Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Apr 27;33(4):1719–1736. doi: 10.1111/1744-7917.70050

Genetic differentiation and adaptive evolution of buff‐tailed bumblebees in Asia

Long Su 1,2, Lele Ding 3, Paul H Williams 4, Yan Liu 1,2, Ruijuan Wang 1,2, Xiaoyan Dai 1,2, Shan Zhao 1,2, Haolin Fu 1,2, Xiaomeng Zhao 5, Quangui Wang 6, Yancan Li 1, Huiling Sang 7, Robert M Waterhouse 8,9, Yifan Zhai 1,2,✉, Cheng Sun 10,✉
PMCID: PMC13440993  PMID: 40287856

Abstract

Bumblebees are important pollinating insects, so their declines resulting from environmental change have received intensive attention. Understanding how environmental factors shape the genetic structure of natural populations and identifying the genetic basis of local adaptation will provide insights into how bumblebees cope with environmental change. The buff‐tailed bumblebee (Bombus terrestris) has a wide natural distribution range and has been domesticated to produce commercial colonies for greenhouse pollination. Previous population genetics studies on B. terrestris have mainly focused on its European populations; however, its Asian populations, representing the eastern side of its natural distribution, remain largely unsampled. To fill this gap, we collected wild B. terrestris samples from Asia, as well as wild B. terrestris samples from Europe and samples from domesticated colonies. We conducted whole‐genome resequencing for 77 collected B. terrestris and performed population genomics analysis. Our results indicate that distinct genetic differentiation (F ST = 0.076) exists between B. terrestris in Europe and Asia, with substantial morphological and physiological differences detected between them; B. terrestris in Asia should represent a distinct genetic resource. Demographic analysis suggests that the effective population size of B. terrestris has increased during historic cold periods, indicating their cold‐adapted characteristics. Selective sweep analysis identified 331 genes under selection in the genomes of Asian B. terrestris, likely involved in their adaptation to the high ultraviolet radiation, low temperature and low precipitation of their habitats. Our research provides insights into the population genetic structure and genetic basis of local adaptation in the buff‐tailed bumblebee, which will be useful for its conservation and management.

Keywords: adaptation, Bombus terrestris, population genomics, positive selection

This study investigates the genetic differentiation and local adaptation of the buff‐tailed bumblebee (Bombus terrestris) across its European and Asian populations. Using whole‐genome resequencing, we identified distinct genetic differences between populations, with Asian B. terrestris representing a unique genetic resource. Selective sweep analysis revealed key genes linked to adaptation to harsh environmental conditions, such as high UV radiation and low temperatures. These findings enhance our understanding of how environmental factors shape the genetic structure of bumblebee populations, aiding in their conservation and management efforts.

graphic file with name INS-33-1719-g007.jpg

Introduction

Bumblebees (genus Bombus) are ecologically and economically important pollinating insects, with approximately 250 extant species worldwide (Williams et al., 2008). They have long been known as excellent pollinators of greenhouse tomato crops, decreasing the cost of labor and improving both the yield and the quality of the fruits (Velthuis & van Doorn, 2006). Today their pollination services have been extended to various other crops grown in polytunnels and open fields (Velthuis & van Doorn, 2006; Garibaldi et al., 2013; Martin et al., 2019). Bumblebees are also ecologically important pollinators, serving as the sole or predominant pollinators of large numbers of wild plants (Fontaine et al., 2005; Goulson et al., 2008).

However, the declines of bumblebees in abundance and distribution, caused by global and local environmental changes, pose a threat to both wild plants and crop pollination (Williams et al., 2008b; Cameron et al., 2020). This decline has raised concerns regarding the implications for bumblebees, food security and ecosystem stability (Bartomeus et al., 2013; Cameron & Sadd, 2020; Soroye et al., 2020). Understanding how environmental factors shape the genetic differentiation of natural populations and identifying the genetic basis of local adaptation will provide insights into how species respond to environmental changes, which is fundamental to bumblebee conservation and breeding (Hoffmann & Sgrò, 2011; Franks & Hoffmann, 2012; Savolainen et al., 2013; Sang et al., 2024).

The buff‐tailed bumblebee, Bombus terrestris (Linnaeus, 1758), is one of the most widely used pollinator species for greenhouse crop pollination and has been artificially reared by commercial companies since the 1980s (Velthuis & van Doorn, 2006; Goulson, 2010). Bombus terrestris is abundant in the Western Palearctic, naturally distributed across the Atlantic islands, North Africa, Mediterranean islands, continental Europe, and the southern Ural Mountains, extending eastward to the Altai Mountains in western Mongolia (Estoup et al., 1996; Rasmont et al., 2008; Williams et al., 2018). Therefore, B. terrestris is a broadly distributed species that is exposed to diverse climatic conditions, which makes it a good model to study population differentiation and local adaptation.

Several studies have been performed to understand the population genetic structure and differentiation of B. terrestris, mainly focusing on continental European and nearby Mediterranean and Atlantic island populations (Estoup et al., 1996; Kraus et al., 2009; Lecocq et al., 2013; Moreira et al., 2015; Silva et al., 2020; Colgan et al., 2022). However, B. terrestris populations in Asia, representing the eastern edge of its global natural distribution, have been sampled only for morphological and cytochrome c oxidase subunit I (COI) barcode variation analyses (Williams et al., 2008).

To address this gap, we collected wild B. terrestris samples from 11 Asian geographic populations, located an average distance of 5100 km from Central Europe. Wild samples from Europe and domesticated samples from a commercial company were also collected. Whole‐genome resequencing was performed on these samples, generating data for population genomic analyses to infer the population genetic structure and demographic history of B. terrestris and to identify candidate genes involved in the local adaptation to its Asian habitats.

Materials and methods

Sample collection

Specifically, we utilized 3 sources of buff‐tailed bumblebee samples for this study: (1) Natural History Museum specimens – dry pinned B. terrestris specimens curated at the Natural History Museum in London represent one of the most comprehensive collections of this species from research institutes worldwide, with specimens collected between 1982 and 2011; (2) field surveys – to incorporate B. terrestris samples from China, our team conducted thorough field surveys from 2018 to 2020 at geographic sites where this species may occur based on previous reports and personal communications; and (3) commercial samples – we also included commercial samples imported from KOPPERT (Berkel en Rodenrijs, the Netherlands), a company that has specialized in the breeding and domestication of buff‐tailed bumblebees for over 30 years.

In total, we collected 44 B. terrestris samples across Asian regions, including 9 geographic sites in China (with an average distance of 254 km between each site) and 3 additional sites located in Tajikistan, Kyrgyzstan and the Asian part of Russia (see Table S1). This study marks the first effort to integrate B. terrestris samples from Asian regions into population genomics analysis, and there should now be sufficient latitudinal representation of the species distribution across Asia to achieve this.

Additionally, we analyzed 9 B. terrestris samples obtained from 6 European countries – France, Germany, Sweden, Switzerland, UK and the European part of Türkiye – as well as 24 commercial B. terrestris samples purchased from KOPPERT (Table S1).

DNA extraction

Prior to DNA extraction, all buff‐tailed bumblebee samples underwent surface cleaning with alcohol. DNA extraction was performed using the Gentra® Puregene® Tissue Kit (QIAGEN, Hilden, Germany), following standard protocols. For specimens obtained from the Natural History Museum, which were dry pinned samples collected several years ago, DNA was extracted from the hind legs, whereas DNA from freshly collected specimens was extracted from head tissue. Given that museum specimens may have suffered DNA damage (Dabney et al., 2013), DNA repair was conducted on DNA extracted from those samples using the NEBNext® FFPE DNA Repair Mix kit (New England Biolabs, NEB, Ipswich, MA, USA). In total, 77 genomic DNA samples of buff‐tailed bumblebees were successfully obtained for downstream sequencing and analysis.

