University at Buffalo - The State University of New York
Skip to Content
Identification of a Novel N-Acetylmuramic Acid Transporter in Tannerella forsythia

Identification of a Novel N-Acetylmuramic Acid Transporter in Tannerella forsythia

  1. Ashu Sharmaa
  1. aDepartment of Oral Biology, University at Buffalo, Buffalo, New York, USA
  2. bInterfaculty Institute of Microbiology and Infection Medicine Tübingen, IMIT, Department of Microbiology & Biotechnology, University of Tübingen, Tübingen, Germany
  3. cDepartment for NanoBiotechnology, NanoGlycobiology Unit, Universität für Bodenkultur Wien, Vienna, Austria
  4. dOral and Maxillofacial Pathology, University of Sheffield, Sheffield, United Kingdom
  1. P. de Boer, Editor
  1. Case Western Reserve University School of Medicine


Tannerella forsythia is a Gram-negative periodontal pathogen lacking the ability to undergo de novo synthesis of amino sugars N-acetylmuramic acid (MurNAc) and N-acetylglucosamine (GlcNAc) that form the disaccharide repeating unit of the peptidoglycan backbone. T. forsythia relies on the uptake of these sugars from the environment, which is so far unexplored. Here, we identified a novel transporter system of T. forsythia involved in the uptake of MurNAc across the inner membrane and characterized a homolog of the Escherichia coli MurQ etherase involved in the conversion of MurNAc-6-phosphate (MurNAc-6-P) to GlcNAc-6-P. The genes encoding these components were identified on a three-gene cluster spanning Tanf_08375 to Tanf_08385 located downstream from a putative peptidoglycan recycling locus. We show that the three genes, Tanf_08375, Tanf_08380, and Tanf_08385, encoding a MurNAc transporter, a putative sugar kinase, and a MurQ etherase, respectively, are transcriptionally linked. Complementation of the Tanf_08375 and Tanf_08380 genes together in trans, but not individually, rescued the inability of an E. coli mutant deficient in the phosphotransferase (PTS) system-dependent MurNAc transporter MurP as well as that of a double mutant deficient in MurP and components of the PTS system to grow on MurNAc. In addition, complementation with this two-gene construct in E. coli caused depletion of MurNAc in the medium, further confirming this observation. Our results show that the products of Tanf_08375 and Tanf_08380 constitute a novel non-PTS MurNAc transporter system that seems to be widespread among bacteria of the Bacteroidetes phylum. To the best of our knowledge, this is the first identification of a PTS-independent MurNAc transporter in bacteria.

IMPORTANCE In this study, we report the identification of a novel transporter for peptidoglycan amino sugar N-acetylmuramic acid (MurNAc) in the periodontal pathogen T. forsythia. It has been known since the late 1980s that T. forsythia is a MurNAc auxotroph relying on environmental sources for this essential sugar. Most sugar transporters, and the MurNAc transporter MurP in particular, require a PTS phosphorelay to drive the uptake and concurrent phosphorylation of the sugar through the inner membrane in Gram-negative bacteria. Our study uncovered a novel type of PTS-independent MurNAc transporter, and although so far, it seems to be unique to T. forsythia, it may be present in a range of bacteria both of the oral cavity and gut, especially of the phylum Bacteroidetes.


