The plasma membrane-associated cation-binding protein PCaP1 of Arabidopsis thaliana is a uranyl-binding protein.

Uranium (U) is a naturally-occurring radionuclide that is toxic to living organisms. Given that proteins are primary targets of U(VI), their identification is an essential step towards understanding the mechanisms of radionuclide toxicity, and possibly detoxification. Here, we implemented a chromatographic strategy including immobilized metal affinity chromatography to trap protein targets of uranyl in Arabidopsis thaliana. This procedure allowed the identification of 38 uranyl-binding proteins (UraBPs) from root and shoot extracts. Among them, UraBP25, previously identified as plasma membrane-associated cation-binding protein 1 (PCaP1), was further characterized as a protein interacting in vitro with U(VI) and other metals using spectroscopic and structural approaches, and in planta through analyses of the fate of U(VI) in Arabidopsis lines with altered PCaP1 gene expression. Our results showed that recombinant PCaP1 binds U(VI) in vitro with affinity in the nM range, as well as Cu(II) and Fe(III) in high proportions, and that Ca(II) competes with U(VI) for binding. U(VI) induces PCaP1 oligomerization through binding at the monomer interface, at both the N-terminal structured domain and the C-terminal flexible region. Finally, U(VI) translocation in Arabidopsis shoots was affected in pcap1 null-mutant, suggesting a role for this protein in ion trafficking in planta.

Determination of the affinity of biomimetic peptides for uranium through the simultaneous coupling of HILIC to ESI-MS and ICP-MS.

Several proteins have been identified in the past decades as targets of uranyl (UO22+) in vivo. However, the molecular interactions responsible for this affinity are still poorly known which requires the identification of the UO22+ coordination sites in these proteins. Biomimetic peptides are efficient chemical tools to characterize these sites. In this work, we developed a dedicated analytical method to determine the affinity of biomimetic, synthetic, multi-phosphorylated peptides for UO22+ and evaluate the effect of several structural parameters of these peptides on this affinity at physiological pH. The analytical strategy was based on the implementation of the simultaneous coupling of hydrophilic interaction chromatography (HILIC) with electrospray ionization mass spectrometry (ESI-MS) and inductively coupled plasma mass spectrometry (ICP-MS). An essential step had been devoted to the definition of the best separation conditions of UO22+ complexes formed with di-phosphorylated peptide isomers and also with peptides of different structure and degrees of phosphorylation. We performed the first separations of several sets of UO22+ complexes by HILIC ever reported in the literature. A dedicated method had then been developed for identifying the separated peptide complexes online by ESI-MS and simultaneously quantifying them by ICP-MS, based on uranium quantification using external calibration. Thus, the affinity of the peptides for UO22+ was determined and made it possible to demonstrate that (i) the increasing number of phosphorylated residues (pSer) promotes the affinity of the peptides for UO22+, (ii) the position of the pSer in the peptide backbone has very low impact on this affinity (iii) and finally the cyclic structure of the peptide favors the UO22+ complexation in comparison with the linear structure. These results are in agreement with those previously obtained by spectroscopic techniques, which allowed to validate the method. Through this approach, we obtained essential information to better understand the mechanisms of toxicity of UO22+ at the molecular level and to further develop selective decorporating agents by chelation.

Optimized Coordination of Uranyl in Engineered Calmodulin Site 1 Provides a Subnanomolar Affinity for Uranyl and a Strong Uranyl versus Calcium Selectivity.

