Antagonistic Activity of Volatile Organic Compounds Produced by Acid-Tolerant Pseudomonas protegens CLP-6 as Biological Fumigants To Control Tobacco Bacterial Wilt Caused by Ralstonia solanacearum

ABSTRACT

Tobacco bacterial wilt, which is caused by Ralstonia solanacearum, is a devastating soilborne disease of tobacco worldwide and is widespread in the continuously acidic fields of southern China. Here, the fumigation activity under different pH conditions, component identification, and bioactivity of the volatile organic compounds (VOCs) produced by an acid-tolerant strain, Pseudomonas protegens CLP-6, were investigated. There was a wide antimicrobial spectrum of the VOCs against phytopathogens, including four bacteria, eight fungi, and two oomycetes. The antagonistic activity of the VOCs against R. solanacearum was proportionally correlated with the concentration of the inoculum, amount, culture time, and culture pH for CLP-6. The number of gene copies of R. solanacearum was significantly inhibited by VOCs produced at pH 5.5 in vivo. The control effect of VOCs emitted at pH 5.5 was 78.91% for tobacco bacterial wilt, which was >3-fold greater than that at pH 7.0. Finally, the main volatile compounds were identified by solid-phase microextraction (SPME)–gas chromatography-mass spectroscopy (GC-MS) as S-methyl thioacetate, methyl thiocyanate, methyl disulfide, 1-decene, 2-ethylhexanol, 1,4-undecadiene, 1-undecene, 1,3-benzothiazole, and 2,5-dimethylpyrazine, and the inhibition rates of 1,3-benzothiazole, 2-ethylhexanolmethyl thiocyanate, dimethyl disulfide, and S-methyl thioacetate were 100%, 100%, 88.91%, 67.64%, and 53.29%, respectively. S-Methyl thioacetate was detected only at pH 5.5. In summary, VOCs produced by P. protegens CLP-6 had strong antagonistic activities against phytopathogens, especially R. solanacearum, under acidic conditions and could be used to develop a safe and additive fumigant against R. solanacearum on tobacco and even other Solanaceae crop bacterial wilt diseases in acidic fields.

IMPORTANCE VOCs produced by beneficial bacteria penetrate the rhizosphere to inhibit the growth of plant-pathogenic microorganisms; thus, they have the potential to be used as biological agents in controlling plant diseases. Tobacco bacterial wilt, which is caused by the acidophilic pathogen R. solanacearum, is a major bacterial disease in southern China and is prevalent in acidic soil. In this study, we discovered that the VOCs produced by P. protegens CLP-6 had excellent inhibitory effects on important plant pathogens. Moreover, two of the VOCs, namely, 1,3-benzothiazole and 2-ethylhexanol, had excellent inhibitory effect on R. solanacearum, and another VOC substance, methyl thiocyanate, was produced only at pH 5.5. The VOCs produced by the acid-tolerant strain P. protegens CLP-6 may have potential as environment-friendly microbial fumigant agents for bacterial wilt of tobacco or even other Solanaceae crops in acidic soils in China.

INTRODUCTION

Volatile organic compounds (VOCs) are commonly released by all organisms, including plants, fungi, and bacteria. They quickly evaporate and disperse under normal temperature and pressure conditions and can penetrate the bulk soil and rhizosphere soil due to their low molecular weights (<300 Da) (1). The VOCs produced by several beneficial bacteria during interactions between plant-pathogenic microorganisms and host plants are novel compounds with antibiotic activity (2–5). The growth of plant-pathogenic microorganisms is inhibited by VOCs produced by soil and plant-related microorganisms, suggesting that they may represent an important biological mechanism of controlling plant diseases (6, 7). The phenomenon of some pathogens being directly or indirectly influenced by low-molecular-weight compounds produced by bacteria was discovered decades ago (8). These compounds were categorized into clusters of ketones, esters, alkanes, terpenoids, alkynes, alcohols, aldehydes, and compounds containing sulfur (9). Among these VOCs, S-methyl thioacetate, methyl thiocyanate, methyl disulfide, 2-ethylhexanol, 1-undecene, 1,3-benzothiazole, and 2,5-dimethylpyrazine were previously reported to possess biocontrol activity (1, 10–14). Tahir et al. (1) indicated that the robust growth of Ralstonia solanacearum in vitro was inhibited by the VOCs emitted by Bacillus artrophaeus LSSC22 and Bacillus velezensis FZB42 and, using these VOCs, tobacco plants were protected from bacterial wilt disease in a potted plant assay. Agrobacterium tumefaciens, a causative agent of crown gall disease in a wide range of plants, was inhibited by VOCs produced by Pseudomonas fluorescens B-4117 and Pseudomonas chlororaphis 449, which suppressed the formation of biofilms and reduced the number of living cells on mature biofilms (13). Pseudomonas protegens is a biocontrol bacterium that has been reported to produce lipopeptide compounds, which constitute valuable antibiotics in agriculture (15); Farag et al. (16) showed that effective antifungal VOCs released by P. protegens AS15 significantly inhibited the mycelial growth, conidial germination, and sporulation of Aspergillus flavus in culture (16).

R. solanacearum, the causal agent of tobacco bacterial wilt, is one of the most devastating soilborne plant pathogens (2). Tobacco bacterial wilt is distributed mainly in the tropics and subtropics and causes serious economic losses, greatly reducing the yield and quality of tobacco crops (4). R. solanacearum migrates toward the organic acids secreted from tobacco root exudates both in vitro and in vivo, and the colonization of the tobacco rhizosphere by R. solanacearum has been shown to contribute to the severe incidence of tobacco bacterial wilt in acidic soil (17). Tobacco bacterial wilt progressively occurs in acidic soil due to the accumulation of R. solanacearum and the continued decreases in the soil pH value. There currently is no safe and effective method of controlling bacterial wilt, especially in acidic soil (4, 18).

