Increase in antioxidant gene transcripts, stress tolerance and biocontrol efficacy of Candida oleophila following sublethal oxidative stress exposure

Abstract

A pretreatment of the yeast, Candida oleophila, with 5 mM H₂O₂ for 30 min (sublethal) increased yeast tolerance to subsequent lethal levels of oxidative stress (50 mM H₂O₂), high temperature (40 °C), and low pH (pH 4). Compared with non-stress-adapted yeast cells, stress-adapted cells exhibited better control of apple fruit infections by Penicillium expansum and Botrytis cinerea and had initially higher growth rates in apple wounds. Suppression subtractive hybridization analysis was used to identify genes expressed in yeast in response to sublethal oxidative stress. Transcript levels were confirmed using semiquantitative reverse transcription-PCR. Seven antioxidant genes were upregulated. The elevated expression of these genes was associated with less accumulation of reactive oxygen species and a lower level of protein and lipid oxidation under subsequent stresses. These data support the premise that induction of abiotic stress tolerance in biocontrol yeast can improve biocontrol efficacy by upregulation of genes involved in the amelioration of oxidative stress.

antioxidant genes, environmental stress tolerance, postharvest diseases, reactive oxygen species, suppression subtractive hybridization

Introduction

The development of biological methods utilizing microbial antagonists for the management of postharvest diseases has been a significant research topic in recent years (Droby et al., 2009; Sharma et al., 2009), and the utilization of yeasts has been emphasized (Spadaro & Gullino, 2004; Wisniewski et al., 2007). Among yeast antagonists, Candida oleophila (strain 182) has been demonstrated to be effective against postharvest diseases of apple (Droby et al., 2003), nectarine (Lurie et al., 1995), peach (Droby et al., 2003), papaya (Gamagae et al., 2003), and grapefruit (Bar-Shimon et al., 2004).

Biocontrol yeasts used to manage postharvest diseases encounter various environmental conditions that can impact their viability and, hence, efficacy. Specific conditions will depend on the standard practices used to process various commodities, as well as the conditions unique to each packinghouse. In some cases, postharvest biocontrol agents are also administered in the field, days to weeks prior to harvest (Lahlali et al., 2011; Zhao et al., 2011), which also exposes them to a wide range of environmental stresses. Environmental conditions such as high temperature, oxidative stress, and low pH play critical roles in biocontrol systems. High temperature can significantly lower the viability of some biocontrol yeasts, especially when yeasts are used prior to harvest under field conditions (Ippolito & Nigro, 2000). The ability to survive and proliferate in wounded tissues is pivotal for postharvest biocontrol agents (Spadaro & Gullino, 2004; Droby et al., 2009); however, wounding of fruit tissue is associated with the accumulation of reactive oxygen species (ROS), which in turn may affect yeast wound competency and thus biocontrol efficacy (Castoria et al., 2003; Macarisin et al., 2010; Tolaini et al., 2010). The growth and viability of biocontrol yeasts are also influenced by the pH values of culture media and fruit tissues (Teixidó et al., 1998; Spadaro et al., 2010).

Previous studies have indicated that adaptation of biocontrol agents to mild stresses improves their tolerance to subsequent lethal stresses (Teixidó et al., 2005; Wang et al., 2010). Cold adaptation improved the tolerance of Cryptococcus flavescens to desiccation by regulating membrane fluidity (Dunlap et al., 2007), while mild heat shock enhanced tolerance to subsequent lethal high temperature and oxidative stress, and improved the biocontrol efficacy of Metschnikowia fructicola (Liu et al., 2011a). In the case of bacteria, it was found that exposure of Pantoea agglomerans to either a heat shock (Cañamás et al., 2008) or mild acidic conditions (Cañamás et al., 2009) could induce resistance to subsequent abiotic stress.

Sublethal oxidative stress can enhance the ability of yeasts to withstand subsequent unfavorable environmental conditions (Vandenbroucke et al., 2008), such as UV (Verma & Singh, 2001), freeze-thaw stress (Park et al., 1997), high temperature (Biryukova et al., 2007), and more severe oxidative stress (Alvarez-Peral et al., 2002). While mild stress adaptation has been reported to induce cross-protection against a variety of abiotic stresses to prokaryotic and eukaryotic microorganisms, in general (Rangel, 2011), limited information is available for biocontrol agents on changes in gene expression during oxidative stress adaptation and the concomitant improvement in subsequent stress tolerance and biocontrol efficacy by the adaptation treatment. The objective of this study was to determine the effect of a mild oxidative stress treatment on gene expression and subsequent stress tolerance and biocontrol efficacy of the yeast antagonist, C. oleophila. More specifically, this study determined (1) the viability of this yeast after exposure to a range of oxidative stress conditions, established by varying concentrations of hydrogen peroxide, and a range of high temperatures; (2) the effect of oxidative stress adaptation on subsequent abiotic stress tolerance, including low pH; (3) the effect of oxidative stress adaptation on differential gene expression of C. oleophila using suppression subtractive hybridization (SSH) and semiquantitative RT-PCR; and (4) the effect of oxidative stress adaptation on yeast growth in apple wounds and biocontrol efficacy against Penicillium expansum and Botrytis cinerea.

