Sugar transport by the bacterial phosphotransferase system. Structural and thermodynamic domains of enzyme I of Salmonella typhimurium.

Enzyme I, the first in the sequence of phosphocarrier proteins of the bacterial phosphoenolpyruvate:glycose phosphotransferase system, is a potential critical point for regulating sugar uptake. The thermal stability of Enzyme I was studied by high sensitivity differential scanning calorimetry. At pH 7.5, thermal unfolding of the protein exhibits two peaks with maxima (Tm) at 47.6 and 55.1 degrees C, indicating that the protein comprises two cooperative unfolding structures. Interaction between the two domains is markedly dependent on pH within the range 6.5-8.5. At pH 7.5, catalytic activity was unaffected by heating through the first transition but was lost by heating through the second. Cleavage of Enzyme I (63.5 kDa) by trypsin, chymotrypsin, or Staphylococcus aureus V8 protease yields a 30-kDa fragment, EI-N, containing the NH2 terminus and the active site, His-189. Protease and differential scanning calorimetry experiments show that EI-N is the structural domain corresponding to the cooperative region in the intact enzyme that unfolds at the higher Tm. EI-N catalyzes one activity of Enzyme I; it accepts a phosphoryl group from phosphohistidine-containing phosphocarrier protein but cannot be phosphorylated by phospho-Enzyme I or phosphoenolpyruvate. The phosphoryl transfer between EI-N and the histidine-containing phosphocarrier protein is reversible. Portions of the Salmonella typhimurium ptsI DNA sequence are known; the complete sequence is presented here and compared to Escherichia coli ptsI.     

Limits to inducer exclusion: inhibition of the bacterial phosphotransferase system by glycerol kinase

The uptake of methyl α- D -glucopyranoside by the phosphoenolpyruvate-dependent phosphotransferase system of Salmonella typhimurium could be inhibited by prior incubation of the cells with glycerol. Inhibition was only observed for glycerol preincubation times longer than 45 s and required the preinduction of both the glucose and the glycerol-catabolizing systems. Larger extents of inhibition by glycerol correlated with higher intracellular levels of glycerol kinase when the glp regulon had been induced to different extents. Preincubation with lactate did not inhibit methyl α- D -glucopyranoside uptake significantly, although both lactate and glycerol were oxidized by the cells. The cellular free-energy state of the cells (intracellular [ATP]/[ADP] ratio) was virtually identical for lactate and glycerol preincubation, suggesting that the inhibition of phosphotransferase-mediated uptake was not a metabolic effect. In vitro , phosphotransferase activity was inhibited to a maximal extent of 32% upon titrating cell-free extracts with high concentrations of commercial glycerol kinase. The results show that uptake systems that have hitherto been regarded merely as targets of the phosphotransferase system component IIA^Glc also have the capacity themselves to retroinhibit the phosphotransferase system flux, presumably by sequestration of the available IIA^Glc, provided that these systems are induced to appropriate levels.     

Sugar transport by the bacterial phosphotransferase system. The intrinsic fluorescence of enzyme I

Enzyme I of the bacterial phosphoenolpyruvate: glycose phosphotransferase system has 2 tryptophan residues/monomer, as determined spectrophotometrically. The tryptophan fluorescence has been investigated with the aid of nanosecond time-resolved techniques. The decay of the fluorescence intensity was analyzed in terms of a biexponential function. The contribution of the emission associated with the shorter decay constant increases from 17-19% at 1 degree C to 43-44% at room temperature. Decay-associated spectra obtained with Enzyme I indicate different spectral distributions associated with the two decay constants. The measurement of tumbling of Enzyme I as a function of temperature revealed a transition of rotational rates between 5 and 15.5 degrees C. Global analysis allowed decomposition of the anisotropy decay into a formulation consistent with monomer and dimer rotational contributions.     

Sodium-dependent methyl 1-thio-beta-D-galactopyranoside transport in membrane vesicles isolated from Salmonella typhimurium.

