Apple (cv. ‘Fuji’) fruit used in this study was collected from the local orchards in Shandong, China. The fruit at a commercial mature stage has not received any postharvest treatment and was kept at 4°C before being used. The apples were disinfected and washed according to the method of Sun et al. (2021). P. syringae B-1, isolated from the healthy apple fruit, was preserved in China Center for Culture Collection (M2015813). The bacteria were cultured in nutrient broth (NB) on a shaker at 28°C for 48 h. B. dothidea LXS030101, isolated from apple fruit with ring rot symptoms, was characterized by our research team based on morphological observation and molecular identification. The strain was cultured on potato dextrose agar (PDA) for routine use (Zhang Q. M. et al., 2016). The conidia of strain LXS030101 was induced with the modified method of Leng et al. (2009). The wounded young apple fruit was inoculated with mycelial plugs of B. dothidea and incubated under UV light (365 nm) at 25°C. After about 2 weeks, the conidia was produced and collected from the lesion around the inoculation sites.

Each apple fruit was punctured with a borer to make three wounds (5 mm wide and 3 mm deep) at the equatorial region. Then, the strain B-1 suspension at different concentrations (1 × 10⁵, 1 × 10⁶, 1 × 10⁷, 1 × 10⁸, and 1 × 10⁹ CFU mL^–1) was dripped into the wounds. Following 48 h of incubation at 25°C, the conidia suspension of B. dothidea (5 × 10⁵ spores mL^–1) was added to the treated wounds with the pipette (Lu et al., 2013; Huang et al., 2021a). All fruits were placed into enclosed plastic boxes and stored at 25°C in a relative humidity of about 95%. Thirty fruits were selected as controls and were treated with sterile water and inoculated only with the pathogen suspension. For each treatment, there were three replicates with 10 apple fruits in each replicate. Disease incidence was calculated as the percentage of infected wounds, and the lesion size around the wounds was recorded at 3 and 5 days post-inoculation (dpi).

The effect of interval time after P. syringae B-1 treatment on control efficiency was determined according to the assay mentioned in Section “2.2. Efficiency of Pseudomonas syringae B-1 in vivo” A volume of 30 μL of strain B-1 suspension at 1 × 10⁸ CFU mL^–1 was added into each wound using a pipette. After 0, 6, 12, 24, 48, and 96 h of incubation at 25°C, respectively, 30 μL of B. dothidea spore suspension at 5 × 10⁵ spores mL^–1 was inoculated to the treated wound. The fruits were incubated at a relative humidity of about 95% at 25°C for 5 days. The diseased symptoms were observed; the disease incidence and lesion diameter were recorded at 5 dpi. The assay was conducted two times, and there were three replicates with 10 apple fruit in each replicate.

Strain B-1 cells were obtained from 2-day-old cultures in NB by centrifuging at 12,000 × g for 15 min and re-suspended in distilled water. The culture filtrates were prepared by a 0.22-μm polycarbonate membrane filter (Sangon Biotech, Shanghai, China). For the inhibiting mycelium growth assay, PDA plates were prepared, which contained 10% (v/v) culture filtrates or different concentrations of strain B-1 cells (1 × 10⁵, 1 × 10⁶, 1 × 10⁷, 1 × 10⁸, and 1 × 10⁹ CFU mL^–1). The mycelial plugs (5 mm in diameter) containing B. dothidea grown on medium for 3 days were transferred to the PDA plates. After incubating for 5 days at 25°C, the colony diameter of B. dothidea was determined and the inhibition rate was calculated. For the spore germination assay, the culture filtrates or different concentrations of strain B-1 cells and the pathogen suspension were dripped into the concavity slides (Zhang Q. M. et al., 2016). Following 12 h of incubation at 25°C, the spore germination was observed using the microscope (DM2500, Leica, German). Conidia were considered germinated when the length of the germ tube was longer than the length of the conidia. For each treatment, sterile distilled water was used as the control. The assays were carried out two times with four replicates.

Four treatments were performed in apple fruit including control and sterile water; B-1, P. syringae B-1 suspension at 1 × 10⁸ CFU mL^–1; pathogen, B. dothidea suspension at 5 × 10⁵ spores mL^–1; B-1 + pathogen, strain B-1 treatment, followed by B. dothidea inoculation. Fruit tissues were collected in different treatments at 0, 12, 24, 48, 72, 96, and 120 h. To reveal the oxidative damage in apple fruit, electrolyte leakage, malondialdehyde (MDA), and hydrogen peroxide (H₂O₂) contents were analyzed. For electrolyte leakage, it was measured according to the modified method of Ji et al. (2020). A total of 0.5 g of fresh tissues were cut into small pieces and dropped into 10 mL of deionized water. After 1 h incubation at room temperature, the conductivity (C1) was determined by a conductivity meter (DDBJ-350, Inesa, China). The tissue samples were then boiled for 15 min and the conductivity (C2) was measured. The formula (C1/C2) × 100% was used to calculate the electrolyte leakage.

