This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fann, M.-C.
Right arrow Articles by Maloney, P. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fann, M.-C.
Right arrow Articles by Maloney, P. C.

 Previous Article  |  Next Article 

Journal of Bacteriology, July 2003, p. 3863-3870, Vol. 185, No. 13
0021-9193/03/$08.00+0     DOI: 10.1128/JB.185.13.3863-3870.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.

Functional Characterization of Cysteine Residues in GlpT, the Glycerol 3-Phosphate Transporter of Escherichia coli

Mon-Chou Fann, Anne Busch, and Peter C. Maloney*

Department of Physiology, Johns Hopkins Medical School, Baltimore, Maryland 21205

Received 6 December 2002/ Accepted 22 April 2003


arrow
ABSTRACT
 
In Escherichia coli, the GlpT transporter, a member of the major facilitator superfamily, moves external glycerol 3-phosphate (G3P) into the cytoplasm in exchange for cytoplasmic phosphate. Study of intact cells showed that both GlpT and HisGlpT, a variant with an N-terminal six-histidine tag, are inhibited (50% inhibitory concentration {approx} 35 µM) by the hydrophilic thiol-specific agent p-mercurichlorobenzosulfonate (PCMBS) in a substrate-protectable fashion; by contrast, two other thiol-directed probes, N-maleimidylpropionylbiocytin (MPB) and [2-(trimethylammonium)ethyl]methanethiosulfonate (MTSET), have no effect. Use of variants in which the HisGlpT native cysteines are replaced individually by serine or glycine implicates Cys-176, on transmembrane helix 5 (TM5), as the major target for PCMBS. The inhibitor sensitivity of purified and reconstituted HisGlpT or its cysteine substitution derivatives was found to be consistent with the findings with intact cells, except that a partial response to PCMBS was found for the C176G mutant, suggesting the presence of a mixed population of both right-side-out (RSO) (resistant) and inside-out (ISO) (sensitive) orientations after reconstitution. To clarify this issue, we studied a derivative (P290C) in which the RSO molecules can be blocked independently due to an MPB-responsive cysteine in an extracellular loop. In this derivative, comparisons of variants with (P290C) and without (P290C/C176G) Cys-176 indicated that this residue shows substrate-protectable inhibition by PCMBS in the ISO orientation in proteoliposomes. Since PCMBS gains access to Cys-176 from both periplasmic and cytoplasmic surfaces of the protein (in intact cells and in a reconstituted ISO orientation, respectively) and since access is unavailable when the substrate is present, we propose that Cys-176 is located on the transport pathway and that TM5 has a role in lining this pathway.


arrow
INTRODUCTION
 
In Escherichia coli, the transporter GlpT enables accumulation of sn-glycerol 3-phosphate (G3P) (14) by an antiport reaction that uses internal phosphate as a countersubstrate (2, 21). Related G3P transporters are widely spread among eubacteria (3, 24, 27, 31), and GlpT homologs have been reported in a number of eukaryotes, including plants (6), Drosophila melanogaster (1), mice (28), and humans (5). GlpT is also closely related to transporters responsible for the accumulation of hexose phosphates or phosphoglycerates (21) in enteric bacteria. Together, such proteins comprise a distinct family of organophosphate transporters within the major facilitator superfamily (MFS), the largest collection of evolutionarily related secondary transport systems (25).

The regulation of expression of GlpT has been well described (20, 30, 35), but there have been fewer structure-function studies of the transporter itself. Thus, while the topology of GlpT conforms to the pattern expected of a member of the MFS (Fig. 1) (11; see also reference 9), there has been little work localizing GlpT residues or domains critical for substrate transport. In prior work with a related protein, UhpT, the hexose phosphate transporter of E. coli, cysteine-scanning mutagenesis implicated transmembrane helix 7 (TM7) (32, 33) and TM11 (12, 13) as lining the substrate translocation pathway. The research reported here employed a similar approach to the study of GlpT. The results of experiments using site-specific mutagenesis along with thiol-directed probes in both intact cells and reconstituted preparations led us to conclude that Cys-176, located in GlpT TM5 (Fig. 1), lies on the substrate translocation pathway. With regard to members of the MFS, therefore, this work identifies a new domain with the conformational accessibility required for translocation.



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 1. The topology of HisGlpT is shown along with the approximate locations of its seven cysteine residues (enlarged circles). Also shown at the extracellular boundary of TM7 is the approximate location of position 290 (square). The N-terminal extension containing the polyhistidine tag is also noted. Positions are numbered as they occur in HisGlpT.

(Part of this work was submitted by A. Busch to the University of Konstanz in fulfillment of the requirements of a Diplomarbeit.)


arrow
MATERIALS AND METHODS
 
Bacterial strains and plasmids. E. coli strain SH1200 (glpT phoR ugpA704::Tn10) (obtained from W. Boos, University of Konstanz) served as the host to the plasmids used in this study. Strain SH1200 failed to transport G3P due to defects in GlpT and in Ugp, the ATP-dependent G3P transporter. Plasmid pQ31 was constructed by introducing into pQE (Qiagen) a BamHI-HindIII fragment that encodes GlpT with an N-terminal polyhistidine (His6) extension (HisGlpT), placing GlpT expression under the control of the tac promoter and in the presence of LacIq. Plasmids expressing mutants of HisGlpT were generated according to the PCR protocol of Ho et al. (16), after which the BamHI-HindIII fragment derived from the PCR product was purified and reinserted into the original vector. The sequence of each inserted fragment was confirmed at the DNA Analysis Facility at Johns Hopkins University.

