Pb(II)-translocating P-type ATPases.

The cad operon of Staphylococcus aureus plasmid pI258, which confers cadmium resistance, encodes a transcriptional regulator, CadC, and CadA, an ATP-coupled Cd(II) pump that is a member of the superfamily of cation-translocating P-type ATPases. The Escherichia coli homologue of CadA, termed ZntA, is a Zn(II)/Cd(II) pump. The results described in this paper support the hypothesis that ZntA and CadA are Pb(II) pumps. First, CadC is a metal-responsive repressor that responds to soft metals in the order Pb>Cd>Zn. Second, both CadA and ZntA confer resistance to Pb(II). Third, transport of 65Zn(II) in everted membrane vesicles of E. coli catalyzed by either of these two P-type ATPase superfamily members is inhibited by Pb(II).

In this report, we show that CadC repression of the cad promoter is relieved upon addition of soft metals, with the order of effectiveness Pb(II) Ͼ Cd(II) Ͼ Zn(II). In E. coli Zn(II) responsiveness could be observed only in a zntA-disrupted strain. The zntA-disrupted strain of E. coli exhibited hypersensitivity to Pb(II) that was complemented by cadA, indicating that both soft metal-translocating P-type ATPases are essential for Pb(II) resistance in bacteria. Everted membrane vesicles from cells expressing either zntA or cadA exhibited ATPdependent 65 Zn(II) accumulation. Since no radioisotopes of Pb(II) are available, direct transport of Pb(II) was not assayed. However, Pb(II) inhibited 65 Zn(II) transport, indicating that Pb(II) is a substrate of the two P-type ATPases. These results support the concept that ZntA and CadA are Pb(II) pumps with physiological functions that include to provide resistance to environmental lead.

EXPERIMENTAL PROCUEDURES
Growth of Cells-The bacterial strains and plasmids used in this study are listed in Table I. Cells were grown in LB medium at 37°C. Ampicillin (50 g/ml), kanamycin (50 g/ml), chloramphenicol (80 g/ ml), isopropyl-␤-D-thiogalactopyranoside (0.1 mM) and 5-bromo-4chloro-3-indolyl-␤-D-galactosidase (80 g/ml) were added as required. For determination of sensitivity to metal ions, a basal salts medium was used (14) with the omission of zinc salts. The pH of the medium was adjusted to 5.5 to prevent precipitation of lead salts. Cells were grown overnight, diluted 50-fold in the same medium containing metal ion salts, and incubated for 24 h at 37°C with shaking. Growth was monitored from the absorbance at 600 nm.
DNA Manipulation-Preparation of plasmid DNA was performed using a Wizard DNA purification kit (Promega). Endonuclease digestions, electrophoretic separations and isolations, ligations, transformations, and Klenow fragment fill in were performed according to standard procedures (15) unless otherwise noted. The conditions for polymerase chain reaction (PCR) 1 were as described previously (16). Restriction endonucleases, T4 DNA ligase, Klenow fragment of DNA polymerase I, and Taq polymerase were from Life Technologies, Inc. For DNA sequencing, double-stranded DNA was isolated with a plasmid minikit from Qiagen and then sequenced by the method of Sanger et al. (17) using an ALFexpress system and a Cy5-labeled sequence kit (Pharmacia Biotech Inc.).
In Vivo Measurement of Inducer Specificity of cadC-The lacZ reporter gene plasmid pYS2 was constructed to monitor the regulatory properties of the cadC gene product. A 121-base pair fragment from plasmid pYPK11 containing the pI258 cad operator/promoter was amplified by PCR. The fragment was engineered with an EcoRI at the 5Ј end and a BamHI site at the 3Ј end. The fragment was ligated into plasmid pMLB1034 that had been digested with EcoRI and BamHI, generating plasmid pYS2, in which a lacZ gene is controlled by the cad operator/promoter. In several steps the pI258 cadC gene was amplified by PCR from plasmid pYPK11 and cloned as a 0.5-kilobase pair fragment into plasmid pACYC184 under control of the T7 promoter.
Overnight LB ϩ 2% glucose cultures of E. coli strains BL21(DE3) or BL21(DE3) zntA::km harboring compatible plasmids pYS2 and pYSC1 were diluted 20-fold into a low phosphate minimal medium (14) containing 2% glucose plus the appropriate antibiotics. Pb(OAc) 2 , ZnSO 4 , Cd(OAc) 2 , HgCl 2 , NaAsO 2 , Bi(NO 3 ) 2 , CuSO 4 , NiCl 2 , or potassium anti-* This work was supported in part by United States Public Health Service Grants GM55425 (to B. P. R.) and GM54102 (to B. M.) and National Research Service Award GM18973 (to C. R.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  For assay of lacZ expression, portions (0.5 ml) of cultures were centrifuged at 13,000 ϫ g and suspended in 0.5 ml of Z buffer (60 mM Na 2 HPO 4 , 40 mM NaH 2 PO 4 , 10 mM KCl, 1 mM MgCl 2 , pH 7) and permeabilized by addition of 30 l of 0.1% SDS and 50 l of chloroform, followed by vigorous mixing by vortexing for 1 min (18). The cells were incubated at 37°C for 10 min. The ␤-galactosidase assay reaction mixture contained 50 1 of permeabilized cells, 0.1 ml of 8 mg/ml o-nitrophenyl-␤-D-galactopyranoside and 0.85 ml of Z buffer. ␤-Galactosidase activity was estimated from the release of nitrophenol at 420 nm at 37°C and expressed in Miller units.
Assay of 65 Zn(II) Accumulation-Everted membrane vesicles were prepared as described (19). Cells were grown overnight at 37°C in 20 ml of LB and diluted 50-fold in prewarmed medium. At an optical density of 0.5 at 600 nm the cultures were induced with 0.  4 and DNase I (20 g/ml), cells were incubated on ice for 30 min. Cell debris was pelleted by centrifugation at 10,000 ϫ g, the supernatant was centrifuged at 100,000 ϫ g for 60 min to isolate the membranes. The pelleted membrane vesicles were washed once with the same buffer without EDTA and stored at Ϫ70°C until use.
Transport assays were performed at room temperature as described previously (6)

