• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of aemPermissionsJournals.ASM.orgJournalAEM ArticleJournal InfoAuthorsReviewers
Appl Environ Microbiol. Mar 2012; 78(6): 1644–1651.
PMCID: PMC3298171

Role for Rhizobium rhizogenes K84 Cell Envelope Polysaccharides in Surface Interactions


Rhizobium rhizogenes strain K84 is a commercial biocontrol agent used worldwide to control crown gall disease. The organism binds tightly to polypropylene substrate and efficiently colonizes root surfaces as complex, multilayered biofilms. A genetic screen identified two mutants in which these surface interactions were affected. One of these mutants failed to attach and form biofilms on the abiotic surface although, interestingly, it exhibited normal biofilm formation on the biological root tip surface. This mutant is disrupted in a wcbD ortholog gene, which is part of a large locus predicted to encode functions for the biosynthesis and export of a group II capsular polysaccharide (CPS). Expression of a functional copy of wcbD in the mutant background restored the ability of the bacteria to attach and form normal biofilms on the abiotic surface. The second identified mutant attached and formed visibly denser biofilms on both abiotic and root tip surfaces. This mutant is disrupted in the rkpK gene, which is predicted to encode a UDP-glucose 6-dehydrogenase required for O-antigen lipopolysaccharide (LPS) and K-antigen capsular polysaccharide (KPS) biosynthesis in rhizobia. The rkpK mutant from strain K84 was deficient in O-antigen synthesis and exclusively produced rough LPS. We also show that strain K84 does not synthesize the KPS typical of some other rhizobia strains. In addition, we identified a putative type II CPS, distinct from KPS, that mediates cell-surface interactions, and we show that O antigen of strain K84 is necessary for normal cell-cell interactions in the biofilms.


Agrobacterium species are soilborne bacteria that cause crown gall disease in a large variety of plants (17). An early step in the infection process is the attachment of agrobacteria to plant tissues (30) which is followed by global root colonization (31, 44). During the bacterium-plant cell interactions in wounded tissues, tumorigenic strains have the ability to transfer a particular DNA segment from the Ti plasmid of the bacterium to the plant genome, and its expression in the transformed plant cell leads to the development of a crown gall tumor (reviewed in references 61 and 62). It is well known that crown gall disease can be controlled by treatment of the root plant system with the biocontrol agent Rhizobium rhizogenes (formerly Agrobacterium radiobacter [56]) strain K84 (reviewed in reference 45). Strain K84 also attaches to root tips ex planta and efficiently colonizes the plant root system (29, 43, 53, 57).

Bacteria can live and proliferate in natural environments either as individual cells (planktonic) or as highly organized multicellular communities encased in self-produced polymeric matrices called biofilms, which are in close association with surfaces or in air-liquid interfaces (19). Aggregates, microcolonies, and highly structured biofilms are common multicellular developmental stages that are observed in both pathogenic and nonpathogenic root-colonizing bacteria (1, 12, 13, 35, 42, 47). Agrobacterium tumefaciens strain C58 forms biofilms on roots that are enhanced by phosphorus, oxygen limitation, and cellulose production (14, 32, 33, 47). Two transcriptional regulators, BigR and ExoR, are necessary for biofilm growth in A. tumefaciens strain C58 (3, 55).

We recently reported that the crown gall biocontrol agent R. rhizogenes K84 binds tightly to polypropylene and efficiently colonizes root surfaces as complex, multilayered biofilm structures (1). However, the genetic basis for these properties had remained undefined. A putative capsular polysaccharide (CPS) produced by R. rhizogenes strain K84, which is distinct from the K-antigen capsular polysaccharide (KPS) produced by some other rhizobium strains, appears to play an important role in cell-surface interactions. Moreover, we have also shown that wild-type lipopolysaccharide (LPS) is necessary for normal cell-cell interactions in the K84 biofilms.


Bacterial strains, plasmids, and growth conditions.

Bacterial strains and plasmids used in this study are listed in Table 1. LB medium (50) was used for growing Escherichia coli. Rhizobium rhizogenes strain K84 was routinely grown on rich MG/L medium (9) supplemented with biotin (2 μg/ml) and occasionally on tryptone-yeast extract (TY) medium (4). Both AT (54) or AB (11) minimal media, with 0.2% mannitol as a carbon source (called ATM or ABM, respectively) and supplemented with biotin (2 μg/ml), were also used. Antibiotics were added to media at the following concentrations: for Rhizobium, tetracycline (Tc), gentamicin (Gm), kanamycin (Km), and neomycin (Nm) at 10, 15, 100, and 100 (μg/ml), respectively; for E. coli, Tc, Gm, Km, and ampicillin (Ap) at 20, 10, 50, and 50 (μg/ml), respectively.

Table 1
Bacterial strains and plasmids used

DNA manipulations.

Plasmid DNA was isolated from E. coli by an alkaline lysis-based procedure kit following the manufacturer's instructions (Qiagen). Standard recombinant DNA techniques were used (50). Plasmids were introduced into Rhizobium strains by electroporation (34) or by biparental cross-streak mating using E. coli S17-1(λpir) harboring the plasmid of interest as the conjugative donor (51). Genomic DNA for Rhizobium strains was isolated using a kit and following the manufacturer's instructions (Epicentre).

Transposon mutagenesis.

