Chromium Oxidation State Imaging in Mammalian ... - ACS Publications

Sep 7, 2005 - Recherche Scientifique (CNRS), Universite´ de Bordeaux 1, BP 120, Chemin du Solarium,. 33175 Gradignan, France, and X-ray Microscopy, ...
0 downloads 0 Views 453KB Size
1512

Chem. Res. Toxicol. 2005, 18, 1512-1519

Articles Chromium Oxidation State Imaging in Mammalian Cells Exposed in Vitro to Soluble or Particulate Chromate Compounds Richard Ortega,*,† Barbara Fayard,‡ Murielle Salome´,‡ Guillaume Deve`s,† and Jean Susini‡ Laboratoire de Chimie Nucle´ aire Analytique et Bioenvironnementale, Centre National de la Recherche Scientifique (CNRS), Universite´ de Bordeaux 1, BP 120, Chemin du Solarium, 33175 Gradignan, France, and X-ray Microscopy, ID-21, European Synchrotron Radiation Facility (ESRF), BP 220, 38043 Grenoble, France Received September 22, 2004

Hexavalent chromium compounds are known carcinogens for the respiratory tract in humans. The mechanism of cell transformation by hexavalent chromium compounds is not fully understood although a role for intracellular reduction is sought. The aim of this study was to determine the distribution of Cr valence states in human cells after in vitro exposure to soluble or particulate chromium compounds. A synchrotron X-ray-based microprobe was used to investigate the cellular reduction of Cr(VI) and to image chromium oxidation states in cells. It was shown that soluble Cr(VI) compounds are fully reduced to Cr(III) in cells. Cr(III) is homogeneously distributed within the cell volume and therefore present within the nucleus. In the case of low solubility particulate chromate compounds, Cr(VI) can coexist in the cell environment, as particles in the perinuclear region, together with intracellular and intranuclear Cr(III). Chemical distribution maps also suggest that intracellular Cr(III) originates from extracellular dissolution and reduction of lead chromate rather than from intracellular engulfed particles. The possible stronger carcinogenicity of low solubility chromate vs soluble chromate compounds may derive from the combinative genotoxic effects of intranuclear Cr(III) and the persistent exposure to a strong oxidant, Cr(VI).

Introduction Solve et coagula, dissolve and coagulate: This ancient adage of alchemists continues to summarize the activity of analytical chemists, which consists of separating and analyzing matter into its elemental components, atoms, molecules, or groups of molecules. The alchemist’s alembics are nowadays substituted by sophisticated instruments, often borrowed from physicists, to perform modern spectroscopy. Until now, there was no analytical tool for the direct chemical mapping of trace metals oxidation states in microscopic samples such as living cells. Recent improvements in hard X-rays optics have enabled the development of X-ray microprobes on third generation synchrotron sources and the conduction of spatially resolved X-ray spectroscopies for microscopic chemical imaging of trace elements valence states (1-4). In this experiment, synchrotron radiation-based microprobe was used to study chromium quantitative distribution and oxidation states in single mammalian cells exposed in vitro to soluble or insoluble Cr(VI) compounds at the subcellular level. * To whom correspondence should be addressed. Tel: +33 557 120 907. Fax: +33 557 120 900. E-mail: [email protected]. † Centre National de la Recherche Scientifique. ‡ European Synchrotron Radiation Facility.

The chromium biological activity depends strongly upon its oxidation state and solubility. Several hexavalent chromium, Cr(VI), compounds are known to be human carcinogens, particularly for the respiratory tract (5, 6), whereas trivalent chromium, Cr(III), compounds are considered nontoxic and noncarcinogenic (5, 7). Cr(VI) as an oxyanion form can penetrate the cell via the sulfate transport system while Cr(III) cannot enter the cell. The mechanism of Cr(VI) compounds carcinogenesis is not yet fully elucidated. Although the genotoxicity of Cr(VI) compounds has been extensively studied (8, 9), their carcinogenic potential in humans usually differs from their genotoxic properties determined in cellular models. For instance, highly soluble chromate compounds induce a broad range of genetic effects in cellular models, whereas poorly soluble and insoluble Cr(VI) compounds induce genotoxic effects in bacteria only when solubilized in acids or in alkali. A number of in vivo animal carcinogenesis bioassays have illustrated that soluble hexavalent chromates are either noncarcinogenic or at best weakly carcinogenic after multiple exposure (5, 10). In contrast, the moderately to highly insoluble particulate forms of chromate are potently carcinogenic (5). It has been proposed that the carcinogenicity of chromium compounds depends on both valence state and solubility, with the least water soluble Cr(VI) compounds being the

10.1021/tx049735y CCC: $30.25 © 2005 American Chemical Society Published on Web 09/07/2005

Chromium Oxidation State Imaging in Mammalian Cells

Chem. Res. Toxicol., Vol. 18, No. 10, 2005 1513

Table 1. Solubility in Water (mol/L) of Chromium(VI) Compounds and Their Respective Carcinogenicity According to IARC Cr(VI) compounds

solubility (mol/L)

carcinogenicity

sodium dichromate Na2Cr2O7 chromium trioxide CrO3 calcium chromate CaCrO4 strontium chromate SrCrO4 barium chromate BaCrO4 lead chromate PbCrO4 zinc chromate ZnCrO4

6.87 1.66 0.027 5.9 × 10-3 1.3 × 10-5 1.1 × 10-7 insoluble

limited evidence limited evidence sufficient evidence sufficient evidence inadequate evidence sufficient evidence sufficient evidence

most active (11-15) (Table 1). The aim of this study was to observe the intracellular distribution of chromium oxidation states after cell exposure to soluble and low solubility chromate compounds, to better understand the differential carcinogenicity of these compounds.

