Silencing of cyp-33C9 Gene Affects the Reproduction and Pathogenicity of the Pine Wood Nematode, Bursaphelenchus xylophilus

Cytochrome P450 genes are very important for plant-parasitic nematodes to reproduce and to metabolize xenobiotic compounds generated by their host plants. The pine wood nematode (PWN), Bursaphelenchus xylophilus, causes very high annual economic losses by killing large numbers of pine trees across Asia and into Europe. In this study, we used RNA interference (RNAi) to analyze the function of the cyp-33C9 gene of PWN. Our results showed that expression of the cyp-33C9 gene was suppressed successfully after soaking nematodes for 24 h in cyp-33C9 double-stranded RNA (dsRNA). The silencing of the cyp-33C9 gene significantly decreased the feeding, reproduction, oviposition and egg hatch of B. xylophilus. Meanwhile, the migration speed of B. xylophilus in Pinus thunbergii was reduced in the early stages when the cyp-33C9 gene was silenced in the nematodes. Moreover, knockdown of the cyp-33C9 gene in B. xylophilus caused a decrease in pathogenicity to pine trees. These results suggest that the cyp-33C9 gene plays an important role in the reproduction and pathogenicity of B. xylophilus. This discovery identified several functions of the cyp-33C9 gene in B. xylophilus and provided useful information for understanding the molecular mechanism behind pine wilt disease caused by PWN.


Introduction
Pine wilt disease (PWD) is one of the most serious diseases of forest trees, affecting conifers around the world [1]. As the causal agent of PWD, the pine wood nematode (PWN), Bursaphelenchus xylophilus (Steinr and Buhrer) Nickle, is a migratory plant endoparasite [2]. B. xylophilus is considered to be native to North America, where it causes little damage to the native pine species which are resistant to the nematode [3]. However, once introduced into a new area where no natural resistance is present in the conifers, B. xylophilus has caused the death of millions of pine trees, resulting in huge economic losses [4]. At the start of the 20th century, the PWN was introduced into Japan, from where it has 2. Results

Effects of RNA Interference (RNAi) on Feeding and Reproduction of Bursaphelenchus xylophilus
The experiments to study the effects of RNAi on feeding and reproduction of B. xylophilus were performed on Botrytis cinerea cultures on potato dextrose agar (PDA) plates. The green fluorescent protein (gfp) gene dsRNA treatment was used to evaluate the influence of non-endogenous dsRNA on feeding and reproduction of B. xylophilus. The nematodes soaked in cyp-33C9 dsRNA solution showed significantly reduced population size compared with that of nematodes soaked in ddH 2 O (control, CK) or gfp dsRNA solution ( Figure 1A). After five days incubation post-treatment on B. cinerea plates at 25 • C, the number of B. xylophilus second-stage juveniles (J2s) per plate after inoculation with J2s treated with cyp-33C9 dsRNA, ddH 2 O (CK) or gfp dsRNA solution were 97,970 and 890, respectively ( Figure 1B). These results suggested that feeding and reproduction of B. xylophilus were significantly suppressed by cyp-33C9 dsRNA interference.  Figure 1B). These results suggested that feeding and reproduction of B. xylophilus were significantly suppressed by cyp-33C9 dsRNA interference.

Effects of RNA Interference (RNAi) on Oviposition of Bursaphelenchus xylophilus
To further assess the influence of the cyp-33C9 gene on the fecundity of B. xylophilus, the number of eggs laid per female nematode was determined. The results showed that the number of eggs laid per female nematode soaked in ddH2O and gfp dsRNA solution were 23 and 22, respectively ( Figure  2). However, the number of eggs laid per female nematode soaked in cyp-33C9 dsRNA solution was only 10, which was much fewer than that from nematodes treated with ddH2O or gfp dsRNA solution. These results indicated that the gene silencing of cyp-33C9 by dsRNA interference has a significant inhibitory effect on oviposition of B. xylophilus.

Effects of RNA Interference (RNAi) on Oviposition of Bursaphelenchus xylophilus
To further assess the influence of the cyp-33C9 gene on the fecundity of B. xylophilus, the number of eggs laid per female nematode was determined. The results showed that the number of eggs laid per female nematode soaked in ddH 2 O and gfp dsRNA solution were 23 and 22, respectively ( Figure 2). However, the number of eggs laid per female nematode soaked in cyp-33C9 dsRNA solution was only 10, which was much fewer than that from nematodes treated with ddH 2 O or gfp dsRNA solution. These results indicated that the gene silencing of cyp-33C9 by dsRNA interference has a significant inhibitory effect on oviposition of B. xylophilus. The bars indicate standard deviation, and different letters indicate significant differences (p < 0.05) among treatments.

