Uranyl Acetate as a Direct Inhibitor of DNA-Binding Proteins

Zinc finger proteins, one of the largest families of DNA-binding proteins in higher eukaryotes, are so named because they require zinc ions for approp...
0 downloads 0 Views 202KB Size
Download PDF
784

Chem. Res. Toxicol. 2007, 20, 784-789

Uranyl Acetate as a Direct Inhibitor of DNA-Binding Proteins Wendy J. Hartsock,†,‡ Jennifer D. Cohen,† and David J. Segal*,†,§ Department of Pharmacology and Toxicology, UniVersity of Arizona, Tucson, Arizona 85721, and Genome Center and Department of Pharmacology, UniVersity of California, DaVis, California 95616 ReceiVed December 4, 2006

Zinc finger proteins, one of the largest families of DNA-binding proteins in higher eukaryotes, are so named because they require zinc ions for appropriate structure and function. Dysregulation of zinc fingercontaining DNA transcription and repair proteins has been proposed as a potential mechanism for the toxic effects of some metal ions. Uranium metal has been reported to be both a cytotoxic and a genotoxic agent. We hypothesized that these toxic effects of uranium might be due to its ability to directly disrupt zinc finger activity. To test this hypothesis, two purified zinc finger proteins, Aart and Sp1, were analyzed by electrophoretic mobility shift in the presence of uranyl acetate. Inhibition of binding was apparent at 10 µM uranyl acetate, while no inhibition was observed with up to 2000 µM the cytotoxic metalloid sodium arsenite. Preincubation of the DNA with uranyl acetate did not inhibit zinc finger protein binding, suggesting that the inhibition was due to direct uranyl interaction with the protein. Surprisingly, uranyl acetate inhibited two nonzinc finger DNA-binding proteins, AP1 and NF-κB, to a similar extent, and zinc finger inhibition was reduced in the presence of bovine serum albumin. These results suggest that uranium can directly inhibit the function of DNA-binding proteins, most likely via a nonspecific protein interaction. Introduction Uranium is the heaviest naturally occurring element and has both radiotoxic and chemotoxic properties. The three major isotopes, U-235, U-235, and U-238, are all radioactive. U-238 is the most abundant, accounting for 99.3% of all natural isotopes. Uranium is most notably used in atomic weapons and as fuel in nuclear power plants, for which the U-235 content is enriched. Depleted uranium (DU),1 in which 50-70% of the U-235 has been removed, is used in armor-piercing weapons because of its high density. By 1879, mining of uranium in Germany and Czechoslovakia (for nonweapon purposes) had already been associated with lung disease (1). Lung cancer in uranium miners in the United States was determined to be due to radioactive radon gas, formed upon decay of U-238 (2). Studies involving DU allow chemotoxic effects to be distinguished from radiotoxicity. Once inside the body, uranium enters the blood primarily as uranyl (UO22+) salts or in complex with proteins (3). Sixty to eighty percent is rapidly excreted in the urine, while tissues such as the kidneys and bones accumulate uranium for months to years (4). In the kidneys, uranium produces chemical damage to renal proximal tubules (5), purportedly by inhibition of Na+, K+-ATPase (6), and oxidative stress (7). Cultured human osteoblasts were transformed to a neoplastic phenotype upon DU exposure (8). More recently, DU was found to be cytotoxic to cultured macrophages and CD4+ T-cells (9) but did not exhibit these effects in cultured rat brain endothelial cells (10). DU exposure resulted in decreased * To whom correspondence should be addressed. Tel: 530-754-9134. Fax: 530-754-9658. E-mail: [email protected]. † University of Arizona. ‡ Current address: Department of Medicinal Chemistry, University of Kansas, Lawrence, Kansas 66045. § University of California. 1 Abbreviations: DU, depleted uranium; UA, uranyl acetate; SA, sodium arsenate; ZFP, zinc finger protein; BSA, bovine serum albumen; EMSA, electrophoretic mobility shift assay.