Protein purification and reconstitution. HisGlpT was purified by metal chelate affinity chromatography, as described previously for UhpT (8, 29). Briefly, cells grown at 37°C were harvested 2 h after addition of 1 mM IPTG (isopropyl-ß-D-thiogalactopyranoside) to induce expression of the protein. Cells were resuspended in 1/20 of the volume of 100 mM KPi (pH 7) containing 0.2 mM phenylmethylsulfonyl fluoride and 20 µg of DNaseI/ml. Cells were disrupted in the cold by using a French press (16,000 lb/in2), and after unbroken cells and cell debris were removed by low-speed centrifugation (10,000 x g, 30 min), membranes were pelleted (180,000 x g, 1 h) and resuspended at pH 8 in 1/50 of the volume of a solubilization buffer containing 100 mM KPi, 200 mM NaCl, 10 mM G3P, 2 mM ß-mercaptoethanol, 20% glycerol, 1.5% dodecylmaltoside, and 0.2% E. coli phospholipid. After incubation for 1 h on ice, insoluble materials were removed by centrifugation (180,000 x g, 30 min), and the supernatant was applied to nickel-nitrilotriacetic acid agarose slurry preequilibrated with the same buffer. After overnight incubation, the resin was washed with 20 ml of the same buffer adjusted to pH 7 and containing 20 mM imidazole. HisGlpT was then eluted at about 1 mg/ml in a small volume of the buffer (pH 7) containing 300 mM imidazole instead of 200 mM NaCl. The purified material (≥70% purity) was stored at -80°C until use.

To prepare proteoliposomes containing HisGlpT, purified protein (10 µg) was mixed for 20 min on ice with 1 ml of a buffer containing 100 mM KPi (pH 7), 1.5% OG (octyl-ß-D-glucoside), and 10 mg of sonicated E. coli phospholipid/ml. Proteoliposomes were then formed by a 20-fold dilution at room temperature with a loading buffer (pH 7) having either 100 mM KPi or 50 mM G3P plus 50 mM K2SO4. They were then isolated by centrifugation (180,000 x g, 60 min), the surface of the pellet was rinsed twice with assay buffer (100 mM K2SO4-50 mM MOPS-K [pH 7.0]), and the proteoliposomes were then washed and resuspended in a small volume of assay buffer prior to assay.

Transport assays. To assay HisGlpT function in cells, induced preparations were harvested, washed, and resuspended as a concentrated stock by using iced assay buffer. Just prior to assay, an aliquot was diluted 10-fold with assay buffer at room temperature, and after a brief period of temperature equilibration, the assay was started by addition of 50 µM labeled G3P. The reaction was terminated after 1 min by membrane filtration (0.45-µm pore size; Millipore HAWP02500) and washing with assay buffer lacking substrate. The same general procedures were used to assay HisGlpT-mediated transport of [14C]G3P or 32Pi after reconstitution in proteoliposomes, except that the filters used had a pore size of 0.22 µm (Millipore).

To monitor inhibition by thiol-directed probes and to assess the protection afforded by G3P, cells or proteoliposomes were preincubated for 5 min with a probe with or without 1 mM G3P. Both the probe and protecting substrate were then removed by filtration (as described above), and residual GlpT activity was measured by overlaying cells or proteoliposomes, on the filter, with labeled substrate. The reaction was terminated by washing after an additional 3 min.

SDS-PAGE and immunoblot analysis. To monitor expression of HisGlpT and its cysteine substitution derivatives, samples were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE; 12.5% acrylamide), transferred to nitrocellulose by a semidry transblotter (Bio-Rad), and probed with a mouse monoclonal antibody (Qiagen) directed against tetrahistidine. Antibody binding was detected by enhanced chemiluminescence (SuperSignal; Pierce) using anti-mouse immunoglobulin G conjugated with horseradish peroxidase.

Materials. p-Mercurichlorobenzosulfonate (PCMBS) was purchased from Sigma Chemical Co., while [2-(trimethylammonium)ethyl]methanethiosulfonate (MTSET) and N-maleimidylpropionylbiocytin (MPB) were obtained from Toronto Research Biochemicals, Inc., and tricarboxyethylphosphine was obtained from Aldrich Chemical Co. Both dodecylmaltoside and OG were obtained from Calbiochem Inc. The E. coli phospholipid was obtained from Avanti Polar Lipids, Inc., and [3H]G3P (12 Ci/mmol), [14C]G6P (57 mCi/mmol), and KH2[32P]O4 (1 Ci/mmol) were obtained from Perkin-Elmer Life Sciences, Inc.


arrow
RESULTS
 
Sensitivity of HisGlpT to thiol-active agents. There are seven cysteine residues native to GlpT (Fig. 1), of which five are in putative transmembrane segments where their presence might influence substrate transport. To determine whether one or more of these cysteines is in a region of functional significance, we combined information from site-directed mutagenesis with that derived from thiol-specific agents (PCMBS, MPT, and MTSET) by using both intact cells and reconstituted preparations. As a preliminary step, we analyzed the kinetic behavior of HisGlpT (Fig. 1) by monitoring the initial rates of G3P transport into intact cells (Fig. 2A). The results of this work indicated that introduction of the polyhistidine tag does not significantly alter GlpT function, since the affinity of HisGlpT for G3P was comparable to that found for GlpT itself (Km of 50 µM G3P) (2, 14). Subsequent study of purified material (see below) confirmed that the kinetic properties of HisGlpT corresponded to those noted earlier for the parent molecule (21).



