IntroductionOverexpression of the oxygen-responsive transcription factor hypoxia-inducible factor 1α (HIF-1α) correlates with poor prognosis in breast cancer patients. The mouse mammary tumor virus polyoma virus middle T (MMTV-PyMT) mouse is a widely utilized preclinical mouse model that resembles human luminal breast cancer and is highly metastatic. Prior studies in which the PyMT model was used demonstrated that HIF-1α is essential to promoting carcinoma onset and lung metastasis, although no differences in primary tumor end point size were observed. Using a refined model system, we investigated whether HIF-1α is directly implicated in the regulation of tumor-initiating cells (TICs) in breast cancer.MethodsMammary tumor epithelial cells were created from MMTV-PyMT mice harboring conditional alleles of Hif1a, followed by transduction ex vivo with either adenovirus β-galactosidase or adenovirus Cre to generate wild-type (WT) and HIF-1α-null (KO) cells, respectively. The impact of HIF-1α deletion on tumor-initiating potential was investigated using tumorsphere assays, limiting dilution transplantation and gene expression analysis.ResultsEfficient deletion of HIF-1α reduced primary tumor growth and suppressed lung metastases, prolonging survival. Loss of HIF-1α led to reduced expression of markers of the basal lineage (K5/K14) in cells and tumors and of multiple genes involved in the epithelial-to-mesenchymal transition. HIF-1α also enhanced tumorsphere formation at normoxia and hypoxia. Decreased expression of several genes in the Notch pathway as well as Vegf and Prominin-1 (CD133)was observed in response to Hif1a deletion. Immunohistochemistry confirmed that CD133 expression was reduced in KO cells and in tumorspheres. Tumorsphere formation was enhanced in CD133hi versus CD133neg cells sorted from PyMT tumors. Limiting dilution transplantation of WT and KO tumor cells into immunocompetent recipients revealed > 30-fold enrichment of TICs in WT cells.ConclusionThese results demonstrate that HIF-1α plays a key role in promoting primary mammary tumor growth and metastasis, in part through regulation of TICs. HIF-1α regulates expression of several members of the Notch pathway, CD133 and markers of the basal lineage in mammary tumors. Our results suggest that CD133, which has not been profiled extensively in breast cancer, may be a useful marker of TICs in the PyMT mouse model. These data reveal for the first time that HIF-1α directly regulates breast TIC activity in vivo.

Highlights d There are now 140 fully inbred BXD strains available, with high-quality genotypes d More strains, new genotypes, and new models have improved power and precision d We have high power even for traits with low heritability or small effect sizes d A phenome of >100 omics datasets and >7,500 classic phenotypes is freely available

GABA type-A (GABA-A) receptors containing the α2 subunit (GABRA2) are expressed in most brain regions and are critical in modulating inhibitory synaptic function. Genetic variation at the GABRA2 locus has been implicated in epilepsy, affective and psychiatric disorders, alcoholism and drug abuse. Gabra2 expression varies as a function of genotype and is modulated by sequence variants in several brain structures and populations, including F2 crosses originating from C57BL/6J (B6J) and the BXD recombinant inbred family derived from B6J and DBA/2J. Here we demonstrate a global reduction of GABRA2 brain protein and mRNA in the B6J strain relative to other inbred strains, and identify and validate the causal mutation in B6J. The mutation is a single base pair deletion located in an intron adjacent to a splice acceptor site that only occurs in the B6J reference genome. The deletion became fixed in B6J between 1976 and 1991 and is now pervasive in many engineered lines, BXD strains generated after 1991, the Collaborative Cross, and the majority of consomic lines. Repair of the deletion using CRISPR- Cas9 -mediated gene editing on a B6J genetic background completely restored brain levels of GABRA2 protein and mRNA. Comparison of transcript expression in hippocampus, cortex, and striatum between B6J and repaired genotypes revealed alterations in GABA-A receptor subunit expression, especially in striatum. These results suggest that naturally occurring variation in GABRA2 levels between B6J and other substrains or inbred strains may also explain strain differences in anxiety-like or alcohol and drug response traits related to striatal function. Characterization of the B6J private mutation in the Gabra2 gene is of critical importance to molecular genetic studies in neurobiological research because this strain is widely used to generate genetically engineered mice and murine genetic populations, and is the most widely utilized strain for evaluation of anxiety-like, depression-like, pain, epilepsy, and drug response traits that may be partly modulated by GABRA2 function.

Assessing the impact of the social environment on health and disease is challenging. As social effects are in part determined by the genetic makeup of social partners, they can be studied from associations between genotypes of one individual and phenotype of another (social genetic effects, SGE, also called indirect genetic effects). For the first time we quantified the contribution of SGE to more than 100 organismal phenotypes and genome-wide gene expression measured in laboratory mice. We find that genetic variation in cage mates (i.e. SGE) contributes to variation in organismal and molecular measures related to anxiety, wound healing, immune function, and body weight. Social genetic effects explained up to 29% of phenotypic variance, and for several traits their contribution exceeded that of direct genetic effects (effects of an individual’s genotypes on its own phenotype). Importantly, we show that ignoring SGE can severely bias estimates of direct genetic effects (heritability). Thus SGE may be an important source of “missing heritability” in studies of complex traits in human populations. In summary, our study uncovers an important contribution of the social environment to phenotypic variation, sets the basis for using SGE to dissect social effects, and identifies an opportunity to improve studies of direct genetic effects.

