The KdpC subunit of the Escherichia coli K+-transporting KdpB P-type ATPase acts as a catalytic chaperone

In Bacteria and Archaea, high-affinity potassium uptake is mediated by the ATP-driven KdpFABC complex. On the basis of the biochemical properties of the ATP-hydrolyzing subunit KdpB, the transport complex is classified as type IA P-type ATPase. However, the KdpA subunit, which promotes K(+) transport, clearly resembles a potassium channel, such that the KdpFABC complex represents a chimera of ion pumps and ion channels. In the present study, we demonstrate that the blending of these two groups of transporters in KdpFABC also entails a nucleotide-binding mechanism in which the KdpC subunit acts as a catalytic chaperone. This mechanism is found neither in P-type ATPases nor in ion channels, although parallels are found in ABC transporters. In the latter, the ATP nucleotide is coordinated by the LSGGQ signature motif via double hydrogen bonds at a conserved glutamine residue, which is also present in KdpC. High-affinity nucleotide binding to the KdpFABC complex was dependent on the presence of this conserved glutamine residue in KdpC. In addition, both ATP binding to KdpC and ATP hydrolysis activity of KdpFABC were sensitive to the accessibility, presence or absence of the hydroxyl groups at the ribose moiety of the nucleotide. Furthermore, the KdpC subunit was shown to interact with the nucleotide-binding loop of KdpB in an ATP-dependent manner around the ATP-binding pocket, thereby increasing the ATP-binding affinity by the formation of a transient KdpB/KdpC/ATP ternary complex.     

Photoaffinity cross-linking of the coupling factor 1 from Micrococcus luteus by 3'-arylazido-8-azido-ATP

The vicinity of nucleotide binding sites and the mechanism of ATP synthesis/hydrolysis have been studied with the bifunctional photosensitive ATP analog 3'-arylazido-8-azido-ATP. 3'-Arylazido-8-azido-ATP is hydrolyzed by the F1-ATPase from Micrococcus luteus in the absence of ultraviolet light. Irradiation, by ultraviolet light, of F1-ATPase in the presence of 3'-arylazido-8-azido-ATP results in the specific formation of cross-links between alpha and beta subunits. The results suggest that a hydrolytic nucleotide binding site is located on a beta subunit at or near an alpha subunit, probably at the interface between these subunits. Such a constellation would permit direct subunit-subunit interactions during ATP synthesis/hydrolysis.     

Solution structure of the KdpFABC P-type ATPase from Escherichia coli by electron microscopic single particle analysis

The K⁺-translocating KdpFABC complex from Escherichia coli functions as a high affinity potassium uptake system and belongs to the superfamily of P-type ATPases, although it exhibits some unique features. It comprises four subunits, and the sites of ATP hydrolysis and substrate transport are located on two different polypeptides. No structural data are so far available for elucidating the correspondingly unique mechanism of coupling ion transport and catalysis in this P-type ATPase. By use of electron microscopy and single particle analysis of negatively stained, solubilized KdpFABC complexes, we solved the structure of the complex at a resolution of 19 Å, which allowed us to model the arrangement of subunits within the holoenzyme and, thus, to identify the interfaces between subunits. The model showed that the K⁺-translocating KdpA subunit is in close contact with the transmembrane region of the ATP-hydrolyzing subunit KdpB. The cytosolic C-terminal domain of the KdpC subunit, which is assumed to play a role in cooperative ATP binding together with KdpB, is located in close vicinity to the nucleotide binding domain of KdpB. Overall, the arrangement of subunits agrees with biochemical data and the predictions on subunit interactions.     

