OUP user menu

Causes and consequences of plant-associated biofilms

Thimmaraju Rudrappa , Meredith L. Biedrzycki , Harsh P. Bais
DOI: http://dx.doi.org/10.1111/j.1574-6941.2008.00465.x 153-166 First published online: 1 May 2008

Abstract

The rhizosphere is the critical interface between plant roots and soil where beneficial and harmful interactions between plants and microorganisms occur. Although microorganisms have historically been studied as planktonic (or free-swimming) cells, most are found attached to surfaces, in multicellular assemblies known as biofilms. When found in association with plants, certain bacteria such as plant growth promoting rhizobacteria not only induce plant growth but also protect plants from soil-borne pathogens in a process known as biocontrol. Contrastingly, other rhizobacteria in a biofilm matrix may cause pathogenesis in plants. Although research suggests that biofilm formation on plants is associated with biological control and pathogenic response, little is known about how plants regulate this association. Here, we assess the biological importance of biofilm association on plants.

Keywords
  • biofilm
  • roots
  • rhizosphere
  • rhizobacteria
  • Bacillus subtilis
  • Arabidopsis thaliana

Introduction

The effect of plant root secreted chemicals on microbial populations in the soil compartment influenced by the root, known as ‘the rhizosphere,’ has been well established (Grayston et al., 1998; Bais et al., 2006). As a result, a number of important beneficial (Biswas et al., 2000; Dong et al., 2001; Ryu et al., 2003, 2004) and harmful interactions (Hawes & Smith, 1989; Timmusk et al., 2005) between plants and microorganisms, localized in the rhizosphere have been studied. From the earliest observations by Rovira et al. (1974) on the colonization of microorganisms in the rhizosphere, to the present, it has been established that most microorganisms often exist in organized structures called biofilms when colonizing different plant surfaces, in contrast to their more well-studied planktonic states (Kearns et al., 2005; Stanley & Lazazzera, 2005), (Table 1). However, with an increasing number of studies on the biochemistry, physiology and genetics of such organized structures (Kearns et al., 2005; Stanley & Lazazzera, 2005), it has become clear that bacterial interactions, growth and formation of biofilms on the root surface involve complex mechanisms (Rudrappa et al., 2007). Although biofilm formation on various biotic and abiotic surfaces has been given much attention in the past (Kearns et al., 2005; O'Toole & Stewart, 2005; Stanley & Lazazzera, 2005), the study of bacterial biofilm formation on root surfaces has not been as prevalent (Bais et al., 2004; Walker et al., 2004; Timmusk et al., 2005; Fujishige et al., 2006; Rudrappa et al., 2007). Nonpathogenic plant growth promoting rhizobacteria (PGPRs) associated with plant root surfaces are known to contribute towards increases in plant yield by mechanisms such as improved mineral uptake, phytohormone production (Biswas et al., 2000; Ryu et al., 2003, 2004; Fujishige et al., 2006) and competitive suppression of pathogens by production of antibiotics (Mazzola et al., 1992) and induction of secondary metabolite-mediated systemic resistance (Ryu et al., 2004). Apart from the root surface, microbial biofilms are also reported on the phyllosphere and vasculature, especially by pathogenic microorganisms (Torres et al., 2006). For example, the pathogenicity of Xylella fastidiosa, which causes Pierce's disease of grapes and citrus fruits, has been attributed to its ability to form extensive biofilms, thereby blocking the plant vasculature (Torres et al., 2006).

View this table:
1

Bacteria known to colonize various plant part surfaces of different plant species

Until 1995, it was generally believed that plants were incapable of being infected by human pathogens. This view changed when Pseudomonas aeruginosa PA14, a human pathogen known to infect burn and cystic fibrosis patients, was recognized as a potent foliar pathogen of a variety of plants, including the model plant Arabidopsis thaliana (Rahme et al., 1995). Additionally, it was determined that many of the same P. aeruginosa virulence factors required for animal pathogenesis are also essential for plant pathogenesis (Rahme et al., 1995). Among the many virulence factors, the ability to form biofilms in P. aeruginosa is crucial for its infection of multiple hosts (Tan et al., 1999). In this review, we describe the mechanisms involved in the initiation and development of plant-associated microbial biofilms in general, with a special emphasis on the root-surface-associated biofilms. Furthermore, examples and references regarding bacterial biofilm formation on aerial plant parts as well as abiotic surfaces are included to enhance the understanding of biofilm formation in general, and to hypothesize and compare and contrast these biofilms to root-associated biofilms.

