Membrane deinsertion of SecA underlying proton motive force-dependent stimulation of protein translocation

The proton motive force (PMF) renders protein translocation across the Escherichia coli membrane highly efficient, although the underlying mechanism has not been clarified. The membrane insertion and deinsertion of SecA coupled to ATP binding and hydrolysis, respectively, are thought to drive the translocation. We report here that PMF significantly decreases the level of membrane-inserted SecA. The prlA4 mutation of SecY, which causes efficient protein translocation in the absence of PMF, was found to reduce the membrane-inserted SecA irrespective of the presence or absence of PMF. The PMF-dependent decrease in the membrane-inserted SecA caused an increase in the amount of SecA released into the extra-membrane milieu, indicating that PMF deinserts SecA from the membrane. The PMF-dependent deinsertion reduced the amount of SecA required for maximal translocation activity. Neither ATP hydrolysis nor exchange with external SecA was required for the PMF-dependent deinsertion of SecA. These results indicate that the SecA deinsertion is a limiting step of protein translocation and is accelerated by PMF, efficient protein translocation thereby being caused in the presence of PMF.     

In vitro translocation of protein across Escherichia coli membrane vesicles requires both the proton motive force and ATP

The energy requirement for protein translocation across membrane was studied with inverted membrane vesicles from an Escherichia coli strain that lacks all components of F1F0-ATPase. An ompF-lpp chimeric protein was used as a model secretory protein. Translocation of the chimeric protein into membrane vesicles was totally inhibited in the presence of carbonyl cyanide m-chlorophenylhydrazone (CCCP) or valinomycin and nigericin and partially inhibited when either valinomycin or nigericin alone was added. Depletion of ATP with glucose and hexokinase resulted in the complete inhibition of the translocation process, and the inhibition was suppressed by the addition of ATP-generating systems such as phosphoenolpyruvate-pyruvate kinase or creatine phosphate-creatine kinase. These results indicate that both the proton motive force and ATP are required for the translocation process. The results further suggest that both the membrane potential and the chemical gradient of protons (delta pH), of which the proton motive force is composed, participate in the translocation process.     

Proton motive force-dependent Hoechst 33342 transport by the ABC transporter LmrA of Lactococcus lactis

The fluorescent compound Hoechst 33342 is a substrate for many multidrug resistance (MDR) transporters and is widely used to characterize their transport activity. We have constructed mutants of the adenosine triphosphate (ATP) binding cassette (ABC)-type MDR transporter LmrA of Lactococcus lactis that are defective in ATP hydrolysis. These mutants and wild-type LmrA exhibited an atypical behavior in the Hoechst 33342 transport assay. In membrane vesicles, Hoechst 33342 transport was shown to be independent of the ATPase activity of LmrA, and it was not inhibited by orthovanadate but sensitive to uncouplers that collapse the proton gradient and to N,N'-dicyclohexylcarbodiimide, an inhibitor of the F0F1-ATPase. In contrast, transport of Hoechst 33342 by the homologous, heterodimeric MDR transporter LmrCD showed a normal ATP dependence and was insensitive to uncouplers of the proton gradient. With intact cells, expression of LmrA resulted in an increased rate of Hoechst 33342 influx while LmrCD caused a decrease in the rate of Hoechst 33342 influx. Cellular toxicity assays using a triple knockout strain, i.e., L. lactis delta lmrA delta lmrCD, demonstrate that expression of LmrCD protects cells against the growth inhibitory effects of Hoechst 33342, while in the presence of LmrA, cells are more susceptible to Hoechst 33342. Our data demonstrate that the LmrA-mediated Hoechst 33342 transport in membrane vesicles is influenced by the transmembrane pH gradient due to a pH-dependent partitioning of Hoechst 33342 into the membrane.     

Proton translocation in cytochrome-deficient mutants of Escherichia coli.

