Antifungal Volatile Organic Compounds from Streptomyces setonii WY228 Control Black Spot Disease of Sweet Potato

ABSTRACT

Volatile organic compounds (VOCs) produced by microorganisms are considered promising environmental-safety fumigants for controlling postharvest diseases. Ceratocystis fimbriata, the pathogen of black spot disease, seriously affects the quality and yield of sweet potato in the field and postharvest. This study tested the effects of VOCs produced by Streptomyces setonii WY228 on the control of C. fimbriata in vitro and in vivo. The VOCs exhibited strong antifungal activity and significantly inhibited the growth of C. fimbriata. During the 20-day storage, VOC fumigation significantly controlled the occurrence of the pathogen, increased the content of antioxidants and defense-related enzymes and flavonoids, and boosted the starch content so as to maintain the quality of the sweet potatoes. Headspace analysis showed that the volatiles 2-ethyl-5-methylpyrazine and dimethyl disulfide significantly inhibited the mycelial growth and spore germination of C. fimbriata in a dose-dependent manner. Fumigation with 100 μL/L 2-ethyl-5-methylpyrazine completely controlled the pathogen in vivo after 10 days of storage. Transcriptome analysis showed that volatiles mainly downregulated the ribosomal synthesis genes and activated the proteasome system of the pathogen in response to VOC stress, while the genes related to spore development, cell membrane synthesis, mitochondrial function, and hydrolase and toxin synthesis were also downregulated, indicating that WY228-produced VOCs have diverse modes of action for pathogen control. Our study demonstrates that fumigation of sweet potato tuberous roots with S. setonii WY228 or use of formulations based on the VOCs is a promising new strategy to control sweet potato and other food and fruit pathogens during storage and shipment.

IMPORTANCE Black spot disease caused by Ceratocystis fimbriata has caused huge economic losses to worldwide sweet potato production. At present, the control of C. fimbriata mainly depends on toxic fungicides, and there is a lack of effective alternative strategies. The research on biological control of sweet potato black spot disease is also very limited. An efficient biocontrol technique against pathogens using microbial volatile organic compounds could be an alternative method to control this disease. Our study revealed the significant biological control effect of volatile organic compounds of Streptomyces setonii WY228 on black spot disease of postharvest sweet potato and the complex antifungal mechanism against C. fimbriata. Our data demonstrated that Streptomyces setonii WY228 and its volatile 2-ethyl-5-methylpyrazine could be a candidate strain and compound for the creation of fumigants and showed the important potential of biotechnology applications in the field of food and agriculture.

INTRODUCTION

Plant-associated microorganisms produce rich metabolites, including diffusible and volatile compounds, which play important roles in plant growth and pathogen control (1–3). In addition to the well-studied diffusible active compounds, the biotechnological applications of volatiles emitted by microorganisms have attracted increased attention in the past 2 decades (4–6). Volatiles produced by microorganisms include inorganic substances such as CO₂, NH₃, and HCN and volatile organic compounds (VOCs) with changeable structures, including alcohols, ketones, aldehydes, terpenes, sulfur-containing and nitrogen-containing compounds, and others (7). They have the physical-chemical characteristics of low molecular weight, high vapor pressure, low boiling point, lipophilic moiety, and volatilizable, are easy to move via the gaseous phase, and diffuse in air and soil for a long distance (8, 9). Previous studies have shown that microbial VOC composition varies depending on microbial species, medium nutrition, temperature, growth period, and other growth conditions, and environmental factors will affect the microbial VOC profile emission (10). Microbial VOCs play multiple functions, such as promoting plant growth, inducing systematic resistance, inhibiting pathogens, and killing insects (11–13). As important plant health modulators and signal molecules, VOCs play important roles in regulating plant-microbe cross talk and microbe-microbe communication and interactions (14).

Actinobacteria are the main inhabitants of rhizosphere microorganisms and play beneficial roles in plant health mediation (15). Streptomyces, as the largest genus within the actinomycetes, has outstanding biocontrol potential. Streptomyces spp. can produce abundant secondary metabolites and hydrolases, which showed great application potential in the fields of drug development, biological control, and plant protection (16, 17). In recent years, it has been found that Streptomyces spp. can produce rich volatile organic compounds, which displayed a variety of biological activities, including pathogen control, plant growth promotion, and insecticidal activity (18–20). For example, the nesting preference of an invasive ant is associated with the volatiles geosmin and 2-methylisoborneol produced by soil actinobacteria (21). VOCs produced by Streptomyces yanglinensis 3-10 can inhibit the growth and toxin production of Aspergillus flavus in stored soybeans (22). However, compared with bacteria and fungi, there are not enough studies and reports on the VOC profile and function of Streptomyces strains, and the application of their VOCs in agriculture and the underlying mechanism of their VOC-mediated killing pathogens has not yet been well investigated.

Sweet potato (Ipomoea batatas [L.] Lam.) is nutritionally balanced and complete. It has been recognized as one of the most nutritious and healthy foods worldwide. However, it can be easily infected by pathogens and thus is difficult to store because of its crisp and tender tissue, rapid postharvest water evaporation, and easy mechanical damage (23, 24). Sweet potato black spot disease is caused by the ascomycete fungus Ceratocystis fimbriata, which mainly endangers potato seedlings and tubers. Black spot is a serious disease of sweet potato which occurs in all sweet potato-producing areas in the world, resulting in significant yield losses. It occurs from the seedling stage to the field growth stage to the postharvest and storage stages. It is estimated that the yield loss of sweet potato in the field and postharvest caused by C. fimbriata is 60% to 80%, which corresponds to more than 150 million dollars (25, 26). In addition to cultivating disease-resistant sweet potato varieties, the main prevention and control measures for postharvest sweet potato black spot disease are low temperature (10 to 15°C) and fungicide treatment, such as formalin, chitosan, carbendazim, and trifloxystrobin (27, 28).

With the enhancement of environmental protection awareness and consumers’ increasing requirements for the quality of sweet potato, biological prevention and control methods have attracted a higher degree of attention (29, 30). Microbial VOC fumigation is an alternative method which has the advantages of high sterilization, low toxicity, low residue, and no pathogen resistance. At present, there are some successful cases of microbial VOCs in the prevention and control of pathogens in postharvest fruits and vegetables, and the development of commercial microbial fumigants has been reported (31–33). However, there are few reports on the control of postharvest sweet potato by microbial VOC fumigation.

