Complex Regulation of Symbiotic Functions Is Coordinated by MucR and Quorum Sensing in Sinorhizobium meliloti

Bacterial strains and growth conditions.S. meliloti strains (Table 1) were grown in Luria-Bertani (LB) broth or agar supplemented with 2.5 mM MgSO₄, 2.5 mM CaCl₂ (MC), and appropriate antibiotics. For RNA isolation, 2 ml of TYC broth (5 g of tryptone, 3 g of yeast extract, and 0.4 g of CaCl₂ liter⁻¹) with streptomycin (500 μg ml⁻¹) was inoculated with colonies grown on LB-MC agar plates and incubated for 48 h at 30°C with constant shaking. The strains were then subcultured (1:100) in 20 ml of minimal glutamate mannitol (MGM) low-phosphate medium (50 mM morpholineethanesulfonic acid [MOPS], 19 mM sodium glutamate, 55 mM mannitol, 0.1 mM K₂HPO₄·KH₂PO₄, 1 mM MgSO₄, 0.25 mM CaCl₂, 0.004 mM biotin, pH 7) and grown at 30°C with constant shaking. The addition of luteolin to 10 μM was included when effects on nod operon gene expression were to be analyzed. When necessary, chloramphenicol (20 μg ml⁻¹), gentamicin (100 μg ml⁻¹ for S. meliloti and 10 μg ml⁻¹ for Escherichia coli), hygromycin (100 μg ml⁻¹), neomycin (200 μg ml⁻¹), tetracycline (10 μg ml⁻¹), or trimethoprim (500 μg ml⁻¹ for S. meliloti and 30 μg ml⁻¹ for E. coli) was added.

Construction of S. meliloti strains and plasmids.Oligonucleotide sequences used in this study are listed in Table S1 in the supplemental material. All mutations for this work were introduced into Rm8530 using the transducing phage φM12 as described previously (22). Expression of expC under a constitutively expressing promoter was accomplished by amplification of a fragment from Rm8530 carrying the open reading frame expC with primers expC+no-linker F and expC+PstI R. Both pJN105 with the stability region from pTR101 (52) and this amplicon were then digested with PstI and ligated together to form pJNexpC. This protocol was also followed for the mucR gene, only using primers mucR+SmaI F and mucR+SmaI R for sequence amplification and the SmaI endonuclease for digestion of both insert and plasmid, resulting in pJNmucR. Due to the redundant presence of gentamicin resistance in the pJN105 plasmid as well as in several strains used in this work, pJNexpC and pJNmucR were modified with an EZ-Tn5 <DHFR-1> Insertion Kit from Epicentre to provide trimethoprim resistance instead. All plasmids were introduced into S. meliloti by triparental mating. A list of plasmids used in this work is provided in Table 1.

Bacterial RNA isolation.Cultures were grown at 30°C to optical densities at 600 nm (OD₆₀₀) of 0.02, 0.1, and 1.2 in either 500 ml or 20 ml of MGM low-phosphate medium. Larger volumes were required for bacteria of low population densities in order to obtain sufficient RNA for analysis. Bacteria were collected by centrifugation (17,000 × g for 10 min at 4°C for 500-ml liquid cultures or 14,000 × g for 2 min at 4°C for 20-ml liquid cultures), and the pellets were frozen in liquid nitrogen. Total RNA was purified by using an RNeasy Mini Kit (Qiagen) as described in Gurich et al. (26). RNA integrity was determined in an Agilent 2100 Bioanalyzer, and the concentrations were measured with Nanodrop spectrophotometer ND-1000.

Quantitative real-time PCR.In order to quantify the presence of specific mRNA transcripts of interest, 1 μg of RNA isolated from bacteria grown in culture was used per Ambion RETROscript reverse-transcription reaction. One microliter of resultant cDNA was used as a template for quantitative real-time PCR analysis. Each reaction mixture was prepared as described previously (26) and included appropriate primers for amplification of the desired mRNA (see Table S1 in the supplemental material). Expression of SMc00128 was used as an internal control for normalization (30). Fold change in expression between strain A versus strain B was calculated using observed threshold cycle (C_T) values of each (C_Ta and C_Tb, respectively) and the following equation: fold change = 2^{CTb − CTa}. Expression levels of expE2, expA1, and expD1 are presented in this work as representations of transcriptional levels of operons since expression of open reading frames within each operon provided similar values (data not shown).

