OUP user menu

Identification of differentially expressed genes associated with changes in the morphology of Pichia fermentans on apple and peach fruit

Stefano Fiori, Barbara Scherm, Jia Liu, Robert Farrell, Ilaria Mannazzu, Marilena Budroni, Bianca E. Maserti, Michael E. Wisniewski, Quirico Migheli
DOI: http://dx.doi.org/10.1111/j.1567-1364.2012.00829.x 785-795 First published online: 1 November 2012

Abstract

Pichia fermentans (strain DISAABA 726) is an effective biocontrol agent against Monilinia fructicola and Botrytis cinerea when inoculated in artificially wounded apple fruit but is an aggressive pathogen when inoculated on wounded peach fruit, causing severe fruit decay. Pichia fermentans grows as budding yeast on apple tissue and exhibits pseudohyphal growth on peach tissue, suggesting that dimorphism may be associated with pathogenicity. Two complementary suppressive subtractive hybridization (SSH) strategies, that is, rapid subtraction hybridization (RaSH) and PCR-based subtraction, were performed to identify genes differentially expressed by P. fermentans after 24-h growth on apple vs. peach fruit. Gene products that were more highly expressed on peach than on apple tissue, or vice versa, were sequenced and compared with available yeast genome sequence databases. Several of the genes more highly expressed, when P. fermentans was grown on peach, were related to stress response, glycolysis, amino acid metabolism, and alcoholic fermentation but surprisingly not to cell wall degrading enzymes such as pectinases or cellulases. The dual activity of P. fermentans as both a biocontrol agent and a pathogen emphasizes the need for a thorough risk analysis of potential antagonists to avoid unpredictable results that could negatively impact the safe use of postharvest biocontrol strategies.

Keywords
  • biocontrol
  • subtractive hybridization
  • yeast pathogenicity
  • postharvest
  • environmental risk
  • morphogenesis

Introduction

Biological control represents an effective alternative to the application of synthetic chemical fungicides for the management of postharvest decay of fruit caused by a wide range of fungal pathogens (Barkai-Golan, 2001; Janisiewicz & Korsten, 2002; Passoth et al., 2006; Droby et al., 2009). Yeasts are particularly suitable as biocontrol agents against several kinds of rot fungi (Wisniewski et al., 2007) because their activity does not usually rely on the production of toxic metabolites (Smilanick, 1994; Spadaro et al., 2002) but rather on their ability to survive on the fruit surface and to rapidly colonize fruit wounds, competing with pathogens for space and nutrients. Other biocontrol mechanisms, such as the production of extracellular hydrolases, the induction of host-resistance response (Wisniewski et al., 2007), and the capability to form a biofilm on the fruit surface or within wounds (Scherm et al., 2003; Severinatne et al., 2008), have been proposed but remain still unexplored at the molecular level. Such lack of information is a significant limitation to the choice and selection of antagonists. A detailed understanding of yeast biology and of their interactions with hosts and pathogens is an essential prerequisite to their use as biocontrol agents. Antagonists should be thoroughly vetted prior to widespread use at high concentrations because human and environmental safety issues are paramount (Migheli, 2001).

A small number of yeast species display dimorphism, that is, the capacity to undergo either budding or a pseudohyphal growth not only in response to different environmental factors such as nutrient availability, temperature and pH (Reynolds & Fink, 2001) but also in response to quorum sensing molecules, hormone-like signals that influence microorganism behavior, inducing a physiologic and morphological shift when cell density changes (Sprague & Winans, 2006; Granek & Magwene, 2010). Dimorphic transition in yeasts can often be related to pathogenicity, where specific morphologies can cause disease. As an example, Candida albicans (C.P. Robin) Berkhout is a commensal fungus that usually colonizes the skin, oral cavity, urogenital, and intestinal system of humans in its yeast-like morphology but may become a serious opportunistic pathogen by switching to a pseudohyphal/hyphal morphology that is more adapted for invasion through host epithelial tissue (Wingard et al., 1979; Rooney & Klein, 2002; D'Enfert, 2009). Its morphogenesis is predominantly determined by environmental conditions and associated with different signaling pathways regulating gene expression (Han et al., 2011).

Another ascomycete, Paracoccidioides brasiliensis (Splend.) F. P. Almeida, which is non-invasive when growing saprophytically in hyphal form in the soil, can switch to a pathogenic yeast-like form after conidia are inhaled into the lungs of human or other mammalian immunocompromised hosts (Felipe et al., 2005). The severity of the disease can range from a localized lung infection to a lethal systemic disease (Felipe et al., 2005). Although not considered a typical dimorphic fungus, the causal agent of cryptococcal meningitis, Cryptococcus neoformans (San Felice) Vuill., can exhibit several different morphologies including yeast-like, chlamydospores, pseudohyphae, and hyphae, but is usually present in the yeast form when infecting human and animal tissues (Lin & Heitman, 2006; Lin, 2009).