For MDA content, it was conducted with the assay kit (#A003) from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). For the analysis of H₂O₂ content, 1 g fresh tissues were homogenized with 3 mL of chilled 0.2% trichloroacetic acid (≥99.0%, Sangon Biotech, Shanghai, China), and then centrifuged at 12,000 × g for 15 min at 4°C. According to the report of Wang et al. (2015), the supernatant was mixed with 0.1 mmol L^–1 phosphate buffer and 1 mol L^–1 potassium iodide (≥99.0%, Sangon Biotech, Shanghai, China), then the absorbance of the reaction system was measured at 390 nm with a spectrophotometer (Philes, T6, China). The units of MDA and H₂O₂ contents were expressed as mmol in 1 kg of fresh weight (mmol kg^–1).

Apple fruit was treated with P. syringae B-1 or sterile water according to the method described in Section “2.5. Estimation of oxidative damage” Fruit tissues were collected near the wound at different time points and were ground to powder with liquid nitrogen. For the antioxidant enzymes, activities of catalase (CAT, #A007), peroxidase (POD, #A084), superoxide dismutase (SOD, #A001), and ascorbate peroxidase (APX, #A123) were measured with the assay kits from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). For the defense-related enzymes, the activities of phenylalanine ammonialyase (PAL), polyphenoloxidase (PPO), β-1,3-glucanase (GLU), and chitinase (CHI) were assayed referring to the report of Zhang Q. M. et al. (2016). The enzyme activity units are expressed on the fresh weight basis as U g^–1.

The total phenolics content was measured with the assay kit (#A143) from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). The lignin content was assayed as described by Li et al. (2019), and the absorbance of the reaction mixture was detected at 280 nm. The lignin content was expressed as A₂₈₀ g^–1 on a fresh weight basis. Quantification of hormones including salicylic acid (SA) and jasmonic acid (JA) in apple fruit was analyzed using the enzyme-linked immunosorbent assay (ELISA) kit from Jingmei Biotechnology (Yancheng, China). SA and JA contents were expressed as μg kg^–1 and ng kg^–1 on the fresh weight basis, respectively.

Fruit tissues were collected near the wounds after being treated with P. syringae B-1 or sterile water at 0, 12, 24, and 48 h. Total RNA was extracted from different tissue samples and the cDNA synthesis was performed using the HiScript 1st Strand cDNA Synthesis kit (Vazyme, Nanjing, China). Quantitative real-time PCR (qPCR) was conducted as reported by Huang et al. (2021a), using a Light Cycler^® 96 PCR Detection System (Roche, Germany). Each reaction consists of at least three biological replicates, with a 25 μL reaction volume. The transcript levels of six genes involved in the defense response and the SA pathway (MdPR1, MdPR5, MdGLU, MdCHI, MdSIT2, and MdPAD4) were analyzed with the 2^–ΔΔCT method (Zhang Q. M. et al., 2016; Huang et al., 2021a). Referenced gene MdEF1a (elongation factor 1-a) was used to normalize the expression levels of target genes. The primers for the qPCR assay are shown in Table 1.

Data analysis was performed using the software SPSS Version 19.0. All the data consisted of at least three replicates and were represented as mean ± standard deviation (SD). Statistical differences were conducted by analysis of variance using Duncan’s multiple range test at a significance level of 0.05.

The biocontrol efficiency of P. syringae B-1 to control B. dothidea rot in apple fruit was influenced by its concentrations (Figure 1). As shown in Figure 1A, the control fruit showed rapidly developing brown lesions around the inoculation sites. In contrast, although strain B-1 at 1 × 10⁵ CFU mL^–1 did not inhibit fruit decay caused by B. dothidea in fruit (Figures 1B, C), strain B-1 at 1 × 10⁶ CFU mL^–1 to 1 × 10⁹ CFU mL^–1 all resulted in a decrease in disease incidence (Figure 1B) and lesion diameter (Figure 1C). When the concentrations of strain B-1 were higher (>10⁸ CFU mL^–1), their efficiency against B. dothidea in vivo had no significant difference. Following storage for 5 days at 25°C, strain B-1 at 1 × 10⁸ CFU mL^–1 reduced the incidence of apple ring rot and lesion diameter by 41.2 and 90.2%, respectively, compared with the control fruit.

