Epigenetic processes and cancer risk assessment

The U.S. Environmental Protection Agency's Guidelines for Carcinogen Risk Assessment encourages the use of mechanistic data in the assessment of human cancer risk at low (environmental) exposure levels. The key events that define a particular mode of action for tumor formation have been concentrated to date more on mutational responses that are broadly the result of induced DNA damage and enhanced cell proliferation. While it is clear that these processes are important in terms of tumor induction, other modes that fall under the umbrella of epigenetic responses are increasingly being considered to play an important role in susceptibility to tumor induction by environmental chemicals and as significant modifiers of tumor responses. Alterations in gene expression, DNA repair, cell cycle control, genome stability and genome reprogramming could be the result of modification of DNA methylation and chromatin remodeling patterns as a consequence of exposure to environmental chemicals. These concepts are described and discussed.     

Induction of malignant transformation by various chemicals in Balb/3T3 clone A31-1-1 cells and biological characterization of some transformants

Balb/c A31-1-1 cells were used for the study of transformation induction by chemicals with different mutagenic specificities. We show that survival of these cells and therefore the calculated transformation frequency per cells at risk is dependent upon the cell density at the time of treatment. It is suggested that equal cell densities should be used for measuring survival values and transformation induction. The quantitative results obtained are discussed in the light of the known mutagenic mechanisms of the chemicals tested. We also characterized morphologically transformed foci induced by different chemicals with respect to some biological properties. Anchorage independence was determined by testing growth in soft agar, loss of contact inhibition was quantitated by measuring maximum cell densities and malignancy was tested by tumor induction in nude mice. Although no very close correlation between these parameters and morphology was observed, the most malignant clones are also the ones with the highest values in the other tests. Our data make one or few genetical targets for transformation induction likely. We therefore speculate that the diverse phenotypes obtained might be due to differential activation of one or very few transforming genes in these cells.     

Reproductive capacity and dominant lethal mutations in female guinea-pigs and Djungarian hamsters following X-rays or chemical mutagens.

The reproductive capacity and induction of dominant lethal mutations in adult female guinea-pigs and Djungarian hamsters were tested following treatment with 400 rad X-rays, 1.6 mg/kg triethylenemelamine (TEM) or 75 mg/kg isopropylmethanesulphonate (IPMS). A fairly high level of dominant lethals were observed in female guinea-pigs mated at the first oestrus after irradiation (23.4 +/- 6.4%) with a lower yield at 3 months (9.6 +/- 8.2%). Neither of the chemicals caused any significant induction of dominant lethals at either mating time. In the reproductive capacity experiments, the mean litter size of irradiated female guinea-pigs was reduced for about 12 months and this was especially marked in the first 6 months following treatment. Neither of the chemicals caused any significant differences in early litter sizes but there was a noticeable reduction in the litter sizes of TEM-treated females in the 18--24 month interval. With Djungarian hamsters a marked effect of X-rays on reproductive capacity was apparent. After 400 rad a smaller proportion of irradiated females littered in the first 25-day interval than after the other treatments, and no irradiated females produced more than one litter. Neither of the chemicals caused such a drastic reduction in fertility but TEM-treated females produced fewer litters and became sterile at an earlier age than control or IPMS-treated females. With IPMS, the number of litters produced was similar to the controls. Both chemicals caused a significant reduction in litter-size but further work is needed to establish whether this was due to induction of dominant lethals. No translocations were observed in the sons of treated female guinea-pigs or hamsters, but the numbers of animals studied were too small for any conclusions to be drawn.     

Relevance of chemical structure and cytotoxicity to the induction of chromosome aberrations based on the testing results of 98 high production volume industrial chemicals

Over a 6-year period (1991-1996), the chromosomal aberration testing of high production volume (HPV) industrial chemicals had been conducted using Chinese hamster lung (CHL/IU) cells according to OECD HPV testing program and the national program in Japan. A total of 98 chemicals were tested for the induction of chromosome aberration (CA), consisting of structural CA and polyploidy. Of the 98 chemicals, structural CA and/or polyploidy were induced by 39 chemicals (40%). Anilines and phenols tended to induce only structural CA. p-tert-Butylphenol had a peculiar feature in inducing not only structural CA but also polyploidy at considerably high frequency (93.2%) after continuous treatment for 48 h, posing an aneugenic potential. Not all, but six of 11 carboxylic acids or esters also showed the simultaneous induction of structural CA and polyploidy. The majority of organic phosphates, alcohols or ethers, alkyl benzenes and non-cyclic alkanes had no CA induction activity. For chemicals which were negative in the bacterial reverse mutation assay (Ames test), the proportion of the chemicals that induced CA at a severely cytotoxic dose (doses manifesting more than 50% cytotoxicity) was similar to that of the CA-negative chemicals manifesting severe cytotoxicity, suggesting that severely cytotoxic chemicals do not always induce CA.     