In our previous study on the biological control potential of actinomycete VOCs, we found that the VOCs produced by endophytic Streptomyces setonii WY228 showed excellent plant growth-promoting and antifungal activities (34), and it has the development potential of agricultural fumigants. Therefore, the aims of this paper are to (i) study the antifungal effect of WY228 VOCs against Ceratocystis fimbriata in vitro and in vivo, (ii) analyze the VOC composition and specific antifungal VOC components, (iii) explore the effect of VOC fumigation on the quality of postharvest sweet potato, and (iv) study the possible antifungal mechanism of VOCs on C. fimbriata by transcriptome sequencing. The purpose of this paper is to provide a feasible new biocontrol strain for the control of black spot disease in postharvest sweet potato and reveal its underlying mechanism.

RESULTS

VOC antifungal activity and fumigation to control black spot disease of postharvest sweet potato.

As shown in Fig. 1a, the VOCs produced by strain WY228 on Trypticase soy agar (TSA) medium exhibited strong antifungal activity. The mycelial growth of black spot pathogen C. fimbriata on a potato dextrose agar (PDA) plate was completely inhibited by WY228 volatiles. Moreover, it can be seen from Fig. 1b that with the extension of storage time, the surface lesion diameter of sweet potato infected with black spot pathogen C. fimbriata gradually increased. The control sweet potato (TR+CF) group showed that the cross section of sweet potato also turned black, and the area infected by C. fimbriata increased with time. However, in the fumigation treatment (TR+CF+WY228) group, there was no expansion of sweet potato epidermal lesions, and there was no fungal infection in the cross section. The weight loss rate of sweet potato with or without WY228 fumigation had little significant change after short-term storage (6 to 10 days). On the 20th day, compared with normal sweet potato (TR), the weight loss rate of C. fimbriata-infected sweet potato increased from 5.92% to 7.03%, but the weight loss rate of the WY228 fumigation group decreased significantly to 3.94% (Fig. 1c). We can see from Fig. 1d that the VOCs produced by WY228 can alleviate the infection degree of sweet potato black spot disease after fumigating on different days. On the 20th day, the incidence of sweet potato without fumigation reached grade 5, while the incidence of the fumigated group was only grade 2. These results showed that WY228-produced VOCs on TSA medium had a positive effect on the control of postharvest black spot disease of sweet potato.

FIG 1.

Effects of mycelial inhibitory activity against C. fimbriata of VOCs produced by strain WY228 on TSA medium, and biological control of sweet potato black spot disease in vivo via fumigation. (a) Antifungal activity by strain WY228 VOCs on TSA medium. (b) Phenotypic photos of VOC-fumigated and nonfumigated sweet potato on the 6th, 10th, and 20th days of storage. (c) The weight loss rate of sweet potato on different days; different letters indicate significant differences (P < 0.05) between different treatment groups. (d) Disease severity (DS) on different days, asterisks (**, P < 0.01) indicate a very significant difference between the two groups. The data represent the mean ± standard deviation. CF, black spot pathogen Ceratocystis fimbriata; TR, sweet potato tuberous roots.

VOC fumigation on the quality and physiology of postharvest sweet potato.

The physiology traits and quality of postharvest sweet potato after VOC fumigation were also investigated. The soluble solid content of sweet potato treated with VOCs did not change significantly (Fig. 2b), but on the 20th day, the browning degree of fumigated sweet potato was significantly lower than that of fungus-infected sweet potato (Fig. 2a). The results showed that the malondialdehyde (MDA) content of the WY228-fumigated group was significantly lower than that of C. fimbriata-infected sweet potato by 33.53% and 23.13% on the 10th and 20th days of storage, respectively (Fig. 2c). The results also showed that the starch content in sweet potato decreased during storage. VOC fumigation had no effect on starch content when there was no black spot disease infection. However, the starch content of fungus-infected sweet potato treated with VOCs was higher than that of the nonfumigated group and increased significantly by 3.15%, 6.73%, and 3.87% on the 6th, 10th, and 20th days, respectively (Fig. 2d). As shown in Fig. 2e, there was no difference in phenylalanine ammonia-lyase (PAL) activity between fumigated sweet potato and control during the whole storage period without C. fimbriata infection. In the presence of C. fimbriata, PAL activity of VOCs treatment group increased by 56.07% and 91.76%, respectively, on the 10th and 20th days. VOC fumigation also increased the polyphenol oxidase (PPO) activity of C. fimbriata-infected sweet potato by 66.67% and 84.54%, respectively, after storage for 10 and 20 days (Fig. 2f). The catalase (CAT) activity of sweet potato inoculated with C. fimbriata and fumigated increased by 44.56% on the 20th day (Fig. 2g). VOC fumigation not only increased the antioxidant enzyme activity of sweet potato, but also significantly increased the content of secondary metabolite flavonoids. The content of flavonoids in C. fimbriata-infected sweet potato was significantly increased by 25.58%, 40.13%, and 61.29%, respectively, on the 6th, 10th, and 20th days of storage (Fig. 2h).

FIG 2.

Effects of strain WY228 VOC fumigation on the physiology and quality of sweet potatoes stored for numbers of different days. (a) Degree of browning; (b) soluble solid content; (c) MDA content; (d) starch content; (e) PAL activity; (f) PPO activity; (g) CAT activity; (h) total flavonoid content. Different letters indicate significant differences (P < 0.05) between different treatment groups. The data represent the mean ± standard deviation. CF, black spot pathogen Ceratocystis fimbriata; TR, sweet potato tuberous roots.

VOC composition analysis.

The volatile compounds produced by strain WY228 on TSA agar medium were analyzed by using the headspace solid-phase microextraction and gas chromatography-mass spectrometry (HS-SPME/GC-MS) method. The results of principal-component analysis (PCA) and orthogonal partial least-squares-discriminant analysis (OPLS-DA) showed a clear spatial separation of the volatiles between the blank medium group and strain WY228 group, and there was a good correlation among biological replicates (see Fig. S1a and b in the supplemental material). A total of 163 volatiles were detected from strain WY228 (Fig. S1c), and the VOC profile mainly consisted of alkane, sesquiterpene, alcohol, and pyrazine. Geosmin (30.75%), 2,2,8,8-tetramethyl-3-oxa-2,8-disila-nonan (22.37%), and 2-ethyl-5-methyl pyrazine (5.24%) were found to be the major compounds. β-Eudesmol (2.61%), 1-dodecanol (2.31%), camphor-10-sulfonyl chloride (2.83%), and β-cadinene (2.76%) were detected in high percentages in the volatile profile (Table 1).

