Detection and characterization of Wolbachia infection in silkworm.

Wolbachia naturally infects a wide variety of arthropods, where it plays important roles in host reproduction. It was previously reported that Wolbachia did not infect silkworm. By means of PCR and sequencing we found in this study that Wolbachia is indeed present in silkworm. Phylogenetic analysis indicates that Wolbachia infection in silkworm may have occurred via transfer from parasitic wasps. Furthermore, Southern blotting results suggest a lateral transfer of the wsp gene into the genomes of some wild silkworms. By antibiotic treatments, we found that tetracycline and ciprofloxacin can eliminate Wolbachia in the silkworm and Wolbachia is important to ovary development of silkworm. These results provide clues towards a more comprehensive understanding of the interaction between Wolbachia and silkworm and possibly other lepidopteran insects.


Introduction
Wolbachia is a cytoplasmically inherited rickettsia that was found in a wide range of arthropods (Jeyaprakash and Hoy, 2000;Werren and Windsor, 2000;Hilgenboecker et al., 2008).Infections were detected in all of the major insect orders, including Coleoptera, Diptera, Hemiptera/Homoptera, Hymenoptera, Lepidoptera, and Orthoptera (Werren and Windsor, 2000).Wolbachia can regulate host reproduction via cytoplasmic incompatibility (CI), feminization, parthenogenesis and male killing (Werren et al., 2008;Blagrove et al., 2012).Zhou et al. (1998) established a general naming system basing on the sequence of the wsp gene, a single copy gene coding for an outer membrane protein of Wolbachia.They classified Wolbachia into two supergroups, supergroup A and B, and within these a total of eight potential groups could be recognized within the A group and four within group B. As wsp shows relatively high genetic divergence among these strains, the gene has been used extensively in phylogenetic analyses and for microtaxonomic subdivision (Van Meer et al., 1999;Baldo et al., 2005).Several recent studies using FISH or Southern blotting methods reported that Wolbachia genes have been horizontally transferred to host chromosomes (Kondo et al., 2002;Nikoh et al., 2008).Such events are present in a variety of insects, including the fruit fly Drosophila ananassae, a parasitoid wasp of the genus Nasonia, the mosquito Aedes aegypti, the pea aphid Acyrthosiphon pisum, the longicorn beetle Monochamus alternates, and the adzuki bean beetle Callosobruchus chinensis (Dunning Hottop et al., 2007;Klasson et al., 2009;Nikoh and Nakabachi, 2009).
In order to identify the function of Wolbachia in its host, a frequently used method is the elimination of Wolbachia through the use of selective antibiotics (Kuriwada et al., 2010;Voronin et al., 2012).Fytrou et al. (2006) found that Wolbachia-infected Drosophila simulans showed a reduced ability to encapsulate parasitoid eggs compared to a tetracycline-treated, bacterium-free line.Chen et al. (2012) used tetracycline to eliminate Wolbachia in the rice water weevil, Lissorhoptrus oryzophilus, showing that Wolbachia is necessary for its host's oocyte production.
The silkworm, Bombyx mori, is an economically important lepidopteran insect, domesticated from the wild silkworm Bombyx mandarina, which occurs in a range from India to China, Korea, Japan and far into the eastern regions of Russia (Goldsmith et al., 2005).Despite its economic importance, surprisingly few studies focused on Wolbachia in the silkworm.One reason may be that Wolbachia was not detected in the silkworm strains used in previous studies (Puttaraju and Madhu, 2002;Prakash and Puttaraju, 2007).In this study, we amplified the Wolbachia-specific wsp gene in silkworm by PCR and cloned the gene.Our results indicate that Wolbachia is present in several strains of silkworm and antibiotics treatment revealed that Wolbachia plays an important role in silkworm ovary development.

Silkworm samples
The silkworms examined in this study, information about strain and collection locations are listed in Supplementary material Table S1.Samples of adult silkworm moths were collected in 2008 and stored in absolute ethanol at -80 °C.Total DNA of was extracted from each single strain (one individual per sample).DNA extraction was performed by using the QIAamp DNA Mini kit (Qiagen) for PCR amplification or by a standard proteinase K/SDS/phenol/chloroform extraction procedure for Southern blotting (Green and Sambrook, 2012).

