| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Molecular Biology and Biotechnology, University of Sheffield, Firth Court, Western Bank, Sheffield S10 2TN, UK
Correspondence
David J. Kelly
d.kelly{at}sheffield.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mol Membr Biol 21, 51–57], but purified YiaO did not bind L- or D-xylulose as judged by fluorescence spectroscopy, circular dichroism or mass spectrometry. Instead, these techniques showed that a breakdown product of L-ascorbate, 2,3-diketo-L-gulonate (2,3-DKG), binds by a simple one-step mechanism with sub-micromolar affinity. The data provide the first evidence for the existence of ESR-dependent transporters for 2-oxoacids and 2,3-DKG, homologues of which appear to be widespread amongst prokaryotes. The results also underline the utility of direct ESR ligand-binding studies for TRAP transporter characterization
Present address: Department of Biology (Area 10), University of York, PO Box 373, York YO10 5YW, UK.
Present address: Technology Facility, Department of Biology (Area 15), University of York, PO Box 373, York YO10 5YW, UK.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Quiocho & Ledvina, 1996). ESRs (commonly called periplasmic binding-proteins in Gram-negative bacteria) are invariably found in bacterial or archaeal solute uptake systems in the well-characterized ABC transporter superfamily, which utilize the free energy of ATP hydrolysis to drive solute uptake via a highly conserved energy-coupling subunit, the ATP-binding cassette or ABC protein (Higgins, 2001). Until recently, no bacterial secondary solute transporters were known which possessed an ESR protein. However, there is now extensive evidence that a novel class of secondary transport system exists, the tripartite ATP-independent periplasmic (‘TRAP’) transporters, which use an ESR in the transport mechanism, but do not possess an ABC protein (Forward et al., 1997; Kelly & Thomas, 2001). Biochemical evidence indicates that solute accumulation by such systems is driven not by ATP hydrolysis but by an electrochemical ion gradient. The discovery of TRAP transporters (Forward et al., 1997) was important in demonstrating that the use of an ESR in solute transport is not exclusive to ABC systems, and indeed there now appear to be other families of secondary transporters like the ‘tripartite tricarboxylate transporters' (TTT), which also use ESR proteins in the transport mechanism (Winnen et al., 2003).
The first TRAP transporter to be identified and fully characterized was the Dct system from Rhodobacter capsulatus, which mediates uptake of the C4-dicarboxylates malate, succinate and fumarate (Forward et al., 1997). In this system, DctP is the ESR, DctQ is a four-helix membrane protein of unknown function (Wyborn et al., 2001) and DctM is a twelve-helix membrane protein which is believed to mediate solute transport across the cytoplasmic membrane. Microbial genome sequencing has now revealed the presence of hundreds of TRAP transport systems in a wide range of bacteria and archaea but not eukaryotes (Kelly & Thomas, 2001). Like the DctPQM system, all TRAP transporters contain two integral membrane domains, which can be on separate proteins or fused into a single polypeptide, in addition to the ESR component (Kelly & Thomas, 2001). The DctM homologues have been shown to be members of the ion-transport superfamily (Rabus et al., 1999; Prakash et al., 2003), which contains many families of well-characterized secondary transporters, and it has been proposed that TRAP transporters evolved by the recruitment of an ESR to an ancestral secondary transport protein (Rabus et al., 1999).
Despite the wealth of bioinformatic evidence for the widespread distribution of TRAP transporters in prokaryotes, the substrates and thus physiological role of only a few of these systems have been elucidated. One other C4-dicarboxylate transporter that has a similar function to the R. capsulatus DctPQM system in the uptake of succinate as a carbon source for growth has been characterized in Wolinella succinogenes (Ullmann et al., 2000). A TRAP transporter involved in movement of the compatible solute ectoine into cells of Halomonas elongata has been characterized in some detail (Grammann et al., 2002). The ESR component, TeaA, has been purified and shown to bind ectoine and hydroxyectoine (Tetsch & Kunte, 2002). A TRAP transporter in Synechocystis has been shown unequivocally to be a glutamate uptake system (Quintero et al., 2001) and is of interest as the ESR is a homologue of GlnH, a glutamine-binding protein from an ABC system. Recently, two studies have shown sialic acid to be a substrate for a TRAP transporter from the pathogen Haemophilus influenzae (Allen et al., 2005; Severi et al., 2005). Biochemical evidence suggests that an ESR-dependent secondary glutamate transporter exists in Rhodobacter sphaeroides (Jacobs et al., 1996), although this has not yet been confirmed by sequence analysis to be a member of the TRAP transporter family. Finally, the substrate for the single TRAP transporter in Escherichia coli (YiaMNO) has been reported to be L-xylulose (Plantinga et al., 2004), although we show here that this is likely to be incorrect.
Ligand-binding kinetics have been determined for several ESRs from ABC systems, and the majority of those studied appear to operate by a simple association of protein and ligand, involving the closure of a binding cleft by the so-called ‘Venus fly trap’ mechanism (Quiocho & Ledvina, 1996). The transition from the open-unliganded to closed-liganded forms of several of these ESRs has been observed by stopped-flow fluorescence spectroscopy (Miller et al., 1980, 1983), and in all cases a linear concentration dependence of the association rate constant has been found. However, the R. capsulatus DctP protein is kinetically distinct: the rate of ligand-induced fluorescence change was found to decrease in a hyperbolic fashion as the concentration of ligand was increased (Walmsley et al., 1992a, b). This is consistent with a slow isomerization of the protein, with the ligand binding to only one form of an equilibrium mixture of two pre-existing protein conformations:
 |
CU represents the non-binding (closed) unliganded conformation, OU the binding (open) unliganded form, L the ligand and CL the closed liganded form. As the rate of ligand binding is much faster than the rate of the conformational change, the rate constant for the fluorescence change decreases with ligand concentration in a hyperbolic manner (Walmsley et al., 1992a). In the absence of ligand, the position of the CU/OU equilibrium strongly favours the CU form of DctP, while with those other ESRs studied the position of this equilibrium appears to strongly favour the OU form.