A transposon mutagenesis library was generated in wild-type R. rhizogenes strain K84 with Mar2×T7, an engineered derivative of the Himar1 transposon of the mariner family (27). Transposon insertions were randomly generated by introducing the suicide plasmid pMAR2xT7 into strain K84 by conjugation with E. coli S17-1(λpir). Transposon-induced mutants of strain K84 were selected on ABM medium supplemented with Gm.

Mar2×T7 insertion site identification.

Transposon insertion sites were identified using a two-round PCR protocol (7) with genomic DNA from mutants, amplifying the sequence adjacent to the transposon insertion with a transposon-specific primer and an arbitrary primer, followed by a second amplification using a nested transposon-specific primer and a primer corresponding to a nonrandom portion of the arbitrary primer used in the first PCR. A third nested transposon-specific primer was used for sequencing reactions. Sequences for the primers used are shown in Table S1 in the supplemental material.

Gene cloning.

A fragment of 1,369 bp containing the wcbD ortholog gene (Arad9146) was obtained by PCR amplification of genomic DNA from strain K84 using a specifically designed primer set (see Table S1) and cloned between the KpnI and BamHI sites of the broad-host-range vector pBBR1-MCS-2 (24). This fragment contained the Arad9146 open reading frame ([ORF] 1,212 bp) and an extended promoter region of the gene. A fragment of 1,538 bp containing the rkpK ortholog gene (Arad3519) was obtained by PCR amplification (see Table S1 in the supplemental material) and cloned between the HindIII and EcoRI sites of pBBR1-MCS-2. This fragment contained the Arad3519 ORF (1,323 bp) and promoter region of the gene.

Biofilm screen.

Transposon-induced mutants of strain K84 were screened in two steps. In the first, large-scale screen, 10 μl of overnight cultures in MG/L liquid medium was added to 200 μl of ATM liquid medium in 96-well microtiter polypropylene plates, using one well per mutant. The amounts of planktonic (optical density at 570 nm [OD570]) and surface-attached (A570) cells were determined by a standard crystal violet (CV) assay after static incubation for 48 h at 26°C (1, 39). Each potential mutant that was compromised in terms of biofilm formation in this first screen was then rescreened by individually adjusting the ATM culture to an OD570 of 0.04, and growth was continued in microtiter plates. Comparative biofilm formation was measured as a function of crystal violet staining as previously described. Experiments were repeated at least twice using 10 replicate wells per strain. Alternatively, biofilm formation on borosilicate culture tubes was also visualized after bacterial cells were grown on ABM liquid medium for 48 h under shaking conditions (50 rpm) with the tube in a slanted position.

Biofilm formation on the root tip surface.

Green fluorescent protein (GFP)-labeled derivative strains of strain K84 and its mutants were evaluated for biofilm formation on the surface of tomato root tips following a previously described methodology (1). Tightly bound bacteria were visualized on the surface of washed root tips by luminescent GFP detection using an LAS-3000 imaging system (Fujifilm). Biofilms on the surface of washed root tips were directly examined with a fluorescence microscope (Nikon Eclipse E800) equipped with a high-resolution digital camera (Nikon DXM1200). All experiments were performed at least twice, with two replicate root tips per strain.

O-antigen and K-antigen capsular polysaccharide analyses.

For purification of O-antigen lipopolysaccharide (LPS), bacteria were grown on solid TY medium for 96 h. LPS extraction, separation by SDS-polyacrylamide gel electrophoresis (PAGE), and silver staining were performed as previously described (6). For purification and visualization of the high-molecular-weight KPS, bacterial cultures were grown on liquid TY medium for 48 h. Cells were washed three times in 0.9% NaCl and pelleted by centrifugation. The bacterial pellet was resuspended and lysed by heating to 100°C in 500 μl of KPS extraction buffer (50 mM Tris-HCl [pH 8.5], 13 mM EDTA, 15 mM H3BO3) for 6 min. The bacterial crude extract was then treated with RNase (60 μg/ml) and DNase (20 μg/ml) for 3 h at 37°C, followed by two 24-h treatments with proteinase K (30 μg/ml) at 37°C. Finally, the treated mixture was centrifuged at 10,000 × g for 5 min, and 400 μl of the supernatant was mixed with 680 μl of 1 M sucrose and 120 μl of absolute ethanol and stored at −20°C. The resulting samples obtained were analyzed by PAGE without SDS, as previously described (40, 49), with the addition of ethanol to the running buffer as well as to the running and stacking gels at a final concentration of 10% (vol/vol). The acrylamide concentration of the running gel was 18% (wt/vol), and the ratio of acrylamide to N,N′-methylenebisacrylamide was 29:1. Gels were fixed using Alcian blue (0.5% in 2% acetic acid) and silver stained.

Extracellular EPS production.

Total carbohydrate amounts of the exopolysaccharide (EPS)-containing supernatants were determined using the anthrone-H2SO4 method, which measures the total reducing sugar content in a given sample (55). Briefly, cultures were grown in 1/20 diluted MG/L medium at 26°C for 72 h, and 1-ml culture samples were prepared by centrifugation to remove cells. This cell-free culture fluid was assayed for EPS content via sulfuric acid hydrolysis in the presence of the colorimetric indicator anthrone. To this end, 250 μl of anthrone solution (2%, wt/vol, in ethyl acetate) was added to 1 ml of culture sample and then hydrolyzed with 2.5 ml of concentrated sulfuric acid. The absorbance at 620 nm was measured, and amounts of carbohydrates were calculated using a standard curve of known concentrations of glucose (0 to 100 μg/ml). Exopolysaccharide values were normalized to the OD600 of each culture.

Transmission electron microscopy.