TABLE 1.

Main VOCs generated by strain WY228 on TSA medium^a,b

No.	Tentatively identified compounds	R.T. (min)	CAS no.	Class	Mean ± SEM	Relative area (%)
1	β-Eudesmol	31.96	473-15-4	Alcohol	254.85 ± 10.15	2.61
2	1-Dodecanol	28.49	110225-00-8	Alcohol	143.83 ± 15.90	2.31
3	Caryophyllenyl alcohol	30.11	913176-41-7	Alcohol	58.76 ± 6.23	0.94
4	Di-epi-1,10-cubenol	30.36	73365-77-2	Alcohol	50.16 ± 4.96	0.81
5	2,2,8,8-Tetramethyl-3-oxa-2,8-disila-nonan	25.56	7140-91-2	Alkanes	1392.85 ± 496.61	22.37
6	2-Ethyl-5-methyl pyrazine	15.20	13360-64-0	Pyrazine	325.95 ± 22.69	5.24
7	Camphor-10-sulfonyl chloride	15.21	21286-54-46	Chloride	176.10 ± 11.23	2.83
8	Dimethyl disulfide	6.41	24-92-0	Disulfide	96.76 ± 21.59	1.55
9	Geosmin	25.66	16423-19-1	Sesquiterpene	1,914.27 ± 93.21	30.75
10	β-Cadinene	24.09	523-47-7	Sesquiterpene	171.75 ± 15.96	2.76
11	1-Tetradecene	15.67	1120-36-1	Sesquiterpene	104.31 ± 9.19	1.68
12	γ-Muurolene	22.53	30021-74-0	Sesquiterpene	71.77 ± 7.64	1.15
13	1-Heptadecene	20.82	6765-39-5	Sesquiterpene	70.45 ± 4.25	1.13
14	α-Calacorene	27.34	21391-99-1	Sesquiterpene	56.26 ± 4.69	0.90
15	Longifolene	21.91	475-20-7	Sesquiterpene	42.78 ± 2.16	0.69
16	Tetradecane	15.33	629-59-4	Sesquiterpene	40.07 ± 2.92	0.64
17	Nonadecane, 2,6,10,14-tetramethyl-	20.94	55124-80-6	Sesquiterpene	34.50 ± 2.35	0.55
18	(E)-9-Octadecene	15.78	7206-25-9	Sesquiterpene	31.96 ± 3.96	0.51

^aMinor compounds (<0.5%) are not listed in this table.

^bR.T., retention time; CAS, chemical abstracts service.

Antifungal activity of single synthetic volatile.

Seven available commercial synthetic compounds were tested for in vitro antifungal activity. The results showed that two volatiles, 2-ethyl-5-methyl pyrazine and dimethyl disulfide, displayed antifungal activity against C. fimbriata. The inhibitory effect of the two compounds on mycelial growth of C. fimbriata was not obvious at low doses (<10 μL), but when the volume reached 50 μL/plate, the mycelial growth inhibition rate reached 100% (Fig. 3a). From the comparison of sporulation inhibition rates, the antagonistic activity of 2-ethyl-5-methyl pyrazine was stronger. It reached 100% inhibition at only 10 μL/plate concentration, compared with a higher concentration of 25 μL/plate of dimethyl disulfide (Fig. 3b and c). The other five volatiles showed almost no inhibitory effect on the mycelial growth of C. fimbriata (Fig. S2).

FIG 3.

Effects of 2-ethyl-5-methylpyrazine and dimethyl disulfide fumigation on control of sweet potato black spot disease in vitro and in vivo. (a) Growth inhibition of two volatiles at different doses; (b) mycelial and sporulation inhibition rates of 2-ethyl-5-methylpyrazine at different doses; (c) mycelial and sporulation inhibition rates of dimethyl disulfide at different doses; (d) photos of sweet potato surface after fumigation with different doses of 2-ethyl-5-methylpyrazine (-1) and dimethyl disulfide (-2); (e) photos of sweet potato cross sections; (f) the weight loss rate of sweet potato; (g) disease severity (DS). Different letters indicate significant differences (P < 0.05) between different treatment groups. The data represent the mean ± standard deviation. CF, black spot pathogen Ceratocystis fimbriata; TR, sweet potato tuberous roots.

C. fimbriata control by 2-ethyl-5-methylpyrazine and dimethyl disulfide fumigation.

We further evaluated the in vivo control effect of the two compounds against black spot disease in postharvest sweet potato. The results showed that both volatiles inhibited the C. fimbriata infection of sweet potato in a dose-dependent manner. An obvious larger fungal infection diameter can be observed on the surface and cross section of control sweet potato (TR+CF). Fumigation with 100 μL/L and 250 μL/L dimethyl disulfide showed better disease control effect compared with a 50 μL/L concentration, and the fungal infection diameter on the surface of sweet potato decreased obviously (Fig. 3d). The C. fimbriata suppressive effect was more pronounced with 2-ethyl-5-methyl pyrazine than dimethyl disulfide. In the treatment with 100 μL/L and 250 μL/L 2-ethyl-5-methyl pyrazine, the fungal growth was completely inhibited (Fig. 3e). The weight loss rate of sweet potato was not clearly distinguishable after volatile fumigation, but black spot disease severity was significantly reduced (Fig. 3f and g). Fumigation with 2-ethyl-5-methyl pyrazine also influenced the quality of postharvest sweet potato. The MDA content was significantly reduced after fumigation with different concentrations of 2-ethyl-5-methyl pyrazine compared with the control (Fig. 4a). Starch content and superoxide dismutase (SOD) activity were significantly increased by 11.76% and 7.21%, respectively, after fumigation with 250 μL/L compound (Fig. 4b and c). In addition, both volatile treatments also increased the total polyphenol metabolism content of sweet potato (Fig. 4d). Therefore, the two compounds, especially 2-ethyl-5-methyl pyrazine, were found to be excellent fumigant candidates for the control of black spot disease in sweet potato.

FIG 4.

