Infection Prevalence of Microsporidia Vairimorpha (Nosema) spp. in Japanese Bumblebees

Simple Summary Microsporidia are intracellular parasites that have detrimental effects on their various invertebrate and vertebrate hosts. The microsporidia pathogenic to honeybees and bumblebees belong to the genus Vairimorpha (previously Nosema). Among these, V. bombi is the most widespread species worldwide and it infects a variety of bumblebee species. In this study, we investigated the infection of V. bombi among Japanese native bumblebees and the alien species, Bombus terrestris, established in Japan, using PCR and microscopy. We discovered that V. bombi infected the subgenus Bombus s. str. species/subspecies with a low prevalence in areas both with and without the presence of B. terrestris, suggesting that V. bombi may be native to Japan. In contrast, a newly discovered microsporidian parasite that is phylogenetically distinct from V. bombi was found to infect the subgenus Diversobombus species/subspecies with a high prevalence. Hence, if the new microsporidian species does not negatively affect the host and is vertically transmitted from queen to offspring, it may coexist with its host. In addition, this may suggest that V. bombi and the new microsporidium have different life history strategies. Abstract Microsporidia are spore-forming intracellular parasites of various invertebrates and vertebrates. Vairimorpha bombi negatively affects the fitness of bumblebees and its prevalence correlates with declining bumblebee populations. The invasive alien species Bombus terrestris colonized Japan and possibly introduced new parasites. To assess the infection prevalence of V. bombi in Japanese bumblebees and B. terrestris, we investigated V. bombi infections using PCR and microscopy. The prevalence of sporulating V. bombi infections in three Bombus s. str. species/subspecies was low, whereas that of non/low-sporulating Vairimorpha sp. infections in three Diversobombus species/subspecies was high. Invasive B. terrestris showed low prevalence of non/low-sporulating V. bombi infections and shared the same V. bombi haplotype with B. hypocrita found in Hokkaido, where B. terrestris is present, and in Honshu, where B. terrestris is absent. Although V. bombi may have been introduced with B. terrestris colonies imported from Europe, it seems to be originally distributed in Japan. Furthermore, a new Vairimorpha sp. was found in Japanese bumblebee species. V. bombi and Vairimorpha sp. showed different organ and host specificities in bumblebees. There are no reports on the specific effects of other Vairimorpha spp. on bumblebees; further studies are needed to clarify the individual characteristics of Vairimorpha spp.


Introduction
Bumblebees (Apidae: Hymenoptera) are eusocial insects with annual colony cycles. Queens emerge from hibernation in spring and establish new colonies, each producing hundreds of workers during the summer. Declines in wild bumblebee populations have been reported mainly in Europe and the United States; four bumblebee species in Europe have disappeared from eleven countries in the last sixty years [1], and the prevalence of 2016; Koshu, Yamanashi (east side of the Okuchichibu mountain area, east OMA) in 2016, 2019, and 2020; and in Hinohara, Tokyo in 2016. Additional collections were conducted in Oshino, Yamanashi, and Hara, Nagano in 2016. The midgut of each specimen was removed and crushed in 200 µL 1× PBS (1370 mM NaCl, 27 mM KCl, 81 mM Na 2 HPO 4 12H 2 O, and 14.7 mM KH 2 PO 4 ; pH 7.4), and homogenates were stored at −35 • C for the following experiments.

