A Phosphoenolpyruvate-Dependent Phosphotransferase System Is the Principal Maltose Transporter in Streptococcus mutans

The fermentation of carbohydrates by bacteria, such as Streptococcus mutans, and the corresponding production of organic acids are central to dental caries in humans (7). Analysis of the completed genome sequence of an S. mutans strain suggested that this bacterium has the capacity to transport and metabolize a wide range of carbohydrates (2). However, the precise role of many gene products is unknown, and many aspects of S. mutans physiology and carbohydrate metabolism are still incompletely understood. For example, although S. mutans can ferment maltose and utilize it as a sole carbon source (19), it is not known how maltose is taken up by this bacterium.

Maltose and maltodextrins are carbohydrates which are present in the oral cavity as dietary starch is broken down by α-amylase, an abundant enzyme in saliva (8). Therefore, fermentation of these carbohydrates by S. mutans may contribute to the competitiveness of S. mutans and to dental caries. BLAST (3) searches indicate that there are two members of the ATP-binding cassette (ABC) transporter superfamily in S. mutans which are similar to the well-characterized Escherichia coli maltose permease (Mal) (5). It was reported that the uptake of melibiose by one of these, the multiple sugar metabolism transporter (Msm), was inhibited by maltose (20), indicating that maltose may also be transported by this permease. Furthermore, the second of these Mal-like ABC transporters in S. mutans, encoded by malX, malF, malG, and SMU.1571, has been annotated as a putative maltose/maltodextrin transporter. However, as far as we know, no experimental evidence exists to support this designation. During the course of related research, we noted that S. mutans strains that lacked both a functional Msm permease and a Mal-like ABC transporter were still able to ferment maltose and use it as a sole carbon source, in a previously described minimal medium (22), to support growth (Fig. 1). Therefore, it is apparent that there is an unidentified maltose transporter in S. mutans and that this may be the principal maltose/maltodextrin permease in this species.

In addition to the Msm and Mal-like ABC transporters, there are 14 phosphoenolpyruvate-dependent phosphotransferase transport systems (PTS) in S. mutans (2). Such transporters are also involved in transporting a range of carbohydrates (21). There is biochemical evidence that some oral streptococci encode a maltose PTS, including strain OMZ176, which, although originally described as S. mutans, is now reclassified as Streptococcus sobrinus (23). However, the identities of streptococcal genes encoding such a maltose PTS are unknown. Therefore, we proposed that one of the S. mutans PTS transporters is the unidentified maltose transporter and sought to examine this hypothesis.

A number of maltose-specific PTS have been found in other bacterial species. In E. coli strains lacking a functional Mal ABC transporter, overexpression of malX, which in E. coli encodes a PTS transporter, restored the ability to grow on maltose, indicating that this PTS could transport maltose (12). However, it was proposed that MalX transports maltose by facilitated diffusion and that maltose is not the natural substrate of this transporter (12). In contrast, it is apparent that maltose is a principal substrate of Enterococcus faecalis MalT, since inactivation of the gene encoding this PTS decreased [¹⁴C]maltose uptake by 97% and markedly impaired growth of E. faecalis on maltose (9). Similarly, in Bacillus subtilis maltose is taken up by a maltose-specific PTS, MalP (17).

Each of the previously characterized maltose PTS belongs to the PTS glucose-glucoside family (transport classification no. 4.A.1.1) (14). Therefore, we compared the EII components of the members of this family present in S. mutans with the characterized maltose PTS transporters and related glucose PTS that belong to the same transporter classification subfamily (Fig. 2). Only 4 of the 14 PTS present in S. mutans UA159 belong to the glucose-glucoside family. These are BglP (a β-glucoside-specific PTS [6]), ScrA (a sucrose-specific PTS [16]), PttB (a putative trehalose-specific PTS), and the uncharacterized SMU.2047 protein. Only one of the S. mutans proteins had particular similarity to a characterized maltose system. The gene encoding this PTS (SMU.2047) was originally named ptsG by the genome annotators. However, the encoded protein is only 33% identical to B. subtilis PtsG. In contrast, S. mutans PtsG is 64% identical to E. faecalis MalT, a PTS that was characterized after the S. mutans genome was annotated (9). Therefore, we propose that rather than being a glucose-specific PTS, S. mutans PtsG is an orthologue of E. faecalis MalT, and as such it should be renamed MalT or EII^mal. As with E. faecalis MalT (9), S. mutans MalT is a complex protein containing fused EIIA, EIIB, and EIIC domains.

Although S. mutans MalT (PtsG) has identity with E. faecalis MalT, the genetic context of the genes encoding this permease in these two species is different. Of particular note is the absence in S. mutans of any apparent orthologue of E. faecalis malP, which encodes a maltose phosphate phosphorylase. There is no evidence that any of the genes surrounding S. mutans malT are associated with maltose metabolism. However, the product of SMU.2046c does have similarity to metal-dependent hydrolases (COG3568). Therefore, it is possible that this gene encodes a maltose-6-phosphate hydrolase and is required for metabolism of maltose transported into the cell by MalT, but this requires experimental confirmation.

