H. glycines cysts were obtained from infected soybean roots and rhizosphere soil by massaging in water and sieving the solution through a set of 850 μm and 250 μm pore-sized nested sieves. Cysts retained on the 250-μm sieve were purified using the sucrose flotation method (Krusberg et al., 1994). Subsequently, the cysts were crushed and passed through with 45-μm and 25-μm pore-size sieves. Eggs recovered from the 25-μm sieve were surface-sterilized with 0.5% (w/v) NaClO and rinsed several times with sterile distilled water (SDW). Sterilized eggs were incubated in a Baermann funnel at 26°C, and the second-stage juveniles (J2s) were collected daily.

Strain Sneb159 was isolated from the crop rhizosphere soil, identified as M. maritypicum and stored at -80°C (Zhao et al., 2019). Activated Sneb159 was cultured in 100 mL lysogeny broth (LB) at 200 rpm at 28°C for 72 h. Bacterial cells were diluted to 10⁸ CFU/mL (OD600 = 0.8) with SDW. Culture filtrates were achieved via centrifugation at 12,000 rpm for 10 min and passed through a 0.22-μm filter to remove additional bacterial cells. Bacterial cells were diluted with SDW to a concentration of 10⁸ CFU/mL.

To evaluate the nematicidal activity of Sneb159 on the J2 of H. glycines, 1 mL of fermentation or LB broth was added to a 2 mL centrifuge tube containing 50 J2s. Besides, the supernatant or bacterial cells of 10⁸ CFU/mL were diluted 10-fold and 100-fold with SDW, respectively. The number of dead nematodes was recorded after 24, 48, or 72 h under a stereomicroscope (Nikon SMZ800, Japan). The experiment was conducted three times with three replications. Nematodes were treated with culture filtrates and bacterial cells diluted 10-fold and 100-fold, and mortality was determined after 48 h. The experiment was conducted twice with three replications. Nematodes were considered dead if they were rigid and did not respond to a small needle probe. The mortality rate was calculated using the following equation:

Soybean seeds (Liaodou15) were disinfected with 1% NaClO for 5 min and washed at least three times with distilled water. An individual soybean seed was cultivated in a 12 cm diameter plastic pot filled with 300 g of sterile sand. At the third true leaf stage, 10 mL of Sneb159 bacterial cells (culture filtrates diluted 0, 10, or 100-fold) were injected around the root. Subsequently, 2000 J2s were inoculated after 48 h. The number of juveniles inside the roots was measured at 5 days post infection (dpi) and 15 dpi by using the acid fuchsin staining method. The experiment was performed twice, and each treatment consisted of five replicates.

The chemotaxis in pot experiments was designed as described by Wang et al. (2019). A total of 250 mL of sterilized sand was filled in each of the plastic cups (9 × 15 cm). Plastic cups containing soybean seeds were grown in a greenhouse at 26°C for a 16-h photoperiod. A 5 mL liquid culture of Sneb159 or sterile distilled water was added to the opposite plastic cups when the seedlings had developed the third true leaf, and both sides added with sterile distilled water were considered blank controls. Two plastic cups were connected with a polyvinyl chloride tube (1.5 cm diameter, 7 cm length). The tube was filled with sterilized sand, and a hole (1 cm in diameter) was drilled in the middle. The J2s (about 1000) were injected through the hole and wrapped in tin foil for protection from light. The roots were collected, stained with acid fuchsin at 10 dpi, and observed using a stereomicroscope. The experiment was conducted twice with eight replications.

The nematicidal activities of Sneb159 and Sneb517 VOCs were assayed using a three-cell Petri dish method described by Zhai et al. (2018) with minor modifications. In brief, 2.5 mL of culture filtrates were added to one compartment, and a layer of 2.5 mL of 2% water agar was added to another compartment with 50 μL of approximately 100 H. glycines J2 on the surface. Control plates contained LB instead of the culture filtrates. Activated charcoal was added to the third compartment to absorb the VOCs generated by Sneb159 and confirm that VOCs paly a main role in nematicidal activity. The plates were immediately sealed and cultured at 26°C. The experiment was conducted three times with three replications. The mortality was measured under a stereomicroscope (Nikon SMZ800, Japan) after 72 h.