We have selected two distinct TRAP-ESR proteins for purification and ligand-binding analysis by intrinsic tryptophan fluorescence spectroscopy, circular dichroism and mass spectrometry. New ligands have been identified from this analysis, which has allowed insights into the physiological function of the cognate TRAP transporters.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
, while strains BL21(DE3) pLysS and ‘Tuner’ (Novagen) were used for expression of pET21 derivatives. E. coli was routinely grown aerobically in Luria medium (broth or agar), containing antibiotics when required. R. capsulatus was grown anaerobically in RCV minimal salts medium (Weaver et al., 1975) with the modifications of Hillmer & Gest (1977). Cultures (30 ml) were grown in McCartney bottles filled to capacity with RCV medium and incubated at 30 °C for 48 h, 30 cm from a 60 W light bulb. Antibiotics were used at the following final concentrations: ampicillin 100 µg ml–1, chloramphenicol 30 µg ml–1.
Preparation of periplasmic extracts.
Periplasmic proteins were released using a modified osmotic shock procedure. Cells were harvested at 4700 g for 15 min at room temperature; the cell pellet was resuspended in 5 ml STE (0·5 M sucrose, 10 mM Tris/HCl, pH 8·0, and 2 mM EDTA, pH 8·0). Lysozyme was added to 1 mg ml–1 and the suspension was incubated for 30 min on ice, after which time 15 ml distilled water was added and the suspension was spun at 18 000 g for 20 min at 4 °C. The supernatants were kept at –20 °C until needed.
Preparation and sequencing of DNA.
Plasmid DNA for screening clones and sequencing was routinely isolated using anion-exchange resin spin-columns (QIAGEN), according to the manufacturer's instructions. Double-stranded sequencing of plasmid DNA was carried out using Applied Biosystems Taq DyeDeoxy terminator cycle sequencing reactions analysed on an ABI model 373A automated DNA sequencer. Primers were obtained commercially. Total genomic DNA was extracted using a modified SDS lysis procedure (Marmur, 1961). Restriction endonucleases, T4 DNA ligase and Taq polymerases were purchased from Promega, and Pfu or Pwo polymerase from Stratagene, and used according to the manufacturer's instructions.
Amplification and cloning of genes encoding ESRs.
DNA sequences of the relevant ESR genes were obtained from publicly available databases. The R. capsulatus sequences were obtained from Integrated Genomics (http://ergo.integratedgenomics.com/ERGO/). The genes encoding the ESR proteins were amplified from genomic DNA by PCR using primers that introduced NdeI and BamHI or EcoRI restriction sites at the 5' and 3' end of the gene, respectively. The primers used to amplify the RRC01191 gene were 1191-F (5'-GGAATTCCATATGGATCGTCGTTCATTTCTGACC-3') and 1191-R (5'-CGGGATCCTCAGATCTTGCCGTTGCG-3'). Restriction sites for NdeI and BamHI are underlined. The primers used to amplify yiaO were yiaO-F (5'-GAGACATATGAAATTACGCTCTGTAACCTACG-3') (NdeI site underlined) and yiaO-R (5'-TTGAATTCTTATTGCACCTCATCCACGG-3'), the latter introducing an EcoRI site (underlined). PCR was performed with Pwo or Pfu polymerase and the products cloned into pET21b(+). The sequences of the genes cloned in this manner were determined by automated methods and found to be identical to those reported in the published genome sequences.
Overexpression and purification of ESR proteins.
Cells were grown overnight in 2 litre batches of LB medium with ampicillin, and induced with 1 mM IPTG when cells had reached an OD650 of 0·5. After overnight growth at 25 °C, cells were harvested by centrifugation, resuspended in 20 ml 50 mM Tris/HCl pH 8, broken by two passages through a French press and centrifuged at 12 000 g for 15 min at 4 °C to remove cell debris and unbroken cells. Aliquots (10 ml) of this extract were applied to a 50 ml DEAE-Sepharose column that had been equilibrated with 50 mM Tris/HCl pH 8 and connected to a BioLogic HP protein purification system (Bio-Rad). The bound protein was eluted using a linear 0–500 mM NaCl gradient and 7 ml fractions were collected. Both proteins required further purification on a Q-Sepharose column, which was done under the same conditions. If necessary, proteins were subjected to a final Superdex S75 gel filtration step using a Pharmacia FPLC system. The column was equilibrated and eluted with 50 mM Tris/HCl pH 8 containing 200 mM NaCl. Pure proteins were dialysed extensively against 50 mM Tris/HCl pH 8, concentrated by treatment with Aquacide (Calbiochem) and stored in aliquots at –20 °C. N-terminal sequences were determined by the automated Edman degradation method, carried out by Dr A. J. G. Moir, University of Sheffield, UK.
Protein analysis by mass spectrometry.