Overnight cultures in ATM liquid medium were resuspended in 0.01 M phosphate-buffered saline (pH 7.2) at 109 CFU/ml. Suspensions were deposited on Formvar-coated 400-mesh copper grids and allowed to settle for a few seconds. Excess liquid was blotted, and then cells were stained with 2% phosphotungstic acid (wt/vol). Grids were washed in distilled water and blotted. Cells were viewed with a JEOL transmission electron microscope model JEM-1010.


Identification of a K84 mutant with defective attachment to polypropylene wells but proficient for biofilm formation on the root tip surface.

A mutant collection of R. rhizogenes K84 carrying the Mar2×T7 transposon was screened for attachment and biofilm formation in 96-well polypropylene microtiter plates. The initial screen of approximately 1,700 transposon-derivative K84 mutants yielded two strains with different surface adhesion phenotypes. One of these mutants, designated M71, grew normally in ATM liquid medium (data not shown), but it was completely defective in attachment and subsequent biofilm formation on polypropylene wells (Fig. 1). In a time course experiment, M71 still did not attach or form biofilms after 4 days of incubation (data not shown). Interestingly, this mutant attached and produced normal biofilms on the root surface of tomato (Fig. 2). GFP-labeled fluorescent wild-type and mutant strains exhibited the same level of bacterial attachment to washed root tips (Fig. 2A) and also showed indistinguishable patterns of biofilm formation as observed by epifluorescence microscopy (Fig. 2B).

Fig 1
The wcbD orthologous gene of R. rhizogenes K84 is required for attachment and subsequent biofilm formation on polypropylene surfaces. CV-stained biofilms on polypropylene wells produced by wild-type strain K84, the wcbD mutant M71, and the complemented ...
Fig 2
The wcbD mutant M71 attaches and forms biofilms on the surface of tomato root tips. (A) Attached bacteria expressing GFP were detected by a luminescent detection imaging system on the surface of 1-cm tomato root tips incubated with GFP-labeled derivatives ...

Another K84 mutant attaches and forms visibly denser biofilms on both abiotic and root tip surfaces.

The mutant designated M108 showed visibly enhanced biofilms when grown in polypropylene wells (548% ± 132% with respect to the normalized data of the ratio between dimethyl sulfoxide [DMSO]-solubilized CV staining [A570] that reflects the surface-adherent cells and the culture density [OD570] that represents the planktonic cells of the wild type) and also on the surface of tomato root tips (Fig. 3). The adhesion pattern carried over to the natural root surface by a GFP fluorescently tagged M108 mutant strain shows that this mutant formed visibly denser biofilms (Fig. 3A). In fact, the mutant biofilms were uniformly thicker than those formed by the wild-type K84 strain on the root surface, as also observed by epifluorescence microscopy (Fig. 3B), with clearly a higher number of GFP-expressing bacterial cells.

Fig 3
The rkpK mutant M108 attaches and forms denser biofilms than the wild-type strain K84 on the surface of tomato root tips. (A) Attached bacteria expressing GFP were detected by a luminescent detection imaging system on the surface of 1-cm tomato root tips ...

The M108 mutant strain also exhibited more pronounced biofilms than the wild-type strain K84 when grown in borosilicate culture tubes under shaking conditions. These biofilms are very dense, but loose, at the liquid-air interface of cultures, while strain K84 grew in suspension mainly as planktonic cells, with minimal signs of biofilm formation (Fig. 4). The mutant also formed remarkable precipitated aggregates at the bottom of the tube that were absent in the wild-type strain (Fig. 4). Specifically, mutant M108 bacteria preferentially grew as surface-adherent biofilms (sessile) that were distinctly different from the wild-type K84 strain that grows, for the most part, as a planktonic suspension culture.

Fig 4
The rkpK mutant M108 grows mainly in a sessile stage (aggregates or biofilms) rather than in a planktonic way. Strain K84 grew in suspension (planktonic cells), forming just a slight biofilm at the surface of the tube at the liquid-air interface, whereas ...

Genetic analyses of the biofilm mutants.

Sequence analysis of the transposon-K84 DNA flanking region from genomic DNA of the mutant M71 localized the insertion site within the coding sequence of the Arad9146 gene. This gene is 1,212 bp long and is located on chromosome 2, which is the 2.65-Mbp megaplasmid replicon of the K84 genome (52). The transposon insertion is located between nucleotides 141 and 142. The projected translated product of the Arad9146 gene shows highest sequence similarity with the WcbD ortholog genes from Erythrobacter sp. (33% identity [I] and 53% similarity [S]) and from Burkholderia cenocepacia (33% I and 52% S) and with the BexC ortholog from Haemophilus influenzae (35% I and 53% S). Based on these amino acid sequence similarities, the Arad9146 coding sequence is predicted to encode a capsular polysaccharide inner membrane export protein (25, 28, 58). In fact, the wcbD ortholog gene in R. rhizogenes strain K84 is part of a locus (extending from GeneID Arad9136 to Arad9152) that is predicted to encode the biosynthesis and export of a group II capsular polysaccharide (CPS) (Fig. 5). Interestingly, this locus is part of a larger genomic region (extending from GeneID Arad9127 to Arad9160) in strain K84 that is virtually missing in the other fully sequenced genome of A. tumefaciens C58 and with only a few orthologous genes in the Agrobacterium vitis S4 genome (see Fig. S1 in the supplemental material). Moreover, the most highly related homologs for this K84 CPS locus genes are not from the Rhizobiaceae family (see Table S2).

