Fanconi anemia complementation group A cells are hypersensitive to chromium(VI)-induced toxicity.

Fanconi anemia (FA) is an autosomal recessive disorder characterized by diverse developmental abnormalities, progressive bone marrow failure, and a markedly increased incidence of malignancy. FA cells are hypersensitive to DNA cross-linking agents, suggesting a general defect in the repair of DNA cross-links. Some forms of hexavalent chromium [Cr(VI)] are implicated as respiratory carcinogens and induce several types of DNA lesions, including ternary DNA-Cr-DNA interstrand cross-links (Cr-DDC). We hypothesized that human FA complementation group A (FA-A) cells would be hypersensitive to Cr(VI) and Cr(VI)-induced apoptosis. Using phosphatidylserine translocation and caspase-3 activation, human FA-A fibroblasts were found to be markedly hypersensitive to chromium-induced apoptosis compared with CRL-1634 cells, which are normal human foreskin fibroblasts (CRL). The clonogenicity of FA-A cells was also significantly decreased compared with CRL cells after Cr(VI) treatment. There was no significant difference in either Cr(VI) uptake or Cr-DNA adduct formation between FA-A and CRL cells. These results show that FA-A cells are hypersensitive to Cr(VI) and Cr-induced apoptosis and that this hypersensitivity is not due to increased Cr(VI) uptake or increased Cr-DNA adduct formation. The results also suggest that Cr-DDC may be proapoptotic lesions. These results are the first to show that FA cells are hypersensitive to an environmentally relevant DNA cross-linking agent.


Metals Toxicity
Certain chromium compounds are wellestablished human respiratory toxins and carcinogens for which adverse health effects are usually associated with occupational exposure (1). Epidemiologic studies carried out in Europe, Japan, and the United States have consistently shown that workers in the chromate production industry have an elevated risk of respiratory disease, perforation of the nasal septum, development of nasal polyps, and lung cancer (2,3). Certain inhaled particulate chromium compounds may persist in the lungs and cause long-term effects. Bifurcations of the lung bronchi of chromate workers have been found to contain millimolar concentrations of chromium more than two decades after cessation of exposure (4). The deposition of chromium in industrial waste, either in the form of dissolved chromium released to surface waters or chromate slag used in landfills, has also raised concerns about chromium as a potential environmental hazard (5). Investigators have identified more than 160 chromate production waste sites within Hudson County, New Jersey, distributed throughout the community in both industrial and residential areas. Frequently, the waste material contained levels of hexavalent chromium in the tens of parts per million (ppm) and sometimes in excess of 100 ppm (6).
If a cell sustains a genotoxic insult, the damage must be repaired or bypassed before replication to prevent death or the outgrowth of cells with potentially oncogenic alterations. If the damage is irreparable, the cell must be removed from the proliferating population. The predominant cellular fates in response to irreparable DNA damage are terminal growth arrest and apoptosis (28,29). Cell populations exposed to Cr(VI) have different spectrums of responses, depending on the extent of DNA damage (29).
Fanconi anemia (FA) is an autosomal recessive disorder characterized by diverse developmental abnormalities, progressive bone marrow failure, and a markedly increased incidence of malignancy. FA cells are hypersensitive to the DNA cross-linking agents diepoxybutane, cisplatin, and mitomycin C, suggesting a general defect in the repair of DNA cross-links (30)(31)(32)(33)(34). Eight genetic complementation groups have been described (FA-A through FA-H), with group A accounting for 60-65% of FA patients (30). FA-A cells are deficient in the repair of DNA interstrand cross-links (35); thus, FA-A fibroblasts were employed in this study. Our intent was to test the hypothesis that human FA-A cells would display an increased sensitivity to Cr(VI)induced cell death. The data suggest that the FA-A gene (FANCA) is essential for survival after exposure to Cr(VI).

Cell Culture
CRL cells (American Type Culture Collection CRL-1634 cells) are normal human foreskin fibroblasts isolated from a newborn black male. FA-A cells (Coriell Cell Repositories GM01309) are FA-A human fibroblasts isolated from a black male 12 years of age. Both CRL and FA-A cells were maintained in minimal essential medium Eagle-Earle media (Gibco, Grand Island, NY, USA) supplemented with 20% fetal bovine serum (Hyclone Laboratories, Inc., Logan, UT, USA), 2× essential and nonessential amino acids, vitamins, and 2 mM L-glutamine (Gibco). Cells were incubated in a 95% air, 5% CO 2 humidified atmosphere at 37°C.

