Purification of a factor for both aerobic-driven and ATP-driven energy-dependent transhydrogenases of Escherichia coli

ATP Synthesis by Oxidative Phosphorylation

The F1F0-ATP synthase (EC 3.6.1.34) is a remarkable enzyme that functions as a rotary motor. It is found in the inner membranes of Escherichia coli and is responsible for the synthesis of ATP in response to an electrochemical proton gradient. Under some conditions, the enzyme functions reversibly and uses the energy of ATP hydrolysis to generate the gradient. The ATP synthase is composed of eight different polypeptide subunits in a stoichiometry of α3β3γδεab2c10. Traditionally they were divided into two physically separable units: an F1 that catalyzes ATP hydrolysis (α3β3γδε) and a membrane-bound F0 sector that transports protons (ab2c10). In terms of rotary function, the subunits can be divided into rotor subunits (γεc10) and stator subunits (α3β3δab2). The stator subunits include six nucleotide binding sites, three catalytic and three noncatalytic, formed primarily by the β and α subunits, respectively. The stator also includes a peripheral stalk composed of δ and b subunits, and part of the proton channel in subunit a. Among the rotor subunits, the c subunits form a ring in the membrane, and interact with subunit a to form the proton channel. Subunits γ and ε bind to the c-ring subunits, and also communicate with the catalytic sites through interactions with α and β subunits. The eight subunits are expressed from a single operon, and posttranscriptional processing and translational regulation ensure that the polypeptides are made at the proper stoichiometry. Recent studies, including those of other species, have elucidated many structural and rotary properties of this enzyme.

Purification and properties of the F1-ATPase from liver mitochondria of Gallus gallus

1. This paper is the first detailed report of the purification of a mitochondrial ATPase from an avian species. 2. The Gallus gallus liver mitochondrial F1-ATPase was purified by chloroform extraction and ion-exchange chromatography. 3. The enzyme shows the five alpha, beta, tau, delta, and epsilon subunits characteristic of mitochondrial F1-ATPases. 4. The Km for ATP is 1 mM and for Mg 0.5 mM with a specific activity of 25.2 mu moles of ATP hydrolyzed x min-1 x mg-1. 5. Unlike mammals enzymes the chicken mitochondrial ATPase shows maximal activity with ITP as substrate, and is strongly inhibited by Cu.

Bacterial adenosine 5'-triphosphate synthase (F1F0): purification and reconstitution of F0 complexes and biochemical and functional characterization of their subunits

Images

Proton ATPase of rat liver mitochondria: a rapid procedure for purification of a stable, reconstitutively active F1 preparation using a modified chloroform method

A method is described for the purification of rat liver F1-ATPase by a modification of the chloroform extraction procedure originally described by Beechey et al. (Biochem. J. (1975) 148, 533). Purified liver membrane vesicles are extracted with chloroform in the presence of ATP and EDTA. The procedure yields pure F1 in only 2-3 h without the necessity of ion-exchange chromatography. The enzyme exhibits the alpha, beta, gamma, delta, and epsilon bands characteristic of F1-ATPase. It has a high ATPase specific activity, and is reconstitutively active, catalyzing high rates of ATP synthesis. Significantly, it can be readily crystallized. If desired, the enzyme can be passed over a gel filtration column to place it in a stabilizing phosphate-EDTA buffer, lyophilized and stored indefinitely at -20 degrees C.

The DCCD-reactive aspartyl-residue of subunit C from the Escherichia coli ATP-synthase is important for the conformation of F0

The effect of various point mutations in subunits a and and c of the E. coli ATP-synthase was characterized. In each of the mutants there was no F0-dependent H+-conduction, but still an ATPase-activity comparable to wildtype activities. In addition, the subunit b could be extracted from the mutant's F0 but not from the F0 of wildtype. The effects are interpreted as a change in the conformation of F0 caused by the different mutations.