Materials and methods

Yeast

Candida oleophila (strain I-182) was isolated from the surface of tomato fruit (Wilson et al., 1993) and grown in a yeast–peptone–dextrose (YPD) broth (10 g of yeast extract, 20 g of peptone, and 20 g of dextrose in 1 L of water). Twenty milliliters of YPD (Fisher Scientific, Fair Lawn, NJ) was placed in 50-mL conical flasks and inoculated with C. oleophila at an initial concentration of 10⁵ cells mL⁻¹ determined by an automated cell counter, Cellometer Vision (Nexcelom Bioscience, Lawrence, MA). Yeast cultures were incubated at 25 °C at 200 r.p.m. on a rotary shaker (New Brunswick Scientific, Edison, NJ).

Pathogens

Penicillium expansum and B. cinerea were isolated from infected apple fruits and maintained on potato dextrose agar (PDA) (Difco, Sparks, MD) at 4 °C. To reactivate the culture and verify their ability to cause disease, the pathogens were inoculated into wounds of apple fruit and re-isolated onto PDA once the infection was established. Spore suspensions of the two pathogens were obtained from 2-week-old cultures on PDA at 25 °C. The spore number was calculated using a Cellometer Vision (Nexcelom Bioscience), and the concentration was adjusted to 1 × 10⁴ spores mL⁻¹ with sterile distilled water.

Fruits

‘Golden Delicious’ apples (Malus× domestica Borkh.) were harvested at commercial maturity. Fruits without wounds or rot were selected based on uniformity of size, disinfected with 2% (v/v) sodium hypochlorite for 2 min, rinsed with tap water, and air-dried (Li et al., 2008). These fruits were used in a subsequent biocontrol assay.

Viability test of C. oleophila under oxidative and heat stresses

Overnight yeast cultures were pelleted at 8000 g for 3 min and washed three times with sterile distilled water to remove any residual medium (Li & Tian, 2006). To simulate oxidative stress, a 10-mL yeast cell suspension at 1 × 10⁸ cells mL⁻¹ in water was placed in a 50-mL conical flask. The yeast cells were then exposed to a final concentration of 20, 50, or 100 mM H₂O₂ at 25 °C on a shaker at 200 r.p.m. for 20, 40, or 60 min. For assay of tolerance to high temperatures, 1-mL cell suspension adjusted to 1 × 10⁸ cells mL⁻¹ was distributed in 1.7-mL eppendorf tubes. The yeast cells were then incubated for 25, 30, or 35 min in separate water baths set at 39, 40, or 41 °C. At the described time points, 50 μL of serial 10-fold dilutions of the samples were spread on YPD agar plates. The plates were incubated at 25 °C for 3 days, and the number of CFUs per plate was determined. Survival rates were expressed as a percentage of CFUs after the oxidative treatment or the high temperature relative to CFUs before the treatment (Liu et al., 2011a). There were three replicates in each treatment, and each experiment was repeated three times.

Effect of mild oxidative stress adaptation on C. oleophila tolerance to subsequent oxidative and heat stresses

The effect of stress adaptation (sublethal H₂O₂ treatment) on survival of C. oleophila under subsequent stresses was determined according to the method of Deveau et al. (2010), with slight modification. Washed cells from overnight culture were resuspended in the same volume of fresh YPD, supplemented with 5 mM H₂O₂ (the concentration was selected based on preliminary data) and incubated at 25 °C on a rotary shaker at 200 r.p.m. for 30 min. Non-stress-adapted (NSA) yeast cells without H₂O₂ treatment served as a control. Cells were harvested by centrifugation at 8000 g for 3 min and washed three times with sterile distilled water to remove any residual H₂O₂ and medium. To assay subsequent oxidative stress tolerance, the stress-adapted (SA) cells (treated with 5 mM H₂O₂ for 30 min) and NSA cells were exposed to 50 mM H₂O₂ at 25 °C on a shaker at 200 r.p.m. for 20, 40, or 60 min. For heat stress, yeast samples (SA or NSA cells) were incubated in a 40 °C water bath for 25, 30, and 35 min. Survival rates at each time point were evaluated using the same method and with the same number of replications as described previously.

Imaging of intracellular ROS

The oxidant-sensitive probe, 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA; Invitrogen, Eugene, OR), was used to assess intracellular ROS production in C. oleophila (Liu et al., 2011a). Yeast cells (NSA or SA) were collected from samples exposed to 50 mM H₂O₂ for 20, 40, and 60 min, or 40 °C for 25, 30, and 35 min. NSA and SA samples taken before exposure to the lethal oxidative or heat stress served as time 0. Cells were washed with phosphate-buffered saline (PBS, pH 7.0) and resuspended in the same buffer containing 25 μM H₂DCFDA (dissolved in dimethyl sulfoxide). The suspension was incubated in the dark at 30 °C for 1 h. After washing twice with PBS buffer, yeast cells were examined under a Zeiss Axioskop microscope (Carl Zeiss, Oberkochen, Germany) equipped with a UV light source using a 485-nm excitation and 530-nm emission filter combination. Five fields of view from each slide (at least 200 cells) were randomly chosen, and the number of cells producing visible levels of ROS in response to oxidative stress was counted for both NSA and SA samples. The ROS level was calculated as a percentage (number of fluorescing cells divided by number of cells present in bright field image × 100). There were three replicates in each treatment, and the experiment was repeated three times.