The KdpFABC complex from Escherichia coli: a chimeric K+ transporter merging ion pumps with ion channels

The KdpFABC complex represents a multi-subunit ATP-driven potassium pump, which is only found in bacteria and archaea. Based on the properties of the ATP-hydrolyzing subunit (KdpB) the transporter has been classified as a type IA P-type ATPase. However, structural and functional properties of the remaining subunits clearly show homologies to members of the potassium channel as well as the ABC transporter family, thus rendering the KdpFABC complex to represent an inimitable chimera of ion pumps and ion channels. Accordingly, this striking juxtaposition entails special features of KdpFABC with respect to typical members of each of the transporter families, involving not only the concepts but also the structures of ion channels and ion pumps. For example, the sites of ATP hydrolysis and substrate transport are spatially separated on two different polypeptides, which, in turn, leads to a unique coupling mechanism. During catalysis, the KdpFABC complex cycles between two main conformational states, each of which comprises different structural properties together with different binding affinities for both ATP and the transport substrate. These structural configurations have recently been directly visualized in the working enzyme. Translocation of potassium is mediated by the KdpA subunit, which comprises structural as well as functional homologies to potassium channels of the MPM-type. The KdpC subunit participates in the binding of ATP, thus acting as a catalytic chaperone, which increases the ATP binding affinity of the KdpB subunit via a mechanism typical of nucleotide binding in ABC transporters.     

The K+-translocating KdpFABC complex from Escherichia coli: A P-type ATPase with unique features

The prokaryotic KdpFABC complex from the enterobacterium Escherichia coli represents a unique type of P-type ATPase composed of four different subunits, in which a catalytically active P-type ATPase has evolutionary recruited a potassium channel module in order to facilitate ATP-driven potassium transport into the bacterial cell against steep concentration gradients. This unusual composition entails special features with respect to other P-type ATPases, for example the spatial separation of the sites of ATP hydrolysis and substrate transport on two different polypeptides within this multisubunit enzyme complex, which, in turn, leads to an interesting coupling mechanism. As all other P-type ATPases, also the KdpFABC complex cycles between the so-called E1 and E2 states during catalysis, each of which comprises different structural properties together with different binding affinities for both ATP and the transport substrate. Distinct configurations of this transport cycle have recently been visualized in the working enzyme. All typical features of P-type ATPases are attributed to the KdpB subunit, which also comprises strong structural homologies to other P-type ATPase family members. However, the translocation of the transport substrate, potassium, is mediated by the KdpA subunit, which comprises structural as well as functional homologies to MPM-type potassium channels like KcsA from Streptomyces lividans. Subunit KdpC has long been thought to exhibit an FXYD protein-like function in the regulation of KdpFABC activity. However, our latest results are in favor of the notion that KdpC might act as a catalytical chaperone, which cooperatively interacts with the nucleotide to be hydrolyzed and, thus, increases the rather untypical weak nucleotide binding affinity of the KdpB nucleotide binding domain.     

Direct photoaffinity labeling of Kir6.2 by [gamma-(32)P]ATP-[gamma]4-azidoanilide

ATP-sensitive potassium (K(ATP)) channels are under complex regulation by intracellular ATP and ADP. The potentiatory effect of MgADP is conferred by the sulfonylurea receptor subunit of the channel, SUR, whereas the inhibitory effect of ATP appears to be mediated via the pore-forming subunit, Kir6.2. We have previously reported that Kir6.2 can be directly labeled by 8-azido-[gamma-(32)P]ATP. However, the binding affinity of 8-azido-ATP to Kir6.2 was low probably due to modification at 8' position of adenine. Here we demonstrate that Kir6.2 can be directly photoaffinity labeled with higher affinity by [gamma-(32)P]ATP-[gamma]4-azidoanilide ([gamma-(32)P]ATP-AA), containing an unmodified adenine ring. Photoaffinity labeling of Kir6.2 by [gamma-(32)P]ATP-AA is not affected by the presence of Mg(2+), consistent with Mg(2+)-independent ATP inhibition of K(ATP) channels. Interestingly, SUR1, which can be strongly and specifically photoaffinity labeled by 8-azido-ATP, was not photoaffinity labeled by ATP-AA. These results identify key differences in the structure of the nucleotide binding sites on SUR1 and Kir6.2.     

