Mechanisms of action of the carcinogenic heterocyclic amine PhIP

Toxicology Letters 168 (2007) 269–277

Mechanisms of action of the carcinogenic heterocyclic amine PhIP N.J. Gooderham ∗ , S. Creton, S.N. Lauber, H. Zhu Biomolecular Medicine, Imperial College London, Sir Alexander Fleming Building, London SW7 2AZ, UK Available online 16 November 2006

Abstract Formed during the cooking of meat, the heterocyclic amine 2-amino-1-methyl-6-phenylimidazo[4-5-b]pyridine (PhIP) is mutagenic and carcinogenic. Although the metabolism and mutational effects of PhIP are well defined, the early cellular and genomic events by which it can induce neoplastic transformation are not yet fully characterised. These early cellular responses to genotoxic doses of PhIP were examined in a human mammary epithelial cell, MCF10A. Using Western blotting, PhIP was shown to induce expression of the DNA damage response proteins p53 and p21WAF1/CIP1 , and to inhibit cell growth while activating G1 cell cycle checkpoint, a consequence of PhIP-induced DNA damage. Using low doses of PhIP (previously shown to activate oestrogenic signalling), PhIP increased proliferation in the oestrogen receptor (ER)-negative MCF10A cell line and to activate the mitogen-activated protein kinase (MAPK) pathway. Inhibition of this pathway significantly reduced the PhIP-induced cell growth of MCF10A cells. The work presented here suggests that, further to its genotoxic properties, at levels close to human exposure PhIP stimulates cellular signalling pathways that are linked to the promotion and progression of neoplastic disease. It is possible that a combination of these DNA damaging and growth promoting properties provide a mechanism for the tumourigenicity of PhIP, and may be key determinants for the tissue specificity of PhIP-induced carcinogenesis. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Heterocyclic amines; Oestrogens; Signal transduction

1. Introduction Diet has long been recognised as one of the major factors that can influence the development of cancer (Doll and Peto, 1981; WCRF, 1997). Consumption of meat is positively correlated with human cancer and the cooking of meat is known to generate chemical carcinogens of extreme genotoxic potency, including a family of heterocyclic amines (Sugimura, 1997). The most abundant of these heterocyclic amines, 2-amino-1-methyl-6phenylimidazo[4-5-b]pyridine (PhIP), has been shown

∗ Corresponding author. Tel.: +44 20 7594 3188; fax: +44 20 7594 3050. E-mail address: [email protected] (N.J. Gooderham).

to specifically induce tumours of the colon, breast and prostate in rats (Ito et al., 1991; Shirai et al., 1997), which, co-incidentally are the three commonest sites of diet-associated cancer in Western society. In realising its mutagenic potential, PhIP requires metabolic activation (by CYP1A2) and induces fundamental changes in the biochemistry of the cell that result in a number of genes being differentially expressed. The initial consequences of exposure to activated PhIP would seem to involve damage to DNA (Zhu et al., 2000). Cells that survive these early events appear to maintain a damaged gene set that leads to their demise by apoptosis (presumably extensive damage) or survival with a mutated genome. Recently, we have shown that PhIP possesses oestrogenic activity at very low doses (10−9 to 10−11 M) that can invoke a mitogenic response (Lauber et al.,

0378-4274/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.toxlet.2006.10.022

270

N.J. Gooderham et al. / Toxicology Letters 168 (2007) 269–277

2004). Since human consumption of cooked meat can lead to PhIP concentrating in body fluids including milk (DeBruin et al., 2001), there is a distinct possibility that low-dose PhIP is oestrogenic and mitogenic in humans. In this context, the following points are striking: (i) PhIP specifically induces tumours of the breast, prostate and colon; (ii) the major sites of diet associated cancer in humans are breast, prostate and colon; (iii) each of these tissues is strongly influenced by hormonal oestrogenic chemicals; (iv) the primary endogenous oestrogens are known to contribute to cancer burden in these tissues. Oestrogens play a crucial role in regulating the growth, differentiation and functioning of many reproductive tissues including the breast, uterus, vagina and ovary (Stone, 1994). The ability of oestrogens to stimulate cell proliferation is well established and the principal physiological oestrogen, 17␤-oestradiol (E2) is considered to be a major risk factor in the development of breast cancer, and has also been implicated in the aetiology of colorectal cancer and prostate cancer (Stone, 1994; English et al., 2001; Noble, 1997). The tissue specific carcinogenicity of both PhIP and oestrogen, the ability of PhIP to mimic selective oestrogen activities and the powerful genotoxicity of PhIP are compelling factors when considering its potential to cause cancer. In eliciting its effects, PhIP may be influencing the fundamental processes of cell signalling, especially those transduction pathways involved in proliferation. 2. Methods 2.1. Cell lines The following cell lines were employed, MCF-7, T47D, MCF10A (all human breast epithelial adenocarcinoma). The MCF-7 and T47D cells are oestrogen receptor positive (ER+ ) and the MCF10A cells are ER− . In some experiments the human lymphoblastoid cell MCL-5 (engineered to express human cytochromes P450) was used to activate PhIP to a genotoxic metabolite in co-culture experiments. 2.2. Co-culture with MCL-5 cells PhIP requires metabolic activation, via CYP1 family enzymes, to its N-hydroxy derivative in order to exert its genotoxic effects. Although MCF-7 and MCF10A cells are reported to express CYP1A1 and CYP1B1, the levels are very low (Sun et al., 2004). Therefore, in order to be able to detect the consequences of PhIP activation, we devised a co-culture system in which these human adherent breast epithelial cells were co-cultured with the human suspension lymphoblastoid cell line, MCL-5. MCL-5 cells constitutively express high levels of CYP1A1, and have also been transfected with a