To confirm that S. mutans MalT (PtsG) is a functional orthologue of E. faecalis MalT, we adopted a targeted mutagenesis approach. It was previously reported that inactivation of S. mutans malT (ptsG) did not alter sugar metabolism (1). However, maltose and maltodextrins were not included in this study. Therefore, we inactivated malT, using standard genetic manipulation techniques (15) and the cloning strategy outlined in Table 1, using the primers described in Table 2, before transforming S. mutans, as previously described (11), with the malT::Spec^r DNA from pKCL163, generating a malT mutant, KCL93. Reverse transcription-PCR (RT-PCR) analysis, with the primers listed in Table 2, confirmed that the insertion into malT did not disrupt the transcription of any adjacent genes, including SMU.2046c (primers P275 and P276) and relA (SMU.2045; primers P277 and P278). An internal fragment of malT was amplified by RT-PCR using primers P273 and P274, and the size of this product was consistent with insertion of the Spec^r gene. This RT-PCR analysis also indicated that malT is cotranscribed with SMU.2046c (primers P281 and P282), relA (SMU.2045; primers P238 and P284), and the putative gene annotated as SMU.2048 (primers P279 and P280). Coexpression with SMU.2048 was not expected, since it is transcribed divergently from malT. However, on closer inspection of the sequence annotated as SMU.2048, we doubt whether this is an actual gene, since it has only 152 nucleotides, it lacks an obvious initiation codon, and it has no significant similarity to known genes.

The abilities of wild-type S. mutans UA159 and the isogenic malT (ptsG) mutant to utilize maltose and maltodextrins as a sole carbon source were determined by observing growth in a semidefined medium (Fig. 3). This semidefined medium was described in a previous publication (22). Growth was determined by incubating inoculated 96-well microtiter plates, previously equilibrated and sealed under anaerobic conditions (80% N₂, 10% H₂, 10% CO₂), at 37°C in a microtiter plate reader (Labsystems iEMS reader MF; Thermo Life Sciences) and recording the optical density (620 nm) at 15-min intervals after shaking.

Although the malT mutant was able to grow on maltose as a sole carbon source, it did so only at a much reduced rate compared to growth of S. mutans UA159 (Fig. 3A). There was also a reduction in its ability to utilize maltotetraose and maltotriose (Fig. 3B and C). In contrast, there was no difference between the wild-type and the malT mutant when glucose (Fig. 3D) or isomaltose (data not shown) was the sole carbon source. Growth analysis of the complemented malT mutant, KCL103, confirmed that the observed phenotype was due to the mutation of this gene and not any secondary mutation (Fig. 3). These data support the hypothesis that MalT is the principal transporter of maltose in S. mutans and not, as originally supposed, a glucose-specific PTS. Thus, S. mutans MalT is an orthologue of E. faecalis MalT.

The slight decrease in growth observed in maltotetraose and maltotriose (Fig. 3) indicates that these maltodextrins may also be substrates of MalT, but alternative transporters of these carbohydrates must be present in S. mutans. It has previously been reported that the principal transporter of isomaltose in S. mutans is the Msm transporter (13, 20). It is possible that maltotetraose or maltotriose is also transported by this or the related Mal-like ABC transporter, but further studies are required to confirm this.

The presented growth data are strong evidence that MalT is involved in maltose uptake. Nonetheless, to confirm this, we performed a series of radioactive uptake assays using [U-¹⁴C]maltose (Amersham GE Healthcare). S. mutans strains were grown in a semidefined medium (described previously [22]) with 20 mM glucose, 10 mM maltose, or 5 mM glucose plus 10 mM maltose as sole carbon sources. Exponentially growing S. mutans cells were harvested by centrifugation when the optical density (620 nm) was approximately 0.5, and uptake of [U-¹⁴C]maltose was assayed by the rapid filtration method described previously, with a final concentration of 25 μM and 0.125 μCi [U-¹⁴C]maltose (22). The S. mutans strains which lacked active malT (KCL93 and KCL104) were unable to accumulate [U-¹⁴C]maltose, whereas the wild-type strain (UA159) and the complemented mutant (KCL103) did so at a rapid rate (Fig. 4). This confirms that MalT is the principal maltose transporter in S. mutans.

The rate of [U-¹⁴C]maltose uptake was slightly enhanced by the inclusion of maltose in the growth medium rather than glucose (Fig. 4), indicating that the presence of this substrate increased expression of malT. However, the high rate of [U-¹⁴C]maltose uptake observed for S. mutans grown on glucose as a sole carbon source (Fig. 4) indicates that malT is expressed under these conditions and it is not subject to significant catabolite repression.

