Binding of Copper and Cisplatin to Atox1 Is ... - ACS Publications

Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Binding of Copper and Cisplatin to Atox1 is Mediated by Glutathione through the Formation of Metal-Sulfur Clusters. Nataliya V. Dolgova, Corey Yu, John P. Cvitkovic, Miroslav Hodak, Kurt Nienaber, Kelly Lynn Summers, Julien Cotelesage, Jerzy Bernholc, George A. Kaminski, Ingrid J. Pickering, Graham N. George, and Oleg Y. Dmitriev Biochemistry, Just Accepted Manuscript • Publication Date (Web): 26 May 2017 Downloaded from http://pubs.acs.org on May 27, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Binding of Copper and Cisplatin to Atox1 is Mediated by Glutathione through the Formation of Metal-Sulfur Clusters. Natalia V. Dolgova1,2, Corey Yu1, John P. Cvitkovic3, Miroslav Hodak4, Kurt H. Nienaber 2, Kelly L. Summers5, Julien J. H. Cotelesage 2, Jerzy Bernholc4, George A. Kaminski3, Ingrid J. Pickering 2,5, Graham N. George 2,5 and Oleg Y. Dmitriev1* 1

Department of Biochemistry, University of Saskatchewan, Saskatoon, SK, Canada, S7N 5E5; 2Department of Geological Sciences, University of Saskatchewan, Saskatoon, SK, Canada, S7N 5E2; 3Department of Chemistry and Biochemistry, Worcester Polytechnic Institute, MA, USA, MA 01609; 4Department of Physics, North Carolina State University, Raleigh, NC, USA, NC 27695-7518; 5Department of Chemistry, University of Saskatchewan, Saskatoon, SK, Canada, S7N 5C9 KEYWORDS: Atox1, copper transport, copper-glutathione polymers, cisplatin ABSTRACT: Copper is an essential nutrient required for many biological processes involved in primary metabolism, but free copper is toxic due to its ability to catalyze formation of free radicals. To prevent toxic effects, in the cell copper is bound to proteins and low molecular weight compounds, such as glutathione, at all times. The widely used chemotherapy agent cisplatin is known to bind to copper-transporting proteins, including copper chaperone Atox1. Cisplatin interactions with Atox1 and other copper transporters are linked to cancer resistance to platinum-based chemotherapy. Here we analyze the binding of copper and cisplatin to Atox1 in the presence of glutathione under redox conditions that mimic intracellular environment. We show that copper(I) and glutathione form large polymers with a molecular mass of approximately 8 kDa, which can transfer copper to Atox1. Cisplatin also can form polymers with glutathione, albeit at a slower rate. Analysis of simultaneous binding of copper and cisplatin to Atox1 under physiological conditions shows that both metals are bound to the protein through copper-sulfur-platinum bridges.

INTRODUCTION Copper is an essential micronutrient, required for many key biological processes, including oxidative phosphorylation, tissue formation, hormone and melanin production, neurotransmission, protection against reactive oxygen species (ROS) and many others1-5. Deregulation of copper distribution and metabolism plays a role in the development of many severe diseases, such as Wilson disease, Menkes disease, Alzheimer’s, Parkinson’s, amyotrophic lateral sclerosis (ALS), and cancer2,6,7,8. Due to its ability to cause oxidative damage through production of reactive oxygen species, copper is not found inside the cells in the free state, but it is always bound to the proteins and low molecular weight compounds such as glutathione9,10. Copper enters the cell through a passive transporter hCtr1 located in the plasma membrane. Atox1, a small ferredoxinlike chaperone protein, accepts copper from hCtr1 and delivers it to ATP-driven transmembrane copper transporters ATP7B and ATP7A for incorporation into the proteins or removal of excess copper from the cell. All these copper transport proteins have been shown to be involved in cell resistance to cisplatin11-13, a platinum-based chemotherapy agent, effective against many solid tumors including ovarian, testicular, cervical cancer, melanoma and others14. Cisplatin binds to the highly conserved copper binding CxxC motif in ATP7B15,16, ATP7A17 and Atox118, and to the methionine-rich motifs presumed to be involved in copper transport in hCtr119,20. Moreover, copper chaperone Atox1 was shown to transfer cisplatin to the metal-binding domain of ATP7B15 and ATP7A17, and Wilson disease ATPase (ATP7B) may transport cisplatin across the membrane21. Deletion of Atox1

leads to resistance to cisplatin22,23. These observations suggest that cisplatin can be transported through the cell along the copper transfer pathways. DNA is considered to be the main pharmacological target of cisplatin: the drug binds to guanine causing intra- and interstrand cross-links, which impair DNA transcription and replication and cause apoptosis24. Interestingly, Atox1 has been shown to deliver copper into the nucleus25 and act as a transcription factor26. This suggests that cisplatin too could be delivered to the nucleus by Atox1, with intriguing implications for drug pharmacology15,27. If platinum and copper bind to the same site in Atox1, one might expect a competitive relationship between the two metals. However, Atox1 can bind both copper and cisplatin simultaneously17,28,29. Moreover, in the presence of glutathione cisplatin was shown to react with Atox1-Cu faster than with apo-Atox117. Binding of copper and cisplatin to Atox1 in the experiments published so far was carried out either in anaerobic conditions without the presence of any reducing agents, or in the presence of low concentrations of dithiothreitol (DTT). The goal of our study was to analyze binding of copper and cisplatin to Atox1 under physiological redox conditions, and to determine the chemical structure of the Atox1-Cu-Pt complex. We modeled the in-cell redox environment by using partially oxidized glutathione with the redox potential of the reduced/oxidized glutathione pair (GSH/GSSG) within the range found in the cytosol of mammalian cells30. Under these conditions, Atox1-Cu reacts with cisplatin, forming complexes in which copper and platinum are bound to the protein through metal-sulfur clusters, which include glutathione. This

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

work establishes glutathione as an important player in metal transfer reactions in the cell. We show that glutathione forms polymers incorporating copper and platinum, and these polymers can transfer metals to Atox1 and possibly to the other metal-binding proteins in the cell. Simultaneous binding of copper and platinum to Atox1 and the complex metal transfer reactions involving glutathione provide a new perspective on the role of copper transport proteins and copper status of the cell in cancer resistance to cisplatin.

Page 2 of 15

To avoid non-linearity of the detector due to the high count rates, filters containing zinc and nickel for platinum and copper respectively were placed between the sample and the detector, to preferentially absorb scattered radiation, with Soller slits placed between the filter and the detector to reduce filter fluorescence. The monochromator was energy-calibrated with reference to the lowest energy LIII inflection point of the platinum metal foil, which was assumed to be 11563.0 eV, or with reference to the lowest energy K inflection point of the copper metal foil, assumed to be 8980.3 eV. The EXAFS (extended X-ray absorption fine structure) oscillations χ(k) were analyzed by curve-fitting using the EXAFSPAK suite of computer programs (http://ssrl.slac.stanford.edu/exafspak.html) employing ab initio theoretical phase and amplitude functions generated with the program FEFF 7.0. No smoothing, Fourier filtering or related manipulations were performed on the data. Fourier transforms were phasecorrected for Pt–S or Cu–S scattering. Molecular structures were plotted using Materials Studio v7 (Accelrys Software Inc).To simulate EXAFS based on the structural models, distances between atoms were measured and used as parameters in the OPT program from EXAFSPAK suite together with coordination numbers determined from the models. Size Exclusion Chromatography Size exclusion chromatography (SEC) was carried out using Superdex 75 16/60 column, in a buffer containing 50 mM HEPES, pH 7.4, 50 mM NaCl, 0.95 mM GSH, 0.05 mM GSSG. Samples were prepared as described above for NMR and XAS, except for omitting the last step of concentrating the samples. Nuclear Magnetic Resonance NMR experiments were performed on a 600 MHz Bruker spectrometer. Combined chemical shift change was calculated as [(Δδ2NH +Δδ2N/25)/2]½, where ΔδNH and ΔδN are the chemical shift changes of the amide proton and nitrogen respectively. Data were processed and analyzed using NMRPipe56. Samples were prepared the same way as for the EXAFS experiments. Before recording the spectra, 1 mM DSS (sodium 2,2-dimethyl-2-silapentane- 5-sulfonate) and 5% (v/v) 2H2O were added to the samples.