As an alpha emitter and chemical toxicant, uranium toxicity in living organisms is driven by its molecular interactions. It is therefore essential to identify main determinants of uranium affinity for proteins. Others and we showed that introducing a phosphoryl group in the coordination sphere of uranyl confers a strong affinity of proteins for uranyl. In this work, using calmodulin site 1 as a template, we modulate the structural organization of a metal-binding loop comprising carboxylate and/or carbonyl ligands and reach affinities for uranyl comparable to that provided by introducing a strong phosphoryl ligand. Shortening the metal binding loop of calmodulin site 1 from 12 to 10 amino acids in CaMΔ increases the uranyl-binding affinity by about 2 orders of magnitude to log KpH7 = 9.55 ± 0.11 (KdpH7 = 280 ± 60 pM). Structural analysis by FTIR, XAS, and molecular dynamics simulations suggests an optimized coordination of the CaMΔ-uranyl complex involving bidentate and monodentate carboxylate groups in the uranyl equatorial plane. The main role of this coordination sphere in reaching subnanomolar dissociation constants for uranyl is supported by similar uranyl affinities obtained in a cyclic peptide reproducing CaMΔ binding loop. In addition, CaMΔ presents a uranyl/calcium selectivity of 107 that is even higher in the cyclic peptide.

A Simple Fluorescence Affinity Assay to Decipher Uranyl-Binding to Native Proteins.

Determining the affinity of proteins for uranyl is key to understand the toxicity of this cation and to further develop decorporation strategies. However, usual techniques to achieve that goal often require specific equipment and expertise. Here, we propose a simple, efficient, fluorescence-based method to assess the affinity of proteins and peptides for uranyl, at equilibrium and in buffered solution. We first designed and characterized an original uranyl-binding fluorescent probe. We then built a reference scale for uranyl affinity in solution, relying on signal quenching of our fluorescent probe in presence of high-affinity uranyl-binding peptides. We finally validated our approach by re-evaluating the uranyl-binding affinity of four native proteins. We envision that this tool will facilitate the reliable and reproducible assessment of affinities of peptides and proteins for uranyl.

In vitro assessment of cobalt oxide particle dissolution in simulated lung fluids for identification of new decorporating agents.

Inhalation of 60Co3O4 particles may occur at the work place in nuclear industry. Their low solubility may result in chronic lung exposure to γ rays. Our strategy for an improved therapeutic approach is to enhance particle dissolution to facilitate cobalt excretion, as the dissolved fraction is rapidly eliminated, mainly in urine. In vitro dissolution of Co3O4 particles was assessed with two complementary assays in lung fluid surrogates to mimic a pulmonary contamination scenario. Twenty-one molecules and eleven combinations were selected through an extensive search in the literature, based on dissolution studies of other metal oxides (Fe, Mn, Cu) and tested for dissolution enhancement of cobalt particles after 1-28 days of incubation. DTPA, the recommended treatment following cobalt contamination did not enhance 60Co3O4 particles dissolution when used alone. However, by combining molecules with different properties, such as redox potential and chelating ability, we greatly improved the efficacy of each drug used alone, leading for the highest efficacy, to a 2.7 fold increased dissolution as compared to controls. These results suggest that destabilization of the particle surface is an important initiating event for a good efficacy of chelating drugs, and open new perspectives for the identification of new therapeutic strategies.

Cyclic Phosphopeptides to Rationalize the Role of Phosphoamino Acids in Uranyl Binding to Biological Targets.

The specific molecular interactions responsible for uranium toxicity are not yet understood. The uranyl binding sites in high-affinity target proteins have not been identified yet and the involvement of phosphoamino acids is still an important question. Short cyclic peptide sequences, with three glutamic acids and one phosphoamino acid, are used as simple models to mimic metal binding sites in phosphoproteins and to help understand the mechanisms involved in uranium toxicity. A combination of peptide design and synthesis, analytical chemistry, extended X-ray absorption fine structure (EXAFS) spectroscopy, and DFT calculations demonstrates the involvement of the phosphate group in the uranyl coordination sphere together with the three carboxylates of the glutamate moieties. The affinity constants measured with a reliable analytical competitive approach at physiological pH are significantly enhanced owing to the presence of the phosphorous moiety. These findings corroborate the importance of phosphoamino acids in uranyl binding in proteins and the relevance of considering phosphoproteins as potential uranyl targets in vivo.

Preorganized Peptide Scaffolds as Mimics of Phosphorylated Proteins Binding Sites with a High Affinity for Uranyl.

