The plasmids and R. solanacearum strains used in this study are listed in Table 1. Escherichia coli strain DH5α (Invitrogen, Carlsbad, CA, United States) was used as a host in gene cloning and vector construction. R. solanacearum strains were maintained at 30°C in casamino acid-peptone-glucose (CPG) plates (Hendrick and Sequeira, 1984), and cultured in CPG broth for testing CWDE activities (Yap et al., 2005), or on minimal medium (MM) agar plate for screening transformants after tri-parental mating (Hussain et al., 2008). E. coli was grown at 37°C in LB medium. Antibiotics were added at the following final concentrations (μg/L): kanamycin, Km (50), gentamicin, Gm (50), and rifampicin, Rif (50). The fungal strains used in this study were maintained in PDA medium unless otherwise stated.

The phcB, rmyA, rmyB, and rmyAB deletion mutants were generated by amplifying two DNA fragments flanking their coding sequences (using primers as described in Table 2), the 859 bp gentamicin resistance gene sequence (Gen) was added between the left and right fragments of phcB or rmyA by using the overlap primers, the left and right DNA fragments of rmyB were ligated without adding the Gen sequence. The two/three fragments were fused using the primer pair L1/R2. All the PCR procedure of this study were amplification using the high fidelity Taq polymerase (PrimeSTAR^® HS DNA Polymerase, Takara Bio Inc., Dalian, China), cycling conditions were set as follows: initial denaturation at 98°C for 1 min, followed by 34 cycles of denaturation at 98°C for 15 s, annealing at 60°C for 20 s, extension at 72°C for 30 s, and a final extension at 72°C for 5 min. The fusion fragments and the suicide plasmid pK18mobsacB (ATCC^®87097^TM) (Schäfer et al., 1994) were digested, respectively, with corresponding restriction enzymes as indicated, followed by purification by using Sangon purification kit (Sangon, Shanghai, China), and then ligated together by using T4 ligase [NEW ENGLAND BioLabs (Beijing) Ltd., China]. The ligation products were transformed into E. coli DH5α competent cells (Life Technologies Corporation, Beijing, China), and the bacterial cells were cultured at 37°C with shaking for 1 h. The transformants were selected in LB medium supplemented with Km and verified by PCR analysis. The plasmid constructs were introduced into R. solanacearum by using tri-parental mating as described previously to generate corresponding in-frame deletion mutants (Ditta et al., 1980). Mutants were selected on CPG plate containing 10% sucrose, antibiotics Gm and Rif, and then confirmed by PCR and DNA sequencing. To construct complementary strains, the DNA fragment containing 235 bp promoter sequence and ORF of phcB was amplified using primers phcB-CF and phcB-CR (Table 2). The purified PCR products were digested by required restriction enzymes, and then purified again prior to ligation with the same enzymes digested expression vector pBBR1MCS2 (Kovach et al., 1995). The ligation products were transformed into E. coli DH5α competent cells, and the transformants were selected on LB plate containing Km. Tri-parental mating was performed as above description, and the complemented strains were selected on MM plates supplemented with Km and Rif, and confirmed by PCR product analysis. The gel electrophoresis results of the mutants and control are provided in Supplementary Information File 1 (SI 1).

Preparation of conidial suspensions of the FOC strain XJZ2 was performed as described (Spraker et al., 2016). For co-culture experiments, R. solanacearum strain GMI1000 and the fungal pathogen were plated as previously described and incubated for 8 days at 28°C. When the colonies met each other, the fungi mycelia of the interaction zone were harvested and examined for FOC chlamydospore formation by using ZEISS Observer Z1 microscopy. For testing bacterial interaction with S. scitamineum, the MAT-1 and MAT-2 (Yan et al., 2016) mating mixture of S. scitamineum were grown on PDA plates for observation of hypha growth. To observe the dose effect of strain GMI1000 on inhibition of S. scitamineum, the GMI1000 (A) was cultured 24 h on the CPG plate before adding GMI1000 (B), other GMI1000 derivatives, and the MAT-1/MAT-2 mating mixture of S. scitamineum on the CPG plate.

Cultures of R. solanacearum strains in CPG broth (OD₆₀₀ = 1.5) were centrifuged and total RNA were isolated by using RNeasy Mini Kit (QIAGEN, Hilden, Germany). The contaminated genomic DNA were removed by treating with DNaseI (Takara, Dalian, China) at 37°C for 1 h, and was confirmed by PCR using the 16S primer pair and visualized on an agarose gel. The qRT-PCR experiments were performed with SuperReal PreMix Color SYBR Green, 2X, (TIANGEN BIOTECH CO. LTD, Beijing, China) on QuantStudio 6 Flex (applied biosystems by life technologies, Carlsbad, CA, United States) following the user’s guide from the manufacturer. The PCR conditions were as follows: 95°C for 15 min, followed by 40 cycles of 95°C for 15 s and 55°C for 31 s, and then the melting curve analysis was carried out to determine the specificity of PCR products. The cDNA samples of each treatment were replicated three times, and the absolute value of –ΔΔCt = –(ΔCt₁ – ΔCt₂) were calculated as described in the 2^-ΔΔCt method (Livak and Schmittgen, 2001).

