Anneal plant samples were randomly collected from multiple locations within Jeju Special Self-Governing Province. Rhizosphere soil of the plants was obtained, and 1 g of root hair was weighted using a chemical balance. The root samples were cut into small pieces and homogenized with 10 mL sterilized H₂O in a mortar. The root suspension was filtered and serially diluted in sterilized H₂O to a final dilution of 10⁻⁴. Each 30 μL from the dilutions were spread on tryptic soy agar (TSA) medium and incubated at 25°C for 48 h. The bacterial colonies exhibiting distinct morphological characteristics such as shape, size, color etc. were separately selected and re-incubated on the TSA medium. The purified bacterial strains were obtained by three-separation method on TSA medium and stored at −80°C until used.

The antibacterial activity of the rhizobacterial strains was evaluated in vitro using artificial medium. Erwinia pyrifoliae (KACC No. 13945) was obtained from the Microbial Bank of the National Institute of Agricultural Sciences. Concentration of E. pyrifoliae suspension was adjusted to 2 × 10⁸ cfu/mL and 30 μL was spread onto TSA plates. Sterilized paper discs (Advantec^®; Toyo Roshi Kaisha, Ltd., Toyo, Japan) with a diameter of 8 mm were placed at the center of the plate. The rhizobacterial strains were applied to the discs at a volume of 50 μL. As control sterilized H₂O and 0.01% streptomycin, which is a recommend concentration to the farm, were treated instead of the rhizobacterial strains. The plates were incubated at 25°C for 48 h, after which the inhibition zones were visually assessed using a ruler.

The three-year-old apple trees (variety Fuji) were used. The leaves of the new shoots were cut with scissors and sterilized by treatment with 1% sodium hypochlorite for 20 s and 70% alcohol for 20 s. Subsequently they were washed three times with sterilized water and dried before used. As inoculum, mixtures of E. pyrifoliae and each rhizobacterial strain were used. The final concentration of the mixtures was adjusted with 1 × 10⁷ for E. pyrifoliae and 1 × 10⁸ cfu/mL for each rhizobacterial strain, respectively. The petiole of the apple leaves was injected with 20 μL of each mixture using a syringe (Becton Dickinson, Franklin Lakes, NJ, USA). As controls, a suspension of E. pyrifoliae and a mixture of 0.01% streptomycin with E. pyrifoliae were used. The inoculated leaves were placed in a petri dish maintaining 100% relative humidity and incubated at 27°C under light conditions for 5 days. Disease severity was assessed as (infected length/total leaf length) × 100 (%), and the protection rate was calculated as (1 − [disease severity of treated / disease severity of untreated control]) × 100 (%).

The inoculated apple leaves including the petiole were cut in size 0.5 × 0.5 cm² using a sterile scalpel. The leaf segment was transferred into a microtube containing 1 mL of sterile water and vortexed for 5 min. The culture broth was centrifuged at 20,238 × g for 5 min at room temperature. After removal of the supernatant, the pellet was resuspended in 1 ml of the sterile DNA-free water (Biosoultion, Seoul, Korea) and stored at −20°C until used.

The number of E. pyrifoliae cells in the inoculated apple leaves was analyzed using quantitative PCR. SYBR Green-Based qPCR was used to perform the analyses. Genomic DNA (5 ng) of E. pyrifoliae, suspension of E. pyrifoliae, and DNA-free sterile water were used as controls. The E. pyrifoliae-specific primer pair RS14185_320F (5′-GGG GCT ACA AGT CGC AAA GAT A-3′) and RS14185_320R (5′-TAA GCC CAT TCG AAC CC AGA C-3′) (Jin et al., 2022) were used. Each qPCR reaction consisted of 7 μL sterile distilled water, 1 μL each of forward and reverse primers (10 pmol/mL), 10 μl of 2× SensiFAST SYBR No-ROX mix (Meridian Bioscience, Cincinnati, OH, USA), and 1 μL of template DNA. Amplification was performed the following cycling conditions: initial denaturation at 95°C for 30 s; 40 cycles of denaturation at 95°C for 5 s and annealing/extension at 61°C for 30 s; followed by a final extension at 95°C for 15 s. Melting curve analysis was conducted by increasing the temperature from 65°C to 95°C at 0.5°C increments every 5 s. A standard curve was generated using serial dilutions of E. pyrifoliae suspensions. Amplification efficiency was derived from the slope of the standard curve using the formula E = 10^(−1/slope), and efficiency was expressed as a percentage using. For leaf samples, qPCR analysis was conducted at 1, 48, and 96 h after inoculation. A logarithmic standard curve was established using E. pyrifoliae suspensions. The obtained threshold cycle (Cq) values were compared with the standard curve to determine the log-transformed concentration of E. pyrifoliae.