In comparison to the 0 h treatment, the diameter and incidence of B. dothidea rot reduced when the interval time between P. syringae B-1 treatment and the pathogen inoculation was 6 h or longer (Figure 2). The efficiency of strain B-1 was improved with the extension of interval time from 6 to 24 h. The expansion of decay symptoms was inhibited at the interval time of 24 h, as the incidence of apple ring rot and lesion diameter in strain B-1 treated fruit were reduced by 30.8 and 85.6%, respectively, compared with the 0 h treatment (Figure 2). There was no obvious difference in the incidence and lesion size of apple ring rot at the interval time from 24 to 96 h (Figures 2B, C).

Pseudomonas syringae B-1 cells ranging from 1 × 10⁶ CFU mL^–1 to 1 × 10⁹ CFU mL^–1 and culture filtrates all exhibited inhibitory effects on the mycelial growth of the fungal pathogen at 25°C. When the concentration of strain B-1 cells reached 1 × 10⁸ CFU mL^–1, the mycelial growth of B. dothidea was almost completely inhibited. Moreover, strain B-1 culture filtrates showed lower antifungal activity against B. dothidea than the bacterial cells at 1 × 10⁶ CFU mL^–1, with 18.28 and 49.64% mycelial growth inhibition, respectively (Figure 3A).

The effect of P. syringae B-1 on B. dothidea spore germination was also determined by mixing strain B-1 cells or culture filtrates with B. dothidea conidia. As shown in Figure 3B, the spore germination rate decreased as the concentration of strain B-1 cells increased, which indicated that the inhibitory efficiency of strain B-1 cells against the spore germination of B. dothidea depended on its concentration. At 12 h of incubation, the germination rate of control reached 88.07%; however, the bacterial cells at 1 × 10⁹ CFU mL^–1 showed the lowest germination rate with only 1.23%. Similarly, P. syringae B-1 culture filtrate showed less inhibition on spore germination than its cells ranging from 1 × 10⁶ CFU mL^–1 to 1 × 10⁹ CFU mL^–1 (Figure 3B).

To analyze the effect of P. syringae B-1 on oxidative damage in apple fruit, the electrolyte leakage, MDA, and H₂O₂ contents were determined in four different treatments (Figure 4). During the storage, no statistical differences in the three factors were detected between strain B-1 treated fruit and the control. The level of electrolyte leakage rose in the fruit inoculated with B. dothidea from 24 h post-inoculation (hpi) and attained the maximum leakage of 97.3% at 120 hpi. However, the level of leakage in strain B-1 plus B. dothidea-treated fruit was only 45.7% at 120 hpi, which had no difference compared with the control fruit (Figure 4A).

In the fruit only inoculated with B. dothidea, MDA content increased from 24 hpi, which was up to 164.9 and 173.1% more than the control at 96 and 120 hpi, respectively (Figure 4B). In strain B-1 plus B. dothidea-treated fruit, although the MDA content rose from 96 hpi, it was only 16.4 and 31.6% more than the control at 96 and 120 hpi, respectively. Consistent with the level of electrolyte leakage and MDA content, the H₂O₂ level sharply increased in B. dothidea-inoculated fruit from 48 hpi and reached the highest value of 14.08 mmol kg^–1 at 120 hpi. Nevertheless, in the fruit treated with strain B-1 plus pathogen, the level of H₂O₂ enhanced slowly and showed no difference compared with the control until 96 hpi, with only 0.90 mmol kg^–1 at 120 hpi (Figure 4C).

To reveal the effect of P. syringae B-1 on the apple fruit, the activities of the four antioxidant enzymes were analyzed (Figure 5). During storage, POD activity in apple fruit varied as time progressed (Figure 5A). An ascending tendency of POD activity was observed in fruit treated with P. syringae B-1 from 24 h and the peaked was at 48 h with 2.21-fold higher than the control. POD activity in strain B-1 treated fruit decreased from 72 h; however, it remained higher than the control during the whole storage. The result of Figure 5B showed that CAT activity was affected by P. syringae B-1 treatment. In strain B-1 treated fruit, the activity of CAT began to rapidly increase from 24 h and reached the peak at 48 h of storage, which was more than 2.85-fold higher compared with the control. Then, the enzyme activity in strain B-1 treated fruit decreased until 72 h, but CAT activity quickly rosed again and the second peak was at 96 h, which was 2.41-fold higher than the control.

