Functional Analysis of Chimeric Proteins of the Wilson Cu(I)-ATPase (ATP7B) and ZntA, a Pb(II)/Zn(II)/Cd(II)-ATPase from Escherichia coli *

ATP7B, the Wilson disease-associated Cu(I)-trans-porter, and ZntA from Escherichia coli are soft metal P1-type ATPases with mutually exclusive metal ion substrates. P1-type ATPases have a distinctive amino-ter-minal domain containing the conserved metal-binding motif G XX C XX C. ZntA has one copy of this motif while ATP7B has six copies. The effect of interchanging the amino-terminal domains of ATP7B and ZntA was inves-tigated. Chimeric proteins were constructed in which either the entire amino-terminal domain of ATP7B or only its sixth metal-binding motif replaced the amino-terminal domain of ZntA. Both chimeras conferred resistance to lead, zinc, and cadmium salts but not to copper salts. The purified chimeras displayed activity with lead, cadmium, zinc, and mercury, which are substrates of ZntA. There was no activity with copper or silver, which are substrates of ATP7B. The chimeras were 2–3-fold less active than ZntA. Thus, the amino-terminal domain of P1-type ATPases cannot alter the metal specificity determined by the transmembrane segment. Also,

ATP7A and ATP7B, respectively (5)(6)(7)(8)(9). Wilson disease is characterized by the accumulation of toxic concentrations of copper in the liver, kidney, and brain. ATP7B, the Wilson diseaseassociated protein, is a 1411 amino acid protein that is localized primarily in the trans-Golgi network where it pumps copper from the cytoplasm into the Golgi lumen (10). ZntA is one of two P1-type ATPases from Escherichia coli; it is specific for Pb(II), Cd(II), and Zn(II) and confers resistance to these metal ions in vivo (11)(12)(13).
A distinctive feature of P1-type ATPases is a highly polar amino-terminal domain of variable length. This domain contains 1-6 repeats of a conserved metal-binding motif; this motif is 70 -100 residues long and usually contains the sequence GXXCXXC. This motif also occurs in metallothioneins, copper chaperone proteins, and the periplasmic mercury-binding protein MerP, among others (14 -16). Both ATP7A and ATP7B have six repeats of the GXXCXXC sequence, whereas the smaller bacterial P1-type ATPases generally have one or two repeats. In some P1-type ATPases, the cysteine-rich amino terminus is replaced by a highly histidine-rich sequence. The isolated amino-terminal domain of ZntA is able to bind different soft metal ions. 1 The amino-terminal domain of ATP7B (WND-Cu(1-6)), 2 comprising residues 1-649, has been shown to bind six atoms of copper (17)(18). Detailed structural analysis of copper binding to this domain showed that each binding site ligates copper in a ϩ1 oxidation state using two cysteine side chains with distorted linear geometry (19). Analysis of copper-induced conformational changes in the amino-terminal domain revealed that both secondary and tertiary structural changes take place upon copper binding. The conformational changes correlate very well with the cooperativity of copper binding to this domain reported earlier (18). These observations together with other functional data led us to postulate that copper-induced conformational changes in WND-Cu (1)(2)(3)(4)(5)(6) stimulate the phosphorylation of the ATPase, thereby initiating the copper transport cycle (19). P1-type ATPases display a high degree of specificity for the 1 B. Mitra, unpublished observation. 2 The abbreviations used are: WND-Cu (1)(2)(3)(4)(5)(6), the amino-terminal domain of ATP7B with all six of its metal-binding motifs; ⌬N-ZntA, a mutant of ZntA with residues 2-106 deleted; WND-Cu(1-6)-ZntA, a chimeric protein in which the first 105 residues of ZntA were deleted and replaced by residues 1-650 of ATP7B together with an extra two residues, GT, at the junction; WND-Cu(6), the sixth metal-binding motif of ATP7B; WND-Cu(6)-ZntA, a chimeric protein in which residues 2-105 of ZntA were deleted and replaced by residues 544 -650 of ATP7B together with an extra two residues, GT, at the junction; PAGE, polyacrylamide gel electrophoresis. metal ions that are transported. For example, ATP7A and ATP7B transport the monovalent cation Cu(I) and possibly Ag(I). On the other hand, ZntA has been shown to be specific for Pb(II), Zn(II), Cd(II), and Hg(II) (20). The determinants of metal ion recognition and specificity for P1-type ATPases are not known. Because there is a high level of sequence similarity among all P1-type ATPases, the basis of metal ion specificity remains an intriguing question. We recently characterized a mutant of ZntA lacking its amino-terminal domain, ⌬N-ZntA (21). ⌬N-ZntA is fully capable of ATP-dependent soft metal transport and has the same metal specificity as the intact protein, although its activity is slightly lower than that of ZntA. Thus, the core transport domain of ZntA is necessary and sufficient for the recognition and transport of specific metal ions.