Phosphatidylserine Translocation
The phosphatidylserine (PS) translocation assay was used to investigate the sensitivity of FA-A and CRL cells to Cr-induced apoptosis. This assay measures PS translocation from the inner (cytoplasmic) leaflet of the plasma membrane to the outer (extracellular) leaflet in the early stages of apoptosis. Annexin V protein has a strong, specific affinity for PS (36), and PS on the outer leaflet is available for binding labeled Annexin V. Positive cells exhibit green fluorescence around the plasma membrane. CRL and FA-A cells were seeded at 10 5 cells/60-mm dish and incubated for 24 hr prior to Cr(VI) exposure. Cells were treated with a final concentration of 0, 1, 3, 6, or 7 µM Cr(VI) for 24 hr in complete media. After 24 hr, the cells were rinsed twice with 1× phosphate-buffered saline (PBS) and incubated for an additional 24 hr in fresh media before analysis. Cells were gently harvested by trypsinization, combined with nonadherent cells from the culture medium, and centrifuged at 600×g for 5 min. Cell pellets were washed once in 1× PBS and resuspended in 100 µL binding buffer [10 mM HEPES (pH 7.4), 150 mM NaCl, 5 mM KCl, 1 mM MgCl 2 , 1.8 mM CaCl 2 ] containing 2 µL Annexin(V)-FLUOS (Roche, Indianapolis, IN, USA). Samples were incubated in the dark at room temperature for 15 min. Thirty microliters of each sample was loaded on a microscope slide and the percentage of Annexin(V)-FLUOS-stained cells was determined by counting five fields that contained at least 30 cells each on an Olympus AX70 microscope (Olympus, Lake Success, NY, USA) with a fluorescent filter set suitable for FLUOS analysis (excitation at 460-490 nm and emission at 515 nm).

Caspase-3 Activity
Caspase-3 is derived from the proenzyme CPP32 at the onset of apoptosis and plays a pivotal role in programmed cell death (37)(38)(39)(40). Caspase-3 exhibits the highest similarity to C. elegans cell death gene of the ICE (interleukin-1β-converting ezyme) proteases (41). Therefore, caspase-3 is an excellent biochemical indicator of apoptosis. The caspase-3 fluorescent assay detects a shift in fluorescence emission of 7-amino-4-trifluoromethyl coumarin (AFC). AFC is conjugated to a specific tetrapeptide sequence that normally emits blue fluorescence. After the substrate is cleaved by caspase-3, the liberated AFC emits a yellow-green fluorescence at 505 nm. Nearly confluent 150-cm 2 flasks were passaged at a 1:3 ratio and incubated for 24 hr prior to Cr(VI) exposure. CRL and FA-A cells were treated with a final concentration of 0, 1, 3, 6, or 7 µM Cr(VI) for 24 hr in complete media. After 24 hr the cells were rinsed twice with 1× PBS, the media was replaced, and the cells were incubated for an additional 24 hr before analysis. Cells were harvested by cell scraping and combined with nonadherent cells from the culture medium. Caspase-3 activity was determined using the FluorAce Apopain Assay Kit (Bio-Rad, Hercules, CA, USA) according to manufacturer's protocol. Cells were centrifuged at 600×g for 5 min. Cell pellets were rinsed once in 1× PBS and resuspended in 100 µL ice-cold apopain lysis buffer [10 mM HEPES, pH 7.4, 2 mM EDTA, 0.1% 3-[(3cholamidopropyl)dimethylammonio]-1propanesulfonic acid (CHAPS), 5 mM dithiothreitol (DTT), 1 mM phenylmethylsulfonyl fluoride (PMSF), supplemented with protease inhibitors (10 µg/mL pepstatin A, 10 µg/mL aprotinin, 20 µg/mL leupeptin) (Sigma, St. Louis, MO, USA)]. Cell suspensions were vortexed gently, then freeze/thawed 4 times by transferring from an isopropanoldry ice bath to a 37°C water bath. Cell suspensions were then centrifuged at 12,000×g for 30 min at 4°C to pellet cell debris. Protein concentrations of the resulting supernatants were determined with the DC Protein Assay II (Bio-Rad). Samples containing an equal amount of protein were diluted to 1 mL and transferred to wells of a CoStar 48-well dish (Corning, Corning, NY, USA). Ten microliters of apopain substrate (Ac-DEVD-AFC) (Bio-Rad kit) and 40-µL 25× reaction buffer (Bio-Rad kit) were added to each sample and control. Caspase-3 enzymatically cleaves the AFC from the peptide and releases free AFC, which then produces a blue-green fluorescence. The samples were gently mixed, and the fluorescence was determined using a Cytofluor 4000 fluorescence multiwell plate reader (PE Biosystems, Foster City, CA, USA) (excitation at 320-400 nm and emission at 505-555 nm). Fluorescence readings were taken at t = 0, t = 30, t = 60, t = 90, and t = 120 min after addition of substrate. The amount of AFC released from the substrate was determined by linear regression of an AFC standard curve (per manufacturer's instructions) at the time point at which the apopain positive control showed the greatest increase. Caspase-3 activity was determined and expressed as the ∆AFC/min.

