Antifungal Activity and the Mechanism of Ozone Against Spoilage Molds, Such as Gibberella Intermedia and Aspergillus Ochraceus Isolated From Freshly-Peeled Garlic


 The antibacterial effect and mechanism of ozone (O3) treatment on freshly-peeled garlic inoculated with Gibberella intermedia (G. intermedia) and Aspergillus ochraceus (A. ochraceus) in different conditions were studied in vivo. The results showed that the specific O3 concentration, treatment time, and humidity significantly affected the garlic's indexes. The optimal treatment conditions of garlic inoculated with both molds were 6 ppm, 15 min, and 90%, respectively. After treatment with those conditions, the incidence etc of garlic were significantly reduced during storage. Differential analysis was performed for the RNA-sequencing and gene expression of the molds subjected to O3 treatment, as well as the samples that were not. The results showed that O3 treatment inhibited the growth of G. intermedia and A. ochraceus. Consequently, 2754 and 2378 differentially expressed genes, 1456 and 1591 up-regulated genes, and 1298 and 787 down-regulated genes were screened, respectively, for these molds, affecting the catalytic activity and various other pathways.


Introduction
Garlic (Allium sativum L.) is a herbaceous plant containing proteins, fats, minerals, polysaccharides, avonoids, and speci c allicin and phenolic components, as well as sulfur-containing compounds (Bozin et al., 2008;Natália et al., 2016). Garlic has been used as a food avoring for thousands of years due to its moderately spicy avor and high levels of organosulfur compounds, such as allicin (Wu et al., 2016). It has historically been used to treat aches and pains, leprosy, deafness, severe diarrhea, constipation, parasitic infections, and asthma, as well as to lower fever, ght infections, and relieve stomachaches (Rana et al., 2011). In their current fast-paced lives, people require convenience when it comes to fruit and vegetables. Consumers increasingly favor freshly-peeled garlic because of exactly this reason. However, when the protective epidermis is removed, freshly-peeled garlic cloves have a shorter shelf life due to enzymatic/nonenzymatic browning, sprouting, moisture loss, microbial spoilage, and surface discoloration (Singh et al., 2019), and is prone to decay, which causes it to lose its commercial value. Previous research isolated and identi ed the main mold strains that cause the spoilage of freshly-peeled garlic as G. intermedia and A. ochraceus. Gibberellin and Aspergillus produced by G. intermedia andA. ochraceus respectively, they all causes rapid germination and excessive growth of freshly-peeled garlic, which leads it to rapid consumption of nutrients, color change, mildew. Consequently, it is important to nd suitable antimicrobial and preservation agents to improve the quality of freshly-peeled garlic and control the growth of G. intermedia and A. ochraceus. Many traditional chemical preservatives are limited because of increasing health risk concerns. This creates a need for developing an antimicrobial agent with no residue and a strong bactericidal effect to inhibit the activity of G. intermedia and A. ochraceus, while ensuring the storage quality of freshly-peeled garlic.
O 3 , as one of the allotropic forms of oxygen, is a powerful antimicrobial agent because of its potential oxidizing capacity (Mylona., 2014). In 1997, O 3 was granted Generally Recognized As Safe (GRAS) status (US-FDA, 1997) and had since received full US-FDA approval as a direct contact food sanitizing agent (US-FDA, 2001) (Tzortzakis., 2006). Recently, O 3 has been gradually applied in the food industry because of its many advantages, such as strong oxidation, while decomposing to oxygen without leaving a trace in the treated substrates (Andrew et al., 2015; Qi et al., 2016). Nowadays, O 3 is attracting increasing attention as a potential method for treating fruits ( Sarig et al., 1996), vegetables, and grain, while reducing mold contamination or mycotoxins in various food products (Freitas-Silva, and Venancio, 2010) ( Ong M.K. and Ali, 2015; Karaca, Velioglu, and Nas, 2010) Furthermore, the bactericidal effects of O 3 on a wide variety of microorganisms have been con rmed, which includes Gram-positive and Gram-negative bacteria, as well as bacterial spores (Brodowska, Nowak, and Smigielski, 2018). Franco C (2008) reported that O 3 could effectively degrade patulin, both in aqueous solutions and in diluted apple juice. Cho et al. (2009) reported that treating fresh vegetable juice with O 3 decreased the microbial numbers, while the treated sample's chemical characteristics displayed no differences compared with the control. Additionally, the antimicrobial activity of O 3 is obtained directly via the progressive oxidation of vital cell components leading to the inhibition of microbial growth (M.E. Parish et al., 2003). Therefore, the proliferation of G. intermedia and A. ochraceus in freshly-peeled garlic can be inhibited by O 3 , while improving the methods used for the storage and preservation of freshly-peeled garlic.

