Divergence with gene flow in a population of thermophilic bacteria: a potential role for spatially varying selection

A fundamental goal of evolutionary biology is to understand how ecological diversity arises and is maintained in natural populations. We have investigated the contributions of gene flow and divergent selection to the distribution of genetic variation in an ecologically differentiated population of a thermophilic cyanobacterium (Mastigocladus laminosus) found along the temperature gradient of a nitrogen‐limited stream in Yellowstone National Park. For most loci sampled, gene flow appears to be sufficient to prevent substantial genetic divergence. However, one locus (rfbC) exhibited a comparatively low migration rate as well as other signatures expected for a gene experiencing spatially varying selection, including an excess of common variants, an elevated level of polymorphism and extreme genetic differentiation along the gradient. rfbC is part of an expression island involved in the production of the polysaccharide component of the protective envelope of the heterocyst, the specialized nitrogen‐fixing cell of these bacteria. SNP genotyping in the vicinity of rfbC revealed a ~5‐kbp region including a gene content polymorphism that is tightly associated with environmental temperature and therefore likely contains the target of selection. Two genes have been deleted both in the predominant haplotype found in the downstream region of White Creek and in strains from other Yellowstone populations of M. laminosus, which may result in the production of heterocysts with different envelope properties. This study implicates spatially varying selection in the maintenance of variation related to thermal performance at White Creek despite on‐going or recent gene flow.     

The advantages of going large: genome‐wide SNPs clarify the complex population history and systematics of the threatened western pond turtle

As the field of phylogeography has matured, it has become clear that analyses of one or a few genes may reveal more about the history of those genes than the populations and species that are the targets of study. To alleviate these concerns, the discipline has moved towards larger analyses of more individuals and more genes, although little attention has been paid to the qualitative or quantitative gains that such increases in scale and scope may yield. Here, we increase the number of individuals and markers by an order of magnitude over previously published work to comprehensively assess the phylogeographical history of a well‐studied declining species, the western pond turtle (Emys marmorata). We present a new analysis of 89 independent nuclear SNP markers and one mitochondrial gene sequence scored for rangewide sampling of >900 individuals, and compare these to smaller‐scale, rangewide genetic and morphological analyses. Our enlarged SNP data fundamentally revise our understanding of evolutionary history for this lineage. Our results indicate that the gains from greatly increasing both the number of markers and individuals are substantial and worth the effort, particularly for species of high conservation concern such as the pond turtle, where accurate assessments of population history are a prerequisite for effective management.     

Massive introgression of major histocompatibility complex (MHC) genes in newt hybrid zones

Variation in the vertebrate major histocompatibility complex (MHC) genes is crucial for fighting pathogen assault. Because new alleles confer a selective advantage, MHC should readily introgress between species, even under limited hybridization. Using replicated transects through two hybrid zones between strongly reproductively isolated European newts, Lissotriton montandoni and L. vulgaris, we demonstrated recent and ongoing MHC class I and II introgression in the Carpathian region. The extent of introgression correlated with the age of contact. In the older zone, MHC similarity between species within transects exceeded similarity between transects within species, implying pervasive introgression ‐ a massive exchange of MHC genes, not limited to specific variants. In simulations, the observed pattern emerged under the combined action of balancing selection and hybridization, but not when these processes acted separately. Thus, massive introgression at advanced stages of divergence can introduce novel and restore previously lost MHC variation, boosting the adaptive potential of hybridizing taxa. In consequence, MHC genes may be the last to stop introgressing between incipient species.     

A test of the central–marginal hypothesis using population genetics and ecological niche modelling in an endemic salamander (Ambystoma barbouri)

The central–marginal hypothesis (CMH) predicts that population size, genetic diversity and genetic connectivity are highest at the core and decrease near the edges of species' geographic distributions. We provide a test of the CMH using three replicated core‐to‐edge transects that encompass nearly the entire geographic range of the endemic streamside salamander (Ambystoma barbouri). We confirmed that the mapped core of the distribution was the most suitable habitat using ecological niche modelling (ENM) and via genetic estimates of effective population sizes. As predicted by the CMH, we found statistical support for decreased genetic diversity, effective population size and genetic connectivity from core to edge in western and northern transects, yet not along a southern transect. Based on our niche model, habitat suitability is lower towards the southern range edge, presumably leading to conflicting core‐to‐edge genetic patterns. These results suggest that multiple processes may influence a species' distribution based on the heterogeneity of habitat across a species' range and that replicated sampling may be needed to accurately test the CMH. Our work also emphasizes the importance of identifying the geographic range core with methods other than using the Euclidean centre on a map, which may help to explain discrepancies among other empirical tests of the CMH. Assessing core‐to‐edge population genetic patterns across an entire species' range accompanied with ENM can inform our general understanding of the mechanisms leading to species' geographic range limits.     