Effects of RNAi on Hatch of Bursaphelenchus xylophilus
The effects of RNAi silencing of cyp-33C9 on percentage hatch of B. xylophilus were similar to those on oviposition. The percentage hatch of B. xylophilus with cyp-33C9 dsRNA treatment was 52%, which was lower than that with ddH2O or gfp dsRNA treatments ( Figure 3). No significant differences were observed between ddH2O and gfp dsRNA treatments, indicating that the non-endogenous dsRNA had no obvious negative effect toward the nematodes. The bars indicate the standard deviation, and different letters indicate significant differences (p < 0.05), according to Tukey's test, among the treatments.

Effects of RNA Interference (RNAi) on Individual Body Length of Bursaphelenchus xylophilus
After being soaked in cyp-33C9 dsRNA solution, the individual body length of both the male and female J2s was slightly but not significantly shorter (p > 0.05) than that following soaking in ddH2O or gfp dsRNA solution ( Figure 4). The body length of B. xylophilus following ddH2O treatment was similar to that following gfp dsRNA treatment. No significant differences in body lengths

Effects of RNAi on Hatch of Bursaphelenchus xylophilus
The effects of RNAi silencing of cyp-33C9 on percentage hatch of B. xylophilus were similar to those on oviposition. The percentage hatch of B. xylophilus with cyp-33C9 dsRNA treatment was 52%, which was lower than that with ddH 2 O or gfp dsRNA treatments ( Figure 3). No significant differences were observed between ddH 2 O and gfp dsRNA treatments, indicating that the non-endogenous dsRNA had no obvious negative effect toward the nematodes.

Effects of RNAi on Hatch of Bursaphelenchus xylophilus
The effects of RNAi silencing of cyp-33C9 on percentage hatch of B. xylophilus were similar to those on oviposition. The percentage hatch of B. xylophilus with cyp-33C9 dsRNA treatment was 52%, which was lower than that with ddH2O or gfp dsRNA treatments ( Figure 3). No significant differences were observed between ddH2O and gfp dsRNA treatments, indicating that the non-endogenous dsRNA had no obvious negative effect toward the nematodes. The bars indicate the standard deviation, and different letters indicate significant differences (p < 0.05), according to Tukey's test, among the treatments.

Effects of RNA Interference (RNAi) on Individual Body Length of Bursaphelenchus xylophilus
After being soaked in cyp-33C9 dsRNA solution, the individual body length of both the male and female J2s was slightly but not significantly shorter (p > 0.05) than that following soaking in ddH2O or gfp dsRNA solution ( Figure 4). The body length of B. xylophilus following ddH2O treatment was similar to that following gfp dsRNA treatment. No significant differences in body lengths The bars indicate the standard deviation, and different letters indicate significant differences (p < 0.05), according to Tukey's test, among the treatments.

Effects of RNA Interference (RNAi) on Individual Body Length of Bursaphelenchus xylophilus
After being soaked in cyp-33C9 dsRNA solution, the individual body length of both the male and female J2s was slightly but not significantly shorter (p > 0.05) than that following soaking in ddH 2 O or gfp dsRNA solution ( Figure 4). The body length of B. xylophilus following ddH 2 O treatment was similar to that following gfp dsRNA treatment. No significant differences in body lengths between the three different treatments were observed, suggesting that cyp-33C9 dsRNA interference had no significant effect on the individual body length of B. xylophilus.
between the three different treatments were observed, suggesting that cyp-33C9 dsRNA interference had no significant effect on the individual body length of B. xylophilus. The bars indicate standard deviation, and different letters within a gender indicate significant differences (p < 0.05), according to Tukey's test, among the treatments.

