Bacterial biofilms as an essential component of rhizosphere plant-microbe interactions

1. Introduction

All multicellular eukaryotes are considered holobionts, or organisms harbouring diverse microbial communities that have a major role in key aspects of their host’s biology (Miller, 2016). These host–microbe interactions are not random and result from long-term coevolution often leading to associations in which the host and its microbiota collaborate in a mutually beneficial way (Koskella & Bergelson, 2020). One of the best studied examples of such intimate and complex host–microbe interactions is represented by the human gut microbiome. Our digestive tract provides a habitat for 10–100 trillion microbial cells that are critically important for immune system development, overall health and predisposition to disease (Ursell, Metcalf, Parfrey, & Knight, 2012). The gut microbiome harbours a combined gene pool 150 times the size of the human genome and contributes to the host’s metabolism by digesting carbohydrate polymers, transforming steroids and synthesizing vitamins, amino acids, short chain fatty acids and numerous other small bioactive metabolites (Hansen & Sams, 2018). The establishment of the gut microflora in early life is vital for the proper development of the innate immune system and its ability to distinguish pathogens from symbionts and commensal microbes (Hooper, Littman, & Macpherson, 2012). Significant alterations in the structure and function of the gut microbiome (i.e., dysbiosis) have been linked to inflammatory bowel disease, susceptibility to enteric pathogens, metabolic syndrome and obesity (Tamboli, Neut, Desreumaux, & Colombel, 2004; Tilg & Kaser, 2011). The alteration of metabolic, endocrine, neural, and immune pathways by gut dysbiosis has also been associated with changes in brain function and behavioural disorders (Diaz Heijtz et al., 2011). In addition to the digestive tract, distinct and diverse microbial communities are associated with other parts of human body. These microbial assemblages, and their metabolites play a critical role in the prevention of bacterial overgrowth and infections of our skin and mucosal surfaces of the nasal cavity, mouth, throat, and urogenital tracts (Reynoso-García et al., 2022).

Plants, like humans, harbour numerous parasitic, commensal and symbiotic microbes that form complex ecological communities comprising the “phytobiome” (a term describing the plant, the environment in which it resides and the associated microorganisms). The association between plants and microorganisms is ancient and thought to have originated over 400 million years ago (Krings et al., 2007). These interactions ultimately led to the emergence of diverse bacterial and fungal communities that reside within plant tissues (endosphere), on aerial plant surfaces (phyllosphere) and especially around plant roots (rhizosphere). The bulk of the phytobiome is located in the rhizosphere, a unique environment characterized by a high level of microbial activity. With over 30,000 estimated prokaryotic species, the rhizosphere can contain up to 10¹¹ microbial cells per gram of plant root (White et al., 2017). The comparison of rhizosphere communities across 30 angiosperm plant species that diverged 140 million years ago revealed a core microbiome dominated by Proteobacteria, Acidobacteria, Actinobacteria, Bacteroidetes, Verrucomicrobia and Firmicutes (Fitzpatrick et al., 2018). However, this and numerous other comparative studies reported distinct variations in bacterial diversity and composition driven by the plant response to biotic and abiotic stressors. Similar to the human gut microbiome, the root-associated microorganisms positively influence plant nutritional status, growth and the ability to resist pathogens and abiotic stress (Trivedi, Leach, Tringe, Sa, & Singh, 2020). Many of these beneficial effects are systemic and involve signalling between the point of origin and other below- or aboveground parts of a plant.

In contrast to the carbon-limited bulk soil, the rhizosphere is a nitrogen-limited environment. By releasing rhizodeposits, plant roots recruit microorganisms that accelerate the decomposition of soil organic matter and improve the availability of nitrogen for plant uptake (the rhizosphere priming effect) (Henneron, Kardol, Wardle, Cros, & Fontaine, 2020). The rhizosphere microbial communities are also enriched in N-cycling genes (Ling, Wang, & Kuzyakov, 2022), and the role of bacterial diazotrophs and mycorrhizae in facilitating plant growth through the acquisition of nitrogen have been studied for decades. Several families of Alpha- and Betaproteobacteria known collectively as rhizobia infect the roots of legumes and certain other non-leguminous plants, where they fix atmospheric nitrogen and convert it to ammonia, which can be assimilated by the plant host (Udvardi & Poole, 2013). Another significant supply of nutrients is provided by arbuscular mycorrhizal (AM) fungi that form symbioses with an estimated 70–90% of plant species (Fitter & Moyersoen, 1996). The AM fungi are obligate biotrophs in the Glomeromycota that form extensive hyphal networks capable of mining ammonium, nitrate ions and phosphate from the soil and delivering these nutrients to the plant via intraradical mycelium (Hodge & Fitter, 2010). Many AM nonhost plants increase their access to nitrogen and essential minerals by forming interactions with diverse ectomycorrhizal, ericoid mycorrhizal and mycorrhizal-like fungal endophytes (Almario, Fabianska, Saridis, & Bucher, 2022). The beneficial effects provided by these fungi are especially evident in poor soils where mycorrhizal plants thrive due to the more efficient absorption of nutrients through the fungal mycelium. The rhizobial and mycorrhizal interactions are mutualistic, and in return for nitrogen and phosphate plants support their microbial partners with photosynthetically fixed carbon. Plants also acquire phosphate by forming associations with rhizobacteria that improve the bioavailability of this essential macronutrient by solubilizing the soil phosphates via production of organic acids (gluconic and citric), acid phosphatases, redox-active secondary metabolites and the release of protons (Alori, Glick, & Babalola, 2017; McRose & Newman, 2021; Richardson, Barea, McNeill, & Prigent-Combaret, 2009). In addition to macronutrients, plants require a variety of microelements for proper growth and development. One of these, iron, is abundant in the environment, but mostly in the insoluble and biologically unavailable oxidized form. To cope with the limited supply of iron, microorganisms scavenge it by producing siderophores, low-molecular weight metabolites with high affinity for Fe³⁺ (Soares, 2022). Siderophores are secreted in the environment, where they bind iron, and the resultant Fe-siderophore complexes are taken up by dedicated membrane receptors. Siderophores produced by rhizobacteria can efficiently supply iron to plants and improve their growth under conditions of heavy metal pollution by reducing free toxic metal concentrations in the environment (Dimkpa, Svatos, Merten, Buchel, & Kothe, 2008; Dimkpa, Merten, Svatos, Buchel, & Kothe, 2009; Lurthy et al., 2020).

Many rhizosphere microorganisms actively modulate host phytohormone levels, which stimulates germination of seed and tubers, promotes stem and root growth, and alleviates the negative effects of abiotic stresses. Treatment of plants with strains producing auxins, cytokinins and/or gibberellins commonly results in better formation of root hairs, increased root growth and branching, and, as a result, in improved mineral and nutrient uptake (Glick, 2012). Certain groups of rhizobacteria also produce an enzyme called 1-aminocyclopropane-1-carboxylate (ACC) deaminase, which cleaves the immediate precursor of the plant hormone ethylene (Nascimento, Rossi, Soares, McConkey, & Glick, 2014). Lower ethylene levels result in longer roots and less inhibition of ethylene-sensitive plant growth following environmentally-induced stress. Rhizosphere bacteria and fungi also produce volatile organic compounds (VOCs), a class of small molecules that diffuse via a gaseous phase and modulate interactions between plants and root-associated microbial communities (Garbeva & Weisskopf, 2020). Several microbial VOCs (e.g., 2,3-butanediol, 2-undecanone, 7-hexanol, 3-methylbutanol) have been shown to positively affect plant growth and tolerance to salinity, drought, and other abiotic stresses.

In addition to abiotic stressors, plants are attacked by various pests and pathogens including bacteria, fungi, oomycetes, and nematodes. Genetic resistance to most soilborne diseases is rare, and plants instead rely on their rhizosphere microbiome for protection from biotic stress. The defence mechanisms associated with rhizobacteria are diverse and include the displacement of pathogens via competition for nutrients (Mendes, Garbeva, & Raaijmakers, 2013). Many rhizosphere microorganisms act as biological control agents by producing metabolites that directly inhibit the activity of plant pathogens. These include the highly specific bacteriocins as well as numerous polyketides, nonribosomal peptides, lytic enzymes and other classes of broad-spectrum antibacterial and antifungal compounds (Mazurier, Corberand, Lemanceau, & Raaijmakers, 2009; Mojgani, 2017; Raaijmakers, de Bruijn, & de Kock, 2006; Raaijmakers & Weller, 1998; Thomashow & Weller, 1988). Rhizosphere microorganisms also promote pathogen resistance through stimulation of the plant immune system. Plant defensive responses to pathogens involve two distinct resistance pathways (Pieterse et al., 2014). The systemic acquired resistance (SAR) pathway is regulated by salicylic acid and requires direct contact with a pathogen. In contrast, the induced systemic resistance (ISR) pathway mediated by jasmonic acid and ethylene is induced by the colonization of plant roots by commensal rhizobacteria, fungi and mycorrhizae. ISR is triggered by specific microbial metabolites, cell surface antigens, effector proteins, pili and flagella, and enhances the plant defence responses against a broad range of pathogens.

