Enterococcus faecalis Mammalian Virulence-Related Factors Exhibit Potent Pathogenicity in the Arabidopsis thaliana Plant Model
http://www.100md.com
感染与免疫杂志 2005年第1期
Department of Horticulture and Landscape Architecture
Cell and Molecular Biology Program, Colorado State University, Fort Collins, Colorado
ABSTRACT
Some pathogenic bacteria belong to a large, diverse group of species capable of infecting plants, animals, and humans. Enterococcus faecalis is an opportunistic human pathogen capable of infecting patients with a deficient immune system. Here we report that three E. faecalis strains (FA-2-2, V583, and OG1RF) are capable of infecting the leaves and roots of the model plant species Arabidopsis thaliana, causing plant mortality 7 days postinoculation. We found that E. faecalis pathogenesis in A. thaliana leaves is determined by the following series of events: attachment to leaf surface, entry through stomata or wounds, and colonization in intercellular spaces, leading to rotting and to the disruption of plant cell wall and membrane structures. The three E. faecalis strains colonize the roots of A. thaliana by forming a mosaic of large clusters of live bacteria on the root surface, as observed by scanning electron microscopy, phase-contrast microscopy, and fluorescence microscopy. To dissect the involvement of mammalian virulence-related factors in plant pathogenicity, we tested E. faecalis mutant strains fsrA (TX5240), fsrB (TX5266), fsrC (TX5242), gelE (TX5264), and sprE (TX5243), which correspond to virulence factors involved in pathogenesis in different animal models. Two E. faecalis virulence-related factors that play an important role in mammalian and nematode models of infection, a putative quorum-sensing system (fsrB) and serine protease (sprE), were also found to be important for plant pathogenesis. The development of an E. faecalis-A. thaliana model system could potentially be used to circumvent certain inherent limitations that an animal model imposes on the identification and study of virulence factors. Furthermore, our study suggests an evolutionary crossover of virulence factors in plant, animal, and nematode pathogenesis.
INTRODUCTION
Enterococcus faecalis is a ubiquitous gram-positive bacterium that is usually found in soil and water (19, 25). E. faecalis causes infections in such varied systems as the gastrointestinal tract, the skin and skin structures, the urinary tract, the bloodstream, and the heart (18, 23, 28), thus making it an important pathogen in hospital settings (9, 19, 23, 24, 26, 39).
E. faecalis presents a significant therapeutic challenge, due to its resistance to a vast array of antimicrobial drugs (cell wall-active agents, all commercially available aminoglycosides, penicillin, ampicillin, and vancomycin). The propensity of E. faecalis to acquire resistance may relate to its ability to participate in various forms of conjugation, which can result in the spread of genes as part of conjugative transposons, pheromone-responsive plasmids, or broad-host-range plasmids. The combination of these attributes suggests that this bacterium and its resistance to antimicrobial drugs will continue to pose a challenge (27, 28, 39). Several virulence-related factors have been described for E. faecalis, including cytolysin, a factor called aggregation substance, a zinc metalloprotease (gelatinase), and Fsr (an E. faecalis regulator), a putative quorum-sensing system thought to be involved in gelatinase and/or serine protease regulation (4, 12, 35, 36, 42, 43). In recent years, an understanding of the conditions that regulate cytolysin expression has been largely worked out, with two genes, cylR1 and cylR2, both lacking homologues of known function, thought to work together to repress transcription of cytolysin genes (17). In a recent report on the complete genome sequence of Enterococcus faecalis V583 (33), a vancomycin-resistant isolate, more than a quarter of the genome was found to consist of probable mobile or foreign DNA. One of the predicted mobile elements is a previously unknown vanB vancomycin-resistance conjugative transposon.
The limited knowledge of enterococcal virulence factors is due to the cumbersome and expensive nature of mammalian models for enterococcal infections. Hence, in the present study we have sought to develop an alternative host, the plant species Arabidopsis thaliana. Despite a vast evolutionary gap between plants and animals, some of the mechanisms of bacterial pathogenesis in the two organisms may be similar; also, a commonality in virulence factors for bacterial pathogenicity in plants and animals has been reported (37, 41, 45). Bacterial proteins involved in the export of proteinaceous virulence factors have been shown to be conserved in both mammalian and plant species (3, 5, 9, 11, 14, 15, 20, 24). Also, bacteria such as Pseudomonas cepacia (10), Pseudomonas aeruginosa (34, 37, 46), and Erwinia species have been found to be pathogenic to both animal and plant hosts. P. aeruginosa (PA14), a multihost pathogen like E. faecalis, infects A. thaliana, causing systemic infection to the vascular parenchyma of leaves and roots which results in the death of the infected plant (34, 46). These findings prompted us to test whether strains of E. faecalis could be capable of inducing disease in a well-defined plant system.
From an evolutionary perspective, it is important to determine if E. faecalis virulence factors are also involved in plant pathogenesis. We found that three strains of E. faecalis exhibit potent pathogenicity in A. thaliana involving a sequential array of events: attachment to the leaf and root surfaces, congregation of bacteria in stomata or wounds, colonization in intercellular spaces, formation of communities on root surfaces, disruption of membrane structures, and maceration and rotting of the petiole. Additionally, our studies also show that some of the mammalian virulence-related factors in E. faecalis are involved in plant pathogenicity. In the present study, we have developed an experimental system for studying E. faecalis pathogenicity by using A. thaliana as the host. The observations presented below suggest that E. faecalis is a facultative pathogen of A. thaliana, capable of causing local and systemic infection leading to death of the infected plant.
MATERIALS AND METHODS
Plant material and growth conditions. Seeds of wild-type Arabidopsis thaliana ecotype Columbia (Col-0, Ler-0) were obtained from Lehle Seeds (Round Rock, Tex.). Seeds were surface sterilized using commercial sodium hypochlorite (0.3% [vol/vol]) for 10 to 12 min and then washed four times in sterile double-distilled water. Surface-sterilized seeds were placed on static Murashige and Skoog (MS) basal solid medium (Invitrogen, Inc.) in petri dishes for germination and incubated in a growth chamber. Fifteen-day-old seedlings of each ecotype were individually transferred to 50-ml culture tubes containing 5 ml of liquid MS basal medium. Plant cultures were maintained on an orbital platform shaker (Lab-Line Instruments) set at 90 rpm with a photoperiod of 16 h light and 8 h dark at 25 ± 2°C.
Bacterial strains and culture conditions. The following strains of E. faecalis were used in this study: FA-2-2 (2), V583 (38), and OG1RF (21), obtained from the laboratory of Frederick M. Ausubel (Harvard Medical School). E. faecalis mutants fsrA (TX5240), fsrB (TX5266), fsrC (TX5242), gelatinase gelE (TX5264), and serine protease sprE (TX5243) were obtained from Barbara E. Murray (Texas Medical Center). All strains were plated on brain heart infusion (BHI) agar medium (Difco) in 35-mm tissue-culture plates (Falcon) and incubated at 37°C. Tetracycline (12.5 μg ml–1) was added to the medium to selectively prevent growth of Escherichia coli. Lawns of bacteria were grown as follows: 2 ml of BHI were inoculated with a single colony of the appropriate strain and grown at 37°C for 6 to 8 h, and 10 μl of the culture was spread on each plate. The plates were incubated at 37°C overnight. Plated cells were suspended in 5 ml of Luria-Bertani broth for overnight growth at 37°C and shaken at 250 rpm.
In vitro root pathogenicity assay. Twenty-five-day-old A. thaliana plants were used for the in vitro root pathogenicity assays. E. faecalis strains were grown to an optical density at 600 nm (OD600) of 0.3 to 0.4 and added separately to the 5 ml of MS medium containing A. thaliana wild-type (Col-0 and Ler-0) plants to reach an initial OD600 of 0.02. MS basal medium (5 ml) without plant material was inoculated with the same volume of each bacterial strain tested. A noninfected plant control was maintained under the same conditions. All the treatments and controls were incubated at 30°C in a controlled environment incubator shaker (New Brunswick Scientific) set at 30 rpm with a photoperiod of 16 h light and 8 h dark. Root tissues (500 mg fresh weight basis) of A. thaliana infected with strains of E. faecalis (FA-2-2, V583, and OG1RF) as well as mutant fsrA (TX5240), fsrB (TX5266), fsrC (TX5242), gelatinase gelE (TX5264), and serine protease sprE (TX5243) strains were homogenized in 1 ml of saline (0.2% sodium chloride) with a tissue grinder (Kontes, size C), and the suspension was serially diluted in saline and plated to determine bacterial cell counts as previously described (1, 37, 46). Each experiment was conducted twice with five replicates.
Leaf and soil pathogenicity assay. Seeds of A. thaliana (Col-0 and Ler-0) were surface sterilized and germinated as described previously (see "Plant material and growth conditions"). Fifteen-day-old seedlings were transplanted from static MS medium to 10-cm black plastic pots containing 50 g (dry weight) of PM-O5 A. thaliana growing medium (Lehle Seeds). Plants were incubated in a growth chamber at 30°C with 12 h of light and watered daily for 2 weeks prior to inoculation with bacteria. For leaf assays, the E. faecalis strains (FA-2-2, V583, and OG1RF) were grown in BHI at 37°C to an OD600 of 0.2 to 0.3 and diluted 1:100. Diluted suspensions were individually injected with the blunt end of a hypodermic needle into intact leaves of A. thaliana at a dose of approximately 103 CFU/cm2 as previously described (34). Infiltrated plants were incubated in a growth chamber at 30°C and 80% relative humidity with 16 h light and 8 h dark. For soil infiltration, 50 g of soil with A. thaliana (Col-0) was flooded with 10 ml of bacterial suspension to give an inoculum concentration of 1 x 107 to 3 x 107 CFU/g of soil. Disease symptoms, such as leaf lesions and water-soaked lesions, were observed upon microscopy (described below). Bacterial cell counts from leaves infected with strains of E. faecalis (FA-2-2, V583, and OG1RF) were performed as described previously (1, 37, 46). Plants were incubated under conditions identical to those used for leaf infiltration assays. Each experiment was conducted twice with five replicates.
Phase-contrast microscopy, CSLM, and SEM. Phase-contrast images of E. faecalis-infected root tissues were captured with a x10 objective on an Olympus BX60 microscope equipped with CoolSnap (San Diego, Calif.) imaging software as described previously (1, 46). For scanning electron microscopy (SEM), segments of A. thaliana roots and leaves were fixed in 4% (wt/vol) paraformaldehyde and passed through increasing concentrations of ethanol (30, 50, 70, 96, and 100%). The fixed roots and leaves were dried in a Samdri-PVT-3B critical point drying apparatus, mounted on stubs, coated with a 12-nm layer of gold-palladium in a Hummer-II sputter coater, and visualized using a scanning electron microscope (Joel JSM-6500F; USA Inc., Peabody, Mass.). To determine whether E. faecalis root colonies were encased in a polysaccharide matrix, we stained the infected roots with Calcofluor, a polysaccharide-binding dye. After several water rinsings, the roots were stained for 30 min with 10 ml of 75-μg ml–1 Calcofluor (Fluostain; Sigma-Aldrich) in wash buffer. The stained bacterial communities were then analyzed by confocal scanning laser microscopy (CSLM) as described previously (1, 46). Phase-contrast microscopy, SEM, and CSLM were performed 4 days postinoculation. Samples were analyzed for fluorescence with a confocal laser microscope (Fluroview LGPS-2; Olympus). Samples were viewed using 488 nm as the excitation wavelength. Phase-contrast, fluorescence, and scanning electron microscopy were done 4 days postinoculation. Each experiment was viewed twice with three replicates each.