Material and Methods Caution: Cr(VI) compounds are human carcinogens, and appropriate precautions should be taken in handling of these materials. Cell Cultures and Samples Preparation. The AA8 Chinese hamster ovary cell line (CHO-AA8)1 is a derivative of the Chinese hamster ovarian K1 cell line (CHO-K1) particularly used for gene mutation assays (16). CHO-AA8 was grown in R-minimum essential medium with penicillin (100 units/mL) and streptomycin (100 µg/mL) and 10% heat-inactivated fetal bovine serum. Cells presented a 12 h doubling time. In addition, the Institut Gustave Roussy ovarian cell line 1 (IGR-OV1) (17) was also analyzed, despite its tumorigenic properties, because of its conserved epithelial morphology and known suitability for chemical element mapping using X-ray microanalysis (4, 18). IGR-OV1 cells were grown in Roswell Park Memorial Institute (RPMI) 1640 with penicillin (100 units/mL) and streptomycin (100 µg/mL) and 10% heat-inactivated fetal bovine serum. Both cell lines were grown at 37 °C, 5% CO2 in air atmosphere saturated with water. According to already published sample preparation protocol (18), cells were grown as a monolayer onto 2 µm thick polycarbonate foil or 0.2 µm Formvar film and exposed to appropriate concentrations of chromate compounds. The Cr(VI) model compounds were lead chromate, PbCrO4 (low solubility compound), and sodium chromate, Na2CrO4‚4H2O (soluble compound). Fine PbCrO4 particles were prepared by grinding the powder using a mortar and pestle. The particles size was smaller than 3 µm as observed by optical microscopy, with a majority of particles of about 1 µm diameter. Lead chromate particles were then washed with distilled water, airdried, weighed, and placed into analytical grade ethanol for sterilization. The particles were maintained in suspension using a vortex mixer, and cells were treated with this freshly prepared stock solution. After exposure to Cr(VI) compounds, cells were twice rinsed vigorously with PBS to remove traces of extracellular chromium. Cells were cryofixed into liquid nitrogen chilled isopentane at -164 °C for 20 s and either stored in their frozen hydrated state into liquid nitrogen until analysis or freeze-dried overnight at -35 °C and stored into a desiccator. Cytotoxicity Assays. Inhibition of cell growth was determined by cell counting with a hemocytometer. First, a defined number of cells, 5 × 104 for IGR-OV1 and 2.5 × 104 for CHOAA8, was seeded on a 25 mm diameter 12 wells culture plate and allowed to grow for 48 h. Cell cultures were treated for 1 h with 10 µM Na2CrO4 or 24 h with 0.1 and 0.5 µg/cm2 PbCrO4 (CHO-AA8) and 1 µg/cm2 PbCrO4 (IGR-OV1). Control samples 1 Abbreviations: CHO-AA8, Chinese hamster ovary AA8 cell line; CHO-K1, Chinese hamster ovary K1 cell line; IGR-OV1, Institut Gustave Roussy ovarian cell line 1; RPMI, Roswell Park Memorial Institute; XANES, X-ray absorption near edge structures; IARC, International Agency for Research on Cancer; HSAE, human small airway epithelial; XRF, X-ray fluorescence.

Figure 1. Cr K-edge XANES spectra from Cr(VI) reference compounds, PbCrO4 and Na2CrO4, and Cr(III) reference compound, CrCl3. The prominent preedge peak at 5993.5 eV is characteristic of the Cr(VI) oxidation state. Energy scan from 5950 to 6100 eV at low energy resolution (1 eV/monochromator step) from 5950 to 5980 eV and from 6050 to 6100 eV and at high energy resolution (0.33/monochromator step) between 5980 and 6050 eV; time scan, 1 s/step. were not treated with chromate compounds. The medium was then removed, cells were rinsed with PBS, and fresh medium was added. After 48 h of growth, cells were counted when control samples were still in the logarithmic growth phase. Cells were treated with trypsin and harvested with PBS, and nonviable cells were stained with Trypan blue. Eight wells were counted for each condition of treatment, and the experiments were repeated twice. X-ray Microscopy and Oxidation State Mapping. X-ray absorption near edge structure (XANES) experiments and oxidation state mapping were performed at the European Synchrotron Radiation Facility (ESRF) using the scanning X-ray microscope on beam line ID21 (19). The electron storage ring was operated at 6.03 GeV, and the electron current was in the range of 170-200 mA in 2 × 1/3 filling mode. A fixed exit doublecrystal monochromator, Si(220), was used in order to produce a highly energy resolved monochromatic beam (10-4). Two photodiodes were used to measure the incident and transmitted beam intensities. The X-ray beam was focused down to a micrometer size using Fresnel zone plates (20). Two zone plates were used for this study, a high flux zone plate (diameter ) 1040 µm), giving a spot size (V × H) of 1.0 × 3.0 µm2 with a 22.9% transmission efficiency at 6.0 keV, and a high-resolution zone plate (diameter ) 70 µm), providing a spot size (V × H) of 0.5 × 1.0 µm2, with 20% transmission efficiency at 6.0 keV. Chromium K-L2,3 lines were detected with a Si(Li) energy dispersive X-ray detector. XANES spectra were calibrated in energy against the absorption edge position of Cr(0) metallic foil, whose first inflection point is 5989 eV. In addition, PbCrO4, Na2CrO4, and CrCl3 were used respectively as references for Cr(VI) and Cr(III) compounds for XANES preedge feature energy calibration (Figure 1). A prominent preedge peak at 5993.5 eV occurred when chromium was present as Cr(VI), caused by a bound state 1s to 3d transition (21). Chromium oxidation state imaging involved scanning a sample in two dimensions with the monochromator set first to the appropriate energy of preedge absorption peak 5993.5 eV, characteristic of Cr(VI) oxidation state, and then scanning again the sample with the monochromator energy position set at a value where the intensities of both oxidation states are observed, 6020 eV for total Cr (Figure 1). XANES and oxidation state mapping experiments were performed at room temperature in air. Blanks with a sample holder without sample were carried out to check that no chromium fluorescence could arise from the microscope chamber or from the sample holder. Cryomicroanalysis at 170 K under a jet stream of liquid nitrogen and dry nitrogen atmosphere was