In the present work, our goal was to investigate whether the amino-terminal domain of P1-type ATPases also determines specificity toward particular metal ions. Toward this end, we constructed chimeric proteins in which the amino-terminal domain of ZntA was replaced by the amino-terminal domain of the Wilson Cu(I)-transporter ATP7B (Fig. 1). In one chimeric protein designated WND-Cu(1-6)-ZntA, the entire amino-terminal domain of ATP7B, containing all six metal-binding motifs, was attached to ZntA lacking its amino-terminal domain (⌬N-ZntA). In the second chimeric protein, WND-Cu(6)-ZntA, only the sixth metal-binding motif of ATP7B was attached to ⌬N-ZntA. It has been reported that the sixth metal-binding domain is necessary and sufficient for the transport activity of ATP7B (22). Both chimeric proteins were able to confer resistance toward Pb(II), Cd(II), and Zn(II) in a zntA-disrupted E. coli strain but not to copper in an E. coli strain disrupted in copA, which encodes a Cu(I)-transporting ATPase (23). Thus, in vivo the chimeras appear to behave in a manner similar to ZntA and ⌬N-ZntA but not to CopA. The purified chimeric proteins were active; the ATPase activity was stimulated by Pb(II), Cd(II), Zn(II), and Hg(II), cations that are substrates for ZntA. No activity was obtained with Cu(I) or Ag(I), cations that are substrates for ATP7B. The V max values obtained for both chimeras are lower than the values obtained for ZntA, especially in the presence of thiolates in the assay buffer. These results confirm our previous conclusion that metal ion specificity is determined by the transmembrane part of the ATPase; the amino-terminal domain cannot override this intrinsic specificity. Additionally, these results demonstrate that the aminoterminal domain interacts with the rest of the transporter in a metal ion-specific manner; the amino-terminal domain of ATP7B cannot replace that of ZntA in restoring full catalytic activity.

EXPERIMENTAL PROCEDURES
Materials-Monoclonal antibodies against the hexahistidyl tag were from CLONTECH, Palo Alto, CA. L-␣-Phosphatidylcholine (asolectin) was purified prior to use (24). All other chemicals were of the highest commercial grade.
