Citrate Does Not Change Uranium Chemical ... - ACS Publications

Dec 18, 2006 - Citrate Does Not Change Uranium Chemical Speciation in Cell Culture Medium but Increases Its Toxicity and Accumulation in NRK-52E Cells...
0 downloads 0 Views 206KB Size
Chem. Res. Toxicol. 2006, 19, 1637-1642

1637

Citrate Does Not Change Uranium Chemical Speciation in Cell Culture Medium but Increases Its Toxicity and Accumulation in NRK-52E Cells Marie Carrie`re,* Ce´line Thiebault, Sarah Milgram, Laure Avoscan, Olivier Proux,† and Barbara Gouget Laboratoire Pierre Su¨e CEA CNRS UMR 9956, CEA Saclay, 91191 Gif sur YVette, France, and Laboratoire de Ge´ ophysique Interne et Tectonophysique, UMR CNRS/UniVersite´ Joseph Fourier, 1381 rue de la piscine, 38400 Saint-Martin-D’He` res, France ReceiVed August 23, 2006

Uranium (U), as a heavy metal, is a strong chemical toxicant, which induces the damage to proximal tubule kidney cells. In order to reproduce U toxicity in Vitro and to avoid precipitation, it is necessary to complex it with a strong ligand such as bicarbonate before dilution with cell culture medium. It was recently shown, in Vitro on the NRK-52E normal renal tubular epithelial cells, that citrate increased the toxicity of U(VI)-bicarbonate complexes. This property was attributed to a change in U speciation, characterized by the occurrence of U(VI)-citrate complexes, which were supposed to be more toxic than U(VI)-bicarbonate. Here, we present the results of extended X-ray absorption fine structure spectroscopy (EXAFS) analyses of the media that were used to expose cells in Vitro. Resulting data show that even when citrate is added to the exposure medium, the predominant species is U(VI)bicarbonate. Nonetheless, citrate increases U(VI) toxicity and accelerates its intracellular accumulation kinetics, without inducing precipitation. This study emphasizes another parameter that modulates U(VI) toxicity for renal tubule cells and further characterizes the mechanisms of U(VI) toxicity. Introduction Uranium (U) is a widely spread heavy metal, naturally present in rocks and surface water. Its intensive use in industry has led to increasing questions about its potential toxicity in the case of accidental exposure of workers and populations. Biokinetics studies have demonstrated that 24 h after acute exposure, U(VI) is detected in the kidneys and bones, where it causes severe toxicity (1). A solution of 1 mM of U(VI) is stable in Vitro in the exposure medium (i.e., MEM1) when it is previously complexed to 10 mM bicarbonate (U(VI)-bicarbonate) or 10 mM citrate (U(VI)citrate) (2). These ligands were chosen because of their physiological relevance: blood contains 25 mM bicarbonate and 0.15 mM citrate (3, 4). Together with serum proteins (5-6), they are the essential ligands of U(VI) in physiological fluids. Citrate is a strong chelator of calcium, continuously reabsorbed by kidney proximal tubules. Its average secretion in the urine flow is 640 mg per day (7-9); it is, thus, permanently present at the vicinity of renal proximal tubule cells, where its concentration varies as a function of metabolism and physiological state. It was proven that 700 µM to 1 mM U is lethal to LLC-PK1 renal cells when prepared as U(VI)-bicarbonate but not as * To whom correspondence should be addressed. Tel: +33 1 69 08 52 35. Fax: +33 1 69 08 69 23. E-mail: [email protected]. † UMR CNRS/Universite ´ Joseph Fourier. 1 Abbreviations: ICP-MS, inductively coupled plasma-mass spectroscopy; EXAFS, extended X-ray absorption fine structure spectroscopy; FAME, French absorption spectroscopy beamline in material and environmental sciences; SEDEM, software package for EXAFS data extraction and modelling; SEM and TEM, scanning and transmission electron microscopy; EDS, energy dispersive spectroscopy; BCA, bicinchoninic acid; MEM, minimum essential medium; DMEM, Dulbecco’s modified Eagle medium; MTT, methyl thiazol tetrazolium.