1514

Chem. Res. Toxicol., Vol. 18, No. 10, 2005

Ortega et al.

Figure 2. Intracellular distribution of K and Cr oxidation state in IGR-OV1 cells exposed to a soluble hexavalent chromium compound, Na2CrO4, at 10 µM for 1 h and corresponding optical microscopy. (a) Group of cells. (b) Single cell analysis. X-ray beam energy, 6020 eV; beam spatial resolution (V × H), 1 × 3 µm2; scan, 1 µm/pixel; 4 s/pixel; color scale in counts per pixel. As shown in Figure 3, no Cr(VI) could be detected at E ) 5993.5 eV. K and Cr(III) are present within the cytosol (C), nucleus (N), and nucleoli (n). The X-ray yield for K and Cr(III) increases from cytosol to nucleus and nucleoli, proportionally to the cell volume. K and Cr(III) intracellular concentration distributions can be considered as homogeneous. also performed to validate the freeze-drying process. Quantitative analysis of data was made with Axil software (22).

Results Intracellular Distribution of Chromium: Soluble Chromate. The distribution of K and Cr in IGR-OV1 cells exposed to a soluble hexavalent chromium compound, Na2CrO4, at 10 µM for 1 h is presented in Figure 2. Potassium distribution is indicative of cell position and is used as an internal reference for chemical element distributions. Using an X-ray beam at 6020 eV, it is shown either on a group of cells (Figure 2a) or for single cell analysis (Figure 2b) that Cr is present within the cytosol and the nucleus. The Cr content is higher in nuclei than in cytosol; however, the Cr concentration distribution (content normalized by volume or by total mass) can be considered as homogeneous within the cell volume, thicker regions of the cell, such as the nuclei, providing a stronger X-ray fluorescence (XRF) signal. By scanning the same area at the preedge peak energy (5993.5 eV), we have found that no Cr(VI) could be detected in cells exposed to the soluble compound Na2CrO4 (Figure 3b). Therefore, the Cr XRF signal measured at 6020 eV (Figure 3a) can be attributed solely to trivalent chromium. In addition, XRF spectra were extracted from distinct zones of the chemical maps corresponding to cellular ultrastructures. As indicated in Figure 4a-c, trivalent chromium was detected in the cytosol, nucleus, and nucleolus of cells. This pattern was observed for both cell lines, IGR-OV1 and CHO-AA8, and for all analyzed cells. Quantitative analysis of Cr/K ratios is presented in Table 2. Potassium was chosen as an intracellular reference element for quantitative determination because similar concentrations are found in the cytosol and the nuclei for a given cell line (23-26), and potassium concentration is not modified during cell cycle (27) or by toxic elements such as heavy metals (28). Cr/K ratios confirm that after chromium soluble compound

Figure 3. Synchrotron radiation XRF spectrum corresponding to the entire cell presented in Figure 2b (a) at 6020 eV and (b) 5993.5 eV. No Cr(VI) could be detected in this cell. Hexavalent chromium is totally reduced by cells to Cr(III).

exposure, the chromium concentration distribution is homogeneous within cells, with chromium concentrations similar in cytosol and nucleus either for IGR-OV1 or for CHO-AA8 cells. An example of an experiment performed at 170 K is presented in Figure 5. Similarly to in-air and room temperature analysis of freeze-dried cells (Figure 2), no Cr(VI) was detected in cells, while the Cr(III) concentration was uniformly distributed. The stability of chemical species in samples is a critical factor during storage, owing to the interconversion and degradation processes that can occur in species characterization. The preservation of chemical species during sample preparation and in the time interval between sampling and analysis had to be verified. This experiment validates the sample preparation protocol, demonstrating that freeze drying does not modify chromium oxidation state or chromium cellular distribution. Intracellular Distribution of Chromium: Low Solubility Chromate. Chemical mapping of K and Cr oxidation state imaging of IGR-OV1 cells exposed to particulate chromium compound, PbCrO4, at 1 µg/cm2 for 24 h, is presented in Figure 6. Oxidation states mapping was performed at 6020.0 eV for total chromium and 5993.5 eV for Cr(VI). The Cr(III) concentration distribution is homogeneous in cells. Particles of Cr(VI) are still visible after 24 h of exposure and are incorporated within the cytosol (Figure 6a). This statement was confirmed using a high-resolution zone plate. The intracellular distribution of Cr(VI) in a single IGR-OV1 cell exposed to 1 µg/cm2 PbCrO4 for 24 h is shown in Figure 6b. A

