Investigating the role of thiosulfate sulfurtransferase in adipose tissue dysfunction in obesity
Mc Fadden, Clare Elizabeth
Obesity is associated with dysfunction of adipose tissue due to oxidative stress and inflammation, leading to insulin resistance. Thiosulfate sulfurtransferase (Tst) was previously identified as an adipose-expressed anti-diabetic gene that protects against diet-induced metabolic impairment when upregulated in adipose tissue of mice. TST is a mitochondrial enzyme involved in the metabolism of cyanide, reactive oxygen species (ROS) and endogenous hydrogen sulfide (H2S). This thesis tested the hypothesis that TST maintains metabolic health in the face of dietary obesity. To do this, I investigated the adipose-tissue phenotypes and metabolic consequences of Tst gene deletion (Tst–/– mice) and of adipose tissue-specific overexpression of human TST (Ad-hTST mice) after exposure to high fat diet (HFD). After 20 weeks of HFD, Tst–/– mice exhibited impaired glucose tolerance despite unchanged adipose tissue inflammatory cell infiltration, protein carbonylation and unfolded protein response activation. However, levels of mRNA encoding mitochondrial antioxidant enzymes including superoxide dismutase 2 and peroxiredoxin 3 were lower in Tst–/– mice on HFD. Unexpectedly, chow-fed Tst-/- mice had lower body weight and fat mass than wild-type controls highlighting a potential effect of Tst on fat accumulation with age. A new mouse model with high expression of human TST genetically targeted to adipose tissue (Ad-hTST) was developed using the LoxP / Cre recombinase expression system, with a parent line expressing Cre under the control of the adiponectin promoter to confer adipose specificity. The Ad-hTST mice were found to gain a similar amount of weight and fat mass to control mice when exposed to 6 weeks of HFD. However, Ad-hTST mice had impaired glucose tolerance with no change in inflammatory cell infiltration, mRNA levels of antioxidant enzymes or unfolded protein response genes. Thus, unexpectedly, overexpression of human TST in adipose tissue of mice results in a detrimental metabolic phenotype. In vivo and in vitro experiments were conducted to test the hypothesis that TST protects against ROS accumulation. Paraquat was tested as an inducer of oxidative stress in vivo in wild-type, Tst-/- and Tst+/- mice. At the doses used (25mg/kg and under), mice became unwell and lost weight, with no increase in markers of oxidative stress in adipose or lung. The production of mitochondrial ROS in response to exogenous hydrogen peroxide (H2O2) exposure was increased in primary adipocytes from Tst-/- mice in vitro. However, primary hepatocytes showed reduced mitochondrial ROS production in response to H2O2 exposure. ROS production in hepatocytes was unaffected by pre-incubation with a H2S donor, an inhibitor of H2S-producing enzyme CSE or N-acetyl-cysteine, an antioxidant. TST may therefore influence mitochondrial ROS production differently in cell types such as adipocytes and hepatocytes. Disposal of exogenous H2O2 was unchanged in primary adipocytes from Tst-/- and Ad-hTST mice, and this was not affected by pre-incubation with sodium thiosulfate, a TST substrate. Metabolic changes in response to HFD may be influenced by alteration in TST expression, however the current data suggest it is unlikely to occur through the prevention of excessive local ROS accumulation in adipose tissue. Mice lacking the Tst gene globally and mice with adipose-specific overexpression of the human TST gene have a similarly impaired metabolic response to HFD. The phenotype of adipose-specific human TST-overexpressing mice does not recapitulate the protective metabolic phenotype produced by overexpression of the endogenous mouse Tst gene. In conclusion, TST may influence adipose tissue due to its role in the oxidation of H2S, however, by the current means, it does not appear to substantially impact the response of this tissue to oxidative stress.