Effects of 2-ethyl-5-methylpyrazine (-1) and dimethyl disulfide (-2) fumigation on the physiology and quality of sweet potato stored for 10 days. (a) MDA content; (b) starch content; (c) SOD activity; (d) total polyphenol content. Different letters indicate significant differences (P < 0.05) between different treatment groups. The data represent the mean ± standard deviation. CF, black spot pathogen Ceratocystis fimbriata; TR, sweet potato tuberous roots.

Overall transcriptional response of C. fimbriata to strain WY228 volatiles.

Next, we explored the possible antifungal mechanism induced by volatiles via transcriptome sequencing. The volatile fumigation for 0 h was designated CK, 6 h was designated CFT1, and 48 h was designated CFT2. By filtering the raw reads, a total of 58.06 Gb clean reads were obtained. The GC content of all samples was more than 50%, the Q30 base percentage was at least 93.51%, and the mapping degree with the reference genome was over 96% (Table S1). Sequencing data analysis and repeatability evaluation showed that transcriptome data are of good quality and are suitable for subsequent analysis (Fig. S3). After applying the classification criteria (fold change of ≥2 [log₂ ratio > |1|] and false-discovery rate [FDR] of <0.05), 2,798 (1,310 up- and 1,488 downregulated) and 3,401 differently expressed genes (DEGs) (1,759 up- and 1,642 downregulated) were identified in groups of CFT1 and CFT2, respectively (Fig. 5a). The Venn diagram in Fig. 5b highlights the unique and shared genes found in each comparison.

FIG 5.

Differentially expressed gene (DEG) analysis of C. fimbriata after WY228 mixture VOC fumigation treatment. (a) Number of upregulated and downregulated DEGs after VOC treatment; (b) Venn diagram of DEGs between the two different treated groups; (c) KEGG pathways significantly enriched (P < 0.05) with DEGs after exposure to VOCs (control versus CFT1, 6 h; control versus CFT2, 48 h). ck, control; CFT1, Ceratocystis fimbriata fumigated with VOCs for 6 h; CFT2, Ceratocystis fimbriata fumigated with VOCs for 48 h.

The identified genes were annotated with 20 Gene Ontology (GO) functional categories, including molecular function, cellular component, and biological process. Catalytic activity and binding ranked as the top two function categories, followed by cellular process, metabolic process, cell part, and membrane part (Fig. S4). The DEGs were also annotated into 21 Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, mainly including metabolism, genetic information processing, environmental information processing, and cellular processes (Fig. S5). In the 6-h treatment group, the dominant pathways were annotated as amino acid metabolism, carbohydrate metabolism, translation, folding, sorting, and degradation, transport and catabolism, but in the 24-h fumigation group, the most enriched pathways were translation, amino acid metabolism, and carbohydrate metabolism. KEGG pathway enrichment analysis indicated that ribosome biogenesis in eukaryotes, and proteasome pathways was significantly enriched in 6 h, and the number of DEGs was the maximum (P  < 0.05), whereas the ribosome pathway was the most enriched after fumigation for 24 h (Fig. 5c).

Functional analysis of differentially expressed genes (DGEs).

Genes related to the cell wall and cell membrane. Genes involved in the cell wall/membrane/envelope biogenesis were affected after fumigation by strain WY228 volatiles. Some genes involved in the chitin synthesis (chs_1, chs_2, and CHS7_0, CHS7_1) were upregulated after VOC treatment. Genes involved in cell wall biogenesis (azaE, OCH1, and GUF1) were downregulated in volatile-treated cells. In addition, volatile treatment also affected the gene expression related to cell membrane synthesis (Fig. 6a). Genes ERG4, erg5, and ARE2, involved in ergosterol biosynthesis, were downregulated 1.99-fold, 4.23-fold, and 3.97-fold, respectively, in C. fimbriata cells fumigated with VOCs for 6 h. After a longer treatment of 48 h, these genes were further downregulated 4.39-fold, 2.87-fold, and 6.34-fold, respectively (Table S2). The genes OLE1, POT1_2, and CFIMG_005736RA, related to cell membrane unsaturated fatty acid biosynthesis were downregulated at both treatment time points (Fig. 6a).

FIG 6.

Heatmaps showing relative expression for selected DEGs at 6 h and 48 h by mixed VOC treatment. The log₂ fold change was colored and standardized. Colors indicate DEGs in VOC-treated cells versus the control. Red, upregulated; black, not differentially expressed; green, downregulated. (a to d) Shown are DEGs involved in (a) cell wall and cell membrane; (b) spore development and toxin biosynthesis; (c) proteasome, and (d) fungal hydrolase.

Genes related to genetic information processing. We found that most of the DEGs related to ribosome biogenesis were downregulated by volatile fumigation. These genes include small ribosomal subunit and large subunit biogenesis, regulator, nucleolar RNA-associated protein, ribosome assembly factor, etc. Some representative genes with more downregulation multiples are shown in Table S2. The gene Mx1, involved in 50S ribosome-binding GTPase synthesis, was downregulated 3.93-fold and 5.68-fold after treatment for 6 h and 48 h, respectively. Expression of the gene rdp1, related to RNA-dependent RNA polymerase 1 synthesis, was downregulated 5.75-fold and 4.81-fold in fumigated cells at 6 h and 48 h, respectively (Fig. S6a). In addition, we detected that some genes related to proteasome subunit synthesis and regulation were upregulated after volatile treatment (Fig. 6c). However, with the extension of treatment time, the number of upregulated genes was decreased, and the downregulated genes were increased. The proteasome synthesized by fungal cells can degrade and remove the damaged proteins in ribosomes, so as to respond to VOC stress in a short time.

Genes related to mitochondrial function. Several genes involved in C. fimbriata mitochondrial function were also downregulated after volatiles treatment (Fig. S6b). Genes encoding mitochondrial cytochrome c₁ heme lyase, ATP-dependent RNA helicase, succinyl-CoA:3-ketoacid coenzyme A transferase, and 3-hydroxyisobutyrate dehydrogenase were downregulated 1.36-fold, 2.07-fold, 3.14-fold, and 2.42-fold, respectively, after treatment for 6 h. These genes were also downregulated 2.71-fold, 2.16-fold, 2.78-fold, and 1.83-fold, respectively, after exposure to volatiles for 48 h. Two genes related to mitochondrial elongation factor G and Tu biosynthesis were downregulated 1.36-fold and 1.32-fold at 6 h, and 2.71-fold and 2.84-fold at 48 h, respectively. S. setonii WY228 volatile treatment also affected the expression of genes involved in the mitochondrial tricarboxylic acid TCA) cycle, including phosphoenolpyruvate carboxykinase, citrate synthase, malate dehydrogenase, succinate-CoA ligase, and 2-methylcitrate synthase (Table S2).

