Abstract 1974 Exploring the Link Between Early Life Adversity, Sleep Patterns, and Eating Using Automated High-Throughput Behavioral Recognition Software

Early social isolation (SI) stress can negatively impact sleep habits and weight management throughout life. Strong evidence from studies in both humans and rodents indicates that SI negatively impacts sleep and increases the preference for high-fat and calorically dense foods. These effects are significant risk factors associated with obesity. However, whether SI-induced sleep shifts predict eating habits later in life has not been evaluated. Furthermore, utilizing home-cage monitoring systems and software capable of unbiasedly tracking and recording high-throughput behavioral data has not been applied to explore this association. This study sought to identify sleep-related behavioral biomarkers predictive of eating using home-cage monitoring and automated behavioral recognition software (BRS). We hypothesized that early-life SI would provide quantifiable associations and strong predictive value between shifts in resting patterns and food consumption later in life. We used an animal model of SI stress where rats were housed in pairs (n=12) or isolated (n=12). Experimental housing began in adolescence and continued into young adulthood for 7 weeks. An array of behavioral assessments was monitored weekly during the rat's active phase/dark cycle using the Noldus Phenotyper/Ethovision XT BRS. In this study, we evaluated resting behavior as a proxy for sleep. Resting behaviors were defined as minimal activity while sitting or lying down, sleep, and lack of interest in the surroundings. In adulthood, we provided all the animals with a high-fat Western-like diet (WD) (43% kcal from fat) to evaluate food preferences and desirability. Results correlating adolescent resting frequencies with adult food consumption demonstrate the association between these two factors. During early adolescence (week 2), paired animals had a higher correlation (r = -0.4134, p = 0.1816) of resting with adult food consumption than isolated animals (r = -0.1794, p = 0.5769). Late adolescence (week 7) demonstrated the highest negative correlation between resting and adult food consumption in isolated animals (r = -0.6094, p = 0.0354) compared to the paired animals (r = -0.001337, p = 0.9967). This shift in correlation coefficients from early adolescence to late adolescence supports the hypothesis that resting behaviors impacted by early SI is a strong indicator of food consumption later in life. Interestingly, when the WD was provided during the daytime, SI animals had a strong negative correlation (r = -0.80, p = 0.002) between resting and WD consumption compared to their paired counterparts (r = -0.37, p = 0.24). Still, this effect was lost when WD was given during the night. These data support the hypothesis that SI-induced disruptions in resting behavior during adolescence can be a predictive factor for food consumption and WD preference in adulthood. In summary, Noldus automated BRS enabled us to shed light on the relationship between early-life social experiences, sleep, and dietary choices. This study opened avenues for future research on early interventions to promote healthier outcomes for those who experience early life adversities. This study was partly supported by the NIH (DK124727, GM060507, and MD006988) and the Loma Linda University School of Medicine GRASP Seed Funds to JDF.