Fig 5
The wcbD ortholog gene in R. rhizogenes strain K84 is part of a locus dedicated to the biosynthesis and export of a group II capsular polysaccharide (CPS). The genetic structure of the wcbD ortholog gene locus (extending from Arad9136 to Arad9152), where ...

Sequence analysis of mutant strain M108 localized the transposon insertion between nucleotides 302 and 303 of the 1,323-bp Arad3519 coding sequence, which is located on chromosome 1 of strain K84. This gene has been designated rkpK based on the predicted amino acid sequence. The K84 rkpK gene shows the highest similarity to mtpE of Mesorhizobium tianshanense (83% I and 89% S) and rkpK orthologs of Rhizobium leguminosarum bv. viciae (82% I and 89% S), Rhizobium etli (83% I and 89% S), and Sinorhizobium meliloti (79% I and 88% S), plus exo5 of R. leguminosarum bv. trifolii (82% I and 89% S). The rkpK gene encodes a UDP-glucose 6-dehydrogenase (21). In R. rhizogenes strain K84, the rkpK gene is part of a two-gene operon called rkp-2 (see Fig. S2 in the supplemental material) that also contains the lpsL gene, which is predicted to specify a UDP-glucuronic acid epimerase function (21). The synteny of the rkp-2 region of R. rhizogenes strain K84 is not conserved among the more closely phylogenetically related species within the Rhizobiaceae family (52), such as R. leguminosarum bv. viciae, R. etli, A. tumefaciens, or A. vitis (see Fig. S2). This rkp-2 operon is required for O-antigen lipopolysaccharide (LPS) synthesis, as well as for the K-antigen capsular polysaccharide (KPS) produced by some S. meliloti strains (8, 21).

wcbD gene of R. rhizogenes K84 is required for attachment and biofilm formation on the abiotic surface.

Expression of a functional copy of wcbD in the M71 mutant [M71(pWcbD) strain] completely restored the attachment and biofilm deficiencies of the mutant (Fig. 1). These results confirm that the wcbD locus plays an important role in the surface adhesion and biofilm development phenotypes of mutant M71.

rkpK gene of R. rhizogenes K84 is required for normal biofilm formation.

Expression of a functional copy of rkpK in the M108 mutant [M108(pRkpK) strain] restored normal biofilm formation on the abiotic polypropylene surface (134% ± 29% with respect to the normalized data of the ratio between DMSO-solubilized CV staining [A570] and the culture density [OD570] of the wild type) and on the surface of tomato root tips (Fig. 3). These results confirm that the rkp-2 locus is required for normal biofilm development in strain K84.

Analyses of cell envelope polysaccharides. (i) rkpK, but not the wcbD gene, is required for O-antigen LPS synthesis.

Because both wcbD and rkpK genes are involved in strain K84 biofilm formation and showed a high degree of similarity to functions involved in cell envelope polysaccharide synthesis, we examined the LPS profiles of these mutants compared to those of the wild-type strain. As shown in Fig. 6A, wild-type strain K84 gives a characteristic LPS banding pattern after gel-electrophoretic separation, showing both smooth (LPS I) and rough (LPS II; lipid A plus the core saccharide) LPS. Interestingly, the wcbD mutant does not show any visible differences in this banding pattern, whereas the rkpK mutant lacked the O-antigen fraction and was concomitantly enriched in the rough LPS band (LPS II) (Fig. 6A). Colonies of the rkpK mutant were slightly less mucoid (drier) than those of the wild type and exhibited a characteristic rough-colony morphology.

Fig 6
(A) O-antigen LPS production of wild-type K84 and its derivative mutans. Proteinase K-treated SDS-lysed cell extracts were separated in PAGE gels, and LPS was visualized after staining as described in Materials and Methods. In wild-type strain K84 (lane ...

Expression of a functional copy of rkpK in the M108 mutant [M108(pRkpK) strain] restored synthesis of smooth LPS I to the mutant (Fig. 6A). These results confirm that the rkpK gene of R. rhizogenes K84 is required for O-antigen LPS biosynthesis and normal biofilm formation.

(ii) Strain K84 does not produce the K-antigen capsular polysaccharide typical of some rhizobia.

Because of the relatedness of both WcbD and RkpK to functions involved in the synthesis of capsular polysaccharides, we were interested to examine whether strain K84 produces the high-molecular-weight K-antigen capsular polysaccharide (KPS) as described for other rhizobia (48). As shown in Fig. 6B, the total LPS extract of S. meliloti strain AK631 contains a prominent KPS fraction that is absent in KPS-deficient strain 1021. Importantly, R. rhizogenes strain K84 does not produce a related high-molecular-weight KPS (Fig. 6B). A faint banding pattern, corresponding to low-molecular-weight polysaccharides, was observed in strain K84 and its wcbD and rkpK derivative mutants.

EPS production.

Because the UDP-glucuronic acid product of the RkpK activity could also be part of the repetitive units for extracellular exopolysaccharide (EPS) biosynthesis, we evaluated the total amount of carbohydrates in EPS-containing supernatants. EPS values from the rkpK mutant supernatants were not significantly different from those of the wild type (277 ± 5.8 mg, EPS/OD600).