Tannerella forsythia is a Gram-negative, obligate anaerobe strongly associated with periodontitis, which affects the soft and hard tissues supporting the teeth, ultimately leading to tooth loss (1, 2). This bacterium is frequently found with the oral bacterial pathogens Treponema denticola and Porphyromonas gingivalis, together forming a pathogenic consortium termed the “red complex” (3), which in turn is part of a much wider dysbiotic microbiota that is thought to cause this widespread inflammatory disease (4). Strikingly, unlike other bacteria, T. forsythia depends on exogenous N-acetylmuramic acid (MurNAc) for growth (5). It was observed 27 years ago by Wyss that the cultivation of T. forsythia required spent broth from Fusobacterium nucleatum (5) or the presence of free MurNAc (6, 7) in the medium. Since MurNAc together with N-acetylglucosamine (GlcNAc) forms the peptidoglycan amino sugar backbone in all bacteria, this indicated that T. forsythia is unable to synthesize its own peptidoglycan amino sugars. The reasons for this auxotrophy for the amino sugar MurNAc became evident after the close inspection of the T. forsythia genome sequence which became available in 2005 (8, 9). It was noted that the MurA and MurB enzyme homologs required for the de novo synthesis of MurNAc and GlcNAc are not present in the bacterium (10). In addition, the bacterium lacks GlmS, GlmM, and GlmU enzymes for biosynthesis of GlcNAc. Furthermore, evidence collected by analyzing genomes of T. forsythia strains deposited at the Human Oral Microbial Database indicated that this bacterial species lacks a canonical phosphotransferase (PTS)-type MurNAc transporter (MurP), which in Escherichia coli and related Gram-negative bacteria is required for MurNAc uptake and concomitant phosphorylation (11). PTS-type sugar transporters generally mediate the uptake and phosphorylation of sugars; a prototypical PTS system consists of enzyme I (EI), a histidine protein (HPr), the sugar-specific components enzyme IA (EIIA) and EIIB, and a transmembrane sugar-specific transporter protein EIIC (12). The lack of PTS systems in T. forsythia suggests that this bacterium utilizes an alternative transport system to utilize exogenous MurNAc from the environment.

Our in silico investigation of the T. forsythia genome revealed genes coding for putative peptidoglycan degradation and recycling functions (10), among these was a homolog (Tanf_08385; GenBank accession no. WP_046825532) of the E. coli MurQ (13) etherase and two adjacent genes encoding a putative integral membrane protein (Tanf_08375; GenBank accession no. WP_046825530.1) and a putative sugar kinase (Tanf_08380; GenBank accession no. WP_046825531.1). Here we report the preliminary characterization of a novel PTS-independent transport system for MurNAc uptake comprising Tanf_08375 and Tanf_08380 proteins in T. forsythia, which we propose be named TfMurT (for T. forsythia MurT) and TfMurK, respectively, and T. forsythia MurQ etherase (TfMurQ) involved in the metabolic conversion of MurNAc-6-phosphate (MurNAc-6-P) to GlcNAc-6-P.


Bacterial strains and growth conditions.The Tannerella forsythia ATCC 43037 wild-type and mutant strains used in this study were grown anaerobically in BF broth or on agar plates as described previously (14). Escherichia coli strains were grown in Luria-Bertani broth (LB) aerobically at 37°C. E. coli strains were also grown in minimal M9 medium (15) supplemented with either 0.2% glucose, 0.2% glycerol, or 0.025% MurNAc, where needed. E. coli ΔmurQ and ΔmurP mutants were from the Keio collection at the Yale Coli Genetic Stock Center ( All bacterial strains and plasmids used in this study are summarized in Table S1 in the supplemental material.

Molecular biology techniques.Standard molecular cloning techniques were performed according to reference 16. All cloning experiments were performed using the electrocompetent recA mutant cloning strain E. coli Stellar (Clontech Laboratories, CA).

Reverse transcription-PCR.Total RNA was isolated from bacteria using the RNeasy kit (Qiagen). Single-stranded cDNA was synthesized using reverse transcriptase (Invitrogen Superscript III) and random hexamer primers per the manufacturer's protocol. The synthesized cDNA was amplified by PCR with primer sets spanning target genes murQ, murT, and murK (see Fig. 5B): region a with primers TF1067F/TF1068R (F stands for forward, and R stands for reverse), region b with TF1068F/TF1069R, and region c with TF1067F/TF1069R. Primer sequences are listed in Table S2 in the supplemental material).