viability and increased frequencies of chromosomal aberrations in cultured Chinese hamster ovary (CHO) cells (11). These effects were enhanced in DNA repair-deficient CHO cells (XRCC1 mutant), implicating DNA strand breaks as a cytotoxic and mutagenic mechanism (12). Indeed, under reducing conditions, DU alone was able to cause strand breaks in purified DNA (13). In mice, oral or subcutaneous administration of DU results in decreased fertility and embryo/fetal toxicity (14). The results of these studies present uranium as both a genotoxic and a cytotoxic agent. Gene expression and DNA repair are mediated by DNAbinding proteins. One of the largest families of DNA-binding proteins in higher eukaryotes is the Cys2-His2 class of zinc finger proteins (ZFPs) (15). Mutations in ZFPs have been implicated in several human diseases, including developmental disorders and cancer (16). Their name relates to the coordination of a zinc ion between two cysteine and two histidine residues, which is essential to their structure and function (17, 18). For example, nitric oxide (NO) can S-nitrosate cysteine thiols, leading to the reversible disruption of ZFP DNA binding (19). Mimics of glutathione peroxidase were shown to catalyze H2O2-induced oxidation of the ZF transcription factor Sp1 in tumor cells, leading to cell death (20). The zinc can also be displaced by other metals. DNA binding was impaired to varying degrees in the presence of Al3+, Cd2+, Co2+, Cu2+, Fe2+, Hg2+, Mn2+, Ni2+, and Pb2+, the exact effect depending on the particular protein and the experimental conditions (21-24). Dysregulation of ZF-containing transcription factors and repair proteins has been proposed as a mechanism for the cytotoxic and carcinogenic effects of these metals (22, 25, 26). In the current study, we investigated the hypothesis that DU might directly act to disrupt ZFP function, thereby suggesting a potential mechanism for the cytotoxic and mutagenic/ carcinogenic effects observed for this metal. The DNA-binding activity of purified ZFP was analyzed by an electrophoretic mobility shift assay (EMSA) in the presence of uranyl acetate

10.1021/tx600347k CCC: $37.00 © 2007 American Chemical Society Published on Web 04/14/2007

Uranium Inhibition of DNA-Binding Proteins

Chem. Res. Toxicol., Vol. 20, No. 5, 2007 785

(UA) under various conditions. UA was found to be a potent inhibitor of two ZFPs, Aart and Sp1. Unexpectedly, two nonzinc finger DNA-binding proteins, AP1 and NF-κB, were inhibited to a similar extent. Finally, the inhibition was reduced in the presence of bovine serum albumin (BSA). These results suggest that DU can directly inhibit DNA-binding proteins via a nonspecific protein interaction.

Experimental Procedures Caution: These chemicals are hazardous and should be handled carefully in a chemical fume hood. [32P] is radioactiVe and should be handled according to institutional guidelines. Chemicals and Reagents. UA was obtained from Spectrum Quality Products (Gardena, CA). Sodium (meta) arsenite (SA) was purchased from Fluka. Procedures for the disposal of radioactive UA and SA wastes were approved by the University of California, Davis, Office of Environmental Health and Safety. Purified recombinant human proteins Sp1, AP1, and NF-κB and their target oligonucleotides were purchased from Promega. The AP1 preparation is a homodimer of c-Jun. The NF-κB preparation is a homodimer of p50. Protein Aart (27) was expressed and purified from BL21 Star (DE3) bacteria (Invitrogen) using expression vector pMAL-c2X (New England Biolabs), which appends an N-terminal maltose-binding protein (MBP) purification tag. Fusion proteins were purified to >90% homogeneity using the standard protocol of the Protein Fusion and Purification System (New England Biolabs) and our purification buffer (20 mM Tris HCl, pH 7.6/70 mM NH4Cl/7 mM MgCl2/4 mM DTT/0.1% NP-40/50% glycerol). The MBP tag was not cleaved off, and all experiments were conducted with the Aart-MBP fusion protein. The protein purity and concentration were determined from Coomassie blue-stained 10-20% gradient SDS-PAGE gels by comparison to BSA standards. Target oligonucleotides for Aart were purchased from Operon. The sequences of the oligonucleotides used for doublestranded DNA targets in these experiments were (complementary oligonucleotides not shown): Aart, 5′-CCC ATG TAG AGA AAA ACC AGG GCC-3′; Sp1, 5′-ATT GGA TCG GGG CGG GGC GAG-3′; NF-κB, 5′- AGT TGA GGG GAC TTT CCC AGG C-3′; and AP1, 5′-CGC TTG ATG AGT CAG CCG GAA-3′. EMSAs. Target DNA was end-labeled with [γ-32P]ATP (MPBiomedicals). In Figures 1-3A, an 18 µL reaction containing metal and protein in binding buffer A [10 mM Tris-HCl, pH 7.5/50 mM NaCl/1 mM MgCl2/0.5 mM EDTA/0.5 mM DTT/4% glycerol/0.05 mg/mL poly dI-dC (Promega)] was incubated at room temperature for 15 min. Target DNA (14 nM) was added, and incubation continued for 30 min. In Figures 3B-5, a 20 µL reaction containing the first two reaction components (metal, protein, or DNA) in binding buffer B (20 mM Tris HCl, pH 7.6/70 mM NH4Cl/7 mM MgCl2/5 mM DTT/0.1% NP-40) was incubated at room temperature for 15 min. From this reaction, 1 µL was added to a 19 µL volume containing the third component in binding buffer B/15% glycerol/ 1% BSA/0.12 mg/mL sheared herring sperm DNA (HS-DNA; Invitrogen), and incubation continued for 30 min. All samples were loaded onto 5% TBE polyacrylamide gels (BioRad) and electrophoresed for 20 min at 120 V in 0.5× TBE buffer. Gels were dried and exposed to X-AR film (Kodak) at -80 °C. Autoradiograms were scanned and quantified using ImageJ software (National Institutes of Health).

