Diffusible signal factor (DSF) (≥99%) was purchased from Shanghai UDChem Technology Co., Ltd (Shanghai, China) and dissolved in methanol to create a stock solution with a concentration of 100 mmol·L^-1. Radishes (Raphanus sativus L.) were purchased from a local market (Guangzhou, China) and healthy plants were selected for the biocontrol experiments.

Ampicillin (AMP, 50 mg·mL^-1), gentamicin (GEN, 50 mg·mL^-1), neomycin sulfate (NEO, 50 mg·mL^-1), carbenicillin (CARB, 50 mg·mL^-1), chloramphenicol (CM50, 30 mg·mL^-1), tetracycline (TC, 5 mg·mL^-1), kanamycin (KAN, 50 mg·mL^-1), and streptomycin (STR, 50 mg·mL^-1), were used for antibiotic susceptibility testing, purchased from Sigma Aldrich Chemicals Co., Ltd (Shanghai, China).

XC1 was provided by the Integrative Microbiology Research Centre, South China Agricultural University, Guangzhou, China. Xcc were cultured on Luria–Bertani (LB) medium (NaCl 10.0 g·L^-1, tryptone 10.0 g·L^-1, and yeast extract 5.0 g·L^-1) with rifampicin (30 μg·mL^-1) at 28°C. The isolates were grown on LB medium or mineral salt medium (MSM: (NH₄)₂SO₄ 2.0 g·L^-1, Na₂HPO₄·12H₂O 1.5 g·L^-1, KH₂PO₄ 1.5 g·L^-1, MgSO₄·7H₂O 0.2 g·L^-1, CaCl₂·2H₂O 0.01 g·L^-1, FeSO₄·7H₂O 0.001 g·L^-1; pH 7.2) with DSF (2 mmol·L^-1) at 30°C. The MSM medium and minimal medium (MM: (NH₄)₂SO₄, 2.0 g; MgSO₄·7H₂O, 0.2 g; CaCl₂, 0.01 g; FeSO₄, 0.005 g; MnCl₂, 0.002 g; K₂HPO₄, 10.5 g; KH₂PO₄, 4.5 g; mannitol, 2.0 g; glycerol, 2.0 g; 1000 mL of H₂O; pH 6.5) were used to test the degradation efficiency of strain F25 on diffusible signal factor (DSF) and for the identification of metabolites.

Soil samples were collected on March 16, 2017, from the surface layer to a depth of 5 cm, at a perennially cultivated sweet potato field in Heshun Lugang, Nanhai District, Foshan City, Guangdong Province (Longitude: 113.14299°; Latitude: 23.02877°). The soil was sampled, bagged, and preserved as a microbial source for strain isolation. MSM medium was prepared, and 50 mL of MSM medium was sterilized in 250 mL triangular flasks. After cooling, DSF mother liquor (mother liquor concentration, 100 mM; methanol as solvent) was added under aseptic conditions, to make the final mass concentration of DSF 0.01 mM. Then, 5 g of soil sample was added, and the solution was incubated at 30°C in a 200 rpm shaker for 7 d. After this, it was transferred to the second batch of MSM medium, with a final mass concentration of DSF of 100 μM at 10% inoculum. After 7 d of incubation under the same conditions, the sample was transferred to MSM medium with a final mass concentration of 200 μM DSF at 10% inoculum, and incubated for another 7 d. The mass concentration of DSF was increased continuously. Then, 1 ml of MSM medium fermentation broth was diluted with sterile water into 10^-1, 10^-2, 10^-3, 10^-4, 10^-5, and 10^-6 fermentation broths in a concentration gradient. The mass concentration of DSF was continuously increased through this method, and the final enrichment of the strain was achieved. Then, 100 μL of diluted fermentation broth from each concentration gradient was evenly spread on LB solid plates and incubated at 30°C in an incubator, and single colonies were picked. The LB solid plate was repeatedly scribed and purified until a single strain was isolated (Bhatt et al., 2020; Zhang et al., 2022). The isolated single strains were stored in a refrigerator at −80°C.

