The worldwide spread of diseases is considered a major threat to biodiversity and a possible driver of the decline of pollinator populations, particularly when novel species or strains of parasites emerge. Previous studies have suggested that populations of introduced European honeybee (Apis mellifera) and bumblebee species (Bombus terrestris and Bombus ruderatus) in Argentina share the neogregarine parasite Apicystis bombi with the native bumblebee (Bombus dahlbomii). In this study we investigated whether A. bombi is acting as an emergent parasite in the non-native populations. Specifically, we asked whether A. bombi, recently identified in Argentina, was introduced by European, non-native bees. Using ITS1 and ITS2 to assess the parasite’s intraspecific genetic variation in bees from Argentina and Europe, we found a largely unstructured parasite population, with only 15% of the genetic variation being explained by geographic location. The most abundant haplotype in Argentina (found in all 9 specimens of non-native species) was identical to the most abundant haplotype in Europe (found in 6 out of 8 specimens). Similarly, there was no evidence of structuring by host species, with this factor explaining only 17% of the genetic variation. Interestingly, parasites in native Bombus ephippiatus from Mexico were genetically distant from the Argentine and European samples, suggesting that sufficient variability does exist in the ITS region to identify continent-level genetic structure in the parasite. Thus, the data suggest that A. bombi from Argentina and Europe share a common, relatively recent origin. Although our data did not provide information on the direction of transfer, the absence of genetic structure across space and host species suggests that A. bombi may be acting as an emergent infectious disease across bee taxa and continents.
Citation: Maharramov J, Meeus I, Maebe K, Arbetman M, Morales C, Graystock P, et al. (2013) Genetic Variability of the Neogregarine Apicystis bombi, an Etiological Agent of an Emergent Bumblebee Disease. PLoS ONE 8(12): e81475. https://doi.org/10.1371/journal.pone.0081475
Editor: Claude Wicker-Thomas, CNRS, France
Received: July 30, 2013; Accepted: October 22, 2013; Published: December 6, 2013
Copyright: © 2013 Meeus et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: GM and IM acknowledge support by the Research Council of Ghent University (BOF-UGent) (www.ugent.be), the Fund for Scientific Research-Flanders (FWO-Vlaanderen, Brussels, Belgium) (www.fwo.be). The Rufford Small Grants for Conservation to CLM and MPA, The National Park Administration of Argentina for sampling permits. MJFB was supported by an IRCSET grant and grant BB/I000151/1 funded jointly by BBSRC, Defra, NERC, the Scottish Government and the Wellcome Trust, under the Insect Pollinators Initiative. Funding for this study was provided by the Natural Environment Research Council (NERC) to PG and WOHH (NE/G012113/1). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: Co authors of the submitting paper GS and WOHH are editors for PLOS ONE and this does not alter the authors' adherence to all the PLOS ONE policies on sharing data and materials.
Bumblebees (Bombus spp.) are essential generalist pollinators , but many of their populations are undergoing major losses. While range decline for some species has been moderate, others are vanishing rapidly , . Although the trend of decline is evident worldwide, the responsible drivers are diverse (for review see ), with each driver potentially interacting with other drivers and acting differently across geographic locations . This complexity makes it difficult to clarify the response to a single driver.
The relatively unspoiled temperate forests of southern Argentina and Chile are the natural habitat of Bombus dahlbomii, the largest bumblebee species in the world , [7,]. Populations of B. dahlbomii appear to be in a steep decline, which has coincided with the recent establishment of the non-native European bumblebee, Bombus ruderatus, and more recently of Bombus terrestris , . Bombus ruderatus was intentionally introduced into southern Chile in 1982–1983 from a New Zealand population, which was also an introduced population originating from the UK , and has subsequently migrated from Chile into Argentina , . Bombus terrestris became established in Chile after the introduction of commercially produced colonies from Belgium and Israel for crop pollination around 1998 . It has been hypothesized that the decline of B. dahlbomii may be partly due to pathogen spillover from introduced bees , and in particular that the introduced bumblebee B. terrestris, may be the carrier of novel pathogens into the environment.
Commercially produced bumblebee colonies have been shown by many studies to carry a wide range of microbial parasites , , , , and such an introduction of new pathogens could induce emergent infectious diseases with dramatic consequences for native populations . Apicystis bombi, a neogregarine parasite (Apicomplexa: Neogregarinorida), has been found to be present in Argentina infesting in native and non-native bumblebees, as well as the honeybee Apis mellifera , , , and thus it may be an important pathogen driving emerging infectious disease, and potentially involved in the decline of B. dahlbomii in this region. However, pathogenicity of A. bombi in B. dahlbomii needs to be determined in order to infer the risk associated with its spillover.