We have recently shown that the biocontrol agent R. rhizogenes (formerly A. radiobacter [56]) strain K84 is able to form biofilms not only on abiotic surfaces but also on plant root surfaces, where biofilms become complex and structured, with cells adhering to the surface and to each other (1). Based on these observations, we hypothesized that biofilm formation is an integral part of strain K84 host root colonization and its ability to act as an efficient biocontrol agent. A genetic screen for R. rhizogenes K84 mutants deficient in adhesion and biofilm formation identified two loci encoding functions consistent with this prediction. The first mutant is disrupted in a wcbD ortholog gene, which is required for attachment to and subsequent biofilm formation on an abiotic surface, but is capable of forming seemingly unaltered biofilms on root tip surfaces. Bacterial cell counts, luminescent detection, and fluorescence microscopy evaluation of GFP-tagged wild-type and mutant strains clearly showed similar numbers of cells within the biofilms formed by both strains on root surfaces and also indicated that wild-type K84 and its wcbD mutant were supported by a tightly adhering subpopulation that remained attached to the root surface even after rigorous washing. Thus, the wcbD mutant lacks an adhesion factor required for the interactions with abiotic polypropylene surfaces by strain K84. The reason for the observed differences in the abiotic versus biotic adhesion phenotypes of the wcbD mutant remains unclear although similar differences in adhesion patterns have also been observed in Pseudomonas putida (18, 60), Pseudomonas fluorescens (2), and Vibrio cholerae (36) mutants. Physicochemical properties of the surfaces are important aspects of bacterial adhesion (23), which could explain the observed difference in attachment to polypropylene, which is hydrophobic, versus the likely hydrophilic root surface. Adhesion to these different substrates may require specific adhesion factors. Thus, we infer that the adhesion factor encoded by the wcbD locus is specific for adhesion to hydrophobic surfaces, with little or no role in adhesion to the root surface. Alternatively, the root surface could be coated with organic material released by root exudates, providing a nutrient source which might also influence bacterial adhesion to the roots (23, 60).

Cell envelope polysaccharides, including capsular polysaccharides (CPSs), are often involved in attachment to inert surfaces and plant tissues (5, 13). Changes in the CPS content might be related to different hydrophobicity characteristics that could play a key role in surface adhesion selectivity, as suggested by De Castro et al. (16). The wcbD gene is part of a predicted CPS locus in strain K84. It is therefore likely that the wcbD mutant has an altered CPS component at the cell surface. The fact that in trans expression of wcbD restored normal adhesion and biofilm formation on the abiotic surface strongly suggests that this locus of strain K84 encodes an adhesion factor that is important for attachment to the abiotic substrate but has minimal, if any, role in adhesion to the natural plant surface. A typical bacterial capsule surrounding the cells was not evident in strain K84 growing in liquid culture, as confirmed by negative staining and transmission electron microscopy. There is the possibility that the predicted K84 capsule is a layer very loosely attached to the external membrane or constitutes a microcapsule that is not readily discerned as described in other bacteria (5, 41). R. rhizogenes is closely related to R. leguminosarum and R. etli and less so to Agrobacterium species (52). The entire wcbD locus of R. rhizogenes strain K84 is virtually missing in the A. tumefaciens C58 and A. vitis S4 genomes. Moreover, each gene present in this wcbD locus in strain K84 has low similarity to genes from the Rhizobiaceae family, suggesting that this predicted CPS locus belongs to a larger DNA region that could be a genomic island in the K84 genome.

The second biofilm mutant we identified has an insertion in the rkpK gene and resulted in the formation of denser biofilms on both the abiotic and root tip surfaces. This gene encodes a UDP-glucose dehydrogenase, which is predicted to catalyze conversion of UDP-glucose to UDP-glucuronic acid (21). The contiguous lpsL gene in this rkp-2 operon is predicted to encode a UDP-glucuronic acid epimerase that converts UDP-glucuronic acid to UDP-galacturonic acid (21). These UDP-saccharides are likely monomeric substrates for LPS (O-antigen) or KPS (K-antigen) biosynthesis since an rkpK mutant of S. meliloti was impaired in normal LPS and KPS production (8, 21). The rkpK mutation in R. rhizogenes strain K84 blocks O-antigen biosynthesis, as indicated by the enriched amounts of rough LPS (LPS II). Moreover, we show that R. rhizogenes strain K84 does not produce the high-molecular-weight K-antigen capsular polysaccharide (KPS). In S. meliloti, three genomic regions, named rkp-1 (rkpA-F, rkpG, rkpH, and rkpIJ), rkp-2 (rkpK-lpsL), and rkp-3 (rkpL-Q and rkpR-T), are required for KPS biosynthesis (21, 37, 46). The fact that the R. rhizogenes K84 genome lacks the rkp-1 and rkp-3 loci is in agreement with our analytical data showing that strain K84 does not produce the characteristic KPS fraction observed in some rhizobia strains.

Interestingly, the rkpK mutant strain preferentially grew in a sessile form as aggregates or biofilms rather than as a planktonic suspension when grown in liquid culture under both static and shaking conditions. This fact suggests that the lack of the O antigen in the LPS from the rkpK mutant enhanced adherence among cells, allowing higher bacterial numbers within the biofilms formed on either the abiotic or the root tip surface. One hypothesis to explore would be to determine if the rkpK mutant lacks the O antigen linked to the core saccharide, and, consequently, the core oligosaccharides are exposed. Some of these core saccharides, such as N-acetylglucosamine, can act as potent intercellular and surface adhesins (20). Although we cannot infer the exact role of the O-antigen LPS in the formation of biofilms by strain K84 based on this rkpK mutant phenotype, it is likely that it could be necessary for efficient root colonization by this biocontrol agent (26). In fact, the lack of the O-antigen LPS in the rkpK mutant could compromise its fitness since other O-antigen mutants of several rhizobial species have shown decreased competitiveness compared with their wild types (15, 38). In this study we also observed that the rkpK gene was not affecting the biosynthesis of EPS, which suggests that UDP-glucuronic acid is not a major monomeric substrate for the EPS pool in the R. rhizogenes K84 strain.

