An 893-bp thermotolerance-related cDNA, homologous to HSP90 and named B40, was found using subtracted cDNA libraries and the BLAST method. The sequence of B40 and those of the products of a spliced leader (SL) primer PCR and a 3′-RACE provided the information necessary to clone a 1206-bp cDNA sequence of the HSP90 gene. The sequence was named Bx-HSP90 and submitted to National Center for Biotechnology Information (NCBI) (accession No. EF490991). Bx-HSP90 contains an open reading frame of 1089 bp, with a 10-bp 5′ untranslated region (UTR) and a 109-bp 3′ UTR featuring a likely polyA-addition signal. Overall, this HSP90 gene shows the greatest similarity to that of Heterodera glycines (AAO14563) (E Value: 1 × 10⁻¹⁴⁷). The encoded protein of 362 amino acids is not predicted to contain a signal peptide, and a PSORTII analysis predicts a cytoplasmic location.

An alignment of Bx-HSP90 with its closest homologs is shown in Figure 1. The molecular architecture and sequence of the Bx-HSP90 protein are similar to those of its closest homologs; the similarity occurs particularly in the C-terminal portion of the protein and is concentrated on the active site residues. A pairwise comparison of the predicted amino acid sequence of Bx-HSP90 with the sequence of each of the other HSP90s revealed a high degree of conservation along the entire length of the HSP90 sequences. The respective percentages of the compared sequences that were identical and similar were 79 and 89 for H. glycines, 80 and 89 for Brugia pahangi, 74 and 86 for Mus musculus, 72 and 85 for Homo sapiens, 73 and 87 for Oxyuranus scutellatus and 79 and 88 for Caenorhabditis elegans. All of the seven conserved amino acid blocks defining the HSP90 protein family signature GVVDS (E/D) DLPLN (I/V) SRE (Figure 1, underlined black-box) and the consensus sequence MEEVD (Figure 1, underlined white-box) at the C-terminus were found in the Bx-HSP90 sequence.

A phylogenetic tree was constructed based on the HSP90 sequences of various organisms (Figure 2) using the MEGA 5.05 program, and the resulting tree showed five major clusters. The Bx-HSP90 sequence was closely clustered with the homologous sequences of other Bursaphelenchus species (B. xylophilus, B. mucronatus and B. doui).

We searched the B. xylophilus genome (CADV01000001-CADV01010432) [12] with the Bx-HSP90 protein sequence using the TBLASTN algorithm. Using an expectation value of 1e-006, this TBLASTN homology search identified seven alignments located in contigs CADV01010217.1 (Score: 280, E Value: 3 × 10⁻⁷⁵), CADV01010388.1 (Score: 75, E Value: 1 × 10⁻¹³), CADV01008131.1 (Score: 44, E Value: 1 × 10⁻⁸), CADV01007052.1 (Score: 39, E Value: 6 × 10⁻⁸), CADV01003002.1 (Score: 39, E Value: 8 × 10⁻⁸), CADV01005604.1 (Score: 38, E Value: 1 × 10⁻⁷) and CADV01005840.1 (Score: 54, E Value: 4 × 10⁻⁷). Except for CADV01010217.1, which is the Bx-HSP90 contig, the other alignments are unlikely to be related to homologous sequences putatively coding for HSP90s. HSP90 is a single-copy gene in the C. elegans and B. pahangi genomes [13,14], and the results presented here indicate that Bx-HSP90 might also be a single-copy gene in the B. xylophilus genome.

In situ hybridization was used to determine the tissue specificity of Bx-HSP90 transcription. In heat-shocked nematodes, the Bx-HSP90 anti-sense probe showed localized hybridization in an area of approximately 350 μm in length and 20 μm in width located over approximately 35% of the body length in the middle of the nematode (Figure 3A–C). The hybridization signals were detected at 25, 30 and 35 °C (Figure 3A–C, respectively). No equivalent hybridization was observed using the control sense probe (Figure 3D). The data showed that the Bx-HSP90 gene was constitutively expressed in response to all of the tested temperatures.