Cytochrome-deficient cells of a strain of Escherichia coli lacking 5-amino-levulinate synthetase have been used to study proton translocation associated with the reduced nicotinamide adenine dinucleotide (NADH) dehydrogenase region of the electron transport chain. Menadione was used as electron acceptor, and mannitol was used as the substrate for the generation of intracellular NADH. The effects of iron deficiency on NADH- and D-lactate-menadione reductase activities were studied in iron-deficient cells of a mutant strain unable to synthesize the iron chelator enterochelin; both activities were reduced. The NADH- menadione reductase activity in cytochrome-deficient cells was associated with proton translocation and could be coupled to the uptake of proline. However proton translocation associated with the NADH-menadione reductase activity was prevented by a mutation in an unc gene. It was concluded that there is no proton translocation associated with the NADH-dehydrogenase region of the electron transport chain in E. coli and that the proton translocation obtained with mannitol as substrate is due to the activity of membrane-bound adenosine triphosphatase.     

Purification of the Escherichia coli secB gene product and demonstration of its activity in an in vitro protein translocation system

Mutations in the Escherichia coli secB gene lead to protein export defects in vivo (Kumamoto, C.A., and Beckwith, J. (1985) J. Bacteriol. 163, 267-274). To demonstrate directly the participation of the secB gene product (SecB) in protein export, SecB was purified, and its effects on in vitro protein translocation were analyzed. SecB was purified from soluble extracts of a strain that overproduced it, by ammonium sulfate precipitation, DEAE-cellulose chromatography, and differential precipitation at acid pH. The chromatographic behavior on gel filtration columns indicated apparent molecular masses of approximately 90 kDa for both purified SecB and SecB in cytosolic extracts of wild type cells. When added to a translocation mixture, purified SecB stimulated pro-OmpA translocation into membrane vesicles. SecB also suppressed the thermoinduced defect in translocating activity of membranes derived from a secY24 mutant. The results of these in vitro studies and of previous in vivo studies demonstrate that SecB plays a direct role in normal protein export in E. coli.     

The determination by radiochemical assay of argininosuccinase produced in an Escherichia coli system in vitro.

A highly sensitive radiochemical assay used to measure the synthesis and regulation of the product of the argH gene, argininosuccinase, in an Escherichia coli system in vitro is described. With L-[guanidino-14C]argininosuccinic acid as a substrate, and in the presence of excess arginase and urease, 14CO2 is collected in a simply designed micro-vessel. With this method less than 1 nmol of product can be measured in the presence of various concentrations of L-arginine.     

The proton motive force lowers the level of ATP required for the in vitro translocation of a secretory protein in Escherichia coli

The role of the electrochemical potential difference of proton (delta mu H+) in protein translocation across the membrane of Escherichia coli was examined in detail using an efficient in vitro assay system (Yamada, H., Tokuda, H., and Mizushima, S. (1989) J. Biol. Chem. 264, 1723-1728). Delta mu H+ reduced the level of ATP necessary for the efficient translocation of OmpF-Lpp, a chimeric model secretory protein. The apparent Km value of the translocation reaction for ATP was lower by 2 orders of magnitude in the presence of delta mu H+ than in its absence. The membrane potential and delta pH, both of which are components of delta mu H+, independently lowered the apparent Km value of the translocation reaction for ATP. An ATP-generating system also lowered the level of ATP required for translocation in the absence of delta mu H+ but not in its presence. It is proposed that ADP formed during protein translocation lowers the affinity of the putative translocation machinery for ATP and that the removal of ADP from the secretory machinery, a possible critical step in the translocation reaction, is stimulated in the presence of either delta mu H+, an ATP-generating system, or a higher concentration of ATP.     

Factors influencing the in vitro translocation of the Escherichia coli maltose-binding protein

An in vitro system has been utilized to study the translocation of newly synthesized Escherichia coli maltose-binding protein (MBP) into inverted membrane vesicles. Approximately 40% of precursor MBP (pMBP) synthesized with a wild-type signal peptide was imported into vesicles. However, MBP species with even minor alterations in the signal peptide hydrophobic core were imported into vesicles with an efficiency much lower than predicted from in vivo studies. Posttranslational import of wild-type pMBP into vesicles could be demonstrated if membranes were added after the termination of protein synthesis. However, if vesicles were present throughout the synthesis reaction, most pMBP import occurred either cotranslationally or very soon after completion of synthesis. The wild-type pMBP rapidly became incompetent for posttranslational translocation upon continued incubation in the absence of membranes, whereas pMBP species with altered folding properties remained competent for significantly longer periods. The rate of in vitro pMBP folding was affected by the nature of the signal peptide. The evidence suggests that one or more soluble factors may interact with the newly synthesized pMBP to help maintain it in a translocation-competent state and to promote its entrance into the export pathway.     

