Studies on the Escherichia coli glucose-specific permease, PtsG, with a point mutation in its N-terminal amphipathic leader sequence

Mohammad Aboulwafa1,2, Yong Joon Chung1,3, Homan Henry Wai1 and Milton H. Saier, Jr1

1 Division of Biological Sciences, University of California at San Diego, La Jolla, CA 92093-0116, USA
2 Department of Microbiology and Immunology, Faculty of Pharmacy, Ain Shams University, Al Khalifa Al Maamoun St, Abbassia, Cairo, Egypt
3 Department of Life Science, Jeonju University, Chonju, South Korea

Correspondence
Milton H. Saier Jr
msaier{at}ucsd.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Previous work has resulted in the isolation of several mutant glucose permeases (IIGlc or PtsG) of the Escherichia coli phosphotransferase system (PTS) with altered N-terminal amphipathic leader sequences. The mutations were reported to (1) broaden permease substrate specificity, (2) promote facilitated diffusion of some sugars and (3) increase ptsG gene transcription. Detailed biochemical analyses were conducted, showing that one such mutant (V12F-IIGlc) (1) contains dramatically increased amounts of IIGlc, (2) displays correspondingly increased in vitro phosphorylation and in vivo transport activities, (3) shows increased utilization of several metabolizable sugars and (4) shows decreased susceptibility to detergent activation. These results are interpreted as suggesting that the V12F substitution in the N-terminal amphipathic leader sequence of IIGlc alters the facility with which the permease is integrated into the membrane. Consequent changes in conformation alter its catalytic properties and increase its affinity for the pleiotropic transcriptional repressor, Mlc. These changes together are proposed to promote transcription of the ptsG gene and account for the observed phenotypic changes.


Abbreviations: PEP, phosphoenolpyruvate; PTS, phosphotransferase system


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In 1988, our laboratory published observations showing that all well-characterized integral membrane sugar permeases of the phosphoenolpyruvate (PEP)-dependent phosphotransferase system (PTS), except those of the mannose (IIMan) family (of the IIABCD type; Saier & Reizer, 1994), exhibit at their extreme N termini strikingly amphipathic {alpha}-helical structures of 12–22 residues (Saier et al., 1988). When these sequences are portrayed in an {alpha}-helical wheel, all strongly hydrophilic residues appear on one side of the helix, while all strongly hydrophobic residues appear on the opposite side (Saier, 1989). In some cases, such as the fructose Enzyme II (IIFru) of Escherichia coli, which possesses hydrophilic domains preceding the transmembrane IIC domain, amphipathic helices immediately precede the first transmembrane helix in the IIC domain (Saier, 1989). These leader sequences resemble mitochondrial targeting sequences, although they differ from the latter sequences in their overall amino acid compositions (Saier et al., 1989b). The different leader sequences exhibit no significant similarity, no conserved residues and no characteristic charge distribution, arguing against a function related to the catalytic potential of these enzymes (Saier et al., 1988, 1989a, b). Synthetic peptides corresponding in sequence to PTS permease leaders were shown to incorporate into artificial lipid membranes with the concomitant induction of secondary structure (Tamm et al., 1989). Moreover, most proteins exhibiting strikingly amphipathic N-terminal sequences proved to function in macromolecular recognition (Saier & McCaldon, 1988). Most non-PTS bacterial membrane transport proteins do not possess leader sequences with large hydrophobic moments (Saier et al., 1989a).

Yamada et al. (1991) introduced point mutations in the N-terminal region of the mtlA gene corresponding to the leader sequence of the mannitol permease, MtlA, also called the mannitol Enzyme II (IIMtl). They determined the effects of these mutations on IIMtl activities: (1) mannitol fermentation, (2) [14C]mannitol uptake in vivo and (3) [14C]mannitol phosphorylation in vitro. Mutations that abolished IIMtl activities eliminated the protein from the membrane fraction, as estimated by Western blot analysis (Yamada et al., 1991). On this basis, it appeared that the leader sequence played a role in the proper insertion or stable integration of the protein into the membrane.

In confirmation of this conclusion, alkaline phosphatase (PhoA) fusion proteins were constructed in which the N-terminal 13 residues (most of the amphipathic leader) or the N-terminal 52 residues (the amphipathic leader as well as the first transmembrane {alpha}-helix) were fused to the otherwise leaderless PhoA protein. Alkaline phosphatase activities of the mutant reporter gene products were then measured. An excellent correlation was observed between the ability of the mutant leaders to properly insert active IIMtl into the membrane and the ability of either the 13 residue or the 52 residue leaders to export mature enzymically active PhoA protein into the E. coli periplasm (Yamada et al., 1991). It was concluded that the N-terminal amphipathic leader serves a targeting/insertion function.