The relationship between yeast dimorphism and pathogenicity occurs not only in connection with human or animal disease but also in plant pathology. Giobbe et al. (2007) demonstrated that a Pichia fermentans Lodder [anamorph: Candida lambica (Lindner & Genoud) Uden & H.R. Buckley ex S.A. Mey. & Ahearn (1983)] strain (DISAABA 726 = DBVPG 3627) isolated from wine must is a very effective biocontrol agent against brown rot and gray mold caused by Monilinia fructicola (G. Winter) Honey and Botrytis cinerea Pers., respectively, in artificially wounded apple fruit (Malus× domestica ‘Golden Delicious’ and ‘Renetta’). When applied to wounded peach fruit (Prunus persica L. Batsch), however, P. fermentans becomes a pathogen causing complete decay of the fruit after few days even in the absence of M. fructicola or other pathogens. Pichia fermentans produces large single yeast-like cells with prominent vacuoles during the colonization of apple tissue while exhibiting pseudohyphal growth on peach tissue. This suggests that pseudohyphal growth may play a key role in the pathogenicity of P. fermentans on peach fruit. However, the genetic regulation of the observed dimorphism has not been studied.

The aim of this study was to identify yeast genes associated with pathogenicity and the transition from a yeast-like to pseudohyphal morphology when P. fermentans is grown on peach. Subtractive suppressive hybridization (SSH) was used to identify genes that were uniquely expressed in P. fermentans grown on peach by subtracting out genes that were commonly expressed in both peach and apple. Two complementary SSH methodologies were used: a rapid subtraction hybridization (RaSH) approach, where unsubtracted cDNAs are selected by matching their ends with the plasmid vector (Jiang et al., 2000); and a PCR-Select cDNA Subtraction Kit (Clontech, Palo Alto, CA), where unsubtracted cDNAs are highlighted by two consecutive PCR.

Materials and methods

Culture conditions and experimental design

A single, fresh colony of P. fermentans DISAABA 726 (=DBVPG 3627) was inoculated into a flask containing 30 mL of 1% yeast extract, 2% bacto peptone, 2% dextrose broth (YPD) and incubated for 24 h at 25 °C on a rotary shaker (200 g). Cells were then recovered by centrifuging at 3000 g for 5 min, washed and resuspended in sterile Ringer's solution (H2O, 0.9% NaCl) and brought to a final concentration of 1 × 109 cells mL−1 by direct counting with a hemocytometer. One hundred microliters of this suspension was spread between two autoclaved circular sheets of cellophane membrane backing sheets (Bio-Rad, Hercules, CA), permeable to liquids and nutrients, which were then placed between either halved apple or peach fruits (Supporting Information, Fig. S1). Fruit selected for the study were at the same maturity and had been disinfected in sodium hypochlorite (0.8% as chlorine) and air-dried prior to being halved. A total of 66 each of apple and peach fruit were treated to recover a high quantity of yeast cells. The fruit was stored for 24 h at 25 °C at 85 ± 5% relative humidity and then the yeast was scraped from the cellophane sheets with a sterile spatula, collected in 1.5-mL Eppendorf tubes and immediately frozen in liquid nitrogen and stored at −80 °C until further processing.

RNA extraction and cDNA reverse transcription

Total RNA was obtained from 0.3 g of yeast cells obtained from either the apple or peach by grinding them in liquid nitrogen and then performing the extraction with the Pure Link Micro-to-Midi Kit (Invitrogen, Carlsbad, CA) using TRIzol® (Invitrogen). For genomic DNA removal, an in-column DNase digestion with the RQ1 RNase-Free DNase (Promega, Wisconsin, MA) set was carried out according to manufacturer's instructions. cDNA reverse transcription was later performed, starting from 1 μg total RNA, using the Super SMART PCR cDNA synthesis kit (Clontech) according to the manufacturer's directions.

