1. The Rhizosphere Habitat: A Significant Trait of Microbes with Multiple Biocontrol Activities
The rhizosphere is a dynamic space and key to the health and productivity of the plant. Roots and their exudates supply nutrients to damaging microbial plant pathogens and a foothold for invasion by pathogenic nematodes and insect larvae. However, the plant counters these effects by nurturing root colonization by beneficial microbes that shift the balance towards root and shoot health. Firmicutes and Gamma-proteobacteria, such as the pseudomonads, are among the most-studied bacteria that are such probiotics for the plant
[1]. Their consumption of simple metabolites, such as the sugars and amino acids in root exudates
[1][2], reduces the ability of root exudation to support the growth of pathogenic microbes, insect larvae, and nematodes. Further, the microbe’s metabolism increases the level of protectant secondary microbial products in the soil pore waters. These protective metabolites may require specific catabolic mechanisms in order to have persistence and to influence microbial community composition
[3]. One example by Stringlis et al.
[4] shows how coumarins in the root exudates favor root colonization by biocontrol-active microbes with a tolerance to coumarin, whereas this root metabolite is toxic to pathogen growth. The pools of biocontrol-active metabolites will function to strengthen the niche of the beneficial microbes, and additionally may trigger systemically protective changes in the plant to further thwart potential pathogens.
The formation of biofilms by the probiotic cells on root cell surfaces is integral to the protection process
[5]. The biofilm patches shield the root surface cells against direct attack and provide a sheltered environment for the cells of the beneficial bacterium. Additionally, the extracellular polymers forming the gel for the matrix encapsulating the cells could retain released microbial products, so that their concentrations are higher than in the soil pore water. When insect larvae and nematodes feed on the colonized plant root, they will ingest these biofilms that are loaded with toxic materials. Interestingly, both resorcinols and phenazines, identified as phenolics active in biocontrol, are correlated with improved biofilm formation
[5][6][7]. The biofilm would also reduce the ingress of pathogenesis factors such as enzymes and toxins from any plant pathogens, so that contact with root cells is lessened.
Studies with a
Paenibacillus polymyxa strain confirm that the matrix exopolysaccharides of the biofilm act as a rhizosphere nutrient source and as a structure that inhibits pathogen attack
[8]. Chan et al.
[9] indicate that reduced motility of nematodes within a
Pseudomonas aeruginosa biofilm impairs their predatory behavior. Earlier work with the biocontrol agent
Pseudomonas brassicacearum DF41 observed that biofilm formation over the head of the nematode contributes to the nematode’s demise
[10]. This group cites the activity as being part of the mechanism that limits predation of the beneficial bacterium, i.e., it is part of the survival features of the biocontrol agent.
Invasion and feeding on root cells are part of the life cycle of the nematodes and insect larvae that damage plants. Being able to kill these organisms would provide nutrients for the biocontrol organisms, thus augmenting the carbon-rich metabolites in the root exudates. A comparison of the root exudate composition shows that colonization of the roots by beneficial microbes alters the composition of amino acids
[11][12][13]. Some amino acids presumably supply essential N for microbial growth, which is limited in the soil pore water. For instance, Boiteau et al.
[12] found that the concentration of two N-containing amino acids in root exudates from
Brachypodium decreases after colonization by
Pseudomonas fluorescens. Microbes that are rhizosphere colonists have higher numbers of amino acid and sugar transporters than non-colonists
[11]. Consequently, the presence of genes enabling the biocontrol agents to kill protein-rich larvae would be advantageous for bacterial multiplication, because of the rich supply of complex N-nutrients from the insect’s body mass
[14][15]. Similarly, enzymatic digestion through proteases of the proteinaceous cuticle of nematodes or the chitin of their eggs would increase the N supply to the bacteria
[16].
Infections with nematodes or insect larvae would result in biofilms on these hosts and the production of the biocontrol-active metabolites and enzymes, providing “hot spot” reservoirs additional to those at the plant root. Indeed, infection with insect larvae by biocontrol-active pseudomonads induces the expression of genes supporting antifungal metabolite production, boosting the value of such pockets of biocontrol microbes within the soil
[17]. Further, any movement by the infected larvae and nematodes would aid in spreading the biocontrol agent. The concept that biocontrol microbes in the rhizosphere are spread by contact with and pathogenicity on insects is discussed well in a research by Pronk et al.