Cyclic peptides with two phosphoserines and two glutamic acids were developed to mimic high-affinity binding sites for uranyl found in proteins such as osteopontin, which is believed to be a privileged target of this ion in vivo. These peptides adopt a ß-sheet structure that allows the coordination of the latter amino acid side chains in the equatorial plane of the dioxo uranyl cation. Complementary spectroscopic and analytical methods revealed that these cyclic peptides are efficient uranyl chelating peptides with a large contribution from the phosphorylated residues. The conditional affinity constants were measured by following fluorescence tryptophan quenching and are larger than 10(10) at physiological pH. These compounds are therefore promising models for understanding uranyl chelation by proteins, which is relevant to this actinide ion toxicity.

Pseudo-peptides based on methyl cysteine or methionine inspired from Mets motifs found in the copper transporter Ctr1.

Most proteins involved in Cu homeostasis bind to intracellular Cu(I) in stable Cu(S-Cys)x environments, thanks to well-conserved cysteine-rich sequences. Similarly, the Cu(I) transport protein Ctr1, responsible for copper acquisition, binds Cu(I) in Cu(S-Met)3 environments in conserved methionine-rich MXMXXM sequences, referred as Mets motifs. Pseudo-peptides based on a nitrilotriacetic acid scaffold and functionalized with three amino acids bearing thioether side chains, either methyl cysteine in T(1) or methionine in T(2), were synthesized as mimics of the Mets sequences found in Ctr1. These two ligands were obtained with good overall yields from commercial amino acids and demonstrate efficient chelating ability for Cu(I). Only one species, the mononuclear [CuT(1,2)](+) complex, was evidenced by electrospray ionization-mass spectroscopy (ESI-MS) and the circular dichroism signature obtained for the most constrained CuT(1) complex having the shortest side chains showed reorganization of the pseudo-peptide scaffold upon Cu(I) complexation. Considering that thioether functions are neutral sulfur donors, the stability constants measured by competition with ferrozine are quite large: log K ≈ 10.2-10.3. The CuT(1,2) complexes are significantly more stable that those formed with linear peptides, mimicking isolated Mets motifs MXMXXM of the Cu transport protein Ctr1 (log K ≈ 5-6). This may be attributed to the preorganized pseudo-peptide scaffold, which arranges the three neutral sulfur donors toward the metal center. Such moderate affinity Cu(I) chelators are interesting for applications in chelation therapy, for instance, to induce minimum disturbance of Cu homeostasis in Wilson's disease treatments.

Engineering short peptide sequences for uranyl binding.

Peptides are interesting tools to rationalize uranyl-protein interactions, which are relevant to uranium toxicity in vivo. Structured cyclic peptide scaffolds were chosen as promising candidates to coordinate uranyl thanks to four amino acid side chains pre-oriented towards the dioxo cation equatorial plane. The binding of uranyl by a series of decapeptides has been investigated with complementary analytical and spectroscopic methods to determine the key parameters for the formation of stable uranyl-peptide complexes. The molar ellipticity of the uranyl complex at 195 nm is directly correlated to its stability, which demonstrates that the ß-sheet structure is optimal for high stability in the peptide series. Cyclodecapeptides with four glutamate residues exhibit the highest affinities for uranyl with log KC =8.0-8.4 and, therefore, appear as good starting points for the design of high-affinity uranyl-chelating peptides.

Design of intrahepatocyte copper(I) chelators as drug candidates for Wilson's disease.

Wilson's disease is an autosomal recessive disease caused by mutations on the ATP7B gene found on chromosome 13. Since the corresponding ATPase is in charge of copper (Cu) distribution and excretion in the liver, its malfunctioning leads to Cu overload. This short review deals with treatments of this rare disease, which aim at decreasing Cu toxicity and are, therefore, based on chelation therapy. The drugs used since the 1950s are described first, then a novel approach developed in our laboratory is presented. Since the liver is the main organ of Cu distribution in the body, we targeted the pool of intracellular Cu in hepatocytes. This Cu pool is in the +1 oxidation state, and therefore soft sulfur ligands inspired from binding sites found in metallothioneins were developed. Their targeting to the hepatocytes by functionalization with ligands of the asialoglycoprotein receptor led to their cellular incorporation and intracellular Cu chelation.