A. bombi was first discovered in Italy in 1988 and has now been recorded in nearly 20 Bombus species ,  including in commercially produced bumblebee colonies , . Although empirical data on the pathology of the parasite are limited, the fat body of infected bumblebees is destroyed due to the proliferation of the pathogen , and its presence correlates with high mortality in infected spring queens, preventing them from establishing colonies , . Consequently, upon entering novel host populations, A. bombi may have the potential to act as an emergent infectious disease. However, to date it is unknown whether the A. bombi present in the populations of European bumblebees and honeybees established in Argentina, was acquired in situ or if it was co-introduced from Europe with them. A previous study using microscopy found A. bombi in the invasive B. terrestris in northwest Patagonia, but not in native bumblebees from regions which are currently free of B. terrestris . Similarly, using molecular techniques, A. bombi-infected honeybees were only observed in regions invaded by B. terrestris, while the pathogen was not detected in honeybees from B. terrestris-free regions . However, B. terrestris-free regions were also geographically and climatologically different from the regions where A. bombi was found and, in the absence of epidemiological knowledge, these observations cannot by themselves definitively prove that A. bombi infections resulted from the introduction of B. terrestris. More recently, Arbetman et al. (2013) found that 14 out 30 B. terrestris, five out nine B. ruderatus and one out of nine B. dahlbomii specimens collected in northwest Patagonia after invasion by B. terrestris, were infected with the neogregarine A. bombi . Conversely, the pathogen could not be detected in any of the 30 of B. ruderatus and 52 B. dahlbomii museum specimens, collected before the invasion of B. terrestris . However, this too is not definitive, because detection limits in the ethanol-stored samples  and thus the original source of this parasite in South America is still controversial. Finally, recent molecular screening of honeybee hives (n = 363) in Europe (Belgium) found 40.8% to be positive for A. bombi  while in Japan two of 69 examined hives (2.9%) were positive . However, whether the A. bombi discovered in honeybees are of the same strain as those found in Bombus species, and whether interspecific transmission is possible, remains unknown.
In order to study the interspecific variability of A. bombi present in non-native bees that are established in Argentina, i.e. B. terrestris, B. ruderatus and Ap. mellifera, we sequenced the highly variable internal transcribed spacer 1 (ITS1) and ITS2 regions as genetic markers. We focused on two questions: (i) do honeybees and bumblebees share the same haplotypes with A. bombi, and (ii) do parasite strain haplotypes found in Argentina and Europe exhibit geographical structure? Answering these will provide insight into the transmission dynamics and native range of A. bombi.
Materials and Methods
Origin of Infected A. bombi Species
To determine the haplotypes of A. bombi present in Argentina, we collected nine A. bombi-positive specimens: B. terrestris (n = 5), B. ruderatus (n = 2) and Ap. mellifera (n = 2). All samples were collected in northwest Patagonia , . The European sampling consisted of eight A. bombi-infected specimens: Bombus pratorum from Ireland (n = 3; , B. terrestris from Belgium (n = 1) and the UK (n = 3), and Ap. mellifera from Belgium (n = 1). Two extra specimens were analyzed: one B. terrestris from a commercially produced colony in Europe , and one B. ephippiatus native from Mexico (Figure 1). All samples from Argentina were ethanol-stored , , and the other samples were stored at −20°C before extraction. All samples were extracted with EZNA® Insect DNA kit (Omega Bio-Tek; Norcross, GA)  with the exception of the samples from the UK which were extracted with a Chelex extraction protocol.
Overview of bumblebee (Bombus pratorum, Bombus terrestris, Bombus ephippiatis and Bombus ruderatus) and honeybee (Apis mellifera) samples included in the study, including the numbers of bees from each species at each location. All bees were infected with Apicystis bombi and the haplotypes found in each bee are given (UNI, EUR1-5, ARG1, ARG2, MEX1 and MEX2).
No national permissions were required to collect samples from the public lands in the locations of Belgium and the UK. Specific permission granted for sampling by the National Botanic Gardens, Ireland, and The Royal Parks, UK. Permits were not required in Argentina for the collection of exotic invertebrates (B. terrestris, and Ap. mellifera) from public lands. Permission to collect samples on his private land was provided by Luis Rovera in Argentina. Permission to collect bumblebees (B. terrestris and B. ruderatus) from Argentine National Parks were obtained from Administracion de Parques Nacionales (Argentina), while permission to export them from Argentina were granted by Secretaria de Desarrollo Sustentable y Medio Ambiente. The field studies did not involve endangered or protected species.