Photoaffinity labeling of the recBCD enzyme of Escherichia coli with 8-azidoadenosine 5'-triphosphate

The recB and recD subunits of the recBCD enzyme (exonuclease V) from Escherichia coli were covalently photolabeled with the ATP photoaffinity analogue [alpha-32P]8-azido-ATP. The labeling was specific for ATP binding sites by the following criteria. Saturation occurs at high 8-azido-ATP concentrations with dissociation constants of 30 and 120 microM for the recD and recB subunits, respectively; ATP strongly inhibits the photolabeling; 8-azido-ATP is hydrolyzed by the recBCD enzyme and supports its double-stranded DNA exonuclease activity; and the label is largely confined to two peptides obtained by tryptic digestion of the photolabeled holoenzyme; one is derived from the recB subunit and the other from the recD subunit.     

Identification of nucleotide binding sites in V-type Na+-ATPase from Enterococcus hirae

A and B subunits of the V-type Na+-ATPase from Enterococcus hirae were suggested to possess nucleotide binding sites (Murata, T. et al., J. Biochem., 132, 789-794 (2002)), although the B subunit did not have the consensus sequence for nucleotide binding. To further characterize the binding sites in the V-ATPase, we did the photoaffinity labeling study using 8-azido-[alpha-32P]ATP. A and B subunits were labeled with 8-azido-[alpha-32P]ATP when analysed with SDS polyacrylamide gel electrophoresis. The peptide fragment of A subunit obtained by lysyl endopeptidase digestion after labeling showed a molecular size of 9 kDa and its amino acid sequencing revealed that it corresponded to residues Arg423-Lys494. The peptide fragment from B subunit after photoaffinity labeling and lysyl endopeptidase digestion showed the size of 5 kDa and corresponded to residues Phe404-Lys443. In our structure model, these peptides were close to the adenine ring of ATP. We suggest that non-catalytic B subunit of E. hirae V-ATPase has a nucleotide binding site, similarly to eukaryotic V-ATPases and F-ATPases.     

Photoaffinity labeling of the nucleotide-binding site of the uncoupling protein from hamster brown adipose tissue.

The nucleotide binding center of the uncoupling protein from brown adipose tissue (UCP) was probed by photoaffinity labeling with 8-azido-ATP. The isolated dimeric UCP in non-ionic detergent was used. 8-azido-ATP binds to UCP with a Kd = 3 microM, i.e. with an only threefold lower affinity than ATP and a maximum number of binding sites of about 12 mumol/g protein corresponding to about 1 mol/mol dimer UCP. UCP is rapidly degraded by ultraviolet radiation, and therefore only near ultraviolet and visible light can be used for photoaffinity labeling. The total covalent incorporation is shown to be dependent on the concentration of azido-ATP and on competing phospholipids. The specific, i.e. ATP-sensitive incorporation only to the binding site depends on the presence of cysteine. With CNBr cleavage the 8-azido-[gamma-32P]ATP insertion within the primary structure was located by identifying ATP-sensitive labeled peptides in SDS/PAGE. A major specific 8-azido-ATP incorporation was found by autoradiography in the smallest CNBr fragments. Identification of the radioactive peptides was difficult since 8-azido-ATP insertion causes a distinct shift in the gels from the stained peptides. Identification was possible by specific disulfide formation at the C-terminal within the UCP dimer which only removed the CB7 (CB, CNBr fragment) portion of the low-molecular-mass peptides but did not move the radioactive band. This excludes the C-terminal CB7 and identifies the labeled peptide as CB6. Also, limited tryptic cleavage of intact UCP at Lys293 did not remove the radioactivity. Cleavage of tryptophanes support localization of 8-azido-ATP between residues 173-280 which includes CB6. Solid-phase sequencing of the labeled CB6 both after serine lactone and carboxyl coupling suggest incorporation into Thr260. These results indicate that the adenine-binding site is within the third domain of the tripartite UCP structure at a putative hydrophilic channel which can be assessed both from the cytosol and matrix of mitochondria.     

Analysis of KdpC of the K(+)-transporting KdpFABC complex of Escherichia coli.