Mechanism of glutamate transport in Escherichia coli B. 2. Kinetics of glutamate transport driven by artificially imposed proton and sodium ion gradients across the cytoplasmic membrane

Simultaneous imposition of a pH gradient (delta pH, interior alkaline) and a sodium gradient (delta pNa, [Na+]out greater than [Na+]in) across cytoplasmic membrane vesicles from Escherichia coli B led to a several hundred fold accumulation of glutamate. Although less effective, delta pH (interior alkaline)( alone caused accumulation of glutamate in the presence of Na+. In addition, delta pNa ([Na+]out greater than [Na+]in) alone also drove the transport system, where the maximum level of glutamate accumulation was affected by the pH of the medium. A membrane potential imposed by valinomycin-induced K+ diffusion (interior negative) enhanced the accumulation, indicating that the system operation in an electrogenic manner. The Michaelis constant of glutamate transport was greatly affected by changes in the concentrations of both Na+ and H+ and could be expressed by a linear combination of the reciprocals of the Na+ and H+ concentrations in the medium. On the contrary, a membrane potential (interior negative) exerted its effect by increasing the maximum velocity. When membrane vesicles were loaded with glutamate and Na+, but not with glutamate alone, rapid efflux of glutamate with Na+ as the cocation down the concentration gradients took place upon dilution. These results indicate that both Na+ and H+ are syn-coupled ions of glutamate transport in E. coli B and that the carrier/Na+/H+/Glu- complex observed in the binding reaction is an intermediate in the transport.

Differential polypeptide synthesis of the proton-translocating ATPase of Escherichia coli

We investigated the regulation of the synthesis of the eight polypeptides of the Escherichia coli proton-translocating ATPase. A plasmid carrying the eight genes of the unc operon was used to direct in vivo and in vitro protein synthesis of the eight polypeptides. Analysis of these data indicates that the ATPase polypeptides are synthesized in unequal amounts both in vitro and in vivo. We identified several regions within the unc operon at which expression of a gene is either increased or decreased from that of the preceding gene. Since genetic information indicates a single polycistronic mRNA for all eight genes of this operon, the observed differential synthesis of the polypeptides is most likely the result of translational regulation. The effect of varying the temperature suggests that the secondary structure in the mRNA may affect the rate of translation initiation in the region between uncE and uncF.

Gene order and gene-polypeptide relationships of the proton-translocating ATPase operon (unc) of Escherichia coli

We have constructed an extensive set of plasmids that carry the genes specifying the eight polypeptides of the proton-translocating ATPase of Escherichia coli. Using detailed restriction analysis and in vitro protein synthesis directed by these plasmids, we have established the order of the eight unc genes to be BEFHAGDC and the corresponding polypeptides to be a, c, b, delta, alpha, gamma, beta, and epsilon. These analyses include determining the location of the gene coding for the delta subunit of the F1 portion of the complex. We call this gene uncH. We have now established the gene order and gene-polypeptide relationships of the unc operon. This approach should be of use for study of other multigene bacterial operons, especially those with genes coding for polypeptides with unknown or unmeasurable catalytic activity.

The nucleotide sequence of the atp genes coding for the F0 subunits a, b, c and the F1 subunit delta of the membrane bound ATP synthase of Escherichia coli

The nucleotide sequence has been determined of a 2,500 base pair segment of the E. coli chromosome located between 3.75 and 6.25 kb counterclockwise of the origin of replication at 83.5 min. The sequence contains the atp genes coding for subunits a-, b-, c-, delta- and part of the alpha-subunit of the membrane bound ATP synthase. The precise start positions of the atpE (c), atpF (b), atpH (delta) and atpA (alpha) genes have been defined by comparison of the potential coding sequences with the known amino acid sequence of the c-subunit and the determined N-terminal amino acid sequences of the respective subunits. The genes are expressed in the counterclockwise direction. Their order (counterclockwise) is: atpB (a), atpE (c), atpF (b), atpH (delta) and atpA(alpha). The coding sequences for subunits b and delta yield polypeptides of 156 and 177 amino acids, respectively, in accordance with the established sizes of these subunits; the one for the c-subunit, the DCCD binding protein, fits perfectly with its known sequence of 79 amino acids. The a-subunit is comprised within a coding sequence yielding a polypeptide of 271 amino acids. It is suggested, however, that the a-subunit (atpB) contains only 201 amino acids, in accordance with its known size, starting from a translation initiation site within the larger coding sequence. The stoichiometry of the F0 sector subunits is discussed and a model is proposed for the functioning of the highly charged b-subunit of the F0 sector as the actual proton conductor.