PCR and cloning of the wsp gene
Wolbachia infection was detected by polymerase chain reaction (PCR) using the following wsp-specifc primers: wsp-81F (5'-TGGTCCAATAAGTGATGAAGA AAC-3'), wsp-691R (5'-AAAAATTAAACGCTACTC CA-3') (Braig et al., 1998;Zhou et al., 1998), resulting in an amplified DNA fragment of about 600 bp.PCR amplications were performed in 25 mL reactions containing 1 mL of DNA, 2.5 mL 10x reaction buffer, 2.0 mL MgCl 2 (25 mM), 1 mL dNTPs (25 uM), 0.2 mL of Taq polymerase (Takara) and 18.3 mL water.PCRs were run under the following cycling conditions: 94 °C for 4 min, followed by 30 cycles of 40 s at 94 °C, 40 s at 55 °C, 1 min at 72 °C and a final extension step of 10 min at 72 °C.The PCR products were electrophoresed on a 1% agarose gel to determine the presence and general size of the amplified DNA.Considering the possibility of false positive PCR results, all the amplified PCR fragments were sequenced.For this, PCR products were purified by a Gel Extraction Kit (Omega) and purified DNA was ligated into a pMD18-T vector (Takara) for transformation of Escherichia coli DH5a competent cells.Positive clones were picked and sequenced.An additional strategy to detect Wolbachia infection was to design specific primers for the Wolbachia ftsZ gene: ftsZ-F: TACTGACTGTTGGAGTTGTAACTAAGCCGT and ftsZ-R: TGCCAGTTGCAAGAACAGAAACTCTAACTC.The resulting PCR fragments were cloned and sequenced in the same way.

Phylogenetic analysis
All Wolbachia wsp gene sequences obtained in this study were aligned by using the program package ClustalW (Thompson et al., 1994) or MEGA3.1 (Kumar et al., 2008).Sequences downloaded from GenBank representing all currently known supergroups of Wolbachia were included in the analysis (Table S2).Phylogenetic analyses were conducted by the Neighbour-Joining algorithm.Bootstrap probabilities were assessed by generating 1000 bootstrap replicates.

Southern blotting analysis
DNA probes for Southern blotting were synthesized by PCR amplification from recombinant wsp plasmid DNA as template using the same wsp gene specific primers.The amplification products were electrophoresed in 1% agarose gels and purified.Digoxigenin labeling was done by using the DIG Random Primed DNA Labeling Kit (Roche).Genomic DNA preparations of Bombyx mori Dazao and wild silkworms were digested with HindIII restriction enzyme.The digested DNA samples of 25-30 mg were electrophoresed in 0.8% agarose gels.Plasmid DNA of recombinant wsp gene vector was used as positive control.The separated DNA fragments were transferred to nylon membranes and fixed by UV cross-linking.Hybridization, stringency washes and detection were performed by using the DIG Detection Kit (Roche) following the manufacturer's manual.

Antibiotic treatment
The Bombyx mori strain Dazao used in this experiment was provided by the Silkworm Gene Resource of Southwest University.Silkworm larvae were reared under standard conditions.Antibiotics used included tetracycline, rifampicin, ciprofloxacin and penicillin (Table S3).Antibiotics were dissolved in double-distilled water at a concentration of 100 mg/mL.After the first day of pupation, each silkworm was injected with a volume of 10 mL (1 mg of the antibiotic) into the eighth abdominal segment by using needles pulled from glass capillaries.Negative-control silkworms were injected with the same volume of water.Individual silkworms were collected 10 days after injection and DNA was extracted as per the above described protocol.Quantitative PCR was used to determine Wolbachia abundance in silkworms.In order to understand the effect of antibiotic treatment, Wolbachia density of untreated samples (Infection free) was also detected by Quantitative PCR.
Given a probable side effect of a high dose of antibiotic, injections were also done with a lower dose of tetracycline (10 mg/mL).Injection volume, time and location were the same as in the former experiment.Furthermore, tetracycline-injected female and male silkworms were mated and the F1 generation was reared to assess the ovary phenotype adult F1 females.

Quantitative PCR
The abundance of Wolbachia after antibiotic treatments was determined by quantitative PCR.Platinum SYBR green qPCR SuperMix-UDG (Invitrogen) was used according to the manufacturer's protocol.Real-time quantitative PCR was performed using the ABI Prism 7000 Sequence Detection System (Applied Biosystems).Primer sequences corresponded to regions of the Bombyx mori reference gene sw22934 (Wang et al., 2008)  Wolbachia wsp gene.The sw22934 primer sequences used were 5'-TTCGTACTGGCTCTTCTCGT-3' and 5'-CAAAGTTGATAGCAATTCCCT-3', while the wsp primer sequences were 5'-AGATAGTGTAACAGCGTT TTCAGGAT-3' and 5'-CACCATAAGAACCAAAAT AACGAG-3'.Template-free qPCRs were included as negative controls.Positive controls for the reference gene and Wolbachia were also included in each qPCR run.The instrument was programmed to provide initial enzyme activation at 95 °C for 10 s, followed by 40 cycles of 95 °C for 5 s and 60 °C for 31 s.Amplification efficiency was calculated according to the method of Tichopad et al. (2003).