The plant root surface and microbial biofilm associations

Root biofilm initiation and development is complex and not well understood due to the dynamic nature of plant root surfaces. In addition to physico-chemical variations throughout the root surface, it is likely that other abiotic factors such as nutrient availability, temperature and relative humidity influence root biofilm associations (Stanley & Lazazzera, 2004). Yet, despite these challenges, diverse bacterial species have adapted to these ever-changing conditions and are capable of starting colonization by forming microcolonies on different parts of the roots from tip to elongation zone. Such microcolonies eventually grow into large population sizes on roots to form mature biofilms. Interestingly, root exudates serve as a major plant-derived factor responsible for triggering root colonization (Lugtenberg et al., 1999) and biofilm associations (Walker et al., 2004). It is estimated that plants secrete between 10% and 44% of their photosynthates as root exudates. These organic compounds may therefore provide a nutrient source for the encroaching microorganisms (Lugtenberg et al., 1999). The root surfaces of plants are continually subjected to the two-way traffic of solutes from plants to the soil and vice versa (Lugtenberg et al., 1999; Dakora & Phillips, 2002). A broad range of environmental factors cause fluctuations in root surface properties and this dynamic environment may therefore make it challenging for two-way communication between plants and microbial communities in the rhizosphere (Bais et al., 2002). This interaction becomes more complicated when more than one bacterium is involved, as observed in the case of multispecies microbial associations (An et al., 2006).

The concept of heterogeneous bacterial colonization has also been reported for the rhizosphere (Beattie & Lindow, 1999; Joseph et al., 2007), because it is known that a pH gradient exists along the root surface from the root tip to basal regions (Fischer et al., 1989). Additionally, it is believed that metabolite requirements and organic compound secretions fluctuate between the root tip and mature root regions; therefore the nutrients available for bacterial use will not be distributed homogeneously throughout the root plane (Hoffland et al., 1989; Eshel & Waisel, 1996). A schematic hypothetical model of the possible bacterial biofilm formation on the different root regions is depicted in Fig. 1a. It is known that the root tip is engaged in maximum carbon turnover, in terms of root secretions and cell debris formation, because of the frequent ‘decapping’ of the root cap (Hawes et al., 2000). It could be speculated that differential bacterial biofilm formation would be least expected at the root tip level compared with the mature root regions. As shown hypothetically in the schematic (Fig. 1a), the biofilm depth in the root tip is less than in mature root regions; this variation may be due to fluctuations in the composition of the root exudates and nutrient availability at the root plane or specific secretion of antimicrobials from the root tip. In addition, involvement of the point of emergence of lateral roots in secretion and subsequent chemoattraction of bacteria leading to microcolony formation (McDougall & Rovira, 1970; Cooley et al., 2003) may be the reason for increased biofilm thickness in mature regions of the root. Concomitantly, Shrout et al. (2006) showed that variations in available nutrients influence Pseudomonas aeruginosa bacterial swarming. A reduction in bacterial swarming is associated with abundant nutrient availability leading to a more ‘structured’ three-dimensional (3-D) biofilm. In contrast, lower nutrient levels influenced swarming of P. aeruginosa, resulting in more ‘flat’ biofilm structures (Shrout et al., 2006) (Fig. 1b). Similarly, our studies have shown that Bacillus subtilis strain FB17, a beneficial bacterium, exhibits differential biofilm formation on A. thaliana roots (Fig. 2) (T. Rudrappa & H.P. Bais, unpublished data), with an interesting pattern of biofilm formation on different root regions. Figure 2 shows that B. subtilis efficiently binds at the mature root region, resulting in a 3-D structural film formation compared with the root tip biofilms (T. Rudrappa & H.P. Bais, unpublished data). These results partially confirm the differential biofilm patterning theory shown in Fig. 1a and a previous report on Pseudomonas fluorescens root colonization (Humphris et al., 2005). However, it is not known whether specific organic compounds secreted by roots also influence biofilm structures and future studies are required to isolate and characterize these compounds from root exudates to unravel more interesting interactions that facilitate or inhibit rhizosphere biofilm formation.

