Role of Rhizobium endoglucanase CelC2 in cellulose biosynthesis and biofilm formation on plant roots and abiotic surfaces
© Robledo et al.; licensee BioMed Central Ltd. 2012
Received: 21 May 2012
Accepted: 6 September 2012
Published: 12 September 2012
The synthesis of cellulose is among the most important but poorly understood biochemical processes, especially in bacteria, due to its complexity and high degree of regulation. In this study, we analyzed both the production of cellulose by all known members of the Rhizobiaceae and the diversity of Rhizobium celABC operon predicted to be involved in cellulose biosynthesis. We also investigated the involvement in cellulose production and biofilm formation of celC gene encoding an endoglucanase (CelC2) that is required for canonical symbiotic root hair infection by Rhizobium leguminosarum bv. trifolii.
ANU843 celC mutants lacking (ANU843ΔC2) or overproducing cellulase (ANU843C2+) produced greatly increased or reduced amounts of external cellulose micro fibrils, respectively. Calcofluor-stained cellulose micro fibrils were considerably longer when formed by ANU843ΔC2 bacteria rather than by the wild-type strain, in correlation with a significant increase in their flocculation in batch culture. In contrast, neither calcofluor-stained extracellular micro fibrils nor flocculation was detectable in ANU843C2+ cells. To clarify the role of cellulose synthesis in Rhizobium cell aggregation and attachment, we analyzed the ability of these mutants to produce biofilms on different surfaces. Alteration of wild-type CelC2 levels resulted in a reduced ability of bacteria to form biofilms both in abiotic surfaces and in planta.
Our results support a key role of the CelC2 cellulase in cellulose biosynthesis by modulating the length of the cellulose fibrils that mediate firm adhesion among Rhizobium bacteria leading to biofilm formation. Rhizobium cellulose is an essential component of the biofilm polysaccharidic matrix architecture and either an excess or a defect of this “building material” seem to collapse the biofilm structure. These results position cellulose hydrolytic enzymes as excellent anti-biofilm candidates.
KeywordsCellulose biosynthesis Rhizobium Cellulases Biofilm Symbiosis
Symbioses between diazotrophic rhizobia and legume plants are of critical agronomic and environmental importance, making crop production possible in nitrogen-limited soils without fertilizer supply. Rhizobia grow as free-living organisms, but can also induce and colonize root nodules in legume plants thereby establishing a partnership that benefits both organisms. This process begins when flavonoids secreted by the plant induce Rhizobium nod genes, which are involved in the synthesis and secretion of lipo-chitooligosaccharide signal molecules, known as Nod factors. In response, plant root hairs deform and exhibit a typical marked curling to facilitate bacteria penetration. The interaction continues with the initiation of the root nodule, where bacterial cells are released into the host cells. Eventually, upon a morphological differentiation into bacteroids, bacteria fix atmospheric dinitrogen into ammonia. Among the many factors involved in development of an effective symbiosis between rhizobia and their host plants, those associated with adherence and colonization of bacteria on the surface of roots and the root hair tip —a key stage for the subsequent entry into the plant— have not been fully characterized. Root attachment and colonization by rhizobia follow the two-phase sequence of events previously described for bacteria in general. Several bacterial proteins, such as adhesins, and flagellar proteins[2, 3], have been proposed to be important factors for the early reversible, specific binding events, whereas bacterial cell-surface polysaccharides are the main components involved in the later, irreversible attachment stages. Some of these cell surface components include exopolysaccharide, lipopolysaccharides, and cyclic β-1,2-glucans[4–6], but mainly cellulose fibrils, which mediate firm adhesion of the bacteria to root hairs and anchor bacteria to the root surface[1, 7, 8]. Hapten-inhibitable root hair lectins are also involved in this dynamic attachment process[9–12].
Attachment of bacteria to a substratum is the initial step in biofilm formation, which is followed by the establishment of micro colonies by clonal propagation, and final maturation into three-dimensional structures that are covered by exopolymer and other matrix materials[3, 5, 13, 14]. These bacterial three-dimensional structures are of considerable biotechnological importance due to their implications not only in bacterial colonization of abiotic surfaces with an economic value, but also in almost all pathogenic infections, making them recalcitrant due to their multidrug resistance. Initially, the proposed function for cellulose in bacteria was not linked to biofilm formation. However, recent studies have revealed that some species of the family Enterobacteriaceae (e.g., Citrobacter spp., Enterobacter spp., and Klebsiella spp.) produce cellulose as a crucial component of the bacterial extracellular matrix (reviewed in 25). In wild-type rhizobia, this polysaccharide of industrial interest is produced in flocculating batch cultures, some constitutively, others at early stationary phase, and upon contact with the roots of the host plant. Cellulose has been proposed to be required for optimal infection of long root hairs and biofilm cap formation at pH 6.5 and 7.5. Other studies have shown that cellulose is involved in anchoring the pathogen Agrobacterium tumefaciens to plant tissue, thereby affecting virulence. In plant culture, Rhizobium leguminosarum bv. trifolii polymerizes β-1-4 glucose residues into cellulose micro fibrils that are anchored to the roots of the host plant Trifolium repens[1, 8].
Biosynthesis of cellulose, a plant process required for cell wall growth, has also been described in some bacteria and even in one group of animals, the urochordates. Although significant in the medical, agricultural and ecological context, the biological role of cellulose biosynthesis and its regulation has not been widely studied in bacteria, apart from Gluconacetobacter xylinum (formerly called Acetobacter xylinum). However, it is widespread in Gram negative and Gram positive bacteria and also in some species of cyanobacteria[15, 22–25].
In both bacteria and plants, two proteins have been identified as directly involved in the biosynthesis of cellulose: CesA and Korrigan in plants, and CelA and CelC in bacteria[23, 26]. CelA and CesA have homology with glycosyl transferases while CelC and Korrigan are homologous to endoglucanases. Orthologs of the putative cellulose biosynthesis genes celABC have been found in a region of the chromosome that is involved in cellulose biosynthesis among a variety of bacteria that synthesize cellulose[16, 19, 27, 28].