number of other human CYP enzymes, including CYP1A2, and as such, are well able to metabolise PhIP to genotoxic species. We have found that irradiation of MCL-5 cells leaves them effectively sterile and unable to replicate, but they retain their metabolic competency (Zhu et al., 2000). The irradiated MCL-5 suspension cells are easily separated by aspiration from the adherent target MCF10A cells following treatment. 2.3. Cell viability assays using trypan blue exclusion Cells (3 × 106 ) were seeded into T75 flasks and allowed to adhere overnight. The growth medium was then removed and replaced with fresh medium containing 6 × 106 MCL-5 cells that had previously been irradiated. Dilutions of PhIP were made up in dimethyl sulphoxide (DMSO) and added to media at a final DMSO concentration of 0.2%. Cells were treated with varying concentrations of PhIP (5–100 ␮M) or DMSO solvent control for 24 h. Following treatment, medium and MCL-5 cells were removed from the flask by aspiration. The target breast epithelial cells were washed with PBS, and then trypsin–EDTA was added in order to detach the cells from the flask. Cells were collected by centrifugation at 1000 rpm for 5 min, and then counted with a haemocytometer using trypan blue exclusion as a marker for cell viability. 2.4. Cell viability using the resazurin reduction assay Cell viability was also measured using the CellTiterBlueTM cell assay. The assay is based on the reduction of resazurin (7-hydroxy-3H-phenoxazin-3-one-10-oxide), a blue non-fluorescent compound, to resorufin which is pink and highly fluorescent. This method is rapid and the compound is non-toxic to cells, allowing repeated measurements with the same cell population. Resorufin fluorescence was measured with a fluorimeter (BMG PolarStar) 560 nm excitation and 590 nm emission. All experiments were performed in triplicate wells and repeated on three separate occasions. 2.5. Cell proliferation assays for ER+ cells MCF-7 cells were routinely maintained in MEM supplemented with 1% non-essential amino acids, 10% FCS, 2 mM l-glutamine and 100 IU/ml penicillin/100 ␮g/ml streptomycin in an incubator maintained at 5% CO2 and 37 ◦ C. For proliferation measurements, the medium was changed to MEM without serum for 24 h to synchronise cells to the resting phase of the cell cycle. The cells were trypsinised then harvested by centrifugation at 1000 rpm for 5 min, and cell number determined by counting. Cells were seeded into 24-well plates at 1 × 105 cells/well in phenol red-free MEM supplemented with 5% dextran-coated charcoal-stripped serum (DCCSS). After 24 h, the medium was changed and the cells were treated with different concentrations of PhIP or estradiol (E2, dissolved in absolute ethanol) in the presence or absence of the ER antagonist ICI 182,780. Dilutions of test compounds