Rapid subtraction hybridization

The rapid subtraction hybridization (RaSH) protocol was based on a method previously reported by Scherm et al. (2009) with slight modifications. Briefly, cDNA of P. fermentans grown on apple fruit was used as the driver to subtract all the genes expressed in common with the tester (cDNA of P. fermentans grown on peach fruit) and highlight those genes expressed only, or at significantly higher levels, in the latter (yeast growing on peach). cDNA of both samples was digested with EcoRII (New England Biolabs, Beverly, MA) at 37 °C for 4 h, purified with the Qiaquick PCR Purification Kit (Qiagen, Hilden, Germany) and then ligated with the adaptors XE-14, XEA-13, and XET-13, complementary to two specific primers XEA-18 and XET-18 (Table 1), by overnight incubation at 4 °C. The fragments were then exponentially amplified by PCR with an Applied Biosystems 9600 thermocycler (Life Technologies, Carlsbad, CA) under the following conditions: TopTaq DNA polymerase reaction buffer (Qiagen), MgCl2 (2.5 mM), dNTP (0.2 mM), XEA-18 (1 mM), XET-18 (1 mM), TopTaq DNA Polymerase reaction buffer (Qiagen), and 1 μL of cDNA. The PCR program was set as follows: 72 °C, 5 min; 25 cycles of 1 min at 94 °C; 1 min at 55 °C; 1 min at 72 °C, with a final extension step of 10 min at 72 °C. PCR products were purified with the Qiaquick PCR Purification Kit (Qiagen, Hilden, Germany) and then 1 μg of amplified tester fragments was digested with XhoI (New England Biolabs) at 37 °C for 4 h and mixed with driver fragments in 10 μL of hybridization solution [0.5 M NaCl, 50 mM Tris pH 7.5, 0.2% SDS, 40% (vol/vol) formamide]. One hundred nanograms of tester was mixed with 3 μg of driver (1 : 30 tester/driver), and after boiling for 5 min, the subtraction mix was incubated for 48 h at 42 °C. The hybridization mix was diluted with sterile water to 100 μL, purified and adjusted to a final volume of 30 μL. Eight microliters of the mix was incubated overnight at 4 °C with XhoI-digested pBlueScript SK (±) vector, dephosphorylated with Shrimp alkaline phosphatase (Fermentas, Vilnius, Lithuania) and then transformed into competent DH5α Escherichia coli cells by chemical transformation. White ampicillin-resistant colonies were directly PCR-amplified with LacZ-plasmid specific primers (RaSH-F and RaSH-R) and visualized on 1.5% agarose gel stained with ethidium bromide using Gel-Doc software (Bio-Rad) to verify the presence and length of the fragments. The newly amplified products were then transferred in duplicate to the same Hybond-N membrane (Amersham Biosciences, Freiburg, Germany) and each blot was hybridized with two different chemiluminescent probes (obtained from driver and tester cDNAs, respectively) to confirm their differential expression and eliminate possible false-positive clones. Labeling and detection were conducted using DIG-High Prime DNA Labeling and Detection Starter Kit II (Roche Applied Science, Mannheim, Germany) and blotted fragments showing a more intense detection with chemiluminescent tester probe (cDNA of P. fermentans grown on peach fruit), hence indicating a higher number of repeated gene copies in the tester cDNA, were chosen and sequenced by C.R.I.B.I. BIO Molecular Research, University of Padova (Italy).

View this table:
Table 1

Oligonucleotide sequences used for the RaSH protocol and for the semi-quantitative PCR

NameForwardReverse
Adaptors
XE-145′-CTGATCACTCGAGA-3′/
XEA-135′-CCAGGTCTCGAG-3′/
XET-135′-CCTGGTCTCGAG-3′/
RaSH primers
XEA-185′-TGATCACTCGAGACCAGG-3′/
XET-185′-TGATCACTCGAGACCTGG-3′/
RaSH5′-ACTCACTATAGGGCGATTG-3′5′-GGAATTCGATATCAAGCTTATC-3′
Semi-qPCR primers
18SrRNA5′-TACATGCGCAAAGCCCCGACT-3′5′-TGCCCCCGACCGTCCCTATT-3′
P-A215′-CGTGATGTTCTGCACCTGCCCAG-3′5′-CTCCGTTTCCGTTCCATTCCCACC-3′
P-A365′-AACGGCAGCACCATAAGCGACA-3′5′-GCCCGGGCAGGTACCTTTGAA-3′
P-B075′-TGGCTTGCACGCATGGAAGGG-3′5′-ACCAGAAAGGTCAGCCTTTGCACA-3′
P-B195′-TCACCGCTTGGGTTGGCAACG-3′5′-TGACCACGGTGCTGATGTCGTC-3′
P-C555′-CGGTGTCAGAGTTGCAGGTGTCG-3′5′-GCTGCCTTGACACCAGATAAGGCC-3′
P-C635′-ACCAGACACCTGGGTTGTCAGCC-3′5′-TCCTGCGTTAGCTACCGTTCCC-3′
P-F555′-CCATTGTCCCCTCTGGTGCATCC-3′5′-TTGCAAGGAACTGGCAGCGTTTG-3′
P-Thi5′-TGCTGGTATGGAATTGAGT-3′5′-CGTAGATGTTGAGAGCTT-3′
P-Pyr5′-CTGACGCTTGTAGATGGA-3′5′-AAATAGGGGAAGAGGGGA-3′
M-B415′-GGTTGGTGACGGCTCTGACAGT-3′5′-TCAGTGTTGCCAGCCTTTCTTGC-3′
M-C515′-CCCCCACGGGCTGTCTTTCGT-3′5′-CCGCAATGCATGGGAAATCTAGCA-3′
M-G435′-GGGGAAGACATGGACAAA-3′5′-CAAAGGTGAAAAGAGCAGG-3′