Protein carbonylation and lipid peroxidation

Yeast samples (NSA or SA) were collected from samples exposed to 50 mM H₂O₂ for 20, 40, and 60 min, or 40 °C for 25, 30, and 35 min. NSA and SA samples before exposure to the lethal oxidative and heat stresses served as time 0. Yeast cells were disrupted in liquid nitrogen by grinding in a mortar with a pestle. Carbonyl content was measured and used as an indicator of oxidative damage on protein (Liu et al., 2011b). Proteins were extracted from the samples using 500 μL of 50 mM KH₂PO₄ buffer (pH 7.5) containing 10 mM Tris, 2 mM MgCl₂, 2 mM EGTA, and 1 mM phenylmethylsulfonyl fluoride. Aliquots of extract were reacted with 500 μL of 10 mM 2,4-dinitrophenylhydrazine (DNPH) dissolved in 2.5 M HCl or 2.5 M HCl without DNPH (blank control) in the dark at room temperature with vortexing every 15 min for 1 h. Proteins were precipitated with 20% trichloroacetic acid (TCA, w/v) and kept on ice for 10 min. After centrifugation at 3000 g for 20 min, protein pellets were washed with ethanol–ethyl acetate (1 : 1, v/v) and dissolved in 6 M guanidine hydrochloride with 20 mM KH₂PO₄ (pH 2.3). The absorbance was recorded at 380 nm after centrifugation at 9500 g for 10 min. The carbonyl content was calculated using the molar absorption coefficient of 22 000 M⁻¹ cm⁻¹ and expressed as nmol per mg protein.

For assay of lipid peroxidation, the method based on the reaction of thiobarbituric acid with malondialdehyde (MDA) was used. Detection of thiobarbituric acid-reactive species was carried out as described by Garre et al. (2010). Disrupted yeast samples were resuspended in 500 μL of 50 mM KH₂PO₄ buffer (pH 6.0) containing 10% (w/v) TCA and centrifuged at 3000 g for 10 min. Supernatants were mixed with 100 μL of 0.1 M EDTA and 600 μL of 1% (w/v) thiobarbituric acid. The reaction mixture was incubated at 100 °C for 15 min and then placed on ice for 10 min. After cooling down, the absorbance was measured at 532 nm. The MDA content was calculated using the molar absorption coefficient of 153 000 M⁻¹ cm⁻¹ and expressed as nmol per mg protein.

Protein content was measured using the Bradford assay (Bradford, 1976). Bovine serum albumin (Sigma-Aldrich, St. Louis, MO) was used as a standard. There were three replicates in each treatment, and the experiment was repeated three times.

RNA isolation and SSH

Total RNA from NSA or SA (by treatment with 5 mM H₂O₂ for 30 min) yeast cells was isolated using the RNeasy Mini kit (Qiagen Science, Germantown, MD) according to the manufacturer's instructions. Extracted RNA was treated with TURBO™ DNase (Ambion, Austin, TX) and purified again with RNeasy. cDNA was synthesized and amplified using the SMARTer cDNA Synthesis kit (Clontech, Mountain View, CA). In this study, 500 ng of RNA was used to generate each cDNA population for use in the subtraction procedure. The manufacturer's recommendations were used throughout the cDNA synthesis procedure. Differentially expressed genes were identified using the PCR-Select cDNA Subtraction kit (Clontech). In brief, the SMARTer cDNAs of NSA and SA yeast samples were subjected to RsaI digestion, adaptor ligation, two rounds of subtractive hybridization, and subsequent PCR amplifications. The forward-subtracted cDNAs (cDNAs from SA sample as tester and cDNAs from NSA sample as driver) were cloned using a PCR 2.1-TOPO TA Cloning kit (Invitrogen), following the manufacturer's protocol. White colonies were isolated from selective media for blue/white colony screening and analyzed by colony PCR using M13 forward and reverse primers to estimate insert sizes in the recombinant plasmids (Bassett et al., 2006). Recombinant plasmids with inserts > 200 bp were isolated with the QIAprep Spin Miniprep kit (Qiagen Science). These DNAs were sequenced by Macrogen Corp. (Rockville, MD) using M13 forward and reverse primers. The obtained sequences were queried for similarity in the NCBI nonredundant protein sequences database (http://www.ncbi.nlm.nih.gov) using blastx sequence comparison algorithms (Liu et al., 2007; Nevarez et al., 2008). The best match was used to assign gene identity. Candida oleophila gene sequences were submitted to GenBank. Accession numbers and gene identity are listed in Table 1.