Genetic rescue: a safe or risky bet?

Small and isolated populations face threats from genetic drift and inbreeding. To rescue populations from these threats, conservation biologists can augment gene flow into small populations to increase variation and reduce inbreeding depression. Spectacular success stories include greater prairie chickens in Illinois (Westermeier et al. 1998 ), adders in Sweden (Madsen et al. 1999 ) and panthers in Florida (Johnson et al. 2010 ). However, we also know that performing such crosses risks introducing genes that may be poorly adapted to local conditions or genetic backgrounds. A classic example of such ‘outbreeding depression’ occurred when different subspecies of ibex from Turkey and the Sinai were introduced to assist recovery of an ibex population in Czechoslovakia (Templeton 1986 ). Despite being fertile, the hybrids birthed calves too early, causing the whole population to disappear. In the face of uncertainty, conservation biologists have tended to respect genetic identity, shying away from routinely crossing populations. In this issue of Molecular Ecology, Frankham ( 2015 ) compiles empirical data from experimental studies to assess the costs and benefits of between‐population crosses (Fig.  1 ). Crosses screened to exclude those involving highly divergent populations or distinct habitats show large heterosis with few apparent risks of outbreeding depression. This leads Frankham to advocate for using assisted gene flow more widely. But do the studies analysed in this meta‐analysis adequately test for latent outcrossing depression?     

Genetic architecture and balancing selection: the life and death of differentiated variants

Balancing selection describes any form of natural selection, which results in the persistence of multiple variants of a trait at intermediate frequencies within populations. By offering up a snapshot of multiple co‐occurring functional variants and their interactions, systems under balancing selection can reveal the evolutionary mechanisms favouring the emergence and persistence of adaptive variation in natural populations. We here focus on the mechanisms by which several functional variants for a given trait can arise, a process typically requiring multiple epistatic mutations. We highlight how balancing selection can favour specific features in the genetic architecture and review the evolutionary and molecular mechanisms shaping this architecture. First, balancing selection affects the number of loci underlying differentiated traits and their respective effects. Control by one or few loci favours the persistence of differentiated functional variants by limiting intergenic recombination, or its impact, and may sometimes lead to the evolution of supergenes. Chromosomal rearrangements, particularly inversions, preventing adaptive combinations from being dissociated are increasingly being noted as features of such systems. Similarly, due to the frequency of heterozygotes maintained by balancing selection, dominance may be a key property of adaptive variants. High heterozygosity and limited recombination also influence associated genetic load, as linked recessive deleterious mutations may be sheltered. The capture of deleterious elements in a locus under balancing selection may reinforce polymorphism by further promoting heterozygotes. Finally, according to recent genomewide scans, balanced polymorphism might be more pervasive than generally thought. We stress the need for both functional and ecological studies to characterize the evolutionary mechanisms operating in these systems.     

Niche divergence builds the case for ecological speciation in skinks of the Plestiodon skiltonianus species complex

Adaptation to different thermal environments has the potential to cause evolutionary changes that are sufficient to drive ecological speciation. Here, we examine whether climate‐based niche divergence in lizards of the Plestiodon skiltonianus species complex is consistent with the outcomes of such a process. Previous work on this group shows that a mechanical sexual barrier has evolved between species that differ mainly in body size and that the barrier may be a by‐product of selection for increased body size in lineages that have invaded xeric environments; however, baseline information on niche divergence among members of the group is lacking. We quantified the climatic niche using mechanistic physiological and correlative niche models and then estimated niche differences among species using ordination techniques and tests of niche overlap and equivalency. Our results show that the thermal niches of size‐divergent, reproductively isolated morphospecies are significantly differentiated and that precipitation may have been as important as temperature in causing increased shifts in body size in xeric habitats. While these findings alone do not demonstrate thermal adaptation or identify the cause of speciation, their integration with earlier genetic and behavioral studies provides a useful test of phenotype–environment associations that further support the case for ecological speciation in these lizards.     