Genes related to spore development and toxin biosynthesis. S. setonii WY228 volatile fumigation not only significantly inhibited fungal spore germination, but also downregulated genes related to spore development (Fig. 6b). Genes encoding conidial pigment polyketide synthase (alb1), sexual development regulator (VELC), C2H2 type master regulator of conidiophore development, and developmental regulatory protein (wetA) were downregulated 7.61-fold, 1.64-fold, 5.95-fold, and 5.03-fold, respectively, at 6 h. A conidiophore development regulator-encoding gene, abaA, was downregulated 1.31-fold after treatment for 48 h. In addition, expression of five genes related to mycotoxin biosynthesis was also downregulated (Table S2). This indicates that VOC fumigation probably inhibits the mycotoxin synthesis in C. fimbriata.

Genes related to C. fimbriata hydrolase. The expression of C. fimbriata hydrolase synthesis genes after volatile treatment is shown in Fig. 6d. In the early stage of fumigation (6 h), the genes encoding endochitinase and endoglucanase were significantly downregulated. For example, five genes encoding endo-1,3(4)-beta-glucanase were downregulated 1.65-fold, 3.90-fold, 4.45-fold, 4.15-fold, and 2.58-fold, respectively, after VOC treatment for 6 h. Two genes encoding endochitinase were downregulated at both time points (6h and 48h) (Table S2). The results indicated that fumigation may reduce the ability of C. fimbriata to infect wounds of sweet potato.

qRT-PCR verification of selected DGEs.

Several DEGs related to proteasome, ribosome, and toxin biosynthesis were randomly selected for quantitative real-time reverse transcription-PCR (qRT-PCR) verification. The results were consistent with the transcriptional profile data generated from transcriptome sequencing (RNA-seq) (Fig. S7).

DISCUSSION

Sweet potato is the fourth most abundant crop globally. In addition, sweet potato also contains lots of functional components and is also a nutritious food (35). As a serious disease of postharvest sweet potato, black spot disease seriously affects the yield and quality of sweet potato. Sweet potato tuberous roots can be chilled or frozen below 10°C. Therefore, they cannot rely on low temperatures alone to resist C. fimbriata infection. At present, the traditional methods of controlling sweet potato black spot disease mainly rely on chemical fungicides, such as fluazinam and carbendazim (25). Considering the impact of chemical fungicides on the environment and human health, an environment-friendly strategy can be used as a sustainable substitute for toxic fungicides to control postharvest fungal diseases of sweet potato. Since the development of Muscodor albus by AgraQuest, Inc. (Davis, CA), as a biofumigant for agricultural uses (36), the research on microbial VOC fumigation and pathogen control has gradually increased (9, 13, 37). Biological fumigation of postharvest fruits and vegetables by microbial VOCs is not only applicable to different storage stages of various products, but is also applicable to products that are easily damaged and cannot be treated with liquid fungicides, such as strawberries and grapes (38, 39). In addition, microbial VOCs are easy to volatilize and degrade at room temperature and do not easily remain on the food surface. Microbial volatiles have become a research hot spot in the field of postharvest storage and pathogen control (2, 4, 13, 30).

The use of VOCs produced by Streptomyces spp. is a method to control postharvest diseases. A recent example is the volatiles emitted from Streptomyces salmonis PSRDC-09 that effectively controlled the anthracnose pathogen Colletotrichum gloeosporioides in postharvest chili (40). The VOCs obtained from Streptomyces philanthi RL-1-178 could be used as a fumigant to protect soybean seeds against aflatoxin-producing fungi (32). However, the knowledge of Streptomyces spp.-produced VOCs for controlling postharvest sweet potato black spot disease caused by C. fimbriata is limited (19). This study provides the latest information about VOCs emitted by a new biocontrol strain, S. setonii WY228, grown on TSA medium to control sweet potato black spot disease in vitro and in vivo. Our study showed that the VOCs released by WY228 exhibited strong antifungal activity against C. fimbriata. During the 20 days of storage, fumigation with WY228 VOCs inhibited fungal growth and effectively controlled the symptom of sweet potato black spot disease. The excellent control effect of the strain may come from direct fungal inhibition of mycelial growth and spore germination, which is similar to the fungicidal effect of VOCs of other Streptomyces strains (41, 42). Studies have shown that PAL is a key enzyme in the synthesis of phenylpropanoid compounds in plants, which is beneficial for biotic and abiotic stress resistance (43). PPO has been frequently implicated in resistance to diverse pathogens (44). Plant flavonoids have also been shown to be related to improving plant disease resistance (45). In addition to direct antifungal activity, VOCs can also trigger the defense response of sweet potato. This can be proved by the observation that VOC treatment increased the content of antioxidant enzymes (PAL and PPO) and total flavonoids in sweet potato (Fig. 2). Similarly, VOC fumigation of Pseudomonas putida BP25 not only controlled postharvest anthracnose, but also increased total flavonoids and PAL activities of mango (46). In addition, WY228 fumigation can maintain the starch content of sweet potato, which is conducive to regulating the quality of sweet potato. Therefore, antifungal VOCs from S. setonii WY228 are promising fumigation agents for the biocontrol of sweet potato black spot disease.

Members of the genus Streptomyces have been reported to release a variety of volatile organic compounds with antifungal activity. The volatiles phenylethyl alcohol, ethyl phenylacetate, methyl anthranilate, α-copaene, caryophyllene, methyl salicylate, and 4-ethylphenol emitted by Streptomyces fimicarius BWL-H1 showed strong inhibitory activity against the litchi downy blight pathogen, Peronophythora litchi (41). Streptomyces albulus strain NJZJSA2 was found to produce the antifungal VOCs 4-methoxystyrene, 2-pentylfuran, and anisole against two fungal pathogens, Sclerotinia sclerotiorum and Fusarium oxysporum (47). It was found that the volatiles produced by WY228 were mainly sesquiterpenes, alkanes, alcohols, and pyrazine (Table 1). Recently, plant-associated terpene VOCs have been detected in many Streptomyces spp. (48). Sesquiterpenes have been reported to have a variety of antifungal and plant resistance functions (49, 50), and geosmin (100 μL/L) has also been reported to inhibit the growth of Aspergillus parasiticus TISTR 3276 and A. flavus 256 PSRDC-4 (32). However, terpene volatiles were not found to inhibit the black spot pathogen, C. fimbriata, in this study. We found that the main volatile, 2-ethyl-5-methylpyrazine, and the trace component dimethyl disulfide could significantly inhibit the growth and sporulation of C. fimbriata in vivo and in vitro in a dose-dependent manner, but they were not solely responsible for the observed antifungal activities. There may be other compounds that were not checked using commercial monomers or detected in the VOC profile, which may play a role.