Construction of WND-Cu(6)-ZntA and WND-Cu(1-6)-ZntA-The chimeric proteins WND-Cu(6)-ZntA and WND-Cu(1-6)-ZntA were constructed as follows. For WND-Cu(6)-ZntA, a fragment of DNA encoding amino acids 543-666 of ATP7B that contains the sixth copper-binding domain was generated by polymerase chain reaction. An NcoI site was generated at the 5Ј-end to provide an initiation codon, and a KpnI site was generated at the 3Ј-end of the fragment using the oligonucleotides 5Ј-GACCATGGCTCAGTTCATCCAGGAC-3Ј and 5Ј-CTGGTACCCTG-CTTTATTTCCATTTTG-3Ј together with the plasmid WCBD-pGEX-6P-2 as template. Plasmid WCBD-pGEX-6P-2 was generated by cloning the cDNA encoding the six Cu(I)-binding domains of ATP7B in pGEX-6P-2 (Amersham Pharmacia Biotech) (18). A fragment of DNA encoding amino acids 106 -732 of ZntA containing a KpnI site at the 5Ј-end and an EcoRI site at the 3Ј-end was generated using the oligonucleotides 5Ј-GAGGTACCAAAGCAGGCTATTCCCTG-3Ј and 5Ј-CTTCGAATTCT-CTCCTGCGCAACAATC-3Ј and the plasmid pZntA as a template (20). The two individual polymerase chain reaction products were subcloned into pGEM-T (Promega). The WND-Cu(6) fragment was excised by restriction digestion using NcoI and KpnI, while the ZntA fragment was excised using KpnI and EcoRI. The two fragments were purified and ligated together with pBAD/MycHis-C (Invitrogen), which was previously digested with NcoI and EcoRI to generate plasmid pWND-Cu(6)-ZntA in which the WND-Cu(6)-ZntA chimera is in-frame with a hexahistidyl tag at the carboxyl terminus.
As a first step toward construction of the DNA fragment encoding the chimera WND-Cu(1-6)-ZntA, the existing EcoRI and NcoI sites, located between nucleotides 1325 and 1336 of the coding sequence of WCBD(1-6)-pGEX-6P-2, were disabled by site-directed mutagenesis by using the U.S.E. Mutagenesis Kit (Amersham Pharmacia Biotech). The oligonucleotide used was 5Ј-GTTTGCACCATACTATTCCCAGCACTGTGG-3Ј in which the altered bases, indicated in bold and italics, replaced GGA in the wild-type DNA. This mutation does not change the amino acid sequence of the amino-terminal domain of ATP7B. The resulting altered plasmid was then used as the template in a polymerase chain reaction to generate a fragment of DNA encoding amino acids 1-666 of ATP7B with NcoI and KpnI sites at the 5Ј-and 3Ј-ends respectively, by using the oligonucleotide 5Ј-TGACCATGGGACCTGAGCAGGAGAGA-CAG-3Ј and the same reverse primer that was used to generate WND-Cu (6). The subsequent steps in creating the plasmid pWND-Cu(1-6)-ZntA were essentially the same as described above for the WND-Cu(6)-ZntA chimera.
To ensure that the polyhistidyl group at the carboxyl terminus can be removed when required, the plasmids carrying the chimeras were further modified. A synthetic oligonucleotide with coding sequence for the cleavage site of PreScission protease (human rhinovirus 3C protease) was introduced by an in-frame insertion between the EcoRI and HindIII sites of both plasmids pWND-Cu(6)-ZntA and pWND-Cu(1-6)-ZntA. This insert also contained an XbaI site to facilitate its identification in the plasmid. The oligonucleotides used were 5Ј-AATTCCTGGAAGTT-CTGTTCCAGGGGCCCTCTACTAGAAA-3Ј and 5Ј-AGCTTCTCTAGA-GGGCCCCTGGAACAGAACTTCCAGG-3Ј. The two oligonucleotides were phosphorylated at the 5Ј-end using the T4 polynucleotide kinase, annealed, then ligated with the plasmids that were precut with EcoRI and HindIII. The presence of the insert in the ligated plasmids was confirmed by XbaI digestion.
The fidelity of the sequences of the chimeras was verified by automated DNA sequencing (DNA Sequencing Facility, Center for Applied Genomics, Hospital for Sick Children, Toronto, Canada).