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 2. HisGlpT function and the effects of thiol-directed agents. (A) Cells were given 50 µM [14C]G3P, and samples were withdrawn at the indicated times for filtration and washing. Solid and open symbols indicate, respectively, cells expressing HisGlpT and cells carrying the vector alone. The inset shows a Lineweaver-Burke plot derived from initial rate measurements (1 min) in three separate experiments, giving a Km of 48 µM and a maximal velocity of 62 nmol/min/mg of cell protein (data are means ± SEM). (B) Cells pretreated for 5 min with PCMBS (•), MPB ({blacksquare}), or MTSET ({square}) prior to assay as described for panel A. It was found that 35 µM PCMBS resulted in a 50% inhibition of G3P transport. The inset shows that the inhibition achieved by 35 µM PCMBS was blocked by coincubation of cells with 1 mM G3P. Means ± SEM for three separate assays are shown.

We next monitored the response of cells to the impermeant SH-active agents, PCMBS, MPB, and MTSET. HisGlpT function proved to be highly sensitive to inhibition by PCMBS (50% inhibitory concentration [IC50] near 40 µM) but not MPB or MTSET (IC50, ≥1 mM). Moreover, inhibition by PCMBS was markedly reduced by coincubation with the substrate, G3P (Fig. 2B, inset), suggesting that at least one cysteine in GlpT (HisGlpT) is accessible to PCMBS in a substrate-protectable fashion. For these same conditions (1 mM substrate), we did not observe protection by Pi, but this differential response was not explored further, since the two substrates have markedly different Km values (see below). Thus, for the conditions used, G3P was present in 25-fold excess relative to its Km (20 to 50 µM) while Pi was tested at levels approximating its Km (1 mM).

To identify the cysteine(s) responsible for such PCMBS sensitivity, we analyzed variants in which HisGlpT cysteines were replaced individually by serine (Fig. 3A and B). In six of seven instances, the C->S derivative displayed a specific activity comparable to (50 to 150%) that of the parent protein, while in the remaining case (C176S), HisGlpT function was reduced by at least 10-fold; a parallel immunoblot suggested that the C176S variant is poorly expressed (if at all), for reasons that are unclear. Among the six functional C->S derivatives, we observed patterns of PCMBS inhibition (Fig. 3C) and substrate protection (data not shown) resembling that of the parental protein, with IC50 values ranging from 30 to 50 µM PCMBS (Fig. 3C) in the absence of substrate. These findings therefore suggested one of two possible explanations for the PCMBS sensitivity of HisGlpT (GlpT): either Cys-176, which remains in each of the C->S derivatives, is largely responsible for the observed PCMBS sensitivity or more than a single cysteine (possibly including Cys-176) contributes to this phenomenon.



View larger version (28K):
[in this window]
[in a new window]
  		FIG. 3. PCMBS inhibition of HisGlpT and its derivatives. (A) HisGlpT expression was evaluated by probing cell extracts with antibody directed against tetrahistidine, as described in Material and Methods. (B) The levels of expression noted in panel A were used to normalize assays of G3P transport to derive the specific activities for the C->S derivatives relative to that of the parental protein, HisGlpT. A similar comparison for the C176G variant is from a separate set of experiments. (C) The graph indicates the concentration dependence of inhibition by PCMBS for HisGlpT (•), its C->S derivatives (symbols as indicated), and the C176G variant ({circ}). For the C->S derivatives, a 50% inhibition was achieved with 15 to 35 µM PCMBS.

To assess the role of Cys-176 in the response to PCMBS, we explored alternative substitutions for this residue, guided in part by consideration of those residues found most frequently at the equivalent position among 13 GlpT homologs identified in BLAST searches: Thr or Val was present in five cases, Cys was found in two examples (including GlpT), and Ser and Ile were each found once (9). In HisGlpT, three substitutions (C176->S, T, or N) yielded no functional product, while four other substitutions (C176->G, A, I, or V) gave derivatives in which cells showed G3P transport at about 30% of the parental value. In preliminary tests, we monitored PCMBS sensitivity for these four derivatives and found that at 100 µM PCMBS, the mean residual activity was 83 ± 6% (mean ± standard error of the mean [SEM]) of the parental controls, indicating a marked resistance to the probe relative to the parental protein. The C176G mutant was selected for detailed study, and as shown in Fig. 3C, HisGlpT activity in this variant was reduced by only 20% at 100 µM PCMBS, a concentration that results in ≥95% inhibition in the other six cysteine substitution mutants. It seems likely, therefore, that the major component of PCMBS sensitivity in (GlpT) HisGlpT reflects an accessibility of Cys-176 to the externally added probe; while not proven by direct physical analysis, this working hypothesis guided subsequent work.