Production of recombinant TfMurQ protein.Recombinant plasmid pET-TfMurQ was constructed by cloning a TfMurQ open reading frame (ORF) fragment in frame with a C-terminal 6×His tag of the pET30a expression vector (Novagen). Briefly, a PCR fragment amplified with primers TF1069-F and TF1069-R (see Table S2 in the supplemental material) from T. forsythia ATCC 43037 genomic DNA was digested with NdeI and XhoI and cloned via NdeI/XhoI sites into pET30a to generate pET-TfMurQ. Subsequently, E. coli BL21(DE3) strain carrying the pET-TfMurQ plasmid was grown in LB medium with kanamycin (50 μg/ml) at 30°C to an optical density at 600 nm (OD600) of 0.3. Protein expression was induced with isopropyl-β-d-1-thiogalactopyranoside (IPTG) (final concentration of 1 mM) for an additional 3 h at 30°C. Bacteria were collected by centrifugation at 7,000 × g for 10 min, washed with phosphate-buffered saline (PBS) twice, and lysed by sonication for 30 s. Lysates were centrifuged at 10,000 × g for 20 min, and supernatants were collected. Supernatants were loaded onto a column containing 500 μl of HIS-Bind resin (Qiagen), and the column was equilibrated with 10 ml of washing buffer (50 mM NaH2PO4, 300 mM NaCl, 20 mM imidazole [pH 8.0]). Bound recombinant protein was eluted with 1 ml of elution buffer (50 mM NaH2PO4, 300 mM NaCl, 250 mM imidazole [pH 8.0]) and dialyzed extensively against phosphate-buffered saline, pH 7.2, at 4°C. The dialyzed protein fraction was analyzed by SDS-PAGE on 12% gels stained with Coomassie brilliant blue R250.

Detection of etherase-catalyzed reaction intermediate.Etherase activity was assessed by utilizing MurNAc-6-P as the substrate in a Morgan Elson reaction (17). This etherase-catalyzed reaction generates a chromogenic intermediate that can be detected by reacting with Ehrlich's reagent dimethylaminobenzaldehyde to yield a purple product. To detect the formation of this chromogen compound in an enzyme-catalyzed reaction, an Ehrlich-Morgan-Elson assay was performed (18). Briefly, 2 μl of purified recombinant TfMurQ (rTfMurQ) enzyme (1, 2, or 4 μg protein) was added to 20 μl of MurNAc-6-P (10 mM in water), and the reaction mixture was incubated for 60 min at 45°C. After the addition of 100 μl of Ehrlich's reagent, incubation was continued for 20 min at 37°C.

Radioactive etherase assay.The 32P-radiolabeled substrates MurNAc-6-P and GlcNAc-6-P were prepared according to a published protocol (19) with minor modifications. Aqueous solutions of 50 mM MurNAc or GlcNAc, respectively, were added to a reaction mixture containing 100 mM Tris-HCl, pH 7.6, 10 mM MgCl2 100 mM ATP, 140 kBq of [γ-32P]ATP, and 20 μg recombinant Clostridium acetobutylicum MurK protein in a total volume of 100 μl, and the reaction mixture was incubated overnight at 37°C. To start the etherase assay, a reaction mixture containing 15 μl of MurNAc-6-P, 0.4 μg of rTfMurQ, and 100 mM Tris-HCl, pH 7.6, in a total volume of 50 μl was incubated at 37°C. Two microliters of this mixture was spotted immediately and after 15 and 30 min of incubation on a thin-layer chromatography (TLC) plate (silica gel 60 F254; Merck, Darmstadt, Germany). Reaction products were separated in a basic solvent with n-butyl alcohol–methanol–25% (wt/vol) ammonium hydroxide–water (5:4:2:1). The radioactive products were detected using a Typhoon Trio biomolecular imager (GE Healthcare).

Construction of expression vectors and complementation of E. coli ΔmurP mutants and ΔmurQ mutant.DNA fragments coding for T. forsythia MurT-MurK, MurT, MurK, and MurQ ORFs were amplified with primer sets listed in Table S2 in the supplemental material from T. forsythia ATCC 43037 genomic DNA, digested with NdeI and HindIII, and cloned into pTrc99 at NdeI/HindIII restriction sites to generate the plasmids pTr-MurTK, pTr-MurT, pTr-MurK, and TfMurQ, respectively. Plasmids were confirmed by sequencing. For complementation, E. coli mutants were transformed with the above plasmids via electroporation (16).