A novel in vitro method for the detection and characterization of photosensitizers

Photoactivation and binding of photoactive chemicals to proteins is a known prerequisite for the formation of immunogenic photoantigens and the induction of photoallergy. The intensive use of products and the availability of new chemicals, along with an increasing exposure to sun light contribute to the risk of photosensitizing adverse reactions. Dendritic cells (DC) play a pivotal role in the induction of allergic contact dermatitis. Human peripheral blood monocyte derived dendritic cells (PBMDC) were thus perceived as an obvious choice for the development of a novel in vitro photosensitization assay using the modulation of cell surface protein expression in response to photosensitizing agents. In this new protocol, known chemicals with photosensitizing, allergenic or non-allergenic potential were pre-incubated with PBMDCs prior to UVA irradiation (1 J/cm(2)). Following a 48 h incubation, the expression of the cell surface molecules CD86, HLA-DR and CD83 was measured by flow cytometry. All tested photosensitizers induced a significant and dose-dependent increase of CD86 expression after irradiation compared to non-irradiated controls. Moreover, the phototoxicity of the chemicals could also be determined. In contrast, (i) CD86 expression was not affected by the chosen irradiation conditions, (ii) increased CD86 expression induced by allergens was independent of irradiation and (iii) no PBMDC activation was observed with the non-allergenic control. The assay proposed here for the evaluation of the photoallergenic potential of chemicals includes the assessment of their allergenic, phototoxic and toxic potential in a single and robust test system and is filling a gap in the in vitro photoallergenicity test battery.     

Induction and prevention of carcinogenesis in rat skin exposed to space radiation

Quantitative cancer incidence data exist for various laboratory animal models, but little of this information is usable for estimating human risks, primarily because of uncertainties about possible mechanistic differences among species. Acceptance and utilization of animal data for human risk assessment will require a much better understanding of the comparative underlying mechanisms than now exists. A dual-lesion, radiation-track model in rat skin has proven to be consistent with tumor induction data with respect to acute radiation doses ranging from 0.5 up to 10 Gy and higher, and average LETs ranging from 0.34 to 150 keV μm⁻¹ according to the form neoplastic risk (D,L) = CLD + BD². A recent result with the ⁵⁶Fe ion beam showed dose-response consistency for malignant (carcinomas) and benign (fibromas) tumor induction with earlier results utilizing argon and neon ion beams. A discrepancy between the model and experiment was found indicating that proportionality of cancer yield with LET did not occur at 150 versus 125 keV μm⁻¹, i.e. tumor yield did not increase in spite of a 20% increase of LET, which suggests that a LET response maximum exists at or within this dose range. Concordance between the model and tumor induction data in rat skin implies that potential intervening complexities of carcinogenic progression fail to obscure the basic radiobiological assumptions underpinning the model. Gene expression microarray analysis shows that vitamin A inhibits the expression of about 80% of the inflammation-related genes induced by the radiation and prevents about 46% of the neoplasms associated with ⁵⁶Fe ion radiation without appearing to interfere with the underlying dose and LET response patterns. Further validation is needed, but the model has the potential to provide quantitative estimates of cancer risk as a function of dose and LET for almost any type of radiation exposure and even for combinations of different radiations provided only three empirical parameters can be established for each type of radiation and organ system.     

In vitro assay of cytogenetic damage induced in bone marrow cells in vivo by chemical carcinogens

Bone marrow cells explanted after the host animal has been exposed to chemicals can be assayed in vitro for cytogenetic damage. Cells were grown in medium for two cell cycles in the presence of 5-bromodeoxyuridine and then analyzed for the frequency of sister chromatid exchanges and cell replication kinetics. The patterns for induction of these endpoints as assayed in vitro were very similar to those observed if the assay was performed totally in vivo. Several chemicals were tested in this system and shown to be positive; hycanthone, ethidium bromide, 4-fluoro-3-nitro-phenyl azide, mitomycin-C, 2-aminoanthracene, benzo(a)pyrene, and cyclophosphamide. Dehydroemetine, known as a nonmutagenic drug, was negative in this assay. Since this assay incorporates host metabolism, is rapid, and has a wide dynamic range, it may be useful to perform to determine the potential of chemicals to induce genetic damage.     

