ProteinExpressionProfile andTranscriptomeCharacterizationof Penicillium expansum Induced by Meyerozyma guilliermondii

Antagonistic yeasts can inhibit fungal growth. In our previous research, Meyerozyma guilliermondii, one of the antagonistic yeasts, exhibited antagonistic activity against Penicillium expansum. However, the mechanisms, especially the molecular mechanisms of inhibiting activity ofM. guilliermondii, are not clear. In this study, the protein expression profile and transcriptome characterization of P. expansum induced by M. guilliermondii were investigated. In P. expansum induced by M. guilliermondii, 66 proteins were identified as differentially expressed, among them six proteins were upregulated and 60 proteins were downregulated, which were associated with oxidative phosphorylation, ATP synthesis, basal metabolism, and response regulation. Simultaneously, a transcriptomic approach based on RNA-Seq was applied to annotate the genome of P. expansum and then studied the changes of gene expression in P. expansum treated with M. guilliermondii. -e results showed that differentially expressed genes such as HEAT, Phosphoesterase, Polyketide synthase, ATPase, andRas-associationwere significantly downregulated, in contrast toCytochromes P450, Phosphatidate cytidylyltransferase, and Glutathione S-transferase, which were significantly upregulated. Interestingly, the downregulated differentially expressed proteins and genes have a corresponding relationship; these results revealed that these proteins and genes were important in the growth of P. expansum treated with M. guilliermondii.


Introduction
Penicillium expansum is one of the most common pathogens in pears which cause blue mold decay and is able to secrete toxic secondary metabolite patulin (PAT), causing serious food safety problems and harming human health [1]. Recently, there are many strategies that have been employed to control the postharvest diseases of pears; among them chemical and physical methods are contribute the major part. However, both methods have some drawbacks. erefore, safe and efficient antagonistic yeasts have been a research hotspot in the control of postharvest diseases of pear. Cryptococcus laurentii, Rhodotorula glutinis, Rhodosporidium paludigenum, and Rhodotorula mucilaginosa [2][3][4][5] are some effective biocontrol agents against P. expansum in pears.
Meyerozyma guilliermondii was reported as an effective antagonistic yeast, which significantly controlled blue mold decay of pears [6] and controlled rice blast disease, cabbage black leaf spot disorder, and bacterial wilt caused by Ralstonia solanacearum, a tomato pathogen [7]. It also showed significant biocontrol of gray mold disease on table grapes caused by Botrytis cinerea [8] and reduced the severity of rot in mangoes during storage [9]. Our previous study showed that blue mold decay caused by P. expansum was significantly inhibited by M. guilliermondii without any change in the fruit quality [6]. e biocontrol efficacy increased with increasing yeast concentrations. M. guilliermondii colonized in pears rapidly and maintained relatively higher numbers to compete for nutrient and space with pathogen. M. guilliermondii also enhanced the defense to pathogens in pears [6]. e proteomics and transcriptomics analysis conducted in pears induced by M. guilliermondii revealed that M. guilliermondii could upregulate the expression of defense-related proteins and genes of pears [6,10].
Several proteins intervene during the interaction of pathogens with antagonistic yeasts and many of them are crucial to explain the inhibition mechanism of antagonistic yeast. Our previous research reported that more than onethird of the proteins differentially expressed in Talaromyces rugulosus in response to Yarrowia lipolytica were associated with essential metabolism, like phosphoglycerate kinase, nucleoside diphosphate kinase, and so on, which showed that the mechanisms by which Y. lipolytica inhibited T. rugulosus involved in the essential metabolism [11]. In the same manner, a general analysis of transcriptome and proteome modification of P. expansum spores during germination was conducted by Zhou et al. using RNA-Seq and iTRAQ approaches.
e corresponding result showed a statistic of 3026 genes and 489 proteins which were differentially expressed [12]. However, as far as we know, the molecular mechanism of M. guilliermondii against P. expansum has not been studied yet. In the present work, we explored the differentially expressed proteins and several defense-related genes of P. expansum cocultured with M. guilliermondii through proteomics and transcriptomics analysis and tried to establish the molecular mechanism of M. guilliermondii inhibiting the growth of P. expansum.