The first gene in the operon is bcsA (bacterial cellulose synthesis), also named acsA or celA. It encodes a cellulose synthase that harbors a β-glycosyltransferase 2 domain and binds the substrate UDP-glucose. It is also the longest and best-conserved protein encoded by the bcs operon among diverse species. The second gene, bcsB (synonyms: acsB, celB), encodes a cyclic diguanylic acid (c-di-GMP) binding protein and is less conserved. The last gene, bcsZ (also called celC in A. tumefaciens and R. leguminosarum bv. trifolii) has been shown to encode a cellulase (family 8 glycosyl hydrolase), which is present in all cellulose-producing species. Rhizobium celC is also located in the celABC operon. BcsZ homologs are also located in the cellulose biosynthesis operons of enterobacterial species, but outside yet adjacent to this operon in several Gluconacetobacter xylinum strains. Other genes such as bcsC and bcsD, both homologs to Rhizobium celE gene, are required for in vivo but not for in vitro cellulose biosynthesis.
The lack of cellulose production in celA or celB mutants of R. leguminosarum bv. trifolii did not affect the ability of these bacteria to nodulate clover under controlled laboratory conditions. The postulated involvement of the Agrobacterium CelC cellulase enzyme in bacterial cellulose biosynthesis is to incorporate a lipid –linked oligosaccharide intermediate into cellulose, but this biochemical function has not been definitely established. R. leguminosarum bv. trifolii ANU843 CelC2 has high substrate specificity for amorphous forms of cellulose (e.g., carboxymethyl cellulose, CMC) and does not degrade other polysaccharides tested so far.
Studies using purified enzyme, CelC2 knockout and overproducing derivative strains of R. leguminosarum bv. trifolii ANU843 show that this cellulase isozyme is essential for primary infection of its symbiotic white clover host (Trifolium repens), being required for the localized tip erosion of the root hair wall allowing the bacteria to breach this host barrier and initiate the infection process[28, 33]. CelC2 has been also shown to be involved in bacterial release from infection threads into nodule cells. Evaluation of the currently defined genomes of rhizobia indicated genes coding for putative cellulases represented by a diversity of glycosyl hydrolase families. The celC gene is widely conserved among Rhizobium species, consistent with its involvement in fundamental processes for survival in the environment. In this work, we provide evidence for a key role of the CelC2 cellulase for optimal cellulose micro fibrils biosynthesis and biofilm formation.
Cellulose production is widespread among rhizobia
A database search revealed that all the currently defined Rhizobiaceae genomes possess genes coding for cellulose synthase belonging to the Glycosyl Transferase family 2 (GT2). Interestingly, these GT2 coding genes are located near endoglucanase celC homologs (belonging to Glycosyl Hydrolase family 8) forming the celABC operon or near cellulase genes from Glycosyl Hydrolase family 26, forming a potential operon that contains a cellulose synthase associated with a cellulase and another hypothetical protein of unknown function, that we have named celIJK. This putative operon has homologs in all currently available genomic sequences of Rhizobiaceae representatives (Additional file2). There are several organisms that share both cellulose production operons, and in Agrobacterium tumefaciens wild-type strain C58, celABC and celIJK are closely located in the genome.
We have sequenced the celABC genes from R. leguminosarum bv. trifolii ANU843 (GeneBank accession no. JN180924, celA; JN180925, celB; AJ561043, celC). Their nucleotide sequence and organization are similar and highly conserved among various Rhizobiaceae members. Orthologs of this operon are also found in other legume-nodulating bacteria (Additional file3).
CelC2 cellulase is involved in cellulose microfibril formation and elongation
Intensity of Congo red staining of ANU843ΔC2 and ANU843C2+ strains was greatly increased or reduced, respectively, as compared to the wild type and ANU843 carrying the empty vector (Figure1D). Also, the intensity of Congo red staining of the celC complemented strain was reduced as compared to ANU843ΔC2, showing a complementation of this phenotype (Figure1D).
The mean widths of line streak colonies for ANU843wt, ANU843ΔC2 and ANU843C2+ when grown on YMA plates (Figures2D-F) were 13.0 ± 1.3, 5.0 ±1.0, and 6.7 ± 1.3 millimeters, respectively, and ANOVA statistics indicated that these differences in width were statistically significant (N = 12, p ≤ 0.03). A similar trend is represented in their mucoidy (Figure2D-F, enlargement).
In addition, quantitative image analysis of fluorescence micrographs of cultures stained with the fluorochrome Calcofluor, which binds to β-linked glucans like cellulose, indicated that the extracellular micro fibrils associated with cells of the ANU843ΔC2 mutant were significantly longer than the average associated with wild-type (up to 15 μm vs. 8.34 μm standardized per cell, respectively), and that they were not detected in association with the ANU843C2+ derivative strain (Figure2G-I).
Thus, the increase in flocculation, enzymatic treatment, Congo red staining, and estimated microfibril length observed in the ANU843ΔC2 mutant are all directly correlated with, hence most likely due to the overproduction of external cellulose micro fibrils.
Cellulose micro fibrils involvement in biofilm formation and maturation
Among the molecules involved in biofilm development are different proteins and some exopolysaccharides, such as cellulose, which is a major component of the biofilm matrix of several bacterial species. Cellulose-production defective mutants of different bacterial species have been shown to be impaired in biofilm development[36–38]. By testing biofilm formation on different abiotic substrata and examining attachment in planta with the host plant clover, we qualitatively and quantitatively compared the biofilm formation ability of the wild type strain ANU843 and its derivatives ANU843ΔC2 and ANU843C2+, based on methods previously described. Different patterns of biofilm formation on both abiotic and biotic substrates were observed for the different strains.
Ring production of ANU843ΔC2 at the glass-liquid-air interface were thicker, more compact, and more easily dislodged from the glass surface compared to the other strains tested (Figure3A, red arrows) probably due to the overproduction of external cellulose micro fibrils.
