Modulation of coupling in the Escherichia coli ATP synthase by ADP and Pi: Role of the ε subunit C-terminal domain.

The ε-subunit of ATP-synthase is an endogenous inhibitor of the hydrolysis activity of the complex and its α-helical C-terminal domain (εCTD) undergoes drastic changes among at least two different conformations. Even though this domain is not essential for ATP synthesis activity, there is evidence for its involvement in the coupling mechanism of the pump. Recently, it was proposed that coupling of the ATP synthase can vary as a function of ADP and Pi concentration. In the present work, we have explored the possible role of the εCTD in this ADP- and Pi-dependent coupling, by examining an εCTD-lacking mutant of Escherichia coli. We show that the loss of Pi-dependent coupling can be observed also in the εCTD-less mutant, but the effects of Pi on both proton pumping and ATP hydrolysis were much weaker in the mutant than in the wild-type. We also show that the εCTD strongly influences the binding of ADP to a very tight binding site (half-maximal effect≈1nM); binding at this site induces higher coupling in EFOF1 and increases responses to Pi. It is proposed that one physiological role of the εCTD is to regulate the kinetics and affinity of ADP/Pi binding, promoting ADP/Pi-dependent coupling.

Mutual information without the influence of phylogeny or entropy dramatically improves residue contact prediction.

MOTIVATION: Compensating alterations during the evolution of protein families give rise to coevolving positions that contain important structural and functional information. However, a high background composed of random noise and phylogenetic components interferes with the identification of coevolving positions. RESULTS: We have developed a rapid, simple and general method based on information theory that accurately estimates the level of background mutual information for each pair of positions in a given protein family. Removal of this background results in a metric, MIp, that correctly identifies substantially more coevolving positions in protein families than any existing method. A significant fraction of these positions coevolve strongly with one or only a few positions. The vast majority of such position pairs are in contact in representative structures. The identification of strongly coevolving position pairs can be used to impose significant structural limitations and should be an important additional constraint for ab initio protein folding. AVAILABILITY: Alignments and program files can be found in the Supplementary Information.

Epsilon-binding regions of the gamma subunit of Escherichia coli ATP synthase.

The interaction between the gamma and epsilon subunits of the F1-ATPase sector of Escherichia coli ATP synthase has been investigated using monoclonal antibodies directed against the gamma subunit and ligand blotting using 125I-epsilon. Monoclonal antibody (MAb) gamma-1 was able to bind to epsilon-depleted F1-ATPase but not to epsilon-replete F1, implying that epsilon blocked access to the epitope. A ligand blot assay for the binding of 125I-epsilon to gamma was developed. Both MAb gamma-1 and a second antibody, MAb gamma II, inhibited binding of 125I-epsilon to gamma in this assay while two other anti-gamma monoclonal antibodies did not. The epitope recognized by MAb gamma-1 was mapped between residues R49 and R70, quite distant in sequence from that of MAb gamma II, which is located C-terminal to residue K199 of the 286-residue polypeptide. The competition of these antibodies with epsilon for binding to gamma implies that their epitopes, quite separate in sequence, are both located in parts of the subunit involved in binding epsilon.

A re-examination of the structural and functional consequences of mutation of alanine-128 of the b subunit of Escherichia coli ATP synthase to aspartic acid.

The effects of mutation of residue Ala-128 of the b subunit of Escherichia coli ATP synthase to aspartate on the structure of the subunit and its interaction with the F(1) sector were analyzed. Determination of solution molecular weights by sedimentation equilibrium ultracentrifugation revealed that the A128D mutation had little effect on dimerization in the soluble b construct, b(34-156). However, the mutation caused a structural perturbation detected through both a 12% reduction in the sedimentation coefficient and also a reduced tendency to form intersubunit disulfide bonds between cysteine residues inserted at position 132. Unlike the wild-type sequence, the A128D mutant was unable to interact with F(1)-ATPase. These results indicate that the A128D mutation caused a structural change in the C-terminal region of the protein, preventing the binding to F(1) but having little or no effect on the dimeric nature of b.