METHODS Protein expression and purification Atox1 was expressed in Escherichia coli BL21(DE3) cells as a fusion with an intein and the chitin-binding domain using pTYB12 vector (New England Biolabs), as described previously55. Protein was purified on the chitin column (New England Biolabs), and the affinity tag was removed by incubation of the fusion protein bound to the column with DTT, which induced intein cleavage and released Atox1, while chitinbinding domain remained bound to the column. Purified protein was dialyzed twice against 50 mM HEPES, pH 7.4, 50 mM NaCl, 0.6 mM TCEP. Protein concentration was measured by BCA assay (Pierce). Sample preparation for NMR and X-ray absorption (XAS) experiments All the samples for NMR and XAS were prepared in a buffer containing 50 mM HEPES, pH 7.4, 50 mM NaCl, 9.5 mM GSH and 0.5 mM GSSG. To prevent protein oxidation as a result of copper disproportionation, copper was added to the buffer with a reducing agent first, and the protein was added last. A 9 mM stock solution of CuCl (Merck) was prepared fresh prior to experiment with degassed 1 M NaCl containing 0.01 M HCl. A 4.5 mM cisplatin (Sigma) stock solution was also prepared fresh immediately prior to experiment using deionized water. Prior to the experiment, Atox1 was reduced by adding 5 mM TCEP, then TCEP was removed by dialysis against a buffer containing 50mM HEPES, pH 7.4, and 50 mM NaCl under anaerobic conditions. Copper and/or cisplatin were added to 50 µM Atox1 in the presence of GSH/GSSG at the equimolar metal: protein ratio. After incubating the protein at 20°C with copper for 1 hour and/or cisplatin overnight, samples were concentrated to 200 µM protein. Free copper and cisplatin was removed by buffer exchange using PD-10 desalting columns (GE Healthcare) in 50 mM HEPES, pH 7.4, 50 mM NaCl, 9.5 mM GSH and 0.5 mM GSSG. Control samples were prepared as described above, but without Atox1. Samples for XAS were mixed with glycerol at a final concentration of 20%, placed into sample holders and immediately frozen in liquid nitrogen. X-ray absorption spectroscopy Copper and platinum X-ray absorption spectra were collected on beamline 7–3 at Stanford Synchrotron Radiation Lightsource with SPEAR storage ring operating at 3 GeV. During measurements, samples were maintained at a temperature of approximately 10 K in a flowing liquid helium cryostat. X-ray absorption was measured by monitoring the X-ray fluorescence excitation spectrum using a 30-element Ge array detector.

COMPUTATIONAL METHODS Structure Preparation In order to expedite molecular mechanics calculations, Atox1 was modeled as residues 9 to 19 of the Cu(I)-loaded crystal structure34 (PBD ID: 1FEE) with the addition of C- and N-terminal capping groups. Use of this truncated model is justified as it includes all the residues showing significant change in NMR chemical shift upon binding of Cu(I) and cisplatin to apo-Atox1 in solution, with the exception of K60 (Fig. 4D), and the solution structure (PDB ID: 1TL4) and crystal structure (PBD ID: 1FEE) of Cu(I)-loaded Atox1 show no major structural differences with an RMSD of ca. 0.9 Å—comparable to the 0.95 Å RMSD of the solution-phase average50. Hydrogen atoms were added and the protonation states of ionizable residues were set to their most stable forms at pH 7 where applicable using the pepz utility of the MCPRO software package59. Cysteine residues involved in copper binding were also set to their deprotonated thiolate forms. The pepz utility was also used to build the naïve starting geometry of

2 ACS Paragon Plus Environment

Page 3 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

glutathione as a linear peptide with idealized bond lengths and angles. Monte Carlo Molecular Mechanics Calculations All nonpolarizable calculations were performed with the MCPRO software package within the OPLS force field framework57. The standard OPLS parameter set58,59 was employed for all atoms except for the thiolate sidechain of cysteine, Cu(I), and Pt(II) which we have previously parameterized6062 . The TIP3P water model57 was used in all aqueous-phase simulations. Protein complexes were built and screened for stability as follows: first, the desired complex was assembled from the truncated Atox1 model, glutathione, and metals ions, then a gas-phase optimization performed with only the intermolecular and cysteine sidechain degrees of freedom free. Next, the complex was solvated in an appropriately sized water droplet with a 1 kcal/mol/Å2 half harmonic restoring potential with equilibrium distance equal to the radius of the droplet. With all solute degrees of freedom fixed, the water droplet was allowed to equilibrate for ca. 1×106 MC steps. Next, large (>150 kcal/mol/Å2) harmonic potentials were applied between metal and coordinated sulfur atoms with equilibrium distances equal to the optimized gas phase distances and the system was allowed to equilibrate for at least an additional 1×106 steps. During this “relaxation” step all intermolecular, internal glutathione, and Atox1 cysteine sidechain degrees of freedom were allowed to vary. Finally, the metal ligand harmonic potentials were removed and at least an additional 2×106 MC steps performed while monitoring key metal-sulfur distances. Complexes in which one or more of these distances diverged, indicating complex dissociation or fragmentation, were deemed unstable and discarded. Stable complexes were subject to an additional 2×106 steps of equilibration followed by 8×106 configurations of averaging. Each averaging run was broken up into batches of 2×105 steps and the variance calculated by the batch means method. Polarizable calculations were performed with our POSSIM software suite utilizing the second order polarizable POSSIM force field63,64. The standard POSSIM parameters and our polarizable POSSIM water model63,65 were employed except for the thiolate sidechain of cysteine and Pt(II) which we have recently parameterized61. Parameters for the Cu(I) ion were obtained in the same manner as for our previous generation polarizable force field62. Due to the combined size of these protein complexes and solvent (>1500 atoms), polarizable calculations were only performed on structures that had passed the nonpolarizable screening process. The equilibrated nonpolarizable complexes and water droplet were used as starting geometries after truncating the glutathione molecules to methyl-capped cysteine residues and removing extraneous water molecules.

which Kohn-Sham equations are solved in real-space and the multigrid technique is used to accelerate convergence of the ground state wavefunctions. The DFT calculations started from MM structures with both Atox1 and glutathione shortened for computational efficiency. Specifically, Atox1 molecules were represented by CGGC fragments with backbone extended on each side until the next α-carbons, where CH3 terminations were applied. The same ends were applied to glutathione, a GCE tripeptide, at GLU and GLY α-carbons. These fragments were fully solvated and structurally optimized within the OF/KS DFT method. A grid with spacing of 0.32 Bohr was used to represent the wavefunctions, corresponding to a kinetic energy cutoff of 50 Ry. Ultrasoft pseudopotentials and generalized gradient approximation in the PBE form were employed.