1

Variations in root-biofilm structure. (a) Differences in 3-D shape of root-biofilms structures dictated by nutrient availability. The upper region depicts biofilm formation in xylem tissues. The biofilm in this region aggregates across the xylem cylinder to absorb nutrients from passing xylem fluids (Koutsoudis et al., 2006). The middle region depicts a mature root region where possible nutrient fluctuations result in a more structured biofilm (Shrout et al., 2006). The lower region illustrates the root tip where lower nutrient availability and possible secretion of antimicrobials allows for a more ‘flat’ biofilm (Shrout et al., 2006). (b) Biofilm structures on an abiotic surface (low nutrient availability) are flat compared with artificially amended nutrient-rich media resulting in more structured biofilm. Note: the figure corresponds to a hypothetical model.

2

Differential biofilm patterning of Bacillus subtilis (strain FB17) biofilm on Arabidopsis thaliana (Col-0) root surface. The figure shows less or no colonization on the root tip (1), followed by slight attachment (arrows) on a central elongation zone (CEZ) (2), region slightly above the CEZ (3) and thick biofilm formation on the mature root surface (4). Magnification of the mature region with a biofilm (5), and its cross section (6) indicating the 3-D depths of the biofilm. Panels 7 and 8 indicate magnification of the root tip and its cross section respectively, revealing no bacterial colonization at the root tip. The square boxes (Panel 1 and 4) show the region magnified and the circles indicate the region cross-sectioned across the Z-axis using a confocal scanning laser microscope. The solid black arrows between the images indicate the progression from the root tip to mature region of the root and the red arrows shows the regions magnified and/or cross sectioned (scale bars=100μm).

The role of bacterial autoinducer signaling and plant mimicry on biofilm initiation and development

Microbial colonization is generally regulated in a population density dependent manner by quorum sensing (QS) (Whitehead et al., 2001). It has been reported that the cells, use autoinducers (AI) to measure limitations to mass transfer or diffusion known as diffusion sensing (DS) (Redfield, 2002). However, the more recent efficiency sensing (ES) theory (Hense et al., 2007), which unifies QS and DS, is most applicable in the context of plant-root-associated bacteria, which will experience spatial distribution and mass transfer limitations under natural settings. In accordance with this hypothesis root-associated bacteria might form small aggregates of clonal cells (microcolonies). Once established, there will be ES between microcolonies especially in the case of sessile bacteria. This leads to expression of specific genes that allows microcolonies/bacteria to coordinate with each other in a manner reminiscent of a multicellular organism (Whitchurch et al., 2002; Kearns et al., 2005; Kolter, 2005) as referred for QS. A number of autoinducer molecules have been identified in both Gram-negative and Gram-positive bacteria (Miller & Bassler, 2001; Lu et al., 2005). Many Gram-negative bacteria produce acyl-homoserine lactones (AHLs) which are known to regulate QS behavior and biofilm formation (Whitehead et al., 2001). Interestingly, it has been reported that AHL production is more frequent in fluorescent pseudomonads isolated from the rhizosphere than in isolates from the bulk soil (Elasri et al., 2001). Diverse polysaccharide and autoinducers production in different strains of organisms allows for variations between biofilm structures (Mack et al., 1996; Miller & Bassler, 2001; Gotz, 2002; Waters & Bassler, 2005). The discussion of biofilms thus far has been supported by studies regarding bacterial aggregation on homogeneous abiotic surfaces, sewage treatment plants and a few plant and animal tissues, but complexity increases for bacterial colonization in the rhizosphere.

As a complex and dynamic organ, the root controls various biochemical and physiological processes that are crucial for the survival of the plant (Mercier et al., 2001; Boru et al., 2003). Among such critical processes, regulation of microbial recruitment and dynamics are most vital. It has been reported that plants regulate microbial processes by deterring pathogenic microorganisms and selectively attracting beneficial microorganisms (Ramey et al., 2004a). How these microorganisms establish themselves as communities on the root surface is a critical question because, like their plant counterpart, microorganisms are equally dynamic and employ various mechanisms to cope with changed conditions (Langer et al., 2004), such as multidrug resistance pumps (Chuanchuen et al., 2001).