The second stalk of Escherichia coli ATP synthase.

Two stalks link the F(1) and F(0) sectors of ATP synthase. The central stalk contains the gamma and epsilon subunits and is thought to function in rotational catalysis as a rotor driving conformational changes in the catalytic alpha(3)beta(3) complex. The two b subunits and the delta subunit associate to form b(2)delta, a second, peripheral stalk extending from the membrane up the side of alpha(3)beta(3) and binding to the N-terminal regions of the alpha subunits, which are approx. 125 A from the membrane. This second stalk is essential for binding F(1) to F(0) and is believed to function as a stator during rotational catalysis. In vitro, b(2)delta is a highly extended complex held together by weak interactions. Recent work has identified the domains of b which are essential for dimerization and for interaction with delta. Disulphide cross-linking studies imply that the second stalk is a permanent structure which remains associated with one alpha subunit or alphabeta pair. However, the weak interactions between the polypeptides in b(2)delta pose a challenge for the proposed stator function.

Localization of the delta subunit in the Escherichia coli F(1)F(0)-ATPsynthase by immuno electron microscopy: the delta subunit binds on top of the F(1).

The binding site of the delta subunit in the F(1)F(0)-ATPsynthase from Escherichia coli has been determined by electron microscopy of negatively stained, antibody-decorated enzyme molecules. The images show that the antibody is bound at the very top of the F(1) domain indicating that at least part of delta is bound in the dimple formed by the N termini of the alpha and beta subunits. The data may explain why there is only one binding site for delta on the F(1) despite there being three identical alphabeta pairs. The finding also implies that the b subunits of the F(0) have to extend all the way from the membrane surface to the very top of the F(1) domain to make contact with the delta subunit.

A cryoelectron microscopy study of the interaction of the Escherichia coli F1-ATPase with subunit b dimer.

A complex between the Escherichia coli F1-ATPase and a truncated form of the ECF0-b subunit was formed and examined by cryoelectron microscopy in amorphous ice. Image analysis of single particles in the hexagonal projection revealed that the polar domain of the b subunit interacts with a beta subunit different from the one which interacts with the epsilon subunit. The cavity in the enzyme, visible in the hexagonal projection, is not filled by the b polypeptide, therefore leaving enough room for extensive conformational changes of the gamma and epsilon subunits within the native F1F0 complex.

Using information theory to search for co-evolving residues in proteins.

MOTIVATION: Some functionally important protein residues are easily detected since they correspond to conserved columns in a multiple sequence alignment (MSA). However important residues may also mutate, with compensatory mutations occurring elsewhere in the protein, which serve to preserve or restore functionality. It is difficult to distinguish these co-evolving sites from other non-conserved sites. RESULTS: We used Mutual Information (MI) to identify co-evolving positions. Using in silico evolved MSAs, we examined the effects of the number of sequences, the size of amino acid alphabet and the mutation rate on two sources of background MI: finite sample size effects and phylogenetic influence. We then assessed the performance of various normalizations of MI in enhancing detection of co-evolving positions and found that normalization by the pair entropy was optimal. Real protein alignments were analyzed and co-evolving isolated pairs were often found to be in contact with each other. AVAILABILITY: All data and program files can be found at http://www.biochem.uwo.ca/cgi-bin/CDD/index.cgi

Removal of the epsilon subunit from Escherichia coli F1-ATPase using monoclonal anti-epsilon antibody affinity chromatography.

