Skip to main content
NCBI home page
Microbial Biotechnology logoLink to Microbial Biotechnology
. 2026 Aug 4;19(8):e70424. doi: 10.1111/1751-7915.70424

TolC Is Required for a Mixed‐Linkage β‐Glucan (MLG) Biosynthesis: Engineering Bacteria for MLG Overproduction

L Ruiz‐Sáez 1, P J Pacheco 1, J Peinado 2, J Lloret 2, S Muñoz 1, J Sanjuán 1, D Pérez‐Mendoza 1,✉
PMCID: PMC13438883  PMID: 42552791

ABSTRACT

Mixed‐linkage β‐glucans (MLGs) are emerging as promising biopolymers with significant biotechnological potential due to their unique structural and rheological properties. In rhizobia, MLG biosynthesis is controlled by the second messenger cyclic di‐GMP (c‐di‐GMP) and mediated by the bicistronic operon bgsBA. However, the full composition of the biosynthetic machinery and strategies for enhanced production remain incompletely understood. In this study, we demonstrate that the outer membrane protein TolC is essential for MLG production in Sinorhizobium meliloti . Genetic disruption of tolC abolished MLG synthesis, while its complementation restored production. We propose that TolC functions as part of a tripartite complex with BgsA and BgsB, enabling efficient polymer synthesis and/or export. Furthermore, co‐overexpression of tolC, bgsBA, and a constitutively active diguanylate cyclase (pleD*) yielded an 8‐fold increase of MLG over a control plasmid without tolC, reaching up to ~10 g/L under bioreactor conditions. Additionally, this genetic module enabled de novo MLG production in otherwise non‐producer rhizobial hosts (e.g., Mesorhizobium japonicum), allowing bacterial chassis exchanges and highlighting its portability and potential for synthetic biology applications. Overall, our findings identify TolC as a key component of the MLG biosynthetic machinery and provide a robust platform for the scalable production of this valuable biopolymer.

Keywords: c‐di‐GMP, exopolysaccharides, mixed‐linkage β‐glucan, tolC

TolC is identified as an essential component of the mixed‐linkage β‐glucan (MLG) biosynthetic machinery in Sinorhizobium meliloti , likely enabling polymer export. Co‐expression of tolC, bgsBA, and pleD* significantly enhances MLG production and enables its synthesis in non‐native bacterial hosts.

graphic file with name MBT2-19-e70424-g001.jpg

1. Introduction

Bacterial exopolysaccharides (EPS) have increasingly captured scientific attention due to their significant commercial relevance in industries such as pharmaceuticals, cosmetics, biotechnology, and biomedicine (Freitas et al. 2011). The versatility of bacterial EPS lies in their unique material properties, including solubility, rheological features, viscoelasticity, crosslinking, gelation, water retention, extensibility, ionic strength, and exceptional stability. These characteristics highlight their industrial potential and position them as promising alternatives to chemical polymers due to their biodegradable and non‐toxic nature (More et al. 2014; Moradali and Rehm 2020). In addition, the ease of working with bacteria has made bacterial EPS preferable over their plant and fungal counterparts (Kaur and Dey 2023).

EPS biosynthesis is a complex intracellular process where sugar residues are polymerized and modified before being secreted (Angelin and Kavitha 2020). Three main pathways mediate EPS synthesis: Wzx/Wzy‐dependent, ABC transporter‐dependent, and synthase‐dependent pathways. The latter, most common in Gram‐negative bacteria, employs a membrane‐embedded glycosyl transferase (GT) that facilitates the simultaneous formation and translocation of the polymer across the inner membrane (Whitney and Howell 2013). EPS production and secretion are tightly regulated processes governed by complex signalling networks that integrate cellular and environmental cues. Among the key regulatory mechanisms, the second messenger bis‐(3′–5′)‐cyclic dimeric GMP (cyclic diguanylate, c‐di‐GMP) has emerged as a ubiquitous bacterial signalling molecule and a central activator of EPS biosynthesis, acting at both transcriptional and post‐translational levels (Liang 2015). This second messenger plays a key role in signal transduction pathways that regulate and enable bacterial adaptation to different changing environments, switching between the motile planktonic state and the sedentary lifestyle associated with biofilms (D'Argenio and Miller 2004; Hengge 2009; Römling et al. 2013). First identified by Benziman and colleagues in 1987 as an activator of cellulose synthase (CS) (Ross et al. 1987), c‐di‐GMP is now recognized as a widespread allosteric regulator of synthase complexes that promote the synthesis and secretion of diverse EPS (Pérez‐Mendoza and Sanjuán 2016). Key proteins involved in c‐di‐GMP metabolism are diguanylate cyclases (DGCs) and phosphodiesterases (PDEs). DGCs, featuring a GGDEF domain, synthesize c‐di‐GMP from two GTP molecules, while PDEs, containing EAL or HD‐GYP domains, degrade c‐di‐GMP into 5′‐pGpG or GMP. The cellular c‐di‐GMP level is tightly regulated by the opposing activities of these enzymes, often modulated by their sensory domains (Tal et al. 1998; Römling et al. 2005).

Rhizobia are a diverse group of soil microorganisms that establish nitrogen‐fixing symbioses with leguminous plants and are distributed across several classes of Proteobacteria, predominantly α‐ and β‐proteobacteria. These bacteria exhibit remarkable adaptability, switching between free‐living and symbiotic lifestyles in response to environmental conditions. Bacterial signal transduction serves as a crucial mechanism for generating the necessary physiological, genetic, and cellular adaptive responses to ensure survival and function under different circumstances. Cyclic‐di‐GMP signalling is essential to coordinate these transitions and regulates motility, biofilm formation, and EPS production (Krol et al. 2020). In Sinorhizobium meliloti 8530, c‐di‐GMP modulates the synthesis of several EPSs, including EPS‐I (succinoglycan) and EPS‐II (galactoglucan), both essential for root colonization, nodule invasion, and host signalling. Additionally, c‐di‐GMP promotes the biosynthesis of an arabinose‐containing polysaccharide (APS) and an adhesive polymer of unknown composition that bears similarities to the unipolar polysaccharide (UPP) observed in Agrobacterium tumefaciens (Schäper et al. 2016, 2017; Krol et al. 2020). Moreover, in S. meliloti , c‐di‐GMP promotes the synthesis and secretion of MLG, a linear mixed‐linkage β‐glucan. Unlike plant‐derived MLGs, this bacterial MLG features a unique primary structure with perfect alternation of β‐(1,3) and β‐(1,4) linkages, making it a β‐glucan with promising biotechnological applications (Pérez‐Mendoza et al. 2015). In Rhizobium etli CFN42, c‐di‐GMP has also been shown to regulate the biosynthesis of MLG, highlighting the conservation of this regulatory mechanism within different rhizobial species. This MLG exhibits structural and rheological characteristics that position it as a novel and promising smart material with strong potential for large‐scale production in pharmaceutical, medical, and food‐related applications (Chang et al. 2021; Caseiro et al. 2022; Flores et al. 2026). Its biosynthesis requires an intact quorum‐sensing system (expR, sinI, and sinR), suggesting cross‐talk between cell density signalling and c‐di‐GMP pathways (Pérez‐Mendoza et al. 2015). In addition to the biotech relevance of this polymer, this conservation underscores the important relevance of MLG in the rhizosphere and symbiotic contexts, possibly contributing to surface architecture, biofilm formation, and plant–microbe interactions. Interestingly, in R. etli CFN42, c‐di‐GMP not only promotes the synthesis and secretion of MLG but also cellulose (Pérez‐Mendoza et al. 2022).

Biosynthesis of MLG is mediated by the bicistronic operon bgsBA (Pérez‐Mendoza et al. 2015). Bioinformatic analyses indicate that BgsA is an integral membrane protein with seven predicted transmembrane domains, a cytoplasmic GT domain responsible for glucan polymerization, and a C‐terminal domain capable of binding c‐di‐GMP. Binding of c‐di‐GMP to this C‐terminal region activates MLG synthesis, a mechanism reminiscent of the role of the PilZ domain in regulating cellulose synthase in other bacteria (Pérez‐Mendoza et al. 2017). BgsB encodes a membrane fusion protein (MFP) containing an HlyD‐like domain. Such proteins are commonly associated with T1SSs, acting as periplasmic adaptors that bridge the inner membrane transporter with Outer Membrane Factors (OMF) to facilitate substrate extrusion from the cell (Pimenta et al. 2005; Zgurskaya et al. 2009). Regarding OMFs, TolC stands out as a trimeric outer membrane protein organized in two main domains: a β‐barrel pore in the outer membrane and an α‐helical domain projecting inward, where it interacts with periplasmic adaptors containing HlyD domains (Lee et al. 2012).

