The highly virulent B. xylophilus strain AMA3 was isolated from diseased Pinus thunbergii seedlings in Nanjing, China. The nematodes were cultured on Botrytis cinerea in potato dextrose agar (PDA) plates at 25°C in the dark for 7–9 days at Qingdao University (Qingdao, Shandong Province, China) (Guo et al., 2017). Then, the mixed-stage nematodes were separated using the Baermann funnel technique (Viglierchio and Schmitt, 1983). The nematodes were placed in Petri dishes for 1 h at 25°C in the dark to obtain eggs, as described previously (Futai, 1980; Tang et al., 2020). The collected eggs were hatched in the absence of food for 24 h in the dark at 25°C to obtain J2 larvae. Subsequently, the J2 larvae were cultured for 48 h at 25°C in the dark to obtain J4 worms. After the male and female J4 nematodes were distinguished, they were transferred to the PDA plates containing B. cinerea separately for 24 h at 25°C in the dark. Then, virgin adult worms were collected using two Baermann funnels (Zhu et al., 2016). Two-year-old Pinus thunbergii (P. thunbergii) seedlings were grown in the greenhouse at 25°C.

DNA was extracted from nematodes with the method described previously (Huang et al., 2010). Total RNA was extracted using TRIzol^® Reagent (Invitrogen, Waltham, MA, United States) and treated with DNase I (Cwbio, Beijing, China), according to the manufacturer’s protocol. DNA and RNA were examined through electrophoresis on a 1.5% denaturing agarose gel and quantified by measuring ultraviolet absorbance at 260/280-nm wavelength (NanoDrop^® ND-2000, Thermo Fisher, USA). The synthesis of first-strand cDNA was performed, as described previously (Tang et al., 2020). According to the UGT440A1 sequence obtained through the transcriptome analysis and the B. xylophilus genome data from WormBase ParaSite², a pair of specific primers (forward: 5′-CCGGAATTCATTTAACCAACATAATATGTTG-3′; reverse: primer 5′-GCTCTAGAAAAGCTATTTATTAATTCTATTCAGA CA-3′) was used to amplify the complete UGT440A1 through polymerase chain reaction (PCR) from B. xylophilus genomic DNA. The PCR conditions were as follows: pre-denaturation at 94°C for 5 min, followed by 30 cycles of denaturation at 94°C for 45 s; annealing at 56°C for 45 s; extension at 72°C for 150 s, and a final extension at 72°C for 10 min. To further analyze the UGT440A1 protein, the coding sequence of the gene was amplified by PCR using cDNA reverse-transcribed from mRNA of B. xylophilus. The primers (forward: 5′-CCCAAGCTTATGCGCGCCTTTCTGC −3′; reverse: 5′-CC GGAATTCTTACTCCGCTTTGACCTTGG-3′) were designed based on the ORF region of UGT440A1. The PCR conditions were as follows: pre-denaturation at 94°C for 5 min, followed by 30 cycles of denaturation at 94°C for 45 s; annealing at 60°C for 45 s; extension at 72°C for 90 s, and a final extension at 72°C for 10 min. The amplified products were cloned into a pMD^®18-T Simple Vector (Takara, Dalian, China) and then sequenced.

The program open reading frame (ORF) Finder³ was utilized to analyze the ORF of UGT440A1. The B. xylophilus genome from WormBase ParaSite was used for the structural analysis of UGT440A1. The protein sequences of UGT440A1 were deduced from the nucleotide sequences by using the Basic Local Alignment Search Tool⁴. Protein sequences of other species utilized in alignments were obtained from the NCBI⁵. Conserved domains were identified using the conserved domain database of NCBI⁶. DNAMAN9 software was used to perform multiple alignments, and MEGA-X software was utilized to construct a phylogenetic tree with the neighbor joining method. Physicochemical characteristics of UGT440A1 were obtained from ProtParam⁷. Signal peptides and transmembrane helices were predicted using SignalP Server⁸ and THHMM Server⁹, respectively. The deduced amino acid sequence of UGT440A1 was modeled using the SWISS-MODEL program¹⁰ and further analyzed using the PyMOL 2.3.2 software.