Identification of Candida oleophila cDNAs induced by sublethal oxidative stress (5 mM H₂O₂ for 30 min)

NCBI accession no. of C. oleophila cDNAs	Size (bp)	BLASTX database match	Identity (%)	Similarity (%)	E-value
JN615130	549	Peroxisomal catalase (Meyerozyma guilliermondii ATCC 6260)	92	97	2e−122
JN615131	485	Cytochrome c peroxidase (Debaryomyces hansenii CBS767)	77	88	6e−91
JN615132	578	Peroxiredoxin TSA1 (Debaryomyces hansenii CBS767)	92	98	4e−87
JN615133	729	Thioredoxin reductase (Scheffersomyces stipitis CBS 6054)	88	97	1e−156
JN615134	486	Glutathione peroxidase (Scheffersomyces stipitis CBS 6054)	79	90	1e−90
JN615135	767	Glutathione reductase (Debaryomyces hansenii CBS767)	75	87	9e−122
JN615136	823	Glucose-6-phosphate dehydrogenase (Candida albicans SC5314)	88	94	4e−164
JN615137	227	Hypothetical protein induced by osmotic stress (Scheffersomyces stipitis CBS 6054)	69	84	4e−27
JN615138	236	Hypothetical protein (Debaryomyces hansenii CBS767)	66	79	1e−24

Semiquantitative reverse transcription (RT)-PCR analysis of gene expression

Seven differentially expressed genes were selected from the SSH library for further analysis using semiquantitative RT-PCR. Yeast cultures were exposed to 50 mM H₂O₂ for 20, 40, and 60 min, or 40 °C for 25, 30, and 35 min. The cDNAs (100 ng) synthesized at each time point were used as templates in PCRs. NSA and SA samples before exposure to the lethal oxidative and heat stresses served as time 0. Transcript levels of 18S rRNA gene served as an internal control (Lee et al., 2008; Pacheco et al., 2009). Primer design for 18S rRNA gene was based on a partial coding sequence of 18S rRNA gene of C. oleophila (NCBI accession no. AB013534). Cycling parameters for each gene amplification were 95 °C for 5 min; 25 cycles of 95 °C for 30 s, annealing temperature for 30 s, and 72 °C for 30 s; and finally 72 °C for 10 min. The primer pairs and annealing temperature for each gene are listed in Table 2. The primer pairs were based on the sequences of genes obtained from cloned products of the subtracted library. PCR products obtained from the RT-PCR were cloned and sequenced to verify their identity. PCR products were separated by agarose gel electrophoresis stained with ethidium bromide (Sigma-Aldrich) and scanned using a Typhoon Trio Variable Mode Imager (GE Healthcare, Fairfield, CT). There were three replicates for each treatment.

Primer sequences and annealing temperatures used in semiquantitative RT-PCR analysis of genes obtained using SSH. Predicted product sizes are also provided

Gene name	Primer sequence	Annealing temperature (oC)	Product size (bp)
Peroxisomal catalase	F: AGGGTAGGAAAGCACGGTCCA, R: TGGGTGTGGATAAAGTGTGGGAACT	55	377
Cytochrome c peroxidase	F: TGTGGACCTTGGGTGGTGTTGT, R: TGGACCCCATGGACCATCGAA	56	269
Peroxiredoxin TSA1	F: TCGCCTCCACTGACTCCGAA, R: ACCTGGTTGCCAGTTAGCTGGACA	56	317
Thioredoxin reductase	F: GTCATCGCCACCGGTGCCTC, R: TGGCAAGTCCTTGGTTTCGTTGCT	57	375
Glutathione peroxidase	F: CGGACTTTTGGCTCTTCAAA, R: GTGTGGTTTCACTCCCCAGT	56	242
Glutathione reductase	F: AGATCCGGCTTCTGGGTGCG, R: TGGCTGTTGTCGGTGCTGGA	57	557
Glucose-6-phosphate dehydrogenase	F: ACAAGGACCACATCAGGCTGGT, R: TGGAACACCATCCCAACGCTCA	56	374
18S rRNA gene	F: TGGCCTTGTGCTGGCGATGG, R: AAGCGGACCGGCCAACCAAG	58	391

Biocontrol assay of C. oleophila against postharvest diseases of apple fruits

Biocontrol activity of C. oleophila was tested according to a previous study (Liu et al., 2011b). Three wounds (4 mm deep × 3 mm wide) were made on the equator of each fruit. A 10-μL suspension of NSA or SA cells of C. oleophila (1 × 10⁷ cells mL⁻¹) was applied to each wound. Sterile distilled water served as a control. After fruits were air-dried for 2 h, 10 μL of P. expansum or B. cinerea suspension (1 × 10⁴ spores mL⁻¹) was administered into each wound. Treated fruits were placed in a covered plastic food tray, and each tray was enclosed in a polyethylene bag to maintain high humidity (about 95% relative humidity) and stored at 25 °C. Disease incidence and lesion diameter of apple fruits were determined after 4 days. Each treatment contained three replicates with 10 fruits per replicate, and the experiment was repeated three times.

Population dynamics of C. oleophila in YPD with pH 4

For determination of in vitro growth dynamics of C. oleophila under low pH condition, basal YPD was adjusted to pH 4 (the same pH value as ‘Golden Delicious’ apple flesh) using buffers (0.1 M citric acid and 0.2 M Na₂HPO₄), according to Teixidó et al. (1998). Yeast samples (SA or NSA) were inoculated to 20 mL of YPD with pH 4 in 50-mL conical flasks, at an initial concentration of 10⁵ cells mL⁻¹, and incubated at 25 °C on a shaker at 200 r.p.m. Cells counts in each treatment were determined by a Cellometer Vision (Nexcelom Bioscience) every 12 h for 48 h. There were three replicates in each treatment, and the experiment was repeated three times.