Single colonies of the purified strains were inoculated in 40 mL of MSM basal medium with DSF as the sole carbon source, resulting in a final mass concentration of 2 mM DSF, then incubated at 30°C for 48 h in a 200 rpm shaker for DSF extraction and high-performance liquid chromatography (HPLC) determination of DSF residues. The strain with the highest DSF degradation rate was finally obtained, which was named F25.

The morphological characteristics of the colonies were studied. Strain F25 was scribed in LB solid medium and incubated at 30°C for 48 h. The genome of strain F25 was extracted as a template and 16S rDNA PCR amplification of the strain was performed using bacterial universal primers 27F (AGAGTTTGATCCTGGCTCAG) and 1492R (GGTTACCTTGTTACGACTT) (Li et al., 2022). The 16S rDNA gene sequence of strain F25, with a length of 1407 bp, was obtained using the above method and then compared with the NCBI (National Center for Biotechnology Information) database (http://www.NCBI.nlm.nih.gov/).

The antibiotic susceptibility of strain F25 was studied to better investigate the biocontrol potential of strain F25. A total of 50 μL of the bacterial solution was added to 5 mL of LB liquid medium, followed by the addition of antibiotics, where the final concentration gradient was 5, 10, 20, 50, 100, 150, 200, 250, 300, 350, and 400 μg·mL^-1. Three sets of replicate trials were set up for each concentration gradient. Cultures were incubated at 30°C and 200 rpm for 16–24 h before measuring and recording the OD₆₀₀ values of the cultures.

A single colony of strain F25 was inoculated in LB medium and pre-cultured to the logarithmic phase, following which the resulting broth was centrifuged at 4000 rpm for 5 min. The supernatant was discarded, and the organism was washed in 0.9% sterile saline, then re-suspended as seed suspension. The bacteria were then inoculated into 50 mL of MSM basal medium at an inoculum of 1:100 and DSF was added to a final concentration of 2 mM. The culture was incubated at 30°C for 72 h at 200 rpm, and samples were taken at regular intervals. The samples were collected at different time points, and the OD₆₀₀ value was measured spectrophotometrically to indicate the growth of strain F25, while the residual amount of DSF was measured by HPLC to indicate the degradation of DSF by strain F25.

To study the antagonistic effect of strain F25 with pathogenic bacteria, studies were carried out on LB solid plates (Li et al., 2020; Zhou et al., 2022). A bacterial solution of Xcc was inoculated at 10% into the melted LB. After cooling, the LB agar plates containing the pathogen were punched with an inactivated punch, and 20 μL each of bacterial solution, metabolite solution, acetonitrile, and sterile water of strain F25 were injected into the punched wells. The plates were incubated at 30°C for 24 h. If antagonism occurred, hyaline circles would appear in the LB plate.

Single colonies of strain F25 and XC1, a DSF-dependent pathogen, were isolated and pre-cultured in LB medium until the logarithmic phase. The resulting broth was centrifuged at 4000 rpm for 5 min, and the supernatant was discarded. The organisms were washed with 0.9% sterile saline and re-suspended as seed suspension. The bacteria were then inoculated into LB medium at an inoculum of 1:100 and incubated at 30°C and 200 rpm until the logarithmic phase. The bacteria were re-suspended in PBS buffer to obtain suspensions of strain F25 and XC1.

The bacterial suspension of strain F25 was mixed with the suspension of XC1 to obtain the mixed bacterial solution. The fleshy roots of white radish were washed with distilled water, then sliced when the surface was dry. The fleshy roots were sliced crosswise to obtain round slices about 0.3 cm thick and placed in Petri dishes (with cotton moistened with sterile water). The OD₆₀₀ of both strain F25 and XC1 was 0.2. The mixture was coated with a spreading rod and incubated at 30°C for 48 h to observe the incidence. XC1 alone and F25 alone were used as positive and negative controls, respectively. In total, there were four experimental groups, divided into XC1+sterile water, XC1+F25+sterile water, F25+sterile water, and sterile water. In the trays in which the radish slices were placed, we included moistened sterile wet water sponges and the trays were sealed with cling-film. The experimental procedure was repeated at least three times for each group. Incubation was carried out at 28°C for 48 h. The extent of the lesion was quantified by measuring the area of maceration, compared to the tissue without the lesion before inoculation (Liao et al., 2015; Liu et al., 2017).

