Bacillus VOCs in the Context of Biological Control

A contemporary agricultural production system relying on heavy usage of agrochemicals represents a questionable outlook for sustainable food supply in the future. The visible negative environmental impacts and unforeseen consequences to human and animal health have been requiring a shift towards the novel eco-friendly alternatives for chemical pesticides for a while now. Microbial-based biocontrol agents have shown a promising potential for plant disease management. The bacteria of the genus Bacillus have been among the most exploited microbial active components due to several highly efficient mechanisms of action against plant pathogens, as well as a palette of additional plant-beneficial mechanisms, together with their suitable properties for microbial biopesticide formulations. Among other bioactive metabolites, volatile organic compounds (VOCs) have been investigated for their biocontrol applications, exhibiting the main advantage of long-distance effect without the necessity for direct contact with plants or pathogens. The aim of this study is to give an overview of the state-of-the-art in the field of Bacillus-based VOCs, especially in terms of their antibacterial, antifungal, and nematicidal action as the main segments determining their potential for biocontrol applications in sustainable agriculture.


Introduction
Microbial volatile organic compounds (mVOCs) represent a diverse group of molecules synthesized through different metabolic pathways by microorganisms, predominantly bacteria and fungi [1,2]. The role of mVOCs was explained through microorganisms' evolution in the context of microbial communities due to their ability to mediate interactions with certain VOCs acting like infochemicals [1,3]. Bacteria use a cell-to-cell communication system (quorum sensing) as a method for monitoring the population density [3,4] relying on the synthesis, release, and subsequent detection of small diffusible signal molecules known as autoinducers [4]. Quorum sensing is variable depending on the bacteria type, allowing both intra-and inter-species communication. The chemical communication enabled by diffusible autoinducers is possible at short distances between cells, also requiring a high concentration of signaling molecules. Recent research, on the other hand, showed that microorganisms also produce different inorganic and organic volatile compounds that can be used as signals in intra-and inter-kingdom interactions, even at lower concentrations and over long distances [2]. This kind of communication system enables the coordinated behavior of a group of organisms to achieve process regulation leading to virulence, biofilm formation, and other developmental processes. The important finding in this matter was that microorganisms employ volatiles during interactions with plants, fungi, nematodes, and bacteria [4].
The beneficial effect of VOCs is additionally supported by several environmentally friendly properties, including biodegradability, meaning they do not leave toxic residues on plant surfaces, low molecular weight (100-500 Da), the lipophilic character with a low boiling point, and high vapor pressure (0.01 kPa), making them easily evaporative at normal temperature and pressure and enabling diffusion through the atmosphere and soil

ISR (Induced Systemic Resistance) Induction via Different Signaling Pathways as the Mechanism of Antibacterial Activity of Bacillus VOCs
Among the pioneers in this field, Ryu et al. [15] have proven the antibacterial effect of 2,3-butanediol produced by Bacillus subtilis GB03 and B. amyloliquefaciens IN937a against Pectobacterium carotovorum subsp. carotovorum (syn. Erwinia carotovora subsp. carotovora). The main mechanism of antibacterial action in the aforementioned study was ISR (induced systemic resistance) in Arabidopsis thaliana after 4 days of exposure to Bacillus VOCs, which was mediated through the ethylene-dependent signaling pathways, while the ISR was independent of the salicylic acid or jasmonic acid signaling pathways. VOCs

ISR (Induced Systemic Resistance) Induction via Different Signaling Pathways as the Mechanism of Antibacterial Activity of Bacillus VOCs
Among the pioneers in this field, Ryu et al. [15] have proven the antibacterial effect of 2,3-butanediol produced by Bacillus subtilis GB03 and B. amyloliquefaciens IN937a against Pectobacterium carotovorum subsp. carotovorum (syn. Erwinia carotovora subsp. carotovora). The main mechanism of antibacterial action in the aforementioned study was ISR (induced systemic resistance) in Arabidopsis thaliana after 4 days of exposure to Bacillus VOCs, which was mediated through the ethylene-dependent signaling pathways, while the ISR was independent of the salicylic acid or jasmonic acid signaling pathways. VOCs

ISR (Induced Systemic Resistance) Induction via Different Signaling Pathways as the Mechanism of Antibacterial Activity of Bacillus VOCs
Among the pioneers in this field, Ryu et al. [15] have proven the antibacterial effect of 2,3-butanediol produced by Bacillus subtilis GB03 and B. amyloliquefaciens IN937a against Pectobacterium carotovorum subsp. carotovorum (syn. Erwinia carotovora subsp. carotovora). The main mechanism of antibacterial action in the aforementioned study was ISR (induced systemic resistance) in Arabidopsis thaliana after 4 days of exposure to Bacillus VOCs, which was mediated through the ethylene-dependent signaling pathways, while the ISR was Antibiotics 2023, 12, 581 5 of 40 independent of the salicylic acid or jasmonic acid signaling pathways. VOCs produced by B. subtilis GB03 and B. amyloliquefaciens IN937a have shown a supreme biocontrol effect by reducing the number of symptomatic leaves to one in Arabidopsis plants, even in comparison to 2,3-butanediol applied as a pure compound (1-2 symptomatic leaves) as well as compared to water used as a control (4-5 symptomatic leaves) [15]. Acetoin (3-hydroxy-2-butanone) was found to be the main VOC produced by B. subtilis FB17 that triggers systemic resistance in Arabidopsis thaliana against Pseudomonas syringae pv. tomato DC3000 via induction of salicylic acid and ethylene signaling pathways, while the jasmonic acid signaling pathway was not essential for ISR. Plants treated with B. subtilis FB17 or acetoin as a pure compound have shown milder disease symptoms in terms of chlorosis, as well as a significant reduction of pathogen incidence measured as CFU/g of fresh weight (around 1 log unit) [16]. The antibacterial activity of 3-pentanol and 2-butanon supplied by the drench treatment in concentrations of 1 mM and 0.1 µM, respectively, was reported through the consistent triggering of systemic resistance in cucumber against Pseudomonas syringae pv. lachrymans, causing bacterial angular leaf spots, whereas 2,3-butanediol had been ineffective in eliciting induced resistance in cucumber plants in the aforementioned concentrations [17]. Later, it was reported that ten Petri dishes containing 10 µM 2,3butanediol (the main VOC produced by B. subtilis GB03) were utilized in the miniature greenhouse system to successfully induce systemic resistance against Pseudomonas syringae pv. lachrymans on cucumber plants, which was mediated via the jasmonic acid signaling pathway. The disease severity score of cucumber angular leaf spot was reduced to 2.7 for B. subtilis GB03 VOCs treatment, compared to 4.2 for plants without VOCs treatment (on the scale in the range 0-5) [18]. The subsequent study revealed that 3-pentanol primes salicylic acid and jasmonic acid signaling pathways responsible for ISR towards Pseudomonas syringae pv. tomato in Arabidopsis plants [19]. Similar results regarding the ISR mechanisms were reported by Choi et al. [20], in which 3-pentanol from B. amyloliquefaciens IN937a has shown a significant reduction in pepper bacterial spots severity caused by Xanthomonas axonopodis pv. vesicatoria in field trials over two years. The disease index, measured 40 days after plant transplantation in the field, was reduced to 0.7 and 1.7 in 3-pentanol treatments, compared to 4.0 and 3.8 for water used as control (scale 0-5) in two subsequent years [20].