Complex trait divergence contributes to environmental niche differentiation in ecological speciation of Boechera stricta

Ecological factors may contribute to reproductive isolation if differential local adaptation causes immigrant or hybrid fitness reduction. Because local adaptation results from the interaction between natural selection and adaptive traits, it is crucial to investigate both to understand ecological speciation. Previously, we used niche modelling to identify local water availability as an environmental correlate of incipient ecological speciation between two subspecies in Boechera stricta, a close relative of Arabidopsis. Here, we performed several large‐scale glasshouse experiments to investigate the divergence of various physiological, phenological and morphological traits. Although we found no significant difference in physiological traits, the Western subspecies has significantly faster growth rate, larger leaf area, less succulent leaves, delayed reproductive time and longer flowering duration. These trait differences are concordant with previous results that habitats of the Western genotypes have more consistent water availability, while Eastern genotypes inhabit locations with more ephemeral water supplies. In addition, by comparing univariate and multivariate divergence of complex traits (Q_ST) to the genomewide distribution of SNP F_ST, we conclude that the aspects of phenology and morphology (but not physiology) are under divergent selection. In addition, we also identified several highly diverged traits without obvious water‐related functions.     

Genomic and phenotypic differentiation of Arabidopsis thaliana along altitudinal gradients in the North Italian Alps

Altitudinal gradients in mountain regions are short‐range clines of different environmental parameters such as temperature or radiation. We investigated genomic and phenotypic signatures of adaptation to such gradients in five Arabidopsis thaliana populations from the North Italian Alps that originated from 580 to 2350 m altitude by resequencing pools of 19–29 individuals from each population. The sample includes two pairs of low‐ and high‐altitude populations from two different valleys. High‐altitude populations showed a lower nucleotide diversity and negative Tajima's D values and were more closely related to each other than to low‐altitude populations from the same valley. Despite their close geographic proximity, demographic analysis revealed that low‐ and high‐altitude populations split between 260 000 and 15 000 years before present. Single nucleotide polymorphisms whose allele frequencies were highly differentiated between low‐ and high‐altitude populations identified genomic regions of up to 50 kb length where patterns of genetic diversity are consistent with signatures of local selective sweeps. These regions harbour multiple genes involved in stress response. Variation among populations in two putative adaptive phenotypic traits, frost tolerance and response to light/UV stress was not correlated with altitude. Taken together, the spatial distribution of genetic diversity reflects a potentially adaptive differentiation between low‐ and high‐altitude populations, whereas the phenotypic differentiation in the two traits investigated does not. It may resemble an interaction between adaptation to the local microhabitat and demographic history influenced by historical glaciation cycles, recent seed dispersal and genetic drift in local populations.     

Extinction,reintroduction, and restoration of a lizard meta‐population equilibrium in the missouri ozarks

In this issue of Molecular Ecology, Neuwald & Templeton (2013) report on a 22‐year study of natural populations of Collared Lizards (Crotaphytus collaris) that evolved on isolated on rock outcrops (‘glades’) in the Ozark Mountains in eastern Missouri. This ecosystem was originally maintained by frequent fires that kept the forest understory open, but fire‐suppression was adopted as official policy in about 1945, which led to a loss of native biodiversity, including local extinctions of some lizard populations. Policies aimed at restoring biodiversity included controlled burns and re‐introductions of lizards to some glades, which began in 1984. Populations were monitored from 1984–2006, and demographic and genetic data collected from 1 679 lizards were used to documents shifts in meta‐population dynamics over four distinct phases of lizard recovery: 1–an initial translocation of lizards drawn from the same source populations onto three glades that were likely part of one meta‐population; 2–a period of isolation and genetic drift associated with the absence of fires; 3–a period of rapid colonization and population increase following restoration of fire; and 4–stabilization of the meta‐population under regular prescribed burning. This study system thus provides a rare opportunity to characterize the dynamics of a landscape‐scale management strategy on the restoration of the meta‐population of a reintroduced species; long‐term case studies of the extinction, founding, increase, and stabilization of a well‐defined meta‐population, based on both demographic and population genetic data, are rare in the conservation, ecological, and evolutionary literature.     

Genetic diversity and recombination in natural populations of the nematode‐trapping fungus Arthrobotrys oligospora from China

Nematophagous fungi can trap and capture nematodes and other small invertebrates. This unique ability has made them ideal organisms from which to develop biological control agents against plant‐ and animal‐parasitic nematodes. However, effective application of biocontrol agents in the field requires a comprehensive understanding about the ecology and population genetics of the nematophagous fungi in natural environments. Here, we genotyped 228 strains of the nematode‐trapping fungus Arthrobotrys oligospora using 12 single nucleotide polymorphic markers located on eight random DNA fragments. The strains were from different ecological niches and geographical regions from China. Our analyses identified that ecological niche separations contributed significantly, whereas geographic separation contributed relatively little to the overall genetic variation in our samples of A. oligospora. Interestingly, populations from stressful environments seemed to be more variable and showed more evidence for recombination than those from benign environments at the same geographic areas. We discussed the implications of our results to the conservation and biocontrol application of A. oligospora in agriculture and forestry.     