Membrane vesicles isolated from Salmonella typhimurium G-30 grown in the presence of melibiose catalyze methyl 1-thio-beta-D-galactopyranoside (TMG) transport in the presence of sodium or lithium, as shown initially with intact cells by Stock and Roseman (Stock, J., and Roseman, S. (1971), Biochem. Biophys. Res. Commun. 44, 132). TMG-dependent sodium uptake is also observed, but only when a potassium diffusion potential (interior negative) is induced across the vesicle membrane. Cation-dependent TMG accumulation varies with the electrochemical gradient of protons generated as a result of D-lactate oxidation, and the vesicles catalyze D-lactate-dependent sodium efflux in a manner which is consistent with the operation of a proton-sodium exchange mechanism. Although the stoichiometry between sodium and TMG appears to be 1:1 when transport is induced by a potassium diffusion potential, evidence is presented which indicates that the relationship may exceed unity under certain conditions. The results are explained in terms of a model in which TMG-sodium (lithium) symport is driven by an electrochemical gradient of protons which functions to maintain a low intravesicular sodium or lithium concentration through proton--sodium (lithium) antiport.     

Mannitol-specific enzyme II of the bacterial phosphotransferase system. II. Reconstitution of vectorial transphosphorylation in phospholipid vesicles

Purified mannitol Enzyme II from Escherichia coli was reconstituted in phospholipid vesicles employing the octylglucoside dilution procedure and was shown to catalyze vectorial mannitol 1-phosphate:mannitol transphosphorylation. Reconstitution of the enzyme into liposomes showed a marked dependency upon the octylglucoside concentration with an optimum at 1.2%. The reconstituted transphosphorylation activity exhibited an absolute dependence upon mannitol 1-phosphate as the phosphoryl donor, was sensitive to N-ethylmaleimide, and had a pH optimum near 6. The intravesicular radiolabeled mannitol phosphate could be released from the proteoliposomes by the addition of either 50 microM unlabeled mannitol or 0.5% sodium dodecyl sulfate. The rate of formation of intraliposomal mannitol phosphate, measured as a function of the mannitol Enzyme II concentration, showed a sigmoidal response, suggesting that at high enzyme concentrations the mannitol Enzyme II exists in an aggregated or oligomeric state and that this form is more active than the monomeric or dissociated form of the enzyme in catalyzing the vectorial mannitol transphosphorylation reaction.     

Sugar transport by the bacterial phosphotransferase system. Fluorescence studies of subunit interactions of enzyme I

Enzyme I of the bacterial phosphoenolpyruvate:glycose phosphotransferase system (PTS) exhibits a temperature-dependent monomer/dimer equilibrium. The accompanying paper (Han, M. K., Roseman, S., and Brand, L. (1990) J. Biol. Chem. 265, 1985-1995) shows that the C-terminal -SH residue (Cys-575) can be modified specifically with fluorescent probes such as pyrene maleimide. The derivative retains full enzyme activity, and is capable of forming dimers at room temperature. In the present studies, Enzyme I labeled in this way is found to exhibit a temperature-, concentration-, and pH-dependent monomer/dimer association. The kinetics of dimer formation of Enzyme I is measured in the following way. A derivatized Enzyme I sample is prepared with a pyrene moiety irreversibly attached to the C-terminal -SH residue and 5,5'-dithiobis-2-nitrobenzoic acid reversibly attached to the other 3 -SH residues. This modified enzyme does not form dimers at room temperature. Addition of dithiothreitol results in total release of the thionitrobenzoate anion within 2 min. After the three -SH groups are unblocked, steady-state and nanosecond time-resolved emission anisotropy measurements indicate the dimer is formed over a period of 30 min. In a similar experiment, little dimer formation is observed at 3 degrees C, at temperature at which the native enzyme also does not form dimers. Tryptophan fluorescence is also examined during the release of the thionitrobenzoate. After the completion of thionitrobenzoate release, additional slow steady-state tryptophan fluorescence changes are observed. These results suggest that dimer formation may be preceded by a conformational change following thionitrobenzoate release.     