Rhizobacterial strains showing direct antibacterial effect and suppressing disease severity were cultured in TSB at 25°C for 48 h with shaking at 130 rpm. Genomic DNA was extracted using the DNeasy Blood & Tissue Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The 16S rDNA gene was amplified using the universal primers 27F (5′-AGA GTT TGA TCC TGG CTC AG-3′, 10 pmol/mL) and 1492R (5′-TAC GGY TAC CTT GTT ACG ACT T-3′, 10 pmol/mL). The reaction mixture consisted of 36.5 μL of SDW, 5 μL of dNTPs, 5 μL of 10× PCR buffer, 1 μL of genomic DNA, 0.5 μL of i-Taq polymerase, and 1 μL each of the primers. The amplification conditions consisted of an initial denaturation at 94°C for 1 min, followed by 30 cycles of denaturation at 94°C for 30 s, annealing at 57°C for 30 s, and extension at 72°C for 1 min, with a final extension at 72°C for 8 min. PCR products were resolved on 1% agarose gels and visualized under UV illumination (Analytik Jena GmbH+Co. KG, Thuringia, Germany). The amplified fragments were purified and sequenced. Sequence identities were determined using the National Center for Biotechnology Information (NCBI) Basic Local Alignment Search Tool (BLAST) database. Phylogenetic analysis was conducted using the Neighbor-Joining method with 1,000 bootstrap replications based on the Tamura-Nei model (Tamura and Nei, 1993) in MEGA software (ver. 11.0.13).

Morphological identification of the rhizobacterial strains was carried out using a scanning electron microscope (SEM). The rhizobacterial strains were incubated on TSA medium for 2 days. The rhizobacteria was harvested with 1 ml sterile H₂O. The bacterial cells were down using a microcentrifuge (Labocene 1730R; Labogene Co., Ltd., Gimpo, Korea) and the pellet was obtained.

Each sample was fixed in 5% glutaraldehyde prepared in 1 M phosphate-buffered saline (PBS, pH 7.2) at 4°C for 2 h. The samples were then washed three times with 0.5 mM PBS buffer (pH 7.2; NaH₂PO₄/Na₂HPO₄) at 4°C for 10 min each. Post-fixation was performed with 1% osmium tetroxide (OsO₄, w/v) dissolved in PBS (pH 7.2) at 4°C for 2 h. The samples were subsequently washed twice with sterile distilled H₂O at room temperature. Dehydration was carried out through a graded ethanol series of 30%, 50%, 70%, 80%, and 90% (v/v) ethanol for 10 min each, followed by three changes of 100% (v/v) ethanol. The specimens were dried twice in 100% hexamethyldisilazane for 15 min each. The dried samples were mounted on metal stubs and coated with platinum using a sputter coater (Q150TS; Quorum Technologies, Laughton, UK). The coated specimens were observed using a field-emission scanning electron microscope (FE-SEM; Mira3, Tescan Ltd., Brno, Czech Republic).

A number of rhizobacterial strains were isolated from the rhizosphere of plants. Among them, 81 rhizobacterial strains were tested for direct antibacterial activity on TSA medium, and some of them formed inhibition zones on the plates (Supplementary Fig. 1). Three rhizobacterial strains, DR2-3, DRB1-1, and DR1-6, showing antibacterial activity were selected, purified, and their antibacterial activity was compared with that of streptomycin.

