Molecular Response of Meyerozyma guilliermondii to Patulin: Transcriptomic-Based Analysis

Patulin (PAT), mainly produced by Penicillium expansum, is a potential threat to health. In recent years, PAT removal using antagonistic yeasts has become a hot research topic. Meyerozyma guilliermondii, isolated by our group, produced antagonistic effects against the postharvest diseases of pears and could degrade PAT in vivo or in vitro. However, the molecular responses of M. guilliermondii over PAT exposure and its detoxification enzymes are not apparent. In this study, transcriptomics is used to unveil the molecular responses of M. guilliermondii on PAT exposure and the enzymes involved in PAT degradation. The functional enrichment of differentially expressed genes indicated that the molecular response mainly includes the up-regulated expression of genes related to resistance and drug-resistance, intracellular transport, growth and reproduction, transcription, DNA damage repair, antioxidant stress to avoid cell damage, and PAT detoxification genes such as short-chain dehydrogenase/reductases. This study elucidates the possible molecular responses and PAT detoxification mechanism of M. guilliermondii, which could be helpful to further accelerate the commercial application of antagonistic yeast toward mycotoxin decontamination.


Introduction
Penicillium expansum is a toxic secondary metabolite of filamentous fungi that causes serious pollution of foodstuffs, produces substantial economic losses worldwide, and harms consumers' health [1]. P. expansum is the most important producer of PAT, which causes the decay and deterioration of fruits and vegetables. PAT contamination in fruits and their products is conventionally controlled by physical, chemical, and biological strategies. The physical methods mainly control the generation of PAT by means of manual screening, highpressure washing, refrigeration, etc. However, quality loss, high costs, and environmental pollution are major drawbacks of physical methods [2]. The chemical PAT control methods can be divided into two types; the first type is the prevention of PAT contamination by killing PAT producers through chemical fungicides, and the other type is the removal of PAT by adding chemicals. However, some PAT-controlling chemicals are toxic in nature and can damage the product quality, taste, and nutritional value [2]. The biological methods refer to antagonistic microorganisms (yeasts, bacteria, mold) to control the generation of PAT by inhibiting pathogenic fungal infection or using microorganisms to adsorb or degrade the generated PAT. The biological control methods have attracted more and more researchers' attention and show a broad application prospect because of their high efficiency, low cost, safety, and non-toxicity [2].
In recent years, the use of antagonistic yeast to remove PAT has become a hot research topic, but the PAT clearance mechanism is unclear. Researchers published two primary PAT scavenging mechanisms of antagonistic yeast; the first is the adsorption of PAT by antagonistic yeast cells without affecting the cell activity. For example, the PAT adsorption capacity of live and heat-inactivated Saccharomyces cerevisiae cells is similar, and the yeast M. guilliermondii was isolated from the pear surfaces from unsprayed orchards by our group and deposited in CCTCC of Wuhan University with the accession number M2017270. In our laboratory, M. guilliermondii isolates were maintained at 4 • C in nutrient yeast dextrose agar (NYDA) medium (nutrient broth 8 g, yeast extract 5 g, glucose 10 g, agar 20 g, in 1 L of distilled water). Liquid cultures of the yeast were cultivated in 50 mL of NYD broth (NYDB) by inoculating a loop of M. guilliermondii cells and incubating on a rotary shaker at 28 • C for 24 h. Later, the cells were washed twice using sterile distilled water and adjusted to an initial concentration of 1 × 10 8 cells/mL before experiments. For PAT exposure, 1 mL of M. guilliermondii suspension was added into 50 mL NYDB containing PAT (adjusted concentration to 10 µg/mL) and then incubated at 28 • C, 180 rpm for 24 h. M. guilliermondii suspension was added into 50 mL NYDB without PAT and was used as a control group. The cells were washed twice with sterile distilled water by centrifugation for 10 min at 4 • C. The yeast cells were frozen by adding liquid nitrogen, and the sample was stored in the −80 • C refrigerator until further use.

