Endophytic bacterium strain Pl7 was previously isolated from the bark of apple (variety named “Kitanosach”) branch at Zhengzhou, Henan province, China in July 2021 by Yuan et al. (2022a) and cultured on LB solid medium (Peptone 10 gL⁻¹, Yeast extract 5 gL⁻¹, Sodium chloride 10 gL⁻¹, Agar 15 gL⁻¹) at 28°C.

CBZ-sensitive B. dothidea strain Bd220 (β-tubulin Genbank accession number: MZ197998) and CBZ-resistant B. dothidea strain Bd7 (β-tubulin Genbank accession number: MZ197999) were used in this study and cultured on potato dextrose agar (PDA) (Potato extracts 200 gL⁻¹, Glucose 20 gL⁻¹, Agar 15 gL⁻¹) in Petri plates at 25°C (Wang et al., 2022).

The dual culture assay based on a previous study (Yuan et al., 2022a) was used to test the antagonistic activity of strain Pl7 against pathogenic pathogens in vitro. A mycelial plug of the pathogen (B. dothidea) with 0.5 cm in diameter was inserted in the center of the Petri dish after 3 μl of overnight cultured-strain Pl7 [1 × 10⁸ colony forming units (CFU) mL⁻¹] were seeded on PDA medium on either side (2 cm from the center). Mycelial colony diameter was recorded at 25°C, 6 days post inoculation (dpi). The PDA medium only inoculated with mycelial plug was used as control (CK). The experiment was performed three times, and each time had three replicates. Inhibition = (colony diameter of CK − colony diameter of treated)/(colony diameter of CK) × 100%. Hyphal morphological characteristics were observed under an ultra-depth three-dimensional microscope (KEYENCE, Japan) at 2 days after dual culture. The experiment was performed three times, and each time examined at least 10 hyphae.

A prior study was used to assess the physiological and biochemical characteristics of strain Pl7 (Holt et al., 1994). This strain was molecularly identified based on 16S rDNA, gyrA, and rpoB sequencing with primers (Supplementary Table S1; Yuan et al., 2022a). The PCR amplification procedure was performed as described previously (Yuan et al., 2022a). The PCR results were sequenced by Sangon Biotech Co., Ltd., Shanghai, China, and the obtained sequences were compared to the NCBI nucleotide database to get the homologous sequences. The MEGA 7.0 software was used to generate phylogenetic trees using the maxium likelihood method with 1,000 bootstrap iterations.

Following a 3 days culture in LB broth medium (Peptone 10 gL⁻¹, Yeast extract 5 gL⁻¹, Sodium chloride 10 gL⁻¹) at 28°C, 200 rpm/min, the samples were run through a 0.22 μm pore size filter to remove bacteria for getting the culture filtrate. The resulting culture filtrate was mixed with PDA medium at the highest culture filtrate concentrations of 5% or 10% (v/v). A B. dothidea plug with 0.5 cm in diameter was inoculated on each plate. The diameter of the mycelial colony was measured at 25°C, 6 dpi. The experiment was performed three times, and each time had three replicates.

Hydrolytic enzyme activity of B. halotolerans strain Pl7, including β-1, 3-glucanase, cellulase, and protease activity, was assessed by incubating strain Pl7 on aniline blue medium, carboxymethylcellulose (CMC) medium, and skim milk medium on Petri dishes (Yuan et al., 2022a). After 3 dpi at 28°C, we observed whether there was a transparent circle around the bacterial colony.

Total DNA was isolated from overnight cultured-strain Pl7 with DNA extraction kit (Solarbio, Beijing, China) according to the manufacturer’ procedure. Target antibiotic genes and primers were listed in Supplementary Table S1 (Lee et al., 2016; Kim et al., 2017). The PCR amplification procedure was performed as described previously (Lee et al., 2016; Kim et al., 2017). The PCR products were sequenced by Sangon Biotech Co., Ltd., Shanghai, China.