Recently, five laboratories independently characterized mutants of the glucose PTS permease, IIGlc, the product of the ptsG gene, with broadened apparent substrate specificities and altered catalytic properties (Gegley et al., 1996; Kornberg et al., 2000; Notley-McRobb & Ferenci, 2000; Plumbridge, 2000; Zeppenfeld et al., 2000). Two of these reports described mutations that occurred in the N-terminal amphipathic leader sequence (Kornberg et al., 2000; Notley-McRobb & Ferenci, 2000). Kornberg et al. (2000) showed that fructose could enter the cell via the genetically altered IIGlc by facilitated diffusion. Translocation of other sugars (mannose, methyl {alpha}-glucoside, 2-deoxyglucose) also appeared to be altered. The mutation resulted in the replacement of a valine at position 12 in the wild-type protein with a phenylalanine (V12F) in the deduced amino acid sequence of the mutant protein.

Mutants isolated by Notley-McRobb & Ferenci (2000) occurred at positions 12 and 13 in the amphipathic leader of IIGlc, and one of these mutants was the same one (V12F) that was independently isolated and described by Kornberg et al. (2000). The other two mutants were V12G and G13C. These three mutations resulted in enhanced transport of glucose and methyl {alpha}-glucoside as well as increased growth rates on mannose and glucosamine. Mutations in the loop and transmembrane regions of IIGlc that appear to broaden its substrate specificity have also been characterized by this group (Notley-McRobb & Ferenci, 2000) as well as by others (Gegley et al., 1996; Plumbridge, 2000; Zeppenfeld et al., 2000). Ribose, fructose and mannitol were shown to be taken up at increased rates by representative mutant cells in various genetic backgrounds. It was proposed that the mutations might in some way affect the protein conformation so as to promote non-specific sugar accessibility to the IIGlc channel (Notley-McRobb & Ferenci, 2000). At least some of the IIGlc mutations exerted secondary effects on gene expression (Notley-McRobb & Ferenci, 2000; Plumbridge, 2000; Zeppenfeld et al., 2000) probably by altering the direct interaction of the global transcriptional repressor, Mlc, with the free (dephosphorylated) form of IIGlc (Lee et al., 2000; Nam et al., 2001; Tanaka et al., 2000). None of the research groups that isolated the IIGlc mutants described above conducted in vitro biochemical analyses of the altered enzymes.

In this paper, we report detailed comparative physiological and biochemical analyses of the V12F IIGlc mutant protein (here called G-F) and the isogenic parental protein (here called G-WT) (Kornberg et al., 2000). We confirm and extend the altered growth and transport properties of this mutant and quantify the effects of the mutation on the phosphorylation activities of the enzyme towards a variety of sugars in in vitro assays. Western blot analyses allowed us to estimate the relative amounts of the protein present in the membrane fraction. We further show that the mutant enzyme exhibits altered sensitivity to detergent treatments. The results reported here, together with previously published results, lead us to suggest that the mutation exerts its multifarious effects by creating an altered protein conformation that results from a change in the facility with which the protein is integrated into the phospholipid bilayer of the membrane. Changes in IIGlc conformation presumably alter both its active site configuration and its Mlc interaction site.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains.
The two isogenic Escherichia coli strains used in this study were HK2240 (wild-type IIGlc, here called G-WT) and HK2237 (fructose-utilizing mutant V12F IIGlc, here called G-F) (Kornberg et al., 2000). The common genotype of these strains is {Delta}fruK fruA {Delta}gutA ptsM : : cm mlc+ mak+ his arg thr leu str (Kornberg et al., 2000; Sproul et al., 2001). We confirmed the auxotrophic amino acid requirements and the chloramphenicol resistance properties of these strains. The mlc null mutant (IBPC1012=JM101 mlc : : tc) used was generously provided by Dr J. Plumbridge (Plumbridge, 1999). The mlc : : tc mutation was transduced into the genetic backgrounds of HK2240 and HK2237 using P1 phage selecting for tetracycline resistance. The relevant phenotypes of the four resultant strains as determined by growth on minimal plates are summarized in Table 1. For all other studies reported, strains G-WT and G-F in the mlc+ genetic background were used.


View this table:
[in this window]
[in a new window]
 
Table 1. Sugar utilization by E. coli wild-type (HK2240) and G-F mutant (HK2237) cells in two isogenic genetic backgrounds: HK2240 (WT) and HK2240 mlc (mlc)

Cells showed insignificant growth differences when other sugars (D-N-acetylglucosamine, D-glucose, D-mannitol, D-galactose, glycerol, D-gluconate, D-ribose, L-arabinose, D-arabinose, D-arabitol, D-ribitol, D-xylose, D-xylitol, D-arbutin and L-sorbose) were tested. GN2 MicroPlate test kits (Biolog) were used to confirm the results reported for many of the carbon sources tested (data not shown). Conditions are described in Methods.

 
Growth conditions.
The minimal medium for the growth studies recorded in Table 1 was 50 mM sodium/potassium phosphate buffer, pH 7·2, 50 mM ammonium chloride, essential salts (4 mg CaCl2.6H2O; 8 mg MgSO4.7H2O; 0·4 mg MnSO4.4H2O; 0·4 mg FeSO4.7H2O per 100 ml), 3 µg thiamine hydrochloride ml-1, 20 µg auxotrophic amino acids ml-1 (his, arg, thr and leu) and 0·5 % of the carbon source (Ashworth & Kornberg, 1966). Incubation was at 37 °C for 24–72 h on Bacto agar plates. Growth of the G-F mutant on mannose and fructose may depend in part on ATP-dependent phosphorylation, as catalysed by a derepressed manno(fructo)kinase (Saier et al., 1971; Sproul et al., 2001), while growth on glucose may depend in part on ATP-dependent phosphorylation by wild-type glucokinase (Saier et al., 1973). The growth studies reported in Table 1 were confirmed using Biolog 96 well assays, BUG agar medium and GN2 MicroPlate test kits (Biolog) following the instructions of the manufacturer.

