l-Rhamnose Transport Is Sugar Kinase (RhaK) Dependent in Rhizobium leguminosarum bv. trifolii

Bacterial strains and culture conditions.Bacterial strains and plasmids used and generated in this work are listed in Table 1. R. leguminosarum strain Rlt100 (original designation, W14-2) (4) and strains derived from this wild-type strain were routinely grown at 30°C on TY (7) as a complex medium and VMM (57) as a defined medium. Carbon sources were filter sterilized and added to defined medium to a final concentration of 15 mM. When required, antibiotics were added to solid or liquid media at the following concentrations: tetracycline, either 10 μg ml⁻¹ or 5 μg ml⁻¹; neomycin, 200 μg ml⁻¹; streptomycin, 200 μg ml⁻¹; ampicillin, 100 μg ml⁻¹; and kanamycin, 50 μg ml⁻¹. Bacterial growth was monitored spectrophotometrically at 600 nm.

DNA manipulations, sequencing, and sequence analysis.Standard techniques were used for plasmid isolation, restriction enzyme digestion, ligation, transformation, and agarose gel electrophoresis (49). Sequencing reactions were done using dye terminators at the University of Calgary Core DNA Services, using an ABI automated sequencer. Sequence data were analyzed using DNASIS (Hitachi Software Engineering Co., San Bruno, CA). Database searches were done using the BLASTX program (3).

Construction of rhaQ::phoA fusions.To generate in-frame rhaQ::phoA fusions, phoA was fused to either a predicted cytoplasmic or periplasmic domain of rhaQ. phoA was amplified from a strain containing TnphoA (36) by using the primers phoA5′ (5′-ATAACCCGGGGACTCTTATACACAAGTA-3′ [containing a SmaI restriction site]) and phoA3′ (5′-ATATGAGCTCTTATTTCAGCCCCAGAG-3′ [containing an SstI restriction site]). The PCR product was amplified, digested, and cloned into pBluescript as a SmaI/SstI fragment, sequenced, and designated pMW36.

Two rhaQ fragments were generated using the 5′ primer Q5Hind (5′-ATAAGCTTAATGAGCACCGTTTCGACGC-3′ [contains a HindIII restriction site and corresponds to the start of rhaQ]) and either the 3′ primer Q3 SmaI A (contains a SmaI restriction site immediately following the sequence that corresponds to amino acid 163 in RhaQ [a predicted periplasmic loop]) or Q3 SmaI B (5′-ATCCCGGGCCGATCGTGCCGGAACC-3′ [contains a SmaI restriction site immediately following the sequence that corresponds to amino acid 282, predicted to be a cytoplasmic loop]). The PCR products were digested with SmaI/HindIII, ligated into pEX18Tc (26), and verified by sequencing. These were named pMR180 and pMR182, respectively.

To generate the fusions, the phoA fragment in pMW36 was taken as a SmaI/SstI fragment and cloned into pMR180 and pMR182, yielding pMR185 and pMR186, respectively (note that these constructs are positioned such that they are not transcribed from the P_lac promoter in pEX18Tc). These constructs were mobilized into Rlt100 and Rlt114, and single-crossover mutants were selected. The resultant constructs were verified by sequencing the fusion junctions and were named Rlt275, Rlt276, Rlt277, and Rlt278.

Overexpression and purification of RhaK.Construction of pMR106 was carried out by amplification of rhaK, using pW3C1 as a template and the primers 5′HindIII/RBS (5′-ATATAAGCTTGGAGAACTGCAGATGACCGCCAGTTCCTATC-3′) and 3′BamHI/PstI (5′-ATATGGATCCCTGCAGCTATGCCATCGCCGCGTA-3′). The resulting fragment was ligated into pBluescript and into the broad-host-range vector prk7813 (31), using BamHI and HindIII restriction enzyme sites within the primers, and the resulting constructs were called pmr106 and pmr110, respectively.