For over 70 years, chemical fungicides have been used in the treatment and control of plant pathogens, and chemical soil pesticides are one of the leading control methods for wilts (11). However, chemical soil pesticides are considered to have strongly biocidal efficacy, and adverse effects have been reported not only on the target pathogens but also nontarget microorganisms. These pesticides are becoming less acceptable due to their potential deleterious effects on human health (5) and the environment. Biofumigation, a biological control method involving VOCs (12, 13), is a type of biocontrol in which beneficial organisms are actively employed to protect the plant from harmful organisms, and it has been sought as an alternative method of protecting crops. Therefore, the use of VOCs to control tobacco wilt disease in acidic soil (19) constitutes a potential development in this direction.

To improve the control of tobacco bacterial wilt in acidic soil, we evaluated the antagonistic activity of VOCs produced by the acid-tolerant strain P. protegens CLP-6 (ZL.201710081711.X, China General Microbiological Culture Collection [CGMCC] 13205), which was previously isolated from the tobacco rhizosphere in acidic soil. In this study, we aimed (i) to clarify the effect on R. solanacearum of VOCs emitted by P. protegens CLP-6 in vitro and in vivo, (ii) to test the inhibitory spectrum of VOCs by means of mycelial growth inhibition and bacterial diffusion methods, and (iii) to identify the components of VOCs produced in culture at pH 5.5 and pH 7.0 and to analyze their different antagonistic activities.

RESULTS

VOCs produced by P. protegens CLP-6 exhibit broad-spectrum antagonistic activity.

The inhibition assays performed in sealed petri plates indicated the relationship between antagonistic activity and the VOCs released by CLP-6. This antagonistic activity might have been due to the volatiles emitted from CLP-6, as there was no physical contact between R. solanacearum and CLP-6 (Fig. 1). The colony growth of R. solanacearum was normal without exposure to any VOCs at 5 days postinoculation (dpi) (Fig. 1a). In contrast, on nonactivated carbon, the growth of R. solanacearum was entirely inhibited in the presence of CLP-6 (Fig. 1b); however, colony growth was surprisingly restored to some extent in the presence of activated charcoal at 5 dpi (Fig. 1c). Therefore, R. solanacearum recovery as a consequence of the adsorption of the VOCs by activated carbon proved that VOCs produced by P. protegens CLP-6 had strongly antagonistic activity against R. solanacearum.

FIG 1.

Activity of VOCs adsorbed by activated carbon in two sealed base plates. The top plate contained CLP-6/blank, while the bottom plate contained R. solanacearum and activated carbon. (a) R. solanacearum and activated carbon. (b) CLP-6 and R. solanacearum. (c) CLP-6, R. solanacearum, and activated carbon.

The antagonistic activity of VOCs was tested in the presence of eight pathogenic fungi and two oomycetes (Fig. 2 and Table 1). The VOCs inhibited almost all mycelial growth of these pathogenic fungi and oomycetes, leading to little or no color pigmentation in colonies, compared with that seen with the control check (CK) treatment. The colony diameter was especially affected. This result indicated that the normal growth of pathogenic fungi and oomycetes was obviously inhibited. VOCs showed excellent inhibition activities for Alternaria alternata and Colletotrichum chlorophyti, with inhibition rates of 80.8% and 71.3%, respectively, while the inhibitory effects on Colletotrichum gloeosporioides, Colletotrichum destructivum, and Phytophthora capsica averaged 69.2%, 68.2%, and 61.9%, respectively. The inhibition of Fusarium graminearum, Fusarium oxysporum f. sp. cucumerinum, Fusarium oxysporum, Phytophthora infestans, and Fusarium solani was relatively low, with inhibition rates of 46.4%, 36.3%, 33.4%, 18.8%, and 17.2%, respectively.

FIG 2.

Antagonistic activity of VOCs against pathogenic fungi evaluated in two sealed base plates. One plate contained CLP-6/blank, while the other contained a culture of pathogenic fungi. (a to j) The species of pathogenic fungi used were Fusarium graminearum (a), F. oxysporum f. sp. cucumerinum (b), F. oxysporum (c), F. solani (d), Alternaria alternata (e), Colletotrichum destructivum (f), C. gloeosporioides (g), and C. chlorophyti (h), with two oomycetes, i.e., Phytophthora capsica (i) and P. infestans (j). (k and l) The experimental method is shown.

TABLE 1.

Antagonistic activity of VOCs against tested pathogenic fungi

Pathogenic fungus	Average inhibition rate (%)
F. oxysporum	33.4
F. graminearum	46.4
C. chlorophyti	71.3
P. infestans	18.8
F. oxysporum f. sp. cucumerinum	36.3
C. gloeosporioides	69.2
F. solani	17.2
P. capsica	61.9
C. destructivum	68.2
A. alternata	80.8

VOCs also showed antagonistic activities against four tobacco-pathogenic bacteria (Fig. 3 and Table 2). The bacterial colonies biofumigated with P. protegens CLP-6 grew slowly and were almost transparent, especially R. solanacearum, which was more strongly inhibited than with the CK treatment (Fig. 3a). The rates of inhibition of R. solanacearum, Pseudomonas syringae pv. tabaci tox⁺, P. syringae pv. tabaci tox⁻, and Dickeya chrysanthemi induced by VOCs were 97.1% (Fig. 3a), 92.6% (Fig. 3b), 92.9% (Fig. 3c), and 80.9% (Fig. 3d), respectively. These results demonstrated that VOCs had antagonistic activities against four main bacterial pathogens of tobacco, and the inhibitory effect on R. solanacearum was the strongest.

FIG 3.

Antagonistic activity of VOCs against pathogenic bacteria evaluated using two inverse face-to-face petri dishes. The top plate contained CLP-6/blank, and the bottom plate contained a culture of pathogenic bacteria. The species of pathogenic bacteria used were Ralstonia solanacearum (a), Dickeya chrysanthemi (b), Pseudomonas syringae pv. tabaci tox⁺ (c), and P. syringae pv. tabaci tox⁻ (d).