Similar to other β‐glucans, such as cellulose and curdlan, whose biosynthetic machineries involve several inner and outer membrane proteins, it would be expected that MLG biosynthesis requires additional proteins besides BgsA and BgsB. The aims of this work were (i) to identify new, yet unknown components of the MLG biosynthesis machinery, and (ii) to engineer bacteria for increased MLG production. We demonstrate that the outer membrane porin TolC is required for MLG production, likely forming with BgsBA a tripartite complex capable of synthesizing and exporting MLG out of the cell. Moreover, its combined overexpression with bgsBA and pleD* enables not only enhanced polymer yields but also de novo synthesis in alternative, naturally non‐producer bacterial chassis.

2. Experimental Procedures

2.1. Bacteria and Culture Conditions

Bacterial strains and plasmids used in this study are listed in Table S1.

Sinorhizobium (Ensifer) meliloti, Mesorhizobium japonicum and Rhizobium etli were routinely grown in Tryptone‐Yeast extract medium supplemented with 3.40 mM CaCl2·2H2O (TY) (Beringer 1974) or in Minimal Medium (MM) (Robertsen et al. 1981). MM contained (per litre): 10 g mannitol, 1.1 g sodium glutamate, 0.3 g KH2PO4, 0.3 g K2HPO4, 0.15 g MgSO4·7H2O, 0.05 g CaCl2·2H2O, 0.006 g FeCl3, and 0.05 g NaCl. Additionally, 1 mL of a 1000× vitamin stock solution (0.2 g L−1 biotin, 0.1 g L−1 thiamine hydrochloride, and 0.1 g L−1 sodium pantothenate) was added.

S. meliloti and M. japonicum strains were incubated at 28°C, whereas R. etli strains were incubated at 30°C. Escherichia coli strains used for cloning and conjugation were routinely grown in LB medium at 37°C. When required, antibiotics and other compounds were added at the following final concentrations: Tetracycline (Tc), 10 μg/mL (5 μg/mL for R. etli ); Kanamycin (Km), 50 μg/mL; Congo Red (CR), 50 μg/mL; Calcofluor White (CF), 200 μg/mL; IPTG, 1 mM. All plasmids used in this study (Table S1) were introduced into rhizobia strains by conjugation using E. coli β2163 or S17.1 donor strains.

2.2. MLG Production and Quantification in Flasks

MLG was quantified by measuring glucose equivalents released after lichenase digestion, following the methodology described by Marchante et al. (2024). Rhizobia were grown in 100 mL Erlenmeyer flasks containing 20 mL of liquid Minimal Medium supplemented with Tc at the appropriate concentration. Cultures were incubated for 3 days with shaking at 170 rpm at 28°C for S. meliloti strains and 30°C for R. etli strains. The initial OD600 was adjusted to 0.05 by resuspending colonies from solid medium.

Cultures with flocs were transferred to 15 mL Falcon tubes and centrifuged at 3739× g for 30 min at room temperature. The supernatant was discarded, and the resulting pellet was washed twice (20 min per wash) with 5 mL of 100 mM sodium phosphate buffer adjust to pH 6.8. The washed flocs were subsequently resuspended in 1 mL of 100 mM sodium phosphate buffer and transferred to 2 mL Eppendorf tubes, for treatment with 6 units of lichenase enzyme (Megazyme, EC Number: 3.2.1.73). Enzymatic digestions were carried out at 40°C for 30 min with constant agitation. After digestion, the tubes were centrifuged at 9290× g for 10 min, and the supernatant was collected for subsequent sugar quantification.

Soluble sugars were quantified using the method described by Masuko et al. (2005). Briefly, 50 μL of the previously obtained supernatant were transferred to a 96‐well flat‐bottom polystyrene microplate (CELLSTAR, Greiner Bio‐One), adding 150 μL of 97% (w/w) sulfuric acid and 30 μL of 5% (v/v) phenol, in that order. The plate was incubated at 90°C for 15 min and then placed in a microplate reader (BioTek) to measure absorbance at 490 nm. Simultaneously, a calibration curve was constructed using different glucose dilutions to calculate glucose‐equivalent concentrations in each sample.

To distinguish extracellular from intracellular MLG, the lichenase digestion protocol described above was modified by introducing sequential enzymatic treatments. Following the initial lichenase digestion, samples were centrifuged at 9290× g for 10 min, and the supernatant was collected. The remaining pellet was washed three times with 100 mM sodium phosphate buffer (pH 6.8), resuspended in 1 mL of the same buffer containing 5 mg mL−1 lysozyme (Roche), and incubated for 5 min at room temperature. Samples were subsequently heated at 100°C for 10 min, centrifuged at 9290× g for 10 min, and the supernatant was collected. The pellet was then washed twice with 100 mM sodium phosphate buffer (pH 6.8), resuspended in 1 mL of the same buffer, and subjected to a second lichenase digestion under the conditions described above. Following centrifugation at 9290× g for 10 min, the resulting supernatant was collected. Glucose equivalents in the supernatants collected after each enzymatic step were quantified as described above.

2.3. Protein Quantification by Bradford Assay

Total protein concentration was determined according to Bradford (Bradford 1976) to normalize the results obtained in sugar quantification. Briefly, cell pellets obtained from 20 mL cultures were resuspended in 300 μL PBS buffer, pH 6.8–7.0. An aliquot of 20 μL from each suspension was diluted 1:1 with sterile distilled water and mixed with 160 μL of Bradford reagent (Bio‐Rad). After 5–10 min of incubation at room temperature, absorbance was measured at 595 nm using a microplate reader. Protein content was calculated using a bovine serum albumin (BSA) standard curve (from 0 to 40 μg/mL), correcting for the dilution factor, and expressed as total mg per pellet.

2.4. Molecular Cloning

The tolC gene from S. meliloti was amplified using primers TolC‐F/TolC‐R (Table S2) with an annealing temperature of 62°C and an extension time of 90 s. The resulting product (1457 bp) was cloned into the pCR2.1‐TOPO vector (Invitrogen) and confirmed by sequencing. The tolC fragment was then digested with BamHI and ligated into the BamHI‐linearized pBBR9091 vector (Table S2), which harbours the bgsBA operon. Insert orientation was confirmed by MluNI digestion. Subsequently, the bgsBA‐tolC fragment was excised from pBBR9091 by XbaI digestion and ligated into the XbaI‐linearized pJBpleD* vector previously treated with 1 μL (5 U) Antarctic Phosphatase (New England Biolabs) according to the manufacturer's instructions to prevent vector self‐ligation. The correct orientation of the insert was confirmed by BamHI digestion, obtaining the plasmid pJBpleD*9091C.

To generate plasmid pJBpleD*lacIq, a 1610 bp fragment containing the lacI q gene was excised from pQE‐80 L by MluNI digestion, purified and ligated with pJBpleD*, which was previously linearized with BsaAI and treated with Antarctic Phosphatase (NEB) as previously mentioned. The ligation mixture was transformed into chemically competent E. coli DH5α cells. Correct plasmid construction was verified by XbaI/EcoRI double digestion, yielding fragments of approximately 6200, 2400, and 1400 bp, consistent with the expected recombinant plasmid.

In‐frame gene deletions were generated in S. meliloti 8530 and IBR505 following the protocol described by Schäfer et al. (1994). Briefly, synthetic DNA fragments containing approximately 750 bp of each gene‐flanking region were cloned into the suicide plasmid pK18mobsacB. Specifically, the exoY deletion construct comprised a 748‐bp upstream region (positions 1,173,540–1,174,287 on pSymB of the S. meliloti 1021 genome) and a 748‐bp downstream fragment (positions 1,172,114–1,172,861). The wgcA deletion construct contained a 747‐bp upstream region (positions 983,879–984,625) and a 747‐bp downstream region (positions 985,874–986,620). The uxs1 construct consisted of a 743‐bp upstream fragment (positions 470,020–470,762) and a 753‐bp downstream fragment (positions 471,393–472,145). In addition, the downstream fragment carried a mutation that affected the next gene of the operon, uxe. The mutation converted the tyrosine codon corresponding to amino acid 14 into an opal stop codon, thereby generating a truncated Uxe protein. All the synthetic DNA fragments were designed with two different unique restriction sites at their 5′ and 3′ ends to facilitate cloning into pK18mobsacB.