The usefulness of two monoclonal antibodies, epsilon-1 and epsilon-4, which recognize the epsilon subunit of Escherichia coli F1-ATPase, for removing that subunit from ATPase was assessed. The epsilon subunit is a tightly bound, but dissociable, inhibitor of the ATPase. epsilon-1 binds epsilon with 10-fold higher affinity than epsilon-4. epsilon-1 recognizes a site on epsilon which is hidden by the quaternary structure of ATPase, while epsilon-4 can recognize epsilon when it is part of ATPase. Each antibody was purified and coupled to Sepharose to generate affinity columns. Solutions of ATPase in a buffer which was designed to reduce the affinity of epsilon for the enzyme were pumped through the columns and the degree of epsilon depletion was determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and by Western blotting. Neither column retained ATPase significantly. At low ATPase concentrations and low flow rates, the epsilon-1 column was more efficient than the epsilon-4 column, removing in excess of 95% of the epsilon in a single passage compared with 93% removal by the epsilon-4 column. At higher protein concentrations or flow rates, however, the performance of the epsilon-1 column was substantially poorer, while that of the epsilon-4 column was much less affected. Very little epsilon emerged from the epsilon-4 column before most of the measured epsilon-binding capacity was filled. A second passage through the epsilon-4 column reduced residual epsilon to less than 2% of that which was originally present. Pure, active epsilon was eluted from either column by 1 M NH4OH, pH 11.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of the modification of transfer buffer composition and the renaturation of proteins in gels on the recognition of proteins on Western blots by monoclonal antibodies.

Two modifications to Western blots which enhance immunochemical recognition have been developed. The first is transfer in carbonate buffer at pH 9.9, rather than the more commonly used Tris-glycine buffer at pH 8.3. This alteration improved the recognition of four of the five subunits of Escherichia coli F1-ATPase by monoclonal antibodies, the smaller subunits showing the greatest effects. Recognition of dinitrophenyl groups attached to the subunits by polyclonal antibodies was improved by the carbonate buffer only for the smallest ATPase subunit, epsilon. The second modification was incubation of the gel in mild buffers, designed to promote the renaturation of proteins, before the electrophoretic transfer step. The most effective buffer was 20% glycerol in 50 mM Tris-HCl, pH 7.4. Improvements in the signal obtained with monoclonal antibodies to all the subunits of ATPase were obtained by this procedure. As the subunits vary markedly in size, isoelectric point, and other properties, this method should be useful for most proteins. The fate of the 15,000-Da epsilon subunit, labeled with 125I, was followed through a blotting experiment. As long as no sodium dodecyl sulfate was added to the transfer buffer, epsilon was bound to nitrocellulose efficiently in either Tris-glycine or carbonate buffer. However, the epsilon was retained much more strongly during the subsequent incubation steps if the transfer was done in the carbonate buffer. The binding of epsilon to the nitrocellulose was even more stable when the gel had been treated with the buffered glycerol solution before transfer. These results indicate that the conditions under which epsilon subunit first encounters the nitrocellulose markedly affect the stability of binding during subsequent steps. The F1-ATPase was partially fragmented by treatment with proteases and then run on a gel and either transferred immediately in Tris-glycine buffer or else treated with the buffered glycerol solution and transferred in the carbonate buffer. The second blot gave stronger recognition of residual alpha subunit and fragments by an anti-alpha monoclonal antibody, with the largest improvement for the smaller fragments. This result suggests that the modified procedure may be particularly useful in enhancing the detection of small proteins.

The polar domain of the b subunit of Escherichia coli F1F0-ATPase forms an elongated dimer that interacts with the F1 sector.

A soluble form of the b subunit of the F0 sector of the F1F0-ATPase of Escherichia coli has been produced, purified, and characterized. In this form of the protein, designated bsol, residues 25-146 (the carboxyl terminus) of b have been fused to an amino-terminal octapeptide extension derived from the vector pUC8. The inferred subunit molecular weight of bsol is 15,459. bsol protein was expressed in E. coli as a soluble cytoplasmic protein and was readily purified to homogeneity by conventional methods. The molecular weight of bsol, determined by sedimentation equilibrium, was 31,200, indicating that the protein is dimeric. Chemical cross-linking studies supported this conclusion. However, bsol sedimented with a coefficient of just 1.8 S and behaved on size exclusion chromatography with an apparent molecular weight of 80,000-85,000. These results indicate that the protein exists in solution as a highly elongated dimer. The circular dichroism spectrum indicated that bsol is highly alpha-helical. Binding of bsol to F1-ATPase was directly demonstrated by size exclusion chromatography. bsol also inhibited the binding of F1-ATPase to F1-depleted membrane vesicles, as measured by reconstitution of energy-dependent quinacrine fluorescence quenching. This result implies that bsol and F0 compete for binding to the same site on F1. The apparently normal interaction of bsol with F1-ATPase strongly suggests that the recombinant protein assumes the correct structure. No substantial effects of bsol on the ATPase activity of purified F1 were observed.