Population dynamics of C. oleophila in wounds of apple fruits

Three wounds (4 mm deep × 3 mm wide) were made on the equator of each apple fruit. A 10-μL suspension of NSA or SA cells of C. oleophila (1 × 10⁷ cells mL⁻¹) was applied to each wound. Fruit samples were collected at different time points after treatment, and yeast populations were measured (Cao et al., 2010). Briefly, yeasts were recovered by removing ten samples of wounded tissues with a cork borer (1 cm diameter × 1 cm deep). Samples were ground with a mortar and pestle in 10 mL of sterile distilled water. Then, 50 μL of serial 10-fold dilutions were spread on YPD agar plates. Samples taken at 1 h after treatment served as time 0. Fruits stored at 25 °C were assessed each day for 4 days. Colonies were counted after incubation at 25 °C for 3 days and expressed as the log₁₀ CFU per wound. There were three replicates in each treatment, and the experiment was repeated three times.

Data analysis

All statistical analyses were performed with spss version 13.0 (SPSS Inc., Chicago, IL). Two-way analysis of variance (anova) was performed on data where treatment and time were variables, and mean separation was determined for NSA and SA samples at each time point using a Student's t-test. Data with a single variable (treatment) were analyzed by one-way anova, and mean separations were performed by Duncan's multiple range tests. Differences at P<0.05 were considered significant. Data presented in this article were pooled across three independent repeated experiments, as the interaction between treatment and experiment variables was not significant.

Results

Survival of C. oleophila under oxidative stress and high temperature

As expected, the viability of C. oleophila cells decreased with exposure to increasing H₂O₂ concentration and time within each concentration (Table 3). At 20 mM H₂O₂ for 20 min, C. oleophila cells exhibited 93.3% viability, while at 100 mM H₂O₂ for 20 min, yeast cells had a viability of 27.8% and only 6.4% after 60 min. Yeast cells exhibited intermediate levels of survival in 50 mM H₂O₂ at all treatment time points. Similar to the trend in viability under oxidative stress, the survival of C. oleophila decreased with exposure to increasing temperature (39–41 °C) and exposure time (Table 3). Based on the results of the viability assays, 50 mM H₂O₂ and 40 °C were chosen to be appropriate conditions for assessing the ability of mild oxidative stress treatment to improve tolerance to a subsequent oxidative stress or high temperature.

Percent viability of Candida oleophila under oxidative stress and high temperatures

Treatment	Treatment time (min)
20	25	30	35	40	60
H2O2 (mM)
20	93.3 ± 2.9a	N/A	N/A	N/A	82.2 ± 2.7a	70.3 ± 2.6a
50	78.4 ± 3.2b	N/A	N/A	N/A	60.8 ± 2.5b	45.1 ± 1.9b
100	27.8 ± 3.5c	N/A	N/A	N/A	10.6 ± 2.4c	6.4 ± 1.2c
Temperature (ºC)
39	N/A	96.5 ± 1.4a	92.7 ± 1.6a	86.3 ± 2.8a	N/A	N/A
40	N/A	74.8 ± 3.4b	64.2 ± 3.3b	48.9 ± 3.6b	N/A	N/A
41	N/A	29.3 ± 2.0c	22.3 ± 1.9c	11.5 ± 0.9c	N/A	N/A

Values are the means of pooled data ± standard deviations (n = 9). Values within a column followed by different letters are significantly different according to Duncan's multiple range test (P<0.05) and reflect differences in viability at different concentrations of H₂O₂ and different temperatures within a specific treatment time.

N/A = not tested.

Effect of mild oxidative stress adaptation on subsequent stress tolerance of C. oleophila and ROS accumulation

Pretreatment of yeast cells with a mild oxidative stress of 5 mM H₂O₂ for 30 min had a significant effect on C. oleophila viability when yeast cells were exposed to a subsequent oxidative or heat stress (P<0.05, two-way anova analysis). As indicated in Fig. 1a, the viability of yeast cells kept at 50 mM H₂O₂ decreased over the exposure time range (20–60 min). However, SA cells had significantly (P<0.05, Student's t-test) higher viability than NSA cells at all three time points. After 20 min at 50 mM H₂O₂, survival of NSA cells was 77.2%, while that of SA cells was 93.2%. A similar pattern of viability was observed in response to heat stress (Fig. 1b). While the viability of cells decreased over time in response to exposure to 40 °C, SA cells exhibited significantly improved tolerance to the heat stress at all time periods assessed.

Percent survival of (a, b) and percent of cells exhibiting visible ROS accumulation (c, d) in NSA and SA (5 mM H₂O₂ for 30 min) yeast cells of Candida oleophila exposed to oxidative stress (50 mM H₂O₂) (a, c) and high temperature (40 °C) (b, d). Data presented are the means of pooled data. Error bars indicate standard deviations of the means (n = 9). Columns with different letters at each time point indicate significant differences between treatments (NSA vs. SA) according to Student's t-test (P<0.05).