Haplotyping of A. bombi
In order to determine the haplotypes of A. bombi, we sequenced both ITS1 and ITS2, which are both highly variable regions of the rRNA. Species-specific primers were designed: ApiITS732F 5′-TGGAAACAAGTCATTTTTGGAA-3′ and ApiITS732R 5′-CCTGTTCACTCGCCGTTACT-3′, amplifying an approximately 730 bp long fragment of A. bombi within a bumblebee DNA extract. The amplicon contained ITS1, 5.8S rDNA, and ITS2. PCR reactions were done in 25 µl-reaction volumes, containing 0.2 mM dNTPs, 0.5 µM primers, 1×PCR buffer, 1.5 mM MgCl2, 1.25 units of Taq polymerase (Taq DNA Polymerase, recombinant/Invitrogen) and 1 µl DNA template. Amplification was performed using one cycle of 94°C for 3 min, followed by 30 cycles of 95°C for 30 s, 55°C for 30 s, and 72°C for 60 s, and a final elongation step of 3 minutes.
PCR products were cloned using the Sticky-End Cloning Protocol (pJET1.2/blunt Cloning Vector, Fermentas UAB, subsidiary of Thermo Fisher Scientific Inc. V. Graiciuno 8 LT-02241, Vilnius, Lithuania). The plasmid was purified with the E.Z.N.A. Plasmid Mini Kit (Omega Bio-Tek) from 10 bacterial colonies containing the ITS regions, and sequenced (LGC Genomics GmbH (Germany, Berlin). In order to analyze the different haplotypes, the sequences were aligned using the BioEdit sequence alignment editor. Not all mutations can be regarded as a different haplotype. It is known that PCR artifacts can occur, with a PCR error rate of ε = 0.555×10−3 (1.85×10−5 miss incorporations per cycle for 30 cycles) and the probability that a single mutation is caused by PCR error for an amplicon with size 730 bp is P = 0.063. This calculation is based on the binomial probability mass function reported by [26,]. To be conservative, we decided to exclude all mutations which occurred only once in the total dataset. Of the remaining mutations, all haplotypes were confirmed at least three independent times. This procedure resulted in eight different haplotypes in our total data set (n = 18). The exclusion of possible PCR errors could lead toward an under-estimate of the true diversity present. Hence PCR products harboring 3 unique mutations (P = 0.0008) were subsequently re-included in the dataset. In this way, we incorporated two additional haplotypes, both found within the single Mexican sample. The ITS1 contained three mononucleotide repeats, one of 10 nucleotides in length and two of more than 10 nucleotides. It is known that these regions are, because of polymerase slippage, sensitive to mutation errors . All sequences deposited at GenBank contain the consensus amount of nucleotides plus an N to indicate the unknown amount of extra nucleotides.
Genetic Variation and Structuring of A. bombi
The pairwise mutational differences between haplotypes was calculated. In order to reveal population structuring, the genetic variation within versus among populations was determined using an analysis of molecular variance (AMOVA; Arlequin v3.1). Differences in the average within-population sequence divergence of haplotypes of Europe and Argentina were tested using a Mann-Whitney U test. To reconstruct the phylogenetic tree and estimate evolutionary relationships of the haplotypes, the Neighbor-Joining method was used . Evolutionary analyses were conducted in MEGA5 . Reduced Median (RM) networks were drawn to visualize the evolutionary relationship between different haplotypes using NETWORK 4.6.1 . The Median Joining (MJ) algorithm resulted in the same topology.
Bombus spp. and Ap. mellifera are Infected with the Same Species of Apicystis
We collected neogregarine infected bees (specimens) from three different locations: Argentina, Europe and Mexico. After sequencing analysis of a part of the 18S region we confirmed that all hosts were infected with A. bombi. All were identical to previous samples of A. bombi found in Argentine B. terrestris, B. ruderatus, B. dahlbomii  and Ap. mellifera , but also to A. bombi found in B. pratorum in Europe . One mutation in the 18S was observed when comparing it with the 18S fragment found in the native Mexican bumblebee B. ephippiatus (KC951279).
With the use of A. bombi specific primers, located in the 18S rRNA and 26S rRNA, we were able to sequence ITS1 and ITS2 (GenBank KF322207–KF322216). From each specimen (n = 18) we sequenced 10 different clones; in total seven bees contained one unique haplotype, nine contained two haplotypes and only two bees contained 3 haplotypes. Most specimens (16 out of 18) therefore contained only one or two different haplotypes, indicating that there is little or no intragenomic variability present in A. bombi. When analyzing the molecular variance of the complete data set, 34% of the variability in the data was explained by variation within one specimen (Table 1a).