Total RNA Extraction from Yeast
The M. guilliermondii in NYDB without PAT induction served as the control group, and the M. guilliermondii cultured in PAT-supplemented NYDB served as the experimental group. Each group contained 3 biological replicates. After PAT exposure, the M. guilliermondii was quickly ground to powder in a pre-cooled mortar with liquid nitrogen. Total RNA was extracted using a column fungal total RNA extraction kit (Sangon Biotech Co., Ltd., Shanghai, China) according to the manufacturer's instructions. The total RNA concentration and purity were determined using NanoDrop One (Thermo Fisher Scientific, Waltham, MA, USA), and the integrity was determined using Agilent 2100 biochip analyzer (Agilent, Santa Clara, CA, USA).
The method of FPKM was used to evaluate the expression levels of genes, and the software of DESeq2 1.40.0 was used to screen differentially expressed genes. The genes with a parameter of false discovery rate (FDR) below 0.05 and absolute fold change greater than or equal to 2 or greater than or equal to 1 were considered differentially expressed genes. Additionally, the method of Gene Ontology (GO) enrichment analysis started with all DEGs being mapped to GO terms in the GO database, then gene numbers were calculated for every term, and significantly enriched GO terms in DEGs comparing to the genome background were defined using the hypergeometric test. The p-value was calculated, and then the p-value went through FDR correction, taking FDR ≤ 0.05 as a threshold. GO terms meeting this condition were defined as significantly enriched GO terms in DEGs. Additionally, the method of the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment is the same as the GO enrichment analysis.

Validation of DEGs by RT-qPCR
Based on RNA-seq analysis, 12 DEGs that play a key role in regulating PAT response were selected (Table 1). In this study, a β-tubulin gene of M. guilliermondii was used as the reference gene. Primers of selected genes were designed using Primer 6.0 and were purchased from Seigon Biotechnology, Shanghai, China. The primers used in RT-qPCR analysis are shown in Table S1. RNA samples from M. guilliermondii cultures exposed with or without PAT were extracted for RT-qPCR analysis as described above. HiFiScript gDNA Removal RT Master Mix (CoWin Biosciences, Beijing, China) was used to synthesize cDNA by reverse transcription. Real-time PCR amplification was performed using a two-step method and was measured in the Applied Biosystems 7300 Real-Time PCR system (USA). The relative expression levels were calculated using 2 −∆∆CT method. The whole experiment was repeated thrice.

Sample Relationship Analysis
The quality control of the transcriptomic data of all the samples (CK-1, CK-2, CK-3, PAT-1, PAT-2, and PAT-3) showed a quality score above Q30, indicating high sequencing quality and good reliability (Figure 1a). The normalized mapping rate of the transcriptomic data of all the samples with the reference genome reached more than 96%, and the samples were mapped predominantly to the exonic regions (more than 84%). These indicators specified that the transcriptomic data of all the samples were of good quality, and the alignment with the selected reference genome was complete (Table S2).
The quality control of the transcriptomic data of all the samples (CK-1, CK-2, CK-3, PAT-1, PAT-2, and PAT-3) showed a quality score above Q30, indicating high sequencing quality and good reliability (Figure 1a). The normalized mapping rate of the transcriptomic data of all the samples with the reference genome reached more than 96%, and the samples were mapped predominantly to the exonic regions (more than 84%). These indicators specified that the transcriptomic data of all the samples were of good quality, and the alignment with the selected reference genome was complete (Table S2).
The Pearson correlation coefficients between the expression levels of two randomly selected samples were calculated and visually displayed in heat maps to show the correlation between the samples. The R-value was positively correlated, and a larger Rvalue indicates a better correlation. The R-value of two freely chosen control group samples was greater than 0.99, and the R-value of two freely chosen experimental groups was greater than 0.95, while the R-value of two freely chosen control and experimental group samples was less than 0.9, which showed that the parallelism of the treatment groups was good and the difference between the treatment groups was significant ( Figure  1b).