The Kdp complex, a high affinity ATP-driven K(+) transport system of Escherichia coli, is composed of the four membrane-bound subunits KdpF, KdpA, KdpB and KdpC. Whereas the role of KdpB (catalytical subunit), KdpA (K(+)-translocating subunit) and KdpF (stabilizing peptide) is well understood, the function of KdpC is still unknown. Therefore, a kdpC deletion strain was constructed and complementation experiments were performed using different kdpC constructs. Truncations of the kdpC gene revealed that only one derivative, which lacks base pairs coding for the four C-terminal amino acids, was able to complement the chromosomal deletion of kdpC. Furthermore, complementation was also observed with kdpC of Mycobacterium tuberculosis, but not with kdpC from Clostridium acetobutylicum or Synechocystis sp. PCC6803. Sequence alignment of 17 different KdpC proteins led to the construction of hybrids between kdpC of E. coli and that of C. acetobutylicum. Complementation revealed that the N-terminal transmembrane segment as well as the C-terminal-third of the protein can be exchanged between both species, but only one after the other. A simultaneous substitution of both regions was not possible.     

Labeling of two different regions of the nucleotide binding site of the uncoupling protein from brown adipose tissue mitochondria with two ATP analogs.

The nucleotide binding site of the uncoupling protein (UCP) from brown adipose tissue was mapped by photoaffinity labeling with 2-azidoadenosine 5'-triphosphate (2-azido-ATP) and by affinity labeling with 3'-O-(5-fluoro-2,4-dinitrophenyl)adenosine 5'-triphosphate (FDNP-ATP). Both analogs bind with high affinity and specificity to the UCP in intact mitochondria, as well as to the isolated solubilized protein. Reversible binding at 4 degrees C in the dark is competitively blocked by GTP. Like the natural ligands ATP and GTP, both analogs are capable of inhibiting the H+/OH- conductance of the UCP as measured in proteoliposomes with reconstituted UCP. 2-azido-ATP was incorporated into UCP in mitochondria in the presence of carboxyatractylate, while FDNP-ATP was inserted into isolated UCP by prolonged incubation at room temperature under pH variation. Both reactions can be blocked by GTP. The incorporation of 2-azido-ATP could be localized between residues 258 and 283 by cleavage with CNBr. Solid-phase sequencing of the homoserine-linked radioactive peptide indicated that the 2-azido-ATP was linked to threonine-263. The incorporation of FDNP-ATP could be assigned by cleavage with CNBr and alternatively with trypsin at a locus of covalent attachment between residues 238 and 255. On the basis of published data that no tyrosine participates in nucleotide binding of the UCP, the probable residue reacting with FDNP-ATP is cysteine-253.(ABSTRACT TRUNCATED AT 250 WORDS)     

Reaction of 2-azido-ATP with beta subunits in the F1-adenosine triphosphatase of Escherichia coli

The three beta subunits of the Escherichia coli F1-ATPase react independently with chemical reagents (Stan Lotter, H. and Bragg, P.D. (1986) Arch. Biochem. Biophys. 248, 116-120). Thus, one beta subunit is readily cross-linked to the epsilon subunit, another reacts with N,N'-dicyclohexylcarbodiimide (DCCD), and the third one is modified by 4-chloro-7-nitrobenzofurazan (NbfCl). The relationship of the binding site for 2-azido-ATP to the three types of beta subunit recognized by chemical labeling was examined. The binding site for 2-azido-ATP was not associated with a specific type of beta-subunit. There was no relationship between the site of nucleotide and the association of the epsilon subunit with a particular beta subunit. It is concluded that the presence of the epsilon subunit (possibly in association with the other minor subunits) does not determine the position of the catalytic site. The possibility that the lack of a specific relationship between the 2-azido-ATP binding site and a specific beta subunit was due to turnover of the enzyme, making each beta a catalytic site in turn, could not be entirely rejected. However, the rate of hydrolysis of 2-azido-ATP by the DCCD-modified ATPase was very low in the presence of EDTA, and was likely due to catalysis at single sites.     

Both ATP sites of human P-glycoprotein are essential but not symmetric.