[18]. Thus, the infectious ability on the soil fauna boosts the sphere of influence of the probiotics further from the root surface
[1]. These potential roles by the biocontrol agents of the biofilm in plant protection add to the need to formulate viable cell preparations that are effective when introduced into agricultural settings.
2. Active Products from Bacillus spp.
Four classes of protectants essential in biocontrol by
Bacillus spp. are peptide toxins, lipopeptides, enzymes, and volatile organic compounds. The most effective of these products are the insecticidal BT toxins, which have been successfully commercialized. The toxins are produced as sporulation of the
Bacillus isolates commences and become concentrated as an inactive crystal structure within bacterial cells
[19]. After ingestion of the crystals by the insect larvae, conversion to an active peptide dependent on the action of a larval protease occurs. The recognition of the active peptide by specific receptors on insect gut cells causes pore formation and loss of selective permeability. The efficacy of digestion is impaired, and the bacteria enter the circulatory systems to eventually kill the larvae. The specificity of the receptors in the gut cell membranes leads to products that target certain insects. The toxins vary in structure and activity; BT from
B. thuringiensis shows an LC50 of 0.28 ppm against the second instar larvae of the diamondback moth
Plutella xylostella, a value lower than the LC50 for the toxin from
Bt subsp.
kurstaki HD1 (LC50 0.47 ppm)
[20]. However, BT toxins are not the only control mode, because additional toxic peptides can be formed even in vegetative cells
[21]. Understanding the additive or synergistic effects of the array of metabolites in biocontrol requires further studies.
Lipopeptides are metabolites commonly secreted by
B. thuringiensis and other
Bacillus species
[22]. Many lipopeptides have potent activity against insects and are also antifungal. For example, the lipopeptide from
B. thuringiensis CMB26 has insecticidal activity on the larvae of the cabbage white butterfly,
Pieris rapae crucivora, and inhibits the growth of the pathogenic fungus,
Colletotrichum gloeosporioides [23]. A single
Bacillus spp. strain may produce several lipopeptides, each with a different potential against plant fungal pathogens. Two lipopeptides produced by
B. subtilis EA-CB0015, iturin A and fengycin C, inhibit the anthracnose pathogen
Colletotrichum acutatum, with a minimum inhibitory concentration (MIC) at 32 ppm and 124 ppm, respectively
[24]. Other
Bacillus species also produce iturin A, and this biocide also disrupts the membranes in cells of the fungal targets
[25]. Studies of iturin from
B. methylotrophicus TEB1 show a MIC of 100 ppm against
Phoma tracheiphila [26], whereas iturin from
B. amyloliquefaciens MG3 inhibits mycelial growth of
C. gloeosporioides at concentrations less than 50 ppm but is ineffective in restricting spore germination
[25]. These examples indicate that timing is crucial for fungal pathogen control on the plant, as it will depend on whether the spores or mycelial growth of a pathogen is targeted. Another lipopeptide from
B. amyloliquefaciens BO5A inhibits mycelia of
F. oxysporum at 10 ppm although there is no effect on the mycelium of
Botrytis cinerea at 100 ppm
[27].
Regarding insecticidal activity, a lipopeptide from
B. amyloliquefaciens AG1 has an LC50 of 180 ng/cm
2 against larvae of the leaf miner,
Tuta absoluta, targeting the membranes in the larval midgut cells
[28]. These same larvae are controlled at 278 ng/cm
2 by the lipopeptide from
B. subtilis V26, which is also an antifungal that halts mycelial growth of the gray mold fungus,
B.
cinerea, at 2 ppm
[29].
The surfactin lipopolypeptide from
B. subtilis Y9 shows aphicidal activity against the green peach aphid
Myzus persicae at 20 ppm
[30]. Similarly, active surfactant lipopeptides from
Bacillus atrophaeus L193 cause the destruction of the cuticle of the aphid
Rhopalosiphum padi [31].
B. atrophaeus L193 also controls fungal diseases caused by
B. cinerea and
Monilinia laxa in cherry fruits by producing 2,3,-butanediol, a weakly volatile metabolite
[32]. Butanediol production is also observed in other plant-associated biocontrol bacteria
[33], again showing the sharing of beneficial traits.