ATP causes a large change in the conformation of the isolated alpha subunit of Escherichia coli F1 ATPase.

The physiochemical properties of the isolated alpha subunit of the Escherichia coli coupling factor ATPase, and changes resulting from the interaction of alpha with ATP, were studied. Amino acid analysis of alpha revealed 42% polar residues, 4 cysteine residues, and a single tryptophan residue. The partial specific volume, v, of alpha was 0.74 cm3 g-1. Molecular weight value of alpha, determined by sedimentation equilibrium, of 55,000 to 59,000 were observed in guanidine hydrochloride, or in nondenaturing buffer in either the presence of absence of ATP, which alpha binds with high affinity. Sedimentation velocity experiments gave a value of s20,w0-3.52 S for alpha. In the presence of ATP, this value increased to 4.00 S, indicating a large conformational change of alpha when ATP is bound. A slow dissociation rate of alpha x ATP was suggested by the finding that a substantial portion of [2-3H]ATP mixed with alpha remained bound to the protein during native polyacrylamide gel electrophoresis, causing alpha to migrate with a higher relative mobility. A dissociation rate constant, koff, of 0.21 min-1 at 22 degrees C was measured by following the rate at which unlabeled ATP displaced [2-3H]ATP from the protein. The properties of the interaction of alpha with ATP suggest that this subunit may be the site of the "tightly bound" nucleotides of the coupling factor ATPase.

An upstream uncD sequence modulates translation of Escherichia coli uncC.

The effect of upstream uncD sequences on expression of the Escherichia coli uncC gene, encoding the epsilon subunit of F1-ATPase, was studied. uncC expression was reduced severalfold in plasmid constructs bearing, in addition to uncC, a region of uncD located between 85 and 119 bases upstream from the uncC initiation codon. This reduction was independent of in-frame translation of the uncD sequences. An mRNA stem-loop structure in which sequences located within the inhibitory region of uncD base pair with the uncDC intercistronic region is suggested to function in modulating uncC expression.

Column centrifugation generates an intersubunit disulfide bridge in Escherichia coli F1-ATPase.

Passage of F1-ATPase through a centrifuge column [Penefsky, H. S. (1979) Methods Enzymol. 56, 527-530] caused formation of a product with a relative molecular mass of 72,000 as determined by sodium dodecyl sulfate/polyacrylamide gel electrophoresis. The product was identified as cross-linked alpha and delta subunits by using Western blots and subunit-specific monoclonal antibodies. The cross-link was reversed by 50 mM dithiothreitol implying that it was a disulfide bridge. Formation of the cross-link was inhibited by 2 mM EDTA and was stimulated in some buffers by the addition of 10 microM CuCl2. Time course experiments indicated that the majority of the cross-link formed while the enzyme was passing through the column. Thus the cross-link induced by column centrifugation arose from the rapid, heavy-metal-ion-catalysed oxidation of two sulfhydryl groups, one on the alpha subunit and one on the delta subunit, to a disulfide. These results demonstrate that care must be exercised when running proteins through centrifuge columns as potentially deleterious disulfide formation can result. An anti-beta monoclonal antibody was capable of immunoprecipitating the entire enzyme including the cross-linked subunits, implying that the cross-linked alpha and delta subunits were still a part of F1. The formation of the cross-link affected neither the hydrolytic activity of the enzyme nor its susceptibility to inhibition by epsilon subunit. The cross-linked enzyme was unable to bind to F1-depleted membranes in experiments in which soluble F1 and membranes were separated by centrifugation. Column centrifugation did not generate the cross-link on membrane-bound enzyme. These results indicate that the alpha-delta cross-link results in a loss of binding affinity between F1 and F0.