Clearly, biofilm formation is highly complex and depends on a number of development-specific functions. The data presented here are a first step in dissecting the biofilm formation process in this important biocontrol agent. The wcbD gene is required for attachment to polypropylene (cell-surface interactions), most likely through the synthesis of a yet uncharacterized capsular polysaccharide (CPS), which is distinct from the K-antigen KPS produced by other rhizobial strains. The K84 genome carries only a subset of the rkp genes (specifically the rkp-2 locus). One of these genes, rkpK, is required for the synthesis of the O antigen in the LPS from strain K84 and to form normal biofilms by this crown gall biocontrol agent.

Supplementary Material

Supplemental material:


We thank M. Martin for sending pHC60.

A. M. Abarca-Grau was the recipient of a predoctoral fellowship from INIA (Spain), and E. Marco-Noales has a contract from the Spanish Ministry of Education and Science (Programa INIA-CCAA) cofunded by the European Social Fund. This work was supported by grants RTA07-112 and RTA10-96 from INIA (Spain) and cofunded by the European Social Fund to R. Penyalver and by NSF grant 1053869 to S. B. von Bodman. Part of this work was conducted during a sabbatical leave of R. Penyalver at the University of Connecticut funded by MEC (Spain). In addition, this material was based on work supported by the National Science Foundation while S. B. von Bodman was working at the Foundation.

Any opinion, finding, and conclusions or recommendations expressed in this article are those of the authors and do not necessarily reflect the views of the National Science Foundation.


Published ahead of print 30 December 2011

Supplemental material for this article may be found at http://aem.asm.org/.