N.J. Gooderham et al. / Toxicology Letters 168 (2007) 269–277

were made up in absolute ethanol and administered so that the final ethanol concentration was 1% in the media. Control cells were incubated with a media containing an equivalent amount of solvent without the test compound. Each treatment was carried out in quadruplicate. After 2 days, the ligands were washed off and replaced with fresh experimental media containing test compounds. The cells were allowed to grow for a further 4 days before the relative cell number was estimated. 2.6. Flow cytometry analysis MCF10A cells were seeded into T25 flasks (106 cells/flask) and allowed to adhere overnight. The following day, medium was removed and replaced with fresh medium containing 2 × 106 irradiated MCL-5 cells. Cells were then treated with PhIP (5–100 ␮M) or DMSO solvent control. Medium, chemicals and MCL-5 cells were removed (aspiration) following treatment and the remaining adherent cells were washed with PBS and then cultured for a further 18 h in fresh medium containing nocodazole (0.4 ␮g/ml). After this period cells were harvested by trypsinisation, washed with PBS and then fixed in ice-cold 70% ethanol at −20 ◦ C for at least 24 h. Following fixation, cells were washed with PBS then incubated in PBS containing 1mg/ml RNase A at 37 ◦ C for 15 min. Propidium iodide (10 ␮g/ml) was then added to the cell suspension and incubated for a further 20 min at 37 ◦ C in order to stain the DNA. A Becton Dickinson FACScan flow cytometer and Cellquest software were used to determine cell cycle distribution, measuring 10,000 cells per treatment. 2.7. Western blotting Cells were treated in T75 flasks as described above, and then harvested for immunoblot analysis of p53, p21WAF1/CIP1 , CYP1A1, CYP1A2, ERK1/2, ELK-1 and ␤-actin protein expression. Briefly, cells were washed twice with ice-cold PBS and then lysed on ice in buffer consisting of 50 mM Tris–HCl pH8.5, 150 mM NaCl, 1% Nonidet-P-40, 5 mM EDTA, supplemented with 10 mM sodium pyrophosphate, 5 mM sodium orthovanadate, 50 ␮g/ml phenylmethylsulfonyl fluoride, 20 ␮g/ml aprotinin and 10 ␮g/ml leupeptin. Lysates were centrifuged at 14,000 × g at 4 ◦ C for 10 min, supernatants were collected and protein content was determined. Proteins were separated by SDS-polyacrylamide gel electrophoresis (7% or 12.5% gels) and electroblotted onto a nitrocellulose membrane. Membranes were blocked for 30 min at room temperature in blocking buffer (PBS containing 0.01% Tween and 5% non-fat dried milk) and then incubated with primary antibody in fresh blocking buffer overnight at 4 ◦ C. Membranes were then washed three times in PBS-T (PBS containing 0.01% Tween) and then incubated in blocking buffer containing secondary antibody for 1 h at room temperature. Membranes were washed as above and proteins were visualised using the enhanced chemiluminescence detection system (Perbio Science, UK). Equal protein loading was confirmed using an antibody to ␤-actin.

271

3. Results 3.1. Effect of treating MCF10A cells with high dose PhIP (10−6 to 10−4 M) on viability and colony forming ability Trypan blue exclusion was used to determine viable MCF10A cells following PhIP treatment. As shown in Fig. 1, the addition of PhIP under co-culture conditions resulted in a marked inhibition of cell growth, which was dose-dependent. At the highest dose (100 ␮M), cell number was reduced by almost 50%, compared to vehicle control. There was very little change in cell viability (dye exclusion) suggesting that the observed drop in cell number was not a consequence of cell death at this early time point. In order to assess the longer-term survival of treated cells, a colony survival assay (clonogenicity) was performed. Treatment with PhIP resulted in impaired longer-term survival, which was more pronounced than the short term growth inhibition induced by PhIP, implying that the compound caused long term damage to the cells. At the highest dose (100 ␮M), survival was reduced to 23.8 ± 5.3% of the control population. In the absence of co-culture with MCL-5 cells, PhIP (100 ␮M) failed to decrease either MCF10A cell number or clonogenicity. 3.2. Induction of cell cycle arrest by PhIP In order to determine whether PhIP had any effect on progression through the cell cycle, MCF10A cells were treated under co-culture conditions for 24 h and then grown for a further 18 h in the presence of nocodazole

Fig. 1. Effect of PhIP on MCF10A cell viability. Cells were treated with PhIP for 24 h in co-culture with metabolically active MCL-5 cells. The MCL-5 cells were removed and the MCF10A cells were counted using trypan blue dye assay. Significantly different (ANOVA) from control, * p < 0.05.

272

N.J. Gooderham et al. / Toxicology Letters 168 (2007) 269–277

Fig. 2. PhIP metabolically activated by MCL-5 cells induces G1 checkpoint in MCF10A cells. Cells were treated with PhIP or DMSO for 24 h, then cultured for a further 18 h in the presence of nocodazole (to prevent mitosis).