Modulation of Pathogens' Gene Expression by Antibacterial Bacillus-Based VOCs
Inhibition of virulence traits and metabolic activity by inhibition of the related genes' expression and modulating the metabolic pathways responsible for pathogenic behavior are among the most important mechanisms of action of Bacillus-generated VOCs in terms of their antibacterial activity. Furthermore, several studies have shown improved antibacterial activity achieved by the mixture of VOCs compared to separate compounds, thereby targeting the expression of different pathogenicity-related genes. The antibacterial activity of the VOC mixture produced by B. amyloliquefaciens SQR-9 was proven against Ralstonia solanacearum ZJ3721 (biovar 3), the causal agent of tomato bacterial wilt, with an efficiency of 70% in bacterial pathogen growth suppression when the VOC mixture was applied as a biocontrol agent, contrary to the 1-11% growth inhibition achieved by the individual VOCs [21]. The underlying mechanisms of antibacterial activity included inhibition of motility, biofilm formation, and root colonization by Ralstonia solanacearum, as well as inhibited production of antioxidant enzymes and exopolysaccharides, with significant down-regulation of genes involved in virulence traits, carbohydrate and amino acid metabolism, protein translation, and folding, as well as antioxidant activity [19]. Almost similar mechanisms of action were observed in the case of B. amyloliquefaciens T-5 VOCs antibacterial activity against the tomato bacterial wilt pathogen, where pathogen growth inhibition rates of 75%, 62%, and 85% were achieved using the mixture of VOCs produced by the Bacillus antagonist on the agar medium, in sterilized soil, and natural soil, respectively [22]. This study also reported down-regulation of several virulence-and metabolism-related genes of Ralstonia solanacearum in the range 27-54%, including transcriptional regulator (phcA), catalase (katG), and SOD (superoxide dismutase-sobB) genes Antibiotics 2023, 12, 581 6 of 40 motility-related genes (twitching motility (pilT) and flagellin (fliC)) as well as metabolismrelated genes, DNA polymerase (polA) and pyruvate dehydrogenase (aceE) [22]. The same research group has also investigated induced VOC production in the presence of organic fertilizers prepared from different animal and plant organic waste materials [23], where the production of VOCs was significantly improved when combined with the application of organic fertilizers. This suggests a possible novel good agricultural practice in tomato bacterial wilt management. It was concluded that maximal production of 2-nonanone, nonanal, xylene, benzothiazole, and butylated hydroxy toluene was achieved by the strain B. amyloliquefaciens T-5 in the presence of organic fertilizer made of amino acid organic fertilizer (prepared from the microbially hydrolyzed oil rapeseed cake) and pig manure compost (1:1, w/w). On the other hand, maximal production of 2-nonanone, nonanal, xylene, and 2-undecanone was achieved by B. amyloliquefaciens SQR-9 in the presence of organic fertilizer composed of 41% vinegar-production residue, 20% rice straw, and 39% cattle dung [23].
The suppression of Ralstonia solanacearum as the causal agent of tobacco bacterial wilt was also investigated by Tahir et al. [12,24]. Albuterol and 1,3-propanediol produced by B. subtilis SYST2 were found to be responsible for ISR and growth promotion of tobacco plants at the level of inhibition of gene expression related to ethylene production, while on the other hand, genes related to expansin, wilt resistance, and plant defense were overexpressed. Moreover, the wilt index in tobacco plants was reduced from 90.66% to 33.00%, 19%, and 27% by applying inoculation with B. subtilis SYST2 or treatments with 1,3-propanediol (1 mM) and albuterol (0.1 mM) [12]. Benzaldehyde, 1,2-benzisothiazol-3(2H)-one, and 1,3-butadiene were found to be major VOCs produced by B. amyloliquefaciens FZB42 and B. atrophaeus LSSC22, which suppressed Ralstonia solanacearum in tobacco plants by reducing the colony size, pathogen cell viability, and motility, and negatively influencing chemotaxis. One of the major mechanisms of action was the induction of morphological and ultrastructural changes/abnormalities in Ralstonia solanacearum cells. Furthermore, expression of pathogenicity-related genes was significantly down-regulated, including PhcA (a global virulence regulator), type III secretion system (T3SS), type IV secretion system (T4SS), extracellular polysaccharides, and chemotaxis-related genes. On the other hand, genes related to pathogen resistance and defense were significantly overexpressed, where the salicylic acid pathway was found to be responsible for the induction of systemic resistance, with up-regulation of the EDS1 and NPR1 genes as its main components. All of the aforementioned mechanisms of action have resulted in the reduced tobacco wilt index during in planta experiments (28% for B. amyloliquefaciens FZB42 and 43.2% for B. atrophaeus LSSC22 VOCs compared to 98% for the non-exposed control) [24].
Growth inhibition of Ralstonia solanacearum was also achieved by 3,5,5-trimethylhexanol produced by B. cereus D13, which has also been successful in suppression of Pseudomonas syringae pv. tomato DC3000 and Xanthomonas oryzae pv. Oryzicola. This is while both 3,5,5trimethylhexanol and decyl alcohol have shown antibacterial activity against Xanthomonas oryzae pv. oryzae with growth inhibition of 60.7% and 53.6% at minimum inhibitory amounts of 0.48 and 2.4 mg, respectively [25]. The same study has revealed several mechanisms of antibacterial activity against Xanthomonas oryzae pv. oryzae as the causal agent of bacterial leaf blight in rice. The first one was the inhibition of pathogen swimming and swarming motility after 3-day exposure to B. cereus D13 VOCs, with significant down-regulation of the genes motA, encoding flagellar motor component MotA, and motC, encoding the flagellar motor protein. Interestingly, while pathogen motility was restricted during the first 3 days of exposure, it took 6 days of exposure to observe a sharp decline in the number of viable pathogen cells, indicating that a certain period of time was required to achieve lethal VOCs concentration, while a certain extent of pathogen spreading inhibition could be achieved in earlier phases of antagonist-pathogen interaction at the sub-lethal level of VOCs concentration. Considering that the rhizosphere is a relatively closed environment, it is a favorable place for maximization of VOCs' antibacterial activity, considering the shorter time necessary to achieve a certain VOCs concentration threshold required for bactericidal Antibiotics 2023, 12, 581 7 of 40 effect. In the case of Xanthomonas oryzae pv. oryzae, rice rhizosphere is therefore a good place to apply antagonist or VOCs solely in order to prevent or inhibit infection via root wounds [25].

Structural and Functional Changes at Cell Level Caused by Antibacterial Bacillus VOCs
Another mechanism of antibacterial activity was observed in the case of B. cereus D13 VOCs, including alteration of Xanthomonas oryzae pv. oryzae cell membrane permeability resulting in changed cell morphology, which could be directed in two ways: (a) increased membrane permeability through membrane distortion, resulting in leakage of intracellular contents, and (b) decreased membrane permeability resulting in concentrated cytoplasm [25]. Hence, the action mode of 3,5,5-trimethylhexanol and decyl alcohol VOCs could be explained by their ability to cause leakage or impede the interchange of materials at the cell membrane surface [4].
On the other hand, one of the possible antibacterial activity mechanisms related to pathogen cells' morphological changes is also inducing several cell abnormalities at the ultrastructural level, as mentioned before [24]. The same was observed by Rajer et al. [5], where benzaldehyde, nonanal, benzothiozole, and acetophenone emitted by B. subtilis FA26 induced a wide range of abnormalities in cells of Clavibacter michiganensis ssp. sepedonicus, a causal agent of potato bacterial ring rot, including distorted colony morphology, misshapen cells and their disintegration, the formation of inclusions, movement of cytoplasmic content toward the ruptured cytoplasmic membrane, and lack of cytoplasmic content or fragmented cytoplasm due to intracellular content leakage. The same authors concluded that the production and activity of B. subtilis FA26 VOCs depend on cultivation conditions and that an increase in their biocontrol potential against Clavibacter michiganensis ssp. sepedonicus is related to the increased VOC concentrations [5].

Effects of Antibacterial Bacillus VOCs Concentration and Treatment Timing on Biocontrol Efficiency
Besides cultivation conditions, the other important variable determining the efficacy of the VOC treatment is the time of application, considering the current pathogen growth and development stage. Han et al. [25] have shown that the mixture of VOCs produced by B. velezensis strains JCK-1618 and JCK-1696 was more successful in suppressing Xanthomonas arboricola pv. pruni when B. velezensis strains were inoculated three days before pathogen inoculation, compared to a one-day window between the antagonist and pathogen inoculation, where the pathogen suppression efficacy ranged between 22.4% and 72.2%.
A mixture of B. velezensis X5-2, B. megaterium X6-3, and Pseudomonas orientalis X2-1P VOCs, including alkanes, alkenes, pyrazines, acids, alcohols, and indoles, was found to be effective during the in vitro and in vivo suppression of the winter oilseed rape black rot pathogen Xanthomonas campestris pv. campestris, with disease reductions of 82.37% and 72.47% in preventive and curative treatments, respectively [26]. Pyrazines (pyrazine, 2-ethyl-3-methyl, pyrazine, 2-ethyl-, pyrazine, 2, 5-dimethyl, and pyrazine, 2-methyl) were also found as active VOCs of B. megaterium BmBP17 against the bacterial wilt pathogen Ralstonia solanacearum, where pyrazine, 2-ethyl-3-methyl, was found to be the most effective compound with a required dosage of 672 µg/mL for complete inhibition of Ralstonia solanacearum [27]. This study also confirmed an increase in antibacterial activity with an increase in VOC concentrations.

Mechanisms of Action of Antifungal Bacillus-Based VOCs
The overview of published literature data investigating the antifungal activity of Bacillus-based VOCs shows a supreme number of studies compared to literature investigating antibacterial and nematicidal effects. Hence, a general overview of the mechanisms involved in the suppression of plant-pathogenic fungi by the Bacillus VOCs will be given here (summarized in Figure 3), emphasizing the most abundant and effective VOCs in terms of antifungal activity. A more detailed overview of the published data regarding the antifungal activity of VOCs produced by different strains of the genus Bacillus can be found in Table 2, while the most abundant Bacillus-based VOCs with antifungal activity towards plant pathogens are presented in Figure 4.
The overview of published literature data investigating the antifungal activity of Bacillus-based VOCs shows a supreme number of studies compared to literature investigating antibacterial and nematicidal effects. Hence, a general overview of the mechanisms involved in the suppression of plant-pathogenic fungi by the Bacillus VOCs will be given here (summarized in Figure 3), emphasizing the most abundant and effective VOCs in terms of antifungal activity. A more detailed overview of the published data regarding the antifungal activity of VOCs produced by different strains of the genus Bacillus can be found in Table 2, while the most abundant Bacillus-based VOCs with antifungal activity towards plant pathogens are presented in Figure 4.   The overview of published literature data investigating the antifungal activity of Bacillus-based VOCs shows a supreme number of studies compared to literature investigating antibacterial and nematicidal effects. Hence, a general overview of the mechanisms involved in the suppression of plant-pathogenic fungi by the Bacillus VOCs will be given here (summarized in Figure 3), emphasizing the most abundant and effective VOCs in terms of antifungal activity. A more detailed overview of the published data regarding the antifungal activity of VOCs produced by different strains of the genus Bacillus can be found in Table 2, while the most abundant Bacillus-based VOCs with antifungal activity towards plant pathogens are presented in Figure 4.