1. Abarca-Grau AM, Penyalver R, López MM, Marco-Noales E. 2011. Pathogenic and non-pathogenic Agrobacterium tumefaciens, A. rhizogenes and A. vitis strains form biofilms on abiotic as well as on root surfaces. Plant Pathol. 60:416–425
2. Barahona E, et al. 2010. Efficient rhizosphere colonization by Pseudomonas fluorescens f113 mutants unable to form biofilms on abiotic surfaces. Environ. Microbiol. 12:3185–3195 [PubMed]
3. Barbosa RL, Benedetti CE. 2007. BigR, a transcriptional repressor from plant-associated bacteria, regulates an operon implicated in biofilm growth. J. Bacteriol. 189:6185–6194 [PMC free article] [PubMed]
4. Beringer JE. 1974. R factor transfer in Rhizobium leguminosarum. J. Gen. Microbiol. 84:188–198 [PubMed]
5. Beveridge TJ, Graham LL. 1991. Surface layers of bacteria. Microbiol. Rev. 55:684–705 [PMC free article] [PubMed]
6. Buendía-Clavería AM, et al. 2003. A purL mutant of Sinorhizobium fredii HH103 is symbiotically defective and altered in its lipopolysaccharide. Microbiology 149:1807–1818 [PubMed]
7. Caetano-Anolles G, Bassam BJ. 1993. DNA amplification fingerprinting using arbitrary oligonucleotide primers. Appl. Biochem. Biotechnol. 42:189–200 [PubMed]
8. Campbell GRO, et al. 2003. Striking complexity of lipopolysaccharide defects in a collection of Sinorhizobium meliloti mutants. J. Bacteriol. 185:3853–3862 [PMC free article] [PubMed]
9. Cangelosi GA, Best EA, Martinetti G, Nester EW. 1991. Methods for genetic analysis of Agrobacterium. Methods Enzymol. 204:384–397 [PubMed]
10. Cheng H-P, Walker GC. 1998. Succinoglycan is required for initiation and elongation of infection threads during nodulation of alfalfa by Rhizobium meliloti. J. Bacteriol. 180:5183–5191 [PMC free article] [PubMed]
11. Chilton M-D, et al. 1974. Agrobacterium tumefaciens DNA and PS8 bacteriophage DNA not detected in crown gall tumors. Proc. Natl. Acad. Sci. U. S. A. 71:3672–3676 [PMC free article] [PubMed]
12. Chin-A-Woeng T, de Priester W, van der Bij AJ, Lugtenberg BJJ. 1997. Description of the colonization of a gnotobiotic tomato rhizosphere by Pseudomonas fluorescens biocontrol strain WCS365, using scanning electron microscopy. Mol. Plant Microbe Interact. 10:79–86
13. Danhorn T, Fuqua C. 2007. Biofilm formation by plant-associated bacteria. Annu. Rev. Microbiol. 61:401–422 [PubMed]
14. Danhorn T, Hentzer M, Givskow M, Parsek M, Fuqua C. 2004. Phosphorous limitation enhances biofilm formation of the plant pathogen Agrobacterium tumefaciens through the PhoR-PhoB regulatory system. J. Bacteriol. 186:4492–4501 [PMC free article] [PubMed]
15. D'Antuono AL, Casabuono A, Couto A, Ugalde RA, Lepek VC. 2005. Nodule development induced by Mesorhizobium loti mutant strains affected in polysaccharide synthesis. Mol. Plant Microbe Interact. 18:446–457 [PubMed]
16. De Castro C, Gargiulo V, Lanceta R, Parrilli M. 2007. Agrobacterium rubiT DSM 6772 produces a lipophilic polysaccharide capsule whose degree of acetylation is growth modulated. Biomacromolecules 8:1047–1051 [PubMed]
17. De Cleene M, De Ley J. 1976. The host range of crown gall. Bot. Rev. 42:389–466
18. Espinosa-Urgel M, Salido A, Ramos JL. 2000. Genetic analysis of functions involved in adhesion of Pseudomonas putida to seeds. J. Bacteriol. 182:2363–2369 [PMC free article] [PubMed]
19. Ghannoum M, O'Toole GA. 2004. Microbial biofilms. ASM Press, Washington, DC
20. Jefferson KK. 2009. Poly-N-acetyl-glucosamine as a mediator of bacterial biofilm formation, p 175–188 In Ullrich M, editor. (ed), Bacterial polysaccharides. Caister Academic Press, Bremen, Germany
21. Kereszt A, et al. 1998. Novel rkp gene cluster of Sinorhizobium meliloti involved in capsular polysaccharide production and invasion of the symbiotic nodule: the rkpK gene encodes a UDP-glucose dehydrogenase. J. Bacteriol. 180:5426–5431 [PMC free article] [PubMed]
22. Kerr A. 1972. Biological control of crown gall: seed inoculation. J. Appl. Microbiol. 35:493–497
23. Korber DR, Lawrence JR, Lappin-Scott HM, Costerton JW. 1995. Growth of microorganisms on surfaces, p 15–45 In Lappin-Scott HM, Costerton JW, editors. (ed), Microbial biofilms. Cambridge University Press, Cambridge, United Kingdom
24. Kovach ME, et al. 1995. Four new derivatives of the broad-host-range cloning vector pBBR1MCS, carrying different antibiotic-resistance cassettes. Gene 166:175–176 [PubMed]
25. Kroll JS, Loynds B, Brophy LN, Moxon ER. 1990. The bex locus in encapsulated Haemophilus influenzae: a chromosomal region involved in capsular polysaccharide export. Mol. Microbiol. 4:1853–1862 [PubMed]
26. Lerouge I, Vanderleyden J. 2002. O-antigen structural variation: mechanisms and possible roles in animal/plant-microbe interactions. FEMS Microbiol. Rev. 26:17–47 [PubMed]
27. Liberati NT, et al. 2006. An ordered, nonredundant library of Pseudomonas aeruginosa strain PA14 transposon insertion mutants. Proc. Natl. Acad. Sci. U. S. A. 103:2833–2838 [PMC free article] [PubMed]
28. Lo RYC, McKerral LJ, Hills TL, Kostrzynska M. 2001. Analysis of the capsule biosynthetic locus of Mannheimia (Pasteurella) haemolytica A1 and proposal of a nomenclature system. Infect. Immun. 69:4458–4464 [PMC free article] [PubMed]
29. Macrae S, Thomson JA, Van Staden J. 1988. Colonization of tomato plants by two agrocin-producing strains of Agrobacterium tumefaciens. Appl. Environ. Microbiol. 54:3133–3137 [PMC free article] [PubMed]
30. Matthysse AG. 1987. Characterization of nonattaching mutants of Agrobacterium tumefaciens. J. Bacteriol. 169:313–323 [PMC free article] [PubMed]
31. Matthysse AG, McMahan S. 1998. Root colonization by Agrobacterium tumefaciens is reduced in cel, attB, attD, and attR mutants. Appl. Environ. Microbiol. 64:2341–2345 [PMC free article] [PubMed]
32. Matthysse AG, et al. 2005. The effect of cellulose overproduction on binding and biofilm formation on roots by Agrobacterium tumefaciens. Mol. Plant Microbe Interact. 18:1002–1010 [PubMed]
33. Merrit PM, Danhorn T, Fuqua C. 2007. Motility and chemotaxis in Agrobacterium tumefaciens surface attachment and biofilm formation. J. Bacteriol. 189:8005–8014 [PMC free article] [PubMed]
