Use of Plant Growth-Promoting Bacteria for Biocontrol of Plant Diseases: Principles, Mechanisms of Action, and Future Prospects

Pathogenic microorganisms affecting plant health are a major and chronic threat to food production and ecosystem stability worldwide. As agricultural production intensified over the past few decades, producers became more and more dependent on agrochemicals as a relatively reliable method of crop protection helping with economic stability of their operations. However, increasing use of chemical inputs causes several negative effects, i.e., development of pathogen resistance to the applied agents and their nontarget environmental impacts (44, 62). Furthermore, the growing cost of pesticides, particularly in less-affluent regions of the world, and consumer demand for pesticide-free food has led to a search for substitutes for these products. There are also a number of fastidious diseases for which chemical solutions are few, ineffective, or nonexistent (62). Biological control is thus being considered as an alternative or a supplemental way of reducing the use of chemicals in agriculture (44, 62, 136, 188).

There has been a large body of literature describing potential uses of plant associated bacteria as agents stimulating plant growth and managing soil and plant health (reviewed in references 63, 70, 143, 165, and 188). Plant growth-promoting bacteria (PGPB) (8) are associated with many, if not all, plant species and are commonly present in many environments. The most widely studied group of PGPB are plant growth-promoting rhizobacteria (PGPR) (82) colonizing the root surfaces and the closely adhering soil interface, the rhizosphere (82, 84). As reviewed by Kloepper et al. (84) or, more recently, by Gray and Smith (65), some of these PGPR can also enter root interior and establish endophytic populations. Many of them are able to transcend the endodermis barrier, crossing from the root cortex to the vascular system, and subsequently thrive as endophytes in stem, leaves, tubers, and other organs (10, 28, 65, 70). The extent of endophytic colonization of host plant organs and tissues reflects the ability of bacteria to selectively adapt to these specific ecological niches (65, 70). Consequently, intimate associations between bacteria and host plants can be formed (28, 70, 84) without harming the plant (70, 83, 84, 92, 191). Although, it is generally assumed that many bacterial endophyte communities are the product of a colonizing process initiated in the root zone (102, 165, 177, 188), they may also originate from other source than the rhizosphere, such as the phyllosphere, the anthosphere, or the spermosphere (70).

Despite their different ecological niches, free-living rhizobacteria and endophytic bacteria use some of the same mechanisms to promote plant growth and control phytopathogens (15, 46, 63, 70, 92, 165). The widely recognized mechanisms of biocontrol mediated by PGPB are competition for an ecological niche or a substrate, production of inhibitory allelochemicals, and induction of systemic resistance (ISR) in host plants to a broad spectrum of pathogens (15, 63, 66, 67, 97, 146) and/or abiotic stresses (reviewed in references 101 and 117). This review surveys the advances of plant-PGPB interaction research focusing on the principles and mechanisms of action of PGPB, both free-living and endophytic bacteria, and their use or potential use for the biological control of plant diseases.

COMPETITIVE ROOT COLONIZATION

Despite their potential as low-input practical agents of plant protection, application of PGPB has been hampered by inconsistent performance in field tests (167); this is usually attributed to their poor rhizosphere competence (153, 189). Rhizosphere competence of biocontrol agents comprises effective root colonization combined with the ability to survive and proliferate along growing plant roots over a considerable time period, in the presence of the indigenous microflora (95, 127, 189, 190). Given the importance of rhizosphere competence as a prerequisite of effective biological control, understanding root-microbe communication (6, 135), as affected by genetic (80, 118) and environmental (128) determinants in spatial (6) and temporal (23) contexts, will significantly contribute to improve the efficacy of these biocontrol agents.

Competition for root niches and bacterial determinants directly involves root colonization.