mRNA FISH was performed to determine the spatial expression patterns of UGT440A1 at different developmental stages of B. xylophilus. A red fluorescence-labeled probe (5′-Cy3-CGCTGATAGTAAGTCATCTCGCTTCCATGATCGTCGTCC ATGCCT-3′) was generated from the cloned sequence of UGT440A1 and designed using Primer Premier 5.0 software (Primer, Canada). B. xylophilus at different developmental stages were collected and centrifuged in 1.5-mL centrifuge tubes. The nematodes were fixed with the RNAse-free paraformaldehyde solution (4%) at 5°C for 16 h and then pretreated at room temperature for 4 h. Smeared 50 μL suspension of nematodes (approximately 200 nematodes) in a microscope slide and baked the nematodes at 56°C on a slide warmer. FISH was performed using FISH in situ hybridization kit C007 (Gefan, Shanghai, China), according to the manufacturer’s protocol. The nematodes were incubated in 200 mM hydrochloric acid for 15 min in a humidified box containing 5 × saline sodium citrate (SSC) and formamide, and then washed twice with diethypyrocarbonate (DEPC)-treated water. Proteinase K (0.5 mg mL^–1) was used to digest the nematodes at 37°C for 20 min, followed by washing with 100 mM glycine for 1 min and twice washing with phosphate buffer saline (PBS) for 2 min. The nematodes were incubated in RNAse-free paraformaldehyde solution (4%) for 10 min, and then washed twice with PBS for 3 min. Incubation of the nematodes with acetic anhydride (0.25%) for 10 min were performed to reduce the non-specific binding, following five times washing with PBS for 5 min and twice washing with 5 × SSC for 2 min. The nematodes then were pre-hybridized in hybridization buffer at 65°C for 1 h. In situ hybridization was performed with rotation and probe concentration of 1.0 μg mL^–1 for 48 h at 65°C. After hybridization, the nematodes were washed three times with a mixture of formamide and 4 × SSC (1:1) at 65°C for 15 min and five times with PBS for 5 min at room temperature. Counterstaining was carried out with 4′,6-diamidino-2-phenylindole (2.0 μg mL^–1) for 5 min. The nematodes were examined under a Nikon light microscope (Eclipse Ci, Nikon, Japan). The sense probe (5′-Cy3-AGGCATGGACGACGATCATGGAAGCGAGATGACTTACTA TCAGCG-3′) was used as a negative control.

The UGT440A1 coding region without the putative signal sequence was amplified through PCR from the original plasmid by using specific primers (forward: 5′- ATGGA GAAGATCTTACTCTTAAATGCGGCGAGAATC-3′; reverse: 5′- ATTGCCGACGCCGTGATTACTT-3′) and inserted into the pET-15b (Invitrogen, United States) vector. Plasmid was transformed into Escherichia coli (E. coli) BL21 (DE3) Plyss (Solarbio, Beijing, China). Expression of the recombinant UGT and collection of inclusion bodies were performed using the method described in a study, with some modifications (Liu et al., 2015). The recombinant UGT440A1 was overexpressed in E. coli BL21 (DE3) harboring pET-15b-UGT through Isopropyl-β-D-thiogalactopyranoside (IPTG) induction at 37°C. The inclusion bodies were dissolved in 10 mL solution buffer (20 mM Tris, 5 mM DTT, 8 M urea, pH 8.0) at 4°C overnight. The supernatant was collected through centrifugation at 10000 g for 15 min at room temperature and diluted with refolding buffer (20 mM Tris-HCl, 150 mM NaCl, pH 8.0) to gradually reduce the concentration of urea from 8 M to 1 M. The refolded protein was purified through Ni-NTA affinity chromatography and analyzed through SDS-PAGE (12% gel) (Liu et al., 2015).