To obtain an indication of the range of solutes transported by MalT, [U-¹⁴C]maltose uptake was carried out in the presence of 1 mM (40-fold excess) competing solutes (Fig. 5). None of the substrates tested inhibited uptake to the same extent as an excess of unlabeled maltose, which reduced uptake to only 5% of the control rate. Therefore, it is likely that MalT has greatest affinity for maltose. However, maltotriose, mannose, galactose, and fructose each inhibit [U-¹⁴C]maltose uptake by 60 to 72%, and maltotetraose, glucose, sucrose, and raffinose inhibit it by 28 to 42%. It is thus possible that these are substrates of MalT, with the latter group being lower-affinity substrates, although it is also possible that some are not actual substrates for this transporter but merely inhibit uptake of maltose. Nevertheless, it is apparent that MalT is at least not the sole transporter in S. mutans for some of these substrates. As discussed above, our growth data (Fig. 3) indicate there are other transporters for maltotriose and maltotetraose. Furthermore, Abranches et al. (1) report that a malT mutant, which they refer to as a ptsG mutant, had no altered glucose, fructose, and mannose PTS activity compared to the wild-type strain. It is therefore unlikely that MalT is the dominant transporter of these carbohydrates.

Orthologues of S. mutans MalT are apparent, as a result of BLAST (3) searches, in the genomes of other Streptococcus species, including Streptococcus pneumoniae TIGR4 and R6, Streptococcus pyogenes MGAS5005, Streptococcus agalactiae 2603V/R, and Streptococcus suis 89/1591. Therefore, MalT orthologues may also play an important role in maltose and maltodextrin uptake in these important pathogens. It has been proposed that maltodextrin transport is a key component for the initial colonization of the oropharynx by S. pyogenes (18). In contrast to our observations with S. mutans, mutation of the gene encoding a single ABC transporter component in S. pyogenes, malE (which is similar to the genes encoding the solute binding proteins of S. mutans Msm and Mal-like ABC transporters), significantly impaired growth on maltose as a sole carbon source (18). We speculate that the MalT orthologue enables the residual growth on maltose in this malE mutant. However, it is clear that MalT alone cannot support optimum growth on maltose for S. pyogenes. Therefore, there are apparent differences in the relative contribution of MalT to maltose transport in S. pyogenes compared to that in S. mutans. Further research is required to determine whether the MalT orthologue in S. pyogenes contributes to colonization.

• Open in new tab
• Download powerpoint

FIG. 1.

Growth of S. mutans UA159 (solid lines), the isogenic msmK::aphA3 mutant (KCL57; dotted lines), the SMU.1571::aphA3 mutant (KCL48; dashed line), and the msmK::aphA3 SMU.1571::ery double mutant (KCL82; dash-dot-dot-dash line) in minimal medium containing 10 mM maltose as the sole carbon source. Data shown are means from at least three independent experiments.

• Open in new tab
• Download powerpoint

FIG. 2.

Phylogenetic tree indicating the relationship between S. mutans MalT (PtsG) and related permeases. A phylogenetic tree was constructed from the amino acid sequences of S. mutans MalT (PtsG) and a number of related EII components that are members of a subclass of PTS (glucose-glucoside family; transport classification no. 4.A.1.1) using Vector NTI Suite (version 10), which uses the ClustalW algorithm, and Treeview (Win32). Protein designations and accession numbers are as follows: bsMalP, Bacillus subtilis MalP (NP_388701); bsPtsG, B. subtilis PtsG (NP_389272); ecPtsG, Escherichia coli PtsG (NP_415619); ecMalX, E. coli MalX (P19642); efMalT, Enterococcus faecalis MalT (NP_814695); SMU.2047 PtsG/MalT, S. mutans PtsG/MalT (NP_722340); SMU.980 BglP, S. mutans BglP (NP_721375); SMU.1841 ScrA, S. mutans ScrA (NP_722158); SMU.2038 PttB, S. mutans PttB (NP_722334).

• Open in new tab
• Download powerpoint

FIG. 3.

Growth of S. mutans UA159 (solid lines), the isogenic malT mutant (KCL93; dotted lines), the complemented malT mutant (KCL103; dashed lines), and the noncomplemented malT mutant plasmid control (KCL104; dash-dot-dot-dash line) in minimal medium containing 10 mM maltose (A), 10 mM maltotetraose (B), 10 mM maltotriose (C), or 20 mM glucose (D) as the sole carbon source. Data shown are means from at least three independent experiments.

• Open in new tab
• Download powerpoint

FIG. 4.

Uptake of [U-¹⁴C]maltose by S. mutans UA159 (A) (filled symbols), KCL93, the isogenic malT mutant (A) (open symbols), KCL103, the complemented malT mutant (B) (closed symbols), and KCL104, the noncomplemented malT mutant plasmid control (B) (open symbols) grown in minimal medium containing 20 mM glucose (squares), 10 mM maltose (triangles) (UA159 and KCL103 only, since the malT mutant cannot grow on maltose as a sole carbon source), or 5 mM glucose plus 10 mM maltose (circles). Data shown are means from at least three independent experiments.

• Open in new tab
• Download powerpoint

FIG. 5.

Inhibition of [U-¹⁴C]maltose uptake by competing solutes. The rate of [U-¹⁴C]maltose uptake by S. mutans UA159 grown in minimal medium containing 20 mM glucose was determined over the initial minute of uptake. Competing solutes were added to a final concentration of 1 mM, 5 s before the addition of 25 μM (0.06 μCi) [U-¹⁴C]maltose.

• Copyright © 2007 American Society for Microbiology