Variation in Extracellular Detoxification Is a Link to Different Carcinogenicity among Chromates in Rodent and Human Lungs.

Inhalation of soluble chromium(VI) is firmly linked with higher risks of lung cancer in humans. However, comparative studies in rats have found a high lung tumorigenicity for moderately soluble chromates but no tumors for highly soluble chromates. These major species differences remain unexplained. We investigated the impact of extracellular reducers on responses of human and rat lung epithelial cells to different Cr(VI) forms. Extracellular reduction of Cr(VI) is a detoxification process, and rat and human lung lining fluids contain different concentrations of ascorbate and glutathione. We found that reduction of chromate anions in simulated lung fluids was principally driven by ascorbate with only minimal contribution from glutathione. The addition of 500 μM ascorbate (∼rat lung fluid concentration) to culture media strongly inhibited cellular uptake of chromate anions and completely prevented their cytotoxicity even at otherwise lethal doses. While proportionally less effective, 50 μM extracellular ascorbate (∼human lung fluid concentration) also decreased uptake of chromate anions and their cytotoxicity. In comparison to chromate anions, uptake and cytotoxicity of respirable particles of moderately soluble CaCrO4 and SrCrO4 were much less sensitive to suppression by extracellular ascorbate, especially during early exposure times and in primary bronchial cells. In the absence of extracellular ascorbate, chromate anions and CaCrO4/SrCrO4 particles produced overall similar levels of DNA double-stranded breaks, with less soluble particles exhibiting a slower rate of breakage. Our results indicate that a gradual extracellular dissolution and a rapid internalization of calcium chromate and strontium chromate particles makes them resistant to detoxification outside the cells, which is extremely effective for chromate anions in the rat lung fluid. The detoxification potential of the human lung fluid is significant but much lower and insufficient to provide a threshold-type dose dependence for soluble chromates.

Introduction

Hexavalent chromium has been firmly recognized as a human respiratory carcinogen.[1−3] Upon dissolution at neutral pH, Cr(VI) exists principally in the form of chromate anion (CrO42–) that structurally is very similar to physiological anions sulfate and phosphate. Because of this structural resemblance, chromate enters a broad variety of cells utilizing their sulfate and phosphate channels. Cr(VI) is unstable inside the cells where it undergoes reduction generating Cr(III) as the thermodynamically stable form.[4] Intracellular reduction of Cr(VI) is required for the formation of Cr-DNA damage, which includes DNA–protein cross-links[5,6] and several types of smaller Cr-DNA adducts.[7−9] Metabolism of Cr(VI) can also result in the production of oxidative DNA damage.[10−12] Cellular reduction of Cr(VI) in vivo is principally driven by direct electron transfer reactions with ascorbate (Asc),[13,14] which is present in low millimolar concentrations in cells,[15,16] glutathione (GSH) and, to a smaller extent, cysteine also contribute to the conversion of Cr(VI) to Cr(III) in tissues. Cultured cells are severely deficient in Asc and their metabolism of Cr(VI) is largely dependent on the nonenzymatic one-electron transfer reactions with GSH.[4] The initial reduction of Cr(VI) by Asc involves transfer of two electrons, which skips the formation of reactive Cr(V) intermediate[17−19] and suppresses oxidative DNA damage.[20,21] In contrast to its activation role inside the cells, reduction of Cr(VI) outside the cells is the detoxification process producing membrane-impermeable Cr(III).[4]

Although all chemical forms of Cr(VI) are classified as human carcinogens,[2] the strength of experimental and epidemiological evidence for individual compounds varies significantly. Two large comparative studies using different strains of rats and delivery methods have found that Cr(VI) compounds of moderate solubility were highly tumorigenic in the lung whereas highly soluble and very poorly soluble chromates were not tumorigenic.[22,23] Epidemiological evidence for carcinogenicity of the most insoluble chromates is also weaker than for the moderately soluble group.[1,3] A frequently referenced first review of chromium carcinogenicity by the International Agency for Research on Cancer, which was released in 1990, did not identify reports with strong evidence for carcinogenicity of soluble chromates in humans.[24] However, subsequent epidemiological studies[25−27] among cohorts of workers exposed to soluble Cr(VI) all found significantly elevated lung cancer risks. The solubility-related dependence of Cr(VI) tumorigenicity in rodents remains unexplained, which limits the use of these standard laboratory animals for investigation of mechanistic aspects of Cr(VI) carcinogenicity and complicates application of animal results for human risk assessment. A comparison of cytotoxicity and chromosomal damage in standard cultures of telomerase-immortalized human fibroblasts has not uncovered large differences among highly soluble sodium chromate, moderately soluble zinc chromate and very poorly soluble barium and lead chromates,[28] suggesting similar causes of genotoxicity for these compounds under GSH-driven reduction conditions. Mechanistic considerations of DNA damage formation by various products of intracellular reduction of chromate anions[10,29] also cannot explain why solubility is such an important factor in rat lungs but not in human lungs.

In this work, we examined the significance of species differences in the concentrations of chromate reducers in the lung lining fluid for uptake and toxicity of Cr(VI) compounds of different solubility. Using human and rat lung epithelial cells, we obtained evidence that link variable carcinogenicity and toxicity of chromates to their different extracellular detoxification.