Microarray experiments and data analyses.Ten micrograms of total RNA from wild-type and mucR-mutant strains of S. meliloti grown to an OD₆₀₀ of 1.2 was used for analysis by microarray to evaluate gene expression levels. The synthesis of cDNA, labeling, and hybridizations to the S. meliloti/Medicago truncatula Affymetrix GeneChip (Santa Clara, CA) were performed by the Core Microarray Facility at UT Southwestern Medical Centre (Dallas, TX). Data were processed using GeneSifter as described by Gurich et al. (26) and uploaded to the NCBI Gene Expression Omnibus (GEO) database. Genes were considered to be differentially expressed if the fold change in expression was >1.5 with a P value < 0.05. In order to observe potential metabolic differences between the wild type and mucR, the complete list of genes differentially expressed greater than 1.5-fold was further analyzed by Kegg Array, version 1.2.3.

EPS II isolation.Succinoglycan mutants (exoY) of S. meliloti grown on LB-MC agar plates were used to inoculate 5 ml of TYC broth with 500 μg ml⁻¹ streptomycin. These were incubated for 48 h with constant shaking at 30°C. TYC cultures were centrifuged and resuspended in 1 liter of mannitol glutamate salts (MGS) medium used for S. meliloti EPS II production (37). MGS cultures were grown for 7 days at 30°C with constant shaking, and EPS II released into the medium was extracted and purified as previously described (24). Cultures were centrifuged, and supernatants were lyophilized. Samples were resuspended in 10 ml of H₂O and precipitated with three volumes of ethanol. This precipitate was resuspended in the smallest volume of H₂O possible and dialyzed for 4 days at 4°C using a membrane with a 500-Da molecular mass cutoff. Samples were then lyophilized and resuspended in H₂O prior to analysis by high-performance anion-exchange chromatography (HPAEC).

High-performance anion-exchange chromatography.Analysis of the EPS II oligosaccharides was performed by HPAEC on a Dionex metal-free BioLC with a CarboPac PA10 (anion-exchange) column (4 by 3 by 250 mm) using a pulsed amperometric detector reporting in nanocolumns (nC) with a gold working electrode and a triple-step carbohydrate waveform (Dionex), as described previously (24). Eluent A consisted of 100 mM NaOH while eluent B consisted of 1 M sodium acetate in 100 mM NaOH. Consistent flow of eluent A was maintained with a gradient of eluent B as follows: 0 min, 10% eluent B; 3 min, 10% eluent B; 10 min, 20% eluent B; 25 min, 100% eluent B; 50 min, 100% eluent B; 55 min, 10% eluent B; and 60 min, 10% eluent B. The flow rate was 1 ml min⁻¹ at room temperature. Reconstitution of the column by passing 100% 300 mM NaOH for 30 min between samples was critical for maintaining the ability of the column to separate fractions of EPS II.

Biofilm formation assay-microtiter plate method.The ability to form biofilm was determined by a modified protocol described in Rinaudi and González (46). Cultures were grown in low-phosphate MGM medium as described above, diluted to an OD₆₀₀ of 1.0, and inoculated into microtiter wells in 100-μl aliquots in at least triplicate. A sterile microporous film was then placed over the entirety of the plate (AeraSeal catalogue number BS-25), and the cover was added to prevent evaporation. The plate was then inverted to prevent the settling of bacteria from interfering with true attachment to the walls of the well due to biofilm formation and then incubated with gentle rocking for 48 h at 30°C and 65% relative humidity. Free liquid culture was gently removed from the wells, and each well was air dried and stained for 15 min with 150 μl of 0.1% crystal violet dye. Each well was then rinsed three times with water, and crystal violet was resuspended with 150 μl ethanol. The absorbance of the solubilized dye at OD₅₆₀ was measured using a microplate reader (Infinite M200; Tecan Trading AG, Männedorf, Switzerland) as a measure of biofilm formation.

Plant nodulation assays.Plant nodulation and invasion assays with the symbiotic host M. sativa were carried out in triplicate sets of 15 plants per strain per experiment. Five-milliliter cultures of S. meliloti were grown in LB-MC broth with 500 μg ml⁻¹ streptomycin at 30°C for 48 h. Cultures were washed four times with sterile water, and 1 ml of a 1:100 dilution was used to inoculate 3-day-old plant seedlings as described previously (26). Plates were incubated at 22°C with 60% relative humidity and a 16-h light cycle. Plants were examined daily or weekly, depending on the observation of nodule development or successful invasion, respectively. To determine the capacity of a strain to produce symbiotically active LMW EPS II, inoculated alfalfa plants were examined after 4 weeks for the presence of pink nodules, indicative of the successful establishment of symbiosis.