TABLE 2.

Antagonistic activity of VOCs against pathogenic bacteria

Pathogenic bacterium	Average inhibition rate (%)
R. solanacearum	97.1
P. syringae pv. tabaci tox+	92.6
P. syringae pv. tabaci tox−	92.9
D. chrysanthemi	80.9

Optimum inoculation conditions for producing VOCs against R. solanacearum.

Among the four different concentrations used, the antagonistic activity of VOCs was stronger when the concentration of CLP-6 was increased. VOCs showed the strongest antagonistic activity, exhibited as a significantly smaller colony diameter than that with CK, when the inoculated concentration of CLP-6 suspension was about 1 × 10⁸ CFU/mL, whereas the activity of VOCs was the weakest when the inoculated concentration was about 1 × 10⁶ CFU/mL (Fig. 4); therefore, there was a positive correlation between the inoculation concentration and the antagonistic activity, and there were significant differences among all treatments.

FIG 4.

Effect of the concentration of the inoculum on the antagonistic activity of VOCs produced by CLP-6 against R. solanacearum (concentration of 10⁸ CFU/mL). One drop equals 10 μL. Means with different letters for each inoculation concentration (10⁵ CFU/mL, 10⁶ CFU/mL, 10⁷ CFU/mL, and 10⁸ CFU/mL) denote significant differences (P < 0.05).

With respect to the fumigation time and inoculation amount, as shown in Fig. 5, the rate of inhibition induced by the VOCs was improved with increased fumigation time. Moreover, the antagonistic activities of VOCs against R. solanacearum were affected by different inoculation amounts. Among the four inoculation amounts, inhibition was less than 20% at 1 dpi. However, the inhibition rates increased and were positively correlated with the inoculation amount after 2 dpi. The inhibition rates according to the inoculation amount of CLP-6 were ranked as follows: 160 μL > 80 μL > 40 μL > 10 μL > CK. The strongest antagonistic activity occurred with 4 days of fumigation, and the inhibition rates for the four inoculation amounts at 4 dpi were 11.4% (10 μL), 54.31% (40 μL), 59.79% (80 μL), and 60.83% (160 μL). The antagonistic activities of 80 μL and 160 μL of CLP-6 were almost the same. Taken together, the antagonistic activity of VOCs reached a maximum at 4 dpi, and their antagonistic activity was positively correlated with the inoculation amount. In summary, the antagonistic activity of VOCs was highest at 4 dpi with 80 or 160 μL of CLP-6.

FIG 5.

Relationships among different volumes of CLP-6 inocula, fumigation time, and antagonistic activity. The top plate contained CLP-6, while R. solanacearum was cultured on the bottom plate. One drop equals 10 μL. (A) Qualitative analysis. (a-d) treated with 10 μL, 40 μL, 80 μL, 160 μL CLP-6 suspensions; (e) Untreated bacteria without CLP-6. (B) Quantitative analysis. Means with different letters of fumigation time (24 h, 48 h, 72 h, 96 h, and 120 h) denote significant differences (P < 0.05).

VOCs can effectively reduce the number of R. solanacearum colonies under pH 5.5 or pH 7.0 conditions.

The population increase for R. solanacearum could be limited by VOCs from CLP-6 cultured at pH 5.5 or pH 7.0. The initial gene copy numbers (GCNs) per gram dry soil for R. solanacearum in soil at pH 5.5 or pH 7.0 was 2.8 × 10⁵. The GCNs of R. solanacearum were 5.47 × 10⁵ without exposure to VOCs and 1.32 × 10⁴ with exposure to VOCs from CLP-6 cultured in medium at pH 5.5, representing a 1/42 decrease, compared with the blank control. The GCNs were 4.58 × 10⁵ without exposure to VOCs and 3.44 × 10⁴ with exposure to VOCs produced by CLP-6 cultured in medium at pH 7.0. The GCN of R. solanacearum in soil with the CK treatment was 13.47-fold greater that detected with the CLP-6 treatment.

The inhibitory effect of VOCs produced at pH 5.5 on R. solanacearum was greater than that of VOCs produced at pH 7.0.

As shown in Fig. 6, when the inoculation time was lengthened, the disease incidence and severity changed to different degrees with all treatments. After 14 dpi and fumigation with VOCs produced at pH 5.5 and pH 7.0, the disease incidence and severity of tobacco bacterial wilt were all decreased, in comparison with the CK; the disease incidence and severity index were 33.33% and 6.88 (pH 5.5) and 38.10% and 8.47 (pH 7.0), respectively, and those of the CK were 61.90% and 33.33 (pH 5.5) and 61.90% and 34.39 (pH 7.0), respectively (Fig. 6A and B). There were significant differences between the VOC treatment and the CK. The disease control effect on tobacco bacterial wilt caused by R. solanacearum of VOCs produced at pH 5.5 or pH 7.0 was 78.91% or 75.39%, respectively (Fig. 6C and D).

FIG 6.

Volatile compounds produced by P. protegens CLP-6 reduced the virulence of R. solanacearum in tobacco. Disease incidence (A), disease index (B), disease control effect (C), and the degree to which tobacco plants grew (D) in soil mixed with R. solanacearum exposed to VOCs produced by P. protegens CLP-6 cultured at pH 5.5 or pH 7.0 were measured. Panel D shows four treatments, i.e., R. solanacearum exposed to VOCs produced by CLP-6 cultured at pH 5.5 (a), R. solanacearum exposed to blank pH 5.5 NA medium (b), R. solanacearum exposed to VOCs produced by CLP-6 cultured at pH 7.0 (c), and R. solanacearum exposed to blank pH 7.0 NA medium (d).

Composition analysis and antagonistic activity of VOCs produced under culture conditions at pH 5.5 or pH 7.0.