Transport, enzyme and protein assays.
Uptake of radioactive sugars, assay of 14C-labelled sugar phosphorylation via the glucose-specific Enzyme II of the PTS, Western blot analysis and protein assays were conducted essentially as described previously (Aboulwafa & Saier, 2002), except that a much higher protein concentration (100x) was used for assay of [14C]fructose phosphorylation. This was required because of the low activity (300-fold lower than the glucose activity) observed with this substrate (see Table 2). For both transport and enzyme analyses, cells were grown in Luria–Bertani (LB) broth with or without a sugar (0·4 %). For the data presented in Figs 2–4, the results of single uptake experiments are shown, but the reproducibility with triplicate samples was within 15 % of the absolute values reported. For the data presented in Tables 2 and 3, PEP-dependent phosphorylation was assayed (unless otherwise indicated) using 20 µM 14C-labelled sugar. For the assay of PEP-dependent IIGlc activities in the pellet fractions, 20 µl of a high-speed supernatant from a cell lysate of strain HK2240 centrifuged at 40 000 r.p.m. for 2 h was used as a source of Enzyme I, HPr and IIAGlc. No high-speed supernatant was included when the transphosphorylation reaction was assayed with 10 mM glucose 6-phosphate as the phosphoryl donor. The total assay volume was always 200 µl. Assays were conducted for 30 min at 37 °C as described previously (Aboulwafa & Saier, 2002). The same assay conditions were used to determine the effects of detergent treatment on IIGlc activities reported in Tables 3 and 4, except that the preparations were diluted 1 to 12, and 10 µl (PEP reaction) or 50 µl (transphosphorylation reaction) aliquots were taken for assay.


View this table:
[in this window]
[in a new window]
 
Table 2. Comparison of specific activities of IIGlc in the pellet fractions of WT and G-F mutant strains

–, Values were too low to be trustworthy.

 


View larger version (10K):
[in this window]
[in a new window]
 
Fig. 2. Retention of radioactivity derived from metabolizable [14C]glucose by cells grown in either (a) LB or (b) LB plus 0·4 % glucose of WT (thin lines) and G-F mutant (thick lines) strains. Cells were grown at 30 °C with shaking (250 r.p.m.), harvested during early stationary phase, washed four times with TM buffer and resuspended in the same buffer to a final cell density (dry wt) of 34 µg ml-1. To a volume of 930 µl cell suspension, arginine (20 mM) and radioactive sugar (20 µM, final concn) were added at -2 min and zero time, respectively, to give a final volume of 1 ml. Uptake was followed at 30 °C with mild shaking. Samples (100 µl) were removed at the times indicated, filtered (25 mm Millipore filters, pore size 0·45 µm), washed three times with cold TM buffer, dried with a heat IR lamp for 15 min and counted using 10 ml scintillation fluid (Biosafe NA).

 

View this table:
[in this window]
[in a new window]
 
Table 3. Effects of treatment with sodium deoxycholate on PEP-dependent sugar phosphorylation and transphosphorylation activities of the G-WT and G-F IIGlc proteins

An assay of the treated samples was performed using PEP-dependent phosphorylation with [14C]methyl {alpha}-glucoside as substrate at a final concentration of 0·67 µM in the case of G-WT and 10 µM in the case of G-F. The PEP concentration was 5 mM. The transphosphorylation assay was performed at a final [14C]glucose concentration of 0·67 µM in the case of G-WT and of 10 µM in the case of G-F. The phosphate donor was glucose 6-phosphate at a final concentration of 10 mM. Aliquots [10 µl (PEP reaction) or 50 µl (transphosphorylation reaction) of a 1- to 12-fold diluted sample] of the detergent-treated enzyme preparations were assayed (30 min at 37 °C) as described in Methods following incubation of the enzyme preparations treated with the detergent concentrations, at the temperatures and for the time intervals indicated above. The accuracy was generally greater than 20 %. The experiment was repeated three times and comparable relative values were always obtained. ND, Not determined.

 

View this table:
[in this window]
[in a new window]
 
Table 4. Effect of treatment with sodium deoxycholate in the presence of 20 % glycerol (a source of osmolytes) on PEP-dependent sugar phosphorylation and transphosphorylation activities of G-WT and G-F IIGlc

An assay of the treated samples was performed using PEP-dependent phosphorylation with [14C]methyl {alpha}-glucoside as substrate at a final concentration of 10 µM in the case of G-F and of 0·67 µM in the case of G-WT. The transphosphorylation assay was performed at a final [14C]glucose concentration of 10 µM in the case of G-F and of 0·67 µM in the case of G-WT, and the phosphate donor was glucose 6-phosphate at a final concentration of 10 mM. See also legend to Table 3. The accuracy was generally greater than 20 %, as indicated by the SD values reported in Table 2. The experiment was repeated three times and comparable relative values were always obtained.