Determination of RNA Interference (RNAi) on Expression of cyp-33C9 in Bursaphelenchus xylophilus
Quantitative reverse transcription PCR (qRT-PCR) was performed to test the effects of RNAi on the expression levels of the cyp-33C9 gene in nematodes soaked in target gene dsRNA solution, compared with the corresponding controls. The average expression level of the cyp-33C9 gene in B. xylophilus with ddH2O and gfp dsRNA treatments were 0.92 and 0.91, respectively ( Figure 5). However, the mean expression level of the cyp-33C9 gene in nematodes soaked in target dsRNA solution was only 0.12 (p < 0.05). These results suggested that the cyp-33C9 gene was silenced by dsRNA interference and that non-endogenous dsRNA had no influence on the expression of the target gene.

Determination of RNA Interference (RNAi) on Expression of cyp-33C9 in Bursaphelenchus xylophilus
Quantitative reverse transcription PCR (qRT-PCR) was performed to test the effects of RNAi on the expression levels of the cyp-33C9 gene in nematodes soaked in target gene dsRNA solution, compared with the corresponding controls. The average expression level of the cyp-33C9 gene in B. xylophilus with ddH 2 O and gfp dsRNA treatments were 0.92 and 0.91, respectively ( Figure 5). However, the mean expression level of the cyp-33C9 gene in nematodes soaked in target dsRNA solution was only 0.12 (p < 0.05). These results suggested that the cyp-33C9 gene was silenced by dsRNA interference and that non-endogenous dsRNA had no influence on the expression of the target gene. between the three different treatments were observed, suggesting that cyp-33C9 dsRNA interference had no significant effect on the individual body length of B. xylophilus.

Figure 4. Effects of RNA interference on individual body length of Bursaphelenchus xylophilus.
The bars indicate standard deviation, and different letters within a gender indicate significant differences (p < 0.05), according to Tukey's test, among the treatments.

Determination of RNA Interference (RNAi) on Expression of cyp-33C9 in Bursaphelenchus xylophilus
Quantitative reverse transcription PCR (qRT-PCR) was performed to test the effects of RNAi on the expression levels of the cyp-33C9 gene in nematodes soaked in target gene dsRNA solution, compared with the corresponding controls. The average expression level of the cyp-33C9 gene in B. xylophilus with ddH2O and gfp dsRNA treatments were 0.92 and 0.91, respectively ( Figure 5). However, the mean expression level of the cyp-33C9 gene in nematodes soaked in target dsRNA solution was only 0.12 (p < 0.05). These results suggested that the cyp-33C9 gene was silenced by dsRNA interference and that non-endogenous dsRNA had no influence on the expression of the target gene.

Effects of RNA Interference (RNAi) on Pine Wood Nematode Migration and Reproduction of Bursaphelenchus xylophilus in Pinus thunbergii
In order to study the migration of pine wood nematodes in the host after silencing of the cyp-33C9 gene by RNAi, we counted the number of B. xylophilus in different parts of pine trees at different sampling times following inoculation of J2s which had been treated with ddH 2 O or dsRNA. For ddH 2 O-treated J2s, there were 66 nematodes at 5 cm above the site of inoculation, but no nematodes existed at 10 or 15 cm above the site at day 5 after inoculation, whereas the number of nematodes at 5, 10 and 15 cm below the site of inoculation were 256, 57 and 9, respectively (Table 1). For cyp-33C9 dsRNA treatment, the distribution of nematodes in P. thunbergii was significantly different from that in the ddH 2 O treatment.
Down-10 cm Down-15 cm The numbers in the first column show the distance from the inoculation sites. The other numbers in the table represent the number of B. xylophilus from the different treatments which were recovered at each site after inoculation of 2000 s-stage juveniles at each inoculation site. The data represent mean values ± standard deviation (SD) from three independent experiments. Different lowercase letters indicate significant differences (p < 0.05), using Tukey's test, among treatments.
These results showed that, after being inoculated into P. thunbergii, B. xylophilus nematodes migrated in two directions (up and down), with the speed of upward migration being slower than that of downward migration. In addition, the migration speed of B. xylophilus treated with ddH 2 O in P. thunbergii was faster than those treated with cyp-33C9 dsRNA (Table 1). At day 10 after inoculation, the nematodes treated with ddH 2 O or cyp-33C9 dsRNA solution had migrated to all monitored sites of the pine trees, but the number of migrated nematodes in P. thunbergii following treatment with the cyp-33C9 dsRNA was significantly lower than that of nematodes treated with ddH 2 O. This result showed that the migration speed of B. xylophilus treated with ddH 2 O in pine trees was faster than that of J2s treated with cyp-33C9 dsRNA, indicating that cyp-33C9 dsRNA treatment has a negative effect on the migration of B. xylophilus. The number of PWN J2s in P. thunbergii increased dramatically after 10 d of inoculation for all three treatments and reached the maximum at day 20 (Table 1). Interestingly, no significant difference was observed between the numbers of J2s from the three treatments at the different monitoring sites at 25 d after inoculation, presumably as the numbers of cyp-33C9 dsRNA-treated J2s had caught up with the control J2s (Table 1). The migration speed of the control J2s was not significantly different from that of the gfp dsRNA-treated J2s throughout the study, indicating that non-endogenous dsRNA had no influence on the migration speed.

