Parasite-manipulated host dispersal: evidence from population genetics and mark-recapture experiments

Parasite-manipulated host dispersal: evidence from population genetics and mark-recapture experiments | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL Oikos This is a preprint and has not been peer reviewed. Data may be preliminary. 23 June 2025 V1 Latest version Share on Parasite-manipulated host dispersal: evidence from population genetics and mark-recapture experiments Authors : Hiroshi Ishii 0000-0003-4119-2398 [email protected] , Shohei Tsujimoto , Akihiro Sakatoku , and Tetsuo Kohyama Authors Info & Affiliations https://doi.org/10.22541/au.175067513.38872580/v1 Published Oikos Version of record Peer review timeline 342 views 233 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Host–parasite interactions are strongly influenced by the dispersal behaviour of both partners, yet direct evidence of parasite-driven active suppression of host dispersal has been lacking. Here, we provide evidence that the parasitic nematode Sphaerularia bombi manipulates its hosts, bumble bee queens (Bombus spp.), to limit their dispersal. We conducted genetic analyses of S. bombi and its hosts in Japan and the Netherlands, and found pronounced genetic differentiation among local S. bombi populations, despite minimal structure among host populations, indicating that infected queens exhibit severely restricted movement. Mark–recapture experiments further confirmed that infected queens tend to remain near their hibernation sites. We also identified cryptic species within S. bombi, based on concordance between mitochondrial haplotypes and nuclear clades, suggesting weak host specificity. Our findings provide the first empirical demonstration that a parasite can actively constrain host dispersal—a strategy likely enhancing mating opportunities and local transmission at the cost of broader range expansion. Parasite populations are probably established through rare, accidental transport of infected hosts, leading to strong founder effects and rapid cryptic speciation. This work highlights a novel mechanism by which parasites can shape host movement ecology and drive their own evolutionary trajectories. Parasite-manipulated host dispersal: evidence from population genetics and mark-recapture experiments Abstract: Host–parasite interactions are strongly influenced by the dispersal behaviour of both partners, yet direct evidence of parasite-driven active suppression of host dispersal has been lacking. Here, we provide evidence that the parasitic nematode Sphaerularia bombi manipulates its hosts, bumble bee queens ( Bombus spp.), to limit their dispersal. We conducted genetic analyses of S. bombi and its hosts in Japan and the Netherlands, and found pronounced genetic differentiation among local S. bombi populations, despite minimal structure among host populations, indicating that infected queens exhibit severely restricted movement. Mark–recapture experiments further confirmed that infected queens tend to remain near their hibernation sites. We also identified cryptic species within S. bombi , based on concordance between mitochondrial haplotypes and nuclear clades, suggesting weak host specificity. Our findings provide the first empirical demonstration that a parasite can actively constrain host dispersal—a strategy likely enhancing mating opportunities and local transmission at the cost of broader range expansion. Parasite populations are probably established through rare, accidental transport of infected hosts, leading to strong founder effects and rapid cryptic speciation. This work highlights a novel mechanism by which parasites can shape host movement ecology and drive their own evolutionary trajectories. Keywords: bumble bee; dispersal; host manipulation; host-parasite interaction; nematode; Sphaerularia bombi The dispersal of both hosts and parasites influences various aspects of host–parasite interactions, including transmission dynamics, spatial distribution, and the population structure of both species (Boots and Sasaki 1999, Iritani and Iwasa 2014, Sweet and Johnson 2018). When parasites have limited ability to move independently, their dispersal largely depends on the movement of their hosts (Prugnolle et al. 2005, Poulin 2007, DiBlasi et al. 2018). In this context, understanding how parasites affect host dispersal is essential for a full understanding of host–parasite dynamics. math_shortcuts When parasites influence host dispersal, three potential scenarios may explain this phenomenon. First, the parasite may reduce the host’s vitality, thereby diminishing its capacity to disperse, or the altered dispersal may simply be a side effect of infection (i.e. a detrimental or incidental outcome of parasitism (Poulin 1995, Weber and Stilianakis 2007, Fellous et al. 2012). Second, the host may disperse more widely to escape areas of high infection pressure, thus enhancing its own fitness (i.e. a host-adaptive response to the parasite: Terui et al. 2017, Baines et al. 2020). Third, parasites may alter host dispersal behaviour to improve their own transmission success (i.e. host manipulation by the parasite: Lion et al. 2006). With regard to the third scenario, Lion et al. (2006) proposed in a theoretical model that parasites might evolve to manipulate host behaviour in order to promote dispersal, thereby facilitating their own spread. Extending this argument, we propose that parasites might evolve to either inhibit or promote host dispersal to enhance their own success, as dispersal may not always be the optimal strategy for organisms (Karlson and Taylor 1995, Nichols et al. 2020, Treep et al. 2021). For example, when suitable habitats for the organism are scarce and scattered, there is no guarantee that a new habitat will be suitable. The uncertainty and potential failure to find a suitable new habitat may lead parasites to evolve to manipulate host behavior to inhibit dispersal. However, there are no reliable empirical examples of the manipulation of host dispersal to date, due to the difficulty of distinguishing whether behavioral changes resulting from infection are the result of manipulation by the parasite, an adaptive response by the host, or merely incidental or pathological side effects (Poulin and Maure 2015, Heil 2016). Sphaerularia bombi Dufour (Nematoda: Sphaerulariidae) is a castrating parasite of bumble bee queens that undergoes a free-living stage in its life history. It has been reported in more than 27 Bombus species across various regions, including Europe (Poinar and van der Laan 1972), New Zealand (MacFarlane and Griffin 1990), North America (McCorquodale et al. 1998), South America (Plischuk and Lange 2012), and Japan (Mitsuhata et al. 1995, Kosaka et al. 2008). Mature females are thought to enter hibernating bumble bee queens through the mouth, anus, or between the tegumental plates (Poinar and van der Laan 1972). A parasitised queen does not reproduce but continues to fly well beyond the time when a healthy queen would have founded a nest, indicating that most overwintered queens active by summer are infected (Poinar and van der Laan 1972, MacFarlane and Griffin 1990, Kadoya and Ishii 2015). Occasionally, a