Growth of ANU843ΔC2 bacteria was slightly delayed and more flocculated than wild-type in microtiter wells containing YMB. The use of either TY or minimal media containing mannitol resulted in more clear-cut results, with no detectable differences in growth among the strains (data not shown), but showing statistically significant reduced biofilm formation for the mutants compared to the wild-type (Figure3B-C). ANU843ΔC2 mutants exhibited a 30–50% decrease in biofilm formation in the microtiter plate assay (Figure3B-C) in minimal media and TY. ANU843C2+ exhibited ca. 30–60% reduced biofilm formation compared to the wild-type strain 24 h and 48 h after the start of the experiment. However, the differences were less pronounced at 72 h.
The most striking difference in biofilm formation was found between the ANU843ΔC2 mutant and wild-type strains because prior to washing, the biofilms of the mutant appeared to be robust and compact. However, ANU843ΔC2 flocs were easily removed with each successive washing step, whereas almost all of the wild-type cells remained firmly attached to the plastic wells (image not shown).
Mutants were also tested for biofilm-forming ability on PVC tabs. Wild type ANU843 developed microcolonies, which progressed into a characteristic three-dimensional biofilm (Figure4D). This biofilm morphology sharply contrasted with that produced by the ANU843ΔC2 mutant in which only thick, tightly appressed mounds of cells were apparent (Figure4E). On the other hand, the three-dimensional biofilm structure of GFP-tagged ANU843C2+ was not detected. This strain only produced a 2-dimensional monolayer biofilm, with just a few bacteria attached to one another (Figure4F). These results are consistent with results obtained in the sand assay and confirm the role of external cellulose micro fibrils in biofilm attachment and architecture. They further suggest that a R. leguminosarum bv. trifolii biofilm does not develop normally if cellulose micro fibril production is altered by elevated or diminished levels of CelC2 endoglucanase (or by the activity of the celC gene product).
Cellulose micro fibrils involvement in biofilm formation and maturation in planta
In ANU843, the majority of root hairs in the growing root hair zone were covered with R. leguminosarum bv. trifolii cells growing in a three-dimensional biofilm (Figure5A, E, H) whereas ANU843ΔC2 formed large aggregates that covered the root in an irregular manner (Figure5B, F, I). In contrast, very few individual cells of the ANU843C2+ derivative strain were attached to root hairs (Figure5C, G, J, K). Quantitative micro densitometry of GFP-dependent cell fluorescence (sum gray/mm root length) in Figures5A, B and C indicated intensity values of 246, 64, and 34 (ratios of 7.2 : 1.9 : 1.0) for wild type, ANU843ΔC2 and ANU843C2+ strains, respectively.
There was attachment of individual GFP-labeled R. leguminosarum bv. trifolii ANU843 wild-type and strain ANU843ΔC2 bacterial cells, followed by “cap” formation, in which additional bacterial cells attach to bacteria that directly bind to root hairs (Figure5A, E, H and B, F, I, respectively). However, ANU843ΔC2 cells more often clustered in discrete clumps on the root surface (Figure5B, F) including both non-root hair and root hair epidermal cells (Figure5I). In contrast, the biofilm caps on root hair tips were significantly reduced in frequency for the ANU843C2+ strain (Figure5C, G, J, K). These results confirm that cellulose is not required for the initial, phase-I hapten-inhibitable lectin-mediated attachment to root hairs, although its production does intensify the three-dimensional Phase 1A cap biofilm formation. The differences in pattern of root and root hair attachment by the cellulose mutants implies that the cellulose fibrils play a significant role in stability of the three-dimensional root hair “cap” biofilm.
Clover roots were also harvested, washed, and sonicated for colony counts. For strain ANU843, an average of 5.8 x 107 rhizobia cells per mg root fresh weight were attached to the root (Figure5D). By contrast, only 3 x 107 ANU843ΔC2 cells per mg root fresh weight attached to clover roots, showing that, as occurred in the abiotic surface experiments, the unstable aggregates on surfaces are easily removed (Figure5D). Regarding ANU843C2+, 4.6 x 107 cells per mg of root fresh weight were attached, showing that the stunted three-dimensional structures led to slightly loose bacterial attachment. However, no significant differences with respect to ANU843 were detected in terms of CFU/mg. It remains to be determined if this difference is due to loss of the recombinant plasmid in the absence of the selector marker in planta, or to the possible regulation of bacterial cellulose production by the plant, perhaps by inhibiting CelC2 activity or by repressing expression of the celC gene.
Cellulose production in bacteria is of potential economic interest because of its impact in medical settings and in the paper and food industry. In addition, cellulose biosynthesis in a wide variety of bacteria has increased the possibility for the elucidation of regulation and molecular mechanisms of cellulose biosynthesis, which still remains largely unknown. The role of cellulose in attachment and biofilm formation during the interaction of the bacteria with the environment and its hosts confers further importance to this polymer as a potential target to design methods against harmful biofilm proliferation. In this work, we report how the alterations in the levels of endogenous cellulase contribute to biofilm architecture.
Although Congo red can also bind outer membrane proteins in some animal pathogens, several authors have noted that among Rhizobium spp., staining with this dye correlated with the cellulose content in the bacterial cultures[16, 40]. The significant reduction in Congo red binding to engineered strains of Rhizobium impaired in cellulose production (Figure1D) suggests that these bacteria do not produce other substances that strongly bind Congo red and contribute to their red colony pigmentation. Both the binding to Congo red in diverse species of the family Rhizobiaceae and the presence of genes encoding cellulose synthases in their sequenced genomes strongly suggest that the ability to synthesize cellulose is fairly common in this taxonomic group of plant-associated bacteria. Several lines of evidence further support that cellulase activity is commonly involved in the cellulose production pathway used by these species: i) the existence of at least a cellulase-encoding gene associated to a glycosil transferase-encoding gene in all Rhizobiaceae species with accessible data in GeneBank, ii) the presence of the celABC operon within Rhizobiaceae, and iii) the high degree of conservation of CelC cellulase-encoding genes between Rhizobium species.
