Microsatellite genotyping
To document temporal and spatial nuclear genetic variation in H. horticola wasps, we genotyped a total of 324 wasps (124 males/194 females/6 unknown) using the 14 microsatellite-loci developed by Couchoux et al. (Couchoux, Seppä, & van Nouhuys, 2015b). The forward primers were labelled with either FAM, HEX, or TAMRA fluorescent dye (DNA Technology A/S), and used in multiplex non-overlapping PCR reactions using Qiagen Multiplex PCR kit as described by Couchoux et al. (Couchoux et al., 2015b). Diluted PCR products were genotyped on an automated ABI 3730 DNA Sequencer (Applied BiosystemsTM, USA). The sizes were called using Genescan-500 ROX size standard. We manually curated the genotypes for each sample using the GeneMapper® Software 5 (Applied BiosystemsTM, USA). Several samples were independently genotyped multiple times to clarify uncertainties in the genotypes.
For the purpose of the genetic analyses described below, we separated the samples into the five localities they were collected from (North Finström, North Föglö, Seglinge-Kumlinge, Saltvik or Sottunga, Figure 1b), or into 12 spatio-temporal groups according to their geographic origin and their collection time interval (Finström 1992-97, 2003-08, 2009-11, Föglö 2000-09, 2010-13, Seglinge-Kumlinge 2000-09, 2010-11, Saltvik 1999, 2005-09, 2010-13, and Sottunga 2002-04, 2005-09, Figure 3, Table 2). Each of the 12 spatio-temporal groups was designed in an attempt to include specimens from either the 90’s, the 2000’s, and the 2010’s from each locality. Due to the lack of samples from other decades, the samples from Sottunga were divided between the early and the late 2000’s groups.
We analysed the genetic structure of wasps collected from the five localities and the spatio-temporal groups using two independent spatial Bayesian clustering analyses, with the ‘clustering of groups of individuals’ settings implemented in the BAPS software (Corander, Marttinen, Sirén, & Tang, 2008; Corander, Waldmann, & Sillanpää, 2003). We ran two spatial Bayesian clustering analyses with respective admixture analyses (Figure 3) to evaluate the degree of admixture at each locality, and in each locality through time. We used the software GenoDive 2.0b27 (Meirmans & van Tienderen, 2004) to calculate population genetics measures using only the samples that were successfully genotyped for at least seven microsatellite markers (N=309). Hymenoptera males are haploid while females are diploid. The analyses described above were thus conducted twice: (1) with both sexes included, but all males considered as homozygote diploids in order to fit the input format of the software used to define the different genetic clusters/genotypes, and (2) with only the females, to avoid overestimating the impact of the haploid males. We tested Hardy-Weinberg equilibrium at each locus in each locality and for each of the 12 spatio-temporal groups (Table 3). We also calculated the FST-values between localities and between the 12 spatio-temporal groups. Finally, using only the female samples, we calculated the observed (HO ) and expected heterozygosity (HE ) at each locus, as well as the inbreeding coefficient (Fis ).