UGT activity was assayed according to a method reported previously, with some modifications (Real et al., 1991). The reaction mixture, with the final volume of 0.2 mL, comprised 100 mM Tris-HCl buffer (pH 7.4), 10 mM MgCl₂, 5 mM D-gluconic acid lactone, 500 μM uridine diphosphate-α-D-glucose (UDPG), 100 mM substrate (kaempferol or quercetin), and 100 μL recombinant UGT. The reaction mixtures were incubated for 1 h at 37°C, and the reaction was terminated by adding 1 vol. methanol. Proteins and insoluble materials were removed through centrifugation at 15,000 g for 5 min at 4°C. Ten microliter of the filtered supernatant fraction was used for high-performance liquid chromatography (HPLC) analysis of the reaction products. Reactions performed in the absence of UDPG or recombinant UGT440A1 served as the control.

By using pET-15b-UGT440A1 as the template, a 368-bp DNA fragment was amplified through PCR with T₇-labeled gene-specific primers (forward: 5′-TAATACGACTCACTATAGGGAAGTAATCACGGCGTCGGC-3′; reverse: 5′-TAATACGACTCACTATAGGGAAGCACATGGT CTCTTCC-3′). A pair of T₇-labeled gene-specific primers was used to amplify the 323-bp DNA fragment of green fluorescent protein gene (Bxy-gfp), with pET-15b-gfp as the template (forward 5′-TAATACGACTCACTATAGGGAACGG CCACAAGTTCAGC-3′; reverse 3′- TAATACGACTCACTATA GGGAAGTCGATGCCCTTCAGC-3′). The amplified products were used as the templates for double-stranded RNA (dsRNA) synthesis. The dsRNA was synthesized using the MEGAscript (TM) RNAi Kit (Invitrogen, Vilnius, Lithuania), according to the manufacturer’s instructions. Fluorescent dsRNA labeled with cyanine dyes 3 (Cy3) was constructed by adding Cy3-dCTP in the reaction system. The integrity of dsRNAs was visualized in a 1.5% agarose gel.

Approximately 4000 B. xylophilus (a mixture of juveniles and adults) were soaked in 50 μL soaking buffer (0.05% gelatin, 3 mM spermidine, 0.25 × Mg²⁺-free M9) containing UGT dsRNA (1.0 μg/μL) and then incubated for 48 h at 25°C (Huang et al., 2019; Park et al., 2008). Afterward, B. xylophilus were washed with sterile water several times. Nematodes soaked in soaking buffer containing ddH₂O and gfp dsRNA were used as double negative controls. The nematodes were rinsed four times with sterilized water and photographed under a fluorescence microscope to assess the efficiency of the dsRNA uptake. QRT-PCR was used to verify the effect of UGT440A1 silencing through RNAi on mRNA levels. PCR was performed using TB Green Premix Ex Taq II (TaKaRa, Dalian, China), with 0.4 μM each of forward and reverse primer (forward: 5′-CTCGCGGAAGCGGGTTACAA-3′; reverse: 5′-CTGGTCGCGACACCCAAGTT-3′). The actin gene amplified by the primer pair (forward: 5′-CTGCTGAGCGTGAAATCGT-3′ and reverse: 5′-GTTGTAGGTGGTCTCGTGGA-3′) was used as the internal control. The PCR program was set as follows: 30 s at 95°C, followed by 40 cycles of 5 s at 95°C and 34 s at 60°C. Each experiment was performed thrice.

Approximately 40 nematodes (a mix of juvenile and adult nematodes) were observed under a stereo microscope after the nematodes were soaked in buffer with dsRNA (as described above) for 48 h at 25°C in the dark. The head thrashing frequency (> 120° within a single head swing) in 30 s was used to measure the vitality of B. xylophilus (Tang et al., 2020). Nematodes soaked in buffer containing ddH₂O and gfp dsRNA were used as double negative controls. Each assay was repeated three times.