The ability of most legume root-nodulating microorganisms to synthesize cellulose implicates the importance of this polymer in their eco-physiology. Furthermore, although the inactivation of cellulose biosynthesis in R. leguminosarum bv. trifolii does not affect the ability to nodulate clover in controlled laboratory conditions, the ability of rhizobia to adhere firmly to a substrate using cellulose micro fibrils facilitates host root colonization[1, 8]. Moreover, it is very possible that cellulose micro fibril-mediated firm adhesion during legume host root colonization is important under natural conditions in the rhizosphere (or rhizospheric soil), where bacteria have to compete for survival and colonization to successfully gain access to plant carbon sources.
During the course of these studies on the CelC2 protein, we detected by Congo Red and Calcofluor staining microscopy and enzymatic treatment that the celC over-expression derivative strain lost the ability to make extracellular cellulose micro fibrils, and that the extracellular micro fibrils of the celC knockout mutant were significantly longer than those seen in the wild type parent (Figure2) suggesting that R. leguminosarum bv. trifolii cellulase CelC2 is involved in determining the longitude of cellulose extracellular micro fibrils.
Ausmees et al. found that cloned genes involved in cellulose biosynthesis have similar homology and the same organization in R. leguminosarum bv. trifolii strain R200 and A. tumefaciens. These authors identified these genes by Tn5 mutagenesis followed by screening with Calcofluor staining for mutants showing less ability to synthesize cellulose. They obtained celA, celB, and celE mutants but not celC, consistent with our finding that a celC knock-out mutant overproduces external cellulose micro fibrils. Therefore, we conclude that the celC gene is involved in the formation and elongation of cellulose micro fibrils. It is likely that cellulose oligomers synthesized by CelA elongate indefinitely in the absence of the CelC endoglucanase, producing very long micro fibrils that entangle ANU843ΔC2 mutant bacterial cells into very large aggregates causing them to flocculate and settle in liquid medium cultures (Figures2A and3A). By contrast, an excess of endoglucanase activity leads to an uncontrolled degradation of CelA-synthesized cellulose oligomers, preventing their transport and subsequent maturation into microcrystalline micro fibrils that extend outside the cell (Figure2I). The fluorescence intensity of these CelC2-degraded oligomer products suggest that they may concentrate at both cell poles (Figure2G-I inserts).
In the model proposed for the synthesis of cellulose in plants, CesA protein is a polymerase that catalyzes ß-1,4-glucan chain elongation by transferring UDP-glucose moieties to the sitosterol-ß-sitosterol-glucoside intermediate, forming cellodextrins. The other protein that has been proposed to participate in this process is Korrigan (Kor) cellulase, which may act by releasing the sitosterol-ß-glucoside from the newly synthesized cellulose polymer chain. How the final cellulose chain is formed, what other molecule(s) intervene in this step, the mechanism of its export and components of its anchoring apparatus on the cell surface all remain unknown. However, in the model for cellulose synthesis in bacteria, CelA activity adds glucose monomers from the UDP-glucose substrate to the intermediate lipid-glucose to form lipid-glucosex (x = 2–4 glucose). Bacterial CelA and plant CesA are both glycosyl transferases, whereas CelC and Korrigan present a glycosyl hydrolase domain. According to Matthysse et al., bacterial CelC cellulase may act in this case as a translocase, by incorporating the lipid-linked oligosaccharide into the cellulose polymer chain being formed. According to our findings, the role of CelC cellulase in cellulose biosynthesis is similar to the role that has been proposed for Korrigan cellulase. This CelC cellulase may catalyze the hydrolysis reaction, in which cellotriose are released from the lipid-glucose4 intermediary, thus providing the substrate for the translocase to transfer it to the internal growing point of another lipid-intermediate, thereby elongating the cellulose microfibril product by 3 glucose units at one time.
This function for the celC gene is distinctly different from its established role as an endoglucanase involved in the infection process. The fact that core celC has both colonization and infection functions, i.e., in cellulose biosynthesis and independently as a hydrolytic enzyme that creates the portal of Rhizobium entry into the host root hair and their liberation from infection threads into the symbiosomes within nodule cells, implies the likely existence of two different sets of control mechanisms, functional designs and target cellular locations. Furthermore, the possible role in cellulose biosynthesis of cellulase CelC1, that it is not involved in plant root hair tip erosion, has not been yet characterized. We propose that cellulose production may reflect an earlier evolutionary development because this property is encoded by an operon common to all nodulating rhizobia, including Burkholderia and Cupriavidus strains, the so-called β-rhizobia, whereas the cellulase celC gene function for the infection may develope later.
Previous results have shown that Tn5 mutation in nodC diminishes extracellular microfibril production by ANU843 on white clover roots and that Rhizobium common nod genes are required for biofilm formation. Nevertheless, the role of cellulose in Rhizobium biofilm establishment has not been fully recognized. To our knowledge, this is the first time that Rhizobium mutants either lacking or over-expressing cellulase have been analysed with respect to cellulose production and biofilm formation on both abiotic surfaces and the root epidermis.
Our results show that biofilm formation ability was markedly reduced in ANU843ΔC2 and ANU843C2+ strains not only on abiotic surfaces (PVC, microtiter plate assays, and sand Figures3 and4), but also on plant roots (Figure5). Since ANU843C2+ is impaired to produce external micro fibrils, it cannot tightly bind to the substrate preventing subsequent mature biofilm formation. On the other hand, it seems that ANU843ΔC2 cells, which produced more cellulose micro fibrils, tended to associate themselves more than to the plastic surfaces. However, no such caps are observed on sand grains. Since the PVC surface is hydrophobic while sand surface is hydrophilic, it might be possible that adhesion to each of these surfaces has different physicochemical constraints, and cellulose might play different roles in these scenarios.