Detection of Vairimorpha spp. Spores
Using 1 µL of the homogenates spread on a glass slide, spore-forming (herein, "sporulating", according to Blaker et al. [11]) Vairimorpha spp. were examined across the center of the cover glass from the left edge to the right edge under a phase-contrast microscope (Nikon, Tokyo, Japan) (400× magnification).
The PCR assay for Vairimorpha spp. detection was performed using 1.0 µL of 10 × Ex Taq Buffer (TaKaRa, Shiga, Japan), 0.8 µL of 2.5 mM each dNTP mixture (TaKaRa), 0.2 µL of each forward and reverse primer (10 µM) for small subunit (SSU) rRNA, Nbombi-SSU-Jfl, and Nbombi-SSU-Jrl [47] (Table S2), 0.05 µL of 5 U/µL Ex Taq HS (TaKaRa), 1 µL of template DNA, and 6.75 µL of sterilized distilled water (SDW). The PCR profile for the detection of Vairimorpha spp. comprised an initial denaturation step at 94 • C for 3 min, followed by 40 cycles at 94 • C for 30 s, 50 • C for 30 s, and 72 • C for 30 s, and a final extension at 72 • C for 5 min. DNA presence was confirmed by PCR using β-actin primers [48] (Table S2) at an annealing temperature of 55 • C. The PCR products were purified using a QIAquick PCR Purification Kit (Qiagen, Hilden, Germany), and sequencing reactions were performed using the BigDye Terminator Kit ver. 3.1 (Applied Biosystems, Foster City, CA, USA). PCR products were directly sequenced using a 3700 DNA analyzer (Applied Biosystems, Foster City, CA, USA). The cloned plasmid DNA was used as a positive control in all PCR assays: Vairimorpha spp.-like spores were harvested from a male B. hypocrita sapporoensis, which was collected in Hokkaido in 2014, and spore DNA was subjected to SSU rRNA gene amplification using the primers Nbombi-SSU-Jfl and Nbombi-SSU-Jrl; the PCR products were purified. The purified PCR products were ligated with the T-vector pMD20 (TaKaRa) using a ligation mix (TaKaRa) as per the manufacturer's instructions. Competent cells (Escherichia coli JM109; TaKaRa) were transformed with plasmids. Plasmid DNA was extracted using the QIAprep ® Spin Miniprep Kit (Qiagen) (Hilden, Germany) according to the manufacturer's protocol and stored at −25 • C.
A phylogenetic tree was constructed based on the SSUr RNA gene sequences amplified from Vairimorpha spp. using the primers SSUrRNA-f1 and SSUrRNA-r1c [17] (Table S2) and gene sequences obtained from GenBank using MEGA 7.0 software [49]. After manual sequence editing, they were aligned and the maximum parsimony method was used to construct a phylogenetic tree for clustering haplotypes. To test the reliability of each clade in the tree, 1000 bootstrap resamplings were performed.

Organs Infected with Vairimorpha spp.
Four organs (midgut, head, thorax, and legs) of bumblebees infected with Vairimorpha spp. were examined for signs of infection using microscopy and PCR to demonstrate its systemic infectivity. Two individuals from each species were used in subsequent experiments. To eliminate spores from the body surface of bumblebees, they were washed with 1× PBS. Then, their head, thorax, and legs were separated using scissors sterilized by autoclaving; only the femur was used for the "legs" sample after separation from other parts. For B. diversus diversus, an ovary was also isolated and checked for Vairimorpha spp. infection to examine the possibility of vertical transmission. Each organ from two individuals was homogenized using a sterile mortar with 200-400 µL of DNAzol and pooled. The homogenates were then transferred to 1.5 mL microtubes.
DNAzol was added to the homogenates up to the 1 mL mark, and 500 mg of glass beads was added. DNA was extracted as described in Section 2.3, with 4 µL proteinase K used here. PCR was performed with 30 ± 5 ng of DNA.

Quantitative PCR
To quantify the severity of Vairimorpha spp. infection among bumblebees, qPCR was performed using a StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). The reaction mixture contained 10 ng DNA, 30 µM of each primer, and 5 µL FastStart Universal SYBR Green Master Mix (Roche, Basel, Switzerland). The cycling conditions were as follows: 2 min at 50 • C and 10 min at 95 • C, followed by 40 cycles of 15 s at 95 • C and 1 min at 60 • C. Dissociation curve analysis of the amplified product was performed after amplification using the specific primers BOMBICAR-F and BOMBICAR-R [50] (Table S2) with the following conditions: 15 s at 95 • C, 20 s at 60 • C, and 15 s at 95 • C. Negative (UPW) controls were used for each run. For each sample, the numbers of Vairimorpha spp. gene copies were calculated using the C t values (mean of two replicates) measured against a standard curve, which was generated using 10-fold serial dilutions from 3.0 × 10 2 to 3.0 × 10 −3 DNA extracted from a solution of 9.8 × 10 5 spores/mL. The spore density, expressed as the number of spores per 100 ng of DNA, was calculated from duplicate measurements per sample.