MurNAc depletion assay.Bacterial cells from overnight cultures of E. coli murP mutant strain harboring either pTrc99, pCS19YfeV, or pTr-MurTK were washed and diluted in M9 minimal medium to an OD600 of 0.05. Before the start of an assay, 10 ml of cell suspension from each strain in triplicate was supplemented with glucose and MurNAc to final concentrations of 5.5 mM and 3.5 μM, respectively, as carbon sources. The cell suspensions were incubated with shaking at 37°C. At regular time intervals, 0.6-ml aliquots were withdrawn, OD600 was determined, and cell-free supernatants were recovered by centrifugation and saved. MurNAc concentration in the cell-free supernatants was then assayed according to a previously described colorimetric assay specific for N-acetyl amino sugars (17). Briefly, 0.1 ml of potassium tetraborate solution was added to 0.5 ml of sample (culture supernatant), followed by boiling for 3 min and cooling to room temperature. Subsequently, 3 ml of p-dimethylaminobenzaldehyde (DMAB) reagent (Sigma) was added, and the tubes were placed in a water bath at 37°C for 20 min. Color developed was read at 585 nm, and the amount of MurNAc was calculated from a standard curve of MurNAc in the range of 0.625 to 5 μM.


T. forsythia contains a putative MurNAc utilization locus.In silico analysis of the T. forsythia ATCC 43037 draft genome (GenBank accession no. JUET00000000.1) identified a three-gene locus (Tanf_08375-Tanf_08385) in the contig_82 DNA sequence (GenBank accession no. NZ_JUET01000082) that included genes coding for an inner membrane protein (Tanf_08375; GenBank accession no. WP_046825530.1), a putative sugar kinase (Tanf_08380; GenBank accession no. WP_046825531.1), and a putative MurQ-type etherase (Tanf_08385; GenBank accession no. WP_046825532). This region is 97% identical to a DNA region of T. forsythia strain 92A2 spanning BFO_0041 to BFO_0044 (NCBI accession no. NC_016610). Interestingly, the Tanf_08375-Tanf_08385 gene cluster is located immediately downstream of a putative peptidoglycan recycling operon, including a muropeptide permease AmpG homolog (Tanf_08365) (20) (Fig. 1A). The product of the putative etherase gene (TfMurQ) shows 46% and 84%, identity with the N-acetylmuramic acid-phosphate (MurNAc-P) etherase MurQ of E. coli (gi:16130353) and predicted MurQ from Bacteroides fragilis (gi:763470620), respectively. The putative T. forsythia MurQ contains a SIS domain (sugar isomerase domain; NCBI Conserved Domain accession no. cd04795) characteristic of phosphosugar isomerases and phosphosugar binding proteins.


T. forsythia MurNAc utilization locus and MurNAc utilization pathway. (A) Genetic organization of the MurNAc utilization locus Tanf_08370-Tanf_08385 (black) of T. forsythia. The genes associated with the locus encode a membrane protein (TfMurT), a sugar kinase (TfMurK), and an etherase (TfMurQ). The locus Tanf_08345-Tanf_08365 (stippled) present immediately upstream is potentially involved in peptidoglycan recycling. Gtf, predicted glycosyltransferase; LytB, predicted amidase enhancer; AmpG, predicted muropeptide transporter; YbbC, hypothetical protein. (B) Schematic model of a MurNAc transport and utilization pathway in E. coli and T. forsythia. Cytoplasmic Memb, cytoplasmic membrane; PEP, phosphoenolpyruvate; E1, enzyme E1; HPR, histidine protein; EIIA, enzyme IIA.