Protein samples were first dialysed into 5 mM Tris pH 8 or water before dilution into solvents for analysis. Electrospray mass spectrometry was performed on an API Qstar mass spectrometer using an ionspray source. For mass determinations, samples were made up in 0·1 % formic acid. To study non-covalent protein–ligand interactions, samples were made up in 20 % (v/v) methanol/water by dilution from stock solutions of protein and ligand in water, and infused at between 1 and 5 µl min–1. They were run within 1 h of preparation. Multiply charged protein signals were observed between 1800 and 3500 m/z under these conditions (signals at much lower m/z values were observed for protein in 1 : 1 : 0·001 acetonitrile/water/formic acid mixtures). Attempts to use 100 % aqueous solutions gave very weak signals. Errors in the mass determinations were approximately 1 in 104 over the range used; calibration was with caesium iodide and pheromone inhibitor peptide (sequence: ALILTLVS; Bachem). Raw m/z data were deconvoluted to mass spectra using the Bayesian Protein Reconstruction routine in the manufacturer's BioAnalyst software. Sodium adducts seen in the spectra were attributed to carry-over from the protein solution.
Steady-state and stopped-flow fluorescence spectroscopy.
Slow-timescale fluorescence spectroscopy was performed using a Hitachi F-2500 spectrofluorimeter with an excitation wavelength of 295 nm (slit width 5 nm) and an emission wavelength of 330 nm (slit width 10 nm). The assay mixture consisted of 0·1 µM protein in 1·5 ml 50 mM Tris/HCl pH 8. The sample cuvette was maintained at 25 °C in the spectrofluorimeter housing and the assay mixture was continuously stirred. Fluorescence changes upon the addition of ligands were monitored until the fluorescence change stabilized. In some experiments, protein photobleaching, which caused baseline drift, was noted. Hence, ligand titrations were performed by removing the cuvette between additions of ligand, which minimized this effect. All fluorescence values were corrected for dilution. For calculation of Kd values, data were averaged from at least three titrations and the hyperbolic curve-fitting algorithms in SigmaPlot used to analyse the data. Rapid reactions were followed using an Applied-Photophysics SX.18MV stopped-flow spectrofluorimeter, routinely operated at 23 °C. For work with YiaO, the temperature was lowered to 10 °C. Tryptophan fluorescence was excited at 295 nm, selected by focusing a 150 W xenon arc lamp onto a f/3·4 monochromator, and the emitted light selected with a WG320 Schott filter, positioned in front of the observation photomultiplier tube. Data were recorded with oversampling enabled and analysed by non-linear regression using the Applied Photophysics software. For the RRC01191 protein, a linear phase that was due to photobleaching was removed from the analysis. At least three stopped-flow traces were averaged prior to non-linear regression analysis using the program KaleidaGraph (Synergy Software).
Circular dichroism.
CD spectra were obtained using a Jasco J810 instrument operating at 20 °C. A 1 cm path-length cuvette was used with a 2 ml volume of purified protein (1 µM final concentration in 50 mM Tris/HCl buffer pH 8). Ligands were added in small volumes and the CD change recorded and corrected for dilution.
Analysis of protein sequences and homologies.
Sequences of TRAP ESR genes were obtained from the ERGO (Overbeek et al., 2003) and UniProt databases. The contigs from the unfinished R. capsulatus genome sequence were analysed using the Artemis package (Berriman & Rutherford, 2003). Sequences for ESR proteins associated with other genes for TRAP transporters and obvious homologues of these were collected from the NCBI database. For phylogenetic analyses, sequences were aligned in CLUSTAL X (Thompson et al., 1997) and the output file used in PHYLIP (Felsenstein, 1989) to produce a distance matrix tree, which was viewed in TREEVIEW (Page, 1996). Signal sequences were identified using SignalP (Bendtsen et al., 2004; Nielsen et al., 1999).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). All these proteins show a statistically significant degree of sequence similarity to the archetypal DctP protein from R. capsulatus, which is contained in cluster 10 along with some other known C4-dicarboxylate binding proteins. The majority of the proteins in the other clusters are uncharacterized with respect to ligand binding. The clustering observed in Fig. 1 is similar to that obtained in a recent phylogenetic analysis (Prakash et al., 2003) of the large membrane components (DctM homologues) of TRAP transporters. We chose to study two proteins (shown boxed in Fig. 1), from clusters 8 and 12, for which there were some previous suggestions of possible ligands. These were the R. capsulatus RRC01191 protein, closely related (81 % amino-acid identity) to the R. sphaeroides SmoM protein (Rsph1195) that has been predicted to bind sorbitol and mannitol (Stein et al., 1997), and the E. coli YiaO protein, which has been suggested to be part of an L-xylulose TRAP transporter (Plantinga et al., 2004, 2005).
|
; Horlacher et al., 1998).
The tryptophan fluorescence emission maximum of RRC01191 was centred at 331·5 nm. Addition of 200 µM sorbitol or mannitol to 0·1 µM protein resulted in no change in fluorescence intensity at the emission maximum, and thus we found no evidence that either sorbitol or mannitol could bind to the protein. Closer examination of the sequence information deposited to support the data in Stein et al. (1997) revealed that the evidence for the function of the R. sphaeroides SmoM protein as a sorbitol/mannitol- binding protein is indirect. There is a 515 bp gap between the end of the confirmed smo operon, encoding known sorbitol/mannitol catabolic genes, and the downstream smoM gene. The region immediately downstream of the smoM gene has been sequenced during the R. sphaeroides genome sequencing project, revealing two genes, transcribed convergently with the smoM gene, that are homologous to dctQ and dctM (Choudhary et al., 1999). Hence these three genes are likely to form a functional TRAP-transport system, but not necessarily related to either sorbitol or mannitol uptake.