The dicyclohexylcarbodiimide-binding protein c of ATP synthase from Escherichia coli is not sufficient to express an efficient H+ conduction

Bacteriophage Mu was inserted into the unc genes of Escherichia coli. The resulting mutation AS12 had a polar effect on the unc operon: membranes of the mutant AS12 contained the dicyclohexylcarbodiimide-binding protein c and the protein a as sole subunits of the ATP synthase. It was shown by peptide mapping and amino acid analysis of the fragments that protein c from mutant AS12 was identical with the wild-type protein c. The absence of subunit b in mutant AS12 drastically lowered the H+ conduction dependent on the membrane-integrated moiety (F0) of the ATP synthase. This suggests that both subunits b and c are necessary for an efficient expression of H+ conduction.

A cross-linking study of the Ca2+, Mg2+-activated adenosine triphosphatase of Escherichia coli

The solubilized Ca2+,Mg2+-activated adenosine triphosphatase of Escherichia coli is composed of five subunits designated alpha, beta, gamma, delta and epsilon in order of decreasing molecular weight. The subunit structure of the enzyme has been investigated by the use of the cleavable cross-linking agents dithiobis(succinimidyl propionate), methyl-4-mercaptobutyrimidate, dimethyl-3,3'-dithiobispropionimidate, disuccinimidyl tartarate, and cupric 1,10-phenanthrolinate. The products of cross-linking were analyzed by two different two-dimensional gel electrophoresis systems. The following cross-linked subunit dimers were observed: alpha 2, beta 2, alpha beta, alpha delta, beta gamma, beta delta, beta epsilon and gamma epsilon. These results, together with other published data, are discussed in relation to a model of the arrangement of the subunits in the ATPase molecule.

Characterization of the inhibitory (epsilon) subunit of the proton-translocating adenosine triphosphatase from Escherichia coli

The inhibitory subunit (epsilon) of the F1 adenosine triphosphatase (ATPase) was purified to homogeneity from the ML 308-225 and K12 (lambda) strains of Escherichia coli. No tryptophan or cysteine was detected in the subunit from either strain. The highly active epsilon from both strains was found to be a globular protein with a Stokes' radius of 18--19 A. Circular dichroism spectra suggested an alpha-helix content of approximately 40%. The molecular weight of epsilon was approximately 15000--16000 by sedimentation equilibrium centrifugation in the presence and absence of guanidinium hydrochloride, molecular sieve chromatography, and gel electrophoresis in the presence of sodium dodecyl sulfate and 8 M urea. The s20,w of epsilon was approximately 1.6 s-1. Inhibition of the purified F1 ATPase by epsilon displayed noncompetitive kinetics with a Ki of approximately 10 nM. The inhibition of the ATPase was rapidly reversed by diluting the enzyme--epsilon mixture. [125I]epsilon which was incorporated into ECF1 was readily displaced by unlabeled epsilon. epsilon had no significant effect on the ATPase activity of "native" or reconstituted everted membrane vesicles under a variety of assay conditions. Combining the epsilon-inhibited F1 ATPase with its hydrophobic portion in everted membrane vesicles reconstituted the reversible proton-translocating ATPase and restored nearly full ATPase activity. These results suggest that epsilon inhibits the enzyme only when the F1 ATPase becomes detached from its hydrophobic subunits.