The solid-phase microextraction (SPME)–gas chromatography-mass spectrometry (GC-MS) analysis indicated that CLP-6 grown at pH 5.5 or pH 7.0 on nutrient agar (NA) medium produced a volatile profile including three olefin compounds (1,4-undecadiene, 1-undecene, and 1-decene), one ester compound (methyl thiocyanate), and four other compounds (methyl disulfide, 2-ethylhexanol, 1,3-benzothiazole, and 2,5-dimethylpyrazine) (Fig. 7 and Table 3, according to the peak retention time sequence). S-Methyl thioacetate was produced only at pH 5.5. The peak area results showed that components of the VOCs differed at pH 5.5 and pH 7.0. 1-Undecene had the largest peak area, i.e., 97.59%, while methyl disulfide, with an average area of 82.42%, ranked second and methyl thiocyanate, with an average area of 37.80%, ranked third. The peak area percentages of 1-undecene at pH 5.5 and pH 7.0 were 99.71% and 97.59%, respectively. Moreover, the amounts of S-methyl thioacetate, 1-decene, and 2-ethylhexanol produced at pH 5.5 were larger than those produced at pH 7.0. These results revealed that the acidic conditions were more suitable for VOC production by CLP-6.

FIG 7.

GC-MS spectra of volatile compounds produced by strain CLP-6 grown on NA plates at pH 5.5 (a) and pH 7.0 (b) for 72 h.

TABLE 3.

Detection of VOCs by GC-MS

Name	Retention time (min)	CAS no.	Molecular formula	Molecular weight	Area (%)		Retention index
					pH 5.5	pH 7.0
S-Methyl thioacetate	2.17	1534-08-3	C3H6OS	90.14	3.59	0
Methyl thiocyanate	2.25	556-64-9	C2H3NS	73.12	14.999	60.60
Methyl disulfide	2.63	624-92-0	C2H6S2	94.2	82.419	87.93	720
1-Decene	7.91	872-05-9	C10H2O	140.26	0.37	0.14	990
2-Ethylhexanol	8.94	104-76-7	C8H18O	130.23	1.15	0.27	1,027
1,4-Undecadiene	10.63	55976-13-1	C11H2O	152	1.92	1.37	1,086
1-Undecene	10.87	821-95-4	C11H22	154.29	99.71	97.59	1,094
1,3-Benzothiazole	14.52	95-16-9	C7H5NS	135.19	0.90	1.41	1,221
2,5-Dimethylpyrazine	17.00	123-32-0	C6H8N2	108.14	1.53	5.42	1,311

As shown in Fig. 8, the pure standard substances of VOCs were tested for antibacterial activity against R. solanacearum. The results showed that 1,3-benzothiazole and 2-ethylhexanol were the two most effective substances, completely preventing colony growth with inhibition rates of 100%; methyl thiocyanate and dimethyl disulfide also showed strong antibacterial activity, with growth inhibition rates of 88.91% and 67.64%, respectively, while the inhibition rates of S-methyl thioacetate, 2,5-dimethylpyrazine, 1-decene, and 1,4-undecadiene were 53.29%, 43.98%, 30.70%, and 15.12%, respectively. The antagonistic effect of 1-undecene was the weakest at <10%.

FIG 8.

Antagonistic activity of component standards of VOCs against R. solanacearum (concentration of 10⁸ CFU/mL). One drop equals 10 μL. (a) S-Methyl thioacetate. (b) Methyl thiocyanate. (c) Methyl disulfide. (d) 1-Decene. (e) 2-Ethyhexanol. (f) (4E)-Undeca-1,4-diene. (g) 1-Undecene. (h) 1,3-Benzothiazole. (i) 2,5-Dimethylpyrazine. (j) CK.

DISCUSSION

To alleviate the occurrence of tobacco bacterial wilt disease in acidic soil, we reported the effects of VOCs produced by an acid-tolerant strain, P. protegens CLP-6, against R. solanacearum and its antimicrobial spectrum. The activated carbon adsorption test verified that R. solanacearum was inhibited by VOCs, and the antagonistic activity of P. protegens CLP-6 against the soilborne pathogen was weakened after treatment with activated carbon to absorb the VOCs. Therefore, this result verified that the VOCs released by P. protegens CLP-6 played a powerful role in inhibiting R. solanacearum and thus show potential as efficient microbial fumigants. The results of activated charcoal absorption showed that the activated charcoal could absorb some VOCs released by CLP-6. A previous report indicated that the Shewanella algae strain YM8 produces volatiles with strong inhibition activity against Aspergillus pathogens and aflatoxins (20). In addition, this paper presented a case investigation of the relationship between VOCs and pathogenic fungi and bacteria. Therefore, we will further screen suitable adsorbents for CLP-6. Moreover, the VOCs exhibited a wide spectrum of inhibition against most target phytopathogens, including 10 pathogenic fungi, two oomycetes, and four pathogenic bacteria. Similarly, Ossowicki et al. (12) indicated that the VOCs produced by the biocontrol bacterium Pseudomonas donghuensis P482 had a broad spectrum of inhibition activity. Some of the VOCs produced by Pseudomonas strains might act as inhibitors of the quorum-sensing cell-to-cell communication network, which regulates the production of antibiotics, pigments, exoenzymes, and toxins (1, 12). Together, these results open up a new avenue of research into disease prevention during crop and vegetable production.

Pseudomonas spp. are important in the biocontrol of a number of soilborne diseases. For instance, the mycelial growth of Metschnikowia laxa, the causal agent of brown rot disease in European plum, is significantly reduced by the VOCs produced by their most potent antagonist, Pseudomonas synxantha P4/16_1 (21). Moreover, P. fluorescens strains have been reported to produce antimicrobial VOCs. P. fluorescens strain WR-1 produces VOCs that restrict colony growth of R. solanacearum and have a concentration-dependent antagonistic effect on this pathogen, restricting its growth and virulence (22). Mannaa et al. (23) argued that aflatoxins produced by Aspergillus flavus on rice grains are significantly reduced by the VOCs produced by P. protegens AS15. However, the inhibitory activity of VOCs produced by P. protegens has not been reported for R. solanacearum.