Other dual biocontrol lipopeptides synthesized by
B. subtilis SPB1 kill larvae of the cotton leafworm
Spodoptera littoralis, initially through the disruption of midgut function
[34]. The lipopeptide also effectively controls two fungal pathogens of potato and tomato,
Fusarium solani and
Rhizoctonia solani, but at high MICs of 3000 ppm and 4000 ppm, respectively, compared with an MIC of 40 ppm for
Rhizoctonia bataticola [35]. Of note is the finding that the
B. subtilis SPB1 lipopeptides have preventative and curative effects
[36]. These findings illustrate the strong potential for lipopeptides to combat plant loss due to insects and fungal pathogens, however, the sensitivities between targets are different.
Although the lipopeptide from
B. subtilis PTB185 effectively controls the gray mold fungus
B. cinerea, its chitinase is the major factor combating the aphids
Aulacorthum solani and
Aphis gossypii [37][38]. Chitinase of
B. firmus also is highlighted as a seed treatment agent to control nematodes
[39]. Several papers discuss the multiple roles of chitinase regarding its antifungal, insecticidal, and nematicidal activities
[40][41]. However, chitinase is not implicated in the control of the nematode
Meloidogyne incognita by a crude supernatant from
B.
firmus YBf-10. The efficacy in pot cultures is comparable to that of the chemical nematicide fenamiphos
[42]. In this case, a serine protease and chemical metabolites seem to be functional in the nematicidal response
[43].
Bacillus strains also produce biocontrol-active volatile organic compounds (VOCs). Two VOCs generated by
B. velezensis CT32, 2,4-dimethyl-6-tert-butylphenol and benzothiazole, have strong antifungal activity against wilt in strawberries caused by
Verticillium dahliae and
F.
oxysporum [44]. These VOCs are documented in the headspace when the bacillus is grown on nutrient agar. Unidentified VOCs produced by another
B. velezensis strain, isolate VN10, inhibit mycelial growth of
Sclerotinia sclerotium with high efficacy (MIC 1 ppm) and reduce the disease caused by this fungus through suppression of its production of oxalic acid
[45].
3. Paenibacillus spp. Metabolites and Enzymes
The biocontrol arsenal of
Paenibacillus elgii HOA73 displays the common finding that a multiplicity of metabolites formed by a single isolate is important in biocontrol. Hydrolytic enzymes (chitinase, protease, and gelatinase), along with the phenolic compounds benzothiazole, methyl 2,3-dihydroxybenzoate, protocatechuic acid, and the volatile 1-octen-3-ol, are among its biocontrol weapons. The phenolic compounds display promising MIC values for fungal pathogen control. Methyl 2,3-dihydroxybenzoate inhibits mycelial growth of
B. cinerea,
R.
solani, and
F. oxysporum f. sp.
lycopersici at 32–64 ppm
[46], and the MIC for protocatechuic acid is similar for
B. cinerea and
R.
solani at 64 ppm. At 100 ppm, protocatechuic acid halts the disease progress of gray mold on strawberry fruits
[47]. The lipopeptides from
Paenibacillus strains also participate in biocontrol. A lipopeptide from
P. polymyxa shows antimicrobial and anti-insect efficacy
[8].
Active enzymes include chitinase, which inhibits spore germination of
Cladosporium tenuissimum and
B.
cinerea at 100 ppm, whereas spores of
Fulvia fulva and
C. gloeosporioides are resilient
[48]. The
P. elgii HOA73 strain exhibits insecticidal and nematicidal performance correlating with chitinase and gelatinase (a type of protease) activities; mortality of
M.
incognita J2 juveniles and a reduction in egg mass and galling in tomatoes occur at 50–400 ppm. Other factors may participate in the nematicidal effects and the killing of second instar larvae of the diamondback moth,
P. xylostella. Combining organic sulfur with the active enzymes acts synergistically
[49]. It will be interesting to see how biocontrol microbes and their products can be combined with other tools to achieve better control.
