Symbiotic microbial communities can affect the health and fitness of the host. Recent studies in humans and mouse models have demonstrated previously unknown roles for the gut microbiota in mammalian digestion, immunity, behavior, and development. These findings suggest that the compositional variation in gut microbiotas may play a fundamental role in host biology. However, the determinants and implications of the gut microbiota in natural populations of mammals are less well understood. My dissertation aims to identify genetic and environmental factors shaping variation in the gut microbiota and to understand the role of gut microbiota in mammalian biology. First, I reviewed studies over the last decade that show links between variation in symbiotic microbiota and variation in host phenotype in natural populations of mammals to motivate the work. Next, I found obesity-associated gut microbial composition tend to be more prevalent in humans living in colder environments consistent with the ecological pattern known as Bergmann’s rule. The results suggest a link between the gut microbiota and climatic adaptation. To further investigate whether the pattern observed in humans is general, I conducted a series of studies where I combined field observations and laboratory experiments to investigate the determinants of the gut microbiota and their roles in environmental adaptation using natural populations of house mice (Mus musculus) as a model system. I have characterized the gut microbiota of wild house mice using 16S amplicon sequencing within individuals’ gastrointestinal tract and between individuals’ cecum across two altitudinal and three latitudinal transects across the Americas. Microbiotas of the lower gastrointestinal tract showed greater individual differences compared to the upper gastrointestinal tract. The individual differences in the cecal microbiota were explained by differences in host genetic distance independent of geographic distance. Several gene-bacteria associations were identified from a microbiome genome-wide association study (mGWAS) using exome sequences. The results from the altitudinal and latitudinal transects suggest that differences in partial pressure of oxygen and host body mass may cause changes in gut microbiota. Beneficial functions of the gut microbiota regulating blood pressure at high altitudes and facilitating energy harvest at high latitudes were proposed. Many of the findings here have been replicated in other mammalian systems, including humans. Together, the results suggest general mechanisms governing the assembly and function of the mammalian gut microbiota.

Understanding how phenotypic plasticity influences adaptive evolution is a major goal in evolutionary biology. In my dissertation, I examined the contributions of phenotypic plasticity to patterns associated with environmental adaptation across introduced populations of house mice (Mus musculus domesticus) collected across North and South America.In Chapter 1, I documented patterns of divergence and plasticity for morphological traits in lab-reared descendants of mice adapted to cold (New York) and warm (Brazil) environments. In particular, under a common lab environment, I observed that mice from New York were larger and had shorter tails and ears than mice from Brazil. These patterns conform to two well-documented ecogeographic rules (Bergmann’s Rule and Allen’s Rule) and likely reflect thermoregulatory adaptations. Under cold laboratory conditions, I observed plasticity for tail and ear length but not for overall body size. These results indicate that adaptive phenotypic plasticity as well as genetic changes underlie major patterns of clinal variation in house mice and likely facilitated their rapid expansion into new environments across the Americas. In Chapter 2, I generated RNA-seq data from liver and brown adipose tissue (BAT) to identify evolved and plastic gene expression differences in mice adapted to warm and cold environments and asked whether gene expression plasticity is generally in the same or opposite direction as evolved differences. First, I demonstrated that gene expression in both BAT and liver revealed evolved differences between mice adapted to warm and cold environments as well as plasticity in response to warm and cold temperatures. Mice from cold environments showed an attenuated response to the cold as they exhibited fewer differentially expressed genes than warm-adapted mice across both tissues. I then determined the direction of plasticity and found that gene expression plasticity in warm-adapted mice is positively correlated with expression divergence, consistent with adaptive plasticity. Interestingly, BAT displayed a larger proportion of non-adaptive plasticity compared to the liver, highlighting tissue-specific differences in expression plasticity. Overall, these data suggest that both adaptive and non-adaptive plasticity play important roles in the rapid colonization of novel environments in house mice. In Chapter 3, I characterized regulatory divergence and plasticity between New York and Brazil house mice. Specifically, using RNA-seq data generated in chapter 2, I studied gene expression in house mice reared in warm and cold environments in parents adapted to warm and cold environments and in their F1 hybrids. I identified strong patterns of expression divergence across environments, largely attributable to cis-regulatory changes. Expression plasticity was largely attributable to trans-effects as trans-effects showed greater sensitivity to temperature change. I also identified genes for which there were significant effects of temperature on regulatory divergence, with genes exhibiting cis x environment and trans x environment effects. Among these genes, I identified two candidates (Scd1 and Cdh13) that show patterns of adaptive plasticity and play important roles in adiposity and thermoregulation. These findings demonstrate the utility of allele-specific expression to identify regulatory mechanisms associated with both environmental adaptation and plasticity.

The human commensal, Mus musculus domesticus, is native to the Eurasian continent, having spread from north of the Indo-Pakistani subcontinent and the Middle East to western Europe by the end of the Iron Age. Wild populations of M. m. domesticus are now distributed across Africa, the Americas, and Oceania, a range consistent with the global migration of western Europeans that began in the late 15th century. Despite its standing as the premier mammalian model organism for biomedical, ecological, and evolutionary research, important details surrounding the population history of wild house mice remain a mystery. To investigate patterns of genetic structure and infer the demographic history of the western house mouse, I analyze collections of genomic and exomic sequences generated from mice sampled in western Europe and the Americas. First, I present evidence that connects the colonization history of house mice in eastern North America to the colonization history of the Americas by European settlers which began in the late 15th century. Then, I analyzed a genomic dataset from an expanded sampling of Western Europe, including 59 new whole genome sequences from historically relevant regions of Western Europe, and present evidence of population structure, population splits, and gene flow between distinct populations of Western European house mice. Lastly, I combine datasets comprised of Western European genomes and North and South American exomes to detail the recent colonization history of house mice in both American continents. Altogether, these results elucidate details around the recent introduction of Western European house mice to North and South America, highlighting the effects of human migration and global colonization on the concurrent spread of an invasive human commensal.