Investigations into the biological relevance of in vitro clastogenic and aneugenic activity

In the current study we present a view of events leading to chemically induced DNA damage in vitro from both a cytogenetic and molecular aspect, focusing on threshold mediated responses and the biological relevance of DNA damaging events that occur at low and high cellular toxicity levels. Current regulatory mechanisms do not take into account chemicals that cause significant DNA damage only at high toxicity. Our results demonstrate a defined threshold for micronucleus induction after insult with the alkylating agent MMS. Other results define a significant change in gene expression following treatment with chemicals that give rise to structural DNA damage only at high toxicity. Pairs of chemicals with a similar mode of action but differing toxicity levels were chosen, the chemicals that demonstrated structural DNA damage only at high levels of toxicity showed an increase in heat shock protein gene expression whereas the chemicals causing DNA damage events at all levels of toxicity did not induce changes in heat shock gene expression at identical toxicity levels. The data presented indicates that there are a number of situations where the linear dose response model is not appropriate for risk estimation. However, deviation from linear risk models should be dependent upon the availability of appropriate experimental data such as that shown here.     

An assessment of the presence and health risks of endocrine-disrupting chemicals in the drinking water treatment plant of Wu Chang,China

An assessment of the presence and health risks of endocrine-disrupting chemicals in the drinking water treatment plant (DWTP) of Wu Chang was performed. A recombinant yeast assay was used to assess the endocrine disrupting activity of the effluents of the DWTP. Agonistic activities of the estrogen receptor and androgen receptor were not detected in any of the effluent samples. However, anti-estrogenic and anti-androgenic activities were observed. In addition, the removal rates of the DWTP for the anti-androgenic activities were limited. A health risk assessment was performed on the basis of the results of the recombinant yeast assay, and the total daily production of hormones was used to evaluate the health risks of these types of endocrine-disrupting chemicals. The predicted effects of the anti-estrogenic and anti-androgenic disrupting activities were below 1.5%. This study suggested that the combined toxicity bioassays with health risk assessment could provide an available method to assess endocrine-disrupting chemicals and to evaluate the potential adverse effects on human health for aquatic environmental samples.     

Reduction of misleading ("false") positive results in mammalian cell genotoxicity assays. I. Choice of cell type

Current in vitro mammalian cell genotoxicity assays show a high rate of positive results, many of which are misleading when compared with in vivo genotoxicity or rodent carcinogenicity data. P53-deficiency in many of the rodent cell lines may be a key factor in this poor predictivity. As part of an European Cosmetics Industry Association initiative for improvement of in vitro mammalian cell assays, we have compared several rodent cell lines (V79, CHL, CHO) with p53-competent human peripheral blood lymphocytes (HuLy), TK6 human lymphoblastoid cells, and the human liver cell line, HepG2. We have compared in vitro micronucleus (MN) induction following treatment with 19 compounds that were accepted as producing misleading or "false" positive results in in vitro mammalian cell assays [6]. Of these, six chemicals (2-ethyl-1,3-hexandiol, benzyl alcohol, urea, sodium saccharin, sulfisoxazole and isobutyraldehyde) were not toxic and did not induce any MN at concentrations up to 10mM. d,l-Menthol and ethionamide induced cytotoxicity, but did not induce MN. o-Anthranilic acid was not toxic and did not induce MN in V79, CHL, CHO, HuLy and HepG2 cells up to 10mM. Toxicity was induced in TK6 cells, although there were no increases in MN frequency up to and above the 55% toxicity level. The other 10 chemicals (1,3-dihydroxybenzene, curcumin, propyl gallate, p-nitrophenol, ethyl acrylate, eugenol, tert-butylhydroquinone, 2,4-dichlorophenol, sodium xylene sulfonate and phthalic anhydride) produced cytotoxicity in at least one cell type, and were evaluated further for MN induction in most or all of the cell types listed above. All these chemicals induced MN at concentrations <10mM, with levels of cytotoxicity below 60% (measured as the replication index) in at least one cell type. The rodent cell lines (V79, CHO and CHL) were consistently more susceptible to cytotoxicity and MN induction than p53-competent cells, and are therefore more susceptible to giving misleading positive results. These data suggest that a reduction in the frequency of misleading positive results can be achieved by careful selection of the mammalian cell type for genotoxicity testing.     