ATPase Assay and Protein Estimation-The metal ion-dependent ATPase activity of purified ZntA and the chimeric proteins was assayed at 37°C either by the pyruvate kinase/lactate dehydrogenase-coupled assay or by a colorimetric method that measures phosphate release at fixed time intervals (26 -27). The purified proteins were incubated with 1-2 mM dithiothreitol at 4°C for 1 h prior to assays. The typical assay buffer used was 0.1 M acetic acid, 0.05 M BisTris, and 0.05 M triethanolamine (pH 7.0) containing 0.1% purified asolectin, 0.2% Triton X-100, 10% glycerol, and 5.0 mM each ATP and MgCl 2 . When assays were performed in the presence of silver ions, care was taken to ensure that nitrate was the only anion present in the assay buffer. Lead acetate, zinc acetate, cadmium chloride, and other soft metal salts as well as cysteine were added as required. For the coupled assay, 0.25 mM NADH, 1.25 mM phosphoenolpyruvate, 6 -9 units of pyruvate kinase, and 13 units of lactate dehydrogenase were also included. For the phosphate release assay, 50-l aliquots of the assay mixture were removed at fixed time intervals and quenched with 10 l of 0.5 M EDTA. The volume was made up to 500 l with water followed by the addition of 172 l of reagent A (28 mM ammonium molybdate in 2.1 M sulfuric acid) and 128 l of reagent B (0.35% polyvinyl alcohol and 0.76 mM malachite green in water). The color was allowed to develop by incubation at room temperature for 20 min followed by reading the absorbance at 610 nm. A standard curve was constructed using sodium phosphate.
The K m for MgATP was measured by taking into account the dissociation constants of MgATP, MgHATP, and the pK a values of ATP (28).
In assays where thiols were added, the concentration of cysteine required to generate a constant metal ion:thiolate ion ratio was calculated using a pK a of 8.33 for cysteine. Protein concentrations were determined using the bicinchoninic acid reagent with bovine serum albumin as standard. Data were analyzed with Kaleidagraph for the Macintosh (Synergy Software).

RESULTS
The Chimeric Proteins Are Able to Confer Resistance to Lead, Zinc, and Cadmium but Not to Copper-We showed earlier that the hypersensitivity of the zntA-disrupted strain LMG194(zntA::kan) to lead, zinc, and cadmium salts can be complemented by transforming it with a plasmid containing the genes for ZntA or ⌬N-ZntA, which is a mutant of ZntA lacking the amino-terminal domain (11, 20 -21). The ability of the chimeric proteins to confer resistance to lead, zinc, and cadmium salts was tested. Fig. 2 shows growth of LMG194, LMG194(zntA::kan), and LMG194(zntA::kan) transformed with plasmid pZntA, p⌬N-ZntA, pWND-Cu(1-6)-ZntA, or pWND-Cu(6)-ZntA in minimal media containing different concentrations of lead, zinc, and cadmium salts. The chimeric proteins were able to complement the sensitivity of the zntAdisrupted strain to the same extent as ZntA or ⌬N-ZntA. Time courses of growth for the same strains in the absence and presence of 5 M lead acetate, 50 M zinc acetate, and 5 M cadmium chloride were also measured (data not shown). Growth rates for LMG194 as well as for LMG194(zntA::kan) containing pZntA, p⌬N-ZntA, pWND-Cu(1-6)-ZntA, or pWND-Cu(6)-ZntA were similar, although all four plasmid-bearing strains exhibited an initial lag in growth compared with the wild-type strain LMG194.
ZntA does not confer resistance to copper salts in the copAdisrupted strain, LMG194(copA::kan) (23). We tested the ability of the chimeric proteins to confer resistance to copper salts. Fig. 3 shows growth of LMG194, LMG194(copA::kan), and LMG194(copA::kan) transformed with plasmids containing CopA, WND-Cu(1-6)-ZntA, or WND-Cu(6)-ZntA in the presence of different concentrations of cupric chloride. Unlike CopA, the chimeric proteins were unable to complement the sensitive phenotype of the copA-disrupted strain.