Results UA Inhibition of Aart DNA-Binding Activity. Aart is a six-zinc finger DNA-binding protein that had been engineered to bind the artificial (nonbiologically relevant) sequence 5′-ATGTAG-AGA-AAA-ACC-AGG-3′ (27). This protein was selected for these studies because of its availability in the laboratory and to demonstrate that any inhibitory effects observed could be generalized to any Cys2-His2 ZFP. Aart has been wellcharacterized biochemically (28) and structurally (29). The

Figure 1. EMSA demonstrating the effect of UA on the ability of Aart to bind target DNA. Aart (12 nM) was incubated with 200 µM UA (lanes 1 and 2), 50 µM UA (lanes 3-5), 10 µM UA (lanes 6-8), or 0 µM UA (lanes 9-12) for 15 min in binding buffer A, prior to 30 min of incubation with Aart target DNA (14 nM).

affinity of this protein for its target sequence was determined to be 7.5 pM (27). To determine the ability of UA to inhibit Aart binding its target DNA, purified Aart protein was incubated with various concentrations of UA for 15 min, followed by the addition of 32P end-labeled oligonucleotide target DNA. After an additional 30 min of incubation, the binding activity was assessed by EMSA (Figure 1). The percent of bound DNA was normalized to control reactions that contained no UA (lanes 9-11). No DNA binding was observed when 200 µM UA was preincubated with 12 nM Aart, prior to the addition of 14 nM DNA (lanes 1 and 2). At 50 and 10 µM UA, binding decreased to 12 and 89% of controls, respectively (lanes 3-8). Reducing the Aart concentration by half (6 nM) resulted in further inhibition of DNA binding, corresponding to 0, 3, and 65% of control levels for 200, 50, and 10 µM UA, respectively (data not shown). UA Inhibition of ZFP Sp1 and non-ZFPs AP1 and NFKB. To determine if UA could inhibit the binding of other ZFPs, the DNA-binding domain of the human transcription factor Sp1 was analyzed. Sp1 is a three-finger Cys2-His2 zinc finger that recognizes GC-rich sequences in over 1000 genes. This protein is involved in several cell regulation events including gene expression, apoptosis, and homeostasis (24, 30). The DNAbinding activity of Sp1 has been shown to decrease in the presence of cadmium and lead (31, 32). These metals were suggested to inhibit zinc finger DNA-binding activity by a mechanism involving displacement of the zinc ion. Such metals would therefore not be expected to inhibit the DNA binding of nonmetalloproteins, which do not require coordination of a metal ion for stability. The nonzinc finger DNA-binding domains of Ap1 and NF-κB were therefore studied to investigate if uraniummediated inhibition required the displacement of zinc. AP1 is a nonmetal-containing dimeric transcriptional activator typically composed of subunits c-Jun and c-Fos, although c-Jun homodimers were used in this study. Similarly, NF-κB is a nonmetal-containing dimeric transcription factor that induces cell proliferation and inhibits apoptosis. Its DNA-binding domain is typically composed of subunits p50 and p65, although homodimers of the p50 subunit were used in these experiments (33). Similar to Aart, no DNA binding was observed when 200 or 50 µM UA was preincubated with 200 nM Sp1, prior to the addition of 14 nM DNA (Figure 2A, lanes 1 and 2). At 10 µM UA, binding decreased to 37% of controls (lane 3). Unexpectedly, dramatic inhibition was also observed for AP1 and NFκB. No DNA binding was observed when 200 or 50 µM UA was incubated with 750 nM AP1 (lanes 5 and 6). At 10 µM UA, binding decreased to 23% of controls (lane 7). AP1 bound poorly under the reaction condition used for these assays. However, this caveat did not alter the conclusions. Additionally, no DNA binding was observed when 200 µM UA was incubated with 240 nM NF-κB (lane 13). At 50 or 10 µM UA, binding

786 Chem. Res. Toxicol., Vol. 20, No. 5, 2007

Hartsock et al.