DNA barcoding analysis

To verify the identity of the samples as B. terrestris, a barcoding PCR experiment was conducted following the protocol outlined by Hebert et al. (2004). The PCR primers were LEPr (5′‐TAAACTTCTGGATGTCCAAAAAATCA‐3′) and LEPf (5′‐ATTCAACCAATCATAAAGATATTGG‐3′). The reaction mixture (15 µL) consisted of 10× buffer with 15 mmol/L MgCl2 (1.5 µL), 10 mmol/L dNTPs (0.3 µL), each primer at 15 µmol/L concentration (1.0 µL each), Taq polymerase (5 U/µL) (0.2 µL), 25 ng/µL DNA template (2 µL) and double‐distilled water (ddH2O) (10.725 µL).

The PCR cycling conditions were as follows: initial denaturation at 95 °C for 5 min, followed by 33 cycles of denaturation at 94 °C for 30 s, annealing at 56 °C for 30 s, extension at 72 °C for 30 s and with a final extension at 72 °C for 5 min. The PCR products then underwent Sanger sequencing to obtain barcoding sequences. Finally, sequences of the 77 samples were submitted to BOLD Systems (http://www.boldsystems.org/index.php) for species identification.

Genome resequencing and read mapping

The quality and integrity of the genomic DNA were assessed using a nucleic acid analyzer and agarose gel electrophoresis. DNA samples of high quality, with good integrity and minimal degradation, were randomly fragmented. Libraries for paired‐end sequencing were constructed in accordance with Illumina sequencing platform requirements. Sequencing was performed using the Illumina HiSeq 2000 sequencer, generating paired‐end sequencing reads with a length of 150 base pairs (bp). Additionally, because some wild samples were museum specimens, to mitigate the effects of depurination reactions that may have occurred during the storage of museum specimens, we trimmed two bases from both ends of each read using FASTQ software (Chen et al., 2018).

Variant calling

Genomic sequencing data from B. terrestris samples were aligned with the reference genome (assembly Bter_1.0 GCA_000214255.1) (Sadd et al., 2015) using BWA‐MEM software (Li & Durbin, 2009a) with default parameters. The resulting alignments were converted into BAM format using SAMtools (Li et al., 2009b) and sorted accordingly. Duplicate reads were removed using the GATK MarkDuplicates module (McKenna et al., 2010). Sequencing depth and coverage were then evaluated using SAMtools and BEDtools (Quinlan & Hall, 2010). For variant calling (including single nucleotide polymorphisms, SNPs, and insertions–deletions, INDELs), the GATK tools HaplotypeCaller, CombineGVCFs, GenotypeGVCFs and SelectVariants were utilized. The variants obtained then underwent rigorous filtering using the GATK VariantFiltration module: quality by depth (QD) < 2.0; mapping quality (MQ) < 40.0; read position rank sum (RP) < −8.0; Fisher strand (FS) > 60.0; haplotype score (HS) > 13.0. SNP variant annotations were generated using ANNOVAR (Wang et al., 2010) and SnpEff (Cingolani et al., 2012).

Kinship analysis

High genomic homogeneity among individuals from the same colony or closely related bees can lead to errors in population genomics analysis. To address this, we employed the KING‐robust method (Manichaikul et al., 2010) using SNP information across the whole genome to compute pairwise kinship coefficients among the 77 samples. As in most bumblebees the queens mate once with a single male, whereas the males can mate with multiple queens (Boomsma & Ratnieks, 1996; Nasei et al., 1998; Schmid‐Hempel & Schmid‐Hempel, 2000), and we applied such parameters in the KING software. Individuals with kinship coefficients of >0.354 were considered full siblings, whereas those with coefficients ranging between 0.177 and 0.354 were considered half‐siblings. Therefore, during sample filtering, individuals with kinship coefficients of ≥0.177 were treated as belonging to the same colony, and only a single individual per colony was retained for subsequent analyses. This resulted in a total of 67 individuals being retained for downstream analysis.

Population genetic structure analysis

To construct the phylogenetic tree of B. terrestris samples, we used Bombus ignitus, which belongs to the same subgenus as B. terrestris, as an outgroup (Williams et al., 2008a). The B. ignitus data were downloaded from the US National Center for Biotechnology Information (NCBI) (PRJNA659133) (Sun et al., 2021). Genome‐wide SNP data from 67 B. terrestris samples and a single B. ignitus sample were converted to phylip files using Plink software (Chang et al., 2015). The data were then used to build a maximum likelihood (ML) tree with randomized axelerated maximum likelihood (RAxML) software (Stamatakis, 2014). Bootstrap analysis with 1000 replicates was performed, and an RAxML_bestTree was generated. The final phylogenetic tree was visualized using FigTree software (http://tree.bio.ed.ac.uk/software/figtree/).

Principal component analysis (PCA) was conducted on genome‐wide SNP data from 67 B. terrestris samples. The data were first formatted into ped and map files using Plink software. PCA analysis was performed using the smartPCA module within EIGENSOFT (Patterson et al., 2006). The results were visualized using scatter plots generated with the ggplot2 package in R.

Using structure analysis, we employed an ML estimation algorithm assuming Hardy–Weinberg equilibrium. Structure assigns B. terrestris samples to distinct populations or admixed groups based on their genetic variation. This approach offers insights into the genetic composition and admixture patterns across various values of K, shedding light on the evolutionary history of B. terrestris populations. The results from the structure analysis were visualized using the ggplot2 package in R, aiding in the interpretation of clustering patterns and genetic admixture among individuals.

Gene flow tests

The ABBA‐BABA test, also known as a D‐test (D‐statistics), is a widely used method in population genetics to detect significant deviations caused by admixture events. To examine gene flow among Asian wild populations, European wild populations and the commercial (domestic) populations of B. terrestris, the ABBA‐BABA test was employed, with B. ignitus used as the outgroup. The analysis included two main combination models: ([Wild‐Europe, Domestic], Wild‐Asia, Outgroup) and ([Domestic, Wild‐Europe], Wild‐Asia, Outgroup). These models were cross‐validated to ensure the robustness of the results. Z‐scores with absolute values of >3 indicate statistically significant gene flow.

TreeMix analysis using whole‐genome allele frequency data was then used to infer population splits and admixture events (Pickrell & Pritchard, 2012). To detect gene flow among different populations of B. terrestris, TreeMix analysis was also employed on the 67 B. terrestris samples, using B. ignitus as the outgroup. Wild populations of B. terrestris were grouped by country of collection (France, Germany, Sweden, Switzerland, Türkiye, UK, Tajikistan, Kyrgyzstan, Russia and China) to examine gene flow patterns between countries.

Pairwise sequentially Markovian coalescent (PSMC) analysis

The PSMC model (Li & Durbin, 2011) is a method used to estimate the time to the most recent common ancestor by assessing the density of heterozygous sites across a diploid genome, which has been used to estimate long‐term demography in many species, including domesticated organisms (Qiu et al., 2015; Wu et al., 2018; Chen et al., 2021). In our study, we utilized the PSMC model to estimate historical population changes in Asian and European populations. Given the stringent requirements of this analysis for sequencing depth (>20×) and extensive coverage, we removed individuals with low coverage. We utilized 41 high‐quality resequencing datasets for Asian samples of B. terrestris and 14 high‐quality resequencing datasets for the European samples.

To correlate the historical dynamics of B. terrestris populations with past climatic conditions, we analyzed how the population size of bumblebees was influenced by climatic changes. We employed mass accumulation rates (MARs) of loess deposits from the Loess Plateau over the past 3 million years as a proxy for global historical temperature changes (Sun & An, 2005). MAR values serve as indicators of global climate patterns: whereas lower MAR values correspond to warm and humid climates, higher values indicate cold and arid conditions.

