|Home | About | Journals | Submit | Contact Us | Français|
Hexavalent chromium (Cr[VI]) is classified by the International Agency for Research on Cancer as a group I carcinogen. Although the U.S. Occupational Safety and Health Administration was obliged to reduce the permissible exposure limit (PEL), it was reported that U.S. workers continue to be exposed to dangerously high Cr(VI) levels. In this study, we examined the role of p53 and target genes in a bronchoalveolar carcinoma isogenic cell line system and in primary human bronchial epithelial cells. p53-Negative parental H358 cell line, the same line in which the wild-type p53 expression vector (pC53-SN3) was introduced, and cells obtained from biopsies of human bronchus were exposed to chromate. Induction of DNA strand breaks were evaluated by alkaline elution assay, and apoptosis was analyzed by gel ladder, annexin V-PI staining, and ELISA, whereas p53 and target genes were evaluated by Western blots. Although Cr(VI) induced DNA strand breaks in both H358 cell clones, apoptosis was present only in the p53-transfected cells (H358p53+/+). In these cells, Cr(VI)-induced apoptosis is mediated by p53 upregulation of p53-upregulated modulator of apoptosis (PUMA), BAX translocation to mitochondria, cytochrome c release, and caspase-3 activation. In primary human bronchial epithelial cells expressing functional p53, Cr(VI) induced expression of PUMA and Noxa, which promote apoptosis through BAX. This result establishes p53 as the “necessary” player in Cr(VI)-induced apoptosis. To the best of our knowledge, this is the first report indicating strict correlation of Cr(VI) apoptosis to PUMA induction on primary human bronchoalveolar cells in short-term cultures.
Hexavalent chromium (Cr[VI]) is classified by the International Agency for Research on Cancer (IARC) as a group I carcinogen (1). Cr(VI)-containing compounds are widespread in cigarette smoke, automobile emissions, and in the environment (e.g., Cr[VI]-contaminated water). These compounds are commonly used in the chemical industry, artistic paints, anticorrosion paints, electroplating, and stainless steel welding (1–4). Although the U.S. OSHA reduced the permissible exposure limit (PEL), a recent study (5) reported that U.S. workers continue to be exposed to dangerously high Cr(VI) levels.
The lower respiratory tract is the target organ of Cr(VI) exposure, and its accumulation in lung tissue is found in workers with occupational exposure and in cigarette smokers (1–4). Epidemiologic studies have consistently shown that occupational exposure is strongly associated with a higher incidence of lung cancer (1, 6, 7). Exposure increases the incidence of lung cancer in cigarette smokers (1, 6, 8), thus the cancer morbidity rate for smokers who were formerly K2CrO4 workers, with 9 or more years of exposure, is 21.6 times higher than that of nonsmokers (1, 6).
Intracellular reduction of Cr(VI) to Cr(III) leads to the extensive formation of DNA-phosphate–based adducts that cause a variety of genetic damage (1, 9–16), which in turn determines the activation of the p53 signaling pathway and cell cycle arrest or apoptosis (9–16). Moreover, Cr(VI) induces the ataxia–telangiectasia mutated (ATM)-dependent DNA damage response pathway, which is paradoxically required for both apoptosis and survival after Cr(VI) insult (16, 17). Nucleotide excision repair (NER) is the principal repair mechanism for Cr(VI)-DNA adducts in human cells (18). However, Cr(VI) exposure causes the inhibition of NER, which enhances carcinogen-induced mutagenicity and cytotoxicity (11, 16). Although the inhibition of NER is an important mechanism for Cr(VI)-induced human carcinogenesis, the underlying molecular mechanisms of Cr(VI)-induced carcinogenesis remain unclear.
In this study, we examined the role of p53 in a bronchoalveolar carcinoma isogenic cell line system consisting in the p53-negative parental H358 cell line and parental line in which the wild-type p53 expression vector (pC53-SN3) was introduced. Because the only difference between these two cell lines is the absence or presence of a single gene, the interpretation of results is particularly straightforward and uncomplicated by the overexpression of exogenous genetic elements. Although Cr(VI) induced DNA strand breaks in both cell clones, the induction of apoptosis, mediated by p53 upregulated modulator of apoptosis (PUMA), Bax translocation to mitochondria, cytochrome c release, and caspase-3 activation, is present only in the p53+/+ clones. This result establishes p53 as the “necessary” player in Cr(VI)-induced apoptosis. To characterize the signaling pathways activated by Cr(VI) in primary human bronchial epithelial cells, we examined its induced effects in short-term primary cultures obtained from biopsies of human bronchus.