Interaction of benzo[a]pyrene, 2,3,3',4,4',5 hexachlorobiphenyl (PCB 156) and cadmium on biomarker responses in flounder (Platichthys flesus L.)

Interactive effects of a mixed pollutant exposure on biomarker responses were studied in European flounder (Platichthys flesus L.). The model chemicals, benzo[a]pyrene (BaP, 2.5 mg kg^-1), 2,3,3',4,4'5 hexachlorobiphenyl (PCB-156, 2.5 mg kg^-1), and cadmium (cadmium, 1 mg kg^-1), were administered to fish by subcutaneous injections. Biomarker responses were quantified both following administration of single chemicals and sequential combinations of the chemicals in pairs. Significant induction of CYP1A protein levels and corresponding ethoxyresorufin-O-deethylase (EROD) activities was observed in BaP and PCB treated flounder after 2 and 8 days, respectively. The strongest induction (44 fold) was caused by BaP. No further induction was observed after additional treatment with PCB 156. CYP1A induction caused by BaP was inhibited (40% compared with BaP treatment alone) in flounder pre treated with cadmium, whereas induction by PCB 156 appeared to be unaffected by pre treatment with cadmium. Flounder treated with cadmium only had significantly elevated hepatic levels of metallothionein (MT) after 15 days. Pre treatment with BaP and PCB prior to cadmium inhibited the MT induction (30-50%) compared with cadmium alone. Furthermore, significantly higher glutathione S transferase activities were observed in flounder administered cadmium alone, and in flounder treated with BaP or PCB 156 prior to cadmium. GST selenium independent peroxidase activities appeared to be unaffected by any of the treatments in the present study. The results indicate that chemical mixtures may affect biomarker responses differently from compounds administered alone, and that the sensitivity of both CYP1A and MT are influenced by pollutants other than their primary inducers.     

Mechanistic investigations of low dose exposures to the genotoxic compounds bisphenol-A and rotenone

A complete hazard and risk assessment of any known genotoxin requires the evaluation of the mutagenic, clastogenic and aneugenic potential of the compound. In the case of aneugenic chemicals, mechanism of action (MOA) and quantitative responses may be investigated by studying their effects upon the fidelity of functioning of components of the cell cycle. These present studies have demonstrated that the plastics component bisphenol-A (BPA) and the natural pesticide rotenone induce micronuclei and modify the functioning of the microtubule organising centres (MTOCs) of the mitotic spindles of cultured mammalian cells in a dose-dependent manner. BPA and rotenone were used as model compounds in an investigation of dose response relationships for the hazard/risk assessment of aneugens. Thresholds of action for the induction of aneuploidy have been predicted for spindle poisons on the basis of the multiple targets, which may need disabling before a quantitative response can be detected. The cytokinesis blocked micronucleus assay (CBMA) methodology was utilised in the human lymphoblastoid cell lines AHH-1, MCL-5 and Chinese hamster V79 cell lines. A no observable effect level (NOEL) at 10.8 microg/ml BPA was observed for MN induction. Rotenone showed a small increase in MN induction with the first significant effect at 0.25 ng/ml in V79 cells but there was no significant effect in the metabolically competent cell line, MCL-5. For a mechanistic evaluation of the aneugenic effects of BPA and rotenone, fluorescently labelled antibodies were used to visualise microtubules (alpha-tubulin) and MTOCs (gamma-tubulin). The NOELs for tripolar mitotic spindle induction in V79 cells were 7 microg/ml for BPA and 80 pg/ml for rotenone (concentrations which produced similar changes to mitotic index (M.I.)). Interestingly there was close proximity to the NOEL of 10.8 microg/ml BPA for micronucleus (MN) induction in the human lymphoblastoid AHH-1 cell. Multiple MTOCs can therefore be predicted as a possible mechanism for MN induction. The similarity in concentration inducing tripolar mitosis, M.I. and MN changes suggests immunofluorescence analysis to be a useful dose setting assay with emphasis on the mechanism.     