Transmission electron microscopy (TEM). Small pieces of infected or uninfected A. thaliana roots and leaves (1 by 3 mm) were fixed overnight in 3% (wt/vol) glutaraldehyde in 0.1 N cacodilate buffer at pH 7.2, washed in the same buffer, postfixed in 2% (wt/vol) osmium tetroxide, dehydrated in an alcohol series (30, 70, 96, and 100%), and embedded in Spurr medium at 60°C for 16 h. After polymerization, ultra-thin sections were cut with an LKB 8 800 Ultratome, stained in 2% (wt/vol) lead citrate and 2% (wt/vol) uranyl acetate, and analyzed by using an electron microscope (Jeol JEM-2000EX; USA Inc.).
RESULTS
In vitro and in-soil root pathogenicity of E. faecalis against A. thaliana. The root pathogenicity of E. faecalis strains FA-2-2, V583, and OG1RF was tested in vitro against two wild-type ecotypes of the model plant A. thaliana (Col-0 and Ler-0). All three E. faecalis strains, when inoculated into the liquid root medium of A. thaliana, caused characteristic disease-like symptoms, such as black necrotic regions at the root tips leading to complete cytoplasmic condensation, a clear signature of rhizotoxicity. A. thaliana infected with strains V583 and OG1RF displayed symptoms of infection, with bacterial cells initially infecting roots, the root-stem transition zone, and fully developed leaves positioned near the base of the plant 2 to 3 days postinoculation and spreading systemically to the top of the plant approximately 4 days postinoculation, with plant mortality occurring 7 days postinoculation (Fig. 1A). The maximum plant mortality was recorded with the OG1RF strain (Fig. 1B). In addition to in vitro studies, we tested the ability of strains FA-2-2, V583, and OG1RF to infect soil-grown A. thaliana plants (Materials and Methods). All strains caused extensive aerial tissue damage leading to plant mortality approximately 7 days postinoculation when infiltrated into the soil immediately surrounding the root system (Fig. 1C).
In planta bacterial growth of E. faecalis on Arabidopsis leaves and roots. Interestingly, all three E. faecalis strains caused condensation of the root meristem and mature regions, leading to complete cell death (unpublished data). In accordance with the rhizotoxicity, all three strains showed stable root colonization along with multiplication of bacteria on root tissues of A. thaliana (Col-0), as reflected by the increasing CFU counts during a time course analysis (Fig. 1D). In previously published work, we showed that A. thaliana ecotype Col-0 is highly susceptible to infection by another opportunistic human pathogen, P. aeruginosa strains PAO1 and PA14 (46). Figure 2A shows E. faecalis strains (FA-2-2, V583, and OG1RF) infiltrating A. thaliana Col-0 leaf surface structures after leaves were incubated at room temperature for 2 h in a suspension of E. faecalis cells and then incubated on the surface of 1.5% (wt/vol) water agar in closed petri dishes. The attached bacteria inflicted disease symptoms and leaf mortality under the above-mentioned conditions (Fig. 2A to C). In order to evaluate the correlation between the appearance of disease symptoms and the size of bacterial colonies infecting the leaf area, we next examined the percentage of infected leaf area in plants challenged with all three strains of E. faecalis (Fig. 2B). Consistent with the plant mortality results, all three strains revealed prominent water-soaked lesions on the upper dorsal leaf surface (adaxial) 7 days postinfection compared to the leaf surface of the untreated control (Fig. 2C). Each of the three E. faecalis strains behaved as a true pathogen, as shown by increasing CFU counts, reflecting in planta bacterial growth on infected leaves during the 7-day time course (Fig. 2D). These results indicate that E. faecalis strains trigger an active physiological response in A. thaliana roots and leaves.
Strains FA-2-2, V583, and OG1RF forms a biofilm-like community on root surfaces. As part of establishing the mode of attachment by the three E. faecalis strains (FA-2-2, V583, and OG1RF) on root surfaces, we determined that these strains adhere to root surfaces of A. thaliana. Four days post-cocultivation of A. thaliana roots and E. faecalis strains (FA-2-2, V583, OG1RF) in MS medium, roots were viewed by phase-contrast microscopy and CSLM. We observed that cells of the three E. faecalis strains (FA-2-2, V583, and OG1RF) had colonized virtually the entire root surface (unpublished data. Phase-contrast microscopy and CSLM revealed that roots of A. thaliana were surrounded by phase-bright material suggestive of an extracellular matrix (unpublished data). Soil-grown A. thaliana, when infected with E. faecalis strains (FA-2-2, V583, and OG1RF), showed a similar root colonization by the bacteria (data not shown), suggesting that E. faecalis strains FA-2-2, V583, and OG1RF form a stable, biofilm-like community on these roots.
Effect of E. faecalis infection on plant ultrastructure. The leaves of A. thaliana were incubated with E. faecalis strains FA-2-2, V583, and OG1RF and then transferred to water agar plates at room temperature for viewing. After cells of the three E. faecalis strains entered the substomatal cavity, they began multiplying and rapidly spread through the leaf mesophyll. As in the case of other well-studied phytopathogenic bacteria, E. faecalis strains FA-2-2, V583, and OG1RF were also found to make a "bacterial ring" (dense populations of bacterial cells that adopt a ring structure) that colonized the intercellular space, presumably by digesting the middle lamellae and separating the plant cells from each other (Fig. 3A to C). At relatively early stages of the infection (2 days), almost all the bacteria were found in intercellular spaces attached to A. thaliana mesophyll walls (Fig. 3D). At later stages of infection (3 to 4 days), many of the mesophyll cells were severely damaged and contained intracellular bacteria (Fig. 3C).
Infection of A. thaliana leaves with E. faecalis strains FA-2-2, V583, and OG1RF had a unique effect on the structures of the host cell walls and membranes and on the location and organization of host cell organelles (Fig. 3A to D). In contrast, uninfected A. thaliana mesophyll cells had a large central vacuole surrounded by a thin layer of cytoplasm containing a nucleus and organelles. The first sign of host cell degeneration following infection was slight plasmolysis and concentration of host membrane structures, including chloroplasts, endoplasmic reticulum, and dictyosomes, at the site of bacterial contact. At this stage a limited number of single bacteria appeared to be able to penetrate into metabolically active plant cells (Fig. 3A to D). In roots, also, the proliferation of bacteria in the intercellular space resulted in alteration of host organelles. Host plasmalemma became highly undulated and disrupted (Fig. 3E to H), and there was swelling and disruption of outer cell membranes and finally the destruction of the whole root structure (Fig. 3F to H). Cell organelles were degraded, and the cell walls were thin and highly convoluted. The bacteria appeared as a colony in the intercellular spaces (Fig. 3). Host cell collapse in both leaves and roots was the final step of the bacterial infection.
Development of the bacterial infection and cellular mode of attachment on A. thaliana leaves and roots. To visualize the bacterial infection and cellular mode of attachment of strains FA-2-2, V583, and OG1RF, the leaf and root surfaces of A. thaliana were viewed by SEM and TEM. SEM of infected leaves and roots showed that E. faecalis cells attached perpendicularly and horizontally to the leaf and root cell walls of A. thaliana (Fig. 4A to C). Bacterial cells oriented in either position appeared to be degrading and penetrating through the outermost layers of the A. thaliana root or leaf cell wall (Fig. 4A to H). Viewing FA-2-2, V583, and OG1RF infection with a higher magnification of SEM revealed that bacterial cells were embedded within and connected together by an extracellular polymeric matrix (Fig. 4C, E, and G). Individual cells of E. faecalis strains FA-2-2, V583, and OG1RF could be seen attaching along the leaf surface towards open stomata (Fig. 4D, F, and H). In the experiments described above, Col-0 leaves were incubated with E. faecalis (FA-2-2, V583, and OG1RF) and then transferred to water agar plates at room temperature. The infection seemed to proceed in a manner similar to that with other well-studied phytopathogenic bacteria, such as P. aeruginosa (PA14), which formed "bacterial threads" (dense populations of bacterial cells that adopt an elongated and branched structure) that colonized the intercellular space, presumably by digesting the middle lamellae and separating the plant cells from each other (34). At relatively early stages of the infection (2 days), almost all the E. faecalis bacteria were found in intercellular spaces attached to A. thaliana mesophyll walls (Fig. 5A and B), although some bacterial cells could also be observed attached to the inner surface of cell walls (Fig. 5). At later stages of infection (3 to 4 days), many of the mesophyll cells were severely damaged and contained intracellular bacteria (Fig. 3 and 5). TEM also revealed that roots and leaves of A. thaliana infected with FA-2-2, V583, and OG1RF showed the presence of bacteria in the intracellular cavities, inflicting pathogenesis (Fig. 5A and B).
Role of fsr quorum sensing in E. faecalis plant pathogenicity. The above-described results strengthened our hypothesis that biofilm-like community development and root colonization contributed to E. faecalis pathogenicity in A. thaliana plants. In E. faecalis, the fsr system positively regulates the expression of several virulence factors (29, 40). The fsr locus is composed of three regulatory genes: fsrA, fsrB, and fsrC (35, 40, 44). To determine whether the previously observed effect of a fsrB mutation on E. faecalis virulence in the Caenorhabditis elegans model (7, 8) was specific to fsrB or representative of the entire fsr locus, we evaluated the ability of fsrA (TX5240) and fsrC (TX5242) deletion mutants to kill A. thaliana. Compared with the wild-type strain OG1RF, the fsrA and fsrC strains showed no significant difference (P < 0.0002) in their ability to kill A. thaliana (Fig. 6A and B). Interestingly, infections with the deletion mutant fsrB revealed an attenuated pathogenicity in the A. thaliana root model (Fig. 6A). A. thaliana plants were susceptible to deletion fsrA and fsrC quorum-sensing mutants, displaying symptoms of infection similar to those observed with infection with wild-type OG1RF and succumbing to infection 7 days postinoculation (Fig. 6A). In accordance with the reduced mortality results with the fsrB strain, fewer bacterial cells were attached to the root surfaces and colonization appeared to be diminished (Fig. 6B to D). Notably, CSLM showed that the fsrB strain appeared to form a nearly diminished biofilm-like community, while the communities formed by the fsrA and fsrC strains appeared full-grown compared to those for the wild type, OG1RF (Fig. 6D). The fact that CSLM of fsrB strain root infectivity on A. thaliana shows the reduced presence of E. faecalis cells authenticates the involvement of a quorum-sensing system in the root model (Fig. 6D). To evaluate the number of bacterial cells associated with biofilm-like community formation in planta, we analyzed the cell counts on the root surface of A. thaliana on the fourth day after infection with all the deletion mutants of E. faecalis (TX5240 [fsrA], TX5266 [fsrB], and TX5242 [fsrC]) (Fig. 6C). The cell counts showed a positive correlation between the bacterial cell counts on the root surface, the degree of community formation, and pathogenicity of E. faecalis fsrB mutant strains on A. thaliana (Fig. 6A to D).