Human bronchoalveolar carcinoma H358 cell line, obtained from the American Type Culture Collection (Rockville, MD), were grown in RPMI 1640 (Gibco BRL, Grand Island, NY) supplemented with 10% FBS (Gibco BRL).
To obtain primary cultures of human bronchus epithelium, samples of bronchial tissue were excised from surgical specimens of resected non–small cell lung cancer from patients presenting normal lung function. The procedure was approved by the ethics committee. Biopsies were assessed by conventional H&E staining. After excision, the bronchial samples were washed and incubated overnight at 4°C with 0.38 mg/ml hyaluronidase, 0.75 mg/ml collagenase, 1 mg/ml protease, and 0.3 mg/ml DNase in RPMI 1640 medium, and then filtered through a 70-mm mesh nylon strainer. After centrifugation, epithelial cells were resuspended in small airway epithelium basal medium (Clonetics; BioWhittaker, San Diego, CA) supplemented with 0.5 μg/ml human recombinant epidermal growth factors, 7.5 mg/ml bovine pituitary extract, 0.5 mg/ml epinephrine, 10 mg/ml transferrin, 5 mg/ml insulin, 0.1 μg/ml retinoic acid, 6.5 μg/ml triiodothyronine, 50 mg/ml gentamicin, 50 μg/ml amphotericin B, and 50 mg/ml BSA/fatty acid–free. Cells were cultured on plates precoated with coating media containing: 29 μg/ml collagen (vitrogen; Collagen Corp., Palo Alto, CA), 10 μg/ml BSA (Biofluids, Inc., Rockville, MD), and 10 μg/ml fibronectin (Calbiochem, La Jolla, CA) for 5 min. Fresh complete medium was replaced every 2–3 d until cells were confluent. Upon confluence, the cells were lifted by 1× trypsin-EDTA (Life Technologies, Inc., Gaithersburg, MD) and subcultured at a 1:2 dilution. Third-fourth passage confluent cultures were used for all the experiments. For in vitro testing of nontumorigenicity of each primary cell line, the anchorage-independent assay was performed using a soft-agar clonogenic method, as described previously (19). The cells were identified as bronchoalveolar by immunocytochemical staining.
Cells were treated as reported by Sugiyama and colleagues (20). Colonies were scored after 10 d.
p53 and p21waf-1 induction were analyzed in FacsCalibur (BD Biosciences, San Jose, CA).
DNA damage was evaluated by the alkaline filter elution assay under deproteinizing conditions, as described previously (21).
Western blotting was performed as described previously (21, 22). Antibodies used were: monoclonal human p53 (DO-1), tubulin, caspase-3, human cleaved caspase-3 (Asp 175), and BCL-XL (Cell Signaling Technology, Beverly, MA); human p21 (polyclonal), PUMA, NOXA, and BAX (Oncogene; Dako Cytomation, Carpinteria, CA); human cytochrome oxidase IV (clone 1A12; Molecular Probes, Inc., Eugene, OR, and Cell Signaling Technology).
Cytosolic fractions (supernatant) and mitochondrial fractions (pellet) were obtained as described previously (23).
A total of 5 × 105 cells were incubated in the presence or absence of 200 μM chromate (K2CrO4) together with the PE-labeled affinity-purified anti-caspase-3 Abs (Pharmingen, San Diego, CA), that binds irreversibly to activated caspases-3, and then analyzed by flow cytometry using FacsCalibur. Alternatively, the enzymatic activity of caspase-8 and caspase-3 was assayed fluorometrically.
Immunofluorescence analysis was performed as described previously (19).
The bronchoalveolar carcinoma isogenic cell line system consisted of the p53-negative H358 parental cell line and the parental cell line in which the wild-type p53 expression vector (pC53-SN3) was introduced, as described previously (24) (parental cell line courtesy of the Medicine Branch (National Cancer Institute, National Institutes of Health, Bethesda, MD).
A different dose–response relationship was seen between parental H358p53−/− cells and transfected H358p53+/+ clones (Figure 1A). Cells transfected with wild-type p53 were sensitive to K2CrO4 treatment with an inhibitory concentration producing a 50% suppression (IC50) value of 200 ± 3.4 μM. On the contrary, in the parental cell line, H358, the IC50 was not reached; treatment with 500 μM K2CrO4 induced a survival fraction of 64%, suggesting that K2CrO4-induced cytotoxicity is p53-dependent (Figure 1A).