To prepare ralsolamycin extracts, bacteria were inoculated in 1 litre CPG broth and cultured at 28°C for 24 h with shaking at 150 rpm (OD₆₀₀ = 1.5). Bacterial supernatants were collected by centrifugation at 10,000 rpm for 10 min, mixed with an equal volume of ethyl acetate with shaking. The upper organic phase was collected and air dried, and the residues were dissolved in methanol for further analysis. LC/MS analysis was conducted using a Waters LC-MS system (Waters, MA, United States) with an ACQUITY UPLC system coupled to the Waters Q-Tof Premier high resolution mass spectrometer. An ACQUITY UPLC BEH C18 column (2.1 mm × 50 mm) was used for chromatography analysis. Solution A was composed of 0.01% formic acid in water; solution B was composed of 0.01% formic acid in CH₃OH. A linear gradient elution from 10 to 100% B over 10 min at 0.4 ml/min, 100% B over 3 min and re-equilibrated with 10% B for an additional 3 min. The injection volume was 1 μl. The entire column elute was introduced into the Q-Tof mass spectrometer. Ion detection was achieved in ESI mode using a source capillary voltage of 1.5 kV, source temperature of 120°C, desolvation temperature of 350°C, cone gas flow of 50 L/h (N₂), and desolvation gas flow of 600 L/h (N₂). Presence of the peak in the bacterial extract showing the accurate mass of ralsolamycin (m/z 1291.7142) was identified, and the corresponding peak areas from wild-type and mutant were compared and calculated.

To evaluate the role of the PhcBSR QS system in regulation of the rmy genes expression and ralsolamycin production, the QS signal synthase gene phcB deletion mutant was constructed in the genetic background of R. solanacearum strain GMI1000 by deletion of its coding sequence. The resulting strain ΔphcB was non-mucoid and nearly avirulent with weak cellulose activity, but was highly motile and with an increased polygalacturonase activity as previously reported (Flavier et al., 1997). The expression levels of rmyA and rmyB in ΔphcB were determined and compared to its parental strain GMI1000. In the ΔphcB, the rmyA transcription was drastically decreased by about 24-fold (Welch t-test, p-value < 3.64e-4; Figure 1), and the gene rmyB transcription was decreased by around 4.6-fold (Welch t-test, p-value < 3.02e-3; Figure 1). Meanwhile, deletion of phcB also resulted in decreasing the expression level of RSp0638 and RSp0639 by approximately 6- and 7-fold, respectively (Welch t-test, p-value < 2.32e-3 and 2.08e-3, respectively; Figure 1), which are the two neighbor genes of the rmy operon. In trans expression of a wild-type phcB gene in ΔphcB restored their expression to the wild-type levels. Consistent with the results of transcriptional assay, LC-MS analysis of the ethyl acetate extracts from wide-type strain GMI1000 and ΔphcB showed that ralsolamycin production in ΔphcB was decreased by about 81-fold compared with the wide-type strain (Welch t-test, p-value < 6.86e-5; Figure 2).

Given that bacterial QS systems modulate target gene expression in a population density dependent manner, the transcriptional levels of rmyA and rmyB in strain GMI1000 was then determined at different growth stages of bacterial cells. The results showed that the transcripts level of rmyA and rmyB in strain GMI1000 at a high cell density (∼10⁸ CFU/ml) was increased by about 2.3- and 4.8-fold, respectively (Welch t-test, p-value < 3.22e-3 and 2.18e-3, respectively; Figure 3), compared with those at a low cell density (∼10⁶ CFU/ml).

Interaction of R. solanacearum strain GMI1000 and fungal pathogen FOC strain XJZ2 were examined by spotting them side by side in the same plate as shown in Figure 4. After 8 days of co-culture, distinct zones of fungal inhibition were formed in the area containing R. solanacearum wide-type strain GMI1000. We also detected the diameter of XJZ2 colony co-cultured with GMI1000 and its derivatives to quantify the inhibition effect, result demonstrated that the XJZ2 colony co-cultured with wide-type strain GMI1000 with the smallest diameter. In comparison to GMI1000, the reduction of phcB deletion mutant inhibition effect reached significant level (Welch t-test, p-value = 0.02; Figure 5). In addition, the chlamydospores of strain XJZ2 were routinely found in the interaction zones with wide type strain GMI1000 Supplementary Information File 2 (SI 2), whereas in the area containing phcB deletion mutant ΔphcB, chlamydospores were hardly found. Complementation of the mutant ΔphcB with a wild-type phcB gene restored the bacterial induction ability on chlamydospore formation Supplementary Information File 2 (SI 2). When the mutant ΔrmyA, or ΔrmyB, or the double deletion mutant ΔrmyAB was co-cultured with the fungal strain XJZ2, the chlamydospore was also hardly found within their interaction zones Supplementary Information File 2 (SI 2).

Distinct inhibition zones were also found between wide-type strain GMI1000 and the sugarcane fungal pathogen S. scitamineum (Figure 6). The colony of strain GMI1000 inoculated 1 day earlier generated a more obvious inhibition zones than the bacterial colony inoculated at the same time with the fungal strain, suggesting a clear dosage effect of the inhibitory compound(s) produced by strain GMI1000. The mutant ΔphcB could also inhibit the growth of S. scitamineum, but was weaker than the wide-type. In contrast, deletion of rmyA, rmyB, or rmyAB completely abolished the growth inhibition ability on S. scitamineum, as the hyphe of S. scitamineum could grow around or cover the colonies of these rmy gene deletion mutants, and the E. coli as a control showed no inhibition effect on the fungi.