U(VI)-citrate (2). U(VI)-bicarbonate toxicity was assessed on various renal cell lines including NRK-52E, which was chosen for future studies because of its high sensitivity to U (10). Recently, concentrations of 200-600 µM of U(VI)-citrate were shown to induce NRK-52E dose-response cell death. The CI50 index (U concentration causing the death of 50% of the cells), which was 600 µM for U(VI)-bicarbonate, fell to 300 µM when citrate was added to the exposure medium: U(VI)-citrate is more toxic than U(VI)-bicarbonate (10). According to computer assisted speciation modeling, during in Vitro exposure, i.e., after dilution in MEM, U(VI)-bicarbonate speciation is dominated by carbonated forms: around 60% of U is complexed as UO2(CO3)34-, 27% as UO2Ca(CO3)3(aq), 7% as UO2Ca(CO3)32-, and 6% as UO2(CO3)22-. Other species, such as UO2PO4- or UO2HPO4(aq) complexes, are also present, but their concentration is minor compared to the concentration of uranyl carbonate complexes. When cells are exposed to U(VI)-citrate, 100% of U was calculated to be in the UO2(CIT)24- complex form (10). It was, thus, postulated that the UO2(CIT)24- complex was more toxic than uranyl carbonate complexes. When cells are exposed to U(VI)-bicarbonate, it intensively accumulates in the cytoplasm (11), partly via the sodium-dependent phosphate cotransporter (12). Inside the cell, it localizes in lysosomes, where it coprecipitates with cell phosphates as pH drops, leading to the appearance of urchin-like electron-dense structures in the cell cytoplasm (2) and to major cell perturbations, particularly in the expression of genes implicated in signal transduction, trafficking, and calcium pathway (13). Here, we present the results of direct speciation analyses (EXAFS) of U(VI)-bicarbonate and U(VI)-citrate diluted in MEM. These solutions were then used to expose NRK-52E cells in order to evaluate citrate influence on U(VI) toxicity, cell accumulation, and repartition. This study permits to further

10.1021/tx060206z CCC: $33.50 © 2006 American Chemical Society Published on Web 12/18/2006

1638 Chem. Res. Toxicol., Vol. 19, No. 12, 2006

characterize the mechanisms of U(VI) interaction with cells and emphasizes the importance of carefully reviewing geochemical databases and models throughout the course of their use and the absolute need to experimentally validate them.

Materials and Methods Caution: Uranyl nitrate is hazardous and should be handled carefully. Chemicals. Classical salts, minimum essential medium (MEM) and Dulbecco’s modified Eagle medium (DMEM) cell culture media were from Sigma-Aldrich; their compositions were described previously (10). Uranyl nitrate was from Labosi. Cell Culture. Normal rat kidney NRK-52E proximal cells (CRL1571) were obtained from the American Type Culture Collection. They were subcultured from 1 to 50 passages. Cells were grown in 8 cm2 petri dishes in DMEM containing 4.5 g/L glucose and supplemented with 2 mM L-glutamine, penicillin/streptomycin (50 IU/mL and 50 µg/mL, respectively), and 10% (v/v) fetal bovine serum. They were maintained at 37 °C in a 5% CO2/air incubator and passed at confluence. Uranium Exposure. A 100 mM U(VI) solution was prepared by dissolving uranyl nitrate crystals in milliQ water. To one volume (v) of this U(VI) stock solution was added 5 v of either 0.2 M Na3-citrate (formula: C3H5O(COO)33-, called citrate in the text) or 0.2 M NaHCO3 (called bicarbonate in the text) aqueous solution. Volume was then adjusted to 10 v with ultrapure water to obtain the two 10 mM U(VI)-bicarbonate or U(VI)-citrate stock solutions. These stock solutions were stirred for 30 min to ensure the total dissolution of U(VI) and then filtered through 0.22 µm filter units to sterilize the solution. U(VI) concentration was not modified by filtration (our own unpublished data). These stock solutions were then diluted in serum and antibiotic free cell culture medium, to which was added, when indicated, Hepes buffer (pH 8.5, 10-20 mM final concentration) or various citrate concentrations. All uranyl solutions were prepared just before exposure and were not stocked. Extended X-Ray Absorption Fine Structure Spectroscopy (EXAFS). EXAFS experiments were performed at the French Absorption Spectroscopy Beamline in Material and Environmental Sciences (FAME, BM30B) (14) at the European Synchrotron Radiation Facility storage ring in Grenoble (France), operating in 2/3 bunches mode at 6 GeV. Spectra were recorded, at room temperature, in fluorescence mode at the U LIII-edge, using a double-crystal Si(220) monochromator. Fluorescence detection was achieved using a 30 elements energy resolved Canberra detector. The size, around 300 × 200 µm2 (HxV fwhm), and the position of the X-ray spot on the sample were kept constant during the acquisition. The samples, (i) U(VI)-bicarbonate stock solution and (ii) 600 µM U(VI)-bicarbonate and 300 µM U(VI)-citrate diluted in MEM, were prepared as described in the previous paragraph, just before analysis. Samples were mixed to 30% (v/v) glycerol and placed in quartz capillaries for analysis. Analysis was done in normal atmosphere, and the experimental hutch was ventilated, that is, it contained 370-380 ppm of CO2. On each sample, 4 to 6 spectra were recorded, leading to a total analysis time of 3-5 h. The first spectrum was identical to the last one, indicating that speciation did not change during this lapse of time. The collected scans for a particular sample were averaged and normalized. The EXAFS oscillations were then isolated by removal of the pre-edge background, followed by µ0-removal via spline fitting techniques using the software package for EXAFS data extraction and modeling (SEDEM) (15). The resulting EXAFS curves in the wavevector (k) space was weighted by k3, Fourier transformed into the R space on the [1.85-14.35 Å-1] range using a Hanning window. Back Fourier transform on the [1.15-2.45 Å-1] range allowed filtering of the first shell’s contribution. Simulations in the k filtered space were performed using theoretical functions calculated with the Feff program (16). Refinement was performed by least-square minimization, and EXAFS structural parameters (N, coordination numbers;