 

   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Growth studies
The WT and G-F mutant strains were examined for their relative abilities to utilize various sugars for growth (Table 1). Glucose was utilized efficiently by both strains, but WT cells could not use, or used poorly, a variety of hexoses and hexitols. Many of those listed in Table 1 exhibited growth enhancement in the G-F mutant. These sugars include fructose, mannose, glucosamine, sorbitol, galactitol, galactosamine and mannosamine. The presence of a null mlc mutation did not affect the results, showing that the growth enhancement could not be due solely to enhanced production of IIGlc. These observations confirm and extend the results reported by Kornberg et al. (2000) and Notley-McRobb & Ferenci (2000).

Western blot analyses of IIGlc in WT and G-F strains
Fig. 1 shows the results of Western blot analyses of IIGlc in membranes isolated from WT and G-F mutant cells grown in LB and LB+0·4 % glucose media. Fig. 1(a) presents the results for exponential-phase cells, while Fig. 1(b) shows the corresponding results for stationary-phase cells. The results were essentially the same for exponential- and stationary-phase cells. Wild-type LB-grown cells exhibited low IIGlc protein, which was enhanced by growth in the presence of glucose. By contrast, the G-F mutant produced much more IIGlc protein than the wild-type when grown in LB medium (2·3–2·4-fold for both exponential- and stationary-phase cells as estimated by densitometry), and the amount decreased when glucose was present. In the presence of glucose, the IIGlc ratio of G-F/G-WT was 1·2–1·3 for both exponential- and stationary-phase cells as estimated by densitometry. These results are in qualitative agreement with the conclusions of Notley-McRobb & Ferenci (2000) based on Northern blot analyses of ptsG-specific mRNAs produced in a different genetic background under different experimental conditions. In their studies, the ptsG mRNA in uninduced (glycerol-grown) cells was enhanced about 5 times by the V12F mutation, presumably owing to increased rates of transcriptional initiation. Our results establish that the IIGlc protein levels reflect the mRNA levels measured previously.



View larger version (56K):
[in this window]
[in a new window]
 
Fig. 1. Western blot analysis of Enzyme IIGlc using antiIICBGlc antibody. The membranes used for the analysis were derived from (a) exponential-phase and (b) stationary-phase cells grown in either LB or LB plus 0·4 % glucose (LB+G) broth. The methods used, including the control bacterial strains lacking or overproducing IIGlc, are described in Aboulwafa & Saier (2002). +ve, Positive control (strain overexpresses the wild-type ptsG gene); -ve, negative control (strain does not express IIGlc).

 
In vitro pellet IIGlc activities
Table 2 presents the results of in vitro phosphorylation activities for both G-WT and G-F forms of IIGlc. Three radioactive sugars were used in the assay. Because both the IIFru and IIMan PTS complexes have been genetically eliminated (Kornberg et al., 2000), we believe that all three sugars primarily measure IIGlc activity. However, the possibility that the maltose Enzyme II, which recognizes glucose (Reidl & Boos, 1991), might contribute to this activity cannot be eliminated. With [14C]glucose as the phosphorylation substrate and PEP or glucose 6-phosphate as the phosphoryl donor, the assays were conducted with both LB- and LB+glucose-grown cells after harvesting the cells in either the exponential or the stationary phase of growth. With [14C]methyl {alpha}-glucoside or [14C]fructose as substrate, only PEP-dependent activities following harvesting of cells in the stationary phase of growth were recorded.

Five conclusions can be derived. First, with [14C]glucose as the substrate, the activity of either G-WT or G-F was essentially the same when cells were harvested during the exponential versus the stationary phase of growth. Second, G-WT activity increased about twofold when glucose was included in the growth medium, but the G-F mutant activity decreased about twofold when glucose was present. These results are in qualitative agreement with the Western blot analyses shown in Fig. 1(a, b) but do not eliminate the possibility that the G-F mutation alters the specific activity of IIGlc. Third, G-F activity was about tenfold higher than G-WT activity when the cells were grown in LB and 2·5-fold higher after growth in LB+glucose. This result also agrees with the Western blot results shown in Fig. 1. Fourth, when [14C]glucose phosphorylation was measured using the transphosphorylation assay with glucose 6-phosphate instead of PEP as the phosphoryl donor (Saier et al., 1977a, b, c), absolute rates of [14C]glucose phosphorylation decreased dramatically as expected, but relative activities were similar. The G-WT/G-F transphosphorylation activity ratios were elevated slightly relative to the PEP-dependent phosphorylation activities, but because the background levels were appreciable, the significance of this observation is not clear. Finally, when [14C]methyl {alpha}-glucoside or [14C]fructose was the substrate, the PEP-dependent phosphorylation activities of the G-F mutant IIGlc increased in parallel with the corresponding activities observed with [14C]glucose as the substrate.