To construct an in-frame His₆-tagged RhaK protein, pW3C1 was used as a template. The primers used in the PCR amplification were kinaseB5′ (5′-ATAAGGATCCATGACCGCCAGTTCCTATC-3′) and kinaseH3′ (5′-ATAAAGCTTCTATGCCATCGCCGCGTA-3′). This fragment was ligated into the pRSETA expression vector by using BamHI and HindIII such that rhaK was under the control of the T7 promoter and in frame with an N-terminal His₆ tag. The resulting construct was confirmed by sequencing and was named pMR45.

To overexpress an RGS-His₆-tagged RhaK protein in R. leguminosarum, both N-terminally and C-terminally tagged versions were constructed. pMR110 was used as a template. The primers 5 His 6H (5′-ATAAGCTTAGGAGAAATAATTATGAGAGGATCGCATCATCATCATCATCATCATACCGCCAGTTCCTATCGC-3′ [contains a HindIII site, a ribosome binding site, the RGS-His₆ tag, and the 5′ annealing portion of rhaK]) and 3 BamHI/PstI (5′-ATATGGATCCCTGCAGCTAAGCCATCGCCGCCGTA-3′ [contains a BamHI and a PstI site and the 3′ annealing portion of rhaK]) were used to generate N-terminally tagged RhaK. The primers 5HindRBS (5′-ATATAAGCTTGGAGAACTGCAGATGACCGCCAGTTCCTATC-3′ [contains a HindIII restriction site, a ribosome binding site, and the 5′ annealing portion of rhaK]) and 3 His 6B (5′-ATGGATCCCTAATGATGATGATGATGATGCGATCCTCTTGCCATCGCCGCGTACC-3′ [contains a BamHI restriction site, a C-terminal RGS-His₆ tag, and the 3′ annealing portion of rhaK]) were used to create C-terminally tagged RhaK.

In both cases, the PCR products were subsequently cut with BamHI and HindIII and cloned into pRK7813 such that they were expressed from the P_lac promoter. The constructs were sequenced and named pMR178 and pMR179 (Table 1).

To isolate His₆-tagged RhaK from E. coli cells, cultures grown to mid-log phase (optical density at 600 nm = 0.5) in LB plus Ap were induced with IPTG (isopropyl-β-d-thiogalactopyranoside; 1 mM) at 37°C while being shaken at 240 rpm for 4 h. Cells were harvested by centrifugation (6,000 × g for 10 min) and resuspended in 2 ml g⁻¹ wet pellet of binding buffer (20 mM Tris, pH 7.8, 300 mM NaCl, 10 mM imidazole, 5 mM β-mercaptoethanol). Cells were lysed by two passages through a French pressure cell (16,000 lb/in²), and cell extracts were prepared by centrifugation (6,000 × g for 10 min). His₆-tagged RhaK was immobilized by passing cell extract through a nickel-affinity column attached to a fast-performance liquid chromatography apparatus (Bio-Rad biologic LP chromatography system). Washing was done with wash buffer (20 mM Tris, pH 7.8, 300 mM NaCl, 20 mM imidazole, 5 mM β-mercaptoethanol) for 60 min at a flow rate of 1 ml min⁻¹. The RhaK fusion was then eluted from the nickel-affinity column (Bio-Rad Ni-nitrilotriacetic acid superflow column) by using elution buffer (20 mM Tris, pH 7.8, 300 mM NaCl, 250 mM imidazole, 5 mM β-mercaptoethanol). The purity of RhaK was confirmed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The presence of His₆-tagged RhaK was confirmed by Western blotting, using mouse monoclonal anti-RGS-His₆ as the primary antibody and a goat anti-mouse horseradish peroxidase-conjugated antibody as the secondary antibody. Detection was carried out using an opti-4CN substrate detection kit as suggested by the manufacturer (Bio-Rad Laboratories, Hercules, CA).

To isolate tagged RhaK from R. leguminosarum, essentially the same protocol was used, except that cultures were grown to late log phase in TY and IPTG was not used to induce expression.