Carrie` re et al. R, interatomic distances; and σ2, Debye-Waller factors) were estimated. The amplitude reduction factor, S02, was estimated at 0.835 with the U-bicarbonate reference and then held fixed for the other samples fits. Cell Viability Test. Cells were subcultured in 96-well plates and seeded at 6,000 cells per well. Exposure was done on 90% confluent monolayers, that is, 3 days after subculture; 100 µL of exposure solutions were deposited on the cells. U(VI) cytotoxicity was evaluated with the classical colorimetric methyl thiazol tetrazolium (MTT) test (17) on the basis of the detection of the reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide to a blue formazan compound by the mitochondrial succinate dehydrogenase enzyme of living cells. At the end of the exposure period, 10 µL of a 20 mg/mL MTT solution (Sigma) was added to the 100 µL exposure solution. After 2 h of incubation at 37 °C, the wells were emptied, blue formazan crystals were dissolved with 100 µL of MTT solubilization solution (10% Triton X-100, 0.1 N HCl in 2-propanol), and absorbance at 570 nm was measured using a Stat Fax-2100 microplate reader (Awareness Technology, Inc.). Quantification of U Precipitation and Intracellular Accumulation. For the quantification of U precipitation, 100 µL of each exposure solution was placed in a 96-well plate that did not contain cells. The plate was incubated for 24 h in the 37 °C 5% CO2 incubator. After the incubation period, the solutions were centrifuged for 30 min at 15,000g, and U(VI) concentration was quantified by inductively coupled plasma-mass spectroscopy (ICPMS) in the supernatant. Quantification of U intracellular accumulation was also measured by ICP-MS. For this purpose, cells were seeded at 75,000 cells per well in 12-well plates and exposed to U(VI). After the exposure period, the cells were washed twice with 10 mM NaHCO3 and lysed with cell culture lysis reagent (Promega). Total protein content was determined using the bicinchoninic acid (BCA) assay (Sigma). For ICP-MS analyses, samples were acidified with ultrapure 65% nitric acid, diluted in ultrapure water, and U concentration was measured using an X7 series quadrupole (Thermo Electron Corporation) calibrated with a SPEX U standard range (0-5 ppb). Scanning and Transmission Electron Microscopy (SEM and TEM). For the SEM observation of U precipitates, semipermeable transwell-clear membranes were exposed for 24 h to uranyl solutions in a 37 °C 5% CO2/air incubator. Membranes were then rinsed with water, dried, covered with a thin carbon film, and observed with a Stereoscan 120 (Cambridge) SEM. Images were captured with Maxview software, and energy dispersive X-ray spectroscopy (EDS) analysis was performed at an acceleration voltage of 20 kV with the IDFix microanalysis software (SAMx Microanalysis Application Software, Levens, France). For the TEM observation of cells, the cells were exposed to 200 µM U(VI)-citrate diluted in MEM medium and prepared and observed as described elsewhere (18).

