- Email: [email protected]
Toxicity of Trp-P-2 to cultured human and rat keratinocytes
Chemico-Biological Interactions 127 (2000) 237 – 253
www.elsevier.com/locate/chembiont
Toxicity of Trp-P-2 to cultured human and rat keratinocytes Hyang-Sook Chun 1, Paul A. Kuzmicky, Norman Y. Kado, Robert H. Rice * Department of En6ironmental Toxicology, One Shields A6enue, Uni6ersity of California, Da6is, CA 95616 -8588, USA Received 14 February 2000; received in revised form 22 May 2000; accepted 1 June 2000
Abstract Keratinocytes cultured from human and rat epidermis exhibited strongly divergent sensitivities to toxicity from the heterocyclic amine food mutagen Trp-P-2. To find a biochemical basis for this difference, the cultured cells were compared in their expression of phase 1 and 2 biotransformation activities, mutagenic activation and macromolecular adducts. The human and early passage rat cells expressed similar levels of ethoxyresorufin O-deethylase and N-acetyl transferase activities, their microsomes were similarly active in inducing bacterial mutagenesis when incubated with Trp-P-2, and the keratinocytes accumulated similar levels of DNA adducts over a 4-day treatment period. However, the human cells expressed an order of magnitude higher cytosolic glutathione S-transferase activity than the rat cells, likely providing enhanced protection. Late passage rat epidermal cells were insensitive to Trp-P-2 toxicity, attributable to their rapid loss of measured cytochrome P450 activity. Rat esophageal and fore-stomach epithelial cells resembled late passage rat epidermal cells in their lack of sensitivity to Trp-P-2 toxicity and lack of P450 activity. Human esophageal epithelial cells expressed substantial P450 activity but, in contrast to human epidermal cells, were sensitive to Trp-P-2 toxicity. Thus keratinocytes provide a valuable
Abbre6iations: EROD, ethoxyresorufin-O-deethylase; hEp, human epidermal cells; hEs, human esophageal epithelial cells; NAT, N-acetyl transferase; rEp, rat epidermal cells; rEs, rat esophageal epithelial cells; rFst, rat fore-stomach epithelial cells; SIK, spontaneously immortalized keratinocytes; TCDD, 2,3,7,8-tetrachlorodibenzo-P-dioxin; Trp-P-1, 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole; Trp-P-2, 3-amino-1-methyl-5H-pyrido[4,3-b]indole. * Corresponding author. Tel.: + 1-530-7525176; fax: +1-530-7523394. E-mail address: [email protected] (R.H. Rice). 1 Present address: Korea Food Research Institute, San 46-1, Songnam-si, 463-420, South Korea. 0009-2797/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S0009-2797(00)00182-4
238
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
system in which to examine the basis for species- and tissue-specific differences in toxicity from this carcinogenic heterocyclic amine. © 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cytochrome P450; EROD; Heterocyclic amines; Bacterial mutagenicity; TCDD
1. Introduction 3-Amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2*) was first isolated in 1977 from tryptophan pyrolysates [1]. This heterocyclic amine is a potent bacterial mutagen [2,3] and, consistent with its genotoxicity in a variety of organs in the mouse [4,5], is a multiorgan animal carcinogen [6,7]. Along with Trp-P-1 and related heterocyclic amines, Trp-P-2 has been found in cooked foods [8,9], tobacco smoke and smoke-polluted indoor air [10,11], river water containing discharges from wastewater treatment plants [12], airborne particulates and rain [13]. Simultaneous exposure to such heterocyclic amines in combination appears additive in producing rat liver cancer [14]. Considerable species differences in genotoxicity from these agents are evident [15], however, emphasizing the importance of elucidating the mechanisms by which they act. Like many other toxic chemicals, Trp-P-2 requires metabolic