Yeast.
e antagonist yeast M. guilliermondii (preserved in the China Center for Type Culture Collection, No. M2017270) was isolated from unsprayed orchards. e yeast was cultured in nutrient yeast dextrose broth medium (NYDB, nutrient broth 8 g/L, yeast extract 5 g/L, and glucose 10 g/L) on an incubator shaker (180 rpm, 28°C) for 20 h. After incubation, yeast cells were collected by centrifugation (6918 ×g for 10 min); the pellets were resuspended in sterile water and adjusted to 1 × 10 8 cells/mL concentration with a hemocytometer.
2.2. Pathogen. P. expansum was maintained on potato dextrose agar medium at 4°C. Before using, the P. expansum strain was inoculated in PDA plates and allowed to grow for seven days at 25°C in an incubator. After seven days, the spores were removed from the Petri dish and suspended in sterile distilled water. A hemocytometer was used to adjust spore concentrations to 1 × 10 7 spores/mL.

Analysis of the Differentially Expressed Proteins of P. expansum and P. expansum Incubated with M. guilliermondii.
Initially, 1 mL of spore suspension (1 × 10 7 spores/mL) of P. expansum was added to 100 mL of PDB inside 500 mL Erlenmeyer volumetric flasks and incubated for two days at 25°C, 120 rpm. en, 1 mL suspension of M. guilliermondii at 1 × 10 8 cells/mL was added. After 1 d incubation, the mycelia were collected and the protein was extracted according to the method described by Yang et al. [11].
2-DE and Image Analysis were conducted in accordance with the method delineated by Yang et al. and Zhang et al. with some modifications [11,13,14]. Isoelectric focusing was used to separate proteins in GE Ettan IPGphor 3, according to the manufacturer's instructions [14].

Identification of Proteins.
Mass calibrations were carried out with a standard peptide mixture. Mass spectra were acquired using Matrix-Assisted Laser Desorption/Ionization Timeof-Flight (MALDI-TOF) mass spectrometer (Bruker Daltonics, Germany). e identification of the vast majority of proteins was performed using search engine MASCOT Peptide Mass Fingerprint of Matrix Science and compared with NCBInr and Swiss-Prot databases. e parameters used for MS search were taxonomy, all series; allowed modifications, carbamidomethyl of cysteine (fixed), oxidation of methionine (variable); and peptide tolerance, ±0.3 Da. Only the highest Mowse score was considered as the most probable identification and was significant (P < 0.05) when protein scores were greater than 88 (NCBInr) or 70 (Swiss-Prot) [14].

Bioinformatics Analysis of RNA-Seq Data.
Transcriptomic data were assembled, after high-quality sequencing data were acquired, using Trinity [14]. e UniGene sequences of P. expansum and P. expansum incubated with M. guilliermondii were searched using BLAST and compared with those obtained using the NR, Swiss-Prot, GO, and KEGG databases for the confirmation of amino acid sequence.

Validation of RNA-Seq Data by RT-qPCR.
RT-qPCR analysis was performed using RNA extracted from P. expansum and P. expansum incubated with M. guilliermondii, in order to validate the data obtained from RNA-Seq. e genes and their specific primers used for RT-qPCR were listed in Supplementary Table 1 and analysis was performed using a Bio-Rad CFX-96 Real-Time PCR System (Bio-Rad, USA). e reaction system that was conducted in accordance with the method delineated by Yang et al. [14] comprised of 12.5 μL SYBR ® Premix Ex Taq ™ II (2x); 0.5 μL e relative expression level of the sample gene was calculated using a 2 − ΔΔCT method [10].

Statistical Analysis.
e data were analyzed by analysis of variance (ANOVA) using the statistical program SPSS/PC version 8 (SPSS Inc., Chicago, Illinois, USA), and Duncan's multiple range test was used for mean separation. e statistical significance was assessed at P < 0.05.