The nematode survival rate was determined after 24 h of incubation at 25, 30 and 35 °C using the loss of spontaneous and touch-provoked movement to indicate death. We found that 100, 99 and 46% of the nematodes were alive at 25, 30 and 35 °C, respectively. Both 30 and 35 °C constitute high-temperature stress for B. xylophilus, although only a slight decrease in survival occurred at 30 °C within the 24-h period.

To validate a relationship between nematode survival and the Bx-HSP90 transcript levels, RT-PCR analyses were performed using the total RNA from B. xylophilus incubated at 25, 30 and 35 °C for 24 h. Three biological replicates were conducted in this experiment, and the signals of the PCR product were measured using BandScan 5.0. The Bx-HSP90 transcript signal was detected in all of the temperature treatments, with the highest level detected at 30 °C (Figure 4). Compared with the level at 25 °C, the Bx-HSP90 transcript level was significantly increased at 30 °C but only slightly increased at 35 °C. After 24 h, Bx-HSP90 transcription in the heat-shocked nematodes displayed a temperature-dependent pattern (Figure 4C Bx-HSP90): The level of Bx-HSP90 mRNA increased as the temperature increased from 25 to 30 °C and decreased as the temperature increased from 30 to 35 °C. The transcription of the Actin gene was used as a control, and the amount of Actin transcript was similar in all of the treatments (Figure 4C Actin). This stable transcription of Actin suggested that the Bx-HSP90 RNA was qualified. The transcripts of the Bx-HSP90 gene appear to be significantly up-regulated in response to heat shock. Together with the survival data, our observations suggest that the production of the highest level of Bx-HSP90 at 30 °C may be adaptive in maintaining a native protein structure under high-temperature stress. At 35 °C, we found that the Bx-HSP90 transcript level also increased compared with the level produced at 25 °C. Contrary to our expectation, however, nematode survival did not increase with this increase in transcript but sharply decreased. These results suggest that Bx-HSP90 was only weakly active at 35 °C such that it could not fulfill a thermoregulatory function. Thus, Bx-HSP90 does play a role in thermoregulation but functions maximally within a restricted temperature range.

There have been reports suggesting that HSP90 is essential in the filarial nematode, B. pahangi, and in the free-living nematode, C. elegans [14]. Because the knockout of HSP90 is lethal in eukaryotes, we used regular temperatures and high-temperature stress environments to examine whether HSP90 had any effect on the mortality of B. xylophilus. The RNA interference of Bx-HSP90 was performed for mixed-stage B. xylophilus using the method developed by Urwin et al. (2002) to allow the octopamine-stimulated uptake of double-stranded RNA (dsRNA). The patterns of fluorescein isothiocyanate (FITC) uptake observed here for the nematodes incubated in octopamine and FITC were similar to those described by Urwin et al. (2002) [15]. RT-PCR experiments were performed to determine the extent of Bx-HSP90 RNAi in B. xylophilus. Although a significant Bx-HSP90 silencing was found after soaking in the relevant dsRNA (Figure 5A Bx-HSP90 at 24 h, 48 h and 3 day), the Bx-HSP90 dsRNA had no effect on the transcription of Actin (Figure 5A Actin). From these data, we conclude that the RNAi by soaking was potent and specific for B. xylophilus.

The survival of the nematodes subjected to Bx-HSP90 RNAi treatment decreased to 50% after incubation for 8 days at 25 °C (Figure 5D). A similar reduction was observed after 48 h at 30 °C and after 24 h at 35 °C (Figure 5B,C). The RNAi treatment produced a 25% reduction in survival at 30 °C after 44 h: more than 90% of the RNAi-treated nematodes were dead upon the exposure to 30 °C for 5 days compared with 7 days for the controls. A t-test for paired samples showed that the survival of the Bx-HSP90 RNAi-treated nematodes differed significantly from that of the control nematodes, and a bivariate correlation analysis indicated a significant correlation between survival and the temperature (p < 0.01, Bx-HSP90 RNAi vs controls). The Actin RNAi negatively affected the survival of the nematodes, but there was no correlation between the survival and the temperature treatment. These data suggest that the HSP90 gene might function in thermoregulation for specific durations within a defined range of high temperatures. Judging from the Bx-HSP90 RNAi results, combined with the RT-PCR results, Bx-HSP90 apparently functions strongly for at least 5 days at 30 °C but poorly at 35 °C.