Experimental Procedures

Materials

l-Ascorbic acid (99.9% pure), dehydro-l-(+)-ascorbic acid dimer, potassium chromate (K2CrO4, 99% pure), l-glutathione (>98% pure), and nitric acid (>99.999% pure) were obtained from Sigma-Aldrich. CaCrO4 (43333) and SrCrO4 (89026) were purchased from Alfa Aesar. RPMI-1640 medium (11875–093) was from ThermoFisher Scientific.

Cell Culture

H460 human lung epithelial and RLE-6TN rat lung epithelial cells were obtained from the American Type Tissue Collection. H460 cells were grown in RPMI-1640 media (11875119, ThermoFisher) containing 10% (v/v) fetal bovine serum (FBS) and penicillin/streptomycin. RLE-6TN rat cells were propagated in DMEM/F12 media (11320–033, ThermoFisher), 10% FBS, and 1% penicillin/streptomycin. Primary human bronchial epithelial (HBE) cells were purchased from Lonza and grown in the supplier’s recommended medium (CC-3170) supplemented with growth factors and antibiotics. All three cell lines were kept in the atmosphere of 95% air/5% CO2. Cells were treated with the indicated concentrations of Cr(VI) next day after seeding. Stock solutions of K2CrO4 were freshly prepared in water for each experiment. CaCrO4 or SrCrO4 were used as suspensions of particles in ethanol.

Particle Preparation

Commercial CaCrO4 or SrCrO4 was placed in glass containers (7-mL scintillation vial for Ca-chromate and Erlenmeyer flask for Sr-chromate), combined with 100% ethanol, sealed with parafilm, and stirred overnight. The next day, chromate solutions were passed through 30 mm nylon syringe filters with 5 μm pores (F2500–50, ThermoFisher) and stored at 4 °C. When necessary, solutions of chromate particles were concentrated at 30 °C using a vacuum centrifuge (Eppendorf Vacufuge 5301). The volumes were monitored over time, and the samples were occasionally taken out and vortexed to remove the accumulated chromium from the side of the vials. When the ethanol was at the desired volume, the solution was vigorously vortexed to break up any aggregates. For the determination of Cr(VI) concentrations, Ca-chromate and Sr-chromate particles were dissolved in 250 mM NaOH (final concentration) followed by absorbance measurements at 372 nm. Sodium chromate solubilized in a mixture of 50% ethanol/50% 500 mM NaOH was used for standard curves. To correct for light scattering and other nonspecific effects, A372 values were adjusted by subtracting absorbance at 550 nm in the same sample. Ethanol suspensions of chromate particles were stored at 4 °C and used within a few days after preparation.

Ascorbate Determination

Asc was extracted with 50 mM methanesulfonic acid/5 mM diethylenetriaminepentaacetic acid and detected as a fluorescent product after conjugation with 1,2-diamino-4,5-dimethoxybenzene dihydrochloride.[30]

Restoration of Cellular Ascorbate

Cells were incubated with dehydroascorbic acid in Krebs-HEPES buffer [30 mM HEPES (pH 7.5), 130 mM NaCl, 4 mM KH2PO4, 1 mM MgSO4, 1 mM CaCl2] supplemented with 10% FBS and 0.5 mM d-glucose. Stock solutions of dehydroascorbic acid were freshly prepared in the same buffer and kept on ice before use.

Chromate Reduction

Dissolved K2CrO4 was used a source of chromate anions in all reactions. Reduction was measured at 37 °C by recording chromate absorbance at 372 nm. Equal volumes of 2× concentrated chromate and reducer solutions were rapidly mixed in 96-well plates followed by the immediate determination of the initial A372. Absorbance values were recorded every 10 s in kinetics experiments. Plates were maintained at 37 °C inside the SpectraMax M5 microplate reader.

Cr(VI) Uptake

The determination of cellular Cr was based on a previously described procedure.[12] Cells were seeded into six-well plates and allowed to grow overnight before the addition of Cr(VI). After removal of Cr-containing media, cells were rinsed twice with warm PBS and collected by trypsinization in the presence of EDTA (Trypsin-EDTA solution from ThermoFisher Scientific, 15400–054). Cells were pelleted and washed twice with cold PBS (5 min at 800 × g, 4 °C) prior to extraction with 5% (v/v) nitric acid at 50 °C for 30 min. Insoluble debris was removed by centrifugation at 10 000 × g for 10 min at 4 °C. Supernatants were diluted to 2% (v/v) nitric acid prior to Cr measurements by graphite furnace atomic absorption spectroscopy (AAnalyst600 Atomic Absorption Spectrometer, PerkinElmer). Cellular precipitates formed after nitric acid extraction were washed twice with cold 5% (v/v) nitric acid (10 000 × g for 5 min, 4 °C), dissolved in 0.5 M NaOH, and used for the determination of protein content of each sample. Final Cr uptake values were protein-normalized. Control experiments showed that Ca-chromate and Sr-chromate were rapidly solubilized in 2% or 5% nitric acid even in the absence of heating, which indicated that the employed nitric extraction procedure did not underestimate the amounts of cellular Cr due to a potential inability to recover internalized but not yet solubilized chromate particles.