Figure 2. EMSA demonstrating the effect of (A) UA and (B) SA on the ability of zinc finger Sp1 and nonzinc fingers AP1 and NF-κB to bind target DNA. Sp1 (200 nM), AP1 (750 nM), or NF-κB (240 nM) was incubated with the indicated concentration of UA or SA for 15 min in binding buffer A, prior to 30 min of incubation with appropriate target DNA (14 nM). The two Sp1 samples in panel A are duplicates.

decreased to 14 and 87% of controls, respectively (lanes 14 and 15). To examine the possibility that any metal or metal-like compound might inhibit DNA binding at the concentrations used here, similar experiments were performed using the metalloid arsenic in the form of SA. Like uranium, arsenic exposure has been shown to alter gene expression in cells (34, 35). Other studies have presented evidence for (36) and against (35, 37) the ability of arsenic to directly inhibit ZFP binding. In contrast to UA, SA produced no detectable inhibition of DNA-binding activity for ZFP Sp1 or non-ZFPs AP1 and NF-κB, even at arsenic concentrations of 2000 µM (Figure 2B). UA-Mediated Inhibition Is Reduced in the Presence of BSA. The ability of UA to inhibit the activity of both zinccontaining and zincless DNA-binding proteins could be explained by several mechanisms. UA could interact nonspecifically with the DNA in a manner that prevented protein-DNA binding. Alternatively, UA could interact nonspecifically with the proteins in a manner that prevented protein-DNA binding. The latter hypothesis suggests that the addition of other proteins to the binding reaction might absorb UA and thereby reduce the inhibitory effects on DNA-binding proteins. To examine this potential mechanism, binding reactions were performed in the presence or absence of 1% BSA. A dramatic reduction in UA-mediated inhibition of DNA binding was observed when 200, 50, or 10 µM UA was preincubated with 750 nM nonzinc finger AP1 in the presence of 1% BSA, prior to the addition of 14 nM target DNA (Figure 3A). A similar reduction of inhibition was observed when UA was preincubated with 38 nM zinc finger Aart, prior to the addition of 1 pM target DNA in a buffer containing 1% BSA (Figure 3B). Inhibition Is Due to UA Interaction with Protein, not DNA. To investigate the alternative hypothesis that UA-DNA complexes were responsible for the observed inhibition of binding, 32P end-labeled oligonucleotide DNA targets were incubated with various concentrations of UA for 15 min, followed by the addition of purified Aart protein. After an additional 30 min of incubation, the binding activity was assessed by EMSA. No inhibition of DNA binding was observed at any concentration (0, 50, or 200 µM) of UA when preincubated with 4 pM DNA, prior to the addition of 61 nM Aart protein (Figure 4).

Figure 3. EMSA demonstrating the effect of BSA on UA-mediated inhibition of (A) nonzinc finger AP1 or (B) zinc finger Aart binding to target DNA. (A) AP1 (750 nM) was incubated with the indicated concentration of UA for 15 min in binding buffer A ( 1% BSA, prior to 30 min of incubation with appropriate target DNA (14 nM). (B) Aart (38 nM) was incubated with the indicated concentration of UA for 15 min in binding buffer B. The reaction was diluted 1:20 in buffer B/Aart target DNA (1 pM)/15% glycerol/0.12 mg/mL HS-DNA/(1% BSA and incubated for an additional 30 min.

Figure 4. EMSA demonstrating the effect of preincubating target DNA with UA, prior to Aart. Aart target DNA (4 pM) was incubated with the indicated concentration of UA for 15 min in binding buffer B. The reaction was diluted 1:12 in buffer B/15% glycerol/0.12 mg/mL HSDNA/1% BSA, followed by Aart (61 nM) and incubation for an additional 30 min.

Inhibition by UA Must Occur before Formation of DNAProtein Complexes. Oxidizing agents such as NO can inhibit ZFP binding only if exposure occurs before the protein binds DNA (19). This result was interpreted to suggest that the vicinal

Uranium Inhibition of DNA-Binding Proteins

Figure 5. EMSA demonstrating the effect of preincubating target DNA with Aart, prior to UA. Aart (0.61 µM) was incubated with Aart target DNA (1 pM) for 15 min in binding buffer B/15% glycerol/0.12 mg/ mL HS-DNA/1% BSA. One microliter of UA was added to create the indicated final concentration, and incubation continued for an additional 30 min.

cysteines were accessible for oxidation when the protein was unbound but were inaccessible in the bound ZFP-DNA complex. In principle, other types of inhibitory mechanism might also be sensitive to structural rearrangements or changes in the accessibility to critical regions that occur upon protein-DNA binding. To investigate if the mechanism of UA-mediated inhibition required the exposure of UA to unbound protein, 32P end-labeled oligonucleotide DNA targets with purified Aart protein were incubated for 15 min, followed by the addition of various concentrations of UA. After an additional 30 min of incubation, the binding activity was assessed by EMSA. No inhibition of DNA binding was observed at any concentration (0, 50, or 200 µM) of UA when incubated with preformed complexes of DNA and Aart protein (Figure 5).