Identification of Differentially Expressed Proteins of P. expansum.
e whole protein expression of P. expansum and P. expansum treated with by M. guilliermondii was shown in Figure 1. In each gel, a total of 66 differentially expressed (average fold change ≥ 2, p < 0.05) proteins were identified. Just 6 of them were significantly upregulated, while 60 proteins were significantly downregulated. Furthermore, 43 spots which showed the best resolution among the significantly differentially expressed proteins were analyzed for identification by mass spectrometry (MS). An elaborated report on lowercase letters about the names of peptides is in Table 1.
e basic information of 43 differentially expressed protein spots, comprising isoelectric point, molecular weight, and peptide matches, was assigned to a class. As reported in Table 1, several proteins were related to secondary metabolite synthesis, which included polyketide synthase (spot 20), enoylreductase (spot 32). Some proteins were associated with ATP synthesis, which included ATP hydrolase (ATPase), delta/epsilon subunit F1, N-terminal (spot 17), ATPase, F1 complex beta subunit/V1 complex, and C-terminal (spot 38). Some proteins were also associated with cellular basal metabolism, which included phosphoesterase (spot 36), glyceraldehyde/erythrose phosphate dehydrogenase family (spot 44), and phosphoglycerate kinase (spot 41) and some proteins were found to be associated with environmental immune response, which included heat shock 70 kDa protein (spot 24); these downregulated proteins were all associated with the basal metabolic process, response, and regulation of P. expansum. Gene ontology (GO) functional annotation examination was performed for the whole identified proteins, which exposed a broad number of molecular functions like biological processes and cellular components ( Figure 2). e results indicated that the largest group of biological processes was metabolic process (25 proteins) and cellular process (25 proteins), and the other two biological processes were single-organism process (16 proteins) and biological regulation (7 proteins); these proteins were all related to the basal metabolic process. All identified proteins were annotated to categories. ere were 98 and 90 proteins involved in cellular and biological processes as well as 56 proteins involved in molecular functioning, respectively. All the differentially expressed proteins were mainly involved in basic metabolism (30%), binding (23%), catalytic (21%), transporter (7%), and hypothetical (7%) processes ( Figure 3).

Transcriptomic Analysis by RNA-Seq.
e transcriptomes of P. expansum and P. expansum cocultured with M. guilliermondii were analyzed using RNA-Seq technology. e transcriptome data indicated a total number of 13 Gb clean data; the Q30 base percentage was 89.75% and 89.92% ( Table 2). A total of 434 differentially expressed genes (DEGs) were compiled, among them 408 genes were upregulated and 26 genes were downregulated in the P. expansum treated with M. guilliermondii (|log 2 (fold change)| ≥ 2, FDR < 0.05) ( Table 3). Using gene ontology (GO), these DEGs were clustered by gene function (Figure 4). e cellular components and molecular functions were further analyzed, each contained 19 subgroups and the biological processes contained 22 subgroups. e main categories concerning cellular component contained cell (23.9%), cell part (23.9%), organelle (13.9%), and membrane (13.4%). e highest percentage of identified differentially expressed genes under molecular function category included catalytic activity (46.1%) and binding (38.5%), whereas the percentage in the biological process category was as follows: basic metabolism (21.0%), cellular process (20.2%), single-organism process (19.0%), and biological regulation (18.5%). e most enriched KEGG pathway of P. expansum transcriptome analysis was shown in Figure 5. ere were four pathways including cellular processes, environmental information processing, genetic information processing, and metabolism. In the cellular processes, there were 5 DEGs in cell growth and death, 16 DEGs in transport and catabolism. In the environmental information processing, there were 22 DEGs in folding, sorting, and degradation, 6 DEGs in transcription, and 6 DEGs in translation. In metabolism, the highest pathway is carbohydrate metabolism (32.14%), energy metabolism (17.26%), amino acid metabolism (16.07%), and lipid metabolism (14.28%).