The apparent lack of binding of sorbitol and mannitol to RRC01191 prompted a bioinformatic approach to derive hypotheses about other potential ligands. The organization of the genes encoding several SmoM-type TRAP transporters is shown in Fig. 1(b)
. We identified pyruvate as a potential ligand from two observations. First, the genes that are divergently transcribed from the dctQ/dctM homologues in R. capsulatus encode an acetolactate synthase. This heterodimeric enzyme catalyses the first step in valine biosynthesis and condenses two molecules of pyruvate to synthesize -acetolactate (Dailey & Cronan, 1986). It can also combine pyruvate and 2-oxobutyrate to form 2-aceto-2-hydroxybutyrate. Secondly, in several bacteria including Brucella melitensis, Vibrio cholerae and Agrobacterium tumefaciens, the smoM orthologues are genetically linked to two-component sensor–regulator genes (Fig. 1b) which encode homologues of the MctRS proteins from Rhizobium leguminosarum encoded by the mctPRS operon (Hosie et al., 2002). MctP is a secondary transport protein that mediates the transport of monocarboxylates, including pyruvate. Therefore, the ability of pyruvate and 2-oxobutyrate to bind to RRC01191 was determined, using tryptophan fluorescence. Addition of 200 µM pyruvate to 0·1 µM protein caused a large quench of the fluorescence of around 23 %, which could be titrated to give a Kd value of 3·4 µM (Table 1). Addition of excess 2-oxobutyrate also caused a large quench (about 28 %) in the fluorescence of the protein, and titration with this ligand revealed a much higher affinity, with a Kd value of 0·32 µM (Table 1).
|
with their respective dissociation constants. Only a monocarboxylate 2-oxoacid structure with an aliphatic backbone allows binding to the protein with physiologically relevant Kd values. The 2-oxo and 1-carboxyl groups are essential, as compounds lacking either or both of these functional groups did not give any fluorescence change. A methyl group bonded to the carbonyl function, as in pyruvate, allows binding, while a hydrogen in this position (as in glyoxylate) does not. Dicarboxylic 2-oxoacids like oxaloacetate and 2-oxoglutarate did not interact with the protein. However, a progressive increase in the chain length of the aliphatic backbone of the monocarboxylate 2-oxoacids increased the ligand-binding affinity, with 2-oxovalerate binding with a Kd value almost an order of magnitude lower than 2-oxobutyrate. Branched-chain aliphatic 2-oxoacids like 2-oxoisovalerate, 2-oxoisocaproate and 2-oxo-3-methylvalerate also bound to the protein, with intermediate Kd values (Table 1).
Mechanism of ligand binding to RRC01191
The mechanism of ligand binding to RRC01191 was investigated using stopped-flow fluorescence spectroscopy with pyruvate. Over the range of concentrations tested, the stopped-flow data could be fitted to a double exponential process, indicative of two optically active phases in the binding mechanism. The first phase had a strong concentration dependence (Fig. 2), with a similar inverted hyperbola as observed previously for DctP (Walmsley et al., 1992a). The second phase was concentration independent, and is most likely due to an optically active but ligand-independent isomerization process subsequent to ligand binding, as has been observed in other systems (e.g. Walmsley et al., 2001). For phase one the k1 and k–1 parameters were 0·4±1·3 s–1 and 32·9±6·6 s–1, respectively. Two phases were also consistently observed using 2-oxobutyrate as the ligand in stopped-flow experiments. The rate of the first phase showed a similar inverse concentration dependence to that seen with pyruvate, but the variation in replicates was too large to yield reliable rate constants (data not shown).
|
; Ibanez et al., 2000). YiaP (LyxK) has been shown to be a kinase that can phosphorylate L-xylulose (Sanchez et al., 1994) and some evidence has been published that YiaMNO encodes an L-xylulose transporter (Plantinga et al., 2004). However, in a study of L-ascorbate fermentation by E. coli, Yew & Gerlt (2002) showed that purified LyxK could also phosphorylate 3-keto-L-gulonate (an intermediate of L-ascorbate catabolism) with kinetic constants comparable to those with L-xylulose. Moreover, they demonstrated that the enzyme encoded by yiaK is a reductase that can form 3-keto-L-gulonate from 2,3-diketo-L-gulonate (2,3-DKG), the latter produced from L-ascorbate via L-dehydroascorbate. Yew & Gerlt (2002) thus suggested that YiaMNO might form a 2,3-DKG transporter that could supply YiaK with its substrate, which would be reduced to 3-DKG, the substrate for LyxK.
We sought to clarify the conflicting data on the function of the YiaMNO TRAP transporter, by determination of the ligand-binding properties of YiaO. The protein was overexpressed and purified to homogeneity by two anion-exchange steps followed by gel filtration. The molecular mass of the pure protein was determined by mass spectrometry to be 33 468·8 Da, which is identical (within the error of the instrument, ±2 Da), to that calculated for the deduced processed form of the protein (33 469·1). When excited at 295 nm, YiaO showed a strong fluorescence emission maximum at 330 nm. Fig. 3 shows a series of experiments in which the change in the steady-state tryptophan fluorescence at 330 nm was measured after the addition of various potential ligands for this protein. The addition of excess (200 µM) L- or D-xylulose did not result in any discernible change in fluorescence emission from the protein (Fig. 3a), while the addition of even a low (5 µM) concentration of 2,3-DKG resulted in an easily measurable fluorescence quench (Fig. 3b). The maximal fluorescence change observed was about 7 % at 200 µM 2,3-DKG. The 2,3-DKG concentration dependence of the fluorescence change is shown in Fig. 3(c). The calculated Kd value was 0·77±0·19 µM, indicating that 2,3-DKG binds to YiaO with reasonably high affinity.
|
) or L-xylulose (Fig. 3e) did not result in any fluorescence change indicative of ligand displacement. However, when 5 µM 2,3-DKG was added to 0·1 µM YiaO which had been pre-incubated with either 200 µM D-xylulose (Fig. 3f) or L-xylulose (Fig. 3g), an immediate fluorescence quench was observed which was of a similar magnitude to that seen when the same concentration of 2,3-DKG was mixed with the protein alone (Fig. 3b). These data indicate that neither L- nor D-xylulose is a ligand for YiaO.