In contrast to the control, P. syringae B-1 treatment also increased the activities of SOD and APX in apple fruit during storage (Figures 5C, D). The activity of SOD in fruit treated with strain B-1 rose steadily from 12 h and reached the peak at 72 h with 5.79 U g^–1. Although the enzyme activity declined from 96 h, SOD activity induced by strain B-1 was still higher than the control throughout the storage period (Figure 5C). As shown in Figure 5D, APX activity was induced by stain B-1 from 12 h and peaked at 48 h, which was 2.09-fold higher than the control. Then, the enzyme activity declined rapidly and it even reduced to the control level at 120 h.

In addition to an enhancement of antioxidant enzyme activities by P. syringae B-1 treatment, Figure 6 shows that the four defense-related enzyme activities were induced by P. syringae B-1 in apple fruit. For PAL and PPO, the activities increased from 12 h in the fruit treated with strain B-1 and they reached the highest values, 1.57-fold and 2.47-fold higher than the control, respectively. After that, the two enzymes’ activities induced by strain B-1 decreased quickly, and they were not different from the control at 96 h for PAL and 120 h for PPO, respectively (Figures 6A, B).

Figures 6C, D show that GLU and CHI activities were affected by P. syringae B-1 treatment in apple fruit. The GLU activity in fruit treated with strain B-1 showed a sharp increase and peaked at 48 h. Thereafter, the enzyme activity decreased from 72 to 120 h; however, it remained higher than the control during the storage phase (Figure 6C). The change of CHI activity induced by strain B-1 was similar to GLU activity. The enzyme activity rose within 24 h after strain B-1 treatment and reached the highest level at 72 h. Then, CHI activity declined from 96 h after treatment, and it was still 2.19-fold higher than the control at 120 h (Figure 6D).

To get a better insight into the impact of P. syringae B-1 on apple fruit and the mechanisms of induced protection against B. dothidea, several metabolites and hormones involved in pathogen resistance were measured. Generally, the contents of total phenolics and lignin contents changed slowly in the control fruit, whereas they were greatly induced in strain B-1 treated fruit. As indicated in Figure 7A, strain B-1 enhanced the accumulation of total phenolics from 24 h (1.31-fold) after treatment during the storage period, and reached the maximum level at 72 h with 1.52-fold higher than the control. In addition, strain B-1 promoted the lignin content from 24 h after treatment compared with the control. Then, the lignin content in strain B-1 treated fruit peaked at 48, with 1.71-fold higher than the control. After that, it decreased only slightly and maintained a higher level throughout the storage (Figure 7B).

Regarding plant hormones, the contents of SA and JA were determined, with or without strain B-1 treatment. During the storage, SA and JA levels had no obvious changes in the control fruit (Figure 8). Strain B-1 treatment slightly enhanced the accumulation of JA at 48 h in apple fruit, with a 1.20-fold higher compared with the control. Nevertheless, the result of Figure 8B illustrated that strain B-1 treatment enhanced the SA levels compared with the control. The content of SA increased from 12 h and continued to accumulate, which was 4.24-fold higher than the control.

Given the increased SA content in strain B-1-treated fruit, we next determined the expressions of genes involved in defense response (MdGLU and MdCHI) and SA pathway (MdPR1, MdPR5, MdSID2, and MdPAD4) by quantitative PCR to reveal the biocontrol mechanism of P. syringae B-1. PR1 and PR5 are widely used molecular markers that correlate with SA signaling activation (Zhang Q. M. et al., 2016). Meanwhile, SID2 (SA induction deficient 2) and PAD4 (phytoalexin deficient 4) are the genes related to SA biosynthesis (Huang et al., 2021b).

Compared with the control, the transcript level of MdPR1 in strain B-1 treated fruit upregulated significantly at 12 h, and then reached the maximum at 48 h, which was 27.50-fold higher than that in the control fruit (Figure 9A). As shown in Figure 9B, the transcript level of MdPR5 in fruit treated with strain B-1 peaked the highest at 24 h, with 9.60-fold higher than the control. Similarly, strain B-1 treatment upregulated the expression of MdSID2 and MdGLU at 12–48 h, which exhibited similar trends with MdPR5 (Figures 9C, E). The result of Figure 9F indicated that the expression of MdPAD4 was triggered by strain B-1, and its transcript level was 7.63-fold higher than the control at 48 h. For MdCHI, compared with the control, the transcript level was activated by strain B-1 and showed 5.15- and 8.28-fold increases at 24 and 48 h, respectively (Figure 9D).