Following conjugation, neomycin‐resistant colonies resulting from plasmid integration were subjected to sucrose counter‐selection to isolate double‐crossover recombinants. Since no antibiotic resistance marker was retained in the genome, deletions were generated sequentially in the following order: wgcA, exoY, and uxs1. Candidate ΔwgcA mutants were identified by the loss of the mucoid phenotype associated with EPS II production, whereas ΔexoY mutants were screened by the loss of Calcofluor fluorescence associated with EPS I production. The Δuxs1‐uxe mutants were identified by PCR analysis. All deletions were confirmed by PCR.

2.5. Random Mutagenesis Strategy

To generate random transposon mutants in S. meliloti 8530 and IBR607 strains, the transposon Tn5 (which confers Km/Sm resistance and is carried by plasmid pSUP2021, Table S1) or the transposon TnSAM (which confers Km resistance and is carried by pSAM_Rl, Table S1) were mobilized from E. coli donor strains S17.1 or β2163, respectively, by conjugation following the filter‐mating procedure with a donor:recipient ratio of 1:1 (Demarre et al. 2005). Transposants were cultured on TY broth supplemented with neomycin or kanamycin. For S. meliloti 8530 mutagenesis, plasmid pJBpleD*lacIq (encoding an exogenous diguanylate cyclase from Caulobacter crescentus , Table S1) was transferred from β2163 to the pool of transposon mutants. Transconjugants were selected on TY or MM plates supplemented with CR, IPTG and Tc. In parallel, S. meliloti 8530 strain harbouring pJBpleD*lacIq or pJB3Tc19 plasmid was plated onto CR‐containing medium also supplemented with Tc and IPTG as positive or negative control, respectively. The CR‐negative colonies obtained from both Tn5 and TnSAM mutagenesis were isolated and their transposon insertion sites were determined by random PCR according to Saavedra et al. (2017) or by whole‐genome sequencing performed by Plasmidsaurus using Oxford Nanopore Technology. Different primer pairs were designed and used depending on the transposon (Tn5 or TnSAM‐harbouring colonies) and the inverted repeat region to amplify (IRR or IRL) (Table S2). The amplified PCR products were purified from agarose gels and sequenced using primers Tn5F2 for Tn5 and IRL3 or SAMIRRSEC for TnSAM (Table S2). DNA sequence homology searches were performed with the BLAST (NCBI) to determine the precise positions of the transposon insertions.

2.6. Bioreactor Culture Conditions for MLG Production

S. meliloti cultures were scaled up progressively. Initially, cells were inoculated into 20 mL of TY broth supplemented with Tc and incubated for 24 h at 28°C (OD600 = 1.0). This pre‐culture was transferred to 50 mL of the same medium and incubated for another 24 h under identical conditions. Subsequently, this culture was used to inoculate a 2.5 L bioreactor at a 2% (v/v) concentration. The laboratory‐scale bioreactor contained MM prepared as described in Section 2.1 (Robertsen et al. 1981). The initial pH was adjusted to 6.8–7.0. The total fermentation time was 96 h, with a fed‐batch addition of 0.5 L of the same medium after 72 h to provide additional nutrients.

Dissolved oxygen (DO) levels were maintained above 95% saturation using a cascade control system with two priority parameters: agitation speed (800–1000 rpm) and aeration rate (0.2–0.4 slpm). Throughout the process, the pH was maintained between 6.8 and 7.0 by the automatic addition of 2 M H2SO4 or 2 M NaOH, while the temperature was kept at 28°C. To prevent foaming, Antifoam 204 (Sigma‐Aldrich) was added at a final concentration of 0.01% (v/v).

2.7. Extraction and Purification of MLG

After completion of the 96 h of fermentation, the culture was distributed into 0.5 L bottles and centrifuged at 14,000× g for 30 min. The supernatant was filtered to recover any released flocs. Pellets were then resuspended in Milli‐Q water and centrifuged again under the same conditions to ensure the complete removal of residual medium. This washing step was repeated until a single MLG‐containing pellet was obtained. The resulting pellet (MLG plus cells) was boiled for 10 min at 100°C and washed with Milli‐Q water. This washing step was repeated three times. The purified MLG was then transferred into a dialysis membrane (6–8 kDa molecular weight cut‐off) and dialyzed against 3 L of Milli‐Q water for 24–48 h, with the water being replaced after the first 24 h. Finally, the dialyzed MLG was frozen at −80°C and lyophilized for 24 h. The resulting dry MLG was weighed to determine the yield.

3. Results

3.1. Effect of Other c‐di‐GMP–Regulated EPS on MLG Production in Rhizobia

Intracellular c‐di‐GMP levels regulate the synthesis of multiple EPS in S. meliloti 8530 (Krol et al. 2020). With the aim of maximizing MLG production, we constructed a triple mutant defective in the synthesis of three c‐di‐GMP–regulated EPS in this bacterium: EPS II, EPS I, and APS. To this end, in‐frame deletions were generated in the following genes: wgcA (EPS II), exoY (EPS I), and uxs1–uxe (APS). These mutations were introduced sequentially and confirmed both genetically by PCR and microbiologically by the absence of the phenotypes associated with each EPS. Once this triple mutant (IBR606) had been constructed (see Experimental Procedures), the MLG yield of this mutant was compared to the wild type (Figure 1A). In line with previous reports, at physiological c‐di‐GMP levels MLG production in the wild type remained below the detection limit, and no relevant increase relative to the wild‐type strain was observed in the IBR606 background (Figure 1A). As expected, the presence of plasmid pJBpleD*, which increases intracellular c‐di‐GMP levels through the expression of a constitutively active diguanylate cyclase in S. meliloti (Pérez‐Mendoza et al. 2015), boosted MLG production in IBR606, which was slightly higher than in the wild type, although this difference was not statistically significant (Figure 1A). These results indicate that the ability of S. meliloti to synthesize other c‐di‐GMP‐regulated EPS, such as EPS I, EPS II and APS, does not have a major impact on MLG yield.

FIGURE 1.

FIGURE 1

Open in a new tab

Effect of other c‐di‐GMP‐regulated EPS on MLG production. Quantification of MLG in (A) Sinorhizobium meliloti 8530 (WT) and the triple mutant IBR606 (ΔwgcA ΔexoY Δuxs1‐uxe) carrying either the control plasmid pJB3Tc19 or pJBpleD*, and (B) Rhizobium etli CFN42 (WT) and the ΔcelAB mutant carrying the indicated plasmids. Data were normalized to total protein content and represent the mean ± SD of independent experiments (n = 8 for S. meliloti ; n = 4 for R. etli ). In panel A, ns = no statistically significant differences according to a two‐tailed Welch's unpaired t‐test. In panel B, statistical significance was determined by one‐way ANOVA followed by Tukey's multiple comparisons test (**p < 0.01; ***p < 0.001).

A similar analysis was performed in R. etli CFN42, which is another MLG‐producer where c‐di‐GMP activates the production of both MLG and bacterial cellulose, two β‐glucans that utilize UDP‐glucose as a substrate (Pérez‐Mendoza et al. 2022). MLG production was compared between the wild type strain and a cellulose‐deficient mutant (ΔcelAB). In contrast to what was observed in S. meliloti , under elevated c‐di‐GMP conditions (pJBpleD*) the ΔcelAB mutant exhibited a significant ~3‐fold increase in MLG production relative to the wild type (Figure 1B). These results indicate that blocking cellulose production is an effective strategy to increase MLG yield in the R. etli genetic background. However, the S. meliloti 8530 background remains a superior bacterial chassis for MLG production compared to R. etli ΔcelAB (~50 mg glucose equivalents/mg protein in the former vs. ~15 mg glucose equivalents/mg protein in the latter; Figure 1).

3.2. Identification of Genes Potentially Involved in MLG Synthesis and Secretion in Sinorhizobium meliloti

To further optimize MLG production in S. meliloti , we investigated whether additional, yet unidentified genes contribute to MLG synthesis and secretion. Other β‐glucans, such as cellulose and curdlan, rely on auxiliary components for full functionality (Stasinopoulos et al. 1999; Whitney and Howell 2013). To this end, we employed a random transposon mutagenesis approach followed by screening on solid medium supplemented with CR (Congo Red), which binds to α‐D‐glucopyranosyl units, neutral or basic polysaccharides, and certain proteins, and has been shown to interact positively with MLG (Pérez‐Mendoza et al. 2015). The goal was then to distinguish between MLG‐producing (red‐stained) and non‐producing (white) colonies on CR plates after the mutagenesis. To maximize screening success, we applied this approach across different variants: (i) two different S. meliloti 8530 genetic backgrounds (wild‐type or a EPS II, EPS I, and APS triple mutant), (ii) two different transposons (Tn5 or TnSAM), and (iii) two different sources of c‐di‐GMP: via heterologous overexpression of pleD* or activation of the cognate DGC (diguanylate cyclase) BgrR (Baena et al. 2019).