Expression and Purification of the Two Chimeric Proteins-The two chimeric proteins were expressed using the same expression vector used for ZntA. WND-Cu(6)-ZntA was expressed at levels similar to ZntA; the level of expression of WND-Cu(1-6)-ZntA was lower than that of ZntA under similar growth and induction conditions. The WND-Cu(6)-ZntA chimera could be purified to near homogeneity; however, the WND-Cu(1-6)-ZntA chimera appeared to be partially degraded into smaller fragments in vivo. The proteins were purified using Ni(II) affinity chromatography; SDS-PAGE analysis of the chimeras is shown in Fig. 4. The molecular masses of WND-Cu(6)-ZntA and WND-Cu(1-6)-ZntA, including the histidyl and Myc tags, are 82 and 139.6 kDa, respectively. It is clear from Fig. 4 that WND-Cu(6)-ZntA could be purified to near homogeneity, but the purified WND-Cu(1-6)-ZntA had smaller degradation products. The size of the smallest fragment corresponded to the size of ⌬N-ZntA.
Soft Metal Cation-dependent ATPase Activity of the Chimeric Proteins-Both chimeric proteins were active and displayed ATPase activities that were stimulated by soft metal ions. As previously observed for ZntA and for ⌬N-ZntA, pretreatment of the chimeras with dithiothreitol as well as with phospholipids in the assay buffer was required for the ATPase activity. Additionally, the thiolate form of cysteine in the assay buffer increased the activity of both chimeric proteins, as has also been observed with ZntA and ⌬N-ZntA (20 -21). The soft metal ion-stimulated ATPase activity was measured as a function of the MgATP concentration in the presence of 100 M Pb(II) (data not shown). The K m values for MgATP at pH 7.0 and 37°C for WND-Cu(1-6)-ZntA and WND-Cu(6)-ZntA were 89 Ϯ 9 and 160 Ϯ 25 M, respectively; the K m for MgATP for ZntA was 106 Ϯ 13 M. Table I summarizes the kinetic parameters of both chimeric proteins together with those of ZntA and ⌬N-ZntA for the soft metal-stimulated ATPase activity in the presence of excess MgATP (5 mM each Mg(II) and ATP) at pH 7.0 and 37°C. For all four proteins, Pb(II) was the best metal ion substrate. In the absence of thiolates, the V max of the Pb(II)-stimulated activity for both chimeric proteins was ϳ2-fold lower compared with the value for ZntA. The Zn(II)-and Cd(II)-stimulated activities of WND-Cu(6)-ZntA were similar to those of ZntA, whereas those of WND-Cu(1-6)-ZntA were 2-fold lower. The K m values for all three metals were similar; these K m values in the absence of thiolates refer to the ATP complexes of soft metal ions given the magnitude of the association constants of PbATP, ZnATP, CdATP, and MgATP (29). When thiolates were present in the assay buffer at a soft metal ion:thiolate ratio of 1:1, both the V max and the apparent K m values were higher for all four proteins. The V max values for both chimeras were ϳ2-3-fold lower than that of ZntA; the apparent K m values were also lower than that of ZntA. These V max and K m values for the chimeras are very similar to those obtained for ⌬N-ZntA (Table I).
The Metal Ion Specificity for the Chimeric Proteins Is Identical to ZntA-Metal cations other than Cd(II), Pb(II), and Zn(II) were tested for their ability to stimulate the ATPase activity of both WND-Cu(6)-ZntA and WND-Cu(1-6)-ZntA. Co(II), Ni(II), Cu(II), Fe(II), Cr(III), and Bi(III) were unable to stimulate the ATPase activity above background levels. As previously observed for ZntA, Hg(II) was able to stimulate the ATPase activity of the chimeric proteins in the presence of added thiolates (20); however, it was less effective than Pb(II), Zn(II), or Cd(II). In the absence of thiolates in the assay medium, Hg(II) displayed no activity with ZntA, ⌬N-ZntA, or either chimeric protein. Of the divalent metals Co(II), Ni(II), Fe(II), and Cu(II), only Cu(II) inhibited the ATPase activity of the chimeras, as was also true for ZntA and ⌬N-ZntA. 5 M Cu(II) completely inhibited the Pb(II)-stimulated ATPase activity of all four proteins. Because Cu(I) and Ag(I) are substrates of ATP7B, we tested the effect of these monovalent metals on the ATPase activity of the chimeras. Cu(I) (generated from Cu(II) in the presence of excess dithiothreitol) and Ag(I) did not increase the activity of the chimeras above background levels. When testing Ag(I), care was taken to ensure that the lack of Ag(I) activity was not due to its insolubility in the assay buffer.