Differential Expression of Genes
When the absolute of log2(Fold Change) greater than or equal to 1 and the FDR less than 0.05 were used as the screening criteria, the up-regulated DEGs were 215 and the down-regulated DEGs were 133. When the absolute of log2(Fold Change) greater than or equal to 2 and the FDR less than 0.05 were used as the screening criteria, the up-regulated DEGs were 54 and the down-regulated DEGs were 8 ( Figure 1c). The DEGs identified in the transcriptome analysis were annotated using NR, Swiss-Port, GO, COG, KOG, KEGG, and other databases (Table S3).

GO Enrichment Analysis of DEGs
The GO enrichment analysis of the DEGs of the PAT-induced M. guilliermondii was enriched in three major categories including biological process, molecular function, and cellular components (Figure 2a). The cell components contained 12 secondary subclasses; among them, membrane and membrane parts might be related to the transport of PAT by M. guilliermondii into cells. The molecular function category included 11 secondary subclasses; among them, the enriched transporter activity subclass might be related to the intramolecular transport of PAT, and the catalytic activity and antioxidant activity might be closely related to the stress resistance of M. guilliermondii. The biological processes consisted of 20 secondary subclasses; among them, signaling, response to stimulus, biological regulation, and detoxification were related to the growth and development of M. guilliermondii and the regulation of PAT and stimulus-response.
Nine secondary subclasses related to the regulation of M. guilliermondii response toward PAT were selected for the GO tertiary enrichment analysis (Figure 2b). Among them, peroxisomal membrane, oxidoreductase activity, glutathione transferase activity, drug transmembrane transporter activity, amino acid transmembrane transporter The Pearson correlation coefficients between the expression levels of two randomly selected samples were calculated and visually displayed in heat maps to show the correlation between the samples. The R-value was positively correlated, and a larger R-value indicates a better correlation. The R-value of two freely chosen control group samples was greater than 0.99, and the R-value of two freely chosen experimental groups was greater than 0.95, while the R-value of two freely chosen control and experimental group samples was less than 0.9, which showed that the parallelism of the treatment groups was good and the difference between the treatment groups was significant ( Figure 1b).

Differential Expression of Genes
When the absolute of log 2 (Fold Change) greater than or equal to 1 and the FDR less than 0.05 were used as the screening criteria, the up-regulated DEGs were 215 and the down-regulated DEGs were 133. When the absolute of log 2 (Fold Change) greater than or equal to 2 and the FDR less than 0.05 were used as the screening criteria, the up-regulated DEGs were 54 and the down-regulated DEGs were 8 ( Figure 1c). The DEGs identified in the transcriptome analysis were annotated using NR, Swiss-Port, GO, COG, KOG, KEGG, and other databases (Table S3).

GO Enrichment Analysis of DEGs
The GO enrichment analysis of the DEGs of the PAT-induced M. guilliermondii was enriched in three major categories including biological process, molecular function, and cellular components (Figure 2a). The cell components contained 12 secondary subclasses; among them, membrane and membrane parts might be related to the transport of PAT by M. guilliermondii into cells. The molecular function category included 11 secondary subclasses; among them, the enriched transporter activity subclass might be related to the intramolecular transport of PAT, and the catalytic activity and antioxidant activity might be closely related to the stress resistance of M. guilliermondii. The biological processes consisted of 20 secondary subclasses; among them, signaling, response to stimulus, biological regulation, and detoxification were related to the growth and development of M. guilliermondii and the regulation of PAT and stimulus-response. Nine secondary subclasses related to the regulation of M. guilliermondii response toward PAT were selected for the GO tertiary enrichment analysis ( Figure 2b). Among them, peroxisomal membrane, oxidoreductase activity, glutathione transferase activity, drug transmembrane transporter activity, amino acid transmembrane transporter activity, and regulation of fungal-type cell wall organization were related to PAT degradation by M. guilliermondii.