Western Blotting

At the time of collection, cells were washed twice with cold PBS and detached from the dishes by scraping in PBS. Attached and floating cells were combined, pelleted, and then boiled for 10 min in a lysis buffer containing 2% SDS, 50 mM Tris, pH 6.8, 10% glycerol, and protease/phosphate inhibitors (78425, ThermoFisher Scientific). Samples were cleared from occasional debris by centrifugation at 10 000 × g for 10 min at room temperature. Proteins were separated on 10 or 12% SDS-PAGE gels and electrotransferred to PVDF membranes (162–0177, Bio-Rad). Primary antibodies were from Santa Cruz for detection of p53 (sc-125, 1:1000 dilution), from Cell Signaling Technology for Ser15-phosphorylated p53 (9284, 1:1000), cleaved caspase 7 (8438, 1:500), PARP (9542, 1:1000), and Ser139-phosphorylated histone H2AX (2577, 1:1000), from Abcam for fibrillarin (ab5821, 1:5000) and from Sigma for γ-tubulin (T6557, 1:2000).

Cell Viability

Cytotoxic effects of Cr(VI) treatments were measured using the CellTiter-Glo luminescent assay (Promega). Cells were seeded into 96-well plates (3000 per well for all cell lines) and treated with Cr(VI) on the next day. Cytotoxicity was determined following 48 h post-Cr recovery for H460 and RLE-6TN cells and 72 h recovery for HBE cells.

Colony Formation

H460 cells were grown on six-well plates overnight followed by preloading with Asc by incubation for 90 min with 2 mM (as monomer) dehydroascorbic acid as described above. Cells were switched to the standard growth media and treated with Cr(VI) compounds for 3 h in the presence of 0, 50, or 500 μM Asc. After 7–8 days of growth with one change of media, colonies were visualized by Giemsa staining.

Statistics

Differences between the groups were evaluated by two-tailed, unpaired t test.

Results

Chromate Reduction by Asc and GSH

Asc and GSH are two most important reducers of chromate in vivo in tissues,[13,14] and they both are present in human and rat bronchoalveolar lining fluids. Cysteine, another biological thiol that is capable of Cr(VI) reduction,[31−33] is generally absent in extracellular milieu. Human lung lining fluid contained on average 109 μM GSH and 40 μM Asc.[34] Relative to humans, respiratory tract lining fluid from rats had approximately two-times lower GSH but almost 10-times higher Asc concentrations.[35] Other investigators have estimated that rat lung lining fluid contained at least 1 mM Asc.[36] The protein content of human respiratory lining fluid is approximately 10% of that in plasma.[34] The rate of chromate reduction in HEPES buffer with physiological pH and ionic strength was dramatically faster by Asc relative to GSH (Figure A). The inclusion of 10% serum to mimic the protein content of human respiratory fluid slightly increased reduction rates for both Asc (1.1-fold) and GSH (1.3-fold). The initial rate of chromate reduction in 10% serum-containing samples by 1 mM Asc was 20.1-times higher than that by 1 mM GSH, which points to a minor role of GSH in Cr(VI) reduction in the bronchoalveolar fluid. The use of both reducers at concentrations that approximate those found in human (50 μM Asc, 100 μM GSH) and rat lung lining fluids (500 μM Asc) confirmed the dominant role of Asc in chromate reduction, as evidenced by practically identical rates of Cr(VI) metabolism in 50 μM Asc versus 50 μM Asc+100 μM GSH reactions (Figure B). The rate of chromate reduction by 500 μM Asc was 7.4-times faster in comparison to 50 μM Asc, which modestly deviated from the linear relationship expected for the first-order reaction kinetics in the Asc-driven reduction.[37] A relatively low absorbance of 5 μM Cr(VI) used to ensure the pseudo-first order conditions for all reactions (≥10-times excess of reducers over Cr) probably diminished the accuracy of our rate measurements, particularly for the fast reduction with 500 μM Asc. To test the relative abilities of extracellular Asc and GSH to convert chromate to nontoxic cell-impermeable Cr(III), we measured accumulation of Cr in H460 human lung epithelial cells in their standard medium supplemented with 10% serum (Figure C). In agreement with chromate reduction results, Asc was also dramatically more effective than GSH in suppressing cellular uptake of Cr(VI). A preincubation of solubilized K2CrO4 (chromate anion-containing solution) with 1 mM Asc in culture media for 1 h prior to the addition to cells completely abolished Cr accumulation, confirming Cr(VI)-dependence of our metal uptake measurements (Figure D).

Figure 1

Impact of ascorbate (Asc) and glutathione (GSH) on reduction and cellular uptake of chromate anion. All reactions used solubilized K2CrO4 as a source of chromate anions. (A) Time-course of 50 μM chromate reduction at 37 °C (buffer, 50 mM HEPES, 100 mM NaCl, pH 7.0; buffer+FBS, HEPES buffer containing 10% fetal bovine serum). Data are means of triplicate measurements. Error bars were smaller than 5% of the mean and are not shown for clarity. (B) Chromate reduction in HEPES buffer by reducer concentrations that are typically found in lung lining fluids. Reactions contained 5 μM Cr(VI). Data are means of triplicate measurements. (C) Uptake of chromate by H460 cells in the presence of extracellular Asc or GSH. Cells were incubated with 10 μM Cr(VI) for 1 h. Data are means ± SD, n = 3, ∗∗∗p < 0.001 relative to 0 mM reducer. (D) Elimination of Cr accumulation by a preincubation of 10 μM chromate in culture media with 1 mM Asc (1 h, 37 °C) prior to the addition to H460 cells. Cells were incubated with Cr-containing media for 1 h. Data are means ± SD, n = 3.