CadC Responds to Pb(II), Cd(II), and Zn(II) in Vivo-In cells of S. aureus
CadC represses transcription from the pI258 cad operator/promoter in the absence of metals (21). The response to metal ions has been examined in both S. aureus and E. coli (12). However, those results are complicated by the presence of chromosomally encoded efflux pumps in S. aureus 2 and E. coli (6,9) that limit the ability of the cells to accumulate intracellular metal ions.
We have developed a method to probe the ability of metals to release CadC from the operator/promoter in vivo. The method relies on the construction of a strain of E. coli in which zntA has been disrupted (6), with introduction of a two-plasmid reporter   (Fig. 1). Since this strain has a chromosomal zntA gene, expression was insensitive to 0.1 mM ZnSO 4 . In contrast, when a zntA disruption was transduced into E. coli strain BL21(DE3), which was subsequently transformed with plasmids pYSC1 and pYS2, repression could be relieved by low concentrations of Zn(II), with maximal induction at 0.8 M ZnSO 4 (Fig. 1), an increase in metal responsiveness of about 3 orders of magnitude. This result demonstrates that the inability to observe derepression in wild type cells of E. coli is the result of ZntA activity. Repression by cadC in the zntA disrupted strain was relieved most effectively by Pb(II), with significant derepression at 25 nM Pb(OAc) 2 and complete response by 200 nM. In cells of the wild type, 3 M Pb(OAc) 2 was required to give the same response, a 15-fold increase. Derepression by Cd(II) was intermediate, with maximal induction at 300 nM. Interestingly, the wild type response to Cd(II) was only shifted by a factor of two relative to the zntA disrupted strain. It is not clear why there is so little difference between wild type and mutant with this metal ion. Other metal ions, including Hg(II), Cu(II), Ni(II), As(III), Sb(III), and Bi(III), showed little or no induction (data not shown).
The zntA and cadA Genes Confer Pb(II) Resistance-Disruption of zntA rendered the cells hypersensitive to Pb(II) (Fig. 2). The wild type could grow in concentrations as high as 200 M (data not shown). In contrast, E. coli strain RW3110 (znt::km) showed growth inhibition even at 100 nM, and no growth was observed at a concentration of 1 M. Thus zntA confers Pb(II) resistance. Pb(II) sensitivity was complemented by plasmid pKJ3, which carries a pI258 cadA gene (Fig. 2). Complementation was also observed with plasmid pKPY11, which carries the entire cadCA operon (data not shown). Thus cadA also confers Pb(II) resistance.
CadA Catalyzes ATP-dependent Transport of 65 Zn(II)-Expression of cadA in either B. subtilis or E. coli has been shown to produce cells and everted membrane vesicles capable of transporting 109 Cd(II) (7). We have shown previously that cadA can complement the Zn(II)-sensitive phenotype of the zntAdisrupted E. coli strain (10). However, the ability of CadA, which is only 30% identical to ZntA, to catalyze transport of Zn(II) has not been examined. Accumulation of 65 Zn(II) was observed in everted membrane vesicles prepared from cells of the zntA-disrupted E. coli strain RW3110 bearing plasmid pKPY11, which has the pI258 cadCA operon (Fig. 3A). In that experiment the cells were induced with 0.2 mM Cd(OAc) 2 . Ex- pression of CadA-catalyzed Zn(II) transport could also be induced with Zn(II) (data not shown). In the absence of a source of energy, no time-dependent uptake of 65 Zn(II) was observed. Addition of MgATP produced time-dependent accumulation of 65 Zn(II) in the strain expressing cadA. Thus CadA, like ZntA, is a Zn(II) pump. 65 Zn Transport by ZntA and CadA Is Inhibited by Pb(II)-The ability of ZntA and CadA to confer resistance to Pb(II) implies that those two cation-translocating P-type ATPases pump Pb(II). Since there are no available lead isotopes, direct measurement of transport was not possible, so the effect of Pb(II) on 65 Zn(II) accumulation in everted membrane vesicles was examined. At 10 M 65 ZnSO 4 , a concentration near the K m value, 10 M Pb(OAc) 2 inhibited the initial rate of ATP-dependent accumulation of 65 Zn (II) catalyzed by either CadA (Fig. 3A) or ZntA (Fig. 3B). The concentration dependence of inhibition of ZntA transport by Pb(II) and Cd(II) was determined (Fig.  3C). With both cations, half-maximal inhibition was observed in a range of 2-4 M, indicating K I values for both Pb(II) and Zn(II) within the same range as the K m for Zn(II). Since ZntA has been shown to transport both Zn(II) and Cd(II), these results suggest that it also transports Pb(II) and that the affinity of the pump is within the same order of magnitude for each of the three cations.