To study A. bombi transmission among different host species, we investigated the Argentine subsample (n = 9). Here three different species were sampled in close proximity to each other (samples were at most 25 km apart). We determined whether haplotypes were distributed randomly among these three different species (B. terrestris, n = 5; B. ruderatus, n = 2; and Ap. mellifera, n = 2). The two most common haplotypes were present in all three host species (Figure 1, 2). The most common haplotype (UNI) was detected in all nine Argentine bees, while the other prevalent haplotype (ARG1) was also found in seven of these specimens (Figure 1, 2). Only 17.3% of the total genetic diversity of A. bombi was explained by differences among host species. It should be noted though, that this percentage is a consequence of the higher sampling of B. terrestris resulting in a unique haplotype (ARG2) for this species, and the majority of the genetic diversity was found within host species: 5.5% among specimens of one host species and 77.2% within a specimen (Table 1b).
The size of the node represents the relative abundance of a haplotype and the different colors indicate the proportion of samples with a haplotype that was present in Bombus terrestris (blue), Bombus ruderatus (yellow) or Apis mellifera (green). The black squares represent unobserved single-nucleotide substitutions.
Argentine and European A. bombi Share the Same Origin
We classified A. bombi haplotypes according to their geographic location, and also included an infected B. terrestris sample from a European commercially produced bumblebee colony (see ). We found three haplotypes in Argentina (ARG1, ARG2 and UNI, in nine hosts), six haplotypes in Europe (EUR1–EUR5 and UNI, in eight hosts), two haplotypes in Mexico (MEX1 and MEX2, in a single host), and two haplotypes in a commercially produced colony (EUR1 and EUR2). Haplotypes EUR1, EUR2, EUR3, EUR4 and EUR5 were only found in European samples, ARG1 and ARG2 were only found in Argentine samples, and the haplotypes MEX1 and MEX2 were only found in Mexico (Figure 3a). The universal haplotype UNI was detected in all Argentine samples, and in 75% of the European samples. The other prevalent haplotype ARG1 occurred in 78% of the samples in Argentina while it remained undetected in our sampling in Europe. Furthermore, as shown in figure 3b, the haplotypes found in Europe and Argentina formed a single phylogenetic clade. The genetic variation among geographic regions (Argentina and Europe) only explained 15.2% of the total genetic variation, with most of the variation found among and within specimens of each region (43% and 42%, respectively) (Table 1c). The genetic structure between Europe and Argentina was therefore small and there was no significant phylogenetic structure across these regions (Figure 3b). The two haplotypes found within the commercially produced bumblebee colony also fell within this European/Argentine cluster.
(a) Bees in Europe (blue), Argentina (green), Mexico (red), and a commercially produced bumblebee colony in Europe (yellow). Circle sizes are proportional to the numbers of bees infected with the haplotype and the circle colors indicate the proportion of bees with the haplotype that came from each location. The black squares represent unobserved single-nucleotide substitutions. (b) The geographic location in which each haplotype was detected is indicated with colored spots for Europe (blue), Argentina (green), Mexico (red), and a commercially produced bumblebee colony in Europe (yellow). The optimal tree with the sum of branch length = 0.0176 is shown. The percentages of replicate trees in which the associated haplotypes clustered together in the bootstrap test (1000 replicates) are shown next to the branches.
Genetic Diversity of A. bombi in Argentina and Europe
The mean sequence divergence of A. bombi among specimens from Argentina (mean ± s.d. = 0.7±0.6, n = 9) was significantly smaller than the mean sequence divergence among European specimens (1.5±1.3, n = 8; Mann Whitney U test: z = −4.98; P<0.001). Indeed, only three haplotypes were found among Argentine specimens separated by one or two mutations, while the European haplotypes were more diverse (Figure 3).
The ITS Region can be used to Haplotype A. bombi
In this study we used ITS1 and ITS2 as a molecular marker to study genetic variability of A. bombi. Although these markers have been widely used, particularly because of their high level of sequence variation, there are caveats regarding their use . Parasites have multiple rDNA loci which makes intragenomic diversity possible, as for example reported for the bumblebee parasite Nosema bombi . Because of concerted evolution, intragenomic diversity is generally low . However, different paralogs are possible, making ITS markers uninformative if intragenomic diversity is higher than the intergenomic diversity within a population. In our data, 34% of the variability was within individuals. This percentage may reflect intragenomic diversity or infection with multiple strains. Although we cannot differentiate between these two explanations, we speculate that there is little intragenomic diversity for A. bombi parasites for two reasons. First, 16 out of 18 specimens contained either one (7out of 16 specimens) or two (9 out of 16 specimens) haplotypes (across 10 clones). Second, it is often reported that multiple parasite strains infect a single host , a phenomenon reported for other bumblebee parasites like Crithidia bombi , .