Sugar transport by the bacterial phosphotransferase system. Regulation of other transport systems (lactose and melibiose)

The role of the phosphoenolpyruvate-dependent phosphotransferase system (PTS) in the phenomenon of inducer exclusion was examined in whole cells of Salmonella typhimurium which carried the genes of the Escherichia coli lactose operon on an episome. In the presence of the PTS substrate methyl alpha-D-glucopyranoside, the extent of accumulation of the lactose analog methyl beta-D-thiogalactopyranoside was reduced. A strain carrying a mutation in the gene for Enzyme I was hypersensitive to the PTS effect, while a crr mutant strain was completely resistant. Influx, efflux, and exchange of galactosides via the lactose "permease" were inhibited by methyl alpha-glucoside. This inhibition occurred in the presence of metabolic energy poisons, and therefore does not involve either the generation of metabolic energy or energy-coupling to the lactose transport system. When the cellular content of the lactose permease was increased by induction with isopropyl beta-D-thiogalactopyranoside, cells gradually became less sensitive to inducer exclusion. The extent of inhibition of methyl beta-thiogalactoside accumulation by methyl alpha-glucoside was shown to be dependent on the relative cellular content of the PTS and lactose system. The data were consistent with an hypothesis involving partial inactivation of galactoside transport due to interaction between a component of the PTS and the lactose permease. By examination of the effects of the PTS and lactose uptake and melibiose permease-mediated uptake of methyl beta-thiogalactoside, it was further shown that the manner in which inducer exclusion is expressed is independent on the routes available to the non-PTS sugar for exit from the cell.     

Reconstitution of the phosphoglycerate transport protein of Salmonella typhimurium

Operation of the phosphoglycerate transport protein (PgtP) of Salmonella typhimurium has been studied in proteoliposomes by using a technique in which membrane protein is solubilized and reconstituted directly from small volumes of cell cultures. When protein from induced cells was reconstituted into phosphate (Pi)-loaded proteoliposomes, it was possible to demonstrate a PgtP-mediated exchange of internal and external phosphate. For this homologous Pi:Pi antiport, kinetic analysis indicated a Michaelis constant (Kt) of 1 mM and a maximal velocity of 26 nmol/min mg of protein; arsenate inhibited with a Ki of 1.3 mM, suggesting that PgtP did not discriminate between these two inorganic substrates. Pi-loaded proteoliposomes also accumulated 3-phosphoglycerate and phosphoenolpyruvate, establishing for each of them a concentration gradient (in/out) of about 100-fold; phosphoenolpyruvate (Ki = 70 microM) rather than 3-phosphoglycerate (Kt = 700, Ki = 900 microM) was the preferred substrate for these conditions. We also concluded that such heterologous exchange was a neutral event, since its rate and extent were unaffected by the presence of a protonophore and unresponsive to the imposition of a membrane potential (positive or negative inside). In quantitative work, we found a stoichiometry of 1:1 for the exchange of Pi and 3-phosphoglycerate, and given an electroneutral exchange, this finding is most easily understood as the overall exchange of divalent Pi against divalent phosphoglycerate. These experiments establish that PgtP functions as an anion exchange protein and that it shares important mechanistic features with the Pi-linked antiporters, GlpT and UhpT, responsible for transport of glycerol 3-phosphate and hexose 6-phosphates into Escherichia coli.     

Mutation in the crp gene of Salmonella typhimurium which interferes with inducer exclusion.

A mutation in the crp gene of Salmonella typhimurium is described which overcame the defects of ptsHI deletion mutants for growth on a number of nonphosphotransferase system compounds. This mutation abolished inducer exclusion in a leaky ptsI mutant. The possible implications for the mechanism of inducer exclusion are discussed.     

Sugar transport by the bacterial phosphotransferase system. In vivo regulation of lactose transport in Escherichia coli by IIIGlc, a protein of the phosphoenolpyruvate:glycose phosphotransferase system