2. Rhizosphere—A microbial habitat shaped by root exudates

The term “rhizosphere” was first coined over a century ago by Lorenz Hiltner to describe the dynamic interface between plant roots and soil influenced by root exudates (Hartmann, Rothballer, & Schmidt, 2008). From the microbial standpoint, the rhizosphere is an extremely diverse habitat characterized by temporal and spatial (sub-micrometre to supra-centimetre) gradients of pH, oxygen, redox potential, enzymes, water, nutrients, and plant exometabolites (Kuzyakov & Razavi, 2019). Most of these parameters are highly dynamic and reflect bidirectional exchange of different substances between the plant roots and the surrounding soil. Plant roots release H⁺ and organic acids that can alter the narrow layer of rhizosphere soil pH by one or two units compared to the surrounding bulk soil (Blossfeld, Gansert, Thiele, Kuhn, & Losch, 2011; Hinsinger, Bengough, Vetterlein, & Young, 2009). This process depends on the plant species and is influenced by nitrogen fertilizers, iron, and phosphorus deficiencies (Lurthy, Pivato, Lemanceau, & Mazurier, 2021; Ma, Liu, Shen, & Kuzyakov, 2021). It should be noted that the soil pH strongly affects the diversity and richness of soil microbial communities, and the bacterial diversity is generally higher in neutral as compared to acidic soils (Fierer & Jackson, 2006). Live plant roots and soil microorganisms consume oxygen and release carbon dioxide, and the levels of O₂ drop sharply within a few millimetres from the root surface, while CO₂ exhibits a reverse trend (Holz, Becker, Daudin, & Oburger, 2020; Larsen, Santner, Oburger, Wenzel, & Glud, 2015). The concentration of both gases depends on the soil depth and especially soil moisture, where flooding and waterlogging may lead to a rapid O₂ depletion and establishment of strictly anaerobic conditions (Pan, Sharif, Xu, & Chen, 2020). The redox potential also forms steep gradients around the root system and reflects changes in the concentration of oxygen, pH and ions produced through the dissolution of Fe³⁺ and Mn⁴⁺ oxides by root exudates (Jauregui & Reisenauer, 1982; Uteau et al., 2015).

One of the most important functions of the root system is the uptake of water to support photosynthesis and other metabolic processes, growth, and the transport of organic and inorganic molecules to the above-ground plant parts. The absorption of water by plant roots is driven by the difference in the water potential between the soil and the air, and leads to the development of radial and longitudinal soil moisture gradients in the rhizosphere (Carminati, Zarebanadkouki, Kroener, Ahmed, & Holz, 2016) (Fig. 1). Due to the continuous flow of water toward plant roots, the rhizosphere becomes drier than the surrounding bulk soil in wet conditions, while in dry bulk soils, the situation reverses. The rhizosphere water content fluctuates in response to drying and rewetting events, circadian rhythms and release of rhizodeposits (Caldeira, Jeanguenin, Chaumont, & Tardieu, 2014; Zarbanadkouki & Carminati, 2014). Many plants respond to water-limited conditions by releasing a hydrating mucilage (see below) that promotes soil aggregation and increases the water retention capability of the rhizosphere (Carminati et al., 2010). Conversely, the secretion of surfactants and hydrophobic metabolites promotes the release of water, especially in relatively wet coarse-textured soils (Read et al., 2003).

FIG. 1.

Water transport from the soil to the roots, xylem, leaf mesophyll, stomata, and ultimately the atmosphere is driven by a series of water potential gradients (ψ in MPa).

The rhizosphere environment is also dramatically affected by organic rhizodeposits composed of border cells and numerous low-molecular-weight (carbohydrates, amino acids, organic acids, phenolics) and high-molecular-weight (polysaccharides, mucilage, proteins) compounds, which differ dramatically in their residence time (Kuzyakov & Xu, 2013). In some plant species, border cells dissociating from the root cap epidermis remain alive in soil for weeks and release low-molecular-weight compounds that attract rhizobacteria and mycorrhizal fungi (Vicre, Santaella, Blanchet, Gateau, & Driouich, 2005; Watson et al., 2015). The mucilage secreted by growing plant roots consists of polysaccharides, proteins and extracellular DNA that collectively reduce friction against soil particles, provide protection against metals and improve root surface hydration (Carminati, Benard, Ahmed, & Zarebanadkouki, 2017) (Table 1). Finally, all plants passively release complex mixtures of sugars, sugar alcohols, amino acids, organic acids, fatty acids and diverse secondary metabolites (Mavrodi et al., 2021; Sasse, Martinoia, & Northen, 2018). The overall amounts of these exudates and relative levels of different classes of exometabolites vary significantly depending on the plant’s genotype, age and physiological state (Chaparro, Badri, & Vivanco, 2013; Haichar, Heulin, Guyonnet, & Achouak, 2016; Monchgesang et al., 2016). In contrast to more stable mucilage that can persist in the soil for days or even weeks (van Veelen, Tourell, Koebernick, Pileio, & Roose, 2018), the low-molecular-weight exudates are rapidly taken up and mineralized by rhizosphere microorganisms (Gunina & Kuzyakov, 2015). It is thought that the changes in the amount and composition of rhizodeposits exert selective pressure on microbes and determine which taxa are recruited into rhizobacterial assemblages, ultimately resulting in better nutrient acquisition and protection against abiotic and biotic stressors (Sasse et al., 2018).

Table 1.

Some Polysaccharides and Glycoproteins Identified in Root Mucilage of Different Plant Species (based on data discussed by Galloway, Knox, & Krause, 2020)

Plant species	Mucilage constituents	References
Arabidopsis thaliana	Arabinogalactan protein, extensin, pectin, xyloglucan	Durant et al. (2009), Galloway et al. (2018)
Barley (Hordeum vulgare)	Arabinogalactan protein, extensin, pectin, xylan, xyloglucan	Galloway et al. (2018)
Cowpea (Vigna unguiculata)	Arabinan, arabinogalactan protein, pectin	Moody et al. (1988)
Garden cress (Lepidium sativum)	Arabinan, arabinogalactan protein, pectin	Ray et al. (1988)
Lupin (Lupinus angustifolius)	Arabinan, arabinogalactan protein, pectin	Read & Gregory (1997)
Maize (Zea mays)	Arabinan, arabinogalactan protein, pectin, xylan, xyloglucan	Bacic et al. (1986), Galloway et al. (2018), Guinel & McCully (1986), Osborn et al. (1999)
Pea (Pisum sativum)	Arabinogalactan protein, extensin, pectin, xyloglucan	Cannesan et al. (2012), Galloway et al. (2018), Knee et al. (2001)
Rapeseed (Brassica napus)	Arabinogalactan protein, xylan, xyloglucan	Cannesan et al. (2012), Galloway et al. (2018)
Tomato (Solanum tuberosum)	Arabinogalactan protein, extensin, pectin, xyloglucan	Galloway et al. (2018)
Wheat (Triticum aestivum)	Arabinogalactan protein, extensin, xylan, xyloglucan	Galloway et al. (2018), Moody et al. (1988)

It has been estimated that up to 80% of all bacteria and archaea live in biofilms, with planktonic cells occurring mostly during biofilm dispersal and the colonization of new habitats (Flemming & Wuertz, 2019). This estimate certainly applies to plant-associated microorganisms that colonize aerial, vascular and root tissues of their host by forming multicellular and multispecies surface-attached assemblies embedded in a self-produced matrix (Danhorn & Fuqua, 2007). The presence of mucilage-embedded microcolonies on plant roots was first reported in the 1970s (Rovira, Newman, Bowen, & Campbell, 1974) and confirmed in numerous later studies that collectively revealed a close resemblance between the processes of rhizosphere colonization and biofilm formation (Fig. 2). Plants secrete up to 40% of their photosynthates through their roots, creating a hot spot of microbial activity and diversity (Bais, Weir, Perry, Gilroy, & Vivanco, 2006). The microorganisms are drawn from the carbon starved bulk soil toward plant roots secreting nutrient-rich exudates, a phenomenon known as the “rhizosphere effect” (Bakker, Berendsen, Doornbos, Wintermans, & Pieterse, 2013). Studies of carbon allocation by isotope tracing estimate the rhizosphere thickness at 2–3 mm (Holz, Zarebanadkouki, Kuzyakov, Pausch, & Carminati, 2018), suggesting that it is a well-structured, competitive and highly-resourced environment conducive to the formation of biofilms. The potential benefits of aggregation and biofilm formation by rhizobacteria include protection from environmental stress (desiccation, unfavourable temperature and pH, antimicrobial agents, etc.), colonization of nutrient-rich habitats, and communal interactions (regulation of biofilm development, division of the metabolic burden, gene exchange, host colonization, defence against competitors and predators) (Jefferson, 2004; Mukherjee & Bassler, 2019).

FIG. 2.

Colonization of wheat roots by the model biocontrol agent Pseudomonas synxantha 2-79. The genetically modified strain used in this experiment is constitutively expressing mCherry (red cells) and carries a phzA::gfp fusion to monitor the expression of phenazine biosynthesis genes (green cells). Cells that express both fluorescent proteins appear yellow. The left panel depicts an extended root section colonized by rhizobacteria, while the right panel shows several microcolonies producing phenazine-1-carboxylic acid. The observations were performed by overlaying differential interference contrast and fluorescent images obtained with an epifluorescence microscope.

3. Experimental approaches for studying rhizosphere biofilms

The experimental approaches for studying rhizosphere biofilms build upon an extensive array of methods and techniques used in general biofilm research. We are not going to discuss all of these approaches since their general principles, strong points and limitations were reviewed elsewhere (Azeredo et al., 2017). Rather, we will briefly discuss how some of these methods were adapted to investigate rhizobacteria and biofilms associated with plant roots. The discussion will focus primarily on Pseudomonas and Bacillus, since these well-studied organisms provided genetically tractable models for unravelling molecular aspects of plant-microbe interactions in the rhizosphere (Table 2).

Table 2.