Three linked wells of a 96-well plate were used to determine the effects of the Sneb159 VOCs on egg hatching. A drop of 300 μL of Sneb159 culture filtrate was added to the middle well. Subsequently, approximately 50 eggs (100 μL) were added to the other two wells. SDW and LB were used as controls. The hatchability rate was investigated at 9 d after incubation. The experiment was conducted three times with four replications.

Sneb159 was cultured in 100 mL LB at 200 rpm at 28°C for 72 h. A total of 9 mL fermentation broth was added to a 20-mL headspace vial. The VOCs were extracted from the headspace vials of inoculated and uninoculated broths using a 65-mm PDMS/DVB SPME fiber (Supelco Inc., Bellefonte, PA, USA). Specifically, the SPME fiber pierced the PTFE septum of the headspace vial and was exposed above the liquid for 1 h while the vial was immersed in a thermostatic water bath at 60°C. Immediately after VOC extraction, the fiber was inserted into the gas chromatograph front inlet and desorbed at 250°C for 5 min. The experiment was conducted using an Agilent 6890N gas chromatograph (Agilent Technologies, Santa Clara, CA, USA) that was connected with an Agilent 5973N mass spectrometer and a DB-5 capillary column (30 m × 0.25 mm × 0.25 μm).

The flow rate of the carrier gas (helium) was 1 mL/min in splitless mode. The oven parameters were set according to (Zhai et al., 2018) and were as follows: 40°C for 2 min, raised to 180°C at 4°C/min rate, and then raised to 250°C at 5°C/min rate, and held for 6 min. The transfer line and ion source temperatures were 150°C and 230°C, respectively. The MS detector ranged between 35 and 520 m/z. The identification of compounds was achieved by using the National Institute of Standards and Technology (NIST98) reference library and the comparison of the retention times. The VOCs from the LB medium were considered the control group. The unique substances produced by strain Sneb159 were identified after eliminating the substances also detected in the LB medium. Each sample was measured twice.

For nematicidal tests, commercial dimethyl disulfide, dimethyl trisulfide, and cyclohexyl(dimethoxy)methylsilane were used. The chemical solution was diluted to 50-800 mg/L with 0.03% Tween-80 (v/v). The 0.5 mL centrifuge tubes containing 10 μL H. glycines J2s (approximately 50 J2s) were filled with different concentrations of test solutions (300 μL). J2s soaked in 0.03% Tween-80 (v/v) solution were used as the control group. The mortality rate was measured after incubation for 24 and 48 h at 26°C. The experiment was conducted three times with three replications.

In order to determine the effect of commercial VOCs on hatchability, 50 eggs suspended in 10 μL SDW were placed in 0.5-mL centrifuge tubes containing 300 μL test solutions. The concentration of dimethyl disulfide and dimethyl trisulfide ranged from 100 to 800 mg/L. The 0.03% Tween-80 (v/v) solution was used as a control. Hatchability was calculated after 9 d at 26°C. Each treatment was conducted in triplicate with three replications.

The fumigation experiments were performed using a 96-well plate. 300 μL of commercial VOC was added to one cell of a 96-well plate, and a 100 μL suspension of nematodes or eggs (approximately 50) was dropped into the two adjacent wells. The final concentration of commercial VOCs was adjusted to 200, 400, or 800 mg/L using 0.03% Tween-80 (v/v). The three wells were sealed with parafilm, and 0.03% Tween-80 (v/v) was used as a control. The number of dead nematodes was counted after 72 h at 26°C. The percentage of hatched eggs was recorded after 9 d at 26°C. The experiment was conducted three times with three replications.

