All treatments with application of P. carribbica at 1 × 10⁸ cells/mL could reduce the disease incidence of Rhizopus decay of strawberries, compared with the control after 3 days of incubation at 20 °C (P < 0.05) (Figure 1). The antagonistic activity of P. carribbica was shown to be greatly enhanced through cultured in the NYDB media amended with 0.5% trehalose. The disease incidence of Rhizopus decay of strawberries treated with P. carribbica which were harvested from the media of NYDB amended with trehalose powder at 0.5% was 5%, which was significantly lower than that of the control strawberries (97.5%) and the strawberries treated with P. carribbica which were harvested from the media of NYDB (27.5%). However, the antagonistic activity of P. carribbica was not enhanced through being incubated in the NYDB media amended with 0.2% or 0.8% trehalose, or NYTB media, compared with that incubated in NYDB without trehalose.

All treatments with application of P. carribbica at 1 × 10⁸ cells/mL could reduce the disease incidence of gray mold decay of strawberries, compared with the control after 3 days of incubation at 20 °C (P < 0.05) (Figure 2). The disease incidence of gray mold decay of strawberries treated with P. carribbica which were harvested from the media of NYDB amended with trehalose at 0.2%, 0.5%, 0.8% or 1% was significantly lower than the strawberries treated with P. carribbica which were harvested from the media of NYDB (45%), especially when amended with trehalose at 0.5% (12.5%).

These results showed that P. carribbica has control efficacy to postharvest Rhizopus decay and gray mold decay of strawberries. Besides, the biological control activity of P. carribbica against Rhizopus decay and gray mold decay of strawberries was greatly enhanced when the yeast was harvested from the culture media of NYDB amended with 0.5% trehalose, compared with the case that P. carribbica harvested from NYDB without trehalose. Since that trehalose has no risks to humans and to the environment (it has been approved as a health food by FDA), and many tests approved that trehalose is non-toxicity, besides trehalose is a cheap and widely available natural resource [19], the utilization of trehalose might be an effective, safe and economic approach to enhance the biocontrol efficacy of P. carribbica.

The PPO activities of strawberries treated with P. carribbica harvested from NYDB or NYDB amended with trehalose at 0.5% or the control increased gradually, and then began to decrease (Figure 3). The PPO activities of strawberries treated with P. carribbica harvested from NYDB or NYDB amended with trehalose at 0.5% were higher than that of the control at all the storage time. There was no significant difference between the PPO activity of strawberries treated with P. carribbica harvested from NYDB amended with trehalose at 0.5% and that of the strawberries treated with P. carribbica harvested from NYDB without trehalose at 1 and 2 days. However, the PPO activity of strawberries treated with P. carribbica harvested from NYDB amended with trehalose at 0.5% was higher than that of the strawberries treated with P. carribbica harvested from NYDB without trehalose at 3 days.

The POD activities of strawberries treated with P. carribbica harvested from NYDB or NYDB amended with trehalose at 0.5% or the control gradually increased, the POD activity reached the highest peak at 2 days, and then began to decrease at 3 days (Figure 4). The POD activity of strawberries treated with P. carribbica either cultured in NYDB media amended with trehalose at 0.5% or cultured in NYDB without trehalose were higher than that of the control at 2 days after storage, and the POD activity of the strawberries treated with P. carribbica cultured in the NYDB media amended with trehalose at 0.5% were higher than that of the strawberries treated with P. carribbica cultured in NYDB without trehalose at 2 days after storage.

The β-1,3-glucanase activity of strawberries treated with P. carribbica harvested from NYDB and water increased quickly at the first day, and began to decrease at 2 days (Figure 5). However, the β-1, 3-glucanase activity of strawberries treated with P. carribbica harvested from NYDB amended with trehalose at 0.5% increased quickly at the first day and the second day, and began to decrease after 2 days. The β-1,3-glucanase activities of strawberries treated with P. carribbica harvested from NYDB and NYDB amended with trehalose at 0.5% were higher than that of the control strawberries at 1 day, and the β-1,3-glucanase activities of strawberries treated with P. carribbica harvested from NYDB amended with trehalose at 0.5% was higher than that of the control strawberries and the strawberries treated with P. carribbica harvested from NYDB at 2 days.