Uptake of Solubilized Chromate and Chromate Particles in H460 Cells

A poorly soluble calcium chromate has been found as one of the most potent tumorigenic Cr(VI) compounds in rat lungs using intrapleural injections[22] and intrabronchial implants.[23] Using filtration through nylon filers with 5 μm pores, we prepared ethanol suspensions of insoluble CaCrO4 particles of the respirable size. In part due to clogging of filters, the yield of CaCrO4 particles was low, which necessitated preparation of fresh batches for each experiment. The use of multiple batches of particles helped avoid sample-specific and “aging”-related effects. As expected, we found that the presence of Asc in media at the rat lung lining fluid concentration (500 μM) strongly inhibited cellular accumulation of Cr using chromate anions (solubilized K2CrO4) as a source of Cr(VI) (Figure A). A small decrease in chromate uptake was also observed with 50 μM Asc, which corresponds to the concentration of this reducer in the human lung fluid. In contrast to chromate anion solutions, cellular accumulation of Cr after incubations with CaCrO4 particles was only minimally decreased by 500 μM Asc, indicating that internalization of particles was the primary uptake mode as particle dissolution and release of chromate anions would have made them susceptible to reduction by Asc to cell-impermeable Cr(III). Consistent with its very slow reduction kinetics, the addition of the human lung fluid-relevant concentration of GSH (100 μM) to 50 μM Asc-supplemented media had no appreciable effects on cellular Cr levels after incubations with solubilized chromate or CaCrO4 particles (Figure B). Uptake and cellular reduction of chromate lead to a long-term retention of Cr due to binding of Cr(III) to various macromolecules.[4] To assess a potential “stickiness” of CaCrO4 particles to cells without the actual uptake, we measured the levels of cellular Cr immediately after Cr(VI) exposures and following 3 h recovery in complete media (Figure C). The amounts of cellular Cr were practically unchanged after 3 h of recovery for both soluble chromate and CaCrO4 particles, which argues against a significant nonspecific absorption of particles in our uptake measurements. To prevent their solubilization, chromate particles were prepared and stored in ethanol, which raises a question whether the addition of this solvent (always <1%) altered chromate metabolism or related cellular functions. We found that the rate of chromate reduction by Asc was not significantly different in the range of 0–2% ethanol concentration (Figure A), excluding a possibility that ethanol interfered with chromate detoxification in the experiments with Cr(VI) particles. Cellular uptake of chromate anions and CaCrO4 particles was also unaffected in the presence of 0–2% ethanol (Figure B). However, the addition of 3% ethanol significantly diminished cellular accumulation of chromate anions (to 73.7% of control) but not that of CaCrO4 particles, further pointing to the different mechanisms of cellular entry for these two forms of Cr(VI). H460 cells incubated with ethanol for 1 h (as in Cr uptake experiments) or 3 h displayed no detectable cytotoxic effects for 0.5 and 1% concentrations but showed small and gradually increasing losses of viability at 1.5% and higher concentrations (Figure C). As all of our experiments with Cr(VI) particles contained <1% ethanol in culture media, its presence at these levels does not affect Cr(VI) metabolism or cell viability.

Figure 2

Uptake of chromate anions and CaCrO4 particles by human H460 cells. Cells were incubated for 1 h with 5 μM Cr(VI) in the form of solubilized K2CrO4 or CaCrO4 particles (0.8% final ethanol concentration). (A) Cellular Cr levels after incubations with soluble and particulate Cr(VI) in the presence of 0, 50, and 500 μM Asc in media. Data are means ± SD, n = 3, ∗p < 0.05, ∗∗∗p < 0.001 relative to 0 μM Asc. (B) Cellular Cr accumulation from soluble and particulate Cr(VI) in the presence of extracellular 50 μM Asc with and without 100 μM GSH. Data are means ± SD, n = 3. (C) Cellular amounts of Cr at 0 and 3 h postexposure. Data are means ± SD, n = 3.

Figure 3

Effects of ethanol on Cr(VI) metabolism and cell viability. (A) Kinetics of chromate reduction by Asc in the presence of ethanol. Reactions contained 0–2% ethanol, 200 μM Asc, 20 μM chromate in HEPES-NaCl buffer, pH 7.0. Data are means of triplicate measurements. (B) Uptake of chromate anions (solubilized K2CrO4) and CaCrO4 particles by H460 cells. Cells were incubated with 5 μM of each Cr(VI) form for 1 h. Means ± SD, n = 3, ∗∗p < 0.01 relative to controls without ethanol. The addition of CaCrO4 gave 0.07% ethanol in media. (C) Viability of H460 cells treated with ethanol for 1 or 3 h. Cell viability measurements were taken at 48 h postexposure. Data are means ± SD, n = 3.