RESULTS AND DISCUSSION Oligomeric state of Atox1 and glutathione in the presence of copper(I) and cisplatin The goal of our work was to study copper and platinum interactions with Atox1 under physiological redox conditions. To this end, reactions of Atox1 with equimolar concentrations of copper and cisplatin were carried out in the presence of 9.5mM reduced glutathione (GSH) and 0.5mM of oxidized glutathione (GSSG) at pH 7.4. The redox potential of the GSH/GSSG pair under these conditions is approximately -230 mV, which is in the range reported for the healthy proliferating mammalian cells31. First, we analyzed the effect of copper and cisplatin on the oligomerization state of Atox1 and glutathione by size-exclusion chromatography. Atox1 can exist both as a monomer and as a dimer, depending on the experimental conditions. Apo-Atox1 was often found to exist as a monomer, while Atox1-Cu was usually reported to be a dimer32-34. Glutathione has been reported to cause dimerization of the copper-bound Atx1, the yeast homolog of Atox135. Size exclusion chromatography profiles of apo-Atox1, Atox1-Cu and Atox1-Pt all show the presence of two peaks, one containing Atox1 (Fig. 1 A, peak 1a, Fig. S1), and the other corresponding to the free glutathione (peak 2a). Atox1 in the presence of partially oxidized glutathione migrates at the apparent molecular mass of 12.3 kDa corresponding to a dimer. Atox1-Cu and Atox1-Pt migrate as dimers as well, although Atox1-Pt shows an increase in the apparent molecular mass of about 1-1.5 kDa (Fig. S1, E). Size exclusion chromatography profile of the mixture of glutathione and copper(I) without Atox1 shows the presence of large polymeric structures with apparent molecular weight of ~8 kDa (Fig 1 B, peak 1b) along with the glutathione monomers and dimers (peak 2b). Glutathione forms polymeric complexes with cisplatin as well, but at a much slower rate than with copper(I). While glutathione polymers with copper form within seconds, platinum-glutathione polymers could be observed only after an incubation time of several hours. Glutathione polymers also form when copper and cisplatin are present together. By comparison, only the low molecular weight component (peak 2b) is observed with metal free glutathione (Fig. 1 B), indicating that glutathione polymerization is caused by copper or platinum, likely through the formation of metal-glutathione complexes. To analyze chemical structure of the metal-glutathione complexes we used X-ray absorption spectroscopy. Near-

Density functional Theory Simulations DFT-level calculations were performed by using a hybrid orbital-free/Kohn-Sham DFT method66, which enables the use of explicit solvent in DFT simulations. In this approach, the chemically active parts of the system, including the first solvation shells, are treated at a Kohn-Sham (KS) DFT level, while an approximate orbital-free (OF) DFT is used for the remaining solvent molecules. The implementation of the hybrid OF/KS DFT approach is based on the RMG code67-69, in


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 15

similar Cu∙∙∙Cu and Cu−S distances, as determined by EXAFS40(Fig. 1 F). To our knowledge, this is the first report of copper(I)-glutathione polymers as large as 8 kDa forming under the redox conditions and glutathione concentrations found in the cytoplasm of the healthy mammalian cells, although some other characteristics of these polymers are consistent with previous reports41,42. At a glutathione: copper ratio of 5:1 small glutathione complexes of 1.35 kDa, apparently tetramers, have been reported with three sulfur ligands per copper atom and Cu∙∙∙Cu distances close to 2.69 Å42. The higher glutathione to copper ratio (50:1 – 200:1 in our experiments) appears to favor formation of larger polymers, but other factors such as pH, ionic strength and redox potential can affect the polymer size as well. At this metal to glutathione ratio, only a small fraction of glutathione can be incorporated into the clusters, while most of it remains in monomeric or dimeric state.

edge X-ray absorption spectra are sensitive to the local geometric and electronic structure of the absorbing atom; those of copper have characteristic appearance, which depends on the coordinating ligands and the geometry of the complex36-39. Comparison of the K-edge spectra of Cu-glutathione complexes to the spectra of model compounds with digonal and trigonal coordination of copper suggests that copper is bound to three sulfur ligands (Fig. 1 C). The Fourier transform of copper K-edge extended X-ray absorption fine structure (EXAFS) of copper(I)-glutathione mixture (Fig. 1 D) shows two prominent distinct peaks: the first corresponds to three Cu−S ligands at 2.267(2) Å, while the second is best fitted as 2 copper atoms at 2.723(3) Å (Fig. 1 E, Table 1, Table S1). From these data, we concluded that Cu-glutathione polymers consist of copper-sulfur clusters with copper atoms in close proximity to each other. The local arrangement of copper and sulfur atoms in these clusters may be similar to the previously described Cu4S6 clusters in yeast metallothionein, which have

Figure 1. Characterization of Atox1 and glutathione complexes with copper and cisplatin. A: SEC chromatography profiles of Atox1 with partially oxidized glutathione with and without copper (Cu) and/or cisplatin (Pt). B: SEC chromatography profiles of partially oxidized glutathione with and without copper and/or cisplatin. Note that the fraction of glutathione polymers cannot be directly determined from the peak ratio due to a large difference in the extinction coefficients of the metal-free and copper-bound glutathione. C: Copper K-edge spectra of copper(I)-glutathione polymers (black) and Atox1-Cu in the presence of glutathione (red), shown together with model compounds for trigonal (blue, Cu(SR)3) and digonal (purple, Cu(SR)2) copper(I). D, E: Copper EXAFS spectrum (k3-weighted) (D) and the corresponding Fourier transform (E) of copper(I)-glutathione polymers. Experimental data is shown as solid line, fitted data as dashed line. F: A model of Cu4S6 cluster37 with copper and sulfur atoms displayed in orange and yellow, respectively.

Table 1: Cu K-edge EXAFS curve fitting parameters for copper-glutathione polymers, and Atox1 complexes with copper and cisplatina Sample

S

Cu

F

N

R

σ2

N

R

σ2

Cu-GSH/GSSG

3

2.267(2)

0.0043(2)

2

2.723(3)

0.0063(3)

0.3819

Atox1-Cu-GSH/GSSG

3

2.251(2)

0.0043(2)

1

2.702(9)

0.0076(9)

0.4806

Atox1-Cu-Pt-GSH/GSSG

3

2.269(1)

0.0054(1)

1

2.694(2)

0.0033(2)

0.2681

σ2,

a Table

(Å2);

columns are: N, coordination number; R, interatomic distance, (Å); Debye-Waller factor, F, goodness of fit parameter. The number in parentheses is three times the estimated standard deviation in the last digit, determined from the diagonal elements of the covariance matrix. Threshold energy shift ΔE0 (eV) was calculated for all samples and its average value of -13.4 eV was used as fixed parameter for all refinements.


Page 5 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Binding of copper(I) to Atox1 in the presence of glutathione Copper(I) has limited solubility in water and is easily oxidized to copper(II). Three methods have been used to stabilize copper(I) for protein binding assays: (i) adding (CH3CN)4Cu(I)]PF6 complex in the presence of acetonitrile; (ii) reducing copper(II) with sodium sulfite or DTT; and (iii) mixing copper(I) chloride with glutathione prior to adding it to the protein. We used the last method as it does not require organic solvents and does not generate non-biological oxidation products. The binding affinity of glutathione to copper is about 500 times lower than of Atox143, so copper should be transferred from glutathione to Atox1. To confirm this prediction, we analyzed copper binding to 15N-labeled Atox1 by NMR. Addition of stoichiometric amounts of copper to Atox1 in the presence of a large excess of glutathione results in complete conversion of apo-Atox1 to Atox1-Cu (Fig. 2 A), as monitored by the chemical shifts of the residues in the loop containing the copper binding site Cys12xxCys15 and residues Thr58-Val62 at the C-terminus of the protein located in spatial proximity to the CxxC motif. Consistent with the NMR

data, comparison of the size exclusion chromatography profiles of the Cu-glutathione (Fig. 1 B) and Atox1 with Cu-glutathione (Fig. 1 A) clearly demonstrates transfer of copper from glutathione polymers to the protein. Only the peaks corresponding to Atox1-Cu (peak 1a) and free glutathione (peak 2a), but not the Cu-glutathione polymers (peak 1b), are observed upon adding Cu-glutathione to Atox1 (Fig. 1 A-B). The local structure of the copper-binding site in Atox1 was characterized by X-ray absorption spectroscopy. Like Cu-glutathione, the near-edge spectrum of copper bound to Atox1 in the presence of glutathione is quite similar to the spectrum of a model compound with copper bound to three sulfur ligands (CuSR3, Fig. 1 C), and very different from CuSR2, suggesting that copper in Atox1-Cu is coordinated by three sulfur atoms. Interestingly, while the near edge spectra of both Atox1-Cu and Cu-glutathione are characteristic of the three-coordinate copper(I), they do show distinctive features that may reflect differences in the geometry of the CuS3 complex. In agreement with previous work on Atox144 and Atx135, NMR data shows that two sulfurs originate from Cys12 and Cys15 in Atox1, with the third sulfur likely from a cysteine in glutathione. .