Sublethal oxidative stress did not significantly increase ROS accumulation in C. oleophila cells (Fig. 1c and d). At time 0, prior to subsequent treatments at 50 mM H₂O₂ or 40 °C, the percentage of both SA and NSA cells exhibiting a visible ROS level, as determined by use of the fluorescent dye H₂DCFDA, was < 10%. However, the percentage of cells exhibiting an accumulation of ROS increased with time of exposure to 50 mM H₂O₂ or 40 °C. Both treatment and duration of exposure to a subsequent lethal stress had a significant (P<0.05, two-way anova analysis) effect on intracellular ROS accumulation in C. oleophila. SA cells exhibited a significantly (P<0.05, Student's t-test) lower percentage of cells exhibiting a visible ROS level for both stresses compared with NSA cells at all three time points after time 0. After 60 min at 50 mM H₂O₂, ROS was detected in 42.6% of SA cells, which was significantly lower than the 66.7% observed for NSA cells. Likewise, after 35 min at 40 °C, the percentage of SA cells stained with H₂DCFDA was 36.8%, while that of NSA cells was 53.6%. While we could not determine whether fluorescent cells were living or dead, the percentage of nonfluorescent cells was very close to the percentage obtained in the viability assay (Fig. 1a and b). Therefore, it appeared that the percentage of fluorescent cells was negatively correlated with cell viability.

Oxidative damage on protein and lipid of C. oleophila under stresses

To investigate the protective effect of sublethal oxidative stress adaptation on oxidative damage to cellular proteins and lipids, carbonyl content as an indicator of protein oxidation and MDA content as an indicator of lipid peroxidation (Saharan & Sharma, 2010) were measured (Fig. 2). The pattern of carbonyl and MDA accumulation was similar to that of ROS. At time 0, prior to subsequent treatments at 50 mM H₂O₂ or 40 °C, carbonyl and MDA contents in both SA and NSA cells were quite low. Sublethal oxidative treatment (5 mM H₂O₂ for 30 min) had a significant (P<0.05, two-way anova analysis) effect on reducing carbonyl and MDA accumulation when yeast cells were exposed to a subsequent oxidative or heat stress. Both 50 mM H₂O₂ and high temperature (40 °C) treatments resulted in a marked increase in carbonyl and MDA contents in yeast cells over the exposure time. SA cells, however, had significantly (P<0.05, Student's t-test) lower carbonyl and MDA contents than NSA cells at all three time points after time 0.

Protein carbonylation (a, b) and lipid peroxidation (c, d) in NSA or SA (5 mM H₂O₂ for 30 min) Candida oleophila cells exposed to oxidative stress (50 mM H₂O₂) (a, c) and high temperature (40 °C) (b, d) for varying periods of time. Data presented are the means of pooled data. Error bars indicate standard deviations of the means (n = 9). Columns with different letters at each time point indicate significant differences according to Student's t-test (P<0.05).

Identification of differentially expressed cDNA fragments

Four hundred clones were obtained from the SSH library where SA cDNA served as the tester and NSA cDNA as the driver. The clones were analyzed by colony PCR using M13 forward and reverse primers to estimate insert sizes in the recombinant plasmids. A total of 110 recombinant plasmids with inserts > 200 bp were isolated for sequencing. Sequence analysis using blastx indicated that the 110 clones represented nine different cDNAs, which were preferentially expressed in the SA yeast cells. Seven of the clones were identified as genes with known functions, all of which were related to antioxidant responses (Table 1). These seven genes were chosen for further analysis using semiquantitative RT-PCR.

Expression of the selected cDNAs in C. oleophila in response to subsequent stress

Results of the semiquantitative RT-PCR (Fig. 3) indicated that all seven genes obtained from the SSH library were upregulated in SA yeast at time 0, thus confirming the results obtained from the SSH. Based on sequence homology, the upregulated genes in C. oleophila with known functions represented the following: peroxisomal catalase, cytochrome c peroxidase, peroxiredoxin TSA1, thioredoxin reductase, glutathione peroxidase, glutathione reductase, and glucose-6-phosphate dehydrogenase (Table 1). Importantly, SA cells had markedly higher expression of these genes than NSA cells at all three time points following exposure to either 50 mM H₂O₂ or 40 °C, although this was more evident in the heat stress treatment.

Expression of seven antioxidant genes in NSA or SA (5 mM H₂O₂ for 30 min) Candida oleophila cells exposed to oxidative stress (50 mM H₂O₂) (left) and high temperature (40 °C) (right) for varying periods of time. The 18S rRNA gene was used as a control for normalizing mRNA quantity.

Biocontrol efficacy of C. oleophila

Candida oleophila significantly (P<0.05, one-way anova analysis) reduced disease incidence caused by P. expansum and B. cinerea on apple fruits stored at 25 °C. Importantly, SA yeast cells exhibited a greater level of efficacy than NSA cells in controlling the two pathogens (Fig. 4a). Additionally, lesion diameters caused by both pathogens were significantly (P<0.05, one-way anova analysis) smaller on apples with SA cells, compared with NSA cells (Fig. 4b).