Analysis of climatic data

We analyzed climatic data from the sampling points of 15 wild Asian bumblebees and 9 wild European bumblebees, based on their geographic coordinates. Using ArcMap 10.2, climatic data for each sampling point were obtained from the WorldClim database (https://www.worldclim.org/). The main climatic variables included minimum temperature, annual mean temperature, maximum temperature, precipitation, elevation and solar radiation. All climatic data were categorized into two groups based on geographic coordinates: Asia and Europe. Subsequently, we used R to conduct Wilcoxon rank‐sum tests on each climatic variable to assess whether there were any significant differences between the Asian and European population locations.

Phenotypic characterization

The morphological traits selected for measurement in this study were chosen because they have been used in the taxonomy of Apidae species and for their potential adaptive functions, and included the proboscis length, forewing length (FL), forewing breadth (FB), cubital vein A, cubital vein B, wing vein angles (A4, B4, D7, E9, G18, J10, J16, K19, L13, N23, O26, new angle), hooks on the right hindwing, femur length (Fe) of the right hind leg, tibia length (Ti) of the right hind leg, metabasitarsus length (ML), metabasitarsus breadth (MT), length of the third abdominal sternum (S3), breadth of the third abdominal sternum, breadth of the fourth abdominal sternum, length of the genitalia and breadth of the genitalia (Table S4); the measured body parts are illustrated in Fig. S2. The samples used for morphological trait measurement included: (1) Asian samples – wild queens were collected from Xinjiang, China, and were reared in laboratory conditions to produce offspring; (2) European samples – commercial queens imported from Europe were reared in laboratory conditions to produce offspring. Notably, queens derived from Asia and queens derived from Europe were reared under the same conditions (24 °C, 50% humidity) and were fed with the same sugar water and pollen. An equal number of B. terrestris from both Asian and European populations (4 queens, 8 workers and 6 males) were used for each measurement. These traits were measured using a stereomicroscope (VHX‐6000; Keyence, Osaka, Japan) and the Micro Measurement and Data Analysis System 1.01 (Keyence).

To measure the density of setae on the face and mesothorax of buff‐tailed bumblebees, we selected 3 representative areas of approximately 0.4 mm2 and counted the puncture density in each area. Measurements were conducted on 4 workers from Asia and 4 workers from Europe. In some cases, particularly for specimens with high setae density or setae that formed clumps through handling during capture and/or preservation, setae at their insertion points (usually indicated by a micropore on the cuticle) were counted. An additional advantage of counting micropores is that this can be applied to specimens that have lost setae (e.g. through aging; Southwick, 1985; or through poor handling) to obtain a measure of the original setae coverage.

Bumblebee workers frequently encounter variable environmental conditions while foraging among flowers. Measuring the supercooling points of these workers can offer valuable insights into how bumblebees physiologically adapt to their habitats. We utilized the SUN‐V Intelligent Insect Supercooling Point Tester (SUN Company, Jinan, China) to determine the supercooling and freezing points of Asian and European populations of B. terrestris. Our sample included 23 randomly selected workers from each population: Asian workers were sourced from colonies established by wild queens collected in Xinjiang, China, whereas European workers came from colonies formed by commercial queens imported from Europe to China. Notably, both sets of queens were reared under identical laboratory conditions. Phenotypic data were correlated with supercooling point data for association and difference analysis.

Detection of positive selection in Asian B. terrestris populations

Forty‐one B. terrestris specimens collected from Xinjiang, China, and 3 specimens collected from Tajikistan, Kyrgyzstan, and Russia represent the Asian population (collection sites in red in Fig. 1A). The European population included 9 B. terrestris specimens collected from France, Germany, Sweden, Switzerland, Türkiye and the UK (collection sites in green in Fig. 1A), along with a single individual of domesticated B. terrestris bought from a European company.

Fig. 1.

Fig. 1

Open in a new tab

Sampling sites and population genomic analyses of Bombus terrestris individuals sampled from across Asia and Europe. (A) Geographic locations of the 53 wild‐sampled B. terrestris individuals (red, collection sites in Asia, green, collection sites in Europe). (B) Maximum‐likelihood phylogenetic tree based on single nucleotide polymorphisms (SNPs) using B. ignitus as the outgroup. (C) Principal component analysis (PCA) plot showing Wild‐Asian (red), Wild‐European (green) and Domestic (yellow) samples on PC1 and PC3. (D) Structure analysis results (K = 2−4) for 67 of the 77 wild and domestic B. terrestris individuals (10 domestic samples excluded based on kinship analysis results). JingS(2025)G0387.

To identify selected genomic regions in the Asian population, we employed 3 metrics: (1) the ratio of genetic diversity between European and Asian populations (ln‐ratio π Europe/ π Asia); (2) the population differentiation coefficient F ST between European and Asian populations; and (3) the composite likelihood ratio (CLR) values in Asian the population. Notably, the European population contains both wild and commercial samples. The ratio of genetic diversity and the population differentiation coefficient F ST were calculated with vcftools 4.2 (Danecek et al., 2011) using a sliding window approach, with a window size of 10 kb and a step size of 5 kb. CLR values were calculated with SweeD (Pavlidis et al., 2013) using a grid size of 1 kb. Based on results from the analyses described above, the genetic diversity ratios (ln‐ratio π Europe/ π Asia), F ST values and CLR values were ranked to identify genomic regions falling within the top 5% for all these 3 metrics (Qiu et al., 2015), which were deemed as regions putatively under selection in the Asian B. terrestris population.

Functional prediction of focal genes

To predict the functions of genes under selection in the Asian population, we first identified the genomic locations of these genes using the genome annotation file (Annotation Release 102). Subsequently, we extracted their amino acid sequences and used InterProScan (Jones et al., 2014) to assign Gene Ontology (GO) terms for each protein‐coding gene. The results of the GO annotation were then subjected to enrichment analysis using the R Package TopGO (Alexa & Rahnenfuhrer, 2024). In addition, we utilized the Kyoto Encyclopedia of Genes and Genomes (KEGG) Orthology Based Annotation System (KOBAS) (Xie et al., 2011) to perform KEGG pathway annotation and enrichment analysis for these genes. Finally, we performed BLAST searches using the amino acid sequences of these genes against the fruit fly database, FlyBase (Marygold et al., 2016), to identify homologous genes in Drosophila. Through this approach, we could infer potential functions of these B. terrestris genes based on their counterparts in fruit flies.

Results

Genome resequencing of buff‐tailed bumblebees collected from Europe and Asia

A total of 77 B. terrestris samples were collected from Asia and Europe for this analysis (Fig. 1A). The identity of all these wild samples was confirmed by DNA barcoding (Hebert et al., 2004). In addition to the wild samples, 24 commercial (domesticated) B. terrestris samples were included in this study, bought from KOPPERT, a company based in the Netherlands.

We performed whole‐genome resequencing for these samples and achieved a mean depth of coverage of 30.80× and a mean breadth of coverage of 99.37% (Table S2). Notably, specimens from the Natural History Museum, London, yielded an average depth of 20.99× (ranging from 11.3× to 33.74×) and a mean breadth of coverage of 98.94% (ranging from 96.9% to 99.54%) (Table S3). The Genome Analysis Toolkit (Mckenna et al., 2010) was used to call SNPs and INDELs for these 77 buff‐tailed bumblebees. Based on the SNPs obtained, we estimated kinship coefficients for these bumblebees and found that 10 commercial samples were full‐ or half‐sisters to other individuals (first‐degree sibling, 0.177−0.354; Manichaikul et al., 2010). As a result, these samples were excluded from further analyses. Following this exclusion, we obtained a total of 4 827 687 high‐quality SNPs and 615 899 high‐quality INDELs for downstream analysis.