The first approach entailed a random transposon mutagenesis of the S. meliloti 8530 wild‐type strain by introducing the Tn5 transposon using the donor strain E. coli S17.1 (pSUP2021) (Simon et al. 1983) or the TnSAM transposon using β2163 pSAM_Rl (Perry and Yost 2014). Approximately 8.5 × 104 Tn5 and 1.6 × 105 TnSAM transposon insertion mutants were obtained, respectively. Then, pleD* was transferred to each pool of transposants by conjugation from the β2163 pJBpleD*lacIq donor strain. Transconjugants were selected on TY or MM plates supplemented with CR, Tc and IPTG. In parallel, the non‐mutagenized S. meliloti 8530 wild‐type strain harbouring either pJBpleD*lacIq or pJB3Tc19 plasmid was plated onto the same selective media as positive and negative controls, respectively. All non‐mutagenized colonies appeared red (pJBpleD*lacIq) or white (pJB3Tc19), respectively. In contrast, some transposant colonies (< 1%) of S. meliloti 8530 mutagenized with Tn5 or TnSAM exhibited a CR‐negative phenotype (white). Eight of these colonies, 3 with Tn5 and 5 with TnSAM, were subjected to further investigation to identify the genomic sites of transposon insertion (Table 1). All transposon insertions were found interrupting genes previously described to be involved in MLG production. Two insertions were mapped to the expR gene (Smc03896), a transcriptional regulator required for the expression of the MLG biosynthetic operon bgsBA (Pérez‐Mendoza et al. 2015), while 4 and 2 insertions were located within the bgsB and bgsA synthesis genes, respectively (Table 1).

TABLE 1.

Mapping transposon insertions.

Genetic background c‐di‐GMP source Transposon used Targeted gene Number and location of the insertions a
Sm 8530 pleD⁎ Tn5 bgsB (Smb20390) 2 (638, 800)
expR (Smc03896) 1 (−6)
TnSAM bgsB (Smb20390) 2 (89, 619)
bgsA (Smb20391) 2 (375, 1123)
expR (Smc03896) 1 (67)
Sm IBR607 BgrR TnSAM bgsB (Smb20390) 3 (943, 1139, ND)
bgsA (Smb20391) 4 (1165, 1491, 1897, ND)
expR (Smc03896) 4 (−1, 9, 368, 655)
Open in a new tab

Note: The position of the insertions was located after random PCR on genomic DNA from CR− colonies harbouring transposon Tn5 or TnSAM.

a

Relative positions to the corresponding ORF are indicated in brackets. ND: exact position not determined.

After failing to identify new genes involved in MLG production using the initial mutagenesis strategy, we decided to repeat the approach in an alternative S. meliloti 8530 genetic background. On one hand, we aimed to minimize potential interference from the synthesis of other EPS (using a triple mutant lacking EPS II, EPS I, and APS), and on the other, to modify c‐di‐GMP‐dependent activation by using a native DGC (BgrR) specifically dedicated to MLG production. For this purpose, we used mutant IBR505, which lacks the bgrV gene, which encodes a repressor of the DGC BgrR, and therefore this mutant displays derepressed MLG synthesis (Baena et al. 2019). Sequential allelic exchange was then used to introduce deletions in wgcA (EPS II), exoY (EPS I), and uxs1–uxe (APS) into this genetic background, yielding the quadruple mutant IBR607 (defective in EPS II, EPS I, APS, and bgrV). IBR607 was subjected to mutagenesis with the mariner transposon TnSAM, generating approximately 6000 transposants. Following screening, 11 colonies showing a CR‐negative phenotype were selected, and the respective transposon insertion sites were determined. As summarized in Table 1, the transposon was found to disrupt again any of the genes already known to be required for MLG production: expR, bgsB, and bgsA.

In summary, both approaches repeatedly recovered mutations in previously known genes, suggesting two possibilities: (i) no additional genes participate in MLG synthesis and secretion, or (ii) such genes do exist but are essential for S. meliloti under our experimental conditions.

3.3. Role of TolC in MLG Production

Other β‐glucans, such as cellulose and curdlan, rely at a minimum on OMFs (Outer Membrane Factor) to mediate EPS extrusion into the extracellular milieu. In this context, TolC has been described as a highly versatile porin involved in several transport systems responsible for exporting diverse molecules and environmental stressors, including a broad range of antimicrobial agents (Wright et al. 2025). Notably, in S. meliloti 2011, TolC is essential for the proper synthesis and secretion of the two symbiotically relevant EPS produced by this strain, namely succinoglycan (EPS‐I) and galactoglucan (EPS‐II). Furthermore, an insertional mutation in tolC in this strain severely compromises resistance to antimicrobials and increases susceptibility to osmotic and oxidative stresses (Cosme et al. 2008). Taken together, these observations led us to ask: (i) whether, as described for succinoglycan (EPS‐I) and galactoglucan (EPS‐II) in S. meliloti strain 2011, TolC might also play a relevant role in MLG synthesis and/or secretion, and (ii) whether the failure to recover tolC in our random mutagenesis screens could be explained by the lack of viability of tolC mutants on CR–supplemented medium.

To address this, we first examined the growth capacity of tolC mutant of S. meliloti 2011 strain on our screening medium supplemented with CR. Unlike their parental wild‐type strain, this mutant showed no appreciable growth on MM or TY plates containing 50 μg/mL CR under our growth conditions (data not shown). These results suggest that TolC is indeed essential under our screening conditions and prompted us to assess MLG production in this mutant. To this end, we introduced plasmid pJBpleD*9091 (Figure 2A), which carries pleD* and bgsBA genes under the control of a constitutive lac promoter (Pérez‐Mendoza et al. 2015), into the 2011 tolC mutant background. Unlike S. meliloti 8530 (expR +), strain 2011 lacks the transcriptional regulator ExpR and thus require the expression of bgsBA from a heterologous promoter for MLG production (Pérez‐Mendoza et al. 2015). As shown in Figure 2B, plasmid pJBpleD*9091 allowed strains 2011 to produce substantial amounts of MLG. In contrast, the tolC mutation in S. meliloti 2011 genetic background resulted in a complete loss of MLG production (Figure 2B). Furthermore, no intracellular accumulation of MLG was detected in the tolC mutant after lysozyme‐mediated cell lysis, with an indistinguishable MLG production in the presence or absence of pJBpleD*9091, arguing against a model in which MLG synthesis proceeds independently of TolC‐mediated export (Figure S1). To corroborate the requirement of TolC for MLG production, we constructed a new plasmid based on pJBpleD*9091 (bgsBA and pleD*), in which the tolC gene was additionally introduced into the same transcriptional unit under the control of the lac promoter, yielding pJBpleD*9091C (bgsBA, pleD* and tolC, Figure 2A). Complementation of 2011 tolC mutant with pJBpleD*9091C plasmid restored MLG production. These results demonstrate that TolC is required for efficient MLG production by S. meliloti .

FIGURE 2.

FIGURE 2

Open in a new tab

TolC is required for MLG production. (A) Plasmid structure of pJBpleD*9091 and its derivative pJBpleD*9091C overexpressing tolC, constructed in this work. (B) Quantification of MLG production in S. meliloti 2011 and their tolC mutant derivative. MLG levels were measured in the wild‐type strains and their corresponding tolC mutant carrying the pJBpleD*9091 plasmid (bgsBA and pleD*) and pJBpleD*9091C (bgsBA, pleD* and tolC). Data were normalized to total protein content and represent the mean ± SD from five independent experiments. Statistical significance between strains was determined using a two‐tailed Welch's unpaired t‐test: *p < 0.05.