Deletion in serine protease (sprE) but not gelE causes attenuation of bacterial infection in the A. thaliana root pathogenicity model. Previous work has shown that a disruption of some of the virulence-related factors of E. faecalis, particularly gelatinase (gelE), causes attenuation of the E. faecalis strain OG1RF in the mouse peritonitis and nematode models (7, 8, 40, 42). Three genes located directly upstream of gelE, fsrA, fsrB, and fsrC, seem to be involved in gelE regulation (35). FsrA and FsrB are homologous to two-component response regulators and sensor kinases (35, 42), respectively, and a nonpolar fsrB deletion blocks the production of gelatinase (35). Based on the homology of FsrA, FsrB, and FsrC to the Staphylococcus aureus quorum-sensing system encoded by agrA, agrB, and agrC, fsrB may encode a processor of a putative E. faecalis signal peptide (42). Figure 6A to D shows that A. thaliana survived when infected with the fsrB gelatinase regulatory mutant (strain TX5266); when infected with the fsrA, fsrC, or gelE mutant or the isogenic parental strain OG1RF, it did not survive (Fig. 6A to D). E. faecalis OG1RF mutants containing disruptions in fsrA, fsrC, and gelE negated prolonged survival of A. thaliana, which was nearly opposite to the response observed with a single disruption (fsrB) (Fig. 6A to D). Notably, it is reported that the fsrB mutant is also less virulent in the mouse intraperitoneal injection and nematode model (7, 8, 35, 40, 42). Interestingly, neither gelatinase nor serine protease activity is reported for the fsr mutants TX5240, TX5242, and TX5266 (35). We evaluated the contribution of E. faecalis gelatinase (gelE) and serine protease (sprE) separately from its phytotoxic activity on the A. thaliana root-infection model. In-frame deletion mutants, TX5264 (gelE) and TX5243 (sprE), were used as reported previously (35). Compared to each other, TX5264 (gelE) and TX5243 (sprE) were significantly different in their ability to cause plant mortality in A. thaliana (P < 0.05) (Fig. 6A to D). The serine protease sprE mutant (TX5243) was completely attenuated in its ability to cause plant mortality compared to the deletion gelatinase gelE mutant (TX5264) (Fig. 6A to D). Interestingly, CSLM showed that the sprE mutant appeared to form poor communities, while the colonization formed by the fsrA, fsrC, and gelE strains appeared developed compared to the colonization of wild-type OG1RF (Fig. 6D). As observed before with fsr deletion mutants, the cell counts showed a positive correlation between the bacterial cell counts on the root surface, the degree of biofilm formation, and pathogenicity of E. faecalis sprE mutant strains on A. thaliana (Fig. 6A to D). These results suggest that an fsr quorum sensing system regulates genes involved in E. faecalis virulence; such a system is probably complemented with serine protease (sprE) in the A. thaliana root pathogenicity model.
DISCUSSION
In the present communication, we have found that E. faecalis strains FA-2-2, V583, and OG1RF can infect roots and leaves of A. thaliana. All three strains of E. faecalis inflict local and systemic infection, leading to the death of the infected A. thaliana plant. Plant infection and subsequent mortality due to E. faecalis were traced to the formation of a pathogenic biofilm-like community colonizing the root surface. This phenomenon is similar to that observed with the opportunistic human pathogen P. aeruginosa and the classical A. thaliana pathogen P. syringae (DC3000), where pathogenic biofilms on the root surface have been shown to be partly responsible for plant mortality (1, 46). E. faecalis infection in A. thaliana, as viewed by various microscopic techniques, starts with bacterial attachment to the plant leaf and root surfaces and entry into the A. thaliana tissues via stomatal openings (in the case of the leaves), followed by proliferation in the substomatal cavities and intercellular space. As the bacteria proliferate, A. thaliana cell walls become undulated and at least some bacteria appear to penetrate the walls. Aggressive E. faecalis propagation in the intercellular space results in maceration and autolysis of plant cells. These data strongly support the observation that the soft-rot symptoms were elicited by E. faecalis on A. thaliana leaves and roots; similar disease symptoms have been reported earlier in the P. aeruginosa-A. thaliana infection model (34, 37, 46). It has previously been reported that strains of E. faecalis are capable of infecting the nematode C. elegans and the mouse peritoneal model (7, 8, 40, 42). This is the first report describing how E. faecalis causes infections in A. thaliana as determined by these sequential characteristics: attachment to leaf or root surface, entry through stomata or wounds, colonization and multiplication in the intercellular spaces leading to plant cell walls, biofilm-like community formation, disruption of membrane structures, and rotting, ultimately leading to complete plant mortality.
The pathogenicity of the tested strains of E. faecalis, FA-2-2, V583, and OG1RF, to the roots of A. thaliana in vitro and in soil indicates that E. faecalis virulence in both settings is similar and that our experimental system is a reliable method for further studying the interaction between E. faecalis and plant roots. Since FA-2-2, V583, and OG1RF were capable of causing the mortality of A. thaliana, we followed the bacterial interactions with the roots by using SEM and TEM, phase-contrast microscopy, and CSLM. As previously reported for P. aeruginosa's mode of infection by attachment on A. thaliana leaves and roots (34, 46), we observed FA-2-2, V583, and OG1RF cells attached perpendicularly to the roots and leaves of A. thaliana (Fig. 3 to 5). Furthermore, CSLM and phase-contrast microscopy of root tissue confirmed the formation of biofilm-like communities on the A. thaliana root surface (unpublished data).
A. thaliana ecotypes exhibit a similar degree of susceptibility to E. faecalis infection. We also show that E. faecalis penetrates and forms larger lesions on the leaves and roots of the susceptible Col-0 and Ler-0 ecotypes (data not shown) but that E. faecalis also attaches at least 1.5-fold more efficiently to the epidermis of Ler-0 roots than to Col-0 roots (data not shown).
One reason that relatively little is known about enterococcal virulence factors is that the mammalian models used to study enterococcal infections are cumbersome and expensive (35, 36, 40, 42). Using a mammalian host to screen enterococcal mutant libraries for avirulent mutants, for example, would be prohibitively time consuming and expensive because of the large number of animals involved. Therefore, we have sought to develop an alternative and defined plant model system like A. thaliana for Enterococcus infections. As we have learned more about the mechanisms and epidemiology of resistance to antimicrobial drugs, it has become clear that bacteria have a remarkable array of tools at their disposal to overcome antibiotics (9, 26). Like other gram-positive microorganisms, enterococci are able to produce biofilms on abiotic surfaces (6, 18), increasing their high innate resistance to antibiotics (6); yet the factors controlling enterococcal biofilm formation and maintenance remain unknown (22, 31). Plant pathogenesis related to biofilm formation is now well documented in the cases of the gram-negative bacteria P. syringae DC3000 and P. aeruginosa (1, 46). Our intriguing observation that E. faecalis forms a pathogenic biofilm-like community for colonization of the biotic surface of A. thaliana roots suggests that E. faecalis, like other gram-negative bacteria, also employs community formation as an important virulence factor in its potent pathogenicity against A. thaliana. Interestingly, not many gram-positive bacteria have been reported as plant pathogens.
To dissect the involvement of mammalian virulence-related factors in plant pathogenicity, we tested E. faecalis mutant fsrA (TX5240), fsrB (TX5266), fsrC (TX5242), gelatinase gelE (TX5264), and serine protease sprE (TX5243) strains. Two E. faecalis virulence-related factors that play an important role in mammalian and nematode models of infection, a putative quorum-sensing system (fsrB) and serine protease (sprE), are also important for plant pathogenesis. Quorum sensing is a cell density-dependent regulatory system that controls a variety of group behaviors in bacteria (13, 32). In E. faecalis, the fsr system positively regulates the expression of gelatinase and serine protease in a cell density-dependent manner, similar to the well-studied regulation of toxins by the S. aureus agr quorum-sensing regulatory locus (16, 32). Qin et al. (35, 36) have characterized three genes in the fsr regulatory locus, fsrA, fsrB, and fsrC. Using a nonpolar fsrB deletion mutant, the same workers showed that fsrB is required for the regulatory function of the Fsr system (35, 36). The expression of the fsr genes in E. faecalis OG1RF is cell density dependent and is most active in the post-exponential phase of growth (30, 35, 36). Supporting the already-existing knowledge about cell density-dependent pathogenicity of E. faecalis using the fsr system, we also found that fsrB plays an important role in regulating bacterial colonization on the A. thaliana root surface (Fig. 6). Accordingly, a fsrB deletion mutant failed to colonize the A. thaliana roots and exhibited an attenuated pathogenicity along with decreased CFU counts (Fig. 6). Contrastingly, it has already been reported that fsrA and fsrC insertion mutants, like the fsrB mutant, were attenuated in their ability to kill C. elegans, though in a rabbit endophthalmitis model, the fsrB mutant showed significantly reduced virulence compared to the wild type (7, 8, 29). The fact that the virulence of the fsrB strain varies between invertebrate and mammalian pathogenicity systems supports the unexpected result obtained in our studies with quorum-sensing mutants (fsrB) and also indicates that the virulence factors regulated by the fsr system may be important for the full expression of E. faecalis pathogenicity towards A. thaliana.
Along with the fsr system in E. faecalis, serine protease (sprE) is an additional virulence factor thought to play a role in systemic disease in mammalian hosts (35, 40). The serine protease gene sprE, which lies immediately downstream of and is cotranscribed with gelE, encodes a secreted 26-kDa serine protease that shares homology with S. aureus V8 protease (35). Insertion disruption of sprE also attenuates virulence in the mouse peritonitis and C. elegans model systems (35, 40). Transcription of the gelE-sprE operon is positively regulated in a growth phase-dependent fashion by the fsr locus, which shares many similarities with the well-studied S. aureus agr regulatory locus (35). The fsr locus is composed of three regulatory genes, fsrA, fsrB, and fsrC, located upstream of the gelE-sprE operon (35). As reported before with the mouse peritonitis and C. elegans models with a deletion mutant of SprE (sprE), we also observed attenuated killing with the sprE mutant in the A. thaliana root pathogenicity model (Fig. 6), suggesting the sharing of similar virulence factors in mammalian, nematode, and plant models.
The E. faecalis fsr system is the second example (after the rhl and las systems of P. aeruginosa) of a quorum-sensing system that regulates virulence gene expression in bacterial infection of both simple model organisms and mammalian hosts. Quorum sensing may be an important mechanism used by many prokaryotes to adapt to different environments encountered during pathogenesis. Our results also raise the possibility that the fsr system in E. faecalis regulates virulence genes in addition to sprE in this pathogen in the A. thaliana root pathogenicity model. In E. faecalis, improvement of the selection of effective antimicrobial agents for use against recalcitrant infections is urgently needed. Taking into account the strong correlation between the presence of fsr and the ability to produce a biofilm-like community, it may be possible to screen for plant-root-exuded products against putative biofilm-forming E. faecalis strains by using the present plant-pathogen model to directly isolate antimicrobial and anti-infective root-exuded metabolites. The bacterial phenotype (acting by adherence or biofilm formation on root surface) presented in our study and the very nature of this screening is advantageous in that it allows preliminary identification of strains which are highly adherent and are thus good candidates for testing the antibiotic susceptibility of biofilms. The development of the E. faecalis-A. thaliana model system could potentially be used to circumvent certain inherent limitations that an animal model imposes on the identification and study of virulence factors. Furthermore, our study unravels some evolutionary crossover of virulence factors in plant pathogenesis which have previously been found relevant for mammalian and nematode models of infection. Finally, it is alluring to speculate that E. faecalis-crop plant interactions may occur in nature and that plants and crops may likely act in the epidemiology of this multihost pathogen.
ACKNOWLEDGMENTS
We thank Barbara E. Murray (Texas Medical Center) for generously providing E. faecalis deletion mutants for our studies.
This work was supported by grants from Colorado State University Agricultural Experiment Station (to J.M.V.) and NSF-CAREER (grant no. MCB 0093014 to J.M.V.).
These two authors contributed equally to this work.