As previously mentioned, plant roots exude photosynthates which include many small molecular weight compounds (Bais et al., 2006). Recent studies have detected specific root-secreted compounds in the rhizosphere that act as plant QS mimics (Teplitski et al., 2000; Mathesius et al., 2003; Keshavan et al., 2005). A study has demonstrated that methanol extracts of pea root exudates activated reporter genes tagged to LuxRI, AhyRI and LasRI QS promoters, indicating that roots are capable of secreting compounds that may be structurally similar to Gram-negative AHLs (Teplitski et al., 2000). Additional evidence for the ability of roots to secrete AHLs has been shown by engineering tobacco and potato to secrete AHLs in the rhizosphere (Toth et al., 2004; Scott et al., 2006). Apart from secretion of AHLs and AHL mimics, plant-derived compounds also influence biofilm formation by interfering with the bacterial QS mechanism (Rasmussen et al., 2005). One system for which chemical and molecular evidence for QS inhibition has been identified is that of the primitive plant, Delisea pulchra, a marine red alga (Yoon et al., 2006). Delisea pulchra produces structural analogs of AHLs, halogenated furanones, which bind competitively to AHL receptors, instigating proteolytic degradation and inhibition of associated QS signals (Teplitski et al., 2004).

While considering natural encounters between plants and bacteria, the disruption of QS regulation by other rhizosphere bacteria may be as important as disruption by QS mimics from the host plant. The best example is the production of AHL lactonases (which degrade AHLs) by the soil bacterium Bacillus thuringiensis (Dong et al., 2001). Various bacterial species that produce AHL-QS signals have been found to activate gene expression in a Pseudomonas reporter strain in native wheat rhizospheres (Steidle et al., 2001; Whitehead et al., 2001). This activation suggests that there is significant QS cross-talk between different bacterial species on plant roots, thereby regulating the outcome of root-associated biofilm formation. Evidence suggests that AHL-responding bacteria do not need to be particularly close to the AHL-producing cells on the host root surface (Pierson et al., 1998; Gantner et al., 2006). Except for the secretion of AHLs/AHL like molecules, there is no evidence to date, however, to suggest that plants make or use AHL-degrading enzymes, and such enzymes might hinder bacterial associations. However, the possibility exists to engineer plants for the production of QS-degrading enzymes, which can considerably change the scenario of root-pathogen interactions (Dong et al., 2001). Thus, the rhizosphere may be considered as a region of overlapping, communicating bacterial populations, each defined by mutual recognition of specific QS signals, and each affected in different ways by the secretion of host plant QS mimics. In addition to their associations with QS mechanisms, plants may also regulate bacterial associations by influencing the structure of biofilms attached to their root surface by varying rhizosphere nutrient status, as suggested by abiotic surface studies (Shrout et al., 2006).

Effects of bacterial and plant-derived secretions on biofilm formation

Bacterial production of secondary metabolites and other bacterial physiological processes are known to occur in a population density dependent manner (Johnson et al., 2005; Barnard et al., 2007). Therefore, it is possible that the benefits available through root-associated biofilms are also dependent upon population density. Recently, it has been demonstrated that the biocontrol ability of B. subtilis against Pseudomonas syringae pv. tomato DC3000 was facilitated by root-surface biofilm formation and the production of surfactin (Bais et al., 2004). Unlike PGPRs, which have been reported to colonize plant root surfaces (Chin-A-Woeng et al., 2001; Ramey et al., 2004b), the symbiotic nitrogen fixing bacteria form infection threads, bacteriodes and diverse biofilm architecture (Gage, 2002, 2004). Additionally, some PGPRs also produce antimicrobial secondary metabolites, which target the competing microorganisms (Mazzola et al., 1992; Raaijmakers et al., 2002; Haas & Keel, 2003). Plant-secreted factors promoting these interactions have been identified in legume-Rhizobium associations (Table 2). In other PGPRs such as some species of Bacillus, Pseudomonas, and Azospirillum, the key mechanisms involved in biofilm formation are different. For example, flagellar motility (Matthysse & McMahan, 2001), cell surface protein (Lugtenberg et al., 2002) and production of polymers/polysaccharides (Hinsa et al., 2003) may be important in colonization. These studies suggest the existence of a plant-derived mechanism in which epiphytic root colonization and biofilm formation are triggered by a plant-derived component similar, as in the legume-Rhizobium and other bacterial interactions (de Ruijter et al., 1999).