Discussion Gene expression and DNA repair are mediated by ZFP and other types of DNA-binding proteins. Disruptions to the function of these proteins could result in inappropriate gene expression, which might be toxic to the cell (16). The inability to repair DNA damage could lead to the accumulation of mutations or other types of genomic rearrangements. Most human cancers exhibit genomic instability and an increased mutation rate due to underlying defects in DNA repair (38). In the current study, we investigated the hypothesis that UA might directly act to disrupt ZFP function, thereby suggesting a potential mechanism for the cytotoxic and mutagenic/carcinogenic effects observed for this metal. To test this hypothesis, the DNA-binding activities of two ZFPs, Aart and Sp1, and two non-ZFPs, AP1 and NFκB, were analyzed by EMSA in the presence of UA under various conditions. When incubated with the protein prior to the addition of DNA, UA was found to be a potent inhibitor of DNA binding for both the zinc finger and the non-ZFPs (Figures 1 and 2). The ability of UA to directly inhibit the activity of both ZFPs and nonmetalloproteins was unexpected. Zinc finger DNAbinding proteins require Zn2+ ions for proper folding and function, and many studies have suggested or demonstrated that other divalent cations could displace the zinc and inhibit DNA binding (21-24). Although the uranyl ion is expected to have an ionic charge of 2+ in aqueous buffers (UO22+), the uranium atom has nearly four times the mass of zinc and the complex geometry of the uranyl ion suggests that it would be an unlikely candidate to substitute or displace zinc. Moreover, direct inhibition of the DNA-binding activity of nonmetalloproteins by divalent cations or uranium has not been described. We therefore anticipated that uranium might inhibit the binding of Aart and Sp1 but not AP1 and NF-κB. The observation that UA was able to disrupt the binding of both zinc finger and non-

Chem. Res. Toxicol., Vol. 20, No. 5, 2007 787

ZFPs strongly suggested that the mechanism of inhibition was unrelated to zinc coordination. The observed inhibition was clearly shown to require some property of the uranyl ion, since even a 10-fold higher concentration of SA did not produce inhibition under identical conditions. As is the case with zinc, uranium and arsenic are fairly different elements. Arsenic is considered a soft Lewis acid and forms bonds with soft Lewis bases such as sulfur and nitrogen in cysteines and histidines (39). Uranium is considered a hard metal and bonds primarily to oxygen-containing functional groups (40). However, arsenic was used in these experiments based on evidence suggesting that it could inhibit the activity of ZFP via a direct interaction with vicinal cysteines. Hartwig and co-workers showed that arsenite could inhibit the activity of zinc finger DNA repair protein Fpg (36). They also reported that arsenite could release zinc from the zinc fingers of the DNA repair protein XPA, although earlier studies failed to show an inhibition of XPA-binding activity (37). Simons et al. showed that arsenic could inhibit the DNA-binding activity of the zinc finger-containing glucocorticoid receptor (41). Kitchin and Wallace demonstrated that arsenite bound strongly to the vicinal cysteines of peptides based on the zinc fingers of estrogen receptor, concluding that such interactions could contribute to arsenic toxicity and carcinogenicity via altered peptide/protein structure and enzyme function (42). We were therefore interested to determine if arsenite could similarly inhibit zinc finger binding in our assay. The reason for the observed lack of inhibition, even at 2 mM SA, is not clear at this time. The zinc fingers of Fpg, XPA, and hormone receptors have a different structure than the Cys2-His2 type zinc finger used here, which might have limited access to the cysteines. Alternatively, some component of the reaction buffer may have interfered with the inhibition, as suggested by Hartwig (36). Uranyl ions are known to form complexes with both DNA and protein (43, 44). UA is routinely used to stain DNA for electron microscopy (44) and, when found in blood, is largely associated with serum proteins (3, 45). One obvious potential mechanism was that uranium could bind to the DNA in a manner that interfered with protein binding, either by steric hindrance or by altering DNA conformation. In addition, Yazzie et al. (13) reported that UA could cause strand breaks in purified DNA under certain buffer conditions. We investigated this potential mechanism by preincubating target DNA with a 12500or 50000-fold molar excess of UA prior to the addition of protein. We observed no inhibition of DNA binding under these conditions (Figure 4), suggesting that the mechanism of inhibition was unrelated to a UA-DNA interaction. Our study did not investigate if such UA-DNA interactions were actually occurring. Yazzie et al. (13) postulated that UA might covalently attach to the negatively charged phosphate backbone. It is therefore possible that UA could have formed adducts with the DNA in a manner that did not interfere with binding. Another potential mechanism was that uranium could interact with the proteins in a manner that interfered with binding. Uranium has been shown to interact with oxygen-containing functional groups (40) as well as free sulfhydryl groups (46, 47), which could be present on surface-exposed amino acid side chains or the cystines involved in zinc coordination. This mechanism was initially disfavored because it was expected to require that UA interact with particular residues, and these residues would have to occur in positions in all four proteins such that UA adducts would always interfere with DNA binding. However, we hypothesized that if UA was indeed able to significantly interact with structurally diverse proteins, the