The root surface and surrounding rhizosphere are significant carbon sinks (143). Photosynthate allocation to this zone can be as high as 40% (34). Thus, along root surfaces there are various suitable nutrient-rich niches attracting a great diversity of microorganisms, including phytopathogens. Competition for these nutrients and niches is a fundamental mechanism by which PGPB protect plants from phytopathogens (50). PGPB reach root surfaces by active motility facilitated by flagella and are guided by chemotactic responses (41, 42, 112, 162, 171, 172). Known chemical attractants present in root exudates include organic acids, amino acids, and specific sugars (188). Some exudates can also be effective as antimicrobial agents and thus give ecological niche advantage to organisms that have adequate enzymatic machinery to detoxify them (reviewed in reference 6). The quantity and composition of chemoattractants and antimicrobials exuded by plant roots are under genetic and environmental control (6). This implies that PGPB competence highly depends either on their abilities to take advantage of a specific environment or on their abilities to adapt to changing conditions. As an example, Azospirillum chemotaxis is induced by sugars, amino acids, and organic acids, but the degree of chemotactic response to each of these compounds differs among strains (142). PGPB may be uniquely equipped to sense chemoattractants, e.g., rice exudates induce stronger chemotactic responses of endophytic bacteria than from non-PGPB present in the rice rhizosphere (5).

Bacterial lipopolysaccharides (LPS), in particular the O-antigen chain, can also contribute to root colonization (35). However, the importance of LPS in this colonization might be strain dependent since the O-antigenic side chain of Pseudomonas fluorescens WCS374 does not contribute to potato root adhesion (43), whereas the O-antigen chain of P. fluorescens PCL1205 is involved in tomato root colonization (35). Furthermore, the O-antigenic aspect of LPS does not contribute to rhizoplane colonization of tomato by the plant beneficial endophytic bacterium P. fluorescens WCS417r but, interestingly, this bacterial determinant was involved in endophytic colonization of roots (57).

It has also been recently demonstrated that the high bacterial growth rate and ability to synthesize vitamin B₁ and exude NADH dehydrogenases contribute to plant colonization by PGPB (35, 157). Another determinant of root colonization ability by bacteria is type IV pili, better known for its involvement in the adhesion of animal and human pathogenic bacteria to eukaryotic cells (69, 162, 163). The type IV pili also play a role in plant colonization by endophytic bacteria such as Azoarcus sp. (49, 162).

Root colonization and site-specific recombinase.

Bacterial traits required for effective root colonization are subject to phase variation, a regulatory process for DNA rearrangements orchestrated by site-specific recombinase (36, 149, 174). In certain PGPB, efficient root colonization is linked to their ability to secrete a site-specific recombinase (36). Transfer of the site-specific recombinase gene from a rhizosphere-competent P. fluorescens into a rhizosphere-incompetent Pseudomonas strain enhanced its ability to colonize root tips (37).

Utilization of root exudates and root mucilage by PGPB.

Since root exudates are the primary source of nutrients for rhizosphere microorganisms (143, 176), rhizosphere competence implies that PGPB are well adapted to their utilization (96). Despite the fact that sugars have often been reported as the major carbon source in exudates, the ability to use specific sugars does not play a major role in tomato root colonization (96). Similarly, although amino acids are present in root exudates, the bioavailability of amino acids alone is considered insufficient to support root tip colonization by auxotrophic mutants of P. fluorescens WCS365 (158). In contrast, Simons et al. (158) reported that amino acid synthesis is required for root colonization by P. fluorescens WCS365, indicating that amino acid prototrophy is involved in rhizosphere competence. In addition, PGPB regulate the rate of uptake of polyamines such as putrescine, spermine, and spermidine, since their high titer could retard bacterial growth and reduce their ability to competitively colonize roots (87). Root mucilage also offers a utilizable carbon source for PGPB (85) to use for the competitive colonization.

BIOCONTROL ACTIVITY MEDIATED BY THE SYNTHESIS OF ALLELOCHEMICALS

Offensive PGPB colonization and defensive retention of rhizosphere niches are enabled by production of bacterial allelochemicals, including iron-chelating siderophores, antibiotics, biocidal volatiles, lytic enzymes, and detoxification enzymes (6, 63, 166).

Competition for iron and the role of siderophores.