Vector sequences were then trimmed and the remaining sequences were used to perform a blastx search in the NCBI database (http://www.ncbi.nlm.nih.gov) to identify them on the basis of the closest ortholog.

PCR-select cDNA Subtraction Kit (SSH)

An additional subtraction method was performed using the PCR-Select cDNA Subtraction Kit (Clontech) on yeast cells grown as previously described and following the manufacturer's protocol. Two different subtractions were conducted: a forward one, where the cDNA of P. fermentans grown on peach fruit was the tester and the cDNA of P. fermentans grown on apple fruit was the driver, and vice versa. After hybridization and two consecutive PCRs, a mixture enriched for differentially expressed cDNAs was obtained. As previously described by Bassett et al., 2006, the PCR products in this mixture were ligated into plasmids with the TOPO® TA Cloning® Kit (Invitrogen) and then transformed into One Shot® TOP10 competent E. coli (Invitrogen) by chemical transformation. White kanamycin-resistant colonies were directly PCR-amplified with M13 forward and reverse primers and the amplified fragments were loaded into a 1.5% agarose gel and visualized with SYBRGold (Molecular Probes, Eugene, OR) in a Storm 860 fluorescence image analyzer (GE Healthware, Piscataway, NJ) to verify the presence and length of the fragments in each clone. All colonies containing fragments longer than 600 pb were selected for further analysis and following extraction with a QIAprep Spin Miniprep kit (Qiagen), individual clones were sequenced in both directions with M13 forward and reverse primers by Macrogen Corp. (Rockville, MD). All sequences were then subjected to blastx on NCBI.

Semi-quantitative PCR analysis

As PCR-based subtraction approaches may result in amplification bias (Jiang et al., 2000), semi-quantitative PCR was performed to confirm differential expression of genes putatively related to dimorphism. The 18S ribosomal RNA gene was used as a reference gene. Specific primers were designed for each fragment using NCBI's Primer-blast tool and the 18S ribosomal RNA gene (Table 1). PCR was carried out with an Applied Biosystems 9600 thermocycler in 60 μL reaction containing 1 × TopTaq DNA Polymerase reaction buffer (Qiagen), MgCl2 (2.5 mM), dNTPs (0.2 mM each), forward and reverse primer (1 mM each), TopTaq DNA Polymerase reaction buffer (Qiagen) and 3 μL of cDNA (100 ng for both cDNA templates). The PCR program was set as follows: 72 °C, 5 min; 30 cycles of 30 s at 94 °C, 30 s at 59 °C, and 40 s at 72 °C with a final extension step of 10 min at 72 °C. After the 15th cycle, an 8-μL aliquot was removed from both samples and subjected to a final extension of 10 min at 72 °C in another thermocycler. The same operation was repeated every three cycles between the 18th and the 30th. The amplified samples collected at different cycles were then loaded side by side on a 1.5% agarose gel and visualized with SYBRGold (Molecular Probes) in a Storm 860 fluorescence image analyzer (GE Healthware) to verify their differential expression.

Results

The main objective of this study was to identify genes that were associated with the observed dimorphism of P. fermentans when grown on peach fruit (pseudohyphal) vs. apple fruit (yeast-like). The first morphological changes are evident after 3 days growth and become clearly evident after 6 days (Fig. 1). Therefore, sampling at later time periods would most likely have identified genes associated with the physiology of the new morphological phenotype rather than the initial changes in gene expression associated with the regulatory switch to pseudohyphal growth.

