Differential proteomic analysis of mouse cerebrums with high-fat diet (HFD)-induced hyperlipidemia

Abstract

Hyperlipidemia is a chronic disease characterized by elevated blood cholesterol and triglycerides and there is accumulated evidence that the disease might affect brain functions. Here we report on a proteomic analysis of the brain proteins in hyperlipidemic mice. Hyperlipidemia was successfully induced in mice by a 20 week high-fat diet (HFD) feeding (model group). A control group with a normal diet and a treatment group with HFD-fed mice treated with a lipid-lowering drug simvastatin (SIM) were established accordingly. The proteins were extracted from the left and right cerebrum hemispheres of the mice in the three groups and subjected to shotgun proteomic analysis. A total of 4,422 proteins were detected in at least half of the samples, among which 324 proteins showed significant difference (fold change >1.5 or <0.67, p < 0.05) in at least one of the four types of comparisons (left cerebrum hemispheres of the model group versus the control group, right cerebrums of model versus control, left cerebrums of SIM versus model, right cerebrums of SIM versus model). Biological process analysis revealed many of these proteins were enriched in the processes correlated with lipid metabolism, neurological disorders, synaptic events and nervous system development. For the first time, it has been reported that some of the proteins have been altered in the brain under the conditions of HFD feeding, obesity or hyperlipidemia. Further, 22 brain processes-related proteins showed different expression in the two cerebrum hemispheres, suggesting changes of the brain proteins caused by hyperlipidemia might also be asymmetric. We hope this work will provide useful information to understand the effects of HFD and hyperlipidemia on brain proteins.

Keywords: Brain; Cerebrum; High-fat diet; Hyperlipidemia; Lipid; Mouse; Proteomics.

Figure 1. Body weight and organ index.

(A) Changes of body weight over time. (B) Final body weight. (C) Percent abdominal fat, calculated by the formula [(abdominal fat weight/ body weight) ×100%]. (D) Liver index, calculated by [(liver weight/body weight) ×100%]. All the values were mean ± standard deviation (SD) (n = 6 in each group). *p < 0.05 comparing the model group with the control group; #p < 0.05, comparing the simvastatin-treated (SIM) group with the model group. The exact p values were given in the figure.

Figure 2. Biochemical analysis.

(A) Serum low-density lipoprotein (LDL). (B) Serum triglyceride (TG). (C) Serum total cholesterol (TC). (D) Serum high-density lipoprotein (HDL). (E) Serum alanine aminotransferase (ALT). (F) Serum aspartate aminotransferase (AST). (G) Serum γ-glutamyltransferase (γ-GGT). (H) Liver superoxide dismutase (SOD). (I) Liver malondialdehyde (MDA). All the values were mean ± SD (n = 6 in each group). *p < 0.05 comparing the model group with the control group; #p < 0.05, comparing the SIM group with the model group. The exact p values were given in the figure.

Figure 3. Categorization and differential analysis of the 4,422 proteins.

(A) Categorization by cellular components (light red) and molecular functions (light blue), using DAVID and PANTHER, respectively. The number of the proteins in each category was given at the right. (B) Volcano plots showing the log₂ (fold change) versus −log₁₀ (p value) of each protein, for (i) the left cerebrums of model group versus control group, (ii) the right cerebrums of model group versus control group, (iii) the left cerebrums of SIM group versus model group, and (iv) the right cerebrums of SIM group versus model group. The proteins with fold change >1.5 and p < 0.05 were indicated in green dots (up-regulated) and those with fold change <0.67 and p < 0.05 were indicated in orange dots (down-regulated).

Figure 4. Principal component analysis of the 324 dysregulated proteins.

The mean values of the technical replicates (n = 3) were used for each biological sample. The “L” represented for the left cerebrums, “R” for the right cerebrums.

Figure 5. Validation of proteomic results by Western blotting.

The levels of hydroxymethylglutaryl-CoA synthase, mitochondrial (Hmgcs2), epidermal growth factor receptor (Egfr), neuropilin-2 (Nrp2) and activity-regulated cytoskeleton-associated protein (Arc) were examined in (A) the left and (B) the right cerebrum hemispheres of the mice (n = 3 in each group). The blots are given in (i) and the densitometric values are given in (ii) to (v), expressed as mean ± standard error of the mean normalized to GAPDH (n = 3 in each group). *p < 0.05 comparing the model group with the control group; #p < 0.05, comparing the SIM group with the model group. The exact p values were given in the figure.

Figure 6. Heat maps of the lipid and brain processes-related proteins in the four types of comparisons.

(A) Lipid processes-related proteins. (B) Brain processes-related proteins. Each column represents a type of comparison, and each row represents one protein (gene name used). Red indicates up-regulation and blue indicates down-regulation, and color density indicates the fold changes in log ₂ scale. The color cells marked with an asterisk (*) indicate the changes were statistically significant (fold change >1.5 or <0.67 and p < 0.05). The lines marked with “ ⧫” indicate these proteins were discussed in detail in Discussion. The solid-line boxes indicate the proteins that showed the same dysregulation in the two hemispheres when the model groups were compared with the control groups, i.e., M-L/C-L and M-R/C-R. The dashed-line boxes indicated those that showed different dysregulation in the two hemispheres. The major biological processes based on UniProtKB records and publications are summarized on the right side for each protein. The details of these statistical results (e.g., the exact fold changes and p values) were provided in Table S5-1.