Generation and overexpression of rhaK P-loop mutations.The generation of rhaK P-loop base pair mutations was accomplished using the megaprimer PCR method (50). The template DNA used in this amplification was from pMR106. The primers used for the megaprimer method were R1 external (5′-GTAAAACGACGGCCA-3′) and F2 external (5′-GGATCCCTGCAGCTATGCCATCGC-3′) and the mutagenic primers lysine-to-methionine M primer (3′-TGTAGCCGTACTGGTT-5′) and lysine-to-aspartate M primer (3′-TGTAGCCGCTGTGGTT-5′). The amplification products were cloned as BamHI/HindIII fragments into pRK7813 by using restriction sites introduced into the primers, subsequent clones were called pMR129 (K to M) and pMR125 (K to D), and these constructs were verified by nucleotide sequencing. Protein overexpression analysis of the rhaK base pair mutants, carried out on pMR129 and pMR125, was accomplished using the same cloning strategy as that for overexpression of pMR45.

Transport assays.Uptake of rhamnose was carried out essentially as previously described (48). Radioactive [³H]rhamnose (5 Ci/mmol) was purchased from American Radiolabeled Chemicals Ltd., St. Louis, MO. Transport assays were initiated by the addition of tritiated rhamnose to a final concentration of 2 μM (125,000 dpm). Aliquots of 0.5 ml were rapidly filtered through a Millipore 0.45-μm Hv filter on a Millipore sampling manifold and washed with 5 ml of defined salts medium. The radioactivity that remained on the filter was determined using a liquid scintillation spectrophotometer (Beckman LS6500). Initial rates were calculated from the linear portion of the data (generally the first 60 seconds) and normalized to the fresh weight (FW) of the bacteria used.

Whereas our standard uptake protocol uses cells that have been induced overnight in defined medium containing glycerol and rhamnose (48), strains carrying either rhaDI or rhaI mutations were unable to grow in this medium for extended time periods, so optimal induction times for strains Rlt105 and Rlt124 were determined experimentally. It was found that 7 h of exposure to rhamnose was sufficient to induce the genes without any detriment to growth due to the sensitivity phenotype exhibited by isomerase or dehydrogenase mutant strains (data not shown).

In vitro RhaK kinase assays.Strains were grown on minimal medium to mid-log phase, harvested by centrifugation (6,000 × g for 10 min), and resuspended in 2 ml g⁻¹ wet pellet of lysis buffer. Resuspended cells were lysed by 2 passages through a French pressure cell (16,000 lb/in²), and cell debris was removed by centrifugation (10,000 × g for 10 min at 5°C in a Sorval SLA-3000 rotor).

Whatman DE81 cellulose chromatography paper was first treated with 1 M formic acid, extensively washed with double-distilled H₂O to remove any impurities, and finally dried prior to use. Isolated, induced cell extracts were combined with 50 mM HEPES buffer, pH 7.6, 5 mM MgCl₂, 10 mM ATP, and 2 mM tritiated rhamnose and incubated at 30°C. Assays were carried out in a final volume of 50 μl. Ten-microliter samples were taken at 1-, 5-, 10-, and 30-minute time points and quenched by placing each sample in a boiling water bath for 3 min and then into an ice bath until all samples were ready to be loaded. The 10-μl quenched aliquots were spotted and separated by descending paper chromatography, using 50 mM formic acid, pH 3.2. To determine the position of the label, the chromatography paper was cut into 1-in.-square pieces and assayed using a liquid scintillation spectrophotometer (Beckman LS6500).

RhaK phylogenetic tree construction.Kinase sequences were selected based on homology of the RhaK amino acid sequence, using BLASTP (3). The sequences were aligned using Clustal X (55). Programs contained within PHYLIP, version 3.6a (15, 16), were used for phylogenetic analysis. The divergence (or distance) between two sequences was calculated by PROTDIST (setting JTT). The resulting distance matrix was used to construct a phylogenetic tree with the NEIGHBOR program (NJ option), and the phylogenetic tree was evaluated using the bootstrap procedure (SEQBOOT; 1,000 replicates). Bootstrap replicates were analyzed with the CONSENSE program to generate the majority rule consensus tree. The phylogenetic trees were drawn using the TREEVIEW (40) program, using the PHYLIP output file as an input file. A carbohydrate kinase from Pirellula sp. strain 1 that appeared to be distantly related on the basis of a BLASTP search was selected as the comparative outgroup.