Expression of expC is critical for the production of LMW EPS II.The molecular-weight content of the EPS II produced by the various mutants of S. meliloti was confirmed by three independent means of detecting the low-molecular-weight form specifically, regardless of overall EPS II production: high-performance anion-exchange chromatography (HPAEC), the capacity of the mutants to form biofilm, and the ability of each to invade M. sativa in the absence of succinoglycan. Work in our laboratory has shown that EPS II, when separated by HPAEC and detected by pulsed amperometry, produces discernible peaks indicative of each fraction. LMW EPS II is identified by repeated peaks from 24 to 27 min of chromatography, while HMW EPS II produces a broad mound at 28 min (24). As expected, analysis of EPS II produced by wild-type S. meliloti grown in mannitol glutamate salts (MGS) indicated the presence of both HMW and LMW fractions (Fig. 3A) (24). Evaluation of the exopolysaccharide produced by the mucR expG mutant indicated the presence of HMW EPS II exclusively. Supplementation of expC expression levels with the introduction of pJTpexpC into this strain restored the production of LMW EPS II. The mucR mutant produced a pattern identical to that of the wild-type strain, as observed previously (24), while the expG mutant synthesized no EPS II for analysis.

• Open in new tab
• Download powerpoint

FIG. 3.

Restored expression of expC results in the biosynthesis of LMW EPS II. (A) Purified EPS II produced from wild-type (WT) and mutant strains was analyzed by HPAEC for the presence of LMW EPS II. Analysis of exopolysaccharide produced by the wild-type S. meliloti confirmed the presence of LMW EPS II based upon distinctive peaks between 24 and 27 min. The mucR expG mutant failed to produce LMW EPS II. Supplementation (by introduction of pJTpexpC) of transcriptional levels of expC in this mutant strain restored the synthesis of the low-molecular-weight fraction. Results were overlaid for comparison. (B) The presence of LMW EPS II produced expected levels of biofilm formation and attachment. The mucR expG mutant strain failed to form biofilm while the strain carrying pJTpexpC attached at equivalent levels to the level of the wild type. (C) In the absence of succinoglycan production, the result of a disruption in exoY, the ability of S. meliloti to invade the host plant and establish a nitrogen-fixing symbiosis, visually discernible by the development of large pink nodules, is a clear indicator of the production of LMW EPS II. The disruption of exoY in the mucR expG mutant abolished the capacity of the strain to invade the plant. Supplementation of expression levels of expC in this strain through the introduction of pJTpexpC resulted in an invasive capacity comparable to that of the exoY strain, confirming the synthesis of LMW EPS II as a result of restored expression of expC.

Recently, studies have shown that the production of LMW EPS II results in biofilm formation and attachment at levels as high as 10-fold greater than those produced by strains of S. meliloti incapable of synthesizing this exopolysaccharide (18, 46). Thus, the ability of the bacteria to establish levels of biofilm comparable to wild-type levels was utilized as a secondary indicator of the presence or absence of the low-molecular-weight form. S. meliloti cells were inoculated in MGM low-phosphate medium and grown in 96-well microtiter dishes. Attachment to the walls of the well due to biofilm formation was measured by crystal violet staining. The deduced presence of LMW EPS II, as observed by the capacity of each strain to form biofilm, coincided with the fractions detected by HPAEC (Fig. 3B). Wild-type levels of attachment were abolished in the mucR expG mutant while supplementation of expC expression levels restored full biofilm formation.

Finally, invasion of M. sativa by S. meliloti is mediated either by succinoglycan (encoded by the exo gene family) or the LMW fraction of EPS II (24, 42). The ability of a strain to establish symbiosis after its capacity to produce succinoglycan has been disrupted is a clear indicator of the production of the low-molecular-weight form, regardless of overall levels of EPS II produced (24). Thus, an exoY mutation was introduced into each strain of S. meliloti, and the invasive capacities of the strains were examined. Disruption of exoY in the wild type abolished succinoglycan production but did not prevent the strain from establishing a successful symbiosis with the host plant due to the continued biosynthesis of symbiotically active LMW EPS II. The mucR expG mutant lost its invasive ability upon mutation of exoY (Fig. 3C) as it failed to produce the symbiotically active fraction of EPS II. The introduction of pJTpexpC into this mucR expG exoY mutant restored the ability to invade the plant through the reestablished production of LMW EPS II. These data confirm that expression levels of expC equivalent to those induced by quorum sensing are critical for the production of symbiotically active EPS II.