Induction of functional Fc receptors in P388 leukemia cells : Requirement for multiple differentiation signals

The development of functional Fc receptors (FcR) during induced differentiation with the tumor promoter, phorbol myristate acetate (PMA), was studied in the murine tumor cell line, P388. PMA induced the appearance of FcR on the membranes of P388 cells as indicated by the binding of IgG-coated sheep red blood cells (IgG-SRBC). Concentrations of PMA as low as 1 ng/ml were sufficient to induce the expression of FcR as well as to inhibit cellular division and to induce adherence in the P388 tumor cell line; however, optimal FcR induction occurred at PMA concentrations of 10-100 ng/ml. Immunofluorescent analysis with heat-aggregated myeloma proteins indicated that PMA induced FcR which were capable of binding IgG2a and IgG2b immunoglobulins, but not IgG1. Adherence to a substratum was determined to be a second required signal for expression of FcR, since PMA induction of P388 tumor cells in teflon dishes failed to fully develop FcR and adherence of P388 cells to poly-L-lysine-coated culture dishes in the absence of PMA was insufficient for FcR expression. FcR which appeared after PMA induction were non-functional in the sense that membrane-bound IgG-SRBC were not ingested to any significant extent by the tumor cells. However, if FcR induction occurred in the presence conA-induced rat spleen cell culture supernatants, phagocytosis of membrane-bound erythrocytes occurred. These findings suggest that for the expression of FcR which are capable of particle internalization, at least three identifiable membrane-transmitted signals are required during differentiation.     

An assay method for the prediction of tumor promoting potential of chemicals by the use of Bhas 42 cells

It has become an important task to develop a simple in vitro method for the detection of non-genotoxic carcinogens, among which tumor promoters are included. Bhas 42 cells are v-Ha-ras-transfected BALB/c 3T3 cells and are regarded as initiated cells in the 2-stage transformation paradigm. We designed a method for detecting tumor promoters by the use of Bhas 42 cells at advanced passage generation. In this method, the cells are cultured in six-well plates for 17 days during which test chemicals are added in the medium for 11 days from days 3 to 14. The end-point of the assay is the induction of transformed foci. When the tumor promoter TPA was used, a significant number of transformed foci were induced concentration-dependently, whereas only a few foci were observed in control cultures. When various chemicals were examined by the method, a reasonable correlation was observed with the reported tumor-promoting ability in animal experiments. We propose that the Bhas 42 cell transformation method is practical and useful for the detection of tumor promoters.     

Mutagenesis and carcinogenesis in nucleotide excision repair-deficient XPA knock out mice

Mice with a defect in the xeroderma pigmentosum group A (XPA) gene have a complete deficiency in nucleotide excision repair (NER). As such, these mice mimic the human XP phenotype in that they have a >1000-fold higher risk of developing UV-induced skin cancer. Besides being UV-sensitive, XPA^−/− mice also develop internal tumors when they are exposed to chemical carcinogens. To investigate the effect of a total NER deficiency on the induction of gene mutations and tumor development, we crossed XPA^−/− mice with transgenic lacZ/pUR288 mutation-indicator mice. The mice were treated with various agents and chemicals like UV-B, benzo[a]pyrene and 2-aceto-amino-fluorene. Gene mutation induction in several tumor target- and non-target tissues was determined in both the bacterial lacZ reporter gene and in the endogenous Hprt gene. Furthermore, alterations in the p53- and ras genes were determined in UV-induced skin tumors of XPA^−/− mice. In this work, we review these results and discuss the applicability and reliability of enhanced gene mutant frequencies as early indicators of tumorigenesis.     

Modulation of cell-cell communication in the cause and chemoprevention/chemotherapy of cancer

Chemopreventive or chemotherapeutic agents have been those that either kill cancer cells to a differential degree over the non-cancer cells or those chemicals that either block the induction of tumors in carcinogen-treated animals or retard transplanted tumors in animals. Carcinogenesis is a multi-stage, multi-mechanism process, involving the irreversible alteration of a stem cell ("initiation"), followed by the clonal proliferation of the initiated cell ("promotion"). To develop a strategy for intervention with chemoprevention/chemotherapeutic chemicals, the basic mechanism(s) of carcinogenesis must be understood. Gap junction intercellular communication (GJIC) regulates cell growth, differentiation, apoptosis and adaptive functions of differentiated cells. Normal cells have functional GJIC while cancer cells do not. Tumor promoters and oncogenes inhibit GJIC, while anti-tumor promoter and anti-oncogene drugs can reverse the down-regulation of GJIC. Transfection of gap junction genes (connexins) has been shown to reverse the tumorigenic phenotype. If prevention/treatment of cancer is to occur, prevention of the chronic down regulation of GJIC by tumor promoters in non-tumorigenic but initiated cells or the up-regulation of GJIC in stably down-regulated GJIC in tumor cells must occur to prevent or to treat cancers.     