Strains carrying a mutant rhaK allele do not transport rhamnose.We previously cloned, sequenced, and characterized a region from Rhizobium leguminosarum bv. trifolii necessary for competition for nodule occupancy and the utilization of rhamnose (39, 48). Nucleotide sequence analysis of this region suggests that it contains three genes necessary for the catabolism of rhamnose (rhaD, encoding a dehydrogenase/aldolase; rhaI, encoding an isomerase; and rhaK, encoding a kinase), all of the components necessary for a functional ABC transporter (rhaS, rhaT, rhaP, and rhaQ, encoding a sugar binding protein, an ABC ATPase, and two transmembrane permeases, respectively), a negative regulator (rhaR), and a small gene that encodes a protein of unknown function (rhaU). These genes are organized in two divergently transcribed operons. One transcript consists of rhaDI, whereas the other consists of rhaRSTPQUK. Expression and transport activity are induced by rhamnose, the transporter has a high affinity for its substrate, and strains carrying mutations that affect the expression of the ABC transporter components are unable to transport rhamnose (48).

In the course of our characterization, we included strain Rlt144, which carries the rhaK50::Tn5-B20 allele, as a putative positive control for rhamnose transport assays. This appeared to be a reasonable choice, since a mutation in rhaK is not polar on any of the core components of the ABC transporter. Surprisingly, strain Rlt144 appeared to have a nontransport phenotype similar to that of the other components of the ABC transport complex (<0.05 nmol/g FW/min). To ensure that this was not a strain-specific artifact, two other strains, Rlt131 and Rlt146, carrying rhaK40 and rhaK52, respectively, were also assayed for the ability to transport rhamnose. These strains were independently isolated and have different points of insertion in rhaK (48). In both cases, these strains did not show any transport activity.

We had previously shown that introduction of a rhaK mutant allele into a strain carrying a polar rhaD mutant (genotypically rhaD rhaI mutant) reverses the inability of this strain to grow on rhamnose-glycerol—presumably a buildup of a phosphorylated intermediate does not occur (48). In light of our result with other rhaK alleles, we wished to ascertain if, in fact, the phenotype reversal was due to an inability to take up rhamnose. Strain Rlt105 (rhaD rhaI) was clearly capable of transporting rhamnose (15.9 ± 0.5 nmol/min/g FW). Note that the rate appears to be somewhat lower than the rate measured for Rlt100 (25 ± 0.5 nmol/min/g FW); however, this may be attributed to the toxicity exhibited by this strain, which required a shorter induction time for sufficient cells to be grown in a glycerol-rhamnose medium to measure uptake rates. Strains Rlt210 (rhaD rhaI rhaK) and Rlt211 (rhaI rhaK), like other strains containing rhaK mutant alleles, were unable to transport labeled rhamnose (<0.05 nmol/min/g FW). The reverse of the glycerol sensitivity phenotype in Rlt1210 and Rlt211 is likely due to the absence of an unmetabolizable intermediate; these results suggest that an inability to transport rhamnose rather than phosphorylation is responsible for the reversal of the sensitivity.

Rhamnose kinase mutants can be complemented with rhaK⁺ in trans.The inability of rhaK mutants to transport rhamnose could be due to the dependence of the transporter on the presence of RhaK. Alternatively, the insertion mutations could cause a transcript instability that leads to an absence of the other components of the ABC transporter. To rule out the latter possibility, it was reasoned that if the rhamnose mutants were complemented with a wild-type copy of rhaK and the ability to grow on rhamnose as a sole carbon source, as well as the ability to transport rhamnose, was restored, then it could be concluded that the mutations did not affect the expression of the other ABC transport components. rhaK was PCR amplified and cloned with a ribosome binding site into the broad-host-range plasmid pRK7813 such that transcription would occur from P_lac, which is part of the plasmid multiple cloning site (31). The resultant construct, pMR110, was conjugally introduced into strains Rlt131, Rlt144, and Rlt146 by selecting for transfer of the plasmid-encoded antibiotic determinant as previously described (17). The transconjugant strains were designated Rlt198, Rlt199, and Rlt200, respectively. These were tested first for the ability to grow using rhamnose as a sole carbon source and subsequently for the ability to transport rhamnose. In all cases, the introduction of rhaK ⁺ into strains carrying a rhaK mutation resulted in the strains being complemented both for the ability to grow on rhamnose as a sole carbon source and for the ability to transport rhamnose at levels comparable to that of wild-type Rlt100 (Table 2).