788 Chem. Res. Toxicol., Vol. 20, No. 5, 2007

addition of excess protein should absorb free UA and reduce its inhibitory effects. To investigate this potential mechanism, we introducing the non-DNA-binding protein BSA, a serum protein commonly used in protein-DNA-binding buffers as a nonspecific protein at a concentration of 1% (10 mg/mL or approximately 150 µM). We observed reduced inhibition of DNA binding in the presence of 1% BSA (Figure 3), suggesting that UA has the ability to interact nonspecifically with a verity of proteins. A similar conclusion was reached by Vidaud et al., who reported that uranyl ions could use a variety of binding sites and coordination strategies to bind metallo- and nonmetalloproteins in serum (45). The data also argue against the possibility that the observed effects were due to a general buffer phenomenon, such as an unexpected change in pH upon addition of uranium. A lower pH could allow Fenton-like chemistry with any reducing agents present. However, the observation that preincubation of the DNA with a 50000-fold molar excess of UA did not inhibit protein binding (Figure 4) suggests that Fenton hydrolysis of the DNA did not occur. Also, the data showing that UA could inhibit the binding of protein to DNA in a dose-dependent manner (Figures 1 and 2) while having a reduced effect in the presence of BSA (Figure 3) and no effect on preformed protein-DNA complexes (Figure 5) would not support an interpretation that the effects were due simply to a change in pH. By providing evidence against the involvement of UA-DNA interactions and evidence for significant nonspecific UA-protein interactions, the data strongly support a model that UA-protein interactions either sterically or conformationally cause the inhibition of DNA binding. A steric inhibition mechanism would require UA to physically interfere with amino acid side chains in the protein-DNA interface. In preformed protein-DNA complexes, side chains in the interface would be protected from UA, while UA would still be free to interact with surfaceexposed residues to cause conformation-dependent inhibition. The observation that UA did not disrupt the interaction of preformed protein-DNA complexes (Figure 5) therefore supports a steric mechanism of inhibition. However, our experiments do not provide direct evidence for steric and/or conformational change of the proteins by uranium. Additional biophysical studies, such as by isothermal titration calorimetry or two-dimentional nuclear magnetic resonance, could identify the specific residues involved in the inhibitory interactions, as has been done with other metals (48). In conclusion, this study provides evidence that DU can directly interact with and inhibit the DNA-binding activity of a variety of DNA-binding proteins. Because such proteins include DNA repair and transcription factors, these inhibitory effects could, in principle, account for some of the genotoxic and cytotoxic properties reported for DU. However, the biological implications of these findings are currently unclear. Because UA was shown to interact with proteins nonspecifically, it is uncertain how or if uranium could achieve inhibitory concentrations within cells of the body. Tissues such as kidneys and bones accumulate uranium for months to years (4); therefore, it is conceivable that conditions might exist under which transcription factors or other proteins might be adversely affected by DU. Further investigation will be required to determine the biological significance of UA as a direct inhibitor of DNA-binding proteins. Acknowledgment. This work was supported by ES06694. We thank R. Clark Lantz at the University of Arizona for his assistance in obtaining UA and SA. Aart was kindly provided by Carlos Barbas at The Scripps Research Institute. We also

Hartsock et al.

thank the members of the Segal lab for their careful reading of this manuscript.

