Objective Obesity is influenced by genetic and environmental factors. Despite the success of human genome‐wide association studies, the specific genes that confer obesity remain largely unknown. The objective of this study was to use outbred rats to identify the genetic loci underlying obesity and related morphometric and metabolic traits. Methods This study measured obesity‐relevant traits, including body weight, body length, BMI, fasting glucose, and retroperitoneal, epididymal, and parametrial fat pad weight in 3,173 male and female adult N/NIH heterogeneous stock (HS) rats across three institutions, providing data for the largest rat genome‐wide association study to date. Genetic loci were identified using a linear mixed model to account for the complex family relationships of the HS and using covariates to account for differences among the three phenotyping centers. Results This study identified 32 independent loci, several of which contained only a single gene (e.g., Epha5, Nrg1, Klhl14) or obvious candidate genes (e.g., Adcy3, Prlhr). There were strong phenotypic and genetic correlations among obesity‐related traits, and there was extensive pleiotropy at individual loci. Conclusions This study demonstrates the utility of HS rats for investigating the genetics of obesity‐related traits across institutions and identify several candidate genes for future functional testing.

There are a number of traits that are thought to increase susceptibility to addiction, and some of these are modeled in preclinical studies. For example, “sensation-seeking” is predictive of the initial propensity to take drugs; whereas “novelty-seeking” predicts compulsive drug-seeking behavior. In addition, the propensity to attribute incentive salience to reward cues can predict the propensity to approach drug cues, and reinstatement or relapse, even after relatively brief periods of drug exposure. The question addressed here is the extent to which these three ‘vulnerability factors’ are related; that is, predictive of one another. Some relationships have been reported in small samples, but here a large sample of 1,598 outbred male and female heterogeneous stock rats were screened for Pavlovian conditioned approach behavior (to obtain an index of incentive salience attribution; ‘sign-tracking’), and subsequently tested for sensation-seeking and novelty-seeking. Despite the large N there were no significant correlations between these traits, in either males or females. There were, however, novel relationships between multiple measures of incentive salience attribution and, based on these findings, we generated a new metric that captures “incentive value”. Furthermore, there were sex differences on measures of incentive salience attribution and sensation-seeking behavior that were not previously apparent.

Obesity is a global health crisis that is influenced by both genetic and environmental factors.Rodent model organisms can be used to understand the biological and genetic basis of obesity and related morphological traits. A major advantage of model organisms is that they can be studied under uniform environmental conditions, thus reducing the complex role of environment and gene by environment interactions. Furthermore, fat pads and other tissues can be dissected and weighed, so that their role in determining body weight can be precisely defined. Highly recombinant populations allow for genetic fine-mapping of complex traits, greatly reducing the number of plausible candidate genes. We performed the largest rat GWAS ever undertaken, using 3,173 male and female adult N/NIH heterogeneous stock (HS) rats, which were created by mixing 8 inbred strains. We identified 31 independent loci for body weight, body length, body mass index, fat pad weight (retroperitoneal, epididymal, and parametrial), and fasting glucose.We observed strong evidence of pleotropic effects across multiple phenotypes. Three loci contained only a single gene (Epha5, Nrg1 and Klhl14), whereas others were larger and contained many genes. We replicated a locus containing Prlhr, and a second locus containing Adcy3, which we had previously identified in a smaller HS rat study. Finally, by subsampling our dataset, we showed an exponential growth of significant loci as sample size increased towards 3,173. Our results demonstrate the potential for rodent studies to add to our understanding of the molecular genetic factors that contribute to obesity-relevant traits and emphasize the importance of sample size.

Sensitivity to cocaine and its associated stimuli (“cues”) are important factors in the development and maintenance of addiction. Rodent studies suggest that this sensitivity is related, in part, to the propensity to attribute incentive salience to food cues, which, in turn, contributes to the maintenance of cocaine self-administration, and cue-induced relapse of drug-seeking. Whereas each of these traits has established links to drug use, the relatedness between the individual traits themselves has not been well characterized in preclinical models. To this end, the propensity to attribute incentive salience to a food cue was first assessed in two distinct cohorts of 2716 outbred heterogeneous stock rats (HS; formerly N:NIH). We then determined whether each cohort was associated with performance in one of two paradigms (cocaine conditioned cue preference and cocaine contextual conditioning). These measure the unconditioned locomotor effects of cocaine, as well as conditioned approach and the locomotor response to a cocaine-paired floor or context. There was large individual variability and sex differences among all traits, but they were largely independent of one another in both males and females. These findings suggest that these traits may contribute to drug-use via independent underlying neuropsychological processes.

