. To develop strategies to tackle this problem the focus is on diet to prevent the development of obesity-associated diseases such as cardiovascular disease (CVD). This will require methods for linking nutrient intake with specific metabolic processes in different tissues.

Methodology/principal finding: Low-density lipoprotein receptor-deficient (Ldlr -/-) mice were fed a high fat/high sugar diet to mimic a westernized diet, being a major reason for development of obesity and atherosclerosis. The diets were supplemented with either beef or herring, and matched in macronutrient contents. Body composition, plasma lipids and aortic lesion areas were measured. Transcriptomes of metabolically important tissues, e.g. liver, muscle and adipose tissue were analyzed by an integrated approach with metabolic networks to directly map the metabolic effects of diet in these different tissues. Our analysis revealed a reduction in sterol metabolism and protein turnover at the transcriptional level in herring-fed mice.

Conclusion: This study shows that an integrated analysis of transcriptome data using metabolic networks resulted in the identification of signature pathways. This could not have been achieved using standard clustering methods. In particular, this systems biology analysis could enrich the information content of biomedical or nutritional data where subtle changes in several tissues together affects body metabolism or disease progression. This could be applied to improve diets for subjects exposed to health risks associated with obesity.

Figure 1. In order to obtain molecular insight into the influence of diet on the metabolism in different tissues, mice were fed with different diets under marcronutrient control.

In the study Ldlr −/− mice were used, as this allowed for evaluation of how diet influences the development of atherosclerosis. The mice were fed with either a beef-based (B) diet or a herring-based (H) diet. The body weights were monitored weekly, and at the end of the study body composition was measured and aortic plaques were detected by en face histology. Furthermore, metabolically important tissues such as liver, muscle and adipose tissue were collected and genome-wide transcription analysis was performed on these samples. After statistical analysis of the data there was performed, in parallel, a standard clustering and dimension reduction analysis with the objective to identify gross patterns within the samples. In the integrated analysis different types of biological network graphs were used. Through this analysis specific metabolic pathways activated in the specific tissues in response to the diet were identified. This information was integrated together with histological data in order to gain new fundamental insight into the molecular effects of diet on whole body metabolism.

Figure 2. Analysis of transcriptome data.

Three mice from each diet group were selected for transcriptome analysis. Liver, muscle and adipose tissue were obtained from these mice, mRNA was extracted from these tissues and the resulting samples were analyzed. A. After normalization Single Value Decomposition (SVD) of the data were performed. This analysis points to a very clear separation of the three tissues analyzed, showing that the tissue effect is larger than the diet effect as expected. The SVD analysis points to good consistency between the samples from the three different mice, giving good statistical power for further analysis of the data. B. Circular mapping plot of Q-values (p-values obtained from a Student t-test and corrected for multiple testing) according to the transcript loci arrangement on the different chromosomes for each of the three tissues. The plot shows the distribution of Q-values in response to diet. The three smaller plots to the right indicate the Q-values for the three different tissues and were overlaid in the figure to the left (more details in Text S1 and Figure S6, for simple boxplot of Q-values see Figure S2). C. For each tissue the reporter Biological Process GO-terms were identified according to the influence of the diets. The reporter GO-terms of cellular component and molecular function category are given in Figure S4. Normalized X-score for all the genes in each GO-term was identified (more details in Text S1). This was done for each of the three tissues in each of the two groups of mice, resulting in a total of 6 categories for each GO-term (3 categories for each GO-terms when consider only tissues factor see Figure S3). The figure illustrates the X-score for each GO-term. The analysis corrects for the size of the group and reporter GO terms with a large number of genes therefore represents a global response, whereas GO terms with few genes represents specific transcriptional changes.

Figure 3. Mapping of metabolic activities in the liver (green and red indicate down- and upregulated based on herring diet, respectively).

A. Overview of genes involved in sterol and lipid biosynthesis that are downregulated in response to herring diet. Besides identification of a key reporter GO terms it is also seen that most genes in the biosynthetic pathway towards sterols and fatty acids are downregulated. B. The downregulation (panel A) is further supported by the identification of several reporter metabolites of the cholesterol and fatty acid biosynthesis. C. Measurements of cholesterol and triacylglyceride in the plasma. It is seen that the levels of both are down in the mice fed with the herring diet, and this effect is seen both after 8 and 16 weeks of feeding. D. For all downregulated genes identified in the reporter GO terms (panel A) there was searched for enrichment of transcription factors and microRNAs. The heat map shows identified transcription factors and microRNAs and their co-occurrence matrix. It is observed that most regulatory effects are due to a single factor. For some of the identified transcription factors the corresponding consensus binding sites were identified, and this resulted in identification of consensus binding sites for Srebf (Srebp), Hnf4, Pparg and Ppara. The Ppar systems are important lipid-activated nuclear receptors involved in lipid and glucose metabolism; Pparg is an important transcription factor in adipocytes and Ppara in hepatocytes. Hnf4a is an important regulator of coordinated nuclear receptor-mediated response to xenobiotics through interaction with Cars/Pxr and through Hnf1 it activates the expression of a large number of liver-specific genes, including those involved in glucose, cholesterol, and fatty acid metabolism. The most frequent binding site for microRNAs is the site mmu.miR.103 which implies its contribution to transcriptional inhibition of hepatic lipid synthesis (see Text S1and Figure S7).

Figure 4. Mapping of metabolic activities in the muscle (green and red indicate down- and upregulated based on herring diet, respectively).

A. Reporter GO terms resulted in the identification of several key processes involved in protein biosynthesis and protein degradation, and genes associated with these processes are downregulated in response to the herring diet. This points a reduced protein turn-over in response to a herring diet. B. Reporter GO terms also show that there is downregulation of genes associated with oxidative stress and muscle contraction in response to a herring diet. This indicates more efficient energy utilization, and the reduced oxidative stress may cause reduced protein misfolding and hence reduced protein turn-over. C. Identification of reporter Reactome processes points to the same overall function and allows for identification of even more specific processes affected by the diet, e.g. start site recognition and binding of activated tRNAs to the ribosome.

Figure 5. Connectivity (topological overlap) matrix for the most differentially expressed genes by the diets in the three tissues.

Based on a two-way ANNOVA, 881 genes were identified to be significantly responding to changes in diet, and these genes were used for the analysis. The rows and columns of the half lower heatmap represent genes in a symmetric fashion. The connectivity strengths were signified by the color intensity, red representing the strongest connection and light yellow representing no connection. The blue color bar delineates the highest interconnected genes module. Within the rectangular frame, the functional terms that show significant enrichment within the blue module is depicted. The colors of the circles indicate the same functional module.