Pyrazines are used as food additives and flavor enhancers in the food industry and have obtained the “generally recognized as safe” (GRAS) certification of the American Flavor and Extract Manufacturers Association (FEMA). They have good biosafety and have been reported to have broad-spectrum antipathogen and insecticidal properties (51–53). Recently, Archana et al. (46) demonstrated that 2-ethyl-5-methylpyrazine released from Pseudomonas putida BP25 mediated suppression of postharvest anthracnose and quality enhancement in mango. To our knowledge, the discovery of 2-ethyl-5-methylpyrazine from the volatiles of Streptomyces species has not been reported previously. We found that 100 μL/L 2-ethyl-5-methylpyrazine can completely control the occurrence of black spot disease. Therefore, 2-ethyl-5-methylpyrazine is a promising alternative chemical fumigant candidate for controlling postharvest diseases. Dimethyl disulfide is an effective soil fumigant, which exhibits excellent efficacy against fungi and nematodes and induces systemic resistance in plants (54–56). Dimethyl disulfide has been found in the volatiles of some biocontrol microorganisms (57–59). For example, in vitro fumigation with dimethyl disulfide produced by Streptomyces alboflavus TD-1 on Gause’s synthetic medium significantly inhibited the growth of Fusarium moniliforme (60). The endophytic bacterium Burkholderia pyrrocinia JK-SH007 was found to mainly release dimethyl disulfide to promote disease resistance in poplar (61). A storage test showed that fumigation with dimethyl disulfide and 2-ethyl-5-methylpyrazine effectively controlled the occurrence of black spot disease, reduced the MDA content of sweet potato, and improved the activity of SOD and total polyphenol contents, which indicates that both volatiles exhibit the function of inducing sweet potato resistance. The detailed molecular mechanism of VOC mixture- and single compound-induced disease resistance in sweet potato needs to be further explored.

Transcriptome sequencing provides a better understand of the mode of action of S. setonii WY228 volatiles on C. fimbriata. Compared with a single-fungicide compound, the current research found that the antifungal mechanism of microbial VOCs has multiple levels, including cellular development, metabolism, and genetic information processing (62, 63). In this study, transcriptional profile analysis showed that the volatiles mainly affected the gene expression related to amino acid and carbohydrate metabolism and ribosome synthesis of the pathogen C. fimbriata, involving the development of cell wall, cell membrane, spores, ribosome, and mitochondrial functions. Therefore, the strong antifungal effect of WY228 volatiles against C. fimbriata may be due to the different actions at the transcriptional level. KEGG enrichment analysis showed that the ribosomal pathway was significantly enriched at both time points (6h and 48h), and the number of differentially expressed genes was the maximum (Fig. 5). Proteasome function is important for cell survival in response to stress (64). Wu et al. (65) found that microbial volatiles induced cytotoxicity in the yeast Saccharomyces cerevisiae and activated transcriptional upregulation of genes involved in the proteasome pathway. The multiple upregulation of proteasome-related genes at 6 h can be explained by the destruction of ribosomal protein synthesis by volatile treatment. The reason why cells can still survive in this state is that C. fimbriata cells can degrade and remove damaged proteins through their own proteasomal regulatory strategy. Therefore, inhibition of ribosome biogenesis at the transcription, translation, and replication levels and the subsequent protein biosynthesis might explain the antifungal activity and main mechanism of S. setonii WY228 volatiles against C. fimbriata.

The inhibition mechanisms of microbial VOCs against fungi usually include inhibition of mycelial growth, spore production (66, 67), destruction of the cell wall (62) and cell membrane (68), and mitochondrial dysfunction (30). Our study showed that WY228 volatiles downregulated the expression of genes related to ergosterol and unsaturated fatty acid biosynthesis of the C. fimbriata cell membrane. This means that VOCs probably cause lesions on the cell membrane by changing membrane component-related genes and may influence membrane permeability, which is consistent with previous observations of Pseudomonas chlororaphis subsp. aureofaciens SPS-41 volatile treatment (30). Fumigation also changed the expression of key C. fimbriata development-related genes, including spore development genes and sexual development regulators. This result can explain the previously observed significant inhibitory effect of VOCs on C. fimbriata mycelial growth and spore germination. Mitochondria produce ATP through the TCA cycle to provide energy for cells. Mitochondria are also potential targets of antifungal antibiotics, because any mitochondrial dysfunction will lead to cell death (69). It was found that genes involved in mitochondrial genetic information processing, elongation factor, and key enzymes of the TCA cycle were downregulated after WY 228 treatment. This is similar to the damage of mitochondrial function caused by hexanal treatment of Aspergillus flavus spores (70). Our study indicated that VOC fumigation probably destroyed the mitochondrial function of C. fimbriata, affected its TCA cycle and energy metabolism, resulted in cell function damage, and inhibited the growth of the pathogen. In addition, we also found that the genes encoding fungal toxin and hydrolases were also downregulated, such as endochitinase and endoglucanase, which could better explain the decline of the ability of C. fimbriata to infect sweet potato after fumigation. Similarly, VOCs produced by Bacillus subtilis CF-3 and its main component, 2,4-di-tert-butylphenol, can inhibit the protease and cellulase activities of pathogen and effectively inhibit the infection of Colletotrichum gloeosporioides on litchi fruits (71). However, these observed DEG profiles were detected for C. fimbriata growing on potato dextrose broth (PDB) medium and did not fully represent the real gene expression in sweet potato as a substrate. In addition, one limitation of our study was that C. fimbriata control samples not exposed to WY228 volatiles at 6 and 24 h were not included. Therefore, we also need to detect the gene expression of fungus-inoculated sweet potato tuberous roots, as well as the gene expression of pathogen grown on sweet potato treated with mixed VOCs and a single volatile, so as to more truly reflect the potential molecular mechanism of antifungal and induced disease resistance.