Distinct genetic differentiation exists between Asian and European buff‐tailed bumblebees

To understand the genome‐wide relationships and divergence among different B. terrestris samples, we constructed the ML tree using these 67 resequenced individuals, with B. ignitus (belonging to subgenus Bombus) used as the outgroup (Fig. 1B).

The phylogenetic tree based on genome‐wide SNPs revealed that the 67 B. terrestris individuals could be divided into two groups: (1) the first group, named Wild‐Asia, consisted of 44 wild B. terrestris samples collected from China, Tajikistan, Kyrgyzstan and Russia; (2) the second group comprised 9 wild samples collected from France, Germany, Sweden, Switzerland, Türkiye and the UK (named Wild‐Europe) plus 14 domesticated samples (named Domestic) (Fig. 1B). In addition, the close relationship observed between the Domestic and Wild‐Europe B. terrestris samples (Fig. 1B) was consistent with the fact that the commercial B. terrestris colonies have been domesticated from wild B. terrestris in Europe (Velthuis & van Doorn, 2006). The relationships inferred from the phylogenetic analysis were supported by Bayesian clustering analysis using PCA (Fig. 1C) and ADMIXTURE analysis (K = 3) (Figs. 1D and S1). There was clear genetic differentiation between B. terrestris samples originating from Asia and those from Europe.

To quantify the genetic differentiation between Asian and European B. terrestris, we calculated pairwise F ST values (genome‐wide mean F ST). The results showed that the pairwise F ST value between Wild‐Asia and Wild‐Europe was 0.076404 (Table 1), indicating that there is moderate genetic differentiation between these groups (Spitze, 1993).

Table 1.

Population genetic diversity summary statistics

Population π Tajima's D F ST
Wild‐Asia Domestic
Wild‐Europe 0.003117 −0.02952 0.076122 0.060032
Wild‐Asia 0.001877 1.8451 − 0.13277
Open in a new tab

Buff‐tailed bumblebees in Asia have no detectable gene flow with Europe

We calculated genetic diversity summary statistics and the results revealed that the nucleotide diversity (π) value for the Wild‐Europe group (π = 0.002906) was slightly higher than that for the Wild‐Asia group (π = 0.002276) (Table 1); however, as European samples are mainly dry pinned museum specimens and Asian samples are freshly collected, we cannot rule out the possibility that differences in decay between European and Asian samples are likely to contribute to the differences detected at the genetic level. The Wild‐Europe group (Tajima's D = −0.027101) had a lower Tajima's D value than that of the Wild‐Asia group (Tajima's D = 1.100971) (Table 1).

To test the possibility that populations of Asian buff‐tailed bumblebees were impacted by the invasion of commercial B. terrestris colonies (imported from Europe), we assessed gene flow between each population, specifically between the commercial samples and the samples collected from Xinjiang Province. First, we used TreeMix (Pickrell & Pritchard, 2012) to infer the migration history among populations (Wild‐Asia, Wild‐Europe and Domestic) by using the joint allele frequencies among all populations in our study. TreeMix employs an ML approach to model the genetic relationships among populations and detect admixture events. Our analysis did not reveal any gene flow between Domestic samples and Wild‐Asia samples (Fig. 2); gene flow was found between Domestic samples and Wild‐Europe samples, which is expected as the former was domesticated from the latter; gene flow was also found between Tajikistan and China‐Xinjiang samples, which is expected as they are geographically adjacent (Figs. 1A and 2A, B). To check this gene flow analysis between populations, we used the ABBA‐BABA gene flow test. The results confirmed that no gene flow was observed between Wild‐Asia (Xinjiang) samples and Domestic samples (Fig. 2C, D).

Fig. 2.

Fig. 2

Open in a new tab

Gene flow analysis for Bombus terrestris samples collected from different geographic locations. (A) TreeMix gene flow analysis between the three main B. terrestris groups. (B) TreeMix gene flow analysis between different collection sites. (C, D) Results from the ABBA‐BABA gene flow test. The topology of Europe_Dom_Asia is ([Wild‐Europe, Domestic], Wild‐Asia, Outgroup), and Dom_Europe_Asia is ([Domestic, Wild‐Europe], Wild‐Asia, Outgroup). (E) Correlation analysis of the genetic and geographic distances among B. terrestris samples.

Finally, we calculated the geographic and genetic distances between wild B. terrestris samples from Europe and Asia. We found that the genetic distance between these samples follows a strict isolation‐by‐distance model (R = 0.76, P < 2.2e−16), and a correlation between genetic distance and geographic distance was found among individuals (Fig. 2E), suggesting that dispersal could explain the current genetic distances between wild B. terrestris samples in Europe and Asia.

The demographic history of buff‐tailed bumblebees

Analysis using the PSMC model (Li & Durbin, 2011) was employed to infer the historical effective population size (Ne) of buff‐tailed bumblebees. The results (Fig. 3) indicate that buff‐tailed bumblebees in Asia and Europe exhibited similar and relatively stable Ne values until approximately 0.3 million years (Myr) ago. The effective population sizes of buff‐tailed bumblebees in these two regions began to diverge at approximately 0.3 Myr ago, roughly at the onset of the Penultimate Glacial Period (Zheng et al., 2002). After this, both buff‐tailed bumblebee populations exhibited increasing Ne values and B. terrestris in Europe showed a consistently higher Ne than that of B. terrestris in Asia.

Fig. 3.

Fig. 3

Open in a new tab

Demographic history of Asian and European Bombus terrestris populations inferred by pairwise sequentially Markovian coalescent (PSMC) software (g = 1,  µ = 0.36 × 10−8). The gray shading marks the Penultimate Glacial Period (0.13–0.3 Myr ago). The gray line represents the mass accumulation rate (MAR) of the Loess Plateau: high MAR values indicate cold and dry conditions, whereas low MAR values indicate hot and wet conditions.

The gray line in Fig. 3 represents the MAR of the Loess Plateau (Sun & An, 2005), which serves as an indicator of climatic conditions, with high MAR values indicating cold and dry periods and low MAR values indicating hot and wet periods. Comparing the climate and effective population size (Ne) estimates over time, we observed two main trends: (1) the increase in Ne between 0.13 and 0.30 Myr ago coincided with the Penultimate Glaciation Period (0.13–0.3 Myr ago) (Zheng et al., 2002), when the climate became cold and dry; (2) during Marine Isotope Stage 5 (MIS5, 0.08–0.13 Myr ago), a period characterized by a hot and wet climate according to the MAR estimates, we observed a decrease in Ne for B. terrestris in both Europe and Asia. Therefore, we infer from the demographic analysis that changes in Ne of buff‐tailed bumblebees are associated with historical climate change, with an increasing Ne during cold periods and a declining Ne during hot periods.

In addition, from the results (Fig. 3) we found that: (1) historical climate change had a more pronounced impact on European populations compared with those in Asia, probably owing to the more severe climate fluctuations during the Last Glacial Maximum (LGM) in Europe (Tallavaara et al., 2015); (2) a marked expansion of Ne occurred exclusively in European populations during the LGM. This continental‐scale divergence suggests that cold and arid LGM conditions selectively favored population growth in Europe – consistent with known cold‐adaptation mechanisms in species of the Northern Hemisphere (Hewitt, 2000; Excoffier et al., 2009). Conversely, the lack of Ne expansion in Asia can be attributed to the comparatively milder glacial conditions in the region, which did not impose the necessary selective pressures to drive expansion in this cold‐adapted genus.