Materials
The garlic was purchased in Bubugao supermarket, Hongguang Town, Pidu District, Chengdu. Peeled garlic was left at room temperature until moldy. The G. intermedia and A. ochraceus was obtained via the separation and puri cation of the moldy garlic.
G.intermedia (SICC3.976) and A. ochraceus (SICC3.975), which were stored in the Southwest Center of Industrial Culture Collection in China), and isolated from spoiled and moldy freshly-peeled garlic, were used throughout the study.

Live inoculation
Freshly-peeled garlic that was uniform in size and free of pests and diseases was selected, washed with tap water, and dried. The samples were divided into four groups, each weighing 2 kg. The samples' surfaces were sterilized with 75% alcohol and placed on standby after ultraviolet irradiation for 30 min. A wound with a 2 mm diameter and a depth of 5 mm was pricked symmetrically in the middle of each piece of garlic using a sterile inoculating needle. After 30 min, 5 µL of scab and A. ochraceus and G. intermedia suspension with spore concentration of 1 × 10 6 CFU/mL were inoculated respectively in each group of garlic wounds. Each group was divided into different fresh-keeping boxes sterilized with O 3 . Freshly-peeled garlic inoculated with different molds was treated with different O 3 concentrations (1 ppm, 2 ppm, 3 ppm, 4 ppm, 5 ppm, and 6 ppm), times (3 min, 6 min, 9 min, 12 min, 15 min, and 18 min), and humidity (85%, 90%, and 95%). The untreated samples were considered the control group. The total number of colonies (TNC), the total number of mold and yeast (TNMY), incidence, lesion diameter, and lesion depth of each group of freshly-peeled garlic were measured every other day .
2.4. The determination of the TNC, TNMY, incidence, diameter, and lesion depth The TNC and TNMY were quanti ed according to the method described by Akshata et al. (2019). The estimation of TNC was performed on plate count agar, while the TNMY was determined using PDA via the spread plate method. The colony-forming units (CFU) were counted after incubation at 37 ℃ for 24 h, while the molds and yeasts were identi ed after incubation at 30 ℃ for 48 h.
The incidence and weight of the garlic were counted and weighed the next day. The incidence was calculated using the following formula: Incidence (%) = (weight of diseased garlic (g)) / (weight of total inoculated garlic (g)) × 100% The lesion diameter was measured the next day. Here, ten garlic samples were used for each measurement.
Calculations were performed using vernier calipers and the cross method, while the garlic's average lesion diameter was also taken.
The lesion depth was measured the next day. Here, 10 garlic samples were used for each measurement and cut along the wound direction. Calculations were performed using vernier calipers, and the average value was taken as the nal result.

Analysis of the differentially expressed genes
Here, 20 mL PDB containing G. intermedia and A. ochraceus conidia was placed on blank Petri dishes (5 dishes per treatment). The dishes' lids were removed to allow the air ow and placed in treatment rooms (20°C, and the room humidity was 95%), where they were exposed to continuous O 3 at concentrations of 0 ppm and 6 ppm for 20 min. At the end of each exposure, the PDB medium containing the G. intermedia and A. ochraceus conidia was transferred to a sterile centrifugal tube, after which it was placed in a shaker at 28°C for 72 h. At the end of incubation, the mycelium of G. intermedia and A. ochraceus was collected for O 3 treatment, and non-O 3 treatment after centrifugation at 4°C for 15 min and repeatedly washed with phosphate buffer saline (PBS).