Mass spectrometry provides evidence that a hydrated form of 2,3-DKG is the ligand for YiaO
A non-optical method that can give additional information about ligand binding to YiaO is mass spectrometry. The purified protein was first examined under denaturing conditions to get an accurate mass estimate, which was determined to be 33 468·8, corresponding to the processed form of the protein (33 469·1). Under less denaturing conditions, in a 20 % methanol/water buffer, the signal from the protein was also still sufficient to be able to accurately and unambiguously determine its mass, even though there is less ionization (Fig. 4a). As well as the major peak, additional peaks, spaced at 23 Da intervals, were also present which probably correspond to sodium adducts on the protein. Identical conditions were then used to examine the mass of the protein after the addition of 50 µM L-xylulose. This did not change the mass spectrum from that seen with the protein alone, indicating no binding of the ligand to the protein in these conditions (data not shown). When the experiment was repeated with the addition of 2,3-DKG, a new species was observed with a mass of 33658·6, a larger peak at a slightly greater mass of 33678·3 and a similar tail of sodium adducts of decreasing intensity (Fig. 4b). The observed mass changes of 191·1 to the first peak and 210·8 to the second are consistent with the binding of 2,3-DKG (191 mass units for the negative ion) and a hydrated gem-diol form of 2,3-DKG (210 mass units), respectively (Deutsch, 1998a). Upon mixing the protein with both L-xylulose and 2,3-DKG, the same mass changes as seen with 2,3-DKG alone were obtained (data not shown). Taken together, the mass spectrometry data support the fluorescence data in showing that L-xylulose is incapable of binding to YiaO, whereas 2,3-DKG does bind.
|
with 2,3-DKG. A number of studies have shown that 2,3-DKG is a spontaneous breakdown product of L-ascorbate and dehydroascorbate (Deutsch, 1997, 2000), which forms in aqueous solutions at neutral pH. Consistent with this, when L-ascorbate or dehydroascorbate solutions were made fresh and used immediately in the experiment, no additional peaks in the mass spectrum were observed. We thus conclude that YiaO specifically binds 2,3-DKG and is unable to bind either L-ascorbate or dehydroascorbate.
Ligand-induced changes in YiaO studied by circular dichroism
Further evidence that YiaO specifically binds 2,3-DKG, but neither xylulose nor ascorbate, was obtained from circular dichroism studies. The room temperature CD spectrum of YiaO in the far-UV region was characterized by troughs centred at 208 nm and 220 nm typical of a protein with a large -helical content (Fig. 5a). Addition of L- or D-xylulose did not result in any significant change in the CD molar ellipticity in this region of the spectrum. In contrast, addition of excess 2,3-DKG led to a decrease in the ellipticity of the 220 nm trough indicative of changes in the secondary structure of the protein (Fig. 5a). The 2,3-DKG-dependent change in the CD signal at 220 nm could be titrated (Fig. 5b) to give a Kd value of 1·29±0·17 µM, very similar to that obtained from the fluorescence quench. CD spectroscopy proved useful to determine if L-ascorbate itself was a ligand for YiaO, as the strong absorbance of ascorbate at 260 nm interferes with fluorescence measurements. However, as with xylulose, no significant changes in the far-UV spectrum were observed (Fig. 5a), suggesting that L-ascorbate is not a ligand for the protein.
|
(a) shows stopped-flow traces obtained at 10 °C with two 2,3-DKG concentrations, 1 µM (trace 2) and 10 µM (trace 3), where in each case the quench data could be fitted to a single exponential with a linear drift. The latter is due to photobleaching, which is more apparent in trace 1, where protein was pushed against buffer. Fig. 6(b) shows the dependence of kobs on 2,3-DKG concentration at 10 °C. In contrast to the kinetic behaviour of RRC01191, a positive linear dependence was observed, consistent with a simple association of ligand and protein. The calculated association rate constant (KA[L]) was 2·4x107 M–1 s–1, and the dissociation rate constant obtained from the intercept of Fig. 6(b) was 6·4 s–1.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
), which has also been observed for the orthologous protein in Haemophilus influenzae (Kolker et al., 2004). This ESR does indeed bind sialic acid with high affinity (Severi et al., 2005) and has been shown to be part of a sialate TRAP transporter (Allen et al., 2005; Severi et al., 2005), extending the substrate range of the TRAP transporters to the sugar acids. Bioinformatic analyses based on gene context suggest that ligands for two additional TRAP transporters are 4-chlorobenzoate and protocatechuate (Chae et al., 2000; Contzen et al., 2001; Kelly & Thomas, 2001). Although these observations may suggest a common function for TRAP transporters in the uptake of carboxylic acids, the discovery of genetic linkage of the genes for a TRAP transporter with those for taurine catabolism in R. sphaeroides suggests a wider range of organic ligands than previously suspected (Bruggemann et al., 2004), although this has yet to be confirmed experimentally.