Proton motive force is not obligatory for growth of Escherichia coli.

When 50 microM carbonyl cyanide-m-chlorophenyl hydrazone (CCCP), a protonophore, was added to growth medium containing glucose at pH 7.5, Escherichia coli TK1001 (trkD1 kdpABC5) started exponential growth after 30 min; the generation time was 70 min at 37 degrees C. Strain AS1 (acrA), another strain derived from E. coli K-12, also grew in the presence of 50 microM CCCP under the same conditions, except that the lag period was ca. 3 h. When this strain was grown in the presence of 50 microM CCCP and then transferred to fresh medium containing 50 microM CCCP, cells grew without any lag. Neither a membrane potential nor a pH gradient was detected in strain AS1 cells growing in the presence of CCCP. When either succinate or lactate was substituted for glucose, these strains did not grow in the presence of 50 microM CCCP. Thus, it is suggested that E. coli can grow in the absence of a proton motive force when glucose is used as an energy source at pH 7.5.     

Proton translocation and the respiratory nitrate reductase of Escherichia coli.

Stoicheometries and rates of proton translocation associated with respiratory reduction of NO3- have been measured for spheroplasts of Escherichia coli grown anaerobically in the presence of NO3-. Observed stoicheiometries [leads to H+/NO3- ratio; P. Mitchell (1966) Chemiosmotic Coupling in Oxidative and Photosynthetic Phosphorylation, Glynn Research, Bodmin] were approx. 4 for L-malate oxidation and approx. 2 for succinate, D-lactate and glycerol oxidation. Measurements of the leads to H+/2e- ratio with formate as the reductant and oxygen or NO3- as the oxidant were complicated by pH changes associated with formate uptake and CO2 formation. Nevertheless, it was possible to conclude that the site of formate oxidation is on the inner aspect of the cytoplasmic membrane, that the leads to H+/O ratio for formate oxidation is approx. 4, and that the leads to H+/NO3- ratio is greater than 2. Measurements of the rate of NO3- penetration into osmotically sensitive spheroplasts demonstrated an electrogenic entry of NO3- anion. The permeability coefficient for nitrate entry at 30 degrees C was between 10(-9) and 10(-10) cm- s(-1). The calculated rate of nitrate entry at the concentration typically used for the assay of nitrate reductase (EC 1.7.99.4) activity was about 0.1% of that required to support the observed rate of nitrate reduction by reduced Benzyl Viologen. Measurements of the distribution of nitrate between the intracellular and extracellular spaces of a haem-less mutant, de-repressed for nitrate reductase but unable to reduce nitrate by the respiratory chain, showed that, irrespective of the presence or the absence of added glucose, nitrate was not concentrated intracellularly. Osmotic-swelling experiments showed that the rate of diffusion of azid anion across the cytoplasmic membrane is relatively low in comparison with the fast diffusion of hydrazoic acid. The inhibitory effect of azide on nitrate reductase was not altered by treatments that modify pH gradients across the cytoplasmic membrane. It is concluded that the nitrate-reducing azide-sensitive site of nitrate reductase is located on the outer aspect of the cytoplasmic membrane. The consequences of this location for mechanisms of proton translocation driven by nitrate reduction are discussed, and lead to the proposal that the nitrate reductase of the cytoplasmic membrane is vectorial, reducing nitrate on the outer aspect of the membrane with 2H+ and 2e- that have crossed from the inner aspect of the membrane.     

SecY and integral membrane components of the Escherichia coli protein translocation system

Genetic approaches can address the question of how integral membrane Sec factors interact with each other and facilitate protein translocation across the cytoplasmic membrane of E. coli. This review summarizes genetic analyses of SecY, SecE and some other protein translocation factors, utilizing 'prl' mutations, 'sec' mutations, 'suppressor-directed inactivation', 'Sec titration', dominant negative mutations and their suppressors. Evidence suggests that co-ordinate participation of SecY, SecE, SecD, SecF, and probably some other factors, is crucial for the process.     