Figure 1

Morphological difference of Pichia fermentans strain 726 grown between two autoclaved circular sheets of cellophane membrane placed between either apple or peach fruit. On apple fruit Pichia fermentans morphology remains yeast-like after 72 (A1) and 144 (A2) h growth at RT, while on peach fruit a morphological switch from yeast-like to pseudohyphal growth is observed on the third day of growth (B1) and becomes clearer on day 6 (B2).

Two complementary suppressive subtractive hybridization (SSH) approaches were used: RaSH and PCR-Select. RaSH is characterized by a mass-driven subtraction with a 1 : 30 ratio of tester and driver during hybridization, and unsubtracted cDNAs are selected by matching their ends with the plasmid vector (Jiang et al., 2000). Using RaSH, a total of 450 clones were obtained and further analyzed by DNA hybridization. Fourteen fragments showed stronger hybridization with the tester cDNA chemiluminiscent probe (Fig. 2), suggesting that they were differentially expressed during pseudohyphal growth. These 14 fragments were sequenced and subjected to a blastx search.

Figure 2

Southern blotting for confirmation of RaSH results: clones were directly amplified with specific RaSH primers and then transferred in duplicate on the same membrane. Each blot was hybridized with two different chemiluminescent probes obtained from cDNAs of Pichia fermentans 726 grown for 24 h on peach (a) or on apple (b) fruit to confirm their differential expression. Fragments showing a stronger hybridization with probe A confirm their higher expression during pseudohyphal growth on peach fruit.

The genes differentially expressed during growth on peach could be roughly clustered into the following categories of putative functions: stress response (heat shock protein 70; 3,4-dihydroxy-2-butanone 4-phosphate synthase; glutathione peroxidase; NADH dehydrogenases and succinyl-CoA synthetase), glycolysis (phosphoglycerate kinase and pyruvate kinase) and amino acid metabolism (NADP+ glutamate dehydrogenase and acetolactate synthase) (Table 2).

View this table:
Table 2

Characteristics and putative function of clones obtained from suppressive subtractive hybridization, confirmed by Southern blotting for RaSH and by semi-quantitative PCR for PCR-Select cDNA Subtraction Kit

ClonePredicted functionIdentity (%)e valueFrequencyGenBank code
RaSH
Pichia fermentans on peach fruit
PI-01Glutathione peroxidase (Candida tenuis) 643e−141JK730582
PI-02Phosphoglycerate kinase (Candida boidinii) 916e−291JK730583
PI-573,4-dihydroxy-2-butanone 4-phosphate synthase (Lodderomyces elongisporus) 733e−221JK730584
PI-144Acetolactate synthase (Ogataea angusta) 953e−241JK730585
PII-01Putative pyruvate kinase (Ogataea angusta) 746e−241JK730586
PII-06GTPase cytoplasmic elongation factor 1 alpha (Pichia fermentans) 952e−211JK730587
PII-26Elongation factor 2 (Ogataea angusta) 942e−263JK730588
PII-59Succynil-CoA synthetase subunit (Ogataea angusta) 902e−291JK730589
PII-60Heat shock protein 70 (Ogataea angusta) 792e−712JK730590
PII-65NADP(+)-glutamate dehydrogenase (Ogataea angusta) 842e−311JK730591
PIII-02NADH dehydrogenase (Ogataea angusta) 748e−211JK730592
PCR-Select™ cDNA Subtraction
Pichia fermentans on peach fruit
P-A21C-4 methyl sterol oxidase (Saccharomyces cerevisiae) 707e−1442JK730593
P-A36Heat shock protein 70-2 (Candida tenuis) 915e−961JK730594
P-B07Glycerol 3-phosphate dehydrogenase NAD+ (Candida glycerinogenes) 866e−891JK730595
P-B19O-acetyl homoserine-o-acetylserine sulfhydrylase (Pichia pastoris) 651e−1052JK730596
P-C55Thiazole synthetase (Ogataea angusta) 667e−863JK730597
P-C63Iron transport multicopper oxidase precursor (Ogataea angusta) 644e−915JK730598
P-F55Enolase (Candida glycerinogenes) 702e−524JK730599
P-ThiThiamine biosynthesis enzyme (Ogataea angusta) 837e−271JK730600
P-PyrPyruvate decarboxylase (Ogataea angusta) 748e−344JK730601
Pichia fermentans on apple fruit
M-B41Peroxisomal primary copper amine oxidase (Ogataea angusta) 688e−942JK730602
M-C51OPT Oligopeptide transporter (Debaryomyces hansenii) 626e−701JK730603
M-G43Glutamate synthase (Ogataea angusta) 873e−1332JK730604

With the PCR-Select™ cDNA Subtraction Kit (Clontech), amplification of cDNAs is subsequent to hybridization. This approach enhances the ability to obtain longer fragments that can be more readily identified by homology to other genes. The use of SSH yielded 560 clones for the forward subtraction (cDNA of P. fermentans grown on peach as tester) and 420 for the reverse subtraction (cDNA of P. fermentans grown on apple as tester). Among these clones, 74 and 56 fragments, respectively, which were >600 bp in length, were chosen for sequencing and subjected to blastx. The expression of 23 fragments of cDNA of P. fermentans grown on peach as tester, and five fragments of cDNA of P. fermentans grown on apple as tester were then confirmed by semi-quantitative PCR (Fig. 3).