The ATP synthetase of Escherichia coli K12: purification of the enzyme and reconstitution of energy-transducing activities

The ATP synthetase of Escherichia coli K12 was purified by a simple procedure. The dicyclohexylcarbodiimide-sensitive ATPase activity was enriched 21-fold. The ATP synthetase preparation contained the eight polypeptides (alpha, beta, gamma, a,delta, b,espilon, c) of the enzyme and a residual contamination (4% of the total protein) as shown by dodecylsulfate/polyacrylamide electrophoresis. The polypeptide c was specifically labelled with [14C]dicyclohexylcarbodiimide. Energy-transducing activities were reconstituted from soybean phospholipids and the purified enzyme. The proteoliposomes exhibited a significantly higher ATP-32Pi exchange activity and a higher proton-translocating activity as compared to the untreated membranes.

Method for isolation of Escherichia coli mutants with defects in the proton-translocating sector of the membrane adenosine triphosphatase complex

A technique for selecting mutants of Escherichia coli in which the proton-translocating sector of the adenosine triphosphatase (ATPase) complex has been inactivated is reported. The procedure uses a strain of E. coli (NR-70) lacking the extrinsic (F1) sector of the ATPase complex and which in consequently permeable to protons (B. P. Rosen, J. Bacteriol. 116:1124--1129, 1973). After growing strain NR-70 under noninducing conditions for the lac operon, cells were mutagenized and plated on minimal medium containing low concentrations of lactose. Several mutants of strain NR-70 were isolated as large colonies on these plates, apparently because they could concentrate lactose more efficiently. A description of one of the mutants, strain KW-1, is reported here. The most distinguishing difference in growth properties of the two strains was that, when transferred to medium containing low concentrations of lactose, strain KW-1 induced the lac operon with a shorter lag time than strain NR-70. The mutation in strain KW-1 leading to more rapid growth on lactose was cotransducible with the asn and unc loci, at 83 min on the E. coli genetic map. Intact cells of strain KW-1 actively transported L-proline as well as did wild-type cells, whereas cells of strain NR-70 were markedly deficient in L-proline transport. The improvement in the transport capacity of strain KW-1 correlated with a marked decrease in proton permeability relative to that of strain NR-70. Based on an acid-base pulse technique that measured the proton conductance of the membranes of intact cells, strain NR-70 was at least 10 times more permeable to protons than was the wild type, whereas strain KW-1 was only 2 times more permeable. The transport properties and proton conductance were also compared with membrane vesicles prepared by osmotic shock. With either D-lactate or ascorbate-N-methylphenazonium methosulfate as respiratory substrates, vesicles of strain KW-1 transported L-proline much more rapidly than did vesicles of strain NR-70, but still at rates less rapid than those of the wild type. The passive proton conductance of the membrane vesicles was quantitated by measuring the rate of H+ influx into vesicles in response to a valinomycin-generated K+ diffusion potential. The proton permeability of vesicles of strain KW-1 was reduced 1.5-fold relative to vesicles of strain NR-70, but these vesicles were still four times more permeable to protons than was the wild type. Vesicles of strain KW-1 corresponded to wild-type vesicles treated with 0.5 micrometer carbonylcyanide m-chlorophenylhydrazone (CCCP) and vesicles of strain NR-70 corresponded to wild-type vesicles treated with 1.4 micrometer CCCP. Treatment of wild-type vesicles with these concentrations of CCCP caused decreases in transport comparable to those observed in the mutants. Strain KW-1 lacked ATPase activity. Cross-reacting material to F1-ATPase was not found in strain KW-1 by double immunodiffusion analysis.

Membrane ATPase of Bacillus subtilis. I. Purification and properties

The membrane ATPase (EC 3.6.1.3) of Bacillus subtilis can be solubilized by a shock-wash process. Two procedures for purifying the solubilized enzyme are reported. A protease inhibitor, phenylmethane sulfonylfluoride, was introduced in the solubilization and purification step. The resultant ATPase purified by density gradient centrifugation has a molecular weight of 315 000, an s20,w of 13,4 and an amino acid composition very similar to bacterial ATPases already studied. After exposure to polyacrylamide gel electrophoresis in presence of sodium dodecyl sulphate (SDS), or 8 M urea or SDS-urea, the purified ATPase can be dissociated in two non-identical subunits of molecular weights 59 000 (alpha) and 57 000 (beta) with different charges. Kinetic studies showed that Ca2+ or Zn2+ are required for ATPase activity, although Mg2+ was uneffective. At optimal Ca2+ concentration, the Mg2+ has an inhibitory effect. The Km for ATP is 1.3 mM. Inhibitors of the oxydative phosphorylation, of the mitochondrial ATPase and of the (Na+ + K+)-ATPase are studied.