Prevalence of Vairimorpha spp.
We examined 903 bumblebee specimens belonging to 12 species and subspecies collected in four regions of Japan (Hokkaido, Tokyo, Nagano, and Yamanashi). The overall prevalence of Vairimorpha spp. across all specimens was 19.4%, varying among bumblebee species (Tables 1 and 2, Figure 1).     PCR analysis revealed that Vairimorpha spp. exhibited a low prevalence in Central Hokkaido: 0.8%, 1.4%, and 2.4% of B. terrestris specimens and 10.3%, 11.9%, and 6.7% of B. hypocrita sapporoensis specimens from Hokkaido were positive for V. bombi in 2014, 2016, and 2018, respectively. Additionally, one B. diversus tersatus worker collected from Central Hokkaido in 2014 was positive for Vairimorpha spp. In Central Japan, a low prevalence of Vairimorpha spp. was found in two consubgeneric (Bombus s. str.) species B. ignites and B. hypocrita hypocrita: 1.6% of B. ignitus and 7.6% of B. hypocrita hypocrita collected from west OMA in 2014 and 2016, respectively, were positive for Vairimorpha spp. The prevalence of Vairimorpha spp. in B. hypocrita hypocrita collected from east OMA was relatively high at 60% in 2016 but decreased to 4.3% in 2020. The prevalence of Vairimorpha spp. in B. honshuensis was low at 10% in specimens from east OMA and 0% in specimens from west OMA. The highest prevalence of Vairimorpha spp. was found in two consubgeneric   PCR analysis revealed that Vairimorpha spp. exhibited a low prevalence in Central Hokkaido: 0.8%, 1.4%, and 2.4% of B. terrestris specimens and 10.3%, 11.9%, and 6.7% of B. hypocrita sapporoensis specimens from Hokkaido were positive for V. bombi in 2014, 2016, and 2018, respectively. Additionally, one B. diversus tersatus worker collected from Central Hokkaido in 2014 was positive for Vairimorpha spp. In Central Japan, a low prevalence of Vairimorpha spp. was found in two consubgeneric (Bombus s. str.) species B. ignites and B. hypocrita hypocrita: 1.6% of B. ignitus and 7.6% of B. hypocrita hypocrita collected from west OMA in 2014 and 2016, respectively, were positive for Vairimorpha spp. The prevalence of Vairimorpha spp. in B. hypocrita hypocrita collected from east OMA was relatively high at 60% in 2016 but decreased to 4.3% in 2020. The prevalence of Vairimorpha spp. in B. honshuensis was low at 10% in specimens from east OMA and 0% in specimens from west OMA. The highest prevalence of Vairimorpha spp. was found in two consubgeneric (Diver-

Organ Specificity of Vairimorpha spp.
The infection of Vairimorpha spp. in four organs of bumblebees (midgut, head, thorax, and legs) was investigated using PCR analysis. Two queens of B. hypocrita hypocrita from Yamanashi were sporulating Vairimorpha spp. PCR-positive specimens, and two queens of B. diversus diversus from Nagano and two workers each of B. ussurensis and B. honshuensis from Yamanashi were non/low-sporulating Vairimorpha spp. PCR-positive specimens.
In the midgut and leg homogenates of two workers of B. hypocrita hypocrita, more than five Vairimorpha spp. spores (++) were observed under a phase-contrast microscope at 400× magnification ( Figure 3A), whereas 1-4 spores (+) were observed in the head and thorax homogenates ( Figure 3A). Vairimorpha spp. spores were not observed in any of the B. diversus, B. ussurensis, or B. honshuensis specimens.