COG Enrichment Analysis of Differentially Expressed Genes
The COG enrichment analysis was performed for the DEGs of the PAT-induced M. guilliermondii (Figure 3). The top seven functional classifications and enriched DEGs in COG enrichment were amino acid transport and metabolism (42 DEGs); carbohydrate transport and metabolism (38 DEGs); inorganic ion transport and metabolism (29 DEGs); posttranslational modification, protein turnover, and chaperones (18 DEGs); energy production and conversion (13 DEGs); lipid transport and metabolism (12 DEGs); and secondary metabolites biosynthesis, transport, and catabolism (12 DEGs).

COG Enrichment Analysis of Differentially Expressed Genes
The COG enrichment analysis was performed for the DEGs of the PAT-induced M. guilliermondii (Figure 3). The top seven functional classifications and enriched DEGs in COG enrichment were amino acid transport and metabolism (42 DEGs); carbohydrate transport and metabolism (38 DEGs); inorganic ion transport and metabolism (29 DEGs); post-translational modification, protein turnover, and chaperones (18 DEGs); energy production and conversion (13 DEGs); lipid transport and metabolism (12 DEGs); and secondary metabolites biosynthesis, transport, and catabolism (12 DEGs).

KEGG Pathway Enrichment Analysis of DEGs
As shown in Figure 4a, all DEGs were annotated to the KEGG database and enriched in 4 primary classifications, such as cellular processes, environmental information processing, genetic information processing, and metabolism, which consisted of 17 secondary subclasses. Among the secondary subclasses, transport and catabolism, signal transduction, replication and repair, amino acid metabolism, carbohydrate metabolism, and metabolism of other amino acids were related to the growth, stress resistance, and PAT-responsive regulation of M. guilliermondii. The DEGs of these 6 secondary subclasses were further analyzed via KEGG tertiary enrichment, and the DEGs were enriched in 22 tertiary subclasses (Figure 4b). Four pathways, such as peroxisome pathways, MAPK signaling pathway yeast, glycolysis/gluconeogenesis pathways, and glutathione metabolic pathways, were significantly enriched and correlated with the PAT response regulation in M. guilliermondii.

KEGG Pathway Enrichment Analysis of DEGs
As shown in Figure 4a, all DEGs were annotated to the KEGG database and enriched in 4 primary classifications, such as cellular processes, environmental information processing, genetic information processing, and metabolism, which consisted of 17 secondary subclasses. Among the secondary subclasses, transport and catabolism, signal transduction, replication and repair, amino acid metabolism, carbohydrate metabolism, and metabolism of other amino acids were related to the growth, stress resistance, and PAT-responsive regulation of M. guilliermondii. The DEGs of these 6 secondary subclasses were further analyzed via KEGG tertiary enrichment, and the DEGs were enriched in 22 tertiary subclasses (Figure 4b). Four pathways, such as peroxisome pathways, MAPK signaling pathway yeast, glycolysis/gluconeogenesis pathways, and glutathione metabolic pathways, were significantly enriched and correlated with the PAT response regulation in M. guilliermondii.

Differential Expression of Genes Involved in PAT Detoxification
As shown in Table 2, the up-regulated expression of ten short-chain dehydrogenase/reductase (SDR) genes and the glutathione S-transferase encoding gene (gedE) was observed in PAT-exposed M. guilliermondii. Among the SDRs, enoyl-(Acyl carrier protein) reductase, NAD-dependent reductase, and NAD(P) H-dependent reductase family genes were more significantly up-regulated.

Validation of RNA-seq Data by RT-qPCR
Twelve DEGs were randomly selected for RT-qPCR verification to prove the reliability of the transcriptome sequencing results. As shown in Figure 5, the expression levels of these DEGs were analyzed via the regression analysis with the expression levels in the transcriptome data. The expression trends of the selected DEGs were consistent between the transcriptomic analysis and the RT-qPCR analysis; the Pearson correlation coefficient was 0.764, indicating that the transcriptome sequencing results were authentic and reliable. reliability of the transcriptome sequencing results. As shown in Figure 5, the expression levels of these DEGs were analyzed via the regression analysis with the expression levels in the transcriptome data. The expression trends of the selected DEGs were consistent between the transcriptomic analysis and the RT-qPCR analysis; the Pearson correlation coefficient was 0.764, indicating that the transcriptome sequencing results were authentic and reliable.