Present address: Department of Biochemistry and Molecular Biology, Colorado State University, Fort Collins, CO 80523-1173.
REFERENCES
1. Bais, H. P., R. Fall, and J. M. Vivanco. 2004. Biocontrol of Bacillus subtilis against infection of Arabidopsis roots by Pseudomonas syringae is facilitated by biofilm formation and surfactin production. Plant Physiol. 134:307-319.
2. Bottone, E. J., L. Patel, P. Patel, and T. Robin. 1998. Mucoid encapsulated Enterococcus faecalis: an emerging morphotype isolated from patients with urinary tract infections. Diagn. Microbiol. Infect. Dis. 31:429-430.
3. Cho, J. J., N. J. Panopoulos, and M. N. Schroth. 1975. Genetic transfer of Pseudomonas aeruginosa R factors to plant pathogenic Erwinia species. J. Bacteriol. 122:192-198.
4. Chow, J. W., L. A. Thal, M. B. Perri, J. A. Vazquez, S. M. Donabedian, D. B. Clewell, and M. J. Zervos. 1993. Plasmid-associated hemolysin and aggregation substance production contribute to virulence in experimental enterococcal endocarditis. Antimicrob. Agents Chemother. 37:2474-2477.
5. Fenselau, S., I. Balbo, and U. Bonas. 1992. Determinants of pathogenicity in Xanthomonas campestris pv. vesicatoria are related to proteins involved in secretion in bacterial pathogens of animals. Mol. Plant-Microbe Interact. 5:390-396.
6. Foley, I., and P. Gilbert. 1997. In-vitro studies of the activity of glycopeptide combinations against Enterococcus faecalis biofilms. J. Antimicrob. Chemother. 40:667-672.
7. Garsin, D. A., C. D. Sifri, E. Mylonakis, X. Qin, K. V. Singh, B. E. Murray, S. B. Calderwood, and F. M. Ausubel. 2001. A simple model host for identifying Gram-positive virulence factors. Proc. Natl. Acad. Sci. USA 98:10892-10897.
8. Garsin, D. A., J. M. Villanueva, J. Begun, D. H. Kim, C. D. Sifri, S. B. Calderwood, G. Ruvkun, and F. M. Ausubel. 2003. Long-lived C. elegans daf-2 mutants are resistant to bacterial pathogens. Science 300:1921.
9. Gold, H. S., and R. C. Moellering, Jr. 1996. Antimicrobial-drug resistance. N. Engl. J. Med. 335:1445-1453.
10. Goldmann, D. A., and J. D. Klinger. 1986. Pseudomonas cepacia: biology, mechanisms of virulence, epidemiology. J. Pediatr. 108:806-812.
11. Gough, C. L., S. Genin, C. Zischek, and C. A. Boucher. 1992. hrp genes of Pseudomonas solanacearum are homologous to pathogenicity determinants of animal pathogenic bacteria and are conserved among plant pathogenic bacteria. Mol. Plant-Microbe Interact. 5:384-389.
12. Haas, W., and M. S. Gilmore. 1999. Molecular nature of a novel bacterial toxin: the cytolysin of Enterococcus faecalis. Med. Microbiol. Immunol. (Berlin) 187:183-190.
13. Haas, W., B. D. Shepard, and M. S. Gilmore. 2002. Two-component regulator of Enterococcus faecalis cytolysin responds to quorum-sensing autoinduction. Nature 415:84-87.
14. Huang, H. C., Y. Xiao, R. H. Lin, Y. Lu, S. W. Hutcheson, and A. Collmer. 1993. Characterization of the Pseudomonas syringae pv. syringae 61 hrpJ and hrpI genes: homology of HrpI to a superfamily of proteins associated with protein translocation. Mol. Plant-Microbe Interact. 6:515-520.
15. Ike, Y., H. Hashimoto, and D. B. Clewell. 1984. Hemolysin of Streptococcus faecalis subspecies zymogenes contributes to virulence in mice. Infect. Immun. 45:528-530.
16. Jarraud, S., C. Mougel, J. Thioulouse, G. Lina, H. Meugnier, F. Forey, X. Nesme, J. Etienne, and F. Vandenesch. 2002. Relationships between Staphylococcus aureus genetic background, virulence factors, agr groups (alleles), and human disease. Infect. Immun. 70:631-641.
17. Johansson, J., and P. Cossart. 2003. RNA-mediated control of virulence gene expression in bacterial pathogens. Trends Microbiol. 11:280-285.
18. Joyanes, P., A. Pascual, L. Martinez-Martinez, A. Hevia, and E. J. Perea. 2000. In vitro adherence of Enterococcus faecalis and Enterococcus faecium to urinary catheters. Eur. J. Clin. Microbiol. Infect. Dis. 19:124-127.
19. Kayser, F. H. 2003. Safety aspects of enterococci from the medical point of view. Int. J. Food Microbiol. 88:255-262.
20. Kominos, S. D., C. E. Copeland, B. Grosiak, and B. Postic. 1972. Introduction of Pseudomonas aeruginosa into a hospital via vegetables. Appl. Microbiol. 24:567-570.
21. Li, X., G. M. Weinstock, and B. E. Murray. 1995. Generation of auxotrophic mutants of Enterococcus faecalis. J. Bacteriol. 177:6866-6873.
22. Mah, T. F., and G. A. O'Toole. 2001. Mechanisms of biofilm resistance to antimicrobial agents. Trends Microbiol. 9:34-39.
23. Moellering, R. C., Jr. 1992. Emergence of Enterococcus as a significant pathogen. Clin. Infect. Dis. 14:1173-1176.
24. Moellering, R. C., Jr. 1991. The Garrod lecture. The enterococcus: a classic example of the impact of antimicrobial resistance on therapeutic options. J. Antimicrob Chemother. 28:1-12.
25. Muller, T., A. Ulrich, E. M. Ott, and M. Muller. 2001. Identification of plant-associated enterococci. J. Appl. Microbiol. 91:268-278.
26. Mundy, L. M., D. F. Sahm, and M. Gilmore. 2000. Relationships between enterococcal virulence and antimicrobial resistance. Clin. Microbiol. Rev. 13:513-522.
27. Murray, B. E. 1998. Diversity among multidrug-resistant enterococci. Emerg. Infect. Dis. 4:37-47.
28. Murray, B. E. 1990. The life and times of the Enterococcus. Clin. Microbiol. Rev. 3:46-65.
29. Mylonakis, E., M. Engelbert, X. Qin, C. D. Sifri, B. E. Murray, F. M. Ausubel, M. S. Gilmore, and S. B. Calderwood. 2002. The Enterococcus faecalis fsrB gene, a key component of the fsr quorum-sensing system, is associated with virulence in the rabbit endophthalmitis model. Infect. Immun. 70:4678-4681.
30. Nakayama, J., Y. Cao, T. Horii, S. Sakuda, A. D. Akkermans, W. M. de Vos, and H. Nagasawa. 2001. Gelatinase biosynthesis-activating pheromone: a peptide lactone that mediates a quorum sensing in Enterococcus faecalis. Mol. Microbiol. 41:145-154.
31. O'Toole, G. A., L. A. Pratt, P. I. Watnick, D. K. Newman, V. B. Weaver, and R. Kolter. 1999. Genetic approaches to study of biofilms. Methods Enzymol. 310:91-109.
32. Otto, M. 2001. Staphylococcus aureus and Staphylococcus epidermidis peptide pheromones produced by the accessory gene regulator agr system. Peptides 22:1603-1608.
33. Paulsen, I. T., L. Banerjei, G. S. Myers, K. E. Nelson, R. Seshadri, T. D. Read, D. E. Fouts, J. A. Eisen, S. R. Gill, J. F. Heidelberg, H. Tettelin, R. J. Dodson, L. Umayam, L. Brinkac, M. Beanan, S. Daugherty, R. T. DeBoy, S. Durkin, J. Kolonay, R. Madupu, W. Nelson, J. Vamathevan, B. Tran, J. Upton, T. Hansen, J. Shetty, H. Khouri, T. Utterback, D. Radune, K. A. Ketchum, B. A. Dougherty, and C. M. Fraser. 2003. Role of mobile DNA in the evolution of vancomycin-resistant Enterococcus faecalis. Science 299:2071-2074.
34. Plotnikova, J. M., L. G. Rahme, and F. M. Ausubel. 2000. Pathogenesis of the human opportunistic pathogen Pseudomonas aeruginosa PA14 in Arabidopsis. Plant Physiol. 124:1766-1774.
35. Qin, X., K. V. Singh, G. M. Weinstock, and B. E. Murray. 2000. Effects of Enterococcus faecalis fsr genes on production of gelatinase and a serine protease and virulence. Infect. Immun. 68:2579-2586.
36. Qin, X., K. V. Singh, G. M. Weinstock, and B. E. Murray. 2001. Characterization of fsr, a regulator controlling expression of gelatinase and serine protease in Enterococcus faecalis OG1RF. J. Bacteriol. 183:3372-3382.
37. Rahme, L. G., E. J. Stevens, S. F. Wolfort, J. Shao, R. G. Tompkins, and F. M. Ausubel. 1995. Common virulence factors for bacterial pathogenicity in plants and animals. Science 268:1899-1902.
38. Sahm, D. F., J. Kissinger, M. S. Gilmore, P. R. Murray, R. Mulder, J. Solliday, and B. Clarke. 1989. In vitro susceptibility studies of vancomycin-resistant Enterococcus faecalis. Antimicrob. Agents Chemother. 33:1588-1591.
39. Schaberg, D. R. 1994. Resistant gram-positive organisms. Ann. Emerg. Med. 24:462-464.
40. Sifri, C. D., E. Mylonakis, K. V. Singh, X. Qin, D. A. Garsin, B. E. Murray, F. M. Ausubel, and S. B. Calderwood. 2002. Virulence effect of Enterococcus faecalis protease genes and the quorum-sensing locus fsr in Caenorhabditis elegans and mice. Infect. Immun. 70:5647-5650.
41. Silo-Suh, L., S. J. Suh, P. A. Sokol, and D. E. Ohman. 2002. A simple alfalfa seedling infection model for Pseudomonas aeruginosa strains associated with cystic fibrosis shows AlgT (sigma-22) and RhlR contribute to pathogenesis. Proc. Natl. Acad. Sci. USA 99:15699-15704.
42. Singh, K. V., X. Qin, G. M. Weinstock, and B. E. Murray. 1998. Generation and testing of mutants of Enterococcus faecalis in a mouse peritonitis model. J. Infect. Dis. 178:1416-1420.
43. Starr, M. P., and A. K. Chatterjee. 1972. The genus Erwinia: enterobacteria pathogenic to plants and animals. Annu. Rev. Microbiol. 26:389-426.
44. Stevens, S. X., H. G. Jensen, B. D. Jett, and M. S. Gilmore. 1992. A hemolysin-encoding plasmid contributes to bacterial virulence in experimental Enterococcus faecalis endophthalmitis. Investig. Ophthalmol. Vis. Sci. 33:1650-1656.
45. Tan, M. W., L. G. Rahme, J. A. Sternberg, R. G. Tompkins, and F. M. Ausubel. 1999. Pseudomonas aeruginosa killing of Caenorhabditis elegans used to identify P. aeruginosa virulence factors. Proc. Natl. Acad. Sci. USA 96:2408-2413.