The total RNA was extracted using an RNA Extraction Kit (Tiangen Biotech CO., LTD., Peking, China) according to the manufacturer's instructions and the method reported bywith some modi cations. The purity and integrity of the RNA samples were detected using the NanoDrop and Agilent 2100 methods after removing residual DNA. After the sample passed the test, the PCR product was heat-denatured into a single strand, and a single-stranded cyclic DNA library was obtained via the cyclization of the single-stranded DNA with a bridge primer. Finally, the library was sequenced using the BGIS EQ-500 Sequencing Platform.
Data from the BGIS EQ-500 Sequencing Platform are known as raw reads or raw data, while it is called clean reads after ltration. The de novo assembly of the clean reads was performed using Trinity, after which the assembled transcripts were clustered for redundancy using TGICL. These sequences were de ned as UniGenes. Subsequently, the assembled UniGenes were annotated by seven kinds of protein databases, including the Kyoto Encyclopedia of Genes and Genomes (KEGG) database, the Gene Ontology (GO) database, the NCBI nonredundant protein (NR) database, the Nonredundant nucleic acid (NT) database, the Swiss-Prot protein database, the Pfam database and the clusters of Karyotic Orthologous Groups (KOG) database.
Using NR annotation, the Blast2GO program was used to obtain the GO annotations for the UniGenes.
Furthermore, the log 2 fold change >4 and p < 0.001 were used as the standard for evaluation and screening to select the differentially expressed genes of G. intermedia and A. ochraceus. To compare the differences in gene expression, the differentially expressed genes were subjected to GO enrichment analysis, and those associated with the O 3 inhibition of G. intermedia growth were identi ed.

Statistical analysis
The tests in this investigation were carried out in triplicate. The test results were analyzed using SPSS 20.0 software (SPSS Inc.) and expressed as mean ± standard deviation. The one-way analysis of variance procedure, followed by the Student-Newman-Keuls test, was used to determine the signi cant difference (p < 0.05) between the treatment means.

The effect of different O 3 treatment conditions on the TNC in freshly-peeled garlic
As shown in Fig. 1(A), during storage, the TNC on the surface of fresh garlic inoculated with G. intermedia and A. ochraceus at different O 3 concentrations was signi cantly lower than that of the blank group (P < 0.05). Cunkun Chen et al. (2020) reported similar ndings that O3 treatment can inhibit the bacteria and fungi that lead to the corruption of Hami melon by postharvest pathogens at appropriate doses. With an increase in storage time, the TNC on the surface of fresh garlic in each experimental group increased. For the fresh garlic inoculated with G. intermedia, the TNC increased with an increase in the O 3 concentration, while 6 ppm was the optimal O 3 concentration level. For freshly-peeled garlic inoculated with A. ochraceus, the TNC displayed a lower level of decline when the O 3 concentration increased from 1 ppm to 3 ppm, while the decrease was higher when the O 3 concentration changed from 4 ppm to 6 ppm. When the garlic was stored for 10 d, no signi cant difference was evident between the freshly-peeled garlic treated with 1 ppm O 3 and the blank group (P > 0.05). Therefore, it can be concluded that an O 3 treatment concentration of 6 ppm was optimal in the freshly-peeled garlic inoculated with A. ochraceus.  humidity and the treatment group with 90% humidity (P < 0.05). During the 10 d of storage, the TNC on the fresh garlic's surface inoculated with the types of bacteria exhibited no signi cant change after treatment with O 3 gas at 90% and 95% humidity, respectively. After storage for 10 d, the TNC on the surface of freshlypeeled garlic inoculated with G. intermedia and A. ochraceus was 3.37 lg CFU/g and 3.92 lg CFU/g, respectively. Therefore, when considering lower costs, the optimal lg CFU/g treatment humidity was 90%. ppm, the TNMY decreased as the treatment concentration continued to increase. Moreover, the TNMY in the fresh garlic was the lowest after treatment with 6 ppm O 3 .