For sequencing, the total DNA of M. maritypicum Sneb159 was extracted using E.Z.N.A^® Bacteria DNA kit (Omega). Pair-end libraries were prepared following the manufacturer instructions with the TruSeq™ Nano DNA Sample Prep Kit (Illumina, USA), and sequenced on the Pacific Biosciences Sequel II technology (PacBio) and the Illumina NovaSeq 6000 platform (150 bp×2, Shanghai Biozeron Biotechnology Co., Ltd., Shanghai, China). The raw paired-end reads were trimmed and quality controlled by Trimmomatic with Parameters (SLIDINGWINDOW: 4: 15 MINLEN: 75) (v0.36; http://www.usadellab.org/cms/uploads/supplementary/Trimmomatic). The collected clean data was used for further analysis. Raw PacBio reads were converted to fastq format with Samtools fastq (http://www.htslib.org/doc/samtools.html). The Illumina data was used to evaluate the complexity of the genome and correct the PacBio long reads. First, we used unicycler v0.4.8 (https://github.com/rrwick/Unicycler) to perform genome assembly with default parameters and received the optimal results of the assembly. GC depth and genome size information were calculated by custom Perl scripts (Biozeron Biotechnology Co., Ltd), which can help us judge whether the DNA sample contained contaminated or not. Finally, the strain genome was circularized with Circlator v1.5.5 (http://sanger-pathogens.github.io/circlator/).

For the prokaryotic organism, we used the ab initio prediction method to get gene models for M. maritypicum Sneb159. Gene models were identified using GeneMark v4.17 (http://topaz.gatech.edu/GeneMark/). Then all gene models were blastp against non-redundant (NR in NCBI) databases, SwissProt (http://uniprot.org), KEGG (http://www.genome.jp/kegg/), and COG (http://www.ncbi.nlm.nih.gov/COG) to do functional annotation by the blastp module. In addition, tRNA was identified using the tRNAscan-SE (v2.04; http://lowelab.ucsc.edu/tRNAscan-SE), and rRNA was determined using the RNAmmer (v1.2; http://www.cbs.dtu.dk/services/RNAmmer/). Homologs to known virulence factors, transport proteins, antibiotic resistance genes, and metal resistance genes were identified using the following databases: the Virulence Factor Database (VFDB: https://www.mgc.ac.cn/VFs/main.htm), the Transporter Classification Database (TCDB: https://tcdb.org/), the Comprehensive Antibiotic Resistance Database (CARD: https://card.mcmaster.ca/), and the antibacterial biocide and metal resistance genes database (BacMet version 2.0: http://bacmet.biomedicine.gu.se/). The second metabolite biosynthetic gene cluster of M. maritypicum Sneb159 was predicted using the antiSMASH 7.0 platform (Blin et al., 2023).

Nematicidal activity of Sneb159 fermentation broth and chemotaxis response in the pot experiment were tested with an independent sample t-test (p<0.05 or p<0.001). Nematicidal activity of filtrate, pot experiment, Sneb159 VOCs on juveniles and eggs was analyzed using one-way analysis of variance with Tukey’s post hoc test (p<0.05). Nematicidal activity data for commercial VOCs were corrected using the Schneider-Orelli formula (Ntalli et al., 2011), and LC₅₀ values were calculated using PROBIT analysis (Cheng et al., 2017). All data are shown as the mean ± standard error. All statistical tests were performed using IBM SPSS Statistics 23 software (IBM Corp., Armonk, NY, USA).