The antagonistic activity of VOCs is usually closely related to the initial concentration and fumigation time. Zhang et al. (24) studied the biological activity of P. chlororaphis SPS-41 against Ceratocystis fimbriata, which causes black root rot of sweet potato, and found that greater inoculation volume and initial inoculation concentration yielded stronger antagonistic activity of the VOCs. The findings from our study concurred with these results. The antagonistic activity of the VOCs was positively correlated with the initial inoculation concentration and amount of CLP-6. Increasing the concentration of the inoculant and the amount of P. protegens CLP-6 was found to be most effective in inhibiting R. solanacearum.

The strongly antagonistic activity of acid-tolerant biocontrol agents has the potential to improve the control of plant diseases in acidic soil. In particular, R. solanacearum has an extreme preference for acidic soil, resulting in its great abundance in these areas and, hence, the serious occurrence of tobacco bacterial wilt. Therefore, we determined the inhibitory effect of VOCs produced at pH 5.5. Data from fluorescence quantitative PCR and pot experiments showed that the inhibitory effect of VOCs produced at pH 5.5 was significantly stronger than that of VOCs produced at pH 7.0. This result not only revealed that the inhibitory effect of CLP-6 was stronger under acidic conditions but also indicated that VOCs are relatively more vigorous under acidic conditions. At 14 dpi, 78.9% of tobacco wilt disease was inhibited by VOCs produced at pH 5.5, whereas only 75.39% of the disease was inhibited by VOCs produced at pH 7.0. Therefore, R. solanacearum previously suppressed by the VOC fumigation of soil might have recovered and infected the tobacco plants. However, it remained clear that VOCs produced at pH 5.5 were more strongly inhibitory than those produced at pH 7.0. Altogether, the VOCs produced at pH 5.5 obviously inhibited the GCN of R. solanacearum, compared with VOCs produced at pH 7.0. This result indicated that acidic conditions are beneficial for exerting the antagonistic activity of the acid-tolerant strain P. protegens CLP-6. The exposure of soil to VOCs produced by CLP-6 may lead to improved tobacco growth as a consequence of enhanced nutrition (25). More work is required to further study this mechanism for promoting plant growth.

SPME–GC-MS is a simple and rapid technique for detecting volatile compounds that has been successfully used to identify the profile of VOCs produced by bacteria, fungi, and yeasts (26–28). Based on SPME–GC-MS technology, the VOC components produced at pH 5.5 and pH 7.0 were detected. Nine components were collected and identified, namely, S-methyl thioacetate, methyl thiocyanate, methyl disulfide, 1-decene, 2-ethylhexanol, 1,4-undecadiene, 1-undecene, 1,3-benzothiazole, and 2,5-dimethylpyrazine, and this is the first report of 1,4-undecadiene and 1-decene produced by Pseudomonas spp. 2-Ethylhexanol, 1-decene, 1,4-undecadiene, and 1-undecene were produced at higher rates at pH 5.5 than at pH 7.0. Interestingly, S-methyl thioacetate was produced only under acidic conditions. This result also indicated that the volatile compounds produced by CLP-6 were variable, depending on the pH of the growing conditions, with pH 5.5 being the most effective. It was previously reported that the type and relative abundance of VOCs produced are affected by many factors, such as the type of substrate, culture conditions, and the physiological state of the microorganisms (29). S-Methyl thioacetate production, through a reaction with the precursor acetic acid in NA medium, might have contributed to the production of S-methyl thiocyanate at pH 5.5, as suggested by their chemical structures. This may be one of the reasons for the stronger antagonistic activity of P. protegens CLP-6 grown at pH 5.5, compared with pH 7.0. Determination of the antagonistic activity of pure standard substances contained in VOCs showed that 1,3-benzothiazole and 2-ethylhexanol were 100% inhibitory against R. solanacearum, and it will be beneficial to further investigate the broad-spectrum inhibition of other Solanaceae crop wilt diseases. Interestingly, the production of 2-ethylhexanol at pH 5.5 increased by >300%, compared with that at pH 7.0. This might have contributed to the stronger inhibitory activity of VOCs produced at the lower pH value. The antifungal activity of 2-ethylhexanol has been verified by Wu et al. (11). Those authors found that 2-ethylhexanol produced by Bacillus amyloliquefaciens could inhibit the growth of F. oxysporum f. sp. niveum, thereby preventing the occurrence of watermelon root rot disease. Similarly, Gao et al. previously confirmed that 1,3-benzothiazole produced by B. velezensis ZSY-1 had antagonistic activity against Botrytis cinerea (3). In our study, the content of 1,3-benzothiazole produced at pH 5.5 was lower than that produced at pH 7.0, with a reduction of 36.17%. The inhibitory activity of the VOCs, however, did not decrease because of the reduction in 1,3-benzothiazole. The reason may be that many antagonistic compounds act synergistically to contribute to the inhibitory effect of the VOCs. The 100% inhibition of R. solanacearum caused by 2-ethylhexanol and 1,3-benzothiazole broadens their inhibitory scope and takes the control of plant bacterial wilt caused by R. solanacearum in a new direction.