Characterization of purified, reconstituted HisGlpT. The work described above established that Cys-176 of HisGlpT is accessible to external PCMBS in a substrate-protectable fashion. It was of interest to examine whether this residue is also accessible when this same probe approaches from the cytoplasmic surface of the protein. If so, then this would indicate that Cys-176 borders an aqueous pathway extending into the protein from both external and internal surfaces. In previous work with UhpT, Yan and Maloney addressed this question by comparing the responses of targeted cysteines in intact cells and everted membrane vesicles (32, 33). Here, to avoid the complexity of a cell- or vesicle-based assay system, we studied purified and reconstituted protein, an approach that was successful in analysis of the bacterial anion (oxalate) exchange protein OxlT (10, 34).

Metal chelate affinity chromatography of HisGlpT (Fig. 4A) yielded about 2 mg of partially purified (ca. 70%) protein per liter of broth-grown cells; while additional steps can give higher purity (4), we judged this partial purification to be sufficient for our biochemical study. The kinetic features of purified HisGlpT were characterized after reconstitution into proteoliposomes loaded with saturating concentrations of either KPi (100 mM) or G3P (50 mM) (see Materials and Methods) (Fig. 4B). This allowed tests of either the homologous (32Pi[out]:Pi[in] or [14C]G3Pout:G3Pin) or heterologous (32Pi[out]:G3Pin or [14C]G3Pout:Pi[in]) exchange reactions and verification that G3P is the preferred substrate. This work (Fig. 4B) suggested that the rate of the homologous self exchange of G3P is significantly greater than heterologous G3P:Pi antiport, a suggestion borne out by the kinetic analysis summarized in Table 1 and consistent with earlier findings with UhpT (21). These experiments also showed, as noted before (21), that the Michaelis constants for the various exchange reactions are largely independent of the trans substrate and that the apparent affinities for the organic (G3P) and inorganic (Pi) substrates differ by about 20-fold.



View larger version (44K):
[in this window]
[in a new window]
  		FIG. 4. Purification and characterization of reconstituted HisGlpT. (A) Purification of HisGlpT, as described in Materials and Methods, was monitored by Coomassie blue staining (left) and immunoblot analysis (right) after SDS-PAGE using samples of 10 µl (solubilized extract, flowthrough, and washes) or 2 µl (eluates). (B) HisGlpT was reconstituted into proteoliposomes loaded with phosphate (solid symbols) or G3P (open symbols) for assays of [14C]G3P transport (circles) or [14C]glucose 6-phosphate (squares). (C) PCMBS inhibition of HisGlpT, its C->S derivatives, and the C176G variant was performed as described in Materials and Methods. Data are means for three or more independent experiments. Without inhibitors present, the specific activities (mean ± SEM for three experiments) for the indicated variants were as follows (values in micromoles per minute per milligram of protein): 0.37 ± 0.02 for HisGlpT, 0.35 ± 0.02 for C143S, 0.88 ± 0.04 for C154S, 0.15 ± 0.01 for C176G, 0.36 ± 0.03 for C209S, 0.55 ± 0.04 for C228S, 1.39 ± 0.09 for C322S, and 0.42 ± 0.04 for C366S.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Kinetics of HisGlpT exchange reactionsa

Sensitivity of purified HisGlpT to thiol-active probes. Using such purified material (Fig. 4), we confirmed and extended the findings made using intact cells (Fig. 2 and 3). For example, we found no response to either MTSET or MPB (data not shown; see below). Also, we verified that HisGlpT and its C->S derivatives show a marked sensitivity to PCMBS (IC50s of 20 to 30 µM). In these same tests, however, the behavior of the C176G variant failed to mimic the response found in cells. Thus, instead of being largely resistant to PCMBS (Fig. 3), the response of this variant suggested the presence of (at least) two components, one of which responded to low concentrations of PCMBS, as in the other C->S mutants, while the other appeared to be unaffected, even at an elevated (100 µM) probe concentration.

The mixed behavior of the C176G mutant (Fig. 4C) suggests the presence of two distinct populations of HisGlpT transporters. Indeed, work using both UhpT (8) and OxlT (10, 34) showed that after reconstitution by detergent dilution (as in this study), about half of the population of transporters is oriented as in the intact cell (right-side out [RSO]),while the other half has an inside-out (ISO) polarity that presents the cytoplasmic surface to the external medium. This likely arises from the generally symmetric structure of these transporters, since the inner and outer surfaces look much the same (15). Moreover, for UhpT and OxlT, these two populations have equivalent kinetic behavior, so the presence of the two forms is not normally evident unless one uses probes with differential reactivities towards one or the other exposed surfaces (cytoplasmic or periplasmic). It seems likely that such a mixed population would also arise on reconstitution of HisGlpT; if so, then the behavior of the C176G mutant (Fig. 4C) has a simple origin. Thus, molecules in the RSO configuration would be resistant to PCMBS (as in intact cells), accounting for the observed 40 to 50% residual activity (Fig. 4C). On the other hand, the PCMBS-sensitive component would reflect the accessibility of one or more of the remaining cysteines on the exposed cytoplasmic surface of the ISO population, a sensitivity that could have been masked by tests with intact cells. This latter finding has been made using UhpT, one of whose cysteines (Cys-143) is modified by PCMBS in everted vesicles but not in intact cells (32).