Location of conserved residue histidine-38 of the epsilon subunit of Escherichia coli ATP synthase.

The function and location of residue His-38 of the epsilon subunit of the Escherichia coli F1-ATPase were investigated. His-38 was replaced by glutamine and cysteine through site-directed mutagenesis to produce epsilon H38Q and epsilon H38C, respectively. Both epsilon H38Q and epsilon H38C fulfilled epsilon function in vivo as determined by growth on nonfermentable carbon sources, growth yield on limiting glucose, and recovery of cells from energy starvation conditions. epsilon H38Q and epsilon H38C were purified and studied in vitro. Pure epsilon H38C reacted rapidly with Ellman's reagent, indicating a surface location of the introduced cysteine. epsilon H38C which had been reconstituted with epsilon-depleted F1-ATPase could be linked specifically to the gamma subunit using two different heterobifunctional sulfhydril-reactive/photoreactive crosslinking agents, indicating that residue 38 lies near gamma. The mutated epsilon subunits were unaltered in their ability to inhibit epsilon-depleted F1-ATPase in vitro, even after modification of epsilon H38C with the bulky reagents fluorescein maleimide and N-(1-anilinonaphthyl-4)maleimide. It seems unlikely, therefore, that residue His-38 of epsilon interacts directly with gamma. Both the epsilon H38Q and epsilon H38C mutations reduced the recognition of epsilon by monoclonal antibody epsilon-1, but recognition of epsilon H38C was not further reduced by reaction with fluorescein maleimide. These results imply that residue 38 is not directly part of the epsilon-1 epitope, but plays a role in its formation.

Orientation of the epsilon subunit in Escherichia coli ATP synthase.

Deletion mutations in the NH2- and COOH-terminal regions of the epsilon subunit of Escherichia coli ATP synthase were constructed making use of the AatII and HincII restriction enzyme sites. The resultant mutated epsilon species were analyzed for in vivo functionality and for recognition by anti-epsilon monoclonal antibodies. Deletion of residues Asp-7 through Gln-14 (epsilon delta D7-Q14) resulted in reduced ability to complement uncC mutants as determined by growth yields on limiting glucose medium and by formation of small colonies on plates with succinate as the source of carbon and energy. None of the other mutants was notably impaired. Upon induction to obtain overexpression, the NH2-terminal deletion mutants were expressed at levels comparable to the wild-type epsilon subunit, but the COOH-terminal deletion mutants were expressed less strongly, suggesting that residues in the latter region are important for protein stability. Monoclonal antibody epsilon-1, which cannot bind to epsilon when it is part of F1-ATPase, recognized the COOH-terminal deletions well, but the NH2-terminal deletions poorly. Additional epitope mapping using epsilon fusion proteins revealed that residues required for the epsilon-1 epitope extend to between Thr-77 and Arg-85. Monoclonal antibody epsilon-4, which can bind to epsilon when it is part of F1-ATPase, recognized the NH2-terminal deletions well, but hardly recognized the COOH-terminal deletions, indicating a role of residues located COOH-terminal to Ile-131 in recognition by this antibody. Epitope mapping using the fusion proteins revealed that the residues required by epsilon-4 begin in the region between Val-78 and Met-95. These results imply a two-domain structure of epsilon and orient the subunit within the enzyme.

Activation and inhibition of the Escherichia coli F1-ATPase by monoclonal antibodies which recognize the epsilon subunit.