Human P-glycoprotein (P-gp) is a cell surface drug efflux pump that contains two nucleotide binding domains (NBDs). Mutations were made in each of the Walker B consensus motifs of the NBDs at positions D555N and D1200N, thought to be involved in Mg(2+) binding. Although the mutant and wild-type P-gps were expressed equivalently at the cell surface and bound the drug analogue [(125)I]iodoarylazidoprazosin ([(125)I]IAAP) comparably, neither of the mutant proteins was able to transport fluorescent substrates nor had detectable basal nor drug-stimulated ATPase activities. The wild-type and D1200N P-gps were labeled comparably with [alpha-(32)P]-8-azido-ATP at a subsaturating concentration of 2.5 microM, whereas labeling of the D555N mutant was severely impaired. Mild trypsin digestion, to cleave the protein into two halves, demonstrated that the N-half of the wild-type and D1200N proteins was labeled preferentially with [alpha-(32)P]-8-azido-ATP. [alpha-(32)P]-8-Azido-ATP labeling at 4 degrees C was inhibited in a concentration-dependent manner by ATP with half-maximal inhibition at approximately 10-20 microM for the P-gp-D1200N mutant and wild-type P-gp. A chimeric protein containing two N-half NBDs was found to be functional for transport and was also asymmetric with respect to [alpha-(32)P]-8-azido-ATP labeling, suggesting that the context of the ATP site rather than its exact sequence is an important determinant for ATP binding. By use of [alpha-(32)P]-8-azido-ATP and vanadate trapping, it was determined that the C-half of wild-type P-gp was labeled preferentially under hydrolysis conditions; however, the N-half was still capable of being labeled with [alpha-(32)P]-8-azido-ATP. Neither mutant was labeled under vanadate trapping conditions, indicating loss of ATP hydrolysis activity in the mutants. In confirmation of the lack of ATP hydrolysis, no inhibition of [(125)I]IAAP labeling was observed in the mutants in the presence of vanadate. Taken together, these data suggest that the two NBDs are asymmetric and intimately linked and that a conformational change in the protein may occur upon ATP hydrolysis. Furthermore, these data are consistent with a model in which binding of ATP to one site affects ATP hydrolysis at the second site.     

Chloroplast F1 ATPase has more than three nucleotide binding sites, and 2-azido-ADP or 2-azido-ATP at both catalytic and noncatalytic sites labels the beta subunit

The photolabeling of chloroplast F1 ATPase, following exposure to Mg2+ and 2-azido-ATP and separation from medium nucleotides, results in derivatization of two separate peptide regions of the beta subunit. Up to 3 mol of the analogue can be incorporated per mole of CF1, with covalent binding of one moiety or two moieties per beta subunit that can be either AMP, ADP, or ATP derivatives. These results, the demonstration of noncovalent tight binding of at least four [3H]adenine nucleotides to the enzyme and the presence of three beta subunits per enzyme, point to six potential adenine nucleotide binding sites per molecule. The tightly bound 2-azido nucleotides on CF1, found after exposure of the heat-activated and EDTA-treated enzyme to Mg2+ and 2-azido-ATP, differ in their ease of replacement during subsequent hydrolysis of ATP. Some of the bound nucleotides are not readily replaced during catalytic turnover and covalently label one peptide region of the beta subunit. They are on noncatalytic sites. Other tightly bound nucleotides are readily replaced during catalytic turnover and label another peptide region of the beta subunit. They are at catalytic sites. No alpha-subunit labeling is detected upon photolysis of the bound 2-azido nucleotides. However, one or both of the sites could be at an alpha-beta-subunit interface with the 2-azido region close to the beta subunit, or both binding sites may be largely or entirely on the beta subunit.     

Recombinant synthesis, purification, and nucleotide binding characteristics of the first nucleotide binding domain of the cystic fibrosis gene product.

The majority of mutations which lead to clinical cystic fibrosis are located within the two predicted nucleotide binding domains of the cystic fibrosis gene product. We have used a prokaryotic expression system to synthesize and purify the first nucleotide binding domain (NBD-1, amino acids 426-588) with and without the most common mutation associated with the disease (the deletion of phenylalanine at position 508, delta F508). Both wild type and delta F508 NBD-1 bind ATP-agarose in a quantitatively comparable manner; this binding was inhibited by excess Na2ATP, trinitrophenol-ATP, or 8-azido-ATP. Irreversible NBD-1 labeling by an ATP analog was demonstrated using [32P]8-azido-ATP. This covalent labeling was inhibited by preincubation with Na2ATP, with half-maximal inhibition for Na2ATP occurring at approximately 5 mM for both the wild type and delta F508 nucleotide binding domain. These experiments are among the first to confirm the expectation that the cystic fibrosis transmembrane conductance regulator NBD-1 binds nucleotide. Since, under the conditions used in our study, NBD-1 without phenylalanine 508 displays very similar nucleotide binding characteristics to the wild type protein, our results support previous structural models which predict that the delta F508 mutation should not cause an alteration in ATP binding.     