Cytotoxicity of Chromate Anions and Chromate Particles in H460 Cells

To assess the effects of extracellular Asc on toxicity of chromate anions, we measured apoptotic and DNA damage-related responses in H460 cells treated with Cr(VI) for 3 h followed by 18 h recovery. We found that the presence of 500 μM Asc in media completely abolished apoptotic cleavage of PARP and the formation of the active form of executioner caspase-7 (Figure A). Even the addition of 50 μM Asc to media was very protective against chromate-induced apoptosis. DNA damage-related markers assessed by levels of the transcription factor p53, its Ser15 phosphorylation, and the appearance of the biochemical marker of DNA double-strand breaks (DSBs), Ser139-phosphorylated histone H2AX (also known as γ-H2AX), showed a complete loss of genotoxicity when cells were treated with chromate in the presence of 500 μM extracellular Asc (Figure B). The suppression of DNA damage responses was also clearly evident for 50 μM extracellular Asc. The protective effects of extracellular Asc were not underestimated due to its instability, as Asc losses during 3-h long incubations were minimal for both 50 and 500 μM concentrations (Figure C). Apoptotic responses reflect only one form of cell death. Therefore, we next examined a longer-term viability of cells, which integrates apoptotic and other forms of cell death. We found that 500 μM extracellular Asc was highly protective against chromate cytotoxicity, which provided a very high viability even at doses of Cr(VI) that were 100% lethal to cells without media supplementation with Asc (Figure D). The lower 50 μM extracellular Asc concentration also produced clear protective effects at a magnitude that was more similar to DNA damage markers (p53 and phospho-H2AX, Figure B) than apoptosis (Figure A). Consistent with the uptake measurements (Figure A), examination of cell viability showed that extracellular Asc offered a much more limited shield against toxicity of CaCrO4 particles using 1 h treatments of cells (Figure E). On the basis of the comparison of the dose–response curves, 500 μM extracellular Asc diminished cytotoxicity of CaCrO4 particles by 1.89-fold whereas its protective effect against chromate anions was 7.5-fold. The presence of Asc in media exhibited a much stronger cytoprotective power when cells were treated with CaCrO4 particles for 3 h (Figure F). In this case, 500 μM Asc decreased cytotoxicity of CaCrO4 by 5.5-fold, which was still less dramatic than protection against chromate anions in 3-h long treatments (18.9-fold, from slopes in Figure D). Longer 3 h treatments with CaCrO4 were clearly much more toxic in the absence of Asc in media (2.8-fold increase in the slope for 3 h versus 1 h), whereas 500 μM Asc-supplemented samples showed no further declines in cell viability (slopes ratio = 1.02) (Figure E,F). These results suggest that Cr(VI) uptake in longer than 1 h incubations is dominated by chromate anions released during dissolution of CaCrO4 particles. In agreement with this interpretation, 500 μM extracellular Asc strongly suppressed cellular accumulation of Cr during 3 h incubations with CaCrO4 (Figure G), which was only minimally affected in shorter 1 h treatments (Figure A). The uptake-suppressive effects of Asc in 3 h incubations were still higher for chromate anions than for CaCrO4 (2.0-fold differences in slopes), which can be attributed to the detoxification-resistant internalization of particles during the early exposure time.

Figure 4

Effects of extracellular Asc on responses to Cr(VI) in H460 cells. Cells were treated for 1 or 3 h in media containing 0, 50, or 500 μM Asc. (A) Westerns for apoptotic markers and (B) DNA damage-related responses at 18 h after 3 h-long treatments with solubilized chromate. (C) Stability of Asc in media during incubations with cells. Data are means ± SD, n = 3. (D) Cell viability at 48 h after treatments with solubilized chromate or (E) CaCrO4 particles for 1 h or (F) 3 h. Data are means ± SD, n = 3. Controls for CaCrO4-treated samples were treated with the corresponding concentrations of ethanol, which was present at <0.7% at the highest Cr(VI) dose. (G) Cr uptake after 3 h incubations with chromate anions (solubilized K2CrO4) and particulate CaCrO4 in the presence of 0, 50, and 500 μM Asc in media. Data are means ± SD, n = 3. When not visible, error bars were smaller than symbols.

Cultured cells contain very low amounts of intracellular Asc due to its absence in the commonly used media and its irreversible oxidation during preparation and storage of serum. We found that the overnight cultures of H460 cells contained only 17 ± 4 μM Asc (n = 3) in comparison to its millimolar concentrations in cells in vivo.[15,16] Asc-driven metabolism of Cr(VI) produces a different spectrum of intermediate Cr forms relative to thiols[17−19] and Asc restoration in cells altered cytotoxic and DNA damage responses to Cr(VI).[20,21] Supplementation of culture media with Asc is expected to raise intracellular concentrations of this reducer, which could have differentially affected cytotoxic responses to chromate anions and chromate particles. We found that the addition of both 50 and 500 μM Asc strongly increased intracellular concentrations of Asc during 1–3 h incubations (Figure A), which was probably responsible for the modest decreases in the extracellular Asc over the same time (Figure C). To assess potential effects of changes in intracellular Asc, we preincubated H460 cells with 500 μM extracellular Asc for 3 h, which raised cellular Asc to 0.84 mM, and then treated them with chromate anions in the presence of different concentrations of Asc in media. This experimental setup also showed dramatically protective effects of extracellular Asc on cytotoxicity of solubilized chromate (Figure B). Next, we preloaded cells with even higher Asc concentrations (6.1 ± 0.3 mM, n = 3), which minimized contributions from the inflow of extracellular Asc, and then assessed their ability to form colonies after treatments with chromate anions and CaCrO4 particles. Again, extracellular Asc was highly protective against chromate anions, with 500 μM Asc rescuing colony formation even at a dose of Cr(VI) that was completely lethal in the Asc-free medium (Figure C). In contrast, 500 μM extracellular Asc offered only a partial protection against lethality of CaCrO4 particles in Asc-loaded cells (Figure D). Thus, protection by extracellular Asc against cytotoxicity of chromate anions and chromate particles is unrelated to changes in intracellular Asc.