Polyphenoloxidase (PPO) catalyzes the oxidation of phenolics to quinines, which are more toxic to pathogens than the former [20]. Increased PPO activity is correlated with disease resistance in plants [21,22]. Peroxidases (POD) as bifunctional enzymes, can oxidize various substrates in the presence of H₂O₂, but also produce reactive oxygen species [23]. High POD activity is associated with the onset of induced resistance, which involves in several plant defence mechanisms, such as lignification and suberization [24,25]. The results showed that the PPO activity of strawberries treated with P. carribbica either cultured in the NYDB media amended with trehalose at 0.5% or cultured in NYDB without trehalose were higher than that of the control at all the storage time, and the POD activity of strawberries treated with P. carribbica either cultured in the NYDB media amended with trehalose at 0.5% or cultured in NYDB without trehalose were higher than that of the control at some storage time. What’s more, the PPO and POD activity of strawberries treated with P. carribbica cultured in the NYDB media amended with trehalose at 0.5% was higher than that of strawberries treated with P. carribbica harvested from NYDB at some of the storage time. These results suggest that treatment of P. carribbica can enhance lignification and suberization of strawberries, which are related to maintenance of integrity and vital functions of the cell, and increase the levels of antimicrobial phenolic compounds, so that result in improvement of disease resistance. Moreover, adding 0.5% trehalose in the media can enhance this activity of P. carribbica to improve disease resistance of fruits. This may be one mechanism by which adding trehalose in the culture medium enhances the biocontrol efficacy of P. carribbica to postharvest Rhizopus decay and gray mold decay of strawberries.

Our study displayed that P. carribbica either cultured in the NYDB media amended with trehalose at 0.5% or cultured in NYDB without trehalose induced more β-1,3-glucanase activity of strawberries compared with the control at some of the storage time, and P. carribbica cultivated in NYDB media amended with trehalose at 0.5% induced more β-1,3-glucanase activity of strawberries compared with P. carribbica cultured in the NYDB at some of the storage time. A previous study indicates that in plants, invasion by a pathogen induces the production of pothogenesis-related (PR) proteins, such as β-1,3-glucanases, and the putative role of β-1,3-glucanases in disease resistance is related to their capacity to degrade fungal cell wall, mainly composed of β-1,3-glucan [26]. This induction can be enhanced by some elicitors with the improvement of resistance to pathogens in plants [27,28]. Therefore, we infer that improving the activity of antagonistic yeast to induce the production of pothogenesis-related (PR) proteins of strawberries may be another reason that a more remarkable upward trend was seen in the efficacy of P. carribbica cultivated in the NYDB media amended with 0.5% trehalose in controlling postharvest decay of strawberries than the case in NYDB.

For protein identification by means of peptide mass fingerprints (PMF), we used MASCOT to search protein database of Viridiplantae. More than 180 protein spots were detected in each gel after ignoring very faint spots and spots with undefined shapes and areas using Image Master 2D Elite software (Figure 6). A total of 78 proteins were differentially expressed when the yeast antagonist P. carribbica were harvested from the NYDB amended with trehalose at 0.5%. Of the 78 proteins, 53 proteins were up-regulated and 25 proteins were down-regulated. Our test focuses on 46 kinds of differences in degree of peak protein (Table 1). Of all the differentially expressed proteins identified, most of them related to basic metabolism. This indicated that the basic metabolism of P. carribbica was improved by trehalose induced incubation, so as to improve its biocontrol efficacy to postharvest diseases of strawberries. Response patterns of P. carribbica to trehalose inducing incubation are complex, as the differentially abundant proteins are involved in multiple metabolic pathways.