Figure 2. Copper binding to Atox1 in the presence of glutathione. A: overlay of 1H,15N-HSQC spectra of apo-Atox1 (black) and Atox1-Cu (red), with residues showing significant chemical shift changes labeled; B,C: copper EXAFS (B) and corresponding Fourier transform (C) of Atox1-Cu (experimental data is shown by the solid line, fitted data is shown by the dashed line); D: comparison of Fourier transforms of EXAFS of copper(I)-glutathione (black) and Atox1-Cu in the presence of glutathione (red).


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Copper EXAFS (Fig. 2 B) shows that copper in Atox1-CuGSH/GSSG sample is bound to 3 sulfur atoms at the average distance of 2.251(2) Å, compared to the average Cu∙∙∙S distance of 2.267(2) Å in Cu-GSH/GSSG polymers (Table 1). This difference is statistically significant and may reflect a change in copper coordination geometry upon transfer from glutathione to Atox1. The Fourier transform of the Atox1-Cu EXAFS spectrum (Fig. 2 C) also contains a peak corresponding to copper-copper interactions at a distance of 2.702(9) Å. This peak is less pronounced in Atox1-Cu-GSH/GSSG sample compared to Cu-GSH/GSSG (Fig. 2 D), another indication of copper transfer from the glutathione polymers to the protein

Page 6 of 15

Fitting of the Atox1-Cu-EXAFS spectrum indicates the presence of only one copper atom in close proximity to the analyzed copper, suggesting binding of two Cu atoms per Atox1 dimer. The (Atox1)2Cu2 stoichiometry is also consistent with the complete copper transfer from glutathione to Atox1 indicated by SEC (Fig 1 A, B) and NMR (Fig. 2 A) data. Yeast copper chaperone Atx1 was previously shown to bind two copper atoms per dimer as well35.

Figure 3. Binding of cisplatin to Atox1 in the presence of glutathione. A: overlay of 1H,15N- HSQC spectra of apo-Atox1 (black) and Atox1Pt (red); B: platinum LIII-edge spectra of Atox1-Pt in the presence of glutathione, cisplatin (dry powder) and cis-dichloro-bis(diethyl-sulfide)platinum(II) (dry powder); C,D: platinum EXAFS spectrum (C) and the corresponding Fourier transform (D) of Atox1-Pt in the presence of glutathione (solid line – experimental data, dashed line - data fit).

Table 2: Pt LIII-edge EXAFS Curve Fitting Parameters for Atox1 complex with cisplatina. sample

S N

R

σ2

Atox1-Pt-GSH/GSSG

4

2.304(1)

0.0047(1)

Atox1-Cu-Pt-GSH/GSSG

4

2.300(1)

0.2588

0.0046(1) σ2,

a Table

F

0.2910 (Å2);

columns are: N, coordination number; R, interatomic distance, (Å); Debye-Waller factor, F, goodness of fit parameter. The number in parentheses is three times the estimated standard deviation in the last digit, determined from the diagonal elements of the covariance matrix. Threshold energy shift ΔE0 (eV) was calculated for all samples and its average value of -8.6 eV was used as fixed parameter for all refinements.


Page 7 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Copper has been previously proposed to bind to glutathione upon entering the cell through Ctr145,46. Glutathione, in turn, was shown to transfer copper to metallothioneins47 and Cu,Zn-SOD48. Our data suggest that stable copper(I)-glutathione polymers can form in the cytosol of mammalian cells and subsequently transfer copper to Atox1, and possibly other copper-binding proteins. Binding of cisplatin to Atox1 in the presence of glutathione Consistent with previous work15,17,18,49 conducted with various reducing agents, NMR shows that, at a physiological GSH/GSSG ratio, cisplatin binds at the same site as copper, with Met10, Cys12 and Cys15 showing the largest chemical shift changes (Fig. 3 A). The reaction of cisplatin with Atox1 at the stoichiometric ratio is much slower than with copper and requires several hours to complete. Comparison of SEC profiles of cisplatin-GSH/GSSG and Atox1-Pt-GSH/GSSG (Fig. 1 A, B) confirms that transfer of platinum from glutathione to Atox1 results in dismantling of glutathione polymers. This data indicates that Atox1 has a higher affinity for cisplatin than glutathione and can extract cisplatin from glutathione polymers in the cytosol. The platinum near-edge spectrum of Atox1-Pt in the presence of glutathione differs both from the spectrum of cisplatin and from the spectrum of the model compound cis-dichlorobis(diethyl-sulfide)platinum(II), suggesting substitution of both amino and chloride ligands of platinum upon cisplatin reaction with Atox1 (Fig. 3 B). Platinum EXAFS (Fig. 3 C) Fourier transform of Atox1-Pt has only one major peak, which is best fitted as four Pt∙∙∙S bonds (Fig. 3 D, Table 2 and Table S2). By comparison, Fourier Transform of cisplatin has two distinct peaks corresponding to nitrogen and chloride ligands (Fig. S2). Attempts to fit the experimental Pt EXAFS spectrum of Atox1-Pt with nitrogen ligands resulted in unrealistic values of Debye-Waller factors for both nitrogen and sulfur atoms (Table S2). Therefore, both near edge and EXAFS spectra of Atox1-Pt indicate the absence of nitrogen ligands in the platinum bound to Atox1 in the presence of glutathione. This finding is in disagreement with the previous report of platinum retaining nitrogen ligands upon binding to Atox1 in the presence of glutathione17. However, that result was based on NMR measurements with 15N-labelled cisplatin and on the molecular weight of the Atox1-cisplatin adduct determined by mass-spectrometry, rather than on the direct detection of platinum ligands by EXAFS. In summary, our X-ray absorption and NMR data taken together suggest that platinum is coordinated by cysteines from the Atox1 CxxC motif and glutathione molecules. Binding of platinum to copper-loaded Atox1 Previous studies showed that, in the presence of DTT, cisplatin binds both to the metal-free and copper-loaded Atox128,29,49. In the presence of physiological concentrations of reduced and oxidized glutathione, binding of copper and cisplatin to Atox1 produces different patterns of chemical shift changes in 1H,15N-HSQC spectra, making it possible to distinguish copper- and platinum-bound forms of the protein by NMR (Figs. 2 A and 3 A). Binding of cisplatin to Atox1-