Screening for Wolbachia infections in silkworm populations
By means of a PCR approach using the general wspspecific primers we screened a total of 21 samples of silkworms for the presence of Wolbachia.The samples were collected in seven provinces of China (Chongqing, Yunnan, Sichuan, Shandong, Zhejiang, Jiangsu and Guangdong) and represented two types of silkworms (domesticated and wild silkworms).The results showed that PCR amplification products with the expected size of about 600 bp were observed (Figure 1).In order to confirm PCR products, we cloned and sequenced the fragments, showing that the 21 silkworm samples share high sequence similarity with differences in only a few sites (Figures S1 and S2).Taken together, these results demonstrate that wsp gene fragments could be amplified in all 21 silkworm samples.The sequence of the silkworm wsp gene was submitted to GenBank (Acc.No. KJ659909).Homology searches with BLASTn indicated that the sequence in the Dali wild silkworm sample were the best hit to the wsp gene of Wolbachia in Orius minutus (Tsukuba), with 98% identity.Additionally, PCR assays were carried out by using ftsZ-specific primers in wild silkworm Dali and Dazao (Figure 1C), and the sequencing results showed that the PCR bands were indeed the ftsZ gene of Wolbachia.The sequence of ftsZ was also submitted to GenBank (Acc.No. KJ659910).In conclusion, the PCR results for the wsp and ftsZ genes are evidence that there are Wolbachia infections in silkworm.

Phylogenetic analysis
Based on previous reports (Zhou et al., 1998;Van Meer et al., 1999), we downloaded 41 wsp sequences of Wolbachia as a data set (Table S2).First, we merged our wsp sequence from the sample of Dali wild silkworm with the data set and performed a phylogenetic analysis.The topology confirmed the division of Wolbachia into the two supergroups A and B, showing that our Wolbachia belongs to supergroup B (Figure 2).Wolbachia strains from Lepidoptera E. kuehniella, E. cautella and silkworm appeared to cluster in different groups.Interestingly, Wolbachia from silkworm was closely related to that of the parasitic wasp T. bedeguaris.Considering the possibility of a horizontal transmission of Wolbachia (Perlman et al., 2006) we hypothesize that Wolbachia detected in wild silkworm may have come from Hymenoptera, such as T. confusum, T. bedeguaris or others.Further experiments are, of course, required to test this hypothesis of horizontal transmission.In a next step we constructed a phylogenetic tree based on our 21 wsp sequences, showing that Wolbachia identified in domesticated and wild silkworms are quite similar to each other (Figure 3), indicating that Wolbachia had infected certain silkworms before strains of this species were domesticated.

Lateral transfer
It was previously reported that Wolbachia genes could laterally transfer into host chromosomes (Kondo et al., 2002;Fenn et al., 2006;Nikoh et al., 2008).The genome sequence of the silkworm, Bombyx mori, has been completed, and the strain for sequencing was Dazao (Xia et al., 2004; International Silkworm Genome Consortium,  as probe.This analysis showed that no wsp hybridization signal was detected in the Dazao genome, but was clearly apparent in the wild type silkworms of Huzhou, Xichan and Pingdu (Figure 4).Since Southern blotting had also previously been used to identify horizontal transfer of Wolbachia (Nikoh et al., 2008), we speculate that Wolbachia did not transfer into the Dazao genome, but that a wsp gene had done so in the wild silkworm genomes of Huzhou, Xichan and Pingdu.

Effects of Wolbachia on silkworm ovary development
In this study, we used four kinds of antibiotics, tetracycline, penicillin, rifampin and ciprofloxacin (Table S3).These were respectively injected into the body of 30 silkworms (half male, half female silkworms).The injected dose was 1 mg per individual.After injection, the survival rate was 100% in all experimental groups.But while in four groups (rifampicin, penicillin, autoclaved H 2 O and untreated controls) the silkworms could all normally emerge as adult moths, those injected with tetracycline or ciprofloxacin did not.By dissecting silkworms injected with tetracycline or ciprofloxacin we noted an abnormal ovary phenotype in female moths with a frequency of 93% and more (Table S3).Oviducts lacked the typical structure and were clearly shorter.The number of developing eggs was also lower than in the control groups (Figure 5).In order to verify the elimination of Wolbachia by the antibiotics, we performed quantitative PCR assays using wsp primers, this showing showed that the density of Wolbachia in the group injected with tetracycline or ciprofloxacin was strongly de-  creased (Figure 6).These results indicated that tetracycline and ciprofloxacin could dramatically reduce the density of Wolbachia in silkworm, and that the presence of Wolbachia is important for normal development of the ovary.
In view of a probable side effect of a high dose of antibiotic treatment, we repeated the experiment injecting a lower dose of tetracycline (0.1 mg per individual).In this experiment, silkworms were furthermore mated to obtain an F1 generation.Compared with the control group, nearly 29% of the F1 females still showed an abnormal ovary morphology (Figure 7), putting in evidence that a lower dose of tetracycline, for which no or less of a side effect was expected, lead to a similar effect.