DISCUSSION
Members of the P1-type ATPase subfamily have a high degree of homology with each other that extends to the hydrophilic metal ion-binding amino-terminal domain, the transmembrane domain, and the ATPase domain. Despite this high level of sequence similarity, P1-type ATPases display stringent specificity in the transport of specific metal ions. For example, ZntA is specific for the ions Pb(II), Zn(II), Cd(II), and Hg(II), while ATP7A, ATP7B, and CopA from E. coli and homologues transport Cu(I) and possibly Ag(I) (5-9, 11-13, 23). Other P1type ATPases with different metal ion specificities include SilP, specific for Ag(I), and CoaT, specific for Co(II) (30 -31). It is also likely that plants have multiple P1-type ATPases that transport different metal ions including Zn(II), Cu(I), Ni(II), and Mn(II) (32). The molecular determinants of metal ion recognition and specificity in P1-type ATPases are not known. We have previously shown that ⌬N-ZntA, a truncated mutant of ZntA that lacks the amino-terminal domain, is an active ATPase with the same metal ion specificity as ZntA, demonstrating that the amino-terminal domain is not essential for function and that the transmembrane domain determines specificity (21). However, ⌬N-ZntA was less active than the fulllength ZntA, suggesting that the amino-terminal domain may have a function in enhancing the overall catalytic activity, possibly by increasing the rate of metal ion binding to the pump. Studies on the isolated amino-terminal domains of a few P1-type ATPases, notably ATP7A and ATP7B, have shown that these domains can bind a variety of soft metals, including those that are not substrates of the full-length pump (17)(18)(19). For example, the amino-terminal domain of ATP7B is able to bind Cu(I), a substrate, as well as Zn(II), Cd(II), and Hg(II), which are ions that are not substrates. Zinc binds to this domain with a stoichiometry of 6:1 and upon binding induces conformational changes that are completely different from those previously observed for copper. Spectra from x-ray absorption spectroscopy indicate that zinc is ligated primarily to nitrogen atoms, which is different from the behavior of copper-binding ligands of this domain. 3 We therefore wanted to investigate the impact of interchanging the amino-terminal domains of P1-type ATPases with mutually exclusive metal ion substrates. In particular, we wanted to determine whether such chimeric proteins are active ATPases and if so, whether the amino-terminal domain is able to alter the specificity of the transmembrane domain. In this study, we have characterized the properties of chimeric proteins constructed by splicing the amino-terminal domains of the mammalian transporter ATP7B with the bacterial transporter ZntA lacking its amino-terminal domain. In one chimera the entire amino-terminal domain of ATP7B, including all six metal-binding motifs, was used; in the second, only the sixth metal-binding domain of ATP7B was used. The sixth metal-binding domain of ATP7B approximately corresponds in size to the amino-terminal domain of ZntA. Also, this domain has been shown to be sufficient for the in vivo transport function of ATP7B in yeast (22).