The properties of two monoclonal antibodies which recognize the epsilon subunit of Escherichia coli F1-ATPase were studied in detail. The epsilon subunit is a tightly bound but dissociable inhibitor of the ATPase activity of soluble F1-ATPase. Antibody epsilon-1 binds free epsilon with a dissociation constant of 2.4 nM but cannot bind epsilon when it is associated with F1-ATPase. Likewise epsilon cannot associate with F1-ATPase in the presence of high concentrations of epsilon-1. Thus epsilon-1 activates F1-ATPase which contains the epsilon subunit, and prevents added epsilon from inhibiting the enzyme. Epsilon-1 cannot bind to membrane-bound F1-ATPase. The epsilon-4 antibody binds free epsilon with a dissociation constant of 26 nM. Epsilon-4 can bind to the F1-ATPase complex, but, like epsilon-1, it reverses the inhibition of F1-ATPase by the epsilon subunit. The epsilon subunit remains crosslinkable to both the beta and gamma subunits in the presence of epsilon-4, indicating that it is not grossly displaced from its normal position by the antibody. Presumably the activation arises from more subtle conformational effects. Antibodies epsilon-4 and delta-2, which recognizes the delta subunit, both bind to F1F0 in E. coli membrane vesicles, indicating that these subunits are substantially exposed in the membrane-bound complex. Epsilon-4 inhibits the ATPase activity of the membrane-bound enzyme by about 50%, and Fab prepared from epsilon-4 inhibits by about 40%. This inhibition is not associated with any substantial change in the major apparent Km for ATP. These results suggest that inhibition of membrane-bound F1-ATPase arises from steric effects of the antibody.

RNase E-dependent cleavages in the 5' and 3' regions of the Escherichia coli unc mRNA.

The endonucleolytic processing of the unc mRNA encoding the eight subunits of the Escherichia coli F1F0-ATPase was studied. Northern (RNA) blots of mRNA expressed from a plasmid which contained the 3'-terminal portion of the operon including the uncDC sequences revealed, in addition to the expected 2-kb mRNA, a 0.5-kb RNA species which hybridized to an uncC antisense RNA probe. An uncD antisense RNA probe hybridized to only the 2-kb mRNA, implying that the upstream 1.5-kb fragment is rapidly degraded. The 5' end of the 0.5-kb fragment was determined by primer extension analysis to be 11 bases into the coding region of the uncC gene. In RNase E-deficient strains, the amount of the 0.5-kb product was strongly reduced while the levels of the precursor uncDC transcript remained high. Similar RNase E-dependent processing was found in the chromosomally encoded unc mRNA. As this RNase E-dependent cleavage directly inactivates uncC and appears to leave uncD susceptible to degradation, it seems unlikely to play a role in differential expression of the gene products but may be an important event in unc mRNA degradation. RNase E mutants also showed altered processing of the chromosomally encoded unc mRNA in the uncB region near the 5' end. The expected full-length (7-kb) transcript was recognized when RNA from the RNase E-deficient strain was subjected to Northern blot analysis with uncB- and uncC-specific probes. RNA from strains with functional RNase E lacked the 7-kb transcript but had a 6.2-kb mRNA detectable with the uncC but not the uncB probe. RNase E is therefore implicated in multiple cleavages of the unc mRNA.

The epsilon subunit and inhibitory monoclonal antibodies interact with the carboxyl-terminal region of the beta subunit of Escherichia coli F1-ATPase.

The epitopes of two classes of monoclonal antibody and the binding site for the epsilon subunit have been mapped to the carboxyl-terminal region of the beta subunit of Escherichia coli F1-ATPase using partial CNBr cleavage, weak acid hydrolysis, and Western blots. One class of antibody, B-I, inhibits ATPase activity; the other class, B-II, recognizes an epitope not exposed on the surface of intact F1. Data from two-dimensional gels and blots of beta cleaved with CNBr/weak acid showed that the B-I epitope lies between Asp-381 and the carboxyl-terminal Leu-459, and the B-II epitope lies between Asp-345 and Met-380. Weak acid hydrolysis of the beta-epsilon product obtained by cross-linking F1 with a water-soluble carbodiimide yielded a fragment containing epsilon and a 13-kDa carboxyl-terminal fragment of beta indicating that epsilon interacts with this portion of beta as well. Fab fragments from the B-I antibody beta-6 could be cross-linked to the epsilon subunit in native F1 by various cross-linking agents demonstrating that the antibody and the epsilon subunit occupy adjacent, nonoverlapping sites on the beta subunit. Implications of these results for the roles of the epsilon subunit and of the carboxyl-terminal region of the beta subunit in F1 are discussed.