(to prevent mitosis). As shown in Fig. 2, exposure to PhIP for 24 h resulted in a dose-dependent increase in the G1 cell population. In the absence of co-culture with MCL-5 cells, PhIP failed to affect cell cycle profile. 3.3. PhIP induces the expression of DNA damage response proteins Enhanced expression of the transcription factor p53 and the downstream cyclin dependent kinase (cdk) inhibitor p21WAF1/CIP1 are often observed following DNA damage, and are known to induce G1 cell cycle arrest. We found that treatment of MCF10A cells with PhIP under co-culture conditions for 24 h, produced a good induction of p21 and to a lesser extent p53 (Fig. 3) that was dose-dependent. In the absence of co-culture with MCL-5 cells, PhIP failed to induce expression of these DNA damage response proteins. 3.4. Effect of treatment of cells with low-dose PhIP (10−10 to 10−6 M)

flow cytometry changes, DNA damage response protein expression such as p53 or p21) in co-cultured target cells be detected (data not shown). However, it is known that both MCF-7 and MCF10A cells, like primary epithelial breast cells, do express very low levels of CYP1A1 and 1B1 (Sun et al., 2004), both of which can metabolise PhIP. Therefore, to mimic the in vivo situation where breast epithelial cells are exposed to circulating PhIP, experiments using low doses of PhIP were performed with target cells (MCF-7 or MCF10A) in the absence of co-culture with MCL-5 cells. 3.5. PhIP induces cell proliferation Like oestradiol, low doses of PhIP (nM) induced proliferation of the oestrogen-dependent cell line MCF7 breast adenocarcinoma cells (Fig. 4). Interestingly, the poor dose–response effect noted is consistent with oestrogen-mediated proliferation where a bell-shaped dose–response curve has often been observed. MCF-7

At low doses (10−10 to 10−6 M), metabolism of PhIP by MCL-5 cells could not be detected nor could any evidence of PhIP-induced toxicity (cell viability, mutation,

Fig. 3. PhIP-induced expression of p53 and p21cip1/waf1 in MCF10A cells co-cultured with MCL-5 cells.

Fig. 4. Effect of E2 and PhIP on MCF-7 cell (ER+ ) proliferation. Viable cells were determined 6 days after addition of ligand. Each point represents the mean ± S.D. of four separate experiments, each performed in quadruplicate (** p < 0.01, significantly different compared to ethanol control). ICI: ICI182 780 (ER␣ antagonist).

N.J. Gooderham et al. / Toxicology Letters 168 (2007) 269–277

273

duction pathway. It is known that E2 can stimulate the mitogen-activated protein kinase signaling pathway in MCF-7 cells (Migliaccio et al., 1996) and we have previously shown that PhIP can mimic the activity of E2 (Lauber et al., 2004). Whether PhIP was able to stimulate proliferation through the MAPK pathway independent of the ER was investigated in MCF10A cells (ER− ) and with the potent MAPK pathway inhibitor PD98059. As may be seen in Fig. 5, PhIP-mediated cell proliferation was prevented by PD98059, supporting the proposal that the proliferative response in the ER− MCF10A cell was via MAPK. 3.7. MAP/ERK kinase (MEK1/2 and ERK1/2 phosphorylation) Fig. 5. Effect of PD98059 (MEK1/2 inhibitor) on PhIP-mediated MCF10A (ER− ) proliferation. Values are mean ± S.D. of three separate experiments, each performed in triplicate (** p < 0.01, * p < 0.05, significantly different compared to ethanol control, # p < 0.01 significantly different compared to treatment with PhIP or ethanol alone, unpaired t-test). PD: PD98059 (MAPK pathway inhibitor).

cells express a functional oestrogen receptor and using a specific inhibitor of this receptor (ICI182 780) we found that the PhIP-induced proliferation of the cell was inhibited (Fig. 4). To examine the suggestion that the ER was essential for this proliferative effect, we also assessed the ability of nM concentrations of PhIP to induce proliferation of MCF10A breast adenocarcinoma cells (ER− ). Fig. 5 shows that PhIP appeared to induce proliferation of the MCF10A cells, although the response was not as prominent as in the MCF-7 cells. Significantly, MCF10A cells do not express ER, therefore the proliferative response in this cell was ER-independent. 3.6. The involvement of the MAPK pathway in PhIP-induced cell proliferation Cell proliferation is dependent upon activation of signal transduction pathways and in particular the mitogen-activated protein kinase (MAPK) signal trans-

Having observed increased proliferation of MCF10A cells following PhIP exposure, the effect of PhIP on the MAP kinase (MEK1/2–ERK1/2) pathway was examined in more detail. This pathway plays a major role in the regulation of cell proliferation and differentiation (Crews et al., 1992; Alessi et al., 1994) and involves a protein kinase cascade following growth factor stimulation that leads to the successive phosphorylation and thus activation of RAF-1, MEK1/2 and then ERK1 and ERK2, which target a number of transcription factors involved in regulating mitogenesis (such as ELK-1). We found evidence that these remarkably low concentrations of PhIP (10−9 to 10−7 M) appeared to induce the activation (phosphorylation) of ERK1/2 in the MCF10A cells within the first hour of exposure (Fig. 6). These preliminary results require further investigation and full evaluation of the dose–response and temporal aspects of this activation are warranted. 3.8. ELK-1 kinase assay In order to further investigate the finding that PhIP increases the phosphorylation of ERK1 and ERK2, a kinase assay was carried out in order to verify the activation of this MAP kinase pathway. The assay involves

Fig. 6. Immunoblot showing the effect of 10−7 and 10−9 M PhIP on the phosphorylation status of ERK1 and ERK2 in MCF10A cells. Phosphorylated and total forms of ERK1/2 were evaluated by Western blotting with protein extracts obtained from MCF10A cells treated with PhIP in serum-free medium for the time periods shown. 1, EtOH; 2, 10−7 M PhIP; 3, 10−9 M PhIP.