Effects of detergent on G-WT versus G-F activities
Table 3 presents the results of experiments designed to test the effects of detergent (deoxycholate) on the activities of G-WT versus G-F. In these experiments, the resuspended membrane preparations were exposed to sodium deoxycholate at various concentrations for 30 or 60 min at either 25 or 37 °C. The final protein concentration in the treatment mixture was 2·4 mg ml-1. In a parallel experiment, conducted under similar conditions, 20 % glycerol was present to provide a source of osmolytes (Table 4). Because G-F exhibits a much higher activity per mg protein than does G-WT, the assay was conducted in two ways: (1) the protein concentration was maintained constant while the sugar substrate concentration differed, and (2) the protein concentration differed while the substrate concentration was held constant. The results were essentially the same. Only the experiments using constant protein concentration and variable substrate concentrations are presented in Tables 3 and 4. Regardless of the conditions used, a detergent concentration could be found that activated G-WT, while either inhibiting or only minimally activating G-F. Under no condition was the reverse observed. Interestingly, in the absence of glycerol, both the PEP- and glucose-6-phosphate-dependent phosphorylation reactions were activated at a low deoxycholate concentration (Table 3), but in the presence of glycerol, only the PEP-dependent reaction was activated and a much higher concentration of deoxycholate was required to observe this activation (Table 4). However, in all cases, the percentage activation was consistently higher for G-WT than for G-F. When the effects of SDS (0·005 %) were examined at an elevated temperature (50 °C) (see Aboulwafa & Saier, 2002), G-WT was again activated more than G-F (data not shown). However, the effects were less dramatic than recorded in Tables 3 and 4. We interpret these results to suggest that in addition to the effects owing to increased G-F production, the conformation of G-F differs from that of G-WT and that detergent allows the latter enzyme, to a greater extent than the former enzyme, to assume a conformation that is optimal for in vitro phosphorylation activity. This might imply that G-F is in a more optimal conformation for catalysing sugar phosphorylation in vitro.

Transport studies
Uptake of [14C]glucose and the non-metabolizable glucose analogues, [14C]methyl {alpha}-glucoside and 2-deoxy[14C]glucose, was studied as shown in Figs 2, 3 and 4, respectively. In quantitative agreement with the growth results (Table 1), but in marked contrast to the in vitro phosphorylation results reported in Table 2, there was little difference in glucose uptake rate. However, the initial rate of uptake by G-F appeared to be somewhat greater than that for G-WT regardless of whether cells were grown in LB (Fig. 2a) or LB+glucose (Fig. 2b). It should be recalled that glucose is a metabolizable sugar, and consequently, net uptake reflects transport, metabolism and efflux of radioactive metabolites.



View larger version (12K):
[in this window]
[in a new window]
 
Fig. 3. Uptake of the non-metabolizable glucose analogue, [14C]methyl {alpha}-glucoside, by cells grown in either (a) LB or (b) LB plus 0·4 % glucose of WT (thin lines) and G-F mutant (thick lines) strains. Cells were grown at 30 °C with shaking (250 r.p.m.), harvested during the early stationary growth phase, washed three (LB) or four (LB+glucose) times and resuspended for uptake studies, as described in the legend to Fig. 2, except that the cell densities were 675 µg dry wt ml-1 (circles) and 135 µg dry wt ml-1 (triangles). Transport was followed with time as shown using a final concentration of the radioactive sugar analogue of 20 µM. Uptake by the cells was followed as indicated in the legend to Fig. 2. WT and the G-F mutant cells grown in LB showed a low uptake of [14C]methyl {alpha}-glucoside when the cell density was 135 µg ml-1 (data not shown).

 


View larger version (11K):
[in this window]
[in a new window]
 
Fig. 4. Uptake of 2-deoxy[14C]glucose by WT (thin lines) and the G-F mutant (thick lines) grown in either (a) LB or (b) LB plus 0·4 % glucose. Cells were grown at 30 °C with shaking (250 r.p.m.), harvested during the early stationary phase, washed and suspended for uptake studies, as described in the legend to Fig. 2, except that the cell densities were (a) 2490 µg dry wt ml-1 and (b) 2136 µg dry wt ml-1, respectively. The final concentration of the radioactive sugar was 20 µM. Uptake was followed as a function of time, as indicated in the legend to Fig. 2.