MucR represses exp gene expression prior to the establishment of quorum.Although disruption of mucR in the absence of an intact quorum-sensing system had been shown to dramatically derepress several exp operons (48), in wild-type S. meliloti, no significant effect on EPS II-related gene expression had been previously reported at any stage of growth. Even at what had been considered to be “low population density” prior to the establishment of quorum (OD₆₀₀ of 0.2), the ExpR/Sin system appeared to completely abolish transcriptional repression by MucR (26). The purpose of the capacity of MucR to repress specific exp genes in S. meliloti was unclear, given that its utilization had never been reported without additional mutations in quorum sensing. However, changes in exp gene expression as a result of a disrupted mucR had not been previously observed below the described low population density.

In order to examine the role of MucR at these earlier points in growth, exp transcriptional activity was measured by collecting RNA from MGM low-phosphate cultures grown to OD₆₀₀ of 0.02, 0.1, and 1.2, followed by quantitative real-time PCR. At the lowest stage of growth (OD₆₀₀ of 0.02), overall expression of the exp gene family in wild-type S. meliloti was similar to that of a sinI mutant incapable of quorum sensing (Fig. 4A). To confirm that these transcriptional levels were specific to the exp operons and not artifacts of overall low gene expression, transcript levels of rem, previously reported to be high at low population density (26), were measured (Fig. 4B). By an OD₆₀₀ of 0.1, full expression of the exp family similar to that observed at an OD₆₀₀ of 1.2 had been restored, suggesting that the quorum required for EPS II production was achieved at some point between OD₆₀₀ of 0.02 and 0.1. Disruption of mucR in the population grown to an OD₆₀₀ of 0.02 resulted in the derepression of operons expE, expA, and expD, responsible for the biosynthesis of EPS II, 4.5-, 2.5-, and 2.4-fold, respectively (Fig. 4C). By an OD₆₀₀ of 0.1, this mutation had no significant effect on expression of these genes. These data suggest that the repressive effects of MucR on EPS II biosynthesis are active only at extremely low population densities previously unexamined. The quorum required for EPS II production is achieved at a particularly low stage of growth between OD₆₀₀ of 0.02 and 0.1, at which point repression by MucR is abolished.

• Open in new tab
• Download powerpoint

FIG. 4.

MucR represses the exp gene family prior to quorum. (A) Wild-type S. meliloti was grown in MGM low-phosphate medium to OD₆₀₀ of 0.02, 0.1, and 1.2. Analysis by real-time PCR and comparison to the sinI mutant indicated that expression levels of the exp gene family, in particular, expE, expA, and expD directly responsible for EPS II biosynthesis, were first induced by quorum at a population density between an OD₆₀₀ of 0.02 and an OD₆₀₀ of 1.2. (B) Expression of rem confirmed that low expression levels observed at an OD₆₀₀ of 0.02 were particular to the exp gene family, not a by-product of low global gene expression at this stage of growth. (C) Disruption of mucR resulted in derepression of expE, expA, and expD operons at an OD₆₀₀ of 0.02. However, the repressive effect of an intact mucR attenuated to negligible levels by an OD₆₀₀ of 0.1, suggesting that MucR facilitates the inhibition of EPS II production at a low population density until quorum is reached. At this point, the ExpR/Sin system derepresses the exp gene family. Expression levels of expG and expC are not shown as this operon is not under the transcriptional regulation of MucR.

MucR induces bacterial sensitivity to plant flavonoids, resulting in increased plant nodulation.At sufficient population densities, quorum sensing in S. meliloti abrogates transcriptional regulation by MucR on 22 open reading frames across expE, expA, and expD operons without repressing the expression of mucR (26). The complex manner by which S. meliloti utilizes the ExpR/Sin system to negate the repressive effects of MucR at each specific binding site upstream of the exp operons (4, 34), rather than simply repress mucR at the transcriptional level, suggested the possibility that the maintained expression of mucR provided the bacterium with significant advantages. In addition, the ability of MucR to induce the production of succinoglycan (9) and repress motility (5) demonstrated its diverse regulatory potentials. Through microarray analysis described later in this article, a variety of alternative effects of MucR were discovered, including a 1.7-fold increase in expression of nodD, encoding the transcriptional activator of the downstream nod genes required for the synthesis of bacterial nod factors and development of root nodules (17, 32) (see Table S2 in the supplemental material). To determine if this modest increase in expression of nodD had a measurable effect on the biosynthesis of nod factor, expression of the nodABC operon responsible for its production was measured using quantitative real-time PCR in wild-type S. meliloti as well as a mucR mutant (Fig. 5). In the absence of the plant flavonoid (luteolin) required for full induction of the nod operon by NodD (32), only a slight decrease in expression of nodABC was observed with the disruption of mucR. However, in the presence of 10 μM luteolin, the handicap in the transcriptional induction of nodABC appeared to be roughly 10-fold in the absence of mucR. Complementation of mucR with the pJTpmucR vector restored this expression.