B. dothidea was inoculated into a potato dextrose broth (PDB) (Potato extracts 200 gL⁻¹, Glucose 20 gL⁻¹) medium for 2 days at 180 rpm/min, 25°C. Afterwards, B. dothidea hyphae were collected and treated with culture filtrate of B. halotolerans strain Pl7 for 24 h. LB broth medium treatment was used as CK. Hyphae were mixed with SYTOX green (1 μM; Biomart, Beijing, China) for 5 min in the dark. After staining, hyphae were rinsed five times with sterile water, and examined by using a Leica Laser Confocal Microscopy (TCS SP5, Germany) under 488 nm excitation and 538 nm emission.

The antifungal activity of B. halotolerans strain Pl7 to control B. dothidea growth in apple (Fuji) fruit was evaluated based on a method reported previously (Yuan et al., 2022a). Each apple (Fuji) was wounded twice with a sterile borer to produce the hole with 0.3 cm deep and 0.5 cm wide. Each hole was inoculated with 40 μl of strain Pl7 cell suspension (1 × 10⁸ CFU mL⁻¹) with pipette. The negative control was sterile water treatment, while the positive control was the fungicide CBZ (0.8 gL⁻¹; Jiangsu Longdeng Chemical Co., Ltd. Jiangsu, China) treatment with the same procedure as strain Pl7 treatment. After dry, fruit wounds were then inoculated with B. dothidea mycelial plugs (diameter: 0.5 cm) from the edge of 5-day-old colony. After that, the fruit were kept at 25°C with an 85% relative humidity in a climate box. At 3 and 5 dpi, a Vernier caliper was used to measure the length of the disease lesion. Disease incidence = the number of diseased sites/the total inoculation sites × 100%. This experiment was repeated three independent times and each time contained five fruits with 10 inoculation sites.

Healthy and equal size apple (Fuji) fruits, collected from an orchard with severe ring rot disease, were used for assay. The fruits were dipped in 1 × 10⁸ CFU mL⁻¹ cell suspension for 2 min and then naturally air dried. LB broth medium treatment was used as control. After treatment, the fruits were put in sterilized plastic boxes and kept at 20°C with an 80% relative humidity in an incubator. At the storage time points (21, 28, and 35 days), the number of rotted fruits was recorded. This assay was carried out twice, with three duplicates in each time. Each replicate consisted of 10 fruits.

B. halotolerans strain Pl7 to colonize apple fruit was assessed based on the method reported previously (Yuan et al., 2022a). As described above, apple fruits were punctured by using a 0.5-cm sterile borer, and each hole (0.3 cm deep and 0.5 cm wide) was inoculated with strain Pl7 cell suspension (40 μl, 1 × 10⁸ CFU mL⁻¹). Fruit tissues (around 0.6 g) were collected with a 0.7 cm core border at 0 (3 h after inoculation) to 9 dpi. After that, the tissue was pulverized with a grinder. 0.1 ml of each dilution was placed onto LB medium on Petri dishes after serial 10-fold dilution. Bacterial colonies in each Petri dishes were enumerated at 28°C, 2 dpi and converted to how many of colonies per wound. This test was carried out twice, with three duplicates in each time. For analysis, colony densities were converted as log₁₀ CFU/wound.

Healthy apples (Fuji) were dipped in cell culture of B. halotolerans strain Pl7 (1 × 10⁸ CFU mL⁻¹) for 2 min, and about 1 g of fruit epidermal tissue from three apples was taken at 2 dpi to determine changes in overall gene expression pattern in apple fruit after strain Pl7 treatment. The assay was repeated three times. As a control, the LB broth treatment was used. Total RNA was extracted by using Trizol kit based on the instructions provided (TransGen Biotech, Beijing, China). Shanghai Personalbia Gene Technology Co., Ltd. performed the transcriptomic sequencing and analysis. Genes with a fold change [Log₂ (fold change)] less than 1 and an adjusted p-value less than 0.05 were considered as differentially expressed genes (DEGs). The topGO software and the cluster profile package were used to perform gene ontology (GO) annotations and KEGG pathway enrichment analyzes, respectively (Alexa and Rahnenfuhrer, 2010; Yu et al., 2012).