46. Walker, T. S., H. P. Bais, E. Deziel, H. P. Schweizer, L. G. Rahme, R. Fall, and J. M. Vivanco. 2004. Pseudomonas aeruginosa-plant root interactions. Pathogenicity, biofilm formation, and root exudation. Plant Physiol. 134:320-331.(Ajay K. Jha , Harsh P. Ba)
Cell and Molecular Biology Program, Colorado State University, Fort Collins, Colorado
ABSTRACT
Some pathogenic bacteria belong to a large, diverse group of species capable of infecting plants, animals, and humans. Enterococcus faecalis is an opportunistic human pathogen capable of infecting patients with a deficient immune system. Here we report that three E. faecalis strains (FA-2-2, V583, and OG1RF) are capable of infecting the leaves and roots of the model plant species Arabidopsis thaliana, causing plant mortality 7 days postinoculation. We found that E. faecalis pathogenesis in A. thaliana leaves is determined by the following series of events: attachment to leaf surface, entry through stomata or wounds, and colonization in intercellular spaces, leading to rotting and to the disruption of plant cell wall and membrane structures. The three E. faecalis strains colonize the roots of A. thaliana by forming a mosaic of large clusters of live bacteria on the root surface, as observed by scanning electron microscopy, phase-contrast microscopy, and fluorescence microscopy. To dissect the involvement of mammalian virulence-related factors in plant pathogenicity, we tested E. faecalis mutant strains fsrA (TX5240), fsrB (TX5266), fsrC (TX5242), gelE (TX5264), and sprE (TX5243), which correspond to virulence factors involved in pathogenesis in different animal models. Two E. faecalis virulence-related factors that play an important role in mammalian and nematode models of infection, a putative quorum-sensing system (fsrB) and serine protease (sprE), were also found to be important for plant pathogenesis. The development of an E. faecalis-A. thaliana model system could potentially be used to circumvent certain inherent limitations that an animal model imposes on the identification and study of virulence factors. Furthermore, our study suggests an evolutionary crossover of virulence factors in plant, animal, and nematode pathogenesis.
INTRODUCTION
Enterococcus faecalis is a ubiquitous gram-positive bacterium that is usually found in soil and water (19, 25). E. faecalis causes infections in such varied systems as the gastrointestinal tract, the skin and skin structures, the urinary tract, the bloodstream, and the heart (18, 23, 28), thus making it an important pathogen in hospital settings (9, 19, 23, 24, 26, 39).
E. faecalis presents a significant therapeutic challenge, due to its resistance to a vast array of antimicrobial drugs (cell wall-active agents, all commercially available aminoglycosides, penicillin, ampicillin, and vancomycin). The propensity of E. faecalis to acquire resistance may relate to its ability to participate in various forms of conjugation, which can result in the spread of genes as part of conjugative transposons, pheromone-responsive plasmids, or broad-host-range plasmids. The combination of these attributes suggests that this bacterium and its resistance to antimicrobial drugs will continue to pose a challenge (27, 28, 39). Several virulence-related factors have been described for E. faecalis, including cytolysin, a factor called aggregation substance, a zinc metalloprotease (gelatinase), and Fsr (an E. faecalis regulator), a putative quorum-sensing system thought to be involved in gelatinase and/or serine protease regulation (4, 12, 35, 36, 42, 43). In recent years, an understanding of the conditions that regulate cytolysin expression has been largely worked out, with two genes, cylR1 and cylR2, both lacking homologues of known function, thought to work together to repress transcription of cytolysin genes (17). In a recent report on the complete genome sequence of Enterococcus faecalis V583 (33), a vancomycin-resistant isolate, more than a quarter of the genome was found to consist of probable mobile or foreign DNA. One of the predicted mobile elements is a previously unknown vanB vancomycin-resistance conjugative transposon.
The limited knowledge of enterococcal virulence factors is due to the cumbersome and expensive nature of mammalian models for enterococcal infections. Hence, in the present study we have sought to develop an alternative host, the plant species Arabidopsis thaliana. Despite a vast evolutionary gap between plants and animals, some of the mechanisms of bacterial pathogenesis in the two organisms may be similar; also, a commonality in virulence factors for bacterial pathogenicity in plants and animals has been reported (37, 41, 45). Bacterial proteins involved in the export of proteinaceous virulence factors have been shown to be conserved in both mammalian and plant species (3, 5, 9, 11, 14, 15, 20, 24). Also, bacteria such as Pseudomonas cepacia (10), Pseudomonas aeruginosa (34, 37, 46), and Erwinia species have been found to be pathogenic to both animal and plant hosts. P. aeruginosa (PA14), a multihost pathogen like E. faecalis, infects A. thaliana, causing systemic infection to the vascular parenchyma of leaves and roots which results in the death of the infected plant (34, 46). These findings prompted us to test whether strains of E. faecalis could be capable of inducing disease in a well-defined plant system.
From an evolutionary perspective, it is important to determine if E. faecalis virulence factors are also involved in plant pathogenesis. We found that three strains of E. faecalis exhibit potent pathogenicity in A. thaliana involving a sequential array of events: attachment to the leaf and root surfaces, congregation of bacteria in stomata or wounds, colonization in intercellular spaces, formation of communities on root surfaces, disruption of membrane structures, and maceration and rotting of the petiole. Additionally, our studies also show that some of the mammalian virulence-related factors in E. faecalis are involved in plant pathogenicity. In the present study, we have developed an experimental system for studying E. faecalis pathogenicity by using A. thaliana as the host. The observations presented below suggest that E. faecalis is a facultative pathogen of A. thaliana, capable of causing local and systemic infection leading to death of the infected plant.
MATERIALS AND METHODS
Plant material and growth conditions. Seeds of wild-type Arabidopsis thaliana ecotype Columbia (Col-0, Ler-0) were obtained from Lehle Seeds (Round Rock, Tex.). Seeds were surface sterilized using commercial sodium hypochlorite (0.3% [vol/vol]) for 10 to 12 min and then washed four times in sterile double-distilled water. Surface-sterilized seeds were placed on static Murashige and Skoog (MS) basal solid medium (Invitrogen, Inc.) in petri dishes for germination and incubated in a growth chamber. Fifteen-day-old seedlings of each ecotype were individually transferred to 50-ml culture tubes containing 5 ml of liquid MS basal medium. Plant cultures were maintained on an orbital platform shaker (Lab-Line Instruments) set at 90 rpm with a photoperiod of 16 h light and 8 h dark at 25 ± 2°C.
Bacterial strains and culture conditions. The following strains of E. faecalis were used in this study: FA-2-2 (2), V583 (38), and OG1RF (21), obtained from the laboratory of Frederick M. Ausubel (Harvard Medical School). E. faecalis mutants fsrA (TX5240), fsrB (TX5266), fsrC (TX5242), gelatinase gelE (TX5264), and serine protease sprE (TX5243) were obtained from Barbara E. Murray (Texas Medical Center). All strains were plated on brain heart infusion (BHI) agar medium (Difco) in 35-mm tissue-culture plates (Falcon) and incubated at 37°C. Tetracycline (12.5 μg ml–1) was added to the medium to selectively prevent growth of Escherichia coli. Lawns of bacteria were grown as follows: 2 ml of BHI were inoculated with a single colony of the appropriate strain and grown at 37°C for 6 to 8 h, and 10 μl of the culture was spread on each plate. The plates were incubated at 37°C overnight. Plated cells were suspended in 5 ml of Luria-Bertani broth for overnight growth at 37°C and shaken at 250 rpm.
In vitro root pathogenicity assay. Twenty-five-day-old A. thaliana plants were used for the in vitro root pathogenicity assays. E. faecalis strains were grown to an optical density at 600 nm (OD600) of 0.3 to 0.4 and added separately to the 5 ml of MS medium containing A. thaliana wild-type (Col-0 and Ler-0) plants to reach an initial OD600 of 0.02. MS basal medium (5 ml) without plant material was inoculated with the same volume of each bacterial strain tested. A noninfected plant control was maintained under the same conditions. All the treatments and controls were incubated at 30°C in a controlled environment incubator shaker (New Brunswick Scientific) set at 30 rpm with a photoperiod of 16 h light and 8 h dark. Root tissues (500 mg fresh weight basis) of A. thaliana infected with strains of E. faecalis (FA-2-2, V583, and OG1RF) as well as mutant fsrA (TX5240), fsrB (TX5266), fsrC (TX5242), gelatinase gelE (TX5264), and serine protease sprE (TX5243) strains were homogenized in 1 ml of saline (0.2% sodium chloride) with a tissue grinder (Kontes, size C), and the suspension was serially diluted in saline and plated to determine bacterial cell counts as previously described (1, 37, 46). Each experiment was conducted twice with five replicates.
Leaf and soil pathogenicity assay. Seeds of A. thaliana (Col-0 and Ler-0) were surface sterilized and germinated as described previously (see "Plant material and growth conditions"). Fifteen-day-old seedlings were transplanted from static MS medium to 10-cm black plastic pots containing 50 g (dry weight) of PM-O5 A. thaliana growing medium (Lehle Seeds). Plants were incubated in a growth chamber at 30°C with 12 h of light and watered daily for 2 weeks prior to inoculation with bacteria. For leaf assays, the E. faecalis strains (FA-2-2, V583, and OG1RF) were grown in BHI at 37°C to an OD600 of 0.2 to 0.3 and diluted 1:100. Diluted suspensions were individually injected with the blunt end of a hypodermic needle into intact leaves of A. thaliana at a dose of approximately 103 CFU/cm2 as previously described (34). Infiltrated plants were incubated in a growth chamber at 30°C and 80% relative humidity with 16 h light and 8 h dark. For soil infiltration, 50 g of soil with A. thaliana (Col-0) was flooded with 10 ml of bacterial suspension to give an inoculum concentration of 1 x 107 to 3 x 107 CFU/g of soil. Disease symptoms, such as leaf lesions and water-soaked lesions, were observed upon microscopy (described below). Bacterial cell counts from leaves infected with strains of E. faecalis (FA-2-2, V583, and OG1RF) were performed as described previously (1, 37, 46). Plants were incubated under conditions identical to those used for leaf infiltration assays. Each experiment was conducted twice with five replicates.
Phase-contrast microscopy, CSLM, and SEM. Phase-contrast images of E. faecalis-infected root tissues were captured with a x10 objective on an Olympus BX60 microscope equipped with CoolSnap (San Diego, Calif.) imaging software as described previously (1, 46). For scanning electron microscopy (SEM), segments of A. thaliana roots and leaves were fixed in 4% (wt/vol) paraformaldehyde and passed through increasing concentrations of ethanol (30, 50, 70, 96, and 100%). The fixed roots and leaves were dried in a Samdri-PVT-3B critical point drying apparatus, mounted on stubs, coated with a 12-nm layer of gold-palladium in a Hummer-II sputter coater, and visualized using a scanning electron microscope (Joel JSM-6500F; USA Inc., Peabody, Mass.). To determine whether E. faecalis root colonies were encased in a polysaccharide matrix, we stained the infected roots with Calcofluor, a polysaccharide-binding dye. After several water rinsings, the roots were stained for 30 min with 10 ml of 75-μg ml–1 Calcofluor (Fluostain; Sigma-Aldrich) in wash buffer. The stained bacterial communities were then analyzed by confocal scanning laser microscopy (CSLM) as described previously (1, 46). Phase-contrast microscopy, SEM, and CSLM were performed 4 days postinoculation. Samples were analyzed for fluorescence with a confocal laser microscope (Fluroview LGPS-2; Olympus). Samples were viewed using 488 nm as the excitation wavelength. Phase-contrast, fluorescence, and scanning electron microscopy were done 4 days postinoculation. Each experiment was viewed twice with three replicates each.