• Open in new tab
• Download powerpoint

FIG. 5.

MucR increases nod operon gene expression in response to luteolin. In the absence of induction by luteolin, a disruption in mucR resulted in a negligible change in nodA expression. Addition of luteolin resulted in an approximately 90-fold induction of the nod operon while disruption of mucR attenuated this effect to roughly 10-fold. Complementation of mucR with pJTpmucR restored this expression. Expression of the first open reading frame (nodA) of the nod operon is shown here, as similar transcriptional levels were observed for downstream nodB and nodC (data not shown).

To confirm whether these changes in nod gene expression resulted in measurable differences in the ability to induce nodule development, 3-day-old alfalfa seedlings were inoculated with S. meliloti carrying either an intact or disrupted mucR gene. The resulting numbers of developing nodules were recorded (Fig. 6A). The prevention of successful invasion was critical in order to avert the potential for established symbioses interfering with continued nodule organogenesis. Furthermore, this isolated any potential effects of the mucR disruption on invasion from nod factor production. In order to accomplish this, mutations in expA and exoY were introduced into all strains to abolish the production of EPS II and succinoglycan, respectively. This permitted the observation of differences in nodule development independent of plant invasion. Beginning with the first observation of nodules 7 days after inoculation, the disruption of mucR resulted in a delay in nodule development persisting throughout the course of the experiment. Complementation of mucR recovered this deficiency. A comparison of the number of nodules produced by each strain to that of the mucR mutant indicates an initial 7-fold advantage provided by carrying an intact copy of mucR, attenuating to roughly 1.3-fold after 30 days (Fig. 6B).

• Open in new tab
• Download powerpoint

FIG. 6.

The presence of an intact mucR provides an advantage for nodule induction in M. sativa. (A) The ability of wild-type and mutant strains of S. meliloti to develop nodules was examined daily after inoculation of bacteria onto seedlings. Disruption of mucR in strains incapable of invading resulted in a decrease in the ability to induce the formation of nodules on the host plant. Complementation of mucR restored this ability to wild-type levels. All strains were disrupted for expA and exoY in order to prevent invasion of the plant and establishment of symbiosis from interfering with nodule development as an indicator of nod factor production. (B) Direct comparison of the ability of strains to produce nodules (panel A) between those carrying either an intact or disrupted mucR indicates an initial 7-fold advantage in nodule induction associated with the presence of an intact mucR gene sequence. This advantage attenuates to roughly 1.3-fold by day 30.

Disruption of mucR affects the transcription of a multitude of genes necessary for symbiosis.The diverse effects of mucR in S. meliloti observed throughout this work warranted a more complete examination of any possible as yet undiscovered additional roles. In order to identify these potential functions, RNA was extracted from wild-type and mucR mutant strains of S. meliloti and analyzed by microarray utilizing Affymetrix GeneChips (see Table S2 in the supplemental material). In total, 802 genes were differentially expressed by at least 2-fold. Of the 154 genes which showed dramatic changes in transcription (10-fold or greater), 138 of these were both induced in the absence of mucR and located on symbiotic megaplasmid, pSymA.

The disruption of mucR resulted in the overexpression of 19 nitrogen fixation and respiration genes by 8- to 1,017-fold, all of which had been previously shown to be induced in bacteroids within the nodule after the establishment of symbiosis (Table 2) (7, 11, 14, 15, 26). Additionally, 14 motility and chemotaxis-related genes increased in expression 1.8- to 11-fold, an effect similarly observed in a quorum-sensing-deficient strain of S. meliloti by Bahlawane et al. (5). In order for bacteria to organize into sedentary biofilm communities, repression of motility is crucial. While Gurich et al. have shown that this is predominantly accomplished in S. meliloti through repression by AHL-bound ExpR at high population density (26), these data suggest that MucR may also contribute to the mitigation of motility.