Power analyses are often used to determine the number of animals required for a genome wide association analysis (GWAS). These analyses are typically intended to estimate the sample size needed for at least one locus to exceed a genome-wide significance threshold. A related question that is less commonly considered is the number of significant loci that will be discovered with a given sample size. We used simulations based on a real dataset that consisted of 3,173 male and female adult N/NIH heterogeneous stock (HS) rats to explore the relationship between sample size and the number of significant loci discovered. Our simulations examined the number of loci identified in sub-samples of the full dataset. The sub-sampling analysis was conducted for four traits with low (0.15 0.03), medium (0.31 0.03 and 0.36 0.03) and high (0.46 0.03) SNP-based heritabilities. For each trait, we sub-sampled the data 100 times at different sample sizes (500, 1,000, 1,500, 2,000, and 2,500). We observed an exponential increase in the number of significant loci with larger sample sizes. Our results are consistent with similar observations in human GWAS and imply that future rodent GWAS should use sample sizes that are significantly larger than those needed to obtain a single significant result.

There are a number of traits that are thought to increase susceptibility to addiction, and some of these are modeled in preclinical studies. For example, "sensation-seeking" is predictive of the initial propensity to take drugs; whereas "novelty-seeking" predicts compulsive drug-seeking behavior. In addition, the propensity to attribute incentive salience to reward cues can predict the propensity to approach drug cues, and reinstatement or relapse, even after relatively brief periods of drug exposure. The question addressed here is the extent to which these three 'vulnerability factors' are related; that is, predictive of one another. Some relationships have been reported in small samples, but here a large sample of 1,598 outbred male and female heterogeneous stock rats were screened for Pavlovian conditioned approach behavior (to obtain an index of incentive salience attribution; 'sign-tracking'), and subsequently tested for sensation-seeking and novelty-seeking. Despite the large N there were no significant correlations between these traits, in either males or females. There were, however, novel relationships between multiple measures of incentive salience attribution and, based on these findings, we generated a new metric that captures "incentive value". Furthermore, there were sex differences on measures of incentive salience attribution and sensation-seeking behavior that were not previously apparent.

Sensitivity to cocaine and its associated stimuli ("cues") are important factors in the development and maintenance of addiction. Rodent studies suggest that this sensitivity is related, in part, to the propensity to attribute incentive salience to food cues, which, in turn, contributes to the maintenance of cocaine self-administration, and cue-induced relapse of drug-seeking. Whereas each of these traits has established links to drug use, the relatedness between the individual traits themselves has not been well characterized in preclinical models. To this end, the propensity to attribute incentive salience to a food cue was first assessed in a large population of 2716 outbred heterogeneous stock rats. We then determined whether this was associated with performance in two paradigms (cocaine conditioned cue preference and cocaine contextual conditioning). These measure the unconditioned locomotor effects of cocaine, as well as conditioned approach and the locomotor response to a cocaine-paired floor or context. There was large individual variability and sex differences among all traits, but they were largely independent of one another in both males and females. These findings suggest that these traits may contribute to drug-use via independent underlying neuropsychological processes.

Genome-wide association studies have demonstrated that most traits are highly polygenic; however, translating these polygenic signals into biological insights remains difficult. A lack of satisfactory methods for translating polygenic results across species has precluded the use of model organisms to address this problem. Here we explore the use of polygenic transcriptomic risk scores (PTRS) for translating polygenic results across species. Unlike polygenic risk scores (PRS), which rely on SNPs, PTRS use imputed gene expression for prediction, which allows cross-species translation to orthologous genes. We first developed RatXcan, which is a framework for transcriptome-wide association studies (TWAS) in outbred rats. Leveraging predicted transcriptome and genotype data from UK Biobank, and the genetically trained gene expression models from RatXcan, we scored more than 3,000 rats using human-derived PTRS for height and BMI. Strikingly, we found that these human-derived PTRS significantly predicted analogous traits in rats ($r = 0.08$, $P = 8.57 \times 10^{-6}$; $r = 0.06$, $P = 8.51 \times 10^{-4}$, respectively). The genes included in the PTRS were enriched for biological pathways including skeletal growth and metabolism and were over-represented in tissues including pancreas and brain. This approach facilitates experimental studies in model organisms that examine the polygenic basis of human complex traits and provides an empirical metric by which to evaluate the suitability of specific animal models and identify their shared biological underpinnings.