MurT-MurK function as a PTS-independent MurNAc transporter.In E. coli and many other bacteria, free MurNAc is transported across the inner membrane and is simultaneously phosphorylated by the PTS-dependent MurP permease, which is the MurNAc-specific IIBC domain of the PTS system (11). Further processing of phosphorylated MurNAc through the action of MurQ leads metabolic products to enter either a glycolytic pathway to generate energy or a biosynthetic pathway to generate peptidoglycan amino sugar GlcNAc (18, 21) (Fig. 1B). As mentioned above, T. forsythia lacks a canonical PTS-type transporter complex and thus utilizes an alternative mechanism to transport and phosphorylate MurNAc. In silico analysis indicated that TfMurT is a membrane protein with 10 putative membrane-spanning helices (see Fig. S1 in the supplemental material), while TfMurK is a putative sugar kinase with a predicted nucleotide binding domain commonly found in sugar kinases and heat shock proteins (NBD_sugar-kinase_HSP superfamily; accession no. cl17037). Taken together, we predicted that in T. forsythia, MurT functions as a MurNAc transporter, and MurK functions as a MurNAc kinase (Fig. 1B).

To determine the functional roles of TfMurT and TfMurK in MurNAc transport, we tested whether providing the T. forsythia murT and murK genes in trans to an E. coli ΔmurP mutant would rescue the inability of the ΔmurP mutant to utilize MurNAc as the sole carbon and energy source. The results showed that while the E. coli ΔmurP mutant (CM103) complemented with a plasmid (pTr-MurTK) coexpressing TfMurT and TfMurK proteins grew on minimal agar supplemented with 0.025% (wt/vol) MurNAc (Fig. 2A) or broth (Fig. 2B), neither the mutant alone nor the mutant complemented with the empty plasmid pTcr99a grew on MurNAc. All strains grew on minimal agar with glucose used as a control. Additionally, as a positive control, complementation with native E. coli murP in trans via pCS19YfeV restored the growth defect of the E. coli ΔmurP strain on MurNAc (Fig. 2A, middle row). Importantly, the growth of E. coli ΔmurP in the presence of MurNAc was rescued with the combined expression of T. forsythia MurT/MurK and was similar to the growth in the presence of native E. coli MurP. To investigate whether MurNAc transport requires TfMurT/TfMurK coexpression, complementation with either TfMurT or TfMurK in E. coli ΔmurP was performed. The results showed that neither TfMurT nor TfMurK alone could confer the ability to grow on MurNAc to the mutant (Fig. 2A).


Growth of E. coli strains MC4100 (parental strain), CM103 (ΔmurP), and CM133 (ΔmurP Δpts) complemented with respective plasmids in M9 minimal agar and liquid medium with 0.2% glycerol or 0.025% MurNAc. (A) Plate legend and growth of E. coli strains on agar. (B) Growth of E. coli strains in medium with MurNAc (Mu) or glycerol (Gl) measured at OD600. Results of one out of three independent cultivations with similar outcomes are given.

Next, since TfMurT and TfMurK proteins do not possess PTS-type signatures, we wanted to confirm that the TfMurT membrane protein and the TfMurK kinase function independently of a canonical PTS system. For this purpose, we provided the T. forsythia murT-murK genes in trans to an E. coli double mutant (CM133) with deletion of the murP (yfeV) gene and the entire pts operon (ptsHIcrr) coding for the components of the PTS system. Strain CM133 was generated by P1 transduction to transfer the ΔptsHIcrr::kan mutation from strain JM-G77 to strain CM103. The results showed that complementation of CM133 with murT-murK restored the ability to grow to the mutant on MurNAc. As shown, CM133 grew on MurNAc-containing agar (Fig. 2A) or broth (Fig. 2B) when complemented with the plasmid pTr-MurTK coexpressing TfMurT and TfMurK but did not grow on MurNAc when complemented with the plasmid pTr-MurT or pTr-MurK expressing either protein alone. As controls, complementation with native murP (pCS19yfeV) or empty vector did not rescue the growth of CM133 on MurNAc; growth was rescued only when glycerol (0.2%) was provided as the sole carbon source (note that this strain is unable to grow on glucose because of its general PTS defect). The parent strain MC4100, from which CM103 and CM133 were derived, carrying either plasmid grew on glycerol as well as MurNAc (Fig. 2B). Together, these data demonstrated that the products of TfMurT and TfMurK function independently of a PTS system for transport and utilization of MurNAc.