The yeast antagonist P. carribbica was isolated from soils of orchard (the central shoal of Yangtze River, Zhenjiang). Classical methods based on colony and cell morphologies were used for a preliminary characterization of the yeast [29]. Subsequently, sequence analysis of the 5.8S internal transcribed spacer (ITS) ribosomal DNA (rDNA) region was used to identify the yeast colony [30]. P. carribbica isolates were maintained at 4 °C on nutrient yeast dextrose agar (NYDA) medium containing 8 g nutrient broth, 5 g yeast extract, 10 g glucose and 20 g agar, in 1 L of distilled water. Liquid cultures of the yeast were grown in 250-mL Erlenmeyer flasks containing 50 mL of nutrient yeast dextrose broth (NYDB) which had been inoculated with a loop of the culture. Flasks were incubated on a rotary shaker at 180 rpm at 28 °C for 24 h. Following incubation, cells were centrifuged at 6000 ×g for 10 min and washed twice with sterile distilled water in order to remove the growth medium. Cell pellets were re-suspended in sterile distilled water and adjusted to an initial concentration of 5 × 10⁸ cells/mL. Then, 1 mL of the above-mentioned suspensions were added and cultivated in nutrient yeast dextrose broth (NYDB) or NYDB amended with trehalose powder at 0.2% (using 0.8% dextrose and 0.2% trehalose as the carbon source) or 0.5% (using 0.5% dextrose and 0.5% trehalose as the carbon source) or 0.8% (using 0.2% dextrose and 0.8% trehalose as the carbon source) or NYTB (trehalose as the sole carbon source instead of dextrose in the media of nutrient yeast dextrose broth) on a rotary shaker at 180 rpm at 28 °C for 24 h. Then the yeast cells were harvested by centrifuging at 6000 ×g for 10 min and were washed twice with sterile distilled water. The yeast cell was counted using a hemocytometer. Cell pellets were re-suspended in sterile distilled water and adjusted to an initial concentration of 1 × 10⁸ cells/mL for experiments.

Strawberries (Fragaria ananassa Duch.) cultivars “fengxiang” were harvested early in the morning and rapidly transferred to the laboratory. Berries were sorted on the basis of size, ripeness and to remove any with the apparent injuries or infections.

The pathogen Rhizopus stolonifer (Ehrenb.: Fr) Vuill. and Botrytis cinerea (Pers.:Fr.) were isolated from infected strawberries. The culture was maintained on potato dextrose agar (PDA: extract of boiled potatoes, 200 mL; dextrose, 20 g; agar, 20 g and distilled water, 800 mL) at 4 °C, and fresh cultures were grown on PDA plates before use. Spore suspensions were prepared by removing the spores from the sporulating edges of a 7-day old culture with a bacteriological loop, and suspending them in sterile distilled water. Spore concentrations were determined with a hemocytometer, and adjusted as required with sterile distilled water.

The surface of strawberries was wounded with a sterile cork borer (approximately 3-mm-diameter and 3-mm-deep) and treated with 30 μL of (1) the cell suspensions of P. carribbica (1 × 10⁸ cells/mL) which were harvested from the media of NYDB; (2) the cell suspensions of P. carribbica (1 × 10⁸ cells/mL) which were harvested from the media of NYDB amended with trehalose powder at 0.2%; (3) the cell suspensions of P. carribbica (1 × 10⁸ cells/mL) which were harvested from the media of NYDB amended with trehalose powder at 0.5%; (4) the cell suspensions of P. carribbica (1×10⁸ cells/mL) which were harvested from the media of NYDB amended with trehalose powder at 0.8%; (5) the cell suspensions of P. carribbica (1 × 10⁸ cells/mL) which were harvested from the media of NYTB and (6) the sterile distilled water as the control. Three hours later, 30 μL of R. stolonifer suspensions (1 × 10⁴ spores/mL) or B. cinerea suspensions (1 × 10⁵ spores/mL) were inoculated to each wound. After air-drying, the strawberries were stored in enclosed plastic trays to maintain a high relative humidity (about 95%) and incubated at 20 °C for 3 days. The number of the infected fruit wounds was examined. There were three replicates per treatment and 20 fruits each replicate. All treatments were arranged in a randomized complete block design, and the experiment was conducted twice.