activation to exert its deleterious biological effects. Although the liver is the major site for metabolism of Trp-P-2, extrahepatic metabolism contributes to its overall elimination from the body. The epidermis provides a major barrier against exposure to toxic environmental pollutants. Keratinocytes are the primary constituent of epidermis and of epithelia lining the esophagus, oropharynx, exocervix, vagina and conjunctiva and are potential peripheral targets of exposure to Trp-P-2 directly through ingestion or indirectly from the circulation. The sensitivity of this cell type to the toxic and mutagenic action of Trp-P-2 has received little attention despite the high levels of cytochromes P4501A1 and 1B1 this cell type can express. Since these isozymes are responsible for the activation of most procarcinogens in extrahepatic: tissues [16], the keratinocyte provides a useful model system for examining the cellular response to procarcinogens which these isozymes activate. Previous work has shown that a line of spontaneously immortalized keratinocytes (SIK) derived from normal human epidermis is sensitive to Trp-P-1 toxicity (stimulated by TCDD) but insensitive to Trp-P-2 [17]. The molecular basis for the difference in toxicity from these compounds, which differ by a single methyl group, is not known. Although a number of attempts have been made to link the toxicity of chemicals (especially aflatoxins) to levels of cytochrome P450 activity, the observed P450 differences alone often cannot account for species differences in toxic response; the influence of other activities and their concentration dependencies must be considered [18]. In some cases, species susceptibility has been correlated with detoxification by conjugating enzymes [19]. The present study documents differences in toxic response to Trp-P-2 between human and rat keratinocytes in culture and explores the relation of toxicity to biotransformation activities, macromolecular adducts and bacterial mutagenicity using microsomal preparations.
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
239
2. Materials and methods
2.1. Chemicals 2,3,7,8-TCDD (99% purity) was purchased from the NCI Chemical Carcinogen Repository, Midwest Research Institute (Kansas City, MO). Trp-P-2 and [14C]TrpP-2 (10 Ci/mol, \ 98% purity) were purchased from Toronto Research Chemicals (Ontario, Canada). Ethoxyresorufin, resorufin, glucose-6-phosphate, glucose-6phosphate dehydrogenase, p-aminobenzoic acid, sulfamethazine, phosphotransacetylase, acetyl phosphate, dithiothreitol, acetyl-CoA, dimethylaminobenaldehyde, 1-chloro-2,4-dinitrobenzene and dimethylsulfoxide were obtained from the Sigma Chemical Co (St Louis, MO). All other chemicals were of analytical grade.
2.2. Cell culture Spontaneously immortalized human keratinocytes (SIK, passages 33–40), normal human epidermal or esophageal epithelial cells (passages 3–5) and Sprague–Dawley rEp (passages 2 – 5), rEs and rFst epithelial cells (passages 2–4) were grown with lethally irradiated 3T3 feeder layer support in a 3:1 mixture of Dulbecco–Vogt Eagle’s and Ham’s F-12 media supplemented with 5% fetal bovine serum, 0.4 mg/ml hydrocortisone, 10 ng/ml epidermal growth factor, 10 ng/ml cholera toxin, 5 mg/ml insulin, 5 mg/ml transferrin, 20 pM triiiodothyronine, 0.18 mM adenine and antibiotics [20]. Late passages (19–25) of rEp were cultured in the same fashion without 3T3 feeder layer support. Primary cultures were established from tissue disaggregated by trypsinization or from tissue explant outgrowths and passaged by routine trypsinization [21,22]. Cultures were treated with the indicated agent or solvent beginning with the first medium change and terminated either by trypsinization and passaging or by fixing and staining with rhodanile blue [23]. Except as noted, experiments were performed a minimum of three times.