Discussion
Blue mold decay caused by P. expansum is a serious disease in fruit, in particular pear fruits [6]. Recently, some pathogenic strains developed resistance against synthetic fungicides; therefore, researchers concern about environmental and food safety. Antagonistic yeasts are gaining considerable attention due to their beneficial environmental and food safety characteristics and also controlling postharvest diseases in fruits. Integrating transcriptomic and proteomic data to achieve meaningful insights into P. expansum inhibited by M. guilliermondii has rarely been reported. erefore, the present work highlights the protein expression profile and transcriptomic changes of P. expansum in order to explore the molecular inhibitory mechanism of M. guilliermondii. GO categorization was compared between differentially expressed proteins from 2-DE and differentially expressed genes from RNA-Seq analysis. A majority of differentially expressed proteins and genes were involved in secondary metabolite synthesis, ATP synthesis, cellular basal metabolism, environmental immune response, genetic information processing, and metabolism.   ere are many enzymes responsible for synthesis of ATP; ATPase was considered as the most important enzyme in ATP synthesis. In the current investigation, protein ATPase, F1 complex (spots 17 and 38) corresponding with ATPase, F0/V0 complex, subunit C and ATPase, F1/A1 complex, alpha subunit, N-terminal, were all downregulated. Similar trend was observed in transcriptomic analysis; the gene corresponding to ATPase, F0/V0 complex, subunit C (TRINITY_DN3524_c0_g2, TRINITY_DN7527_c0_g1) was downregulated significantly. Yang et al. reported that the protein related to ABC transporters (PAAT), which is a novel ATPase and a trans-regulator of mitochondrial ABC transporters, is responsible for cell survival and maintenance of mitochondrial homeostasis [15]. Additionally, Yang et al. reported that the poplar buds active fraction (PBAF) depresses ATPase activity of Penicillium italicum and explained that the inhibitory effect of PBAF on fungal growth is due to the inhibition of ATPase by PBAF [16]. Castellote et al. investigated the molecular expression of Geotrichum candidum overripening of Reblochon-type cheese and observed that the genes responsible for F1F0 ATP synthase subunits (ATP1, ATP2, ATP5, and ATP7) were quiescent which leads to the reduced ATP production and consumption, normally less energy needed for quiescent cells [17]. From these reports, we anticipated that ATPase did not only provide energy for the basic metabolism of P. expansum, but also was the main carrier for the life activities of P. expansum. In the present investigation, the ATP activity of P. expansum treated with M. guilliermondii was reduced; thus, the energy supply was prevented and the growth was inhibited.
Heat shock protein 70 (HSP70) was a main part of the cell's machinery for protein folding and contributed to protection of cells against stress [18,19]. In addition to the involvement of HSP70 in maintaining and improving the protein integrity, it directly inhibits apoptosis [20]. In our results, the heat shock 70 kDa protein (point 24) corresponded with the transcriptome genes Heat, type 2 (TRI-NITY_DN5224_c0_g4), and both were downregulated. e intracellular heat shock protein and gene expression of P. expansum was downregulated because of the effect of M. guilliermondii. Zhang et al. studied the underlying molecular mechanisms of HSP70 in the environmental stress response of coral through transcriptome expression and reported that HSP70 (PdHSP70) was an essential stress regulatory protein in the stony coral Pocillopora damicornis; therefore, diverse environmental stress could induce HSP70 mRNA expression and its activity could remain stable under heat stress [21]. Kim et al. identified 10 Calmodulin (CaM)-binding protein in Beauveria bassiana; one of its targets was HSP70 and their results also suggested that ATP was involved in the inhibition of molecular interaction between CaM and HSP70 [22]. From the earlier studies, it was evidenced that downregulation of HSP70 weakens the adaptation ability and immune response of P. expansum to the external environment; the cells were apoptotic and the growth of P. expansum was inhibited by M. guilliermondii.
In biological systems, phosphoesterase plays a vital role in DNA fragmentation, RNA replication, human body medicine, metabolism, chemotherapy, and bioremediation [23,24]. In the present results, the phosphoesterase protein (point 36) that corresponded with the transcriptome genes phosphoesterase (TRINITY_DN431_c0_g2) was downregulated. It could be speculated that the phosphate kinase activity of P. expansum was reduced and substance metabolism process was blocked due to the effect of M. guilliermondii, which resulted in shortage of phosphate needed for the survival of P. expansum and thus the normal life activity of P. expansum was inhibited by M. guilliermondii.