 
The differences observed between G-WT and G-F were substantially more pronounced when methyl {alpha}-glucoside (Fig. 3) or 2-deoxyglucose (Fig. 4) uptake was studied. With either one of these non-metabolizable sugars as the substrate, only transport was measured. Thus, when grown in LB (Fig. 3a) or in LB+glucose medium (Fig. 3b), the initial rates and extents of [14C]methyl {alpha}-glucoside uptake were substantially greater in the G-F mutant cells than in the WT cells, and these differences were augmented when 2-deoxyglucose was the substrate (Fig. 4). It can be concluded therefore that IIGlc activities are greatly enhanced both in vivo and in vitro, either directly or indirectly (or both), by the G-F mutation.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In the late 1980s and early 1990s, our laboratory published a series of nearly forgotten papers showing that all IIABC-type PTS permeases exhibit at their N-termini strikingly amphipathic helical structures that provide a targeting/membrane insertion function (see Introduction). Recently, Kornberg et al. (2000) and Notley-McRobb & Ferenci (2000) isolated mutants in the amphipathic leader sequence of the glucose-specific Enzyme II, IIGlc, and showed that these mutants exhibit a broader growth phenotype, transporting mannose and fructose in a IIGlc-dependent fashion. Mutations altering other parts of the protein, possibly also in regions that contain topological signals for membrane insertion, could similarly broaden the apparent substrate specificity of IIGlc (Gegley et al., 1996; Notley-McRobb & Ferenci, 2000). Kornberg et al. (2000) showed that the V12F mutant IIGlc could transport fructose by facilitated diffusion, suggesting either that the mutant IIGlc quantitatively exhibited an altered mode of transport or that the inherent facilitated diffusion activity of the wild-type IIGlc was unmasked by overproduction. In view of the difficulty in detecting this mode of transport in wild-type cells, the former explanation is favoured. Interestingly, Bogdanov & Dowhan (1999) have found that the E. coli lactose permease, LacY, is dependent on ‘lipid chaperones' for proper membrane insertion, and LacY inserted in the absence of phosphatidylethanolamine exhibits an altered conformation that cannot catalyse sugar : H+ symport, although it retains the ability to catalyse sugar-facilitated diffusion (Bogdanov & Dowhan, 1999; Bogdanov et al., 2002). These observations may be relevant to the facilitated diffusion mode of IIGlc-mediated transport observed for G-F IIGlc by Kornberg et al. (2000). This vectorial process may be a direct consequence of altered IIGlc conformation, resulting from an altered insertional process.

Notley-McRobb & Ferenci (2000) as well as Plumbridge (2000) reported that several IIGlc mutations increased ptsG gene transcription, and Plumbridge (2000) demonstrated that the expression of other genes under the control of the IIGlc-interacting pleiotropic transcriptional regulator, Mlc, was also altered. However, Notley-McRobb & Ferenci (2000) presented results arguing that the broadened apparent specificity of the mutant IIGlc could not be due solely to increased ptsG expression and we have confirmed this conclusion (see Table 1). Thus, single-point mutations in ptsG appear to alter (1) the transport specificity of IIGlc towards sugars, (2) the mode of transport (i.e. group translocation vs facilitated diffusion) catalysed by IIGlc and (3) the Mlc-dependent transcription of several genes. This last observation is not entirely surprising when it is considered that the binding of Mlc to IIGlc determines its activity as a transcriptional repressor (Lee et al., 2000; Nam et al., 2001; Tanaka et al., 2000).

We studied two isogenic pairs of parental and V12F IIGlc mutant strains with the following results. (1) The V12F mutation dramatically increases the amount of IIGlc found in the membranes of LB-grown cells in the absence of an mlc mutation as determined by Western blotting. (2) The in vitro phosphorylation activities of the pellet fraction were correspondingly increased, suggesting that increased IIGlc protein levels largely accounted for the increased activities. (3) The wild-type strain (mlc+) exhibited glucose induction, but the isogenic V12F mutant exhibited glucose repression. Thus, the V12F mutation apparently rendered ptsG expression largely constitutive (independent of Mlc) so that catabolite repression became apparent (Plumbridge, 2001). (4) Detergents activated the wild-type protein for sugar phosphorylation in vitro far more effectively than the V12F mutant protein, suggesting that the V12F mutation affects the membrane-embedded IIGlc conformation so that the protein is more active for in vitro phosphorylation. (5) The increased phosphorylation activities of G-F IIGlc as compared with G-WT IIGlc observed in vitro were paralleled by increased transport activities observed when either [14C]methyl {alpha}-glucoside or 2-deoxy[14C]glucose was the transport substrate. (6) The slight effect of the G-F mutation on retention of 14C label derived from glucose correlated with our inability to observe a difference in the rate of glucose utilization by intact cells (Table 1). (7) Several sugars (see Table 1) became effective growth substrates upon introduction of the G-F mutation (but not the mlc null mutation), including sugars that had not previously been known to be transported by IIGlc.

Although other investigators studying these mutant strains did not report in vitro phosphorylation activities or examine detergent sensitivities, most of our physiological observations are in agreement with the available published results. Taken together, our results confirm and extend those of Kornberg et al. (2000), Notley-McRobb & Ferenci (2000) and Plumbridge (2000). They extend the earlier efforts by presenting detailed biochemical analyses that lead to the proposal that the V12F mutation exerts its primary effect by altering insertion of IIGlc into the cytoplasmic membrane of E. coli.