Results and Discussion U(VI) Speciation in Cell Exposure Solution. To validate the accuracy of U(VI) speciation model which suggested that UO2(CIT)24- is more toxic to NRK-52E cells than uranyl carbonate complexes (10), we analyzed U speciation before cell exposure when it was prepared as U(VI)-bicarbonate and U(VI)-citrate and then diluted in MEM. The coordination of U(VI) was studied by EXAFS spectroscopy at the U LIII-edge. The three k3-weighted U EXAFS spectra, corresponding to the 10 mM U(VI)-bicarbonate aqueous stock solution and U(VI)-citrate and U(VI)-bicarbonate diluted in MEM, were found to be similar (Figure 1). Spectra closely resembled the already described U(VI)-bicarbonate spectrum (19) but not the U(VI)-citrate spectrum (20). Data were analyzed and modeled (Table 1). Concerning U(VI)bicarbonate aqueous reference (pH 7.1), the presence of two

Citrate: U(VI) Cytotoxicity Enhancer

Chem. Res. Toxicol., Vol. 19, No. 12, 2006 1639

k3-weighted

Figure 1. U LIII-edge EXAFS spectra of U solutions. A 10 mM U(VI)-bicarbonate reference solution (A), this same solution diluted to 600 µM in MEM cell culture medium (B), and a 10 mM U(VI)-citrate reference solution diluted to 300 µM in MEM (C) were analyzed. Table 1. EXAFS Analysis of U(VI) Speciation in Exposure Mediuma sample

bond

N

R (Å)

σ2 (Å-2)

reference

U-bicarbonate U-O ax. 2 (fixed) 1.83 ( 0.01 0.0014 ( 0.0002 this work aqueous stock solution U-O equ. 6 (fixed) 2.46 ( 0.02 0.0053 ( 0.0005 MEM U-BIC U-O ax. 2.1 ( 0.2 1.82 ( 0.01 0.0012 ( 0.0002 U-O equ. 6.9 ( 0.7 2.44 ( 0.02 0.0076 ( 0.0007 MEM U-CIT U-O ax. 2.5 ( 0.2 1.82 ( 0.01 0.0012 ( 0.0002 U-O equ. 6.6 ( 0.6 2.46 ( 0.02 0.0055 ( 0.0005 U-bicarbonate U-O ax. 2 aqueous U-O equ. 6 solution

1.79 2.46

0.0026 0.0075

Allen (1995)

U-citrate aqueous solution

1.78 2.38

0.0035 0.0068

Allen (1996)

U-O ax. 2.3 U-O equ. 5.0

a The shift in threshold energy, ∆E0, was allowed to vary as a global parameter in each of the fits (i.e., the same ∆E0 was used for each shell). k-range ) 1.85-14.35 Å-1; dk ) 2 Å-1; k-window ) hanning; k-weight ) 3; and R-range ) 1.15-2.45. N, coordination number; R, interatomic distance; σ2, Debye-Waller factor; MEM U-BIC, 600 µM U(VI)bicarbonate prepared in MEM medium; MEM U-CIT, 300 µM U(VI)citrate prepared in MEM; U-O ax., axial U-O bond (uranyl); U-O equ., equatorial U-O bond.

axial oxygen atoms was hypothesized; their distance to the U atom was calculated to be 1.83 ( 0.01 Å. The second coordination shell was composed of 6 oxygen atoms (fixed value), at a distance of 2.46 ( 0.02 Å. These values are consistent with published data (19). According to Allen et al. (20), the U local structure in the U(VI)-citrate complex is the following: 2.3 axial oxygen atoms at 1.78 Å and 5 equatorial oxygen atoms at 2.38 Å. The distance of equatorial oxygen atoms in these two complexes are, thus, sufficiently different to permit the identification of one or the other chemical form. U was then prepared as U(VI)-bicarbonate or U(VI)-citrate, then diluted in MEM at a final concentration of 600 µM for U(VI)-bicarbonate (pH 7.2, MEM U-BIC in Table 1) and 300 µM for U(VI)-citrate (pH 7.1, MEM U-CIT in Table 1). Exposure to these concentrations of U caused the death of 50% of NRK-52E cells (10). After the EXAFS analysis of these solutions, the best fit to data was obtained with a first coordination shell consisting of 2.1 ( 0.2 (MEM U-BIC) or

Figure 2. Uranyl citrate cytotoxicity. (A) Cell death induced by U(VI) was evaluated in various conditions. U(VI) was prepared as a citrate stock solution and diluted in MEM (9) or as a bicarbonate stock solution and diluted in MEM (2), in MEM containing 2 mM citrate (4), 4 mM citrate (×), or 4 mM citrate and 17.5 mM Hepes (b, O). (B) The kinetics of cell death induced by 320 (b, O), 340 (9), or 400 µM (2, 4) U(VI) in the presence of 4 mM citrate (B).