Biocontrol efficacy of Candida oleophila against Penicillium expansum and Botrytis cinerea on apple fruits stored at 25 °C. Fruits were wounded and inoculated with 10 μL of sterile water (control), NSA or SA (5 mM H₂O₂ for 30 min) cell suspension of C. oleophila at 1 × 10⁷ cells mL⁻¹. After 2 h, wounds were inoculated with 10 μL of P. expansum or B. cinerea at 1 × 10⁴ spores mL⁻¹. Disease incidence (a) and lesion diameter (b) in apple fruits were measured 4 days after inoculation. Data presented are the means of pooled data. Error bars indicate standard deviations of the means (n = 9). Pathogen specific columns with different letters indicate significant differences according to Duncan's multiple range test (P<0.05).

Growth dynamics of C. oleophila in YPD at pH 4 and in apple wounds

The mild oxidative stress treatment (5 mM H₂O₂, 30 min) had a significant impact on growth of C. oleophila in both YPD at pH 4 and apple wounds (P<0.05, two-way anova analysis). In both conditions, C. oleophila proliferated quickly in the first 24 h. In YPD at pH 4, the population of SA cells was significantly (P<0.05, Student's t-test) higher than that of NSA cells at 12 and 24 h, but not at 36 and 48 h (Fig. 5a). In apple wounds, C. oleophila multiplied quickly, and the number of the yeast increased more than 10-fold after 1 day. The population of SA cells was significantly (P<0.05, Student's t-test) higher than the population of NSA cells at 1 and 2 days, but not at 3 and 4 days when the cells reached the stationary phase (Fig. 5b).

Growth dynamics of NSA and SA (5 mM H₂O₂ for 30 min) Candida oleophila in YPD with pH 4 (a) and in wounds of apple fruits (b) at 25 °C. YPD was inoculated with NSA or SA (30 min at 5 mM H₂O₂) C. oleophila at initial concentration of 1 × 10⁵ cells mL⁻¹. Fruits were wounded and inoculated with 10 μL suspension of NSA or SA C. oleophila at 1 × 10⁷ cells mL⁻¹. Data presented are the means of pooled data. Error bars indicate standard deviations of the means (n = 9). Different letters at each time point indicate significant differences according to Student's t-test (P<0.05).

Discussion

Biocontrol agents must perform reliably under a wide range of environmental conditions to be commercially acceptable (Droby et al., 2009). Improving the stress tolerance of postharvest biocontrol agents is one approach being explored to address this need (Teixidó et al., 2005; Li & Tian, 2006). Previous research demonstrated that exposure to a sublethal heat stress can improve the stress tolerance of yeast to a subsequent oxidative stress (as represented by exposure to hydrogen peroxide) and that exogenous glycine betaine can confer oxidative stress tolerance. The induced resistance resulted in faster initial growth rates of yeasts in inoculated wounds of apple and increased efficacy. Heat shock pretreatment induced the upregulation of a TPS1 (trehalose-6-phosphate synthase) gene in M. fructicola (Liu et al., 2011a). Glycine betaine pretreatment resulted in a lower level of protein oxidation and an increase in the activity of several antioxidant enzymes (catalase, superoxide dismutase, and glutathione peroxidase) when Cystofilobasidium infirmominiatum cells were exposed to a subsequent oxidative stress (Liu et al., 2011b). In both studies, a lower level of ROS was observed when yeasts were subsequently exposed to hydrogen peroxide after the pretreatment.

Cross-protection or tolerance to a variety of abiotic stresses can be obtained when microorganisms are transiently exposed to a sublethal stress. This phenomenon has been observed in different types of microorganisms, including yeasts (Mitchell et al., 2009; Liu et al., 2011a), filamentous fungi (Delgado-Jarana et al., 2006; Rangel et al., 2008), and bacteria (Teixidó et al., 2005; Mitchell et al., 2009). In this study, transient exposure of C. oleophila to a sublethal oxidative stress treatment (5 mM H₂O₂ for 30 min) provided a significant increase in tolerance to a subsequent higher level of oxidative stress (50 mM H₂O₂). Similar results have been reported for other microorganisms, such as Saccharomyces cerevisiae (Davies et al., 1995), Fusarium decemcellulare (Medentsev et al., 2001), and Listeria monocytogenes (Lou & Yousef, 1997). In addition to increasing oxidative stress tolerance, the mild oxidative pretreatment also provided increased tolerance to high temperature (40 °C) indicating some level of cross-protection. The present results support previous reports on cross-protection in S. cerevisiae induced by sublethal oxidative stress against freeze-thaw stress (Park et al., 1997) and hyperosmotic stress (Lu et al., 2005).

Exposure of yeast cells to abiotic stresses such as oxidative and temperature stress leads to the production of intracellular ROS which can cause cellular damage resulting in decreased viability (Reverter-Branchat et al., 2004). Sublethal oxidative stress treatment (5 mM H₂O₂ for 30 min) did not significantly increase ROS accumulation in C. oleophila cells compared with untreated cells at time 0; however, the percentage of cells with a visible intracellular ROS level increased in both SA and NSA cells with time of exposure to a subsequent oxidative (50 mM H₂O₂) or high temperature (40 °C) stress. Moreover, the percentage of SA cells exhibiting ROS accumulation was significantly lower than the percentage in NSA cells. These results support previous reports that intracellular levels of ROS in yeast cells were dependent on the level and duration of exposure to an oxidative stress (Pekmez et al., 2008; Liu et al., 2011a, b). Importantly, the lower percentage of ROS in SA cells compared with NSA control cells was reflective of the higher viability of SA yeast cells exposed to subsequent oxidative or heat stress.