3.4. TolC Overexpression Enhances MLG Production

The above complementation experiments with pJBpleD*9091C plasmid not only restored MLG production in the tolC mutant but also resulted in substantial increases over the wild‐type levels (Figure 2B). From a biotechnological perspective, this finding is highly relevant for maximizing MLG yield. To confirm the enhancing effect of tolC overexpression on MLG production, we measured MLG levels in the S. meliloti 8530 ΔbgsBA background carrying the pJB3Tc19, the pJBpleD*9091 or the pJBpleD*9091C plasmids (Figure 3A). As shown in Figure 3A, multicopy tolC expression yielded a significant ~25% increase in MLG production—the highest levels measured in our laboratory to date, exceeding 250 mg glucose equivalents per mg of protein.

FIGURE 3.

FIGURE 3

Open in a new tab

TolC overexpression increases MLG yields in S. meliloti . (A) MLG production was quantified in the S. meliloti 8530 ΔbgsBA background harbouring pJB3Tc19, pJBpleD*9091, or pJBpleD*9091C plasmids. Data were normalized to total protein content and represent the mean ± SD from five independent experiments (n = 5). Statistical significance between pJBpleD*9091 and pJBpleD*9091C was determined using a two‐tailed Welch's unpaired t‐test (*p < 0.05). (B) MLG yields obtained under biotechnological fermentation conditions (2.5‐L laboratory bioreactor) with the ΔbgsBA background with native TolC levels (pJBpleD*9091) or overexpressing tolC (pJBpleD*9091C). Statistical significance between pJBpleD*9091 and pJBpleD*9091C was determined using a two‐tailed Welch's unpaired t‐test (**p < 0.01).

Having demonstrated the enhancing effect of tolC overexpression at flask scale, we evaluated its impact on MLG production at a larger, more controlled scale using a 2.5‐L laboratory bioreactor. Throughout the process, dissolved oxygen was maintained above 95% saturation, pH was held stable at 6.8–7.0, and temperature was controlled at 28°C, with agitation and aeration automatically adjusted to sustain these parameters. After a 96‐h fermentation, MLG was isolated and purified following the methodology described previously (Pérez‐Mendoza et al. 2015). Under these conditions, we compared MLG production in the S. meliloti 8530 ΔbgsBA background hosting pJBpleD*9091C (containing tolC) or the control plasmid pJBpleD*9091. Consistent with the indirect quantification of glucose equivalents (Figure 3A), tolC overexpression resulted in nearly an 8‐fold increase over the control (8.04 ± 0.25), achieving up to ~10 g/L of MLG (Figure 3B). In summary, identifying the essential role of TolC and leveraging its overexpression has led to outstanding yields, paving the way for the future biotechnological exploitation of this promising biopolymer.

3.5. Exploring MLG Production Across Distinct Genetic Background

The excellent MLG production levels obtained with the plasmid carrying the biosynthetic genes bgsBA, the c‐di‐GMP source (pleD*), and the S. meliloti porin tolC within a single transcriptional unit (pJBpleD*9091C; Figure 2A), raised the question of whether this plasmid could be used to engineer alternative bacterial chassis to produce MLG. To address this, pJBpleD*9091C was introduced into two distinct backgrounds: (i) a R. etli derivative unable to produce cellulose and MLG (ΔcelAB ΔbgsA, Pérez‐Mendoza et al. 2022), and (ii) Mesorhizobium japonicum MAFF303099, a rhizobium that naturally lacks the bgsBA biosynthetic operon.

First, the resulting transconjugants were grown on different media supplemented with CF and compared to their corresponding negative controls carrying the empty vector pJB3Tc19 (Figure 4A). CF binds to various pure and mixed‐linkage β‐glucans, including MLG and cellulose, and exhibits a characteristic blue‐green fluorescence under long‐wavelength UV light (Pérez‐Mendoza et al. 2014). As shown in Figure 4A, the presence of pJBpleD*9091C induced clear fluorescence under UV illumination in both genetic backgrounds. Subsequent quantification in liquid MM medium confirmed that the ΔcelAB ΔbgsA double mutant of R. etli and M. japonicum MAFF303099 are both capable of producing MLG in presence of pJBpleD*9091C, in contrast to their respective controls with the empty vector (Figure 4B). From a biotechnological standpoint, these results open the possibility of exploring additional bacterial chassis for MLG overproduction, not only to maximize yield but also to exploit metabolic flexibility and optimize production costs.

FIGURE 4.

FIGURE 4

Open in a new tab

MLG production across distinct bacterial genetic backgrounds. MLG production in Rhizobium etli ΔcelAB ΔbgsA and Mesorhizobium japonicum MAFF 303099 strains carrying pJBpleD*9091C or the empty vector pJB3Tc19. (A) Calcofluor phenotype of R. etli ΔcelAB ΔbgsA and M. japonicum MAFF 303099 strains hosting pJB3Tc19 (left side of the plates) or pJBpleD*9091C (right side of the plates). Strains were grown on minimal medium supplemented with Calcofluor (200 μg mL−1) and photographed under UV illumination after 72 h. (B) Quantification of MLG in liquid cultures of both bacterial chassis containing pJB3Tc19 or pJBpleD*9091C. Data were normalized to total protein content and represent the mean ± SD from 3 independent experiments. Statistical significance between pJB3Tc19 and pJBpleD*9091C was determined using a two‐tailed Welch's unpaired t‐test (**p < 0.01; *p < 0.05).

4. Discussion

The second messenger c‐di‐GMP has been proposed in recent years as an important regulatory molecule controlling the production of bacterial EPS. This regulation can reach high levels of complexity in some bacteria, such as S. meliloti , where c‐di‐GMP can influence the biosynthesis of at least five different EPS (Krol et al. 2020). How this second messenger can regulate the biosynthesis of multiple EPS within the same bacterium remains a difficult question to answer. This question lies at the heart of the ongoing discussion about how c‐di‐GMP can simultaneously control a wide variety of processes in a given bacterium, for example, motility, biofilm formation, cell cycle progression, etc. Recent studies establish that such high specificity and flexibility arise from the combination of local and global modes of c‐di‐GMP signalling within complex regulatory networks (Junkermeier and Hengge 2023). On EPS regulation, it could be explained by signalling systems that activate specific DGCs, which synthesize c‐di‐GMP in proximity to the biosynthetic complex involved in the synthesis of a given polysaccharide, thereby conferring spatial specificity to the system. From a biotechnological perspective, then, the situation could be completely different when a heterologous DGC (e.g., pleD*) is artificially overexpressed with the aim of broadly raising the intracellular levels of c‐di‐GMP for increasing the production of a specific biopolymer of interest and no others (Pérez‐Mendoza et al. 2026). With this in mind, and with the biotechnological goal of maximizing MLG production, we decided to investigate the potentially beneficial impact of funnelling metabolic flux toward MLG by blocking the synthesis of other c‐di‐GMP–regulated EPS in S. meliloti . However, the results in S. meliloti 8530 suggest that MLG biosynthesis is largely independent of other c‐di‐GMP regulated EPS pathways, despite the high metabolic cost of producing multiple polysaccharides simultaneously. The absence of a significant increase in MLG production in the triple mutant IBR606 (defective in EPS II, EPS I and APS) suggests that the biosynthetic flux toward MLG is sufficiently robust to accommodate normal resource allocation even in the presence of other active polysaccharide pathways. This finding aligns with previous reports indicating that MLG synthesis is regulated primarily by c‐di‐GMP rather than by competition for precursors among co‐produced EPS (Pérez‐Mendoza et al. 2014; Krol et al. 2020).

In stark contrast, in R. etli CFN42, MLG production appears to compete with bacterial cellulose biosynthesis. This effect can be attributed to the structural and metabolic similarities between cellulose and MLG, both β‐glucans assembled by homologous glycosyltransferases from the same precursor (UDP‐glucose) and both allosterically activated by c‐di‐GMP (Pérez‐Mendoza et al. 2017, 2022). Indeed, R. etli CFN42 appears to prioritize cellulose synthesis when c‐di‐GMP levels are high, likely due to the critical role of cellulose in biofilm formation and host plant colonization, which identified cellulose as the primary β‐glucan contributing to these processes (Pérez‐Mendoza et al. 2022). Only when cellulose production is abolished (ΔcelAB mutant) does MLG become overproduced in this genetic background, highlighting a hierarchy in β‐glucan biosynthesis and suggesting that cellulose exerts a dominant pull on the shared metabolic precursors and regulatory signals. In any case, considering the MLG production levels achieved in both species, the genetic chassis of S. meliloti 8530 remains a superior platform for the hyperproduction of this c‐di‐GMP–dependent biopolymer. With the aim of further enhancing MLG production in this strain, we set out to identify new genetic factors that might be involved in MLG biosynthesis. The mutagenesis approaches used to identify additional genes involved in MLG production consistently yielded hits in bgsA, bgsB, and expR, supporting the idea that these genes constitute the core machinery for MLG synthesis and regulation. The absence of other candidates suggests either that the minimal MLG system relies primarily on these genes, or that essential components could not be recovered due to lethality or pleiotropic effects on the selective media used in this screening.

