Pollution-Driven Selection in a Non-Biting Midge: Genome-Wide Responses to Bacillus thuringiensis israelensis and Copper

Abstract

Riparian ecosystems are vital interfaces between aquatic and terrestrial environments but are increasingly impacted by anthropogenic pollution. In these systems, merolimnic insects serve as crucial ecological links, occupying aquatic habitats as larvae and terrestrial environments as adults, thus being an essential food source in both. Consequently, pollutant exposure during the aquatic larval stage can have cascading effects across ecosystem boundaries. While the ecological consequences of such exposure are well documented, the evolutionary potential of merolimnic insects to adapt to chronic pollution remains poorly understood. To address this, we previously conducted a selection experiment exposing populations of the non-biting midge Chironomus riparius to the mosquito larvicide Bacillus thuringiensis israelensis (Bti) or heavy metal copper over approximately eight generations, which revealed only limited evidence of consistent phenotypic adaptation. Here we use whole-genome sequencing of these populations to assess their genomic responses to chronic pollutant exposure. Despite similar phenotypic sensitivity in pre-exposed and naïve populations, we detected distinct stressor-specific genomic responses. Copper exposure induced a significant genome-wide reduction in nucleotide diversity and evidence of selection-driven allele frequency changes, while Bti exposure was associated with heterogeneous, replicate-specific shifts, potentially reflecting drift or selection on multiple redundant pathways. Functional enrichment analyses indicated early-stage adaptation: immune- and apoptosis-related pathways were enriched under Bti, while metal detoxification and DNA repair pathways were enriched under copper, highlighting distinct adaptive mechanisms despite weak genome-wide signals of selection. Our findings demonstrate that Evolve and Resequencing approaches enable the detection of early genomic signals of adaptation even when phenotypic change is subtle or absent, offering a powerful framework for studying evolutionary responses to environmental pollution.

Keywords: ecotoxicogenomics; ecotoxicology; evolve and resequencing (E&R); experimental evolution; microevolutionary dynamics; pool‐sequencing.

FIGURE 1

PCA of allele frequencies from 119,556 SNPs across the genome where colour and shape distinguish treatment groups. The founding population is F0; all others are from F8.

FIGURE 2

Candidate SNPs exceeding drift expectations identified from Wright–Fisher simulations in at least four out of six replicates. The horizontal bars indicate the total number of candidate loci in each group while the vertical bars show the number in each category. Black points show treatments in a category where multiple groups are indicated by connected lines. Note that the counts in each group are exclusive; for example, the number of loci in the control‐alone category are those loci not shared with any other group.

FIGURE 3

Convergent correlations of allele frequency change from F0 to F8. Higher values indicate a more similar change in allele frequency between two replicates. Small symbols represent convergent correlations from pairwise population comparisons; large symbols show the mean for each group comparison.

FIGURE 4

The contribution of laboratory adaptation, treatment selection, and drift plus replicate‐specific responses to the total variance of allele frequency change from F0 to F8. The laboratory component was determined using the covariance of allele frequency change between the control and each of the two treatment groups while selection was identified as the covariance within a treatment group minus the laboratory adaptation. The remaining variance was attributed to drift and replicate‐specific responses to selection.

FIGURE 5

Scatterplot of 45 significantly enriched Gene Ontology (GO) terms, visualised in a two‐dimensional semantic space using REVIGO (Supek et al. 2011). The x‐ and y‐axes represent semantic space, where closer terms have higher functional similarity. Each coloured circle represents a GO term, with colours indicating the treatment(s) in which the term was significantly enriched (see legend for colour coding). Some functionally related GO terms are grouped within dashed‐lined circles. Each coloured circle or group is assigned a number, and below the figure, a corresponding list provides higher‐level biological processes associated with the respective GO term(s). A full list of specific GO terms is available in Table S1.