Polyketide synthases (PKSs) belong to a multidomainenzyme or complex-enzyme family which are responsible for the production of polyketides, a major class of secondary metabolites, in microorganisms, plants, and some animal lineages [25,26]. Polyketide synthases also play a definite role in the production of naturally occurring small molecules employed in chemotherapy [27]; many of the commonly used antibiotics and other industrially important polyketides were produced by polyketide synthases [28]. Our results demonstrated that the polyketide synthase, enoylreductase protein (point 20 and point 32), was downregulated. RNA-Seq analysis also showed that the gene responsible for polyketide synthases (TRINITY_DN10607_c0_g1) was downregulated significantly. Moreover, in P. expansum, the polyketide synthases are involved in the synthesis of patulin, a hazardous mycotoxin derived from polyketides [29]. Role of patulin in the pathogenicity and virulence of P. expansum was already proved in postharvest apples infected by P. expansum [30]. Patulin synthesis in P. expansum is carried out by a biosynthetic gene cluster consisted of 15 genes (PatA-PatO) in ten different enzymatic reactions [31]. Among them, the initial seven reactions in patulin synthesis are catalyzed by polyketide synthases [32]. erefore, our  results may confirm that the polyketide synthase activity of P. expansum was reduced due to the inhibition of M. guilliermondii; thus the normal synthesis of patulin and other secondary metabolites of P. expansum was prevented. Likewise, Ras-association gene (TRINITY_DN6243_c0_g3) was also downregulated in P. expansum treated with M. guilliermondii. In most of the pathogenic fungi, Ras pathway signaling is determined as critical virulence factor [33]. e Ras pathway signaling is involved in pathogenesis, morphological transitions, nutrient sensing and acquisition, sexual reproduction, and stress responses of fungi [34]. e results also evidenced that the pathogenicity and virulence of P. expansum was weakened, and its growth was eventually inhibited by M. guilliermondii.
Phosphatidate cytidylyltransferase is an enzyme, involved in lipid transport and metabolism, that majorly takes part in the phospholipid metabolism in cell membranes. Lipids are essential components in membrane trafficking, cytoskeletal rearrangement, and secretion, which are reported to be the important mechanisms of stress tolerance [35]. In the present transcriptomic analysis, the gene responsible for phosphatidate cytidylyltransferase (TRINITY_DN3203_c0_g2) was upregulated in P. expansum. Bernardo et al. reported the upregulation and accumulation of phosphatidate cytidylyltransferase during drought stress in the roots of wheat [36]. Other than stress tolerance, phosphatidate cytidylyltransferase was also known for its role in antibiotic resistance mechanism against antimicrobial lipopeptides [37]. It was evidenced that in order to overcome the stress created by M. guilliermondii and to resist the activity of the yeast, the phosphatidate cytidylyltransferase gene was upregulated in P. expansum. In the same way, the upregulation of striatin, N-terminal (TRINITY_DN6439_c0_g1) was observed in the transcriptomic analysis. Striatin orthologs were directly related to the virulence of some filamentous fungi like Fusarium verticillioides and F. graminearum against plants [38]. Hence, we hypothesized that though some pathogenic related factors were inhibited by M. guilliermondii, some other genes related to stress tolerance and virulence were upregulated.
To sustain cellular redox homeostasis, the thioredoxin system assures a crucial function and the thioredoxin protein (point 2) was downregulated in the present study. Viefhues et al. indicated that thioredoxin had a strict influence on virulence of Botrytis cinerea, proving that redox processes were determinant for host-pathogen interactions in this pathogen; in addition, the balanced redox status established by the thioredoxin system was vital for growth and pathogenesis of B. cinerea [39]. Ianiri et al. investigated the modification of gene expression in Sporobolomyces sp. under patulin exposition; they noted the major increase in transcript levels of antioxidant molecules glutathione and thioredoxin in Sporobolomyces genes. e research team further subclassified the metabolic process of patulin exposed group and found that 6 DEGs were involved in glutathione biosynthesis, and 3 DEGs were involved in the thioredoxin system [40]. In contrast, by downregulation in the present study, we speculated that the thioredoxin system, which is involved in the defense response to ROS, was inhibited by M. guilliermondii and the growth of P. expansum was controlled. e ribosomal S21 protein as part of eukaryotic ribosomes had proved to be strongly related to ribosome-associated protein in human and Drosophila cells [41][42][43]. In our results, the 40S ribosomal protein S21 (spot 4) was downregulated and was involved in nucleotide transport and metabolism. Sato et al. examined the functions of the ribosomal proteins S0 and S21 in Schizosaccharomyces pombe and found that the ribosomal protein played a major physiological role in 18s rRNA stability which was essential for the survival of S. pombe cells [44]. erefore, we hypothesized that, in the presence of M. guilliermondii, 40S ribosomal protein S21 P. expansum was inhibited so that the growth and survival of the fungi was reduced dramatically.