VOCs from
P. polymyxa BMP-11 are implicated in biocontrol. Two VOCs, 1-octen-3-ol (LC50, 16.75 ppm) and benzothiazole (LC50, 3.5 ppm), are active against the adult red flour beetle
Tribolium castaneum [50]. In vitro work with authentic 1-octen-3-ol shows potent inhibition of bacteria and fungi at 1–2 ppm
[51]. At 100 ppm, 1-octen-3-ol effectively inhibits mycelial growth of the brown rot fungus,
Monilinia fructicola, but fruit treatment requires about 50 ppm to slow disease progress. Fumigation with 1-octen-3-ol (LD50, 27.7 mL/L air) reduces the effects of the maize weevil
Sitophilus zeamais and the production of mycotoxin by the fungus
Fusarium verticillioides (MIC 81.5 mL/L air)
[52]. Citronellol, also produced by
P.
polymyxa strain BMP-11, inhibits
F. oxysporum. Additionally, VOCs from
P.
polymyxa KM2501-1 can kill second-stage juveniles of
M.
incognita through mechanisms that include acting as a repellent and a fumigant active on contact
[53].
These examples suggest that certain VOCs can be applied as fumigants. However, boosting the plant-associated bacteria in agricultural soils by inoculation would also promote beneficial volatile production in the rhizosphere. These methods could be valuable when raising crops in enclosed spaces, such as in greenhouse cultivation.
4. Pseudomonad Products
Among the phenolic-based products with biocontrol activity, phenazines are well-known for their potential as antifungal compounds
[54][55]. In general, phenazine-1-carboxylic-acid (PCA) has higher antifungal activity than phenazine-1-carboxamide (PCN) or 1-hydroxyphenazine (1HP)
[56][57]. Phenazines of different structures can be produced simultaneously by pseudomonads, dependent on their gene pools
[58].
The commercial product Shenqinmycin contains PCA as the major metabolite and has high efficacy in the field against
Phoma infections
[59]. The PCA produced by
P.
fluorescens 2–79 limits the mycelial growth of several plant pathogens;
Cochliobolus sativus,
G. graminis var.
tritici, and
R.
solani are inhibited with MICs of 1 ppm, but
Fusarium spp. requires 25–30 ppm. Species-dependent sensitivity is seen in the oomycete
Pythium, with MIC values for PCA of 1 ppm for
Pythium aristosporum, 25–30 ppm for
Pythium ultimate, and 80–100 ppm for
Pythium ultimum var.
sporangiifurum [60].
Although mycelial exopolysaccharide production from
B. cinerea is inhibited at 3 ppm PCA, the control of postharvest gray mold caused by this fungus requires 25 ppm
[61]. The PCA synthesized by
P.
aeruginosa GC-B26 is active against
Colletotrichum orbiculare as well as the oomycetes
Pythium capsici (MIC 5 ppm) and
P.
ultimum (MIC 5 ppm)
[62]. Studies with PCA from
P. aeruginosa report activities against
Sclerotium rolfsii (MIC 29 ppm),
F.
oxysporum (MIC 40 ppm), and
Colletotrichum falcatum (MIC 50 ppm)
[63]. The effective dose of PCA is often similar to the doses required for commercial pesticides, for example, for protection against Phytophthora blight on pepper and anthracnose on cucumber
[62].
The control of tomato foot and root rot, caused by
F. oxysporum sp.
radices-lycopersici, with
P. chlororaphis PCL1391 uses PCN as the major phenazine
[64]. As for PCA, effective doses of PCN are similar to those of chemical products. Extracted PCN, produced by
P.
aeruginosa MML2212, effectively controls rice sheath blight caused by
R.
solani at 5 ppm, exceeding protection from the chemical fungicide carbendazim. Control of rice bacterial leaf blight,
Xanthomonas oryzae pv.
oryzae, by 5 ppm PCN is similar to that by the pesticide refamycin
[65], and control of
B. cinerea by PCN from
P.
aeruginosa at 108 ppm has almost the same inhibition rate as the chemical fungicide carbendazim. The effective dosages for PCN and chemical fungicides are higher for effects on spore germination and mycelial growth of
Sclerotinia, requiring about 700 ppm for both PCN and carbendazim
[66].