Microtiter well, sand attachment and microscopy assays of GFP-tagged bacterial cultures confirmed the role of cellulose in affecting biofilm structure. We propose that cellulase CelC is associated with cellulose cleavage and processing, and that extracellular cellulose micro fibrils, but not in excess, are needed to build the three-dimensional structure of a mature biofilm of R. leguminosarum bv. trifolii on plastic, sand and white clover host root surfaces.
Taking all together, these results show that celC gene deletion lead to an increase of external cellulose micro fibrils production and that its overexpression results in the opposite effect, elucidating the role of this gene in the celABC cellulose biosynthesis operon. Furthermore, the investigation of the mutants ability to form biofilms in different surfaces indicates that celC gene suitable expression is necessary for biofilm formation and maturation in Rhizobium. All this data confirm that Rhizobium cellulose is an essential component of the biofilm polysaccharidic matrix architecture and that either an excess or a defect of this “building material” seem to collapse the biofilm structure. These results position cellulose hydrolitic enzymes as excellent anti-biofilm candidates.
Bacterial strains, plasmids and growth conditions
Representative type strains of bacteria that form legume root-nodule symbioses and Rhizobium leguminosarum bv. trifolii strains used in this study are listed in Additional file1 and Additional file4 respectively. The former share the ability to form 1–5 mm colonies after 5–7 days on Yeast Mannitol Agar. R. leguminosarum strains were also grown at 28°C in Triptone Yeast extract (TY), Yeast extract Mannitol Broth (YMB) or minimal medium containing 1% mannitol (MM) or 0.5% inositol (BINOS) as the carbon source. Escherichia coli S17.1 strain was grown at 37°C in LB medium. These media were supplemented with kanamycin (50 μg/ml) or tetracycline (10 μg/ml) as required. Plasmid pHC60 was introduced via biparental mating to yield GFP-tagged bacteria.
Cellulose detection assays
Cellulose production was assayed by growing the strains in YMA plates containing 25 mg/l Congo red for 7 days to stain for cellulose, followed by background-subtracted quantitation of colony red luminosity by image analysis. For direct microscopy visualization of cellulose micro fibrils, cells grown on YMA plates were suspended in 0.025% Calcofluor (Sigma), placed onto Teflon printed slides and examined by fluorescence microscopy. Flocculation assays were performed basically as described in. Overnight cultures were adjusted to an OD600 of 0.6 by dilution with YMB medium. Three-ml aliquots of cultures were incubated in 10 ml standing tubes at room temperature, and the OD600 of the upper parts of the cultures was measured 24 h after settling of flocs at 1 x g. These tests were performed in triplicate.
Amplification and sequencing of the Rhizobium leguminosarum bv. trifolii ANU843 celABC operon
The region upstream celC gene containing celA and celB genes in R. leguminosarum bv. trifolii was amplified and sequenced by using primers designed in this work (Additional file5) from conserved gene sequences available in databases. DNA was extracted according to. PCR DNA amplification and agarose DNA electrophoresis were performed by using standard procedures. The sequence reaction was performed on an ABI377 sequencer (Applied Biosystems Inc., USA) using a Big Dye terminator v3.0 cycle sequencing kit as supplied by the manufacturer. The celABC sequences obtained were compared with those present in other bacterial genomes available in GenBank using the BLASTN program, while BLASTP was employed to compare the encoding proteins.
Determination of biofilm formation
Ring formation at the glass-air-liquid interface were qualitatively scored after 3 to 5 days of bacterial growth in 5 ml of TY medium in glass tubes shaken at 180 rpm in an orbital shaker followed by static growth for 15 days. In vitro biofilms were established as described earlier. Basically, to prepare the initial inocula, GFP-tagged bacteria were grown in the corresponding liquid medium for 2 days (extrapolated OD600 of approximately 2.0). They were then washed and diluted in the same medium to OD600 = 0.2 (ca. 1x107 cells/ml) for all subsequent assays. Bacterial attachment to PVC plates was assayed by pipetting 100 μl of this culture into individual PVC wells in a 96-well plate (Falcon 3911, Becton Dickinson, Franklin Lakes, NY). The plates were sealed with parafilm and incubated at 28°C. At defined times, the growth media and unbound cells were removed, and attached bacteria were quantified by staining them with 0.3% (wt/vol) Crystal Violet (CV) (Sigma) for 10 min. After the excess dye was washed away and the wells allowed to dry, the remaining dye was solubilized with 80% ethanol-20% acetone and quantified by measuring the absorbance at 570 nm in an ASYS (Biochrom, UK) Microtiter Plate reader (Model No. UVM340).
Biofilm formation on abiotic surfaces was assayed by transferring 0.5 ml of each strain’s TY culture to individual wells of a 24-well Costar PS (polystyrene) plate containing either a PVC tab or 1ml of sterile sand substrata. After 1 week at 28°C, either the tabs or an alicuot of the sand particles previously washed with sterile water were examined by epifluorescence microscopy to evaluate the two- and three-dimensional architectures of the biofilms. In parallel, equal amounts of sand particles (aprox. 100 μl) were taken with a pipette harbouring a cut edge, weighted, and washed three times with sterile water. The cells were finally suspended in 1 ml of TE buffer (pH 7.0), dissociated at 25°C by 2 sonication pulses (37 kHz, 30 W) of 1 min each with pause time of 1 min between the pulses, in an Elmasonic (Singen, Germany) sonication bath and quantified by counting CFUs (normalized to the amount of sand weighted).
Root biofilms were prepared by dipping one-week-old seedling roots into a Rhizobium suspension of OD600 = 0.02 (ca. 1 x 106 cells/ml) in Fähraeus N-free medium for 5 min. The seedlings were then grown on a Whatman paper support placed inside glass tubes containing 20 ml Fähraeus N-free medium. The colonization of the bacteria to the root and the root hairs was evaluated using confocal microscopy at different time points. To count the number of colonized rhizobia three days after inoculation, the roots were washed three times with sterile water under vigorous shaking to remove loosely associated cells, then dried at room temperature for one minute on sterile Whatman paper to eliminate the excess of water, and weighed. The dried roots were immersed in 1 ml of sterile TE buffer and sonicated as it was described above to release attached cells, which were quantified by colony counting (normalized to the roots weight). For each strain tested, we counted rhizobial attachment to at least four roots. All binding experiments were repeated at least three times.