Spore Numbers of Vairimorpha spp. among Bumblebees
The number of Vairimorpha spp. spores in the midguts per 100 ng DNA containing the host gene were estimated using quantitative PCR in the following specimens: (1) sporulating Vairimorpha spp. PCR-positive bees, (2) non/low-sporulating Vairimorpha spp. PCRpositive bees, and (3) non/low-sporulating Vairimorpha spp. PCR-negative bees (Table S3, Figure 4). The highest spore numbers were found in the sporulating Vairimorpha spp. PCRpositive bees: approximately 1.6 × 10 5 to 1.1 × 10 6 spores/100 ng DNA in bees with more than five spores/per visual field of the microscope, and approximately 2.5 × 10 4 to 1.4 × 10 5 spores/100 ng DNA in bees with less than five spores/per visual field of the microscope. Conversely, low spore numbers, approximately 8.9 to 1.2 × 10 3 spores/100 ng DNA, were found in non-sporulating Vairimorpha spp. PCR-positive bees. The detection limit for spore concentration as assessed by microscopy was <10 3 spores/100 ng DNA. According to Blaker et al. [11], this number is the threshold for Vairimorpha spp. to be considered non/low-sporulating. As expected, no spores were detected in V. bombi PCR-negative bees, because these bees were not infected with Vairimorpha spp. Using DNA extracted from B. hypocrita hypocrita, PCR analysis revealed amplified Vairimorpha spp.fragments of approximately 300 bp from all four organs ( Figure 3B). For B. honshuensis, B. diversus diversus, and B. ussurensis, similar fragments were detected in the midguts, thoraxes, and legs, but not in the heads ( Figure 3B). DNA was also extracted from the ovaries of B. diversus diversus, and an amplified Vairimorpha spp. fragment of 300 bp was detected.

Spore Numbers of Vairimorpha spp. among Bumblebees
The number of Vairimorpha spp. spores in the midguts per 100 ng DNA containing the host gene were estimated using quantitative PCR in the following specimens: (1) sporulating Vairimorpha spp. PCR-positive bees, (2) non/low-sporulating Vairimorpha spp. PCR-positive bees, and (3) non/low-sporulating Vairimorpha spp. PCR-negative bees (Table S3, Figure 4). The highest spore numbers were found in the sporulating Vairimorpha spp. PCR-positive bees: approximately 1.6 × 10 5 to 1.1 × 10 6 spores/100 ng DNA in bees with more than five spores/per visual field of the microscope, and approximately 2.5 × 10 4 to 1.4 × 10 5 spores/100 ng DNA in bees with less than five spores/per visual field of the microscope. Conversely, low spore numbers, approximately 8.9 to 1.2 × 10 3 spores/100 ng DNA, were found in non-sporulating Vairimorpha spp. PCR-positive bees. The detection limit for spore concentration as assessed by microscopy was <10 3 spores/100 ng DNA. According to Blaker et al. [11], this number is the threshold for Vairimorpha spp. to be considered non/low-sporulating. As expected, no spores were detected in V. bombi PCR-negative bees, because these bees were not infected with Vairimorpha spp.

Phylogenetic Analysis Using Microsporidian SSU rRNA
Sequences of 502 bp were directly amplified from the SSU rRNA genes of 10 individuals of B. hypocrita hypocrita, 11 B. hypocrita sapporoensis, 1 B. terrestris, 16 B. diversus diversus, 6 B. ussurensis, and 2 B. honshuensis. BLAST analysis showed the sequences (GenBank accession number: LC756306-LC756308) from the three Bombus s. str. species were the same and more than 99% identical to the SSUr RNA of V. bombi from Europe, the United States, China, and Russia. The sequences (GenBank accession number: LC756309-LC756312) from the three Diversobombus species and B. honshuensis were the same and shared more than 98% similarity with Nosema sp. from India and China. Phylogenetic analysis using the maximum parsimony method showed that Vairimorpha sp. found in Japan belonged to the same clade as N. bombi, whereas Vairimorpha sp. was close to the clade of Nosema sp. D (Figure 5), suggesting that Vairimorpha sp. is endemic to Japanese bumblebees, and is hereafter referred to as Vairimorpha sp. J.