Another three components of VOCs, namely, methyl thiocyanate, methyl disulfide, and S-methyl thioacetate, also had strong antagonistic activities against R. solanacearum. The inhibitory roles of S-methyl thioacetate, methyl thiocyanate, and 1-undecene, components of the VOCs produced by the antagonistic bacterium P. donghuensis isolated from tomato rhizosphere soil, have been reported. These three volatile substances have excellent applications in preventing the growth of pathogenic fungi, oomycetes, and bacterial diseases (13). The results of this study further suggested that S-methyl thioacetate and methyl thiocyanate had strong inhibitory effects on R. solanacearum. All of these results revealed that the main antagonistic components of VOCs play an important role in interfering with the growth of R. solanacearum and enhancing the inhibitory effect of CLP-6, especially under acidic conditions. In addition, it was noteworthy that the 1-undecene content was greatest in the VOCs produced by CLP-6, but this compound had almost no antagonistic activity. The results address some of the explanations related to the results reported by Plyuta et al. (13), who found that 1-undecene, which was produced by P. fluorescens B-4117 and P. chlororaphis 449 as the main VOC, did not affect the pathogenic biofilm. Our result was consistent with this finding. However, Velivelli et al. suggested that large amounts of 1-undecene and 2,5-dimethylpyrazine inhibit the growth of fungi but these compounds have no antagonistic activity when used in small amounts (30). Conversely, Hunziker et al. (31) pointed out that 1-undecene significantly reduces the growth and spore production of P. infestans. Of course, the inhibitory activity of 1-undecene might be apparent with high doses (32). The inhibitory effects of 1-undecene vary for different kinds of target phytopathogen and in different amounts; although 1-undecene has almost no antagonistic activity against R. solanacearum, it might be involved in a metabolic process in which antagonistic strains secrete active VOCs. This suggestion requires further research.

Finally, for the first time, the present study investigated the fumigant activity of the acid-tolerant P. protegens strain CLP-6 and its volatile components against phytopathogens, especially R. solanacearum, under acidic condition. The VOCs had a broad antagonistic spectrum, particularly for R. solanacearum, both in vitro and under acidic conditions. The component analysis revealed that the content and abundance of VOCs were high when CLP-6 was cultured at pH 5.5, and two new volatiles (2-ethylhexanol and 1,3-benzothiazole) completely inhibited the growth of R. solanacearum. This study extends our knowledge about the efficient, broad-spectrum fumigation of P. protegens CLP-6, which would especially improve the control effect on tobacco wilt in acidic soil, and new antagonistic volatile compounds that could be efficient fumigants to inhibit Solanaceae crop bacterial wilt caused by R. solanacearum.

MATERIALS AND METHODS

Microorganisms and chemicals.

The antagonistic strain P. protegens CLP-6 was isolated from acidic tobacco rhizosphere soil (pH 5.6) in Jiangya, Cili, Zhangjiajie, Hunan Province, China (29°24′N, 110°47′E), and stored at the Tobacco Research Institute of the Chinese Academy of Agricultural Sciences (Qingdao, China). The pathogen R. solanacearum was isolated from the stem tissue of diseased tobacco in Shuanghe Village, Fenghe, Cili, China. The pathogen was identified as physiological race 1 and biochemical type III. The pathogenic fungi, including Fusarium graminearum, F. oxysporum f. sp. cucumerinum, F. oxysporum, F. solani, Alternaria alternata, Colletotrichum destructivum, C. gloeosporioides, and C. chlorophyti, two oomycetes, i.e., Phytophthora capsica and P. infestans, and four pathogenic bacteria, i.e., R. solanacearum, Dickeya chrysanthemi, Pseudomonas syringae pv. tabaci tox⁺, and P. syringae pv. tabaci tox⁻, were from the Tobacco Research Institute, CAAS/Key Laboratory of Tobacco Pest Monitoring, Controlling, and Integrated Management.

S-Methyl thioacetate (1.024 g/mL, >99%), methyl thiocyanate (1.08 g/mL, >99%), methyl disulfide (1.062 g/mL, >95%), 1-decene (0.75 g/mL, >99%), 2-ethylhexanol (0.833 g/mL, >99%), 1,4-undecadiene (0.765 g/mL, >99%), 1-undecene (1.887 g/mL, >99%), 1,3-benzothiazole (1.238 g/mL, >99%), and 2,5-dimethylpyrazine (0.99 g/mL, >99%) were purchased from Shanghai Macklin Biochemical Co., Ltd. (China).

Inhibitory effect of VOCs produced by P. protegens CLP-6.

(i) Antagonistic activity assay of VOCs against R. solanacearum. The antibacterial activity of VOCs produced by CLP-6 against R. solanacearum was evaluated in divided petri dishes (90 mm in diameter). Activated charcoal, a type of charcoal that absorbs VOCs (20), was used in this assay to prove that VOCs were directly responsible for the observed antagonistic activity. The two-sealed-base-plate method was used to test the antagonistic activity of the VOCs. Three treatments were carried out, including (i) activated charcoal plus R. solanacearum, (ii) CLP-6 plus R. solanacearum, and (iii) CLP-6 plus activated charcoal plus R. solanacearum. Two-compartment petri dishes with a partition down the center were used; one section contained NA medium mixed with R. solanacearum (1 × 10⁸ CFU/mL), and the other section was supplemented with or without 5 g of activated charcoal. The other plate was streaked with strain CLP-6 on NA medium to fill the plate or left blank as a control. The two plates were sealed with parafilm and cultured for 120 h at 28°C. Every experiment was performed in triplicate.

(ii) Determination of the antimicrobial spectrum of VOCs. The antagonistic activity of VOCs against plant-pathogenic fungi was also evaluated using the two-compartment-petri-dish method. The pathogen was grown on one side and could not spread to the other side. However, gas exchange proceeded normally (24). The pathogen was grown on 7 mL potato dextrose agar medium in one section and on 7 mL NA medium in the other. P. protegens CLP-6 was streaked onto the full NA medium surface, and 7-mm-diameter spots of the pathogenic fungi F. graminearum, F. oxysporum f. sp. cucumerinum, F. oxysporum, F. solani, A. alternata, C. destructivum, C. gloeosporioides, and C. chlorophyti and two oomycetes, P. capsica and P. infestans, were placed in the center of the potato dextrose agar plate. One plate containing the fungus served as a control (CK). All plates were sealed with Parafilm and cultured for 72 h at 28°C in the dark. Every experiment was performed in triplicate. The diameter of the colony was measured using the cross method (24). The inhibition rates of VOCs were evaluated by calculating the percent inhibition using the following formula (24):

$$\mathrm{Inhibition}\mathrm{rate}(\%)=\left[\mathrm{1}-\left(\mathrm{Da}-\mathrm{Dc}\right)/\left(\mathrm{Db}-\mathrm{Dc}\right)\right]\times \mathrm{1}00\%$$ (1)

where the average diameter of the mycelial colony on the CK plate is D_a, that on the test plate is D_b, and that on the inoculated mycelium plate is D_c.