Proton translocation coupled to nitrite reduction in anaerobically grown Escherichia coli

Proton translocation coupled to the reduction of nitrite was studied in anaerobically grown Escherichia coli. Extrusion of protons occurred by adding nitrite to an anaerobic suspension of wild-type cells. This extrusion was sensitive to a proton conductor, 3,5-di-tert-butyl-4-hydroxybenzylidenemalononitrile (SF6847) or carbonylcyanide-p-trifluoromethoxyphenylhydrazone. Dicyclohexylcarbodiimide, an inhibitor of H+-ATPase, prevented the proton extrusion linked to nitrite reduction, whereas this reagent had no effect on respiratory nitrate reduction to nitrite. Proton extrusion was undetectable when nitrite was added to a suspension of mutant cells defective in H+-ATPase. These results indicate that the proton extrusion associated with nitrite reduction to ammonia is not by redox pumps but by H+-ATPase. From the results obtained by the measurement of proton extrusion in nitrite reductase-deficient mutants, NADH-nitrite reductase system is suggested to involve the proton extrusion in whole cells of E. coli.     

Lipid involvement in protein translocation in Escherichia coli

Signal peptides play an essential role in protein translocation. This review summarizes the current knowledge of the structure of signal peptides and signal peptide-lipid interactions and addresses the possibility that signal peptide-lipid interactions initiate membrane translocation of precursor proteins. A new model for protein translocation in Escherichia coli is proposed, which includes as central features conformational changes of the signal peptide and signal-peptide-induced local changes in membrane organization (non-bilayer lipids).     

Construction and characterization of mutated LEA peptides in Escherichia coli to develop an efficient protein expression system

To develop an efficient protein expression system, we designed a late embryogenesis abundant (LEA) peptide by mutating the LEA peptide constructed in our previous study (LEA‐I). The peptide is based on the repeating units of an 11mer motif characteristic of LEA proteins from Polypedilum vanderplanki larvae. In the amino acid sequence of the 13mer LEA peptide, glycine at the 6th and 12th positions was replaced with other amino acids via point mutations. Glutamic acid, lysine, leucine, and asparagine in the LEA peptide at the 6th and 12th positions increased green fluorescence protein (GFP) expression. The GFP expression of the mutated LEA peptide was 1.5 to 2.0 times higher than that without LEA peptide. In contrast, the serine‐containing mutated LEA peptide has low GFP expression levels. We hypothesize that the position of amino acids and the nature of amino acid in LEA peptide are important for our coexpression system. These data suggest that the size, structure, and charge of amino acids in the LEA peptide improve the protection and expression of the target protein. The amino acid balance also plays an important role in the expression of the target protein.     

Proton translocation coupled to formate oxidation in anaerobically grown fermenting Escherichia coli

Proton translocation, coupled to formate oxidation and hydrogen evolution, was studied in anaerobically grown fermenting Escherichia coli JW136 carrying hydrogenase 1 (hya) and hydrogenase 2 (hyb) double deletions. Rapid acidification of the medium by EDTA-treated anaerobic suspension of the whole cells or its alkalization by inverted membranes was observed in response to application of formate. The formate-dependent proton translocation and 2H(+)-K(+) exchange coupled to H(2) evolution were sensitive to the uncoupler, carbonylcyanide-m-chlorophenylhydrazone, and to copper ions, inhibitors of hydrogenases. No pH changes were observed in a suspension of formate-pulsed aerobically grown ("respiring") cells. The apparent H(+)/formate ratio of 1.3 was obtained in cells oxidizing formate. The 2H(+)-K(+) exchange of the ATP synthase inhibitor N,N'-dicyclohexylcarbodiimide-sensitive ion fluxes does take place in JW136 cell suspension. Hydrogen formation from formate by cell suspensions of E. coli JW136 resulted in the formation of a membrane potential (Deltapsi) across the cytoplasmic membrane of -130 mV (inside negative). This was abolished in the presence of copper ions, although they had little effect on the value of Deltapsi generated by E. coli under respiration. We conclude that the hydrogen production by hydrogenase 3 is coupled to formate-dependent proton pumping that regulates 2H(+)-K(+) exchange in fermenting bacteria.     