Figure 5

Cytotoxicity of Cr(VI) in H460 cells preloaded with Asc. (A) Accumulation of Asc in cells during incubations in culture media supplemented with Asc. Data are means ± SD, n = 3. (B) Viability of Asc-preincubated cells treated with solubilized K2CrO4 (chromate anions) for 3 h in the presence of different concentrations of Asc. Cells were incubated with 500 μM Asc for 3 h before the addition of chromate. Cell viability was measured at 48 h after chromate removal. Data are means ± SD, n = 3. (C) Colony formation by cells preloaded with 6.1 mM Asc and then treated for 3 h with chromate anions or (D) CaCrO4 particles in the presence of different concentrations of extracellular Asc. Cells were preloaded with Asc by incubations with 2 mM dehydroascorbic acid as described in Materials and Methods. In the experiment with CaCrO4, control dishes were treated with 0.7% ethanol to match its concentration in Cr(VI) samples.

Uptake and Cytotoxicity of Cr(VI) in Primary Human Bronchial Epithelial (HBE) Cells

Bronchial cells are the main transformation target for Cr(VI) in occupationally exposed individuals.[3] Therefore, we next examined interactions of Cr(VI) compounds with primary HBE cells. To expand investigation of poorly soluble chromates, we also prepared respirable-size particles of SrCrO4, which is another potent carcinogenic Cr(VI) compound in rat lungs.[22,23] Uptake measurements in normal HBE cells found striking effects of extracellular Asc on suppression of the cellular entry of chromate anions, as essentially no uptake was detectable in the presence of 500 μM Asc in culture media (Figure A). Although less dramatic, supplementation of media with 50 μM Asc was also very effective in inhibition of cellular accumulation of chromate anions, as evidenced by a significantly (p < 0.01) decreased uptake during 1 h incubation and no further changes during additional 2 h treatments. In contrast, uptake of CaCrO4 particles by HBE cells was practically unaffected by either 50 or 500 μM extracellular Asc (Figure B). Cellular accumulation of Cr during incubations with SrCrO4 particles was also unchnaged by the addition of low or high Asc concentrations to the media (Figure C). To validate our findings on Cr accumulation, we examined a long-term viability of HBE cells after Cr(VI) treatments. In full agreement with uptake measurements, we found that 500 μM extracellular Asc completely eliminated cytotoxicity of chromate anions, whereas 50 μM Asc was strongly cytoprotective (Figure D). Although uptake of chromate anions by H460 (Figure B) and HBE cells (Figure A) was similar, cytotoxic responses in HBE cultures without Asc addition were much lower than those in H460 cells treated under the same conditions. For example, 3 h-long treatment of H460 cells with 20 μM chromate resulted in almost complete loss of viability (Figure D), whereas the same dose produced less than 50% loss of viability in HBE cells (Figure D). We attribute a lower toxicity of chromate in HBE cells to their slow proliferation (typical for cultures of primary cells). The formation of DNA-double stranded breaks (DSBs) in human lung cells by chromate anions has been found to be replication-dependent,[38,39] indicating that poorly cycling cells would produce fewer toxic lesions.

Figure 6

Uptake and cytotoxicity of different forms of Cr(VI) in primary HBE cells. All panels show means ± SD, n = 3. (A) Time-dependent accumulation of Cr in cells incubated with chromate anions (solubilized K2CrO4), (B) CaCrO4 particles, or (C) SrCrO4 particles in the presence of 0, 50, or 500 μM Asc in culture media. Panels A–C use the same symbols and colors for line labeling. (D) Viability of cells treated with chromate anions (solubilized K2CrO4), (E) CaCrO4 particles, or (F) SrCrO4 particles in the presence of 0, 50, or 500 μM Asc in culture media. Cells were treated with Cr(VI) compounds for 3 h and their viability was measured 72 h later. Panels D–F use the same symbols and colors for line labeling.

Effects of Extracellular Asc in Rat Alveolar Cells

Since rats were used as the model species for testing lung carcinogenicity of chromates of different solubility,[22,23] we next assessed the effects of extracellular Asc in spontaneously immortalized rat lung epithelial cells, the RLE-6TN (rat lung epithelial-T-antigen negative) cell line. Similar to human cells, rat cells accumulated progressively less Cr when incubated with chromate anions in the presence of increasing concentrations of Asc in media (Figure A). The presence of 500 μM Asc severely suppressed uptake of Cr during the initial 1 h incubation and completely prevented entry of chromate anions into cells during the additional 2 h-long incubation. Using CaCrO4 particles, we found that extracellular Asc only minimally diminished Cr accumulation by cells during the first 1 h, but it was much more effective in inhibition of Cr uptake during the next 2 h (Figure B). As found for chromate anions, 500 μM Asc completely suppressed further Cr accumulation during the post-1 h treatment time. These results indicate that uptake of Cr(VI) from CaCrO4 initially occurs via internalization of particles whereas the subsequent Cr accumulation in cells is caused by entry of chromate anions arising from dissolution of particles. Consistent with its lower water solubility in comparison to CaCrO4, accumulation of cellular Cr during incubations with SrCrO4 was more modestly diminished by extracellular Asc (Figure C). In agreement with uptake results, cytotoxic effects of chromate anions in rat cells were significantly decreased by the addition of 50 μM Asc to the media and completely prevented by 500 μM extracellular Asc (Figure D). We used 6 h-long incubations with chromate anions for examination of their cytotoxicity in rat cells because shorter exposure times, such as 1 h or even 3 h used for human cells, did not produce sufficient losses of viability that are necessary for a reliable assessment of cytoprotective effects of extracellular Asc. Since prolonged incubations with Ca-chromate or Sr-chromate particles results in their dissolution, cytotoxicity studies with these Cr(VI) forms would not be informative and have not been performed.