Escherichia coli and Salmonella typhimurium preferentially utilize sugar substrates of the phosphoenol-pyruvate:glycose phosphotransferase system (PTS) when the growth medium also contains other sugars. This phenomenon, diauxic growth, is regulated by the crr gene, which encodes the PTS protein IIIGlc (Saffen, D.W., Presper, K.A., Doering, T.L., and Roseman, S. (1987) J. Biol. Chem. 16241-16253). We have proposed that non-PTS permeases are regulated by their interaction with IIIGlc, and in vitro studies from other laboratories have provided support for this model, but the in vivo effects of excess IIIGlc are not known. In the present studies, transformed cells that overproduced IIIGlc 2- and 10-fold, respectively, were constructed from a pts+ strain of E. coli and plasmids containing the crr gene. In the 2-fold overproducer, fermentation of, and growth on the non-PTS carbohydrates glycerol, lactose, maltose, and melibiose was generally more sensitive to the glucose analogue methyl-alpha-D-glucopyranoside than in a control strain containing normal levels of IIIGlc. In addition, inhibition of lactose permease activity by methyl-alpha-glucoside (inducer exclusion) was more effective in the 2-fold overproducer than in the control strain, particularly when the permease activity was high. The 10-fold IIIGlc overproducing strain had a requirement for the amino acids methionine, isoleucine, leucine, and valine that may or may not be related to the increased concentration of IIIGlc. Fermentation of non-PTS carbohydrates was also poor in the latter strain. Finally, lactose permease activity was 50% of that in control cells containing the same levels of beta-galactosidase, and the lactose permease activity in the IIIGlc overproducer was reduced to an extremely low level in the presence of methyl alpha-glucoside. Thus there is an inverse relationship between the cellular concentration of IIIGlc and the ability to metabolize non-PTS substrates. The results are consistent with the model where inducer exclusion is affected by a direct interaction between IIIGlc and a non-PTS transport system.     

Reconstitution of periplasmic transport in inside-out membrane vesicles. Energization by ATP

The periplasmic histidine permease of Salmonella typhimurium has been reconstituted in inside-out vesicles (IOV) of Escherichia coli by disrupting the cells with a French press in the presence of a high concentration of the periplasmic histidine-binding protein, HisJ. Efflux from IOV, which is equivalent to uptake in whole cells, is induced by ATP. The reconstituted system depends on the presence of the membrane-bound permease proteins, HisQ, HisM, and HisP, and does not function if reconstitution is performed in the presence of a mutant HisJ protein, HisJ5625, that can bind histidine normally but can't interact properly with the membrane complex. Efflux is not induced by the nonhydrolyzable ATP analog, adenyl-5'-yl imidodiphosphate, supporting the contention that ATP hydrolysis is necessary. 8-Azido ATP inactivates IOV, indicating that the ATP effect occurs through the HisP protein, which has previously been shown to be modified by 8-azido ATP (Hobson, A., Weatherwax, R., and Ames, G.F.-L. (1984) Proc. Natl. Acad. Sci. U. S. A. 81, 733-7337). The estimated Km of the vesicles for ATP is about 200 microM. Vanadate, an inhibitor of phosphohydrolase enzymes, inhibits ATP-induced efflux. We conclude that ATP is likely to be the proximal energy source for periplasmic permeases.     

Reconstitution of the binding protein-dependent galactose transport of Salmonella typhimurium in proteoliposomes.

Binding protein-dependent transport systems mediate the accumulation of diverse substrates in bacteria. The binding protein-dependent galactose transport of Salmonella typhimurium has been reconstituted in proteoliposomes. The proteoliposomes were made with proteins solubilized and renatured from inclusion bodies produced by a bacterial strain containing a plasmid with the mgl (methylgalactose permease) operon of Salmonella typhimurium. Galactose transport is dependent both on the addition of the purified galactose binding protein to the transport assay, and on ATP. The interaction between the liganded galactose binding protein and proteoliposomes displays Michaelis type kinetics with a Km of around 15 microM. Galactose transport is coupled to ATP hydrolysis with a stoichiometry (ATP/galactose) of 2.5:1. Galactose transport in proteoliposomes is not significantly inhibited by the uncoupler carbonylcyanide m-chlorophenylhydrazone, but is inhibited by 0.5 mM vanadate. The present reconstitution of galactose transport in proteoliposomes suggests that the MglA, MglC and MglE proteins have been solubilized and renatured in an active form from the inclusion bodies of the mgl hyperproducing strain.     

Evidence for regulation of gluconeogenesis by the fructose phosphotransferase system in Salmonella typhimurium.

A genetic locus designated fruR, previously mapped to min 3 on the Salmonella typhimurium chromosome, gave rise to constitutive expression of the fructose (fru) regulon and pleiotropically prevented growth on all Krebs cycle intermediates. Regulatory effects of fruR were independent of cyclic AMP and its receptor protein and did not prevent uptake of Krebs cycle intermediates. Instead, the phosphotransferase system appeared to regulate gluconeogenesis by controlling the activities of phosphoenolpyruvate carboxykinase and phosphoenolpyruvate synthase.     