A list of selected studies featuring different experimental models and approaches used in the analysis of root-associated biofilms and assessment of their contribution to rhizosphere plant-microbe interactions

Rhizobacterium	Type of organism	Plant host	Study synopsis	References
Agrobacterium tumefaciens	Pathogen	Tomato, Arabidopsis thaliana	Bacterial cellulose was identified as a critical contributor to the ability of A. tumefaciens to attach, colonize and form biofilms on roots of tomato and A. thaliana	Matthysse et al. (2005)
Bacillus amyloliquefaciens	Commensal, mutualist	Maize	Root exudates were shown to enhance biofilm formation by modulating the expression of genes that control extracellular matrix production and motility	Zhang et al. (2015)
Bacillus amyloliquefaciens	Commensal, mutualist	A. thaliana	A combination of mutagenesis, in vitro and in planta assays was used to demonstrate the role of collagen-like proteins in auto-aggregation of B. amyloliquefaciens FZB42 and biofilm formation on roots of A. thaliana	Zhao et al. (2015)
Bacillus subtilis	Commensal, mutualist	A. thaliana	Plant polysaccharides were shown to induce rhizosphere biofilms of B. subtilis and act as an environmental cue and C source for the production of biofilm matrix	Beauregard et al. (2013)
Bacillus subtilis	Commensal, mutualist	Tomato	Several conserved B. subtilis regulatory and matrix genes were shown to function in biofilm formation, root colonization and protection of tomato plants against the plant pathogen Ralstonia solanacearum	Chen et al. (2013)
Bacillus subtilis	Commensal, mutualist	A. thaliana	The first report of a microfluidic chip set-up to investigate the dynamics of B. subtilis root colonization	Massalha et al. (2017)
Bacillus subtilis	Commensal, mutualist	A. thaliana	A combination of gene replacement mutagenesis, in vitro and in planta assays, and confocal microscopy was used to demonstrate that genes involved in turnover of c-di-AMP affect colony morphology, biofilm gene expression and colonization of plant roots by B. subtilis NCIB3610	Townsley et al. (2018)
Bacillus subtilis and Pseudomonas fluorescens	Commensal, mutualist	Populus tremuloides	A microfluidic RIM-chip was used for extended observations of root development and rhizosphere interactions between Populus tremuloides (aspen tree) and strains of B. subtilis and P. fluorescens	Noirot-Gros et al. (2020)
Bacillus velezensis	Commensal, mutualist	Cucumber	A syntrophy between B. velezensis and P. stutzeri was implicated in the formation of mixed rhizosphere biofilms that promote plant growth and alleviate salt stress in cucumber	Sun et al. (2022)
Gluconacetobacter diazotrophicus	Mutualistic endophyte	Rice	An exopolysaccharide-deficient mutant of G. diazotrophicus was defective in biofilm formation, root surface attachment and endophytic colonization of rice plants	Meneses et al. (2011)
Paenibacillus polymyxa	Commensal, mutualist	A. thaliana	A combination of fluorescent and electron scanning microscopy was used to visualize biofilms formed by P. polymyxa on A. thaliana roots	Timmusk et al., (2005)
Pseudomonas fluorescens	Commensal, mutualist	Tomato	One of the first studies to visualize rhizosphere biofilms via simultaneous imaging of mixed bacterial populations tagged with different autofluorescent proteins	Bloemberg et al. (2000)
Pseudomonas fluorescens	Commensal, mutualist	A. thaliana	A set of experiments identified putrescine as a biofilm inducer and showed that a putrescine utilization Pseudomonas mutant formed enhanced biofilms, but also inhibited Arabidopsis by activating pattern-triggered immunity (PTI)	Liu et al. (2018)
Pseudomonas putida	Commensal	Maize	This study characterized contribution of LapF to plant root colonization and architecture of biofilms formed by P. putida KT2440	Martinez-Gil et al. (2010)
Pseudomonas putida	Commensal	Maize	The analysis of chemoreceptor mutants of P. putida KT2440 revealed that polyamine receptor genes control biofilm formation and colonization of maize roots by regulating c-di-GMP levels	Corral-Lugo et al. (2016)
Pseudomonas putida	Commensal	Maize	Seed adhesion and biofilm formation assays revealed the role of the large adhesin LapA in the colonization of abiotic and biotic surfaces by P. putida KT2440	Yousef-Coronado et al. (2008)
Pseudomonas chlororaphis	Commensal, mutualist	Avocado	Surface attachment, flow cell and plant-based assays paired with confocal laser scanning microscopy were used to show the contribution of Psl exopolysaccharide and Fap-like fibres to biofilm formation, niche competition, and plant pathogen suppression by P. chlororaphis PCL1606	Heredia-Ponce (2021)
Pseudomonas simiae	Commensal	Brachypodium distachyon	A microfabricated ecosystem (EcoFAB) was used to investigate the spatiotemporal colonization dynamics of plant roots by fluorescent tagged Neurospora crassa and P. simiae	Jabusch et al. (2021)
Sinorhizobium meliloti	Mutualistic endophyte	Alfalfa	Inactivation or overproduction of EPS leads to a reduction in biofilm formation and alfalfa nodulation by S. meliloti	Fujishige et al. (2006)

The spatial heterogeneity of root-associated communities, unpredictability of field conditions and opacity of the soil matrix makes the direct real-life characterization of rhizosphere biofilms difficult. To circumvent these challenges, relevant experiments are often conducted in a controlled environment using various biofilm cultivation devices such as microtiter plates, flow chambers, drip flow reactors, and microfluidic platforms (Harding & Daniels, 2017). Some of these techniques, especially the combination of microplate-based assays and crystal violet staining, are very popular due to their low cost and simplicity and have been used in numerous descriptive papers focused on the isolation and characterization of rhizobacteria with biocontrol and/or plant growth-promoting properties. However, the results of such studies should be interpreted cautiously since these biofilm experiments are often conducted in rich culture media under conditions vastly different from those observed in the rhizosphere.

3.1. Growing biofilms in the presence of root exudates

The environmental relevance of in vitro assays can be improved by growing biofilms in the presence of root exudates that play a key role in the communication between the plant and soil microorganisms and mediate the selection and maintenance of the rhizosphere microbiome. As detailed above, root exudates contain numerous primary and secondary metabolites and have complex chemical profiles that are difficult to match in an artificially constructed culture medium (Sasse et al., 2018). These problems can be overcome by performing biofilm assays in the presence of root exudates collected from axenically grown plants. The seeds for starting axenic plants are surface sterilized by applying various combinations of antimicrobial agents, which helps to limit the interference from soil chemical inputs and microbial metabolites. One practical sterilization technique involves treating seeds with chlorine gas generated by mixing fresh commercial chlorine bleach (7.5%–8.25% NaClO) with concentrated HCl (Lindsey, Rivero, Calhoun, Grotewold, & Brkljacic, 2017). With this approach, treating the seeds for 1 h with 6.1% Cl₂ yielded optimal sterilization efficacy and viability. Other successfully tested sterilization protocols use different combinations of sodium hypochlorite with ethanol or hydrogen peroxide (Miché, & Balandreau, 2001; Davoudpour, Schmidt, Calabrese, Richnow, & Musat, 2020). Finally, a recent study by Munkager et al. (Munkager et al., 2020) demonstrated that microorganisms present on the surface of barley grain could be efficiently removed by a brief soaking in 1% (w/w) silver nitrate. The germination rate of surface-sterilized seeds can be monitored by incubating them for 2–3 days on water agar or moistened germination paper.

The exudates can be collected by growing axenic plants hydroponically or in containers containing glass beads, silica or quartz sand that simulate the mechanical contact of roots with soil particles (Vranova, Rejsek, Skene, Janous, & Formanek, 2013). An interesting and ecologically relevant extension of these techniques involves growing plants in soil and then extracting, rinsing their roots and transferring them to a hydroponic setup for exudate collection. Such an approach was successfully used to collect and compare root exudates of velvet grass (Holcus lanatus), sorrel (Rumex acetosa) and white clover (Trifolium repens) (Williams et al., 2021). A critical step of this hybrid collection method involved a 3-day recovery period after removing plants from soil to prevent changes in exudate profiles associated with osmotic shock and damage to plant roots. Recent studies indicate that the hydroponic and hybrid approaches are robust and can be applied to diverse plant species to yield root exudates with reproducible profiles of primary and secondary metabolites (Strehmel, Böttcher, Schmidt, & Scheel, 2014; van Dam, & Bouwmeester, 2016; Mavrodi et al., 2021; Song, Pieterse, Bakker, & Berendsen, 2021).

Although exudate-based experimental systems lack the microflora and gradients of nutrients, oxygen and water present in the real soil, they reliably generate exudates in quantities sufficient for instrumental profiling and microbial transcriptomic and proteomic studies (Kierul et al., 2015; Mavrodi et al., 2021; Pantigoso, He, DiLegge, & Vivanco, 2021; Vora, Ankati, Patole, Podile, & Archana, 2022). Using such axenically-collected exudates and microplate assays, multiple studies have demonstrated the stimulatory effect of root exudates on biofilms formed by different species of Bacillus (Liu et al., 2020; Sharma et al., 2020; Xie et al., 2022; Yuan et al., 2015), Pseudomonas (Lopez-Farfan et al., 2019), actinomycetes (Jiang, Long, Xu, & Han, 2023), and other groups of rhizobacteria. These observations are explained by the results of Zhang et al. (2015), who performed an RNA-seq profiling of B. amyloliquefaciens biofilms exposed to root exudates and demonstrated downregulation of flagellar motility genes and induction of pathways involved in the production of biofilm matrix. We have recently analysed transcriptome responses of the model rhizosphere biocontrol agent P. synxantha 2–79 (Weller & Cook, 1983) to root exudates of barley (Hordeum vulgare) cv. Golden Promise. The processing of bacterial cultures incubated in the presence of 20-fold concentrated barley exudates generated a total of 236.2 million high-quality Illumina reads (~19.7 million reads per sample) that were aligned onto the 2–79 genome, assembled, and subjected to differential expression analysis with a false discovery rate (FDR) cutoff of 0.05 and an absolute log2 fold change threshold of 1.5. The analysis revealed a total of 179 genes that were differentially expressed in response to root exudates. Similar to Zhang et al. (Zhang et al., 2015), we observed an induction of genes involved in the synthesis of fimbriae and type IVb pili, which are predicted to function in surface attachment and early stages of biofilm formation (Fig. 3). We also observed an induction of genes involved in phenazine biosynthesis and uptake and catabolism of quaternary ammonium and aromatic compounds, whereas pathways involved in the metabolism of zinc, iron, and synthesis of siderophores were significantly downregulated.