The period homolog genes Per1, Per2 and Per3 are important components of the circadian clock system. In addition to their role in maintaining circadian rhythm, these genes have been linked to mood disorders, stress response and vulnerability to addiction and alcoholism. In this study, we combined high-resolution sequence analysis and quantitative trait locus (QTL) mapping of gene expression and behavioral traits to identify Per3 as a compelling candidate for the interaction between circadian rhythm, alcohol and stress response. In the BXD family of mouse strains, sequence variants in Per3 have marked effects on steady-state mRNA and protein levels. As a result, the transcript maps as a cis-acting expression QTL (eQTL). We found that an insertion/deletion (indel) variant in a putative stress response element in the promoter region of Per3 causes local control of transcript abundance. This indel results in differences in protein binding affinities between the two alleles through the Nrf2 transcriptional activator. Variation in Per3 is also associated with downstream differences in the expression of genes involved in circadian rhythm, alcohol, stress response and schizophrenia. We found that the Per3 locus is linked to stress/anxiety traits, and that the basal expression of Per3 is also correlated with several anxiety and addiction-related phenotypes. Treatment with alcohol results in increased expression of Per3 in the hippocampus, and this effect interacts with acute restraint stress. Our data provide strong evidence that variation in the Per3 transcript is causally associated with and also responsive to stress and alcohol.

Summary How lifespan and body weight vary as a function of diet and genetic differences is not well understood. Here we quantify the impact of differences in diet in a genetically diverse family of female mice, split into matched isogenic cohorts fed a low-fat chow (CD, n = 663) or a high-fat diet (HFD, n = 685). We further generate key metabolic data in a parallel cohort sacrificed at four time points. HFD feeding shortens lifespan by 12%— equivalent to a decade in humans. Initial body weight and early weight gains account for longevity differences of ~4–6 days/g. At 500 days, animals on a HFD typically gain 4× as much weight as control, but variation in weight gain does not correlate with lifespan. Classic serum metabolites, often regarded as health biomarkers, are not necessarily strong predictors of longevity. Our data indicate that responses to a high fat diet are substantially modulated by gene-by-environmental interactions, highlighting the importance of genetic variation in making accurate individualized dietary recommendations.

The challenge of precision medicine is to model complex interactions among DNA variants, sets of phenotypes, and complex environmental factors and confounders.We have expanded the BXD family, creating a powerful and extensible test bed for experimental precision medicine and an ideal cohort to study gene-by-environmental interactions.These BXD segregate for over 6 million variants, with a mean minor allele frequency close to 0.5. We have increased the family two-fold to 150 inbred strains, all derived from C57BL/6J and DBA/2J. We have also generated updated and comprehensive genotypes and an unrivaled deep phenome.Approximately 10,000 recombinations have been located, allowing precision of quantitative trait loci mapping of ±2.0 Mb over much of the genome and ±0.5 Mb for Mendelian loci. The BXD phenome includes more than 100 'omics data sets and >7000 quantitative and clinical phenotypes, all of which is publicly available.The BXD family is an enduring, collaborative, and replicable resource to test causal and mechanistic links between genomes and phenomes at many stages and under a wide variety of treatments and interventions. Background The origin of the BXD familyRecombinant inbred strains of mice, and the BXD family in particular, have been used for 45 years to map Mendelian and quantitative trait loci 1-5 . Production was started in 1971 by Benjamin A. Taylor by crossing a female C57BL/6J (B6 or B) and a male DBA/2J (D2 or D)-hence BXD. The first set of 26 (expanded in the 1990s to 35) recombinant inbred strains were intended mainly for mapping Mendelian loci 1,6 , but the family was soon used to map more complex quantitative traits-cancer susceptibility 7-9 , neuroanatomical traits 10-13 , and behavioral differences 14,15 , and pharmacological responses to toxicants and drugs [16][17][18][19] . All of these strains are still available from The Jackson Laboratory (JAX) and carry the strain suffix "/TyJ".Production of BXD43 through BXD102 started in the late 1990s at the University of Tennessee Health Science Center (UTHSC) 20,21 . These new strains were derived from advanced intercross (AI) progeny that had been bred for as many as 14 generations before inbreeding 22 (Figure 1; Supplementary figure 1; Supplementary table 1). AI-derived BXDs incorporate roughly twice as many fixed cross-over events (recombinations) between B and D parental genomes compared to F2-derived BXDs-80 versus 40 [22][23][24][25][26] (Figure 2). This improved both power and precision. statistic (LRS) scores at three cut-offs: 15 (suggestive), 20 (significant) and 25 (highly significant) ( Table 1). The new genotype files increase the number of QTLs detected at suggestive and significant LRS levels, increasing the percentage detected by 17-19% and 24-26% respectively. However, there is a much smaller increase (2% and 10% respectively) at the highly significant cutoff (LRS >25) because large effect QTLs and Mendelian loci are relatively insensitive to low marker density.

No abstract