Cultivation and Treatment Variables Affecting Antifungal Efficiency of Bacillus VOCs
The effect of the growth medium on VOCs production and antibacterial activity was reported by Huang et al. [29], where 90% growth inhibition of Rhizoctonia solani and Pythium aphanidermatum (Edson) was achieved using B. mycoides grown on TSA (tryptic soy agar) or SPMA (soy powder milk agar). This is while the application of KBA (King's B agar) and LBA (Luria-Bertani agar) resulted in moderate pathogen growth inhibition and NA (nutrient agar) and PDA (potato dextrose agar) showed almost no inhibitory activity against fungal pathogens. Similar results in terms of the suitability of TSA as a growth medium were obtained by Gotor-Vila et al. [30] for B. amyloliquefaciens CPA-8, whose antifungal activity was the highest compared to NYDA (nutrient yeast glucose agar) and NAglu20 (nutrient agar supplemented with glucose) in the cases of Monilinia laxa (>78.6%), Botrytis cinerea (86.8%), and Monilinia fructicola (68.6%) suppression. In the study by Fujimoto et al. [31], TSB (tryptone soya broth) and TSA favored the production of VOCs by Bacillus sp. ACB-65 and Bacillus sp. ACB-73, which exhibited antifungal activity against Phyllosticta citricarpa, the orange-black spot pathogen, over the NA, PDA, and King B media. Specific VOCs were produced by B. amyloliquefaciens strains UCMB5033, UCMB5036, UCMB5113, and FZB42 using different cultivation media: 2,3-butanedione and acetoin on M9A medium, while 5-methyl-heptanone, 2-methylpyridine, and 2-pentanone were produced on TSA and LBA media [32]. The highest production of antifungal VOCs (chloroacetic acid, tetradecyl ester, octadecane and hexadecanoic acid, and methyl ester) by B. atrophaeus HAB-5 in submerged cultivation was obtained in LB medium, followed by BPY (beef peptone yeast) and MH (Mueller Hinton) medium, inhibiting the growth of Colletotrichum gloeosporioides by 47.81, 45.51, and 44.55%, respectively [33]. Different types of growth media (NA, TSA, LBA, and TMEA (TM Enterprise agar) were investigated for their effect on VOCs production by B. pumilus TM-R and their antifungal activity against Alternaria alternata, Aspergillus niger, Cladosporium cladosporioides, Curvularia lunata, Fusarium oxysporum, and Penicillium italicum in both small-and large-scale tests (plate and 12L-tests, respectively). TMEA medium resulted in the strongest antifungal activity, supported by the production of methyl isobutyl ketone, ethanol, 5-methyl-2-heptanone, and S-()-2-methylbutylamine as the predominant antifungal VOCs [34]. MOLP medium (medium for optimum lipopetide production) used for the production of VOCs by B.velezensis BUZ-14 and B. ginsengihumi S38 resulted in 90% suppression of Botrytis cinerea in table grapes, while grape juice was the least favorable medium for VOC efficacy [35].
Furthermore, cultivation time also affects the type and content of the bacterial VOCs produced. The largest diversity of VOCs in the cultivation broth of B. subtilis CF-3 was detected after 48 hof cultivation, while the peak of the antifungal efficiency (73.46% on Monilinia fructicola, causing peach brown rot, and 63.63% on Colletotrichum gloeosporioides, causing litchi antrachnose), was achieved after 24 h of cultivation. Benzothiazole and 2,4-ditert-butylphenol showed a strong inhibitory effect on both pathogens in vitro and vivo [36], with EC 50 values of 9.90 × 10 −4 mol/L for Monilinia fructicola and 1.26 × 10 −2 mol/L for Colletotrichum gloeosporioides [37]. Similar results were obtained by Ni et al. [28], where B. atrophaeus JZB120050 produced 29 alkanes, three alkenes, four acids, one aldehyde, and one phenol after 24 h, while 41 compounds were detected via GC-MS analysis, including 34 alkanes, three alkenes, two acids, one benzene, and one phenol after 48 h of cultivation. Zheng et al. [38] investigated the mixture content of VOCs produced by B. amyloliquefaciens PP19 to suppress Peronophythora litchi and found significant differences in VOCs production over the course of the cultivation, where a total of 9, 33, 14, 28, and 17 compounds were detected at each of the five investigated time points (24,36,48,60 and 72 h, respectively). However, only two compounds (2-nonanone and 6-methyl-2-heptanone) were common at all time points and constantly produced during the cultivation. Zhang et al. [39] have found that there was no significant difference in biocontrol activity regarding the in vivo suppression of raspberry postharvest diseases caused by Botrytis cinerea and Rhizopus stolonifer when VOCs produced by B. siamensis G-3 in NA medium and at pH 7.0 were applied for treatment after 72 h and 84 h of cultivation. This points out the necessity to optimize cultivation time in order to achieve maximal VOCs biocontrol efficacy while minimizing production operational costs. The timing of inoculation/VOCs in correspondence with the pathogen infection time frame has been considered an important aspect affecting the efficiency of plant pathogen suppression, pointing out the importance of choosing between the preventive/curative plant disease management strategies. Previously mentioned molecular signaling between antagonists and plant pathogens and the related induction of VOC production in the pathogen's presence need to be understood in more detail to be able to define the appropriate timing of the treatment to maximize its efficiency. For example, Han et al. [25] have shown that the inoculation of antagonists B. velezensis JCK-1618 and JCK-1696 3 days before the fungal pathogen inoculation has significantly increased the treatment efficiency compared to a shorter time period (1 day) between the subsequent antagonist and pathogen inoculation. The efficiency of the VOCs produced by B. velezensis JCK-1696 was increased from 67% to 75.7% inhibition for Mycosphaerella cerasella and from 44.9% to 66.8% inhibition for Epicoccum tobaicum, taking into account a 1-day and 3-day inoculation interval [25]. These findings support several mechanisms of action described for Bacillus spp. biocontrol activity, including competition for growth space and nutrients with extra time given to establish antagonist populations at the target site of application as well as the longer time necessary to achieve a fungicidal level of VOC content. Jangir et al. [40] have reported an increase in the antifungal activity of Bacillus sp. B44 VOCs against Fusarium oxysporum f. sp. lycopersici from 20% on the 1st day of the in vitro experiment to 70% on the 7th day of incubation, while the in vivo test showed a 36% reduction in tomato disease incidence. An increase in the antifungal activity of VOCs produced by B. amyloliquefaciens ALB629 and UFLA285-including 3-methylbutanoic acid, 2-methylbutanoic acid, isovaleric acid, and 2-methyl butyric acid-was observed in the reduction of common bean antrachnose from 83% on the first day to 93% disease incidence reduction on the 11th day [41]. This was also shown by Arrebola et al. [42], where VOCs produced by B. amyloliquefaciens PPCB004 exhibited the highest inhibition of fungal radial growth of Penicillium crustosum (73.3%) in vitro after 10 days. This is while a significant in vivo decrease of decay incidence and severity in Valencia orange was observed after a 12-day treatment. Similar results were observed by Leelasuphakul et al. [43], where the disease incidence of green mold caused by Penicillium digitatum Sacc. in citrus fruit was decreased by 86.7% and disease symptoms were delayed by 6 days and decay symptoms to day 9 by the VOCs produced by the B. subtilis 155 suspensions, which was inoculated 24 h before pathogen inoculation. B. amyloliquefaciens NJN-6 inhibited the mycelial growth of Fusarium oxysporum f. sp. cubense by 30% to 40% compared with the control after 3 days of treatment, while in the soil test the number of Fusarium oxysporum f. sp. cubense was significantly reduced (10 2 spores/g compared to 10 4 spores/g in control samples) after 45 days of treatment [44]. The same study identified benzene (2,3,6-trimethyl-phenol) and ketone (2-undecanone, 2-dodecanon, and 2-tridecanone) compounds as the most effective in terms of antifungal activity, where the number of carbon atoms in ketones negatively correlated with their antifungal activity against Fusarium oxysporum f. sp. cubense [44]. B. subtilis Bs 8B-1 has produced a sufficient amount of antifungal VOCs after 5 days of incubation with Phytophthora capsici, causing cucumber damping-off, and Rhizoctonia solani, causing radish damping-off [45]. B. velezensis RDA1 VOCs inhibited the growth of Rosellinia necatrix by approximately 60-70% compared to control after 10 days of treatment [46].