Chromium Oxidation State Imaging in Mammalian Cells

Chem. Res. Toxicol., Vol. 18, No. 10, 2005 1515

morphology. The mapping of chromium oxidation states reveals that Cr(III) concentration is homogeneously distributed within cells, while a Cr(VI) particle appears within the cytosol of one CHO-AA8 cell. These observations are similar to those previously mentioned for IGROV1 cells (Figure 6). Single cells could be easily identified to perform quantitative analysis as reported in Table 2. Quantitative analysis indicates that the concentration of Cr(III) is similar in the cytosol and the nucleus of IGROV1 and CHO-AA8 cells (Table 2). IGR-OV1 cells exposure to 10 µM Na2CrO4 for 1 h or to PbCrO4 at 1 µg/cm2 for 24 h resulted in similar intracellular and intranuclear chromium concentrations. The exposure of cells to PbCrO4 resulted, however, in increased cytotoxicity (Table 2).

Discussion

Figure 4. Synchrotron radiation XRF spectra corresponding to the cell presented in Figure 2b in different subcellular compartments: (a) cytosol, [Cr]/[K] ) 0.023; (b) nucleus, [Cr]/ [K] ) 0.023; and (c) nucleolus, [Cr]/[K] ) 0.023. X-ray beam energy E ) 6020 eV.

fine particle of Cr(VI) is located within the cytosol, in the perinuclear region. However, the high resolution but low flux zone plate did not allow chromium imaging in the cytosol because it is present at a lower concentration. Similar results were observed for CHO-AA8 cells (Figure 7). The micrographs taken before and after analysis indicate that irradiation conditions did not alter cellular

According to an IARC (International Agency for Research on Cancer) report (5), the moderately soluble compounds such as calcium chromate, strontium chromate, lead chromate, and basic zinc chromate are associated with the greatest risk of developing lung cancer (Table 1). The differential carcinogenicity in experimental animals of Cr(VI) compounds according to their solubility has led to the hypothesis that particulate Cr(VI) could be stronger carcinogens than soluble compounds (9-14). A number of in vivo animal carcinogenesis bioassays have illustrated that soluble hexavalent chromates are either noncarcinogenic or at best weakly carcinogenic after multiple exposure (10). In contrast, the moderately to highly insoluble particulate forms of chromate are potently carcinogenic, presumably due to their persistence as a chronic source of genotoxic oxyanions, with the deposition site serving as a focal point of exposure (29). Analysis of chromium in the bronchi of workers exposed to particulate chromate revealed hot spots of chromium accumulation at bifurcations of bronchi (30) demonstrating long-term retention of chromium in the bronchial walls. The microscopic analysis of chromium accumulation in chromate workers concluded in a significant correlation between the chromium accumulation and the progression of the malignant change of the bronchial epithelium (31). In addition, a meta-analysis of painting exposure and cancer mortality concluded that the excess deaths from lung cancer among workers exposed to paint might be from exposure to particles containing lead chromate (32). These observations attest the strong carcinogenic potential of low solubility particulate chromates. To better understand the differential carcinogenicity of soluble and low solubility chromate compounds, it is essential to know the intracellular distribution of chromium and chromium oxidation states after cell exposure to these compounds. The classical method used so far in this purpose consisted of isolation of cells from their culture substrate, fractionation of cellular compartments by differential centrifugation, and analysis of chromium oxidation state in cell fractions. This method may be prone to artifacts, and chromium oxidation state and chromium distribution could be altered by the numerous sample preparation steps. This is particularly critical for low solubility compounds as the centrifugation will result in a mixing of their subcellular distribution. Using the recently developed synchrotron radiation X-ray microprobe, we could observe the distribution of chromium

1516

Chem. Res. Toxicol., Vol. 18, No. 10, 2005

Ortega et al.

Table 2. Chromate Compounds Cytotoxicity in IGR-OV1 and CHO-AA8 Cells and Corresponding Cytosolic and Nuclear Chromium Levelsa cell line

exposure conditions

% cell survival

Cr/K within cytosol (mean ( SD)

Cr/K within nuclei (mean ( SD)

IGR-OV1 IGR-OV-1 CHO-AA8 CHO-AA8 CHO-AA8

10 µM Na2CrO4, 1 h PbCrO4, 1 µg/cm2, 24 h 10 µM Na2CrO4, 1 h PbCrO4, 0.1 µg/cm2, 24 h PbCrO4, 0.5 µg/cm2, 24 h

83 ( 11 30 ( 5 89 ( 12 90 ( 9 30 ( 8

0.0223 ( 0.0049 (n ) 11) 0.0178 ( 0.0048 (n ) 22) 0.0176 ( 0.0007 (n ) 3) 0.0174 ( 0.0009 (n ) 3) 0.0408 ( 0.0035 (n ) 18)

0.0223 ( 0.0026 (n ) 11) 0.0176 ( 0.0022 (n ) 22) 0.0173 ( 0.0025 (n ) 3) 0.0180 ( 0.0009 (n ) 3) 0.0398 ( 0.0037 (n ) 18)

a Potassium was used as an internal standard for Cr quantitative determination because potassium is known to be homogeneously distributed in cells and its concentration constant for a given cell line.