Honeybees and Bumblebees Share A. bombi Haplotypes
Our results indicate that honeybees and the two European bumblebee species, B. terrestris and B. ruderatus, sampled from the same area in Argentina, were all infected by the same haplotypes of A. bombi. This parasite was originally described as a bumblebee parasite, implicated in increased mortality rates in infected queens (B. pratorum; ). These authors found that the parasite re-emerges in bumblebee workers late in the season, suggesting that some queens survive infection and transmit the parasite to the progeny and/or alternate reservoirs exist for the parasite .
Recent studies suggest that Ap. mellifera could represent an alternative reservoir and/or vector of A. bombi , , [24,]. Although we have demonstrated that the same parasite species -and haplotypes within the species- is present in both taxa, we cannot conclude whether A. bombi found in honeybees can indeed re-infect bumblebees. Furthermore, it still needs to be determined whether the A. bombi found in Argentina represented genuine infections of Ap. mellifera. The fact that honeybee and bumblebee populations are largely sympatric throughout the geographic range of Bombus, which is a first requirement for parasites to jump between host species , makes the presence of shared pathogens probable. Indeed, several honeybee parasites are now known to infect bumblebees , , , , , , showing that species jumps between bumblebees and honeybees do not necessarily result in a dead-end host. Furthermore, recent work has confirmed that A. bombi from bumblebees can readily infect honeybees . Although knowledge about multi-host parasites is ‘scarce’, the high rate of interspecies transmission plays a central role in the evolution of virulence in host-parasite networks .
The Same A. bombi Haplotypes are Shared between Argentina and Europe
Our fundamental question was whether A. bombi detected in Argentina is indigenous or from introduced bees originating from Europe. If A. bombi were indigenous in two geographically separated continents, two clearly differentiated clusters should be seen, where one contains parasites of Argentina and the other of Europe. Two separate clusters would also imply that the introduced European bees became infected with A. bombi already present in Argentina. Instead we found that the European parasites clustered together with those from Argentina. However, we note that the detection of only one cluster could be the consequence of insufficient sampling in Argentina and/or Europe. Higher sample numbers could reveal separating clusters. However, the homogeneity in the data between Europe and South America strongly suggests that parasites identified in the analyzed samples share the same origin.
We detected low geographic structuring between Europe and Argentina, with only 15.2% of the genetic variation being explained by location. Indeed both locations share the most frequent haplotype (UNI). The haplotypes detected in Argentina (ARG1 and ARG2) are one and two point mutations different from UNI, respectively (Figure 3a). This actually makes them indistinguishable from most of the haplotypes (EUR1, EUR2, EUR3 and EUR4) exclusively found in Europe. Thus the differences detected between Europe and Argentina are of the same magnitude as the those found within European samples. Geographical structuring in parasites between land masses separated by an ocean has been observed in other taxa; a good example is Atractolytocestus huronensis (Cestoda: Caryophyllidea), an invasive parasite of common carp introduced from North America to Europe, which exhibits at least 8 mutations or 1.2% sequence diversity in ITS2 of the two geographic locations . We recognize here that one has to be careful when comparing mutational rates of the ITS region of different species. Ideally, reference samples of A. bombi from isolated locations without non-native bumblebees or honeybees would be required to quantify the natural genetic structuring of A. bombi within its natural habitats. With the intense and worldwide transport of honeybees and bumblebees such samples are very difficult to obtain. However, it is evident from our data that the weak structuring of A. bombi between European and Argentine bumblebees does not reflect the historical Palearctic, Nearctic, and Neotropic separation of the Old World Bombus ancestor . Rather, it supports a common and more recent origin, which is consistent with the hypothesis that introduced European bees (be it Apis and/or Bombus) carried Apicystis into Argentina.
The A. bombi haplotypes found in B. ephippiatus, native to Mexico, are more distantly related, demonstrating that the ITS region of A. bombi is variable enough to identify differentiation at these spatial scales where present. These data also represent the first record of A. bombi in this country. A. bombi was originally described as Mattesia bombi ,  and has been found in Ontario (Canada) , suggesting its presence in North America before bumblebee transport started. However, the single sample from Mexico is insufficient to draw any conclusion regarding genetic differentiation within Mexico. The native B. ephippiatus is found in a region where non-native commercially produced bumblebees, Bombus impatiens, have become established , . It is uncertain whether the A. bombi haplotypes in the Mexican bee represent indigenous, North American haplotypes, or haplotypes spilling over into Mexico from other locations.