Morphological and Ultrastructural Abnormalities in Fungal Cells Caused by Bacillus VOCs
Mycelia morphological abnormalities were observed in Sclerotinia sclerotiorum by Liu et al. [56], together with spore cracking causing the brownish color of sporaceous inclusion and its effusion after 24-48 h of treatment using Paenibacillus polymyxa BMP-11, B. subtilis BL02, B. pumilus BSH-4, and B. pumilus ZB13 VOCs. Several VOCs from B. megaterium USB2103 have caused ultrastructural alterations at cell organelles, mostly membranes, mitochondria, and endoplasmic reticulum of Sclerotinia sclerotiorum [49]. Hemolytic activity of the majority of the pure VOCs confirmed that the membrane appeared to be one of the primary targets in terms of the antifungal mechanism of action since 2-nonanone caused damage to the cytoplasmic membrane, resulting in complete or partial hyphae emptying. Detachment of the membrane from the outer cell wall resulted in strong vacuolization with internal residues of membranes in the cytoplasm. DL-limonene treatment has also led to cytoplasmic membrane detachment from the cell wall, as well as cytoplasm granulation, the absence of organelles, multi-vesciculation, and the accumulation of protein and lipidic material in the cytoplasm. Dimethyl disulfide has shown strong ultrastructural modifications, including missing or altered cytoplasm, hyper-vesiculation, hypocrested and vesiculated mitochondria, and accumulation of protein and lipidic material in the cytoplasm [49]. Ultrastructural damage of Sclerotinia sclerotiorum hyphae due to the VOCs of B. velezensis VM11 included abnormalities on cell membranes, mitochondria, nucleus, multivesicular structures, and cytoplasm, together with increased vacuoles' size and disorganized cytoplasmic materials to the extent of non-descript cell organelles [57]. On the other hand, Monteiro et al. [58] have not observed any hyphae alteration in Sclerotinia sclerotiorum causing white mold in Lactuca sativa after contact with VOCs produced by B. subtilis, although the reduction of mycelial growth was 83.84%, indicating the fungistatic effect of the produced VOCs.
Abnormalities on conidiophores of Penicillium crustosum were observed in treatment with VOCs produced by B. subtilis PPCB001, while complete loss of conidiophore structures was detected in combined treatment with B. amyloliquefaciens PPCB004 VOCs, with the presumed dominant role of 3-hydroxy-2-butanone (acetoin), which has induced reduction of the multiple phialides at the end of each hyphae, as well as vacuolation and swelling in hyphae and sporangium [41]. Besides the reduced radial growth in vitro, dimethyl disulphide and ammonia as the main VOCs produced by B. mycoides had reduced the damping-off disease of cabbage seedlings caused by Pythium aphanidermatum Edson by 45% in greenhouse experiments, while the disease incidence caused by Rhizoctonia solani had not been reduced. The bacterial VOCs affected the morphology of fungal hyphae, causing poor rigidity, shrinkage, curling, swelling, and hyphal deformation, together with organelle degeneration [29]. Abnormal swelling and increased branching of the Fusarium oxysporum f. sp. niveum hyphae were induced by VOCs produced by B. subtilis IBFCBF-4 [59].
O-anisaldehyde produced by B. atrophaeus CAB-1 showed a higher in vitro inhibitory effect than L-alaninol on hyphal elongation of Botrytis cinerea when applied at the concentration of 80 µL/plate (70.2% growth inhibition). Interestingly, the individual VOCs produced by B. atrophaeus CAB-1 have not been successful in the biocontrol of cucumber powdery mildew (Sphaerotheca fuliginea), while the mixture of VOCs has shown supreme biocontrol efficacy (71.54%) under greenhouse conditions compared to the other bioactive compounds of the same strain, including lipopeptides (29.57%) and crude secreted proteins (41.94%) [60]. Several VOCs produced by B. subtilis M29, including 2,6-diisocyanato-1methyl-benzene, 1-propoxy-2-propanol, and benzophenone, were reported to destroy normal hyphae morphology and induce mycelial fragmentation and crumpling in Botrytis cinerea [61]. Botrytis cinerea hyphae treated with B. velezensis XT1 and B. atrophaeus L193 volatiles produced in the MOLP medium showed severe cytoplasmic cavitation and vacuolation, and no organelles were identified, while the reduction of fungal growth was 27% and 46%, respectively [62].
Chaves-Lopez et al. [1] have reported changes in colony morphology, spore production, and microstructural changes of the hyphae in different fungal pathogens induced by the VOCs produced by B. subtilis, B. amyloliquefaciens, and B. cereus. The presence of VOCs from B. subtilis SV75-1 has significantly contributed to degenerative changes in the hyphal morphology of Moniliophthora perniciosa, including the observation of flaccid hyphae presenting retracted protoplasm and the formation of empty segments and a thinner wall. Retraction of protoplasm was also observed in Aspergillus parasiticus exposed to VOCs from B. cereus SV40, while exposure to B. amyloliquefaciens SV20-2 VOCs resulted in shortened and swollen somatic hyphae. Fusarium oxysporum f. sp. lactucae MA284 exposure to the VOCs of B. subtilis SV75-1 has resulted in a reduction in conidia number and granulation of the mycelia [1].
Malformations, vacuolations, and swellings were observed in hyphae of Verticillium dahliae, causing tomato Verticillium wilt as induced by B. velezensis C2 VOCs and resulting in an approximately 70% reduction in disease incidence [63]. The hyphae of Verticillium dahliae were completely lysed/dissolved after a 5-day treatment with styrene produced by Bacillus sp. T6 [64].
The VOCs produced by B. vallismortis 12a and B. altitudinis 14b seriously decomposed the cell walls and damaged the protoplast of Monilinia fructicola, causing peach brown rot, with a special emphasis on the outer cell walls, whose structure was thin or gapped, which might allow leaking out of the cell contents. Additionally, the result was 77.1% and 50% disease suppression in fumigation treatment with cultivation broth of these two strains, respectively [65]. Similar thin or gapped structures of the uneven cell wall surface presenting a retracted protoplasm were also observed in B. cereus CF4-51 VOCs-treated mycelia of Sclerotinia sclerotiorum [66]. Acetic acid (20.68%), propanoic acid (33.30%), butanoic acid (26.87%), valeric acid (43.71%), and isovaleric acid (53.10%) produced by Bacillus sp. LPPC170 significantly inhibited the mycelial growth of Fusarium kalimantanense, causing Panama disease of banana, by damaging the vegetative and reproductive structures of the fungus and causing a lower density of mycelium with dehydrated, considerably deformed, withered, and malnourished hyphae [67].
Hyphae with wrinkled surface cells, surface swelling, and thinner cell walls were observed as a consequence of 6-methyl-2-heptanone (emitted by B. subtilis ZD01) treatment in Alternaria solani. As a result, the cytoplasm was shrunken, with an increased number of inclusions and larger liquid droplets, together with cytoplasmic content movement toward the ruptured cell walls or cytoplasmic membranes [68]. Hyphae of Alternaria alternata (tobacco brown spot) without any attached conidia appeared ruptured, shrunken, and twisted after being treated with 2-methylbutanoic acid and 3-methylbutanoic acid produced by B. siamensis LZ88 [69]. The mycelium of Alternaria solani, the causal agent of tomato early blight, became thin, twisted, deformed, bifurcated, and fractured, with a wrinkled and cracked surface, followed by intracellular content leakage after the treatment with B. velezensis ZJ1 VOCs led by isooctanol and 2-nonanol as the major active compounds [70]. Similar results were found for Alternaria solani causing early potato blight when exposed to the VOCs produced by B. subtilis ZD01 [71], as well as for Mucor circinelloides, Fusarium arcuatisporum, Alternaria iridiaustralis, and Colletotrichum fioriniae due to 2,3-butanedione and 3-methylbutyric acid produced by B. subtilis CL2 [72]. Twisted, flattened, and enlarged hyphae that lost their linearity were observed in Alternaria iridiaustralis, caused by the VOCs produced by B. velezensis L1 with 2,3-butanedione as the leading VOC [73].
Twisted hyphae were also observed in Colletotrichum acutatum, Colletotrichum coccodes, Colletotrichum dematium, and Colletotrichum gloeosporioides after treatment with the VOC mixture produced by B. velezensis CE 100 [74]. VOCs of B. velezensis JRX-YG39 have caused hyphae abnormalities such as coiling and discoloration in Colletotrichum gloeosporioides (walnut and jujube anthracnose), while the pure compounds 5-nonylamine and 3-methylbutanoic acid have caused similar morphological changes followed by cell wall lysis and fracturing [75].
Hyphal swelling, cytoplasm and protoplasm aggregation, and distortion of large amounts of balloon-shaped cells were reported for Fusarium verticillioides, Fusarium graminearum, and Rhizoctonia solani after treatment with B. mojavensis I4, with mycelial growth inhibition in the range of 16-76% [76]. Hyphae of Phyllosticta citricarpa were fattened, twisted, and deformed, with a reduced amount of conidia at the lesion site on the surface of orange plants after treatment with the VOCs produced by Bacillus sp. ACB-65 and Bacillus sp. ACB-73 [31].
In Monilinia laxa and Monilinia fructicola, the hyphal membrane and cell walls were thinner and degraded, while the cytoplasmic content was completely coagulated and no organelles could be identified [62]. One of the main mechanisms of action related to the degradation of the structure of fungal cell walls and cell membrane relies on the modification of fatty acids and ergosterol content. Benzothiazole produced by B. subtilis CF-3 reduced the content of long-chain saturated fatty acids (trans-linoleic acid, oleic acid, and palmitic acid) in the Monilinia fructicola membrane, indicating decreased cell membrane unsaturation resulting in the gradual weakening of membrane fluidity and leakage of intracellular content. Furthermore, benzothiazole treatment also reduced ergosterol content in the cell wall of Monilinia fructicola, affecting the integrity of the fungal cell wall and thus weakening the cell membrane and material transport [77].
B. subtilis PPCB001 and B. amyloliquefaciens PPCB004 VOCs affected the germination and germ tube elongation of Penicillium crustosum [42]. Inhibition of spore germination and germ tube elongation were also observed in Penicillium digitatum and Penicillium italicum in the presence of the VOCs produced by B. amyloliquefaciens JBC36, which resulted in 57.8% and 54.1% in vitro inhibition of mycelial growth of Penicillium digitatum and Penicillium italicum, respectively, as well as a reduced incidence of green and blue mold on wounded mandarin fruits with control efficacies of 88% and 80.2%, respectively [78]. Inhibition of conidial germination of Thielaviopsis paradoxa ranged from 12% to 68.8% during the 9-h exposure to the VOCs of B. siamensis N-1 [79].