The crude enzyme solution of strain F25 was prepared by the following method. The overnight culture of strain F25 was centrifuged at 4°C for 10 min at 10,000 rpm, and the resulting supernatant was the extracellular enzyme. Washing and suspending the cellular sediment was achieved by rinsing it three times with PBS. The cell suspension of strain F25 was then sonicated and centrifuged. The supernatant obtained was the intracellular enzyme to be collected. To verify its effect on the control of Xcc black rot, experiments were performed with the resulting crude enzyme extract, with four designs: (1) XC1-treated radish slices; (2) radish slices treated with 1 µl of XC1 and intracellular enzyme of strain F25; (3) treatment with 1 µl of extracellular enzyme of strain F25 and XC1 radish slices; and (4) radish slices treated with sterile water as a control. Three replicate trials were performed for all treatments. A final assessment of the disease level severity was performed.

To determine whether strain F25 presents acylase activity (Bao et al., 2020), we analyzed the degradative enzymes of F25 using an Acylase Activity Assay Kit (Solarbio, Beijing Solarbio Science & Technology Co., Ltd., China). Strain F25 was incubated in LB for 12 h. Intracellular and extracellular enzymes of strain F25 were extracted and collected separately. Four experimental groups were designed, with one intracellular enzyme and one extracellular enzyme setup, a negative control group without enzyme addition, a blank control group without enzyme addition, and a blank control group with distilled water. The acylase catalyzed the transfer of the acetyl group of acetyl coenzyme A to butanol and the simultaneous reduction of 5,5’-dithiobis-(2-nitrobenzoic acid) (DTNB), to produce 2-nitro-5-thiobenzoic acid (TNB), which is a yellow compound.

Experimental data were analyzed by one-way analysis of variance (ANOVA), and means were compared by Bonferroni’s multiple comparison test using the GraphPad Prism software (Version 6.0). Experiments were arranged as a completely randomized design, and P < 0.05 was considered to indicate statistical significance.

In this study, strains with DSF degradation ability were screened from the collected soil samples by enrichment culture. MSM with DSF as the sole carbon source was prepared and strains from different soil samples were enriched in the MSM. Nine morphologically different strains capable of degrading DSF were attained by streaking plate method and named as F20~28, respectively. Among them, strain F25 with rapid DSF degradation potential was selected and used in further studies. The F25 strain was deposited at the Guangdong Microbial Culture Collection Center (GDMCC) with the collection number GDMCC NO: 60346.

The above strain F25 was scribed in LB solid medium and incubated at 30°C for 48 h. As shown in Figure 1A , the colonies were yellowish-green in color, slightly elevated, and had smooth and opaque surfaces and neat edges. Strain F25 was diffusely turbid and aerobic in LB liquid medium. Scanning electron microscopy indicated that the strain was rod-shaped (or short spherical), with a size of 0.5–1.0 × 0.3–0.5 μm ( Figure 1B ).

The genome of strain F25 was extracted as a template, and 16S rDNA PCR amplification of the strain was performed using bacterial universal primers, in order to obtain the 16S rDNA gene sequence of strain F25, which was 1407 bp in length. It was compared with the NCBI database (http://www.NCBI.nlm.nih.gov/), and we found that strain F25 had 99% high homology with Burkholderia cepacia TC62 (AY677087.1). The standard strains in the BLAST results were compared, and a phylogenetic tree was constructed using MEGA 6.0. The constructed phylogenetic tree is shown in Figure 1C .

Therefore, strain F25 was identified as Burkholderia sp., in accordance with its morphological characteristics, 16S rDNA gene sequence, and phylogenetic analysis. The antibiotic susceptibility of strain F25 was further investigated. The results showed that strain F25 was resistant to 400 μg·mL^-1 or more of gentamicin, neomycin, carbenicillin, ampicillin, and streptomycin; 300 μg·mL^-1 of kanamycin; and 50 μg·mL^-1 of tetracycline and chloramphenicol ( Figure 2 ).