To eliminate the remote possibility that RhaK may have cryptic transport functions independent of the ABC transporter, Rlt213 was constructed by introducing pMR110 into strain Rlt106, which carries the rhaT2 allele. The rhaT2 mutation is a lesion in the gene that encodes the ABC protein, a critical component of a functioning ABC transporter. Moreover, the mutation is also polar on rhaPQUK, thus eliminating all components of the ABC transport complex except for the periplasmic sugar binding protein encoded by rhaS. Neither Rlt106 nor Rlt213 could grow on minimal medium supplemented with rhamnose, nor could they transport tritiated rhamnose from the growth medium (Table 2). Together, these results suggest that transport of rhamnose in Rlt100 is dependent on both the components of the ABC transporter and RhaK.

RhaK does not affect transcription or translation of ABC components.It is possible that the metabolite produced by RhaK or a subsequent catabolic step may alter the transcription of the ABC transporter genes and thus lead to an absence of rhamnose transport. To address this, a series of constructs were made to contain the rhaP36::Tn5-B20 lacZ fusion allele either integrated into the genome or present on a plasmid in the presence or absence of rhaK expressed from either a plasmid or the genome (Table 3). The results clearly show that fusion activity from rhaP36 is unaffected by the presence or absence of rhaK. The data suggest that the transcription of rhaSTP, and presumably rhaQ, occurs in strains carrying rhaK alleles that are transport deficient.

To verify that the ABC transcript was in fact being translated, phoA translational fusions to rhaQ were constructed with either a predicted cytoplasmic or periplasmic loop of RhaQ. The resultant strains, Rlt275 and Rlt276, were assayed for alkaline phosphatase activity. As anticipated, the cytoplasmically localized phoA fusion rhaQ59 in Rlt275 did not have activity discernible from the background on either glucose- or rhamnose-containing medium. However, the periplasmically targeted fusion rhaQ60 in strain Rlt276 showed induction similar to that of the transcriptional fusions (Table 4). To ensure that there was no expression of rhaK, these phoA fusions were also introduced into a rhaK mutant background (48). The rhaQ59 and rhaQ60 alleles in strains Rlt277 and Rlt278 gave results comparable to those in the wild-type background (Table 4 and data not shown). Together, these data suggest that the absence of RhaK does not affect transcription or translation of the transcript that encodes the components of the ABC transporter.

Detection of a kinase-dependent rhamnose intermediate.The inability of strains carrying rhaK mutations to take up rhamnose poses yet another question, i.e., does RhaK need to be present as a structural entity or in a biochemically active form for transport to occur? The genetic evidence is consistent with RhaK acting on rhamnose before either the rhaD or rhaI gene product (48). Since rhamnose kinase activity has not been reported previously, a standard assay for RhaK does not exist. Therefore, direct testing of the genetic data has not been possible. To answer these questions, an in vitro assay was developed.

Initial assays relied on overexpressing and purifying RhaK containing either an N-terminal or C-terminal His₆ tag from either E. coli or R. leguminosarum and incubating the protein in the presence of an appropriate buffer, ATP, and radiolabeled rhamnose. The reaction products were then separated by paper chromatography. Under the chromatography conditions utilized, phosphorylated compounds are retarded in their migration, whereas unphosphorylated sugars migrate at or very near the solvent front (62).