When both cell lines were incubated for 2 h at the equimolar concentration of 200 μM K2CrO4, a large amount of protein-associated DNA single-strand breaks (SSB) (Figure 1B) was generated, both in the parental cell line (p53−/−) and in the transfected clones, suggesting that p53 is not involved in DNA-SSB formation induced by Cr(VI). In these cells, p53 protein levels started to increase after 1 h of recovery, and were maximum after 4 h of recovery. It was observed only in H358p53+/+ clones (Figure 1C). Interestingly no p21waf-1, the expression of which is largely regulated at the transcriptional level by p53-dependent mechanism (25), was induced (Figure 1C).
All the following experiments were performed using the concentrations of K2CrO4 needed to reach the IC50 value in H358p53+/+ clones and considered equimolar for both the parental and the transfected cells. Cells were treated with 200 μM K2CrO4 for 2 h, then the K2CrO4 was washed out and cells were incubated until 24 h of recovery in K2CrO4-free medium.
DNA ladder formation, characteristic of apoptosis, was detectable after 4 h of recovery and increased significantly until 24 h of recovery in H358p53+/++/+ clones; low levels of DNA ladder was detectable after 24 h of recovery in the parental cell line H358p53−/− (Figure 2A). Annexin–V-PI assay, performed after 24 h of recovery (Figure 2B), confirmed these observations.
Because these data suggested that Cr(VI) triggered apoptosis in H358p53+/++/+ clones via p53 signal, p53 target gene products (25), such as PUMA, NOXA, BAX, and BCL-XL, were analyzed. According to p53 induction, PUMA and NOXA proteins were induced in a time-dependent manner (Figure 3A), but only in H358p53+/++/+ clones.
In these clones, a significant fraction of BAX translocated from the cytosol to the mitochondria (Figure 3B), and BCL-XL was downregulated (Figure 3C). As a result, some second mitochondria-derived activator of caspases, such as Smac/Diablo (Figure 3B), cytochrome c (Figure 4A), AIF, and endonuclease G, were released from mitochondria to the cytoplasm (Figure 4B). As a final consequence, caspase-3 was activated (Figures 5A–5D). When H358p53+/+ clones were incubated with 200 μM K2CrO4 in the presence of the specific inhibitors of caspase-3 z-DEVD.fmk 30 nM (26), no induction of apoptosis was observed (annexin V/PI assay) (Figure 5E) and, as expected, caspase-3 activation was inhibited (Figure 5D).
In conclusion, K2CrO4 induced cell cytotoxicity and mitochondria-dependent caspase-3–mediated apoptosis via p53 induction.
Hexavalent chromium exposure experiments were performed in human bronchoalveolar cells freshly isolated from surgical specimens. Short-term culture cells were obtained from six human bronchial biopsies identified as non-neoplastic bronchoalveolar cells after H&E staining (Figure 6A). The failure of these cells to proliferate in soft agar (anchorage-independent assay) supported the nontumorigenicity of each primary cell line (Figure 6A). These primary cells were considered bronchial epithelial cells after immunohistochemistry analysis. These cells strongly stained with the mouse antihuman cytokeratin monoclonal antibodies 34βE12 and RCK102, while no significant immunostaining was found with mouse anti-human vimentin monoclonal antibodies, anti-CD3, or anti-CD68, thus excluding contamination of cultures by fibroblasts, lymphocytes, or macrophages (Figure 6B). These cells displayed normal p53 function, as p53 protein levels were increased after treatment with campthotecin, a positive p53-inducing agent (27), at 1.0 μM for 2 h (Figure 6C).
A dose–response relationship was seen in primary cells treated with different concentrations of K2CrO4 (Figure 7). Primary human bronchoalveolar cells were more sensitive to K2CrO4 than was the bronchoalveolar carcinoma cell line (compare Figure 1A with Figure 7A). The IC50 values covered a range from 18 to 82 μM (Figure 7B), suggesting that primary cells are very sensitive to K2CrO4.
When cells were incubated for 2 h with K2CrO4 at their respective IC50 values, followed by 2 h of recovery, p53, PUMA, and NOXA (Figures 8A, 8B, and 8D), but not p21waf-1, proteins were induced (Figure 8C). In the same pool of cells, caspase-3 was activated (Figure 9A). Apoptosis started 3 h after recovery, increased with time, and rose to the maximum after 9 h of recovery (Figures 9B–9C), as evaluated by ELISA and ladder assays.