View this table:
2

A selection of plant root secreted compounds implicated in induction of nod genes and promoting legume-Rhizobium interactions

The factors influencing biofilm formation are most likely diverse, including proteins, secondary metabolites, organic acids, amino acids and small peptides (Charon et al., 1997). These factors may function in a differentially selective manner to enhance the competitive ability of a particular species from a heterogeneous rhizosphere microbial community when compared with bulk soil (Small et al., 2001). Possible selective mechanisms include any plant root secreted molecule(s) with a bacterial cell wall receptor or a diffusible factor that which interact(s) with the regulatory proteins in the bacterial cytoplasm, selectively activating key biofilm genes. Plant roots which create a rhizosphere enriched with a specific PGPR for its own advantages may utilize such mechanisms. The molecular mechanisms involved in many of the above named processes still require elucidation.

Root surface chemistry and bacterial biofilm formation

Recently, through the use of various A. thaliana disease-resistance mutants, it has been shown that B. subtilis strain FB17 is not perceived by plant roots through any of the known disease resistance pathways (Rudrappa et al., 2007). Interestingly, B. subtilis colonization is dependent on root surface chemistry and the generation of reactive oxygen species (ROS) on the root surface. Knowledge about the functional activity of ROS in nonpathogenic plant–microorganism associations is scarce. In pathogenic plant–microorganism interactions, ROS activate signal transduction pathways that regulate cell proliferation and modification of the extracellular matrix and are regarded as positive regulators of disease resistance (Suh et al., 1999; Lambeth, 2004; Nasr et al., 2005). Recently, ROS has been implicated in the regulation of mutualistic interactions between the fungus Epichloe festucae and a perennial ryegrass (Lolium perenne) (Shin et al., 2004).

Reduced biofilm formation on A. thaliana NahG roots [salicylic acid (SA) deficient transgene] has been observed (vanWees & Glazebrook, 2003), accompanied by an increased production of catechol on the root surface. Elevated catechol concentrations resulted in increased basal ROS production, which further mediates the down regulation of biofilm formation genes in B. subtilis (Rudrappa et al., 2007). This is surprising but not unexpected because SA is involved in balancing the total redox potential in different plant systems (e.g. rice and tomato) (Yang et al., 2004a). Therefore, it is possible that its degradation product, catechol, generates higher levels of ROS in NahG plants (Rudrappa et al., 2007). Studies have demonstrated in rice and Arabidopsis systems that higher catechol levels result in the production of superoxide anions and H2O2 leading to increased ROS generation (Kojo et al., 2006; Tanaka et al., 2006). A fundamental question in this regard is whether the biofilm formation and bacterial colonization, per se, are significantly impacted by varying ROS levels in the rhizosphere? This concept is supported by data that indicate that plants respond to different biotic and abiotic stimuli by producing higher endogenous levels of ROS (vanWees & Glazebrook, 2003). Additionally, nitric oxide (NO) (an important plant defense signal) is involved in the dispersal of P. aeruginosa biofilms, although data were obtained for biofilm colonization on an abiotic glass surface (Barraud et al., 2006). Nevertheless, superoxide dismutase produced by bacteria in the rhizosphere might also play a role in altering the overall ROS level in addition to plant-directed process (Barloy-Hubler et al., 2004). Future studies are required to elucidate the novel plant genes and signaling events that inhibit the formation of beneficial biofilms on plant roots.