Like the broadened substrate specificity, the facilitated diffusion reaction documented by Kornberg et al. (2000) is probably a direct consequence of altered IIGlc membrane insertion. However, we can ask if the effect of the V12F mutation on Mlc repressor activity is due to (1) altered active site recognition of sugar substrates (extracellular glucose, for example, dephosphorylates IIBCGlc-phosphate and promotes Mlc binding to IIGlc; see Introduction), (2) an altered IIGlc conformation resulting from altered membrane integration that renders the protein more effective at binding Mlc or (3) increased membrane IIGlc levels that allow sequestration of increased amounts of Mlc. We reason that increased IIGlc production must be a secondary consequence of a direct effect of the V12F mutation on IIGlc conformation and propose that the V12F mutation alters the N-terminal leader sequence so that the mutant IIGlc is differently (possibly more efficiently) integrated into the membrane. It is this effect that we suggest gives rise to an altered IIGlc conformation.

Other mutations in IIGlc, both in its leader sequence and in topological determinants found throughout the protein, may give rise to similarly altered IIGlc conformations. It should be recalled that the various mutants isolated by the different research groups were produced by the application of similar positive selection procedures. We therefore suggest that (1) the mutations alter membrane integration, (2) altered membrane integration alters the IIGlc conformation and (3) the altered conformations of the mutant IIGlc proteins are responsible for the observed phenotypes of the mutant strains. Finally, the conformationally altered proteins bind Mlc with increased affinity, either because they more closely resemble the dephospho form than the phospho form of wild-type IIGlc or because of an altered ratio of the phospho to dephospho forms of the enzyme. Overproduction of the mutant protein would be expected to contribute to the observed phenotypes (the broad substrate specificity and the quantitatively altered mode of transport). Further studies will be necessary to determine the detailed molecular mechanism responsible for the effects of the V12F mutation and of other related mutations on IIGlc activities, stability and membrane insertional propensity, and to establish which effects are primary versus secondary consequences of the mutations.


   ACKNOWLEDGEMENTS
 
We thank Sir Hans Kornberg for the two isogenic ptsG strains used in this study, Dr Jacqueline Plumbridge for the mlc : : tc mutant used, Professor Bernhard Erni for the polyclonal antibody used for the Western blot analyses and Mary Beth Hiller for her assistance in the preparation of this manuscript. We are also grateful to Drs J. Deutscher, T. Ferenci, G. Jacobson, H. Kornberg and J. Plumbridge for critically reading the manuscript. This work was supported by NIH grant #GM64368.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Aboulwafa, M. & Saier, M. H., Jr (2002). Dependency of sugar transport and phosphorylation by the phosphoenolpyruvate-dependent phosphotransferase system on membranous phosphatidyl glycerol in Escherichia coli: studies with a pgsA mutant lacking phosphatidyl glycerophosphate synthase. Res Microbiol 153, 667–677.[CrossRef][Medline]

Ashworth, J. M. & Kornberg, H. L. (1966). The anaplerotic fixation of carbon dioxide by Escherichia coli. Proc R Soc Lond Ser B 65, 179–188.

Bogdanov, M. & Dowhan, W. (1999). Lipid-assisted protein folding. J Biol Chem 274, 36827–36830.[Free Full Text]

Bogdanov, M., Heacock, P. N. & Dowhan, W. (2002). A polytopic membrane protein displays a reversible topology dependent on membrane lipid composition. EMBO J 21, 2107–2116.[Abstract/Free Full Text]

Gegley, G. S., Warner, K. A., Arents, J. C., Postma, P. W. & Jacobson, G. R. (1996). Isolation and characterization of a mutation that alters the substrate specificity of the Escherichia coli glucose permease. J Bacteriol 178, 940–942.[Abstract]

Kornberg, H. L., Lambourne, L. T. M. & Sproul, A. A. (2000). Facilitated diffusion of fructose via the phosphoenolpyruvate/glucose phosphotransferase system of Escherichia coli. Proc Natl Acad Sci U S A 97, 1808–1812.[Abstract/Free Full Text]

Lee, S.-J., Boos, W., Bouché, J.-P. & Plumbridge, J. (2000). Signal transduction between a membrane-bound transporter, PtsG, and a soluble transcription factor, Mlc, of Escherichia coli. EMBO J 19, 5353–5361.[Abstract/Free Full Text]

Nam, T. W., Cho, S. H., Shin, D. & 8 other authors (2001). The Escherichia coli glucose transporter enzyme IICB(Glc) recruits the global repressor Mlc. EMBO J 20, 491–498.[Abstract/Free Full Text]

Notley-McRobb, L. & Ferenci, T. (2000). Substrate specificity and signal transduction pathways in the glucose-specific Enzyme II (EIIGlc) component of the Escherichia coli phosphotransferase system. J Bacteriol 182, 4437–4442.[Abstract/Free Full Text]

Plumbridge, J. (1999). Expression of the phosphotransferase system both mediates and is mediated by Mlc regulation in Escherichia coli. Mol Microbiol 33, 260–273.[CrossRef][Medline]

Plumbridge, J. (2000). A mutation which affects both the specificity of PtsG sugar transport and the regulation of ptsG expression by Mlc in Escherichia coli. Microbiology 146, 2655–2663.[Abstract/Free Full Text]

Plumbridge, J. (2001). Regulation of PTS gene expression by the homologous transcriptional regulators, Mlc and NagC, in Escherichia coli (or how two similar repressors can behave differently). J Mol Microbiol Biotechnol 3, 371–380.[Medline]

Reidl, J. & Boos, W. (1991). The malX malY operon of Escherichia coli encodes a novel Enzyme II of the phosphotransferase system recognizing glucose and maltose and an enzyme abolishing the endogenous induction of the maltose system. J Bacteriol 173, 4862–4876.[Medline]

Saier, M. H., Jr (1989). Protein phosphorylation and allosteric control of inducer exclusion and catabolite repression by the bacterial phosphoenolpyruvate : sugar phosphotransferase system. Microbiol Rev 53, 109–120.