Inhibition, by a protease inhibitor, of the solubilization of the F1-portion of the Mg2+-stimulated adenosine triphosphatase of Escherichia coli

The effects of two protease inhibitors on the solubilization of the membrane-bound Mg2+-adenosine triphosphatase (Mg-ATPase) of Escherichia coli were investigated. p-Aminobenzamidine prevented the solubilization of the Mg-ATPase during treatment of membranes with low-ionic-strength buffers containing ethylenediaminetetraacetic acid. p-Aminobenzamidine did not prevent subsequent solubilization of the Mg-ATPase by treatment of the membranes with chloroform. This method of solubilization yielded a preparation of similar apparent molecular weight but with a 10-fold-increased specific activity as compared with the Mg-ATPase solubilized by washing with low-ionic-strength buffer. However, in contrast to the latter preparation, the chloroform-solubilized Mg-ATPase did not reconstitute ATP-dependent energization of stripped membranes, which were prepared by low-ionic-strength washing in the absence of p-aminobenzamidine. Another protease inhibitor, epsilon-amino-n-caproic acid, did not effect the solubilization of the Mg-ATPase, but did inhibit the loss of activity occurring during concentration, by ultrafiltration, of the Mg-ATPase solublized by the low-ionic-strength treatment.

Adenosine 5'-triphosphate-linked transhydrogenase in cytoplasmic membranes of colicin-treated and untreated Escherichia coli

The adenosine 5'-triphosphate (ATP)-linked transhydrogenase reaction, present in the particulate fractions of Escherichia coli, was previously shown to be inhibited in these fractions when the bacteria were treated with colicins K or El. The purpose of this study was to characterized the ATP-linked transhydrogenase reaction and the colicin-caused inhibition of the reaction in purified cytoplasmic membranes. Particulate fractions from bacteria treated or untreated with colicins were separated on sucrose gradients into cell wall membrane and cytoplasmic membrane fractions. The ATP-linked transhydrogenase reaction was found to be exclusively associated with the cytoplasmic membrane fractions. The reaction was inhibited by carbonylcyanide m-chlorophenlhdrazone, dinitrophenol, N,N'-dicyclohexylcarbodiimide, and trypsin. Although the cytoplasmic membrane fractions were purified from the majoriy of the cell wall membrane and its bound colicins, they showed the inhibitory effects of colicins K and El on the ATP-linked transhydrogenase reaction. The inhibition of ATP-linked transhydrogenase reaction induced by the colicin could not be reversed by subjection the isolated membranes to a variety of physical and chemical treatments. Cytoplasmic membranes depleted of energy-transducing adenosine triphosphatase ATPase) complex (coupling factor) lost the ATP-linked transhydrogenase activity. The ATPase complexes isolated from membranes of bacteria treated or untreated with colicins El or K reconstituted high levels of ATP-linded transhydrogenase activity to depleted membranes of untreated bacteria. The same ATPase complexes reconstituted low levels of activity to depleted membranes of the treated bacteria.

A mutant ATP synthetase of Escherichia coli with an altered sensitivity to N,N' -dicyclohexylcarbodiimide: characterization in native membranes and reconstituted proteoliposomes