Mapping of ATP binding regions in poly(A) polymerases by photoaffinity labeling and by mutational analysis identifies a domain conserved in many nucleotidyltransferases

We have identified regions in poly(A) polymerases that interact with ATP. Conditions were established for efficient cross-linking of recombinant bovine and yeast poly(A) polymerases to 8-azido-ATP. Mn2+ strongly stimulated this reaction due to a 50-fold lower Ki for 8-azido-ATP in the presence of Mn2+. Mutations of the highly conserved Asp residues 113, 115, and 167, critical for metal binding in the catalytic domain of bovine poly(A) polymerase, led to a strong reduction of cross-linking efficiency, and Mn2+ no longer stimulated the reaction. Sites of 8-azido-ATP cross-linking were mapped in different poly(A) polymerases by CNBr-cleavage and analysis of tryptic peptides by mass spectroscopy. The main cross-link in Schizosaccharomyces pombe poly(A) polymerase could be assigned to the peptide DLELSDNNLLK (amino acids 167-177). Database searches with sequences surrounding the cross-link site detected significant homologies to other nucleotidyltransferase families, suggesting a conservation of the nucleotide-binding fold among these families of enzymes. Mutations in the region of the "helical turn motif" (a domain binding the triphosphate moiety of the nucleotide) and in the suspected nucleotide-binding helix of bovine poly(A) polymerase impaired ATP binding and catalysis. The results indicate that ATP is bound in part by the helical turn motif and in part by a region that may be a structural analog to the fingers domain found in many polymerases.     

Crystal structure of the archaeal A1Ao ATP synthase subunit B from Methanosarcina mazei Gö1: Implications of nucleotide-binding differences in the major A1Ao subunits A and B

The A1Ao ATP synthase from archaea represents a class of chimeric ATPases/synthases, whose function and general structural design share characteristics both with vacuolar V1Vo ATPases and with F1Fo ATP synthases. The primary sequences of the two large polypeptides A and B, from the catalytic part, are closely related to the eukaryotic V1Vo ATPases. The chimeric nature of the A1Ao ATP synthase from the archaeon Methanosarcina mazei G?1 was investigated in terms of nucleotide interaction. Here, we demonstrate the ability of the overexpressed A and B subunits to bind ADP and ATP by photoaffinity labeling. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry was used to map the peptide of subunit B involved in nucleotide interaction. Nucleotide affinities in both subunits were determined by fluorescence correlation spectroscopy, indicating a weaker binding of nucleotide analogues to subunit B than to A. In addition, the nucleotide-free crystal structure of subunit B is presented at 1.5 A resolution, providing the first view of the so-called non-catalytic subunit of the A1Ao ATP synthase. Superposition of the A-ATP synthase non-catalytic B subunit and the F-ATP synthase non-catalytic alpha subunit provides new insights into the similarities and differences of these nucleotide-binding ATPase subunits in particular, and into nucleotide binding in general. The arrangement of subunit B within the intact A1Ao ATP synthase is presented.     

Derivatization of the catalytic subunits of the chloroplast ATPase by 2-azido-ATP and dicyclohexylcarbodiimide. Evidence for catalytically induced interchange of the subunits