Discussion
Pome fruits, including pears, are vulnerable to mechanical damage during cultivation, transportation, storage, and selling, which facilitates pathogen infestations and mycotoxin contamination. PAT is one of the critical mycotoxins that can cause severe ill effects to consumers and economic losses to producers. P. expansum is the primary producer of PAT. In recent years, PAT detoxification using antagonistic yeasts has become a hot research topic. Our previous research proved the antagonistic effect and PAT degradation ability of M. guilliermondii, but the PAT removal, especially the molecular mechanism, is unclear. In this study, we analyzed the molecular responses of M. guilliermondii with or without PAT stimulation through transcriptome sequencing and validated the results via the RT-qPCR analysis. The RT-qPCR results confirmed that the expression trend of the selected DEGs in RT-qPCR and the transcriptomic sequencing results were consistent ( Figure 6). The gene annotation analysis of DEGs in GO, COG, KEGG, Swiss-Port, and other databases provided the essential information about the critical genes involved in the molecular regulation of M. guilliermondii during PAT exposure. Notably, the genes related to post-translational modification, DNA damage repair, resistance and drug resistance, oxidative stress resistance, growth and reproduction, transcription, and translation regulation and transport ( Figure 6) were differentially regulated in M. guilliermondii.

Discussion
Pome fruits, including pears, are vulnerable to mechanical damage during cultivation, transportation, storage, and selling, which facilitates pathogen infestations and mycotoxin contamination. PAT is one of the critical mycotoxins that can cause severe ill effects to consumers and economic losses to producers. P. expansum is the primary producer of PAT. In recent years, PAT detoxification using antagonistic yeasts has become a hot research topic. Our previous research proved the antagonistic effect and PAT degradation ability of M. guilliermondii, but the PAT removal, especially the molecular mechanism, is unclear. In this study, we analyzed the molecular responses of M. guilliermondii with or without PAT stimulation through transcriptome sequencing and validated the results via the RT-qPCR analysis. The RT-qPCR results confirmed that the expression trend of the selected DEGs in RT-qPCR and the transcriptomic sequencing results were consistent ( Figure 6). The gene annotation analysis of DEGs in GO, COG, KEGG, Swiss-Port, and other databases provided the essential information about the critical genes involved in the molecular regulation of M. guilliermondii during PAT exposure. Notably, the genes related to posttranslational modification, DNA damage repair, resistance and drug resistance, oxidative stress resistance, growth and reproduction, transcription, and translation regulation and transport ( Figure 6) were differentially regulated in M. guilliermondii.
The gene STE14 directs a range of post-translational reactions, including isopentenylation, endoproteolysis, and carboxymethylation [10]. The gene PPQ1 regulates mating signaling by targeting at or upstream of the terminal MAP kinase Fus3 in the cascade and is also associated with the dephosphorylation of target pathway proteins [11]. In our results, both the STE14 and PPQ1 genes related to the protein modification were up-regulated in M. guilliermondii. This result indicated that M. guilliermondii encounters the PAT induction by up-regulating the genes related to protein modification, ensuring the carboxymethylation and phosphorylation of post-translational proteins and signal processing ( Figure 6).
PAT causes ROS accumulation inside the cells, which leads to oxidative DNA damage [12,13]. DNA damage response (DDR) plays a key role in maintaining genome integrity and stability. The protein encoded by SPBC2A9.02 genetically interacted with the DNA replication initiation proteins Abp1 is a guarantee for the efficient initiation of DNA replication [14]. Likely, the DNA repair protein encoded by RAD14 is a DNA damage recognition factor in nucleotide excision repair [15]. The DNA repair protein encoded by MAG1 is associated with a regulatory factor (RPN4)-dependent DNA repair pathway. Similarly, the DNA repair factor IIH helicase subunit encoded by SSL2 [16] is a component of the DNA repair factor and participates in DNA damage repair. In this study, the expression of SPBC2A9.02, RAD14, MAG1, and SSL2 were all up-regulated, which suggested that the DNA damage caused by PAT stress was repaired by M. guilliermondii, which ensured the genome integrity and stability ( Figure 6). protein could reduce 80% of PAT from apple juice. The CgSDR could transform toxic PAT into non-toxic E-ascladiol in vitro with NADPH as a coenzyme [50]. In the present study, the significant up-regulation of several SDRs was observed in M. guilliermondii under PAT stress, which speculated that SDRs could be involved in PAT detoxification [51]. Similarly, the role of glutathione S-transferase (GST) in PAT detoxification was described from the proteomic analysis of PAT-induced R. mucilaginosa. GST could catalyze the conjugation of a reduced form of GSH with PAT in R. mucilaginosa upon PAT induction [49]. In our results, the GST expression was induced in M. guilliermondii under PAT stress. Collectively, our study supports the previously proposed hypothesis of PAT detoxification into E-ascladiol via SDRs and the involvement of GST. Studies on the direct participation of SDRs and GST of M. guilliermondii in PAT detoxification and the degradation products by heterogeneous expression and purification are ongoing.  