Buff‐tailed bumblebees in Asia and Europe differ in morphology and physiology

To understand whether some environmental factors significantly differ between the two habitats (Europe and Asia) of buff‐tailed bumblebees, we collated and compared the environmental factors associated with their habitats. Specifically, environmental data on temperature, elevation, solar radiation and precipitation were extracted from the global natural environment database (WorldClim) for the 24 B. terrestris collection sites in Europe and Asia. After comparison, we found that several environmental factors significantly differed between the two habitats of B. terrestris (Fig. 4). Both the mean and minimum temperatures of Asian habitats were significantly lower than that of European habitats (P Ann‐Mean‐Temp = 0.02942; P Mean‐Temp‐Cold = 0.02942; P Min‐Temp = 0.009444). Additionally, the elevation and sun radiation in Asian habitats were significantly higher than in European habitats (P Elevation = 0.007926; P Sun‐Radiation = 0.004594). Moreover, the precipitation in Asian habitats was significantly lower than in European habitats (P Precipitation = 0.001733).

Fig. 4.

Fig. 4

Open in a new tab

Comparison of climate factors between the habitats of Asian and European Bombus terrestris, which include: (A) annual mean temperature; (B) mean temperature of coldest month; (C) minimum temperature of coldest month; (D) elevation; (E) solar radiation; and (F) annual precipitation. P‐values from Wilcoxon rank‐sum tests are indicated for each comparison.

To understand whether buff‐tailed bumblebees from the two habitats exhibit any phenotypic differences, we measured and compared a selection of morphological characters. The results indicate that buff‐tailed bumblebees in Asia had a shorter proboscis length compared with their European counterparts (Fig. 5B, E). The proboscis lengths of Asian queens (n = 4), workers (n = 8) and males (n = 6) were 9.75 ± 0.44, 6.75 ± 1.11 and 8.33 ± 1.44 mm, respectively, whereas these lengths for European queens (n = 4), workers (n = 8) and males (n = 6) were 11.74 ± 0.85, 9.24 ± 0.81 and 10.03 ± 0.72 mm, respectively. We also measured the setae density of the buff‐tailed bumblebees, which refers to the number of small pores within an area of 0.04 cm in diameter over a surface, and found that Asian samples exhibited a higher setae density (Fig. 5A, C, D, G). The setae densities on the face and mesothorax of Asian bumblebees (n = 4) were 775.1 ± 39.6 and 863.4 ± 50.2 setae/mm2, respectively, whereas the corresponding densities for European bumblebees (n = 4) were 675.97 ± 78.75 and 594.5 ± 53.56 setae/mm2, respectively. However, as we conducted these measurements using bumblebees reared in the laboratory, we cannot rule out the possibility that artificial rearing conditions – particularly pronounced for commercial B. terrestris – may have contributed to the observed morphological differences. All the morphometric characters are presented in Table S4.

Fig. 5.

Fig. 5

Open in a new tab

The morphological and physiological differences detected between Asian and European Bombus terrestris. The overall appearance (A), proboscis (B), setae density on the face (C) and setae density on the mesothorax (D) of Asian B. terrestris. (E) Comparative analysis of proboscis lengths for the queens, workers and drones of Asian and European B. terrestris. (F) The freezing points and supercooling points of Asian and European B. terrestris. (G) The setae density on the face and mesothorax of Asian and European B. terrestris. P‐values: *P < 0.05; **P < 0.01; ***P < 0.005.

The ambient temperature of the Asian habitats of buff‐tailed bumblebees was lower than that of the European buff‐tailed bumblebees (Fig. 5F), so Asian B. terrestris may have an enhanced capacity for cold tolerance. The phenomenon of supercooling refers to an insect's ability to maintain a liquid state even when temperatures fall below its freezing point. This ability is common among insects and plays a vital role in their survival in cold environments (Renault et al., 2002; Lombardo & Elkinton, 2017). As the supercooling point and freezing point were important parameters related to the survival of insects in cold environments, we conducted measurements of the supercooling point and freezing point for buff‐tailed bumblebees from Asia and Europe. Results suggested that, as expected, the supercooling point and freezing point of Asian buff‐tailed bumblebees were significantly lower than those of their European counterparts (Fig. 5F). The supercooling point of B. terrestris from Asia (n = 23) was −10.17 ± 1.26 °C and the freezing point was −7.02 ± 1.5 °C. In contrast, the supercooling point of European domestic B. terrestris (n = 23) was −8.85 ± 1.04 °C, with a freezing point of −5.58 ± 1.34 °C.

Genomic signatures of local adaptation for buff‐tailed bumblebees in Asia

To identify genomic regions that experienced positive selection in Asian B. terrestris, we employed three metrics (described earlier in the text). A total of 181 genomic regions were identified as being under selection in Asian buff‐tailed bumblebees, within which there are 331 protein‐coding genes (Fig. 6A; Table S7).

Fig. 6.

Fig. 6

Open in a new tab

Genomic regions and genes showing evidence of positive selection in Asian Bombus terrestris. (A) Distribution of ln‐ratio π Europe/ π Asia and fixation statistics F ST of 10‐kb windows with 5‐kb steps. Red dots represent windows fulfilling the selected requirements for the region (based on the top 5% for ln‐ratioπ Europe/π Asia [>0.6635] and fixation statistics F ST [>0.196113]). (B) The plot shows the Gene Ontology (GO) categorization of the identified genes in significant pathways (P < 0.05). The color of the circle represents the −log P‐value. The size of the circle represents the number of genes under selection in the enriched pathway. (C) Two focal genes – Eiger and Wengen – showing selection signals in Asian B. terrestris. ln‐ratio π Europe/ π Asia, F ST and SweeD (>0.686) values are plotted using a 10‐kb sliding window.

We performed GO and KEGG enrichment analysis for the 331 genes and the results indicate that signal transduction (GO:0007165), cell communication (GO:0007154), the mitogen‐activated protein kinase (MAPK) signaling pathway, the glycerophospholipid (GPL) metabolism pathway and the apoptosis pathway (Fig. 6B; Tables S5 and S6) were significantly enriched. Notably, both Eiger and Wengen genes in the MAPK signaling pathway and Apoptosis pathway were identified as being under selection in the Asian population (Figs. 6C and 7). Previous reports have indicated that Eiger and Wengen genes could elicit adaptive changes in response to intense ultraviolet (UV) radiation; additionally, studies demonstrated that Eiger can influence cellular insulin‐mediated nutritional responses in Drosophila, thereby enhancing the adaptability of these insects to varying nutritional environments (Agrawal et al., 2016). Furthermore, other studies have revealed the crucial involvement of the GPL metabolic pathway in the low‐temperature adaptation of the insect Pyrrhocoris apterus (Han & Bauce, 1995; Tomcala et al., 2006).

Fig. 7.

Fig. 7

Open in a new tab

The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway for mitogen‐activated protein kinase (MAPK) signaling. Genes in the MAPK signaling pathway that exhibited patterns of positive selection in Asian Bombus terrestris are highlighted in red.

Discussion

The buff‐tailed bumblebee (B. terrestris) is an important pollinator of wild plants and is one of the most widely used pollinator species for greenhouse crop pollination (Velthuis & van Doorn, 2006; Goulson, 2010). Intensive studies have been performed to understand its population structure, genetic differentiation and adaptation using samples from European and nearby Mediterranean and Atlantic island populations (Estoup et al., 1996; Kraus et al., 2009; Lecocq et al., 2013; Moreira et al., 2015; Silva et al., 2020; Colgan et al., 2022). However, less was known about B. terrestris populations in Asia, which represent the eastern edge of its global natural distribution (Williams et al., 2008a). In this study, we collected wild B. terrestris samples from 11 geographic populations in Asia, as well as wild samples from Europe and domesticated samples from a commercial company based in Europe. After whole‐genome resequencing of these samples, we investigated their population genetic structure, demographic history and local adaptation.