The surface of strawberries was wounded with a sterile cork borer (approximately 3-mm-diameter and 3-mm-deep) and treated with 30 μL of a cell suspension of P. carribbica (1 × 10⁸ cells/mL) which were harvested from the media of NYDB or NYDB amended with trehalose powder at 0.5% (using 0.5% dextrose and 0.5% trehalose as the carbon source) after 24 h incubation. Treatment with sterile distilled water served as the control. After air-drying, the strawberries were stored in enclosed plastic trays to maintain a high relative humidity (about 95%) and incubated at 20 °C. Samples were taken at 0, 1, 2, and 3 days after treatment. After removing the wound tissue with a sterile borer (6-mm-diameter and 5-mm-deep), the fresh tissue around the wound was picked up by another sterile borer (9-mm-diameter and 10-mm-deep). Two grams of the fresh tissue from six fruits were homogenized with 10 mL of cold (4 °C) 50 mM (pH 7.8) sodium phosphate buffer containing 1.33 mM EDTA and 1% PVPP. The homogenates were then centrifuged at 12,000 ×g at 4 °C for 15 min and the supernatants were assayed. There were three replicates per treatment. The experiment was conducted twice. The testing methods of enzyme activities were described below.

PPO activity was measured as the method described by Aquino-Bolaños and Mercado-Silva [31] with some modifications, using catechol as a substrate. The reaction mixtures contained 2.9 mL of 0.1 M catechol (prepared by 50 mM sodium phosphate buffer, pH 6.4, incubated at 30 °C for 5 min) as a substrate and 100 μL of enzymatic extract. The change in absorbance was read per minute at 398 nm, and determined 3 min continuously. The PPO activity was expressed as U per g of fresh tissue weight (U/g FW), which one unit (U) of PPO was defined as the change of absorbance ascent 0.01 in 1 min.

POD activity was measured as the method described by Lurie et al. [24] with some modifications, using guaiacol as a substrate. The reaction mixture contained 0.2 mL of crude enzyme extract (supernatant extract), 2.2 mL of 0.3% guaiacol (prepared by 50 mM sodium phosphate buffer, pH 6.4) and was incubated for 5 min at 30 °C. The reaction was then initiated immediately by adding 0.6 mL of 0.3% H₂O₂ (prepared by 50 mM sodium phosphate, pH 6.4, and incubated at 30 °C for 5 min) and the activity was determined by measuring at A₄₇₀ once every 1 min for 3 min. A cuvette containing all components except adding 0.6 mL of distilled water was used as a control. The POD activity was expressed as U per g fresh tissue weight (U/g FW). One unit was defined as an increase in A ₄₇₀ of 0.01 per minute.

β-1,3-glucanase was assayed by measuring the amount of reducing sugar released from the substrate by the method reported by Ippolito et al. [32] with some modifications. Crude enzyme sample of 250 μL was routinely added to 250 μL of 0.2% laminarin (w/v) in 50 mM, pH 5.0, and potassium acetate buffer and incubated at 37 °C for 1 h. For the control, the same mixture was similarly diluted at zero incubation time. The reaction was stopped by adding 1.5 mL of 3, 5-dinitro-salicylate and boiling for 5 min on a water bath. And the amount of reducing sugars was measured spectrophotometrically at 500 nm using a UV-1601 spectrophotometer (Shimadzu, Japan). One unit of the β-1,3-glucanase was defined as the formation of 1 mg glucose equivalents per hour and the specific activity was expressed as the U per gram FW.