Figure 5. Cryomicroanalysis of IGR-OV1 cells exposed to 10 µM Na2CrO4. Cells were cultured onto sample holder, then cryofixed, and maintained in their frozen hydrated state until microanalysis performed at 170 K under a jet stream of liquid nitrogen and dry nitrogen atmosphere. Similarly to in-air and room temperature analysis of freeze-dried cells, no Cr(VI) was detected in cells (E ) 5993.5 eV), while Cr(III) distribution (E ) 6020 eV) was proportional to the cell volume, suggesting an homogeneous intracellular concentration. Cryomicroanalyses confirm the preservation of chemical species during sample preparation and in the time interval between sampling and analysis.

valence states directly in frozen hydrated cultured human cells and reveal differences in the intracellular reduction of Cr(VI) soluble and low solubility compounds. Intracellular Distribution of Chromium: Soluble Chromate. Soluble Cr(VI) compounds are taken up quickly by cells by sulfate transport systems (33). The anion exchange is a passive transport process independent of metabolic energy and controlling influx and efflux of inorganic anions. With the chromate anion, however, no efflux has been observed. This intracellular trapping of chromium is due to the rapid reduction of Cr(VI) by cellular constituents leading to its intracellular accumulation, known as the uptake-reduction model (34). Cellular fractionation studies, after exposure to soluble Cr(VI) compounds either in vitro or in vivo, have shown that chromium accumulates in the cytosol as well as in the nuclei (35, 36). To eliminate possible sample preparation artifacts that can arise in cellular fractionation studies, we used a synchrotron-based in situ analysis on whole cells. Our results on chromium distribution at the cellular level clearly demonstrate the intracellular and intranuclear localization of chromium in CHO-AA8 and IGR-OV1 cells after exposure to a soluble chromate Na2CrO4. The high concentration of chromium accumulated in cell nuclei, as observed in this study, certainly contributes to the genotoxic effects observed, since trivalent chromium interacts directly with the phosphate backbone of DNA and also creates DNA protein cross-links (37). Chromium intracellular distribution on V79 cells exposed

Figure 6. Chemical mapping and Cr oxidation states mapping of IGR-OV1 cells exposed to particulate chromium compound, PbCrO4, 1 µg/cm2 for 24 h. (a) Potassium, total Cr obtained at E ) 6020 eV, Cr(VI) obtained at 5993 eV, and optical microscopy. Beam spatial resolution (V × H), 1 × 3 µm2; scan, 1 µm/ pixel; 1.5 s/pixel; color scale in counts per pixel. Cr(III) is distributed proportionally to the cell volume, suggesting an homogeneous intracellular concentration. Cr(VI) particles are still visible after 24 h of exposure and are incorporated within the cytosol. PbCrO4-engulfed particles were found only in a limited number of cells. Therefore, intracellular Cr(III) might originate mainly from dissolution and reduction of extracellular lead chromate rather than from intracellular engulfed particles (see also Figure 7). (b) Potassium, Cr(VI) obtained at E ) 5993 eV and optical microscopy. Beam spatial resolution (V × H), 0.5 × 1 µm2; color scale in counts per pixel. Cr(VI) distribution shows a perinuclear localization.

to a Cr(VI) soluble compound was also found uniformly distributed by micro-PIXE analysis (38). More recently, synchrotron radiation XRF applied with a hard X-ray microprobe enabled the direct mapping, with 1 µm spatial resolution, of chromium distribution in Chinese hamster lung cells exposed to 0.5 µM of soluble chromate during 4 h (39). Both studies confirm that chromium concentration is evenly distributed in cells, including the nucleus, and follows the distribution of ubiquitous elements such as phosphorus and potassium. Intracellular Distribution of Chromium: Low Solubility Chromate. Chromium is present in the atmosphere in the form of particulate and droplet aerosols. The Cr-containing particles emitted by various sources differ considerably in size, from 0.2 to 50 µm in diameter. The size of particles is of importance when chromium toxic effects are considered. Only the particles from 0.2 to 10 µm are respirable, and their retention in the lung can pose a carcinogenic risk (40). The smallest particles are produced by combustion of coal or by chromate and brick production (particle diameter less than 1 µm). The two stable oxidation states of chromium, Cr(III) and Cr(VI), are commonly found in the atmosphere. Sparingly soluble chromates with some metal ions such as Pb2+, Cu2+, or Zn2+ are only found under particulate forms. In our study, PbCrO4 particles size was

Chromium Oxidation State Imaging in Mammalian Cells

Figure 7. Optical microscopy and potassium and chromium oxidation states maps of CHO-AA8 cells. Scan size, 150 µm × 60 µm. Red arrow indicates the position of a Cr(VI) particle as seen on Cr map at E ) 6020 eV. From the micrographs taken before and after analysis, it can be concluded that irradiation conditions did not alter cellular morphology. Potassium distribution is indicative of cell position; Cr(III) is distributed proportionally to the cell volume, suggesting an homogeneous intracellular concentration. A Cr(VI) particle appears within the cytosol of one CHO-AA8 cell in this example. These observations are similar to those previously mentioned for IGR-OV1 cells. Cells could be easily identified to perform quantitative analysis as reported in Table 2.