Overall, VOC fumigation with WY228 can significantly control the occurrence of black spot disease of postharvest sweet potato and affect the physiology and maintain the quality of sweet potato. The volatiles 2-ethyl-5-methylpyrazine and dimethyl disulfide play strong antifungal roles and the potential function of inducing resistance. Mixed VOC fumigation mainly downregulated the expression of genes related to ribosome synthesis and caused the response of proteasome function genes. It also inhibited the expression of genes encoding cell wall function, cell membrane synthesis, spore development, TCA cycle and mitochondrial function, and toxin and hydrolase synthesis. However, the detailed mechanisms of antifungal and inducing host resistance of individual compounds need to be further clarified.

MATERIALS AND METHODS

Strains and culture conditions.

Endophyte Streptomyces setonii WY228 (stored as a patent strain in the Guangdong Microbial Culture Center [GDMCC 60274], Guangzhou, China), isolated from the roots of Arctium lappa L. was cultured at 28°C on intracellular serine protease (ISP) 2 agar and stored in glycerol at –80°C. Sweet potato black spot pathogen, Ceratocystis fimbriata (voucher specimen no. CF1.01127), was isolated from infected sweet potato and identified by the Xuzhou Sweet Potato Research Institute of Chinese Academy of Agricultural Sciences. The fungus was preserved in our laboratory and cultured at 28°C on PDA medium.

Antifungal activity of Streptomyces setonii WY228 VOCs.

The antifungal activity of S. setonii WY228 VOCs against C. fimbriata was carried out using the bipartite petri plate (I-plate) method (72). Briefly, strain WY228 was first cultured on the TSA medium of one side of the I-plate (9-cm diameter) at 28°C for 2 days. Then, the fungal mycelial discs (6-mm diameter) were inoculated on PDA medium on the other side of the I-plate. The plates without strain WY228 were used as controls. All the I-plates were completely sealed with parafilm and incubated at 28°C. Once the fungi in the control plate were full of the PDA medium, the mycelial inhibition rate was measured and calculated according to the formula described previously (19). Five plate replicates were set for the fungus C. fimbriata.

Effect of VOC fumigation against C. fimbriata in postharvest sweet potato.

First, the PDA plate of black spot fungus grown at 28°C for 5 days was washed repeatedly with 2 mL sterile water to prepare the spore suspension of C. fimbriata. After spores were counted through the blood cell counting plate under the microscope, the final concentration of the C. fimbriata spore suspension was adjusted to 1 × 10⁶/mL. Healthy Tianmu sweet potato tuberous roots (TRs) (10 to 15 cm long and 4 to 7 cm wide), purchased from a local market (Xuzhou, China) were used for pathogen inoculation. The sweet potato TRs were surface-sterilized by soaking in 2% sodium hypochlorite solution for 2 min and then washed twice with sterile water and dried naturally at room temperature for use. About 18 microwounds (1 to 2 mm diameter, 5 mm deep) were made using the sterile needle tip in the middle of two sides of the TRs to form artificial wounds, and then 10 μL C. fimbriata spore suspension were injected into the wounds. Strain WY228, precultured on a TSA plate (35 mm) at 28°C for 2 days, was used for fumigation. The pathogenicity assay was set in sterilized plastic boxes (volume, 0.75 L) with four groups: (i) only one sweet potato in a plastic box, (ii) one sweet potato and four small plates (35 mm) cultured with WY228, (iii) one sweet potato inoculated with C. fimbriata spore suspension, and (iv) one sweet potato inoculated with black spot pathogen and four small plates (35 mm) cultured with WY228. Each group consisted of five replicates. All the boxes were sealed with parafilm and cultured at 28°C. The diameter of the infected wounds and the weight of TRs were measured at 6, 10, and 20 days. The TR disease severity (DS) was calculated according to the formulas described previously (72) based on the following empirical scale: (1) no obvious symptoms; (2) few wounds infected with C. fimbriata; (3) lesion diameter 1 mm to ≤10 mm; (4) lesion diameter, 10 mm to ≤15 mm; and (5) lesion diameter, >15 mm.

Effect of VOC fumigation on the quality and physiology of postharvest sweet potato.

After fumigation with WY228, the browning degree of sweet potato tuberous roots was checked according to the reported method (73). The total soluble solids were measured by using a hand-held digital refractometer according to the manufacturer’s instructions and were expressed as Brix (%). The content of malondialdehyde (MDA) was determined by the thiobarbituric acid method (74). The starch content was measured by the HCl hydrolysis-anthrone colorimetric method (75). For the total polyphenol metabolism content assay, 0.5 g roots sample was put into a precooled mortar, and 2 mL of 1% hydrochloric acid methanol solution was added. After being ground into homogenate, samples were put into dark treatment at 4°C for 30 min and then centrifugation at 10,000 rpm at 4°C for 20 min. The absorbance value at an optical density at 280 nm (OD₂₈₀) of the supernatant was determined, and the total polyphenol content was expressed as OD₂₈₀ g⁻¹ fresh weight. The total flavonoid content was detected using the detection kit method previously described (76). The defense-related enzyme activities were also determined; 0.2 g of the roots was homogenized in a precooled mortar with 2 mL of 0.1 M phosphate buffer (pH 7.4) on ice in a dark environment. After centrifugation at 10,000 rpm at 4°C for 20 min, the supernatant was collected for superoxide dismutase (SOD), catalase (CAT), polyphenol oxidase (PPO), and phenylalanine ammonia-lyase (PAL) enzyme activity detection according to the previously described methods (77–80).

Volatile organic compound composition analysis.

The VOCs emitted by strain WY228 grown on TSA medium in 20-mL sterile glass headspace vials at 28°C for 5 days were analyzed by using the headspace solid-phase microextraction and gas chromatography-mass spectrometry (HS-SPME/GC-MS) method by Shanghai Lu-Ming Biotech Co. (Shanghai, China). The VOC extraction and analysis were conducted on the Agilent 7890B GC system coupled to an Agilent 5977A MSD system (Agilent Technologies, Inc., California, USA). A DB-5MS fused-silica capillary column (30 m by 0.25 mm by 0.25 μm; Agilent J&W Scientific, Folsom, CA, USA) was utilized to separate the derivatives. The detailed HS-SPME/GC-MS analysis and quantification using the internal standard were performed as previously described (72). Data were transformed by log₁₀, and the resulting data matrix was then imported into the R ropls package. Principal-component analysis (PCA) and orthogonal partial least-squares-discriminant analysis (OPLS-DA) were performed to visualize the metabolic difference among the control and experimental groups, after mean centering and unit variance scaling. The differential metabolites were selected based on the statistically significant threshold of variable influence on projection (VIP) values of >1 obtained from the OPLS-DA model, P values of <0.05 from a two-tailed Student’s t test, and a fold change (FC) of >2 compared with the control of three independent biological repeats. The NIST17 (National Institute of Standards and Technology, USA) database was used for the putative identification of volatiles with a similarity index higher than 800.

