Multispecies genetic diversity investigations on Andrena using SSR-GBAS microsatellite analysis

Abstract

With about 150 species in central Europe, the genus Andrena covers a variety of ecological groups, comprising widely distributed generalists as well as specialists. Phylogenetic and taxonomic analyses of some species were inconclusive because mitochondrial loci used for barcoding can comprise only a low number of variable characters. In particular, for species which diverged recently, multilocus markers like microsatellites can be informative for the determination of species concepts, reproductive isolation and potential hybridization. In this study, we use genetic structure of multiple Andrena species to verify species concepts and investigate patterns of relationship. The main focus was on the cross-species functionality of the marker panel, which can be applied to differentiate on species and population level. Long-term, it could be used for cross-species comparisons of genetic diversity to investigate evolutionary ecology questions. We developed 46 microsatellite markers, 22 for the Andrena cineraria species complex and 24 for Andrena flavipes. The cross-species applicability was then tested on 37 selected Andrena species of different subgenera (555 samples). Genotyping was conducted using the SSR-GBAS approach (short sequence repeat-genotyping by amplicon sequencing), a high-throughput method using the Illumina technology. The marker panel discriminated most of the analysed 37 species and successfully resolved the A. cineraria species complex. Not all species in the subgenera Micrandrena and Notandrena could be differentiated, indicating that with increasing phylogenetic distance to the source of markers, alleles show lower levels of divergence. However, we identified one individual which appears to be a first-generation hybrid between A. danuvia and A. cineraria. While the microsatellites confirm general divergence between these species, occasional hybridization exists. Our investigation shows the suitability of genetic structure analysis to verify taxonomic concepts, especially for species groups with ambiguous barcoding outcomes. Wild bees are particularly threatened by land use change, causing habitat fragmentation and loss of population connectivity. Resulting genetic erosion might also lead to widespread bee decline. Our multispecies approach allows to implement comparative investigations on genetic diversity. It could be further developed into a genetic monitoring approach which can serve as an early warning system for the detection of species at risk.