Twenty pairs of virgin male and female nematodes were soaked separately in buffer containing dsRNA (1.0 μg/μL) for 48 h (Huang et al., 2019). Thereafter, the nematodes were washed several times to remove external dsRNA. Then, 10 pairs of nematodes were cultured in a PDA plate with B. cinerea at 25°C in the dark. The feeding area of B. xylophilus was photographed daily. After culturing for 9 days, the nematodes were isolated from the PDA plates and counted under an optical microscope. The other 10 pairs were transferred onto a glass dish with ddH₂O for 48 h in the dark at 25°C to allow the nematodes to lay eggs, and the number of eggs was counted under an optical microscope. To determine the effect of UGT dsRNA on egg hatching of B. xylophilus, thirty pairs of virgin male and female nematodes were soaked separately in buffer containing dsRNA for 48 h. Following treatment, the nematodes were transferred onto a glass dish for 16 h to lay eggs, then eggs were incubated at 25°C for 20 h for hatching, and the hatching rate was calculated (Huang et al., 2019). The same quantity of nematodes soaked in buffer containing ddH₂O and gfp dsRNA served as negative controls. Each experiment was performed three times.

Approximately 3000 B. xylophilus (a mix of juvenile and adult nematodes) were soaked in 50 μL soaking buffer containing UGT dsRNA (1.0 μg/μL) and then incubated for 48 h at 25°C (Huang et al., 2019). Afterward, B. xylophilus were washed with sterile water several times. Thereafter, the nematodes were inoculated into 2-year-old P. thunbergii seedlings. Sterile water without nematodes and nematodes soaked in buffer containing ddH₂O and gfp dsRNA were used as negative controls. Each treatment was replicated three times. Each P. thunbergii seedling was inoculated with 200 μL suspension of mixed-stage nematodes (approximately 1,000 nematodes). Approximately 2-cm-long wounds on P. thunbergii at 30–50 cm above the soil level were inoculated and sealed with parafilm (Qiu et al., 2016). The inoculated P. thunbergii seedlings were grown in the greenhouse at 25°C. Wilting symptoms developing on the pine woods that were inoculated with the nematodes were observed daily. The PWD symptoms were evaluated and categorized into four groups: 0 = all of the needles were green; 1 = 0%–25% of the needles were discolored and turning yellow; 2 = 26%–50% of the needles had turned yellow; 3 = 51%–75% of the needles had turned yellow; and 4 = 76%–100% of the needles had turned yellow (Yu et al., 2012). The infection rates and the disease severity index (DSI) of the P. thunbergii seedlings were calculated with a method described previously (Yu et al., 2012).

All experiments involving B. xylophilus included three replicates, with each treatment replicated three times. All values of the repeated experiments are expressed as means ± standard deviation (S.D.), and all statistical analyses were performed using SPSS 19.0 software (SPSS, Chicago, IL, USA). The independent sample t-test and one-way analysis of variance were used to assess differences between the groups. A P value of < 0.05 was considered to denote statistical significance.

The DNA sequence of UGT440A1 was successfully amplified by PCR (Supplementary Figure 1). The sequence analysis showed that the PCR product comprises 2,231 bp containing a 1,566-bp ORF which encoded 521 amino acids. The structural analysis showed that UGT440A1 (BXY_1088500.1) comprised seven exons and six introns and contained 50 bp of the 5′ untranslated region and 50 bp of the 3′ untranslated region (Figure 1A). The coding sequence of UGT440A1 was submitted to NCBI Genbank under the accession number MZ467299. The conserved domain search showed that UGT440A1 belonged to the glycosyltransferases B (GTB) superfamily (the members of this family share a common GTB topology) of glycosyltransferase and the coding sequence of UGT440A1 consisted of the GT1-Gtf-like Pfam domain, which was localized between amino acids 22 and 443. The alignment of multiple amino acid sequences revealed the homology of UGT440A1 with a range of UGTs (Figure 1B). Furthermore, sequence analysis revealed the presence of a typical UGT signature sequence in UGT440A1: [FVA]-[LIVMF]-[TS]-[HQ]-[SGAC]-G-x(2)-[STG]-x(2)-[DE]-x(6)-P-[LIVMFA]-[LIVMFA]-x(2)-P-[LMVFIQ]-x(2)-[DE]-Q (the amino acids in the square brackets can be arbitrary, and x stands for any base) (Mackenzie et al., 1997; Figure 1B). The further bioinformatics analysis showed the presence of a signal peptide and a transmembrane region in the deduced amino acid sequence of UGT440A1 (Figure 1B), which indicated that UGT440A1 is a transmembrane protein. The phylogenetic relationship of UGT440A1 with other UGT proteins revealed that UGT440A1 formed a well-founded cluster with UGT of other nematodes (Figure 2). And UGT proteins in nematode species apparently separated those of plant-parasitic nematodes, including B. xylophilus, Aphelenchus avenae, Meloidogyne enterolobii and Meloidogyne graminicola from those of other nematode species. Notably, the UGT protein of Aphelenchus avenae was the closest protein to UGT440A1 among the 18 family members examined. In addition, four outgroups, namely Chordata, Arthropoda, Bacteria, and Fungi were observed (Figure 2). The analysis showed that the UGT440A1 sequences are more closely related to fungal and bacterial sequences compared with animal sequences.