Cancer risk assessment for 1,3-butadiene: data integration opportunities

The US Environmental Protection Agency recently released its new guidelines for carcinogen risk assessment together with supplemental guidance for assessing susceptibility from early-life exposure to carcinogens. In particular, these guidelines encourage the use of mechanistic data in support of dose-response characterization at doses below those at which an increase in tumor frequency over background levels might be detected. In this context of the utility of mechanistic data for human cancer risk assessment, the International Life Sciences Institute (ILSI) has developed a human relevance framework (HRF) that can be used to assess the plausibility of a mode of action (MoA) described for animal models operating in humans. The MoA is described as a sequence of key events and processes that result in an adverse outcome. A key event is a measurable precursor step that is in itself a necessary element of the MoA or is a bioindicator for such an element. A number of cellular and molecular perturbations have been identified as key events whereby DNA-reactive chemicals can produce tumors. These include DNA adducts in target tissues, gene mutations and/or chromosomal alterations in target tissues and enhanced cell proliferation in target tissues. This type of data integration approach to quantitative cancer risk assessment can be applied to 1,3-butadiene, for example, using data on biomarkers in exposed Czech workers [1]. For this study, an extensive range of biomarkers of exposure and response was assessed, including: polymorphisms in metabolizing enzymes; urinary concentrations of several metabolites of 1,3-butadiene; hemoglobin adducts; HPRT mutations in T-lymphocytes; chromosomal aberrations by FISH and conventional staining procedures; sister chromatid exchanges. Exposure levels were monitored in a comprehensive fashion. For risk assessment purposes, these data need to be considered in the context of how they inform the MoA for leukemia, the tumor type reported to be increased in synthetic rubber workers exposed to 1,3-butadiene. Also, for the HRF it is necessary to establish key events for a MoA in rodents for the induction of tumors by 1,3-butadiene. There is clearly a species difference in sensitivity to tumor induction, with mice being much more sensitive than rats; key events need to explain this difference. For butadiene, the MoA is DNA-reactivity and subsequent mutagenicity and so following the EPA's cancer guidelines, a linear extrapolation is used from the point of departure (POD), unless additional data support a non-linear extrapolation. For the present case, the human bioindicator data are not informative as far as dose-response characterization is concerned. Mouse chromosome aberration data for in vivo exposures might be used for establishing a POD, with linear extrapolation from this POD. The available cytogenetic data from rodent studies appear to be sufficiently extensive and consistent for this to be a viable approach. This approach of using MoA and key events to establish the human relevance can lead to the development of specific informative bioindicators of response that can be used as surrogates to predict the shape of the tumor dose response curve at low doses. Truly informative predictors of tumor responses should be able to provide estimates of human tumor frequencies at low, environmental exposures to 1,3-butadiene.     

Interaction of benzo[a]pyrene, 2,3,3′,4,4′,5 hexachlorobiphenyl (PCB 156) and cadmium on biomarker responses in flounder (Platichthys flesus L.)

Interactive effects of a mixed pollutant exposure on biomarker responses were studied in European flounder (Platichthys flesus L.). The model chemicals, benzo[a]pyrene (BaP, 2.5 mg kg^-1), 2,3,3′,4,4′5 hexachlorobiphenyl (PCB-156, 2.5 mg kg^-1), and cadmium (cadmium, 1 mg kg^-1), were administered to fish by subcutaneous injections. Biomarker responses were quantified both following administration of single chemicals and sequential combinations of the chemicals in pairs. Significant induction of CYP1A protein levels and corresponding ethoxyresorufin-O-deethylase (EROD) activities was observed in BaP and PCB treated flounder after 2 and 8 days, respectively. The strongest induction (44 fold) was caused by BaP. No further induction was observed after additional treatment with PCB 156. CYP1A induction caused by BaP was inhibited (40% compared with BaP treatment alone) in flounder pre treated with cadmium, whereas induction by PCB 156 appeared to be unaffected by pre treatment with cadmium. Flounder treated with cadmium only had significantly elevated hepatic levels of metallothionein (MT) after 15 days. Pre treatment with BaP and PCB prior to cadmium inhibited the MT induction (30-50%) compared with cadmium alone. Furthermore, significantly higher glutathione S transferase activities were observed in flounder administered cadmium alone, and in flounder treated with BaP or PCB 156 prior to cadmium. GST selenium independent peroxidase activities appeared to be unaffected by any of the treatments in the present study. The results indicate that chemical mixtures may affect biomarker responses differently from compounds administered alone, and that the sensitivity of both CYP1A and MT are influenced by pollutants other than their primary inducers.