Effects of RNA Interference (RNAi) on Pathogenicity of Bursaphelenchus xylophilus
After inoculated with B. xylophilus for 25 d, the symptoms in P. thunbergii were observed to determine whether treatment with RNAi affected the pathogenicity of the nematodes. Of the three nematode treatments, the 2-year-old P. thunbergii seedlings inoculated with B. xylophilus soaked in ddH 2 O appeared to wilt first, with the percentage needle wilting reaching 100% at 25 d after inoculation (Figure 6). At the same time, the percentage needle wilting of the P. thunbergii seedlings inoculated with B. xylophilus soaked in cyp-33C9 dsRNA solution was significantly lower at 44%. In contrast, the CK P. thunbergii seedlings mock-inoculated with water (no J2s) grew well throughout the whole experiment process. These results indicated that silencing of the cyp-33C9 gene decreased the percentage wilting of P. thunbergii seedlings and reduced the pathogenicity of B. xylophilus. values ± standard deviation (SD) from three independent experiments. Different lowercase letters indicate significant differences (p < 0.05), using Tukey's test, among treatments.

Effects of RNA interference (RNAi) on Pathogenicity of Bursaphelenchus xylophilus
After inoculated with B. xylophilus for 25 d, the symptoms in P. thunbergii were observed to determine whether treatment with RNAi affected the pathogenicity of the nematodes. Of the three nematode treatments, the 2-year-old P. thunbergii seedlings inoculated with B. xylophilus soaked in ddH2O appeared to wilt first, with the percentage needle wilting reaching 100% at 25 d after inoculation (Figure 6). At the same time, the percentage needle wilting of the P. thunbergii seedlings inoculated with B. xylophilus soaked in cyp-33C9 dsRNA solution was significantly lower at 44%. In contrast, the CK P. thunbergii seedlings mock-inoculated with water (no J2s) grew well throughout the whole experiment process. These results indicated that silencing of the cyp-33C9 gene decreased the percentage wilting of P. thunbergii seedlings and reduced the pathogenicity of B. xylophilus. Figure 6. Wilting symptoms of P. thunbergii seedlings at 25 d after inoculation with Bursaphelenchus xylophilus soaked in ddH2O ("ddH2O") or cyp-33C9 dsRNA solution ("cyp-33C9"), respectively. Controls ("CK") were represented by P. thunbergii seedlings inoculated with ddH2O alone (no nematodes).

Discussion
The cytochrome P450s are considered to be involved in reproductive development in nematodes [17,18]. In the present study, we found that the cyp-33C9 gene was silenced when B. xylophilus was soaked 1 μg·μL -1 in cyp-33C9 dsRNA solution. Xu et al. observed a similar result (with gene expression of 0.05) when the cyp-33C9 dsRNA solution was 800 ng·μL -1 [16]. These results indicated that different cyp-33C9 dsRNA solutions did not influence RNAi. Our results showed that silencing of the cyp-33C9 gene decreased feeding and reproduction of B. xylophilus. In order to shed further light on the influences of cyp-33C9 gene on B. xylophilus, we tested the effects of RNAi on the oviposition, percentage hatch and individual body length of B. xylophilus after silencing the cyp-33C9 gene, to determine whether knockdown of the cyp-33C9 gene affected growth and development of B. xylophilus.
Eggshell formation is one of the most important factors affecting the hatching characteristics of phytoparasitic nematodes. Previous studies had shown that the eggshell of phytoparasitic nematodes is a structure with three crucial layers: vitelline layer, chitin layer and lipid-rich layer [35][36][37]. Benenati et al. studied the functions of cyp-31A2 and cyp-31A3 genes in C. elegans, using RNAinterference technology, and they discovered that both of these genes played an essential role in the development of the eggshell, especially the lipid-rich layer [17]. Their studies also showed that