Although we were capable of isolating full-length protein products from both E. coli and R. leguminosarum, the experiments were never successful. Even though both the N-terminally (pMR178) and C-terminally (pMR179) His₆-tagged versions of RhaK were able to complement rhaK mutations, the isolated RhaK proteins did not show any in vitro activity (data not shown). Moreover, if transport activity was dependent upon RhaK, we were not convinced that RhaK activity might not be dependent on ABC transport components. An additional component(s) required for kinase activity might be present in a cell extract, whereas it may be missing from an overexpressed purified protein isolated from either E. coli or R. leguminosarum. We therefore abandoned the overexpression system and developed an assay using a cell lysate of Rlt100.

Assays either utilizing labeled ATP or measuring the amount of ATP hydrolyzed were not practical because the assays were carried out using a crude extract. For the assay to be specific, we set up the following criteria. Activity from the cell extract would need to be dependent upon rhamnose, ATP, and induced extract, the signal of the intermediate(s) would need to be well over the background, and the accumulation of reaction products would need to be time dependent. Over a 30-min assay, a labeled peak accumulated with an R _f value of 0.24 (Fig. 1). The biochemical conversion of the labeled substrate into a rhamnose derivative was time dependent over the course of the assay, and the peak observed on the chromatography paper was consistent with the behavior of a phosphorylated intermediate (Fig. 1). This peak was dependent on the presence of induced extract, ATP, and rhamnose (data not shown).

• Open in new tab
• Download powerpoint

FIG. 1.

A rhamnose-induced wild-type Rlt100 cell extract was isolated and assayed as described in Materials and Methods. Samples were taken and quenched at 1, 5, 10, 15, and 30 min and then were counted. Samples were corrected for background. Symbols: filled bars, 1 min; filled bars with white spots, 5 min; white bars with few black spots, 10 min; white bars with dense black spots, 15 min; white cross-hatched bars, 30 min. Note that counts over background were not seen at 1 min and that the tritiated rhamnose peak that runs at the solvent front (R _f > 0.8) is not shown. One representative set of data is presented; experiments were replicated at least three times.

Accumulated peak is rhaK dependent.Although the peak that accumulated with an R _f value of 0.24 fulfilled the criteria for a RhaK assay, it was necessary to show that this peak was dependent upon rhaK. To show this, we used extracts from strains carrying various alleles that were unable to grow using rhamnose as the sole carbon source.

When rhamnose-induced cell extracts from strains Rlt105 (rhaD1), Rlt146 (rhaK52), Rlt211 (rhaD1 rhaK58), and Rlt212 (rhaI31 rhaK58) were assayed, activity was present in Rlt105, whereas it was abolished in Rlt146, Rlt211, and Rlt212 (Table 5). To determine if this activity could be restored with the addition of wild-type rhaK, strain Rlt200 (rhaK52 rhaK ⁺) was also assayed (Table 5). The data show that the activity of the RhaK assay is independent of rhaD and rhaI, its absence is correlated with the presence of a rhaK mutant allele, and complementation with a wild-type copy of rhaK restores activity. Together, these data show that the activity we assayed was due to RhaK.

Transport and biochemical activities are dependent on the presence of a functional kinase.RhaK affects both the interconversion and the whole-cell transport of rhamnose (Tables 2 and 5). From these data, it is not clear if transport and interconversion are interdependent. To resolve this, we constructed site-directed mutants of RhaK that should have compromised biochemical activity and determined whether these mutants could transport rhamnose.

Database searches using BLASTP (3) as well as CDART (20) clearly recognized RhaK to be a sugar kinase (E value = 8 × 10⁻⁴⁶). Searching the RhaK sequence against the Prosite database (14) for conserved functional domains yielded very few hits with convincing scores. The only conserved domain that was easily recognizable was an ATP/GTP binding site motif A (P loop; Prosite accession no. PS00017) located at amino acids 10 to 17. This motif consists of the sequence A/G-X(4)-G-K-S/T (51). The lysine from this motif is thought to interact with the beta or gamma phosphate of ATP (51). Since RhaK activity is dependent on ATP (data not shown), the highly conserved lysine residue within the phosphate-binding loop (P loop) of RhaK was targeted for mutagenesis. The large, positively charged lysine residue (amino acid 16) within the P loop was changed to either an aspartate or a methionine. Both of these substitutions in the conserved portion of the P loop should result in an abolition of ATP-dependent kinase activity. The mutated copies of the gene were cloned into the broad-host-range vector pRK7813, sequenced, and introduced into strain Rlt146, generating strains Rlt206 and Rlt207.