To confirm that this putative transport complex was indeed involved in MurNAc utilization, an experiment was designed where depletion of MurNAc by E. coli strains was assessed in a minimal medium with glucose or MurNAc as a carbon source. Under these conditions, the E. coli ΔmurP mutant carrying either an empty plasmid or plasmid expressing the E. coli MurP (pCS19YfeV) or the T. forsythia MurTK (pTr-MurTK) grew as expected, and MurNAc depletion in the medium was not observed for E. coli cells bearing empty plasmid. However, significant depletion of MurNAc was observed in the case of the E. coli ΔmurP mutant complemented with pTr-MurTK expressing TfMurT/TfMurK or pCS19YfeV expressing native E. coli MurP transporter (Fig. 3). Taken together, these data demonstrate that TfMurK and TfMurT act in concert and TfMurT is a unique transporter for the utilization of exogenous MurNAc in T. forsythia.


MurNAc depletion in minimal medium incubated with the E. coli ΔmurP mutant complemented with different plasmids. E. coli strains were incubated in minimal medium supplemented with glucose and MurNAc, and every 2 h after incubation, spent medium for each strain was assayed for MurNAc using a chromogenic assay specific for N-acetyl-amino sugars.

Tanf_08385 encodes T. forsythia MurQ etherase and is cotranscribed with murTK.Since the MurQ etherase is important in the utilization of MurNAc in bacteria (Fig. 1B), we confirmed the activity of Tanf_08385 as a functional MurNAc-6-P etherase (TfMurQ). For this purpose, TfMurQ expressed as a His6-tagged recombinant protein (rTfMurQ) in E. coli was purified to homogeneity by nickel affinity chromatography (see Fig. S2 in the supplemental material) and confirmed the etherase activity using the Elson-Morgan enzymatic assay and conversion of MurNAc-6-P to GlcNAc-6-P by a radioactive assay using 32P-labeled MurNAc-6-P (18). The Elson-Morgan assay showed that the purified rTfMurQ had etherase activity since a color change was seen with Ehrlich's reagent when rTfMurQ was incubated with MurNAc-6-P (see Fig. S3). Furthermore, rTfMurQ protein catalyzed the formation of a radioactive GlcNAc-6-P product when incubated with MurNAc-6-P in a TLC-based assay employing 32P-labeled MurNAc (Fig. 4A). In addition, the functionality of TfMurQ was tested by trans complementation in an E. coli ΔmurQ mutant. For this purpose, the E. coli ΔmurQ mutant JW2421-1 was transformed with either an IPTG-inducible plasmid harboring the T. forsythia murQ gene (pTr-MurQ) or an empty plasmid vector (pTrc99) and plated on minimal agar plates with glucose or MurNAc as the sole carbon source. Growth of the E. coli ΔmurQ mutant JW2421-1 complemented with pTr-MurQ was rescued on minimal agar plates containing MurNAc (Fig. 4B). The E. coli ΔmurQ mutant complemented with empty pTrc99 did not grow on plates containing MurNAc but grew on media supplemented with glucose. In contrast, the E. coli parent strain BW25113 harboring pTrc99 grew on minimal medium containing glucose and MurNAc as the sole carbon sources (Fig. 4B). These data suggested that Tanf_08385 is the T. forsythia MurQ etherase (TfMurQ) involved in the metabolic conversion of MurANc-6-P to GlcNAc-6-P. Next, we wanted to determine whether mur genes are cotranscribed. For this purpose, RNA from T. forsythia ATCC 43037 cells was extracted, and cotranscription of the mur genes was analyzed using reverse transcription-PCR (RT-PCR) as outlined in Fig. 5. The data demonstrated that the T. forsythia murT, murK, and murQ genes were transcribed as a single transcript (Fig. 5B), since PCR products of the expected size were obtained with primer pairs (see Table S2) designed to bridge the ends between the open reading frames (ORFs) of adjacent genes and thus yielding amplification products only when cotranscription was occurring. Taken together, our data showed that the murT, murK, and murQ genes form an operon (murTKQ) involved in MurNAc utilization.