Iron is an essential growth element for all living organisms. The scarcity of bioavailable iron in soil habitats and on plant surfaces foments a furious competition (93). Under iron-limiting conditions PGPB produce low-molecular-weight compounds called siderophores to competitively acquire ferric ion (191). Although various bacterial siderophores differ in their abilities to sequester iron, in general, they deprive pathogenic fungi of this essential element since the fungal siderophores have lower affinity (94, 122). Some PGPB strains go one step further and draw iron from heterologous siderophores produced by cohabiting microorganisms (19, 92, 94, 137, 186, 191).

Siderophore biosynthesis is generally tightly regulated by iron-sensitive Fur proteins, the global regulators GacS and GacA, the sigma factors RpoS, PvdS, and FpvI, quorum-sensing autoinducers such as N-acyl homoserine lactone, and site-specific recombinases (31, 141). However, some data demonstrate that none of these global regulators is involved in siderophore production. Neither GacS nor RpoS significantly affected the level of siderophores synthesized by Enterobacter cloacae CAL2 and UW4 (148). RpoS is not involved in the regulation of siderophore production by Pseudomonas putida strain WCS358 (86). In addition, GrrA/GrrS, but not GacS/GacA, are involved in siderophore synthesis regulation in Serratia plymuthica strain IC1270, suggesting that gene evolution occurred in the siderophore-producing bacteria (123). A myriad of environmental factors can also modulate siderophores synthesis, including pH, the level of iron and the form of iron ions, the presence of other trace elements, and an adequate supply of carbon, nitrogen, and phosphorus (52).

Antibiosis.

The basis of antibiosis as a biocontrol mechanism of PGPB has become increasingly better understood over the past two decades (191). A variety of antibiotics have been identified, including compounds such as amphisin, 2,4-diacetylphloroglucinol (DAPG), hydrogen cyanide, oomycin A, phenazine, pyoluteorin, pyrrolnitrin, tensin, tropolone, and cyclic lipopeptides produced by pseudomonads (33, 40, 114, 115, 138) and oligomycin A, kanosamine, zwittermicin A, and xanthobaccin produced by Bacillus, Streptomyces, and Stenotrophomonas spp. (72, 81, 103, 104, 110). Interestingly, some antibiotics produced by PGPB are finding new uses as experimental pharmaceuticals (45, 75, 192), and this group of bacteria may offer an untapped resource for compounds to deal with the alarming ascent of multidrug-resistant human pathogenic bacteria.

Regulatory cascades of these antibiotics involve GacA/GacS or GrrA/GrrS, RpoD, and RpoS, N-acyl homoserine lactone derivatives (15, 21, 68, 131) and positive autoregulation (17, 151). Antibiotic synthesis is tightly linked to the overall metabolic status of the cell, which in turn is dictated by nutrient availability and other environmental stimuli (167), such as major and minor minerals, type of carbon source and supply, pH, temperature, and other parameters (11, 51, 52, 61, 78, 103, 104, 124, 125). Trace elements, particularly zinc, and carbon source levels influence the genetic stability/instability of bacteria, affecting their ability to produce secondary metabolites (53). It is important to note that many strains produce pallet of secondary antimicrobial metabolites and that conditions favoring one compound may not favor another (52). Thus, the varied arsenal of biocontrol strains may enable antagonists to perform their ultimate objective of pathogen suppression under the widest range of environmental conditions. For example, in P. fluorescens CHA0 biosynthesis of DAPG is stimulated and pyoluteorin is repressed in the presence of glucose as a carbon source. As glucose is depleted, however, pyoluteorin becomes the more abundantly antimicrobial compound produced by this strain (52). This ensures a degree of flexibility for the antagonist when confronted with a different or a changeable environment. Biotic conditions can also influence antibiotic biosynthesis (51, 54, 68, 116, 128). For example bacterial metabolites salicylates and pyoluteorin can affect DAPG production by P. fluorescens CHA0 (151). Furthermore, plant growth and development also influence antiobiotic production, since biological activity of DAPG producers is not induced by the exudates of young plant roots but is induced by the exudates of older plants, which results in selective pressure against other rhizosphere microorganisms (129). Plant host genotype also plays a significant role in the disease-suppressive interaction of plant with a microbial biocontrol agent, as demonstrated by Smith et al. (160, 161).

