Rooting for success : Microbial Biotechnology (original) (raw)

INTRODUCTION

With the increasing world population and the impact of global changes, agriculture is under pressure. One major concern is that global change is predicted to reduce the pool of beneficial microorganisms with which plants can interact (Custer, ). In parallel, the traditional use of chemicals to sustain crop growth and protection poses challenges for both the health of the ecosystems and the population (Good & Beatty, ; Lini et al., ). For many years, plant‐beneficial bacteria including diverse Bacillus strains have been sold as sustainable biocontrol agents and fertilizers (Glare et al., ; Santos et al., ). The interest in Bacillus stem from their capacity to secrete high amount of diverse secondary metabolites such as antimicrobial molecules, and their spore‐forming ability which ensures long shelf‐life (Blake, Christensen, & Kovács, ; Caulier et al., ; Fira et al., ; Zhang et al., ). Recent meta‐analysis demonstrated that microbial inoculants significantly enhance plant productivity and reduce microbial‐mediated disease severity in a wide variety of plants (Li et al., ; Serrão et al., ). However, in unfavourable conditions, the beneficial effect of inoculated bacteria is often incomplete or absent, which constitutes a barrier to adoption for the farmers who rely on reliable results to combat pests (Moser et al., ; O'Callaghan et al., ).

The inconsistent efficacy of bioinoculants in the field might be explained, at least partially, by a weak association between the bacteria and the plant (Gange & Gadhave, ; Martínez‐Viveros et al., ; O'Callaghan et al., ). The colonization capacity of bacteria with strong plant‐beneficial activities is only occasionally examined, yet many reports demonstrate that robust plant colonization is a prerequisite for the bacteria to provide plant‐beneficial activities (Bais et al., ; Chen et al., ; Liu et al., ; Santoyo et al., ). For the Bacillus subtilis species complex, robust plant colonization results first from attraction of the bacteria to the roots via chemotaxis, a directed motility towards increasing concentration of attracting molecules secreted by the plant (Allard‐Massicotte et al., ; Feng et al., ; Gao et al., ; Jiao et al., ; Perea‐Molina et al., ). Formation of a biofilm on the root surface, that is, a multicellular community engulfed in a self‐secreted matrix, then provides a durable association with the plant (Beauregard et al., ; Chen et al., ; Li et al., ; Xu et al., ). In addition to these two active roles, other conditions will impact the success of colonization such as fitness towards the pre‐existing community and metabolic adaptation to the abiotic conditions prevailing on the root (Carlström et al., ; Debray et al., ; Fukami, ). To artificially improve plant root colonization, many of these phenotypic traits could be genetically manipulated. However, the wide adoption of a bio‐inoculant formulated using a genetically modified organism (GMO) might elicit from various groups a similar outcry than for GMO plants. Genome‐wide association studies between plants and their microbiome suggest a co‐evolution process, but the nature of the beneficial bacteria's natural adaptations to the roots have never been examined (Trivedi et al., ). A better understanding of this process would allow us to make more informed choices about the bacterial species and traits that will promote efficient root colonization in diverse conditions.

ADAPTIVE LABORATORY EVOLUTION OF BACILLUS ON PLANTS

Adaptive laboratory evolution (ALE) are experiments in which an organism is cultured under a specific condition for many generations, allowing the emergence and selection of evolved strains with an increased fitness (Dragosits & Mattanovich, ). The mutations responsible for the improved phenotype can be determined by whole‐genome sequencing, allowing the identification of the genetic drivers for adaptation. ALE has been extensively used with model organisms such as Escherichia coli and Saccharomyces cerevisiae to gain mechanistic insights into specific processes and to develop strains with improved traits for biotechnology purposes (Dragosits & Mattanovich, ; Mavrommati et al., ; Sandberg et al., ). In this perspective, ALE was used in recent years to develop B. subtilis and other Bacilli strains with an increased root colonization capacity, and to understand the cellular adaptations underlying this phenotype.

Blake, Nordgaard, et al. () evolved B. subtilis on Arabidopsis thaliana seedlings in a minimal medium with a little amount of glycerol (Blake, Nordgaard, et al., ). Cycles were performed by transferring the colonized plant in a new growth medium containing a sterile seedling. This ALE led to the emergence of three morphotypes, with different plant colonization capacity. Many evolved isolates had enhanced colonization capacity on A. thaliana roots, but not on tomato seedlings (Solanum lycopersicum), showing a plant‐specific adaptation. Inoculation of A. thaliana with all three morphotypes simultaneously provided stronger root colonization than predicted, suggesting a complementarity between the strains. The ALE performed by Nordgaard et al. () had a similar set up, but was performed in mildly shaking conditions. Rapid adaptation to the root was also observed, resulting in increased colonization and emergence of different morphotypes. Multiple evolved isolates displayed increased individual root colonization, lower motility and stronger biofilm formation in presence of the plant polysaccharide xylan. Whole‐genome sequencing revealed mutations in genes related to motility, cell wall metabolism and biofilm formation, suggesting that these cellular pathways are interconnected with root colonization efficacy. One evolved isolate also showed better root establishment compared to the ancestor in the presence of a soil‐derived synthetic community. Finally, Lin et al. () also obtained Cry−Bacillus thuringiensis strains with stronger plant colonization on A. thaliana. Whole genome sequencing revealed that a nonsense mutation in rho, present in certain evolved isolates, led to reduced motility and sporulation efficiency, and increased plant polysaccharide usage, leading to increased aggregation and establishment on plants.