Clonogenicity
CRL and FA-A cells were seeded at 10 5 cells/100-mm dish and incubated for 24 hr prior to Cr(VI) exposure. Cells were treated with a final concentration of 0, 0.01, 0.1, 0.5, 1, 2, or 3 µM Cr(VI) for 24 hr in complete media. Cells were washed twice with 1× PBS, collected by trypsinization, counted, and reseeded at 2,000 cells/100-mm dish in triplicate. The plates were incubated for 7-8 days and then rinsed with 1× PBS and incubated with crystal violet stain (80% methanol, 2% formaldehyde, and 2.5 g/L crystal violet) for 15 min at room temperature. The plates were thoroughly rinsed with distilled water and allowed to dry. Colonies were counted and the means ± SE of triplicate cultures were used to determine clonogenic survival as a percentage of control cultures.

Chromium Uptake
CRL and FA-A cells were seeded at 3 × 10 5 cells/100-mm dish and incubated for 24 hr prior to Cr(VI) exposure. Three extra dishes of each cell type were seeded for determining final cell number. Cr(VI) was prepared as above and spiked with Na 2 51 CrO 4 (ICN, Irvine, CA). CRL and FA-A cells were treated with a final concentration of 0, 3, or 7 µM Cr(VI) for 3 hr at 37°C. Following Cr(VI) treatment, cells were harvested by trypsinization and centrifuged at 300×g for 5 min at 4°C. Cell pellets were washed twice in 1× PBS and lysed in 500 µL lysis buffer (10 mM Tris-Cl, 0.5% sodium dodecyl sulfate [SDS], 0.5% Triton X-100). One hundred microliters of each sample was combined with Ecolite scintillation cocktail (ICN, Irvine, CA, USA). Disintegrations per minute (DPM) were determined on a Beckman LS3801 scintillation counter (Beckman Instruments, Fullerton, CA, USA). Final cell number was determined in replicate dishes on a Coulter cell counter (Coulter, Louton, UK). Data were normalized to cell number.

Chromium-DNA Adducts
Cr(VI) was prepared as above and spiked with Na 2 51 CrO 4 (ICN). CRL and FA-A cells were treated with 0, 3, or 7 µM Cr(VI) for 2 hr at 37°C. Following Cr(VI) treatment, cells were harvested by trypsinization and centrifuged at 300×g for 5 min at 4°C. Cell pellets were washed twice in 1× PBS and lysed in 500 µL lysis buffer (10 mM Tris-Cl, 0.5% SDS, 0.5% Triton X-100) containing 20 mg/mL proteinase K overnight in a 55°C water bath. Samples were then incubated with 10 mg/mL RNase for 4 hr in a 37°C water bath. DNA was extracted from the cell lysates, using phenol-chloroform extraction and ethanol precipitation. The DNA was then quantified by its spectrophotometric absorbance at 260 nm. DNA-bound 51 Cr was quantified by scintillation counting of an aliquot of each sample in EcoLite scintillation cocktail (ICN). DPM were then used to calculate Cr-DNA adducts per 10,000 DNA base pairs.

Statistics
Statistical significance was calculated using the Student t-test. Significance was defined as p < 0.05.