274

N.J. Gooderham et al. / Toxicology Letters 168 (2007) 269–277

Fig. 7. Effect of 10−7 M PhIP on the kinase activity of ERK1 and ERK2 in MCF10A cells. ERK1/2 kinase activity was determined by the detection of ELK-1 phosphorylation (p-ELK-1) by active ERK1 and ERK2 immunoprecipitated from MCF10A cells treated with PhIP for 30 min in serum-free medium. Bands represent separate kinase assays performed with different cell lysates.

evaluation of the amount of phosphate incorporated into an ELK-1 fusion protein that serves as a substrate for immunoprecipitated active ERK1/2. This assay confirmed that ERK1 and ERK2 are activated by 30 min PhIP (10−7 M) treatment (Fig. 7). 4. Discussion The genetic toxicology of PhIP is dependent upon CYP-mediated metabolic activation. Human liver is particularly efficient at this activation. We have previously described the high toxicity associated with incubating human cells with liver S9 and microsomal preparations and have devised a co-culture system using irradiated metabolically competent cells to overcome these problems of toxicity (Zhu et al., 2000). We therefore used this approach to examine PhIP toxicity in target cells. Our data show that exposure to metabolically activated PhIP results in induction of the DNA damage response protein p53, and its downstream target p21WAF1/CIP1 . It is well established that p53 and p21WAF1/CIP1 mediate hypophosphorylation of pRB and subsequent G1 arrest following DNA damage (Kastan et al., 1991). This is consistent with the induction of G1 checkpoint response observed here with MCF10A cells co-incubated with PhIP and MCL-5 cells. While protective DNA damage responses are perhaps to be expected given the well characterised genotoxicity of metabolically activated PhIP, these initial studies were carried out at comparatively high doses of PhIP (␮M). This is consistent with the majority of published experimental studies where PhIP has been used at doses substantially higher than those found in the human diet. Few studies have examined the effects of PhIP at concentrations of the compound systemically available after a cooked meat meal. Layton et al. (1995) has estimated that total HCA intake in humans ranges from around <1 to 17 ng/kg body weight (b.w.) per day. At these dietary realistic doses, the mutagenic potential of PhIP is undetectable using conventional assays. Such differ-

ences make extrapolation of the toxicological effects observed in high dose animal experiments to the effects of the low doses found in the human diet, very difficult. Nevertheless, PhIP clearly does have the potential to exert genotoxic effects in humans, as evidenced by the detection of PhIP–DNA adducts in mammary gland samples (Zhu et al., 2003). The genotoxicity of PhIP has been assumed to be the major mode of action by which the compound induces carcinogenicity, however mechanisms other than adduction and damage of DNA may also be important in explaining the tumourigenic properties of this amine. While the formation of DNA adducts is believed to be of critical importance for induction of cancer by genotoxic carcinogens, there is not a simple relationship of cause and effect between the presence of adducts and cancer development. Feeding studies have shown that although PhIP–DNA adducts do arise in the target organs of the compound (Shirai et al., 1997; Snyderwine et al., 1998), adducts are also distributed in other non-target tissues such as the pancreas and lungs (Takayama et al., 1989), without induction of tumours in these tissues. It is therefore likely that PhIP possesses additional mechanisms of action that also influence its potential for cancer formation. The pharmacokinetics and elimination profile of PhIP in humans suggests that absorption, and therefore internal exposure to PhIP, would be maximal shortly after a meal, before the concentrations are reduced by metabolic, renal and faecal clearance mechanisms (Lynch et al., 1992). However, unchanged PhIP is excreted in the urine for at least 12 h following consumption of a cooked meat meal indicating that first pass clearance is not efficient and circulating levels of unchanged PhIP and its metabolites continues for some time (Lynch et al., 1992). Taken together, this suggests that cells are exposed to low levels of PhIP for prolonged periods, and that this exposure would be reinforced by daily cooked meat intake. Given the rapid and extensive bioavailability of PhIP, circulating plasma levels are likely to be in the ∼nanomolar range following cooked meat consumption. Supporting this estimate, PhIP has been detected in human breast milk sample at concentrations averaging at ∼10−10 M (DeBruin et al., 2001), and it is possible that local levels may be higher. This evidence further indicates that ductal mammary epithelial cells are directly exposed to this carcinogen. It is known that both MCF-7 and MCF10A cells, like primary epithelial breast cells, express very low levels of CYP1A1 and 1B1 (Sun et al., 2004), both of which can metabolise PhIP. Therefore, to mimic the in vivo situation where breast epithelial cells are exposed to circulating PhIP,