smaller than 3 µm as observed by optical microscopy, with a majority of particles of about 1 µm diameter. The size range of the particles, between 0.2 and 5 µm, is consistent with inhalation models describing particle deposition in the lower respiratory tract and deep lung, which are the most frequent sites of chromium-induced cancers (7). The particle size is critical to this experiment, as suspension of large aggregates was not found to be cytotoxic to CHO cells (41). Lead chromate is a prototype of particulate forms of chromium, and it is clastogenic and cytotoxic and can induce neoplastic transformation of cell cultures (11, 13). Although lead chromate particles are actively internalized, it is cell to particulate contact and cell-enhanced extracellular dissolution that are responsible for its genotoxic activity (13, 42). It was also shown that hexavalent chromium is the proximate genotoxicant of lead chromate and that Pb2+ does not contribute to its genotoxicity (43, 44). The extracellular concentration of chromium increased 7-fold when lead chromate was incubated in the presence of CHO cells as compared to incubation with culture medium alone (13). A specific degree of solubility in biological fluids and tissues, as needed to obtain a slow release of chromate ions in the target cells over a long period, seems necessary for a carcinogenic response (12). The solubility of more or less slightly water soluble Cr(VI) compounds was greatly increased in growth medium (12). The mechanism of cellenhanced extracellular solubility has also been described by other authors (13, 42). It has been suggested that biological effects of particulate Cr(VI) originated in extracellularly solubilized chromate (12). Particulate lead chromate induces apoptosis as the mode of cell death in CHO-AA8 and human small airway epithelial cells (HSAE) (43, 45). Only few studies report in situ characterization of PbCrO4 in cells. Lead chromate particles were observed by transmission electron microscopy within cytoplasmic vacuoles in CH0-AA8 cells treated with 350 µM PbCrO4 for 24 h (43) or in HSAE cells treated with 10 µg/cm2 lead chromate for 24 h (45). HSAE cells showed phagocytosed intracellular particles and presented large vacuoles

Chem. Res. Toxicol., Vol. 18, No. 10, 2005 1517

containing angular, jagged lead chromate particles, as confirmed by energy dispersive X-ray microanalysis. Using RAW 264.7 cells exposed to PbCrO4, confocal microscopy showed that particles could become either bound to the cell surface or engulfed over a 120 min time period (46). In our study, we observed intracellular Cr(VI) particles in CHO-AA8 and IGR-OV1 cells exposed in vitro to PbCrO4, located in the perinuclear region. In addition to previously published results, intracellular Cr(III) was also observed in these cells and its concentration was similar in the cytosol and in the nucleus. Because PbCrO4-engulfed particles were found only in a limited number of cells, whereas Cr(III) was found in all cells, intracellular Cr(III) might originate mainly from dissolution and reduction of extracellular lead chromate rather than from intracellular engulfed particles (Figures 6 and 7). This new result suggests that the mechanism of carcinogenicity of low solubility compounds may involve a combinative mechanism of intracellular Cr(III), a genotoxic effector, and Cr(VI), a pro-oxidant effector. Differential Carcinogenicity of Cr(VI) Soluble and Low Solubility Compounds. Once inside the cell, Cr(VI) is rapidly converted by intracellular reducing agents such as vitamin C to Cr(III). Cr(VI) reduction in cells is supposed to play an important role in Cr(VI) carcinogenesis (34). The intracellular Cr(VI) reduction, leading to generation of reactive species, may be viewed as an activation process when it occurs in the proximity of DNA. Alternatively, reduction is a detoxification process when it occurs far away from DNA, and the reactive species can be trapped by a large number of ligands, nucleophiles, and antioxidants, which are present in the intracellular environment. Therefore, the cellular site of reduction is crucial in affecting the fate of the cell taking up Cr(VI). It has been proposed that the carcinogenicity of chromium compounds depends on both valence and solubility, with the least water soluble, hexavalent compounds being the most active (11-14). This has led to the hypothesis that the carcinogenicity of these compounds is due to deposition and slow dissolution, resulting in prolonged exposure of local cells to hexavalent chromium (12). The comparison between intracellular and intranuclear chromium concentrations and cytotoxicity is presented in Table 2. The cytotoxicity and potential carcinogenicity of chromate compounds do not follow the intracellular accumulation of chromium, which is indirect evidence for the active role of particulate chromium. It is important to note that no lead was found in cells as measured by micro-PIXE (particle-induced X-ray emission) analysis (47): The cytotoxicity of PbCrO4 is therefore probably due mainly to chromium. It has been reported that both Na2CrO4 and PbCrO4 induce concentration-dependent cytotoxicity in primary human bronchial fibroblasts (48). In the case of IGR-OV1 cells exposed to Na2CrO4, we observed 83% cell survival, while the intracellular Cr/K was 0.0223 (Table 2). In the same way, cells exposed to PbCrO4 resulted in only 30% cell survival for a lower Cr/K ratio of 0.0178. This result strongly suggests that the cellular effects of lead chromate are not simply due to intracellular Cr(III) diffusion but also due to the presence of Cr(VI) particles in the cellular environment, either within the cell, as observed in the perinuclear region, or also possibly outside the cells, adsorbed to the plasma membrane. The result is not so clear for CHO-AA8 cells, probably because the tested conditions gave lower cyto-

1518

Chem. Res. Toxicol., Vol. 18, No. 10, 2005

toxicity, with almost 100% cell survival for Cr/K of about 0.017 for soluble and insoluble chromate compounds (Table 2). A recent study reports that in human bronchial cells lead chromate clastogenesis was mediated by the extracellular dissolution of the particles and not their internalization (49). Both studies clearly suggest that the extracellular dissolution of lead chromate is the first step of the carcinogenic mechanism of lead chromate.