Our results indicate that buff‐tailed bumblebees could be divided into two groups: samples originating from Asia were clustered into one group and samples originating from Europe were clustered into another group (Fig. 1). That is, previous studies on buff‐tailed bumblebees largely missed the diversity represented by Asian populations. In addition, distinct genetic differentiation was detected between Asian and European populations (Table 1), with an F ST value of 0.076404, indicating moderate genetic differentiation (Spitze, 1993). Moreover, buff‐tailed bumblebees in Asia and Europe differ in morphology and physiology. For example, compared with their European counterparts, buff‐tailed bumblebees collected in Asia (more specifically, Xinjiang Province, China) had a shorter proboscis length, a higher setae density, and a lower supercooling and freezing point (Fig. 5; Table S4). Taken together, and considering the moderate genetic differentiation and distinct morphological/physiological differences, buff‐tailed bumblebees in Asia could be considered substantially divergent from those in Europe, and they should represent a distinct genetic resource.

Local adaptation is an important ability required for organisms to survive in changing environments (Fuse, 2017). Both the mean and minimum temperatures of B. terrestris habitat in Asia were significantly lower than that in Europe (Fig. 4). Bumblebees employ a variety of strategies to adapt to low‐temperature challenge, including the use of their body setae, which provide insulation and aid in heat retention, color patterns and muscle activity, used to generate heat (Heinrich, 1974; Jackson et al., 2020; Maebe et al., 2021). Bombus terrestris in Asia exhibited phenotypes that were detectably different from those in Europe, such as increased setae density (Fig. 5). The increased setae density in Asian B. terrestris could provide them with an enhanced ability to withstand cold temperatures. The phenomenon of supercooling refers to the ability of insects to maintain a liquid state even when the temperature drops to below their freezing point, which is widespread among insects and is crucial for their survival in cold environments (Renault et al., 2002; Lombardo & Elkinton, 2017). A comparison between Asian B. terrestris and European B. terrestris revealed that the former group had a lower supercooling point (Fig. 5), suggesting that B. terrestris in Asia likely tolerate lower temperatures.

Genes related to local adaptation will show positive selection signals within the genome. Here, we identified genes under positive selection in B. terrestris in Asia. Our results indicate that the MAPK pathway and GPL metabolic pathway in B. terrestris in Asia had undergone selection (Table S6). Specifically, we observed that Eiger and Wengen in these two pathways have been subject to positive selection (Table S7). Previous research has shown that Eiger, a homolog of tumor necrosis factor in fruit flies, is released in epidermal cells undergoing apoptosis in response to intense UV radiation; Wengen, on the other hand, acts as the receptor for Eiger on sensory neurons, mediating the transmission of adaptive signals that enable the organism to cope with UV radiation (Agrawal et al., 2016). Therefore, it is likely that Eiger and Wengen contributed to the adaptation of B. terrestris in Asia to the high elevation and UV‐intensive environments of their habitats. Additionally, studies have demonstrated that Eiger can modulate insulin‐mediated nutritional responses in fruit flies, enhancing their adaptability to changing nutritional environments (Agrawal et al., 2016). Considering the habitat characteristics of B. terrestris in Asia, such as low precipitation, high elevation and limited vegetation, leading to reduced food availability (Fig. 4), Eiger may play a role in enhancing the adaptability of Asian B. terrestris to their nutritional environments. Genes in the GPL metabolic pathway were also found to be under positive selection (Table S6). Previous studies have demonstrated the association between the GPL metabolism pathway and low‐temperature adaptation in the hemipteran insect P. apterus (Han & Bauce, 1995; Tomcala et al., 2006). Also, it was reported that GPLs play a crucial role in regulating the supercooling point in insects (Tomcala et al., 2006). Therefore, genes in the GPL metabolic pathway that showed selection signals are likely to have enhanced the cold tolerance of B. terrestris in Asia.

Conclusion

Through collecting and re‐sequencing buff‐tailed bumblebees from their Asian habitats, which were less well represented in previous studies, we performed the first multi‐locus population genetic structure analysis of this important pollinator. Our results revealed that buff‐tailed bumblebees in Asia dispersed from Europe and accumulated moderate genetic differentiation from their European counterparts. Buff‐tailed bumblebees in Asia have adapted to their habitats morphologically and physiologically, and we identified genes that were likely involved in their adaptation to the high UV radiation, low temperature and low precipitation of their habitats. The new genomic resources and insights revealed through this study will be valuable for the conservation and improved management of this important pollinator.

Disclosure

All authors have seen and agree with the contents of the article and report no conflicts of interest, including specific financial interests and relationships and affiliations relevant to work.

Supporting information

Table S1 Summary information for wild samples.

Table S2 Quality control data.

Table S3 Resequencing mapping data.

Table S4 Morphological parameters of Bombus terrestris from Asia and Europe.

Table S5 GO enrichment analysis of selected genes from Asian samples.

Table S6 KEGG enrichment analysis of selected genes from Asian samples.

Table S7 Selected genes from Asian samples.

INS-33-1719-s002.docx (54.4KB, docx)

Fig. S1 Admixture CV error for K values of 1‐6.

INS-33-1719-s003.pdf (4.7KB, pdf)

Fig. S2 Fig. S2 Phenotypic traits of Bombus terrestris measured.

INS-33-1719-s001.pdf (32MB, pdf)

Acknowledgments

This work was supported by the National Key R&D Program of China (2023YFF1103800), the National Natural Science Foundation of China (32270445 and 31971397), the Swiss National Science Foundation (SNSF) (202669 and 196125, to R.M.W.), Shandong Provincial Natural Science Foundation (ZR2021YQ21, ZR2021QC218); Shandong Provincial Agriculture Research System (SDAIT‐24); Taishan Scholars Program of Shandong Province (tsqn202312293); and Shandong Provincial Key R&D Program (2023LZGC017). This work was also supported by the Shandong Academy of Agricultural Sciences, Institute of Plant Protection, Shandong Key Laboratory for Green Prevention and Control of Agricultural Pests, and the Shandong Provincial Engineering Research Center for Development and Utilization of Beneficial Insects.

Contributor Information

Yifan Zhai, Email: saaszyifan@163.com.

Cheng Sun, Email: cheng.sun@cnu.edu.cn.