The identification of a distinct subfamily of closely related TRAP-ESR proteins that contain the SmoM protein from R. sphaeroides (Fig. 1a) is significant, as SmoM has been suggested to be involved in the uptake of sorbitol and mannitol (Stein et al., 1997), implying the existence of a novel type of sugar-alcohol transporter. However, we have demonstrated that the closely related R. capsulatus orthologue binds 2-oxoacids but neither sorbitol nor mannitol. In R. capsulatus, the TRAP transporter associated with this type of ESR is genetically linked to acetolactate synthase (Fig. 1b), and we suggest that the physiological role of this transporter is to supply intermediates in the synthesis of valine and isoleucine. Since R. capsulatus is not auxotrophic for branched-chain amino-acids, it is possible that the transporter plays a role in supplementing intracellular pools of the pathway intermediates from external sources. Determination of the most physiologically important substrate will require phenotypic analysis of transport mutants. This group of TRAP transporters (clade 12; see Fig. 1a) is currently one of the largest and may reflect a more widespread but hitherto unsuspected role for 2-oxoacid uptake in bacterial biosynthetic pathways. The ligand-binding mechanism of RRC01191 appears to be similar to that of DctP (Walmsley et al., 1992a), although the rate constants for the forward transition between the closed unliganded and open unliganded form are rather lower (less than 1 s–1) than for DctP. Many of the genes encoding orthologues of RRC01191 are called smoM and are annotated as encoding sorbitol/mannitol-binding proteins, but this now seems unlikely in the light of our data. A phylogenetic and bioinformatic analysis of the SmoM family of TRAP transporters revealed that in Synechocystis the gene encoding the SmoM ESR has been displaced away from the gene encoding the membrane subunits by a different ESR and is now distantly located relative to the gene encoding the membrane components (Kelly & Thomas, 2001). In a non-orthologous gene displacement event, the expected ESR has been replaced by an ESR normally found associated with ABC transporters, which is homologous to the glutamine ESR family (GlnP). Experimental evidence from Flores and colleagues (Quintero et al., 2001) has demonstrated that this TRAP transporter mediates glutamate transport rather than having a function in 2-oxoacid transport. This raises interesting questions about how ligand-binding specificity is imposed on the TRAP transporters and the important role of the ESR in determining this property.
The substrate for the single E. coli K-12 TRAP transporter (YiaMNO) has been proposed by Plantinga et al. (2004) to be L-xylulose. However, although these authors showed that purified YiaO apparently bound L-xylulose, their assay method was indirect, involving measurement of the residual unbound ligand. With very high (100 µM) protein concentrations and 50 µM L-xylulose, only 9 % of the ligand was removed from solution by their YiaO preparation. Given that the authors had purified the YiaO protein, it is notable that they did not use any biophysical methods to examine ligand binding. Using a range of such techniques, we could find no evidence that L-xylulose was able to bind, but we were able to unequivocally demonstrate that YiaO specifically binds 2,3-DKG. Significantly, Plantinga et al. (2004) did not use 2,3-DKG directly in any of their experiments. In addition, they performed only indirect transport assays for L-xylulose uptake in intact cells, and the difference in rates between a yiaMNO deletion strain and its isogenic parent was minimal. In a separate study, Plantinga et al. (2005) also found that L-xylulose did not induce expression of the yiaM gene, and that deletion of the yiaMNO genes in strain MC4100 led to complex changes in growth, salt tolerance and biofilm formation that are difficult to explain but seemingly unrelated to L-xylulose transport. Hence, we conclude that 2,3-DKG and not xylulose is the physiological substrate for the YiaMNO transporter, as predicted by Yew & Gerlt (2002). Unfortunately, 2,3-DKG is relatively unstable and is unavailable commercially in a radiolabelled form, so that future transport and growth studies with the appropriate mutants will not be trivial to perform.
As ascorbate (vitamin C) is absorbed from the gut, it is likely that dehydroascorbate and hence 2,3-DKG will also be present in this environment (Bode et al., 1990; Cioffi et al., 2000; Deutsch, 1998a, b). Although no reliable estimates of its concentration in vivo are known, the use of a high-affinity TRAP system would allow scavenging of this common breakdown product of ascorbate. It is apparent from Fig. 1(a) that YiaMNO homologues are conserved in several important pathogens including Salmonella, Haemophilus and Pasteurella species, which may imply a significant general role for gulonate transport and metabolism in vivo. Expression of the yiaK operon in wild-type E. coli K-12 cells is repressed by the yiaJ gene product, a member of the IclR family. As yet, the inducer for this operon has not been identified (Ibanez et al., 2000), and we suggest that 2,3-DKG, the identified substrate for YiaO, could be a potential inducer. Interestingly, in a separate study, strains of E. coli were isolated that had reduced fitness when grown on M9 glucose medium (Badarinarayana et al., 2001). Some of these strains had mutations in yiaJ, which suggests that unregulated expression of the yiaK operon is detrimental to growth under these conditions.
Recently, Zhang et al. (2003) have described an ascorbate transporter in E. coli as being the products of the sgaTBA genes; deletion of these genes led to a complete loss of ascorbate transport in K-12 strains (Zhang et al., 2003). These data are complementary to those in this study, where we have shown that ascorbate itself is not a ligand for YiaO protein. Thus E. coli appears to contain separate transporters for both ascorbate and its breakdown product 2,3-DKG.
Finally, the kinetic behaviour of the two proteins characterized in this study was not identical, reflecting variations in the position of the equilibrium between the closed and open forms of the proteins. A study of the effect of ionic strength and pH on the DctP CU/OU equilibrium suggested that one or more salt-bridges were involved in stabilizing the closed unliganded conformation (Walmsley et al., 1992b). Relatively few proteins appear to undergo an isomerization before ligand binding, a recent example being the discovery of isomerization of antigen-binding sites in antibodies (James et al., 2003). The structural basis for the observed differences in kinetic behaviour will partly be resolved by future crystallographic studies of these and other TRAP ESR proteins.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Badarinarayana, V., Estep, P. W., III, Shendure, J., Edwards, J., Tavazoie, S., Lam, F. & Church, G. M. (2001). Selection analyses of insertional mutants using subgenic-resolution arrays. Nat Biotechnol 19, 1060–1065.[CrossRef][Medline]
Badia, J., Ibanez, E., Sabate, M., Baldoma, L. & Aguilar, J. (1998). A rare 920-kilobase chromosomal inversion mediated by IS1 transposition causes constitutive expression of the yiaK-S operon for carbohydrate utilization in Escherichia coli. J Biol Chem 273, 8376–8381.