Figure 3

Semi-quantitative PCR confirmation of expression levels for 12 of the cloned fragments obtained using a PCR-Select cDNA Subtraction Kit (Clontech). Specific primers were designed to specifically amplify each of the fragments using the cDNA of both Pichia fermentans 726 grown on apple and on peach fruit after 24 h as templates. The PCR products were obtained after 25 cycles of amplification.

Differentially expressed genes, identified using SSH could be grouped into the following functional categories: stress response (heat shock protein); alcoholic fermentation (pyruvate decarboxilase, enolase and glycerol 3-phosphate dehydrogenase NAD+, the two latter being also involved in stress response); amino acid metabolism (o-acetyl homoserine-o-acetylserine sulfhydrylase, peroxisomal primary copper amine oxidase, and glutamate synthase); thiamine biosynthesis (thiamine biosynthesis enzyme and thiazole synthetase); transport (iron transport multicopper oxidase and OPT oligopeptide transporter); and cellular membrane biosynthesis (C-4 methyl sterol oxidase) (Table 2).

Discussion

A strain of P. fermentans is an effective antagonist against M. fructicola on apple fruit, but exhibits unexpected pathogenic traits on peach fruit (Giobbe et al., 2007). On peach, P. fermentans causes a rapid decay of tissues and undergoes a rapid increase in growth and a morphological change from yeast-like to pseudohyphal growth (Figs S2 and S3), suggesting that the latter form plays a role in governing its pathogenic behavior. To better understand the mechanisms involved in the dimorphic transition/pathogenicity of P. fermentans, we utilized two subtractive methods to identify genes expressed differentially during growth on apple vs. peach fruit.

Pichia fermentans represents one of the few examples ever reported of a pathogenic yeast species on plants (Giobbe et al., 2007). Therefore, speculation on the potential function of genes identified in the present study can be only inferred by comparison with dimorphic yeasts associated with mammalian systems. The functions of the identified genes are diverse but can be generally grouped into the categories of stress response, glycolysis, amino acid biosynthesis, and alcoholic fermentation (Table 1).

The higher expression of stress genes in P. fermentans when it is grown on peach fruit may indicate that peach fruit as a substrate represents a stressful environment that triggers a shift to the pseudohyphal growth form. Heat shock proteins are expressed in response to an array of stresses, including hyperthermia, exposure to oxygen radicals, heavy metals, ethanol, and amino acid analogs (De Maio, 1999). Felipe et al. (2005) reported that heat shock genes were among the upregulated genes possibly involved in P. brasiliensis dimorphism. They suggested that stress response was because of the temperature of the mammalian host, which triggered a change in growth form. While an increase in temperature within the peach fruit tissues during colonization by P. fermentans is clearly not applicable, the upregulation of heat shock genes may reflect a general stress response that plays a role in triggering a change from the yeast-like to pseudohyphal growth form. In stress conditions, de novo synthesis of proteins is required and heat shock proteins may assist protein folding preventing aggregation of unfolded polypeptides. In performing its task as a molecular chaperone, the HSP70s have several essential roles important to dimorphic, pathogenic fungi (Moraes Nicola et al., 2005).

The upregulation of a putative enolase gene may also reflect a stress response because an isoprotein of enolase was reported to be a heat shock protein expressed by Saccharomyces cerevisiae Meyen in response to heat stress (Franklyn & Warmington, 1994). In C. albicans, enolase does not increase during heat stress although its amino acid sequence is quite similar to hsp70 class proteins (Franklyn & Warmington, 1994). On the other hand, enolase belongs to the so-called ‘moonlighting proteins,’ having more than one catalytic function in a single polypeptide chain (Jeffery, 1999): in S. cerevisiae, enolase is associated with vacuolar protein trafficking machinery and secretory pathways (Gancedo & Flores, 2008).