Saier, M. H., Jr & McCaldon, P. (1988). Statistical and functional analyses of viral and cellular proteins with N-terminal amphipathic {alpha}-helices with large hydrophobic moments. Importance to macromolecular recognition and organellar targeting. J Bacteriol 170, 2296–2300.[Medline]

Saier, M. H., Jr & Reizer, J. (1994). The bacterial phosphotransferase system: new frontiers 30 years later. Mol Microbiol 13, 755–764.[Medline]

Saier, M. H., Jr, Young, W. S., III & Roseman, S. (1971). Utilization and transport of hexoses by mutant strains of Salmonella typhimurium lacking enzyme I of the phosphoenolpyruvate-dependent phosphotransferase system. J Biol Chem 246, 5838–5840.[Abstract/Free Full Text]

Saier, M. H., Jr, Bromberg, F. G. & Roseman, S. (1973). Characterization of constitutive galactose permease mutants in Salmonella typhimurium. J Bacteriol 113, 512–514.[Medline]

Saier, M. H., Jr, Cox, D. F. & Moczydlowski, E. G. (1977a). Sugar phosphate : sugar transphosphorylation coupled to exchange group translocation catalyzed by the Enzyme II complexes of the phosphoenolpyruvate : sugar phosphotransferase system in membrane vesicles of Escherichia coli. J Biol Chem 252, 8908–8916.[Medline]

Saier, M. H., Jr, Feucht, B. U. & Mora, W. K. (1977b). Sugar phosphate : sugar transphosphorylation and exchange group translocation catalyzed by the Enzyme II complexes of the bacterial phosphoenolpyruvate : sugar phosphotransferase system. J Biol Chem 252, 8899–8907.[Medline]

Saier, M. H., Jr, Newman, M. J. & Rephaeli, A. W. (1977c). Properties of a phosphoenolpyruvate : mannitol phosphotransferase system in Spirochaeta aurantia. J Biol Chem 252, 8890–8898.[Medline]

Saier, M. H., Jr, Yamada, M., Suda, K., Erni, B., Rak, B., Lengeler, J., Stewart, G. C., Waygood, E. B. & Rapoport, G. (1988). Bacterial proteins with N-terminal leader sequences resembling mitochondrial targeting sequences of eukaryotes. Biochemie 70, 1743–1748.[Medline]

Saier, M. H., Jr, Schnierow, B., Yamada, Y. & Daniels, G. A. (1989a). Structures, evolution and membrane insertion of the integral membrane permease proteins of the bacterial phosphotransferase system. In Highlights of Modern Biochemistry, vol. 1, pp. 771–780. Edited by A. Kotyk, J. Skoda, V. Paces & V. Kostk. Zeist: VSP International Science.

Saier, M. H., Jr, Werner, P. & Müller, M. (1989b). Insertion of proteins into bacterial membranes: mechanism, characteristics and comparisons with the eukaryotic process. Microbiol Rev 53, 333–366.

Sproul, A. A., Lambourne, L. T., Jean-Jacques, D. J. & Kornberg, H. L. (2001). Genetic control of manno(fructo)kinase activity in Escherichia coli. Proc Natl Acad Sci U S A 98, 15257–15259.[Abstract/Free Full Text]

Tamm, L. K., Tomich, J. M. & Saier, M. H., Jr (1989). Membrane incorporation and induction of secondary structure of synthetic peptides corresponding to the N-terminal signal sequences of the glucitol and mannitol permeases of Escherichia coli. J Biol Chem 264, 2587–2592.[Abstract/Free Full Text]

Tanaka, Y., Kimata, K. & Aiba, H. (2000). A novel regulatory role of glucose transporter of Escherichia coli: membrane sequestration of a global repressor Mlc. EMBO J 19, 5344–5352.[Abstract/Free Full Text]

Yamada, Y., Chang, Y.-Y., Daniels, G. A., Wu, L.-F., Tomich, J. M., Yamada, M. & Saier, M. H., Jr (1991). Insertion of the mannitol permease into the membrane of Escherichia coli. Possible involvement of an N-terminal amphiphilic sequence. J Biol Chem 266, 17863–17871.[Abstract/Free Full Text]

Zeppenfeld, T., Larisch, C., Lengeler, J. W. & Jahreis, K. (2000). Glucose transporter mutants of Escherichia coli K-12 with changes in substrate recognition of IICBGlc and induction behaviour of the ptsG gene. J Bacteriol 182, 4443–4452.[Abstract/Free Full Text]

Received 10 May 2002; revised 12 November 2002; accepted 18 November 2002.