Aims/hypothesis Uncoupling protein (UCP) 3 is a mitochondrial inner membrane protein expressed predominantly in glycolytic skeletal muscles. Its role in vivo remains poorly understood. The aim of the present work was to produce a mouse model with moderate overproduction and proper fibre-type distribution of UCP3. Methods Transgenic mice were created with a 16 kb region encompassing the human UCP3 gene. Mitochondrial uncoupling was investigated on permeabilised muscle fibres. Changes in body weight, adiposity and glucose or insulin tolerance were assessed in mice fed chow and highfat diets. Indirect calorimetry was used to determine wholebody energy expenditure and substrate utilisation.Results Transgenic mice showed a twofold increase in UCP3 protein levels specifically in glycolytic muscles. Mitochondrial respiration revealed an increase of uncoupling in glycolytic but not in oxidative muscles. Transgenic mice gained less weight than wild-type littermates due to lower adipose tissue accretion when fed a high-fat diet. Animals showed a sexual dimorphism in metabolic responses. Female transgenic mice were more glucose-sensitive than wild-type animals, while male transgenic mice with high body weights had impaired glucose and insulin tolerance. Measurements of RQs in mice fed chow and high-fat diets suggested an impairment of metabolic flexibility in transgenic male mice. Conclusions/interpretation Our data show that physiological overproduction of UCP3 in glycolytic muscles results in mitochondrial uncoupling, resistance to high-fat dietinduced obesity and sex specificity regarding insulin sensitivity and whole-body substrate utilisation.

The mu opioid receptor (MOR) plays a critical role in modulating social behavior in humans and animals. Accordingly, MOR null mice display severe alterations in their social repertoire as well as multiple other behavioral deficits, recapitulating core and secondary symptoms of autism spectrum disorder (ASD). Such behavioral profile suggests that MOR dysfunction, and beyond this, altered reward processes may contribute to ASD etiopathology. Interestingly, the only treatments that proved efficacy in relieving core symptoms of ASD, early behavioral intervention programs, rely principally on positive reinforcement to ameliorate behavior. The neurobiological underpinnings of their beneficial effects, however, remain poorly understood. Here we back-translated applied behavior analysis (ABA)-based behavioral interventions to mice lacking the MOR (Oprm1−/−), as a model of autism with blunted reward processing. By associating a positive reinforcement, palatable food reward, to daily encounter with a wild-type congener, we were able to rescue durably social interaction and preference in Oprm1−/− mice. Along with behavioral improvements, the expression of marker genes of neuronal activity and plasticity as well as genes of the oxytocin/vasopressin system were remarkably normalized in the reward/social circuitry. Our study provides further evidence for a critical involvement of reward processes in driving social behavior and opens new perspectives regarding therapeutic intervention in ASD.

No abstract