FIG. 3.

Transcriptome responses of P. synxantha 2-79 to root exudates of barley (cv. Golden Promise). (A) Volcano plot of bacterial genes that respond to root exudates. Each point on the chart represents an individual gene. The log₂-fold differences between treatments are plotted on the x-axis, while the −log₁₀ P-value differences are plotted on the y-axis. The significantly up- or downregulated genes (FDR-adjusted P <0.05) are highlighted in green (abs. log₂ FC <1.5) and red (abs. log₂ FC >1.5), respectively. (B) KEGG Orthology classification of proteins encoded by differentially expressed genes. The KO annotations were assigned with BLASTKOALA (Kanehisa, Sato, & Morishima, 2016) (C) The response to root exudates involved induction of genes for the biogenesis of type IVb pili and fimbriae, which is consistent with surface attachment and transition to the biofilm growth mode.

3.2. The use of microfluidics in the rhizosphere biofilm research

Root exudates can be used to grow biofilms of rhizosphere bacteria in various reactors and flow chambers. When combined with fluorescent reporters and confocal microscopy (see next section), such assays allow to perform long-term nondestructive observations that reveal temporal changes in biofilm development and gene expression (Azeredo et al., 2017). Unfortunately, the conventional flow chamber experiments require moderate to large volumes of culture media and cannot be used with roots exudates, which are difficult to mass-produce. Microfluidics helps to overcome this limitation by growing biofilms in micrometric volumes containing precisely controlled gradients of nutrients and shear forces. These technical capabilities are superior to assays based on agar-grown plants and make the microfluidic technology particularly attractive for dissecting environmental factors that affect the bacterial behaviour and physiology in rhizosphere biofilms (Perez-Rodriguez, Garcia-Aznar, & Gonzalo-Asensio, 2022). For example, Massalha, Korenblum, Malitsky, Shapiro, & Aharoni (2017) designed a microfluidic platform termed TRIS (tracking root interactions system) for real time imaging of root–bacteria interactions and used it to monitor the colonization of Arabidopsis roots by fluorescently labelled B. subtilis. The TRIS system experiments revealed chemotactic behaviour of B. subtilis toward the root elongation zone and confirmed the importance of bacterial exopolysaccharides for the formation of rhizosphere biofilms. The authors also assembled a double-channel TRIS device and used it to compare the bacterial response to the wild type Arabidopsis and its mutant root genotypes. A different microfluidic chip was employed to perform extended month-long observations of root development and plant-microbe rhizosphere interactions in aspen tree (Populus tremuloides) (Noirot-Gros et al., 2020). Using the mNeonGreen- and dsRed-tagged derivatives of P. fluorescens SBW25, the authors revealed dynamic formation and dispersal of root-associated biofilms, while the use of B. subtilis MMB1023-based Gfp biosensors demonstrated the presence of xylose and reactive oxygen species in root exudates. An agar-embedded microfluidic platform was used to study the colonization of Arabidopsis seedlings by Gfp- and mCherry-tagged Variovorax sp. CF313 and Pantoea sp. YR343 isolated from the endosphere and rhizosphere of P. deltoides (Aufrecht et al., 2018). The experimental setup allowed precise tracking and quantification of the colonization kinetics and spatial distribution of the bacteria and revealed distinct associations of the two strains with plant roots. A different approach was pursued by Bonebrake and co-authors (Bonebrake et al., 2018), who used high-flux hollow fibre membranes (HFMs) as root mimetics to study the effect of artificial exudates and CuO and ZnO nanoparticles on biofilms of the rhizobacterium P. chlororaphis O6 and the plant endophyte B. subtilis Bs309. Scanning electron microscopy and atomic force microscopy confirmed that both strains successfully colonized the outer surface of HFM fibres and differentially responded to the artificial root exudates and nanoparticle challenges. Another interesting adaptation of microfluidic technology is exemplified by the “rhizosphere-on-a-chip” device, a synthetic microhabitat that encourages the development of root exudate hotspots by mimicking the structural heterogeneity of soil. The device allowed the profiling of exudates by liquid micro-junction surface sampling probe mass spectrometry and revealed hotspots of concentrated carbon in the form of amino acids forming around the B. distachyon roots (Aufrecht et al., 2022). The rhizosphere-on-a-chip platform will help to unravel how these exudation hotspots affect the recruitment of soil microorganisms and establishment of rhizosphere biofilms.

3.3. Integration of rhizosphere biofilm research with microscopy

In addition to advances in the area of specialized instrumentation for culturing microorganisms on solid surfaces, our understanding of the nature and properties of rhizosphere biofilms benefitted from the use of confocal laser scanning microscopy (CLSM) and bacterial and fungal strains tagged with fluorescent proteins (Cardinale, 2014). Although the technical details of CLSM are beyond the scope of this chapter and can be found elsewhere (McNamara, Difilippantonio, Ried, & Bieber, 2017), a key advantage of this optical imaging technique over conventional fluorescence microscopy is the ability to take optical sections and combine them to generate detailed 3D renderings of biofilm architecture and spatial arrangement of microbial cells. In contrast to molecular techniques that require the destructive removal of biofilms, CLSM can provide crucial spatial information by observing live specimens in situ. The first use of CLSM in rhizosphere research dates back to the late 1990s/early 2000s when several strains of Pseudomonas putida and P. fluorescens were tagged with genes encoding enhanced green, cyan, and yellow variants of GFP and red fluorescent protein (DsRed) and applied to axenic seedlings of barley and tomato (Normander, Hendriksen, & Nybroe, 1999; Bloemberg, Wijfjes, Lamers, Stuurman, & Lugtenberg, 2000). The CLSM analysis of the colonized plants revealed the presence of mixed biofilm microcolonies on mature upper parts of the root, whereas root tips contained predominantly single cells. These pioneering efforts were followed by dozens of reports that broadened the range of studied microorganisms and plant hosts (many of these studies are cited in this chapter). There was also a significant expansion of genetic delivery tools (stable plasmids, site-specific integration vectors) and fluorescent proteins for tagging rhizobacteria and creating reporters, which allow studying biofilms formed by multi-species consortia (Miller, Leveau, & Lindow, 2000; Lambertsen, Sternberg, & Molin, 2004; Bisicchia, Botella, & Devine, 2010; Schlechter et al., 2018; Wilton et al., 2018). In addition to fluorescent proteins, microorganisms attached to plant roots were visualized with fluorescent stains, probes (see next section) or even fluorescent lectins that bind selectively to certain carbohydrates present in microbial exopolysaccharides (Santaella, Schue, Berge, Heulin, & Achouak, 2008; Yang et al., 2021). These developments were complemented by efforts to reduce the autofluorescence from plant tissues, which improves the sensitivity of CLSM during probing rhizosphere plant-microbe interactions (Jiang, Pees, & Reinhold-Hurek, 2022).

Because of the inherent opacity of soil, the in situ imaging of living roots and rhizosphere microorganisms is often performed in gnotobiotic systems. Several different gnotobiotic growth systems were used to investigate functional plant-microbe interactions, microbial community establishment mechanisms, and the impact of environmental factors on plant microbiomes. The less complicated approaches involve growing inoculated seedlings hydroponically, on agar or in germination pouches, which provides easy access to intact root systems but at the same time lacks soil-like physical structure (Ma, Ordon, & Schulze-Lefert, 2022). Other gnotobiotic systems use sterile quartz sand (Simons et al., 1996), perlite (Ma, Ordon, & Schulze-Lefert, 2022), peat matrix (Kremer et al., 2021), or transparent artificial soil comprised of Nafion, a sulfonated tetrafluoroethylene-based fluoropolymer-copolymer with low refractive index (Downie et al., 2012). Yet another interesting approach is represented by the development of microfabricated ecosystems (EcoFABs) designed for studies of rhizosphere plant-microbe interactions using model plants such as A. thaliana, Brachypodium distachyon, or Panicum virgatum (Gao et al., 2018). The original EcoFAB chambers were recently modified to permit nondestructive high-resolution (20×, 40×) microscopic imaging of the entire plant root system for a period of up to 3 weeks (Jabusch et al., 2021). The device was successfully tested with different solid media and allowed investigations of the spatiotemporal colonization dynamics of a B. distachyon rhizosphere by fluorescent tagged Neurospora crassa and nine different strains of Pseudomonas simiae. In another recent study, gnotobiotic chambers constructed of polydimethylsiloxane and microscope glass slides and filled with Nafion were used to examine the behaviour of GFP-tagged B. subtilis in the rhizosphere of lettuce seedlings (Liu et al., 2021). The authors used light sheet fluorescence microscopy, a new technique that uses a sheet of laser light to illuminate a thin slice of the sample perpendicularly to the direction of observation. The fluorescence and images are then captured by a wide-field fluorescence microscope coupled to a full-frame camera. Compared to conventional CLSM, light-sheet microscopy offers reduced acquisition time, photo-bleaching, and excellent temporal and 3D-spatial resolution (Ovečka et al., 2022). The new imaging technique captured unprecedented details of the dynamics of B. subtilis in the rhizosphere through high-resolution tracking of samples up to 3600mm³ in size at a rate of one scan every 30 min.