Such considerations made it important to resolve the question of whether HisGlpT exists in both RSO and ISO configurations after reconstitution. To address this issue, tests of UhpT or OxlT exploited either the accessibility of an extracellular protease cleavage site, the use of a differential protease sensitivity, or the introduction of cysteines whose specific modification marks the internal or external surface (8, 10). For the work reported here, we chose the last approach, and for this reason we introduced a cysteine (P290C) into the extracellular loop connecting TM7 and TM8, a region known from previous work with UhpT to be critical for function (8, 22). Tests with intact cells verified that HisGlpT harboring the P290C mutation showed activity comparable to that of the parental protein (data not shown) and that HisGlpT function became highly sensitive to MPB and MTSET (IC50, ≤1µM) (Fig. 5A), unlike what was seen with the parental strain (IC50, ≥1 mM) (Fig. 2). More important, after purification and reconstitution, treatment by these same probes yielded the biphasic response expected of a mixed population of RSO and ISO orientations (Fig. 5B). To verify that these patterns of inhibition were of the expected character and reflected inhibition at the external surface of proteoliposomes, in a separate experiment (Fig. 5C) we followed treatment by these agents with exposure to tricarboxyethylphosphine, an impermeant reducing agent whose specificity for disulfide bond cleavage should lead to reversal of inhibition mediated by MTSET but not MPB. Indeed, exposure to tricarboxyethylphosphine had the expected result. These findings, together with those noted earlier (Fig. 2 and5A and B), led us to conclude that reconstitution of HisGlpT gives a preparation with both RSO and ISO orientations, as found with related transporters (see above).



View larger version (25K):
[in this window]
[in a new window]
  		FIG. 5. The P290C mutation confers sensitivity to thiol-direct agents. (A) Cells expressing the P290C mutant were monitored for G3P transport in the presence or absence of 1 µM MTSET or 1 µM MPB, as indicated. In the absence of inhibitors, G3P transport was 3.2 ± 0.14 nmol/min/mg of protein (mean ± SEM for three experiments). (B) HisGlpT (squares) and its P290C derivative (circles) were purified and reconstituted in Pi-loaded proteoliposomes and pretreated with MTSET or MPB at the indicated concentrations prior to assays of G3P transport. For three separate experiments, we recorded specific activities (means ± SEM) of 130 ± 14 and 230 ± 7 nmol/min/mg of protein for G3P transport by HisGlpT and its P290C derivative, respectively. (C) Proteoliposomes were treated with excess (1 mM) MTSET or MPB, as described for panel B. After a wash to remove the inhibitor, proteoliposomes were assayed in the absence or presence of 1 mM tricarboxyethylphosphine (TCEP), an impermeant reducing agent.

An appreciation that HisGlpT reconstitutes with random orientation aided the design of studies to monitor Cys-176 accessibility. Thus, we treated P290C-containing variants with MPB to eliminate the RSO configuration, ensuring that the response to PCMBS would reflect only that of the ISO population. We then examined whether the absence of Cys-176 (the P290C/C176G double mutant) correlates with a change in reactivity towards PCMBS. Our findings (Fig. 6) show that after elimination of the RSO configuration, treatment of proteoliposomes for 5 min with 25 µM PCMBS resulted in a ≥90% loss of activity when Cys-176 is present. By contrast, without Cys-176 (the double mutation C176G/P290C) this same protocol resulted in no more than a 50% inhibition. A similarly increased resistance was found at other PCMBS concentrations (data not shown), leading us to conclude that for the ISO population, as for the RSO population, C176 contributes a significant target for PCMBS modification.



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 6. PCMBS sensitivity of Cys-176. The P290C and C176G/P290C variants were purified and reconstituted into proteoliposomes loaded with 50 mM G3P and then exposed to 200 µM MPB for 5 min at room temperature to give a preparation in which only the ISO configuration was active. After a wash to remove MPB, pretreated proteoliposomes were given 25 µM PCMBS for 5 min at room temperature in the presence (solid bars) or absence (open bars) of 1 mM G3P. Subsequently, proteoliposomes were placed on Millipore filters, washed to remove additives, and overlaid with assay buffer containing [14C]G3P for 3 min for monitoring of residual activity. Data (means ± SEM for three experiments) are expressed relative to results with MPB-treated controls that were not exposed to either G3P or PCMBS. In the absence of treatments, the specific activity of the P290C derivative was 140 ± 12 nmol/min/mg of protein and that of the C176G/P290C variant was 79 ± 11 nmol/min/mg of protein.

In these experiments (Fig. 6), we used proteoliposomes loaded with G3P so that we could also investigate whether addition of external G3P protects against modification by PCMBS. This approach was necessary because the differing velocities of the heterologous and homologous exchanges (Fig. 3) (Table 1) might confound interpretation of an experiment using proteoliposomes loaded with Pi and then given G3P during trials of substrate protection. Our findings clearly show that in the presence of Cys-176 (the P290C variant), PCMBS inhibition of the ISO configuration is relieved by the presence of 1 mM G3P, while in the absence of this residue (the P290C/C176G double mutant) there is no substrate protection (Fig. 6B). It may be recalled that PCMBS inhibition in intact cells is largely the result of an attack at Cys-176 (Fig. 3) and that this, too, is moderated by the substrate (Fig. 2). Accordingly, we conclude that Cys-176 is accessible to PCMBS added at either surface of the protein and that in both instances this accessibility is blocked by the presence of substrate.