Role of IIIGlc of the phosphoenolpyruvate-glucose phosphotransferase system in inducer exclusion in Escherichia coli.

The phosphoenolpyruvate-D-glucose phosphotransferase system of Enterobacteriaceae is thought to regulate the synthesis and activity of a number of catabolite uptake systems, including those for maltose, lactose, and glycerol, via the phosphorylation state of one of its components, IIIGlc. We have investigated the proposal by Kornberg and co-workers (FEBS Lett. 117(Suppl.):K28-K36, 1980) that not IIIGlc, but an unknown protein, the product of the iex gene, is responsible for the exclusion of the above-mentioned compounds from the cell. The iex mutant HK738 of Escherichia coli contains normal amounts of IIIGlc as measured by specific antibodies, in contrast to crr mutants that lack IIIGlc. The IIIGlc of the iex strain functions normally in glucose and methyl alpha-glucoside transport, and the specific activity in in vitro phosphorylation is approximately 60% of that of the parent. The IIIGlc activity of the iex strain is, however, heat labile, in contrast to the parental IIIGlc, suggesting that the mutant contains an altered IIIGlc. This is supported by the observation that IIIGlc from the iex strain cannot bind to the lactose carrier. Thus it cannot inhibit the carrier, and this explains why the uptake of non-phosphotransferase system compounds in an iex strain is resistant to phosphotransferase system sugars. The introduction of a plasmid containing a wild-type crr+ allele into the iex strain restores the iex phenotype to that of the iex+ parent. The IIIGlc produced from the plasmid in the iex strain is heat stable and binds normally to the lactose carrier. These results lead to the conclusion that the iex mutation is most likely allelic with crr and results in an altered, temperature-sensitive IIIGlc that is still able to function D-glucose and methyl alpha-glucoside uptake and phosphorylation and in the activation of adenylate cyclase, but is unable to bind to and inhibit the lactose carrier.     

Sugar transport by the bacterial phosphotransferase system. Characterization of the sulfhydryl groups and site-specific labeling of enzyme I

Enzyme I is the first protein of the phospho transfer sequence in the bacterial phosphoenolpyruvate:glycose phosphotransferase system. This protein exhibits a temperature-dependent monomer/dimer equilibrium. The nucleotide sequence of Escherichia coli ptsI indicates four -SH residues per subunit (Saffen, D. W., Presper, K. A., Doering, T. L., and Roseman, S. (1987) J. Biol. Chem. 262, 16241-16253). In the present experiments, the sulfhydryl groups of the E. coli enzyme were studied with various -SH-specific reagents. Titration of Enzyme I with 5,5'-dithiobis-2-nitrobenzoic acid also revealed four reacting -SH groups. The kinetics of the 5,5'-dithiobis-2-nitrobenzoic acid reaction with Enzyme I exhibit biphasic character, with pseudo-first order rate constants of 2.3 x 10(-2)/s and 2.3 x 10(-3)/s at pH 7.5, at room temperature. Fractional amplitudes associated with the rate constants were 25 +/- 5% for the fast and 75 +/- 5% for the slow rate. The "slow" rate was influenced by ligands that react with Enzyme I (the protein HPr, Mg2+, Mg2+ plus P-enolpyruvate), and also by temperature (at the temperature range where the monomer/dimer association occurs). The fractional ratio of the two rates remained at 1:3 under these conditions. Thus, under all conditions tested, two classes of -SH groups were detected, one reacting more rapidly than the other three -SH groups. Modification of the "fast" -SH group results in an active enzyme capable of forming dimer, whereas modification of the slow -SH groups results in inactive and monomeric Enzyme I. The enzyme was labeled with pyrene maleimide under conditions where only the more reactive sulfhydryl group was derivatized. Hydrolysis by trypsin followed by reverse-phase high performance liquid chromatography analysis of the peptide mixture resulted in only one fluorescent peak. This peak was not observed when the more reactive sulfhydryl residue was protected prior to pyrene maleimide labeling. Amino acid sequencing of the fluorescent peak indicated that the more reactive residue is the C-terminal amino acid residue, cysteine 575. The results provide a means for selectively labeling Enzyme I with a fluorophore at a single site while retaining full catalytic activity.