The tertiary structure of UGT440A1 was predicted with the crystal structure of macrolide glycosyltransferase (PDB ID: 2iya.1) as a template by using the online program SWISS-MODEL (Figure 3A). The theoretical tertiary structure consisted of 15 alpha-helices and 13 beta-strands. The N-terminal domain had 7 beta-strands and surrounded by 8 alpha-helices, and the C-terminal domain was made up of 7 alpha-helices and 6 beta-strands. The molecular surface of UGT440A1 was assessed using the PyMOL program, as the protein was in a cell environment. Molecular surface modeling revealed mainly the presence of white patches, along with slight red and blue patches, which indicated the presence of both neutral and charged regions (Figure 3B).

Fluorescence in situ hybridization (FISH) was used to identify the spatiotemporal expression of UGT440A1. Hybridization signals were detected from embryo until adult B. xylophilus were formed. The red fluorescence-labeled probe stained nearly the whole embryo (Figure 4A), whereas a restricted staining pattern was observed in juvenile nematodes (J2, J3, and J4 females), wherein the hybridization signals were detected in the head, intestine, and subcutaneous tissues (Figures 4B,C,E). However, the hybridization signals were detected nearly throughout the body of J4 males and adults (Figures 4D,F,G). No hybridization signal was observed in the negative controls (Figures 4H,I).

The UGT440A1 coding region without the putative signal sequence was amplified through PCR and cloned into the vector pET-15b to construct pET-15b-UGT. The recombinant UGT440A1 was overexpressed in E. coli BL21 (DE3) through IPTG induction at 37°C. The SDS-PAGE analysis indicated that the recombinant protein had the expected molecular mass of 53 kDa, which corresponded to 51.5 kDa from the UGT440A1 ORF and 1.5 kDa encoded by the expression vector, including a His₆tag (Figure 5A). The expressed recombinant protein was found mainly in the inclusion bodies, which could be refolded by stepwise dilution with refolding buffer (Figure 5A). The renatured recombinant UGT440A1 was then purified through Ni2 + affinity chromatography, and the electrophoretic homogeneity was verified through SDS-PAGE (Figure 5A). To assay the activity of UGT440A1, two types of flavonols, quercetin and kaempferol, which are the parent nuclei of flavonoids in pine wood, were selected as substrates to detect whether recombinant UGT440A1 had the catalytic activity. Because the polarity of flavanoids increases when they combine with glucose, the product is expected to peak earlier in case of a glycosylation reaction. The HPLC results indicated that the expressed recombinant protein could catalyze the two substrates, as more than one products were detected (Figure 5B).