Effect of synthetic volatiles on C. fimbriata mycelial growth and sporulation.

The mycelial growth and spore germination assay of C. fimbriata were detected using the I-plate. Fungal mycelial discs (6-mm diameter) were inoculated on PDA medium of one compartment of an I-plate. Seven available synthetic commercial volatiles, geosmin (98%; Smart Solutions, USA), 2-ethyl-5-methylpyrazine (98%; Adamas-beta, China), β-eudesmol, 1-dodecanol, 1-tetradecene, 1-heptadecene, and dimethyl disulfide (>98%; Aladdin, China), with different volumes (1 μL, 5 μL, 10 μL, 25 μL and 50 μL) were dropped on a filter paper (diameter, 0.8 cm) in the other compartment of the I-plate. After incubation at 28°C for 5 days, the mycelial growth was measured, and the inhibition rate was calculated; the spores were also collected, and spore production was calculated (72). Five plate replicates were set for each treatment.

Effect of 2-ethyl-5-methylpyrazine and dimethyl disulfide fumigation against C. fimbriata in postharvest sweet potato.

Two volatiles, 2-ethyl-5-methylpyrazine and dimethyl disulfide, which showed strong antifungal activity against C. fimbriata in vitro, were used for fumigation of postharvest sweet potato. The fumigation experiment was carried out in sterilized plastic boxes (volume, 0.75 L). In each box, there was a surface-sterilized sweet potato tuberous root and a glass petri dish (diameter, 60 mm) with different volumes of pure monomer volatiles (50, 100, and 250 μL/L). The infection of sweet potato by C. fimbriata was described as in the above-mentioned method. The plastic boxes were sealed and cultured at 28°C. After 10 days of fumigation, the sweet potatoes were taken out, the infection of TRs was counted, and the physiological and quality characteristics of the sweet potatoes were also detected.

Transcriptome analysis of the C. fimbriata response to Streptomyces setonii WY228 volatiles.

Strain WY228 was first cultured on TSA medium of one compartment of an I-plate at 28°C for 2 days. Then, the C. fimbriata spore suspension (1 × 10⁶/mL) was added to the PDB medium of the other compartment, and the I-plates were sealed and cultured for at 28°C for 6 and 24 h. All treatments were repeated three times. After fumigation, the spore suspension of C. fimbriata was collected by centrifugation at 8,000 rpm for 20 min at 4°C and then rinsed twice with sterile water. The total RNA extraction, library construction, and sequencing using and Illumina HiSeq 2500 platform were entrusted to Shanghai Majorbio Co., Ltd. After removing adaptor sequences and low-quality reads from the raw data, high-quality clean reads were obtained and assembled. The clean reads were mapped to the C. fimbriata genome (https://www.ncbi.nlm.nih.gov/genome/?term=Ceratocystis+fimbriata) using Tophat2 (81). The fragments per kilobase per million (FPKM) values were used to estimate the expression patterns of genes. The differently expressed genes (DEGs) were analyzed by DESeq (82), and the fold change of ≥2 (log₂ ratio > |1|) and FDR of <0.05 were used as the criteria for DEG screening. The enrichment analysis of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) were performed using the free online Majorbio Cloud Platform (www.majorbio.com).

qRT-PCR analysis.

The same RNA samples extracted as above were used for quantitative real-time reverse transcription-PCR (qRT-PCR), which was performed using a 2× Universal SYBR green fast qPCR mixture (ABclonal, Wuhan, China). The qRT-PCR (total 10 μL volume) contained the following components: 1 μL cDNA, 0.2 μL forward primer, 0.2 μL reverse primer, 5 μL master mix, and 3.6 μL double-distilled water (ddH₂O). Gene expression was normalized to the reference gene actin relate protein 3 (arp3) and calculated using the 2^–ΔΔCT method (83). All qRT-PCRs were repeated three times, and the primers used are listed in Table 2.

TABLE 2.

RT-qPCR primer sequences used in this study

Gene	Gene annotation	Forward (F) and reverse (R) primers
arp3	Actin-related protein 3	F 5′-CGATGTTGGTTACGAGCGTTTCC-3′ R 5′-ACTACGACGGGTAGAGGGGTCAGGA-3′
SCL1	Proteasome subunit alpha type-1	F 5′-AGAGACTGGCCAACATCAGC-3′ R 5′-GAGGTCCAAACTCCGAGTCC-3′
PRE3	Proteasome subunit beta type-1	F 5′-TGGTGCTGCCGCTCTGGCTC-3′ R 5′-GAGCCAGAGCGGCAGCACCA-3′
rhp23	UV excision repair protein rhp23	F 5′-AAAGATGATGACAAGGTCGA-3′ R 5′-TCGACCTTGTCATCATCTTT-3′
NAT10	RNA cytidine acetyltransferase	F 5′-TTGACCCAGAACAAGTCGGT-3′ R 5′- ACCGACTTGTTCTGGGTCAA-3′
RAC1	Ras-related C3 botulinum toxin substrate 1	F 5′-CGCCTGCGTCCGCTGTCCTA-3′ R 5′-TAGGACAGCGGACGCAGGCG-3′
ABCC11	ATP-binding cassette subfamily C member 11	F 5′-CAGTTTGCGTCAGATGCTTG-3′ R 5′-CAAGCATCTGACGCAAACTG-3′

Statistical analysis.

Mean values and standard deviations were calculated using Excel 2010. Data were analyzed by analysis of variance (ANOVA) and Student’s t-tests, with a P value of <0.05 as generally significant. The results are expressed as the mean ± standard deviation. GraphPad Prism 8.0 was used for data mapping.

Data availability.

The transcriptome data have been deposited in the NCBI under accession number PRJNA768780.