Cytochromes P450 was distributed in almost all organisms and involved in diverse metabolic processes, which were highly diverse and contributed to the metabolism of xenobiotic compounds such as bioconversion of xenobiotics, biotransformation of drugs, and so on [45]. In the current investigation, the Cytochrome P450 (TRINI-TY_DN5730_c0_g1) was upregulated. Trippe et al. studied the genetic underpinnings which permit Graphium sp. to catalyze the early step in the alkane and ether oxidation pathway; furthermore, they identified CYP52L 1 as an alkaneoxidizing Cytochrome P450, posttranscriptional ds-RNAmediated gene silencing of CYP52L 1 and proved that the gene silencing disrupted the ability of Graphium sp. to grow on alkanes and ethers [46]. Similarly, our results revealed that the upregulation of Cytochrome P450 (TRINI-TY_DN5730_c0_g1) could be activated by the basic metabolic process of P. expansum after it was cocultured with M. guilliermondii.
Glutathione S-transferases (GSTs) are well known for their capacity to reduce the oxidative potency of xenobiotic substrates for the purpose of detoxification [47,48]. e main role of GSTs is to reduce the toxicity of xenobiotics compounds by catalyzing the GSH nucleophilic attack on electrophilic atoms of said nonpolar xenobiotic substrates, thereby avoiding their interaction with essential cellular proteins as well as nucleic acids [49]. In the present study, the Glutathione S-transferase (TRINITY_DN8139_c0_g1)   was upregulated. Ouyang et al. showed that citral possibly triggered a reduction in the mitochondrial membrane potential (MMP), intracellular ATP, and glutathione content, in contrast to an increase in the glutathione S-transferase activity and the accumulation of reactive oxygen species (ROS); these results indicated that the addition of citral probably leads to the oxidative damage of Penicillium digitatum and inhibited the growth of P. digitatum [50]. Hence, from the upregulation of GST, it could be postulated that the GST activity was activated in order to overcome the increased ROS in the presence of M. guilliermondii. e results also evidenced that M. guilliermondii created oxidative stress against P. expansum; hence, the level of GST (antioxidative enzyme) was increased. Overall, we have hypothesized that though some stress tolerance mechanism of P. expansum was activated, M. guilliermondii inhibit the growth and pathogenesis of P. expansum by downregulating certain pathways of virulence mechanism, cellular signal transduction, mycotoxin production, and cell organization.
On the other hand, the biocontrol efficacy of M. guilliermondii and pathogenic ability of P. expansum might be altered under availability of different nutritional sources  such as carbon (C), nitrogen (N), and environmental factors such as pH value, temperature, and water activity. Jianjie et al. studied the effect of nutrition and environmental factors on the biocontrol potential of Esteya vermicola and reported that both of them have great influence on the biocontrol potential [51]. Similarly, the growth and mycotoxin producing ability of P. expansum was altered by different glucose-containing sugars, complex N sources, and acidic conditions were favorable conditions for patulin production [52]. Since the stress tolerance of M. guilliermondii in extreme salinity stress was proved already [53], we speculated that the antagonistic potential of M. guilliermondii might be worthy in a diverse range of nutritional supplements and environmental factors. However, in the future prospect, the effect of various nutritional sources and environmental factors in the biocontrol mechanism of M. guilliermondii are needed to be studied.
In conclusion, several genes, such as HEAT, Phosphoesterase, Polyketide synthase, and ATPase, disclosed a corresponding relationship between the transcript and protein levels and revealed that those genes could play an important role in the growth regulation of P. expansum. Moreover, the data from transcriptomic and proteomic analysis discussed in the present study have not only underlined a set of genes and proteins that were essential in the growth of P. expansum, but also gave fundamental knowledge of molecular mechanism behind the inhibitory activity of M. guilliermondii against the growth of mold.

Data Availability
e data used to support the findings of this study are included within the article.

Conflicts of Interest
e authors have no conflicts of interest to declare.