Animal models of substance use disorders have been criticized for their limited translation. One important factor behind seeking and taking that has so far been largely overlooked is the availability of alternative non-drug rewards. We recently reported that only about 15% of outbred Wistar rats will choose alcohol over a sweet solution of saccharin. It was also shown using a novel operant model of choice of drugs over social rewards that social interaction consistently attenuates self-administration and incubation of craving for stimulants and opioids. Whether this is also true for alcohol and choice of alcohol over a sweet reward translates to social rewards is currently unknown. We therefore evaluated choice between alcohol and a social reward in different experimental settings in both male and female Wistar rats. We found, in contrast to prior work that employed discrete choice of drugs vs. social reward, that rats almost exclusively prefer alcohol over social interaction, irrespective of the nature of the social partner (cagemate vs. novel rat), the length of interaction, housing conditions and sex. Alcohol choice was reduced when the response requirement for alcohol was increased. However, rats persisted in choosing alcohol, even when the effort required to obtain it was 10-16 times higher (for females and males respectively) than the one for the social reward. Altogether, these results indicate that the social choice model may not generalize to alcohol, pointing to the possibility that specific interactions between alcohol and social reward, not seen when a sweet solution is used as an alternative to the drug, may play a crucial role in alcohol vs. social choice experiments.

Post-traumatic stress disorder (PTSD) is a common, debilitating condition with limited treatment options. Extinction of fear memories through prolonged exposure therapy, the primary evidence-based behavioral treatment for PTSD, has only partial efficacy. In mice, pharmacological inhibition of fatty acid amide hydrolase (FAAH) produces elevated levels of anandamide (AEA) and promotes fear extinction, suggesting that FAAH inhibitors may aid fear extinction-based treatments. A human FAAH 385C-greater thanA substitution encodes an FAAH enzyme with reduced catabolic efficacy. Individuals homozygous for the FAAH 385A allele may therefore offer a genetic model to evaluate the impact of elevations in AEA signaling in humans, helping to inform whether FAAH inhibitors have the potential to facilitate fear extinction therapy for PTSD. To overcome the challenge posed by low frequency of the AA genotype (appr. 5%), we prospectively genotyped 423 individuals to examine the balanced groups of CC, AC, and AA individuals (n = 25/group). Consistent with its loss-of-function nature, the A allele was dose dependently associated with elevated basal AEA levels, facilitated fear extinction, and enhanced the extinction recall. Moreover, the A-allele homozygotes were protected against stress-induced decreases in AEA and negative emotional consequences of stress. In a humanized mouse model, AA homozygous mice were similarly protected against stress-induced decreases in AEA, both in the periphery, and also in the amygdala and prefrontal cortex, brain structures critically involved in fear extinction and regulation of stress responses. Collectively, these data suggest that AEA signaling can temper aspects of the stress response and that FAAH inhibition may aid the treatment for stress-related psychiatric disorders, such as PTSD.