Apparent inhibition zones were formed around paper discs treated with the rhizobacterial strains on TSA plates, whereas no inhibition zone was observed on the sterile water-treated control plates (Fig. 1), indicating direct suppression of E. pyrifoliae growth by all tested rhizobacterial strains. The inhibitory effects on the growth of E. pyrifoliae differed among the rhizobacterial strains.

Strains DR2-3 and DRB1-1 inhibited E. pyrifoliae growth more strongly than 0.01% streptomycin, a commercial antibiotic. In contrast, the diameter of the inhibition zone produced by strain DR1-6 was similar to that produced by streptomycin (Fig. 1).

Severe disease symptoms developed in leaves inoculated with E. pyrifoliae alone. The main vein turned brown, the leaf blade exhibited chlorosis, and bacterial ooze was observed in some leaves (Fig. 2A). At 5 days after inoculation, disease severity reached approximately 30% (Fig. 2F).

In contrast, no visible symptoms were observed in leaves inoculated with a mixture of E. pyrifoliae and 0.01% streptomycin (Fig. 2B), resulting in 100% protection (Fig. 2F). Similarly, leaves inoculated with E. pyrifoliae and either isolate DR1-6 or DR2-3 exhibited no apparent symptoms (Fig. 2C, 2D) and showed low disease severities, corresponding to protection rates of 93% and 96%, respectively (Fig. 2F).

In contrast, inoculation with a mixture of E. pyrifoliae and stain DRB1-1 resulted in slight browning of the main vein, although the symptoms were less severe than those observed in control leaves inoculated with E. pyrifoliae alone (Fig. 2E). This treatment showed a moderate protective effect, with a protection rate of approximately 61% (Fig. 2F).

Quantitative PCR (qPCR) enabled measurement of the E. pyrifoliae population in inoculated apple leaves. The number of E. pyrifoliae cells increased markedly in leaves inoculated with E. pyrifoliae alone, reaching nearly a 1,000 fold increase at 48 h after inoculation (Fig. 3).

In contrast, in leaves inoculated with mixtures of E. pyrifoliae and either rhizobacterial strain DR1-6 or DR2-3, the population of E. pyrifoliae increased only slightly (Fig. 3). Similarly, in leaves inoculated with a mixture of E. pyrifoliae and 0.01% streptomycin, the pathogen population did not increase during the same period (Fig. 3).

Although the reduction was less pronounced, inoculation with the mixture of E. pyrifoliae and rhizobacterial strain DRB1-1 also resulted in suppression of the pathogenic bacterial population (Fig. 3).

The PCR products were visualized by electrophoresis on 1% agarose gel, in which approximately 1.5 kb fragments were amplified from the genomic DNA of all rhizobacterial strains (Fig. 4A). Analysis of the PCR products using the NCBI BLAST database revealed that rhizobacterial strain DR1-6 was identified as Pseudomonas fluorescens, DR2-3 as Serratia nematodiphila, and DRB1-1 as Paenibacillus polymyxa (Table 1). Phylogenetic analysis showed that each rhizobacterial strain clustered with reference strains of the corresponding species, consistent with the results of the NCBI BLAST analysis (Fig. 4B-4D).

The rhizobacterial strains were characterized based on morphological features, including cell shape, size, and surface texture, using SEM. Cells of P. polymyxa DRB1-1 were rod-shaped and exhibited two distinct phenotypes: one with a smooth surface and the other with a wrinkled surface (Fig. 4E). The smooth type measured approximately 0.7 μm × 2.5 μm, whereas the wrinkled type measured approximately 1.0 μm × 2.0 μm (Fig. 4E).

S. nematodiphila DR2-3, which was relatively smaller than the other two rhizobacterial strains, also exhibited rod-shaped cells with a slightly rough surface, measuring approximately 0.5-0.6 μm × 1.5 μm (Fig. 4F). P. fluorescens DR1-6 exhibited rod-shaped cells with a smooth surface, measuring approximately 0.3-0.4 μm × 2.0-2.2 μm (Fig. 4G).