2.5 ( 0.2 (MEM U-CIT) axial oxygen atoms at a distance of 1.82 ( 0.01 Å (MEM U-BIC and MEM U-CIT). On the second shell, corresponding to the equatorial ligands of the uranyl group, the U atom is surrounded by 6.9 ( 0.7 (MEM U-BIC) or 6.6 ( 0.6 (MEM U-CIT) oxygen atoms, at a distance of 2.44 ( 0.02 Å (MEM U-BIC) and 2.46 ( 0.02 Å (MEM U-CIT). Spectral comparison with already published data concerning U(VI)-bicarbonate and U(VI)-citrate references, thus, showed that the predominant species in both of our samples is U(VI)-bicarbonate. The UO2(CIT)24- complex was not observed in these experimental conditions, or its concentration was below the detection limit, that is, about 5% of the total U(VI) concentration. The discrepancy between U(VI) speciation modeling and experimental speciation can be explained by the complexity of U(VI) chemistry in aqueous media, which contain multiple potential ligands, and/or the uncertainty in the modeling scenario and the model constants (21). The complexation constant of citrate with either uranyl or other ions such as calcium may be unreliable. Moreover, thermodynamic constant databases were originally constructed for geological purposes in order to model chemical speciation of the elements when equilibrium is reached and not for dynamic biological systems. Influence of Citrate Addition on U(VI) Toxicity. Because speciation did not explain the difference in toxicity between U(VI)-bicarbonate and U(VI)-citrate, we further investigated the influence of the presence of citrate during cell exposure. U was prepared as U(VI)-citrate and diluted in MEM or as U(VI)-bicarbonate and diluted in MEM to which was added citrate and/or Hepes. As previously described, the concentration leading to a 50% loss of viability (CI50) was 450 µM for U(VI)bicarbonate, without the addition of citrate in the exposure medium (Figure 2A). The addition of 4 mM citrate led to an increase of U(VI)-bicarbonate cytotoxicity: The CI50 value decreased to 275 µM. The addition of Hepes did not change U(VI) cytotoxicity. The CI50 value was also 275 µM when U(VI) was directly prepared as U(VI)-citrate. Dose-response cell death was observed regardless of the citrate concentration, and U(VI) toxicity increased as citrate concentration rose. The CI50 value varied from 275 µM at 4 mM citrate to 325 µM at 2 mM

1640 Chem. Res. Toxicol., Vol. 19, No. 12, 2006

Carrie` re et al.

Figure 4. U(VI) intracellular accumulation. U(VI) concentration was measured on total cell lysates after exposure to increasing U(VI) concentrations for 2, 4, and 6 h. The results were normalized to the total cell protein concentration.

Figure 3. Variation of U(VI) solubility in the exposure solution. (A) Precipitation of U(VI) in the exposure medium was measured as a function of U(VI) concentration, when the exposure medium contained 400 µM (9), 800 µM (4), or 4 mM (O) citrate. (B) Precipitates were observed with a scanning electron microscope coupled to EDS analysis, at an accelerating voltage of 20 kV (C).