parasitised queen was observed flying close to the ground, landing, and either crawling on the surface or digging small depressions in the soil (see videos in Supporting information), during which third-stage juvenile nematodes of the next generation are released (Poinar and van der Laan 1972). After developing to the adult stage in the soil, the nematodes mate, and females become ready to enter a new host. By definition (Lefèvre et al. 2009), S. bombi is considered a manipulative parasite, as the prolonged foraging period of its host is not advantageous to the host but is essential for the parasite’s life cycle (Poinar and van der Laan 1972). A recent study reported changes in gene expression in bumble bee queens infected with S. bombi (Colgan et al. 2020). Given that S. bombi significantly alters host behaviour, and that bumble bees are key pollinators in temperate to cold ecosystems, the ecological impact of this parasite is expected to be substantial. Indeed, a study conducted on Hokkaido Island, Japan, demonstrated that S. bombi , by extending the foraging periods of host queens, increases competition for floral resources between host queens and workers, the latter being the offspring of uninfected queens (Kadoya and Ishii 2015). This, in turn, modifies interactions between the plant community and non-host flower visitors. However, aside from this study, ecological research on this nematode remains limited, and many aspects of its biology and ecological role have yet to be explored. One unclear aspect is how nematodes are distributed across the hibernation sites of bumble bee queens, where infections occur. Some earlier studies suggested that infected queens may return to their hibernation sites (Cumber 1949, Alford 1969). However, no concrete evidence has been found to support this hypothesis (Poinar and van der Laan 1972). Thus, it remains uncertain whether queens actively return to or remain at their own hibernation sites, pause at hibernation sites during dispersal, or randomly release the next generation of nematodes at arbitrary locations. Another unresolved issue concerns the existence of cryptic species within S. bombi , and, if present, whether each cryptic species exhibits host specificity. Consequently, it remains unclear whether S. bombi infections can be transmitted between different bumble bee species. These questions are significant, as they may influence parasite-mediated apparent competition among host species (Price et al. 1988, Hudson and Greenman 1998). This study primarily addresses the following question: Do bumble bee queens infected with S. bombi return to, or remain at, their own hibernation sites? Additionally, we explore a supplementary question: Does S. bombi consist of cryptic species, and if so, do these cryptic species exhibit host specificity? To investigate these questions, we examined the population genetic structures of S. bombi infecting various bumble bee species collected from multiple locations in Toyama and Hokkaido, Japan, and from two locations in the Netherlands, using mitochondrial and nuclear genetic markers. For the Japanese populations, we also assessed the genetic structures of their bumble bee hosts for comparison. Bumble bee queens typically exhibit high dispersal capabilities, with little or no population genetic differentiation observed across spatial scales of tens to hundreds of kilometres, unless their habitats are fragmented by geographical barriers e.g., in Europe: Lecocq et al. 2013, Blasco-Lavilla et al. 2019; in North America: Koch et al. 2017; in Japan: Takeuchi et al. 2018). However, if S. bombi inhibits the dispersal of infected queens, we would expect to observe significant genetic differentiation among S. bombi populations. In addition, we conducted a behavioural tracking study using the mark–recapture method to determine whether infected individuals tend to remain confined to specific areas. Based on these results, we discuss the implications of our findings for understanding the unique ecology of this parasite. Materials and Methods (a) Sample collection and DNA extraction Specimens of Sphaerularia bombi and their host bumble bee species for genetic analyses were collected from multiple locations in Toyama and Hokkaido Prefectures, Japan, between May and July from 2010 to 2020, and from two locations in the Netherlands in July 2019. Geographic data (site name, latitude, longitude, elevation, etc.) for these sites are provided in Supporting information . At each location, various species of bumble bee queens were captured using insect nets. All collected bumble bees, except for Bombus terrestris in Hokkaido, were native to their respective sampling locations. B. terrestris has been introduced to Hokkaido from Europe since at least 1995 and has since become a dominant species in the open land zones of Hokkaido (Ishii et al. 2008, Kadoya et al. 2009). The collected queens were dissected under a stereoscopic microscope. Both nematode and bumble bee specimens were preserved in 99% ethanol at −30 °C until DNA extraction. DNA was extracted using the Wizard® Genomic DNA Purification Kit (Promega, USA) following the manufacturer’s protocol (Hanum et al. 2018). (b) Amplification and sequencing of S. bombi mitochondrial DNA A partial fragment (417 bp) of the mitochondrial cytochrome c oxidase subunit I ( COI ) gene was amplified using the primers JB3 5′-TTTTTTGGGCATCCTGAGGTTTAT-3′ and JB4.5 5′-TAAAGAAAGAACATAATGAAAATG-3′ (Bowles et al. 1992). PCR amplification was performed in a 20 μl reaction volume containing 0.5 units of Ex Taq HS polymerase (Takara Bio, Japan), 1× Ex Taq buffer, 200 μM of each dNTP, 1 μM of each primer, and approximately 50 ng of genomic DNA. The thermal cycler (TaKaRa PCR Thermal Cycler Dice Gradient; Takara Bio) was programmed for an initial denaturation at 96 °C for 5 minutes, followed by 30 cycles of denaturation at 96 °C for 1 minute, annealing at 50 °C for 1 minute, and extension at 72 °C for 1 minute, with a final extension at 72 °C for 5 minutes. PCR products were purified using the QIAquick PCR Purification Kit (QIAGEN, Germany) after verification of successful amplification by agarose gel electrophoresis. Direct sequencing of the PCR products was performed using an ABI 3130xl Genetic Analyzer (Thermo Fisher Scientific, USA) with the BigDye® Terminator v3.1 Cycle Sequencing Kit (Thermo Fisher Scientific). COI fragment sequence data were obtained for a total of 542 S. bombi individuals collected from various host species across multiple locations. (c) Amplification and sequencing of S. bombi nuclear DNA To assess mito-nuclear discordance or concordance, nuclear DNA sequences were additionally analysed for 166 of the 542 aforementioned S. bombi individuals. Primers nSSU_F_02 (5′-GGAAGGGCACCACCAGGAGTGG-3′; The Blaxter Lab website: http://www.nematodes.org) and 28SR (5′-TTTCACTCGCCGTTACTAAGG-3′ ( Vrain et al. 1992 ) were used to amplify partial fragments of the nuclear rRNA gene region, including 18S, ITS1, 5.8S, ITS2, and 28S (fragment lengths ranged from 1370 to 1463 bp, excluding primer sequences). PCR amplification was performed in a 20 μl reaction volume containing 0.5 units of PrimeSTAR GXL polymerase (Takara Bio, Japan), 1× PrimeSTAR GXL