Increased levels of protein and lipid oxidation occur in cells exhibiting ROS accumulation and have a deleterious effect on structure and function of protein and lipid (Lee et al., 2001; Saharan & Sharma, 2010). The presence of carbonyl groups of proteins can be used as a marker of oxidative protein damage (Abegg et al., 2010), while MDA content serves as a marker of lipid peroxidation (Li et al., 2010). Data from the present study indicate lower carbonyl and MDA accumulation in SA cells compared with NSA cells when yeast was exposed to a subsequent oxidative or heat stress. Sublethal oxidative treatment reduced oxidative damage and increased the viability of C. oleophila cells exposed to subsequent oxidative and heat stresses. The lower level of oxidative damage may be because of the stimulation of antioxidant systems in SA.

The enzymatic detoxification of ROS is dependent on the upregulation of antioxidant genes such as peroxisomal catalase (Nakagawa et al., 2010), cytochrome c peroxidase (Demir & Koc, 2010), peroxiredoxin TSA1 (Iraqui et al., 2009), thioredoxin reductase (Demir & Koc, 2010), glutathione peroxidase (Collinson et al., 2002), glutathione reductase (Grant et al., 1996), and glucose-6-phosphate dehydrogenase (Lee et al., 2001). Upregulation of glutathione oxidoreductase expression in S. cerevisiae increased resistance to hydroperoxides including H₂O₂, and glutaredoxins were active as glutathione peroxidases (Collinson et al., 2002). Peroxiredoxin TSA1 has been reported to be the key peroxidase suppressing genome instability and protecting S. cerevisiae from cell death (Iraqui et al., 2009). Saccharomyces cerevisiae mutants lacking the mitochondrial antioxidant genes peroxiredoxin,thioredoxin reductase, and cytochrome c peroxidase were more sensitive to the thiol oxidant diamide than the wild type (Demir & Koc, 2010). Moreover, S. cerevisiae mutants lacking either cytosolic CuZnSOD or MnSOD, or both SODs also had lower activities of antioxidant enzymes such as catalase, glucose-6-phosphate dehydrogenase, and glutathione reductase, as well as a higher ROS level (Lee et al., 2001).

This study identified seven antioxidant genes that were more highly expressed in SA yeast cells. Importantly, the higher expression of these genes was maintained under subsequent oxidative and heat stress. Data indicate that the SA-induced antioxidant gene expression in C. oleophila may be a key factor in the improvement of stress tolerance, lower ROS accumulation, and the lower level of oxidative damage to proteins and lipids observed when cells were exposed to subsequent oxidative stress or high temperatures.

Biocontrol efficacy of C. oleophila was also enhanced by the sublethal oxidative stress treatment. High population density is an advantage to microbial antagonists in competing for nutrients and space (Wisniewski et al., 2007; Sharma et al., 2009). Results indicate that SA cells of C. oleophila multiplied more rapidly than NSA cells in wounds of apple fruits during log phase of growth. This time period is important for pathogen germination and infection (Li et al., 2008) and may explain the increased efficacy observed with SA cells against P. expansum and B. cinerea compared with the use of NSA cells. Additionally, the in vitro growth assay showed that the SA yeast grew more rapidly in YPD at pH 4, which is the same pH value as ‘Golden Delicious’ apple flesh. Oxidative stress resistance may be necessary for yeast to remain viable and maintain biocontrol efficacy in wounded fruit (Castoria et al., 2003). Yeast biocontrol agents underwent a significant oxidative burst when applied to apple (Macarisin et al., 2010) or grapefruit (Hershkovitz et al., 2011) and induced host tissues to produce elevated levels of ROS. The oxidative stress tolerance induced in SA cells observed in this study may allow them to survive and compete better in a high oxidative stress environment.

In summary, this study provides additional information on induced stress tolerance in biocontrol yeast by demonstrating that sublethal oxidative stress can directly improve stress tolerance to a subsequent stronger oxidative stress and provide cross-protection to other stresses such as high temperature and low pH. Lower levels of ROS accumulation, as well as protein and lipid oxidation, were observed in the SA yeast cells compared with NSA cells and may be because of the activation of antioxidant enzyme genes in SA cells. Stress adaptation of formulated yeast biocontrol agents may represent a viable approach for improving their reliability under the variable conditions found in orchards, packinghouses, and storage facilities.

Additional corresponding author

Shiping Tian, Key Laboratory of Photosynthesis and Environmental Molecular Physiology, Institute of Botany, Chinese Academy of Sciences, Beijing 100093, China. Tel.: [phone]; fax: [phone]; e-mail:

Acknowledgements

This research was supported by a grant (IS-4268-09) from the US–Israel Binational Agricultural Research and Development (BARD) Fund to S.D. and M.W., and a grant (31030051) from National Natural Science Foundation of China (NNSFC) to Tian SP.

Author notes

Editor: Christoph Tebbe