Both chimeric proteins were able to mediate resistance to lead, zinc, and cadmium salts in a zntA-disrupted strain but not to copper salts in a copA-disrupted strain. Thus, physiologically the chimeras behave like ZntA and not like ATP7B and CopA. Therefore, the amino-terminal domain of ATP7B cannot override the metal ion specificity of ZntA. Additionally, these in vivo results suggested that the chimeric proteins are active transporters. This was confirmed by measuring the ATPase activities of the purified chimeras. Both WND-Cu(1-6)-ZntA and WND-Cu(6)-ZntA could be expressed in functional forms in E. coli, although WND-Cu(1-6)-ZntA appeared to be degraded to some extent in vivo. The purified hybrid proteins were able to catalyze the soft metal ion-dependent hydrolysis of ATP. The K m values of MgATP for WND-Cu(1-6)-ZntA and WND-Cu(6)-ZntA for the Pb(II)-stimulated activity at pH 7.0 were similar to that of ZntA (90 -160 M). ZntA is specific for the divalent soft metal cations Pb(II), Zn(II), Cd(II), and Hg(II), with Pb(II) displaying the highest activity (20). The chimeric proteins showed the same substrate specificity, with Pb(II) again displaying the highest activity. In particular, no activity was observed with either chimera for Cu(I) or Ag(I).
A comparison of the V max and K m parameters for the ATPase activities of the purified chimeras with those for the ATPase activities of ⌬N-ZntA and ZntA in the absence of thiolates shows that the Pb(II)-and Zn(II)-stimulated activities of the chimeras and of ⌬N-ZntA are ϳ2-3-fold lower than those of ZntA. Interestingly, the Cd(II)-stimulated activity of WND-Cu(6)-ZntA is consistently higher than that of ZntA, ⌬N-ZntA, or WND-Cu(1-6)-ZntA. We have previously reported that thiolates increase the ATPase activity of ZntA and ⌬N-ZntA, possibly by increasing the rate of metal ion release from the transporter (20 -21). The effect is most pronounced for the Cd(II)-ATPase activity. Thiolates were able to increase the activities of both chimeras for all three metal ions. The thiolatestimulated activities of WND-Cu(1-6)-ZntA, WND-Cu(6)-ZntA, and ⌬N-ZntA were highly similar to each other and 2-3-fold lower than those for ZntA (Table I). The thiolatestimulated activity is likely to be a more accurate reflection of the in vivo activity of ZntA, given that glutathione is abundant inside the cell and that metal ion chaperones present in the periplasm may assume the role of thiolates. Thus, the data in Table I strongly suggest that the chimeric proteins resemble ⌬N-ZntA in terms of their catalytic abilities. Therefore, the amino-terminal domain of ATP7B cannot replace that of ZntA in restoring full catalytic activity.
It has been suggested that the full-length amino-terminal domain of ATP7B may play a regulatory role; metal ion binding releases an inhibitory interaction of this domain with the hydrophilic ATPase domain (33). In this case, it would be expected that the WND-Cu(1-6)-ZntA chimera would be inactive until copper ions were added to the assay medium. In other words, we would observe copper-stimulated Pb(II)-ATPase activity for WND-Cu(1-6)-ZntA. Contrary to this expectation, Cu(II) in the presence and absence of dithiothreitol did not increase the Pb(II)-stimulated ATPase activity; it completely inhibited the Pb(II)-ATPase activity. This observation suggests that at least in the case of the WND-Cu(1-6)-ZntA chimera the amino-terminal domain does not play a regulatory role.
Conclusion-We constructed two chimeric proteins in which either the entire amino-terminal domain of ATP7B or only the sixth metal-binding domain of ATP7B was attached to ZntA lacking its amino-terminal domain. The chimeras were able to mediate resistance to lead, zinc, and cadmium salts but not to copper salts in vivo. Thus, physiologically, they resemble ZntA. The purified chimeras were active ATPases. The soft metal cation-dependent ATPase activity was specific for Pb(II), Cd(II), Zn(II), and Hg(II), metal ion substrates of purified ZntA; the highest activity was obtained with Pb(II). There was no activity with Cu(I) or Ag(I), which are substrates of ATP7B. The V max values for the chimeras are ϳ3-fold lower than the values for ZntA. Thus, the amino-terminal domain of ATP7B does not override the specificity of the transmembrane segment of ZntA. Also, it cannot replace the amino-terminal domain of ZntA in restoring full catalytic activity to ZntA.