Understanding the genomic architecture of complex adaptive traits in natural populations has long been a central goal in evolutionary biology. Of particular interest is the extent of the “genetic toolbox,” that is: how many evolutionarily viable solutions are there to a common evolutionary problem? To get at this question, I looked at convergent adaptation to desert environments across the order Rodentia. In mammals, the problem of living in an extremely arid environment with limited access to free water has been solved in a variety of different ways. For granivorous desert rodents, many have developed ultra-efficient osmoregulatory systems, including highly modified kidneys, to handle the problem of salt and water homeostasis. Here, I examined the genomic basis of adaptation across multiple evolutionary timescales.

In chapter one, I studied the recent invasion of a North American desert by the house mouse, Mus musculus. I compared this desert population with a non-desert population and examined organismal phenotypes and gene expression profiles in the lab for experimentally dehydrated mice and for control mice. I discovered significant differences in the response to water stress in these two populations. Mice derived from a desert population showed a less extreme response to water stress, suggesting they may have adapted to live in deserts. Non-desert mice showed shifts towards desert-like gene expression after water stress, consistent with adaptive plasticity. Further, I identified candidate genes underlying desert adaptation in this population.

In chapter two, I leveraged three phylogenetically distinct species pairs, one a desert specialist and the other a mesic representative, from three families of rodents over ~70 million years of divergence and found convergent patterns of evolution using sequences of expressed genes (RNA-seq data). First, I looked for convergent shifts in gene expression associated with desert living in each species pair and identified genes displaying this pattern. Next I used multiple methods to test for aspects of convergence at the DNA sequence level and found that few genes showed both convergence in DNA expression and convergence in protein structure. I found that a greater proportion of genes tested show convergent patterns in gene expression suggesting that changes in the regulation of genes may have a greater impact than changes in protein structure during adaptation to desert environments.

In chapter three, I sequenced and assembled de novo the genome of the rock pocket mouse, Chaetodipus intermedius (Heteromyidae) using MaSuRCA. This species, found in the Sonoran and Chihuahuan deserts of North America, is a desert specialist and a model for understanding the genetics of coat color variation in wild populations. This genome is a resource to be used to further understand the genomics of adaptation in this unique lineage.

Gene expression is a molecular phenotype that is essential to organismal form and fitness. However, how gene regulation evolves over evolutionary time and contributes to phenotypic differences within and between species is still not well understood. In my dissertation, I examined the role of gene regulation in adaptation and speciation in house mice (Mus musculus).

In chapter 1, I reviewed theoretical models and empirical data on the role of gene regulation in the origin of new species. I discuss how regulatory divergence between species can result in hybrid dysfunction and point to areas that could benefit from future research.

In chapter 2, I characterized regulatory divergence between M. m. domesticus and M. m. musculus associated with male hybrid sterility. The major model for the evolution of post-zygotic isolation proposes that hybrid sterility or inviability will evolve as a product of deleterious interactions (i.e., negative epistasis) between alleles at different loci when joined together in hybrids. As the regulation of gene expression is inherently based on interactions between loci, disruption of gene regulation in hybrids may be a common mechanism for post-zygotic isolation. To test this question, I compared expression differences between house mouse subspecies with expression patterns in sterile and fertile male F1 hybrids. I identified extensive regulatory divergence between the subspecies in the testis and found that compensatory cis-and trans- changes were non-randomly associated with genes that were misexpressed in sterile hybrids, but not in fertile hybrids. These results support the idea that such regulatory interactions may contribute to hybrid incompatibilities and may be major drivers of speciation.

In my third chapter, I used expression quantitative trait locus (eQTL) mapping in tandem with a genome scan for selection to identify adaptive regulatory variation in house mice (M. m. domesticus) on the east coast of North America. Mice on the east coast of North America show adaptive differences in body mass in response to latitudinal variation in temperature. I identified genes with clinally varying cis-eQTL for which expression level is correlated with latitude. Among these clinal outliers, I identified two genes (Adam17 and Bcat2) with cis-eQTL of large effect that are associated with adaptive body mass variation and for which expression is correlated with body mass both within and between populations. Adam17 and Bcat2 deletions affect body mass in mice and these genes have also been linked to obesity in humans. These findings provide strong evidence for cis- regulatory elements as essential loci of environmental adaptation in natural populations.

In chapter 4, I used low-coverage whole genome sequencing data from the same individuals to identify and characterize copy number variation in natural populations of house mice on the east coast of North America. Consistent with a role for copy number variation in local adaptation, I identified two regions where copy number is significantly correlated with latitude. One of these regions contains the gene Trpm8, which has previously been shown to affect physiological responses to environmental cold in other species. These results suggest that copy number variation significantly contributes to genetic variation in North American house mice and that copy number variation may play an important role in local adaptation.

Finally, in my fifth chapter, I examined the relationship between gene co-expression networks and molecular evolution in natural populations of M. m. domesticus. I found that genes that are more central to their co-expression networks (i.e., have more and greater associations with other genes) and genes whose co-expression relationships are conserved across tissues show lower rates of protein evolution, have fewer polymorphisms, and are less likely show regulatory variation. Together, these results are consistent with gene co-expression network structure being a source of evolutionary constraint.