Discussion
The cytochrome P450s are considered to be involved in reproductive development in nematodes [17,18]. In the present study, we found that the cyp-33C9 gene was silenced when B. xylophilus was soaked 1 µg·µL −1 in cyp-33C9 dsRNA solution. Xu et al. observed a similar result (with gene expression of 0.05) when the cyp-33C9 dsRNA solution was 800 ng·µL −1 [16]. These results indicated that different cyp-33C9 dsRNA solutions did not influence RNAi. Our results showed that silencing of the cyp-33C9 gene decreased feeding and reproduction of B. xylophilus. In order to shed further light on the influences of cyp-33C9 gene on B. xylophilus, we tested the effects of RNAi on the oviposition, percentage hatch and individual body length of B. xylophilus after silencing the cyp-33C9 gene, to determine whether knockdown of the cyp-33C9 gene affected growth and development of B. xylophilus.
Eggshell formation is one of the most important factors affecting the hatching characteristics of phytoparasitic nematodes. Previous studies had shown that the eggshell of phytoparasitic nematodes is a structure with three crucial layers: vitelline layer, chitin layer and lipid-rich layer [35][36][37]. Benenati et al. studied the functions of cyp-31A2 and cyp-31A3 genes in C. elegans, using RNA-interference technology, and they discovered that both of these genes played an essential role in the development of the eggshell, especially the lipid-rich layer [17]. Their studies also showed that simultaneous depletion of cyp-31A2 and cyp-31A3 genes of C. elegans resulted in osmotic imbalance within the egg, impaired establishment of polarity and incorrect execution of meiosis, which eventually led to the death of C. elegans embryos. Interestingly, their findings mirrored those of Piano et al. who reported that the silencing of the cyp-31A2 gene of C. elegans caused embryonic lethality [38].
Cytochrome P450 genes also play an important role in the oviposition and egg hatch of the parasite. In this study, our results showed that the number of eggs laid per female B. xylophilus was significantly lower after the cyp-33C9 gene was silenced, and that approximately 50% of the B. xylophilus eggs produced were not capable of hatching. These results indicated that cyp-33C9 gene had a marked effect on reproduction by negatively impacting both the egg-laying capacity and the percentage egg hatch of B. xylophilus. The explanation for this may be that, at the early stage of B. xylophilus embryo development, the cyp-33C9 gene was required for formation of the lipid-rich layer of the embryo. The absence of the lipid-rich layer would have resulted in an inappropriate construction of the eggshell, which would then affect the egg-laying capacity of B. xylophilus [17]. Meanwhile, deletion of the cyp-33C9 gene of B. xylophilus appeared to result in abnormal cell division and cortical polarization of eggs, which consequently resulted in a decline in the number of hatched eggs of B. xylophilus [17]. In the present study, knockdown of the cyp-33C9 gene of B. xylophilus may be responsible directly for the abnormal development of the eggshell, which represents the permeability barrier for the embryos, ultimately leading to the decrease in oviposition number and percentage hatch of B. xylophilus eggs. In addition, cytochrome P450s have been reported to play important roles in hormone biosynthesis and metabolism [39]. The inhibited growth (as body length) and reproduction exhibited by the cyp-33C9-silenced B. xylophilus may be because the silencing of the cyp-33C9 gene inhibited the biosynthesis of hormones. The data from the present study showed that silencing of the cyp-33C9 gene, reduced the individual body length of B. xylophilus by 5-10%, albeit not significantly. It has been reported that the body size of B. xylophilus is driven by somatic polyploidy and cell proliferation simultaneously [40]. Consequently, we speculate that the cyp-33C9 gene of B. xylophilus appears to have little effect on the polyploidization and proliferation of somatic cells.
After nematodes invade a plant, the host will generate a wide range of nematicidal and nematistatic substances as part of a natural resistance response to attempted infection by the nematodes [41,42]. As a consequence, the nematodes must resist or metabolize these secondary metabolites in order to invade the host successfully [23,43]. Cytochrome P450s play an important role in the metabolism of xenobiotic compounds in general, and this activity may enable B. xylophilus to overcome the host resistance. Our data showed that, after the knockdown of the cyp-33C9 gene of B. xylophilus, the percentage needle wilting of the P. thunbergii seedlings was markedly lower than that induced by the control nematodes. This result indicated that the cyp-33C9 gene may play an important role in the pathogenic process of B. xylophilus. A possible reason for this may be that silencing of the cyp-33C9 gene impaired the metabolism of xenobiotic compounds, subsequently slowing the migration of the nematodes and delaying the time course of wilting by the infected trees [44,45].
When B. xylophilus J2s had been soaked in cyp-33C9 dsRNA, the nematodes took a longer time to migrate through the host trees ( Table 1). Silencing of the cyp-33C9 gene with RNAi decreased the number of nematodes in P. thunbergii seedlings compared with the control seedlings ( Table 1). The results obtained by this study demonstrated that the knockdown of the cyp-33C9 gene interrupted the invasiveness, fitness, motility or reproduction of B. xylophilus during invasion of the host. These results are consistent, in part, with those reported by Cheng et al. who discovered that silencing of the Bx-eng-1 gene reduced the migration ability of B. xylophilus in Pinus bungeana [32].
In summary, this study investigated the function of the cyp-33C9 gene of B. xylophilus by dsRNA interference. The results showed that the cyp-33C9 gene could be effectively silenced by the dsRNA soaking method. We found that the silencing of the cyp-33C9 gene suppressed feeding, reproduction, oviposition, percentage hatch and pathogenicity of B. xylophilus. However, silencing of the cyp-33C9 gene had no significant effect on the individual body length of B. xylophilus. The migration speed of B. xylophilus treated with cyp-33C9 dsRNA was reduced in the early stages when nematodes were inoculated into pine trees. These data provide fundamental information to increase our understanding of pine wilt disease (PWD) and may help to develop effective strategies to prevent and control PWD. However, further experiments are needed to investigate the molecular mechanisms by which the cyp-33C9 gene influences eggshell formation, and the correct execution of meiosis and polarization of the embryo.