Genetic diversity of small populations: Not always “doom and gloom”?

Is a key theory of evolutionary and conservation biology—that loss of genetic diversity can be predicted from population size—on shaky ground? In the face of increasing human‐induced species depletion and habitat fragmentation, this question and the study of genetic diversity in small populations are paramount to understanding the limits of species’ responses to environmental change and to providing remedies to endangered species conservation. Few empirical studies have investigated to what degree some small populations might be buffered against losses of genetic diversity. Even fewer studies have experimentally tested the potential underlying mechanisms. The study of Schou, Loeschcke, Bechsgaard, Schlotterer, and Kristensen ( 2017 ) in this issue of Molecular Ecology is elegant in combining classic common garden experimentation with population genomics on an iconic experimental model species (Drosophila melanogaster). The authors reveal a slower rate of loss of genetic diversity in small populations under varying thermal regimes than theoretically expected and hence an unexpected retention of genetic diversity. They are further able to hone in on a plausible mechanism: associative overdominance, wherein homozygosity of deleterious recessive alleles is especially disfavoured in genomic regions of low recombination. These results contribute to a budding literature on the varying mechanisms underlying genetic diversity in small populations and encourage further such research towards the effective management and conservation of fragmented or endangered populations.     

The role of structural genomic variants in population differentiation and ecotype formation in Timema cristinae walking sticks

Theory predicts that structural genomic variants such as inversions can promote adaptive diversification and speciation. Despite increasing empirical evidence that adaptive divergence can be triggered by one or a few large inversions, the degree to which widespread genomic regions under divergent selection are associated with structural variants remains unclear. Here we test for an association between structural variants and genomic regions that underlie parallel host‐plant‐associated ecotype formation in Timema cristinae stick insects. Using mate‐pair resequencing of 20 new whole genomes we find that moderately sized structural variants such as inversions, deletions and duplications are widespread across the genome, being retained as standing variation within and among populations. Using 160 previously published, standard‐orientation whole genome sequences we find little to no evidence that the DNA sequences within inversions exhibit accentuated differentiation between ecotypes. In contrast, a formerly described large region of reduced recombination that harbours genes controlling colour‐pattern exhibits evidence for accentuated differentiation between ecotypes, which is consistent with differences in the frequency of colour‐pattern morphs between host‐associated ecotypes. Our results suggest that some types of structural variants (e.g., large inversions) are more likely to underlie adaptive divergence than others, and that structural variants are not required for subtle yet genome‐wide genetic differentiation with gene flow.     

Landscape features along migratory routes influence adaptive genomic variation in anadromous steelhead (Oncorhynchus mykiss)

Organisms typically show evidence of adaptation to features within their local environment. However, many species undergo long‐distance dispersal or migration across larger geographic regions that consist of highly heterogeneous habitats. Therefore, selection may influence adaptive genetic variation associated with landscape features at residing sites and along migration routes in migratory species. We tested for genomic adaptation to landscape features at natal spawning sites and along migration paths to the ocean of anadromous steelhead trout (Oncorhynchus mykiss) in the Columbia River Basin. Results from multivariate ordination, gene–environment association and outlier analyses using 24,526 single nucleotide polymorphisms (SNPs) provided evidence that adaptive allele frequencies were more commonly associated with landscape features along migration paths than features at natal sites (91.8% vs. 8.2% of adaptive loci, respectively). Among the 45 landscape variables tested, migration distance to the ocean and mean annual precipitation along migration paths were significantly associated with adaptive genetic variation in three distinct genetic groups. Additionally, variables such as minimum migration water temperature and mean migration slope were significant only in inland stocks of steelhead that migrate up to 1,200 km farther than those near the coast, indicating regional differences in migratory selective pressures. This study provides novel approaches for investigating migratory corridors and some of the first evidence that environment along migration paths can lead to substantial divergent selection. Consequently, our approach to understand genetic adaptation to migration conditions can be applied to other migratory species when migration or dispersal paths are generally known.     