citrate (Figure 2A). Interestingly, when the U concentration exceeded 600-700 µM (U(VI)-citrate), toxicity was abolished. The kinetics of cell death occurrence was analyzed. U(VI)bicarbonate was prepared and diluted in MEM medium to which was added 4 mM citrate. When cells were exposed to 400 µM U(VI), mortality appeared rapidly, and cell death reached 50% after 10 h. Mortality was maximal, that is, 80%, after 13 to 14 h of exposure. Fifty percent cell mortality was observed after 16 h of exposure to 340 µM U(VI) or after 22 h of exposure to 320 µM U(VI) (Figure 2B). In comparison, when cells were treated with 400 µM U(VI)-bicarbonate without the addition of citrate, only 25% of the cells were dead after 24 h of exposure (10). U(VI), thus, induced cell mortality more rapidly when the exposure medium contained citrate. The citrate effect was drastic because the CI50 value dropped from 450 to 275 µM, that is, 50% of the cells died when they were exposed to the half U(VI) dose. Uranium Toxicity Is Due to Soluble Chemical Forms. In some conditions of exposure, precipitates were observed on cells. In a former study, it was shown that U(VI) toxicity on rat alveolar macrophages was mainly due to insoluble forms (22). It was of great importance to determine the implication of these precipitates in U(VI)-citrate toxicity in Vitro and identify their chemical composition. U(VI) concentration was measured after the incubation of exposure media for 24 h at 37 °C and 5% CO2. U(VI) precipitation was shown to occur between 400 and 500 µM when the exposure medium contained 4 mM of citrate, between 450 and 500 µM when the exposure medium contained 800 µM of citrate, and between 550 and 600 µM when the exposure medium contained 400 µM of citrate (Figure 3A). At these concentrations, cell mortality was already maximal (80%). This also means that 80% cell death occurred with non-precipitated U(VI) and consequently that toxicity was caused by soluble chemical forms of U(VI). The observed precipitates were spherical, urchin-like, and around 1-2 µm in diameter (Figure 3B). EDS analysis showed that they contained U, oxygen, and phosphorus (Figure 3C). They then must be U(VI)-phosphate precipitates. Such precipitates would certainly be lethal if they appeared in ViVo, but the conditions of contamination would be extreme

Figure 5. Transmission electron micrograph of NRK-52E cells exposed to U(VI). Cells were treated for 24 h with 200 µM U(VI) and 2 mM citrate. Electron-dense precipitates are present inside and outside the cells. Magnification: ×10,000.

because the U concentration needed for them to occur are very high. In environmental conditions of contamination, U toxicity is, thus, mainly caused by soluble chemical forms. Strikingly, when 4 mM citrate was added to the exposure medium, no precipitation was observed for U concentrations greater than 700 µM. Influence of Citrate Addition on U(VI) Intracellular Accumulation. We then wondered whether the increase in U(VI) toxicity was the consequence of an increased intracellular accumulation, which would mean that the citrate increased U(VI) bioavailability, or whether U(VI) was accumulated to the same extent but on another chemical form, which would target it to other intracellular compartments or metabolic pathways, leading to more severe damages in the cell. First, U(VI) intracellular accumulation was quantified. When exposed for 6 h to 400 µM U(VI) in a medium containing 4 mM citrate, NRK-52E cells accumulated 2.0 ( 0.3 mg U/g proteins. Accumulation increased as a function of time and U(VI) concentration. When cells were exposed to 500 µM U(VI), intracellular accumulation of U(VI) could be observed as early as after 4 h of exposure, with 6.3 ( 0.3 mg U/g proteins, and reached 90 ( 5 mg U/g proteins after 6 h. For U(VI) concentrations higher than 600 µM, accumulation decreased (Figure 4). TEM observation of cells after the accumulation of U(VI) showed that U(VI) precipitated as dense urchin-like structures in intracellular compartments (Figure 5A), as already observed by Mirto et al. on LLC-PK1 cells exposed to 600 µM U(VI)-

Citrate: U(VI) Cytotoxicity Enhancer

bicarbonate (2). Nevertheless, in this study, the precipitates were observed inside the cells for U concentrations ranging from 200 to 500 µM of U(VI)-citrate, that is, for low U(VI) concentrations. These precipitates were progressively expulsed from cells (Figure 5B, C, and D). As observed with U(VI) toxicity, the citrate effect was strong. U(VI) storage in cells exposed to 500 µM started more than three times faster than that in the absence of citrate. No U(VI) accumulation was observed before 12 h of exposure to 600 µM U(VI)-bicarbonate, and after 12 h exposure, accumulation only reached 8.0 ( 0.3 mg U/g proteins (18). U intracellular accumulation decreased when cells were exposed to more than 600 µM of U. As described in previous paragraphs, no toxicity and no U precipitation were detected for U concentrations greater than 700 µM in the presence of citrate. Mirto et al. did not observe intracellular precipitates when they exposed LLC-PK1 cells to 700 µM U(VI)-citrate (2), nor did they observe any toxicity. One possible explanation for this phenomenon is that uranyl can form polymeric complexes, either with citrate in stock solution or with carbonate in diluted solutions (19, 23-24). These complexes, still soluble in exposure media, potentially polymerize and flocculate. They would consequently reduce the activity of uranyl ion outside the cell, that is, its bioavailability. They are not eliminated with U precipitates, but they may trap U in a nontoxic, non bioavailable chemical form.