In addition, expression of 11 open reading frames of the virB operon involved in type IV secretion systems increased 15- to 605-fold, and four genes encoding transport-related proteins were induced 6- to 55-fold (see Table S2 in the supplemental material). Seven of these open reading frames related to transport or secretion were also observed to have increased expression within the nodule (Table 2). Differentially expressed nitrogen fixation, motility, type IV secretion, and transport genes described here were confirmed by quantitative real-time PCR (data not shown). These data show a multitude of downstream effects from a disrupted mucR across a variety of bacterial functions. Furthermore, the derepression of numerous genes that are also known to be induced during symbiosis indicates a possible broad role of mucR as an inhibitor of bacteroid gene expression in planktonic S. meliloti.

For a more global understanding of the extensive microarray data, the entire list of genes differentially expressed more than 1.5-fold was analyzed using a Kegg Array, presenting all the known metabolic pathways affected. Compared to the wild type, the mucR mutation impacted a diverse set of genes involved in numerous metabolic processes (see Fig. S1 in the supplemental material). While the consequences of each are unclear, similar results were obtained when S. meliloti was analyzed in the bacteroid state during symbiosis to the independent planktonic form (26) (see Fig. S2). Few of the pathways affected by the mucR disruption were not also observed as differentially expressed in the same direction after the establishment of symbiosis (see Fig. S3). This, in addition to the observed repression of gene expression related to nitrogen fixation, suggests that a significant portion of the role of MucR may involve repression of genes in planktonic S. meliloti that are intended for maximal transcription only after symbiosis.

EPS II production in response to population density.Wild-type S. meliloti produces two exopolysaccharides which allow for host-plant invasion: succinoglycan and EPS II. While succinoglycan appears to act as a signal, inducing appropriate responses in the host plant (29), the method by which EPS II functions is more elusive. Recent work by Rinaudi and González shows that EPS II production, specifically the synthesis of the low-molecular-weight fraction, results in organized biofilm formation and attachment at dramatically greater levels than those observed in the absence of this exopolysaccharide (46). In bacteria, biofilms permit the utilization of high population densities for the development of complex social organizations, providing various advantages, from antibiotic resistance to the channeling of nutrients (1, 2, 12). The development of these structures in S. meliloti is coincident with the invasive ability mediated by LMW EPS II. In wild-type S. meliloti, production of all fractions of EPS II requires an active ExpR/Sin quorum-sensing system, paralleling the population-dependent nature of biofilms. In accordance with our real-time PCR expression data as well as previous observations (36) (Fig. 1), induction of expE, expA, and expD required for EPS II production is downstream of increased expression of expG-expC by the ExpR/Sin quorum-sensing system. The transcriptional regulator encoded by expG derepresses these three operons from MucR (4), while evidence presented in the present study shows that expression of expC encodes a glycosyl transferase required for the production of the symbiotically active low-molecular-weight form of EPS II. Mutant strains incapable of LMW EPS II synthesis can regain this capacity by transcriptional supplementation of expC expression in trans to wild-type levels (Fig. 3). The direct mechanism by which increased levels of expC results in decreased molecular weights of EPS II is unclear. However, direct disruption of expC results in the complete termination of EPS II production (23) while basal levels permit only the high-molecular-weight form (Fig. 3). This suggests the possibility that ExpC is required for initiation of EPS II biosynthesis while factors encoded by expE, expA, and expD are responsible for the actual polymerization. Increased levels of ExpC with maintained expression of expE, expA, and expD, as demonstrated in this work, would result in a greater number of exopolysaccharide chains of lower molecular weight, distributed among more points of initiation. By tethering expression of both expG and expC to the same transcript, S. meliloti ensures that the quorum-sensing activation of EPS II biosynthesis through ExpG occurs alongside induced expression levels of the ExpC glycosyl transferase, allowing for the production of the low-molecular-weight fraction. This linkage results in the synchronization of overall EPS II production with the synthesis of its symbiotically active form. These data suggest that the primary utility of this exopolysaccharide is achieved through functions of the low-molecular-weight fraction. Furthermore, the production of LMW EPS II is concurrent with the population density-dependent abolishment of motility, a factor known to conflict with biofilm formation and recently shown to interfere with plant invasion (26). The fact that the synthesis of the necessary exopolysaccharides for biofilm formation is coincident with the termination of interfering behaviors, such as motility, suggests that bacterial aggregation and attachment may be primary functions of EPS II.