buffer, 200 μM of each dNTP, 0.2 μM of each primer, and approximately 50 ng of genomic DNA. Thermal cycling conditions consisted of 35 cycles of 98 °C for 10 seconds, 60 °C for 15 seconds, and 68 °C for 25 seconds. PCR product purification and sequencing followed the same procedure as described above for the mitochondrial COI gene, but two additional internal primers were used for sequencing: nSSU_F_04 (5′-GCTTGTCTCAAAGATTAAGCC-3′) and nSSU_F_11 (5′-AAGTCTGGTGCCAGCAGCCGC-3′; The Blaxter Lab). Sequence assembly and multiple sequence alignment were performed using MEGA 11 (Tamura et al. 2021). (d) Microsatellite genotyping of bumble bees To determine the genetic structure of bumble bees in Japan, specimens of various bumble bee species collected from multiple locations were genotyped at eight microsatellite loci: BTMS0125, BTMS0044, BTMS0043, BTMS0045, BTMS0136 (Stolle et al. 2009); 0918_31j22, 0336_1l24 (Stolle et al. 2011); and B124 (Estoup et al. 1995). To obtain fluorescently labelled DNA fragments of microsatellite markers via multiplex PCR, we employed the three-primer PCR approach, which involves the use of a fluorescently labelled universal primer in combination with modified locus-specific primers bearing 5′ universal sequence tails (Blacket et al. 2012). To reduce non-specific amplification during multiplex PCR, the reaction was performed in two steps: amplification with locus-specific primers was conducted in the first step, followed by the addition of fluorescently labelled universal primers in the second step. The Multiplex PCR Assay Kit Ver. 2 (Takara Bio) was used to prepare an 8 μl first-step reaction mixture containing 0.05 μl of Multiplex PCR Enzyme Mix, 1× Multiplex PCR Buffer, 0.05 μM of a mixture of forward primers with universal sequence tails (Blacket et al. 2012), 0.2 μM of ‘pig-tailed’ reverse primers (Brownstein et al. 1996), and 50 ng of DNA. The first-step reaction was carried out under the following thermal conditions: 94 °C for 1 minute, followed by 15 cycles of 94 °C for 30 seconds, 57 °C for 1 minute, and 72 °C for 30 seconds. Immediately after the first step, 2 μl of a second-step reaction solution containing 0.15 μM of fluorescently labelled universal primer and 1× Multiplex PCR Buffer was added. An additional 13 cycles were run using the same thermal conditions to incorporate fluorescent labels. Fragment analyses were performed using an ABI 3130xl Genetic Analyzer (Thermo Fisher Scientific), with the GeneScan 600 LIZ Size Standard (Thermo Fisher Scientific) as a size marker. The output data were analysed using GENEMAPPER v4.0 (Thermo Fisher Scientific). math_shortcuts (e) Phylogenetic and population genetic analyses We constructed a median-joining haplotype network using mitochondrial COI sequences of S. bombi with Network 10 (https://fluxus-engineering.com/). Maximum likelihood inference of S. bombi phylogenetic trees for both the mitochondrial COI gene and the nuclear rRNA gene sequences was performed using IQ-TREE v2.3.2 (Minh et al. 2020). The mitochondrial COI sequence was partitioned by codon, and the best-fit substitution models selected by the ModelFinder programme (Kalyaanamoorthy et al. 2017) were applied for phylogenetic tree inference: HKY+F+I for the merged first and second codon positions, and HKY+F+G4 for the third codon position. Pairwise distances among COI haplotypes were calculated using Kimura’s two-parameter (K2P) model (Kimura 1980), as implemented in MEGA 11. Nuclear rRNA gene sequences were also partitioned into transcribed and non-transcribed regions, and substitution models were selected as follows: JC+I for the merged [18S, 5.8S, 28S] partition, and TIM3+F+I for the merged [ITS1, ITS2] partition. Branch support was assessed using the ultrafast bootstrap method implemented in IQ-TREE, with 10,000 replicates. Summary statistics for population genetic diversity ( Supporting information ) were computed in R v4.3.3 (R Core Team 2024), using the adegenet v2.1.10 (Jombart 2008) and diveRsity v1.9.90 (Keenan et al. 2013) packages. Genetic differentiation among S. bombi populations collected from the same host species at different sites, as well as among populations from different host species within the same region, was quantified using Jost’s D (Jost 2008) and Weir & Cockerham’s F ST (Weir and Cockerham 1984), based on mitochondrial COI sequences. Likewise, genetic differentiation among populations of host bumble bee species was calculated using genotypes at the eight microsatellite loci. (f) Mark-recapture experiments To assess whether infected queens return to or remain within specific areas, we conducted mark–recapture experiments at two sites (Site A and Site B). Site A is a pre-season campground encircled by dense forest, located in Kareisawa Forest Park, Toyama Prefecture, Japan (36.82°N, 137.55°E, 735 m a.s.l.; Supporting information ). Site B is the precinct of Inari-jinja Shrine, surrounded by trees and located in Higashigakura Town, Hokkaido Prefecture, Japan (43.68°N, 142.46°E, 173 m a.s.l.; Supporting information). These sites are presumed to be overwintering habitats for bumble bees, as we observed several hibernating queens in the soil, in leaf litter, or beneath log piles during early spring (personal observations). From late spring to mid-summer—the period during which a healthy queen would typically have established a nest—we frequently observed bumble bee queens (primarily Bombus diversus at Site A and B. terrestris at Site B) flying low to the ground, landing, crawling, or excavating small holes in the soil at these locations. Due to the scarcity of flowers preferred by bumble bees at these sites, floral visits were infrequent. However, in the surrounding areas, bumble bee queens were frequently observed foraging on flowers. During the experiments, we captured and dissected a subset of these queens and confirmed that most were infected with S. bombi (100.0% at Site A, n = 25; 98.2% at Site B, n = 56). The experiment was conducted at both sites over five consecutive days: 7–11 June 2014 at Site A, and 25–29 June 2014 at Site B. The observation period at Site A was from 09:00 to 15:00, except on 11 June, when it ended at 12:00; at Site B, observations were conducted from 07:00 to 13:00. At Site A, we established a survey area measuring 170 × 70 metres ( Supporting information ). In contrast, at Site B, we designated the entire precinct of Inari-jinja Shrine—approximately 28,000 m²—as the survey area ( Supporting information ). To facilitate location tracking, 50-cm-tall flags were installed at the intersections of a 5-metre grid across Site B, enabling coordinates to be visually estimated from any point within the site. Throughout the experimental period, three observers independently and simultaneously patrolled the survey areas, capturing every bumble bee queen encountered where possible. For each capture, we recorded the species identity, behaviour, and identification number on the thorax if the individual had been previously tagged. At Site B, capture location coordinates were also recorded by visual estimation to the nearest metre, using the flag grid as a reference. At both sites, previously tagged bees (i.e. recaptures) were released at the location of