Conclusions The major action of citrate was to increase U(VI) toxicity and intracellular accumulation, which can be defined as U(VI) in Vitro bioavailability. The implicated mechanisms can be multiple: citrate can have a direct action on cell membrane permeability. Citrate can also act on the cell phenotype, inducing the expression or the relocalization to cell membrane of transporters such as sodium-dependent phosphate cotransporters, whose implication in U(VI) internalization have already been described (12). U(VI) can also be co-transported with citrate via a specific transporter such as the renal sodium-dicarboxylate (NADC1) co-transporter (25), whose opening may be induced by the presence of citrate in the exposure medium. No U(VI) influence has been shown on citrate metabolism (13); however, that citrate modifies U(VI) metabolization it is not excluded. Finally and more probably, competition between uranyl and calcium for the same sites on the cell membrane has been suggested earlier (1). These sites might be membrane transporters or endocytosis initiation sites. It is possible that citrate, while complexing calcium, increases the accessibility of uranyl to these sites and, thus, enhances its bioavailability. This hypothesis is further reinforced by previously published results, which showed that when Ca concentration was increased in the exposure medium, U(VI)-bicarbonate cytotoxicity and intracellular accumulation decreased (11). This result was confirmed in the presence of citrate in the exposure medium (our own unpublished data). Moreover, it was recently shown that uranyl created perturbations in the calcium signalling pathway (13). These hypotheses will sooner be tested. Finally, this study confirms that it is important to take into account the presence of citrate in biological fluids (blood, urine) when studying radionuclide toxicity, even if citrate concentration in the bloodstream or in kidney tubules will never reach more than transiently such high concentrations. It also enlightens the need to carefully review thermodynamic data and modeling and to validate modeling results by experimental data in order to

Chem. Res. Toxicol., Vol. 19, No. 12, 2006 1641

obtain clear conclusions on the influence of speciation on radionuclide bioavailability and toxicity. Acknowledgment. This work was partly supported by the French “Environmental Nuclear Toxicology” program. We thank J. M. Verbavatz and R. Gobin for TEM sample preparation and observation and C. Mariet and F. Carrot for ICP-MS analyses.