Cu produces an Atox1-Cu-Pt adduct (Fig. 4 A, D) that is distinct from both Atox1-Cu and Atox1-Pt forms: the HSQC spectrum of 15N-Atox1-Cu-Pt (red, Fig. 4 A-C) does not match either 15N-Atox1-Cu (black, Fig. 4 B) or 15N-Atox1-Pt (black, Fig. 4C). Comparison of the amino acid residues perturbed by the simultaneous binding of copper and cisplatin (Atox1-Cu-Pt, Fig. 4F) to the effects of copper (Atox1-Cu, Fig. 4 D) and cisplatin (Atox1-Pt, Fig. 4 E) added separately, shows that the individual chemical shift signatures of both bound copper and cisplatin are present in the Atox1-Cu-Pt spectrum. This fact suggests that platinum binds to the same site in apo-Atox1 and Atox1-Cu. The third cysteine in Atox1, located outside of the CxxC motif (Cys41) did not show any chemical change compared to apo-Atox1, and therefore is not involved in cisplatin binding to Atox1-Cu. Comparison of the size exclusion chromatography profiles of Cu-Pt-glutathione samples with and without Atox1 (Fig. 1 A, B) indicates that both copper and platinum are transferred to the protein, because there are no detectable metal-glutathione polymers left in the presence of Atox1. Taken together, NMR and SEC results provide evidence for simultaneous binding of copper and platinum to the copper-binding site of Atox1. These results support data obtained previously by other groups17,28,29 and show that both metals are able to bind to the protein independent of the type of reducing agent or copper source used in the experiment. The binding of platinum to copper-loaded Atox1 (Atox1Cu-Pt) causes small changes in the near edge copper X-ray absorption spectra. The most pronounced change is observed in the region of the spectrum sensitive to the angles between Cu-S bonds in the trigonal configuration37 (Fig. 5 A), suggesting a different arrangement of bonds to copper when platinum is present. Platinum LIII-edge spectra of the Atox1-Pt and Atox1-Cu-Pt are almost identical (Fig. 5 B), and show no evidence of Pt-Cu interaction, leading us to the conclusion that copper and platinum do not form a direct bond with each other as has been proposed previously28, but rather connect through the sulfur bridges. Fitting of the platinum EXAFS of Atox1-Cu-Pt sample shows that platinum is bound to 4 sulfur atoms with the bond lengths of 2.300(1) Å (Fig. 5 E, Table 2). Pt∙∙∙S bond length in Atox1-Cu-Pt sample is very similar to the average Pt∙∙∙S distance in Atox1-Pt (Table 2), suggesting that rigid square planar conformation of platinum does not undergo any major change. Fitting of the copper EXAFS shows that in Atox1-Cu-Pt, as in Atox1-Cu and in Cu-GSH/GSSG polymers, copper is bound to three sulfur atoms (Fig. 5 C, Table 1). The average length of Cu∙∙∙S bond in Atox1-Cu-Pt samples is larger than in Atox1-Cu (2.269(1) Å vs 2.251(2) Å) and is closer to the Cu∙∙∙S bond length in Cu-GSH/GSSG polymers (Fig. 5 D, Table 1). An outer shell peak, which is intermediate in intensity between that of the Atox1-Cu-GSH/GSSG sample and the copper-glutathione polymer (Fig. 5 D), is best fitted as a Cu∙∙∙Cu interaction (Table 1). There are no peaks that could be attributed to Cu∙∙∙Pt interaction. In summary, our data indicate that platinum is bound to Atox1 in the copper-binding site through sulfur bridges formed with cysteine residues from protein and glutathione.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 15

Figure 4. Binding of cisplatin to Atox1-Cu in the presence of glutathione. A-C : overlays of 1H,15N-HSQC spectra of Atox1-Cu-Pt (red) and (A) apo-Atox1 (black), (B) Atox1-Cu (black), and (C) Atox1-Pt (black), with peaks showing largest chemical shift differences labeled in black, and C41 labeled in blue; D-F: Combined chemical shift change, relative to apo-Atox1, as a function of residue number, caused by (D) copper, (E) cisplatin and (F) copper followed by cisplatin treatment. Chemical shift change is capped at 0.4 ppm. Insets show combined chemical shift change (∆δ>0.05 ppm), mapped on the structure of Atox1 monomer, in Atox1-Cu (D, orange), Atox1-Pt (E, blue) and Atox1-Cu-Pt (F, magenta). The cysteines in the conserved CxxC motif are shown as spheres.

Chemical structure of Atox1-Cu-Pt complex From our experimental data we conclude that in the presence of glutathione and Cu(I) Atox1 exists as a dimer, and, after the addition of cisplatin, an Atox1-Cu-Pt complex is formed, in which copper is bound to three sulfur atoms and platinum is bound to four. Cu and Pt bind in close proximity to each other in the metal-binding motif of Atox1, but do not form a metal-metal bond. We have considered possible chemical structures that satisfy these conditions, and tested several models by molecular mechanics and DFT calculations. The copper EXAFS Fourier transform peaks at around 2.7 Å are characteristic of the Cu∙∙∙Cu interactions in Cu4S6 clusters found in the copper-bound metallothioneins37,40 (Fig. 1 F). The Cu-glutathione complexes and Atox1-Cu-Pt show very similar 2.7 Å peaks in their spectra, and therefore our initial models were based on metal arrangement in the Cu4S6 clusters.

As a starting point and a benchmark, hydrated structures of an 11-residue fragment of Atox1 monomer containing the CGGC metal-binding motif and flanking residues, with a bound Cu(I) atom, were generated with Monte Carlo molecular dynamics using both the fixed-charge OPLS and polarizable POSSIM force fields. Both models reproduced the copper bound geometry with similar levels of accuracy. The average Cu-S distances were 2.21(5) Å and 2.28(7) Å for the OPLS and POSSIM models respectively. Both models exhibited nearly linear (ca. 150°) S-Cu-S coordination. This compares favorably to previous NMR and EXAFS studies of the Atox1Cu monomer that indicated a digonal copper coordination with a Cu-S distance of 2.17 Å and S-Cu-S angle of about 160 degrees44,50, but differs from our experimental data that consistently show tri-coordinate copper with Cu-S distances of 2.25-2.26 Å in the complexes of Atox1 and glutathione with copper.


Page 9 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 5. Binding of cisplatin to Atox1-Cu in the presence of glutathione. A: A comparison of copper K-edge spectra of Atox1-Cu (black) and Atox1-Cu-Pt (red); B: A comparison of platinum LIII-edge spectra of Atox1-Pt (black) and Atox1-Cu-Pt (red); C: copper EXAFS Fourier transforms of Atox1-Cu-Pt (solid line – experimental data, dashed line - data fit) ); D: A comparison of copper EXAFS Fourier transforms of copper-glutathione polymers (blue), Atox1-Cu (black) and Atox1-Cu-Pt (red); E: platinum EXAFS Fourier transforms of Atox1-Cu-Pt (solid line – experimental data, dashed line - data fit); F: A comparison of platinum EXAFS Fourier transforms of Atox1-Pt (black) and Atox1-Cu-Pt (red).


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Additionally, structures of the Atox1-Cu monomer with a coordinated glutathione molecule were investigated. Evidence of such complexes formed in large excess of glutathione or other thiols has been reported44. Interestingly, neither the non-polarizable nor the polarizable model resulted in a stable coordination structure as the S– of glutathione consistently migrated to the second solvation shell (~3.6 Å) of the copper atom. Attempts to produce dimeric structures of copperloaded Atox1 with a bridging glutathione ligand were also unsuccessful as these structures consistently disassociated into copper-loaded monomers. This strong preference for dicoordinated structures in disagreement with experimental data, which indicates CuS3 coordination mode, is best explained by the limitations of the truncated Atox1 model used in our simulations. Specifically, the model does not include Lys60, which is proximal to the metal binding domain (Fig 4 D, inset), and has been shown to be necessary for optimal copper transfer to ATP7B51. Presumably, the positively charged lysine sidechain stabilizes the net negative charge of the copper-thiolate complex. Atox1 dimerization may also provide additional stabilization for tricoordinate copper coordination by sulfur observed experimentally.

Page 10 of 15

We have then screened a number of potential Atox1-Cu-Ptglutathione structures for stability. In order to reduce the number of possible structures tested, two key factors were considered. First was agreement with the number and type of metalcoordinated ligands as shown by our spectroscopic data. Second, as indicated by NMR data (Fig. 4), only structures in which copper is bound to one or more of the cysteine residues in Atox1 dimer were considered. Due to the sub-femtomolar binding affinity of Atox1 for Cu(I)52, Pt(II) would not be expected to completely displace copper from the binding site. Monte Carlo molecular mechanics results were further refined in density functional theory (DFT) simulations. This yielded two types of stable Cu-S-Pt clusters. The first type contains two copper and one platinum atom (Cu2PtS6, Fig. 6 A), while the second has two copper and two platinum atoms (Cu2Pt2S8 , Fig. 6 D). The Cu2PtS6 model was qualitatively more stable in the simulations than Cu2Pt2S8 and could better resist small perturbations of coordination geometry.