Among such essential components, TolC emerged as a compelling candidate. TolC is a well‐characterized outer membrane protein and plays a critical role in T1SSs and multidrug efflux pumps, which are essential for the export of diverse metabolites and macromolecules (Koronakis et al. 1997; Nikaido 2011). This functional versatility further supports its potential involvement in the MLG biosynthetic machinery. Indeed, TolC homologues are generally required for the secretion of various EPS in Gram‐negative bacteria, including S. meliloti (Cosme et al. 2008; Bina et al. 2023). This observation is particularly relevant given that different studies highlight the involvement of TolC‐like proteins in c‐di‐GMP–regulated biological processes, such as biofilm formation, through the secretion of different adhesins via T1SSs (Monds et al. 2007; Pérez‐Mendoza et al. 2011). In this context, our results show that tolC disruption in S. meliloti 2011 completely abolishes detectable MLG levels, and that its production was restored when tolC was provided in trans. The lack of detectable intracellular MLG accumulation in the 2011 tolC pJBpleD9091 strain, suggests that TolC is not merely required for outer membrane export but is functionally linked to productive MLG biosynthesis, consistent with the coupled synthesis–secretion mechanism proposed for synthase‐dependent exopolysaccharide secretion systems (Whitney and Howell 2013). This intriguing result raises the question of whether the requirement for an intact tolC copy stems from the direct involvement of this porin in extruding the β‐glucan chain through the outer membrane to the cell exterior forming a tripartite complex together with BgsB (HlyD) and BgsA (GT). Similar tripartite secretion systems have been reported for cyanobacterial extracellular cellulose production (Maeda et al. 2018; An et al. 2025). Furthermore, regarding other EPS, it seems reasonable to assume that MLG secretion requires an additional outer membrane component that interacts with BgsB to form a continuous channel through which MLG can be extruded outside the cell. Similar outer membrane porins have been described for other EPS, such as BcsC for cellulose, PgaA for poly‐β‐1,6‐N‐acetyl‐D‐glucosamine (pNAG), or AlgE for alginate (Tan et al. 2014; Wang et al. 2016; Acheson et al. 2019). All of them are outer membrane porins characterized by a 16‐ to 18‐stranded transmembrane β‐barrel architecture, forming channels through which their respective EPS are secreted to the extracellular milieu. The pore diameter varies depending on the porin but has been estimated to be around 15 Å diameter within the membrane plane in some cases, such as in BcsC (Acheson et al. 2019). In turn, TolC proteins consist of homotrimeric 12‐stranded β‐barrel structures (four β‐strands from each protomer), forming a channel–tunnel with an average accessible interior diameter of about 19.8 Å (30 Å backbone‐to‐backbone) spanning the outer membrane and most of the tunnel (Koronakis et al. 2004). This suggests a degree of lumen compatibility that could potentially accommodate the extrusion of a β‐glucan such as MLG. However, we also cannot rule out that TolC's role in MLG production is indirect, for example, through the secretion of another macromolecule necessary for the proper assembly of MLG outside the cell. In this sense, TolC is, for example, required for the secretion of ExpE1, a critical factor for efficient EPS‐II synthesis or secretion in S. meliloti (Cosme et al. 2008). There is little doubt that future experiments capable of confirming or refuting the direct involvement of TolC in the export of certain EPS, such as MLG, could represent a significant step forward in our understanding of this broadly conserved family of bacterial proteins.

Whether directly or indirectly, our results undoubtedly demonstrate, not only that TolC is essential for MLG production in S. meliloti , but also that its heterologous overexpression, together with the biosynthetic genes (bgsBA) and the source of its allosteric activation by c‐di‐GMP (pleD*), greatly enhances MLG production. This was confirmed both with an indirect quantification by an approximately 25% increase in free glucose equivalents released after MLG lichenase digestion, and—more importantly—directly by an almost eightfold increase in the amount of MLG obtained under biotechnological fermentation conditions. The successful “bench to market” transfer of knowledge of a novel bacterial biopolymer like MLG critically depends on our capacity to scale up its production from the milligram quantities required for initial characterization to the amounts necessary for comprehensive evaluation of its biotechnological potential. This scale‐up must be rationally carried out, prioritizing cost‐effective and sustainable production conditions. Achieving up to 10 g/L of MLG through the specific genetic modifications described here under bioreactor conditions provides a robust foundation for future optimization of the physicochemical characteristics of the scale up fermentation process (e.g., carbon and nitrogen sources, aeration and temperature).

Additionally, the construction in this work of the plasmid pJBpleD*9091C, which carries all the known genetic factors required for MLG production, has not only enabled an increase in yield but also opens the possibility of using diverse natively non‐producing bacterial chassis that may be more suitable for MLG biosynthesis. The use of alternative microbial platforms is crucial for coupling the synthesis of this biopolymer to other industrial by‐products, enabling its generation from sustainable carbon and nitrogen sources that could reduce operational expenses and promote a circular economy. Future studies will evaluate this promising biotechnological approach by employing bacterial hosts beyond rhizobia. The goal is to leverage their metabolic flexibility to minimize production costs by utilizing diverse low‐cost and renewable industrial by‐products as feedstock.

5. Conclusions

This study identifies TolC as an essential factor for MLG production in S. meliloti , expanding the current understanding of the molecular machinery involved in bacterial β‐glucan biosynthesis. Our results support a model in which TolC participates in a tripartite complex with BgsA and BgsB, likely facilitating polymer export across the outer membrane. Importantly, overexpression of tolC in combination with bgsBA and pleD⁎ significantly enhances MLG production, enabling yields compatible with industrial‐scale applications. Moreover, the successful transfer of this genetic module to non‐native hosts demonstrates its versatility and opens new avenues for engineering alternative microbial platforms for MLG production. These findings provide a solid foundation for future optimization strategies aimed at sustainable and cost‐effective biopolymer manufacturing.

Author Contributions

P. J. Pacheco: investigation, methodology, writing – original draft. L. Ruiz‐Sáez: investigation, writing – original draft, methodology. J. Sanjuán: conceptualization, funding acquisition, writing – review and editing, project administration. J. Peinado: investigation, methodology. D. Pérez‐Mendoza: conceptualization, funding acquisition, writing – original draft, project administration, supervision. J. Lloret: investigation, methodology, writing – original draft. S. Muñoz: investigation, methodology.

Funding

This work was supported by Agencia Estatal de Investigación (Grants TED2021‐129640B‐I00, BIO2017‐83533‐P, PID2022‐140168NB‐I00).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Bacterial strains and plasmids used in this study.

Table S2: Oligonucleotides used in this work.

Figure S1: Extracellular and intracellular mixed‐linkage glucan (MLG) quantification in Sinorhizobium meliloti tolC mutant strains. Upper panel, schematic representation of the sequential enzymatic digestion protocol. Following lichenase digestion, extracellular MLG was quantified by measuring the sugars released into the supernatant (1, +lich). The remaining cell pellet was washed, lysed by lysozyme treatment followed by boiling, and the soluble sugars released were quantified (2; +lich +lys). Finally, the remaining cellular debris, including putative insoluble polysaccharides, was washed and subjected to a second lichenase digestion to release residual cell‐associated MLG (3, +lich +lys +lich). Lower panel, released sugars were quantified as glucose equivalents in three different tolC mutant strains under the indicated experimental conditions. Data were normalized to total protein content represent the mean ± SD from 4 independent experiments. Statistical analysis was performed using an ordinary two‐way ANOVA followed by Tukey's multiple‐comparison test. Different letters indicate significant differences among strains within each treatment (p < 0.05).