Modifications of the catalytic beta subunits of the chloroplast ATPase (CF1-ATPase) are reported which support the proposal that all three subunits participate sequentially during catalysis. The beta subunits of the CF1-ATPase are sufficiently homogeneous to allow detection of their derivatization with dicyclohexylcarbodiimide (DCCD) or the substrate analog 2-azido-ATP by two-dimensional isoelectric focusing. Whether the DCCD reacts with the same beta subunit that tightly binds ATP has not been known. Our results show that when CF1-ATPase is covalently labeled with 2-azido-ATP followed by reaction with DCCD, different beta subunits are labeled. The DCCD labeling does not stop catalytic cooperativity of the CF1-ATPase and allows slow enzyme turnover. When the DCCD-modified enzyme catalyzes 2-azido-ATP cleavage and the enzyme with tightly bound nucleotide is photolyzed, both DCCD-modified and unmodified subunits are randomly labeled by the azido nucleotide. This result is as expected if during the catalytic cycle one beta subunit with unique properties is replaced by another subunit that gains these properties. The participation of all three subunits in the catalytic cycle is suggested by the apparent retention of catalytic cooperativity by the two remaining subunits after one subunit has already catalyzed 2-azido-ATP cleavage and been labeled.     

Demonstration of an Mg2+-induced conformational change by photoaffinity labelling of the high-affinity ATP-binding site of (Na+ + K+)-ATPase with 8-azido-ATP

8-Azido-ATP (8-N3ATP) is a substrate of (Na+ + K+)-ATPase from pork kidney and photoinactivates it by binding to the Mr = 100 000 alpha-subunit. The photoinactivation requires the presence of Mg2+ even though 8-azido-ATP is recognized by the high-affinity ATP binding site (Kd = 3.1 microM). K+ ions protect the enzyme against photoinactivation as does excess ATP. To see whether the Mg2+-requirement of the photoinactivation is due to the action of free Mg2+ or to the existence of an Mg X 8-azido-ATP complex, the action of the stable Mg X ATP complex analogue, chromium X 8-N3ATP (Cr X 8-N3ATP), was studied. Cr X 8-N3ATP photoinactivates (Na+ + K+)-ATPase in the absence of Mg2+, but the photoinactivation is enhanced by Mg2+, indicating that the formation of a Mg X ATP complex is an absolute requirement for photoinactivation. However, the interaction of Mg2+ with a low-affinity site also enhances the photoinactivation. It is therefore concluded that interactions with MgATP and free Mg induce conformational changes in the purine subsite of the high-affinity ATP binding site. Controlled trypsinolysis of the [alpha-32P]8-N3ATP-photolabelled enzyme in the presence of K+ results in the formation of an Mr = 56 000 radioactive peptide, whereas trypsinolysis of a [gamma-32P]Cr X ATP-labelled enzyme under identical conditions forms an Mr = 41 000 radioactive peptide. Extensive trypsinolysis of the [alpha-32P] 8-N3ATP-photolabelled alpha-subunit leads to the formation of a radioactive peptide of Mr = 1800.     

Analysis of the properties of the N-terminal nucleotide-binding domain of human P-glycoprotein

Human P-glycoprotein, the MDR1 gene product, requires both Mg(2+)-ATP binding and hydrolysis to function as a drug transporter; however, the mechanism(s) defining these events is not understood. In the present study, we explored the nature of Mg(2+)-ATP binding in the N-terminal nucleotide-binding domain of human P-glycoprotein and identified the minimal functional unit required for specific ATP binding. Recombinant proteins encompassing amino acids within the region beginning at 348 and ending at 707 were expressed in Escherichia coli, purified from inclusion bodies under denaturing conditions, and renatured by rapid dilution. The ability of ATP to interact with these proteins was examined by use of the photoactive ATP analogue [alpha-(32)P]-8-azido-ATP. Photoaffinity labeling of recombinant proteins identified the region between amino acids 375 and 635 as the region necessary to obtain specific ATP-binding properties. Specific protein labeling was saturable, enhanced by Mg(2+), and inhibited by ATP. Recombinant proteins confined within the region beginning at amino acid 392 and ending at amino acid 590 demonstrated nonspecific [alpha-(32)P]-8-azido-ATP labeling. Nonspecific labeling was not enhanced by Mg(2+) and was inhibited only by high concentrations of ATP. Using a D555N mutated protein, we found that the conserved aspartate residue in the Walker B motif plays a role in magnesium-enhanced ATP-binding. Taken together, these data define the region of the N-terminal nucleotide-binding domain of P-glycoprotein that is required for specific ATP binding and suggest that magnesium may play a role in stabilizing the ATP-binding site.