The ExpR/Sin quorum-sensing system in S. meliloti derepresses EPS II production across a broad range of population densities (26). However, at early stages of growth previously unexamined (OD₆₀₀ of 0.02), overall expression of the exp gene family closely paralleled that observed in sinI mutants incapable of detecting quorum (Fig. 4A). The disruption of mucR in these cultures restored expression to roughly 50% of full induction, suggesting a role of MucR in tightening the repression of EPS II production at inadequate population densities for its utilization (Fig. 4C). The impact of the mucR mutation attenuates significantly at high population densities due to the dominance of activation by the ExpR/Sin quorum-sensing system. The observation that MucR actively represses EPS II production at extremely low stages of growth supports the argument that the major functions of this exopolysaccharide, potentially bacterial organization and attachment coincident with plant invasion, require sufficient population densities. Furthermore, these data suggest that quorum may be achieved for certain regulatory circuits at earlier stages of growth than previously thought. Bacterial functions, such as EPS II production, have been acknowledged as dependent on the presence of an intact quorum-sensing system but independent of population density (26). However, a reexamination of the definition of low population density has shown this to be imprecise. At the OD₆₀₀ of 0.02 studied in this work, the quantification of viable bacteria by serial dilutions and plating indicates roughly 10⁷ bacteria per ml of liquid culture, suggesting an average of approximately 50 μm between each bacterium. While this population density appears insufficient for quorum in relation to EPS II production, it is not unfeasible that other functions may activate at yet lower thresholds. Caution must be taken in defining regulation as population density independent as dilute cultures under laboratory conditions may still be too populous.

Quorum-sensing-independent functions of MucR.The global examination of potential roles of MucR completed in this study suggests a variety of functions beneficially affecting the establishment of symbiosis. One such example is the observation that the presence of an intact mucR gene sequence results in a modest decrease in the expression of 14 motility-related genes, as similarly observed in a quorum-sensing-deficient strain (5). Gurich and González showed that the presence of flagella interferes with the invasion of nodules and that quorum sensing at a high population density abolishes motility at the transcriptional level (26). These data suggest two mechanisms by which S. meliloti regulates this behavior: mitigation by MucR independent of growth, followed by complete termination by quorum sensing. In addition, transcription of the exo operon is increased, resulting in the previously reported induction of succinoglycan production (see Table S2 in the supplemental material) (9).

The potential for successful symbiosis is further enhanced by MucR through an increase in the biosynthesis of nod factor for the induction of nodule formation (Fig. 6). Expression of the nodABC operon responsible for the production of bacterial nod factor requires induction by the NodD transcriptional regulator in the presence of the plant signal, luteolin. With only a slight increase in expression of nodD (see Table S2 in the supplemental material), MucR significantly enhances induction of nod factor biosynthesis by luteolin, as measured by expression of the nodABC operon (Fig. 5). Although in most cases MucR acts as a repressor, the capacity for this transcriptional regulator to increase expression of genes such as exoY has been demonstrated (9). In addition, features typical of MucR binding sites as described by Becker et al. (9, 10) have been observed upstream of nodD, suggesting the potential for a direct effect. This positive role of MucR on nod factor production is carefully regulated as MucR increases expression of nodD, rather than the nodABC operon directly. This thereby allows for an increase in nod factor production only in response to the detection of plant-produced luteolin, preventing excess biosynthesis in the absence of a host plant. During plant nodulation assays, a sharp advantage in nodule development coincident with an intact mucR is initially observed although this quickly attenuates. This is possibly a consequence of the gradual accumulation of nod factor from excess inoculated S. meliloti. Ultimately, 3 to 4 weeks after inoculation, the numbers of nodules developed and invaded become essentially equivalent between the wild type and the mucR mutant under laboratory conditions. However, in a more unfavorable environment, either due to harsh conditions or lower numbers of bacteria, this difference in luteolin sensitivity may bear greater consequences.