Results
Cr(VI) caused a dose-dependent increase in the number of Annexin V-positive FA-A cells (Figure 1). The percent apoptosis in the CRL cells, as determined by PS translocation, was 12.1 and 16.4% after treatment with 6 and 7 µM Cr(VI), respectively, for 24 hr followed by a 24-hr recovery period. This is similar to results shown in a previous report using human lung fibroblasts (42). In comparison, the percentage of apoptotic FA-A cells was significantly more than that of CRL cells at 43.9 and 52.4% at 6 and 7 µM doses, respectively.
To confirm the hypersensitivity of FA-A cells to Cr(VI)-induced apoptosis using a biochemical assay, we measured caspase-3 activity of extracts from Cr(VI)-treated CRL and FA-A cells. Cr(VI) caused a dose-dependent increase in caspase-3 activation in the FA-A cells (Figure 2). The FA-A cells again showed significantly more apoptosis than the CRL cells at 6 and 7 µM Cr(VI). The caspase-3 activity/µg protein in the CRL cells was 0.06 and 0.22 units after treatment with 6 or 7 µM Cr(VI), respectively, for 24 hr, followed by a 24-hr recovery period. In comparison, caspase-3 activity/µg protein in the FA-A cells was 0.75 and 1.59 units at 6 and 7 µM doses, respectively (12-to 14-fold increase).
Clonogenicity is an indicator of long-term cell survival and replicative potential after exposure to a toxic agent. Cr(VI) caused a dose-dependent decrease in clonogenic survival in both the CRL and the FA-A cells (Figure 3). The FA-A cells showed significantly less clonogenic survival after 1, 2, and 3 µM Cr(VI) treatment compared with CRL cells. The percent clonogenic survival for 1, 2, and 3 µM Cr(VI) was 28.7, 5.4, and 0.2% for the FA-A cells compared with 89.6, 18.6, and 4.4% for the CRL cells, respectively.
We performed Cr(VI) uptake analysis to determine if the differences in apoptosis and clonogenicity could be explained by differential Cr(VI) uptake. There was a dosedependent increase in Cr(VI) uptake from 0 through 7 µM Cr(VI) (Figure 4). However, there was no significant difference in Cr(VI) uptake between CRL and FA-A cells at any of the concentrations tested. Three and 7 mm Cr(VI) resulted in approximately 8.5 × 10 -4 and 1.4 × 10 -3 nmol Cr per cell, respectively.
We also measured total Cr-DNA binding in both the CRL and FA-A cells to determine if the differences in apoptosis and clonogenicity could be explained by differences in Cr-DNA adduct formation. Although Cr(VI) caused a dose-dependent increase in Cr-DNA adduct formation in both CRL and FA-A cells ( Figure 5), there were no significant differences in Cr-DNA adduct formation between CRL and FA-A cells at either the 3 or 7 µM dose. There was a slight but not statistically significant trend toward a higher adduct level in the CRL cells than in the FA-A cells. Three micromolar Cr(VI) resulted in approximately 0.02-0.03 Cr-DNA adducts per 10,000 base pairs. Seven micromolar Cr(VI) resulted in approximately 0.04-0.08 Cr-DNA adducts per 10,000 base pairs. Previous studies using a 10-fold higher concentration (75 µM) Cr(VI) resulted in approximately 2 Cr-DNA adducts/10,000 base pairs (27).

