Museum genomics suggests long-term population decline in a putatively extinct bumble bee.

Abstract

Pollinator declines globally threaten ecosystem stability and agricultural productivity. Reconstructing pollinator historic demographies provides an evolutionary perspective to understand contemporary population declines. The Franklin bumble bee (Bombus franklini), once endemic to Oregon and California and last observed alive in 2006, is emblematic of this phenomenon. We collected whole-genome sequence data from museum specimens spanning four decades to elucidate the genetic and demographic history of this potentially extinct species. Heterozygosity estimates of 25 individuals were remarkably low, and runs of homozygosity (ROH) patterns identified short segments suggestive of historical inbreeding, with some individuals having almost entire chromosomes in ROH. Demographic reconstructions revealed a marked decline in effective population size beginning in the late Pleistocene, with further declines in the last 400 y, which may have been influenced by fire and drought stressors. We found little to no genomic evidence implicating pathogens in the species' decline and used coalescent simulations to show that we would be able to detect recently reduced heterozygosity only when colony-level survival rates are 15 to 30%. We conclude that a combination of historically low effective population size and genetic diversity along with environmental stochasticity heightened this species' extinction vulnerability prior to recent anthropogenic stressors. This study demonstrates the utility of museum collections for clarifying genetic and demographic dynamics of rare species and suggests that B. franklini may have already been on a trajectory of decline prior to human impacts.