References (1) Ha¨rting, F. H., and Hesse, W. (1879) Der Lungenkrebs, die Bergkrankheit in den Schneeberger Gruben, Vol. 30-31, pp 102129, 296-308, and 313-337, Vieteljahrsschrift fu¨r Gerichtliche Medicin und O ¨ ffentliches Sanita¨tswesen. (2) Harley, J. H. (1953) Sampling and measurement of airborne daughter products of radon. Nucleonics 11, 12-15. (3) Eidson, A. F. (1994) The effect of solubility on inhaled uranium compound clearance: A review. Health Phys. 67, 1-14. (4) Brugge, D., de Lemos, J. L., and Oldmixon, B. (2005) Exposure pathways and health effects associated with chemical and radiological toxicity of natural uranium: A review. ReV. EnViron. Health 20, 177193. (5) Tannenbaum, A. (1951) Toxicology of Uranium Compounds, McGrawHill, New York. (6) Brady, H. R., Kone, B. C., Brenner, R. M., and Gullans, S. R. (1989) Early effects of uranyl nitrate on respiration and K+ transport in rabbit proximal tubule. Kidney Int. 36, 27-34. (7) Pavlovic, D., Vlahovic, P., Cvetkovic, T., Savic, V., and Kocic, G. (1998) Is oxidative stress the biochemical basis of uranyl nitrateinduced acute renal failure? Renal Failure 20, 539-542. (8) Miller, A. C., Blakely, W. F., Livengood, D., Whittaker, T., Xu, J., Ejnik, J. W., Hamilton, M. M., Parlette, E., John, T. S., Gerstenberg, H. M., and Hsu, H. (1998) Transformation of human osteoblast cells to the tumorigenic phenotype by depleted uranium-uranyl chloride. EnViron. Health Perspect. 106, 465-471. (9) Wan, B., Fleming, J. T., Schultz, T. W., and Sayler, G. S. (2006) In vitro immune toxicity of depleted uranium: Effects on murine macrophages, CD4+ T cells, and gene expression profiles. EnViron. Health Perspect. 114, 85-91. (10) Dobson, A. W., Lack, A. K., Erikson, K. M., and Aschner, M. (2006) Depleted uranium is not toxic to rat brain endothelial (RBE4) cells. Biol. Trace Elem. Res. 110, 61-72. (11) Lin, R. H., Wu, L. J., Lee, C. H., and Lin-Shiau, S. Y. (1993) Cytogenetic toxicity of uranyl nitrate in Chinese hamster ovary cells. Mutat. Res. 319, 197-203. (12) Stearns, D. M., Yazzie, M., Bradley, A. S., Coryell, V. H., Shelley, J. T., Ashby, A., Asplund, C. S., and Lantz, R. C. (2005) Uranyl acetate induces hprt mutations and uranium-DNA adducts in Chinese hamster ovary EM9 cells. Mutagenesis 20, 417-423. (13) Yazzie, M., Gamble, S. L., Civitello, E. R., and Stearns, D. M. (2003) Uranyl acetate causes DNA single strand breaks in vitro in the presence of ascorbate (vitamin C). Chem. Res. Toxicol. 16, 524-530. (14) Domingo, J. L. (2001) Reproductive and developmental toxicity of natural and depleted uranium: A review. Reprod. Toxicol. 15, 603609. (15) Venter, J. C., et al. (2001) The sequence of the human genome. Science 291, 1304-1351. (16) Ladomery, M., and Dellaire, G. (2002) Multifunctional zinc finger proteins in development and disease. Ann. Hum. Genet. 66, 331342. (17) Frankel, A. D., Berg, J. M., and Pabo, C. O. (1987) Metal-dependent folding of a single zinc finger from transcription factor IIIA. Proc. Natl. Acad. Sci. U.S.A. 84, 4841-4845. (18) Miller, J., McLachlan, A. D., and Klug, A. (1985) Repetitive zincbinding domains in the protein transcription factor IIIA from Xenopus oocytes. EMBO J. 4, 1609-1614. (19) Kroncke, K. D. (2001) Zinc finger proteins as molecular targets for nitric oxide-mediated gene regulation. Antioxid. Redox Signaling 3, 565-575. (20) Giles, N. M., Gutowski, N. J., Giles, G. I., and Jacob, C. (2003) Redox catalysts as sensitisers towards oxidative stress. FEBS Lett. 535, 179182. (21) Hanas, J. S., and Gunn, C. G. (1996) Inhibition of transcription factor IIIA-DNA interactions by xenobiotic metal ions. Nucleic Acids Res. 24, 924-930. (22) Hartwig, A. (2001) Zinc finger proteins as potential targets for toxic metal ions: Differential effects on structure and function. Antioxid. Redox Signaling 3, 625-634. (23) Petering, D. H., Huang, M., Moteki, S., and Shaw, C. F., 3rd (2000) Cadmium and lead interactions with transcription factor IIIA from Xenopus laevis: A model for zinc finger protein reactions with toxic metal ions and metallothionein. Mar. EnViron. Res. 50, 89-92. (24) Watkin, R. D., Nawrot, T., Potts, R. J., and Hart, B. A. (2003) Mechanisms regulating the cadmium-mediated suppression of Sp1 transcription factor activity in alveolar epithelial cells. Toxicology 184, 157-178.