The antagonistic activity of the VOCs against four plant-pathogenic bacteria was evaluated using two inverse face-to-face petri dishes (90 mm in diameter) (24). Each dish contained 15 mL NA; the upper one was densely streaked with CLP-6, and the bottom one was inoculated with 5 μL of fermentation broth (1 × 10⁸ CFU/mL) containing one of the four tobacco-pathogenic bacteria, i.e., R. solanacearum, Dickeya chrysanthemi, Pseudomonas syringae pv. tabaci tox⁺, and P. syringae pv. tabaci tox⁻. The control contained pathogenic bacteria on one-half of the plate and NA medium only on the other one-half. Every experiment was performed in triplicate. All plates were sealed with Parafilm and cultured for 48 h at 28°C. The pathogenic bacteria were resuspended by washing with 3 mL of sterile double-distilled water (ddH₂O), and the optical density at 600 nm (OD₆₀₀) of the culture was measured. The rate of inhibition by VOCs was determined by calculating the percent inhibition using the following formula (24):

$$\mathrm{Inhibition}\mathrm{rate}\left(\%\right)=\left(\mathrm{Da}-\mathrm{Db}\right)/\mathrm{Da}\times \mathrm{1}00\%$$ (2)

where D_a is the average OD₆₀₀ of the CK plate containing pathogenic bacteria, and D_b is that of the test plate.

Determination of the optimal amount of CLP-6 needed to produce VOCs against R. solanacearum.

Different inoculation conditions of CLP-6, including various inoculation concentrations, amounts, and fumigation times, were used in two inverse face-to-face petri dishes (90 mm in diameter) (24). For the inoculation concentration assay, 5 μL of R. solanacearum suspension (1 × 10⁸ CFU/mL) was inoculated on the bottom NA plate, while 5 μL of CLP-6 suspension at concentrations of 0, 10⁵, 10⁶, 10⁷, and 10⁸ CFU/mL was inoculated at the center of the upper NA plate. The two plates were sealed with Parafilm and cultured for 48 h at 28°C. For the quantitative analysis of the inoculation amounts and fumigation time, 0, 1, 4, 8, and 16 drops, containing 10 μL of CLP-6 suspension (1 × 10⁸ CFU/mL) per drop, were spotted on the surface of the upper NA plate, and 10 μL of R. solanacearum suspension (1 × 10⁸ CFU/mL) was placed on the bottom NA plate. The two plates were sealed with Parafilm and incubated for 24, 48, 96, or 120 h at 28°C. Each experiment was repeated in triplicate. The R. solanacearum was washed twice and resuspended in 3 mL of sterile ddH₂O, and the OD₆₀₀ of the suspension was measured. The inhibition rate was calculated using equation 2 as noted above.

Inhibitory effect of VOCs emitted under different pH culture conditions on R. solanacearum.

(i) Establishing a standard curve. A fluorescence quantitative detection system was used to establish a standard curve. SYBR green I technology (33), with a 20-μL fluorescent quantitative PCR system containing 10.0 μL SYBR green Supermix (2×), 0.4 μL ROX dye II (50×), 0.8 μL RSF (10 μM), 0.8 μL RSR (10 μM) (RSF and RSR means forward and Reverse primer names respectively), 2.0 μL model DNA, and 6.0 μL ddH₂O, was employed. The standard curve for the plasmid copy number was established by plotting the cycle threshold (C_T) value for the fluorescence intensity on the ordinate and the logarithm of the plasmid copy number for different concentrations on the abscissa.

To carry out real-time PCR of R. solanacearum, DNA fragments were amplified on a commercial instrument (ABI 7500/7500 fast real-time PCR system; Applied Biosystems, USA). The product of this PCR was expected to be 159 bp (34) (Table 4). Positive amplification was scored when the SYBR green fluorescence exceeded a threshold of 0.015 arbitrary units (AU) at any cycle during the PCR process. The PCR conditions were as follows: preheating at 95°C for 30 s and then 95°C for 5 s and 60°C for 30 s for 40 cycles; each treatment was repeated five times, and a dissolution curve analysis was performed immediately after PCR. The specificity of the PCR product amplification was verified based on the melting temperature (T_m) values for 15 s at 95°C and 1 min at 60°C.

TABLE 4.

Primers and sequencing conditions used in this study

Parameter	Finding
Gene	UDP-3-O-acetylglucosamine deacetylase
Primer sequence (5′ to 3′)
Forward	GTGCCTGCCTCCAAAACGACT
Reverse	GTGCCTGCCTCCAAAACGACT
Tm (°C)	60
Product size (bp)	159

(ii) Effects of VOCs on the quantity of R. solanacearum in soil. The fluorescence quantitative PCR method was used to determine the GCN for R. solanacearum in soil exposed to VOCs by conducting experiments in two inverse face-to-face petri dishes (120 mm in diameter). CLP-6 and R. solanacearum were incubated with shaking at 28°C to reach concentrations of 2 × 10⁸ CFU/mL. A total of 100 μL of CLP-6 was fully spread on NA medium at a pH of 5.5 or 7.0 (120 mm diameter). Soils were autoclaved for 1 h at 121°C, inoculated with 10 mL of a suspension of 2 × 10⁸ CFU/mL of R. solanacearum after cooling, and divided equally into two parts; they were challenged with strain CLP-6 (grown on NA medium at pH 5.5 or pH 7.0) or blank NA medium and cultured in an incubator for 4 days at 28°C. The soil was sampled at 0 dpi as the initial CK and at 4 dpi. Each treatment was repeated five times, and a 2-g sample of the rhizosphere soil was used each time. The soil bacterial genomic DNA was extracted using a soil genomic DNA extraction kit (Solarbio). The fluorescence quantitative PCR was performed according to the manufacturer’s instructions, and the dynamic curves of growth and decline and the C_T values for each sample were recorded. The C_T value was introduced into the standard curve equation to calculate the initial GCN of the sample, to determine the GCN per gram of dry soil.