Dicyclohexylcarbodiimide-resistant mutants of Escherichia coli were isolated and characterized In one mutant the unc genes and affects the membrane-integrated part of the ATP synthetase. The sensitivity of ATP synthetase functions to N,N' -dicyclohexylcarbodiimide was compared in wild-type and mutant membranes. The membrane-integrated part of the wild-type ATP synthetase is highly sensitive to ATP-dependent membrane energization and restoration of lactate-dependent energization of ATPase-depleted membranes. In mutant membranes this concentration has only a slight effect on these activities whereas a severe inhibition is obtained at 200 muM. Using the highly water-soluble 1-ethyl-3(3-dimethylaminopropyl)-carbodiimide theactivities of wild-type and mutant membranes are inhibited to the same extent. TheATP synthetase of wild-type and mutant was partially purified and incorporated muM. Uinto liposomes. These showed an uncoupler-sensitive ATP-32Pi exchange and ATP-dependent quenching of acridine-dye fluorescence. The activities of mutant and wild-type proteoliposomes exhibit the same pattern of sensitivity to dicyclohexylcarbodiimide as the corresponding membranes.

Transhydrogenase activity in the marine bacterium Beneckea natriegens

The marine bacterium, Beneckea natriegens, which has previously been reported not to form transhydrogenase, has been shown to synthesize a soluble energy-independent transhydrogenase (NADPH:NADP+ oxidoreductase, EC 1.6.1.1), though no energy-linked activity could be detected. The transhydrogenase is induced maximally in stationary phase cells and its formation is 70-90% repressed by raising the medium phosphate level from 0.33 to 3.3 mM. The enzyme is inhibited by arsenate, inorganic ortho- and pyrophosphate and by a range of organic phosphate-containing compounds, including 2'-AMP, which is an activator of several bacterial transhydrogenases.

The use of several energy-coupling reactions in characterizing mutants of Escherichia coli K12 defective in oxidative phosphorylation

Oxidative phosphorylation, ATP-32Pi exchange, ATP-dependent quenching of acridine-dye fluorescence, ATP-dependent transhydrogenase and ATP-dependent transport of thiomethyl beta-D-galactoside are shown to be experimentally equivalent tools to study the functional state of the ATPase complex in Escherichia coli wild-type and mutant strains defective in oxidative phosphorylation. According to these criteria ten mutants in the ATPase complex were classified having lesions in the unc A,B region of the chromosome. The first mutant type lacks ATPase activity, but the membrane-integrated part of the complex remains functional (class I). The second mutant type lacks a functional membrane-integrated part, but retains ATPase activity (class II). The third mutant type is shown to be defective in both parts of the ATPase complex (class III).

Identification of the dicyclohexylcarbodiimide-reactive protein component of the adenosine 5'-triphosphate energy-transducing system of Escherichia coli

Membranes of Escherichia coli contain an adenosine 5'-triphosphate (ATP) energy-transducing system that is inhibited by treatment with dicyclohexylcarbodiimide (DCCD). The carbodiimide-reactive protein component of this system has been identified after treatment with [14C]DCCD. This protein has an apparent molecular weight of 9,000 as judged from acrylamide gel electrophoresis in the presence of sodium dodecyl sulfate and is extracted from the membrane with chloroform-methanol (2:1). These properties are similar to the analogous protein previously identified in mitochondria (Cattell et al., 1971). A mutant strain, RF-7, has been isolated which derives energy from oxidative phosphorylation in the presence of 5 mM DCCD. The ATP hydrolase activity of the membraned system in the mutant was considerably less sensitive to inhibition by DCCD than that in the wild type. The carbodiimide-reactive protein, which was easily labeled by [14C]DCCD in the wild type, was labeled much less rapidly in the carbodiimide-resistant mutant. It is thus concluded that the reaction of DCCD with this specific protein leads to inhibition of the ATP energy-transducing reactions. The mutation causing carbodiimide resistance in strain RF-7 was mapped. It is cotransduced with the uncA gene at a frequency exceeding 90%. The mutationally altered protein causing the carbodiimide resistance was not conclusively identified. However, reconstitution experiments indicate that the altered protein is not one of the subunits of the soluble ATP hydrolase activity, which can be removed from the membrane by washing with 1 mM tris(hydroxymethyl)aminomethane buffer lacking Mg2+. The carbodiimide-reactive protein remains with the membrane residue after removal of the soluble ATP hydrolase and is thus distinct from these subunits as well.