capture. Newly captured bees were anaesthetised by chilling at approximately 1–4 °C, and then marked by attaching a numbered tag to the thorax with adhesive. At Site A, newly tagged bees were evenly divided into two groups: one was released at a designated point within the survey area, and the other at a location 300 metres outside the area (see map in Supporting Information). At Site B, all newly tagged bees were released at a single location within the shrine precinct (Supporting Information). In both experiments, all marked individuals were released within one hour of capture and recovered within 15 minutes after removal from chilling. In the analysis of the mark–recapture experiments, we used only data from B. diversus (Site A) and B. terrestris (Site B) to eliminate potential confounding effects due to interspecific differences. Sample sizes for other Bombus species were too small to allow separate analysis. Analysis for the Site A experiment – We calculated the proportion of individuals recaptured one, two, three, and four days after their initial capture, and compared these proportions between the group released within the survey area and the group released outside it. An individual was considered recaptured if it had been marked on the previous day or earlier. Analysis for the Site B experiment – We computed the distances between initial capture locations and recapture locations. To avoid pseudoreplication, only the first capture of each bee on each day was included in the analysis. Distances were grouped according to the number of days elapsed between initial capture and recapture. To determine whether individuals tended to be recaptured near their original capture locations, we compared the observed mean distances against a null distribution. This null distribution was generated from the mean distances of 10,000 permuted datasets, each created by randomly reassigning the initial capture and recapture location pairs among individuals. Results (a) COI haplotype distribution in S. bombi We detected 31 haplotypes (DDBJ accession nos. XXXXXX–XXXXXX), which can be broadly categorised into eight haplogroups (A–H; Fig. 1). The K2P distances between haplotypes within the same haplogroup ranged from 0.24% to 2.70%, while distances between haplotypes from different haplogroups ranged from 4.97% to 14.03% (Supporting Information). These haplotypes were generally specific to each sampling region: D2 and C4 were detected in both Toyama and Hokkaido, but A5–7, B1, C4–8, and E1–3 were detected only in Toyama; A1–4, C1–3, and D1–2 were detected only in Hokkaido; and F1–6, G1, and H1–3 were detected only in the Netherlands. All bumble bee species, except for B. ardens sakagamii and B. hypocrita sapporoensis from Hokkaido, were infected with multiple haplogroups of S. bombi . (b) Genetic differentiation among S. bombi populations Jost’s D among pairs of S. bombi populations from the same host species ranged from 0.014 to 1.000 (median = 0.360), while Weir & Cockerham’s F ST ranged from 0.009 to 0.883 (median = 0.186) (Table 1). In most cases, the 95% confidence intervals (95% CI) of these values did not include zero, indicating significant genetic differentiation among the populations. In some population pairs, the 95% CI included zero (e.g., To vs Ka in B. diversus and Ki/Bi vs Ur in B. terrestris ); however, these cases are likely attributable to small sample sizes or the low number of haplotypes, leading to inaccurate estimates. We note that genetic differentiation was also found among local populations even when focusing on nematodes belonging to the same haplogroup. For example, among nematodes of haplogroup E parasitising B. ardens in Toyama Prefecture, Japan, the Ka population was dominated by E2, whereas the Ta population, located 28 km from Ka, was dominated by E1. Similarly, among nematodes of haplogroup F parasitising B. pascuorum in the Netherlands, the Wa population was predominantly composed of F4, whereas the Ke population, located 85 km from Wa, had a higher proportion of F2 and F5 in addition to F4. Jost’s D and F ST among S. bombi population pairs from different host species within the same region ranged from –0.013 to 0.940 (median = 0.474) and from –0.048 to 0.301 (median = 0.182), respectively (Table 2). In most cases, the 95% CI of these values did not include zero, again suggesting significant genetic differentiation, with some exceptions likely due to the low number of haplotypes and small sample sizes. (c) Genetic differentiation among Bombus populations In general, Jost’s D and F ST among population pairs of the same bumble bee species were very small, ranging from 0.001 to 0.018 (median = 0.009) and from 0.000 to 0.014 (median = 0.007), respectively (Table 3). In most cases, the 95% CI of these values included zero, indicating frequent gene flow among local populations. (d) Mito-nuclear discordance/concordance Seven well-supported haplogroups (A–G) of mitochondrial COI sequences, as presented in the haplotype network graph (Fig. 1), were reconfirmed in the phylogenetic tree inferred by the maximum likelihood method for the mitochondrial COI sequences (left side of Fig. 2). In addition, five well-supported clades (a–e) were identified in the nuclear rRNA tree (right side of Fig. 2), reconstructed using 25 unique sequence variants (DDBJ accession nos. XXXXXX–XXXXXX) obtained from 166 specimens. These clades of rRNA gene alleles can be clearly distinguished by insertions and deletions within the internal transcribed spacers, although this is not fully reflected in the phylogenetic tree topology. Except for clade e, all haplogroups and clades were detected exclusively in either Japan or the Netherlands. Within clade e, however, no identical sequences were shared between Japan and the Netherlands: sequences e1–7 were detected only in Japan, while sequences e8–9 were detected only in the Netherlands. Despite topological differences between the mitochondrial and nuclear phylogenies, there was remarkable correspondence between mitochondrial haplogroups and nuclear clades, with some exceptions. In general, mitochondrial haplogroup F corresponded with nuclear clade b; haplogroups E and D corresponded with clade a; haplogroups G, C, and B corresponded with clade e; haplogroup H corresponded with clade d; and haplogroup A corresponded with clade c. math_shortcuts (e) Mark-recapture experiments At Site A, the proportion of recaptured individuals did not differ significantly between the group released within the survey area and the group released outside the survey area, regardless of the number of days elapsed since the first release ( P > 0.05; Fisher’s exact test) (Fig. 3a). The proportions decreased only slightly with increasing time since release. At Site B, the mean distances between the initial capture location and the recapture location were approximately 60 metres, regardless of the number of days between initial capture and recapture (Supporting Information). These distances were shorter than the 2.5th percentile values of the null distributions of the means, indicating a significant tendency for individuals to be recaptured in close proximity to their initial capture locations. math_shortcuts Discussion (a) Do