Prevention of Fungal Plant Attachment and Colonization by Bacillus VOCs
The VOCs of B. subtilis C9, with a special emphasis on DG4 (an isomer of acetyl butanediol) as an antifungal compound, have significantly reduced the incidence of stemsegment colonization by Rhizoctonia solani in Zoysia grass [48]. Sharifi and Ryu [84] have concluded that VOCs produced by B. subtilis GB03 might interfere with mycelial attachment to the hydrophobic cuticular surface of the Arabidopsis leaves, hence causing epiphytic mycelial growth of Botrytis cinerea and its inability to penetrate and colonize host tissue. Castro et al. [85] have made a relation between B. velezensis XT1 VOCs and the reduced number of Verticillium dahliae microsclerotia in the soil. B. velezensis OEE1 VOCs supported the reduction of Verticillium dahliae microsclerotia density in the naturally infested soil around olive trees [86]. The VOCs produced by B. amyloliquefaciens UQ154, B. velezensis UQ156, and Acinetobacter sp. UQ202 (isovaleraldehyde, 2-ethylhexanol, 2-heptanone, benzyl alcohol, and 3-methylbutanol) in a concentration of 10 µg/mL inhibited sporangia production and zoospore motility of Phytophthora capsici [87].

Altering the Expression of Genes Related to Pathogenicity, Metabolism and Antioxidant Activity of Fungal Pathogens
Zhang et al. [64] have reported the down-regulation of genes related to transport and catabolism, cell growth, and biosynthesis, especially peptidases, lipases, proteases, and chitinases, which act as plant cell wall-degrading enzymes, as well as methionyl-tRNA synthetases, in Verticillium dahliae, causing Verticillium cotton wilt, while the gene expression was modulated via styrene produced by Bacillus sp. T6.
Wang et al. [88] investigated differentially expressed genes and proteins in Colletotrichum gloeosporioides after treatment with B. subtilis CF-3 VOCs. The results revealed significant down-regulation of expression of genes related to cell membrane fluidity, cell wall integrity, energy metabolism, and production of cell wall-degrading enzymes, with a special emphasis on the biosynthesis of unsaturated fatty acids and ergosterol as the significant components of cell membranes, where 2,4-di-tert-butylphenol has been detected as the major antifungal VOC [88]. B. subtilis CF-3 VOC 2,4-di-tert-butylphenol can inhibit the activity of the pathogenic enzymes (pectinase and cellulase) secreted by Colletotrichum gloeosporioides to reduce the decomposition of plant tissues in litchi fruits [89], as well as benzothiazole with the same mechanism of antifungal activity against Monilinia fructicola in peaches [77]. The VOCs of Bacillus endophytes (B. velezensis VM11, B. velezensis VM10, and B. amyloliquefaciens VM42) induced strong ROS (reactive oxygen species) production in Sclerotinia sclerotiorum mycelial cells [57]. Down-regulation of the SOD gene, which plays a significant role in the SOD (superoxide dismutase) synthetic pathway in Alternaria solani, was found as a consequence of in vivo treatment by the VOCs produced by B. subtilis ZD01, suggesting the inhibition of the pathogen's antioxidant metabolism [71]. ROS can damage DNA replication and cell membranes, leading to cell death. In the study by Xie et al. [83], isopentyl acetate produced by B. subtilis DZSY21 caused ROS accumulation in the conidia of Curvularia lunata (maize leaf spot), while 2-methylbutyric acid and 2-heptanone did not affect ROS accumulation in conidia and mycelia. Here, it is suggested that these VOCs target different germination-related processes in Curvularia lunata conidia [83].

ISR Induced by Bacillus VOCs as Antifungal Mechanism of Action
One of the ways to induce resistance in plants against pathogens is physical inhibition of pathogen entrance to plant inner tissues by closing stomata. Acetoin and 2,3-butanediol as VOCs of B. amyloliquefaciens FZB42 modulated stomatal closure in Arabidopsis thaliana and Nicotiana benthamiana, where root absorption of VOCs was more effective than volatilization considering that VOC concentration of 250 µL was enough for stomata closure via root treatment, while 1 mM of VOCs was required for volatilization treatment. Furthermore, these VOCs have been successful in targeting pathogen entry points into hosts by triggering salicylic acid and abscisic acid signaling pathways and inducing the accumulation of hydrogen peroxide and nitric oxide, which are required for stomata closure [90]. Sharifi and Ryu [84] have found that the optimized concentration of B. subtilis GB03 that did not directly inhibit fungal growth of Botrytis cinerea successfully protected Arabidopsis from fungal infection, which indicates that ISR elicited by the bacterial VOCs has a more important role in biocontrol than direct inhibition of fungal growth on Arabidopsis plants. It was further confirmed that the ISR proportion was 90.63% and direct inhibition of fungi was 9.36% of the overall biocontrol activity, which means that ISR had the main role in suppressing Botrytis cinerea on Arabidopsis plants in conditions of low VOCs concentration. An almost three-fold increase in expression levels of plant defensin PDF1.2 indicated that the jasmonic acid signaling pathway had a key role in VOC-elicited plant defense responses, followed by the salicylic acid signaling pathway, as determined by the 2.8-fold increase in the expression level of PR1 (pathogenesis-related protein 1). On the other hand, the ethylene signaling pathway was not included in the Arabidopsis defense response since the expression level of the ChiB (basic endochitinase) gene did not display statistically significant differences between the VOCs-treated and control plants. This was also the first study to make a distinction between direct and indirect mechanisms of fungal pathogen suppression [84]. Pepper priming using the B. velezensis strains 5YN8 and DSN012 as VOC producers has resulted in more rapid transcription of the three pathogenicity-related genes (NPR1, PR1, and peroxidase gene), thus enhancing pepper resistance to Botrytis cinerea by activating the salicylic acid-mediated defense signaling pathway. In this way, gray mold biocontrol efficacy was above 50% in greenhouse experiments, with increased leaf number, stem diameter, and chlorophyll content in pepper seedlings [91]. Zheng et al. [38] have concluded that α-farnesene produced by several Bacillus isolates is probably related to ISR considering that it had not shown any antifungal activity in vitro against Peronophythora litchi while in vivo it had suppressed litchi downy blight with an efficacy of 52.34%.
B. siamensis LZ88 VOCs induced plant basal immunity through the induction of defense-related enzymes against Alternaria alternata, including peroxidase and polyphenol oxidase, thus contributing to the reduction of brown spots in tobacco leaves [92]. B. velezensis XT1 VOCs increased polyphenol oxidase activity by 395%, indicating induced resistance against Verticillium wilt of olive (Verticillium dahliae) in plant tissues, resulting in reduced disease severity in young olives by almost 80% [85]. Activation of antioxidant enzymes (peroxidase, polyphenol oxidase, catalase, and superoxide dismutase) in litchi fruit to eliminate excessive reactive oxygen species to reduce plant cell damage and activate disease resistance enzymes (phenylalanine ammonia-lyase, chitinases, β-1,3-glucanase) and enhance the resistance of litchi fruits to Colletotrichum gloeosporioides by inhibiting its growth was observed to be enhanced by 2,4-di-tert-butylphenol produced by B. subtilis CF-3 [89]. Similar mechanisms of action were observed for benzothiazole produced by the same Bacillus isolate in the suppression of Monilinia fructicola peach rot [77].

Inhibition of Fungal Pigments Production by Bacillus VOCs
Another interesting mechanism of action of Bacillus-produced VOCs is the inhibition of the production of different fungal pigments. Inhibition of pigment formation by the volatiles produced by B. subtilis G 8 [47], Paenibacillus polymyxa BMP-11, B. subtilis BL02, B. pumilus BSH-4, and B. pumilus ZB13 [56] was observed in Ascochyta citrullina, Alternaria solani, and Alternaria brassicae. The VOCs of B. amyloliquefaciens M49 inhibited the production of pink pigment by Fusarium oxysporum f. sp. lactucae [1]. B. velezensis SBB VOCs (2-nonanol, 2-heptanone, 6-methyl-2-heptanone, and 2-nonanone) inhibited melanin production by Verticillium dahliae [80]. Melanin plays an important role in providing the mechanical strength required for germ tubes, obtained by conidia germination, to penetrate host tissues [93]. The VOCs produced by B. subtilis DZSY21 caused inhibition of the expression of SCD and brn1 genes involved in the synthesis of melanin in Curvularia lunata (maize leaf spot), where the inhibitory effect of isopentyl acetate was higher than that of 2-heptanone [83].