To verify that the dsRNA could be transferred effectively into B. xylophilus, nematodes were observed under a fluorescence microscope. The patterns of fluorescein indicated that Cy3 labeled dsRNA was taken up by the nematodes effectively (Figures 6B,D). No fluorescence was detected in the nematode soaked in sterilized water (Figure 6F). And no phenotypic changes were detected in the nematodes after RNAi of UGT440A1 (Figures 6A,C,E). The RNAi efficiency on the UGT440A1 expression level was evaluated through QRT-PCR. The results showed that compared with the expression level of UGT440A1 in the nematodes soaked in sterilized water, that in the nematodes soaked in UGT440A1 dsRNA was only 0.15 (Figure 7A) (p < 0.001). This result indicated that the expression of UGT440A1 in nematodes could be strongly inhibited by soaking the nematodes in UGT440A1 dsRNA. The gfp dsRNA, as one of the negative controls, demonstrated no significant effects on the expression level of UGT440A1 compared with sterilized water (Figure 7A) (p > 0.05).

To analyze the influence of UGT440A1 on the locomotion of B. xylophilus, the head thrashing frequency of the nematodes was observed under a stereo microscope. The nematodes soaked in dsRNA exhibited difficulties in motility. The head thrashing frequencies in the UGT440A1 dsRNA, sterilized water, and gfp dsRNA groups were 14.53 ± 0.19, 25.82 ± 0.37, and 25.59 ± 0.38 times per 30 s, respectively (Figure 7B). These results indicated that UGT440A1 is crucial for B. xylophilus motility.

The nematode feeding areas were observed after the nematodes soaked in the dsRNA of UGT440A1 were transferred to the PDA plate. The feeding rate of the treatment groups was obviously slower than that of the negative controls (Figure 7C). At the ninth day, almost all of the hyphae had been consumed by B. xylophilus in the sterilized water and gfp dsRNA groups, whereas in the UGT440A1 dsRNA group, only a part of the hyphae was consumed (Figure 7C). The same day, the nematodes were collected and counted in each group; the number of nematodes in the UGT440A1 dsRNA, sterilized water and gfp dsRNA groups was 3,192 ± 111, 9,572 ± 182, and 9,642 ± 210, respectively (Figure 7D). Sterilized water and gfp dsRNA were not found to have significant influences on the feeding and reproduction of B. xylophilus (p > 0.05). These results indicated that UGT440A1 is essential for feeding and reproduction of B. xylophilus.

Furthermore, to evaluate the effects of UGT440A1 on egg laying and hatching abilities of the nematodes, we treated the virgin adults with UGT440A1 dsRNA. The number of eggs laid by an individual female nematode in the treatment, sterilized water, and gfp dsRNA groups was 9.33 ± 1.52, 22.33 ± 4.04, and 21.33 ± 2.08, respectively (Figure 7E). The hatching rates for the UGT440A1 dsRNA, sterilized water, and gfp dsRNA groups were 29.11% ± 3.01%, 81.33% ± 3.53%, and 80.78% ± 3.49%, respectively (Figure 7F). These results suggested that UGT440A1 is vital for the egg laying and hatching potential of B. xylophilus.

To examine the virulence of the nematodes, B. xylophilus treated with UGT440A1 dsRNA were inoculated into 2-year-old P. thunbergii seedlings (Figure 8A). Twelve days after inoculation, the pine seedlings inoculated with nematodes that were treated with sterile water and gfp dsRNA exhibited clear symptoms of PWD. The infection rate of both the sterilized water and gfp dsRNA treatment groups was 33.3%, whereas the DSI for the two groups was 16.67 ± 16.67 and 8.33 ± 8.33, respectively (Table 1) (p < 0.05). The UGT440A1 dsRNA treatment group exhibited no PWD symptom in at this time point; however, this group exhibited leaf yellowing 18 days after inoculation. Furthermore, the pine seedlings inoculated with nematodes that were treated with sterilized water and gfp dsRNA exhibited higher infection rates and DSI, respectively, than those treated with UGT440A1 dsRNA at 20 and 40 days after inoculation (Table 1 and Figures 8B,C) (p < 0.05). On day 60, all the pine seedlings in the sterilized water and gfp dsRNA groups died, whereas the UGT440A1 dsRNA-treated seedlings were still alive, despite the obvious symptoms of PWD (Figure 8D). The pine seedlings inoculated with sterile water without nematodes did not exhibit any PWD symptom throughout the experimental period. These results indicated that PWD onset was delayed after RNAi of UGT440A1.