The fermentation of Sneb159, cultivated in LB medium, showed strong nematicidal activity against J2 of H. glycines with mortalities of 88.43, 93.7, and 95.53% at 24, 48, and 72 h, respectively. The mortality rate for the LB treatment was 9.57, 12.88, and 13.45% at 24, 48, and 72 h, respectively, which were all significantly lower than those of the Sneb159 treatment according to the t-test (P<0.001) ( Figure 1A ). Moreover, the nematicidal activity of fermentation filtrate and bacterial cells against nematodes was determined after separation. The results revealed that the mortality rates for fermentation filtrate diluted 10 and 100 folds were 88.57, and 85.95%, which were significantly higher than the control using Tukey’s post hoc test (p<0.05) ( Figure 1B ). However, no notable nematicidal properties were observed in the bacterial cell treatment, both bacterial cells diluted 10 and 100 folds with no significant difference compared to the control (P<0.05).

The effect of Sneb159 on the invasion and development of H. glycines juveniles was evaluated by pot experiments. At 5 dpi, pre-inoculation with bacterial cells or filtrate led to a significant reduction in the number of infected J2s (P<0.05). Compared with the control, the reduction in the filtrate, filtrate dilution 10-fold, and filtrate dilution 100-fold was 70.67%, 68.3% and 56.88% respectively. The inhibitory effect decreased as the dilution ratio increased ( Figure 2A ). However, no differences were observed among bacterial cells, filtrate, 10-fold and 100-fold filtrate.

At 15 dpi, the number of different stages of juveniles was under investigation. Most juveniles have already developed to the third and fourth stages of juveniles (J3 and J4) ( Figure 2B ). Both J3 and the total number of juveniles in filter-treated roots were remarkably decreased compared to the control (p<0.05), with reduction rates of 70.60% and 65.43% accordingly. However, no statistical significance was observed when the filtrate was diluted 10 and 100-fold for the number of J3, J4, and total. After treatment with bacterial cells, the number of J3 was decreased by 42.27%, the J4 number increased by 46.65% compared to the control, and there was no significant difference in the total number of juveniles between them.

The effect of Sneb159 on nematode chemotaxis in soil was determined ( Figure 3A ). J2s were repelled by the Sneb159 fermentation broth, with more juveniles moving to the other side of the sterile water treatment. The greatest decrease was 53.41% compared to the control (P < 0.001) ( Figure 3B ). For control, there was no significant difference in the number of J2s between the two sides. No significant difference in total nematode numbers between the treatment group and the control group (P < 0.05) means that repelling rather than killing caused the reduction in nematodes on the Sneb159 side.

To evaluate the influence of VOCs from M. maritypicum Sneb159 fermentation broth on H. glycines J2s, three-cell Petri dishes were used. After 24 h of incubation, the mortality in the Sneb159 VOCs was 68.61%, which was significantly higher than that in the LB control (12.35%) ( Figure 4A ). The mortality rate reached 76.84% in Sneb159 treatment after 48 h. Activated charcoal was added to absorb the volatiles, and the mortality was 5.33% and 8.28% after 24 and 48 h, respectively. The juveniles were coiled and mobile in the LB control at 48 h. While most of the juveniles in the Sneb159 treatment were stiff and immobile, their organs were destroyed and dissolved, and several severe vacuolations appeared in their bodies ( Figure 4C ). After 9 d, the egg hatchability of the Sneb159 volatiles treatment was 10.25%, which was greatly reduced compared to the LB control (25.81%) and the SDW control (25.54%) (p<0.05) ( Figure 4B ).

The specific components of the VOCs collected via headspace-solid phase microextraction were identified by GC/MS analysis (Figure 5). Seven compounds (Qual > 80) were identified in Sneb159 after the removal of the common substance in LB, including dimethyl disulfide, 2,5-dimethyl pyrazine, dimethyl trisulfide, 2-ethylhexanol, 3-ethyl-2,5-dimethylpyrazine, phenylacetone, and cyclohexyl(dimethoxy)methylsilane ( Table 1 ). The retention of Sneb159 spanned from 4.483 to 30.755 minutes. Dimethyl disulfide was the most abundant compound in Sneb159, with a 50.25% relative area, followed by cyclohexyl(dimethoxy)methylsilane (10.02%), 2-ethylhexanol (2.1%), phenylacetone (1.36%), and dimethyl trisulfide (1.04%). Therefore, dimethyl disulfide, cyclohexyl(dimethoxy)methylsilane and dimethyl trisulfide were selected for further analysis.