Biological Materials
The BxJJ01 strain of B. xylophilus was isolated using Baermann funnels from infested P. thunbergii samples collected in Jiujiang City, Jiangxi Province, China. The nematodes were cultured on PDA plates covered with B. cinerea for five days at 25 • C. Subsequently, the nematodes were isolated from the plates with Baermann funnels and centrifuged at 1500× g for 5 min. About 10,000 mixed-stage nematodes were decanted into a burette, which contained 25 mL 0.3% carboxymethyl cellulose, sodium salt (CMC) solution. After 12 h, the second-stage larvae were collected from the top of the burette [46].
Two-year-old P. thunbergii seedlings used in this study were obtained from a forest farm in Jiujiang City and were transplanted into pots (18 cm in diameter, 11 cm in height), which were watered on alternate days. The heights of the seedlings were 40-50 cm and they were grown in an air-conditioned greenhouse with a relative humidity of 70% at 30 • C during the daytime and 25 • C at night.

RNA Isolation and cDNA Synthesis of Bursaphelenchus xylophilus
The total RNA was extracted from nematodes using RNeasy Mini Kit (Qiagen, Valencia, CA, USA) and purified with the RNAclean Kit (Tiangen, Beijing, China), following the manufacturer's instructions. The total RNA concentration was detected at 260 nm using an ultraviolet spectrophotometer and its quality examined by electrophoresis on a 1% agarose gel. RNA was reverse transcribed to cDNA using a PrimeScript 1st strand cDNA Synthesis Kit (TaKaRa, Shuzo, Japan), according to the manufacturer's instructions.