Calcofluor-stained bacteria were examined using epifluorescence optics (100 w HBO light source, 365/395/420 nm filter set) in a Zeiss universal microscope. Scanning electron microscopy (SEM) was performed as previously described, except that the samples were coated with osmium instead of gold.
GFP-tagged rhizobia growing on sand particles or plastic tabs were aseptically removed from the wells, and placed into the wells of depression slides, topped with a covers lip, and examined by confocal microscopy.
Bacterial attachment to the root hairs was observed by laser scanning confocal microscopy with a Leica microscope using 488-nm argon laser excitation and a 500-nm long-pass emission filter to allow observation of GFP-labeled bacteria, and phase contrast microscopy to observe root hairs. Images were processed into loss-less montages using Leica confocal software. Biofilms of GFP-labelled bacteria on roots were examined on a Zeiss LSM510 confocal microscope. Roots were stained with 10 μM of propidium iodide (Sigma). Projections were made from adjusted individual channels in the image stacks using Zeiss LSM510 imaging software.
Digital image analysis
Center for Microbial Ecology Image Analysis (CMEIAS) software was used to measure the intensity of Congo red uptake and mean diameter of bacterial colonies, the calcofluor-fluorescent extracellular micro fibril length normalized to cell number, the % substrata coverage by the biofilms, and the in situ local density of GFP-tagged bacteria colonized on roots[54–56].
We thank Nancy Fujishige and Peter De Hoff for help in the biofilm assays. This work was supported by Junta de Castilla y León Grant GR49 and MICINN Grants AGL2008-03360 and AGL2011-29227. M.R. and L.R. were supported by a PhD fellowship from the Spanish government.
- Dazzo FB, Truchet GL, Sherwood JE, Hrabak EM, Abe M, Pankratz SH: Specific phases of root hair attachment in the Rhizobium trifolii-clover symbiosis. Appl Environ Microbiol. 1984, 48: 1140-1150.Google Scholar
- Ausmees N, Jacobsson K, Lindberg M: A unipolarly located, cell-surface-associated agglutinin, RapA, belongs to a family of Rhizobium-adhering proteins (Rap) in Rhizobium leguminosarum bv. trifolii. Microbiology. 2001, 147: 549-559.View ArticleGoogle Scholar
- Fujishige NA, Kapadia NN, De Hoff PL, Hirsch AM: Investigations of Rhizobium biofilm formation. FEMS Microbiol Ecol. 2006, 56: 195-206. 10.1111/j.1574-6941.2005.00044.xView ArticleGoogle Scholar
- Sherwood JE, Vasse JM, Dazzo FB, Truchet GL: Development and trifoliin A-binding ability of the capsule of Rhizobium trifolii. J Bacteriol. 1984, 159: 145-152.Google Scholar
- Davey ME, O'Toole GA: Microbial biofilms: from ecology to molecular genetics. Microbiol Mol Biol Rev. 2000, 64: 847-867. 10.1128/MMBR.64.4.847-867.2000View ArticleGoogle Scholar
- Dazzo FB, Brill WJ: Bacterial polysaccharide which binds Rhizobium trifolii to clover root hairs. J Bacteriol. 1979, 137: 1362-1373.Google Scholar
- Smit G, Swart S, Lugtenberg BJ, Kijne JW: Molecular mechanisms of attachment of Rhizobium bacteria to plant roots. Mol Microbiol. 1992, 6: 2897-2903. 10.1111/j.1365-2958.1992.tb01748.xView ArticleGoogle Scholar
- Mateos PF, Baker B, Philip-Hollingsworth S, Squartini A, Peruffo A, Nuti M, Dazzo F: Direct in situ identification of cellulose micro fibrils associated with Rhizobium leguminosarum biovar. trifolii attached to the root epidermis of white clover. Can J Microbiol. 1995, 41: 202-207. 10.1139/m95-028.View ArticleGoogle Scholar
- Dazzo FB, Napoli CA, Hubbell DH: Adsorption of bacteria to roots as related to host specificity in the Rhizobium-clover symbiosis. Appl Environ Microbiol. 1976, 32: 166-171.Google Scholar
- Dazzo FB, Brill WJ: Receptor site on clover and alfalfa roots for Rhizobium. Appl Environ Microbiol. 1977, 33: 132-136.Google Scholar
- Dazzo FB, Brill WJ: Regulation by fixed nitrogen of host-symbiont recognition in the Rhizobium-clover symbiosis. Plant Physiol. 1978, 62: 18-21. 10.1104/pp.62.1.18View ArticleGoogle Scholar
- De Hoff PL, Brill LM, Hirsch AM: Plant lectins: the ties that bind in root symbiosis and plant defense. Mol Genet Genomics. 2009, 282: 1-15. 10.1007/s00438-009-0460-8View ArticleGoogle Scholar
- Webb JS, Givskov M, Kjelleberg S: Bacterial biofilms: prokaryotic adventures in multicellularity. Curr Opin Microbiol. 2003, 6: 578-585. 10.1016/j.mib.2003.10.014View ArticleGoogle Scholar
- Costerton JW, Cheng KJ, Geesey GG, Ladd TI, Nickel JC, Dasgupta M, Marrie TJ: Bacterial biofilms in nature and disease. Annu Rev Microbiol. 1987, 41: 435-464. 10.1146/annurev.mi.41.100187.002251View ArticleGoogle Scholar
- Napoli C, Dazzo F, Hubbell D: Production of cellulose micro fibrils by Rhizobium. Appl Microbiol. 1975, 30: 123-131.Google Scholar