(iii) Pot experiment. Forty-day-old tobacco plants were transplanted into pots (5 cm in diameter) containing 20 g of the aforementioned mixed-bacteria soil treated with CLP-6 cultured at pH 5.5 or pH 7.0, with blank NA medium as the CK. The soil from the three plates was put into one pot, in which the tobacco seedlings were planted. Each treatment was repeated in triplicate. All seedlings were maintained in a greenhouse at 28°C (80% relative humidity, with a 14-h light/10-h dark cycle) and inspected periodically until disease symptoms appeared. The plants were evaluated for disease severity due to R. solanacearum infection at 14 dpi. The disease severity was evaluated according to an empirical scale with six degrees based on the grade and investigation method of tobacco diseases and insect pests (GB/T 23222-2008) (23), as follows: level 0, tobacco plants without visible symptoms; level 1, striped necrosis on stems occasionally or less than one-half of the leaves wilted on unilateral stems; level 3, black streaks less than one-half the height of the stem or one-half to two-thirds of the leaves wilted on unilateral stems; level 5, black streaks over one-half the length of the stem but not reaching the top of the stem or more than two-thirds of the leaves wilted on unilateral stems; level 7, black streaks reaching the top of the stem or all leaves wilted; level 9, the plant is dead. The disease severity index was calculated according to the following formula:

$$\text{Disease\hspace{0.17em}severity\hspace{0.17em}index}\left(\mathrm{DSI},\%\right)=\left[\Sigma \left(\mathrm{x}\times \mathrm{y}\right)/\left(\mathrm{z}\times \mathrm{9}\right)\right]\times \mathrm{1}00\%$$ (3)

$$\text{Disease\hspace{0.17em}control\hspace{0.17em}effect}\left(\%\right)=[\left(\mathrm{CLP}-{\mathrm{6}}_{\mathrm{DSI}}-\mathrm{C}{\mathrm{K}}_{\mathrm{DSI}}\right)/\left(\mathrm{C}{\mathrm{K}}_{\mathrm{DSI}})\right]\times \mathrm{1}00\%$$ (4)

where x is the number of different degrees of infected plants in the treatments, y is the relative degree value, and z is the number of total plants in the treatments.

Collection and identification of VOCs produced by P. protegens CLP-6 under pH 5.5 and pH 7.0 culture conditions.

We identified the VOCs emitted by P. protegens CLP-6 using headspace SPME–GC-MS. CLP-6 was incubated for 96 h at 28°C in a 20-mL headspace vial containing 5 mL of NA medium at pH 5.5 or pH 7.0. Blank NA medium (pH 5.5 or pH 7.0) was used as a CK to eliminate the interference of the VOCs released by the NA medium and the loss of the extraction head coating in the results. Each experiment was performed in triplicate.

The incubated samples were equilibrated at room temperature for 20 min, a 50/30-μm divinylbenzene/Carboxen/polydimethylsiloxane (DVB/CAR/PDMS) SPME fiber (purchased from Supelco) was used for sampling, and the fiber was activated according to the manufacturer's instructions. To sample the VOCs, the SPME fiber was exposed to the headspace of the upper one-third of the headspace vital, to avoid contact with P. protegens CLP-6 colonies, and incubated at room temperature for 30 min. Later, the fiber was pulled away from the sample vial and immediately introduced into the GC injector for desorption for 5 min at 250°C using an RTX-5MS column (Agilent 7890 GC system coupled with a 5977 mass spectrometer and matched with a DB-5MS capillary column [30 m by 0.25 mm by 0.25 μm]). Helium was used as the carrier gas, with the flow velocity adjusted at 11.0 mL/min. The oven temperature was initially kept at 40°C for 2 min, raised to 180°C at 5°C/min, further raised to 180°C for 10 min, finally raised from 180 to 280°C at 10°C/min, and kept at 280°C for 15 min. The transfer line, ion source, and quadruple analyzer temperatures were 280, 230, and 150°C, respectively. The mass spectra were recorded in the positive ionization mode at 70 eV and 220°C and were acquired using the full scan monitoring mode with a mass scan range of 35 to 500 m/z.

The mass spectra of VOCs were compared with those in the NIST/EPA/NIH MS library with respect to spectra in the Mainlib and/or Replib databases. Tentative compound identification was based on ≥80% quality match with the NIST database information for each compound and retention indices reported in the literature when reference standards were not available. The retention indices were calculated according to the retention times of C7 to C30 n-alkane retention markers under the same chromatographic conditions.

The antagonistic activities of nine pure standards were analyzed to define the bioactivity of individual components of VOCs. R. solanacearum (10⁸ CFU/mL) was sprayed onto the NA medium surface, and 10 μL of each pure standard compound was placed in the center of the NA medium; an equal amount of ddH₂O was used as the control. The concentration of R. solanacearum was measured according to the method described above, and the inhibition rate was calculated using equation 2, as indicated above, at 2 dpi. Each treatment was repeated in triplicate.

Statistical analysis.

Differences among treatments were identified by one-way analysis of variance (ANOVA). Duncan’s multiple-range test was applied when one-way ANOVA revealed significant differences (P < 0.05). All statistical analyses were performed using Data Processing Station version 9.50 (China).