The qRT-PCR experiment was used to verify the accuracy of the RNA-seq data. Apple fruit samples were prepared and collected in the same manner as previously indicated for total RNA extraction. First-stand cDNA was created based on the instructions provided (Monad, Suzhou, China). The primers mentioned in Supplementary Table S1 were used in a qRT-PCR experiment conducted on a Roche PCR apparatus with a kit (Monad, Suzhou, China). The reference gene was elongation factor 1-a (EF1a). Six randomly selected DEGs were analyzed based on the 2^−ΔΔCt technique (Livak and Schmittgen, 2001). Three replicates were used in each of the two assay sets.

Freshly picked apples (Fuji) were dipped in strain Pl7 cell suspension (1 × 10⁸ CFU mL⁻¹) for 2 min and then stored at 20°C in a transparent plastic in a incubator to evaluate the impact of strain Pl7 on the postharvest quality of apples. LB-treated apples were used as the CK. According to the previous study (Yuan et al., 2022a), four characteristics were assessed: soluble sugar content, titratable acidity, ascorbic acid content, and fruit hardness. The experiment was repeated twice and each test consisted of three replicates, with three fruit samples in each replicate.

All data were analyzed by using DPS package. Following a Shapiro–Wilk test for normality, one-way ANOVA with Duncan’s multiple range test or Student’s t-test (p < 0.05) was used to determine significant differences between CK and treatments. All data are expressed as mean ± standard deviation (SD) of at least three biological replicates.

The dual culture assay showed that strain Pl7 had substantial antagonistic action against B. dothidea, inhibiting about 69% of mycelial development, with a 0.57 cm inhibitory zone (Figure 1A). Additional findings showed that strain Pl7-treated hyphae were unusually stretched and distorted with tenuous and ruptured compared to control hyphae (Figure 1B). These findings demonstrated that, collectively, strain Pl7 had strong antifungal activity against B. dothidea in vitro.

Supplementary Table S2 highlighted the biochemical and physiological characteristics of strain Pl7, revealing that it belonged to the genus Bacillus. When combined with closely comparable sequences, phylogenetic trees based on the 16S rDNA (accession number: OP692716), gyrA (accession number: OP712701), or rpoB (accession number: OP712700) sequences of strain Pl7 demonstrated that the strain Pl7 was more closely related to B. halotolerans (16s rDNA accession number: CP029364.1, MK509936.1; gyrA accession number: CP041357.1, CP054584.1; rpoB accession number: CP041357.1, CP054584.1; Figures 2A–C). As a result, strain Pl7 was identified as B. halotolerans.

B. dothidea mycelial plugs were inoculated on a PDA medium containing 5 or 10% of the culture filtrate of B. halotolerans strain Pl7 to examine if it would exhibit antagonistic activity. The outcome showed that strain Pl7 culture filtrate greatly reduced the mycelial growth of B. dothidea (Figures 3A,B). The inhibition by 5 and 10% strain Pl7 culture filtrate, respectively, was 39 and 53% (Figure 3C). This result suggested that strain Pl7 culture filtrate had inhibition activity toward B. dothidea.

The results revealed that strain Pl7 formed distinct translucent circles on different medium (Figures 4A–C), indicating protease, cellulase, and β-1, 3-glucanase activity in strain Pl7 produced enzymes or metabolites.

Lipopeptide is one of the major antimicrobial compounds secreted by Bacillus species. The result of PCR amplification showed that expected size of PCR products associated with ituD, srf, fen, and ituC were amplified from strain Pl7 (Figure 5). Further sequencing analysis confirmed the result that strain Pl7 harbored gene clusters involved in the synthesis of iturin, surfactin, and fengycin.

SYTOX green staining was applied to determine whether B. halotolerans strain Pl7 affected the permeabilization of B. dothidea cell plasma membrane. The result of GFP fluorescence observation showed that no obvious GFP fluorescent signal was observed in CK-treated B. dothidea hyphae (Figure 6). In contrast, a bright GFP fluorescent signal was detected in hyphae after B. halotolerans strain Pl7 culture filtrate treatment (Figure 6). These results indicated that culture filtrate of B. halotolerans strain Pl7 could cause cell membrane permeabilization of B. dothidea.