- Ausmees N, Jonsson H, Hoglund S, Ljunggren H, Lindberg M: Structural and putative regulatory genes involved in cellulose synthesis in Rhizobium leguminosarum bv. trifolii. Microbiology. 1999, 145 (Pt 5): 1253-1262.View ArticleGoogle Scholar
- Laus MC, van Brussel AA, Kijne JW: Role of cellulose fibrils and exopolysaccharides of Rhizobium leguminosarum in attachment to and infection of Vicia sativa root hairs. Mol Plant Microbe Interact. 2005, 18: 533-538. 10.1094/MPMI-18-0533View ArticleGoogle Scholar
- Williams A, Wilkinson A, Krehenbrink M, Russo DM, Zorreguieta A, Downie JA: Glucomannan-mediated attachment of Rhizobium leguminosarum to pea root hairs is required for competitive nodule infection. J Bacteriol. 2008, 190: 4706-4715. 10.1128/JB.01694-07View ArticleGoogle Scholar
- Matthysse AG, White S, Lightfoot R: Genes required for cellulose synthesis in Agrobacterium tumefaciens. J Bacteriol. 1995, 177: 1069-1075.Google Scholar
- Matthysse AG, Deschet K, Williams M, Marry M, White AR, Smith WC: A functional cellulose synthase from ascidian epidermis. Proc Natl Acad Sci USA. 2004. 10.: 986-991.View ArticleGoogle Scholar
- Ross P, Weinhouse H, Aloni Y, Michaeli D, Weinberger-Ohana P, Mayer R, Braun S, de Vroom E, van der Marel GA, van Boom JH, Benziman M: Regulation of cellulose synthesis in Acetobacter xylinum by cyclic diguanylic acid. Nature. 1987, 325: 279-281. 10.1038/325279a0View ArticleGoogle Scholar
- Deinema MH, Zevenhuizen LP: Formation of cellulose fibrils by Gram-negative bacteria and their role in bacterial flocculation. Arch Mikrobiol. 1971, 78: 42-51. 10.1007/BF00409087View ArticleGoogle Scholar
- Ross P, Mayer R, Benziman M: Cellulose biosynthesis and function in bacteria. Microbiol Rev. 1991, 55: 35-58.Google Scholar
- Nobles DR, Romanovicz DK, Brown RM: Cellulose in cyanobacteria. Origin of vascular plant cellulose synthase?. Plant Physiol. 2001, 127: 529-542. 10.1104/pp.010557View ArticleGoogle Scholar
- Lasa I, Penades JR: Bap: a family of surface proteins involved in biofilm formation. Res Microbiol. 2006, 157: 99-107. 10.1016/j.resmic.2005.11.003View ArticleGoogle Scholar
- Doblin MS, Kurek I, Jacob-Wilk D, Delmer DP: Cellulose biosynthesis in plants: from genes to rosettes. Plant Cell Physiol. 2002, 43: 1407-1420. 10.1093/pcp/pcf164View ArticleGoogle Scholar
- Wong HC, Fear AL, Calhoon RD, Eichinger GH, Mayer R, Amikam D, Benziman M, Gelfand DH, Meade JH, Emerick AW, et al: Genetic organization of the cellulose synthase operon in Acetobacter xylinum. Proc Natl Acad Sci USA. 1990, 87: 8130-8134. 10.1073/pnas.87.20.8130View ArticleGoogle Scholar
- Robledo M, Jiménez-Zurdo JI, Velázquez E, Trujillo ME, Zurdo-Piñeiro JL, Ramírez-Bahena MH, Ramos B, Díaz-Minguez JM, Dazzo F, Martínez-Molina E, Mateos PF: Rhizobium cellulase CelC2 is essential for primary symbiotic infection of legume host roots. Proc Natl Acad Sci USA. 2008. 10.: 7064-7069.View ArticleGoogle Scholar
- Lin FC, Brown RM, Drake RR, Haley BE: Identification of the uridine 5'-diphosphoglucose (UDP-Glc) binding subunit of cellulose synthase in Acetobacter xylinum using the photoaffinity probe 5-azido-UDP-Glc. J Biol Chem. 1990, 265: 4782-4784.Google Scholar
- Romling U: Molecular biology of cellulose production in bacteria. Res Microbiol. 2002, 153: 205-212. 10.1016/S0923-2508(02)01316-5View ArticleGoogle Scholar
- Saxena IM, Kudlicka K, Okuda K, Brown RM: Characterization of genes in the cellulose-synthesizing operon (acs operon) of Acetobacter xylinum: implications for cellulose crystallization. J Bacteriol. 1994, 176: 5735-5752.Google Scholar
- Matthysse AG, Thomas DL, White AR: Mechanism of cellulose synthesis in Agrobacterium tumefaciens. J Bacteriol. 1995, 177: 1076-1081.Google Scholar
- Mateos PF, Baker D, Petersen M, Velázquez E, Jiménez-Zurdo JI, Martínez-Molina E, Squartini A, Orgambide G, Hubbell D, Dazzo FB: Erosion of root epidermal cell walls by Rhizobium polysaccharide-degrading enzymes as related to primary host infection in the Rhizobium-legume symbiosis. Can J Microbiol. 2001, 47: 475-487.Google Scholar
- Robledo M, Jiménez-Zurdo JI, Soto MJ, Velázquez E, Dazzo F, Martínez-Molina E, Mateos PF: Development of functional symbiotic white clover root hairs and nodules requires tightly regulated production of rhizobial cellulase celC2. Mol Plant Microbe Interact. 2011, 24: 798-807. 10.1094/MPMI-10-10-0249View ArticleGoogle Scholar
- Robledo M, Velázquez E, Ramírez-Bahena MH, García-Fraile P, Pérez-Alonso A, Rivas R, Martínez-Molina E, Mateos PF: The celC gene, a new phylogenetic marker useful for taxonomic studies in Rhizobium. Syst Appl Microbiol. 2011, 34: 393-399. 10.1016/j.syapm.2011.01.010View ArticleGoogle Scholar