Fungal Diseases Caused by Colletotrichum gloeosporioides
Zheng et al. [94] have reported successful inhibition of Colletotrichum gloeosporioides, the causal agent of mango anthracnose, in vitro (88.87% and 80.07%) and in vivo (94.28% and 87.06%) using the VOCs produced by B. pumilus TB09 and B. thuringiensis TB72, respectively. The main identified bioactive VOCs and their minimal inhibitory concentrations for mycelial growth inhibition were 2-nonanone, b-benzeneethanamine, 2-decanone, and 2-methylpyrazine in a concentration of 100 µL/L and thymol in a concentration of 50 µL/L. The disease incidence of mango fruit anthracnose, caused by Colletotrichum gloeosporioides and treated by B. siamensis N-1, was reduced by 44.6%, while the litchi fruit disease index and browning index were reduced by 57.8% and 82.3% through mediation by the VOCs [73].

Fungal Diseases Caused by Fusarium spp.
Different Bacillus strains produced VOCs that inhibited Fusarium solani, including B. amyloliquefaciens subsp. plantarum YAU B9601-Y2, Bacillus spp. 041, 285, 033, 355 and B. subtilis XF-1, which had shown exceptionally strong inhibitory activities (75-82% inhibition), as well as B. velezensis FZB42 and B. subtilis 168, whose VOCs demonstrated 67% and 56% inhibition, respectively [95]. B. subtilis IBFCBF-4 VOCs successfully inhibited the mycelial growth of Fusarium oxysporum f. sp. niveum in vitro by 47.9%, while the reduction of Fusarium wilt in watermelon in greenhouse experiments was 51.1% [59]. B. amyloliquefaciens L3 VOCs (2-nonanone and 2-heptanone) were found to suppress Fusarium wilt of watermelon (Fusarium oxysporum f. sp. niveum) in greenhouse pot experiments, while acetoin and 2,3-butanediol were responsible for watermelon plant growth promotion [96]. Greenhouse experiments confirmed the efficacy of B. mojavensis I4 VOCs in the suppression of Fusarium verticillioides, Fusarium graminearum, and Rhizoctonia solani in Arabidopsis thaliana plants, with significant improvements in plant growth, biomass production, and chlorophyll content [76]. The VOCs of B. cereus MH778713 (hentriacontane and 2,4-di-tertbutylphenol) reduced the disease severity of tomato Fusarium wilt (Fusarium oxysporum) from 88.1 ± 4.1% to only 23 ± 8.2%, followed by a several-fold increase in tomato root and shoot length as well as in the fresh and dry weight of plants in treatment with 50 µg of the bioactive VOCs [97]. In-planta assays showed that B. mycoides BM02 treatment mediated by phenylacetic acid and methylphenyl acetate reduced spore attachment and germination of Fusarium oxysporum f. sp. lycopersici and increased the formation of swollen hyphae, thus protecting tomato seedlings against Fusarium wilt [81].

Fungal Diseases Caused by Sclerotinia spp.
Shifa et al. [98] were the first to report the production of tridecane by B. subtilis G-1, which participated in the suppression of Sclerotium rolfsii, which causes stem rot or white mold of groundnut. Wu et al. [99] have identified toluene, phenol, and benzothiazole as the main VOCs produced by B. amyloliquefaciens NJZJSB3, which have shown comparable effects to chemical fungicides in terms of reduction of canola stem rot incidence (by 83.3%), caused by Sclerotinia sclerotiorum, in pot experiments. Southern blight of Aconitum carmichaelii Debx., caused by Sclerotium rolfsii, was successfully inhibited (30% with a longacting duration of up to 62 days) in a field trial by fermentation culture VOCs of B. subtilis JY-7-2L [100].

Fungal Diseases Caused by Monilinia spp.
Gotor-Vila et al. [30] have identified 1,3 pentadiene, acetoin (3-hydroxy-2-butanone), and thiophene as the main VOCs produced by B. amyloliquefaciens CPA-8, whereas thiophene was the most effective VOC in the suppression of Monilinia laxa, Monilinia fructicola, and Botrytis cinerea in vitro. On the other hand, during in vivo tests, it has not shown a biocontrol effect as a pure compound, while the mixture of VOCs decreased the disease incidence of cherry decay to 25% (compared to 65% in the control). The efficacy of VOCs produced by B. amyloliquefaciens SF14, B. amyloliquefaciens SP10, Alcaligenes faecalis ACBC1, and Pantoea agglomerans ACBP1 against Monilinia fructigena and Monilinia laxa causing apple brown spot was confirmed in a semi-commercial large-scale trial, with efficacy comparable to the commercial biocontrol agents (B. subtilis Y1336 and Pantoea agglomerans P10c) [101]. B. subtilis CF-3 VOCs combined with heat treatment could significantly reduce the rot index of peach and litchi fruit caused by Monilinia fructicola and Colletotrichum gloeosporioides, respectively, and effectively maintained fruit firmness and soluble solids content, reducing the fruit weight loss [102].

Fungal Diseases Caused by Botrytis cinerea
Dibutyl phthalate produced by B. velezensis JRX-YG39 has significantly reduced disease incidence in Arabidopsis thaliana caused by Botrytis cinerea up to 19.38% [103]. Zhang et al. [41] have reported 2,6-di-tert-butyl-4-methylphenol (BHT) and 2,4-di-tert-butylphenol (2,4-DTBP) produced by B. siamensis G-3 as biofumigants for controlling raspberry postharvest diseases caused by Botrytis cinerea and Rhizopus stolonifer in vivo, with biocontrol efficacy of 52.38% and 93.33%, respectively. This results in the possibility of storing rasp-berries for 20 days at 0 • C with the disease rate maintained at~10% after the 2,4-DTBP treatment. Diacetyl and benzaldehyde produced by B. velezensis strains BUZ-14, I3, and I5 have been reported as promising VOCs for active packaging during the postharvest commercialization of fruit. Furthermore, VOCs of B. velezensis I3 suppressed gray mold (Botrytis cinerea) in grapes by 50%, while the VOCs of B. velezensis BUZ-14 decreased brown rot severity in apricots (Monilinia fructicola) from 60 to 4 mm. Diacetyl was shown as suitable for biocontrol of gray mold with a concentration of only 0.02 mL/L and blue rot in mandarins at the same dose up to 60% [35].  [104]. In vivo assays on maize ears resulted in an 88% reduction in Aspergillus flavus growth and complete inhibition of fungal sporulation and aflatoxin accumulation by the VOCs produced by B. licheniformis BL350-2, dominated by 3-methyl-1-butanol [105].

Suppression of Other Fungal Diseases by Bacillus-Based VOCs
VOCs emitted by bacterial antagonists B. amyloliquefaciens UQ154, B. velezensis UQ156, and Acinetobacter sp. UQ202 negatively influenced the mycelial growth of the soil-borne phytopathogenic oomycete Phytophthora capsici by 35%, with significant positive effects on the increase in chili biomass (shoot and root fresh weights) and the primary root length, as well as the promotion of both root hair growth and lateral root development. The most important VOCs with antifungal activity were isovaleraldehyde, 2-ethylhexanol, 2-heptanone, benzyl alcohol, and 3-methylbutanol [87]. The leaf spot disease indexes of maize leaves sprayed with conidia of Curvularia lunata treated with the VOCs produced by B. subtilis DZSY21 were reduced from 60.52 to 26.64% [83]. In vivo application of 2,3-butanedione and 3-methylbutyric acid produced by B. subtilis CL2 significantly reduced the weight loss rate of wolfberry fruits caused by the pathogenic fungus Mucor circinelloides, as well as the decay incidence rate caused by Fusarium arcuatisporum, Alternaria iridiaustralis, and Colletotrichum fioriniae [72]. The VOCs produced by B. methylotrophicus BCN2 and B. thuringiensis BCN10 played complementary roles in controlling Fusarium oxysporum, Botryosphaeria sp., Trichoderma atroviride, Colletotrichum gloeosporioides, and Penicillium expansum, providing freshness to loquat fruits for ten days with a disease incidence of 20.19% compared to 54.17% in the control group [106]. The VOCs produced Bacillus sp. ACB-65 and Bacillus sp. ACB-73 cultured in TSB culture medium provided 86% inhibition of freckles that evolved into hard spots, with more superficial citrus black spot (Phyllosticta citricarpa) lesions in oranges [31]. The importance of the applied VOC concentration was pointed out by Zheng et al. [38], who discovered that 1-(2-aminophenyl)ethanone, benzothiazole, and α-farnesene had shown the highest efficacy in suppression of litchi downy blight (Peronophythora litchi) when applied in the lowest concentration (100 mg/L) from the examined concentration range (100-1000 mg/L). It suggested that VOCs in small concentrations could achieve the priming effect in plants, thus sensitizing plants for faster and/or stronger responses to successive pathogen attacks by acting as signaling molecules for environmental stresses.