bumble bee queens infected with S. bombi return to and remain near their own hibernation sites? Evidence from population genetic analyses, combined with results from mark–recapture experiments, suggests that bumble bee queens infected with the Sphaerularia nematode repeatedly return to and remain near their hibernation sites. First, our mitochondrial DNA analyses revealed marked genetic differentiation among local populations of the nematode at spatial scales of less than 100 km (Table 1, Fig. 1), whereas weak or no differentiation was observed among local populations of their hosts, the bumble bees, based on microsatellite analyses (Table 3). The latter finding is consistent with previous studies, which have shown that, in the absence of geographical barriers such as seas or mountain ranges, weak or no population genetic differentiation arises over spatial scales of tens to a few hundred kilometres in bumble bee populations (Lecocq et al. 2013, Koch et al. 2017, Takeuchi et al. 2017, Blasco-Lavilla et al. 2019). These results likely reflect the high dispersal abilities of bumble bee queens, as suggested by several observational and empirical studies (Mikkola 1984, Morales et al. 2013). In contrast, parasitic nematodes in their free-living stage are likely to have very limited dispersal capacity on their own. Thus, their spatial dispersion—and consequently their genetic structure—must largely depend on the mobility of their hosts. Anderson et al. (1998) reviewed the population genetic structure of parasitic nematodes and found that very few species exhibit genetic differentiation. Such structures are observed only when the host’s dispersal ability is low (Cole and Viney 2018) or when host movement is hindered (Blouin et al. 1995). Therefore, the significant genetic structure observed in Sphaerularia nematodes, despite the high migratory dispersal ability of their host species, suggests that infected queens, unlike healthy individuals, do not disperse widely but instead remain near their hibernation sites where they were infected. Second, our mark–recapture experiments demonstrated a clear tendency for infected bumble bee queens to be recaptured within the survey area—presumed to serve as their hibernation site—even when they had been released outside it (Fig. 3a). We also found that individuals tended to be recaptured in close proximity within the survey area over several consecutive days (Fig. 3b, Supporting Information). There were few flowers within the experimental areas, and foraging activity was rarely observed during the experimental periods, indicating that the queens foraged outside the area but returned to it. Kadoya and Ishii (2015) reported that areas where infected queens were frequently captured remained stable for at least three years, within spatial scales of only several kilometres. We also note that the two locations in the Netherlands where we captured infected bumble bee queens (Ka and Wa) are the same sites where infected queens were recorded approximately half a century ago (Poinar and van der Laan 1972). During our sampling period at these locations (31 May to 9 June 2019), we frequently observed overwintered bumble bee queens flying low to the ground, excavating holes in the soil, or foraging from flowers. The overall infection rate among queens at these sites was 68.1% (N = 167). By contrast, at other locations in the Netherlands, overwintered queens were infrequently observed during the same period, although many worker bees were present (personal observation). This suggests that Ka and Wa have remained areas of high infection rates for approximately fifty years. Taken together—the population genetic evidence, the mark–recapture experiment results, the findings of Kadoya and Ishii (2015), and the rediscovery of infection at historically recorded sites—our results strongly support the conclusion that bumble bee queens infected with Sphaerularia nematodes return to and remain near their own hibernation sites. Then, what could be the underlying cause of suppressed dispersal in infected queens? We propose three possible scenarios to explain this phenomenon. The first is a reduction in the vitality of bumble bee queens due to infection. However, this seems unlikely, as infected queens are actively flying and visually indistinguishable from uninfected queens (see videos in Supporting Information). Infected queens continue to forage and fly for over one to two months after emerging from hibernation (Kadoya and Ishii 2015), making it implausible that they are incapable of traveling several tens of kilometres. Moreover, this scenario cannot account for the mark–recapture experiment results, where individual queens returned to close proximity even after being released either within or outside the survey area. Such localized homing behavior is inconsistent with general debilitation or lethargy due to pathology. The second scenario is that the suppressed dispersal represents an adaptive response by the host. However, this too appears implausible, as infected queens are functionally sterile ( Poinar and van der Laan 1972, MacFarlane and Griffin 1990, Kadoya and Ishii 2015 ) and thus unlikely to benefit from adaptive behaviours such as conserving energy or investing in immune function—key assumptions of the “host adaptation” hypothesis. Given that infected queens have no reproductive future, the plausibility of this explanation is effectively ruled out. The third scenario, which we consider the most convincing, is adaptive host manipulation by the parasite. Sphaerularia nematodes are released into the soil as third-stage larvae, where they mature, mate, and infect new hosts ( Poinar and van der Laan 1972 ). However, their mobility and dispersal capacity in the soil are likely limited. Thus, manipulating the host to return to its hibernation site, thereby releasing the nematodes there, would enhance opportunities for new infections, which occur during host hibernation. Furthermore, if a nematode is not released near another potential mate, it may lose the opportunity to reproduce. Although multiple nematodes are released from the same host, these individuals are likely siblings. To avoid inbreeding, it would be preferable for nematodes to encounter unrelated individuals released from different hosts. Given that the infection rates of Sphaerularia nematodes vary significantly even across small spatial scales ( Kadoya and Ishii 2015 ), it is advantageous for many unrelated individuals to be released in close proximity to maximise outcrossing opportunities. Thus, manipulating host behaviour to ensure that numerous infected queens release nematodes near each other would confer a clear advantage to Sphaerularia nematodes. (b) Does S. bombi have cryptic species, and if so, do these cryptic species exhibit host specificity? The general correspondence between mitochondrial haplogroups and major nuclear rRNA clades (Fig. 2) suggests restricted gene flow among Sphaerularia nematodes of different haplogroups. The presence of individuals that deviate from this pattern provides evidence for occasional hybridisation between different lineages. However, the overall maintenance of this correspondence implies reduced fitness in