A proteomic study of PAT-exposed C. guilliermondii indicated that PAT causes adverse stress and affects various metabolic pathways of yeast cells [6]. In the present study, PAT induces adverse stress on M. guilliermondii and stimulates the stress and drug resistance mechanisms of the yeast. Some genes associated with stress and drug resistance were up-regulated in M. guilliermondii. APD1 plays a critical role in cellular defense, and the loss of APD1 leads to the loss of cellular sensitivity and intracellular redox homeostasis [17]. The ubiquitin-binding enzyme E2 encoded by UBC2 helps the cells adapt to cold, salt, and toxicity stress and improves the ability of environmental stress resistance [18,19]. Likewise, the drug-resistant protein encoded by MDR1 is associated with the multipotent and pleiotropic drug resistance of yeast [20]. In our research, under PAT stimulation, the expression of APD1, UBC2, and MDR1 related to adverse environmental stress and drug resistance was significantly up-regulated in M. guilliermondii. The up-regulation of these genes could enhance the resistance of M. guilliermondii to PAT and help M. guilliermondii to resist PAT toxicity [21] (Figure 6).
After entering the yeast cells, PAT causes sulfhydryl GSH depletion, induces intracellular ROS accumulation, and eventually leads to redox homeostasis imbalance, thus poisoning the cells [6,22]. In M. guilliermondii, the oxidative stress-related genes, such as GRP2 (encoding NADPH dependent methylglyoxal reductase), NPY1 (encoding peroxisome), and PST2 (encoding oxidative stress protein) [23,24] were up-regulated to maintain intracellular redox homeostasis. Iron reductase encoded by CFL1 plays an important role in oxidative stress, and Xu et al. confirmed that mutation in the CFL1 gene leads to high levels of ROS production in C. albicans [25]. The mitochondrial transporter (FLX1) encoded by FLX1 catalyzes the movement of the redox cofactor FAD through the mitochondrial membrane and affects the ATP production efficiency, ROS homeostasis, and longevity of S. cerevisiae. The deletion of FLX1 gene showed significant ATP deficiency and ROS imbalance in S. cerevisiae [26]. The peroxisome membrane signal receptor encoded by PEX5 promotes the peroxisome matrix protein's introduction by shuttling between the cytoplasm and peroxisome membrane [27]. The genes GRX2 and GRX3 encode small redox proteins called glutaredoxins (GRXs), which reduce glutathione as an electron donor and are vital components of the antioxidant system of cells. The loss of GRXs reduces the expression of stress-reactive proteins, resulting in an increased accumulation of ROS in cells [28]. The gene Snz1 encoding the phosphosynthase subunit is associated with vitamin B6 (VB6) biosynthesis, a potent antioxidant, and plays a vital role in development and stress response [29]. In the present study, the up-regulated expression of GRP2, NPY1, PST2, CFL1, FLX1, PEX5, GRX2, GRX3, and Snz1 indicated that M. guilliermondii eliminated intracellular ROS and prevented cell damage caused by PAT toxicity by up-regulating the expression of genes related to the redox process ( Figure 6).
A series of genes related to growth and reproduction were also up-regulated in PATinduced M. guilliermondii. The gene mel1 is a metabolism-associated gene that encodes α-galactosidase, a key enzyme in the catabolic pathway of galactose and glucose disaccharide [30], and the gene ETR1 that encodes enoyl reductase is associated with yeast fatty acid synthesis (fatty acid synthesis type II) and respiratory metabolism [31]. The flavodoxin-like genes YCP4 regulate the expression of several metabolism-related genes during the late growth stage of yeast [32]. The PKR1 gene encodes the V-type ATPase assembly factor (PKR1) associated with iron ion utilization, and PKR1 deficiency leads to V-ATPase levels and defective Fet3p, a component of the high-affinity iron transport system [33]. The gene MIA40 encodes the mitochondrial intermembrane space import and assembly protein 40, which uses cytochrome oxidase copper chaperone as an important substrate (encoded by COX17) that plays a crucial role in the import, oxidation, and folding of other mitochondrial proteins [34]. Mitochondrial proteins encoded by NFU1 play an important role in the assembly of mitochondrial Fe-S clusters and intracellular iron homeostasis in yeast [35]. In our research, after PAT stimulation, the expression of mel1, ETR1, YCP4, PKR1, MIA40, COX17, and NFU1 were all increased to provide enhanced energy utilization, regulated mitochondrial function, promoted respiratory metabolism, and enhanced cell vitality in M. guilliermondii, thus reducing cell damage caused by PAT stress (Figure 6). Similarly, in the case of C. guilliermondii, PAT stimulation induced the accumulation of heat shock protein 70 to prevent the damage caused by PAT to the yeast cell [6].