In both cases, the mutant strains Rlt206 and Rlt207 were unable to grow on minimal medium supplemented with rhamnose as the sole carbon source (Table 6). These strains were also unable to import labeled rhamnose in radioactive transport assays, and both had transport rates of <0.05 nmol/g free weight/min (Table 6). Additionally, neither mutant had any observable kinase activity in the radioactive chromatography assay (Table 6). Moreover, Western blot analysis of the P-loop mutants showed that each construct was capable of being translated into a protein of the correct size (data not shown). Together, these data suggest that transport of rhamnose by the rhamnose ABC transporter is dependent upon biochemically active RhaK.

Phylogenetic and domain analysis of RhaK.Since the only motif that we were able to identify was the P loop, it was thought that a phylogenetic analysis of kinases with high amino acid identities to RhaK might be useful in identifying other sugar kinases that may affect transport function. It was reasoned that this analysis might also identify regions that are conserved and thus may be necessary for either transport or kinase activity.

BLASTP analysis of RhaK identified RHE_PE00284 and pRL110408 as having E values of zero. In addition, they are carried in operons that are nearly identical to that of Rlt100. Although it is not possible to identify orthologues by a reciprocal best-match method because genomic sequence data for Rlt100 do not exist (42), it is highly probable that these genes and their operons are orthologous. Since these two proteins are so closely related, they were not used in a phylogenetic analysis of 35 related kinases (Fig. 2). However, searches for homologues of RhaK in the individual proteomes of Agrobacterium tumefaciens, M. loti, and S. meliloti by using BLASTP showed that the proteins delineated by the shaded box in Fig. 2 are the best matches in the respective organisms. Moreover, an examination of the flanking sequences of the respective genes showed an identical arrangement of genes and putative transcripts at this locus in each organism. In addition, insertion mutations isolated in the S. meliloti gene Smc03003 (GenBank accession no. NP_384741.1) also cause an inability to grow by utilizing rhamnose as a sole carbon source (N. Poysti, J. S. Richardson, and I. J. Oresnik, unpublished data). Using this protein sequence to carry out a reciprocal best-hit analysis against the genomes of M. loti and A. tumefaciens showed that the delineated area represents proteins that are orthologous to the rhaK product and, more than likely, are involved in rhamnose catabolism in these organisms (Fig. 2).

• Open in new tab
• Download powerpoint

FIG. 2.

RhaK phylogenetic tree based on predicted kinase sequences. Related kinase sequences were identified using BLASTP, and the tree was constructed as described in Materials and Methods. Sequence accession numbers for the predicted sequences are listed next to the corresponding strains. Bootstrap values of <700 are not shown.

To facilitate the identification of regions that may play an important role in the function of this protein, the most closely related proteins were aligned using both CLUSTALX and PRALINE (54). Whereas the former program aligns on the basis of statistical probability, using the entire protein, PRALINE utilizes a PSI-BLAST step that constructs an alignment by recognizing localized regions of identity (54). Both alignments gave congruent results showing that there are regions of conservation.

CDART analysis clearly defined RhaK as a sugar kinase that aligns with the FGGY-type sugar kinases. One of the better-characterized members of this family of proteins is E. coli GlpK. To better understand RhaK, the protein sequences of RhaK and GlpK were aligned (data not shown). Utilizing GlpK crystal structure data, we were able to putatively identify residues that were important for substrate and ATP binding, as well as active-site residues (29, 44). Consistent with what is known about GlpK, where the active-site aspartate is found in a hydrophobic pocket (44), the RhaK aspartate 246 is also found in a highly hydrophobic region of the protein and aligns with the GlpK active-site aspartate (data not shown). Also, the majority of the residues necessary for ATP/ADP binding by GlpK are also conserved in RhaK and its homologues.