3.4. Other emerging approaches for mapping rhizosphere biofilms

The confocal laser scanning microscopy can be combined with fluorescence in situ hybridization (FISH) to identify specific microorganisms or colocalized species associated with hotspots of microbial activity in the rhizosphere. Phylogenetic FISH uses fluorescent oligonucleotide probes matching conserved or variable regions in small and large ribosomal subunit RNAs (rRNAs) of bacteria, archaea and eukaryotes (Young, Jackson, & Wyeth, 2020). The probes can be highly specific and target certain closely related species or broadly conserved, thus permitting tracking members of an entire phylum of interest in a biological sample. The probes can also be adapted to study various functional genes of interest (Kawakami et al., 2012). In rhizosphere research, FISH was successfully used to identify and enumerate microorganisms of interest (Schmidt & Eickhorst, 2014; Darriaut et al., 2022), monitor shifts in microbial communities (Gamez, Ramirez, Montes, & Cardinale, 2020), or analyse rhizosphere plant-microbe interactions (Scholz, Muller, Koren, Nielsen, & Meckenstock, 2019). A modification of this method with an error-correction strategy called sequential error-robust fluorescence in situ hybridization (SEER-FISH) was recently used to produce a highly multiplexed and detailed spatial profiling of microbial communities associated with Arabidopsis roots (Cao et al., 2023). Finally, the use of an enzymatic signal amplification step in the catalysed reporter deposition FISH (CARD-FISH) significantly enhances detection sensitivity and permits targeting not only rRNAs but also much less abundant mRNAs. The CARD-FISH technique was recently successfully tested to visualize bacteria in the rhizosphere of Zea mays (Bandara et al., 2021). Other exciting emerging ways of studying rhizosphere biofilm processes involve various combinations of fluorescence in situ hybridization and isotope imaging approaches. Detailed reviews of these experimental techniques are available elsewhere (Moran & McGrath, 2021; Oburger & Schmidt, 2016).

4. Sensing the plant host: Bacterial chemotaxis towards plant roots exudates

The sensing of root exudates by rhizobacteria followed by chemotaxis and movement towards the root system, encompass the early stages in the establishment of rhizosphere biofilms (Fig. 4). A number of studies conducted with strains of Pseudomonas, Bacillus and Sinorhizobium have revealed that the rhizobacteria locate their host by sensing low-molecular-weight compounds released into the soil by plant roots (Allard-Massicotte et al., 2016; Meier, Muschler, & Scharf, 2007; Oku, Komatsu, Tajima, Nakashimada, & Kato, 2012). The detection of exudate constituents is controlled by a special class of chemoreceptors called methyl-accepting chemotaxis proteins (MCPs) (Ortega, Zhulin, & Krell, 2017). The MCPs are homodimeric transmembrane proteins containing a sensory periplasmic moiety and a conserved cytosolic methyl-accepting domain. The binding of a specific ligand by the sensory domain determines the MCP methylation state. This information is transmitted down a phosphorelay system to the flagellar machinery and controls bacterial movement towards chemoattractants by changing the direction of rotation of the flagellum. Many rhizobacteria harbour diverse repertoires of MCPs that reflect their ability to persist in the bulk soil and colonize plant roots. For example, genomes of Sinorhizobium meliloti carry nine MCP genes, whereas different strains of Bacillus amyloliquefaciens and B. subtilis have on average seven to ten genes encoding methyl-accepting chemotaxis proteins (Gumerov, Ortega, Adebali, Ulrich, & Zhulin, 2020). On the other hand, species of the Pseudomonas fluorescens group, which includes numerous well-characterized rhizosphere strains of P. fluorescens, P. chlororaphis, P. protegens, and P. synxantha, harbour between 25 and 58 MCP genes.

FIG. 4.

Recruitment of soil microorganisms by root exudates and establishment of rhizosphere biofilms.

A growing number of functional studies help to relate the results of genome sequence mining with chemotactic responses to exometabolites present in root exudates. For example, the recruitment of the nitrogen-fixing symbiont S. meliloti by alfalfa and other legume hosts involves chemotaxis towards certain amino acids, organic acids, carbohydrates, and quaternary ammonium compounds (Compton, Hildreth, Helm, & Scharf, 2018; Meier et al., 2007; Shrestha et al., 2018; Webb et al., 2017). Analysis of chemotactic responses of the biological control agent Bacillus velezensis SQR9 to cucumber root exudates revealed that 39 out of the 98 identified plant metabolites acted as chemoattractants (Feng et al., 2018) that included amino and organic acids, carbohydrates, sugar alcohols and other compounds. Studies of soil and rhizosphere pseudomonads such as Pseudomonas putida KT2440 revealed chemotactic responses to tricarboxylic acid cycle intermediates and other organic acids, various amino acids, polyamines, and carbohydrates (Corral-Lugo et al., 2016; Fernandez, Morel, Corral-Lugo, & Krell, 2016; Garcia et al., 2015; Reyes-Darias et al., 2015). The chemosensory response of P. putida KT2440 to root exudates was concentration-dependent, and high levels of exudates repressed the MCP genes suggesting that chemotaxis is less useful in close proximity to plant roots (Lopez-Farfan, Reyes-Darias, Matilla, & Krell, 2019). Another model rhizosphere strain, P. fluorescens Pf0-1, exhibited a positive chemotactic response towards common L-amino acids, many of which are present in root exudates (Oku et al., 2012). Interestingly, Pf0-1 also strongly discriminated between the L- and D-isomers of malate illustrating yet another layer of complexity in the chemotactic responses to exudate components (Oku, Komatsu, Nakashimada, Tajima, & Kato, 2014).

Chemotactic responses to root exudates involve flagella- and pili-driven motility towards a chemical gradient which are controlled by distinct but similar signal transduction mechanisms. In the canonical flagella-driven chemotaxis, cytoplasmic domains of MCPs interact with an adaptor protein, CheW, and control the histidine kinase CheA. When no attractant is present, the CheA kinase phosphorylates a protein called CheY, which then interacts with a rotor protein resulting in clockwise flagellum rotation and cell tumbling (Sourjik & Wingreen, 2012). The binding of an attractant inhibits CheA kinase activity, leading to lower CheY-P levels and triggering counterclockwise flagellum rotation, which enables swimming towards the attractant. The system also involves a S-adenosyl methionine-dependent methyltransferase, CheR, that constitutively methylates MCPs, thus resetting the CheA kinase. CheA reactivation triggers tumbles until an even higher concentration of attractant is encountered. The CheR activity is counteracted by CheB which, upon phosphorylation by CheA, removes methyl groups and re-sensitizes the system as the cell moves away from the attractant. Flagella-driven motility theoretically allows the bacteria to move rapidly towards root exudates in the presence of sufficient soil moisture. The chemotactic movement speed of E. coli (approx. 30 μm s⁻¹) suggests that motile bacteria can reach sources of exudation in the rhizosphere by swimming up to 3 mm in under 2 min (Macnab & Koshland, 1972; Olson, Ford, Smith, & Fernandez, 2004). Another chemosensory system resembling flagella-driven chemotaxis regulates twitching motility; it uses type IV pili (Bertrand, West, & Engel, 2010). Type IV pili gene clusters are present in the genomes of some rhizosphere and endophytic bacteria, where they contribute to the colonization of plant tissues (Dorr, Hurek, & Reinhold-Hurek, 1998; Yan et al., 2008).

The significance of chemotactic responses to root exudates for rhizosphere fitness has been supported by multiple studies involving isogenic mutants of taxonomically diverse rhizobacteria. For example, the early-stage colonization of the Arabidopsis thaliana root surface (rhizoplane) by Bacillus subtilis requires chemotaxis and flagella-mediated motility (Allard-Massicotte et al., 2016). Isogenic mutants of Rhizobium leguminosarum with inactivated methyl-accepting chemotaxis mcpB and mcpC genes were less fit than the wild type parent in the nodulation of pea roots (Yost, Rochepeau, & Hynes, 1998), while chemotaxis to L-malate proved to be essential for the infection of tomato roots by the soilborne pathogen Ralstonia pseudosolanacearum (Hida et al., 2015). Similarly, null mutations in different chemotaxis- and motility-related genes were associated with reduced rhizosphere fitness in several species of Pseudomonas. In P. fluorescens Pf0-1, mutants lacking multiple MCP systems (CtaABC, McpS and McpT) and flagella-driven chemotaxis (CheA) exhibited reduced fitness in the colonization of the tomato rhizosphere (Oku et al., 2012). The importance of motility and chemotaxis towards root exudates was confirmed in several other biocontrol P. fluorescens strains (WCS365, WCS374, OE 28.3, SBW 25, F113) and P. putida, where isogenic cheA or fli mutants were up to 100- to 1000-fold less competitive in rhizosphere colonization of different crops (Corral-Lugo et al., 2016; de Weert et al., 2002; Lugtenberg, Dekkers, & Bloemberg, 2001). Interestingly, a large-scale comparative analysis of sequenced genomes revealed the widespread presence of chemotaxis and flagellum biosynthesis operons in plant- and root-associated microorganisms, thus underscoring the key importance of these genomic features for bacterial adaptation to plants (Levy et al., 2017).

5. Colonizing the plant root

When rhizobacteria reach the plant, they switch from flagella-based swimming to type IV pili-mediated motility and adhere to root surfaces by upregulating adhesins and forming biofilms (Morris & Monier, 2003). Rhizosphere biofilms contain bacterial cell aggregates embedded in a matrix of secreted polysaccharide polymers with entrapped extracellular DNA, proteins, lipids and inorganic material. This highly hydrated biofilm shell protects the rhizobacteria from desiccation, protozoan grazing and bactericides (Danhorn & Fuqua, 2007). Biofilms also improve the uptake of nutrients, enhance genetic exchange and stimulate the expression of cell density-dependent phenotypes involved in the interactions with the host plant and other microbes. Similar to biofilms formed in other environments, the colonization of plant root surfaces progresses via several stages that include reversible and irreversible attachment, formation of microcolonies, maturation, and a dispersal process that generates new planktonic cells. The molecular details of these processes in rhizosphere settings have been unravelled using a variety of Gram-positive and Gram-negative model organisms including several species of Bacillus, Pseudomonas, Agrobacterium and some rhizobia.