Bendtsen, J. D., Nielsen, H., von Heijne, G. & Brunak, S. (2004). Improved prediction of signal peptides: SignalP 3.0. J Mol Biol 340, 783–795.[CrossRef][Medline]
Berriman, M. & Rutherford, K. (2003). Viewing and annotating sequence data with Artemis. Brief Bioinform 4, 124–132.
Bode, A. M., Cunningham, L. & Rose, R. C. (1990). Spontaneous decay of oxidized ascorbic acid (dehydro-L-ascorbic acid) evaluated by high-pressure liquid chromatography. Clin Chem 36, 1807–1809.
Bruggemann, C., Denger, K., Cook, A. M. & Ruff, J. (2004). Enzymes and genes of taurine and isethionate dissimilation in Paracoccus denitrificans. Microbiology 150, 805–816.
Chae, J. C., Kim, Y., Kim, Y. C., Zylstra, G. J. & Kim, C. K. (2000). Genetic structure and functional implication of the fcb gene cluster for hydrolytic dechlorination of 4-chlorobenzoate from Pseudomonas sp. DJ-12. Gene 258, 109–116.[CrossRef][Medline]
Choudhary, M., MacKenzie, C., Mouncey, N. J. & Kaplan, S. (1999). RsGDB, the Rhodobacter sphaeroides Genome Database. Nucleic Acids Res 27, 61–62.
Cioffi, N., Losito, I., Terzano, R. & Zambonin, C. G. (2000). An electrospray ionization ion trap mass spectrometric (ESI-MS-MSn) study of dehydroascorbic acid hydrolysis at neutral pH. Analyst 125, 2244–2248.[CrossRef][Medline]
Contzen, M., Burger, S. & Stolz, A. (2001). Cloning of the genes for a 4-sulphocatechol-oxidizing protocatechuate 3,4-dioxygenase from Hydrogenophaga intermedia S1 and identification of the amino acid residues responsible for the ability to convert 4-sulphocatechol. Mol Microbiol 41, 199–205.[CrossRef][Medline]
Dailey, F. E. & Cronan, J. E., Jr (1986). Acetohydroxy acid synthase I, a required enzyme for isoleucine and valine biosynthesis in Escherichia coli K-12 during growth on acetate as the sole carbon source. J Bacteriol 165, 453–460.
Deutsch, J. C. (1997). Gas chromatographic/mass spectrometric measurement of ascorbic acid and analysis of ascorbic acid degradation in solution. Methods Enzymol 279, 13–24.[Medline]
Deutsch, J. C. (1998a). Spontaneous hydrolysis and dehydration of dehydroascorbic acid in aqueous solution. Anal Biochem 260, 223–229.[CrossRef][Medline]
Deutsch, J. C. (1998b). Oxygen-accepting antioxidants which arise during ascorbate oxidation. Anal Biochem 265, 238–245.[CrossRef][Medline]
Deutsch, J. C. (2000). Dehydroascorbic acid. J Chromatogr A 881, 299–307.[CrossRef][Medline]
Felsenstein, J. (1989). PHYLIP – Phylogeny Inference Package, version 3.2. Cladistics 5, 164–166.
Forward, J. A., Behrendt, M. C., Wyborn, N. R., Cross, R. & Kelly, D. J. (1997). TRAP transporters: a new family of periplasmic solute transport systems encoded by the dctPQM genes of Rhodobacter capsulatus and by homologs in diverse gram-negative bacteria. J Bacteriol 179, 5482–5493.
Grammann, K., Volke, A. & Kunte, H. J. (2002). New type of osmoregulated solute transporter identified in halophilic members of the bacteria domain: TRAP transporter TeaABC mediates uptake of ectoine and hydroxyectoine in Halomonas elongata DSM 2581T. J Bacteriol 184, 3078–3085.
Higgins, C. F. (2001). ABC transporters: physiology, structure and mechanism – an overview. Res Microbiol 152, 205–210.[Medline]
Hillmer, P. & Gest, H. (1977). H2 metabolism in the photosynthetic bacterium Rhodopseudomonas capsulata: H2 production by growing cultures. J Bacteriol 129, 724–731.
Horlacher, R., Xavier, K. B., Santos, H., Diruggiero, J., Kossman, M. & Boos, W. (1998). Archaeal binding protein-dependent ABC transporter: molecular and biochemical analysis of the trehalose/maltose transport system of the hyperthermophilic archaeon Thermococcus litoralis. J Bacteriol 180, 680–689.
Hosie, A. H., Allaway, D. & Poole, P. S. (2002). A monocarboxylate permease of Rhizobium leguminosarum is the first member of a new subfamily of transporters. J Bacteriol 184, 5436–5448.
Ibanez, E., Campos, E., Baldoma, L., Aguilar, J. & Badia, J. (2000). Regulation of expression of the yiaKLMNOPQRS operon for carbohydrate utilization in Escherichia coli: involvement of the main transcriptional factors. J Bacteriol 182, 4617–4624.
Jacobs, M. H., van der Heide, T., Driessen, A. J. & Konings, W. N. (1996). Glutamate transport in Rhodobacter sphaeroides is mediated by a novel binding protein-dependent secondary transport system. Proc Natl Acad Sci U S A 93, 12786–12790.