Conclusion The processes following deposition of chromiumcontaining materials in the lung and governing the biological activation of Cr(VI) can be divided in three main processes: (i) solubilization of Cr(VI); (ii) cellular uptake of Cr(VI), either as soluble chromate anion or as particulate matter; and (iii) intracellular reduction of solubilized Cr(VI). In the case of poorly soluble or insoluble particulate chromate, the chronic inflammation and/or oxidative processes could enhance cellular damages explaining the higher cytotoxicity of particulate chromate compounds as compared to soluble compounds and also their higher carcinogenic potential. Cell exposure to Cr(VI) compounds resulted in intracellular accumulation of reduced forms of chromium for both soluble and low solubility chromate compounds. These reduced forms of chromium were found homogeneously distributed within the cell volume, including in the nucleus and nucleoli. Cr(VI) was observed in the cell environment only after exposure to low solubility compound, in the form of PbCrO4-engulfed particles. Cr(III) was found in all cells, including cells that did not contain PbCrO4-engulfed particles. Lead was not found in cells exposed to PbCrO4 except within engulfed particles. These results suggest that intracellular Cr(III) originates from dissolution and reduction of extracellular lead chromate. Cell-enhanced dissolution of PbCrO4 has been clearly evidenced in this study as most of PbCrO4 particles are solubilized after 24 h in culture medium with living cells, while this compound is practically insoluble at neutral pH. These results strongly support some recently published data on lead chromate clastogenicity (49) indicating that extracellular dissolution was the first step for lead chromate clastogenicity and carcinogenesis. Considering that Cr(VI) low solubility compounds can persist longer than soluble compounds at sites of exposure (30, 31) and that they can mediate cellular responses via reactive oxygen species (46), the stronger carcinogenicity of Cr(VI) low solubility compounds may derive from the combinative genotoxic effects of intracellular Cr(III), partially bound to DNA, and longterm exposure to Cr(VI), a strong oxidant at physiological pH.

Acknowledgment. We are grateful to the European Synchrotron Radiation Facility (ESRF) staff for their assistance, in particular Robert Baker, Gilles Berruyer, and Nicolas Pascal. We are also grateful to the ESRF for beam time allocation and financial support.

References (1) Brown, G. E., Foster, A. L., and Ostergren, J. D. (1999) Proc. Natl. Acad. Sci. U.S.A. 96, 3388-3395. (2) Bertsch, P. M., and Hunter, D. B. (2001) Chem. Rev. 101, 18091842.