Figure 7

Treatments of rat alveolar cells with different Cr(VI) forms. Data in all panels are means ± SD (n = 3). (A) Accumulation of Cr by RLE-6TN cells after incubations with 5 μM solubilized K2CrO4, (B) CaCrO4 particles, or (C) SrCrO4 particles in the presence of 0, 50, or 500 μM extracellular Asc. (D) Viability of RLE-6TN cells treated with solubilized K2CrO4 for 6 h. Cell viability was measured at 48 h post-Cr treatments.

Formation of DNA Double-Stranded Breaks (DSBs)

DSBs are critical genetic lesions produced by chromate anions in Asc-restored human lung cells.[38−40] Mechanistically, DSBs arise from recognition and processing of Cr-DNA adducts by mismatch repair in replicated DNA.[38,39,41] Thus, formation of DSBs can serve as an important test for the assessment of genotoxic activity of Cr(VI) compounds and intracellular delivery of chromate anions either directly or as a result of particle dissolution inside the cell. We have previously validated Ser139-phosphorylated H2AX (also known as γ-H2AX) as a marker of DSBs in chromate-treated cells by parallel physical measurements of these lesions.[39,40] Using westerns for phospho-Ser139- H2AX, we found that chromate anions (solubilized K2CrO4) and CaCrO4 and SrCrO4 particles produced similar levels of DSBs in Asc-restored H460 cells at 3 h recovery after 3 h-long treatments (Figure A). For example, in 5 μM Cr-treated cells, the combined amounts of mono- and diubiquitinated forms of phospho-H2AX were 100, 94.1, and 106.3% for K2CrO4, CaCrO4, and SrCrO4, respectively. At this concentration, all three Cr(VI) compounds also showed similar cellular levels of Cr accumulation at the end of 3 h-long incubations (Figure B). The observed abundance of mono- and diubiquitinated forms of phospho-H2AX at short postexposure times is typical for chromate anions-induced DSBs in cells with physiological levels of Asc.[42] This nonproteolytic ubiquitination acts as signal for the recruitment of DNA proteins involved in repair of DSBs.[43] Apoptotic DNA cleavage products can also stimulate H2AX phosphorylation but not its ubiquitination (as shown earlier in Figure B). A lack of apoptosis at the time of collection of cells for DSB detection is further confirmed by the absence of a caspase-mediated cleavage of PARP (Figure C). Thus, comparable levels of phospho-H2AX forms for all three Cr(VI) compounds indicate similarity in their genotoxic potencies when cellular uptake is not modulated by the presence of extracellular Asc. Since solubilized and particulate chromates deliver Cr(VI) into cells via different routes and in different forms, we also examined the formation of DSBs immediately after 3 h-long treatments. We found that at this time the amounts of ubiquitinated forms or total phospho-H2AX were the highest for chromate anions and the lowest for the less soluble SrCrO4 (Figure D,E). Although at the end of 3 h-long treatments the overall Cr accumulation was comparable for all three compounds, uptake of CaCrO4 and SrCrO4 particles occurred faster in comparison to chromate anions (Figure B). A gradual dissolution of internalized particles releasing Cr(VI) is the likely cause for the observed slower kinetics of DSB formation by SrCrO4 despite its faster uptake in comparison to chromate anions-treated cells.

Figure 8

Histone H2AX phosphorylation by solubilized and particulate Cr(VI) in Asc-restored cells. H460 cells were preincubated with 0.2 mM dehydroascorbic acid and then treated with Cr(VI) for 3 h. (A) Ser139-phosphorylated histone H2AX (p-H2AX) in cells collected at 3 h recovery post-Cr. p-H2AX-ub1 and p-H2AX-ub2 indicate mono- and diubiquitinated forms of phospho-H2AX. Tubulin was used as a loading control. (B) Time-dependent accumulation of Cr in cells treated with different Cr(VI) forms (all at 5 μM Cr). Data are means ± SD (n = 3). (C) Absence of apoptotic PARP cleavage in cells treated as in panel A. “Cleaved” indicates the expected position of the caspase-generated 89 kDa product. Fibrillarin was used as a loading control. (D) Representative Western blot for Ser139-phosphorylated histone H2AX in cells treated with 5 μM Cr(VI) for 3 h and collected immediately. K, solubilized K2CrO4; Ca, CaCrO4 particles; Sr, SrCrO4 particles. (E) Relative amounts of phospho-H2AX in cells treated as in panel D. The amounts of ubiquitinated (ub1+ub2) or all three p-H2AX forms (total) for K2CrO4 were taken as 100%. Means ± SD, n = 3, ∗p < 0.05, ∗∗p < 0.01 relative to K2CrO4.