The Mexico sample does further emphasize the need for a worldwide prevalence and haplotyping study in which regions with intense bumblebee and/or honeybee importation are compared with regions without importation.
Geographic Transmission Routes of A. bombi and Perspectives
Our molecular haplotyping strongly suggests that the origin of the A. bombi currently found in non-native Argentine bees is the same as those found in Europe. What mechanisms could explain this? A first possibility is that A. bombi originated from Europe and was subsequently introduced into Argentina either by Ap. mellifera, B. terrestris or B. ruderatus. We have one infected bumblebee sample originating from a commercially-produced bumblebee colony, and its A. bombi haplotypes fall within the European/Argentine cluster. This again indicates there is a shared origin of A. bombi, but remains uninformative about the original native region of A. bombi. Furthermore, the original native locality of commercially produced bumblebees is not always clear, and thus also of the parasites they carry. The two subspecies most commonly used in European breeding factories are B. t. terrestris and B. t. dalmatinus, with the former originating from a variety of European countries and the latter from Greece and Turkey . In addition, Israeli facilities also supplied bumblebees to Chile. Together with the fact that A. bombi has been detected in Turkey , this makes the Mediterranean region a good candidate to retrieve the original location of the ARG1, ARG2 haplotypes.
A second possibility is that Ap. mellifera became infected with A. bombi from a South American host and global honeybee queen transport subsequently homogenized the genetic variation of this parasite. Although this seems a less probable explanation, it is a possibility that the current results cannot exclude. At a minimum, however, A. bombi can be considered as an emergent disease in Argentina, in either the native or non-native bees depending on the direction of transfer, with the implication that it may have also rapidly increased in prevalence in other regions where non-native honeybees and bumblebees have been introduced. It has been recognized that emergent diseases of wild populations pose a substantial threat to the conservation of global biodiversity , . With bee diversity already being threatened by multiple factors , monitoring of this emergent infectious disease is crucial. Here we present molecular tools to study intra- and inter-species diversity to untangle the spread of this parasite. In order to detect the exact origin and transmission dynamics of this parasite, multiple locations now need to be sampled worldwide, especially the Mediterranean region, preferably including regions with no honeybee and/or bumblebee imports. In parallel, pathological studies are also urgently needed to assess the virulence of the parasite across native and non-native hosts.
Conceived and designed the experiments: JM IM GM. Performed the experiments: JM IM KM. Analyzed the data: JM IM KM MA CM PG WOHH SP CEL DCDG JJDLR TEM MJFB GM. Contributed reagents/materials/analysis tools: MA CM PG WOHH SP CEL DCDG NZ JJDLR TEM MJFB. Wrote the paper: JM IM KM MA CM PG WOHH SP CEL DCDG NZ JJDLR TEM MJFB GM.
- 1. Goulson D (2010) Bumblebees: their behaviour, ecology and conservation; Goulson D, editor. New York: Oxford university press.
- 2. Cameron SA, Lozier JD, Strange JP, Koch JB, Cordes N, et al. (2011) Patterns of widespread decline in North American bumble bees. Proc Natl Acad Sci U S A 108: 662–667.
- 3. Potts SG, Biesmeijer JC, Kremen C, Neumann P, Schweiger O, et al. (2010) Global pollinator declines: trends, impacts and drivers. Trends Ecol Evol 25: 345–353.
- 4. Goulson D, Lye GC, Darvill B (2008) Decline and conservation of bumble bees. Annu Rev Entomol 53: 191–208.
- 5. Brown MJF, Paxton RJ (2009) The conservation of bees: a global perspective. Apidologie 40: 410–416.
- 6. Abrahamovich AH, Díaz NB (2001) Distribución geográfica de las especies del género Bombus Latreille (Hymenoptera, Apidae) en Argentina. Rev Bras Entomol 45: 99–117.
- 7. Abrahamovich AH, Diaz NB, Morrone JJ (2004) Distributional patterns of the Neotropical and Andean species of the genus Bombus (Hmenoptera: Apidae). Acta Zool Mex 20: 99.