Lytic enzyme production.

A variety of microorganisms also exhibit hyperparasitic activity, attacking pathogens by excreting cell wall hydrolases (26). Chitinase produced by S. plymuthica C48 inhibited spore germination and germ-tube elongation in Botrytis cinerea (58). The ability to produce extracellular chitinases is considered crucial for Serratia marcescens to act as antagonist against Sclerotium rolfsii (121), and for Paenibacillus sp. strain 300 and Streptomyces sp. strain 385 to suppress Fusarium oxysporum f. sp. cucumerinum. It has been also demonstrated that extracellular chitinase and laminarinase synthesized by Pseudomonas stutzeri digest and lyse mycelia of F. solani (91). Although, chitinolytic activity appears less essential for PGPB such as S. plymutica IC14 when used to suppress S. sclerotiorum and B. cinerea, synthesis of proteases and other biocontrol traits are involved (77). The β-1,3-glucanase synthesized by Paenibacillus sp. strain 300 and Streptomyces sp. strain 385 lyse fungal cell walls of F. oxysporum f. sp. cucumerinum (159). B. cepacia synthesizes β-1,3-glucanase that destroys the integrity of R. solani, S. rolfsii, and Pythium ultimum cell walls (59). Similar to siderophores and antibiotics, regulation of lytic enzyme production (proteases and chitinases in particular) involves the GacA/GacS (30, 60, 111, 147) or GrrA/GrrS regulatory systems (123) and colony phase variation (97).

Detoxification and degradation of virulence factors.

Another mechanism of biological control is the detoxification of pathogen virulence factors. For example, certain biocontrol agents are able to detoxify albicidin toxin produced by Xanthomonas albilineans (9, 183, 194, 195). The detoxification mechanisms include production of a protein that reversibly binds the toxin in both Klebsiella oxytoca (183) and Alcaligenes denitrificans (9), as well as an irreversible detoxification of albicidin mediated by an esterase that occurs in Pantoea dispersa (194, 195). Several different microorganisms, including strains of B. cepacia and Ralstonia solanacearum, can also hydrolyze fusaric acid, a phytotoxin produced by various Fusarium species (169, 170). More often though, pathogen toxins display a broad-spectrum activity and can suppress growth of microbial competitors, or detoxify antiobiotics produced by some biocontrol microorganisms, as a self-defense mechanism against biocontrol agents (55, 152).

Recently, it has been discovered that certain PGPB quench pathogen quorum-sensing capacity by degrading autoinducer signals, thereby blocking expression of numerous virulence genes (47, 48, 105, 106, 113, 173). Since most, if not all, bacterial plant pathogens rely upon autoinducer-mediated quorum-sensing to turn on gene cascades for their key virulence factors (e.g., cell-degrading enzymes and phytotoxins) (181), this approach holds tremendous potential for alleviating disease, even after the onset of infection, in a curative manner.

Although biocontrol activity of microorgansims involving synthesis of allelochemicals has been studied extensively with free-living rhizobacteria, similar mechanisms apply to endophytic bacteria (92), since they can also synthesize metabolites with antagonistic activity toward plant pathogens (24). For example, Castillo et al. (20) demonstrated that munumbicins, antibiotics produced by the endophytic bacterium Streptomyces sp. strain NRRL 30562 isolated from Kennedia nigriscans, can inhibit in vitro growth of phytopathogenic fungi, P. ultimum, and F. oxysporum. Subsequently, it has been reported that certain endophytic bacteria isolated from field-grown potato plants can reduce the in vitro growth of Streptomyces scabies and Xanthomonas campestris through production of siderophore and antibiotic compounds (154). Interestingly, the ability to inhibit pathogen growth by endophytic bacteria, isolated from potato tubers, decreases as the bacteria colonize the host plant's interior, suggesting that bacterial adaptation to this habitat occurs within their host and may be tissue type and tissue site specific (164). Aino et al. (1) have also reported that the endophytic P. fluorescens strain FPT 9601 can synthesize DAPG and deposit DAPG crystals around and in the roots of tomato, thus demonstrating that endophyte can produce antibiotics in planta.