References (1) Leggett, R. W. (1989) The behaviour and chemical toxicity of U in the kidney: a reassessment. Health Phys. 57, 365-383. (2) Mirto, H., Henge-Napoli, M. H., Gibert, R., Ansoborlo, E., Fournier, M., and Cambar, J. (1999) Intracellular behaviour of uranium(VI) on renal epithelial cell in culture (LLC-PK1): influence of uranium speciation. Toxicol. Lett. 104, 249-256. (3) Zeevaart, J. R., Jarvis, N. V., Louw, W. K. A., and Jackson, G. E. (2001) Metal-ion speciation in blood plasma incorporating the tetraphosphonate, N,N-dimethylenephosphonate-1-hydroxy-4-aminopropilydenediphosphonate (APDDMP), in therapeutic radiopharmaceuticals. J. Inorg. Biochem. 83, 57-65. (4) Letkeman, P. (1996) Computer-modelling of metal speciation in human blood serum. J. Chem. Educ. 73, 165-170. (5) Vidaud, C., Dedieu, A., Basset, C., Plantevin, S., Dan, I., Pible, O., and Que´meneur, E. (2005) Screening of human serum proteins for uranium binding. Chem. Res. Toxicol. 18, 946-953. (6) Van Horn, J. D., and Huang, H. (2006) Uranium(VI) bio-coordination chemistry from biochemical, solution and protein structural data. Coord. Chem. ReV. 250, 765-775. (7) Melnick, J. Z., Preisig, P. A., Alpern, R. J., and Baum, M. (2000) Renal citrate metabolism and urinary citrate excretion in the infant rat. Kidney Int. 57, 891-897. (8) Kramer, L., Bauer, E., Joukhadar, C., Strobl, W., Gendo, A., Madl, C., and Gangl, A. (2003) Citrate pharmacokinetics and metabolism in cirrhotic and noncirrhotic critically ill patients. Crit. Care Med. 31, 2450-2455. (9) Nicar, M. J., Skurla, C., Sakhaee, K., and Pak, C. J. (1983) Low urinary citrate excretion in nephrolithiasis. Urology 21, 8-14. (10) Carrie`re, M., Avoscan, L., Collins, R., Carrot, F., Khodja, H., Ansoborlo, E., and Gouget, B. (2004) Influence of uranium speciation on normal rat kidney (NRK-52E) proximal cell cytotoxicity. Chem. Res. Toxicol. 17, 446-452. (11) Carrie`re, M., Khodja, H., Avoscan, L., Carrot, F., and Gouget, B. (2005) Uranium(VI) complexation in cell culture medium: influence of speciation on Normal Rat Kidney (NRK-52E) cell accumulation. Radiochim. Acta 93, 691-697. (12) Muller, D., Houpert, P., Cambar, J., Henge´-Napoli, M. H. (2006) Role of the sodium-dependent phosphate co-transporters and of the phosphate complexes of uranyl in the cytotoxicity of uranium in LLCPK1 cells. Toxicol. Appl. Pharmacol. 214, 166-177. (13) Pratt, O., Berenguer, F., Malard, V., Tavan, E., Sage, N., Steinmetz, G., and Quemeneur, E. (2005) Transcriptomic and proteomic responses of human renal HEK293 cells to uranium toxicity. Proteomics 5, 297306. (14) Proux, O., Biquard, X., Lahera, E., Menthonnex, J.-J., Prat, A., Ulrich, O., Soldo, Y., Tre´visson, P., Kapoujvan, G., Perroux, G., Taunier, P., Grand, D., Jeantet, P., Deleglise, M., Roux, J. P., and Hazemann, J. L. (2005) FAME: A new beamline for X-ray absorption investigations of very-diluted systems of environmental, material and biological interests. Phys. Scr. 115, 970-973. (15) Aberdam, D. (1998) SEDEM, a software package for EXAFS data extraction and modelling. J. Synchrotron Radia. 5, 1287-1297. (16) Rehr, J. J., Zabinski, S. I., and Albers, R. C. (1992) High-order multiple scattering calculations of X-ray-absorption fine structure. Phys. ReV. Lett. 69, 3397-3400. (17) Mossman, T. (1983) Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55-63. (18) Carrie`re, M., Gouget, B., Gallien, J. P., Avoscan, L., Gobin, R., Verbavatz, J. M., and Khodja, H. (2005) Cellular distribution of uranium after acute exposure of renal epithelial cells: SEM, TEM and nuclear miscroscopy analysis. Nucl. Instrum. Methods Phys. Res., Sect. B 231, 268-273. (19) Allen, P. G., Bucher, J. J., Clark, D. L., Edelstein, N. M., Ekberg, S. A., Gohdes, J. W., Hudson, E. A., Kaltsoyannis, N., Lukens, W. W.,

1642 Chem. Res. Toxicol., Vol. 19, No. 12, 2006 Neu, M. P., Palmer, P. D., Reich, T., Shuh, D. K., Tait, C. D., and Zwick, B. D. (1995) Multinuclear NMR, RAMAN, EXAFS, and X-ray diffraction studies of uranyl carbonate complexes in near-neutral aqueous solutionsX-ray structure of [C(NH2)(3)](6)[(UO2)(3)(CO3)(6)]-6.5H(2)O. Inorg. Chem. 34, 47974807. (20) Allen, P. G., Shuh, D. K., Bucher, J. J., Edelstein, N. M., Reich, T., Denecke, M. A., and Nitsche, H. (1996) EXAFS determinations of uranium structures: the uranyl ion complexed with tartric, citric, and malic acids. Inorg. Chem. 35, 784-787. (21) Denison, F. H., and Garnier-Laplace, J. (2005) The effects of database parameter uncertainty on uranium(VI) equilibrium calculations. Geochim. Cosmochim. Acta 69, 2183-2191.

Carrie` re et al. (22) Lizon, C., and Fritsch, P. (1999) Chemical toxicity of some actinides and lanthanides towards alveolar macrophages: an in vitro study. Int. J. Radiat. Biol. 75, 1459-1471. (23) Pasilis, S. P., and Pemberton, J. E. (2003) Speciation and coordination chemistry of uranyl(VI)-citrate complexes in aqueous solution. Inorg. Chem. 42, 6793-6800. (24) Thue´ry, P. (2006) Uranyl ion complexation by citric and tricarballylic acids: hydrothermal synthesis and structure of two- and threedimensional uranium-organic frameworks. Chem. Commun. 8, 853855. (25) Pajor, A. M. (1999) Sodium-coupled transporters for Krebs cycle intermediates. Annu. ReV. Physiol. 61, 663-682.

TX060206Z