2.3. Preparation of subcellular fractions Cells near confluence were harvested after removal of remaining 3T3 feeder cells with isotonic 0.5 mM EDTA [24] and washed with ice-cold phosphate-buffered saline (pH 7.4). After centrifugation for 10 min at 1850 × g the cells were resuspended in ice-cold homogenization buffer containing 10 mM Hepes (pH 7.9), 1.5 mM MgCl2, 10 mM KCl and 0.5 mM DTT. After homogenization using a loose fitting Potter – Elvehjem homogenizer, the homogenate was centrifuged successively at 3300×g for 15 min, 12 000× g for 5 min and finally 100 000× g for 1 h. The resulting supernatant was used as a cytosolic fraction after addition of 0.11 volume of cytoplasmic buffer containing 0.3 M Hepes (pH 7.9), 1.4 M KCl and 30 mM MgC12. The microsomal pellet was resuspended in homogenization buffer containing 0. 11 volume of cytoplasmic buffer and 20% (v/v) glycerol [25].
240
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
The S-9 fraction was obtained by centrifugation of the cell homogenate at 9000 × g for 1 h. Rat liver microsomes were obtained by further centrifugation of the S-9 fraction at 100 000 ×g for 1 h and were suspended in the original volume of buffer solution. The cytosolic and microsomal protein concentrations were estimated using bicinchoninic acid with bovine serum albumin as a protein standard [26].
2.4. Enzyme acti6ity 7-Ethoxyresorufin-O-deethylase (EROD), a marker for P4501A1, 1A2 and 1B1 activity, was determined spectrofluorometrically as described [27] with slight modification. Briefly, each reaction mixture had a total volume of 1.25 ml and contained 0.5 ml of the reconstituted system (5 mM glucose-6-phosphate, 2 units of glucose-6phosphate dehydrogenase, 0.8 mg bovine serum albumin, 5 mM MgSO4), 0.15 ml of microsomal suspension, 0.58 ml of 0.1 M Hepes buffer and 10 ml of ethoxyresorufin (dissolved in methanol) giving a final concentration of 1.5 mM. The solution was allowed to equilibrate at 37°C for 5 min. The reaction was initiated by addition of NADPH to 0.6 mM and stopped by addition of 2.5 ml of methanol after 30 min. After centrifugation at 2000 rpm for 15 min, the fluorescence of the supernatant was measured with an excitation wavelength of 530 rim and an emission wavelength of 585 nm. The EROD activity was normalized to microsomal protein content. N-acetyltransferuse (NAT) activity was measured spectrophotometrically using an acetyl-CoA regenerating system [28]. Sulfamethazine and p-aminobenzoic acid were used at concentrations of 0.8 mM. The incubation mixture had a total volume of 0.09 ml and contained 0.05 ml of cytosol fraction, 0.02 ml of a solution containing 25 mM Tris –Cl (pH 7.5 at 37°C), 4.5 mM dithioerythritol, 4.5 mM EDTA, 22.5 mM acetyl phosphate, 2.25 units/ml phosphotransacetylase and 0.8 mM arylamine. After preincubation at 37°C for 5 min, the reaction was started by addition of 0.02 ml of 1 mM acetyl-CoA. The reaction was terminated by addition of 0.05 ml of 20% (w/v) trichloroacetic acid and then 0.5 ml of 5% (w/v) dimethylaminobenzaldehyde in acetonitrile was added. The samples were recentrifuged and incubated for at least 10 min at room temperature. Blank values were obtained by substituting water for acetyl-CoA in the incubation mixture. Samples were prepared in triplicate and the absorbance at 450 mm was recorded. The units of NAT activity are expressed as nm of amine acetylated/ min. Glutathione S-transferase activity was measured spectrophotometrically [29] using 1-chloro-2,4-dinitrobenzene as a substrate. The specific activity was expressed as nmol of glutathione conjugate formed/min per mg cytosolic protein.