N.J. Gooderham et al. / Toxicology Letters 168 (2007) 269–277

experiments using low doses of PhIP were performed with target cells (MCF-7 or MCF10A) in the absence of co-culture with MCL-5 cells. Work in our laboratory has shown that nanomolar concentrations of PhIP are able to stimulate cell proliferation and invasion in ER+ human breast cancer cells. These effects appear to be a result of oestrogenic properties possessed by PhIP, as administration of anti-oestrogen inhibited these effects, and PhIP was also shown to induce transcription of an oestrogen receptor ␣ reporter gene (Gooderham et al., 2002; Lauber et al., 2004). Given that oestrogens are regarded as a major risk factor in the development of breast cancer, and have also been implicated in the aetiology of colorectal cancer (Di Leo et al., 2001) and prostate cancer (Bonkhoff et al., 2001), the ability of PhIP to mimic selective oestrogen activity at such low doses suggests that the oestrogenicity of PhIP may be an important contributing factor to the site specific carcinogenicity of the compound. Such effects could be expected to exert a promotional influence on cells with an initiated phenotype. Our current and previous findings (Lauber et al., 2004) demonstrate that exposure to low doses of PhIP results in markedly different cellular responses than the DNA damage and cell cycle arrest observed at high dose. The ERK signal transduction cascade plays a central role in the stimulation of cell proliferation. The ERK isoforms ERK1 and ERK2 are ubiquitously expressed and represent a point of convergence of mitogenic signals from a diverse range of receptors (Su and Karin, 1996). Once activated, ERKs can also regulate the actin cytoskeleton, cell migration and proteinase induction (Reddy et al., 2003). These proliferative and migratory effects are therefore believed to play a role in tumour promotion and progression, and indeed, ERK signalling is stimulated by a number of tumour promoters including 12-O-tetradecanoylphorbol-13-acetate (TPA) (Suzukawa et al., 2002). Activated ERKs have been shown to be elevated in human breast (Maemura et al., 1999) and prostate tumours (Gioeli et al., 1999), and have also been reported to be overexpressed in tissue samples from human colorectal tumours (Hoshino et al., 1999). Given the importance of ERK1 and ERK2 signalling in carcinogenesis, the ability of PhIP (at concentrations likely to occur in humans consuming cooked meat) to stimulate this MAP kinase cascade in an ER-negative human breast cell line was examined. In preliminary experiments we found that nanomolar concentrations of PhIP do indeed appear to activate this pathway, and influence mechanisms favouring cell proliferation rather than inhibition.

275

With MCF10A cells, an ER-negative human breast epithelial cell line with the phenotypic characteristics of normal cells, PhIP only marginally increased proliferation under serum-free conditions. This is consistent with the report of Spink et al. (2006) who describe the resistance of MCF10A cells to oestradiol-mediated proliferation. However, our data shows that PhIP does influence the ERK signalling pathway, with a rapid but transient activation of ERK1 and ERK2, and the regulators of these kinases, MAP/ERK kinase 1/2, as assessed by examination of the phosphorylation status of these proteins. The proliferative effects of PhIP were inhibited by inhibitors of this signalling cascade, demonstrating the critical role of the pathway in the mitogenicity of low-dose PhIP. These preliminary experiments must be confirmed with full dose–response and time course experiments, but they are consistent with the report by Wang and Miller (2000) who described the ability of oestradiol to stimulate signal transduction pathways in MCF10A cells. Stimulation of mitogenesis is commonly associated with tumour promoting agents, suggesting that in addition to its role as an initiating agent, PhIP may also play a role in the promotion of neoplasia. Work confirming the effects of PhIP on proliferation and ERK signalling in other cell lines and in vivo is essential. It seems highly likely that if PhIP were to stimulate ERK and proliferation in cells of its target tissues, such promotional effects could well play a significant role in PhIP’s tissue specific carcinogenicity. It is important to emphasise that the data presented in the present study have been obtained in vitro and although very low doses have been used, extrapolation of these findings to those that pertain in vivo after consumption of cooked meat is exceedingly difficult. Given that PhIP is bioavailable in humans, is metabolically activated and has been shown to form DNA adducts in the mammary gland and colon, there is considerable evidence to support the concept that PhIP has initiating potential in humans. The results reported here that low-dose PhIP is able to stimulate mammary cell proliferation and in combination with previous reports of the oestrogenic properties of PhIP at similar concentrations, suggest that the parent amine also has the potential to exert a promotional influence on cells previously initiated either by PhIP or other genotoxic agents. In connection with this, it is acknowledged that the present study has considered PhIP in isolation and the impact of the other complex components of cooked food, including other heterocyclic amines, have not been considered. Clearly, food is likely to contain other chemicals that could act as ER antagonists or agonists and the consequences of exposure to such mixtures could range