Transmission electron microscopy (TEM). Small pieces of infected or uninfected A. thaliana roots and leaves (1 by 3 mm) were fixed overnight in 3% (wt/vol) glutaraldehyde in 0.1 N cacodilate buffer at pH 7.2, washed in the same buffer, postfixed in 2% (wt/vol) osmium tetroxide, dehydrated in an alcohol series (30, 70, 96, and 100%), and embedded in Spurr medium at 60°C for 16 h. After polymerization, ultra-thin sections were cut with an LKB 8 800 Ultratome, stained in 2% (wt/vol) lead citrate and 2% (wt/vol) uranyl acetate, and analyzed by using an electron microscope (Jeol JEM-2000EX; USA Inc.).
RESULTS
In vitro and in-soil root pathogenicity of E. faecalis against A. thaliana. The root pathogenicity of E. faecalis strains FA-2-2, V583, and OG1RF was tested in vitro against two wild-type ecotypes of the model plant A. thaliana (Col-0 and Ler-0). All three E. faecalis strains, when inoculated into the liquid root medium of A. thaliana, caused characteristic disease-like symptoms, such as black necrotic regions at the root tips leading to complete cytoplasmic condensation, a clear signature of rhizotoxicity. A. thaliana infected with strains V583 and OG1RF displayed symptoms of infection, with bacterial cells initially infecting roots, the root-stem transition zone, and fully developed leaves positioned near the base of the plant 2 to 3 days postinoculation and spreading systemically to the top of the plant approximately 4 days postinoculation, with plant mortality occurring 7 days postinoculation (Fig. 1A). The maximum plant mortality was recorded with the OG1RF strain (Fig. 1B). In addition to in vitro studies, we tested the ability of strains FA-2-2, V583, and OG1RF to infect soil-grown A. thaliana plants (Materials and Methods). All strains caused extensive aerial tissue damage leading to plant mortality approximately 7 days postinoculation when infiltrated into the soil immediately surrounding the root system (Fig. 1C).
In planta bacterial growth of E. faecalis on Arabidopsis leaves and roots. Interestingly, all three E. faecalis strains caused condensation of the root meristem and mature regions, leading to complete cell death (unpublished data). In accordance with the rhizotoxicity, all three strains showed stable root colonization along with multiplication of bacteria on root tissues of A. thaliana (Col-0), as reflected by the increasing CFU counts during a time course analysis (Fig. 1D). In previously published work, we showed that A. thaliana ecotype Col-0 is highly susceptible to infection by another opportunistic human pathogen, P. aeruginosa strains PAO1 and PA14 (46). Figure 2A shows E. faecalis strains (FA-2-2, V583, and OG1RF) infiltrating A. thaliana Col-0 leaf surface structures after leaves were incubated at room temperature for 2 h in a suspension of E. faecalis cells and then incubated on the surface of 1.5% (wt/vol) water agar in closed petri dishes. The attached bacteria inflicted disease symptoms and leaf mortality under the above-mentioned conditions (Fig. 2A to C). In order to evaluate the correlation between the appearance of disease symptoms and the size of bacterial colonies infecting the leaf area, we next examined the percentage of infected leaf area in plants challenged with all three strains of E. faecalis (Fig. 2B). Consistent with the plant mortality results, all three strains revealed prominent water-soaked lesions on the upper dorsal leaf surface (adaxial) 7 days postinfection compared to the leaf surface of the untreated control (Fig. 2C). Each of the three E. faecalis strains behaved as a true pathogen, as shown by increasing CFU counts, reflecting in planta bacterial growth on infected leaves during the 7-day time course (Fig. 2D). These results indicate that E. faecalis strains trigger an active physiological response in A. thaliana roots and leaves.
Strains FA-2-2, V583, and OG1RF forms a biofilm-like community on root surfaces. As part of establishing the mode of attachment by the three E. faecalis strains (FA-2-2, V583, and OG1RF) on root surfaces, we determined that these strains adhere to root surfaces of A. thaliana. Four days post-cocultivation of A. thaliana roots and E. faecalis strains (FA-2-2, V583, OG1RF) in MS medium, roots were viewed by phase-contrast microscopy and CSLM. We observed that cells of the three E. faecalis strains (FA-2-2, V583, and OG1RF) had colonized virtually the entire root surface (unpublished data. Phase-contrast microscopy and CSLM revealed that roots of A. thaliana were surrounded by phase-bright material suggestive of an extracellular matrix (unpublished data). Soil-grown A. thaliana, when infected with E. faecalis strains (FA-2-2, V583, and OG1RF), showed a similar root colonization by the bacteria (data not shown), suggesting that E. faecalis strains FA-2-2, V583, and OG1RF form a stable, biofilm-like community on these roots.
Effect of E. faecalis infection on plant ultrastructure. The leaves of A. thaliana were incubated with E. faecalis strains FA-2-2, V583, and OG1RF and then transferred to water agar plates at room temperature for viewing. After cells of the three E. faecalis strains entered the substomatal cavity, they began multiplying and rapidly spread through the leaf mesophyll. As in the case of other well-studied phytopathogenic bacteria, E. faecalis strains FA-2-2, V583, and OG1RF were also found to make a "bacterial ring" (dense populations of bacterial cells that adopt a ring structure) that colonized the intercellular space, presumably by digesting the middle lamellae and separating the plant cells from each other (Fig. 3A to C). At relatively early stages of the infection (2 days), almost all the bacteria were found in intercellular spaces attached to A. thaliana mesophyll walls (Fig. 3D). At later stages of infection (3 to 4 days), many of the mesophyll cells were severely damaged and contained intracellular bacteria (Fig. 3C).
Infection of A. thaliana leaves with E. faecalis strains FA-2-2, V583, and OG1RF had a unique effect on the structures of the host cell walls and membranes and on the location and organization of host cell organelles (Fig. 3A to D). In contrast, uninfected A. thaliana mesophyll cells had a large central vacuole surrounded by a thin layer of cytoplasm containing a nucleus and organelles. The first sign of host cell degeneration following infection was slight plasmolysis and concentration of host membrane structures, including chloroplasts, endoplasmic reticulum, and dictyosomes, at the site of bacterial contact. At this stage a limited number of single bacteria appeared to be able to penetrate into metabolically active plant cells (Fig. 3A to D). In roots, also, the proliferation of bacteria in the intercellular space resulted in alteration of host organelles. Host plasmalemma became highly undulated and disrupted (Fig. 3E to H), and there was swelling and disruption of outer cell membranes and finally the destruction of the whole root structure (Fig. 3F to H). Cell organelles were degraded, and the cell walls were thin and highly convoluted. The bacteria appeared as a colony in the intercellular spaces (Fig. 3). Host cell collapse in both leaves and roots was the final step of the bacterial infection.
Development of the bacterial infection and cellular mode of attachment on A. thaliana leaves and roots. To visualize the bacterial infection and cellular mode of attachment of strains FA-2-2, V583, and OG1RF, the leaf and root surfaces of A. thaliana were viewed by SEM and TEM. SEM of infected leaves and roots showed that E. faecalis cells attached perpendicularly and horizontally to the leaf and root cell walls of A. thaliana (Fig. 4A to C). Bacterial cells oriented in either position appeared to be degrading and penetrating through the outermost layers of the A. thaliana root or leaf cell wall (Fig. 4A to H). Viewing FA-2-2, V583, and OG1RF infection with a higher magnification of SEM revealed that bacterial cells were embedded within and connected together by an extracellular polymeric matrix (Fig. 4C, E, and G). Individual cells of E. faecalis strains FA-2-2, V583, and OG1RF could be seen attaching along the leaf surface towards open stomata (Fig. 4D, F, and H). In the experiments described above, Col-0 leaves were incubated with E. faecalis (FA-2-2, V583, and OG1RF) and then transferred to water agar plates at room temperature. The infection seemed to proceed in a manner similar to that with other well-studied phytopathogenic bacteria, such as P. aeruginosa (PA14), which formed "bacterial threads" (dense populations of bacterial cells that adopt an elongated and branched structure) that colonized the intercellular space, presumably by digesting the middle lamellae and separating the plant cells from each other (34). At relatively early stages of the infection (2 days), almost all the E. faecalis bacteria were found in intercellular spaces attached to A. thaliana mesophyll walls (Fig. 5A and B), although some bacterial cells could also be observed attached to the inner surface of cell walls (Fig. 5). At later stages of infection (3 to 4 days), many of the mesophyll cells were severely damaged and contained intracellular bacteria (Fig. 3 and 5). TEM also revealed that roots and leaves of A. thaliana infected with FA-2-2, V583, and OG1RF showed the presence of bacteria in the intracellular cavities, inflicting pathogenesis (Fig. 5A and B).
Role of fsr quorum sensing in E. faecalis plant pathogenicity. The above-described results strengthened our hypothesis that biofilm-like community development and root colonization contributed to E. faecalis pathogenicity in A. thaliana plants. In E. faecalis, the fsr system positively regulates the expression of several virulence factors (29, 40). The fsr locus is composed of three regulatory genes: fsrA, fsrB, and fsrC (35, 40, 44). To determine whether the previously observed effect of a fsrB mutation on E. faecalis virulence in the Caenorhabditis elegans model (7, 8) was specific to fsrB or representative of the entire fsr locus, we evaluated the ability of fsrA (TX5240) and fsrC (TX5242) deletion mutants to kill A. thaliana. Compared with the wild-type strain OG1RF, the fsrA and fsrC strains showed no significant difference (P < 0.0002) in their ability to kill A. thaliana (Fig. 6A and B). Interestingly, infections with the deletion mutant fsrB revealed an attenuated pathogenicity in the A. thaliana root model (Fig. 6A). A. thaliana plants were susceptible to deletion fsrA and fsrC quorum-sensing mutants, displaying symptoms of infection similar to those observed with infection with wild-type OG1RF and succumbing to infection 7 days postinoculation (Fig. 6A). In accordance with the reduced mortality results with the fsrB strain, fewer bacterial cells were attached to the root surfaces and colonization appeared to be diminished (Fig. 6B to D). Notably, CSLM showed that the fsrB strain appeared to form a nearly diminished biofilm-like community, while the communities formed by the fsrA and fsrC strains appeared full-grown compared to those for the wild type, OG1RF (Fig. 6D). The fact that CSLM of fsrB strain root infectivity on A. thaliana shows the reduced presence of E. faecalis cells authenticates the involvement of a quorum-sensing system in the root model (Fig. 6D). To evaluate the number of bacterial cells associated with biofilm-like community formation in planta, we analyzed the cell counts on the root surface of A. thaliana on the fourth day after infection with all the deletion mutants of E. faecalis (TX5240 [fsrA], TX5266 [fsrB], and TX5242 [fsrC]) (Fig. 6C). The cell counts showed a positive correlation between the bacterial cell counts on the root surface, the degree of community formation, and pathogenicity of E. faecalis fsrB mutant strains on A. thaliana (Fig. 6A to D).