INDIRECT PLANT GROWTH PROMOTION THROUGH INDUCED SYSTEMIC RESISTANCE

Biopriming plants with some PGPB can also provide systemic resistance against a broad spectrum of plant pathogens. Diseases of fungal, bacterial, and viral origin, and in some instances even damage caused by insects and nematodes, can be reduced after application of PGPB (79, 135, 139, 146, 165).

Induced systemic resistance.

Certain bacteria trigger a phenomenon known as ISR phenotypically similar to systemic acquired resistance (SAR). SAR develops when plants successfully activate their defense mechanism in response to primary infection by a pathogen, notably when the latter induces a hypersensitive reaction through which it becomes limited in a local necrotic lesion of brown, desiccated tissue (175). As SAR, ISR is effective against different types of pathogens but differs from SAR in that the inducing PGPB does not cause visible symptoms on the host plant (175). PGPB-elicited ISR was first observed on carnation (Dianthus caryophillus) with reduced susceptibility to wilt caused by Fusarium sp. (178) and on cucumber (Cucumis sativus) with reduced susceptibility to foliar disease caused by Colletotrichum orbiculare (187). Manifestation of ISR is dependent on the combination of host plant and bacterial strain (80, 175). Most reports of PGPB-mediated ISR involve free-living rhizobacterial strains, but endophytic bacteria have also been observed to have ISR activity. For example, ISR was triggered by P. fluorescens EP1 against red rot caused by Colletotrichum falcatum on sugarcane (182), Burkholderia phytofirmans PsJN against Botrytis cinerea on grapevine (2, 3) and Verticllium dahliae on tomato (156), P. denitrificans 1-15 and P. putida 5-48 against Ceratocystis fagacearum on oak (18), P. fluorescens 63-28 against F. oxysporum f. sp. radicis-lycopersici on tomato (109) and Pythium ultimum and F. oxysporum f. sp. pisi on pea roots (12), and Bacillus pumilus SE34 against F. oxysporum f. sp. pisi on pea roots (13) and F. oxysporum f. sp. vasinfectum on cotton roots (29).

Determinants of ISR.

The ability to act as bioprotectants via ISR has been demonstrated for both rhizobacteria and bacterial endophytes, and considerable progress has been made in elucidating the mechanisms of plant-PGPB-pathogen interaction. Several bacterial traits (i.e., flagellation and production of siderophores and lipopolysaccharides) have been proposed to trigger ISR (73, 88, 90, 175, 179), but there is no compelling evidence for an overall ISR signal produced by bacteria (67). It has recently been reported that volatile organic compounds may play a key role in this process (135, 145). For example, volatiles secreted by B. subtilis GBO3 and B. amyloquefaciens IN937a were able to activate an ISR pathway in Arabidopsis seedlings challenged with the soft-rot pathogen Erwinia carotovora subsp. carotovora (144). A major distinction often drawn between ISR and SAR is the dependence of the latter on the accumulation of salicylic acid (SA) (128). Some PGPB do trigger an SA-dependent signaling pathway by producing nanogram amounts of SA in the rhizosphere (38, 39). However, the majority of PGPB that activate ISR appear to do so via a SA-independent pathway involving jasmonate and ethylene signals (128, 133). ISR is associated with an increase in sensitivity to these hormones rather than an increase in their production, which might lead to the activation of a partially different set of defense genes (71, 134).

Defense mechanisms of ISR-mediated by PGPB.