The nematicidal activity of dimethyl disulfide and dimethyl trisulfide against H. glycines was determined using LC₅₀/24 h values of 268.58 and 63.3 mg/L. The LC₅₀ value for 48 h was 212.04 mg/L for dimethyl disulfide and 9.80 mg/L for dimethyl trisulfide ( Figure 6A ). No nematicidal activity was observed for cyclohexyl(dimethoxy)methylsilane with a mortality rate of less than 12% even at a concentration of 800 mg/L (data not shown).

In vitro exposure of H. glycines eggs to dimethyl disulfide and dimethyl trisulfide resulted in a reduction in the number of hatched J2s. The inhibition rate of dimethyl disulfide varied from 24.94% to 63.36% as the concentration ranged from 100 mg/L to 800 mg/L. A hatching reduction of 46.87% was observed at a concentration of 800 mg/L for dimethyl trisulfide ( Figure 6B ).

The fumigant activities of dimethyl disulfide and dimethyl trisulfide against H. glycines were determined. The results showed that the corrected mortality of dimethyl disulfide at concentrations of 400 and 800 mg/L was 47.02% and 60.32% at 72 h, respectively. And the corrected mortality of dimethyl trisulfide was 52.63% at 800 mg/L after 72 h exposure ( Figure 7 ). Besides, dimethyl disulfide and dimethyl trisulfide have a negative effect on egg hatching. The inhibition of dimethyl disulfide at 400 and 800 mg/L was 34.34% and 48.28%; and a 39.38% reduction in egg hatchability of dimethyl trisulfide at 800 mg/L ( Figure 7 ).

To better understand the genomic features of M. maritypicum Sneb159, the third generation was conducted by combining the Illumina TruSeq system and PacBio Sequel II high-throughput sequencing technologies. The genome assemblies of Sneb159 showed a 3.89 Mb genome with an average G + C content of 68.63%, consisting of one circular chromosome ( Table 2 , Figure 8). In total, 3,746 protein-coding genes with an average gene length of 965 bp were present in Sneb159. In addition, 6 rRNA genes and 48 tRNA genes were also identified. The genome sequence data have been uploaded to the GenBank database with accession number CP168030.

A total of 3161 CDS were assigned to COG (Clusters of Orthologous Groups) by EggNOG. The functional classification of COG showed that the largest proportions were unknown functions, which accounted for 14.96% of the entire CDSs ( Figure 9 ). Genes related to amino acid transport and metabolism were the second largest, with 12.59%, and 400 genes related to carbohydrate transport and metabolism accounted for 11.29%. 397 genes related to the transcript and 258 genes related to inorganic ion transport and metabolism were also commented on. It has been demonstrated that the transport and metabolism of amino acids and carbohydrates are fundamental to all life entities and play a crucial role in sustaining life activities.

Groups of genes that may be responsible for the virulence of the strain were investigated. The M. maritypicum Sneb159 genome was annotated with 715, 75, and 7 genes for known transporters, antibiotic resistance genes, and virulence factors. Most of them were associated with resistance to macrolides (55.7%), glycopeptides, and peptides (7.1%) ( Table 2 ).

Three putative biosynthetic gene clusters of M. maritypicum Sneb159 were predicted in the genome using antiSMASH. They were assigned to three types, including hydrogen-cyanide, terpene, and T3PKS. In two established biosynthetic gene clusters, carotenoid and frankobactin, the similarity was 28% and 12%, respectively. In the results of comparison with known gene clusters, one gene cluster showed 50% similarity with the cluster carotenoid, the other one showed 12% homology with the cluster frankobactin, and one failed to match the known cluster.