Liquid cultures of the yeast were grown in NYDB or NYDB+ 0.5% trehalose, respectively as described above (Antagonist and growth conditions). The method of protein sample preparation was described by Li et al. [33] with some modifications. The yeast cells were harvested from NYDB or NYDB+ 0.5% trehalose by centrifuging at 10,000 ×g for 10 min (4°C) and washed three times with cold double-distilled water to remove residual medium. Then grind sample (yeast cells) into a fine powder in a mortar pestle under liquid nitrogen, a fine powder is important for effective contaminant removal and protein extraction. Transfer the powder into a 10-mL tube, filled the tube with TE buffer containing 10 mM Tris-HCL, pH 8.0, 1 mM EDTA, and 1 mM PMSF, incubated at 4 °C for 30 min. Mixed well by vortexing, and take 0.5 mL to 1.5 mL EP tube, added 8.7 μL of 10 mg mL⁻¹ PMSF and several 0.5 mm glass beads, vortex oscillator 5 × 30 s, and interval ice bath for 1 min. Then added 25 μg of RNase A and 100 μg of DNase, incubated at 4 °C for 30 min. Centrifuge at 15,000 ×g for 30 min (4 °C), take the supernatant 400 μL, added three volumes of 20% TCA/acetone (−20 °C pre-cooled at least 30 min), mixed well, incubated at −20 °C for 12–16 h. Centrifuge at 15,000 ×g for 30 min (4 °C), discard the supernatant, washed with acetone (−20 °C pre-cooled) several times, and after washed every time, centrifuge at 15,000 ×g for 30 min (4 °C), and discard the supernatant. Air-dry at room temperature at least 10 min to remove residual acetone. Then solubilized in 100 μL of thiourea/urea/lysis buffer containing 2 M thiourea, 7 M urea, 4% (w/v) CHAPS, 65 mM DTT, 0.2% (w/v) Bio-Lyte. Protein samples were kept at −70°C until use. The protein concentration was determined according to Bradford’s method using bovine serum albumin as standard [34].

Two-dimensional electrophoresis (2-DE) and image analysis were performed as described by Wang et al. [35]. 380 μg of each protein extract was separated by iso-electrophoresis using IPG strips (pH 3–10, 17 cm) in the Protean system (Bio-Rad). The second dimension was run on a 12.5% polyacrylamide gel using the Multiphor system (Amersham Biosciences). Gels were visualized by Coomassie Blue staining. The stained gels were scanned and analyzed using PDQuest software (version 7.2, Bio-Rad, Hercules, CA, USA). Proteins that were increased 2-fold at least at one point after treatment, as well as exhibiting the same expression pattern among the replicates, were considered as significant and reproducible changes proteins. These proteins were subjected to identification. At least three biological replicates were performed for each treatment.

Differentially expressed protein spots were excised from the gel and were washed twice by double-distilled H₂O, then destained with 50 mM NH₄CO₃/acetonitrile solution. Washed with 25 mM NH₄CO₃, 50% acetonitrile and acetonitrile until the gels became full white, then vacuum drained 5 min. Added 2 μL 10 μg·μL⁻¹ trypsin (Sigma–Aldrich, St Louis, MO, USA), incubated at 4 °C for 30 min, then added 10 μL 25 mM NH₄CO₃, the gels were incubated overnight at 37 °C. The supernatant was collected for MS analysis [36].

The peptide solution was analyzed using MALDI TOF/TOF mass spectrometer (Ultraflex III, Bruker-Daltonics). The resulting monoisotopic peptide masses were queried against protein database in NCBInr using MASCOT software [37] and the following search parameters: all entries, trypsin, up to one missed cleavage, carbamidomethyl(C), oxidation(M) and Gln-Pyro-glu, peptide tolerance 0.3 Dal, mass value MH⁺, and monoisotopic [38].

The data were analyzed by the analysis of variance (ANOVA) in the statistical program SPSS/PC version II.x, (SPSS Inc. Chicago, Illinois, USA) and the Duncan’s multiple range test was used for means separation. In addition, when the group of the data was two, the independent samples t test was applied for means separation. The statistical significance was assessed at the level P = 0.05.