Figure 6. Model structures of Atox1-GSH-Cu-Pt complexes and corresponding EXAFS spectra. A-C: Cu2PtS6 model; D-F Cu2Pt2S8 model ; G-I Cu2Pt2S10 model; A, D, G – ball-and-stick structure models with copper shown in orange, platinum in navy, sulfur in yellow, and the other atoms in CPK colors. The Cu2PtS6 model (A) is shown with fragments of Atox1 and GSH molecules, while the Cu2Pt2S8 (D) and Cu2Pt2S10 (G) models use methyl groups bonded to sulfur. B,E,H- experimental Fourier transformed Cu K-edge EXAFS (solid lines) and corresponding simulated spectra (dashed lines); C,F,I-experimental Fourier transformed Pt LIII EXAFS (solid lines) and corresponding simulated spectra (dashed lines).

We simulated EXAFS Fourier transforms (dashed lines in Fig. 6 B, C, E, F) from the atomic coordinates of the two model structures. In either case, simulated EXAFS did not

reproduce the experimental peak corresponding to Cu∙∙∙Cu interaction at the distance of ~2.7 Å, because both models place copper atoms farther away from each other, at approximately


Page 11 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

3.3 Å. The average Cu-S and Pt-S distances in both models were also longer than reported by EXAFS. To account for the short Cu∙∙∙Cu distance observed experimentally, we designed a Cu2Pt2S10 model, which puts copper atoms at 2.7 Å from each other, with two platinum atoms bound through sulfur bonds to each copper (Fig. 6 G). Although this model has not been tested using molecular mechanics, it was refined by DFT. Simulated EXAFS for both copper K-edge and platinum LIII-edge for Cu2Pt2S10 model are in good agreement with the experimental data: Fourier transform of copper EXAFS contains ~2.25 Å peak corresponding to Cu∙∙∙S bond and ~2.7 Å peak corresponding to Cu∙∙∙Cu interaction (Fig. 6 H), while Fourier transform of Pt EXAFS accurately reproduces the experimental Pt-S peak. According to the Cu2Pt2S10 model, distances between copper and platinum are in the range of 3.7-3.9 Å. At such a distance, Debye-Waller factor is usually quite high, decreasing the peak amplitude. Moreover, platinum is connected to copper through a single sulfur atom, which allows certain degree of flexibility in the relative positions of copper and platinum, further increasing Debye-Waller factor, and decreasing amplitude of the peak corresponding to Cu∙∙∙Pt interaction to the noise level (Fig. 6 H, I). Thus Cu2Pt2S10 model provides the best description of the metal atom arrangement in Atox1 dimer at the stoichiometric ratio of Atox1, copper and platinum. Under the conditions of chemotherapy, a variety of Atox1 complexes with glutathione, copper and platinum may form in the cell, depending on cisplatin concentration and copper status. A rough estimate based on Atox1 abundance in the cell53, and cisplatin concentration dynamics in blood plasma under the standard chemotherapy regimen54 suggests that the numbers of Atox1 and cisplatin molecules in the cell can be of the same order of magnitude, and therefore structures similar to the Cu2PtS6 model, which was shown to be stable in simulations, can occur along with Cu2Pt2S10 and single-metal Atox1 complexes. In summary, we have shown that: (i) Atox1 exists in a dimeric form and binds copper and platinum in the same site formed by the proximal CxxC motifs of the two Atox1 monomers; (ii) copper is coordinated by three and platinum by four sulfur atoms; (iii) glutathione molecules participate in metal binding to Atox1; and (iv) in Atox1-Cu-Pt, copper and platinum do not show a well-defined metal-metal interaction, but are connected through the shared sulfur ligands.

and the metal balance is likely to affect the effectiveness of anticancer chemotherapy with platinum based drugs.

ABBREVIATIONS DFT – density functional theory; DTT – dithiothreitol; EXAFS – Extended X-ray Absorption Fine Structure; GSH – glutathione; GSSG – oxidized glutathione; HSQC- heteronuclear single quantum coherence experiment; NMR – nuclear magnetic resonance; ROS – reactive oxygen species; SEC – size exclusion chromatography; TCEP - tris(2-carboxyethyl)phosphine.

AUTHOR INFORMATION Corresponding Author Oleg Y. Dmitriev, 107 Wiggins Rd, Saskatoon, SK, S7N 5E5, Canada, Phone: 1-306-966-4377, E-mail: [email protected] Funding Sources This research was supported by Canadian Institutes of Health Research and Saskatchewan Health Research Foundation, and by Telus Ride for Dad funding to O.D. GNG and IJP are Canada Research Chairs and acknowledge funding from the Natural Sciences and Engineering Research Council of Canada and from the Saskatchewan Innovation and Science Fund. KHN and KLS are Fellows in the CIHR Training grant in Health Research Using Synchrotron Techniques (CIHRTHRUST). KLS is a recipient of Alexander Graham Bell Canada Graduate Scholarship (NSERC). MH and JB were funded by NSF ACI-1339844.

ACKNOWLEDGEMENTS We thank Dr. Svetlana Lutsenko (John Hopkins University, Baltimore, MA) for providing us with plasmid for Atox1 expression. The supercomputer time was provided by NSF grant OCI1036215 at the National Center for Supercomputing Applications (NSF OCI-0725070 and ACI-1238993). NVD and JJHC are CIHR-THRUST Associates. Use of the Stanford Synchrotron Radiation Lightsource, SLAC National Accelerator Laboratory, is supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Contract No. DE-AC02-76SF00515. The SSRL Structural Molecular Biology Program is supported by the DOE Office of Biological and Environmental Research, and by the National Institutes of Health, National Institute of General Medical Sciences (including P41GM103393). The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official views of NIGMS or NIH.

CONCLUSION Analysis of copper and cisplatin binding to copper chaperone Atox1 under physiological redox conditions revealed the formation of large copper- and platinum-glutathione complexes that are able to transfer both metals to the protein. Binding of copper and cisplatin to Atox1 was found to occur through the formation of copper-sulfur-platinum bridges, where copper is coordinated by three and platinum by four sulfur atoms. These data offer a new perspective on copper and cisplatin metabolism in the cells where glutathione and Atox1 both participate in platinum transport across the cell,

ASSOCIATED CONTENT Supporting Information: Two figures showing SEC profiles and EXAFS data, and two tables with EXAFS fit data. This material is available free of charge via the Internet at http://pubs.acs.org.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 15