Modulating uranium binding affinity in engineered calmodulin EF-hand peptides: effect of phosphorylation.

To improve our understanding of uranium toxicity, the determinants of uranyl affinity in proteins must be better characterized. In this work, we analyzed the contribution of a phosphoryl group on uranium binding affinity in a protein binding site, using the site 1 EF-hand motif of calmodulin. The recombinant domain 1 of calmodulin from A. thaliana was engineered to impair metal binding at site 2 and was used as a structured template. Threonine at position 9 of the loop was phosphorylated in vitro, using the recombinant catalytic subunit of protein kinase CK2. Hence, the T(9)TKE(12) sequence was substituted by the CK2 recognition sequence TAAE. A tyrosine was introduced at position 7, so that uranyl and calcium binding affinities could be determined by following tyrosine fluorescence. Phosphorylation was characterized by ESI-MS spectrometry, and the phosphorylated peptide was purified to homogeneity using ion-exchange chromatography. The binding constants for uranyl were determined by competition experiments with iminodiacetate. At pH 6, phosphorylation increased the affinity for uranyl by a factor of ∼5, from K(d) = 25±6 nM to K(d) = 5±1 nM. The phosphorylated peptide exhibited a much larger affinity at pH 7, with a dissociation constant in the subnanomolar range (K(d) = 0.25±0.06 nM). FTIR analyses showed that the phosphothreonine side chain is partly protonated at pH 6, while it is fully deprotonated at pH 7. Moreover, formation of the uranyl-peptide complex at pH 7 resulted in significant frequency shifts of the ν(as)(P-O) and ν(s)(P-O) IR modes of phosphothreonine, supporting its direct interaction with uranyl. Accordingly, a bathochromic shift in ν(as)(UO(2))(2+) vibration (from 923 cm(-1) to 908 cm(-1)) was observed upon uranyl coordination to the phosphorylated peptide. Together, our data demonstrate that the phosphoryl group plays a determining role in uranyl binding affinity to proteins at physiological pH.

Chelation therapy in Wilson's disease: from D-penicillamine to the design of selective bioinspired intracellular Cu(I) chelators.

Wilson's disease is an orphan disease due to copper homeostasis dysfunction. Mutations of the ATP7B gene induces an impaired functioning of a Cu-ATPase, impaired Cu detoxification in the liver and copper overload in the body. Indeed, even though copper is an essential element, which is used as cofactor by many enzymes playing vital roles, it becomes toxic when in excess as it promotes cytotoxic reactions leading to oxidative stress. In this perspective, human copper homeostasis is first described in order to explain the mechanisms promoting copper overload in Wilson's disease. We will see that the liver is the main organ for copper distribution and detoxification in the body. Nowadays this disease is treated life-long by systemic chelation therapy, which is not satisfactory in many cases. Therefore the design of more selective and efficient drugs is of great interest. A strategy to design more specific chelators to treat localized copper accumulation in the liver will then be presented. In particular we will show how bioinorganic chemistry may help in the design of such novel chelators by taking inspiration from the biological copper cell transporters.

A series of tripodal cysteine derivatives as water-soluble chelators that are highly selective for copper(I).

A series of tripodal ligands derived from nitrilotriacetic acid and extended by three converging, metal-binding, cysteine chains was synthesised. Their ability to bind soft metal ions thanks to their three thiolate functions was investigated by means of complementary analytical and spectroscopic methods. Three ligands that differ by the nature of the carbonyl group next to the coordinating thiolate functions were studied: L(1) (ester), L(2) (amide) and L(3) (carboxylate). The negatively charged derivative L(3), which bears three carboxylate functions close to the metal binding site, gives polynuclear copper(I) complexes of low stability. In contrast, the ester and amide derivatives L(1) and L(2) are efficient Cu(I) chelators with very high affinities, close to that reported for the metal-sequestering metallothioneins (log K≈19). Interestingly, these two ligands form mononuclear copper complexes with a unique MS(3) coordination in water solution. An intramolecular hydrogen-bond network involving the amide functions in the upper cavity of the tripodal ligands stabilises these mononuclear complexes and was evidenced by the very low chemical-shift temperature coefficient of the secondary amide protons. Moreover, L(1) and L(2) display large selectivities for the targeted metal ion that is, Cu(I), with respect to bioavailable Zn(II). Therefore the two sulfur-based tripods L(1) and L(2) are of potential interest for intracellular copper detoxication in vivo, without altering the homeostasis of the essential metal ion Zn(II).