When both H358 p53+/+ and primary human bronchoalveolar cells in short-term cultures were preincubated with PUMA antisense for 2 h, then treated with K2CrO4 for an additional 2 h at their respective IC50 values, followed by 2 h of recovery, no induction of PUMA proteins was observed (Figure 10A). Under these experimental conditions, no induction of apoptosis was observed (Figure 10B). These experiments showed that PUMA antisense inhibited apoptosis induced by Cr(VI).
The analysis of the apoptosis induced by Cr(VI) in the isogenic human bronchoalveolar carcinoma cell system consisting of H358p53−/− parental cells and their derivative H358p53+/+, and in human primary bronchoalveolar cells displaying normal functional p53, revealed that p53 is necessary for Cr(VI)-induced expression of the BH3 domain–containing proteins, PUMA and NOXA. These proteins promote apoptosis through the multidomain Bcl-2 family member, BAX. When PUMA expression was inhibited by specific antisense, no apoptosis was observed.
The tumor suppressor protein, p53, mediates some parts of the response of mammalian cells to DNA damage, either by stimulating DNA repair or, beyond a certain threshold of DNA damage, by initiating apoptosis. The primary role of activated p53 is to act as a transcription factor to induce the expression of proteins involved in cell cycle arrest or apoptosis. The transcriptional activation of the cyclin-dependent kinase inhibitor CDKN1A (p21waf-1/cip1) is the key event for growth arrest after p53 activation (25). Alternatively, p53 transactivates a series of proapoptotic proteins from the BCL2 family; in particular, BAX, PUMA, and NOXA (25, 28, 29), which induce mitochondrial-membrane permeabilization and, therefore, release apoptogenic factors from the mitochondrial intermembrane space (25, 28, 29). p53 Can also induce apoptosis in a transcription-independent manner by acting directly at the mitochondrial level (25), although the extent of this for DNA damage–induced apoptosis is controversial. Whatever the mechanisms, p53 activates the intrinsic mitochondrial apoptotic pathways in response to DNA damage by inducing the expression of at least three BCL-2 proapoptotic family members (BAX, PUMA, and Noxa), shifting the balance toward proapoptotic effects. The formation of apoptosomes requires the release of cytochrome c and APAF-1 from mitochondria. The cytoplasm efflux of cytochrome c is the key event, with consequent mitochondrial release of Smac/DIABLO (second mitochondria-derived activator of caspases), which promotes the transactivation of caspase-9 and downstream effector caspase-3. Caspase-3 executes cell death (25).
In our experiments, Cr(VI) increased the expression of p53 protein levels, but did not induce p21waf-1, the principal mediator of G1 growth arrest, whereas p53 increases strongly promote apoptosis. It has been proposed (30) that the two outcomes induced by p53 (growth arrest or apoptosis) have a profound influence on the tumorigenic effects of genotoxic damage. As an example, O'Reilly and colleagues (31) suggested that p21waf-1 protects the lung from oxidative stress, in part by inhibiting DNA replication and thereby allowing additional time to repair damaged DNA.
The proapoptotic and cell cycle–arresting functions of p53 have been attributed to distinct transcriptional profiles (e.g., increased transcription of BAX, as in the case of Cr[VI]-treated cells , for apoptosis induction versus transcription of p21waf-1 for cell-cycle arrest). These profiles correlate, to some extent, with the phosphorylation of p53 on serine (Ser) 15, 20, or 46 (30). Experiments in transgenic mice have indicated that lack of phosphorylation of residues 15 or 20 does not affect growth arrest, but reduces the apoptotic response (30). Cr(VI) induced phosphorylation of p53 at Ser15 (33). A further potential mechanism for determining apoptosis versus p21waf-1 induction is the coexpression of proteins that act to repress p21waf-1 expression; one example is c-jun (34), which is enhanced after Cr(VI) exposure (32).