2.5. Northern blot Confluent cultures were incubated with 5 nM TCDD or solvent alone (0.1% DMSO final concentration) for 24 h. Cultures were rinsed twice with isotonic
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
241
phosphate buffer, lysed by addition of Trizol reagent, and the total cellular RNA was then isolated. The RNA obtained was electrophoresed (30 mg/lane) in a 1.5% agarose gel in 0.66 M formaldehyde and transferred to a nitrocellulose membrane. The membrane was pre-hybridized for 4 h at 42°C in a solution containing 50% formamide, 5 × SSC, 5× Denhardt’s solution, 50 mM sodium phosphate (pH 6.5), 0.1% SDS and 0.1 mg/ml salmon sperm DNA. cDNA probes were labeled with [32P]dCTP by random priming using Klenow polymerase [30] and then added to the hybridization mixture for 24 h at 42°C. The membrane was washed at 62°C twice with 2 ×SSC-0.1% SDS and once with 0.1× SSC-0.1% SDS. Specific hybridization signals were visualized by autoradiography and quantified by scanning densitometry. Rat CYP1A1 and 1B1 cDNA probes were prepared by RTPCR of full length coding regions using template RNA from early passage rEp. Glyceraldehyde-3-phosphate dehydrogenase mRNA was used for normalization.
2.6. Adduct measurement After treatment with [14C]Trp-P-2 (1 mg/ml, 10 Ci/mol) for 4 days in the presence or absence of 5 nM TCDD, cultures were rinsed twice with serum free medium. Residual 3T3 cells were removed from near confluent cultures using isotonic EDTA and adducts were measured [17]. To this end, cells were lysed in 0.5% SDS, treated with 20 mg/ml pancreatic ribonuclease A at 37°C for 15 min, and extracted with buffer saturated phenol (pH 8.0). After phase separation, the aqueous and phenol phases were adjusted to 0.02 M sodium acetate and precipitated separately with 2.5 vol of 100% ice-cold ethanol. The protein was rinsed in 70% ethanol, redissolved in 0.1 N NaOH quantitated using bicinchoninic acid and counted by liquid scintillation. DNA was precipitated from the aqueous phase, redissolved in water and reprecipitated with ethanol. After estimation of the DNA concentration based on absorbance at 260 nm and confirmation of purity by A260/A280, aliquots were scintillation counted.
2.7. Mutagenicity testing A simple microsuspension modification [31] of the standard Salmonella assay [32] with increased sensitivity was used as described [33]. Keratinocyte cultures were treated with TCDD prior to assay. Briefly 0.1 ml of microsomes (25 mg of protein) containing enzyme cofactors, 0.1 ml of Salmonella typhimurium TA98 (1010 cells/ml) and 5 mg of Trp-P-2 dissolved in DMSO were preincubated at 37°C with shaking. After 90 min, top agar was added and the mixture was poured onto minimal glucose plates. After incubation at 37°C for 48 h, histidine revertants on each plate were counted. All experiments were performed in duplicate using duplicate plates except as noted.
242
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
3. Results
3.1. Toxicity 6ersus phase 1 biotransformation Human and rat epidermal keratinocytes were compared in sensitivity to the toxic effects of Trp-P-2 according to expansion of colonies at low density. Compared to control cultures treated with solvent alone, observed effects were judged to range from marginal to highly toxic as illustrated in Table 1. Trp-P-2 at 0.5 mg/ml was highly toxic to early passage rEp but virtually nontoxic to human epidermal cells (SIK and hEp). In contrast to the early passage rEp, those at late passage exhibited no toxic response to Trp-P-2. Human epidermal cultures treated with 5 nM TCDD (not toxic alone) showed a slight increase in sensitivity to Trp-P-2 toxicity, while there was no stimulation of Trp-P-2 toxicity by TCDD in any of the rat cultures. Keratinocytes, from the same rat but of different epithelial origin were also treated with Trp-P-2. In contrast to early passage rEp cells, rEs and rFst were affected little if at all. On the other hand, hEs were very sensitive, and the toxicity was markedly greater in the presence of TCDD. EROD activities were measured in microsomes from human and rat keratinocytes following treatment with or without TCDD. As shown in Fig. 1, similar Table 1 Trp-P-2 toxicity toward human and rat keratinocytesa Cells
Trp-P-2 (mg/ml) 0.05
0.1
0.5
−TCDD SIK hEp hEs Early rEp Late rEp rEs rFst
– – – – – – –
0.05
0.1
0.5
+TCDD – + + + – – –
+ 9 ++ +++ – 9 –
9 9 + – – – –
9 9 ++ + – – –
+ + +++ +++ – 9 –
a Cultures were treated for 10 days with Trp-P-2 at the indicated concentrations. The degree of toxicity (inhibition of colony growth) was judged in comparison with control cultures treated with solvent alone (0.1% dimethylsulfoxide) according to the scale shown below.