Discussion

All Cr(VI)-containing compounds are considered as human respiratory carcinogens based on a large number of epidemiological studies demonstrating elevated lung cancer risks among occupational groups with inhalation exposures to more than one form of Cr(VI), including soluble compounds.[1−3] Because of their superior quality, data from two cohorts of US workers exposed to soluble Cr(VI)[27,44] were used by the Occupational Safety and Health Administration in the United States to estimate lung cancer risks for airborne Cr(VI) under a current permissible exposure limit.[45] However, comparative studies using intrapleural injections or intrabronchial implantations in rats have found a striking dependence of lung tumorigenicity on the solubility of Cr(VI) compounds.[22,23] Readily solubilized Cr(VI) forms releasing chromate anions upon dissolution produced no tumors, whereas moderately soluble Cr(VI) compounds such as CaCrO4 and SrCrO4 were highly tumorigenic. Our results suggest that the disparity between humans and rats in carcinogenicity of Cr(VI) compounds of different solubility was probably caused by the species differences in Asc concentrations in the lung extracellular fluid. We found that the Asc concentration corresponding to that in the rat lung fluid practically completely eliminated uptake of chromate anions and their toxicity in human and rat lung epithelial cells. In contrast, uptake and cytotoxicity of CaCrO4 and SrCrO4 particles were significantly less sensitive to the presence of high extracellular Asc, indicating their internalization in the form of particles. The protective effects of the 10-times lower Asc corresponding to its level in human lung fluids were proportionally smaller. Thus, the dramatically lower Asc content of the human lung extracellular fluid makes it a much less effective at detoxification of chromate anions through their reduction to cell-impermeable Cr(III). A relatively slow dissolution of CaCrO4 and SrCrO4 particles and their rapid internalization limit the detoxification potential of even high extracellular Asc, which explains high lung tumorigenicity of these compounds in rats.

Similar to highly soluble chromate salts, very poorly soluble Pb-chromate and Ba-chromate were also nontumorigenic in rats using intrapleural injections[22] or intrabronchial implantation models.[23] We suggest that high carcinogenicity of moderately soluble chromate forms in the rat lung can be attributed to two factors (Figure ):

Figure 9

Model for carcinogenicity of Cr(VI) compounds of different solubility in rat and human lungs. High concentrations of ascorbate (Asc) in the rat lung lining fluid rapidly detoxify extracellular chromate (CrO42–) produced by highly soluble Cr(VI) compounds or during a gradual dissolution of less soluble Cr(VI) particles. Dissolution of internalized Cr(VI) particles releases chromate that undergoes reduction to Cr(III) leading to Cr-DNA damage. A much lower (∼1/10th) concentration of Asc in the human lung fluid provides a more slow and limited detoxification of chromate anions, permitting their uptake by lung epithelial cells and the subsequent formation of Cr-DNA damage.

Relative resistance to detoxification by high extracellular Asc due to their limited dissolution outside of cells, permitting their uptake during early exposure times in the form of particles.

Dissolution and release of chromate anions inside the cells leading to DNA damage.

Intracellular dissolution of CaCrO4 or SrCrO4 particles was evidenced by their cytotoxicity even in the presence of high extracellular Asc. The release of chromate anions from internalized particles appeared to be relatively rapid, as the amounts of DNA breaks induced by exposures to particulate and solubilized chromates were similar at 3 h postexposure (Figure A). Studies with very poorly water-soluble PbCrO4 showed that its genotoxicity was caused by cellular uptake of chromate anions released during prolonged incubations in culture media, which promoted the solubilization process.[46] Although PbCrO4 particles were also internalized, they remained insoluble inside the cells and did not produce toxic effects.[46,47] High extracellular Asc in the rat ling fluid can effectively detoxify slowly released chromate anions preventing carcinogenicity of PbCrO4. This slow release of chromate anions should promote their detoxification even by low Asc levels in the human lung fluid, which is consistent with epidemiological findings on weaker carcinogenicity of PbCrO4 relative to more soluble chromates.[1,2]

Our data showed that high extracellular Asc corresponding to its concentration in the rat lung lining fluid acted as a very effective shield against uptake and toxicity of chromate anions. Exposure to moderately insoluble forms of Cr(VI) avoided this defense mechanism by delivery of chromate anions into the cells via internalization and dissolution of particles. These observations lead us to suggest that the differences in rat lung tumorigenicity of Cr(VI) compounds are largely caused by different intracellular doses of chromate anions. One prediction of this model is that soluble chromates should be tumorigenic even in rats when they are delivered in sufficiently high doses to deplete extracellular Asc. The results of a study by Steinhoff et al.[48] on a high threshold dose dependence of tumorigenicity of solubilized dichromate delivered via intratracheal instillations are consistent with this prediction. Intratracheal instillations deposited concentrated Cr(VI) solutions onto a small area, which would create a localized exhaustion of Asc (due to its oxidation by massive amounts of Cr-6) permitting uptake of chromate anions into the cells. Despite their much lower levels, Asc concentrations found in the human lung fluid were clearly protective against chromate anions, particularly in primary HBE cells. A preferential deposition of inhaled Cr(VI) in a small region of human lungs (area of bronchial bifurcation)[49] increases a risk for depletion of extracellular Asc during periods of particularly heavy exposures or in situations with coexposures to other oxidants, further weakening the detoxification potential of the lining fluid. Bypass of extracellular detoxification by chromate particles suggests that the effectiveness of a frequently used chemoprotective agent N-acetylcysteine, which primarily acts via reduction of chromate anion outside the cells,[50] will be beneficial largely for exposures to Cr(VI) compounds with high solubility. Effective Cr(VI) detoxification by extracellular Asc raises a question whether its levels in the lung lining fluid could be manipulated via dietary intake of vitamin C. This possibility has never been tested directly but the concentrations of Asc in plasma and the lung fluid did not show any significant correlation,[34] suggesting that the amounts of extracellular Asc in the lung are unrelated to its systemic levels. Concentrations of GSH in the lung fluid was approximately 100-times higher in comparison to plasma,[34] further demonstrating that the levels of antioxidants in the human lung fluid are regulated independently and do not exist in equilibrium with plasma levels.