Once symbiosis has been established, differentiation of S. meliloti into the bacteroid state leads to a multitude of changes in gene expression. Interestingly, a disruption of mucR in planktonic S. meliloti results in many of these same effects, including increased expression of multiple operons required for nitrogen fixation and respiration, as well as numerous type IV secretion system and putative transport-related genes (Table 2). While the functions of nitrogen fixation and respiration within the nodule are clear, the purposes of secretion and transport-related gene expression are more elusive. These may contribute to the exchange of plant signals and peptides reported to regulate bacterial growth and other cellular processes during symbiosis (49, 50). A more complete analysis by Kegg Array of all known cellular processes affected by both the transition into the bacteroid state and the disruption of mucR presents remarkable parallels in the impacted pathways (see Fig. S1, S2, and S3 in the supplemental material). Due to the expansiveness of affected functions, a complete understanding of the purpose of each modulation in gene expression requires exhaustive analysis. However, the prevalent similarities in global changes strongly suggest a role of MucR in planktonic bacteria as a suppressor of premature bacteroid behavior. Previous data from our laboratory have shown that within the nodule, mucR expression is repressed roughly 3.3-fold (26). While this appears to be a fairly modest decrease in transcription, the effects of disrupting mucR are more significant in amplitude than those observed with differentiation into the bacteroid state. This suggests the likelihood that a similar decrease in mucR transcription, as opposed to complete disruption, may produce equivalent values.

The ability of the ExpR/Sin system to modulate specific regulatory effects produces both quorum-sensing-dependent and-independent functions of MucR (Fig. 7). By this mechanism, derepression of EPS II production, suppression of premature bacteroid-gene expression, and maximal synthesis of nodules are simultaneously achieved in planktonic S. meliloti.

• Open in new tab
• Download powerpoint

FIG. 7.

Schematic model for the quorum-sensing (QS)-dependent and -independent roles of MucR. At a low population density (OD₆₀₀ of ≤0.02), MucR prevents the unnecessary production of EPS II prior to the quorum required for attachment and biofilm formation. At a high population density (OD₆₀₀ of ≥0.1), AHL-bound ExpR induces expression of the expG-expC operon encoding both the ExpG transcriptional regulator and the ExpC glycosyl transferase. ExpG derepresses operons expE, expA, and expD from MucR, allowing for the biosynthesis of EPS II (4). Concurrently, increased levels of ExpC result in the production of the symbiotically active low-molecular-weight fraction. Throughout all stages of growth prior to symbiosis, MucR increases production of nod factor as well as the symbiotically active exopolysaccharide succinoglycan while repressing premature expression of bacteroid genes and mitigating motility.

Trends in the roles of symbiotic megaplasmids pSymA and pSymB.Wild-type S. meliloti carries three major genetic elements: the chromosome and symbiotic megaplasmids pSymA and pSymB. Throughout this study, clear patterns emerged in the spatial organization of the genes involved in each regulatory network. Prior to symbiosis, at a low population density, expression of motility persists as S. meliloti seek nutrients or host plants with which to establish symbiosis. Maintained expression of expR and sinI allows for the constitutive ability of the bacteria to detect quorum and respond appropriately. Furthermore, transcription of mucR suppresses bacteroid gene expression in this planktonic state. Each of these genes resides on the chromosome of S. meliloti. As the bacterial population density increases, AHL-bound ExpR permits the induction of the exp gene family present on pSymB. Once symbiosis has been established, MucR is suppressed, and nitrogen fixation, type IV secretion, specific respiration, and transport-related gene expression, all on pSymA, are induced. In addition, 90% of the 154 genes most significantly differentially expressed (greater than 10-fold) as a result of a mucR mutation reside on this megaplasmid. These observations suggest key utilization of pSymB as population densities increase and expression from pSymA during the maintenance of symbiosis.

Furthermore, the unique organization of each of these regulatory circuits provides a broad understanding of the sophisticated developmental history of S. meliloti, as well as the evolutionary versatility of the MucR repressor. With the acquisition of pSymB, the ability to abolish motility and produce EPS II in response to population allowed for the development of biofilms and the advantages associated with these communities. However, this required the precise adjustment of the roles of MucR: a repressive capacity on EPS II production at early stages of growth, followed by the abolishment of this effect at sufficient population densities. Additionally, the adoption of pSymA expanded the functions of pSymB. Synthesis of symbiotically active exopolysaccharides and the ability to develop biofilm could now be utilized for plant invasion and subsequent nitrogen fixation. With this new capacity, the additional responsibility of preventing premature symbiotic gene expression was included among the already diverse functions of MucR. The complex roles of this single transcriptional regulator provide for more speculation than insight, but extensive analyses in greater detail may result in a clearer understanding of the means and order of adoption of each megaplasmid, as well as the evolutionary progression to the modern S. meliloti.