- Zogaj X, Nimtz M, Rohde M, Bokranz W, Romling U: The multicellular morphotypes of Salmonella typhimurium and Escherichia coli produce cellulose as the second component of the extracellular matrix. Mol Microbiol. 2001, 39: 1452-1463. 10.1046/j.1365-2958.2001.02337.xView ArticleGoogle Scholar
- Saldaña Z, Xicohtencatl-Cortes J, Avelino F, Phillips AD, Kaper JB, Puente JL, Giron JA: Synergistic role of curli and cellulose in cell adherence and biofilm formation of attaching and effacing Escherichia coli and identification of Fis as a negative regulator of curli. Environ Microbiol. 2009, 11: 992-1006. 10.1111/j.1462-2920.2008.01824.xView ArticleGoogle Scholar
- Solano C, Garcia B, Valle J, Berasain C, Ghigo JM, Gamazo C, Lasa I: Genetic analysis of Salmonella enteritidis biofilm formation: critical role of cellulose. Mol Microbiol. 2002, 43: 793-808. 10.1046/j.1365-2958.2002.02802.xView ArticleGoogle Scholar
- Theisen M, Rioux CR, Potter AA: Molecular cloning, nucleotide sequence, and characterization of lppB, encoding an antigenic 40-kilodalton lipoprotein of Haemophilus somnus. Infect Immun. 1993, 61: 1793-1798.Google Scholar
- Zevenhuizen LP, Bertocchi C, van Neerven AR: Congo red absorption and cellulose synthesis by Rhizobiaceae. Antonie Van Leeuwenhoek. 1986, 52: 381-386. 10.1007/BF00393465View ArticleGoogle Scholar
- Peng L, Kawagoe Y, Hogan P, Delmer D: Sitosterol-beta-glucoside as primer for cellulose synthesis in plants. Science. 2002, 295: 147-150. 10.1126/science.1064281View ArticleGoogle Scholar
- Moulin L, Munive A, Dreyfus B, Boivin-Masson C: Nodulation of legumes by members of the beta-subclass of Proteobacteria. Nature. 2001, 411: 948-950. 10.1038/35082070View ArticleGoogle Scholar
- Dazzo FB, Hollingsworth R, Philipp-Hollingsworth S, Robeles M, Olen T, Salzwedel J, Djordjevic M, Rolfe B: Recognition processes in the Rhizobium trifolii-white clover symbiosis. In Nitrogen fixation: Hundred years after. Edited by: Bothe FdB H, Newton W. 1988, 431-436. Justav Fischer Publisher, Stuttgard,Google Scholar
- Fujishige NA, Lum MR, De Hoff PL, Whitelegge JP, Faull KF, Hirsch AM: Rhizobium common nod genes are required for biofilm formation. Mol Microbiol. 2008, 67: 504-515. 10.1111/j.1365-2958.2007.06064.xView ArticleGoogle Scholar
- Jordan DC: Family III. Rhizobiaceae. 1984, Willians and Wilkins, Baltimore,Google Scholar
- Beringer JE: R factor transfer in Rhizobium leguminosarum. J Gen Microbiol. 1974, 84: 188-198. 10.1099/00221287-84-1-188Google Scholar
- Somasegaran P, Hoben HJ: Handbook for rhizobia: methods in legume-Rhizobium technology. 1994, Springer, New York,View ArticleGoogle Scholar
- O'Gara F, Shanmugam KT: Regulation of nitrogen fixation by Rhizobia. Export of fixed N2 as NH + 4. Biochim Biophys Acta. 1976, 437: 313-321. 10.1016/0304-4165(76)90001-5View ArticleGoogle Scholar
- Mateos PF, Jiménez-Zurdo JI, Chen J, Squartini AS, Haack SK, Martínez-Molina E, Hubbell DH, Dazzo FB: Cell-associated pectinolytic and cellulolytic enzymes in Rhizobium leguminosarum biovar trifolii. Appl Environ Microbiol. 1992, 58: 1816-1822.Google Scholar
- Sambrook J, Fritsch EF, Maniatis T: Molecular cloning: a laboratory manual. 1989, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y, 2,Google Scholar
- Cheng HP, Walker GC: Succinoglycan is required for initiation and elongation of infection threads during nodulation of alfalfa by Rhizobium meliloti. J Bacteriol. 1998, 180: 5183-5191.Google Scholar
- Da Re S, Le Quere B, Ghigo JM, Beloin C: Tight modulation of Escherichia coli bacterial biofilm formation through controlled expression of adhesion factors. Appl Environ Microbiol. 2007, 73: 3391-3403. 10.1128/AEM.02625-06View ArticleGoogle Scholar
- Rivas R, Velazquez E, Valverde A, Mateos PF, Martínez-Molina E: A two primers random amplified polymorphic DNA procedure to obtain polymerase chain reaction fingerprints of bacterial species. Electrophoresis. 2001, 22. 10.6-1089.View ArticleGoogle Scholar
- Liu J, Dazzo FB, Glagoleva O, Yu B, Jain AK: CMEIAS: a computer-aided system for the image analysis of bacterial morphotypes in microbial communities. Microb Ecol. 2001, 41: 173-194.View ArticleGoogle Scholar
- Dazzo F: Applications of quantitative microscopy in studies of plant surface microbiology. In Plant surface microbiology. Edited by: Varma A, Abbott D, Werner D, Hampp R. 2004, 503-550. Springer- Verlag, Germany,Google Scholar
- Gross CA, Reddy CK, Dazzo FB: CMEIAS color segmentation: an improved computing technology to process color images for quantitative microbial ecology studies at single-cell resolution. Microb Ecol. 2010, 59: 400-414. 10.1007/s00248-009-9616-7View ArticleGoogle Scholar
- Rolfe BG, Gresshoff PM, Shine J: Rapid screening method for symbiotic mutants of Rhizobium leguminosarum biovar trifolii on white clover plants. Plant Sci Lett. 1980, 19: 277-284. 10.1016/0304-4211(80)90082-6.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.