arrow
DISCUSSION
 
The work described here used cysteine substitution mutagenesis to identify a region that lies on the translocation pathway in GlpT. Experimentally, such domains may be defined as the collection of residues exposed (to solvent) in the inward- and outward-facing conformations of the transporter, a set that includes, but is not limited to, residues involved in substrate binding. While study of the intact cell allows one to characterize accessibility from the extracellular medium (e.g., Fig. 2), it is more difficult to examine accessibility from the internal, cytoplasmic face. Previous efforts have used cells and everted vesicles (18, 32, 33), where one could be assured that the transporter was fully RSO or ISO in orientation. By contrast, the present work relies on purified and reconstituted material, which although simpler biochemically, requires an understanding of the distribution of protein orientations after reconstitution (10, 34). It now appears that GlpT behaves as does UhpT and OxlT, each of which is distributed about equally in kinetically equivalent RSO and ISO conformations after reconstitution by detergent dilution (Fig. 5) (8, 10, 34). Consequently, in reconstituted preparations, hydrophilic probes such as PCMBS have access to both the extracellular surface of the protein (in an outward-facing configuration) and the intracellular surface of the protein (in the inward-facing configuration).

Studies with intact cells (Fig. 2 and 3) strongly suggested that Cys-176 of HisGlpT is accessible to PCMBS from the extracellular phase, and use of reconstituted protein made it possible to examine whether this residue is also accessible to PCMBS from the cytoplasmic surface (Fig. 5 and 6). It was difficult to address this issue directly, because we were unsuccessful in obtaining a functional GlpT variant containing only Cys-176. As a result, our conclusions are drawn from experiments comparing responses recorded in the presence and absence of Cys-176, using variants with and without the C176G mutation. While this approach is less direct than use of a single target cysteine, as in previous work (10, 32-34), study of the response to PCMBS by the reconstituted system showed convincingly that in HisGlpT Cys-176 is accessible to PCMBS in both outward- and inward-facing conformations. By itself, this behavior fits the experimental definition of a residue on a translocation pathway (32). But we also note that in both situations the inferred PCMBS modification of Cys-176 is prevented by the presence of substrate. Substrate protection reflects a complex mix of phenomena, including steric blockage by the substrate and closure of the pathway to the target due to initiation of turnover. In either circumstance, however, this behavior is fully consistent with the idea of Cys-176 belonging to the set of residues whose conformational plasticity allows them to occupy a position along the translocation pathway. In reconstituted preparations, one also finds PCMBS inhibition of HisGlpT even in the absence of Cys-176 (e.g., the C176G variant). This is attributed to modification of one (or more) of the six remaining cysteine residues by PCMBS that approaches from the cytoplasmic surface (Fig. 5). We did not pursue this observation to identify the responsive cysteine(s), since this phenomenon has been noted before; in UhpT, for example, one finds cysteine residues that are targets of thiol-specific probes added to either the outside or the inside surface but not both (32, 33).

GlpT, UhpT, and OxlT each have 12 transmembrane helices (7, 11, 15), as do most members of the MFS, and identification of helices that delineate the transport pathway relies on inferences from genetic, biochemical, or bioinformatic studies. Thus, in work with UhpT and OxlT, we probed strategically placed cysteine residues with hydrophilic and impermeant thiol-active agents, such as those used here. This directly implicated TM2, TM7, and TM11 (10, 12, 13, 32-34) as helices containing residues exposed to solvent in both inward- and outward-facing forms of the transporter. The same tactics have now been used to study (GlpT) HisGlpT, leading to positive identification of TM5 as contributing to the translocation pathway. The functional relevance of TM5 and Cys-176 is further underscored by the fact that within the organophosphate transporter family, this helix contains the highly (>95%) conserved motif W173NXXHN178 (numbering as in HisGlpT) (9). Recent studies of other members of the MFS (GLUT1, TetA, and LacY) have also implicated TM5 (17, 19, 23, 26, 36) as having a role consistent with the conclusions reported here. Such work, largely based on a biochemical approach, is now further strengthened by the low-resolution (6.5 Å) crystallographic profile of OxlT (15), which identifies TM5 as one of eight helices (TM1, -2, -4, -5, -7, -8, -10, and -11) positioned to contribute to the translocation pathway.


arrow
ACKNOWLEDGMENTS
 
We are grateful to Nobuhisa Kimura for help in constructing several single-serine replacement mutants and in conducting tests of their function. We also thank W. Boos (University of Konstanz) for helpful comments.

This research was supported by grant GM24195 from the National Institutes of Health.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Department of Physiology, Johns Hopkins Medical School, Baltimore, MD 21205. Phone: (410) 955-8325. Fax: (410) 955-0461. E-mail: pmaloney{at}jhmi.edu. Back