As shown in Fig. 2(B), the TNMY on the surface of the fresh garlic decreased with an increase in the treatment time and storage period at the same storage time. When the O 3 treatment time increased from 15 min to 18 min, no signi cant difference was evident in the TNMY (P > 0.05) levels, but it was signi cantly lower than in the blank group and the previous experimental group (P < 0.05), while the better O 3 treatment time was 15 min. P. Sarig et al. (1996) found that when berries were exposed to O 3 for 20 min, the number of CFU of the fungi, yeast, and bacteria that naturally exist on the surface of the fruit was signi cantly reduced. O 3 treatment substantially reduced the level of decay caused by fungi after cold storage and prolonged the quality assurance period of the berries. This is consistent with the results indicating that the quantity of mold and yeast decreased after the fresh garlic was treated with O 3 for a particular time at a speci c concentration. Fig. 2(C) shows that after O 3 treatment at 90% and 95% humidity, respectively, the TNMY in the freshlypeeled garlic inoculated with G. intermedia and A. ochraceus was always lower than in the treatment group subjected to 85% humidity. After 10 d of storage, the TNMY in the O 3 treatment group at 90% humidity was 3.55 lg CFU / g and 3.48 lg CFU / g, respectively, which was considerably lower than in the treatment group exposed to 85% humidity.

The effect of different O 3 treatment conditions on the incidence of freshly-peeled garlic
As shown in Fig. 3(A), O 3 treatment signi cantly inhibited the incidence of freshly-peeled garlic inoculated with G. intermedia and A. ochraceus. After inoculation with G. intermedia, the garlic incidence decreased in conjunction with an increase in the O 3 concentration at the same storage time. After 10 d of storage, the fresh garlic incidence in the 6 ppm treatment group was the lowest at only 44.85%, which was substantially lower than in the blank group (98.68%). Furthermore, when the fresh garlic was inoculated with A. ochraceus, the differences between the blank, the 1 ppm, and the 2 ppm O 3 treatment groups were small, and the total incidence was high at the same storage time. Additionally, when the O 3 concentration increased to 6 ppm, the fresh garlic incidence was the lowest of all the storage times, while the germicidal effect was the best.
After O 3 treatment at different times, the incidence of the freshly-peeled garlic inoculated with G. intermedia and A. ochraceus during storage is shown in Fig. 3 (B(a)) and Fig. 3 (B(b)), respectively. With an extended storage time, the incidence of the garlic increased gradually. After 10 d of storage, the incidence of the blank group reached 100%. With an increase in the treatment time, the garlic incidence decreased slowly at the same storage time. After 15 min of O 3 treatment, the incidence of the fresh garlic inoculated with the two types of mold remained low. When the treatment time continued to increase, no signi cant reduction was evident. After 10 d of storage, the incidence was 49.75% and 44.55%, respectively. Fig. 3(C) indicated that during the entire storage period, the curve of the O 3 treatment group at 85% humidity exceeded the curve of the O 3 treatment groups at 90% and 95% humidity, and the incidence of fresh garlic treated with 90% and 95% O 3 was signi cantly lower than that of the O 3 treatment group at 85% humidity. At the end of the storage period, the freshly-peeled garlic incidence exposed to 90% humidity was 49.75% and 44.55%, respectively. From the perspective of saving resources, 90% O 3 humidity was selected as appropriate, which was consistent with the results suggesting that O 3 could reduce the incidence rate of papaya and red peppers, according to Marcin (2016). concentration became higher, while the lesion diameter of the blank group always remained at the maximum level. After O 3 treatment at different concentrations, the 6 ppm group displayed the best bacteriostatic effect and the smallest lesion diameter during storage. After 10 d of storage, the lesion diameter of the freshlypeeled garlic inoculated with G. intermedia and A. ochraceus was 3.07 mm and 2.67 mm, respectively, which was substantially lower than in the blank group (5.53 mm and 5.90 mm respectively).    5(C) shows that from the beginning of storage and at the same storage time point, when the humidity increased from 85% to 90%, the disease spot depth of the freshly-peeled garlic inoculated with G. intermedia decreased signi cantly. When the humidity continued to increase to 95%, the disease spot depth remained unchanged. Regarding the freshly-peeled garlic inoculated with A. ochraceus when the storage time was less than 4 d, no signi cant differences were evident between the three O 3 treatments. From the 6th day of storage, the lesion depth of the freshly-peeled garlic treated with 85% humidity was signi cantly higher than that of freshly-peeled garlic exposed to 90% and 95% humidity. Based on these results, 90% humidity was selected as optimal during the O 3 treatment of G. intermedia. 