PGPB-triggered ISR fortifies plant cell wall strength and alters host physiology and metabolic responses, leading to an enhanced synthesis of plant defense chemicals upon challenge by pathogens and/or abiotic stress factors (117, 139). After inoculation of tomato with endophytic P. fluorescens WCS417r, a thickening of the outer tangential and outermost part of the radial side of the first layer of cortical cell walls occurred when epidermal or hypodermal cells were colonized (57). In Burkholderia phytofirmans PsJN-grapevine interaction, a host defense reaction coinciding with phenolic compound accumulation and a strengthening of cell walls in the exodermis and in several cortical cell layers was also observed during endophytic colonization of the bacterium (28). The type of bacterized plant response induced after challenge with a pathogen resulted in the formation of structural barriers, such as thickened cell wall papillae due to the deposition of callose and the accumulation of phenolic compounds at the site of pathogen attack (13, 14, 109). Biochemical or physiological changes in plants (139) include induced accumulation of pathogenesis-related proteins (PR proteins) such as PR-1, PR-2, chitinases, and some peroxidases (76, 100, 109, 126, 139, 182). However, certain PGPB do not induce PR proteins (73, 132, 139, 180) but rather increase accumulation of peroxidase, phenylalanine ammonia lyase, phytoalexins, polyphenol oxidase, and/or chalcone synthase (25, 120, 139, 178). Recent evidence indicates that induction of some of these plant defense compounds (e.g., chalcone synthase) may be triggered by the same N-acyl homoserine lactones that bacteria use for intraspecific signaling (99). The revelation that some PGPB genes involved in antibiotic biosynthesis (e.g., phlD) are highly homologous with some plant genes involved in defense (e.g., chalcone synthase) (4, 7) raises the intriguing but as yet unexplored possibility that the products of these DeVriesien-like pangens may have interspecies activity benefiting plant protection, in addition to their currently known functions.

CONCLUSIONS AND FUTURE PROSPECTS TO MAKE BETTER USE OF PGPB

Research into the mechanisms of plant growth promotion by PGPB have provided a greater understanding of the multiple facets of disease suppression by these biocontrol agents. Still, most of the focus has been on free-living rhizobacterial strains, especially to Pseudomonas and Bacillus. Much remains to be learned from nonsymbiotic endophytic bacteria that have unique associations and apparently a more pronounced growth-enhancing effect on host plants (6, 22, 29, 135).

Revelations about the mechanisms of PGPB action open new doors to design strategies for improving the efficacy of biocontrol agents (107, 108, 184). Identification of key antimicrobials produced by superior agents, such as 2,4-diacetylphloroglucinol, can be exploited for streamlining strain discovery by targeting selection of new isolates that carry relevant biosynthetic genes (193). Determination of the role of edaphic parameters favorable for disease suppression, particularly those that stimulate antibiotic production and activity, can be exploited by targeting inoculants for soils that are more likely to support biocontrol. For example, amending soils or growth substrates with minerals such as zinc or priming inoculants with media amendments during fermentation (51, 53, 125) can be very effective. Similarly, modulation of the rhizosphere bacteria consortia can be accomplished by soil aeration, hydrogenation, and delivery of molasses, sugars and by appropriate crop rotations (reviewed in reference 188).

Identifying different mechanisms of action facilitate the combination of strains, bacteria with bacteria or bacteria with fungi, to hit pathogens with a broader spectrum of microbial weapons (32, 56, 80, 89, 98, 119, 130, 140, 150). Along this same line, biotechnology can be applied to further improve strains that have prized qualities (e.g., formulation ease, stability, or otherwise exceptionally suited to plant colonization) by creating transgenic strains that combine multiple mechanisms of action (27, 74, 168). For example, transforming the 1-aminocyclopropane-1-carboxylic acid deaminase gene, which directly stimulates plant growth by cleaving the immediate precursor of plant ethylene (64) into P. fluorescens CHAO, not only increases plant growth but can also increase biocontrol properties of PGPB (185). Continued work with endophytic bacteria also holds potential for developing biocontrol agents that may be self-perpetuating by colonizing hosts and being transferred to progeny much as is the case with associative nitrogen-fixing PGPB on sugarcane (16) or the nonsymbiotic endophyte bacterium Burkholderia phytofirmans PsJN (117, 155).