276

N.J. Gooderham et al. / Toxicology Letters 168 (2007) 269–277

from inhibition to synergy. One of the requirements of tumour promotion is repeated exposure to the promoting agent. Notwithstanding the complex nature of diet, regular intake of cooked meat would be expected to provide exposure to PhIP on a regular basis, and given that low levels of PhIP are thought to be present in the circulation for at least 12 h following meat consumption, this exposure is expected to be prolonged. It is therefore possible that low-dose dietary PhIP could act as a tumour initiator and as a promoter and exposure to this compound at such concentrations could be potentially hazardous. Acknowledgments This work was funded in part by the UK Food Standards Agency. References Alessi, D.R., Saito, Y., Campbell, D.G., Cohen, P., Sithanandam, G., Rapp, U., Ashworth, A., Marshall, C.J., Cowley, S., 1994. Identification of the sites in MAP kinase kinase-1 phosphorylated by p74raf-1. EMBO J. 1, 1610–1619. Bonkhoff, H., Fixemer, T., Hunsicker, I., Remberger, K., 2001. Progesterone receptor expression in human prostate cancer: correlation with tumor progression. Prostate 48, 285–291. Crews, C.M., Alessandrini, A., Erikson, R.L., 1992. Erks: their fifteen minutes has arrived. Cell Growth Differ. 3, 135–142. DeBruin, L.S., Martos, P.A., Josephy, P.D., 2001. Detection of PhIP (2amino-1-methyl-6-phenylimidazo[4,5-b]pyridine) in the milk of healthy women. Chem. Res. Toxicol. 14, 1523–1528. Di Leo, A., Messa, C., Cavallini, A., Linsalata, M., 2001. Estrogens and colorectal cancer. Curr. Drug Targets Immune Endocr. Metabol. Disord. 1, 1–12. Doll, R., Peto, R., 1981. The causes of cancer: quantitative estimates of avoidable risks of cancer in the United States today. J. Natl. Cancer Inst. 66, 1191–1308. English, M.A., Stewart, P.M., Hewison, M., 2001. Estrogen metabolism and malignancy: analysis of the expression and function of 17␣-hydroxysteroid dehydrogenases in colonic cancer. Mol. Cell. Endocrinol. 171, 56–60. Gioeli, D., Mandell, J.W., Petroni, G.R., Frierson Jr., H.F., Weber, M.J., 1999. Activation of mitogen-activated protein kinase associated with prostate cancer progression. Cancer Res. 59, 279–284. Gooderham, N.J., Zhu, H., Lauber, S., Boyce, A., Creton, S., 2002. Molecular and genetic toxicology of 2-amino-1-methyl6-phenylimidazo[4,5-b]pyridine (PhIP). Mutat. Res. 506–507, 91–99. Hoshino, R., Chatani, Y., Yamori, T., Tsuruo, T., Oka, H., Yoshida, O., Shimada, Y., Ari-i, S., Wada, H., Fujimoto, J., Kohno, M., 1999. Constitutive activation of the 41-/43-kDa mitogen-activated protein kinase signaling pathway in human tumors. Oncogene 18, 813–822. Ito, N., Hasegawa, R., Sano, M., Tamano, S., Esumi, H., Takayama, S., Sugimura, T., 1991. A new colon and mammary carcinogen in cooked food, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP). Carcinogenesis 12, 1503–1506.