REFERENCES 1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

Horn, D., Barrientos, A. (2008) Mitochondrial copper metabolism and delivery to cytochrome c oxidase, IUBMB Life, 60(7), 421-429. Lutsenko, S., Barnes, N.L., Bartee, M.Y., Dmitriev, O.Y. (2007) Function and regulation of human coppertransporting ATPases. Physiol. Rev., 87(3), 1011-1046. Opazo, C.M., Greenough, M.A., Bush, A.I. (2014) Copper: from neurotransmission to neuroproteostasis, Front. Aging Neurosci., 2014, 6, 143. Parvez, S., Kang, M., Chung, H., Bae, H. (2007) Naturally occurring tyrosinase inhibitors: mechanism and applications in skin health, cosmetics and agriculture industries, Phytother. Res., 21, 805-816. Watanabe, K., Shibuya, S., Ozawa, Y., Nojiri, H., Izuo, N., Yokote, K., Shimizu, T. (2014) Superoxide dismutase 1 loss disturbs intracellular redox signaling, resulting in global age-related pathological changes, BioMed Res. Int., 2014, 140165. Ahuja, A., Dev, K., Tanwar, R., Selwal, K., Tyagi, P.J. (2015) Copper mediated neurological disorder: Visions into amyotrophic lateral sclerosis, Alzheimer and Menkes Disease. Trace Elem. Med. Biol., 29, 11-23. Brewer, G. (2014) The promise of copper lowering therapy with tetrathiomolybdate in the cure of cancer and in the treatment of inflammatory disease, J. Trace Elem. Med. Biol., 28(4), 372-378. Myhre, O., Utkilen, H., Duale, N., Brunborg, G., Hofer, T. (2013) Metal dyshomeostasis and inflammation in alzheimer’s and parkinson’s diseases: possible impact of environmental exposures, Oxid. Med. Cell. Longev., 2013, 726954. Halliwell, B., Gutteridge, J.M. (1990) Role of free radicals and catalytic metal ions in human disease: an overview, Methods Enzymol., 186, 1-85. Rae, T.D., Schmidt, P.J., Pufahl, R.A., Culotta, V.C., O'Halloran, T.V. (1999) Undetectable intracellular free copper: the requirement of a copper chaperone for superoxide dismutase, Science, 284, 805-808. Howell, S.B., Safaei, R., Larson, C.A., Sailor, M.J. (2010) Copper transporters and the cellular pharmacology of the platinum-containing cancer drugs, Mol. Pharm., 77(6), 887-894. Ishida, S., Lee, J., Thiele, D.J., Herskowitz, I. (2002) Uptake of the anticancer drug cisplatin mediated by the copper transporter Ctr1 in yeast and mammals, Proc. Natl. Acad. Sci. USA, 99(22), 14298-14302. Komatsu, M., Sumizawa, T., Mutoh, M., Chen, Z.-S., Terada, K., Furukawa, T., Yang, X.-L., Gao, H., Miura, N., Sugiyama, T., Akiyama, S. (2000) Copper-transporting Ptype adenosine triphosphatase (ATP7B) is associated with cisplatin resistance. Cancer Research, 60, 1312-1316. Apps, M.G., Choi, E.H.Y., Wheate, N.J. (2015) The stateof-play and future of platinum drugs, Endoc .Relat. Cancer, 22, R219-R233. Dolgova, N., Nokhrin, S., Yu, C., George, G., Dmitriev, O. (2013) Copper chaperone Atox1 interacts with the metal-binding domain of Wilson’s disease protein in cisplatin detoxiﬁcation. Biochem. J., 454, 147-156. Dolgova, N.V., Olson, D., Lutsenko, S., Dmitriev, O. (2009) The soluble metal-binding domain of the copper transporter ATP7B binds and detoxifies cisplatin. Biochem. J., 419, 51-56. Galliani, A., Losacco, M., Lasorsa, A. Natile, G., Arnesano, F. (2014) Cisplatin handover between copper transporters: the effect of reducing agents, J. Biol. Inorg. Chem., 19(4-5), 705-14.

18. Boal, A.K., Rosenzweig, A.C. (2009) ) Crystal structures of cisplatin bound to a human copper chaperone, J. Am. Chem. Soc., 131, 14196-14197. 19. Ma, G., Wu, Q., Wu, X., Arnesano, F., Natile, G., Sletten, E., Liu, Y. (2015) The reaction of a platinated methionine motif of CTR1 with cysteine and histidine is dependent upon the type of precursor platinum complex. J. Inorg. Biochem., 153, 239-246. 20. Nguen, T.H., Arnesano, F., Scintilla, S., Rosetti, G., Ippoliti, E., Carloni, P., Natile, G. (2013) Structural determinants of cisplatin and transplatin binding to the Met-rich motif of Ctr1: a computational spectroscopy approach, J. Chem. Theory Comput., 8(8), 2912-2920. 21. Safaei, R., Otani, S., Larson, B.J., Rasmussen, M.L., Howell, S.B. (2008) Transport of cisplatin by the copper efflux transporter ATP7B. Mol. Pharm., 73(2), 461-468. 22. Hua, H., Günther, V., Georgiev, O., Schaffner, W. (2011) Distorted copper homeostasis with decreased sensitivity to cisplatin upon chaperone Atox1 deletion in Drosophila, Biometals, 24(3), 445-453. 23. Safaei, R., Maktabi, M.H., Blair, B.G., Larson, C.A., Howell, S.B. (2009) Effects of the loss of Atox1 on the cellular pharmacology of cisplatin, J. Inorg. Biochem., 103(3), 333-341. 24. Jamieson, E.R., Lippard, S.J. (1999) Structure, recognition, and processing of cisplatin-DNA adducts, Chem. Rev., 99, 2467–2498. 25. Beaino, W., Guo, Y., Chang, A., Anderson, C.J. (2014) Roles of Atox1 and p53 in the trafficking of copper-64 to tumor cell nuclei: implications for cancer therapy, J. Biol. Inorg. Chem., 19(3), 423-438. 26. Itoh, S., Kim, H.W., Nakagawa, O., Ozumi, K., Lessner. S.M., Aoki, H., Akram, K., McKinney, R.D., Ushio-Fukai, M., Fukai, T. (2008) Novel role of antioxidant-1 (Atox1) as a copper-dependent transcription factor involved in cell proliferation, J. Biol. Chem., 283, 9157–9167. 27. Dmitriev, O.Y. (2011) Mechanism of tumor resistance to cisplatin mediated by the copper transporter ATP7B, Biochem. Cell Biol., 89, 138-147. 28. Palm-Espling, M.,E., Andersson, C.D., Bjorn, E., Linusson, A., Wittung-Stafshede, P. (2013) Determinants for simultaneous binding of copper and platinum to human chaperone Atox1: hitchhiking not hijacking. PloS One, 8(7), e70473. 29. Xi, Z., Guo, W., Tian, C., Wang, F., Liu, Y. (2013) Copper binding promotes the interaction of cisplatin with human copper chaperone Atox1. Chem. Commun., 49, 1119711199. 30. Deponte, M. (2013) Glutathione catalysis and the reaction mechanisms of glutathione-dependent enzymes, Biochim. et Biophys. Acta 1830, 3217-3266. 31. Schafer, F.Q., Buettner, G.R. (2001) Redox environment of the cell as viewed through the redox state of the glutathione disulfide/glutathione couple, Free Radic. Biol. Med., 30(11), 1191-1212. 32. Narindrasorasak, S., Zhang, X., Roberts, E.A., Sarkar, B. (2004) Comparative analysis of metal binding characteristics of copper chaperone proteins, Atx1 and Atox1. Bioinorg. Chem. Appl., 2(1-2), 105-123. 33. Tanchou, V., Gas, F., Urvoas, A., Cougouluegne, F., Ruat, S., Averseng, O., Quemeneur, E. (2004) Copper-mediated homo-dimerisation for the HAH1 metallochaperone, Biochem. Biophys. Res. Comm., 2004, 388-394. 34. Wernimont, A.K., Huffman, D.L., Lamb, A.L., O’Halloran, T.V., Rosenzweig, A.C. (2000) Structural basis for copper transfer by the metallochaperones for the Menkes/Wilson disease proteins, Nature Struct. Biol., 7, 766-771.