hybrids. Additionally, the K2P distances among the COI haplogroups ranged from 4.97% to 14.03%, substantially exceeding the intraspecific COI K2P distances typically observed in various taxa (generally less than 2–3%: Mabragaña et al. 2011, Hawlitschek at al. 2017) , including nematodes ( Ferri et al. 2009 ). Therefore, we conclude that S. bombi includes cryptic species. Nematodes from each lineage were found to infect multiple bumble bee species (Fig.s 1 and 2), suggesting that host specificity within each lineage is either absent or weak. math_shortcuts (c) Are S. bombi infections transmitted between different bumble bee species in nature? Nematode populations infecting different host species within the same region exhibited substantial differences in mitochondrial haplotype composition (Fig. 1) and a high degree of genetic differentiation (Table 2). These results suggest that interspecies transmission among bumble bee hosts is restricted, even within the same collection site. Nematode infections occur during the hibernation of bumble bee queens (Poinar and van der Laan 1972). Thus, differences in hibernation sites among bumble bee species (Alford 1969) may limit cross-species transmission, although our understanding of these differences remains limited. (d) Conclusion Our study highlights the unique ecology of Sphaerularia bombi . By restricting host dispersal, S. bombi likely increases its mating opportunities and chances of infecting new hosts. This alteration in host behaviour limits the parasite’s ability to colonise new locations, which may explain the significant variation in infection rates observed at small spatial scales (Kadoya and Ishii 2015). The establishment of S. bombi populations in new areas is likely driven by the accidental transport of hosts by strong winds or other factors from their original locations, with a small number of individuals acting as founders. The low genetic diversity observed within each population is probably a consequence of founder effects in these instances. Repeated founder events, followed by isolation resulting from the parasite’s non-dispersal strategy via host manipulation, are likely to accelerate speciation ( Gavrilets 2004, Templeton 2008 ), ultimately leading to the existence of multiple cryptic species of S. bombi even at small spatial scales. Compared with free-living consumers such as predators, herbivores, and detritivores, the ecology of parasites remains less well understood. Parasitic organisms with previously unknown relationships with their hosts are likely to continue to be discovered, revealing intricate and complex interactions that challenge our current understanding. The accumulation of such knowledge will not only fill existing gaps in parasitic biology but also provide critical insights into the co-evolutionary dynamics between parasites and their hosts. References Alford DV. 1969 A study of the hibernation of bumble bees (Hymenoptera: Bombidae) in southern England. J Anim Ecol 38, 149–170. math_shortcuts Anderson TJ, Blouin MS, Beech RN. 1998 Population biology of parasitic nematodes: applications of genetic markers. Adv Parasitol 41, 219–283. Baines CB, Diab S, McCauley SJ. 2020 Parasitism risk and infection alter host dispersal. Am Nat 196, 119–131. Blacket MJ, Robin C, Good RT, Lee SF, Miller AD. 2012 Universal primers for fluorescent labelling of PCR fragments—an efficient and cost-effective approach to genotyping by fluorescence. Mol Ecol Resour 12, 456–463. Blasco-Lavilla N, Ornosa C, Michez D, De la Rúa P. 2019 Contrasting patterns of genetic and morphological diversity in the bumble bee Bombus lucorum (Hymenoptera: Apidae: Bombus) along a European gradient. J Insect Conserv 23, 933–943. math_shortcuts Blouin MS, Yowell CA, Courtney CH, Dame JB. 1995 Host movement and the genetic structure of populations of parasitic nematodes. Genetics 141, 1007–1014. Boots M, Sasaki A. 1999 ’Small worlds’ and the evolution of virulence: infection occurs locally and at a distance. Proc R Soc B 266, 1933–1938. Bowles J, Blair D, McManus DP. 1992 Genetic variants within the genus Echinococcus identified by mitochondrial DNA sequencing. Mol Biochem Parasitol 54, 165–173. Brownstein MJ, Carpten JD, Smith JR. 1996 Modulation of non-templated nucleotide addition by Taq DNA polymerase: primer modifications that facilitate genotyping. BioTechniques 20, 1004–1010. math_shortcuts Colgan TJ, Carolan JC, Sumner S, Blaxter ML, Brown MJF. 2020 Infection by the castrating parasitic nematode Sphaerularia bombi changes gene expression in Bombus terrestris bumble bee queens. Insect Mol Biol 29, 170–182. Cole R, Viney M. 2018 The population genetics of parasitic nematodes of wild animals. Parasites Vectors 11, 590. Cumber RA. 1949 Humble-bee parasites and commensals found within a thirty-mile radius of London. Proc R Entomol Soc Lond 24, 119–121. DiBlasi E, Johnson KP, Stringham SA, Hansen AN, Beach AB, Clayton DH, Bush SE. 2018 Phoretic dispersal influences parasite population genetic structure. Mol Ecol 27, 1–10. Estoup A, Scholl A, Pouvreau A, Solignac M. 1995 Monandry and polyandry in bumble bees (Hymenoptera: Bombinae) as evidenced by highly variable microsatellites. Mol Ecol 4, 89–93. Fellous S, Duncan AB, Quillery E, Vale PF, Kaltz O. 2012 Genetic influence on disease spread following arrival of infected carriers. Ecol Lett 15, 186–192. Ferri E, Barbuto M, Bain O, Galimberti A, Uni S, Guerrero R, Ferté H, Bandi1C, Martin C, Casiraghi M. 2009 Integrated taxonomy: traditional approach and DNA barcoding for the identification of filarioid worms and related parasites (Nematoda). Front Zool 6 , 1. Gavrilets S. 2004 Fitness landscapes and the origin of species . Princeton, NJ: Princeton University Press. Hanum L, Windusari Y, Setiawan A, Muharni, Adriansyah F, Mubarok AA. 2018 Comparison of CTAB method and Wizard genomic DNA purification system kit from Promega on DNA isolation of local varieties of rice of South Sumatera. Sci Technol Indones 3, 26–29. Hawlitschek O, Morinière J, Lehmann GUC, Lehmann AW, Kropf M, Dunz A, Glaw F, Detcharoen M, Schmidt S, Hausmann A, Szucsich NU, Caetano-Wyler SA, Haszprunar G. 2017 DNA barcoding of crickets, katydids and grasshoppers (Orthoptera) from Central Europe with focus on Austria, Germany and Switzerland. Mol Ecol Resour 17, 1037-1053. Heil M. 2016 Host manipulation by parasites: cases, patterns, and remaining doubts. Front Ecol Evol 4, 80. Hudson P, Greenman J. 1998 Competition mediated by parasites: biological and theoretical progress. Trends Ecol Evol 13, 387–390. Ishii HS, Kadoya T, Kikuchi R, Suda S, Washitani I. 2008 Habitat and flower resource partitioning by an exotic and three native bumble bees in central Hokkaido, Japan. Biol Conserv 141, 2597–2607. Iritani R, Iwasa Y. 2014 Parasite infection drives the evolution of state-dependent dispersal of the host. Theor Popul Biol 92, 1–13. Jombart T. 2008 adegenet: an R package for the multivariate analysis of genetic markers. Bioinformatics 24, 1403–1405. Jost L. 2008 G(ST) and its relatives do not measure differentiation. Mol Ecol 17, 4015–4026. math_shortcuts Kadoya EZ, Ishii HS. 2015 Host manipulation of bumble bee queens by Sphaerularia nematodes indirectly affects foraging of non-host