Kastan, M.B., Onyekwere, O., Sidransky, D., Vogelstein, B., Craig, R.W., 1991. Participation of p53 protein in the cellular response to DNA damage. Cancer Res. 51, 6304–6311. Lauber, S., Ali, S., Gooderham, N.J., 2004. The cooked food derived carcinogen 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine is a potent oestrogen: a mechanistic basis for its tissue-specific carcinogenicity. Carcinogenesis 25, 2509–2517. Layton, D.W., Bogen, K.T., Knize, M.G., Hatch, F.T., Johnson, V.M., Felton, J.S., 1995. Cancer risk of heterocyclic amines in cooked foods: an analysis and implications for research. Carcinogenesis 16, 39–52. Lynch, A.M., Knize, M.G., Boobis, A.R., Gooderham, N.J., Davies, D.S., Murray, S., 1992. Intra- and interindividual variability in systemic exposure in humans to 2-amino3,8-dimethylimidazo[4,5-f]quinoxaline and 2-amino-1-methyl-6phenylimidazo[4,5-b]pyridine, carcinogens present in cooked beef. Cancer Res. 52, 6216–6223. Maemura, M., Iino, Y., Koibuchi, Y., Yokoe, T., Morishita, Y., 1999. Mitogen-activated protein kinase cascade in breast cancer. Oncology 57, 37–44. Migliaccio, A., Di Domenico, M., Castoria, G., de Falco, A., Bontempo, P., Nola, E., Auricchio, F., 1996. Tyrosine kinase/p21ras/MAPK-kinase pathway activation by estradiolreceptor complex in MCF7 cells. EMBO J. 15, 1292–1300. Noble, R.L., 1997. The development of prostatic adenoma in Nb cells following prolonged sex hormone administration. Cancer Res. 37, 1929–1933. Reddy, K.B., Nabha, S.M., Atanaskova, N., 2003. Role of MAP kinase in tumor progression and invasion. Cancer Metastasis Rev. 22, 395–403. Shirai, T., Sano, M., Tamano, S., Takahashi, S., Hirose, M., Futukuchi, M., Hasegawa, R., Imaida, K., Matsumoto, K., Wakabayashi, K., Sugimura, T., 1997. The prostate: a target for carcinogenicity of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) derived from cooked foods. Cancer Res. 52, 195–198. Snyderwine, E.G., Thorgeirsson, U.P., Venugopal, M., RobertsThomson, S.J., 1998. Mammary gland carcinogenicity of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine in SpragueDawley rats on high- and low-fat diets. Nutr. Cancer 31, 160– 167. Spink, B.C., Cole, R.W., Katz, B.H., Gierthy, J.F., Bradley, L.M., Spink, D.C., 2006. Inhibition of MCF-7 breast cancer cell proliferation by MCF-10A breast epithelial cells in coculture. Cell Biol. Int. 30, 227–238. Stone, R., 1994. Environmental estrogens stir debate. Science 265, 308–310. Su, B., Karin, M., 1996. Mitogen-activated protein kinase cascades and regulation of gene expression. Curr. Opin. Immunol. 8, 402–411. Sugimura, T., 1997. Overview of carcinogenic heterocyclic amines. Mutat. Res. 376, 211–219. Sun, Y.-W., Guengerich, F.P., Sharma, A.K., Boyiri, T., Amin, S., El-Bayoumy, K., 2004. Humans cytochromes CYP1A1 and 1B1 catalyse ring oxidation but not nitroreduction of environmental pollutant mononitropyrene isomers in primary cultures of human breast cells and cultured MCF-10A and MCF-7 cell lines. Chem. Res. Toxicol. 17, 1077–1085. Suzukawa, K., Weber, T.J., Colburn, N.H., 2002. AP-1, NF-kappa-B, and ERK activation thresholds for promotion of neoplastic transformation in the mouse epidermal JB6 model. Environ. Health Perspect. 110, 865–870. Takayama, K., Yamashita, K., Wakabayashi, K., Sugimura, T., Nagao, M., 1989. DNA modification by 2-amino-1-methyl-6-

N.J. Gooderham et al. / Toxicology Letters 168 (2007) 269–277 phenylimidazo[4,5-b]pyridine in rats. Jpn. J. Cancer Res. 80, 1145–1148. Wang, B., Miller, F.R., 2000. Tyrosine phosphorylation of an erbB-2 related p90 protein induced by estrogen in human breast epithelial cells. Int. J. Oncol. 16, 1035–1042. WCRF (World Cancer Research Fund/American Institute for Cancer Research), 1997. Food Nutrition and the Prevention of Cancer: A Global Perspective. American Institute for Cancer Research, Washington, DC, pp. 1–670.

277

Zhu, J., Chang, P., Bondy, M.L., Sahin, A.A., Singletary, S.E., Takahashi, S., Shirai, T., Li, D., 2003. Detection of 2-amino-1methyl-6-phenylimidazo[4,5-b]-pyridine-DNA adducts in normal breast tissues and risk of breast cancer. Cancer Epidemiol. Biomarkers Prev. 12, 830–837. Zhu, H., Boobis, A.R., Gooderham, N.J., 2000. The food carcinogen 2-amino-1-methyl-6-phenylimidazo [4,5-b]pyridine activates S-phase checkpoint and apoptosis and induces gene mutation on human lymphoblastoid TK6 cells. Cancer Res. 60, 1283–1289.