The genetic structure of the earth bumblebee (Bombus terrestris L.) was examined across 22 wild populations and two commercially reared populations using eight microsatellite loci and two mitochondrial genes. Our study included wild bumblebee samples from six populations in Ireland, one from the Isle of Man, four from Britain and 11 from mainland Europe. A further sample was acquired from New Zealand. Observed levels of genetic variability and heterozygosity were low in Ireland and the Isle of Man, but relatively high in continental Europe and among commercial populations. Estimates of Fst revealed significant genetic differentiation among populations. Bayesian cluster analysis indicated that Irish populations were highly differentiated from British and continental populations, the latter two showing higher levels of admixture. The data suggest that the Irish Sea and prevailing south westerly winds act as a considerable geographical barrier to gene flow between populations in Ireland and Britain; however, some immigration from the Isle of Man to Ireland was detected. The results are discussed in the context of the recent commercialization of bumblebees for the European horticultural industry.
Table S1
Microsatellite loci in Bombus terrestris typed in multiplex reactions in this study. Table contains list of primers for respective loci as well as information on the labelling dye used.
Table S2
Table S2– Microsatellite allele frequencies in populations of Bombus terrestris from 22 European sites and two commercial populations. Population codes are represented in the first row and number of individuals genotyped for each microsatellite locus are presented in parentheses. Observed (Ho) and expected (He) heterozygosities and allelic richness are presented for each locus and population. The column on the left represents the allele sizes.
Table S3
Table S3 – Genetic differentiation between European and commercial populations of Bombus terrestris. Pairwise Fst is presented above the diagonal with corresponding Euclidean geographical distances (km) between sites presented below the diagonal
Table S4
Table S4 – STRUCTURE assignments of population probability to membership of K = 3 clusters and K = 7 clusters.
Table S5
Table S5 – Correlations between genetic and geographic distance, as measured by Rousset’s distance (Fst/1-Fst) and the natural logarithm of geographical separation for Bombus terrestris
Table S6
Table S6 – Migration rates among B. terrestris populations estimated using BAYESASS. Values shown are the means of the posterior distributions of m, the migration rate into each population, and their respective 95% CI in parenthesis. Values along the diagonal (in bold) are the proportion of individuals derived from the source population each generation. The expected mean and CI of non-migrants is 0.833 (0.675, 0.992), the expected mean and CI for migration is 0.013 (1.31x10-15, 0.104).
Figure S1
Figure S1 – Fifty-percent rule consensus tree showing relationships among Bombus terrestris based on maximum likelihood analysis and Bayesian inference analysis of a 633bp segment of the mitochondrial COI gene. The tree was rooted with Bombus ignitus and Apis mellifera. Numbers above the branch are clade credibility for maximum likelihood and for Bayesian analysis below the branch.
Figure S2
Figure S2– A) Estimated likelihood of the posterior probability, LnP(X|K), for values of K (the number of genetic clusters) ranging from 1 to 12 as calculated using STRUCTURE. The mean LnP(X| K) for each K over 20 permutations is represented by solid squares. Bars correspond to standard deviations over 20 permutations. B) modal value quantity of ∆K based on the second order rate change is calculated according to Evanno et al (2005).
Figure S3
Figure S3 – Proportional membership of 24 European Bombus terrestris populations to K=7 genetic clusters. Individuals are represented by a single vertical column divided into 7 (=K) colors. The relative lengths of the colored segments correspond to the individual’s estimated proportion of membership in that cluster.
Code used for Bayesian analysis
Code used for Bayesian analysis
Code used for maximum likelihood analysis
Code used for maximum likelihood analysis
Table 1
Abundance of bumblebee species (Bombus s.str) in assemblages sampled throughout Europe. Species were separated according to the RFLP-based method of Murray et al. (2008).
COI aligned sequences for all individuals
COI aligned sequences for all individuals
Table 2 - GPS locations of all sampled populations
GPS locations of all sampled populations
Table 3 - Full genotyped dataset of all populations of Bombus terrestris.
Full genotyped dataset of all populations of Bombus terrestris.
Abundance of bumblebee species (Bombus s.str) in assemblages sampled throughout Europe.
Abundance of bumblebee species (Bombus s.str) in assemblages sampled throughout Europe. Species were separated according to the RFLP-based method of Murray et al. (2008).
Table 1 - Abundance of bumblebee species (Bombus s.str) in assemblages sampled throughout Europe..doc
COI aligned sequences for all individuals_PAUP
COI aligned sequences for all individuals. PAUP input file.
COI aligned sequences for all individuals_MrBayes
COI aligned sequences for all individuals. MrBayes input file.
MrBayes Tree File
MrBayes output file.
Paup Tree File
Paup output file.