Long-term metabolic effects of stress and antidepressants : a novel paradigm of antidepressant-induced weight gain in the post-stress acclimation period

Abstract

Despite the frequent use of antidepressants and the high incidence of obesity in Australia and many other developed countries, the relationship between major depressive disorder, obesity and antidepressant use remains complex. The association between antidepressant use and body weight gain has been widely reported. However, the pathophysiological mechanisms of this association are still poorly understood. In this thesis, an animal paradigm that addresses the "paradoxical" weight loss of rodents during antidepressant treatment was developed. The paradigm referred here as the SAD model and consisting of short-term exposure to recurrent restraint stress (RRS) and antidepressant treatment for two weeks, followed by long-term high-fat diet was evaluated in male Sprague-Dawley rats for 296 days. The phenotypic characteristics of the SAD model were compared to the diet-induced obesity (DIO) model, a model of obesity solely induced by high-fat diet. Immediately after the RRS period, groups of animals that underwent RRS had significantly lower body weight in comparison to the DIO group. The rate of weight gain during the catch up growth phase was significantly greater in the SAD model than in the DIO model. During the post-restraint period, the SAD model of fluoxetine-treated group became heavier than the DIO model, however this happened only in the animals in the upper half of body weight gain, i.e. the obesity prone group. Body weight gain in the SAD model in comparison to the DIO model was referred as "SAD-induced weight gain". The SAD model of fluoxetine-treated group had significantly greater body length and enhanced morphological features of the trabecular and cortical bones in comparison to the DIO model, these findings suggest that the SAD-induced weight gain was accompanied by growth via the growth hormone (GH)/Insulin-like growth factor-I (IGF-I) somatotropic axis. However, pituitary GH and plasma IGF-I levels were not different at the end point of the experiment. Igf-1 mRNA expression in the liver was significantly higher in groups that underwent RRS, which may have occurred as a compensatory mechanism for the decreased IGF-I levels during the RRS period. We also investigated whether "SAD-induced weight gain" was accompanied by a metabolic/lipid dysregulations. Plasma leptin, cholesterol, vanilmandelic acid, and free fatty acid levels were not significantly different between groups. Thus, SAD-induced weight gain is not accompanied by further increases in fat mass or metabolic dysregulations. Our results showed that the RRS and the fluoxetine treatment result in decreased plasma triglyceride levels when compared to the DIO model. Thus, the "protective" effect of fluoxetine in metabolic dysregulations may be sustained long-term. Our results suggest that individuals who are already prone to obesity may have increased risk of weight gain with antidepressant treatment. Antidepressant-induced weight gain is more likely to be associated with growth during the post-stress acclimation period, rather than with increases in fat mass and metabolic dysregulations. However, mechanisms behind the antidepressant-induced weight gain require further investigation. This thesis has contributed towards a better understanding of long-term effects of environmental factors such as stress and diet, and antidepressants in post-stress acclimation weight gain.