workers. Ecology 96, 1361–1370. Kadoya T, Ishii HS, Kikuchi R, Suda S, Washitani I. 2009 Using monitoring data gathered by volunteers to predict the potential distribution of the invasive alien bumblebee Bombus terrestris . Biol Conserv 142, 1011–1017. math_shortcuts Kalyaanamoorthy S, Minh BQ, Wong TKF, von Haeseler A, Jermiin LS. 2017 ModelFinder: fast model selection for accurate phylogenetic estimates. Nat Methods 14, 587–589. Karlson RH, Taylor HM. 1995 Alternative predictions for optimal dispersal in response to local catastrophic mortality. Theor Popul Biol 47, 321–330. Keenan K, McGinnity P, Cross TF, Crozier WW, Prodöhl PA. 2013 diveRsity: an R package for the estimation and exploration of population genetics parameters and their associated errors. Methods Ecol Evol 4, 782–788. Kimura M. 1980 A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J Mol Evol 16, 111–120. Koch JB, Looney C, Sheppard WS, Strange JP. 2017 Patterns of population genetic structure and diversity across bumble bee communities in the Pacific Northwest. Conserv Genet 18, 507–520. Kosaka H, Sayama K, Kanzaki N, Takahashi J, Makino S. 2008 Sphaerularia sp. isolated from native Bombus spp. in Hokkaido. Jpn J Nematol 38, 115. (in Japanese) Lecocq T, Vereecken NJ, Michez D, Dellicour S, Lhomme P, Valterová I, Rasplus JY, Rasmont P. 2013 Patterns of genetic and reproductive traits differentiation in mainland vs. Corsican populations of bumblebees. PLoS ONE 8, e65642. Lefèvre T, Lebarbenchon C, Gauthier-Clerc M, Missé D, Poulin R, Thomas F. 2009 The ecological significance of manipulative parasites. Trends Ecol Evol 24, 41–48. Lion S, van Baalen M, Wilson WG. 2006 The evolution of parasite manipulation of host dispersal. Proc R Soc B 273, 1063–1071. Mabragaña E, Díaz de Astarloa JM, Hanner R, Zhang J, González Castro M. 2011 DNA barcoding identifies Argentine fishes from marine and brackish waters. PloS one , 6, e28655. math_shortcuts MacFarlane RP, Griffin RP. 1990 New Zealand distribution and seasonal incidence of the nematode, Sphaerularia bombi Dufour, a parasite of bumble bees. N Z J Zool 17, 191–199. McCorquodale DB, Beresford RG, Francis JM, Thomson CE, Bartlett CM. 1998 Prevalence of Sphaerularia bombi (Nematoda: Tylenchida: Sphaerulariidae) in bumble bee queens (Hymenoptera: Apidae) on Cape Breton Island, Nova Scotia, Canada. Can Entomol 130, 877–882. math_shortcuts Mikkola K. 1984 Spring migrations of wasp and bumblebee queens across the Gulf of Finland (Hymenoptera: Vespidae and Apidae). Notulae Entomol 64, 125–128. Minh BQ, Schmidt HA, Chernomor O, Schrempf D, Woodhams MD, von Haeseler A, Lanfear R. 2020 IQ-TREE 2: new models and efficient methods for phylogenetic inference in the genomic era. Mol Biol Evol 37, 1530–1534. Mitsuhata M, Asada S, Ono M. 1995 About Sphaerularia sp. parasitizing Bombus ardens . Proc Annu Meet Jpn Soc Appl Entomol Zool 39, 91. (in Japanese) Morales CL, Arbetman MP, Cameron SA, Aizen MA. 2013 Rapid ecological replacement of a native bumble bee by invasive species. Front Ecol Environ 11, 529–534. Nichols BS, Leubner-Metzger G, Jansen VAA. 2020 Between a rock and a hard place: adaptive sensing and site-specific dispersal. Ecol Lett 23, 1370–1379. Plischuk S, Lange CE. 2012 Sphaerularia bombi (Nematoda: Sphaerulariidae) parasitizing Bombus atratus (Hymenoptera: Apidae) in southern South America. Parasitol Res 111, 947–950. Poinar GO, van der Laan PA. 1972 Morphology and life history of Sphaerularia bombi . Nematologica 18, 239–252. math_shortcuts Poulin R. 1995 ’Adaptive’ changes in the behaviour of parasitized animals: a critical review. Int J Parasitol 25, 1371–1383. math_shortcuts Poulin R. 2007 Evolutionary ecology of parasites . 2nd ed. Princeton, NJ: Princeton University Press. Poulin R, Maure F. 2015 Host manipulation by parasites: a look back before moving forward. Trends Parasitol 31, 563–570. Price PW, Westoby M, Rice B. 1988 Parasite-mediated competition: some predictions and tests. Am Nat 131, 544–555. math_shortcuts Prugnolle F, Manica A, Charpentier M, Guégan JF, Guernier V, Balloux F. 2005 Pathogen-driven selection and worldwide HLA class I diversity. Curr Biol 15, 1022–1027. R Core Team. 2024 R: A language and environment for statistical computing . Vienna, Austria: R Foundation for Statistical Computing. Stolle E, Rohde M, Vautrin D, Solignac M, Schmid-Hempel P, Schmid-Hempel R, Moritz RFA. 2009 Novel microsatellite DNA loci for Bombus terrestris (Linnaeus, 1758). Mol Ecol Resour 9, 1345–1352. math_shortcuts Stolle E, Wilfert L, Schmid-Hempel R, Schmid-Hempel P, Kube M, Reinhardt R, Moritz RFA. 2011 A second generation genetic map of the bumblebee Bombus terrestris (Linnaeus, 1758) reveals slow genome and chromosome evolution in the Apidae. BMC Genomics 12, 48. Sweet AD, Johnson KP. 2018 The role of parasite dispersal in shaping a host-parasite system at multiple evolutionary scales. Mol Ecol 27, 5104–5119. Takeuchi T, Takahashi M, Nishimoto M, Kiyoshi T, Tsuchida K, Nomura T, Takahashi J. 2018 Genetic structure of the bumble bee Bombus hypocrita sapporoensis , a potential domestic pollinator for crops in Japan. J Apic Res 57, 203–212. math_shortcuts Tamura K, Stecher G, Kumar S. 2021 MEGA11: molecular evolutionary genetics analysis version 11. Mol Biol Evol 38, 3022–3027. Templeton AR. 2008 The reality and importance of founder speciation in evolution. BioEssays 30, 470–479. Terui A, Ooue K, Urabe H, Nakamura F. 2017 Parasite infection induces size-dependent host dispersal: consequences for parasite persistence. Proc R Soc B 284, 20171491. Treep J, de Jager M, Bartumeus F, Soons MB. 2021 Seed dispersal as a search strategy: dynamic and fragmented landscapes select for multi-scale movement strategies in plants. Mov Ecol 9, 4. Vrain TC, Wakarchuk DA, Levesque A, Hamilton RL. 1992 Intraspecific rDNA restriction fragment length polymorphism in the Xiphinema americanum group. Fundam Appl Nematol 15, 563–573. math_shortcuts Weber TP, Stilianakis NI. 2007 Ecologic immunology of avian influenza (H5N1) in migratory birds. Emerg Infect Dis 13, 1139–1143. Weir BS, Cockerham CC. 1984 Estimating F-statistics for the analysis of population structure. Evolution 38, 1358–1370. Table 1. Genetic differentiation among S. bombi populations parasitising the same host species at different locations, estimated using mitochondrial COI haplotype data. Values represent pairwise estimates of population differentiation based on Jost’s D and Weir & Cockerham’s F ST , with 95% confidence intervals (CIs) obtained via a non-parametric bootstrap method (shown in brackets). Values for which the 95% CIs do not include zero are shown in bold . Data from the Fu and Ku populations, as well as from the Ki and Bi populations, were pooled for analysis due to their geographical proximity (Fig. 1: less than 15 km apart) and the absence of significant differences in haplotype composition (Fisher’s exact test, P > 0.05). Sample sizes for each population are indicated in parentheses. Populations with fewer than 10 individuals were excluded from the analysis. Population pair by locations ( N ) Distance (km) Jost’s D Weir & Cockerham’s Fst B. diversus (Toyama Pref.) To (27) vs. Ta (31) 38 0.659 [0.405–0.877] 0.427 [0.234–0.613] To (27) vs. Ka (30) 57 