Phylogenetic Analysis Using Microsporidian SSU rRNA
Sequences of 502 bp were directly amplified from the SSU rRNA genes of 10 individuals of B. hypocrita hypocrita, 11 B. hypocrita sapporoensis, 1 B. terrestris, 16 B. diversus diversus, 6 B. ussurensis, and 2 B. honshuensis. BLAST analysis showed the sequences (GenBank accession number: LC756306-LC756308) from the three Bombus s. str. species were the same and more than 99% identical to the SSUr RNA of V. bombi from Europe, the United States, China, and Russia. The sequences (GenBank accession number: LC756309-LC756312) from the three Diversobombus species and B. honshuensis were the same and shared more than 98% similarity with Nosema sp. from India and China. Phylogenetic analysis using the maximum parsimony method showed that Vairimorpha sp. found in Japan belonged to the same clade as N. bombi, whereas Vairimorpha sp. was close to the clade of Nosema sp. D ( Figure 5), suggesting that Vairimorpha sp. is endemic to Japanese bumblebees, and is hereafter referred to as Vairimorpha sp. J.

Discussion
This study showed that PCR is a more sensitive and specific method for V. bombi detection than microscopy is. The prevalence of sporulating V. bombi was low (6.8%) among the three Bombus s. str. species/subspecies, B. hypocrita hypocrita, B. hypocrita sapporoensis, and B. ignitus, and the prevalence of non/low-sporulating Vairimorpha sp. J was high (96.1%) among the three Diversobombus species/subspecies, B. diversus diversus, B. diversus tersatus, and B. ussurensis. The invasive B. terrestris showed low prevalence of non/low-