MBT2-19-e70424-s001.docx (235.5KB, docx)

Acknowledgements

We would like to thank Miguel Redondo Nieto (UAM) for his invaluable assistance with the genomic localization of transposon sequences. This work is part of the project TED2021‐129640B‐I00 funded by MICIU/AEI/10.13039/501100011033 and by European Union NextGenerationEU/PRTR, BIO2017‐83533‐P funded by MICIU/AEI/10.13039/501100011033 and by ERDF A way of making Europe and PID2022‐140168NB‐I00 funded by MICIU/AEI/10.13039/501100011033.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  1. Acheson, J. F. , Derewenda Z. S., and Zimmer J.. 2019. “Architecture of the Cellulose Synthase Outer Membrane Channel and Its Association With the Periplasmic TPR Domain.” Structure 27: 1855–1861.e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. An, X. , Ramírez V., and Pauly M.. 2025. “Cellulose Synthesis in Cyanobacteria: Shared Pathways and Distinct Features With Bacteria and Plants.” Plants (Basel, Switzerland) 14: 2655. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Angelin, J. , and Kavitha M.. 2020. “Exopolysaccharides From Probiotic Bacteria and Their Health Potential.” International Journal of Biological Macromolecules 162: 853–865. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Baena, I. , Perez‐Mendoza D., Sauviac L., et al. 2019. “A Partner‐Switching System Controls Activation of Mixed‐Linkage Beta‐Glucan Synthesis by c‐di‐GMP in Sinorhizobium meliloti .” Environmental Microbiology 21: 13. [DOI] [PubMed] [Google Scholar]
  5. Beringer, J. E. 1974. “R Factor Transfer in Rhizobium leguminosarum .” Journal of General Microbiology 84: 188–198. [DOI] [PubMed] [Google Scholar]
  6. Bina, X. R. , Weng Y., Budnick J., Allen M. E. V., and Bina J. E.. 2023. “ Klebsiella pneumoniae TolC Contributes to Antimicrobial Resistance, Exopolysaccharide Production, and Virulence.” Infection and Immunity 91: e0030323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bradford, M. M. 1976. “A Rapid and Sensitive Method for the Quantitation of Microgram Quantities of Protein Utilizing the Principle of Protein‐Dye Binding.” Analytical Biochemistry 72: 248–254. [DOI] [PubMed] [Google Scholar]
  8. Caseiro, C. , Dias J. N. R., de Andrade Fontes C. M. G., and Bule P.. 2022. “From Cancer Therapy to Winemaking: The Molecular Structure and Applications of β‐Glucans and β‐1, 3‐Glucanases.” 23: 3156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chang, S.‐C. , Saldivar R. K., Liang P.‐H., and Hsieh Y. S. Y.. 2021. “Structures, Biosynthesis, and Physiological Functions of (1,3;1,4)‐β‐d‐Glucans.” Cells 10: 510. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cosme, A. M. , Becker A., Santos M. R., Sharypova L. A., Santos P. M., and Moreira L. M.. 2008. “The Outer Membrane Protein TolC From Sinorhizobium meliloti Affects Protein Secretion, Polysaccharide Biosynthesis, Antimicrobial Resistance, and Symbiosis.” Molecular Plant‐Microbe Interactions 21: 947–957. [DOI] [PubMed] [Google Scholar]
  11. D'Argenio, D. A. , and Miller S. I.. 2004. “Cyclic di‐GMP as a Bacterial Second Messenger.” Microbiology 150: 2497–2502. [DOI] [PubMed] [Google Scholar]
  12. Demarre, G. , Guerout A. M., Matsumoto‐Mashimo C., Rowe‐Magnus D. A., Marliere P., and Mazel D.. 2005. “A New Family of Mobilizable Suicide Plasmids Based on Broad Host Range R388 Plasmid (IncW) and RP4 Plasmid (IncPalpha) Conjugative Machineries and Their Cognate Escherichia coli Host Strains.” Research in Microbiology 156: 245–255. [DOI] [PubMed] [Google Scholar]
  13. Flores, A. , Zaera J., Ruiz‐Saez L., et al. 2026. “Crystal Structure of Alternating Mixed‐Linkage β‐Glucan by X‐Ray Diffraction.” Preprint, SSRN. 10.2139/ssrn.6170295. [DOI]
  14. Freitas, F. , Alves V. D., and Reis M. A.. 2011. “Advances in Bacterial Exopolysaccharides: From Production to Biotechnological Applications.” Trends in Biotechnology 29: 388–398. [DOI] [PubMed] [Google Scholar]
  15. Hengge, R. 2009. “Principles of c‐di‐GMP Signalling in Bacteria.” Nature Reviews Microbiology 7: 263–273. [DOI] [PubMed] [Google Scholar]
  16. Junkermeier, E. H. , and Hengge R.. 2023. “Local Signaling Enhances Output Specificity of Bacterial c‐di‐GMP Signaling Networks.” Microlife 4: uqad026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kaur, N. , and Dey P.. 2023. “Bacterial Exopolysaccharides as Emerging Bioactive Macromolecules: From Fundamentals to Applications.” Research in Microbiology 174: 104024. [DOI] [PubMed] [Google Scholar]
  18. Koronakis, V. , Eswaran J., and Hughes C.. 2004. “Structure and Function of TolC: The Bacterial Exit Duct for Proteins and Drugs.” Annual Review of Biochemistry 73: 467–489. [DOI] [PubMed] [Google Scholar]
  19. Koronakis, V. , Li J., Koronakis E., and Stauffer K.. 1997. “Structure of TolC, the Outer Membrane Component of the Bacterial Type I Efflux System, Derived From Two‐Dimensional Crystals.” Molecular Microbiology 23: 617–626. [DOI] [PubMed] [Google Scholar]
  20. Krol, E. , Schäper S., and Becker A.. 2020. “Cyclic di‐GMP Signaling Controlling the Free‐Living Lifestyle of Alpha‐Proteobacterial Rhizobia.” Biological Chemistry 401: 1335–1348. [DOI] [PubMed] [Google Scholar]
  21. Lee, M. , Jun S.‐Y., Yoon B.‐Y., Song S., Lee K., and Ha N.‐C.. 2012. “Membrane Fusion Proteins of Type I Secretion System and Tripartite Efflux Pumps Share a Binding Motif for TolC in Gram‐Negative Bacteria.” PLoS One 7: e40460. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Liang, Z. X. 2015. “The Expanding Roles of c‐di‐GMP in the Biosynthesis of Exopolysaccharides and Secondary Metabolites.” Natural Product Reports 32: 663–683. [DOI] [PubMed] [Google Scholar]
  23. Maeda, K. , Tamura J., Okuda Y., Narikawa R., Midorikawa T., and Ikeuchi M.. 2018. “Genetic Identification of Factors for Extracellular Cellulose Accumulation in the Thermophilic Cyanobacterium Thermosynechococcus vulcanus: Proposal of a Novel Tripartite Secretion System.” Molecular Microbiology 109: 121–134. [DOI] [PubMed] [Google Scholar]
  24. Marchante, J. A. , Ruiz‐Sáez L., Muñoz S., Sanjuán J., and Pérez‐Mendoza D.. 2024. “Quantification of Mixed‐Linkage β‐Glucan (MLG) in Bacteria.” In Host‐Pathogen Interactions: Methods and Protocols, edited by Medina C. and López‐Baena F. J., 133–143. Springer US. [DOI] [PubMed] [Google Scholar]
  25. Masuko, T. , Minami A., Iwasaki N., Majima T., Nishimura S.‐I., and Lee Y. C.. 2005. “Carbohydrate Analysis by a Phenol–Sulfuric Acid Method in Microplate Format.” Analytical Biochemistry 339: 69–72. [DOI] [PubMed] [Google Scholar]
  26. Monds, R. D. , Newell P. D., Gross R. H., and O'Toole G. A.. 2007. “Phosphate‐Dependent Modulation of c‐di‐GMP Levels Regulates Pseudomonas fluorescens Pf0‐1 Biofilm Formation by Controlling Secretion of the Adhesin LapA.” Molecular Microbiology 63: 656–679. [DOI] [PubMed] [Google Scholar]
  27. Moradali, M. F. , and Rehm B. H. A.. 2020. “Bacterial Biopolymers: From Pathogenesis to Advanced Materials.” Nature Reviews Microbiology 18: 195–210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. More, T. T. , Yadav J. S. S., Yan S., Tyagi R. D., and Surampalli R. Y.. 2014. “Extracellular Polymeric Substances of Bacteria and Their Potential Environmental Applications.” Journal of Environmental Management 144: 1–25. [DOI] [PubMed] [Google Scholar]