T. forsythia MurQ (TfMurQ) is a MurNAc-6-P etherase. (A) TLC analysis of MurNAc-[6-32P]phosphate conversion by TfMurQ. MurNAc was radioactively phosphorylated at position C-6 by using recombinant Clostridium acetobutylicum MurK and [γ-32P]ATP. MurNAc-[6-32P]phosphate was then incubated with purified rTfMurQ etherase, and MurNAc-[6-32P]phosphate to GlcNAc-[6-32P]phosphate conversion was monitored. Samples from different time points (lanes 1 to 3) were spotted on a TLC plate together with the standards MurNAc-6-P (lane 4) and GlcNAc-6-P (lane 5). The radioactive products were detected using a phosphorimager. (B) Complementation of an E. coli ΔmurQ mutant (JW2421-1) with TfMurQ. The E. coli ΔmurQ mutant, empty vector control (pTrc99), and complemented strain were plated on minimal agar with MurNAc (0.02% [wt/vol]) or glucose (0.2% [wt/vol]) as a control.


(A) RT-PCR analysis with primer sets spanning adjacent genes (fragments a, b, and c). (B) PCR products were separated on a 1% agarose gel. Lanes: 1, no reverse transcription (RNA only) controls; 2, genomic DNA as the template; 3, cDNA as the template for each primer set. Lane MW contains DNA ladder. The positions of molecular size markers are shown to the left of the gel.


T. forsythia, a common pathogen present in dental biofilms, is implicated in periodontitis. Its role in the disease process has been confirmed in animal models (1), and it has been demonstrated that the bacterium's ability to induce disease is enhanced when coinfected with other bacteria such as Fusobacterium nucleatum (22). Strikingly, T. forsythia depends on exogenous MurNAc, an essential peptidoglycan amino sugar, for growth. Its inability to de novo synthesize the peptidoglycan amino sugars MurNAc and GlcNAc was first described by Wyss (5), who noted that growth of T. forsythia could be rescued when spent media from cultures of F. nucleatum or free MurNAc was supplied exogenously. Since MurNAc is not known to be synthesized by the human host, scavenging on peptidoglycan by-products (muropeptides and anhydro-MurNAc) released by cohabiting oral bacteria during their cell wall recycling is a plausible mechanism by which T. forsythia obtains MurNAc in vivo. Therefore, growth and, thus, the virulence potential of T. forsythia depend on its ability to obtain and utilize MurNAc or MurNAc-containing peptidoglycan fragments from the environment. To our knowledge, no other bacterium has such a strict requirement for MurNAc. Moreover, despite its clear ability to utilize exogenously supplied MurNAc, the T. forsythia genome lacks homologs of PTS-type MurNAc transporters present in bacteria (12). In E. coli and the majority of bacteria, the MurP PTS system is responsible for phosphorylation and import of MurNAc (11, 21), and further utilization of MurNAc transported as MurNAc-6-P proceeds through the action of MurQ etherase (18). MurP contains both the PTS domains EIIB and EIIC and requires enzyme I, histidine protein HPr, and the phosphoryl transfer protein EIIA (EIIAGlc) for function. We searched the T. forsythia ATCC 43037 genome for a similar PTS-type MurNAc transport system, but our search identified no MurP or any of the PTS homologs in the genome of T. forsythia. However, we identified a genetic cluster (Tanf_08375-Tanf_08385) in the genome that contained ORFs for a membrane protein (TfMurT), sugar kinase (TfMurK), and etherase (TfMurQ). This genetic cluster is located immediately downstream from a locus likely to be involved in peptidoglycan recycling as suggested by the presence of an ORF for a putative peptidoglycan permease AmpG in the locus (Fig. 1A). Since TfMurT and TfMurK ORFs were present in close association with an ORF for a MurQ-like etherase (TfMurQ), we hypothesized that TfMurT and TfMurK might be involved in MurNAc transport and utilization functions. During peptidoglycan recycling in bacteria, MurNAc is released as 1,6-anhydro-MurNAc (anhMurNAc) and is phosphorylated to MurNAc-6-P by the kinase AnmK (13). MurNAc-6-P is converted by the MurQ etherase into GlcNAc-6-P, and both these sugars are reused for synthesis of new peptidoglycan or enter general carbohydrate metabolism (13).

