Hepatic Mitochondrial Dysfunction and Risk of Liver Disease in an Ovine Model of "PCOS Males"

Abstract

First-degree male relatives of polycystic ovary syndrome (PCOS) sufferers can develop metabolic abnormalities evidenced by elevated circulating cholesterol and triglycerides, suggestive of a male PCOS equivalent. Similarly, male sheep overexposed to excess androgens in fetal life develop dyslipidaemia in adolescence. Dyslipidaemia, altered lipid metabolism, and dysfunctional hepatic mitochondria are associated with the development of non-alcoholic liver disease (NAFLD). We therefore dissected hepatic mitochondrial function and lipid metabolism in adolescent prenatally androgenized (PA) males from an ovine model of PCOS. Testosterone was directly administered to male ovine fetuses to create prenatal androgenic overexposure. Liver RNA sequencing and proteomics occurred at 6 months of age. Hepatic lipids, glycogen, ATP, reactive oxygen species (ROS), DNA damage, and collagen were assessed. Adolescent PA males had an increased accumulation of hepatic cholesterol and glycogen, together with perturbed glucose and fatty acid metabolism, mitochondrial dysfunction, with altered mitochondrial transport, decreased oxidative phosphorylation and ATP synthesis, and impaired mitophagy. Mitochondrial dysfunction in PA males was associated with increased hepatic ROS level and signs of early liver fibrosis, with clinical relevance to NAFLD progression. We conclude that excess in utero androgen exposure in male fetuses leads to a PCOS-like metabolic phenotype with dysregulated mitochondrial function and likely lifelong health sequelae.

Keywords: NAFLD; NASH; androgens; hepatic cholesterol; liver fibrosis; male PCOS; mitochondrial dysfunction; oxidative phosphorylation; prenatal programming.

Figure 1

Prenatally androgen-exposed males have a decreased hepatic ATP level that is independent of mitochondria number. (A) Hepatic ATP level in adolescent prenatally androgen-exposed males (PA; n = 14) and control males (C; n = 14). (B) Relative copy number of mitochondrial DNA to nuclear DNA in adolescent prenatally androgen-exposed males (PA; n = 14) and control males (C; n = 14). Differences were analyzed by unpaired, two-tailed t-test. (* p < 0.05). Box-plot whiskers are lowest and highest observed values, and box is the upper and lower quartile, with median represented by line in box.

Figure 2

Prenatally androgen-exposed males have decraesed TCA and OXPHOS cycles. Schematic representation of (A) TCA and (B) OXPHOS cycles in adolescent prenatally androgen-exposed males. In dark blue boxes, significantly decreased genes as compared with control males with q < 0.05. In light blue boxes, trend for decreased gene expression with q < 0.2. In green boxes, genes with no difference in expression between control and PA males. RNAseq data were examined by pairwise comparisons with false discovery rate (q-value) determined by Benjamini–Hochberg method.

Figure 3

Decreased expression of OXPHOS genes is not evident in fetal life. Expression of selected hepatic OXPHOS genes, identified as differentially expressed during adolescence, was examined by using qRT-PCR in control (C; n = 10) and prenatal androgen excess (PA; n = 6) male fetal livers (day 90 of gestation). Differences were analyzed by unpaired, two-tailed t-test. Box-plot whiskers are lowest and highest observed.

Figure 4

Prenatally androgen-exposed male offspring have increased plasma free fatty acids and develop increased hepatic glycogen and total and free cholesterol accumulation. Plasma and liver samples were collected at 6 months postnatal age from control (C; n = 14) and prenatal androgen excess (PA; n = 14) male offspring. Plasma free fatty acids and glucose level in adolescent control and PA males (A,C). Level of hepatic triglycerides, glycogen, total and free cholesterol in adolescent control and PA males (B,D–F). Statistical testing by unpaired, two-tailed Student’s t-test (* p < 0.05). Box-plot whiskers are lowest and highest observed values, and box is the upper and lower quartile, with median represented by line in box.

Figure 5

Prenatally androgen-exposed male offspring have increased hepatic ROS and collagen levels. (A,B) Liver samples from control (C; n = 14) and prenatal androgen excess (PA; n = 14) were collected at 6 months postnatal age from male offspring. ROS and DNA damage were assessed by using commercially available assays. (C) Representative liver sections from control and PA males stained with Sirius Red. Scale bars (white rectangles) represent 200 μm. (D) Sirius Red staining was quantified by using digitized ImageJ analysis. Differences were analyzed by unpaired, two-tailed t-test (* p < 0.05; **** p < 0.0001). Box plot whiskers are lowest and highest observed values; box is the upper and lower quartile, with median represented by line in box.