The type III protein translocation system of enteropathogenic Escherichia coli involves EspA-EspB protein interactions

Enteropathogenic Escherichia coli (EPEC), like many bacterial pathogens, use a type III secretion system to deliver effector proteins across the bacterial cell wall. In EPEC, four proteins, EspA, EspB, EspD and Tir are known to be exported by a type III secretion system and to be essential for 'attaching and effacing' (A/E) lesion formation, the hallmark of EPEC pathogenicity. EspA was recently shown to be a structural protein and a major component of a large, transiently expressed, filamentous surface organelle which forms a direct link between the bacterium and the host cell. In contrast, EspB is translocated into the host cell where it is localized to both membrane and cytosolic cell fractions. EspA and EspB are required for translocation of Tir to the host cell membrane suggesting that they may both be components of the translocation apparatus. In this study, we show that EspB co-immunoprecipitates with the EspA filaments and that, during EPEC infection of HEp-2 cells, EspB localizes closely with EspA. Using a number of binding assays, we also show that EspB can bind and be copurified with EspA. Nevertheless, binding of EspA filaments to the host cell membranes occurred even in the absence of EspB. These results suggest that following initial attachment of the EspA filaments to the target cells, EspB is delivered into the host cell membrane and that the interaction between EspA and EspB may be important for protein translocation.     

Efficient in vitro translocation into Escherichia coli membrane vesicles of a protein carrying an uncleavable signal peptide. Characterization of the translocation process

The translocation into Escherichia coli cytoplasmic membrane vesicles of a protein containing an uncleavable signal peptide was studied. The signal peptide cleavage site of the ompF-lpp chimeric protein, a model secretory protein, was changed from Ala-Ala to Phe-Pro through oligonucleotide-directed site-specific mutagenesis of the ompF-lpp gene on a plasmid. The mutant protein was no longer processed by the signal peptidase. When proteinase K treatment was adopted as a probe for protein translocation into inverted membrane vesicles, the mutant protein exhibited rapid and almost complete translocation, most likely due to the lack of premature cleavage of the signal peptide before the translocation. This result also indicates that cleavage of the signal peptide is not required for translocation of the mature domain of the protein. The establishment of an efficient system made it possible to perform precise and quantitative analysis of the translocation process. The translocation was time-dependent, vesicle-dependent, and required ATP and NADH. Translocation into membrane vesicles was also observed with the uncleavable precursor protein purified by means of immunoaffinity chromatography, although the efficiency was appreciably low. The translocation required only ATP and NADH. Addition of the cytosolic fraction did not enhance the translocation.     

Flexibility in actin-myosin motility system revealed by in vitro motility assay

Flexibility of myosin molecule was studied by in vitro motility assay in terms of the direction of actin movement. Electron microscopy showed that HMM scattered on a nitrocellulose surface can bind actin filaments and form arrowhead-like patterns. Actin filaments can move in both directions on tracks of HMM made on a nitrocellulose surface. Further, actin filaments can move bidirectionally along native thick filaments over their central bare zone. These observations indicate that there is considerable flexibility in a myosin molecule and that the direction of the movement is determined by the polarity of actin filaments.     

A high concentration of SecA allows proton motive force-independent translocation of a model secretory protein into Escherichia coli membrane vesicles

The in vitro translocation of OmpF-Lpp, a model secretory protein, into inverted membrane vesicles of Escherichia coli obligatorily requires the proton motive force (delta mu H+) in the conventional assay system (Yamada, H., Tokuda, H., and Mizushima, S. (1989) J. Biol. Chem. 264, 1723-1728). The translocation, however, took place efficiently, even in the absence of delta mu H+, when the system was supplemented with additional SecA. With the stripped membrane vesicles, which are permeable to protons, or in the absence of NADH, the supplementation of SecA remarkably stimulated the translocation activity. The further addition of NADH did not significantly enhance the translocation activity under the SecA-enriched conditions. OmpF-Lpp thus translocated could be recovered from the vesicular lumen by sonication, indicating that complete translocation occurred in the absence of delta mu H+. It is suggested that delta mu H+ is required for high affinity interaction of SecA with the presumed secretory machinery in the cytoplasmic membrane and that a high concentration of SecA modulates the delta mu H+ requirement.