Effects of bacterial and root produced chemicals on bacterial attachment and binding

As mentioned earlier, bacterial attachment is the preliminary step in bacterial biofilm formation on plants, and several mechanisms have been suggested (Leite et al., 2002; de Souza et al., 2004). During X. fastidiosa biofilm initiation the exposed thiol groups impart a net negative charge to the bacterial cell surface promoting divalent ion bridging between the host cells and aiding in bacterial attachment (Leite et al., 2002; de Souza et al., 2004). However, other plant pathogenic bacteria have been shown to utilize different mechanisms for attachment on host plant surfaces. For example, Ralstonia solonacearum uses type-IV pili for surface attachment and motility while virulence is dependent on exopolysaccharides production, causing lethal wilt in many plants (Leite et al., 2004). The role of plant root surface lectins and their interaction with bacterial exopolysaccharides in bacterial attachment have been studied in detail in Rhizobium-legume interactions (see review by Hirsch, 1999) and Azospirillum (Burdman et al., 1998; Rodríguez-Navarro et al., 2007).

Although many studies highlight the importance of bacterial exopolysaccharides production for effective colonization on root surfaces, few studies have considered the role of root surface chemicals and polysaccharides on initial attachment and binding (Burdman et al., 1998; Hirsch, 1999; Rodríguez-Navarro et al., 2007). Therefore, this topic is worth addressing in detail, as the possibility exists for complex interactions in terms of initial attachment and binding between other plant and bacterial species. The possibility for these undiscovered interactions increases immensely when considering the abundance of small molecular weight compounds on the root surface (Bais et al., 2006). Many polysaccharides produced by bacteria modulate the chemical and physical properties of P. aeruginosa biofilms on abiotic surfaces (Friedman & Kolter, 2004) and the plant might also secrete specific compounds, which can suppress pathogenic interactions by reducing attachment and binding (Teplitski et al., 2000).

Root-exuded biofilm inhibitors and antimicrobials

A major concern in developing an antimicrobial therapy against opportunistic pathogens like P. aeruginosa is their inherent potential to develop antibiotic resistance through biofilm formation. Disturbingly, recent evidence also suggests that some aminoglycoside antibiotics induce biofilm formation in P. aeruginosa at subinhibitory concentrations (O'Toole & Stewart, 2005). Conversely, plants combat these opportunistic pathogens by secreting antimicrobials through roots (Walker et al., 2004). To counter these antimicrobials, microorganisms in the biofilm stage synthesize compounds which can bind and sequester antibiotics in the periplasm. The sequestering of antibiotics prevents their through the periplasmic space to their site of action in the cytoplasm as reported for P. aeruginosa (Mah et al., 2003). Alternatively, if these compounds diffuse in a controlled manner or can degrade the established biofilm matrix, they could potentially be more effective. Very few studies have targeted the complex possibilities of this rhizosphere interaction. Rosmarinic acid (RA) secreted by Ocimum basilicum (sweet basil), was potent against the planktonic form of P. aeruginosa, therefore preventing biofilm initiation, but lacked the ability to penetrate mature biofilms on the root surfaces (Walker et al., 2004). This could be because the pathogen might acquire resistance to the secretions either by altered lipopolysaccharide production, enhanced expression of antibiotic efflux pumps or by detoxification of the antimicrobial compound (Chuanchuen et al., 2001). The above-mentioned lines of research encourage the concept that specific root secretions may function as natural plant pathogenic biofilm inhibitors. Screening of root exudates for potential biofilm inhibitors against several plant pathogens may lead to the discovery of the regulatory pathways controlling biofilm formation in plant pathogens by identifying compounds that inactivate bacterial binding and/or QS.

Some human food-borne pathogens can also colonize plant surface and tissues and cause enteric illness upon consumption of the contaminated plant product. According to the World Health Organization (Fact sheet No. 237, http://www.who.int/mediacentre/factsheets/fs237/en/), 76 million cases of food borne diseases occur annually in the United States. Several of the pathogens responsible for these epidemics form biofilms on aerial plant surfaces, and these studies may provide insight on how they would infect root tissues. Enteric bacteria such as Salmonella enterica and Escherichia coli are most often associated with produce crops including sprouts, fruits, and leafy vegetables (Hogan & Kolter, 2002). Their ability to form biofilms on the plant surfaces prevents removal of the bacteria during sanitizing procedures and allows colonization of interior plant surfaces (Cooley et al., 2003; Brandl, 2006; Kutter et al., 2006). The recent issue of spinach contaminated with E. coli O157:H7 is one of the best examples (http://www.fda.gov/oc/opacom/hottopics/spinach.html). Several of the genes in S. enterica responsible for biofilm formation and virulence in humans are those that cause attachment to plant surfaces (Cooley, 2007).