3.6. Analysis of the transcriptome differences in the putrefactive fungi in fresh garlic 3.6.1. Sample RNA test results of G. intermedia and A. ochraceus The RNA of the O 3 treated, and untreated Gibberella and ochratoxin were extracted, and the nanodrop UV-Vis spectrophotometer was used to verify the results (Table 1). Fig. 6 shows that the two treatment groups' baselines were stable, and the 5S peak was normal. Furthermore, the purity, concentration, and integrity of the extracted RNA were consistent with subsequent analysis requirements.  Table 2 shows the quality assessment of the O 3 -treated and untreated gibberellin and ochratoxin transcriptome. The sequencing data of the processed and untreated G. intermedia and A. ochraceus were good, meeting subsequent analysis requirements.  Table 3 shows the results after aggregating the output data, indicating that the assembly quality of G. intermedia and A. ochraceus were good. The statistics for the gene annotation rate of the transcriptome sequencing data of G. intermedia and A. ochraceus are shown in Fig. 7. The overall annotation rate of the transcriptome sequencing data of G. intermedia and A. ochraceus reached respective levels as high as 99.53% and 89.89%, providing a database for subsequent gene analysis. 3.6.6. The GO annotation of the differentially expressed genes for G. intermedia and A. ochraceus There were 16, 13, and 8 annotation classi cations for the biological process, cell composition, and molecular functionality, respectively (Fig. 9). According to the classi cation of differential gene GO annotations, the gene expression and regulation of Gibberella were in uenced by the metabolic process, cell process, and localization of the biological process after O 3 treatment. The gene expression and regulation of Gibberella were in uenced by the cell, cell part, organelle, membrane, and membrane part of the cell composition. The gene expression and regulation of Gibberella were in uenced by the catalytic activity, binding activity, transport activity, and transcriptional regulation activity of the molecular functions. The pathways with more differentially expressed genes include the metabolic process, cell process, localization, cell, cell part, membrane, membrane part, binding, and catalytic activity.
The pathways with more differentially expressed genes include the metabolic process, the cell process, localization, cell, cell part, membrane, membrane part, binding, and catalytic activity. Furthermore, there were 18, 13, and 10 annotation classi cations for the differentially expressed genes in the biological process, cell composition, and molecular functionality, respectively. According to the classi cation of differential gene GO  Table 4. First, regarding cell composition, the nodes with signi cant enrichment in Gibberella were the membrane, the cell, the organelle, the macromolecule complex, and nucleoid, while the number of differentially expressed genes was 1, 2, 3, 2, and 1, respectively. The nodes with signi cant enrichment in the molecular functions were the catalytic activity, binding transport activity, signal sensor activity, and transcriptional regulation activity, while the number of differentially expressed genes was 19, 13, 6, 1, and 1, respectively. There were 18, 7, 9, 1, and 5 differentially expressed genes in the cell process, biological regulation process, localization, stress response, and cell component tissue or biogenesis. Second, regarding the ochratoxin, the nodes with signi cant enrichment in the cell composition were the cells and organelles, while the number of differentially expressed genes was 9, 10, and 1, respectively. The nodes with signi cant enrichment in molecular functions were the catalytic activity, binding, signal sensor activity, and structural molecular activity, while the number of differentially expressed genes was 50, 11, 3, and 1, respectively.
There were 48, 16, 8, 4, 1, 1, and 6 differentially expressed genes in the biological processes, including the cell process, biological regulation process, tissue or biogenesis of the cell components, metabolic process, and multiple biological processes stress response and localization. 3.6.8. The annotation of the differentially expressed gene KEGG of G. intermedia and A. ochraceus The annotation classi cation map of the KEGG pathway of the O 3 -treated and untreated gibberellins is shown in Fig. 10(a). Here, 2503 differentially expressed genes were annotated into the KEGG database, 22 metabolic pathways were annotated by the KEGG pathway of which two metabolic pathways belonged to the cell process, two to the environmental information process, and four to the genetic information process, while 13 were metabolic processes, and one denoted the organic system. There were 176 differential gene annotations to the transport and catabolism pathway, 117 to the cell growth and death pathway, 118 to the signal transduction pathway, 64 to the gene transcription pathway, and 167 to the gene transcription pathway during environmental information processing. In the translation process, 171 differential gene annotations were included in the gene folding, classi cation, and degradation pathways, 77 in the DNA replication and repair pathways, 237 in the carbohydrate metabolism pathway, and 552 in the overall overview pathway.