Double-stranded RNA (dsRNA) Synthesis and in Vitro RNA Interference (RNAi)
A fragment of cyp-33C9 gene (1200-bp) was amplified by PCR from the cDNA templates of pine wood nematodes with the following primers: forward primer 5 -ACTTTCCTGGTAACACTG-3 and reverse primer 5 -CTTTGATTCTTTGGACGA-3 . The cyp-33C9 fragments were cloned into Pmd TM 18-T vector and then amplified by PCR using T 7 -labeled gene-specific primers as follows: forward primer 5 -TAATACGACTCACTATAGGGACTTTCCTGGTAACACTG-3 and reverse primer 5 -TAATACGACTCACTATAGGGCTTTGATTCTTTGGACGA-3 . The amplified products were used as template for synthesis of the single-strand RNA (ssRNA). The ssRNA was synthesized using a 20 µL reaction volume, which contained 2 µL template, 2 µL 10× T7 reaction buffer, 2µL ATP solution, 2 µL CTP solution, 2 µL GTP solution, 2 µL UTP solution, 2 µL T7 enzyme mix, with nuclease-free water added to a final volume of 20 µL. The ssRNAs were mixed together and incubated at 75 • C for 5 min to synthesize dsRNA. In addition, the dsRNA of green fluorescent protein gene (gfp) was synthesized with the following primers: forward primer 5 -GTACTCGAGTGGGTTATGGTGTTCTATGCT-3 and reverse primer 5 -GAATCTAGAGTGGTCTCTCTTTTCGTTGG-3 [32]. Then, the dsRNA was purified to remove proteins, free nucleotides and nucleic acid degradation products. Approximately 5000 J2 nematodes were immersed in 500 µL cyp-33C9 dsRNA solution (1 µg·µL −1 ) and incubated in a shaking incubator at 180 rpm for 24 h at 25 • C. Equal numbers of nematodes immersed in gfp dsRNA solution or ddH 2 O without dsRNA were treated as controls. There were three replicates for each treatment. Samples from each treatment were washed with sterile water three times before use and the nematodes were collected by centrifuging at 1500× g for 3 min.

Quantitative Reverse Transcription PCR (qRT-PCR)
Quantitative reverse transcription PCR (qRT-PCR) was used to assess the effect of RNAi on cyp-33C9 mRNA levels. Total RNA was extracted from B. xylophilus following cyp-33C9 dsRNA treatment or ddH 2 O or gfp dsRNA controls. The first-strand cDNA was synthesized and used as a template for PCR. qRT-PCR was carried out using TransStart Green qPCR SuperMix (Trans GEN Biotech, Beijing, China) in a thermal cycler (ABI Prism 7500; Applied Biosystems, Foster City, CA, USA). The primers for the cyp-33C9 gene were as follows: forward primer 5 -TCGGTTGTGGCGTGGATG-3 and reverse primer 5 -TGAATTATGTTCAGGCGGTT-3 . The thermocycler conditions comprised one cycle of denaturation at 94 • C for 30 s, followed by 35 cycles at 94 • C for 5 s, 60 • C for 34 s, and 72 • C for 1 min. The actin gene of B. xylophilus was used as the internal control. The primers for the actin gene were as follows: forward primer 5 -GCAACACGGAGTTCGTTGTA-3 and reverse primer 5 -GTATCGTCACCAACTGGGAT-3 . The experiment had three biological replicates (with respect to independent RNA preparations) and three technical replicates.

Effect of RNA Interference (RNAi) on Feeding and Reproduction of Bursaphelenchus xylophilus
Two hundred nematodes (second-stage juveniles, J2s) were soaked in ddH 2 O (CK), gfp dsRNA solution or cyp-33C9 dsRNA solution for 48 h, respectively. The nematodes were washed three times with sterile water. Subsequently, the nematodes were transferred onto a culture of B. cinerea on a PDA plate and cultured at 25 • C for 5 d. There were three replicates for each treatment. Feeding of B. xylophilus on B. cinerea was observed and photographed. Then, the nematodes were isolated from the PDA plates using Baermann funnels and counted with an optical microscope (CX31; Olympus, Tokyo, Japan) [47].

Effect of RNA Interference (RNAi) on Oviposition of Bursaphelenchus xylophilus
Approximately one thousand J2 nematodes were soaked in cyp-33C9 dsRNA solution, ddH 2 O or gfp dsRNA solution for 48 h. After soaking, the nematodes were washed three times with sterile water and transferred to B. cinerea cultures on PDA plates in the dark at 25 • C for 24 h. Ten pairs of each of female and male adult nematodes were selected and mixed together for mating in small petri dishes (30-mm diameter) in the dark at 25 • C for 12 h. Subsequently, the resulting eggs laid by female nematodes were counted under an optical microscope (CX31; Olympus, Tokyo, Japan). Each treatment was replicated three times. The effects of the three different treatments on the egg-laying capacity of B. xylophilus were determined by comparing the number of eggs laid per female nematode.