34. Mersereau M, Pazour GJ, Das A. 1990. Efficient transformation of Agrobacterium tumefaciens by electroporation. Gene 90:149–151 [PubMed]
35. Morris CE, Monier JM. 2003. The ecological significance of biofilm formation by plant-associated bacteria. Annu. Rev. Phytopathol. 41:429–453 [PubMed]
36. Mueller RS, et al. 2007. Vibrio cholera strains possess multiple strategies for abiotic and biotic colonization. J. Bacteriol. 189:5348–5360 [PMC free article] [PubMed]
37. Müller MG, Forsberg LS, Keating DH. 2009. The rkp-1 cluster is required for secretion of Kdo homopolymeric capsular polysaccharide in Sinorhizobium meliloti Rm1021. J. Bacteriol. 191:6988–7000 [PMC free article] [PubMed]
38. Noel KD, Forsberg LS, Carlson RW. 2000. Varying the abundance of O-antigen in Rhizobium etli and its effect on symbiosis with Phaseolus vulgaris. J. Bacteriol. 182:5317–5324 [PMC free article] [PubMed]
39. O'Toole GA, Kolter R. 1998. Initiation of biofilm formation in Pseudomonas fluorescens WCS365 proceeds via multiple, convergent signalling pathways: a genetic analysis. Mol. Microbiol. 28:449–461 [PubMed]
40. Parada M, et al. 2006. Sinorhizobium fredii HH103 mutants affected in capsular polysaccharide (KPS) are impaired for nodulation with soybean and Cajanus cajans. Mol. Plant Microbe Interact. 19:43–52 [PubMed]
41. Patrick S, Houston S, Thacker Z, Blakely GW. 2009. Mutational analysis of genes implicated in LPS and capsular polysaccharide biosynthesis in the opportunistic pathogen Bacteroides fragilis. Microbiology 155:1039–1049 [PubMed]
42. Pearce Bazin DMJ, Lynch JM. 1995. The rhizosphere as a biofilm, p 207–220 In Lappin-Scott HM, Costerton JW, editors. (ed), Microbial biofilms. Cambridge University Press, Cambridge, United Kingdom
43. Peñalver R, Serra MT, Durán-Vila N, López MM. 1996. Attachment of Agrobacterium tumefaciens B6 and Agrobacterium radiobacter K84 to tomato root tips. Appl. Environ. Microbiol. 62:3530–3534 [PMC free article] [PubMed]
44. Penyalver R, López MM. 1999. Co-colonization of the rhizosphere by pathogenic Agrobacterium strains and nonpathogenic strain K84 and K1026, used for crown gall biocontrol. Appl. Environ. Microbiol. 65:1936–1940 [PMC free article] [PubMed]
45. Penyalver R, Vicedo B, López MM. 2000. Use of the genetically engineered Agrobacterium strain K1026 for biological control of crown gall. Eur. J. Plant Pathol. 106:801–810
46. Putnoky P, et al. 1990. Rhizobium meliloti lipopolysaccharide and exopolysaccharide can have the same function in the plant-bacterium interaction. J. Bacteriol. 172:5450–5458 [PMC free article] [PubMed]
47. Ramey BE, Koutsoudis M, von Bodman SB, Fuqua C. 2004. Biofilm formation in plant microbe associations. Curr. Opin. Microbiol. 7:602–609 [PubMed]
48. Reuhs BL, Carlson RW, Kim JS. 1993. Rhizobium fredii and Rhizobium meliloti produce 3-deoxy-d-manno-2-octulosonic acid-containing polysaccharides that are structurally analogous to group II K antigens (capsular polysaccharides) found in Escherichia coli. J. Bacteriol. 175:3570–3580 [PMC free article] [PubMed]
49. Rodríguez-Carvajal MA, et al. 2001. Determination of the chemical structure of the capsular polysaccharide of strain B33, a fast-growing soya bean-nodulating bacterium isolated from an arid region of China. Biochem. J. 357:505–511 [PMC free article] [PubMed]
50. Sambrook J, Fritsch EF, Maniatis TA. 1989. Molecular cloning: a laboratory manual, 2nd ed Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
51. Simon R, Priefer U, Pühler A. 1983. Vector plasmid for in vivo and in vitro manipulations of gram-negative bacteria, p 98–106 In Pühler A, editor. (ed), Molecular genetics of the bacteria-plant interaction. Spring-Verlag KG, Berlin, Germany
52. Slater SC, et al. 2009. Genome sequences of three Agrobacterium biovars help elucidate the evolution of multichromosome genomes in bacteria. J. Bacteriol. 191:2501–2511 [PMC free article] [PubMed]
53. Stockwell VO, Moore LW, Loper JE. 1993. Fate of Agrobacterium radiobacter K84 in the environment. Appl. Environ. Microbiol. 59:2112–2120 [PMC free article] [PubMed]
54. Tempé J, Petit A, Holsters M, Van Montagu M, Schell J. 1977. Thermosensitive step associated with transfer of the Ti plasmid during conjugation: possible relation to transformation in crown gall. Proc. Natl. Acad. Sci. U. S. A. 74:2848–2849 [PMC free article] [PubMed]
55. Tomlinson AD, Ramey-Hartung B, Day TV, Merrit PM, Fuqua C. 2010. Agrobacterium tumefaciens ExoR represses succinoglycan biosynthesis and is required for biofilm formation and motility. Microbiology 156:2670–2681 [PMC free article] [PubMed]
56. Velázquez E, et al. 2010. Analysis of core genes supports the reclassification of strains Agrobacterium radiobacter K84 and Agrobacterium tumefaciens AKE10 into the species Rhizobium rhizogenes. Syst. Appl. Microbiol. 33:247–251 [PubMed]
57. Vicedo B, Peñalver R, Asíns MJ, López MM. 1993. Biological control of Agrobacterium tumefaciens, colonization, and pAgK84 transfer with Agrobacterium radiobacter K84 and the Tra mutant strain K1026. Appl. Environ. Microbiol. 59:309–315 [PMC free article] [PubMed]
58. Warawa JM, Long D, Rosenke RD, Gardner Gherardini F. 2009. Role for the Burkholderia pseudomallei capsular polysaccharide encoded by the wcb operon in acute disseminated melioidosis. Infect. Immun. 77:5252–5261 [PMC free article] [PubMed]
59. Reference deleted.
60. Yousef-Coronado F, Travieso ML, Espinosa-Urgel M. 2008. Different, overlapping mechanisms for colonization of abiotic and plant surface by Pseudomonas putida. FEMS Microbiol. Lett. 288:118–124 [PubMed]
61. Zhu J, et al. 2000. The bases of crown gall tumorigenesis. J. Bacteriol. 182:3885–3895 [PMC free article] [PubMed]
62. Zupan J, Muth TR, Draper O, Zambryski P. 2000. The transfer of DNA from Agrobacterium tumefaciens into plants: a feast of fundamental insights. Plant J. 23:11–28 [PubMed]

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...