Page 13 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry 35. Miras, R., Morin, I., Jacquin, O., Cuillel, M., Guillain, F., Mintz, E.(2008) Interplay between glutathione, Atx1 and copper. 1. Copper(I) glutathionate induced dimerization of Atx1, J. Biol. Inorg. Chem., 13, 195-205. 36. Kau, L.S., Spira-Solomon, D.J., Penner-Hahn, J.E., Hodgson, K.O., Solomon, E.I. (1987) X-ray absorption edge determination of the oxidation state and coordination numer of copper: application to the type 3 site in Rhus vernicifera laccase and its reaction with oxygen. J. Am. Chem. Soc., 109 (1987), 6433-6442. 37. Pickering, I.J., George, G.N., Dameron, C.T., Kurz, B., Winge, D.R., Dance, I.G. (1993) X-ray absorption spectroscopy of cuprous-thiolate clusters in proteins and model systems, J. Am. Chem. Soc., 115, 9498-9505. 38. Pushie, M.J., Zhang, L., Pickering, I.J., George, G.N. (2012) The fictile coordination chemistry of cuprous-thiolate sites in copper chaperones, Biochim. Biophys. Acta., 1817, 938-947. 39. Rubino, J.T., Franz, K.J. (2012) Coordination chemistry of copper protens: how nature handles toxic cargo for essential function, J. Inorg. Biochem., 107, 129-143. 40. Zhang, L., Pickering, I.J., Winge, D.R., George, G.N. (2008) X-ray absorption spectroscopy of cuprous-thiolate clusters in Saccharomyces cerevisiae metallothionein. Chem. Biodiv., 5, 2042–2049. 41. Corazza, A., Harvey, I., Sadler, P.J. (1996) 1H,13C-NMR and X-ray absorption studies of copper(I) glutathione complexes, Eur. J. Biochem., 236, 697-705. 42. Poger, D., Fillaux, C., Miras, R., Crouzy, S., Delangle, P., Mintz, E., Den Auwer, C., Ferrand, M. (2008) Interplay between glutathione, ATX1 and copper: X-ray absorption spectroscopy determination of Cu(I) environment in an Atx1 dimer, J. Biol. Inorg. Chem., 13, 1239-1248. 43. Banci, L., Bertini, I., Ciofi-Baffoni, S., Kozyreva, T., Zovo, K., Paluma, P. (2010) Affinity gradients drive copper to cellular destinations. Nature, 465, 645-650. 44. Ralle, M., Lutsenko, S., Blackburn, N. (2003) X-ray absorption spectroscopy of the copper chaperone HAH1 reveals a linear two-coordinate cu(i) center capable of adduct formation with exogenous thiols and phosphines. J. Biol. Chem., 278(25), 23163-23170. 45. Kaplan, J.H., Maryon, E.B. (2016) How mammalian cells acquire copper: an essential but potentially toxic metal. Biophys. J., 110, 7-13. 46. Maryon, E.B., Molly, S.A., Kaplan, J.H. (2013) Cellular glutathione plays a key role in copper uptake mediated by human copper transporter 1. Am. J. Physiol. Cell. Physiol., 304, C768-C779. 47. Ferreira, A.M., Ciriolo, M.R., Marcocci, L., Rotillo, G. (1993) Copper(I) transfer into metallothioneins mediated by glutathione, Biochem J., 292(Pt 3), 673-676. 48. Ciriolo, M.R., Desideri, A., Paci, M., Rotilio, G. (1990) Reconstitution of Cu,Zn-superoxide dismutase by the Cu(I)-glutathione complex. J. Biol. Chem., 265, 1103011034. 49. Palm, M.E., Weise, C.F., Lundin, C., Wingsle, G., Nygren, Y., Björn, E., Naredi, P., Wolf-Watz, Wittung-Stafshede, P. (2011) Cisplatin binds human copper chaperone Atox1 and promotes unfolding in vitro. Proc. Natl. Acad. Sci. USA, 108(17), 6951-6956. 50. Anastassopoulou, I., Banci, L., Bertini. I., Cantini, F., Katsari, E., Rosato, A. (2004) Solution structure of the apo and copper(i)-loaded human metallochaperone HAH1. Biochemistry, 43, 13046-13053. 51. Hussain, F., Rodriguez-Granillo, A., Wittung-Stafshede, P. (2009) Lysine-60 in copper chaperone Atox1 plays an essential role in adduct formation with a target Wilson Disease domain. J. Am. Chem. Soc., 131, 16371-16373. 52. Xiao, Z., Brose, J., Schimo, S., Ackland, S.M., La Fontaine, S., Wedd, A.G. (2011) Unification of the copper(I) binding affinities of the metallo-chaperones Atx1, Atox1,

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

and related proteins: detection probes and affinity standards, J. Biol. Chem., 286(13), 11047-11055. Beck, M., Schmidt, A., Malmstroem, J., Claassen, M., Ori, A., Szymborska, A., Herzog, F., Rinner, O., Ellenberg, J., Aebersold, R. (2011) The quantitative proteome of human cell line, Mol. Sys. Biol., 7, 549. Himmelstein, K.J., Patton, T.F., Belt, R.J., Taylor, S., Repta, A.J., and Sternson, L.A. (1981) Clinical kinetics on intact cisplatin and some related species, Clin. Pharmacol. Ther. 29(5), 658-664. Morgan, C.T., Tsivkovskii, R., Kosinsky, Y.A., Efremov, R.G., Lutsenko, S. (2004) The distinct functional properties of the nucleotide-binding domain of ATP7B, the human copper-transporting ATPase: analysis of the Wilson disease mutations E1064A, H1069Q, R1151H, and C1104F, J. Biol. Chem, 279, 36363-36371. Delaglio, F., Grzesiek, S., Vuister, G., Zhu, G., Pfeifer, J., Bax, A. (1995) NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J. Biomol. NMR, 6, 277-293. Jorgensen, W.L., Chandrasekhar, J., Madura, J.D., Impey, R.W., Klein, M.L. (1983) Comparison of simple potential functions for simulating liquid water, J. Chem. Phys., 79, 926-935. Jorgensen, W.L. and Tirado-Rives, J. (2005) Molecular modeling of organic and biomolecular systems using BOSS and MCPRO, J. Comput. Chem., 26, 1689–1700. Kaminski, G.A., Friesner, R.A., Tirado-Rives, J., Jorgensen, W.L. (2001) Evaluation and Reparametrization of the OPLS-AA Force Field for Proteins in Comparison with Accurate Quantum Chemical Calculations on Peptides. J. Phys. Chem. B 105, 28, 6474-6487. Click, T.H., Ponomarev, S.Y., Kaminski, G.A. (2012) Importance of electrostatic polarizability in calculating cysteine acidity constants and copper(I) binding energy of Bacillus subtilis CopZ, J. Comput. Chem., 33, 1142–1151. Cvitkovic, J.P., Kaminski, G.A. (2017) Developing multisite empirical force field models for Pt(II) and cisplatin, J. Comput. Chem., 38, 161-168. Ponomarev, S.Y., Click, T.H., Kaminski, G.A. (2011) Polarizable simulations with second-order interaction model (possim) force field: developing parameters for alanine pepties and protein backbone, J. Phys. Chem. B, 115(33), 10079–10085. Li, X., Ponomarev, S.Y., Sigalovsky, D.L., Cvitkovic, J.P., Kaminski, G.A. (2014) POSSIM: parameterizing complete second-order polarizable force field for proteins, J. Chem. Theory Comput., 10(11), 4896-4910. Ponomarev, S.Y., Kaminski, G.A. (2011) Polarizable simulations with second-order interaction model (POSSIM) force field: developing parameters for alanine peptides and protein backbone, J. Chem. Theory Comput., 7(5), 14151427. Kaminski, G.A., Ponomarev, S.Y., Liu, A.B. (2009) Polarizable simulations with second order interaction model – force field and software for fast polarizable calculations: parameters for small model systems and free energy calculations. J. Chem. Theory Comput., 5(11), 2935-2943. Hodak, M., Lu, W., Bernholc, J. (2008) Hybrid ab initio Kohn-Sham density functional theory/frozen-density orbital-free density functional theory simulation method suitable for biological systems, J. Chem. Phys., 128(1), 014101-9. Bernholc, J., Hodak, M., Lu, W. (2008) Recent developments and applications of the real-space multigrid method, J. Phys. Condens. Matter, 20(29). Briggs, E.L., Sullivan, D.J., Bernholc, J. (1996) Realspace multigrid-based approach to large-scale electronic structure calculations, Phys. Rev. B Condens. Matter, 54(20), 14362-14375.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

69. Hodak, M., Wang, S., Lu, W., Bernholc, J. (2007) Implementation of ultrasoft pseudopotentials in large-scale gridbased electronic structure calculations, Phys. Rev. B, 76, 085108.


Page 14 of 15

Page 15 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

TOC graphic

81x43mm (300 x 300 DPI)


Binding of Copper and Cisplatin to Atox1 Is ... - ACS Publications

Recommend Documents