In addition to growth-related genes, our transcriptome results also found differential regulation of a range of genes related to transcription and translation. Cyclins (encoded by Pch1) form a complex with cyclin-dependent kinase 9 (encoded by Cdk9), a forward transcription extension factor, to regulate the capping and elongation of transcripts [36]. The BUD31 of yeast contributes to spliceosome assembly, thus promoting effective pre-mRNA splicing [37]. The gene RPC1 encodes the RNA polymerase III subunit, an important part of RNA polymerase III, involved in tRNA, rRNA, and other essential RNA synthesis [38]. MPP10 encodes the U3 small nucleolar RNA protein (MPP10), an instantaneous correlation factor of eukaryotic ribosomal synthesis, and MPP10 forms a protein complex in 90S ribosomal precursors, which carries out early processing of 18S rRNA [39]. In our results, Pch1, Cdk9, BUD31, RPC1, and MPP10 were up-regulated in M. guilliermondii after PAT stimulation. The up-regulation of these genes could reduce and prevent the transcription and translation blockage caused by PAT to ensure a smooth transcription and translation process ( Figure 6).
PAT exposure in yeasts significantly up-regulates the genes that regulate transporters [40]. PAT stimulation in M. guilliermondii up-regulated 11 transport-related genes, including the YOR1 [41], which encodes oligomycin-resistant ATP-dependent permease, and dotC [42], which encodes an efflux pump. Both YOR1 and dotC belong to the genes encoding the ATP binding box or major promoter superfamily transporter associated with pumping toxic substances out of cells. Likewise, the nicotinic acid transporter (encoded by TNA1) helps the extracellular quinolinic acid to enter cells, thereby increasing intracellular NAD+ concentration [43]. NAD+ biosynthesis is associated with the yeast's lifespan, and increased intracellular NAD+ levels could prolong the yeast's lifespan [44]. Remy et al. showed that TPO1 endows Arabidopsis with multiple drug resistance (MDR). Yeast expressing TPO1 showed higher tolerance to many herbicides and fungicides. In contrast, yeast mutants lacking TPO1 showed sensitivity to many drugs [45]. Zn is an important cofactor of transcription factors and enzymes and is essential to all organisms. The gene zrt1 encodes Zn-regulating membrane proteins that can maintain the dynamic balance of Zn in cells and maintain cell homeostasis [46]. Similarly, the pantothenic acid transporter (encoded by liz1) is responsible for the transport of vitamin pantothenic acid and CoA, located in the plasma membrane, which is necessary for cell pantothenic acid uptake [47]. In our research, the expression of YOR1, dotC, TNA1, TOP1, and zrt1 all showed up-regulation, which might be associated with inorganic ions and nutrient utilization. These molecular events could maintain the homeostasis of M. guilliermondii cells and enhance cell viability. At the same time, the up-regulation of proteins related to drug transport might be related to the intracellular transport of PAT ( Figure 6).
Short-chain dehydrogenase/reductases (SDR) are the largest and most diverse enzyme superfamilies in all life forms, including bacteria, fungi, plants, and animals. SDRs possess broad substrate specificity and multiple biological functions, such as lipid, amino acid, steroid hormone biosynthesis, and xenobiotic metabolism [48]. The role of SDRs in PAT detoxification and induced expression of SDR genes under PAT stress was already reported in several yeasts. The SDRs of Sporobolomyces sp. [40], Candida guilliermondii [6], R. mucilaginosa [49], and S. cerevisiae were found to be up-regulated upon PAT stress. Xing et al. (2021) studied the direct involvement of SDRs in PAT detoxification by cloning the CgSDR gene from Candida guilliermondii into E. coli. They found that the purified CgSDR protein could reduce 80% of PAT from apple juice. The CgSDR could transform toxic PAT into non-toxic E-ascladiol in vitro with NADPH as a coenzyme [50]. In the present study, the significant up-regulation of several SDRs was observed in M. guilliermondii under PAT stress, which speculated that SDRs could be involved in PAT detoxification [51]. Similarly, the role of glutathione S-transferase (GST) in PAT detoxification was described from the proteomic analysis of PAT-induced R. mucilaginosa. GST could catalyze the conjugation of a reduced form of GSH with PAT in R. mucilaginosa upon PAT induction [49]. In our results, the GST expression was induced in M. guilliermondii under PAT stress. Collectively, our study supports the previously proposed hypothesis of PAT detoxification into E-ascladiol via SDRs and the involvement of GST. Studies on the direct participation of SDRs and GST of M. guilliermondii in PAT detoxification and the degradation products by heterogeneous expression and purification are ongoing.