Discussion
This study showed that PCR is a more sensitive and specific method for V. bombi detection than microscopy is. The prevalence of sporulating V. bombi was low (6.8%) among the three Bombus s. str. species/subspecies, B. hypocrita hypocrita, B. hypocrita sapporoensis, and B. ignitus, and the prevalence of non/low-sporulating Vairimorpha sp. J was high (96.1%) among the three Diversobombus species/subspecies, B. diversus diversus, B. diversus tersatus, and B. ussurensis. The invasive B. terrestris showed low prevalence of non/lowsporulating V. bombi infections and shared the same haplotype of V. bombi with B. hypocrita collected in Central Hokkaido and Honshu, Japan.
Previous studies have reported the introduction of alien parasites with B. terrestris commercial colonies [51][52][53][54]. For example, V. bombi and other parasitic species have been detected in worker bees and pollen of commercial colonies from Europe [55]. Therefore, it is possible that V. bombi could have been introduced to and subsequently spread with bees escaping from greenhouses to the field. Consequently, it could have horizontally infected the native bumblebee species. Takahashi et al. [44] reported that the prevalence of V. bombi was 7.5% (n = 120) and 0% (n = 100) in B. terrestris and native Bombus spp., respectively, in other areas of Hokkaido and Nagano from 2010 to 2013, suggesting that V. bombi may have been brought with B. terrestris colonies imported from Europe later. However, considering that V. bombi was detected not only in B. hypocrita sapporoensis in Hokkaido, where B. terrestris is present, but also B. hypocrita hypocrita in Honshu, where B. terrestris is absent, and that the higher prevalence of V. bombi in B. hypocrita than B. terrestris suggests that V. bombi originates from Japan. In the future, it will be necessary to distinguish whether V. bombi is introduced or native to clarify the status of the accompanying V. bombi infestation using molecular genetic techniques.
The prevalence of V. bombi in wild bumblebees is often low. Studies using microscopy reported that the overall prevalence of nine different Bombus spp. in the United States was 2.9% (n = 9909) with the highest prevalence being 37.2% in declining B. occidentalis (n = 172) [15]. In Europe, 56.1% of the natural population of B. terrestris in Sweden (n = 380) [14] and 24.7% of B. terrestris/lucorum with 9.0% overall prevalence in 10 different Bombus spp. and Psithrus spp. (n = 446) in Switzerland [13] were infected with V. bombi. Conversely, PCR analysis suggested the mean prevalence was 1.2% (n = 83) for B. terrestris in Spain [21] [22], and 13.6% for four different Bombus spp. (n = 280) in Thailand [23]. The 6.8% overall prevalence of V. bombi in our study was similar to that reported in other areas of the world. However, a high prevalence of V. bombi was found in B. hypocrita hypocrita collected from Koshu, Yamanashi from 2016 to 2019, suggesting that the spreading of V. bombi occurred at the site and ended in 2020 for unknown reasons.
While V. bombi has been reported across a wide range of the Northern Hemisphere [3,11,15,56], other Vairimorpha (Nosema) spp. found in bumblebees have been reported in Asian regions-four species from China and two from Russia [16,17]-indicating a high diversity of microsporidia in bumblebees in Asia. Bumblebees also show high species diversity in Asia, and their geographic origin is considered to be the mountainous region of Asia [57,58]. Based on these facts, the microsporidia of bumblebees may have originated in Asia and then dispersed to Europe and North America as bumblebees expanded their habitats. Specifically, this is a possible explanation for the worldwide occurrence of V. bombi.
In this study, a new Vairimorpha sp. J was detected in species endemic to Japan, such as B. diversus, B. honshuensis, and in B. ussurensis, which is also distributed in far east Russia and Korea [59]. Organ specificity in bumblebees differed between V. bombi and Vairimorpha sp. J: V. bombi showed systemic infectivity, whereas Vairimorpha sp. J did not seem to infect the head tissue of bees ( Figure 3). It is possible that spores contaminated on the surface of the body were detected, but considering differences in spore density between these two species, this might be a species-specific difference in infections for target organs. For example, nine Nosema species of grasshoppers have organ specificity: N. locustae and N. chorthippi infect only the fat body, showing host and tissue specificity, whereas N. pyrgomorphae infects the midgut, fat body, muscle tissue, and gonad [60]. Alternatively, the difference in infected organs might be related to spore density; the spore density of some specimens with sporulating V. bombi was high, whereas that of others with non/lowsporulating Vairimorpha sp. J was low. Vairimorpha sp. J was highly prevalent in B. diversus and B. ussurensis. Since V. bombi infects multiple subgenera of bumblebees [2,3,11,16,17] and can be artificially inoculated into B. diversus [43], it has a wide host range. In contrast, Vairimorpha spp. reported in China and Russia also infect various bumblebee subgenera with a low prevalence [16,17]. Of these, Nosema sp. D contained microsporidia detected from B. trifasciatus (Diversobombus), B. lepides (Pyrobombus), B. friseanus (Melanobombus), and B. festivus (Festivobombus), and 83% of B. trifasciatus specimens were infected with Nosema sp. D, albeit with a small sample size [16]. Vairimorpha sp. J was mostly found in long-tongued Diversobombus species. These infection patterns, combined with molecular phylogenetic analysis, suggest that Vairimorpha sp. J is a distinct species from V. bombi and is more closely related to Nosema sp. D.
The negative effects of V. bombi infection on bumblebees [9,61] and its high prevalence in declining bumblebee species [2] have been reported; however, there are no reports on the effects of other Vairimorpha spp. Vairimorpha sp. J., which is highly prevalent in Diversobombus species/subspecies, but shows low spore density, may be a covert infection, possibly vertically transmitted from queen to offspring. In this scenario, it may provide a fitness benefit to its host by establishing a mutualistic relationship to spread throughout a host population. To clarify the characteristics of Vairimorpha spp., it is necessary to conduct a comparative study of Vairimorpha spp.-infected and non-infected bumblebees in the future.
In conclusion, we revealed that V. bombi and a new microsporidium, Vairimorpha sp. J., are infecting the Japanese native bumblebees and discovered that they may have different life history strategies.
Supplementary Materials: The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/insects14040340/s1, Table S1: List of collection sites; Table S2: Primer list of the study; Table S3: The list of the bumblebees used for quantitative PCR. References [17,43,45,47]  Funding: This study was supported by a research grant provided by The Oshita Foundation.

Data Availability Statement:
The datasets analyzed during the current study are available from the corresponding author upon reasonable request.