James, L. C., Roversi, P. & Tawfik, D. S. (2003). Antibody multispecificity mediated by conformational diversity. Science 299, 1362–1367.
Kapatral, V., Anderson, I., Ivanova, N. & 22 other authors (2002). Genome sequence and analysis of the oral bacterium Fusobacterium nucleatum strain ATCC 25586. J Bacteriol 184, 2005–2018.
Kelly, D. J. & Thomas, G. H. (2001). The tripartite ATP-independent periplasmic (TRAP) transporters of bacteria and archaea. FEMS Microbiol Rev 25, 405–424.[CrossRef][Medline]
Kolker, E., Makarova, K. S., Shabalina, S. & 9 other authors (2004). Identification and functional analysis of ‘hypothetical’ genes expressed in Haemophilus influenzae. Nucleic Acids Res 32, 2353–2361.
Marmur, J. (1961). A procedure for the isolation of deoxyribonucleic acid from microorganisms. J Mol Biol 3, 208–218.
Miller, D. M., Olson, J. S. & Quiocho, F. A. (1980). The mechanism of sugar binding to the periplasmic receptor for galactose chemotaxis and transport in Escherichia coli. J Biol Chem 255, 2465–2471.
Miller, D. M., III, Olson, J. S., Pflugrath, J. W. & Quiocho, F. A. (1983). Rates of ligand binding to periplasmic proteins involved in bacterial transport and chemotaxis. J Biol Chem 258, 13665–13672.
Nielsen, H., Brunak, S. & von Heijne, G. (1999). Machine learning approaches for the prediction of signal peptides and other protein sorting signals. Protein Eng 12, 3–9.
Overbeek, R., Larsen, N., Walunas, T. & 19 other authors (2003). The ERGO genome analysis and discovery system. Nucleic Acids Res 31, 164–171.
Page, R. D. (1996). TREEVIEW: an application to display phylogenetic trees on personal computers. Comput Appl Biosci 12, 357–358.
Plantinga, T. H., van der Does, C., Badia, J., Aguilar, J., Konings, W. N. & Driessen, A. J. M. (2004). Functional characterization of the Escherichia coli K-12 yiaMNO transport protein genes. Mol Membr Biol 21, 51–57.[CrossRef][Medline]
Plantinga, T. H., van der Does, C., Tomkiewicz, D., van Keulen, G., Konings, W. N. & Driessen, A. J. M. (2005). Deletion of the yiaMNO transporter genes affects the growth characteristics of Escherichia coli K-12. Microbiology 151, 1683–1689.
Prakash, S., Cooper, G., Singhi, S. & Saier, M. H., Jr (2003). The ion transporter superfamily. Biochim Biophys Acta 1618, 79–92.[Medline]
Quintero, M. J., Montesinos, M. L., Herrero, A. & Flores, E. (2001). Identification of genes encoding amino acid permeases by inactivation of selected ORFs from the Synechocystis genomic sequence. Genome Res 11, 2034–2040.
Quiocho, F. A. & Ledvina, P. S. (1996). Atomic structure and specificity of bacterial periplasmic receptors for active transport and chemotaxis: variation of common themes. Mol Microbiol 20, 17–25.[Medline]
Rabus, R., Jack, D. L., Kelly, D. J. & Saier, M. H., Jr (1999). TRAP transporters: an ancient family of extracytoplasmic solute-receptor-dependent secondary active transporters. Microbiology 145, 3431–3445.
Rohrbach, M. R., Braun, V. & Koster, W. (1995). Ferrichrome transport in Escherichia coli K-12: altered substrate specificity of mutated periplasmic FhuD and interaction of FhuD with the integral membrane protein FhuB. J Bacteriol 177, 7186–7193.
Sanchez, J. C., Gimenez, R., Schneider, A., Fessner, W. D., Baldoma, L., Aguilar, J. & Badia, J. (1994). Activation of a cryptic gene encoding a kinase for L-xylulose opens a new pathway for the utilization of L-lyxose by Escherichia coli. J Biol Chem 269, 29665–29669.
Severi, E., Randle, G., Kivlin, P., Whitfield, K., Young, R., Moxon, R., Kelly, D., Hood, D. & Thomas, G. (2005). Sialic acid transport in Haemophilus influenzae is essential for lipopolysaccharide sialylation and serum resistance and is dependent on a novel tripartite ATP-independent periplasmic transporter. Mol Microbiol 58, 1173–1185.[CrossRef][Medline]
Stein, M. A., Schafer, A. & Giffhorn, F. (1997). Cloning, nucleotide sequence, and overexpression of smoS, a component of a novel operon encoding an ABC transporter and polyol dehydrogenases of Rhodobacter sphaeroides Si4. J Bacteriol 179, 6335–6340.
Tam, R. & Saier, M. H., Jr (1993). Structural, functional, and evolutionary relationships among extracellular solute-binding receptors of bacteria. Microbiol Rev 57, 320–346.
Tetsch, L. & Kunte, H. J. (2002). The substrate-binding protein TeaA of the osmoregulated ectoine transporter TeaABC from Halomonas elongata: purification and characterization of recombinant TeaA. FEMS Microbiol Lett 211, 213–218.[CrossRef][Medline]
) and second (
) phase are plotted as a function of pyruvateconcentration.
), 50 µM sodium ascorbate (
) or 50 µM 2,3-DKG (
). (b) Titration of the molar ellipticity at 220 nm as a function of 2,3-DKG concentration. The line through the data points is the fit to a single rectangular hyperbola.