Invasion facilitates hybridization with introgression in the Rattus rattus species complex

Biological invasions result in novel species interactions, which can have significant evolutionary impacts on both native and invading taxa. One evolutionary concern with invasions is hybridization among lineages that were previously isolated, but make secondary contact in their invaded range(s). Black rats, consisting of several morphologically very similar but genetically distinct taxa that collectively have invaded six continents, are arguably the most successful mammalian invaders on the planet. We used mitochondrial cytochrome b sequences, two nuclear gene sequences (Atp5a1 and DHFR) and nine microsatellite loci to examine the distribution of three invasive black rat lineages (Rattus tanezumi, Rattus rattus I and R. rattus IV) in the United States and Asia and to determine the extent of hybridization among these taxa. Our analyses revealed two mitochondrial lineages that have spread to multiple continents, including a previously undiscovered population of R. tanezumi in the south‐eastern United States, whereas the third lineage (R. rattus IV) appears to be confined to Southeast Asia. Analyses of nuclear DNA (both sequences and microsatellites) suggested significant hybridization is occurring among R. tanezumi and R. rattus I in the United States and also suggest hybridization between R. tanezumi and R. rattus IV in Asia, although further sampling of the latter species pair in Asia is required. Furthermore, microsatellite analyses suggest unidirectional introgression from both R. rattus I and R. rattus IV into R. tanezumi. Within the United States, introgression appears to be occurring to such a pronounced extent that we were unable to detect any nuclear genetic signal for R. tanezumi, and a similar pattern was detected in Asia.     

Molecular population genetics of the melanic plumage polymorphism in Arctic skuas (Stercorarius parasiticus): evidence for divergent selection on plumage colour

The Arctic skua (Stercorarius parasiticus) is a classic example of an avian plumage polymorphism, with variation in melanin‐based ventral plumage coloration defining pale, intermediate and dark morphs in adults of both sexes. However, despite several decades of field research, there is an incomplete understanding of how the polymorphism in ventral plumage colour is maintained and the selective forces involved. Here, we investigate selection on a locus (MC1R) that is strongly associated with plumage colour variation in Arctic skuas using patterns of nucleotide variation and comparison to neutral loci (nuclear introns and mtDNA). We find that three linked nonsynonymous mutations in MC1R, including the single mutation described previously, are associated with plumage colour in the Arctic skua. The position of nonsynonymous mutations on a MC1R haplotype network implies that divergent selection drove the initial evolution of the colour morphs. Comparisons of F_STs of MC1R vs. nuclear introns among five skua populations differing in proportion of dark morphs along an approximate north–south cline reveal a signature of divergent selection on MC1R. In contrast, we find limited evidence for balancing selection on MC1R within populations, although the power is low. Our results provide strong evidence for both past and ongoing selection on MC1R, and, by implication, plumage colour in Arctic skuas. The results suggest that a fruitful avenue for future ecological studies will be analysis of selection on morphs in colonies at the extremes along the morph ratio cline.     

Statistical methods in spatial genetics

The joint analysis of spatial and genetic data is rapidly becoming the norm in population genetics. More and more studies explicitly describe and quantify the spatial organization of genetic variation and try to relate it to underlying ecological processes. As it has become increasingly difficult to keep abreast with the latest methodological developments, we review the statistical toolbox available to analyse population genetic data in a spatially explicit framework. We mostly focus on statistical concepts but also discuss practical aspects of the analytical methods, highlighting not only the potential of various approaches but also methodological pitfalls.     

All models are wrong

As the field of phylogeography has continued to move in the model‐based direction, researchers continue struggling to construct useful models for inference. These models must be both simple enough to be tractable yet contain enough of the complexity of the natural world to make meaningful inference. Beyond constructing such models for inference, researchers explore model space and test competing models with the data on hand, with the goal of improving the understanding of the natural world and the processes underlying natural biological communities. Approximate Bayesian computation (ABC) has increased in recent popularity as a tool for evaluating alternative historical demographic models given population genetic samples. As a thorough demonstration, Pelletier & Carstens ( 2014 ) use ABC to test 143 phylogeographic submodels given geographically widespread genetic samples from the salamander species Plethodon idahoensis (Carstens et al. 2004 ) and, in so doing, demonstrate how the results of the ABC model choice procedure are dependent on the model set one chooses to evaluate.     

Identifying the number of population clusters with structure: problems and solutions

The program structure has been used extensively to understand and visualize population genetic structure. It is one of the most commonly used clustering algorithms, cited over 11 500 times in Web of Science since its introduction in 2000. The method estimates ancestry proportions to assign individuals to clusters, and post hoc analyses of results may indicate the most likely number of clusters, or populations, on the landscape. However, as has been shown in this issue of Molecular Ecology Resources by Puechmaille ( 2016 ), when sampling is uneven across populations or across hierarchical levels of population structure, these post hoc analyses can be inaccurate and identify an incorrect number of population clusters. To solve this problem, Puechmaille ( 2016 ) presents strategies for subsampling and new analysis methods that are robust to uneven sampling to improve inferences of the number of population clusters.