Discussion
The carcinogenic and transformagenic effects of Cr(VI) have been associated with Metals Toxicity • Hypersensitivity of FA-A cells to Cr(VI)  the cytotoxicity of certain Cr(VI)-containing compounds at the site of administration in vivo (1) and in cell culture systems (12,29). In human occupational settings, exposure to levels of particulate Cr(VI) compounds usually associated with lung cancers also effect high levels of cell death, which manifest as nasal perforations and respiratory tract ulcerations (1). Several investigations have established the induction of apoptosis by Cr(VI) in treated normal human fibroblasts (29,42,43); thus, understanding the mechanisms of Cr-induced toxicity and cell death may help elucidate mechanisms of carcinogenicity. The intracellular reduction of Cr(VI) generates several DNA-reactive species that can lead to myriad genetic lesions. Although a wealth of information exists concerning the genotoxicity of Cr(VI), the specific types of DNA damage that may be responsible for the toxic effects of Cr(VI) have not yet been determined. At least one type of genetic lesion formed as a result of Cr(VI) reduction (Cr-DDC) results in guanine-specific arrest of DNA replication (26,27). Although Cr-DDC may represent lethal lesions formed by Cr(VI), information is lacking on the effects of Cr(VI) on DNA cross-link repair-deficient cells. The focus of this investigation was to examine the cellular effects of Cr(VI) on FA-A cells, which are specifically deficient in the repair of DNA interstrand cross-links (35).
The ultimate fate of a cell exposed to a genotoxin such as Cr(VI) is heavily dependent upon the severity of the initial insult. At low concentrations, cells exposed to Cr(VI) primarily undergo growth arrest, presumably to allow time for repair. As the concentration of Cr(VI) is increased, the predominate cell fate becomes terminal growth arrest followed by apoptosis (29). The data from the present study are consistent with this model. The clonogenicity assay measures the long-term survival of a population after Cr(VI) exposure and accounts for both growth arrest and apoptosis. We found that low doses of Cr(VI) (0.1-3 µM) produced a dramatic decrease in clonogenic survival (Figure 3). However, a marked increase in the number of apoptotic cells measured at a specific point in time [24 hr after the Cr(VI) exposure] was not detected at these doses (Figures 1, 2). These results suggest that terminal growth arrest, not apoptotic cell death, is the primary cell fate at this low Cr(VI) concentration, and that FA-A cells are more susceptible to undergo growth arrest at these low doses compared with control cells. At higher concentrations (6-7 µM), FA-A cells demonstrated a statistically significant increase in apoptosis. These data show that FA-A cells are markedly hypersensitive to both the growth-inhibiting and apoptotic effects of Cr(VI). These results are the first to show that FA cells are hypersensitive to an environmentally relevant DNA cross-linking agent.
The hypersensitivity of FA-A cells toward Cr(VI)-induced growth inhibition and apoptosis was not the direct result of higher Cr-DNA binding or Cr(VI) uptake. There were no differences found in Cr(VI) uptake or Cr-DNA adduct formation between FA-A and CRL cells. It is interesting however, that human FA-A fibroblasts were markedly hypersensitive to Cr(VI)induced apoptosis compared with CRL cells. These data suggest that fewer Cr-DNA adducts (mono-and bifunctional) are required to induce equivalent growth arrest and apoptosis in FA-A cells relative to CRL cells. Additionally, the similar steady-state Cr-DNA adduct levels observed in this study suggest that either Cr-DNA adducts are resistant to removal in repair-competent cells or FA-A cells do not display a deficiency in Cr-DNA adduct removal.
Although the difference in Cr-DNA adduct levels between FA-A and CRL cells was not significant, it is interesting that there was a trend showing a decreased level of Cr-DNA adducts in FA-A cells exposed to relatively high doses of Cr(VI). This result may be caused by the marked increase in cytotoxicity in FA-A cells exposed to this dose of Cr(VI). The measurement of adduct levels and the magnitude of statistical error may be affected as the majority of cells progress through apoptosis. Alternatively, perhaps the lack of a functional DNA cross-link repair mechanism in FA-A cells causes an altered cellular response to Cr-DNA adduct formation. It may be possible that other molecular pathways designed to prevent or repair the formation of DNA adducts are upregulated in FA-A cells. Whatever the case, our results show that FA-A cells are more susceptible to Cr(VI)induced growth arrest and apoptosis without exhibiting an increase in Cr-associated DNA.
Cr(VI) produces several types of DNA damage, and it is therefore possible that lesions other than Cr-DDC may contribute to the differential sensitivity of FA-A cells to chromium toxicity. FA-A cells are specifically deficient in repair of DDC, but one must consider the possible role of chromium-induced DPC (Cr-DPC) as well. We have shown that Cr-DPC are preferentially formed in and repaired from nuclear matrix DNA (44). Others have shown that the vast majority of Cr-DPC are actually individual amino acids or small peptides (such as glutathione) coordinated to DNA phosphate through ionic interactions (45), and that this lesion is premutagenic (46). We and others have found that this lesion correlates strongly with altered gene expression but not DNA replication (47). We have recently shown that DPC are not polymerase-arresting lesions and that their formation actually precludes the formation of polymerase-arresting lesions (48). Combining information published by our laboratory and others, we can identify experimental manipulations that alter DPC and toxicity in an inverse fashion (14,21,43,49). Finally, although there is no information on FA-A cells and repair of DPC, we think it is unlikely that the mechanism of repair of this single-stranded, non-polymerase-arresting, complex monoadduct would use a repair pathway designed to remove or bypass lethal DDC, which block progression of the replication fork.
In summary, human FA-A fibroblasts were markedly hypersensitive to Cr(VI)-induced apoptosis compared with CRL cells, although there were no differences found in Cr(VI) uptake or Cr-DNA adduct formation between FA-A and CRL cells. Because Cr(VI) produces Cr-DDC (24)(25)(26)(27), and FA-A cells are known to be deficient in the repair of these lesions, these results suggest that Cr-DDC are terminal growth-arresting and/or proapoptotic lesions. We are currently exploring the differential formation and repair of polymerasearresting lesions in FA-A and CRL cells.