Deletion in serine protease (sprE) but not gelE causes attenuation of bacterial infection in the A. thaliana root pathogenicity model. Previous work has shown that a disruption of some of the virulence-related factors of E. faecalis, particularly gelatinase (gelE), causes attenuation of the E. faecalis strain OG1RF in the mouse peritonitis and nematode models (7, 8, 40, 42). Three genes located directly upstream of gelE, fsrA, fsrB, and fsrC, seem to be involved in gelE regulation (35). FsrA and FsrB are homologous to two-component response regulators and sensor kinases (35, 42), respectively, and a nonpolar fsrB deletion blocks the production of gelatinase (35). Based on the homology of FsrA, FsrB, and FsrC to the Staphylococcus aureus quorum-sensing system encoded by agrA, agrB, and agrC, fsrB may encode a processor of a putative E. faecalis signal peptide (42). Figure 6A to D shows that A. thaliana survived when infected with the fsrB gelatinase regulatory mutant (strain TX5266); when infected with the fsrA, fsrC, or gelE mutant or the isogenic parental strain OG1RF, it did not survive (Fig. 6A to D). E. faecalis OG1RF mutants containing disruptions in fsrA, fsrC, and gelE negated prolonged survival of A. thaliana, which was nearly opposite to the response observed with a single disruption (fsrB) (Fig. 6A to D). Notably, it is reported that the fsrB mutant is also less virulent in the mouse intraperitoneal injection and nematode model (7, 8, 35, 40, 42). Interestingly, neither gelatinase nor serine protease activity is reported for the fsr mutants TX5240, TX5242, and TX5266 (35). We evaluated the contribution of E. faecalis gelatinase (gelE) and serine protease (sprE) separately from its phytotoxic activity on the A. thaliana root-infection model. In-frame deletion mutants, TX5264 (gelE) and TX5243 (sprE), were used as reported previously (35). Compared to each other, TX5264 (gelE) and TX5243 (sprE) were significantly different in their ability to cause plant mortality in A. thaliana (P < 0.05) (Fig. 6A to D). The serine protease sprE mutant (TX5243) was completely attenuated in its ability to cause plant mortality compared to the deletion gelatinase gelE mutant (TX5264) (Fig. 6A to D). Interestingly, CSLM showed that the sprE mutant appeared to form poor communities, while the colonization formed by the fsrA, fsrC, and gelE strains appeared developed compared to the colonization of wild-type OG1RF (Fig. 6D). As observed before with fsr deletion mutants, the cell counts showed a positive correlation between the bacterial cell counts on the root surface, the degree of biofilm formation, and pathogenicity of E. faecalis sprE mutant strains on A. thaliana (Fig. 6A to D). These results suggest that an fsr quorum sensing system regulates genes involved in E. faecalis virulence; such a system is probably complemented with serine protease (sprE) in the A. thaliana root pathogenicity model.
DISCUSSION
In the present communication, we have found that E. faecalis strains FA-2-2, V583, and OG1RF can infect roots and leaves of A. thaliana. All three strains of E. faecalis inflict local and systemic infection, leading to the death of the infected A. thaliana plant. Plant infection and subsequent mortality due to E. faecalis were traced to the formation of a pathogenic biofilm-like community colonizing the root surface. This phenomenon is similar to that observed with the opportunistic human pathogen P. aeruginosa and the classical A. thaliana pathogen P. syringae (DC3000), where pathogenic biofilms on the root surface have been shown to be partly responsible for plant mortality (1, 46). E. faecalis infection in A. thaliana, as viewed by various microscopic techniques, starts with bacterial attachment to the plant leaf and root surfaces and entry into the A. thaliana tissues via stomatal openings (in the case of the leaves), followed by proliferation in the substomatal cavities and intercellular space. As the bacteria proliferate, A. thaliana cell walls become undulated and at least some bacteria appear to penetrate the walls. Aggressive E. faecalis propagation in the intercellular space results in maceration and autolysis of plant cells. These data strongly support the observation that the soft-rot symptoms were elicited by E. faecalis on A. thaliana leaves and roots; similar disease symptoms have been reported earlier in the P. aeruginosa-A. thaliana infection model (34, 37, 46). It has previously been reported that strains of E. faecalis are capable of infecting the nematode C. elegans and the mouse peritoneal model (7, 8, 40, 42). This is the first report describing how E. faecalis causes infections in A. thaliana as determined by these sequential characteristics: attachment to leaf or root surface, entry through stomata or wounds, colonization and multiplication in the intercellular spaces leading to plant cell walls, biofilm-like community formation, disruption of membrane structures, and rotting, ultimately leading to complete plant mortality.
The pathogenicity of the tested strains of E. faecalis, FA-2-2, V583, and OG1RF, to the roots of A. thaliana in vitro and in soil indicates that E. faecalis virulence in both settings is similar and that our experimental system is a reliable method for further studying the interaction between E. faecalis and plant roots. Since FA-2-2, V583, and OG1RF were capable of causing the mortality of A. thaliana, we followed the bacterial interactions with the roots by using SEM and TEM, phase-contrast microscopy, and CSLM. As previously reported for P. aeruginosa's mode of infection by attachment on A. thaliana leaves and roots (34, 46), we observed FA-2-2, V583, and OG1RF cells attached perpendicularly to the roots and leaves of A. thaliana (Fig. 3 to 5). Furthermore, CSLM and phase-contrast microscopy of root tissue confirmed the formation of biofilm-like communities on the A. thaliana root surface (unpublished data).
A. thaliana ecotypes exhibit a similar degree of susceptibility to E. faecalis infection. We also show that E. faecalis penetrates and forms larger lesions on the leaves and roots of the susceptible Col-0 and Ler-0 ecotypes (data not shown) but that E. faecalis also attaches at least 1.5-fold more efficiently to the epidermis of Ler-0 roots than to Col-0 roots (data not shown).
One reason that relatively little is known about enterococcal virulence factors is that the mammalian models used to study enterococcal infections are cumbersome and expensive (35, 36, 40, 42). Using a mammalian host to screen enterococcal mutant libraries for avirulent mutants, for example, would be prohibitively time consuming and expensive because of the large number of animals involved. Therefore, we have sought to develop an alternative and defined plant model system like A. thaliana for Enterococcus infections. As we have learned more about the mechanisms and epidemiology of resistance to antimicrobial drugs, it has become clear that bacteria have a remarkable array of tools at their disposal to overcome antibiotics (9, 26). Like other gram-positive microorganisms, enterococci are able to produce biofilms on abiotic surfaces (6, 18), increasing their high innate resistance to antibiotics (6); yet the factors controlling enterococcal biofilm formation and maintenance remain unknown (22, 31). Plant pathogenesis related to biofilm formation is now well documented in the cases of the gram-negative bacteria P. syringae DC3000 and P. aeruginosa (1, 46). Our intriguing observation that E. faecalis forms a pathogenic biofilm-like community for colonization of the biotic surface of A. thaliana roots suggests that E. faecalis, like other gram-negative bacteria, also employs community formation as an important virulence factor in its potent pathogenicity against A. thaliana. Interestingly, not many gram-positive bacteria have been reported as plant pathogens.
To dissect the involvement of mammalian virulence-related factors in plant pathogenicity, we tested E. faecalis mutant fsrA (TX5240), fsrB (TX5266), fsrC (TX5242), gelatinase gelE (TX5264), and serine protease sprE (TX5243) strains. Two E. faecalis virulence-related factors that play an important role in mammalian and nematode models of infection, a putative quorum-sensing system (fsrB) and serine protease (sprE), are also important for plant pathogenesis. Quorum sensing is a cell density-dependent regulatory system that controls a variety of group behaviors in bacteria (13, 32). In E. faecalis, the fsr system positively regulates the expression of gelatinase and serine protease in a cell density-dependent manner, similar to the well-studied regulation of toxins by the S. aureus agr quorum-sensing regulatory locus (16, 32). Qin et al. (35, 36) have characterized three genes in the fsr regulatory locus, fsrA, fsrB, and fsrC. Using a nonpolar fsrB deletion mutant, the same workers showed that fsrB is required for the regulatory function of the Fsr system (35, 36). The expression of the fsr genes in E. faecalis OG1RF is cell density dependent and is most active in the post-exponential phase of growth (30, 35, 36). Supporting the already-existing knowledge about cell density-dependent pathogenicity of E. faecalis using the fsr system, we also found that fsrB plays an important role in regulating bacterial colonization on the A. thaliana root surface (Fig. 6). Accordingly, a fsrB deletion mutant failed to colonize the A. thaliana roots and exhibited an attenuated pathogenicity along with decreased CFU counts (Fig. 6). Contrastingly, it has already been reported that fsrA and fsrC insertion mutants, like the fsrB mutant, were attenuated in their ability to kill C. elegans, though in a rabbit endophthalmitis model, the fsrB mutant showed significantly reduced virulence compared to the wild type (7, 8, 29). The fact that the virulence of the fsrB strain varies between invertebrate and mammalian pathogenicity systems supports the unexpected result obtained in our studies with quorum-sensing mutants (fsrB) and also indicates that the virulence factors regulated by the fsr system may be important for the full expression of E. faecalis pathogenicity towards A. thaliana.
Along with the fsr system in E. faecalis, serine protease (sprE) is an additional virulence factor thought to play a role in systemic disease in mammalian hosts (35, 40). The serine protease gene sprE, which lies immediately downstream of and is cotranscribed with gelE, encodes a secreted 26-kDa serine protease that shares homology with S. aureus V8 protease (35). Insertion disruption of sprE also attenuates virulence in the mouse peritonitis and C. elegans model systems (35, 40). Transcription of the gelE-sprE operon is positively regulated in a growth phase-dependent fashion by the fsr locus, which shares many similarities with the well-studied S. aureus agr regulatory locus (35). The fsr locus is composed of three regulatory genes, fsrA, fsrB, and fsrC, located upstream of the gelE-sprE operon (35). As reported before with the mouse peritonitis and C. elegans models with a deletion mutant of SprE (sprE), we also observed attenuated killing with the sprE mutant in the A. thaliana root pathogenicity model (Fig. 6), suggesting the sharing of similar virulence factors in mammalian, nematode, and plant models.
The E. faecalis fsr system is the second example (after the rhl and las systems of P. aeruginosa) of a quorum-sensing system that regulates virulence gene expression in bacterial infection of both simple model organisms and mammalian hosts. Quorum sensing may be an important mechanism used by many prokaryotes to adapt to different environments encountered during pathogenesis. Our results also raise the possibility that the fsr system in E. faecalis regulates virulence genes in addition to sprE in this pathogen in the A. thaliana root pathogenicity model. In E. faecalis, improvement of the selection of effective antimicrobial agents for use against recalcitrant infections is urgently needed. Taking into account the strong correlation between the presence of fsr and the ability to produce a biofilm-like community, it may be possible to screen for plant-root-exuded products against putative biofilm-forming E. faecalis strains by using the present plant-pathogen model to directly isolate antimicrobial and anti-infective root-exuded metabolites. The bacterial phenotype (acting by adherence or biofilm formation on root surface) presented in our study and the very nature of this screening is advantageous in that it allows preliminary identification of strains which are highly adherent and are thus good candidates for testing the antibiotic susceptibility of biofilms. The development of the E. faecalis-A. thaliana model system could potentially be used to circumvent certain inherent limitations that an animal model imposes on the identification and study of virulence factors. Furthermore, our study unravels some evolutionary crossover of virulence factors in plant pathogenesis which have previously been found relevant for mammalian and nematode models of infection. Finally, it is alluring to speculate that E. faecalis-crop plant interactions may occur in nature and that plants and crops may likely act in the epidemiology of this multihost pathogen.
ACKNOWLEDGMENTS
We thank Barbara E. Murray (Texas Medical Center) for generously providing E. faecalis deletion mutants for our studies.
This work was supported by grants from Colorado State University Agricultural Experiment Station (to J.M.V.) and NSF-CAREER (grant no. MCB 0093014 to J.M.V.).
These two authors contributed equally to this work.
Present address: Department of Biochemistry and Molecular Biology, Colorado State University, Fort Collins, CO 80523-1173.