Ortega et al. (3) Ortega, R. (2002) In Heavy Metals in the Environment (Sarkar, B., Ed.) pp 35-68, Marcel Dekker, New York. (4) Ortega, R., Bohic, S., Tucoulou, R., Somogyi, A., and Deve`s, G. (2004) Anal. Chem. 76, 309-314. (5) IARC, International Agency for Research on Cancer (1990) IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 49, IARC Scientific Publications, Lyon. (6) U.S. Environmental Protection Agency (1998) Toxicological Review of Hexavalent Chromium. National Centre for Environmental Assessment, Office of Research and Development, Washington, DC; Available on-line at http://www.epa.gov/iris. (7) Barceloux, D. G. (1999) Clin. Toxicol. 37, 173-194. (8) De Flora, S., Bagnasco, M., Serra, D., and Zanacchi, P. (1990) Mutat. Res. Rev. Mutat. 238, 99-172. (9) O’Brien, T. J., Ceyrak, S., and Patierno, S. R. (2003) Mutat. Res. 533, 3-36. (10) Glaser, U., Hochrainer, D., Kloppel, H., and Oldiges, H (1986) Toxicology 15, 219-232. (11) Patierno, S. R., Bahn, D., and Landolph, J. R. (1988) Cancer Res. 48, 5280-5288. (12) Elias, Z., Poirot, O., Baruthio, F., and Daniere, M. C. (1991) Carcinogenesis 12, 1811-1816. (13) Wise, J. P., Orenstein, J. M., and Patierno, S. R. (1993) Carcinogenesis 14, 429-434. (14) Singh, J., Carlisle, D., Pritchard, D. E., and Patierno, S. R. (1998) Oncol. Rep. 5, 1307-1318. (15) O’Brian, T. J., Ceryak, S., and Patierno, S. R. (2003) Mutat. Res. 533, 3-36. (16) Thompson, L. H., Fong, S., and Brookman, K. (1980) Mutat. Res. 74, 21-36. (17) Be´nard, J., Da Silva, J., De Blois, M.-C., Boyer, P., Duvillard, P., Chiric, E., and Riou, G. (1985) Cancer Res. 45, 4970-4979. (18) Ortega, R., Moretto, P., Fajac, A., Be´nard, J., Llabador, Y., and Simonoff, M. (1996) Cell. Mol. Biol. 42, 77-88. (19) Susini, J., Salome´, M., Fayard, B., Ortega, R., and Kaulich, B. (2002) Surf. Rev. Lett. 9, 203-211. (20) Di Fabrizio, E., Romanato, F., Gentili, M., Cabrini, S., Kaulich, B., Susini, J., and Barret, R. (1999) Nature 401, 895-898. (21) Peterson, M. L., Brown, G. E., Parks, G. A., and Stein, C. L. (1997) Geochim. Cosmochim. Acta 61, 3399-3412. (22) Vekemans, B., Janssens, K., Vincze, L., Adams, F., and Van Espen, P. (1994) X-ray Spectrom. 23, 278-285. (23) Cameron, I. L., Smith, N. K. R., and Pool, T. B. (1979) J. Cell Biol. 80, 444-450. (24) Zierold, K., Schafer D., and Pietruschka F. (1984) Histochemistry 80, 333-337. (25) Deve`s, G., and Ortega, R. (2002) Anal. Bioanal. Chem. 374, 390394. (26) Ortega, R., Bohic, S., Tucoulou, R., Somogyi, A., and Deve`s, G. (2004) Anal. Chem. 76, 309-314. (27) Warley, A., Stephen, J., Hockaday, A., and Appleton, T. C. (1983) J. Cell Sci. 62, 339-350. (28) Zierold, K. (2000) Toxicol. In Vitro 14, 557-563. (29) Blankenship, L. J., Manning, F. C. R., Orenstein, J. M., and Patierno, S. R. (1994) Toxicol. Appl. Pharmacol. 129, 75-83. (30) Ishikawa, Y., Nakagawa, K., Satoh, Y., Kitagawa, T., Sugano, H., Hirano, T., and Tsuchiya, E. (1994) Cancer Res. 54, 2342-2346. (31) Kondo, K., Takahashi, Y., Ishikawa, Y., Uchihara, H., Hirose, Y., Yoshizawa, K., Tsuyugushi, M., Takizawa, H., Miyoshi, T., Sakiyama, S., and Monden, Y. (2003) Cancer 98, 2420-2429. (32) Chen, R., and Seaton, A. (1998) Cancer Detect. Prev. 22, 533539. (33) Norseth, T. (1986) Br. J. Ind. Med. 43, 649-651. (34) Connett, P., and Wetterhahn, K. E. (1983) Struct. Bonding 54, 93-124. (35) Sehlmeyer, U., Hechtenberg, S., Klyszcz, H., and Beyersmann, D. (1990) Arch. Toxicol. 64, 506-508. (36) Clodfelder, B. J., Emamaullee, J., Hepburn, D. D., Chakov, N. E., Nettles, H. S., and Vincent, J. B. (2001) J. Biol. Inorg. Chem. 6, 608-617. (37) Salnikow, K., Zhitkovich, A., and Costa, M. (1992) Carcinogenesis 13, 2341-2346. (38) Cholewa, M., Dillon, C. T., Lay, P. A., Phillips, D., Talarico, T., Lai, B., Balaic, D. X., Barnea, Z., Cai, Z., Deacon, G. B., Ilinski, P., Legnini, P., Rainone, S., Shea-McCarthy, G., Stampfl, A. P. J., Webster, L. K., and Yun, W. (2001) Nucl. Instrum. Methods B 181, 715-722. (39) Dillon, C. T., Lay, P. A., Kennedy, B. J., Stampfl, A. P. J., Cai, Z., Ilinski, P., Rodrigues, W., Legnini, D. G., Lai, B., and Maser, J. (2002) J. Biol. Inorg. Chem. 7, 640-645. (40) Kotas, J., and Stasicka, Z. (2000) Environ. Pollut. 107, 263-283. (41) Wise, J. P., Leonard, J. C., and Patierno, S. R. (1992) Mutat. Res. 278, 69-79.

Chromium Oxidation State Imaging in Mammalian Cells

Chem. Res. Toxicol., Vol. 18, No. 10, 2005 1519

(42) Wise, J. P., Stearns, D. M., Wetterhahn, K. E., and Patierno, S. R. (1994) Carcinogenesis 15, 2249-2254. (43) Blankenship, L. J., Carlisle, D. L., Wise, J. P., Orenstein, J. M., Dye, L. E., III, and Patierno, S. R. (1997) Toxicol. Appl. Pharm. 146, 270-280. (44) Wise, S. S., Holmes, A. L., Ketterer, M. E., Hartsock, W. J., Fomchenko, E., Katsifis, S., Thompson, W. D., and Wise, J. P., Sr. (2004) Mutat. Res. 560, 79-89. (45) Singh, J., Pritchard, D. E., Carlisle, D. L., McLean, J. A., Montaser, A., Orenstein, J. M., and Patierno, S. R. (1999) Toxicol. Appl. Pharmacol. 161, 240-248.

(46) Leonard, S. S., Roberts, J. R., Antonini, J. M., Castranova, V., and Shi, X (2004) Mol. Cell. Biochem. 255, 171-179. (47) Ortega, R., Deve`s, G., Fayard, B., Salome´, M., and Susini, J. (2003) Nucl. Instrum. Methods B 210, 325-329. (48) Wise, J. P., Wise, S. S., and Little, J. E. (2002) Mutat. Res. 517, 221-229. (49) Xie, H., Holmes, A. L., Wise, S. S., Gordon, N., and Wise, J. P. (2004) Chem. Res. Toxicol. 17, 1362-1367.

TX049735Y