The annotation classi cation diagram of the KEGG pathway denoting the differential expression genes between the O 3 -treated and the untreated Aspergillus ochre is shown in Fig. 10(b). Here, 1934 differentially expressed genes were annotated into KEGG by Aspergillus ochre, including 22 metabolic pathways in total, of which two were due to the cellular process, two due to environmental information processing, four due to genetic information processing, 12 due to the metabolic process, and two due to the organic system.
Furthermore, there were 156 differential gene annotations during the process of cell transportation and catabolism, 40 during the process of cell growth and death, 76 during the process of environmental information processing, 35 during the process of genetic information processing, and 107 during the process of gene transcription. In the process of translation, there were 80 differential gene annotations into the gene folding, classi cation, and degradation pathways, 30 into DNA replication and repair pathways, 229 into the carbohydrate metabolism pathway, 496 into the whole body overview pathway, 182 into the amino acid metabolism pathway, and 123 into the lipid metabolism pathway.
3.6.9. The enrichment of allogeneic KEGG for G. intermedia and A. ochraceus As shown in Table 5, the statistics regarding the signi cant KEGG enrichment (p-value < 0.05) of the differentially expressed genes between the O 3 -treated and untreated Gibberella and ochratoxin were studied, and the metabolic process of the genes involved in the inhibition of Gibberella and ochratoxin growth by O 3 was explored. The results indicated that O 3 affects the growth of Gibberella via the KEGG pathway process mentioned above, and even causes its death. In addition, O 3 mainly affects the growth and death of ochratoxin by impacting its amino acids, overall overview, and lipid metabolism. The primary target of ozone is the cell surface where degradation of unsaturated lipids of the cell envelope occurs, followed by the leakage of cellular contents and bacterialcells lyses (Komanapalli & Lau, 1996). Komanapalli and Lau(1996) further asserted that ozone could in ltrate the microorganisms to oxidize inner contents such as proteins, nucleic acids and enzymes. analysis of treated and untreated rot-causing molds were carried out to explore the inhibition mechanism of ozone on freshly-peeled garlic rot-causing molds, which provides a theoretical basis for further study and ozone preservation of fresh-peeled garlic.
In this study, it was found that ozone treatment could signi cantly reduce the total number of colonies, molds and yeasts, incidence, spot diameter and spot depth of freshly peeled garlic during storage, and achieved a better germicidal and inhibitory effect. So far, the germicidal effect of ozone on fruits and vegetables has been applied to broccoli, pepper, strawberry, guava, papaya and other fruits and vegetables (Nur aida MP et al ,2011), as well as fresh-cut lotus root slices, fresh-cut pineapple, fresh-cut kiwifruit and other fresh-cut fruits and vegetables (Tzortzakis N et al, 2007;Karaca H et al, 2014) . It reacts with the components of microbial cells on the surface of fruits and vegetables, destroys the components of the membrane, causes metabolic imbalance, further destroys the tissue inside the cell membrane, and produces irreversible destruction, thus killing microorganisms. The results of the study will provide a theoretical basis for the further study of the bacteriostatic mechanism of ozone, and provide a certain theoretical basis and technical guidance for the application of ozone in the processing industry and commercial development of freshly-peeled garlic. it has guiding signi cance for the industrialization of ozone fresh-keeping processing of fresh-peeled garlic.

Abbreviations
The total number of colonies TNC  Figure 1 The The gene annotation rate statistics for (a) G. intermedia and (b) A. ochraceus The numerical statistics of the differentially expressed genes for (a) G. intermedia and (b) A. ochraceus The GO annotated statistical map of the differentially expressed genes for (a) G. intermedia and (b) A.