Little is known of how gene expression and its plasticity evolves as populations adapt to different environmental regimes. Expression is expected to evolve adaptively in all populations but only those populations experiencing environmental heterogeneity are expected to show adaptive evolution of plasticity. We measured the transcriptome in a cadmium-enriched diet and a salt-enriched diet for experimental populations of Drosophila melanogaster that evolved for ~130 generations in one of four selective regimes: two constant regimes maintained in either cadmium or salt diets and two heterogeneous regimes that varied either temporally or spatially between the two diets. For populations evolving in constant regimes, we find a strong signature of counter-gradient evolution; the evolved expression differences between populations adapted to alternative diets is opposite to the plastic response of the ancestral population that is naïve to both diets. Based on expression patterns in the ancestral populations, we identify a set of genes for which we predict selection in heterogeneous regimes to result in increases in plasticity and we find the expected pattern. In contrast, a set of genes where we predicted reduced plasticity did not follow expectation. Nonetheless, both gene sets showed a pattern consistent with adaptive expression evolution in heterogeneous regimes, highlighting the difference between observing “optimal” plasticity and improvements in environment-specific expression. Looking across all genes, there is evidence in all regimes of differences in biased allele expression across environments (“allelic plasticity”) and this is more common among genes with plasticity in total expression.

Environmental heterogeneity has been hypothesized to influence levels of genetic variation but the effect of heterogeneity depends on (i) the form of heterogeneity, (ii) whether ecologically relevant or neutral loci are being considered, and (iii) the genetic basis of ecological adaptation. We surveyed genome-wide SNP diversity in replicate experimental Drosophila melanogaster populations with equal census sizes that evolved for 42 generations under one of four selection regimes: (i) salt-enriched environment (Salt), (ii) cadmium-enriched environment (Cad), (iii) temporally (Temp) or (iv) spatially (Spatial) variable environments. There was significant differentiation between all pairs of treatments but the greatest differentiation occurred between the two homogenous treatments (Cad and Salt). For sites likely under differential ecological selection (and those closely linked to them), the pattern of within-population diversity π followed the expectation from classic antagonistic selection theory: Spatial>Temp>Salt≈Cad. However, neutral diversity unlinked to selected sites followed a different pattern: Spatial>Salt≈Cad>Temp. As implicated by the latter result, measures of FST among replicate populations within treatments are consistent with differences in effective population sizes among selective regimes despite equal census sizes. Though there are clear changes in the rank order of treatments when contrasting selected and neutral sites with respect to π, the rank ordering of treatments with respect to FST appears reasonably consistent between site categories. These results demonstrate that alternative selective regimes affect within- and among-population diversity differently for different site types.

Heterogeneous environments are typically expected to maintain more genetic variation in fitness within populations than homogeneous environments. However, the accuracy of this claim depends on the form of heterogeneity as well as the genetic basis of fitness traits and how similar the assay environment is to the environment of past selection. Here, we measure quantitative genetic (QG) variance for three traits important for fitness using replicated experimental populations of Drosophila melanogaster evolving under four selective regimes: constant salt-enriched medium (Salt), constant cadmium-enriched medium (Cad), and two heterogeneous regimes that vary either temporally (Temp) or spatially (Spatial). As theory predicts, we found that Spatial populations tend to harbor more genetic variation than Temp populations or those maintained in a constant environment that is the same as the assay environment. Contrary to expectation, Salt populations tend to have more genetic variation than Cad populations in both assay environments. We discuss the patterns for QG variances across regimes in relation to previously reported data on genome-wide sequence diversity. For some traits, the QG patterns are similar to the diversity patterns of ecological selected SNPs, whereas the QG patterns for some other traits resembled that of neutral SNPs.

Changes in gene regulation at multiple levels may comprise an important share of the molecular changes underlying adaptive evolution in nature. However, few studies have assayed within- and between-population variation in gene regulatory traits at a transcriptomic scale, and therefore inferences about the characteristics of adaptive regulatory changes have been elusive. Here, we assess quantitative trait differentiation in gene expression levels and alternative splicing (intron usage) between three closely-related pairs of natural populations of Drosophila melanogaster from contrasting thermal environments that reflect three separate instances of cold tolerance evolution. The cold-adapted populations were known to show population genetic evidence for parallel evolution at the SNP level, and here we find evidence for parallel expression evolution between them, with stronger parallelism at larval and adult stages than for pupae. We also implement a flexible method to estimate cis- versus trans-encoded contributions to expression or splicing differences at the adult stage. The apparent contributions of cis- versus trans-regulation to adaptive evolution vary substantially among population pairs. While two of three population pairs show a greater enrichment of cis-regulatory differences among adaptation candidates, trans-regulatory differences are more likely to be implicated in parallel expression changes between population pairs. Genes with significant cis-effects are enriched for signals of elevated genetic differentiation between cold- and warm-adapted populations, suggesting that they are potential targets of local adaptation. These findings expand our knowledge of adaptive gene regulatory evolution and our ability to make inferences about this important and widespread process.

Environmental heterogeneity helps maintain genetic variation in fitness. Therefore, one might predict that populations living in heterogeneous environments have higher adaptive potential than populations living in homogeneous environments. Such a prediction could be useful in guiding conservation priorities without requiring detailed genetic studies. However, this prediction will be true only if the additional genetic variation maintained by environmental heterogeneity can be used to respond to novel selection. Here we examine the effect of environmental heterogeneity on future adaptability using replicated experimental Drosophila melanogaster populations that had previously evolved for ∼100 generations under one of four selective regimes: constant salt-enriched larvae medium, constant cadmium-enriched larvae medium, and two heterogeneous regimes that vary either temporally or spatially between the two media. Replicates of these experimental populations were subjected to a novel heat stress while being maintained in their original larval diet selection regimes. Adaptation to increased temperature was measured with respect to female productivity and male siring success after ∼20 generations. For female productivity, there was evidence of adaptation overall and heterogeneous populations had a larger adaptive response than homogeneous populations. There was less evidence of adaptation overall for male siring success and no support for faster adaptation in heterogeneous populations.