REFERENCES
1. Bais, H. P., R. Fall, and J. M. Vivanco. 2004. Biocontrol of Bacillus subtilis against infection of Arabidopsis roots by Pseudomonas syringae is facilitated by biofilm formation and surfactin production. Plant Physiol. 134:307-319.
2. Bottone, E. J., L. Patel, P. Patel, and T. Robin. 1998. Mucoid encapsulated Enterococcus faecalis: an emerging morphotype isolated from patients with urinary tract infections. Diagn. Microbiol. Infect. Dis. 31:429-430.
3. Cho, J. J., N. J. Panopoulos, and M. N. Schroth. 1975. Genetic transfer of Pseudomonas aeruginosa R factors to plant pathogenic Erwinia species. J. Bacteriol. 122:192-198.
4. Chow, J. W., L. A. Thal, M. B. Perri, J. A. Vazquez, S. M. Donabedian, D. B. Clewell, and M. J. Zervos. 1993. Plasmid-associated hemolysin and aggregation substance production contribute to virulence in experimental enterococcal endocarditis. Antimicrob. Agents Chemother. 37:2474-2477.
5. Fenselau, S., I. Balbo, and U. Bonas. 1992. Determinants of pathogenicity in Xanthomonas campestris pv. vesicatoria are related to proteins involved in secretion in bacterial pathogens of animals. Mol. Plant-Microbe Interact. 5:390-396.
6. Foley, I., and P. Gilbert. 1997. In-vitro studies of the activity of glycopeptide combinations against Enterococcus faecalis biofilms. J. Antimicrob. Chemother. 40:667-672.
7. Garsin, D. A., C. D. Sifri, E. Mylonakis, X. Qin, K. V. Singh, B. E. Murray, S. B. Calderwood, and F. M. Ausubel. 2001. A simple model host for identifying Gram-positive virulence factors. Proc. Natl. Acad. Sci. USA 98:10892-10897.
8. Garsin, D. A., J. M. Villanueva, J. Begun, D. H. Kim, C. D. Sifri, S. B. Calderwood, G. Ruvkun, and F. M. Ausubel. 2003. Long-lived C. elegans daf-2 mutants are resistant to bacterial pathogens. Science 300:1921.
9. Gold, H. S., and R. C. Moellering, Jr. 1996. Antimicrobial-drug resistance. N. Engl. J. Med. 335:1445-1453.
10. Goldmann, D. A., and J. D. Klinger. 1986. Pseudomonas cepacia: biology, mechanisms of virulence, epidemiology. J. Pediatr. 108:806-812.
11. Gough, C. L., S. Genin, C. Zischek, and C. A. Boucher. 1992. hrp genes of Pseudomonas solanacearum are homologous to pathogenicity determinants of animal pathogenic bacteria and are conserved among plant pathogenic bacteria. Mol. Plant-Microbe Interact. 5:384-389.
12. Haas, W., and M. S. Gilmore. 1999. Molecular nature of a novel bacterial toxin: the cytolysin of Enterococcus faecalis. Med. Microbiol. Immunol. (Berlin) 187:183-190.
13. Haas, W., B. D. Shepard, and M. S. Gilmore. 2002. Two-component regulator of Enterococcus faecalis cytolysin responds to quorum-sensing autoinduction. Nature 415:84-87.
14. Huang, H. C., Y. Xiao, R. H. Lin, Y. Lu, S. W. Hutcheson, and A. Collmer. 1993. Characterization of the Pseudomonas syringae pv. syringae 61 hrpJ and hrpI genes: homology of HrpI to a superfamily of proteins associated with protein translocation. Mol. Plant-Microbe Interact. 6:515-520.
15. Ike, Y., H. Hashimoto, and D. B. Clewell. 1984. Hemolysin of Streptococcus faecalis subspecies zymogenes contributes to virulence in mice. Infect. Immun. 45:528-530.
16. Jarraud, S., C. Mougel, J. Thioulouse, G. Lina, H. Meugnier, F. Forey, X. Nesme, J. Etienne, and F. Vandenesch. 2002. Relationships between Staphylococcus aureus genetic background, virulence factors, agr groups (alleles), and human disease. Infect. Immun. 70:631-641.
17. Johansson, J., and P. Cossart. 2003. RNA-mediated control of virulence gene expression in bacterial pathogens. Trends Microbiol. 11:280-285.
18. Joyanes, P., A. Pascual, L. Martinez-Martinez, A. Hevia, and E. J. Perea. 2000. In vitro adherence of Enterococcus faecalis and Enterococcus faecium to urinary catheters. Eur. J. Clin. Microbiol. Infect. Dis. 19:124-127.
19. Kayser, F. H. 2003. Safety aspects of enterococci from the medical point of view. Int. J. Food Microbiol. 88:255-262.
20. Kominos, S. D., C. E. Copeland, B. Grosiak, and B. Postic. 1972. Introduction of Pseudomonas aeruginosa into a hospital via vegetables. Appl. Microbiol. 24:567-570.
21. Li, X., G. M. Weinstock, and B. E. Murray. 1995. Generation of auxotrophic mutants of Enterococcus faecalis. J. Bacteriol. 177:6866-6873.
22. Mah, T. F., and G. A. O'Toole. 2001. Mechanisms of biofilm resistance to antimicrobial agents. Trends Microbiol. 9:34-39.
23. Moellering, R. C., Jr. 1992. Emergence of Enterococcus as a significant pathogen. Clin. Infect. Dis. 14:1173-1176.
24. Moellering, R. C., Jr. 1991. The Garrod lecture. The enterococcus: a classic example of the impact of antimicrobial resistance on therapeutic options. J. Antimicrob Chemother. 28:1-12.
25. Muller, T., A. Ulrich, E. M. Ott, and M. Muller. 2001. Identification of plant-associated enterococci. J. Appl. Microbiol. 91:268-278.
26. Mundy, L. M., D. F. Sahm, and M. Gilmore. 2000. Relationships between enterococcal virulence and antimicrobial resistance. Clin. Microbiol. Rev. 13:513-522.
27. Murray, B. E. 1998. Diversity among multidrug-resistant enterococci. Emerg. Infect. Dis. 4:37-47.
28. Murray, B. E. 1990. The life and times of the Enterococcus. Clin. Microbiol. Rev. 3:46-65.
29. Mylonakis, E., M. Engelbert, X. Qin, C. D. Sifri, B. E. Murray, F. M. Ausubel, M. S. Gilmore, and S. B. Calderwood. 2002. The Enterococcus faecalis fsrB gene, a key component of the fsr quorum-sensing system, is associated with virulence in the rabbit endophthalmitis model. Infect. Immun. 70:4678-4681.
30. Nakayama, J., Y. Cao, T. Horii, S. Sakuda, A. D. Akkermans, W. M. de Vos, and H. Nagasawa. 2001. Gelatinase biosynthesis-activating pheromone: a peptide lactone that mediates a quorum sensing in Enterococcus faecalis. Mol. Microbiol. 41:145-154.
31. O'Toole, G. A., L. A. Pratt, P. I. Watnick, D. K. Newman, V. B. Weaver, and R. Kolter. 1999. Genetic approaches to study of biofilms. Methods Enzymol. 310:91-109.
32. Otto, M. 2001. Staphylococcus aureus and Staphylococcus epidermidis peptide pheromones produced by the accessory gene regulator agr system. Peptides 22:1603-1608.
33. Paulsen, I. T., L. Banerjei, G. S. Myers, K. E. Nelson, R. Seshadri, T. D. Read, D. E. Fouts, J. A. Eisen, S. R. Gill, J. F. Heidelberg, H. Tettelin, R. J. Dodson, L. Umayam, L. Brinkac, M. Beanan, S. Daugherty, R. T. DeBoy, S. Durkin, J. Kolonay, R. Madupu, W. Nelson, J. Vamathevan, B. Tran, J. Upton, T. Hansen, J. Shetty, H. Khouri, T. Utterback, D. Radune, K. A. Ketchum, B. A. Dougherty, and C. M. Fraser. 2003. Role of mobile DNA in the evolution of vancomycin-resistant Enterococcus faecalis. Science 299:2071-2074.
34. Plotnikova, J. M., L. G. Rahme, and F. M. Ausubel. 2000. Pathogenesis of the human opportunistic pathogen Pseudomonas aeruginosa PA14 in Arabidopsis. Plant Physiol. 124:1766-1774.
35. Qin, X., K. V. Singh, G. M. Weinstock, and B. E. Murray. 2000. Effects of Enterococcus faecalis fsr genes on production of gelatinase and a serine protease and virulence. Infect. Immun. 68:2579-2586.
36. Qin, X., K. V. Singh, G. M. Weinstock, and B. E. Murray. 2001. Characterization of fsr, a regulator controlling expression of gelatinase and serine protease in Enterococcus faecalis OG1RF. J. Bacteriol. 183:3372-3382.
37. Rahme, L. G., E. J. Stevens, S. F. Wolfort, J. Shao, R. G. Tompkins, and F. M. Ausubel. 1995. Common virulence factors for bacterial pathogenicity in plants and animals. Science 268:1899-1902.
38. Sahm, D. F., J. Kissinger, M. S. Gilmore, P. R. Murray, R. Mulder, J. Solliday, and B. Clarke. 1989. In vitro susceptibility studies of vancomycin-resistant Enterococcus faecalis. Antimicrob. Agents Chemother. 33:1588-1591.
39. Schaberg, D. R. 1994. Resistant gram-positive organisms. Ann. Emerg. Med. 24:462-464.
40. Sifri, C. D., E. Mylonakis, K. V. Singh, X. Qin, D. A. Garsin, B. E. Murray, F. M. Ausubel, and S. B. Calderwood. 2002. Virulence effect of Enterococcus faecalis protease genes and the quorum-sensing locus fsr in Caenorhabditis elegans and mice. Infect. Immun. 70:5647-5650.
41. Silo-Suh, L., S. J. Suh, P. A. Sokol, and D. E. Ohman. 2002. A simple alfalfa seedling infection model for Pseudomonas aeruginosa strains associated with cystic fibrosis shows AlgT (sigma-22) and RhlR contribute to pathogenesis. Proc. Natl. Acad. Sci. USA 99:15699-15704.
42. Singh, K. V., X. Qin, G. M. Weinstock, and B. E. Murray. 1998. Generation and testing of mutants of Enterococcus faecalis in a mouse peritonitis model. J. Infect. Dis. 178:1416-1420.
43. Starr, M. P., and A. K. Chatterjee. 1972. The genus Erwinia: enterobacteria pathogenic to plants and animals. Annu. Rev. Microbiol. 26:389-426.
44. Stevens, S. X., H. G. Jensen, B. D. Jett, and M. S. Gilmore. 1992. A hemolysin-encoding plasmid contributes to bacterial virulence in experimental Enterococcus faecalis endophthalmitis. Investig. Ophthalmol. Vis. Sci. 33:1650-1656.
45. Tan, M. W., L. G. Rahme, J. A. Sternberg, R. G. Tompkins, and F. M. Ausubel. 1999. Pseudomonas aeruginosa killing of Caenorhabditis elegans used to identify P. aeruginosa virulence factors. Proc. Natl. Acad. Sci. USA 96:2408-2413.
46. Walker, T. S., H. P. Bais, E. Deziel, H. P. Schweizer, L. G. Rahme, R. Fall, and J. M. Vivanco. 2004. Pseudomonas aeruginosa-plant root interactions. Pathogenicity, biofilm formation, and root exudation. Plant Physiol. 134:320-331.(Ajay K. Jha , Harsh P. Ba)