The DSF degradation ability of the strains was tested by growing the strains in MSM spiked with DSF. The solution was extracted at regular intervals and stored for the detection of residual DSF. The HPLC results are shown in Figure S1 , from which it can be seen that the amount of DSF decreased with time and finally disappeared completely. At 12, 24, 36, 48, and 60 h, strain F25 degraded DSF up to 16.35%, 29.20%, 32.63%, 83.00%, and 100%, respectively. The growth and degradation curves of the corresponding strain F25 with DSF as the only carbon source are shown in Figure 3 , where the degradation of DSF was positively correlated with the growth of the strain. In the presence of DSF, the strain growth had no lag period and rapidly entered the logarithmic phase of growth. The fastest phase of DSF degradation by this strain was 36-48 h and, when the strain was cultured to 60 h, the DSF was completely decomposed. The natural degradation rate of DSF in the control was 20% within 60 h.

Strain F25 was inoculated into MSM medium spiked with DSF, and multiple time points were set for sampling, in order to investigate the pathway of DSF degradation by strain F25. Five metabolic products were identified during the biodegradation of DSF. Figure S2 shows the results of the GC-MS analysis. A distinct peak was detected at a retention time (RT) of 17.459 min with a characteristic mass fragment [M+] at m/z = 99.0 and a major fragment ion at m/z = 43.1; this important compound was identified as DSF ( Figure 4A ). With time, the peaks of DSF decreased and a new compound appeared, at a RT of 15.003 min with m/z = 73.0 as the base peak, which was characterized as n-decanoic acid, based on the elution time and how well the molecular ion matched the corresponding authentic compound in the NIST database ( Figure 4B ). In addition to this, we detected several degradation products of DSF ( Figures 4C–F ). It is worth noting that these metabolites were transiently present. Finally, DSF was completely decomposed into carbon dioxide and water.

The metabolic pathway of DSF in strain F25 was proposed based on the intermediates formed during degradation and the chemical structure of DSF, as follows. First, the degradation of DSF begins with the oxidation of branched carbon atoms to form fatty acids with one less carbon atom. Then, the cis-double bond of the formed intermediate is converted to a trans-double bond by isomerase. Through β-oxidation of the fatty acid, the intermediate product forms a trans-2-decenoic acid with two fewer carbon atoms. The unsaturated fatty acids then undergo hydrogenation to form saturated fatty acids. Finally, these compounds disappeared with the complete degradation of DSF by strain F25.

Antagonism experiments were performed on strains F25 and XC1, in order to verify the presence of antagonism between them. LB agar plates containing pathogenic bacteria were prepared, and bacterial suspensions of strain F25 were injected into the plates. The results demonstrated that, when strain F25 was grown on the same plate as XC1, the plate showed no obvious transparent circle, indicating no zone of inhibition. The other two controls also showed no zone of inhibition. The antagonism experiment indicated that strain F25 has no antagonistic effect on Xcc.

We tested the biocontrol ability of strain F25 against Xcc on radish slices, where the severity of the disease was reflected by the size of the maceration zone. The results are shown in Figure 5 . The severity of the disease was high in radish slices treated with XC1 alone ( Figure 5C ). The dipping zone was significantly smaller in radish slices treated with the F25 and XC1 mixture ( Figure 5D ), while treatment of radish slices with strain F25 alone indicated that the strain was not harmful to the radishes ( Figure 5B ). Treatment of radish slices with only distilled water did not show any disease ( Figure 5A ). These results indicate that strain F25 exhibited good biological control potential against Xcc and, so, can be used as a biological agent to prevent infection by bacterial plant diseases.

To verify the biocontrol effect of strain F25 crude enzyme on Xcc, the pathogenic bacteria XC1, intracellular enzyme, and extracellular enzyme were inoculated onto radish slices under ex vivo conditions and incubated for 48 h. The experimental results are shown in Figure 6 , where a slight decay was observed in the crude enzyme-treated group; however, the area of the macerated zone was significantly lower, compared to the XC1-treated group. These results indicate that the crude enzyme of strain F25 also has a good biocontrol effect and has great potential to be developed for the treatment of plant diseases caused by Xcc.

To verify whether strain F25 presents acylase activity, intracellular and extracellular enzymes of strain F25 were extracted and tested using an analytical kit. From the results, we observed that the control group was colorless, while the experimental group with the addition of the intracellular enzyme exhibited a distinct yellow color ( Figure S3 ), indicating that strain F25 possesses acylase activity.