Another over-expressed gene, glycerol 3-phosphate dehydrogenase NAD+, may reflect an osmotic stress response for P. fermentans when grown on peach fruit because stimulation of glycerol biosynthesis and its accumulation under osmotic stress growth conditions is a common mechanism in most yeasts (Hua et al., 2008). Moreover, strains of C. albicans deficient in glycerol-3-phosphate dehydrogenase were not able to induce damage in epithelial cells even though they maintained their invasive behavior, suggesting that glycerol accumulation within the hyphae may contribute to cell damage, likely influencing turgor pressure and morphogenic plasticity (Naglik et al., 2011). We also observed over-expression of a 3,4-dihydroxy-2-butanone 4-phosphate synthase gene, involved in the synthesis of riboflavin, a compound typically overproduced in some yeast species in response to oxidative stress and iron deficiency (Boretsky et al., 2007). The higher expression of NADH dehydrogenase, one of the three energy-transducing enzymes of the mitochondrial electron transport chain, suggests an increase in respiration that could also imply a higher production of reactive oxygen species (ROS), likely further enhanced by growth of the yeast on wounded fruits (Macarisin et al., 2010). Thus, oxidative stress may also be associated with the shift toward pseudohyphal morphology. In this regard, Müller et al. (2008) reported that ROS production from NADH dehydrogenase was higher during reverse electron transfer in the presence of elevated levels of succinate. Glycerol 3-phosphate dehydrogenase NAD+ and succinate dehydrogenase serve as electron donors to reduce NAD+ to NADH. Increased oxidative stress in the yeast grown on peach could also explain the higher expression of the antioxidant enzyme glutathione peroxidase, which protects cells from oxidative damage by reducing hydrogen peroxide while converting two molecules of reduced glutathione into one molecule of glutathione disulfide. In C. albicans, the amount of the reduced form of glutathione becomes lower during the yeast-to-mycelium transition (González-Párraga et al., 2005).

Han et al. (2011) suggested that higher glycolytic enzyme expression was related to the hyphal growth form of C. albicans because several glycolytic genes are regulated by signaling pathways associated with morphogenesis. Although expression of glycolytic enzyme genes is not regulated tightly and fluctuates during the yeast-to-hyphal switch in C. albicans (Swoboda et al., 1994), differences in their activities have been correlated with yeast dimorphism (Schwartz & Larsh, 1982). Phosphoglycerate kinase and pyruvate kinase are among the upregulated enzymes present in pseudohyphal growth forms (Yin et al., 2004; Enjalbert et al., 2006); elongation factor 1 also seems to play an important role in the upregulation of glycolytic enzymes and hence in morphogenesis (Doedt et al., 2004; García-Sánchez et al., 2005; Sexton et al., 2007). Reduced expression of elongation factor 1 or the deletion of this gene results in an inability to form germ tubes and true hyphae (Lo et al., 1997; Stoldt et al., 1997). A C. albicans hyphal-deficient mutant efg1Δ/Δ exhibited a lack of adhesion and the inability to invade and damage epithelial tissues (Naglik et al., 2008).

Additional evidence of the importance of glycolytic enzymes during morphogenesis was reported by Shirtliff et al. (2009), who observed that enolase and pyruvate kinase were downregulated when C. albicans was exposed to farnesol, an acyclic sesquiterpene alcohol that works as a quorum sensing signal able to block hyphal differentiation of the yeast.

Even the expression of oxidoreductase enzymes, such as C-4 methyl sterol oxidase (Ying-Ying et al., 2005), involved in the sterol biosynthetic pathway and hence in cellular membrane composition, and iron transport enzymes, such as multicopper oxidase (Nett et al., 2009), are lower when C. albicans hyphal formation is inhibited by farnesol. The cited reports may explain why all these enzyme genes are upregulated in P. fermentans when the pseudohyphal form is induced by growing it on peach fruit.

The higher expression of pyruvate decarboxylase also suggests that P. fermentans may exhibit a higher level of fermentation when growing on peach fruit than on apple. This enzyme is indeed involved in alcoholic fermentation and its catalytic activity depends on the presence of the cofactor thiamine diphosphate (König et al., 2009), a derivative of thiamine. Synthesis of the latter requires both thiamine biosynthesis enzymes and thiazole synthetase, a gene that was more highly expressed in the pseudohyphal form of P. fermentans growing on peach fruit than the yeast-like form growing on apple. Thiamine derivatives could also have noncoenzymatic functions and have been suggested to play a role in the regulation of gene expression in response to adverse environmental conditions through as yet unidentified signal transduction pathways (Tylicki & Siemieniuk, 2011).

