Host association, environment, and geography underlie genomic differentiation in a major forest pest

Abstract

Diverse geographic, environmental, and ecological factors affect gene flow and adaptive genomic variation within species. With recent advances in landscape ecological modelling and high‐throughput DNA sequencing, it is now possible to effectively quantify and partition their relative contributions. Here, we use landscape genomics to identify determinants of genomic differentiation in the forest tent caterpillar, Malacosoma disstria, a widespread and irruptive pest of numerous deciduous tree species in North America. We collected larvae from multiple populations across Eastern Canada, where the species experiences a diversity of environmental gradients and feeds on a number of different host tree species, including trembling aspen (Populus tremuloides), sugar maple (Acer saccharum), red oak (Quercus rubra), and white birch (Betula papyrifera). Using a combination of reciprocal causal modelling (RCM) and distance‐based redundancy analyses (dbRDA), we show that differentiation of thousands of genome‐wide single nucleotide polymorphisms (SNPs) among individuals is best explained by a combination of isolation by distance, isolation by environment (spatial variation in summer temperatures and length of the growing season), and differences in host association. Configuration of suitable habitat inferred from ecological niche models was not significantly related to genomic differentiation, suggesting that M. disstria dispersal is agnostic with respect to habitat quality. Although population structure was not discretely related to host association, our modelling framework provides the first molecular evidence of host‐associated differentiation in M. disstria, congruent with previous documentation of reduced growth and survival of larvae moved between natal host species. We conclude that ecologically mediated selection is contributing to variation within M. disstria, and that divergent adaptation related to both environmental conditions and host association should be considered in ongoing research and management of this important forest pest.

Genome-wide markers show continental structuring and mitonuclear discordance in the forest tent caterpillar (Malacosoma disstria Hübner) (Lepidoptera: Lasiocampidae)

“…1). However, recent work on the eastern population of M. disstria identified signals of host-associated genomic divergence within M. disstria that was not apparent in patterns of population structuring (MacDonald et al 2022). It is unclear whether M. disstria is currently undergoing divergence related to host association that will eventually be reflected in patterns of population clustering or whether gene flow among host-associated groups is simply too high to allow discrete host-associated structuring to precipitate.…”

“…For example, southern populations of M. disstria have distinct regional host associations (Parry and Goyer 2004) and show latitudinal clines in reproductive traits (Parry et al 2001). Given that genomic differentiation was linked to both host association and variation in environmental conditions (MacDonald et al 2022), it is plausible that additional population structure remains undetected throughout the range of M. disstria. Increased sampling across the entire range of M. disstria is needed to fully describe the population variation in this species and would provide insight to the historical biogeography and evolutionary history of this important forest pest.…”

“…Furthermore, larval host plant was not explored as a potential factor structuring genetic diversity, despite M. disstria populations exhibiting significant performance differences on different host plants (e.g., Nicol et al 1997;Parry and Goyer 2004). Recent work by MacDonald et al (2022) used thousands of nuclear genomic markers to identify spatial, environmental, and host-associated divergence among regional populations of M. disstria in eastern Canada. This work explored the fine-scale genomic structuring among M. disstria collected on distinct host plants across a latitudinal gradient.…”

“…Recent work by MacDonald et al. (2022) used thousands of nuclear genomic markers to identify spatial, environmental, and host-associated divergence among regional populations of M. disstria in eastern Canada. This work explored the fine-scale genomic structuring among M. disstria collected on distinct host plants across a latitudinal gradient.…”

“…filtering and SNP identification. Bioinformatic processing of raw sequence reads followedMacDonald et al (2022). Briefly, we used Stacks 2.0(Rochette et al 2019) to demultiplex Illumina single-end, 75-bp reads, removing those with quality scores of less than 20 within a sliding window equal to 15% of the read length.…”