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
243
Fig. 1. EROD activities using microsomes from cultured human and rat keratinocytes treated with 5 nM TCDD. Each value represents the mean and standard deviation of triplicate determinations.
activities were seen in microsomes, from TCDD-treated SIK, hEp, hEs and early passage rEp cultures, indicating that CYP1A1 or 1B1-associated EROD activity was not well correlated with species differences in toxic response to Trp-P-2. In contrast, microsomes, prepared from late passage rEp treated with TCDD had little or no inducible EROD activity. The activities measured in rEs and rFst cultures were readily detected but similar to the low levels in the rEp cultures. In parallel experiments with microsomes from cultures not treated with TCDD, EROD activity was not detectable (data not shown).
3.2. Loss of toxicity and EROD with passage The effect of serial subcultivation of rEp on Trp-P-2 toxicity was examined further. As illustrated in Fig. 2, the epidermal cells lost their initial sensitivity after approximately five passages. Passages 6–14 exhibited slight to marginal toxicity, and by passage 15 the sensitivity was judged completely lost. Similar results were obtained in numerous trials using rat epidermal cells from three different animals. TCDD-inducible EROD activities were examined in rEp cultures upon serial subcultivation. As shown in Fig. 3, the activities decreased markedly through passage 5 (:50 generations) and continued to decline to background levels over
244
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
Fig. 2. Decreasing Trp-P-2 toxicity with increasing passage of rat epidermal keratinocytes. Cultures were inoculated with 200–500 cells at the indicated passages, treated with 5 ml of dimethylsulfoxide containing 0 (− ) or 2.5 mg (+ ) of Trp-P-2 at each medium change, and fixed and stained after 10 days of treatment.
Fig. 3. Diminished microsomal enzyme expression with increasing passage of rat epidermal cells. Shown are the means of duplicate or triplicate EROD determinations using TCDD-treated cultures. Cells from different animals are represented by the different symbols (circle, square, triangle). Three independent lines were initiated and passaged from one rat (circles). Inset: diminished expression of CYP1A1 (hatched bars) and CYPIB1 (solid bars) in a representative experiment upon passage of one line of rat epidermal cells. TCDD-treated cells at the indicated passages were examined by Northern blotting.
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
245
Table 2 N-acetyl transferase activities in human and rat keratinocytes Cells
SIK hEp Early rEp Late rEp
N-Acetyltransferasea (nmol/min mg−1) −TCDD
+TCDD
2.0 90.1 3.2 90.4 1.6 90.2 0.5 90.02
2.1 9 0.1 3.2 90.3 1.7 90.1 0.7 9 0.1
Each value is the mean 9 SD of three determinations. For each type of cell, the values 9TCDD were not significantly different (t-test, a= 0.05). Values for the four types of cell examined were significantly different from each other at PB0.05 by Duncan’s multiple range test. a
the next dozen passages. The rapid initial decrease was apparent using independently derived epidermal samples from three different rats. The results of the EROD assays were consistent with CYP1A1 and 1B1 expression in the rat and human epidermal cultures. As observed previously [17,34], the mRNAs for both P450s were clearly detectable in SIK, hEp and early passage rEp, but they were not detected in the late passage rEp or in any of the cultures in the absence of TCDD stimulation. As shown for a representative experiment in Fig. 3 (inset), rEp mRNA levels decreased essentially in parallel with measured EROD activities upon serial subcultivation.