It is known that morphogenesis in S. cerevisiae and in C. albicans can be controlled by the production of aromatic alcohols derived from amino acids (Chen & Fink, 2006). Methionine itself plays a role in the induction of pseudohyphal growth in C. albicans as it was observed that the presence of methionine was the signal for the onset of the transition to filamentous growth (Maidan et al., 2005). A similar scenario is also plausible in P. fermentans. It was recently reported that methionine, probably after its conversion to methanol, induces dimorphism in vitro (Sanna et al., 2012). Over-expression of o-acetyl-homoserine sulfhydrylase could therefore be implicated as a regulator of pseudohyphal growth as it is involved in methionine biosynthesis. It is also worth noting that both NAD- and NADP-dependent glutamate dehydrogenases, the biosynthetic enzymes involved in the reductive amination of α-ketoglutarate in the urea cycle (Peters & Sypherd, 1979), play an important role in linking carbon and nitrogen metabolism during the biosynthesis of chitin (Amin et al., 2004). Chitin is a key fungal cell wall polymer and the likely increase in its synthesis during the morphological switch of P. fermentans may strengthen the cell wall of pseudohyphae driving the penetration of peach tissue.

No induction of cell wall lytic enzyme expression, which could explain the fruit decay, was observed in our samples after 24 h of growth on peach fruit. Nonetheless, these could act at a second stage, as tissue rot was observed later than 1 day following inoculation, the time at which samples were collected in the present study. Peach tissue invasion by P. fermentans may also depend on the mechanical pressure applied by the growing pseudohyphae against the fruit tissues which become softer than those of apple in storage because of the activity of endogenous endo-polygalacturonases during ripening (Pressey & Avants, 1973; Tatsuki, 2010). Ethylene, a hormone involved in fruit ripening and whose biosynthesis begins with methionine, has a major influence on the ripening of peach fruit (Hayama et al., 2006). Although we have no evidence for a relationship between the dimorphic behavior of P. fermentans and ethylene, it is interesting to note that ethylene was shown to regulate the hyphal growth of B. cinerea (Kępczynska, 1993) and as previously noted, methionine is one of the signals for the induction of pseudohyphal growth in C. albicans and, likely, could play the same role in P. fermentans.

Our results indicate that P. fermentans dimorphism is characterized by changes in gene expression most likely influenced by the commodity on which it is growing. Pathogenicity on peach fruit, however, seems to be complex and associated with several linked events rather than the simple production of lytic enzymes or toxins typically associated with postharvest decay fungi, such as Penicillium, Botrytis, and Monilinia. Nevertheless, the dimorphic and decay inducing characteristics displayed by P. fermentans underscore the need to widely test and characterize the microorganisms chosen as postharvest biocontrol agents. How these antagonists interact with different species of fruit and how they respond to environmental stress can affect their behavior and stability. Functional studies utilizing transcriptomic, proteomic, and metabolomic approaches may help to predict and to monitor changes in the physiologic status of microorganisms associated with fruit and used for the management of postharvest diseases (Droby et al., 2009). As more in-depth studies of postharvest biocontrol agents are conducted, a greater understanding of ‘good’ and ‘bad’ traits associated with biocontrol activity will be forthcoming.

Supporting Information

Additional Supporting Information may be found in the online version of this article:

Fig. S1. A total of 100 μL Pichia fermentans strain 726 suspension (1 × 109 cells mL−1) was spread between two autoclaved circular sheets of cellophane membrane backing sheets (Bio-Rad) [1], permeable to liquids and nutrients, which were placed between either halved apple or peach fruits [2]. When P. fermentans is allowed to grow within halved fruits for more than 6 days at room temperature, apple fruit do not undergo decay and the cut surface becomes dried out and oxidized [3], while peach fruit are rotten and completely invaded by the pseudohypal form of P. fermentans [4].

Fig. S2. Growth curve (cells mL−1) and cell viability (%) of Pichia fermentans strain 726 grown between two autoclaved circular sheets of cellophane membrane placed between the two halves of an apple or peach fruit.

Fig. S3. SEM image of Pichia fermentans strain 726 grown on wounded peach fruit after 6 days at RT.

Please note: Wiley-Blackwell is not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

Acknowledgements

This work was partially funded by MURST PRIN Anno 2007 – Prot. N 2007FRBK9N, by Research Program MiPAAF – CIPE ‘FRU.MED.’ – Project ‘DAFME’ (Publication no. 93), and by NPRP grant # 4-259-2-083 from the Qatar National Research Fund (a member of Qatar Foundation). The statements made herein are solely the responsibility of the authors.

Footnotes

  • Editor: Cletus Kurtzman

References

View Abstract