Uranium Inhibition of DNA-Binding Proteins (25) Witkiewicz-Kucharczyk, A., and Bal, W. (2006) Damage of zinc fingers in DNA repair proteins, a novel molecular mechanism in carcinogenesis. Toxicol. Lett. 162, 29-42. (26) Zawia, N. H., Crumpton, T., Brydie, M., Reddy, G. R., and Razmiafshari, M. (2000) Disruption of the zinc finger domain: A common target that underlies many of the effects of lead. Neurotoxicology 21, 1069-1080. (27) Dreier, B., Beerli, R. R., Segal, D. J., Flippin, J. D., and Barbas, C. F., III (2001) Development of zinc finger domains for recognition of the 5′-ANN-3′ family of DNA sequences and their use in the construction of artificial transcription factors. J. Biol. Chem. 276, 29466-29478. (28) Segal, D. J., Beerli, R. R., Blancafort, P., Dreier, B., Effertz, K., Huber, A., Koksch, B., Lund, C. V., Magnenat, L., Valente, D., and Barbas, C. F., III (2003) Evaluation of a modular strategy for the construction of novel polydactyl zinc finger DNA-binding proteins. Biochemistry 42, 2137-2148. (29) Segal, D. J., Crotty, J. W., Bhakta, M. S., Barbas, C. F., 3rd, and Horton, N. C. (2006) Structure of Aart, a designed six-finger zinc finger peptide, bound to DNA. J. Mol. Biol. 363, 405-421. (30) Shi, Y., and Berg, J. M. (1995) A direct comparison of the properties of natural and designed zinc-finger proteins. Chem. Biol. 2, 83-89. (31) Hanas, J. S., Rodgers, J. S., Bantle, J. A., and Cheng, Y. G. (1999) Lead inhibition of DNA-binding mechanism of Cys(2)His(2) zinc finger proteins. Mol. Pharmacol. 56, 982-988. (32) Tabatabai, N. M., Blumenthal, S. S., and Petering, D. H. (2005) Adverse effect of cadmium on binding of transcription factor Sp1 to the GC-rich regions of the mouse sodium-glucose cotransporter 1, SGLT1, promoter. Toxicology 207, 369-382. (33) Ghosh, G., van Duyne, G., Ghosh, S., and Sigler, P. B. (1995) Structure of NF-kappa B p50 homodimer bound to a kappa B site. Nature 373, 303-310. (34) Yang, C., and Frenkel, K. (2002) Arsenic-mediated cellular signal transduction, transcription factor activation, and aberrant gene expression: Implications in carcinogenesis. J. EnViron. Pathol. Toxicol. Oncol. 21, 331-342. (35) Wijeweera, J. B., Gandolfi, A. J., Parrish, A., and Lantz, R. C. (2001) Sodium arsenite enhances AP-1 and NFkappaB DNA binding and induces stress protein expression in precision-cut rat lung slices. Toxicol. Sci. 61, 283-294. (36) Schwerdtle, T., Walter, I., and Hartwig, A. (2003) Arsenite and its biomethylated metabolites interfere with the formation and repair of stable BPDE-induced DNA adducts in human cells and impair XPAzf and Fpg. DNA Repair (Amsterdam) 2, 1449-1463.

Chem. Res. Toxicol., Vol. 20, No. 5, 2007 789 (37) Asmuss, M., Mullenders, L. H., Eker, A., and Hartwig, A. (2000) Differential effects of toxic metal compounds on the activities of Fpg and XPA, two zinc finger proteins involved in DNA repair. Carcinogenesis 21, 2097-2104. (38) Kennedy, R. D., and D’Andrea, A. D. (2006) DNA repair pathways in clinical practice: Lessons from pediatric cancer susceptibility syndromes. J. Clin. Oncol. 24, 3799-3808. (39) Rensing, C., Ghosh, M., and Rosen, B. P. (1999) Families of softmetal-ion-transporting ATPases. J. Bacteriol. 181, 5891-5897. (40) Pearson, R. (1963) Hard and soft acids and bases. J. Am. Chem. Soc. 85, 3533-3539. (41) Simons, S. S., Jr., Chakraborti, P. K., and Cavanaugh, A. H. (1990) Arsenite and cadmium(II) as probes of glucocorticoid receptor structure and function. J. Biol. Chem. 265, 1938-1945. (42) Kitchin, K. T., and Wallace, K. (2005) Arsenite binding to synthetic peptides based on the Zn finger region and the estrogen binding region of the human estrogen receptor-alpha. Toxicol. Appl. Pharmacol. 206, 66-72. (43) Lombardi, L., Prenna, G., Okolicsanyi, L., and Gautier, A. (1971) Electron staining with uranyl acetate. Possible role of free amino groups. J. Histochem. Cytochem. 19, 161-168. (44) Gray, C. W. (2001) Electron microscopy of protein-nucleic acid complexes. Uniform spreading of flexible complexes, staining with a uniform thin layer of uranyl acetate, and determining helix handedness. Methods Mol. Biol. 148, 579-587. (45) Vidaud, C., Dedieu, A., Basset, C., Plantevin, S., Dany, I., Pible, O., and Quemeneur, E. (2005) Screening of human serum proteins for uranium binding. Chem. Res. Toxicol. 18, 946-953. (46) Gibson, J. K. (1999) Gas-phase transuranium chemistry: Reactions of actinide ions with alcohols and thiols. J. Mass Spectrom. 34, 11661177. (47) Kleinman, J. G., McNeil, J. S., Schwartz, J. H., Hamburger, R. J., and Flamenbaum, W. (1977) Effect of dithiothreitol on mercuric chloride- and uranyl nitrate-induced acute renal failure in the rat. Kidney Int. 12, 115-121. (48) Razmiafshari, M., Kao, J., d’Avignon, A., and Zawia, N. H. (2001) NMR identification of heavy metal-binding sites in a synthetic zinc finger peptide: Toxicological implications for the interactions of xenobiotic metals with zinc finger proteins. Toxicol. Appl. Pharmacol. 172, 1-10.

TX600347K