Nematicidal Action of Bacillus-Based VOCs
Besides the antimicrobial activity against bacterial and fungal pathogens, representatives of the Bacillus genus are recognized as potential biocontrol agents effective in the suppression of plant-parasitic nematodes. The overview of published literature data focused on the nematicidal activity of Bacillus-based VOCs is summarized and examples of active VOCs are given in Table 3. The mode of action of Bacillus strains is defined through several approaches, including regulation of nematode behavior, competition for nutrients, and interference with nematode-host recognition [125]. Some of the examples reported in the scientific literature include B. firmus YBf-10, which shows nematicidal activity against the root-knot nematode Meloidogyne incognita, B. cereus C1-7, which inhibits root gall formation and reduces egg production of the carrot and tomato parasite Meloidogyne hapla [126]; and B. subtilis, which is active even under high temperatures, making it applicable in greenhouses as a biocontrol agent of root-knot nematode [125]. The previous studies also reported that secondary metabolites produced by Bacillus strains including B. megaterium, B. cereus, B. thuringiensis, and B. pumilus are effective against Meloidogyne exigua, Bursaphelenchus xylophilus, and Ditylenchus destructor. The production of volatile metabolites was also recognized as a potential mechanism of action for nematode suppression. B. nematocida B16 was one of the examples reported in the literature, where it was explained that it lures nematodes by producing seven VOCs that attract worms and afterward enter the intestine of the nematodes [126].

Styrene as the Nematicidal Bacillus-Based VOC
In the study by Luo et al. [126], a total of 45 members of the Bacillus genus were isolated from the root-knot nematode-infested tomato rhizosphere soil, and 5 of them were positive in terms of nematicidal activity with a mortality rate over 60%. The strain exhibiting the highest potential of nematicidal activity, causing a 98.1% mortality rate, was identified as B. mycoides R2. Styrene was defined as the primary nematicidal substance produced by the investigated strain, expressing high nematicidal activity against both free-living nematodes (Caenorhabditis elegans) and the root-knot nematode Meloidogyne incognita. In comparison with other nematicides, for example, chloronicotinyl insecticide thiacloprid for the nematode Meloidogyne incognita J2, with the LC50 of 36.2 mg/L in tomato crops, this value for styrene was 4.55 µg/mL [126]. The additional unique advantage for the control of nematodes with this VOC in the field lies in the fact that it fully evaporates in the soil and quickly and efficiently repels nematodes from crops [126].

Interference of Bacillus-Based VOCs with Nematodes' Chemotaxis
In another study, the potential nematicidal strain, B. subtilis Bs-1, was isolated from tomato rhizosphere and tested for in vitro nematicidal and ovicidal activities against Meloidogyne incognita, but the study also included efficiency evaluation in the pot experiments and in the field [127]. The results indicated high nematicidal activity, with an egg mortality rate of 100%, and a positive outcome even in field conditions. B. subtilis Bs-1 successfully reduced the disease index and stimulated the growth and yield of cucumber. Previous studies also confirmed that the B. subtilis isolate repelled second-stage juveniles (J2s) and reduced nematode production in tomatoes [127]. The major VOC produced by B. subtilis Bs-1 was CO 2 , and other identified active substances were acetic acid, 2-heptanone, pyrazine 2,5-dimethyl-and dimethyl disulfide, which were reported in previous studies as leading to chemotaxis in Meloidogyne incognita or Caenorhabditis elegans [128,129]. Considering the above-mentioned components' roles, it should be findings pointed out that Meloidogyne incognita was attracted by low concentrations of CO 2 and that acetic acid (0.1%) caused 100% mortality of Meloidogyne incognita. Organic sulfide also showed strong antimicrobial and insecticidal activities, and phenol, cyclohexanol, 2-octanol, and 2-undecanone expressed suppressive activity against nematodes [125]. The VOCs of another strain, B. cereus Bc-cm103, caused the mortality rates of Meloidogyne incognita J2s to be 90.8% and 97.2% after 24 h and 48 h of incubation, respectively. The identification of VOCs revealed the presence of 21 compounds, including alkanes, alkenes, esters, and sulfides, but the nematocidal activity was observed in the case of dimethyl disulfide (30.63%) and S-methyl ester butanethioic acid (30.29%) [130]. As it was suggested in previous studies, the mechanism of action included strong interference of VOCs with the chemotaxis of Meloidogyne incognita to cucumber roots.

Interference of Bacillus-Based VOCs with Nematodes' Antioxidant Metabolism
Ayaz et al. [131] conducted the experiments with Bacillus GBSC56 isolated from the Tibet region of China investigating the potential nematicidal activity against Meloidogyne incognita. The VOCs profile analysis revealed the presence of 10 compounds, while 3 of them, including dimethyl disulfide, methyl isovalerate, and 2-undecanone indicated strong nematicidal activity with mortality rates of 87%, 83%, and 80%, respectively. The activity of VOCs was based on severe oxidative stress caused by nematodes, resulting in rapid death. Additionally, the activity of antioxidant enzymes SOD, CAT, POD, and APX has enhanced in Meloidogyne incognita-infested roots in the presence of volatiles, which might reduce the adverse effect of oxidative stress induced after infection.

Specific Bacillus-Based VOCs Exhibiting Nematicidal Action
B. licheniformis JF-22 was another Bacillus strain recognized as a potential biocontrol agent active against Meloidogyne incognita. It was isolated from the tomato rhizosphere in the area where healthy plants were grown in the presence of the tomato root-knot nematode, and the study pointed out that acetoin, 2,3-butanediol, and hexamethylcyclotrisiloxane are the main components among the produced VOCs [132]. The VOCs produced by B. altitudinis AMCC 1040 were analyzed in the study by Ye et al. [133] and grouped into four major categories: ethers, alcohols, ketones, and organic acids. Out of eight compounds, six exhibited different levels of nematicidal activity, including 2,3-butanedione, acetic acid, 2-isopropoxy ethylamine, 3-methylbutyric acid, 2-methylbutyric acid, and octanoic acid. All four organic acids showed strong nematicidal activity, with octanoic acid showing the highest, followed by the activity of acetic acid as the second best, while two isomeric organic acids, 2-methyl-butanoic acid and 3-methyl-butanoic acid, had the same suppressive effect. Among the two ketones tested, only 2,3-butanedione showed activity [133].

Future Outlook on Bacillus-Based VOCs Research and Application
As presented in the previous sections, recent research on Bacillus-based VOCs has provided useful results in terms of (a) screening of Bacillus strains producing VOCs exhibiting antibacterial, antifungal, and nematicidal activity; (b) identification of the VOCs present in the volatile mixtures emitted by Bacillus spp.; (c) targeting the specific VOCs responsible for the antibacterial, antifungal, and nematicidal effects, as well as evaluation of the bacteriostatic/bactericidal and fungistatic/fungicidal concentrations of the specific VOCs; (d) in vitro, in vivo, and field testing of the specific VOCs/VOCs mixtures on the suppression of bacterial, fungal, and nematode plant pathogens. While the first studies in this field focused on identifying Bacillus strains producing VOCs with possible biocontrol applications, the research was later directed towards the precise identification of the produced VOCs and promotion of the ones responsible for biocontrol activity. Recent and currently ongoing studies are mostly focused on understanding the underlying mechanisms of antibacterial, antifungal, and nematicidal activity, both at the levels of cell structure and metabolism. However, a more profound understanding of the aforementioned mechanisms is required to maximize the potential of Bacillus-based VOCs biocontrol applications, leaving a significant space for further research in this area. Furthermore, in vivo testing procedures have been mostly based on model plants so far, requiring the widening of the host palette to increase the applicability and test the VOCs' biocontrol potential in realistic application conditions, with more studies needed to be performed in greenhouses and in the field, followed by an investigation of the VOCs' possible negative effects to the environment, animal-and human health. One of the possible research directions is also related to the biotechnological production of VOCs, directly affecting the mode of their biocontrol applications. Specific remarks related to all of the aforementioned aspects are given in the following subsections.

The Necessity to Investigate the Effects of Microbial Communities on Bacillus-VOCs Synthesis and Vice-Versa
Considering the previously mentioned fact that VOCs serve as signaling molecules for intra-and interspecies interactions as well as communication mediators across the kingdoms, it is necessary to better understand the exact mechanisms of microbial VOCsmediated communication, considering the complexity of microbial communities in different ecological niches. So far, it has been revealed that VOCs help microorganisms make distinctions between neighboring microorganisms (friend, foe, or prey) and adjust their behavior and performance (persist, invade, escape, or defend) accordingly [11,135]. Furthermore, it is necessary to understand the mechanisms underlying volatile emission and perception, since it has been very challenging to detect the signal senders, receivers, and putative eavesdroppers [10]. Biodiversity is a key driver of several ecosystem functions [136], hence the interactions among the ecosystem members affect the production and function of VOCs. Diverse examples of shifts in VOC production by bacteria in the presence of other organisms involved in different communication pathways could be found so far, although this field is still under-investigated. An increase or decrease of the produced VOCs' diversity and amount could be observed depending on the ecosystem community structure and the types of interactions, ranging from beneficial to antagonistic, resulting in the triggering/silencing of the VOCs' production (mostly mediated by quorum sensing). Beneficial interactions result in additive/complementary/facilitative effects on VOC production and could arise from the induction of certain VOC production in the presence of specific strains/their metabolites, possibly acting as molecular inducers or precursors for VOC synthesis. On the other hand, antagonistic interactions are usually related to direct antagonism as well as the necessity to provide a competitive advantage in terms of growth space and nutrients in a limited ecological niche [7]. The aforementioned interactions in complex microbial communities could result in the production of novel VOCs that were not detected in the respective monocultures, and consequently, in a higher diversity of VOCs with distinct biological functions observed separately and together. One of the reasons for that could be the horizontal acquisition of the genes for the synthesis of volatile secondary metabolites by bacteria [3], which should be further investigated. Hence, there is a necessity to increase the diversity of the investigated microbial communities and shift from pure monocultures to more complex systems that include (micro)organisms representative of certain applications of the investigated VOCs. Although the soil communities are the most complex in terms of diversity and the number of present species, it is also important to direct investigations to the plant phyllosphere as well as to endophytic microbial interactions resulting in VOC production.