In contrast to their high efficacy for inhibiting various plant fungal pathogens, phenazines are not as efficient in suppressing insect or nematode larvae. They may impair egg hatching and promote increased J2 larval mortality of the root-knot nematode
M.
incognita [67][68]. Among the secondary metabolites produced by
P. chlororaphis strain PA23, hydrogen cyanide (HCN) and the phenolic-based pyrrolnitrin (PRN) are cited as active agents in killing the nematode
Caenorhabditis elegans, for which they also act as repellents. Growing
P. chlororaphis strain PA23 with a nematode as the food source enhances the production of HCN and PRN
[69]. However, mutational analysis of
P.
chlororaphis PA23 reveals that, among the secondary metabolites produced by this bacterium, pyrrolnitrin is the major metabolite involved in inhibiting
S.
sclerotiorum [6]. Studies with root-knot nematode juveniles found that HCN production is essential for the nematicidal effects of
P. chlororaphis O6
[70][71].
Induced mortality of the root-knot nematode by
P. fluorescens CHA0 requires another suite of products involving a protease and two phenolic-based structures, pyoluteorin and 2,4-diacetylphloroglucinol (2,4-DAPG), each with demonstrated antifungal activities
[72][73]. The production of 2,4-DAPG by
P. fluorescens CHA0 also suppresses populations of another nematode,
Meloidogyne javanica [72], while 2,4-DAPG from
P. fluorescens is toxic to
Xiphinema americanum adults with an LC50 of 8.3 ppm. Apparently, 2,4-DAPG does not harm beneficial entomopathogenic nematodes
[74].
The antifungal activity of 2,4-DAPG from pseudomonad isolates is well documented. It is effective at 50 ppm against the citrus postharvest fungi
Penicillium digitatum and
Penicillium italicum, a dose that compares favorably to 600 ppm for the chemical fungicide albesilate
[75]. Production of 2,4-DAPG by
P. fluorescens CHA0 is a major factor in suppressing the soil-borne disease black root rot in tobacco, caused by
Thielaviopsis basicola, and against
G. graminis var.
tritici in wheat
[76].
Other metabolites play roles in the insecticidal activities displayed by the multi-biocontrol pseudomonads
[77][78]. Flury et al.
[17] proposed that a mixture of the cyclic lipopeptide orfamide A and volatile HCN contributes to the insecticidal activity of
P. fluorescens CHA0 and other pseudomonads
[17]. Interestingly, the growth of
P. fluorescens CHA0 on insect larvae promotes the expression of genes involved in producing an array of antifungal metabolites from this strain
[17]. This finding suggests the to this rich nutrition of the bacterium helps its arsenal of metabolites provide security against competition with invasive microbes in the rhizosphere and bulk soil. Even though HCN is produced only at low levels by
P. protegens Pf-5, the synthesis of analogs of rhizoxin adds to the power of orfamide A and a chitinase in the oral toxicity of this isolate to the larvae of the common fruit fly,
Drosophila melanogaster [79].
Proteins termed Fit toxins are another part of the pathogenic mechanism of
P. protegens and
P. chlororaphis isolates on lepidopteran larvae
[77][78][80]. Ruffner et al.
[81] discussed how the
fit gene clusters in the pseudomonads are related to those in two insect pathogens,
Photorhabdus and
Xenorhabdus, suggesting horizontal gene transfer of the clusters into limited pseudomonad genomes.
Hydrogen cyanide, HCN, is an important biocontrol volatile from the pseudomonads and other rhizobacteria including
Bacillus isolates
[82][83][84], contributing to a broad spectrum of inhibitory activities against microbes, insects, and nematodes
[17][85]. Root exudates may exogenously supply the glycine that the microbes use as the substrate for HCN synthase; carbon dioxide is the other product of the synthase activity
[85]. Production of HCN is observed in the airspace of closed growth boxes supporting plants with roots colonized by
P. chlororaphis O6
[71][85]. Dependent on concentration, HCN may be toxic due to the disruption of cellular function through the inhibition of heme group function
[86]. Indeed, root colonization and HCN production by certain isolates leads to herbicidal activity
[87]. However, lower doses of HCN may also be important through synergistic interaction with other antimicrobial metabolites, such as lipopeptides or pyrrolnitrin, in enhancing the biocontrol effects.