To investigate the relationship between Bx-HSP90 and the nematode population growth, we measured the hyphal mat consumed by control, Bx-HSP90 RNAi-treated and Actin RNAi-treated B. xylophilus, which were cultured on fungi (B. cinerea) at 25, 30 and 35 °C. The measurement of the area of the hyphal mat that was consumed permitted a quantitative estimation of the population growth of the nematode. At the 6th day at 25 °C, the Bx-HSP90 RNAi-treated B. xylophilus consumed only 25% of the hyphae, whereas the Actin RNAi-treated and the control nematodes consumed 58 and 100%, respectively. The Bx-HSP90 RNAi-treated B. xylophilus required 11 days to consume all of the hyphae at 25 °C, and the Actin RNAi-treated B. xylophilus required 8 days. These results demonstrate that both the Bx-HSP90 and Actin RNAi-treated B. xylophilus grew normally at 25 °C but that their population growth rates were lower than that of the control. At the 14th day, no hyphae were consumed in the culture dishes containing the Bx-HSP90 RNAi-treated nematodes at 30 °C. The population growth rates of both the Bx-HSP90 RNAi-treated and Actin RNAi-treated B. xylophilus were slower at 30 than at 25 °C, with the rate being slower for the Bx-HSP90 RNAi-treated group. After 14 days, the populations of the Bx-HSP90 RNAi-treated B. xylophilus had not grow at 30 °C, whereas the control populations grew at that temperature beginning at the 2nd day (Figure 6B). Regardless of the treatment, no consumption of hyphae occurred at 35 °C during the 18 days observation period.

The temperature regulation of gene expression is critical for the survival and success of organisms living in thermally variable environments. Heat shock proteins function during cell stress as molecular chaperones, interacting with diverse protein substrates to assist in repairing damaged proteins by refolding or degrading them, thereby restoring protein homeostasis and promoting cell survival [16]. Our results have implications for the roles that Bx-HSP90 may play in the environmental adaptation processes of B. xylophilus; for example, the protein may help to buffer the stress of high temperatures (within a certain high-temperature range, which includes 30 °C). However, further studies are necessary to clarify this point and to elucidate the possible involvement of other stress proteins in mitigating a vulnerability to the multitude of environmental stresses affecting development.

Using RNAi, we showed that Bx-HSP90 is essential for the survival of the nematode B. xylophilus. HSP90 can be considered important for nematodes because silencing it reduces their survival at all temperatures. These results demonstrate that the inhibition of HSP90 using such agents as geldanamycin may represent a novel approach for the treatment of pine wilt disease. HSP90 or some of its client proteins of mosquito and mammalian parasites, such as the filarial nematode, B. pahangi, have been proposed as drug targets [15]. Because HSP90 is associated with diverse tissues and functions within nematodes, there is the prospect of affecting a range of processes using the control agents that target this protein. The inhibition of HSPs leads to reduced growth and fecundity of plant parasitic nematodes [17]. If the survival of nematodes can be reduced at warmer temperatures, even by a small percentage, this approach may be effectively used against B. xylophilus in warmer climates.

B. xylophilus specimens were isolated from wilted Pinus massoniana in the Guangdong Province of China using the Baermann funnel technique. The nematodes were cultured on the Botrytis cinerea for 5 days at 25 °C and then divided into two groups. One group was heat shocked at 30 °C for 24 h to induce the expression of thermotolerance-related genes, and a second group was cultured at 25 °C as a control.

For the RNA extractions, the shocked nematodes and controls were separately frozen in a mortar with liquid nitrogen and powdered using a pestle. For each group, the total RNA was extracted from 100 μg of mixed nematode stages (a mixture of adult and juvenile nematodes in a male to female to juvenile ratio of approx. 1:1:2) using the TRIzol reagent (Invitrogen, USA, cat. No. 15596-026). The RNA was then purified using DNase I (RNase-free) (TaKaRa, Japan, cat. No. D2270A) treatment. Double-stranded cDNAs were generated using a SMART (switching mechanism at the 5′ end of the RNA transcript) kit according to instructions of the manufacturer (BD Clontech, USA, cat. No. 634902).