Effect of RNA Interference (RNAi) on Percentage Hatch of Bursaphelenchus xylophilus
In order to obtain synchronous eggs, adult nematodes were transferred to petri dishes (3 cm) where the depth of sterile water was less than 5 mm. After 0.5 to 1.0 h, some nematodes had laid eggs and most of the eggs were stuck to the bottom of the petri dish. By gently removing the water and nematodes to another petri dish, the synchronous eggs were collected by washing the petri dish several times with sterile water. Subsequently, one hundred eggs were soaked in cyp-33C9 dsRNA solution, ddH 2 O or gfp dsRNA solution at 25 • C for 48 h. Each treatment was replicated three times. If an egg hatched within 36 h, the egg was considered to have hatched successfully; otherwise, an unhatched embryo was considered to be dead [48].

Effects of RNA Interference (RNAi) on Individual Body Length of Bursaphelenchus xylophilus
Two hundred J2 nematodes were soaked in cyp-33C9 dsRNA solution, ddH 2 O or gfp dsRNA solution for 48 h. After soaking, the nematodes were washed three times with sterile water and transferred to B. cinereal cultures on PDA plates in the dark at 25 • C for 24 h. The nematodes were isolated from the plates with Baermann funnels and centrifuged at 1500× g for 5 min. Subsequently, nematodes were killed by warming up the solution to 55 • C. The nematodes were washed three times with sterile water and the individual body length of adult nematodes was measured under an optical microscope (CX31; Olympus, Tokyo, Japan). Forty female and forty male nematodes were measured for each treatment.

Effects of RNA Interference (RNAi) on Migration and Reproduction of Bursaphelenchus xylophilus in Pinus thunbergii
Approximately two thousand J2 nematodes soaked in cyp-33C9 dsRNA solution, gfp dsRNA solution or ddH 2 O (as control) for 48 h were inoculated into P. thunbergii seedlings. For each treatment, three P. thunbergii seedlings were cut for observation every 5 d. After removing the branches and leaves, the trunks of P. thunbergii seedlings were cut into 5-cm long sections in either direction from the site of inoculation. Subsequently, B. xylophilus in the trunks of P. thunbergii were collected by Baerman funnels and counted under an optical microscope (CX31; Olympus, Tokyo, Japan) [34].

Effects of RNA Interference (RNAi) on Pathogenicity of Bursaphelenchus xylophilus
The 2-year-old P. thunbergii seedlings were inoculated with three different preparations at about 15-20 cm above the soil level, and the preparations were as follows: (i) 200 µL suspension of nematodes soaked in ddH 2 O; (ii) 200 µL suspension of nematodes treated with cyp-33C9 dsRNA; (iii) 200 µL suspension of ddH 2 O without nematodes. The pine trees inoculated with B. xylophilus soaked in ddH 2 O or inoculated with ddH 2 O without nematodes were utilized as controls. Wilting symptoms of P. thunbergii were observed after inoculation for 25 d. A P. thunbergii seedling was deemed to be dead if all its needles turned yellow. Each treatment group contained nine P. thunbergii seedlings. The percentage wilting was calculated using the following equation [49]: The percentage wilting = Number of wilted trees Total number of trees × 100% (1)

Statistical Analysis
The results shown are the means and standard deviation (SD) of three replicates. Statistical analysis was determined using SPSS Statistics 13.0 software (IBM, Armonk, NY, USA), using one-way analysis of variance, with multiple pairwise comparisons conducted by Tukey's test.

Conclusions
In summary, this study investigated the function of the cyp-33C9 gene of B. xylophilus by dsRNA interference. The results showed that the cyp-33C9 gene could be effectively silenced by the dsRNA soaking method. We found that the silencing of the cyp-33C9 gene suppressed feeding, reproduction, oviposition, percentage hatch and pathogenicity of B. xylophilus. However, silencing of the cyp-33C9 gene had no significant effect on the individual body length of B. xylophilus. The migration speed of B. xylophilus treated with cyp-33C9 dsRNA was reduced in the early stage when nematodes were inoculated in pine trees. These data provide fundamental information to increase our understanding of pine wilt disease (PWD) and may help to develop effective strategies to prevent and control PWD. However, further experiments are needed to investigate the molecular mechanisms by which the cyp-33C9 gene influences eggshell formation, and the correct execution of meiosis and polarization of the embryo.