References

  1. Alexa, A. and Rahnenfuhrer, J. (2024) TopGO: enrichment analysis for gene ontology. R package version 2.56.0.
  2. Agrawal, N. , Delanoue, R. , Mauri, A. , Basco, D. , Pasco, M. , Thorens, B. et al. (2016) The Drosophila TNF Eiger is an adipokine that acts on insulin‐producing cells to mediate nutrient response. Cell Metabolism, 23, 675–684. [DOI] [PubMed] [Google Scholar]
  3. Bartomeus, I. , Ascher, J.S. , Gibbs, J. , Danforth, B.N. , Wagner, D.L. , Hedtke, S.M. et al. (2013) Historical changes in northeastern US bee pollinators related to shared ecological traits. Proceedings of the National Academy of Sciences USA, 110, 4656–4660. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Boomsma, J.J. and Ratnieks, F.L. (1996) Paternity in eusocial Hymenoptera. Philosophical Transactions of the Royal Society B – Biological Sciences, 351, 947–975. [Google Scholar]
  5. Cameron, S. and Sadd, B. (2020) Global trends in bumble bee health. Annual Review of Entomology, 65, 209–232. [DOI] [PubMed] [Google Scholar]
  6. Chang, C.C. , Chow, C.C. , Tellier, L.C. , Vattikuti, S. , Purcell, S.M. and Lee, J.J. (2015) Second‐generation PLINK: rising to the challenge of larger and richer datasets. GigaScience, 4, 7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chen, S. , Zhou, Y. , Chen, Y. and Gu, J. (2018) fastp: an ultra‐fast all‐in‐one FASTQ preprocessor. Bioinformatics, 34, i884–i890. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chen, Z.H. , Xu, Y.X. , Xie, X.L. , Wang, D.F. , Aguilar‐Gómez, D. , Liu, G.J. et al. (2021) Whole‐genome sequence analysis unveils different origins of European and Asiatic mouflon and domestication‐related genes in sheep. Communications Biology, 4, 1307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cingolani, P. , Platts, A. , Wang, L.L. , Coon, M. , Nguyen, T. , Wang, L. et al. (2012) A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff: SNPs in the genome of Drosophila melanogaster strain w1118; iso‐2; iso‐3 . Fly, 6, 80–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Colgan, T.J. , Arce, A.N. , Gill, R.J. , Ramos Rodrigues, A. , Kanteh, A. , Duncan, E.J. et al. (2022) Genomic signatures of recent adaptation in a wild bumblebee. Molecular Biology and Evolution, 39, msab366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dabney, J. , Knapp, M. , Glocke, I. , Gansauge, M.‐T. , Weihmann, A. , Nickel, B. et al. (2013) Complete mitochondrial genome sequence of a Middle Pleistocene cave bear reconstructed from ultrashort DNA fragments. Proceedings of the National Academy of Sciences USA, 110, 15758–15763. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Danecek, P. , Auton, A. , Abecasis, G. , Albers, C.A. , Banks, E. , Depristo, M.A. et al. (2011) The variant call format and VCFtools. Bioinformatics, 27, 2156–2158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Estoup, A. , Solignac, M. , Cornuet, J.M. , Goudet, J. and Scholl, A. (1996) Genetic differentiation of continental and island populations of Bombus terrestris (Hymenoptera: Apidae) in Europe. Molecular Ecology, 5, 19–31. [DOI] [PubMed] [Google Scholar]
  14. Excoffier, L. , Foll, M. and Petit, R.J. (2009) Genetic consequences of range expansions. Annual Review of Ecology Evolution and Systematics, 40, 481–501. [Google Scholar]
  15. Fontaine, C. , Dajoz, I. , Meriguet, J. and Loreau, M. (2005) Functional diversity of plant–pollinator interaction webs enhances the persistence of plant communities. PLoS Biology, 4, e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Franks, S. and Hoffmann, A. (2012) Genetics of climate change adaptation. Annual Review of Genetics, 46, 185–208. [DOI] [PubMed] [Google Scholar]
  17. Fuse, N. (2017) Genome research elucidating environmental adaptation: dark‐fly project as a case study. Current Opinion in Genetics and Development, 45, 97–102. [DOI] [PubMed] [Google Scholar]
  18. Garibaldi, L. , Steffan‐Dewenter, I. , Winfree, R. , Aizen, M. , Bommarco, R. , Cunningham, S. et al. (2013) Wild pollinators enhance fruit set of crops regardless of honey bee abundance. Science, 339, 1608–1611. [DOI] [PubMed] [Google Scholar]
  19. Goulson, D. (2010) Impacts of non‐native bumblebees in Western Europe and North America. Applied Entomology and Zoology, 45, 7–12. [Google Scholar]
  20. Goulson, D. , Lye, G. and Darvill, B. (2008) Decline and conservation of bumble bees. Annual Review of Entomology, 53, 191–208. [DOI] [PubMed] [Google Scholar]
  21. Han, E. and Bauce, É. (1995) Non‐freeze survival of spruce budworm larvae, Choristoneura fumiferana, at sub‐zero temperatures during diapause. Entomologia Experimentalis et Applicata, 75, 67–74. [Google Scholar]
  22. Hebert, P. , Penton, E. , Burns, J. , Janzen, D. and Hallwachs, W. (2004) Ten species in one: DNA barcoding reveals cryptic species in the neotropical skipper butterfly Astraptes fulgerator . Proceedings of the National Academy of Sciences USA, 101, 14812–14817. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Heinrich, B. (1974) Thermoregulation in endothermic insects. Science, 185, 747–756. [DOI] [PubMed] [Google Scholar]
  24. Hewitt, G. (2000) The genetic legacy of the Quaternary ice ages. Nature, 405, 907–913. [DOI] [PubMed] [Google Scholar]
  25. Hoffmann, A. and Sgrò, C. (2011) Climate change and evolutionary adaptation. Nature, 470, 479–485. [DOI] [PubMed] [Google Scholar]
  26. Jackson, J. , Pimsler, M. , Oyen, K. , Strange, J. , Dillon, M. and Lozier, J. (2020) Local adaptation across a complex bioclimatic landscape in two montane bumble bee species. Molecular Ecology, 29, 920–939. [DOI] [PubMed] [Google Scholar]
  27. Jones, P. , Binns, D. , Chang, H.Y. , Fraser, M. , Li, W. , McAnulla, C. et al. (2014) InterProScan 5: genome‐scale protein function classification. Bioinformatics, 30, 1236–1240. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kraus, F.B. , Wolf, S. and Moritz, R.F. (2009) Male flight distance and population substructure in the bumblebee Bombus terrestris . Journal of Animal Ecology, 78, 247–252. [DOI] [PubMed] [Google Scholar]
  29. Lecocq, T. , Vereecken, N.J. , Michez, D. , Dellicour, S. , Lhomme, P. , Valterová, I. et al. (2013) Patterns of genetic and reproductive traits differentiation in Mainland vs. Corsican populations of bumblebees. PLoS ONE, 8, e65642. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Li, H. and Durbin, R. (2009a) Fast and accurate short read alignment with Burrows–Wheeler transform. Bioinformatics, 25, 1754–1760. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Li, H. and Durbin, R. (2011) Inference of human population history from individual whole‐genome sequences. Nature, 475, 493–496. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Li, H. , Handsaker, B. , Wysoker, A. , Fennell, T. , Ruan, J. , Homer, N. et al. (2009b) The sequence alignment/map format and SAMtools. Bioinformatics, 25, 2078–2079. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Lombardo, J.A. and Elkinton, J.S. (2017) Environmental adaptation in an asexual invasive insect. Ecology and Evolution, 7, 5123–5130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Maebe, K. , Hart, A. , Marshall, L. , Vandamme, P. , Vereecken, N. , Michez, D. et al. (2021) Bumblebee resilience to climate change, through plastic and adaptive responses. Global Change Biology, 27, 4223–4237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Manichaikul, A. , Mychaleckyj, J.C. , Rich, S.S. , Daly, K. , Sale, M. and Chen, W.M. (2010) Robust relationship inference in genome‐wide association studies. Bioinformatics, 26, 2867–2873. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Martin, C. , Fountain, M. and Brown, M. (2019) Varietal and seasonal differences in the effects of commercial bumblebees on fruit quality in strawberry crops. Agriculture, Ecosystems & Environment, 281, 124–133. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Marygold, S.J. , Crosby, M.A. and Goodman, J.L. (2016) Using FlyBase, a database of Drosophila genes and genomes. In Drosophila. Methods in Molecular Biology (ed. Dahmann C.), pp. 1–31. Humana Press, New York, NY. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. McKenna, A. , Hanna, M. , Banks, E. , Sivachenko, A. , Cibulskis, K. , Kernytsky, A. et al. (2010) The Genome Analysis Toolkit: a MapReduce framework for analyzing next‐generation DNA sequencing data. Genome Research, 20, 1297–1303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Moreira, A. , Horgan, F. , Murray, T. and Kakouli‐Duarte, T. (2015) Population genetic structure of Bombus terrestris in Europe: isolation and genetic differentiation of Irish and British populations. Molecular Ecology, 24, 3257–3268. [DOI] [PubMed] [Google Scholar]