The PCR-select cDNA subtraction was performed using the BD PCR-Select cDNA Subtraction Kit (BD Clontech, USA, cat. No. PT1117-1) according to established protocols [18]. The cDNA from the heat-shocked nematodes was used as the tester, and the cDNA from the untreated controls was used as the driver. The cDNAs were digested with Rsa I and then ligated to different adapters. Two rounds of hybridization and PCR amplification were processed to normalize and enrich the differentially expressed cDNAs. The products of the secondary PCR from the forward subtraction were directly inserted into the pGEM-T vector and transformed into Escherichia coli JM109 competent cells (Promega, USA, cat. No. A3610). All of the recombinant clones were included to establish the subtracted cDNA library. Random sequencing of the library using the T7 primer yielded 89 successful sequencing reactions.

The BLAST analysis of all 89 of the sequences revealed that one sequence of 893 bp (sequence No. B40) was homologous to the HSP90 gene of other species. This sequence was selected for further cloning of the full-length cDNA of the HSP90 gene from B. xylophilus. None of the other sequences showed a direct relationship with HSP90 (data not show).

Most nematode messenger RNAs (mRNAs) have a short sequence at their 5′ end that is not contiguously encoded in the genome. This leader sequence is trans-spliced to the 5′ end of primary transcripts during the maturation of the pre-mRNA. This spliced leader (SL) provides a 5′ anchor, which both simplifies the process of second-strand cDNA synthesis and allows the isolation of cDNA molecules that are guaranteed to be full length [19]. The SL primer PCR used a spliced leader sequence 5′ (forward) primer, SL1 [20], and the specific primer SLHSP90-A2L was used to amplify the entire transcriptome. To obtain a complete sequence of this gene, 3′ RACE (rapid amplification of cDNA ends) techniques were performed using the specific primer SLHSP90-S1L with the SMART RACE cDNA Amplification Kit (Clontech, USA, cat. No. K1811-1) according to the manufacturer’s instructions.

The sequences obtained from the spliced leader primer PCR and the 3′-RACE were joined with that of B40 to generate a full cDNA sequence, which was named Bx-HSP90. The full-length cDNA obtained was PCR amplified using the forward primer EHSP90S2L and the reverse primer EHSP90A1L, which contained start and stop codons at the 5′ and 3′ ends, respectively.

Searches for nucleotide and amino acid sequence similarities were conducted using the BLAST programs at the National Center for Biotechnology Information (NCBI) [21]. Multiple alignments of HSP90 were performed using the ClustalW multiple alignment program [22]. The similarity percentages of the full-length amino acid sequences between HSP90 and other HSP90s were calculated using the Identity and Similarity Analysis program [23]). The signal peptide was predicted with SignalP 3.0 [24]. To find the conserved HSP90 domains, the amino acid sequences of the HSP90 proteins that have been analyzed (H. glycines, B. pahangi, M. musculus, H. sapiens, O. scutellatus and C. elegans) were aligned with Bx-HSP90 using the BioEdi program. Those aligned sequences and other homologous sequences from plant nematode species (Ditylenchus destructor, ADZ13510; Meloidogyne incognita, ADD10372; M. artiellia, CAU15484), including several Bursaphelenchus species (B. xylophilus, ACZ13352 and ACY01918; B. mucronatus, ACU00686 and ADK26462; B. doui, ADC79631) were all retrieved from NCBI and then used to generate a phylogenetic tree using the maximum-likelihood method with MEGA 5.05.

Homologous sequences were found using Bx-HSP90 as a query sequence in a TBLASTN search of the B. xylophilus complete genome sequence databases (the B. xylophilus whole-genome shotgun project consists of sequences CADV01000001-CADV01010432) [12].

The nematodes were heat shocked at 25, 30 or 35 °C for 24 h to induce Bx-HSP90 expression. In situ hybridization was performed on the heat-shocked B. xylophilus essentially as described in De Boer et al. [25]. A digoxigenin-labeled RNA probe was synthesized using the cDNA sequence following the general protocol of the Roche DIG RNA Labeling Kit (SP6/T7) (Roche, Germany, cat. No. 1175025). The specificity of the probe was verified by Northern blotting. The hybridization signals were detected by incubation with the DIG High Prime DNA Labeling and Detection Starter Kit I (Roche, Germany, cat. No. 11745832910) and photographed under a microscope (Olympus Bx51, Japan). The backgrounds of the images were cropped using Adobe Photoshop CS2.