arrow
REFERENCES
 
    1
  1. Adams, M. D., et al. 2000. The genome sequence of Drosophila melanogaster. Science 287:2185-2195.[Abstract/Free Full Text]
    2. 2
  2. Ambudkar, S. V., T. J. Larson, and P. C. Maloney. 1986. Reconstitution of sugar phosphate transport systems of Escherichia coli. J. Biol. Chem. 261:9083-9086[Abstract/Free Full Text]
    3. 3
  3. Andersson, S. G. E., A. Zomorodipour, J. O. Andersson, T. Sicheritz-Ponten, U. C. M. Alsmark, R. M. Podowski, A. K. Naeslund, A. S. Eriksson, H. H. Winkler, and C. G. Kurland. 1998. The genome sequence of Rickettsia proqazekii and the origin of mitochondria. Nature 396:133-140.[CrossRef][Medline]
    4. 4
  4. Auer, M., M. J. Kim, M. J. Lemieux, A. Villa, J. Song, X. D. Li, and D. N. Wang. 2001. High-yield expression and functional analysis of Escherichia coli glycerol 3-phosphate transporter. Biochemistry 40:6628-6635.[CrossRef][Medline]
    5. 5
  5. Bartoloni, L., M. Wattenhofer, J. Kudoh, A. Berry, K. Shibuya, K. Kawasaki, J. Wang, S. Asakawa, I. Talior, B. Bonne-Tamir, C. Rossier, J. Michaud, E. R. McCabe, S. Minoshima, N. Shimizu, H. S. Scoot, and S. E. Antonarakis. 2000. Cloning and characterization of a putative human glycerol 3-phosphate permease gene (SLC37A1 or G3PP) on 21q22.3: mutation analysis in two candidate phenotypes, DFNB10 and a glycerol kinase deficiency. Genomics 70:190-200.[CrossRef][Medline]
    6. 6
  6. Bevan, M., et al. 1998. Analysis of 1.9 Mb of contiguous sequence from chromosome 4 of Arabidopsis thaliana. Nature 391:485-488.[CrossRef][Medline]
    7. 7
  7. Eiglmeier, K., W. Boos, and S. T. Cole. 1987. Nucleotide sequence and molecular characterization of the glpT gene encoding the glycerol-3-phosphate permease of Escherichia coli: extensive sequence homology with components of the hexose-6-phosphate transport system. Mol. Microbiol. 1:251-258.[CrossRef][Medline]
    8. 8
  8. Fann, M.-C., and P. C. Maloney. 1998. Functional symmetry of UhpT, the sugar phosphate transporter of Escherichia coli. J. Biol. Chem. 273:33735-33740.[Abstract/Free Full Text]
    9. 9
  9. Fann, M.-C., A. H. Davies, A. Varadhachary, T. Kuroda, C. Sevier, T. Tsuchiya, and P. C. Maloney. 1998. Identification of two essential arginine residues in UhpT, the sugar phosphate antiporter of Escherichia coli. J. Membr. Biol. 164:187-195.[CrossRef][Medline]
    10. 10
  10. Fu, D., R. I. Sarker, K. Abe, E. Bolton, and P. C. Maloney. 2001. Structure/function relationships in OxlT, the oxalate-formate transporter of Oxalobacter formigenes. II. Assignment of transmembrane helix 11 to the translocation pathway. J. Biol. Chem. 276:8753-8760.[Abstract/Free Full Text]
    11. 11
  11. Gott, P., and W. Boos. 1988. The transmembrane topology of the sn-glycerol-3-phosphate permease of Escherichia coli analyzed by phoA and lacZ protein fusions. Mol. Microbiol. 2:655-663.[CrossRef][Medline]
    12. 12
  12. Hall, J. A., and P. C. Maloney. 2001. Transmembrane segment 11 of UhpT, the sugar phosphate carrier of Escherichia coli, is an {alpha}-helix that carries determinants of substrate specificity. J. Biol. Chem. 276:25107-25113.[Abstract/Free Full Text]
    13. 13
  13. Hall, J. A., M.-C. Fann, and P. C. Maloney. 1998. Altered substrate selectivity in cysteine substitution mutants of an intrahelical salt bridge in UhpT, the sugar phosphate transporter of Escherichia coli. J. Biol. Chem. 274:6148-6153.[Abstract/Free Full Text]
    14. 14
  14. Hayashi, S., J. P. Koch, and E. C. C. Lin. 1964. Active transport of L-{alpha}-glycerophosphate in Escherichia coli. J. Biol. Chem. 239:3098-3105.[Free Full Text]
    15. 15
  15. Hirae, T., J. A. W. Heymann, D. Shi, R. Sarker, P. C. Maloney, and S. Subramaniam. 2002. Three-dimensional structure of a bacterial oxalate transporter. Nat. Struct. Biol. 9:597-600.[Medline]
    16. 16
  16. Ho, S. N., H. D. Hunt, R. M. Morton, J. K. Pullen, and L. R. Pease. 1989. Site-directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 77:51-59.[CrossRef][Medline]
    17. 17
  17. Hruz, P. W., and M. M. Mueckler. 1999. Cysteine-scanning mutagenesis of transmembrane segment 7 of the GLUT1 glucose transporter. J. Biol. Chem. 274:36176-36180.[Abstract/Free Full Text]
    18. 18
  18. Hu, L. A., and S. C. King. 1999. Identification of the amine-polyamine-choline transporter superfamily 'consensus amphipathic region' as the target for inactivation of the Escherichia coli GABA transporter GabP by thiol modification reagents. Biochem. J. 339:649-655.
    19. 19
  19. King, S. C., and S. Li. 1998. Suppressor scanning at positions 177 and 236 in the Escherichia coli lactose/H+ cotransporter and stereotypical effects of acidic substituents that suggest a favored orientation of transmembrane segments relative to the lipid bilayer. J. Bacteriol. 180:2756-2758.[Abstract/Free Full Text]