The inhibition of fungal growth of the foliar plant pathogens
Septoria tritici and
Puccinia recondita f. sp.
tritici on wheat is attributed to HCN production by
P. putida BK8661
[88]. Furthermore, cyanide-producing pseudomonads suppress fungal diseases on canola and rice
[89][90]. Mutation of
P. fluorescens CHA0 revealed that HCN is a major player in controlling black root rot of tobacco, caused by the fungus
Thielaviopsis basicola [91][92]. The combination of HCN and 2,4-DAPG syntheses by
Pseudomonas sp. LBUM300 protects tomatoes against bacterial canker caused by
Clavibacter michiganensis subsp.
michiganensis [93].
Insecticidal effects of HCN production have been revealed by the observed mortality of second instar nymphs of the green peach aphid
Myzus persicae when exposed to volatiles from
P. chlororaphis O6
[94]. The study, and others with different pseudomonad isolates, showed that HCN production is controlled by quorum sensing
[94]. Other studies extend the insecticidal effects of HCN produced by pseudomonads to flies and termites
[95][96][97]. Combining HCN with lipopeptides of
P. protegens CHA0 and
P. chlororaphis PCL1391 shows activity against insects
[17]. The nematicidal activities of HCN include killing of the nematode
Caenorhabditis elegans by
P.
chlororaphis PA23, in combination with pyrrolnitrin
[90]. A primary role for HCN is implicated in the mortality of J2 juveniles of the root-knot nematode, where for tomato, the efficacy of the biocontrol agent was as strong as that of the chemical nematicide fosthiazate in commercial greenhouses
[70][71].
These studies illustrate that the biocontrol-active pseudomonads, like the firmicutes, strongly influence rhizosphere health, their strength being dependent upon a multiplicity of potential control mechanisms incited by different metabolites.
5. Streptomyces, Brevibacillus, Serratia, and Photorhabdus
As discussed above, secondary products from the firmicutes and pseudomonads produce many different structures with multiple biocontrol activities. Only a few highlights are provided to illustrate the expansive diversity in structure and function. The Gram-positive
Streptomyces hydrogenans DH16 secretes metabolites that kill
M.
incognita J2 juveniles at 100 ppm
[98], inhibit fungal pathogens, and induce mortality to second instar larvae of
Spodoptera litura [98]. A chitinase produced by
Brevibacillus laterosporus Lak1210 has dual performance against larvae of the diamondback moth
P. xylostella and the pathogenic fungi
Fusarium equiseti [99].
For other Gram-negative isolates, the volatile benzaldehyde produced by
Photorhabdus temperata M1021 contributes to the antimicrobial effects on
Phytophthora capsici,
R. solani,
Corynespora cassiicola, and
Bacillus spp., as well as insecticidal activity on the larvae of the greater wax moth
Galleria mellonella [100].
Serratia species are a source of many deterrents for plant protection and represent an understudied resource for soil and plant associations
[101]. Ordentlich et al.
[102] reported that chitinase from
Serratia marcescens is key for controlling
Sclerotium rolfsii; direct lysis of mycelia of the pathogen was observed.
Serratia entomophila AB2 shows a dual effect on pathogenic fungi and the pod borer
Heliothis armigera [103]. Haterumalides, novel and complex metabolites from
Serratia plymuthica A153, suppress the apothecial formation in sclerotia of
Sclerotium with MIC of 0.5 ppm; additionally, oomycetes also are sensitive to these metabolites
[104][105].
Serratia isolates are amongst the several bacterial genera that synthesize pyrrolnitrin, previously. In combination, pyrrolnitrin and the haterumalides inhibit spore germination of various fungal plant pathogens at MICs of 0.4–50 ppm
[104]. The array of volatiles synthesized from
Serratia isolates also offers biological control potential
[106]. Recent work on the plant-growth-promoting isolate of
S. plymuthica A153 correlates impaired growth of several plant fungal pathogens with the volatiles for which ammonia is identified as an important component
[106]. Activation of the WRKY18-plant stress pathway was noted with volatiles of
Serratia [107].
These examples illustrate that the soil contains a diversity of microbial genera that can contribute to plant health by affecting the growth and metabolism of potentially harmful microbe, nematode, and insect challenges. The chosen examples highlight that many soil isolates thrive in the rhizosphere partly due to the carbon and nitrogen nutrition from plant metabolites in the root exudates.