Discussion
Wolbachia infects a wide variety of insects and plays important roles in host reproduction.As it was previously reported that Wolbachia was not detected in silkworm (Puttaraju and Madhu, 2002;Prakash and Puttaraju, 2007), we conducted a study on both wild type and domesticated silkworm strains using PCR assays for the Wolbachia-spe-cific wsp gene and subsequent sequencing.This is the first study to detect the presence of Wolbachia in silkworm.Interestingly, in the PCR screening of the wsp gene we found that the band of a PCR fragment in lanes 1 and 6-8 in Figure 1B showed a stronger signal, suggesting that the wsp gene could easily be amplified and detected by PCR in these four samples.It has previously been reported that Wolbachia infections in some species could be detected by common PCR, while in others long PCR assays under strict conditions were necessary (Jeyaprakash and Hoy, 2000).These results may imply a difference in the density of Wolbachia infecting the hosts.
As reported, Wolbachia can play important roles in lepidopteran insects.In the butterfly Eurema hecabe, a female-biased sex-ratio distortion was observed due to feminization of genetic males by Wolbachia (Hiroki et al., 2002).In the Mediterranean flour moth, Ephestia kuehniella Lewis et al. (2011) showed that Wolbachia-infected males transfer fewer fertile sperm at mating than uninfected ones in their heteromorphic, sperm, and that Wolbachia may affect fertile sperm production.In the adzuki bean borer, Ostrinia scapulalis (Walker), Wolbachia selectively kills male progeny.This Wolbachia strain appears to have a feminizing effect since antibiotic treatment of infected female moths gave rise to male progeny with sexually mosaic phenotypes (Kageyama et al., 2003).In Wolbachia-induced sexual mosaics of O. scapulalis, which are genetically male, the female-specific isoform of the sex determining gene Osdsx was shown to be expressed in addition to the male-specific isoform (Sugimoto et al., 2010).This finding indicated that Wolbachia manipulates the sex of its host by interfering either with the    sex-specific splicing of Osdsx itself or with another upstream sex determination process..By injection of different antibiotics for elimination of Wolbachia in the silkworm, we could show that the elimination of large amounts of Wolbachia by tetracycline and ciprofloxacin can, cause abnormal development of the silkworm ovary.Apparently, Wolbachia thus plays an important role in silkworm ovary development and that the biological role of Wolbachia in silkworm is different from that in other lepidopteran insects.These results provide some first clues to address the role of Wolbachia silkworm biology and possibly also in other lepidopteran insects.

Figure 2 -
Figure 2 -Phylogenetic tree of Wolbachia based on wsp gene sequences.The name of the host arthropod species was followed by the group designation.The name `Bombyx mandarina(Dali)' represents Wolbachia in wild silkworm from Dali in Yunnan.The tree was constructed by the Neighbour-Joining algorithm.Numbers on the nodes indicated percentages of 1,000 bootstrap replicates.

Figure 3 -
Figure 3 -Phylogenetic tree of Wolbachia in silkworms.All 21 wsp sequences from samples of silkworms were used to construct the tree.

Figure 4 -
Figure 4 -Southern blotting analysis of the wsp gene using genomic DNA from different Wolbachia infected silkworms.The blot was hybridized with a probe corresponding to the wsp gene.Lane C represents a positive control from plasmid DNA of a recombinant wsp gene.Lanes 1-3 represent wild silkworm samples from Huzhou, Xichan and Pingdu, respectively.Lane 4 represents F1 silkworms from a Male Dali x Female Dazao cross.Lanes 5 and 6 represent female and male samples of the Dazao strain, respectively.

Figure 5 -
Figure 5 -Phenotype of ovaries after antibiotic treatments.The two groups of H 2 O and untreated were used as control.The groups of tetracycline and ciprofloxacin treatment showed an abnormal ovary phenotype compared with that in the control groups.

Figure 6 -
Figure 6 -Relative Wolbachia abundance after antibiotic treatments.The graphs show Wolbachia abundance relative to the host silkworm reference gene (sw22934) as determined by quantitative PCR.The error bars represent standard deviations.

Figure 7 -
Figure7-Ovary phenotype of F1 generation females after tetracycline treatment.The injected H 2 O group was used as control.Tetracycline was injected with a lower dose of 0.1 mg per silkworm.After antibiotic treatments, tetracycline-injected females were crossed with tetracyclineinjected males.The F1 generation was reared and used to identify the ovary phenotype.Nearly 29% of the F1 female still showed an abnormal ovary morphology.