    20. 20
  20. Larson, T. J., S. Ye, D. L. Weissenborn, H. J. Hoffmann, and H. Schweizer. 1987. Purification and characterisation of the repressor for the sn-glycerol-3-phosphate regulon in Escherichia coli K12. J. Biol. Chem. 262:15869-15874.[Abstract/Free Full Text]
    21. 21
  21. Maloney, P. C., S. Ambudkar, V. Anantharam, L. Sonna, and A. Varadhachary. 1990. Anion-exchange mechanisms in bacteria. Microbiol. Rev. 54:1-17.[Abstract/Free Full Text]
    22. 22
  22. Matos, M., M.-C. Fann, R.-T. Yan, and P. C. Maloney. 1996. Enzymatic and biochemical probes of residues external to the translocation pathway of UhpT, the sugar phosphate carrier of Escherichia coli. J. Biol. Chem. 271:18571-19575.[Abstract/Free Full Text]
    23. 23
  23. Mueckler, M., and C. Makepeace. 1999. Transmembrane segment 5 of the GLUT1 glucose transporter is an amphipathic helix that forms part of the sugar permeation pathway. J. Biol. Chem. 274:10923-10926.[Abstract/Free Full Text]
    24. 24
  24. Nilsson, R. P., L. Beijer, and B. Rutberg. 1994. The glpT and glpQ genes of the glycerol regulon in Bacillus subtilis. Microbiology 140:723-730.[Abstract/Free Full Text]
    25. 25
  25. Pao, S. S., I. T. Paulsen, and M. H. Saier, Jr. 1998. The major facilitator superfamily. Microbiol. Mol. Biol. Rev. 62:1-32.[Abstract/Free Full Text]
    26. 26
  26. Saraceni-Richards, C. A., and S. B. Levy. 2000. Evidence for interactions between helices 5 and 8 and a role for the interdomain loop in tetracycline resistance mediated by hybrid Tet proteins. J. Biol. Chem. 275:6101-6106.[Abstract/Free Full Text]
    27. 27
  27. Song, X.-M., A. Forsgren, and H. Janson. 1998. Glycerol-3-phosphate transport in Haemophilus influenzae: cloning, sequencing and transcription analysis of the glpT gene. Gene 215:381-388.[CrossRef][Medline]
    28. 28
  28. Takahashi, Y., M. Miyata, P. Zheng, T. Imazato, A. Orwitz, and J. D. Smith. 2000. Identification of cAMP analogue inducible genes in RAW264 macrophages. Biochim. Biophys. Acta 1492:385-394.[Medline]
    29. 29
  29. Tamai, E., M.-C. Fann, and P. C. Maloney. 1998. Purification of UhpT, the sugar phosphate transporter of Escherichia coli. Protein Expr. Purif. 10:275-282.
    30. 30
  30. Tommassen, J., K. Eiglmeier, S. T. Cole, P. Overduin, T. J. Larson, and W. Boos. 1991. Characterization of two genes, glpQ and ugpQ, encoding glycerophosphoryl diester phosphodiesterases of Escherichia coli. Mol. Gen. Genet. 226:321-327.[CrossRef][Medline]
    31. 31
  31. Venkatesan, M. M., W. A. Alexander, and C. Fernandez-Prada. 1996. A Shigella flexneri invasion plasmid gene, ipgH, with homology to IS629 and sequences encoding bacterial sugar phosphate transport proteins. Gene 175:23-27.[CrossRef][Medline]
    32. 32
  32. Yan, R. T., and P. C. Maloney. 1993. Identification of a residue in the translocation pathway of a membrane carrier. Cell 75:37-44.[Medline]
    33. 33
  33. Yan, R. T., and P. C. Maloney. 1995. Residues in the pathway through a membrane transporter. Proc. Natl. Acad. Sci. USA 92:5973-5976.[Abstract/Free Full Text]
    34. 34
  34. Ye, L., and P. C. Maloney. 2002. Structure/function relationships in OxlT, the oxalate/formate antiporter of Oxalobacter formigenes. IV. Assignment of transmembrane helix two (TM2) to the translocation pathway. J. Biol. Chem. 277:20372-20378.[Abstract/Free Full Text]
    35. 35
  35. Zeng, G., S. Ye, and T. J. Larson. 1996. Repressor for the sn-glycerol 3-phosphate regulon of Escherichia coli K-12: primary structure and identification of the DNA-binding domain. J. Bacteriol. 178:7080-7089.[Abstract/Free Full Text]
    36. 36
  36. Zhao, M., K.-C. Zen, W. L. Hubbell, and H. R. Kaback. 1999. Proximity between Glu126 and Arg144 in the lactose permease of Escherichia coli. Biochemistry 38:7407-7412.[CrossRef][Medline]

Journal of Bacteriology, July 2003, p. 3863-3870, Vol. 185, No. 13
0021-9193/03/$08.00+0     DOI: 10.1128/JB.185.13.3863-3870.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Pirch, T., Landmeier, S., Jung, H. (2003). Transmembrane Domain II of the Na+/Proline Transporter PutP of Escherichia coli Forms Part of a Conformationally Flexible, Cytoplasmic Exposed Aqueous Cavity within the Membrane. J. Biol. Chem. 278: 42942-42949 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fann, M.-C.
Right arrow Articles by Maloney, P. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fann, M.-C.
Right arrow Articles by Maloney, P. C.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»