  40. Nasei, J. , Moinard, C. , Moreau, L. , Himpens, B. and Guyonnaud, S. (1998) Relationship between aging, mating and sperm production in captive Bombus terrestris . Journal of Apicultural Research, 37, 107–113. [Google Scholar]
  41. Patterson, N. , Price, A.L. and Reich, D. (2006) Population structure and eigenanalysis. PLoS Genetics, 2, e190. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Pavlidis, P. , Živkovic, D. , Stamatakis, A. and Alachiotis, N. (2013) SweeD: likelihood‐based detection of selective sweeps in thousands of genomes. Molecular Biology and Evolution, 30, 2224–2234. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Pickrell, J.K. and Pritchard, J.K. (2012) Inference of population splits and mixtures from genome‐wide allele frequency data. PLoS Genetics, 8, e1002967. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Qiu, Q. , Wang, L. , Wang, K. , Yang, Y. , Ma, T. , Wang, Z. et al. (2015) Yak whole‐genome resequencing reveals domestication signatures and prehistoric population expansions. Nature Communications, 6, 10283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Quinlan, A.R. and Hall, I.M. (2010) BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics, 26, 841–842. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Rasmont, P. , Copp, E. , Michez, D. and Meulemeester, T. (2008) An overview of the Bombus terrestris (L. 1758) subspecies (Hymenoptera: Apidae). Annales de la Société entomologique de France, 44, 243–250. [Google Scholar]
  47. Renault, D. , Salin, C. , Vannier, G. and Vernon, P. (2002) Survival at low temperatures in insects: what is the ecological significance of the supercooling point? Cryo Letters, 23, 217–228. [PubMed] [Google Scholar]
  48. Sadd, B.M. , Barribeau, S.M. , Bloch, G. , de Graaf, D.C. , Dearden, P. , Elsik, C.G. et al. (2015) The genomes of two key bumblebee species with primitive eusocial organization. Genome Biology, 16, 76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Sang, H. , Li, Y. , Tan, S. , Gao, P. , Wang, B. , Guo, S. et al. (2024) Conservation genomics analysis reveals recent population decline and possible causes in bumblebee Bombus opulentus . Insect Science, 31, 1631–1644. [DOI] [PubMed] [Google Scholar]
  50. Savolainen, O. , Lascoux, M. and Merilä, J. (2013) Ecological genomics of local adaptation. Nature Reviews Genetics, 14, 807–820. [DOI] [PubMed] [Google Scholar]
  51. Schmid‐Hempel, R. and Schmid‐Hempel, P. (2000) Female mating frequencies in Bombus spp. from Central Europe. Insectes Sociaux, 47, 36–41. [Google Scholar]
  52. Silva, S. , Seabra, S. , Carvalheiro, L. , Nunes, V. , Marabuto, E. , Mendes, R. et al. (2020) Population genomics of Bombus terrestris reveals high but unstructured genetic diversity in a potential glacial refugium. Biological Journal of the Linnean Society, 129, 259–272. [Google Scholar]
  53. Soroye, P. , Newbold, T. and Kerr, J. (2020) Climate change contributes to widespread declines among bumble bees across continents. Science, 367, 685–688. [DOI] [PubMed] [Google Scholar]
  54. Spitze, K. (1993) Population structure in Daphnia obtusa: quantitative genetic and allozymic variation. Genetics, 135, 367–374. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Stamatakis, A. (2014) RAxML version 8: a tool for phylogenetic analysis and post‐analysis of large phylogenies. Bioinformatics, 30, 1312–1313. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Sun, C. , Huang, J. , Wang, Y. , Zhao, X. , Su, L. , Thomas, G.W.C. et al. (2021) Genus‐wide characterization of bumblebee genomes provides insights into their evolution and variation in ecological and behavioral traits. Molecular Biology and Evolution, 38, 486–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Sun, Y. and An, Z. (2005) Late Pliocene‐Pleistocene changes in mass accumulation rates of eolian deposits on the central Chinese Loess Plateau. Journal of Geophysical Research, 110, D23101. [Google Scholar]
  58. Tallavaara, M. , Luoto, M. , Korhonen, N. , Järvinen, H. , and Seppä, H. (2015) Human population dynamics in Europe over the Last Glacial Maximum. Proceedings of the National Academy of Sciences, 112, 8232–8237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Tomcala, A. , Tollarová, M. , Overgaard, J. , Simek, P. , and Kostál, V. (2006) Seasonal acquisition of chill tolerance and restructuring of membrane glycerophospholipids in an overwintering insect: triggering by low temperature, desiccation and diapause progression. The Journal of Experimental Biology, 209, 4102–4114. [DOI] [PubMed] [Google Scholar]
  60. Velthuis, H. and van Doorn, A. (2006) A century of advances in bumblebee domestication and the economic and environmental aspects of its commercialization for pollination. Apidologie, 37, 421–451. [Google Scholar]
  61. Wang, K. , Li, M. and Hakonarson, H. (2010) ANNOVAR: functional annotation of genetic variants from high‐throughput sequencing data. Nucleic Acids Research, 38, e164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Williams, P.H. , Cameron, S.A. , Hines, H.M. , Cederberg, B. and Rasmont, P. (2008) A simplified subgeneric classification of the bumblebees (genus Bombus). Apidologie, 39, 46–74. [Google Scholar]
  63. Williams, P.H. , Lobo, J.M. and Meseguer, A.S. (2018) Bumblebees take the high road: climatically integrative biogeography shows that escape from Tibet, not Tibetan uplift, is associated with divergences of present‐day Mendacibombus . Ecography, 41, 461–477. [Google Scholar]
  64. Williams, S. , Shoo, L. , Isaac, J. , Hoffmann, A. , Langham, G. and Mathur, H. , (2008b) Towards an integrated framework for assessing the vulnerability of species to climate change. PLoS Biology, 6, e325. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Wu, D.D. , Ding, X.D. , Wang, S. , Wójcik, J.M. , Zhang, Y. , Tokarska, M. et al. (2018) Pervasive introgression facilitated domestication and adaptation in the Bos species complex. Nature Ecology & Evolution, 2, 1139–1145. [DOI] [PubMed] [Google Scholar]
  66. Xie, C. , Mao, X. , Huang, J. , Ding, Y. , Wu, J. , Dong, S. et al. (2011) KOBAS 2.0: a web server for annotation and identification of enriched pathways and diseases. Nucleic Acids Research, 39, W316–W322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Zheng, B. , Xu, Q. and Shen, Y. (2002) The relationship between climate change and Quaternary glacial cycles on the Qinghai–Tibetan Plateau: review and speculation. Quaternary International, 97, 93–101. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1 Summary information for wild samples.

Table S2 Quality control data.

Table S3 Resequencing mapping data.

Table S4 Morphological parameters of Bombus terrestris from Asia and Europe.

Table S5 GO enrichment analysis of selected genes from Asian samples.

Table S6 KEGG enrichment analysis of selected genes from Asian samples.

Table S7 Selected genes from Asian samples.

INS-33-1719-s002.docx (54.4KB, docx)

Fig. S1 Admixture CV error for K values of 1‐6.

INS-33-1719-s003.pdf (4.7KB, pdf)

Fig. S2 Fig. S2 Phenotypic traits of Bombus terrestris measured.

INS-33-1719-s001.pdf (32MB, pdf)

Articles from Insect Science are provided here courtesy of Wiley

RESOURCES