Whole genomes from the extinct Xerces Blue butterfly can help identify declining insect species

Abstract

The Xerces Blue (Glaucopsyche xerces) is considered to be the first butterfly to become extinct at global scale in historical times. It was notable for its chalky lavender wings with conspicuous white spots on the ventral wings. The last individuals were collected in their restricted habitat, in the dunes near the Presidio military base in San Francisco, in 1941. We sequenced the genomes of four 80 to 100-year-old Xerces Blue, and seven historical and one modern specimens of its closest relative, the Silvery Blue (G. lygdamus). We compared these to a novel annotated genome of the Green-Underside Blue (G. alexis). Phylogenetic relationships inferred from complete mitochondrial genomes indicate that Xerces Blue was a distinct species that diverged from the Silvery Blue lineage at least 850,000 years ago. Using nuclear genomes, both species experienced population growth during the Eemian interglacial period, but the Xerces Blue decreased to a very low effective population size subsequently, a trend opposite to that observed in the Silvery Blue. Runs of homozygosity and deleterious load in the Xerces Blue were significantly greater than in the Silvery Blue, suggesting a higher incidence of inbreeding. These signals of population decline observed in Xerces Blue could be used to identify and monitor other insects threatened by human activities, whose extinction patterns are still not well known.

The last days ofAporia crataegi(L.) in Britain: evaluating genomic erosion in an extirpated butterfly

“…Our results suggest that a consistently small N e on the order of ≈10 4 , inbreeding in the recent past and accumulation of deleterious variants could be viable indicators of population decline in species with generally large N e . Indeed, similar signals of genomic erosion have also been detected in the genomes of the extinct G. xerces (de-Dios et al ., 2023).…”

“…We, as others (e.g., Hohenlohe et al ., 2021), recommend that single metrics such as genetic diversity (individual heterozygosity in our case) may not be sufficient on their own to delineate threatened populations and need to be supplemented with estimates of inbreeding and genetic load. Here and elsewhere (de-Dios et al ., 2023), we demonstrate that this achievable for insect species with a relatively small number of samples.…”

“…In contrast, no category of F RoH was significantly different between British and continental European P. icarus suggesting this population likely recovered from the initial bottleneck at the time of post-glacial colonisation of Britain. Morever, F roh for British A. crataegi are substantially larger in size and proportion than those estimated from historical genomes of the extinct G. xerces (de-Dios et al ., 2023).…”

“…However, purging of deleterious variants through purifying selection should lead to a decrease in strongly deleterious variants (LoF) in both homozygous and heterozygous state, whereas homozygosity of mildly (missense) or weakly (synonymous) can still increase and inflate the genetic load. The accumulation of genetic load in small populations has been observed in many vertebrate populations (see references in Dussex et al 2023) and in the Xerces blue butterfly (de-Dios et al, 2023). However, there has also been evidence documentation of purging of the most deleterious variants and accumulation of less deleterious variants in the same system in Alpine Ibex (Grossen et al, 2020) and Indian tigers (Khan et al, 2021) and is supported by simulations (Dussex et al, 2023).…”

“…There was a reduction in genome-wide heterozygosity in British specimens of both species, but reductions (compared to European populations) in both the extirpated A. crataegi and the extant and stable P. icarus were roughly comparable (17.9 and 14.5%, respectively). As a comparison, the observed reduction in genome-wide heterozygosity in the extinct Glaucopsyche xerces (Xerces Blue butterfly; de-Dios et al ., 2023) was 22%, hence greater than what we observed for British A. crataegi . However, the former was calculated in comparison to a sister species ( Glaucopsyche lygdamus ) and hence it is unclear whether these estimates of reductions in genome-wide heterozygosity are strictly comparable.…”