Associations between fatty acid oxidation, hepatic mitochondrial function, and plasma acylcarnitine levels in mice

Abstract

Background: The 4-thia fatty acid tetradecylthiopropionic acid (TTP) is known to inhibit mitochondrial β-oxidation, and can be used as chemically induced hepatic steatosis-model in rodents, while 3-thia fatty acid tetradecylthioacetic acid (TTA) stimulates fatty acid oxidation through activation of peroxisome proliferator activated receptor alpha (PPARα). We wished to determine how these two compounds affected in vivo respiration and mitochondrial efficiency, with an additional goal to elucidate whether mitochondrial function is reflected in plasma acylcarnitine levels.

Methods: C57BL/6 mice were divided in 4 groups of 10 mice and fed a control low-fat diet, low-fat diets with 0.4% (w/w) TTP, 0.4% TTA or a combination of these two fatty acids for three weeks (n = 10). At sacrifice, β-oxidation and oxidative phosphorylation (OXPHOS) capacity was analysed in fresh liver samples. Hepatic mitochondria were studied using transmission electron microscopy. Lipid classes were measured in plasma, heart and liver, acylcarnitines were measured in plasma, and gene expression was measured in liver.

Results: The TTP diet resulted in hepatic lipid accumulation, plasma L-carnitine and acetylcarnitine depletion and elevated palmitoylcarnitine and non-esterified fatty acid levels. No significant lipid accumulation was observed in heart. The TTA supplement resulted in enhanced hepatic β-oxidation, accompanied by an increased level of acetylcarnitine and palmitoylcarnitine in plasma. Analysis of mitochondrial respiration showed that TTP reduced oxidative phosphorylation, while TTA increased the maximum respiratory capacity of the electron transport system. Combined treatment with TTP and TTA resulted in a profound stimulation of genes involved in the PPAR-response and L-carnitine metabolism, and partly prevented triacylglycerol accumulation in the liver concomitant with increased peroxisomal β-oxidation and depletion of plasma acetylcarnitines. Despite an increased number of mitochondria in the liver of TTA + TTP fed mice, the OXPHOS capacity was significantly reduced.

Conclusion: This study indicates that fatty acid β-oxidation directly affects mitochondrial respiratory capacity in liver. As plasma acylcarnitines reflected the reduced mitochondrial β-oxidation in TTP-fed mice, they could be useful tools to monitor mitochondrial function. As mitochondrial dysfunction is a major determinant of metabolic disease, this supports their use as plasma markers of cardiovascular risk in humans. Results however indicate that high PPAR activation obscures the interpretation of plasma acylcarnitine levels.

Keywords: Acylcarnitine; Beta-oxidation; Mitochondrial function; Oxidative phosphorylation; Tetradecylthioacetic acid; Tetradecylthiopropionic acid.

Fig. 1

Weight gain and organ weights in male C57/BL6 mice fed 0.4% (w/w) TTP, 0.4% (w/w) TTA, or TTP + TTA for 3 weeks. a Final body weight, (b) index of epididymal white adipose tissue (WAT weight/body weight*100), (c) index of perirenal WAT, (d) index of subcutaneus WAT, (e) weekly feed intake per mouse, and (f) liver index (liver weight/body weight*100). Values are means with standard deviations (n = 8–10). Significant difference from controls was determined using one-way ANOVA with Dunnett’s post hoc test, comparing treatment groups to the control group (**p ≤ 0.01, ***p ≤ 0.001)

Fig. 2

Liver enzyme activity in male C57/BL6 mice fed 0.4% (w/w) TTP, 0.4% (w/w) TTA, or TTP + TTA for 3 weeks. a β-oxidation of palmitoyl-Coenzyme A (CoA) (n = 6), (b) plasma acetylcarnitine (n = 3–6), (c) hepatic Crat gene expression, (d) hepatic acyl-CoA oxidase (ACOX) activity (n = 6), (e) hepatic Acox1 gene expression, (f) hepatic Cpt1 gene expression, (g) hepatic Cpt2 gene expression, (h) hepatic Hmgcs2 gene expression, and (i) hepatic Cd36 gene expression. Values are means with standard deviations (n = 8 unless otherwise stated). Significant difference from controls was determined using one-way ANOVA with Dunnett’s post hoc test (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001)

Fig. 3

Plasma L-carnitine, carnitine percursors and acylcarnitines in male C57/BL6 mice fed 0.4% (w/w) TTP, 0.4% (w/w) TTA, or TTP + TTA for 3 weeks. a L-carnitine, (b) trimethyllysine, (c) g-butyrobetaine, (d) hepatic Bbox1 gene expression (n = 8), (e) palmitoylcarnitine, (f) propionylcarnitine, and (g) iso−/L-valerylcarnitine. Values are means with standard deviations (n = 3–6, unless otherwise stated). Significant difference from controls was determined using one-way ANOVA with Dunnett’s post hoc test (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001)

Fig. 4

Tissue lipid levels in male C57/BL6 mice fed 0.4% (w/w) TTP, 0.4% (w/w) TTA, or TTP + TTA for 3 weeks. a Hepatic triacylglycerol (TAG), (b) hepatic cholesterol, (c) hepatic phospholipids, (d) total hepatic lipids, (e) linear regression of liver TAG and β-oxidation of palmitoyl-coenzyme A (n = 18, black circles - control, light gray circles – TTP, grey circles - TTA), (f) heart TAG (n = 5). Values are means with standard deviations (n = 8, unless otherwise stated). Significant difference from controls was determined using one-way ANOVA with Dunnett’s post hoc test (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001). g Representative images of liver cryo-sections stained with Oil Red O and captured at 40× magnification using a light microscope (n = 3–4). The black line indicates 100 μm

Fig. 5

Transmission electron microscopy (TEM) and respiration in liver samples from male C57/BL6 mice fed 0.4% (w/w) TTP, 0.4% (w/w) TTA, or TTP + TTA for 3 weeks. a Representative images from TEM, 10,000 x magnification and, (b) estimation of number of mitochondria from 3 to 4 TEM images per animal (n = 4). c Oxygen consumption at the different stages in the SUIT protocol (n = 4-6). d Oxygen consumption divided by the average number of mitochondria in the different treatment groups (n = 4–6). e Hepatic gene expression of Ndufs, (f) Sdha, and (g) Ucp2 (n = 8). Values are means with standard deviations. Significant difference from controls was determined using one-way ANOVA with Dunnett’s post hoc test (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001)