5.1. Attachment to plant roots

The initial contact of bacteria with plant roots involves weak hydrophobic and electrostatic attractive forces developing between the bacterial cell envelope and the root surface (Kimkes & Heinemann, 2020). This interaction is dynamic and may be followed by detachment and return to the planktonic mode (Hinsa, Espinosa-Urgel, Ramos, & O’Toole, 2003). On the other hand, this initial reversable contact may be followed by a stronger and more specific attachment that is induced by certain particular environmental signals and leads to the production of various cell surface appendages and cell aggregation. The attachment of rhizobacteria to plant roots employs a combination of several common and species-specific molecular mechanisms. Some rhizobacteria use hydrophobic flagella as adhesins, a strategy demonstrated in the beneficial diazotroph Azospirillum brasilense and the human pathogen Salmonella, which can contaminate vegetables by interacting with their roots (Cooley, Miller, & Mandrell, 2003; Croes, Moens, van Bastelaere, Vanderleyden, & Michiels, 1993). In addition, the interference with flagellar rotation activates signalling pathways that control biofilm formation, as has been shown in Bacillus subtilis (Cairns, Marlow, Bissett, Ostrowski, & Stanley-Wall, 2013).

In other species of rhizobacteria, surface attachment may employ secretion of specialized polysaccharides. For example, Rhizobium leguminosarum and Bradyrhizobium japonicum use for attachment a polarly localized polysaccharide adhesin comprised mainly of glucose and mannose (Laus et al., 2006; Lodeiro & Favelukes, 1999). This glucomannan interacts with lectins produced by root hairs and contributes to the establishment of the Rhizobium-legume symbiosis required for nitrogen fixation. A similar glucomannan adhesin called the unipolar polysaccharide (UPP) is produced by another member of the Rhizobiaceae, the plant pathogen A. tumefaciens. The UPP adhesin plays an important role in the surface attachment and formation of A. tumefaciens biofilms and may be required for the interaction of this pathogen with the host plant (Xu, Kim, Danhorn, Merritt, & Fuqua, 2012).

Yet another common surface attachment strategy relies on the use of pili, fimbriae or adhesive proteins. For example, the genome of A. tumefaciens carries the ctpABCDEFGHI gene cluster encoding the biosynthetic components and secretion apparatus for type IVb pili. The inactivation of selected ctp genes correlated with the loss of thin surface filaments on transmission electron micrographs and a significant decrease in the surface attachment and biofilm formation by A. tumefaciens (Wang, Haitjema, & Fuqua, 2014). Type IV pili were also implicated in the interactions of the nitrogen-fixing plant endophyte Azoarcus sp. with the root surface of rice seedlings (Dorr et al., 1998) and strains of P. fluorescens (Vesper, 1987). The filamentous proteinaceous appendages known as fimbriae represent yet another primary root attachment factor deployed by plant-associated bacteria. The analysis of transposon mutants of S. enterica revealed that this human pathogen uses the aggregative (agf) fimbriae as adhesins for the attachment to alfalfa roots (Barak, Gorski, Naraghi-Arani, & Charkowski, 2005).

Finally, some well-studied rhizobacteria rely on specialized adhesion proteins for their interactions with plant root surfaces. Certain plant growth-promoting species of Bacillus produce collagen-like proteins (CLPs), which are located on the bacterial cell surface. The site-directed mutagenesis of the CLP biosynthesis genes in B. amyloliquefaciens FZB42 caused defects in auto-aggregation and biofilm formation on roots of A. thaliana, suggesting that these adhesins play an important role in rhizosphere plant-microbe interactions (Zhao et al., 2015). Rhizosphere pseudomonads attach to plant roots using nonfimbrial large adhesion proteins secreted via a type I secretion system (T1SS) (Chagnot, Zorgani, Astruc, & Desvaux, 2013). The common features of these adhesins include an N-terminal secretion signal, a repetitive core domain, and a glycine-rich C-terminal part that facilitate, respectively, adhesion to hydrophobic and hydrophilic surfaces (El-Kirat-Chatel, Beaussart, Boyd, O’Toole, & Dufrene, 2014). Many strains of P. fluorescens and P. putida produce the 520-kDa large adhesion protein LapA, which is secreted by the LapEBC type I secretion system and remains associated with the cell-surface (Collins, Smith, Sondermann, & O’Toole, 2020). Mutational studies have revealed that LapA functions in the transition between reversible to irreversible adhesion, a key step leading to the formation of microcolonies and the transition to mature biofilms (Hinsa et al., 2003). Interestingly, the TolC-like outer membrane pore of the Lap T1SS has an agglutinin-like structure and also plays a role in the attachment of bacteria to surfaces.

5.2. The rhizosphere biofilm matrix

The establishment of rhizosphere microcolonies triggers repression of the motility genes and activation of genes involved in the biosynthesis of the biofilm matrix, creating a complex microenvironment fundamental to the biofilm lifestyle (Flemming et al., 2016). The matrix includes extracellular polymeric substances (EPS) such as exopolysaccharides, extracellular DNA (eDNA) and extracellular RNA (eRNA), proteins, lipids, amyloids, cellulose and minerals, but the exact EPS structure can vary both between and within species and between single- and mixed-species biofilms (Karygianni, Ren, Koo, & Thurnheer, 2020). The hydrated and viscous EPS matrix performs an important structural function by keeping cells adherent to each other and the bacterial cell surface. It promotes intercellular interactions and enhances tolerance to desiccation and antimicrobials. The EPS also affects diffusion of solutes and forms heterogeneous microenvironments by creating localized gradients of oxygen, pH, inorganic ions, nutrients, signalling molecules and other metabolites (Jo, Price-Whelan, & Dietrich, 2022).

The exopolysaccharide components of the biofilm matrix can be firmly or more loosely associated with the cell surface and vary in charge and adhesiveness depending on the chemical composition and chain conformation (Karygianni et al., 2020). Although some species rely on a discrete type of exopolysaccharide for surface adhesion, most rhizobacteria are capable of producing multiple exopolysaccharides, which provides adaptability under changing environmental conditions. For example, Bacillus spp. carry the epsA–O operon encoding a biosynthetic pathway for an exopolysaccharide composed of N-acetylgalactosamine, glucose and galactose (Branda, Gonzalez-Pastor, Ben-Yehuda, Losick, & Kolter, 2001). Studies of eps mutants have revealed a crucial role of this exopolysaccharide in the formation of Bacillus biofilms and swarming motility (Nagorska, Ostrowski, Hinc, Holland, & Obuchowski, 2010). Nitrogen-fixing rhizobia, some plant-associated Burkholderia and Pseudomonas, and the soilborne plant pathogens Dickeya dadantii and A. tumefaciens harbour different sets of bsc genes that control the production of cellulose, a linear polymer of β-(1→4)-d-glucose (Romling & Galperin, 2015). The bacterial cellulose and its derivatives constitute a significant component of the biofilm matrix and contribute to the colonization of plant surfaces and the ecological fitness of both beneficial and pathogenic rhizobacteria (Heindl et al., 2014; Nielsen, Li, & Halverson, 2011; Spiers & Rainey, 2005). In addition to cellulose, most pseudomonads produce other well-characterized exopolysaccharides such as alginate, a linear polymer of O-acetylated (1→4)-linked β-D-mannuronic acid and α-L-glucuronic acid (Remminghorst & Rehm, 2006). In plant-associated Pseudomonas, the alginate biosynthetic machinery is highly conserved, but the exopolysaccharide itself plays a minor role in the ability to form rhizosphere biofilms and resist desiccation stress (Bianciotto, Andreotti, Balestrini, Bonfante, & Perotto, 2001; Chang et al., 2007). Some rhizosphere strains of the P. fluorescens species complex also carry pathways for the biosynthesis of the exopolysaccharides Psl (comprised of pentasaccharide units of mannose, glucose, and rhamnose) and Pel (composed of partially acetylated N-acetylgalactosamine and N-acetylglucosamine) (Heredia-Ponce et al., 2021). Although the role of Pel and Psl as primary biofilm matrix components has been studied extensively in the opportunistic human pathogen P. aeruginosa (Mann & Wozniak, 2012), their contributions to biofilms formed in the rhizosphere remains to be established. However, a recent study of the biocontrol strain P. chlororaphis PCL1606 revealed that the Psl-deficient mutant was impaired in the formation of biofilms, niche competition, and pathogen suppression on avocado plants (Heredia-Ponce et al., 2021). Finally, strains of P. putida secrete a biofilm matrix comprised of alginate, bacterial cellulose (Bcs) and two species-specific exopolysaccharides, Pea and Peb (putida exopolysaccharides A and B) (Costa-Gutierrez, Adler, Espinosa-Urgel, & de Cristobal, 2022). As indicated above, alginate is involved in the response to water stress (Chang et al., 2007), while other exopolysaccharides seem to stabilize the biofilm structure under water-replete conditions. Functional studies have revealed that isogenic Bcs, Pea and Peb mutants are less competitive than the corresponding wild type parents, suggesting that these exopolysaccharides are important and not redundant for competitive rhizosphere colonization by P. putida (Martinez-Gil et al., 2013; Nilsson et al., 2011).

In addition to assorted exopolysaccharides, biofilms formed by rhizobacteria also contain eDNA released via cell lysis (Moshynets, Pokholenko, Iungin, Potters, & Spiers, 2022; Peng et al., 2020; D. Wang, Yu, Dorosky, Pierson, & Pierson, 2016). eDNA provides structural integrity to the biofilm matrix by interacting with other biopolymers and serves as a source of nutrients and genetic material for gene transfer (Flemming et al., 2016). Although most of these important functions were demonstrated under laboratory conditions, it is known that extracellular DNA is abundant in soil (Carini et al., 2016), which suggests its likely contribution to biofilms formed on plant roots. The integrity of mature biofilms is further strengthened by diverse protein adhesins such as the large adhesion protein LapF, which is required for intercellular interactions during the later phases of the development and maturation of P. putida biofilms (Martinez-Gil, Yousef-Coronado, & Espinosa-Urgel, 2010). The RTX adhesion protein MapA produced by many species of plant associated Pseudomonas functions in deeper layers of mature biofilms by supplying cells with oxygen and nutrients (Collins, Pastora, Smith, & O’Toole, 2020). Yet another interesting example of rhizobacterial adhesins is provided by the amyloid protein TasA of B. subtilis, which forms fibres that tether bacterial cells and strengthen the structural integrity of the biofilm matrix (Romero, Aguilar, Losick, & Kolter, 2010). The inactivation of tasA correlates with a drop in viability of B. subtilis under biofilm conditions, reduced biocontrol activity and decreased colonization of plant surfaces (Camara-Almiron et al., 2020; Chen et al., 2013).