Conclusions
In summary, through transcriptome sequencing and the subsequent bioinformatics analysis, this study revealed the changes in the gene expression during the response regulation of M. guilliermondii to PAT. Based on the transcriptome results, the genes related to post-translational modification, DNA damage repair, resistance and drug resistance, oxidative stress resistance, growth and reproduction, transcription, and translation regulation and transport were excavated, and the genes involved in PAT detoxification were distinguished. Collectively, the findings presented in this study will serve as a foundation for the further understanding of the molecular responses and the PAT detoxification mechanisms of M. guilliermondii. Transcriptome analysis will provide more insights for further research and development of PAT biodetoxification strategies.
Author Contributions: Conceptualization, Q.Y., Y.F. and H.Z.; methodology, Y.F. and D.S.; formal analysis, Y.F. and X.Z.; data curation, Y.F. and X.Z.; writing-original draft preparation, Q.Y., X.Z. and D.S.; writing-review and editing, Q.Y., X.Z., D.S. and H.Z.; project administration, Q.Y. and H.Z. All authors have read and agreed to the published version of the manuscript. Data Availability Statement: The transcriptome data of M. guilliermondii cultured with and without PAT used in this study were deposited in the NCBI Sequence Read Archive (SRA) Sequence Database with accession numbers SRR24233557 to SRR24233562.

Conflicts of Interest:
The authors declare that they have no known competing financial interest or personal relationships that could have appeared to influence the work reported in this paper.