DISCUSSION
In humans chronic exposure to low levels of lead may cause neurological, reproductive, and developmental problems. Lead exposure is especially harmful to children, and nearly one million American children below the age of 5 years have bloodlead levels that exceed those considered as elevated by the Centers for Disease Control and Prevention (22). Even though lead affects virtually every organ and tissue in the body, little is known about the routes of lead ion uptake and extrusion. Even less is known about Pb(II)-regulated gene transcription, and there are no genetic markers for lead exposure. We have undertaken a study of Pb(II)-responsive genes and transporters. The CadC repressor and CadA/ZntA pumps represent the first proteins demonstrated to have a physiological function that includes providing the host organism with a protective response to environmental lead stress.
CadA and ZntA are members of the superfamily of P-type cation-translocating ATPases, but belong to a group of soft metal transporters that includes bacterial enzymes such as the CopA Cu(I) pump (23), the yeast CCC2 Cu(I) pump (24), and the human Cu(I)-transporting ATPases such as MNK (25) and WND (26). The soft metal pumps can be further subdivided into the Cu(I)/Ag(I)-translocating ATPases and the Zn(II)/Cd(II)/ Pb(II) ATPases (1). While none of these proteins has yet been demonstrated to catalyze ATP hydrolysis, several have been shown to have properties consistent with being cation-translocating ATPases. First, transport requires ATP and is inhibited by orthovanadate, a classical inhibitor of P-type ATPases (5). Second, P-type ATPases form a ␤-acylphosphate intermediate, and several soft metal pumps have been shown to form these intermediates (27,28).
CadA had been shown to catalyze ATP-coupled, vanadatesensitive Cd(II) transport (7) and to form a phosphoenzyme intermediate (27). ZntA has been shown to transport both Zn(II) and Cd(II) in an ATP-requiring, vanadate-sensitive reaction (6). In this paper we report that CadA also transports Zn(II). Despite the unavailability of lead isotopes for direct transport assays, the results described here are consistent with Pb(II) transport catalyzed by members of this subgroup of P-type ATPases. First, Pb(II) induces expression of CadA (Fig.  1). Second, both ZntA and CadA confer Pb(II) resistance (Fig.  2). Third, Pb(II) inhibits 65 Zn(II) transport by both ZntA and CadA (Fig. 3, A and B). Fourth, inhibition of ZntA activity by Pb(II) was essentially identical to inhibition by Cd(II), a known pump substrate (Fig. 3C).
Copper pumps are widely distributed in nature, and genetic diseases such as Menkes and Wilsons result from mutations in the genes for these pumps (25,26). We predict that Zn(II)/ Cd(II)/Pb(II) P-type ATPases exist in humans, and it is not unreasonable to expect that there are diseases related to defects in the genes for these pumps. Elucidation of these bacterial model systems may also lead to the development of biomarkers for lead exposure and susceptibility in humans.