With all of the aforementioned human pathogens, extensive research will be needed to fully understand how these bacteria operate in diverse hosts and inflict virulence through biofilm formation. Continued research into plant-associated bacteria, identification of genes responsible for biofilm formation and possible strategies for preventing or curing human illness should spearhead future research.

Multitrophic interactions

In addition to human pathogenic-plant-associated bacteria, a deeper description of multitrophic interactions is necessary to understand the life cycles and virulence factors in many biofilm-forming bacteria. As mentioned above, an example of a tritrophic interaction involves E. coli that survives and form biofilms on plants, and can later infect humans and other higher animals (Cooley et al., 2003; Annous et al., 2005).

Biofilms involved in multitrophic interactions are economically important for several agricultural crops. An interesting example of a plant-associated biofilm derived through multitrophic interactions is the previously discussed X. fastidiosa; although this example is specific to aerial portions of the plant, it is likely that root pathogens colonize through a similar means. Xylella fastidiosa relies on an insect vector to transfer the disease in plants. Once transmitted to the plant host, the bacteria form biofilms in vascular tissues and blocks water flow (Newman et al., 2004). Interestingly, in this interaction, the rpfF gene of X. fastidiosa has been implicated to regulate colonization in the insect vector (Graphocephala atropunctata) as well the plant hosts (Newman et al., 2004). The rpfF gene encodes for production of diffusible signaling factors (DSF), which lead to the regulation of exopolysaccharides- and pathogenesis-related enzymes. These DSF have been implicated in causing differences in biofilm architecture in the insect gut in comparison to the biofilms formed in the plant xylem (Newman et al., 2004). Additionally, mutation in the rpfF gene promotes hypervirulence in plants that are artificially inoculated but prevents bacterial attachment in insects, thereby preventing natural transmission of the bacteria to the plants (Newman et al., 2004). Further molecular investigation to regulate this obligate tritrophic interaction can elucidate means to control the outspread of this pathogenic bacterium.

In contrast to pathogenic multitrophic interactions, biofilm formation can be beneficial for many organisms. Biofilm formation in Bradyrhizobium elkanii SEMIA 5019 and Penicillium sp. significantly increases nodulation and nitrogen accumulation in soybeans compared with planktonic inocula (Jayasinghearachchi & Seneviratne, 2004). Further molecular investigations to understand the regulation of this obligate tritrophic interaction will be vital in understanding how these bacteria are transmitted and colonized. Concomitantly, rhizosphere nematodes often eavesdrop on chemical communication between microorganisms and plants (Horiuchi et al., 2005). Unlike plants and microorganisms, rhizosphere nematodes are highly mobile and may readily avoid or respond to underground chemical signals. Pea root border cells attract nematodes and border cell specific secretions inhibit their motility (Hawes et al., 2000). Given this evidence, there is a possibility that the outcome of plant–microorganism biofilm interactions may also be detrimental to nematode survival in the rhizosphere. Knowledge about the pathogenesis or advantageous aspects of these vector transmitted bacterial biofilms and their environmental interactions will greatly benefit the scientific community.

Concluding remarks

The plant root with complex surface chemistry is capable of directing various interactions with the rhizosphere microbial community, of which bacterial biofilm formation is one of the most significant. The chemical complexity of root secretions implies the existence of extremely complex mechanisms facilitating beneficial and suppressing pathogenic biofilms. Both of these aspects of root bacterial interaction likely involve complex regulatory mechanisms, which modulate gene expression in both the plant and the associated bacteria. Therefore, there is a need for continued effort in elucidating the mechanisms involved in root surface colonization and biofilm formation. This would not only contribute towards better understanding of the biology of root-biofilm interactions but also contributes to the field of rhizosphere biology.

Acknowledgements

We apologize to colleagues whose work could only be covered and cited because of space limitations. H.P.B. acknowledges the support from the University of Delaware through a new faculty initiation grant, University of Delaware Research Foundation and NSF-EPSCoR seed grant.

Footnotes

  • Editor: Jim Prosser

References

View Abstract