The Necessity to Better Understand the VOCs' Mechanisms of Action against Broader Spectrum of Plant Pathogens and Hosts
Although a lot of literature data could be found regarding VOCs' effect in terms of antifungal activity, their effects against other types of plant pathogens are still not sufficiently investigated. Furthermore, the majority of the studies deal only with the confirmation of antifungal/antibacterial/nematicidal activity of the mixture of VOCs produced by the certain Bacillus strain, while a minority of the studies involve a precise investigation of the activity of separate VOCs from the emitted mixture. There could be observed significant differences in terms of biocontrol efficacy between VOC blends (with specific concentrations and ratios or naturally produced) and single components, both in terms of general efficacy as well as in terms of the range of pathogenic targets. Currently, <10% of mVOCs have been assigned a function [13]. Taking into account overlaps in the biological roles of many VOCs as well as the raising evidence on different antimicrobial activities of chiral VOCs/enantiomers, there is a necessity for complete chemical characterization and molecular targets' profiling of the VOCs in future investigations [10]. Furthermore, limited knowledge about the perception of microbial volatiles by other (micro)organisms is currently available. Therefore, further research should focus on mechanisms of VOC cell entry and cell targets, and here microbial mutants could offer a useful toolbox to detect genes responsible for the perception of VOCs, as well as their cellular/molecular targets [11]. In general, a greater understanding of VOCs' mechanisms of action across a broader range of pathogenic microorganisms is required, besides the necessity to include a wider range of plant hosts going behind the model plant species, considering that most of the current research refers to Arabidopsis thaliana and Nicotiana benthamiana [13].

Research Directions Related to VOCs Production by Bacillus spp.
VOCs' potential application is directly related to the selected VOC source-pure compounds, Bacillus-based captured VOCs, or the use of viable cells for on-site VOC production and application. Some of the VOCs produced by Bacillus strains could be chemically synthesized and used as pure compounds in different modes of application. On the other hand, VOCs produced by Bacillus spp. could be captured in a closed system and used for plant pathogen suppression at the same or a different location. Moreover, simultaneous production and application of Bacillus-based VOCs could be achieved, either by direct inoculation of biocontrol strains to plants or the surrounding soil or by maintaining cultures producing VOCs in plants' close vicinity, which would be the suitable mode of application in closed and controlled systems, such as greenhouses. Independently of the application mode, it is necessary to consider the main bioprocess variables affecting VOC production by Bacillus strains. On the upstream side of the bioprocess, the main bioprocess development step is the selection of suitable nutrients to be included in the solid or liquid cultivation medium, depending on the metabolic pathways required for the production of targeted VOCs and the substrates required in these metabolic processes, as well as optimization of the medium pH value as one of the most important physiology-affecting parameters. The next step is to optimize nutrients' ratios and concentrations to maximize target VOC production, as well as to decide between the utilization of solid or liquid medium, which significantly affects other bioprocess variables, such as mass and oxygen transfer. This leads us to the cultivation step itself, where it is necessary to determine the production system design, significantly depending on the previous selection between solid and liquid cultivation medium, and to optimize the cultivation parameters (inoculum preparation and inoculation strategy, medium volume, T, mixing and aeration rate, duration) to minimize resource utilization and achieve cost-effectiveness in terms of energy consumption. Furthermore, downstream and analytical methods require constant advancements to collect and detect the produced volatiles correctly, especially in cases where the collected volatiles are further used as biocontrol agents. Most of the currently available studies deal with VOCs produced by a single strain cultivated on artificial nutrient-rich media [137], not taking into account the real conditions where the VOCs will be applied, both in terms of the previously mentioned problems related to the diversity of natural microbial communities and environmental conditions. Hence, future research should focus on the imitation of environmental conditions at the VOC application site in terms of nutrient supply, physicochemical properties of the soil, and climate factors shaping the growth and development of the bacteria but also their volatile emission [3]. The factors that should be taken into account include soil aeration and limited oxygen availability in the rhizosphere, together with regional, seasonal, or climate dependencies and nutritional profiling in terms of the dependency of microbial growth and metabolism on root exudates. On the other hand, poor soil nutrient status is a raising trend in different geographic regions across the globe. Simulation of VOCs application environmental conditions should be implemented into the general bioprocessing strategy, starting from the cultivation medium whose composition strongly affects VOCs profile and amount and choice between solid and liquid media, mostly affecting oxygen availability [138]. Furthermore, different cultivation conditions (T, pH, mixing and aeration rates, duration) should be investigated to predict the VOC emission in a real-world application environment as well as to maximize their efficiency in terms of optimal diversity and concentration. Another important aspect for future investigations includes the monitoring and analysis of volatile secondary metabolites related to specific stages of bacterial growth and development, especially in terms of multicellular behaviors of Bacillus spp., including motility of freely available cells, sporulation, and biofilm formation [3].

VOCs Application-Related Remarks
Recent research of Bacillus-based VOCs has revealed varying differences in their effects from the lab to the field, mostly due to significant differences between in vitro production and testing conditions and real application conditions, including abiotic soil conditions (temperature, pH, moisture content, and soil texture), as well as the availability and quality of organic energy sources [137]. It points out the requirement for more field studies in terms of VOCs efficacy, which are currently scarce. Considering that the exact VOC concentration in the complex volatile mixture emitted by Bacillus spp. is usually not determined, it is very important to incorporate a wider concentration range in the testing protocols as many VOCs reveal dose-dependent biocontrol effects. The perception of a microbial VOC by other (micro)organisms could be related to its ratio/concentration in the complex volatile mixture, besides its presence. Therefore, this complex background should be incorporated as much as possible in the experimental setups designed to assess the biological activity of VOCs, facilitating the transition from laboratory research to real-life application conditions [10]. The incorporation of sensors for continuous monitoring of VOC concentration and big data computing analytics represent promising auxiliary tools for sustainable agricultural practice [2]. Bacillus-based VOCs could be applied, e.g., as repellents or biopesticides against plant diseases caused by microbial pathogens, soil fumigants, and fumigants in the management of post-harvest diseases. Their mode of use depends a lot on how far these VOCs travel before they cause a biological response in other organisms [11]. It opens a chapter of possibilities for VOCs delivery, including injection, dripping, drench versus spray application, companion cropping systems, etc. [13]. Utilization of different carriers for immobilization of Bacillus-producing VOCs, which could be supported by, e.g., 3D bioprinting, is worthy of investigation in future studies. One of the possible investigation routes includes the utilization of organic soil amendments that promote the production of VOCs during decomposition, where the necessity of the presence of VOC precursors in organic amendments should be further examined [137].

Possible Risks of Bacillus-Based VOCs Application in Biocontrol of Plant Diseases and Pathogens
Due to their biological origin, Bacillusand other microbially-derived VOCs are usually perceived as safe for the environment, plants, animals, and humans due to their lower (eco)toxicity and biodegradability. Some studies have observed the phytotoxic effects of higher VOC concentrations at different plant growth stages. Hence, there is a necessity to optimize VOC dosage, delivery, and time of treatment. Furthermore, when applying a mixture of VOCs emitted by a certain Bacillus strain present directly on the plant or in its close vicinity, it is necessary to investigate the strain's overall capacity for VOC production depending on the growth conditions, i.e., to predict the strain's volatile profile expected under certain application conditions to prevent the emergence of (phyto)toxic volatiles.
Moreover, it is necessary to predict the strain's volatile profile independently of the growth conditions by knowing and understanding the strain's genetic basis for VOC production, where whole genome sequencing technology could be useful together with databases of genomic data on Bacillus strains capable of VOC production to provide the basis for the selection of safe strains in terms of possible VOC release risks. Due to high evaporation rates and a lack of stability, VOC application may require high initial concentrations, raising the risk of dose-dependent toxicity against plants and other non-target organisms [137]. Dimethyl disulfide, produced by many Bacillus strains, is commercialized as one of the most potent microbial VOCs, exhibiting antifungal and nematicidal action [139,140]. However, recent studies have reported possible toxicity-related issues, including potential eye-and skin-irritable properties during acute exposure, while exposure to large concentrations may cause nausea, headache, dizziness, and irritation of the upper respiratory tract [141]. Therefore, besides the utilization of VOCs as pure compounds, which have been previously proven to be safe for humans and animals, the toxicity and non-target effects of the VOCs need to be documented before any field application, including effects on mycorrhizal development, non-target beneficial soil organisms, and a possible increase in other harmful soil organisms, to overcome the translational gap related to Bacillus-VOCs unexpected effects [10,11,137].