Figure 7A showed that strain Pl7 treatment significantly suppressed the disease development of apple ring rot caused by B. dothidea, as compared with the CK treatment. CBZ treatment as a positive control remained disease-free (Figure 7A). Even though there was no discernable difference in disease incidence between the strain and CK treatments (Figure 7B), the average lesion diameter of fruit treated with strain Pl7 was 0.67 cm and 1.16 cm at 3 or 5 dpi, which was much smaller than CK (Figure 7C). Therefore, B. halotolerans strain Pl7 treatment effectively inhibited the development of ring rot in the apple fruit.

We also tested the capacity of B. halotolerans strain Pl7 to prevent the growth of ring rot in apple fruit caused by a CBZ-resistant isolate. Results indicated that strain Pl7 cell suspension treatment dramatically reduced the average lesion diameter in apple fruit infected with the CBZ-resistant isolate (Figure 8). CBZ treatment did not inhibit CBZ-resistant isolate-caused apple ring rot (Figure 8). At 3 dpi or 5 dpi, the average lesion diameter of CK was 1.28 cm or 2.07 cm, while the strain Pl7-treated fruits was significantly lower (0.42 cm or 1.07 cm; Figure 8C). These data revealed that B. halotolerans strain Pl7 was similarly effective in preventing CBZ-resistant isolate-caused apple ring rot.

During apple storage period, the incidence of apple decay was significantly reduced after strain Pl7 treatment, as compared to CK (Figure 9). At 21 days storage period, the disease incidence of strain Pl7 treated-apple fruits was 27%, which was significantly lower than CK (50%). At 35 days storage period, the disease incidence of CK reached to 77%, while the Pl7 treated-apple fruits was still lower with 43% (Figure 9). These results indicated that strain Pl7 was effectively inhibited apple fruit natural decay.

At 1 dpi, there were 8.57 × 10⁶ CFU of strain Pl7 colonies, which is a 29.54-fold increase from 2.90 × 10⁶ CFU at 0 dpi (Figure 10). At 5 dpi, the number of strain Pl7 colonies reached the peak with the population 3.54 × 10⁸ CFU (Figure 10). After that, the colony number slightly decreased but remained stable (Figure 10). At 9 dpi, strain Pl7 colony number still kept high (2.74 × 10⁸ CFU; Figure 10). These findings suggested that strain Pl7 might successfully colonize apple fruit wounds.

A transcriptome investigation of apple fruit inoculation with B. halotolerans strain Pl7 was also carried out to assess overall apple gene expression changes. 489 and 884 DEGs were upregulated and downregulated compared to control samples, respectively (Figure 11A; Supplementary Figure S1). The qRT-PCR result revealed a similar expression pattern of 6 randomly selected genes with RNA-seq (Supplementary Figure S2). This demonstrated the reliability of the transcriptome data used in this study. According to GO analyzes, these DEGs were related to cellular components (CC), especially the extracellular region, cell wall, and cell periphery, biological processes (BP) primarily including oxidation–reduction process, photosynthesis, light reaction, and photosynthesis, molecular functions (MP) primarily including xyloglucosyl transferase activity, oxidoreductase activity and copper ion binding (Figure 11B). Further, KEGG enrichment studies showed that these DEGs were mainly enriched in plant secondary metabolite biosynthesis pathways, such as biosynthesis of various secondary metabolites, flavonoid biosynthesis, riboflavin metabolism, diterpenoid biosynthesis, phenylpropanoid biosynthesis, isoquinoline alkaloid biosynthesis (Figure 11C). In addition, DEGs were also enriched in plant-pathogen interaction (Figure 11C).

Following treatment with strain Pl7, additional analysis showed that 37 DEGs associated with plant secondary metabolism were elevated to varying degrees in apple fruit (Supplementary Table S3). Several potential defense-related genes, such as pathogen-related genes and receptor protein genes, were also greatly induced in apple fruit treated with strain Pl7 (Supplementary Table S4). Taken together, B. halotolerans strain Pl7 enhanced apple resistance by inducing genes involved in plant secondary metabolite biosynthesis and plant-pathogen interaction.

Assays for fruit firmness, ascorbic acid, soluble sugar, and titratable acid revealed no significant differences between apple fruit treated with strain Pl7 and those from the CK group (Table 1). These findings suggested that B. halotolerans strain Pl7 had no discernible impact on the apple postharvest quality.