In this study, we showed that expression of the TfMurT and TfMurK bipartite pair in an E. coli ΔmurP mutant restored bacterial growth in minimal medium supplemented with MurNAc. In addition, TfMurQ trans complementation in an E. coli ΔmurQ mutant restored the ability to utilize MurNAc, and the purified recombinant TfMurQ protein converted MurNAc-6-P to GlcNAc-6-P in vitro. These data show that TfMurT and TfMurK, coding for an integral membrane transporter and a putative MurNAc sugar kinase, respectively, constitute a unique PTS-independent system for MurNAc transport and phosphorylation. Furthermore, TfMurQ is involved in the metabolic conversion of MurNAc-6-P to GlcNAc-6-P. The functionality of TfMurT and TfMurK was confirmed via trans complementation in the E. coli host. Deletion of these ORFs in T. forsythia was potentially lethal, as no mutants were recovered. While we predict that TfMurT and TfMurK proteins are likely present in close association or have direct physical interactions as a bipartite pair (Fig. 1B) to carry out the function of transport and phosphorylation of MurNAc, we have no experimental evidence to support this notion, and the presence of TfMurK as a cytoplasmic protein cannot be ruled out. Future studies will be needed to biochemically characterize the structure-function relationship of MurT/MurK proteins. Our preliminary attempts to obtain a soluble active form of 6×His-tagged recombinant TfMurK protein have been unsuccessful, as the recombinant protein is expressed in an insoluble, inactive form, even after attempted refolding from insoluble material. Alternative expression approaches are under way to obtain the protein in a soluble form. TfMurT/TfMurK proteins do not possess PTS-type signatures and together represent a novel transport system for MurNAc in T. forsythia. PTS-independent sugar transporters, not as common as PTS-dependent systems, have been reported previously in bacteria. However, such systems have not been characterized at the molecular level. For instance, in streptococci (23, 24) and corynebacteria (25), there is evidence of PTS-independent glucose uptake. We predict that this mode of sugar uptake and utilization might be prevalent at least in the Bacteroidetes phylum of bacteria, since homologs of the murT and murK genes of T. forsythia are present in the genomes of a range of several gut Bacteroides spp. and oral Prevotella spp. (see Fig. S4 in the supplemental material). Strikingly, T. forsythia and Prevotella spp. seem to have a minimal gene set, as the others have extra genes in the cluster, including kinases, ferredoxin, and a β-lactamase, which may reflect their unique niches. Thus, the TfMurTK system is the first evidence of a PTS-independent MurNAc transporter system thus far, and although so far it is unique to T. forsythia, it may be present in a range of Gram-negative bacteria both of the oral cavity and gut.


We thank Tsuyoshi Uehara for helpful discussions during the development stages of this study.

This work was supported in part by U.S. Public Health grants DE14749 and DE22870 (both to A.S.) and the Austrian Science Fund project P24317-B22 (C.S.). A.R. is supported by a T32 training grant (DE023526).


HHS | National Institutes of Health (NIH) SharmaHHS | National Institutes of Health (NIH) RuscittoAustrian Science Fund (FWF) Schäffer


    • Received 16 June 2016.
    • Accepted 2 September 2016.
    • Accepted manuscript posted online 6 September 2016.
  • Address correspondence to Christoph Mayer, christoph.mayer{at}, or Ashu Sharma, sharmaa{at}
  • Citation Ruscitto A, Hottmann I, Stafford GP, Schäffer C, Mayer C, Sharma A. 2016. Identification of a novel N-acetylmuramic acid transporter in Tannerella forsythia. J Bacteriol 198:3119–3125. doi:10.1128/JB.00473-16.

  • Supplemental material for this article may be found at


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.