Per-Arnt-Sim Kinase (PASK) Deficiency Increases Cellular Respiration on a Standard Diet and Decreases Liver Triglyceride Accumulation on a Western High-Fat High-Sugar Diet

Abstract

Diabetes and the related disease metabolic syndrome are epidemic in the United States, in part due to a shift in diet and decrease in physical exercise. PAS kinase is a sensory protein kinase associated with many of the phenotypes of these diseases, including hepatic triglyceride accumulation and metabolic dysregulation in male mice placed on a high-fat diet. Herein we provide the first characterization of the effects of western diet (high-fat high-sugar, HFHS) on Per-Arnt-Sim kinase mice (PASK^-/-) and the first characterization of both male and female PASK^-/- mice. Soleus muscle from the PASK^-/- male mice displayed a 2-fold higher oxidative phosphorylation capacity than wild type (WT) on the normal chow diet. PASK^-/- male mice were also resistant to hepatic triglyceride accumulation on the HFHS diet, displaying a 2.7-fold reduction in hepatic triglycerides compared to WT mice on the HFHS diet. These effects on male hepatic triglyceride were further explored through mass spectrometry-based lipidomics. The absence of PAS kinase was found to affect many of the 44 triglycerides analyzed, preventing hepatic triglyceride accumulation in response to the HFHS diet. In contrast, the female mice showed resistance to hepatic triglyceride accumulation on the HFHS diet regardless of genotype, suggesting the effects of PAS kinase may be masked.

Keywords: PAS kinase; PASK; electron transport chain; female; hepatic; high-fat high-sugar diet; lipidomics; lipids; liver; mice; respiration; sexual dimorphism; triglycerides; western diet.

Figure 1

PAS kinase-deficient mice (PASK^−/−) exhibit increased respiration when on a normal chow diet. (A–F) Soleus muscle and liver tissue mitochondrial O₂ consumption determined according to the protocol in Materials and Methods. Results shown are a basal respiration rate (GMS) and oxidative phosphorylation capacity (GMSD). (A,B) three-factor analysis (sex, genotype, and diet) for (A) soleus and (B) liver tissue. (C,E) two-factor analysis (sex and genotype, sex and diet or genotype and diet) for (C) soleus and (E) liver tissue. (D,F) one-factor analysis (sex, genotype, or diet) for (D) soleus and (F) liver tissue. NC is Normal Chow, HFHS is High-Fat High-Sugar diet. For all figures, error bars represent standard error of the mean (SEM). Three-factor ANOVA was performed using JMP Pro14 software. Significant differences were further analyzed by Tukey post-hoc test for three-factor and two-factor comparisons and students t-test for one-factor comparisons. * p < 0.05 are reported.

Figure 2

Quantification of 5 electron transport chain complexes using homogenized soleus muscle reveals no significant differences in the PAS kinase-deficient male (PASK^−/−) mice compared to the WT. (A) Soleus tissue was homogenized and analyzed by western blot using the OxPhosBlue Native WB Antibody Cocktail (ThermoFisher Scientific, Waltham, MA, USA) containing mouse monoclonal NDUFA9 (complex I), SDHA (complex II), UQCRC2 (complex III, core II), COX IV (complex IV, subunit IV) and ATP5A (complex V alpha subunit) antibodies. Protein concentration was determined before loading using the Pierce Coomassie Plus (Bradford) Assay Reagent (ThermoFisher Scientific, Waltham, MA, USA). The same control sample (cntrl) was loaded on each gel for normalization between gels. (B) Plots of one-factor (genotype or diet) analysis of complex I. (C) Representative western blots for each complex are shown. Each biological replicate n > 4 was run in duplicate. Error bars represent SEM. Two-factor (genotype and diet) ANOVA was performed using JMP Pro14 software with students t-test performed on significant differences, * p < 0.05 is shown.

Figure 3

PAS kinase deficiency (PASK^−/−) protects against HFHS-induced accumulation of hepatic triglycerides. (A) Body weight of male and female mice at the start of the diet (week 1, 12-week-old mice) and the end of the diet (week 25). (B) Retroperitoneal fat, (C) gonadal fat and (D) liver weight as a percentage of Body Weight (BW). (E) Hepatic triglyceride quantification for female and male mice using BioVision Triglyceride Quantification kit. (F) Factorial ANOVA analysis (sex, genotype, and diet) results for the data presented in (A–E). NC is Normal Chow diet, HFHS is High-Fat High-Sugar diet. For all figures, error bars represent SEM. Three-factor ANOVA was performed using JMP Pro14 software with Tukey post-hoc test for three-factor and two-factor (sex and genotype, sex and diet, or genotype and diet) comparisons and students t-test for one-factor (sex, genotype, or diet) comparisons. * p < 0.05 are shown in (A–E) for the three-factor analysis.

Figure 4

LC/MS triglyceride analysis of WT and PASK^−/− male mice on NC and HFHS reveal significant changes in individual triglycerides. A heat map LC/MS triglyceride analysis from male WT and PASK^−/− mice placed on a NC and HFHS diet (n = 6 for each of 4 sample groups) *, #, or X, p < 0.05 when analyzed by two-factor ANOVA and Tukey post-hoc test. Two-factor (genotype and diet) and one-factor (genotype or diet) interaction analysis is provided in a table on the right with p-values and false discovery rates (FDR) given. Significant p-values (p < 0.05) are shown in read, with alternative FDR q-value cutoff (p < 0.0599) provided in red for comparison.

Figure 5

Saturated fatty-acid side chains are elevated in WT male mice on the HFHS diet but not PASK^−/− male mice on the HFHS diet. (A) Examples of PAS kinase-dependent protection from triglyceride accumulation. (B) Examples of triglycerides that were not significantly affected by the HFHS diet. (C) One triglyceride that increased in response to the HFHS diet in the PASK^−/− mouse. Bars represent SEM. * p < 0.05 when analyzed by two-factor ANOVA and Tukey post-hoc test. (D–F) Data in Figure 4 was analyzed for saturated fatty acid (SFA), monounsaturated fatty acid (MUFA), and polyunsaturated fatty acid (PUFA) side chains within each triglyceride. Side-chain abundance was calculated using mole percent ratio (percentage of moles of each fatty-acid side chain compared to total mole concentration). p < 0.1 when analyzed by two-factor ANOVA and Tukey post-hoc test. (G–I) One-factor (genotype or diet) analysis of (D–F). p < 0.05 when analyzed by student’s t-test.