The expression of the Bx-HSP90 transcript in the heat-shocked B. xylophilus was measured using RT-PCR. The nematodes were cultured at 25, 30 and 35 °C for 24 h; three biological replicates were used in this experiment. The fresh living nematodes for the RNA extraction were all collected at the same temperature using the Baermann funnel technique. Two Bx-HSP90 gene-specific primers, HSP90rtR1L and HSP90rtF1L (Table 1), were used to amplify a 405-bp product. A constitutively expressed gene, Actin, was used as an internal control to verify the quantitative RT-PCR reaction using primers (RsACT_F and RsACT_R) (Table 1) [26]. PCR-grade water replaced the cDNA template in the negative control. The RT-PCR was performed for 28 cycles, and the signals of the RT-PCR products were measured using BandScan 5.0. The 150-ng (750 bp) band of the DL 2,000 DNA marker (TaKaRa, Japan, cat. no. D500A) was used as a quantitative standard band. The survival of the heat-shocked nematodes was recorded to measure the relationship between Bx-HSP90 and mortality.

The dependence of the observed thermotolerance on the Bx-HSP90 gene was estimated by the RNA interference of the function of the gene. The RNAi was performed using mixed stages of the nematode (a mixture of adult and juvenile nematodes in a male-to-female-to-juvenile ratio of approx. 1:1:2), as outlined in Urwin, et al. [13]. Double-stranded RNA corresponding to Bx-HSP90 was prepared using the MAXIscript T7/T3 RNA Synthesis Kit (Ambion, Japan, cat. no. AM1324M). The treated nematodes were soaked in M9 buffer with 10 mM octopamine and dsRNA (5 mg/mL) corresponding to the Bx-HSP90 sequence. The uptake of the soaking solution was monitored by a final nematode treatment of M9 buffer containing 10 mM octopamine and 1 mg/mL FITC. The control nematodes were soaked in M9 buffer only, M9 buffer with 10 mM octopamine or Actin dsRNA (5 mg/mL). After 12 h of soaking with intermittent agitation at 25 °C, the nematodes were washed copiously with sterile water to remove the external dsRNA and then observed fluoroscopically to detect the uptake of FITC.

The RNAi-treated nematodes were divided into three groups: The first group was used to determine the extent of Bx-HSP90 RNAi; the second was used to assess survival under 25, 30 and 35 °C heat shock; and the third was used to assess the population growth under 25, 30 and 35 °C heat shock.

RNA was extracted from the control nematodes and the RNAi-treated nematodes after 24 h, 48 h and 3 days. RT-PCR was then performed using the following primers: HSP90rtR1L, HSP90rtF1L, RsACT_F and RsACT_R (Table 1).

The survival of the RNAi-treated nematodes and controls was assessed during their culture at 25, 30 and 35 °C (n = 500 for each temperature) and was recorded every 4 h for 2 days and then daily for 12 days. Each treatment was repeated four times. A t-test for paired samples and a bivariate correlation analysis (SPSS 13.0) were used to determine the extent to which RNAi influenced the number of nematodes surviving the exposure to the experimental temperatures.

Samples of 100 RNAi-treated nematodes (a mixture of adults and juveniles in a male-to-female-to-juvenile ratio of approx. 1:1:2) or controls were cultured on B. cinerea at 25, 30 and 35 °C to measure the relationship between the function of Bx-HSP90 and mortality. Each treatment was cultured in 10 dishes, and the process repeated twice. Images of the culture dishes were obtained using a digital camera (Nikon, Japan) every day for 18 days. The area of the hyphal mat that was consumed in each dish was measured using ImageJ 1.45 m [27,28]. Because the area of the B. cinerea hyphal mat that is consumed is proportional to the number of B. xylophilus feeding on the mat, this technique quantitatively estimated the population growth of the nematodes in each dish.