- 8. Morales CL, Arbetman MP, Cameron SA, Aizen MA (2013) Rapid ecological replacement of a native bumble bee by invasive species. Front Ecol Environ:
- 9. Ruz L (2002) Bee pollinators introduced to Chile. Pollinating Bees - The Conservation Link Between Agriculture and Nature - Ministry of Environment/Brasília: 155–167.
- 10. Arretz PV, Macfarlane RP (1986) The introduction of Bombus ruderatus to Chile for red-clover pollination. Bee World 67: 15–22.
- 11. Arbetman MP, Meeus I, Morales CL, Aizen MA, Smagghe G (2013) Alien parasite hitchhikes to Patagonia on invasive bumblebee. Biol Invasions 15: 489–494.
- 12. Colla SR, Otterstatter MC, Gegear RJ, Thomson JD (2006) Plight of the bumble bee: Pathogen spillover from commercial to wild populations. Biol Conserv 129: 461–467.
- 13. Meeus I, Brown MJF, De Graaf DC, Smagghe G (2011) Effects of invasive parasites on bumble bee declines. Conserv Biol 25: 662–671.
- 14. Graystock P, Yates K, Evison SEF, Darvill B, Goulson D, et al. (2013) The Trojan hives: pollinator pathogens, imported and distributed in bumblebee colonies. J Appl Ecol 50: 1207–1215.
- 15. Murray TE, Coffey MF, Kehoe E, Horgan FG (2013) Pathogen prevalence in commercially reared bumble bees and evidence of spillover in conspecific populations. Biol Conserv 159: 269–276.
- 16. Plischuk S, Lange CE (2009) Invasive Bombus terrestris (Hymenoptera: Apidae) parasitized by a flagellate (Euglenozoa: Kinetoplastea) and a neogregarine (Apicomplexa: Neogregarinorida). J Invert Pathol 102: 261–263.
- 17. Plischuk S, Meeus I, Smagghe G, Lange CE (2011) Apicystis bombi (Apicomplexa: Neogregarinorida) parasitizing Apis mellifera and Bombus terrestris (Hymenoptera: Apidae) in Argentina. Environ Microbiol Rep 3: 565–568.
- 18. Lipa JJ, Triggiani O (1992) A newly recorded neogregarine (Protozoa, Apicomplexa), parasite in honey bees (Apis mellifera) and bumble bees (Bombus Spp). Apidologie 23: 533–536.
- 19. Lipa JJ, Triggiani O (1996) Apicystis gen. nov. and Apicystis bombi (Liu, Macfarlane & Pengelly) comb. nov. (Protozoa: Neogregarinida), a cosmopolitan parasite of Bombus and Apis (Hymenoptera: Apidae). Apidologie 27: 29–34.
- 20. Schmid-Hempel P (2001) On the evolutionary ecology of host-parasite interactions: addressing the question with regard to bumblebees and their parasites. Naturwissenschaften 88: 147–158.
- 21. Macfarlane RP, Lipa JJ, Liu HJ (1995) Bumble bee pathogens and internal enemies. Bee World 76: 130–148.
- 22. Rutrecht ST, Brown MJF (2008) The life-history impact and implications of multiple parasites for bumble bee queens. Int J Parasitol 38: 799–808.
- 23. Ravoet J, Maharramov J, Meeus I, De Smet L, Smagghe G, et al. Comprehensive bee pathogen screening in Belgium reveals Crithidia mellificae as a new contributory factor to winter mortality. PLoS ONE 8(8): e72443
- 24. Morimoto T, Kojima Y, Yoshiyama M, Kimura K, Yang B, et al. (2013) Molecular detection of protozoan parasites infecting Apis mellifera colonies in Japan. Environ Microbiol Rep 5: 74–77.
- 25. Meeus I, de Graaf DC, Jans K, Smagghe G (2010) Multiplex PCR detection of slowly-evolving trypanosomatids and neogregarines in bumblebees using broad-range primers. J Appl Microbiol 109: 107–115.
- 26. Cummings SM, McMullan M, Joyce DA, van Oosterhout C (2010) Solutions for PCR, cloning and sequencing errors in population genetic analysis. Conserv Genet 11: 1095–1097.
- 27. Clarke LA, Rebelo CS, Goncalves J, Boavida MG, Jordan P (2001) PCR amplification introduces errors into mononucleotide and dinucleotide repeat sequences. Mol Pathol 54: 351–353.
- 28. Saitou N, Nei M (1987) The neighbor-joining method - a new method for reconstructing phylogenetic trees. Mol Biol Evol 4: 406–425.
- 29. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, et al. (2011) MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol 28: 2731–2739.