0.025 [-0.029 – 0.157] 0.016 [-0.057–0.165] Ta (31) vs. Ka (30) 28 0.473 [0.197–0.725] 0.269 [0.081–0.470] B. arden (Toyama Pref.) Fu/Ku (54) vs. Ka (14) 35-43 0.908 [0.689–1.003] 0.883 [0.776–0.966] Fu/Ku (54) vs. Ta (19) 29-33 1.000 [1.000–1.000] 0.841 [0.791–0.912] Ta (19) vs. Ka (14) 28 0.360 [0.026–0.717] 0.242 [-0.033–0.565] B. hypocrita (Toyama Pref.) Ta (35) vs. Ka (21) 28 0.621 [0.354–0.841] 0.186 [0.067–0.335] B. terrestris (Hokkaido Pref.) Hi (57) vs. Ki/Bi (39) 117-131 0.115 [0.022–0.252] 0.102 [0.021–0.208] Hi (57) vs. Ur (10) 46 0.168 [0.065–0.311] 0.141 [0.073–0.217] Ki/Bi (39) vs. Ur (10) 164-178 0.014 [-0.005–0.057] 0.009 [-0.047–0.076] B. pascuorum (Netherlands) Ke (41) vs. Wa (31) 85 0.275 [0.071–0.522] 0.069 [0.013–0.141] Table 2. Genetic differentiation among S. bombi populations parasitising different host species captured at the same location, estimated using mitochondrial COI haplotype data. Values indicate population differentiation based on Jost’s D and Weir & Cockerham’s F ST , along with 95% confidence intervals (CIs) obtained via a non-parametric bootstrap method (shown in brackets). Values for which the 95% CIs do not include zero are shown in bold . Sample sizes for each population are indicated in parentheses. Populations with fewer than 10 individuals were excluded from the analysis. Population pair by Host species ( N ) Jost’s D Weir & Cockerham’s Fst Ta (Toyama Pref.) B. diversus (31) vs. B. arden (19) 0.016 [-0.060–0.187] -0.010 [-0.071–0.120] B. diversus (31) vs. B. hypocrite (35) 0.503 [0.251–0.737] 0.196 [0.082–0.332] B. diversus (31) vs. B.beaticola (28) 0.940 [0.801–0.997] 0.301 [0.228–0.391] B. diversus (31) vs. B.honshuensis (12) -0.010 [-0.092–0.189] -0.038 [-0.104–0.116] B. arden (19) vs. B. hypocrita (35) 0.380 [0.097–0.664] 0.159 [0.007–0.345] B. arden (19) vs. B.beaticola (28) 0.840 [0.622–0.994] 0.290 [0.172–0.421] B. arden (19) vs. B.honshuensis (12) -0.013 [-0.105–0.241] -0.048 [-0.132–0.168] B. hypocrite (35) vs. B.beaticola (28) 0.368 [0.119–0.642] 0.099 [0.024–0.199] B. hypocrite (35) vs. B.honshuensis (12) 0.484 [0.224–0.724] 0.178 [0.047–0.340] B.beaticola (28) vs. B.honshuensis (12) 0.894 [0.711–0.993] 0.277 [0.168–0.400] Ka (Toyama Pref.) B. diversus (30) vs. B. arden (14) 0.734 [0.493–0.917] 0.423 [0.225–0.633] B. diversus (30) vs. B. hypocrita (21) 0.594 [0.297–0.848] 0.274 [0.093–0.484] B. arden (14) vs. B. hypocrita (21) 0.044 [-0.070–0.276] -0.002 [-0.081–0.142] Wa (Netherlands) B. pascurum (41) vs. B. ypocrite (13) 0.463 [0.140–0.745] 0.185 [0.010–0.379] Table 3. Genetic differentiation among populations of four bumblebee species captured at different localities, estimated using genotype data from eight microsatellite loci. Values indicate population differentiation based on Jost’s D and Weir & Cockerham’s F ST , along with 95% confidence intervals (CIs) obtained via a non-parametric bootstrap method (shown in brackets). Values for which the 95% CIs do not include zero are shown in bold . Sample sizes for each population are indicated in parentheses. Population pair by locations ( N ) Distance (km) Jost’s D Weir & Cockerham’s F ST B. diversus (Toyama Pref.) Fu/Ku (32) vs. Ka (32) 35-43 0.018 [-0.012–0.058] 0.009 [-0.003–0.025] Fu/Ku (32) vs. Ta (32) 29-33 0.011 [-0.017–0.050] 0.010 [-0.002–0.026] Fu/Ku (32) vs. To (32) 18-25 0.011 [-0.015–0.047] 0.009 [-0.002–0.025] Ka (32) vs. Ta (32) 28 0.001 [-0.008–0.023] 0.002 [-0.006–0.013] Ka (32) vs. To (32) 57 0.006 [-0.006–0.033] 0.010 [0.001–0.023] Ta (32) vs. To (32) 38 0.010 [-0.001–0.035] 0.014 [0.004–0.027] B. arden (Toyama Pref.) Fu/Ku (32) vs. Ka (32) 35-43 0.011 [-0.028–0.063] 0.005 [-0.006–0.017] Fu/Ku (32) vs. Ta (32) 29-33 0.008 [-0.029–0.057] 0.002 [-0.008–0.013] Fu/Ku (32) vs. To (32) 18-25 0.011 [-0.028–0.063] 0.008 [-0.003–0.021] Ka (32) vs. Ta (32) 28 0.004 [-0.035–0.058] 0.001 [-0.008–0.013] Ka (32) vs. To (32) 57 0.004 [-0.036–0.057] 0.000 [-0.009–0.011] Ta (32) vs. To (32) 38 0.001 [-0.031–0.045] 0.002 [-0.007–0.014] B. hypocrite (Toyama Pref.) Ka (32) vs. Ta (32) 28 0.007 [-0.029–0.054] 0.004 [-0.006–0.016] B. terrestris (Hokkaido Pref.) Hi (32) vs. Ki/Bi (32) 117-131 0.023 [-0.019–0.078] 0.012 [0.001–0.025] Figure 1. COI haplotype network of Sphaerularia bombi and the spatial and host species distribution of haplotypes across regions. (a) Toyama Prefecture, (b) Hokkaido Prefecture, (c) Netherlands. In each network, circle size indicates the number of individuals with a given haplotype. Colored areas represent individuals from the focal region, while black areas represent individuals from other regions. Lines between haplotypes indicate single mutational steps; small white circles represent unsampled (inferred) haplotypes. Maps show pie charts for each host species, illustrating haplotype composition. Colors in the pies correspond to haplotypes in the network. Figure 2. Mito-nuclear concordance and discordance in Sphaerularia bombi lineages. This figure compares maximum likelihood trees based on mitochondrial COI (left) and nuclear rRNA sequences (right) for 542 individuals. Each line connects tips representing the same individual across the two trees. For heterozygous individuals, one mitochondrial tip connects to two nuclear tips. Tip colors indicate host species, while line colors denote the collection region. Only bootstrap support values above 75% are shown at nodes. Figure 3. Results of mark–recapture experiments. (a) Site A: Proportion of re-observed individuals released within vs. outside the survey area over time. Error bars indicate 95% confidence intervals (CI) based on the binomial distribution; sample sizes are indicated beside points. (b) Site B: Mean distance between initial capture and recapture sites over time. Error bars show 95% CI of the mean; sample sizes are shown beside points. Blue histograms represent null distributions of means from 10,000 permutations (see Methods for details). Triangles indicate the 50th and 2.5th percentiles of the null distribution. Information & Authors Information Version history V1 Version 1 23 June 2025 Peer review timeline Published Oikos Version of Record 5 Jan 2026 Published Copyright This work is licensed under a Non Exclusive No Reuse License. Collection Oikos Keywords bumble bee dispersal host manipulation host-parasite interaction nematode sphaerularia bombi Authors Affiliations Hiroshi Ishii 0000-0003-4119-2398 [email protected] University of Toyama View all articles by this author Shohei Tsujimoto University of Toyama View all articles by this author Akihiro Sakatoku University of Toyama View all articles by this author Tetsuo Kohyama The University of Tokyo View all articles by this author Metrics & Citations Metrics Article Usage 342 views 233 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Hiroshi Ishii, Shohei Tsujimoto, Akihiro Sakatoku, et al. Parasite-manipulated host dispersal: evidence from population genetics and mark-recapture experiments. Authorea . 23 June 2025. DOI: https://doi.org/10.22541/au.175067513.38872580/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.175067513.38872580/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a00289808a4b300f',t:'MTc3OTUyMzMzMw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();