Plant‐specific adaptation of B. subtilis was further examined by Hu et al. (), using ALE on A. thaliana and S. lycopersicum. Mutations from evolution regimen with both plants in alternance had more similarity with mutations accumulated in ALE performed with S. lycopersicum, suggesting that this plant species exerted a stronger selection pressure. Pomerleau et al. () also evolved B. subtilis on S. lycopersicum, in presence or not of a competing Pseudomonas fluorescens, another plant‐beneficial bacterium. One key difference to the previously described ALEs was the usage of sonication to remove the bacteria from the colonized roots before re‐inoculation on a sterile seedling. This approach, combined with shaking conditions, likely diminished the selective pressure for motility traits. Despite these differences, rapid root adaptation also emerged in that study. Many evolved isolates displayed enhanced root colonization on both S. lycopersicum and A. thaliana, with or without the presence of the competitor. These evolved strains all possessed loss‐of‐function or mutations in genes encoding biofilm transcriptional repressors, which caused a moderate increase in biofilm formation.

All these Bacillus ALE on plants revealed that increased biofilm formation, either constitutively or in presence of plant‐specific carbon source, might be one of most important traits for increased root colonization (Blake, Nordgaard, et al., ; Hu et al., ; Lin et al., ; Pomerleau et al., ). Biofilm was also correlated with increased secondary metabolite secretion, many of which promote plant health and antagonize pathogens (Pisithkul et al., ; Schoenborn et al., ). Thus, a general strategy to increase Bacillus root colonization could be to positively affect biofilm productivity. However, a motility versus biofilm trade‐off was observed in most evolved isolates. This is not surprising, since biofilm formation in B. subtilis was shown to be mutually exclusive with motility (Vlamakis et al., ). Additionally, many ALEs were performed in mildly shaking conditions, which might have rendered motility and chemotaxis superfluous since stochastic contact triggered by agitation is likely sufficient for initial attachment to the root. The rhizosphere fitness of these strains in a natural substrate would need to be investigated to examine the real impact of motility loss. Direct inoculation of the bacteria on the root by drenching or seed coating might bypass the issue, but it could become problematic when bacteria need to reach the root before colonizing, such as in soil applications. Another parameter closely linked with biofilm formation is sporulation, since both are controlled by the same genetic regulation pathway. Sporulation was shown to be induced rapidly in the rhizosphere (Charron‐Lamoureux et al., ; Charron‐Lamoureux & Beauregard, ), and while it can provide persistence and stress tolerance, the extent of the beneficial activities provided by that dormant form is likely drastically diminished. Indeed, no secondary metabolites or nutrients solubilization can be performed by an endospore. In B. thuringiensis, the increase in biofilm formation was associated with a delay in sporulation for certain isolates (Lin et al., ). Thus, it would be of interest to examine if increased biofilm formation on root also affects the dynamic of spore formation in other Bacillus.

THE POTENTIAL USAGE OF BACILLUS ADAPTIVE EVOLUTION TO GENERATE NEW, EFFECTIVE STRAINS

The site of origin of most bacterial strains used to formulate biofertilizers is unclear (Mehnaz, ). Thus, we can speculate that many of them were probably not isolated from the plant they are inoculated on, and this is particularly true for biofertilizers with a wide array of usage. As revealed by ALE experiments, certain plants exert stronger selection pressure than others on Bacillus, which might impair colonization by an ill‐adapted biofertilizer strain. Therefore, a well‐designed ALE with the plant of interest could provide a powerful tool to naturally adapt those strains to the plant they are used on, and to promote their colonization efficacy.

Experimental conditions in which ALE are performed, including factors such as agitation, nutritional conditions, growth substrate or process of selection, can apply selective pressure that may predominantly drive adaptation towards a specific laboratory condition rather than to root colonization (Figure 1). At the same time, reproducing the complex rhizosphere environment to obtain stable evolved isolates that are adapted to the different, varying conditions may pose significant challenges. One potential approach could be to determine the fitness of different evolved isolates in natural settings to evaluate which evolution conditions yield strains better adapted to the constraints of the rhizosphere. Such strains would present stronger colonization levels and higher persistence in the rhizosphere despite the presence of a pre‐existing microbiota. If no single strains show improvement, an alternative would be to combine various, phenotypically different isolates stemming from the same ALE. Indeed, it was reported that such a consortium can exhibit stronger colonization capacities compared to individual isolates, and thus could show better fitness in the natural rhizosphere (Blake, Nordgaard, et al., ). This strategy would ensure that all the beneficial tasks provided by the original Bacillus strain are accomplished by one or another member of the consortium. Importantly, the capacity of these evolved isolates to better promote plant growth and antagonize pathogens would still need to be validated, similarly to what was performed with Pseudomonas protegens (Li et al., ). In an eco‐evolution perspective, the impact of the evolved isolates on the rhizosphere microbial community, including fungi, protist, archaea and bacteria, should also be examined since the evolved strain might impact its ecosystem differently. Thus, while ALE could become a powerful approach to generate new, more efficient plant growth‐promoting Bacillus, certain parameters need to be examined before bringing these strains in the fields.

Open multimedia modal

FIGURE 1Open multimedia modal

The many parameters of adaptative laboratory experiments. The various parameters chosen in an ALE protocol (in bold green) can influence the outcome of the evolution, leading to adaptations with various fitness to the natural rhizosphere. Subsequent steps should include validation of the colonization efficacy and whole‐genome sequencing to determine the driver of adaptations.

AUTHOR CONTRIBUTIONS

PBB conceived and wrote the draft. VCL draw Figure 1. VCL, SLB and MP wrote specific subsections, and revised and bonified the draft.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interests.

ACKNOWLEDGEMENTS

We thank Julie Beaudin for the critical reading of the manuscript, Andréa Marini for contribution to Figure 1, and members of the Beauregard laboratory for their helpful discussions. This work was supported by the NSERC discovery grant RGPIN‐2020‐07057 to PBB and a J.A. DeSève fellowship (Université de Sherbrooke) to SL‐B.

REFERENCES