In conclusion, the response to genotoxic stress mediated by p53 is a highly complex process, as it must be to maintain viability. Tumors are very complex, and their initiation and progression involve disruption of interacting mechanisms that normally act to maintain tissue homeostasis after genotoxic damage. Although apoptosis eliminates the majority of damaged cells, a stochastic fraction of cells may survive (Figure 11) with unrepaired DNA damage. These cells may exhibit a predisposition to genomic instability and neoplasia. Hirose and colleagues (35) reported that a longer period of K2CrO4 exposure was associated with a tendency toward a higher frequency of microsatellite instability (MSI). In their study, 30 (78.9%) out of 38 tumors present in the K2CrO4 lung cancer group exhibited MSI. In contrast, only 4 (15.4%) out of 26 tumors present in the non-K2CrO4 lung cancer group exhibited MSI. The authors suggested that MSI may play a role in chromium-induced carcinogenesis, and that the carcinogenic mechanism of K2CrO4 lung cancer may differ from that of non-K2CrO4 lung cancer.
Apoptosis is a survival mechanism aimed at maintaining good cellular events; however, adaptive responses to genotoxic stress may paradoxically generate mechanisms that can drive to carcinogenesis. In culture, a progressive change was observed in the phenotype of cells exposed repetitive times to Cr(VI); one such observation was of an increased resistance to apoptosis (36). The altered apoptotic susceptibility in these cells may be relevant to the response of Cr(VI) exposure in individuals with long-term repetitive exposures. Lung tissue that has undergone cellular turnover in response to recurrent cytotoxic Cr(VI) exposure may present an attenuated response to subsequent exposures, This implies that individuals with longer-term occupational exposure may be more susceptible to Cr(VI)-induced lung cancer due to selection for apoptosis-resistant cells.
Finally, considerable literature supports a role of apoptosis in the remodeling of lung tissue after acute lung injury (37–40). For example, Cr(VI)-related lung cancers are often associated with respiratory toxicity involving high levels of cell death, such as perforation of the nasal septum and/or respiratory tract ulcerations (1). In response to injury to the bronchial epithelium, there is an immediate requirement to initiate tissue repair and restore barrier function (41). When the degree of lung injury is mild, damaged tissue will normally be repaired, whereas excess cell death may lead to irreparable lung damage and pulmonary fibrosis. Pulmonary fibrosis is associated with an increased risk of lung cancer (40).
We have explored the possibility that pulmonary toxicity from Cr(VI) is due in part to apoptosis caused by DNA damage. Several studies have shown that intracellular reduction of Cr(VI) to Cr(III) leads to extensive formation of DNA phosphate–based adducts that cause different genetic damage, such as chromosomal aberration, sister-chromatid exchange, DNA strand breaks, and DNA-DNA and DNA-protein cross-links (9, 14). Using isogenic human bronchoalveolar carcinoma cells (H358 cell system) that differ only in their p53 status, we have shown that the induction of DNA strand breaks is independent of the status of p53, but, importantly, the resolution of these breaks is p53-dependent. In p53 wild-type cells (transfected cell clone and primary human bronchoalveolar cells), PUMA induction is essential for Cr(VI)-induced apoptosis, as demonstrated by experiments in the presence of PUMA antisense (Figure 11). To the best of our knowledge, this is the first report indicating that Cr(VI)-apoptosis strictly correlates to PUMA induction on primary human bronchoalveolar cells. Moreover, Cr(VI) determined the release of Smac/DIABLO, as well as apoptosis-inducing factors, such as AIF and endoprotein G.
PUMA was initially identified as a gene activated by p53 in cells undergoing p53-induced apoptosis (28, 42, 43), and as a protein interacting with Bcl-2 (29). PUMA and another p53 target, Noxa, share homology with Bcl-2 family proteins. PUMA is extremely effective in inducing apoptosis: when expressed, it kills cancer cells within a few hours (28, 44). Importantly, gene knockouts in human colorectal cancer cells showed that PUMA was required for apoptosis induced by p53, hypoxia, and DNA-damaging agents (43). Finally, PUMA acts by modulating Bax activity to facilitate cytochrome c release from the mitochondria, thereby triggering the apoptotic cascade (44).
Overall, our results suggest that the mitochondrial death pathway mediating apoptosis is activated by Cr(VI), and that Cr(VI) is directly genotoxic to relevant lung target cells, such as bronchoalveolar cells. Cr(VI)-induced DNA damage in p53 wild- type cells triggers complex cellular signaling pathways that determine cell fate. These responses include increases in p53 protein levels and transcriptional activation of p53 targets genes, such as PUMA and NOXA, which, in turn, trigger mitochondrial apoptosis.
Originally Published in Press as DOI: 10.1165/rcmb.2005-0213OC on September 15, 2005
Conflict of Interest Statement: None of the authors have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.