5.3. Regulatory mechanisms that govern the biofilm lifestyle

The process of attachment to root surfaces is triggered in planktonic cells by specific environmental signals. In several well-studied rhizobacteria, root surface sensing triggers a switch between the motile and sessile lifestyle and is regulated by cyclic di-guanosine monophosphate (c-di-GMP). Low c-di-GMP levels are associated with the planktonic phase, whereas high levels correlate with surface-attached growth and biofilm formation (Jenal, Reinders, & Lori, 2017). Rhizobacteria often carry extensive sets of c-di-GMP-modulating enzymes, suggesting the importance of c-di-GMP homeostasis for the rhizosphere lifestyle. For example, A. tumefaciens has 33 genes encoding c-di-GMP metabolizing proteins, while genomes of some rhizobia and species of the Pseudomonas fluorescens complex carry over 50 such genes (Gao et al., 2014; Heindl et al., 2014; Romling, Galperin, & Gomelsky, 2013). The synthesis of this conserved second messenger is controlled by enzymes of the diguanylate cyclase (DGC) family that carry a conserved GGDEF domain (named after the conserved central sequence pattern) catalysing the condensation of two GTP molecules to form c-di-GMP (Dahlstrom & O’Toole, 2017). The DGCs often have complex structure and contain PAS, Cache or other domains that may allosterically regulate the activity of the catalytic GGDEF domains in response to specific ligands (Cruz, Huertas, Lozano, Zarate, & Zambrano, 2012; Galperin, Nikolskaya, & Koonin, 2001). Diguanylate cyclase activity is counterbalanced by phosphodiesterases (PDEs), which contain EAL or HD-GYP domains and degrade c-di-GMP (Galperin, Natale, Aravind, & Koonin, 1999; Simm, Morr, Kader, Nimtz, & Romling, 2004). There are also proteins that contain GGDEF and EAL domains that may exhibit both diguanylate cyclase and phosphodiesterase functions. Interestingly, a recent study identified citrate and succinate, which are present in plant root exudates (Mavrodi et al., 2021), as ligands sensed by CACHE domains of some diguanylate cyclases of P. fluorescens (Giacalone et al., 2018). This finding suggests a mechanism by which plant exometabolites directly influence rhizosphere colonization by modulating the intracellular c-di-GMP pools and helps to explain the induction of biofilm formation by root exudates (Martins, Medeiros, Lakshmanan, & Bais, 2017; Sharma, Saleh, Charron, & Jabaji, 2020; Yuan et al., 2015).

Cyclic-di-GMP exerts its biological effects by interacting with diverse effectors (e.g., riboswitches, transcription factors, and proteins with PilZ, GGDEF or EAL domains) that function in several important cellular processes (Jenal et al., 2017). Although the connections between c-di-GMP levels and these pathways were established in E. coli and several human pathogens, the presence of highly conserved homologous genes suggests the involvement of similar mechanisms in many rhizobacteria. One of these relevant cellular processes is represented by flagella-driven motility, which is downregulated in the presence of elevated c-di-GMP levels. In E. coli and S. enterica, the effector protein YcgR binds c-di-GMP and favours surface attachment by interfering with the function of the flagellar motor (Boehm et al., 2010; Paul, Nieto, Carlquist, Blair, & Harshey, 2010). YcgR activity is counterbalanced by the phosphodiesterase PdsH, which enables motor function by breaking down c-di-GMP and inactivating YcgR. Similar mechanisms control flagellar function in B. subtilis and some Pseudomonas spp. that carry YcgR homologues (Baker et al., 2016; Chen, Chai, Guo, & Losick, 2012).

Another group of c-di-GMP-regulated pathways participate in the synthesis of surface adhesins, appendages and components of the biofilm matrix. For example, c-di-GMP coordinates the formation of type IV pili that function in the irreversible surface attachment of E. coli and other Gram-negative and Gram-positive bacteria (Bordeleau et al., 2015; Jones et al., 2015; Kazmierczak, Lebron, & Murray, 2006; Skotnicka et al., 2016). An increase in the c-di-GMP concentration stimulates the synthesis of bacterial cellulose by activating BcsA/BcsB cellulose synthase (Lindenberg, Klauck, Pesavento, Klauck, & Hengge, 2013; Morgan, McNamara, & Zimmer, 2014). Finally, c-di-GMP also contributes to the production of the alternative exopolysaccharide poly-β1,6-N-acetyl-glucosamine (PGA) which, together with cellulose, functions as a key biofilm matrix component of E. coli. PGA biosynthesis is induced by short-chain fatty acids via the action of the BarA-UvrY two-component system and the small RNAs CsrB and CsrC that collectively antagonize the carbon storage regulator Csr, a master regulator that normally represses pga genes (Vakulskas, Potts, Babitzke, Ahmer, & Romeo, 2015). In addition, this regulatory mechanism enables the expression of the diguanylate cyclases DgcT and DgcZ, increasing intracellular levels of c-di-GMP. This, in turn, allosterically activates the PGA secretion machinery including the PgaC glycosyltransferase that functions in the biosynthesis and export of the PGA polymer (Steiner, Lori, Boehm, & Jenal, 2013). A transposon mutagenesis screen of the model rhizobacterium P. protegens Pf-5 identified several strains with altered biofilm formation, swimming and swarming motility patterns (Ueda, Ogasawara, & Horiuchi, 2020). Among the characterized mutants was a strain carrying a transposon insertion in a gene encoding a GGDEF domain-containing protein.

As biofilms mature, the growing shortage of nutrients and oxygen depletion act as important cues that reduce adhesiveness, increase motility, and trigger the dissolution of extracellular polymeric substances (Rumbaugh & Sauer, 2020). These processes help rhizobacteria to escape the biofilm matrix and return to the planktonic mode of growth needed to locate and colonize a better niche. In contrast to relatively well characterized biofilm formation mechanisms, the processes driving biofilm dispersion are less understood. However, a few distinct biofilm dissolution strategies have been described in rhizosphere pseudomonads. For example, in P. fluorescens and P. putida, the escape mechanism involves the large adhesion protein LapA, which participates in surface adhesion and biofilm stabilization (see Section 5.1). During biofilm growth, the inner membrane protein LapD binds c-di-GMP and sequesters the periplasmic protease LapG (Newell, Monds, & O’Toole, 2009). When c-di-GMP levels drop, the LapD protein is inactivated, permitting the protease to cleave the periplasmic domain of LapA, thereby releasing the adhesin and weakening the biofilm. In the Lap regulatory system, the reduction in c-di-GMP levels is mediated by the phosphodiesterase BifA, which is activated when bacteria begin to adapt to nutrient starvation (Jimenez-Fernandez, Lopez-Sanchez, Calero, & Govantes, 2015). This process, known as the stringent response, involves rapid growth arrest and physiological adjustment accompanied by the accumulation of guanosine tetra- and pentaphosphate (abbreviated collectively as (p)ppGpp) (Dalebroux & Swanson, 2012). The accumulation of (p)ppGpp induces the phosphodiesterase gene bifA, leading to a decrease in the concentration of c-di-GMP and ultimately, inactivation of the adhesin LapA (Jimenez-Fernandez et al., 2015). Interestingly, the c-di-GMP-mediated release of LapA may also involve the phosphodiesterase RapA, a member of a large set of genes responding to the availability of phosphate and known as the Pho regulon (Collins, Smith, et al., 2020). RapA-based regulation triggers the biofilm dispersion and the transition to planktonic growth mode in response to low phosphate conditions. Other mechanisms that contribute to the disintegration of the biofilm matrix involve the breakdown of eDNA by endonucleases and degradation of polysaccharides by glycoside hydrolases (Cherny & Sauer, 2019, 2020; Kaplan, Ragunath, Velliyagounder, Fine, & Ramasubbu, 2004).

Recent genome surveys have revealed that, in addition to c-di-GMP metabolizing enzymes, members of many eubacterial and archaeal lineages carry proteins involved in the metabolism of a second messenger cyclic di-adenylate monophosphate (c-di-AMP) (de Purificação, de Azevedo, de Araujo, de Souza, & Guzzo, 2020). The genomic and experimental data also suggest that many bacteria use both c-di-GMP and c-di-AMP to relay environmental cues to downstream receptors regulating fundamental physiological processes including biofilm formation (Corrigan & Grundling, 2013). Among processes regulated by the c-di-AMP-based signalling are cell wall homeostasis, synthesis of the biofilm matrix and secretion of eDNA (Bowen & Koo, 2011; Corrigan, Abbott, Burhenne, Kaever, & Grundling, 2011; DeFrancesco et al., 2017). In rhizobacteria, the contribution of c-di-AMP to the formation of in planta biofilms has been demonstrated in Gram-positive Firmicutes. For example, Townsley, Yannarell, Huynh, Woodward, & Shank (2018) demonstrated a link between intracellular levels of c-di-AMP and colony morphology, biofilm gene expression and colonization of Arabidopsis thaliana roots by B. subtilis NCIB3610. The study employed a set of isogenic mutants deficient in genes with predicted di-adenylate cyclase and phosphodiesterase activities. The authors also identified two transporters, YcnB and YhcA, involved in the secretion of c-di-AMP. A mutant strain lacking both permeases secreted less c-di-AMP and was deficient in the attachment to plant surfaces. To what extent such c-di-AMP and other nucleotide-based signalling contributes to the establishment of biofilms in other groups of rhizobacteria remains to be determined.