A rigorous framework to interpret water relaxivity. The case study of a Gd(III) complex with an alpha-cyclodextrin derivative.

We present a general theoretical framework suitable for an economical, but rigorous, analysis of the relaxivity and EPR data of paramagnetic metal complexes. This framework is based on the so-called Grenoble method that properly accounts for the fluctuations of the "static" zero-field splitting Hamiltonian and avoids the misinterpretation of experimental data, which occurs with the Solomon, Bloembergen, and Morgan (SBM) formalism and may lead to erroneous conclusions. The applicability of the SBM approximation is discussed. Our approach is implemented in the case of a new Gd(3+) chelate with a cyclodextrin derivative ligand hexakis(2-O-carboxymethyl-3,6-anhydro)-alpha-cyclodextrin (ACX), designed to obtain lanthanide complexes of enhanced stability in comparison to natural cyclodextrins. The introduction of carboxymethyl units on the six residual hydroxyl groups of an alpha-per-3,6-anhydro cyclodextrin leads to mono- and binuclear Ln(3+) complexes with log beta(110) approximately = 7.5. The GdACX complex induces large water proton relaxivity in 0.1 M KCl aqueous solution. The molecular parameters governing the longitudinal (r1) and transverse (r2) relaxivities above 1 T are obtained through simple SBM-like theoretical expressions and complementary experimental techniques. The metal hydration state, the translational diffusion coefficient of the complex, and its rotational correlation time are derived from luminescence lifetime studies, pulse-field gradient NMR, and deuteron quadrupolar relaxation, respectively. The high relaxivity induced by the GdACX complex is attributed to its high hydration state in the presence of potassium ions and to a rotational correlation time lengthened by the hydrophilic character of the ACX scaffold.

Model peptides based on the binding loop of the copper metallochaperone Atx1: selectivity of the consensus sequence MxCxxC for metal ions Hg(II), Cu(I), Cd(II), Pb(II), and Zn(II).

The amino acid sequence MxCxxC is conserved in many soft-metal transporters that are involved in the control of the intracellular concentration of ions such as Cu(I), Hg(II), Zn(II), Cd(II), and Pb(II). A relevant task is thus the selectivity of the motif MxCxxC for these different metal ions. To analyze the coordination properties and the selectivity of this consensus sequence, we have designed two model peptides that mimic the binding loop of the copper chaperone Atx1: the cyclic peptide P(C) c(GMTCSGCSRP) and its linear analogue P(L) (Ac-MTCSGCSRPG-NH2). By using complementary analytical and spectroscopic methods, we have demonstrated that 1:1 complexes are obtained with Cu(I) and Hg(II), whereas 1:1 and 1:2 (M:P) species are successively formed with Zn(II), Cd(II), and Pb(II). The complexation properties of the cyclic and linear peptides are very close, but the cyclic compound provides systematically higher affinity constants than its unstructured analogue. The introduction of a xPGx motif that forms a type II beta turn in P(C) induces a preorganization of the binding loop of the peptide that enhances the stabilities of the complexes (up to 2 orders of magnitude difference for the Hg complexes). The affinity constants were measured in the absence of any reducing agent that would compete with the peptides and range in the order Hg(II) > Cu(I) >> Cd(II) > Pb(II) > Zn(II). This sequence is thus highly selective for Cu(I) compared to the essential ion Zn(II) that could compete in vivo or compared to the toxic ions Cd(II) and Pb(II). Only Hg(II) may be an efficient competitor of Cu(I) for binding to the MxCxxC motif in metalloproteins.