  29. Nikaido, H. 2011. “Structure and Mechanism of RND‐Type Multidrug Efflux Pumps.” In Advances in Enzymology and Related Areas of Molecular Biology, 1–60. Wiley. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Pérez‐Mendoza, D. , Aragón I. M., Prada‐Ramírez H. A., et al. 2014. “Responses to Elevated c‐di‐GMP Levels in Mutualistic and Pathogenic Plant‐Interacting Bacteria.” PLoS One 9: e91645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Pérez‐Mendoza, D. , Bertinetti D., Lorenz R., Gallegos M. T., Herberg F. W., and Sanjuan J.. 2017. “A Novel c‐di‐GMP Binding Domain in Glycosyltransferase BgsA Is Responsible for the Synthesis of a Mixed‐Linkage Beta‐Glucan.” Scientific Reports 7: 8997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Pérez‐Mendoza, D. , Coulthurst S. J., Humphris S., et al. 2011. “A Multi‐Repeat Adhesin of the Phytopathogen, Pectobacterium atrosepticum , Is Secreted by a Type I Pathway and Is Subject to Complex Regulation Involving a Non‐Canonical Diguanylate Cyclase.” Molecular Microbiology 82: 719–733. [DOI] [PubMed] [Google Scholar]
  33. Pérez‐Mendoza, D. , Rodríguez‐Carvajal M. A., Romero‐Jiménez L., et al. 2015. “Novel Mixed‐Linkage Beta‐Glucan Activated by c‐di‐GMP in Sinorhizobium meliloti .” Proceedings of the National Academy of Sciences of the United States of America 112: E757–E765. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Pérez‐Mendoza, D. , Romero‐Jiménez L., Rodríguez‐Carvajal M. Á., et al. 2022. “The Role of Two Linear β‐Glucans Activated by c‐di‐GMP in Rhizobium etli CFN42.” Biology 11: 1364. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Pérez‐Mendoza, D. , and Sanjuán J.. 2016. “Exploiting the Commons: Cyclic Diguanylate Regulation of Bacterial Exopolysaccharide Production.” Current Opinion in Microbiology 30: 36–43. [DOI] [PubMed] [Google Scholar]
  36. Pérez‐Mendoza, D. , Schmid J., Döring M., et al. 2026. “Bioprospecting c‐di‐GMP‐Activated Exopolysaccharides in Bacteria: Isolation of a Novel Atypical Sphingan.” Trends in Biotechnology 44: 2012–2037. [DOI] [PubMed] [Google Scholar]
  37. Perry, B. J. , and Yost C. K.. 2014. “Construction of a Mariner‐Based Transposon Vector for Use in Insertion Sequence Mutagenesis in Selected Members of the Rhizobiaceae.” BMC Microbiology 14: 298. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Pimenta, A. L. , Racher K., Jamieson L., Blight M. A., and Holland I. B.. 2005. “Mutations in HlyD, Part of the Type 1 Translocator for Hemolysin Secretion, Affect the Folding of the Secreted Toxin.” Journal of Bacteriology 187: 7471–7480. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Robertsen, B. K. , Aman P., Darvill A. G., Mcneil M., and Albersheim P.. 1981. “The Structure of Acidic Extracellular Polysaccharides Secreted by Rhizobium leguminosarum and Rhizobium trifolii .” Plant Physiology 67: 389–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Römling, U. , Galperin M. Y., and Gomelsky M.. 2013. “Cyclic di‐GMP: The First 25 Years of a Universal Bacterial Second Messenger.” Microbiology and Molecular Biology Reviews 77: 1–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Römling, U. , Gomelsky M., and Galperin M. Y.. 2005. “C‐di‐GMP: The Dawning of a Novel Bacterial Signalling System.” Molecular Microbiology 57: 629–639. [DOI] [PubMed] [Google Scholar]
  42. Ross, P. , Weinhouse H., Aloni Y., et al. 1987. “Regulation of Cellulose Synthesis in Acetobacter xylinum by Cyclic Diguanylic Acid.” Nature 325: 279–281. [DOI] [PubMed] [Google Scholar]
  43. Saavedra, J. T. , Schwartzman J. A., and Gilmore M. S.. 2017. “Mapping Transposon Insertions in Bacterial Genomes by Arbitrarily Primed PCR.” Current Protocols in Molecular Biology 118: 15.15.11–15.15.15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Schäfer, A. , Tauch A., Jäger W., Kalinowski J., Thierbach G., and Pühler A.. 1994. “Small Mobilizable Multi‐Purpose Cloning Vectors Derived From the Escherichia coli Plasmids pK18 and pK19: Selection of Defined Deletions in the Chromosome of Corynebacterium glutamicum .” Gene 145: 69–73. [DOI] [PubMed] [Google Scholar]
  45. Schäper, S. , Krol E., Skotnicka D., et al. 2016. “Cyclic Di‐GMP Regulates Multiple Cellular Functions in the Symbiotic Alphaproteobacterium Sinorhizobium meliloti .” Journal of Bacteriology 198: 521–535. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Schäper, S. , Steinchen W., Krol E., et al. 2017. “AraC‐Like Transcriptional Activator CuxR Binds c‐di‐GMP by a PilZ‐Like Mechanism to Regulate Extracellular Polysaccharide Production.” Proceedings of the National Academy of Sciences of the United States of America 114: E4822–E4831. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Simon, R. , Priefer U., and Pühler A.. 1983. “A Broad Host Range Mobilization System for In Vivo Genetic‐Engineering‐Transposon Mutagenesis in Gram‐Negative Bacteria.” Biotechnology 1: 784–791. [Google Scholar]
  48. Stasinopoulos, S. J. , Fisher P. R., Stone B. A., and Stanisich V. A.. 1999. “Detection of Two Loci Involved in (1→3)‐Beta‐Glucan (Curdlan) Biosynthesis by Agrobacterium sp. ATCC31749, and Comparative Sequence Analysis of the Putative Curdlan Synthase Gene.” Glycobiology 9: 31–41. [DOI] [PubMed] [Google Scholar]
  49. Tal, R. , Wong H. C., Calhoon R., et al. 1998. “Three Cdg Operons Control Cellular Turnover of Cyclic di‐GMP in Acetobacter xylinum: Genetic Organization and Occurrence of Conserved Domains in Isoenzymes.” Journal of Bacteriology 180: 4416–4425. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Tan, J. , Rouse S. L., Li D., et al. 2014. “A Conformational Landscape for Alginate Secretion Across the Outer Membrane of Pseudomonas aeruginosa .” Acta Crystallographica. Section D, Biological Crystallography 70: 2054–2068. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Wang, Y. , Andole Pannuri A., Ni D., et al. 2016. “Structural Basis for Translocation of a Biofilm‐Supporting Exopolysaccharide Across the Bacterial Outer Membrane.” Journal of Biological Chemistry 291: 10046–10057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Whitney, J. C. , and Howell P. L.. 2013. “Synthase‐Dependent Exopolysaccharide Secretion in Gram‐Negative Bacteria.” Trends in Microbiology 21: 63–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Wright, M. , Kaur M., Thompson L. K., and Cox G.. 2025. “A Historical Perspective on the Multifunctional Outer Membrane Channel Protein TolC in Escherichia coli .” NPJ Antimicrobials and Resistance 3: 6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Zgurskaya, H. I. , Yamada Y., Tikhonova E. B., Ge Q., and Krishnamoorthy G.. 2009. “Structural and Functional Diversity of Bacterial Membrane Fusion Proteins.” Biochimica et Biophysica Acta 1794: 794–807. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1: Bacterial strains and plasmids used in this study.

Table S2: Oligonucleotides used in this work.

Figure S1: Extracellular and intracellular mixed‐linkage glucan (MLG) quantification in Sinorhizobium meliloti tolC mutant strains. Upper panel, schematic representation of the sequential enzymatic digestion protocol. Following lichenase digestion, extracellular MLG was quantified by measuring the sugars released into the supernatant (1, +lich). The remaining cell pellet was washed, lysed by lysozyme treatment followed by boiling, and the soluble sugars released were quantified (2; +lich +lys). Finally, the remaining cellular debris, including putative insoluble polysaccharides, was washed and subjected to a second lichenase digestion to release residual cell‐associated MLG (3, +lich +lys +lich). Lower panel, released sugars were quantified as glucose equivalents in three different tolC mutant strains under the indicated experimental conditions. Data were normalized to total protein content represent the mean ± SD from 4 independent experiments. Statistical analysis was performed using an ordinary two‐way ANOVA followed by Tukey's multiple‐comparison test. Different letters indicate significant differences among strains within each treatment (p < 0.05).

MBT2-19-e70424-s001.docx (235.5KB, docx)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from Microbial Biotechnology are provided here courtesy of Wiley

RESOURCES