- 30. Bandelt HJ, Forster P, Sykes BC, Richards MB (1995) Mitochondrial portraits of human populations using median networks. Genetics 141: 743–753.
- 31. Harris DJ, Crandall KA (2000) Intragenomic variation within ITS1 and ITS2 of freshwater crayfishes (Decapoda : Cambaridae): Implications for phylogenetic and microsatellite studies. Mol Biol Evol 17: 284–291.
- 32. O'Mahony EM, Tay WT, Paxton RJ (2007) Multiple rRNA variants in a single spore of the microsporidian Nosema bombi.. J Eukaryot Microbiol 54: 103–109.
- 33. Read AF, Taylor LH (2001) The ecology of genetically diverse infections. Science 292: 1099–1102.
- 34. Ruiz-Gonzalez MX, Bryden J, Moret Y, Reber-Funk C, Schmid-Hempel P, et al. (2012) Dynamic transmission, host quality, and population structure in a multihost parasite of bumblebees. Evolution 66: 3053–3066.
- 35. Schmid-Hempel P, Puhr K, Kruger N, Reber C, Schmid-Hempel R (1999) Dynamic and genetic consequences of variation in horizontal transmission for a microparasitic infection. Evolution 53: 426–434.
- 36. Woolhouse MEJ, Haydon DT, Antia R (2005) Emerging pathogens: the epidemiology and evolution of species jumps. Trends Ecol Evol 20: 238–244.
- 37. Evison SEF, Roberts KE, Laurenson L, Pietravalle S, Hui J, et al. (2012) Pervasiveness of parasites in pollinators. PLoS ONE 7(1): e30641
- 38. Genersch E, Yue C, Fries I, de MirandaJR (2006) Detection of deformed wing virus, a honey bee viral pathogen, in bumble bees (Bombus terrestris and Bombus pascuorum) with wing deformities. J Invertebr Pathol 91: 61–63.
- 39. Graystock P, Yates K, Darvill B, Goulson D, Hughes WOH (2013) Emerging dangers: deadly effects of an emergent parasite in a new pollinator host. J Invert Pathol 114: 114–119
- 40. Meeus I, Vercruysse V, Smagghe G (2012) Molecular detection of Spiroplasma apis and Spiroplasma melliferum in bees. J Invert Pathol 109: 172–174
- 41. Singh R, Levitt AL, Rajotte EG, Holmes EC, Ostiguy N, et al. (2010) RNA viruses in hymenopteran pollinators: evidence of inter-taxa virus transmission via pollen and potential impact on non-Apis hymenopteran species. PLoS ONE 5(12): e14357
- 42. Rigaud T, Perrot-Minnot MJ, Brown MJF (2010) Parasite and host assemblages: embracing the reality will improve our knowledge of parasite transmission and virulence. Proc Biol Sci. 277: 3693–3702.
- 43. Bazsalovicsova E, Kralova-Hromadova I, Stefka J, Scholz T, Hanzelova V, et al. (2011) Population study of Atractolytocestus huronensis (Cestoda: Caryophyllidea), an invasive parasite of common carp introduced to Europe: mitochondrial cox1 haplotypes and intragenomic ribosomal ITS2 variants. Parasitol Res 109: 125–131.
- 44. Hines HM (2008) Historical biogeography, divergence times, and diversification patterns of bumble bees (Hymenoptera : Apidae : Bombus). Syst Biol 57: 58–75.
- 45. Liu HJ, Macfarlane RP, Pengelly DH (1974) Mattesia bombi n. sp. (Neogregarinida: Ophrocystdae), a parasite of Bombus (Hymenoptera: Apidae) J Invertebr Pathol 23: 225–231.
- 46. Torres-Ruiz A, Jones RW (2012) Comparison of the efficiency of the Bumble Bees Bombus impatiens and Bombus ephippiatus (Hymenoptera: Apidae) as pollinators of tomato in greenhouses. J Econ Entomol 105: 1871–1877.
- 47. Vergara CH, Fonseca-Buendía P (2012) Pollination of greenhouse tomatoes by the mexican bumblebee Bombus ephippiatus (Hymenoptera: Apidae) J Pol Ecol. 7: 27–30.
- 48. Velthuis HHW, 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.
- 49. Cankaya NE, Kaftanoglu O (2006) An investigation on some diseases and parasites of bumblebee queens (Bombus terrestris L.) in Turkey. Pak J Biol Sci 9: 1282–1286.
- 50. Daszak P, Cunningham AA, Hyatt AD (2000) Wildlife ecology - Emerging infectious diseases of wildlife - Threats to biodiversity and human health. Science 287: 443–449.