6. The importance of rhizosphere biofilms

It is clear that the biofilm mode of growth represents a key facet of the bacterial survival and persistence in the environment and the ability to interact with diverse eukaryotic hosts (Flemming et al., 2016). Bacterial biofilms are viewed as crucial contributors to many important animal and plant diseases since biofilm-residing pathogens efficiently colonize their hosts and are resilient to immune responses and antimicrobial agents. Similar ideas have been advanced in the field of rhizosphere research, where biofilms are considered an important trait contributing to the efficient colonization and persistence on root surfaces, and the formation of biofilms in response to root exudates, antimicrobials or stress factors was demonstrated for both beneficial and pathogenic rhizobacteria (Bais, Fall, & Vivanco, 2004; Espinosa-Urgel, Kolter, & Ramos, 2002; Haggag & Timmusk, 2008; Heindl et al., 2014; Selin et al., 2010; Timmusk, Grantcharova, & Wagner, 2005; Zhang et al., 2015). These observations are supported by the discovery of numerous often redundant genes that govern the transition from the planktonic to the sessile mode of growth or control the synthesis of biofilm matrix components and are required for the efficient colonization of root surfaces and persistence in the rhizosphere (see studies cited in Section 5). However, these diverse biofilm formation determinants must be carefully regulated, since isogenic mutants with altered motility or enhanced in vitro biofilm formation often exhibit rhizosphere fitness defects (Liu et al., 2018). Based on these findings, it has been suggested that the ability to form biofilms should be taken into account during the selection of biological control agents or strains with plant growth promotion properties (Pandin, Le Coq, Canette, Aymerich, & Briandet, 2017).

Many rhizobacteria are capable of communicating with their kin or other species by producing and responding to small extracellular molecules via a process called “quorum sensing” (QS) (Mukherjee & Bassler, 2019). The accumulation of signalling molecules is proportional to the density of producing cells, making QS a mechanism by which microorganisms can monitor their own numbers and engage in activities that require certain population levels to work effectively. Biofilms provide a closed structured environment that facilitates intercellular communication due to the ability of matrix components to bind, spatially restrict and concentrate quorum sensing signalling molecules (Seviour et al., 2015). These QS signals include N-acylhomoserine lactones (AHLs) of Gram-negative bacteria, autoinducer peptides of Gram-positive bacteria, furanosyl borate diester (AI-2) that mediates inter-species communication found in both groups, as well as other types of signalling molecules (Aframian & Eldar, 2020). It has been estimated that the levels of QS signals can be orders of magnitude higher in biofilms compared to environments inhabited by planktonic cells (Hense et al., 2007), which may explain why many quorum sensing phenotypes are specific to the biofilm mode of growth. Cellular pathways regulated by QS function occur in surface motility, bioluminescence, stress tolerance, competence for DNA uptake, and secretion of virulence factors (Bassler & Losick, 2006; Von Bodman, Bauer, & Coplin, 2003). Many of these QS-dependent traits are present in beneficial and pathogenic bacteria inhabiting the plant rhizosphere. Finally, in many beneficial rhizobacteria, quorum sensing regulates the production of diverse secondary metabolites that have been linked with survival in highly competitive niches and the ability to suppress soilborne diseases (Danhorn & Fuqua, 2007; Pandin et al., 2017).

Interestingly, QS signals produced by plant-associated bacteria also directly affect the growth and physiology of the host plant. Several studies documented the uptake of water-soluble AHLs by plant roots and their systemic transport through the vascular system (Gotz et al., 2007; Sieper et al., 2014). The exposure of plants to these AHLs is correlated with changes in root length, number of lateral roots and phytohormonal balance (Ortiz-Castro, Martinez-Trujillo, & Lopez-Bucio, 2008; Schenk et al., 2014; von Rad et al., 2008). The treatment of A. thaliana roots with different AHLs induced several defence-related genes and primed resistance to the hemibiotrophic foliar pathogen Pseudomonas syringae pv. tomato (Shrestha, Grimm, Ojiro, Krumwiede, & Schikora, 2020). Similarly, treatment of barley and Arabidopsis with 3-oxo-C₁₄-HSL conferred resistance to the biotrophic fungus Golovinomyces orontii, the barley powdery mildew pathogen Blumeria graminis, and P. syringae pv. tomato DC3000 (Schikora et al., 2011). Interestingly, in Arabidopsis, defence priming was accompanied by activation of the protein kinase AtMPK6 and was compromised in a mpk6 mutant. These inducible resistance effects were also observed in experiments with QS sensing rhizobacteria. For example, the colonization of tomato by C₆- and C₈-HSL-producing Serratia liquefaciens MG1 and P. putida IsoF primed defence against the fungal pathogen Alternaria alternata (Schuhegger et al., 2006). Both AHL-producing bacteria increased the levels of salicylic acid and induced ethylene-dependent PR1a and chitinase defence genes. The effect was absent from plants treated with AHL-deficient mutants and could be reproduced by the application of C₆- and C₈-HSLs. In another interesting study, the inoculation of two barley lines with the 3-oxo-C₁₄-HSL-producing rhizobacterium Ensifer meliloti reduced feeding and the reproductive response of the bird cherry-oat aphid Rhopalosiphum padiaphids compared to the AHL-deficient mutant (Wehner, Schikora, Ordon, & Will, 2021).

Yet another interesting aspect of rhizosphere biofilms relates to the ability of plant-associated bacteria to alleviate water stress in plants. Recent studies have provided clear evidence that drought-adapted bacterial strains and soil microbial communities positively influence root length, biomass, leaf surface area, accelerate flowering and increase seed yield in plants under water-stressed conditions (Bhattacharyya et al., 2021). Among mechanisms mediating the ability of rhizobacteria to provide drought relief is the production of hydrating biofilm matrices that can absorb and retain a significant amount of water (Chenu, 1993). Soil drying stimulates the production of bacterial exopolymeric substances (EPSs), as was shown by LeTourneau et al. (LeTourneau et al., 2018), who employed a combination of stable isotope probing, nano-scale resolution secondary ion mass spectrometry (NanoSIMS) and bioreporters to investigate the release of EPS by P. synxantha 2–79 in the rhizosphere of dryland wheat. In 2–79 and certain other pseudomonads, biofilm formation is influenced by the synthesis and secretion of redox-active phenazines, which act as extracellular electron shuttles. The authors demonstrated that under the conditions of water deficit, the wild-type strain formed robust rhizosphere biofilms and secreted significantly more EPS than the phenazine-deficient mutant, suggesting the importance of hydrating extracellular matrices for the ability of rhizobacteria to resist drought stress. Soil moisture influenced microbial biomass turnover and rates of incorporation of bacterial ¹⁵N into plant root tissues.

The application of EPS-producing rhizobacteria improves the uptake of water and nutrients, tolerance to increased salinity, as well as growth promotion and survival under drought stress. These effects were demonstrated for numerous combinations of rhizobacteria and plants including sunflower seedlings treated with Pseudomonas sp. GAP-P45 (Sandhya, Ali, Grover, Reddy, & Venkateswarlu, 2009), quinoa inoculated with halotolerant Enterobacter and Bacillus (Yang et al., 2016), chickpea bacterized with strains of Halomonas and Planococcus (Qurashi & Sabri, 2012), and maize inoculated with EPS-producing Proteus, Pseudomonas, and Alcaligenes and their exopolysaccharides (Naseem & Bano, 2014). The beneficial effects of microbial EPSs include the formation of soil aggregates that decrease evaporation, enhance moisture retention, and maintain the flow of water and nutrients to the roots of drought- and salt-stressed plants (Woo Suk Chang et al., 2007; Guo et al., 2018; Schmid, Sieber, & Rehm, 2015). In the rhizosphere, the bacterial EPSs combine with plant mucilage and bind soil particles contributing to the formation of rhizosheaths. First discovered in desert grasses, the rhizosheaths were later described in many crop plants such as wheat, barley, and corn (Alsharif, Saad, & Hirt, 2020). Their development correlates positively with water uptake and nutrient acquisition in dry soils and is considered an important adaptive mechanism by plants to tolerate drought stress.

7. Conclusions

It is now apparent that the complex microbial biofilms associated with roots are an essential interface between plants and the environments in which they reside. The dense microbial populations that inhabit these biofilms orchestrate plant growth and resistance to a myriad of biotic and abiotic stresses via molecular signals that mediate communication not only locally, but also throughout the plant. New tools have become available that circumvent some of the challenges inherent in characterizing rhizosphere biofilms in soil, revealing intimate details of biofilm form and function and providing insights that complement and broaden what knowledge can be gleaned from other experimental systems. This progress in understanding rhizosphere biofilms has opened the way to new questions about the diversity of cues and signals that mediate plant well-being across environments and the resilience of these systems in the face of environmental change.

Abbreviations

ACC

1-aminocyclopropane-1-carboxylate

AHLs

N-acylhomoserine lactones

AM

arbuscular mycorrhizae

c-di-AMP

cyclic di-adenylate monophosphate

c-di-GMP

cyclic di-guanosine monophosphate

CLSM

confocal laser scanning microscopy

DGC

diguanylate cyclase

eDNA

extracellular DNA

EPS

extracellular polymeric substances

FDR

false discovery rate

FISH

fluorescence in situ hybridization

ISR

induced systemic resistance

MCP

methyl-accepting chemotaxis protein

(p)ppGpp

guanosine tetra- and pentaphosphate

QS

quorum sensing

SAR

systemic acquired resistance

T1SS

type I secretion system

TRIS

tracking root interactions system

UPP

unipolar polysaccharide