3.3. Phase 2 biotransformation NAT activities were determined in the cytosolic fractions of human and rat keratinocytes with and without TCDD treatment. Using the NAT1-specific substrate p-dimethylaminobenzaldehyde, the measured activities were slightly higher in human than rat epidermal cells (Table 2). In contrast to the stimulation of EROD activity in some cells, NAT activity was not affected in any of the samples by TCDD treatment. Using the NAT2-selective substrate sulfamethazine with or without TCDD treatment, no activity was detected, suggesting this isozyme was expressed at low or negligible levels and did not contribute to the differential sensitivity of the keratinocytes to Trp-P-2. Cytosolic glutathione S-transferase activities were determined using the substrate 1-chloro-2,4-dinitrobenzene. Although the P-isozyme is reportedly the most prevalent form in human and rat skin and cultured keratinocytes [35], this substrate reacts at least moderately well with isoforms of the A, M and P classes [36,37]. As shown in Fig. 4, measured specific activites were an order of magnitude higher in the human than in the rat cells. Consistent with other reports that Ah receptor agonists increase glutathione S-transferase expression [38], the specific activities in cultures treated with TCDD were 1.5–2 (human) or 3–4 (rat) times those in untreated cultures. Thus, although TCDD was unable to induce certain P450 isozymes in late passage rEp, induction of other activities was still possible.
246
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
3.4. Mutagenic acti6ation and macromolecular adduct formation The number of revertants induced in the Salmonella tester strain TA98 was measured using microsomes from cultured keratinocytes to activate Trp-P-2 to mutagenic metabolites. Consistent with results of EROD assays, the highest number of revertants was detected in the presence of microsomes from SIK cultures, followed closely by those from early passage rEp and hEp (Table 3). Thus, the striking contrast in sensitivities of rEp and the human epidermal cells to Trp-P-2 toxicity could not be attributed to markedly different abilities to produce mutagenic metabolites. Microsomes from late passage rEp gave little or no mutagenic activity, consistent with the lack of EROD activity and toxicity seen in these cells. The amount of Trp-P-2 covalent binding to DNA and protein was examined in cultured human and rat keratinocytes. As shown in Table 4, the degree of DNA adduct formation was similar in the human epidermal and early passage rEp cells and was at least doubled by TCDD treatment. Values for late passage rEp were low and barely above the nonspecific background of the measurement (1 vs. 0.4 ng Trp-P-2/mg DNA). Protein adducts were highest in early passage rEp, 30–50% higher than in the human epidermal cells, and in each case were :50% higher in
Fig. 4. Glutathione S-transferase activities in human vs rat epidermal cells. Cultures grown in the presence (solid bars) or absence (clear bars) of TCDD were assayed.
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
247
Table 3 Mutagenic activation of Trp-P-2 by microsomes from human and rat keratinocytes TA98 Revertants/platea Trp-P-2 (ng/tube) Cells
1
10
100
SIK hEpb Early rEp Late rEp Rat liverc
769 48 43(99) 8 94 494 10549352
319 9228 142( 91) 859 55 12 9 6 2246 9520
490 9 284 195 958 460 9 158 26 9 4 2313 925
a Data are the means and standard deviations of duplicate samples from two independent experiments except hEp at 1 and 10 ng/ml (range of duplicate samples in one experiment shown in parentheses). Background rates of spontaneous reversion subtracted in different trials ranged from 12–18 per plate. b In a parallel experiment performed in duplicate, Trp-P-2 at 12.5, 25 and 50 ng/tube gave 94 ( 92), 122 (9 1) and 154 (9 11) revertants/per plate, respectively. c Positive control microsomes obtained from Arochlor 1254-treated rat liver.
Table 4 Trp-P-2 adducts of DNA and protein in human and rat keratinocytes Cells
SIK hEp Early rEp Late rEp
DNA adductsa
Protein adductsa
−TCDD
+TCDD
−TCDD
+TCDD
1.39 0.4 2.09 0.9 2.09 0.6 0.99 0.9
3.7 91.4 5.7 93.3 6.7 94.7 1.2 90.3
17 94.3 19 9 1.7 25 9 7.1 18 95.9
23 95.7 25 93.0 37 94.8 19 93.8
a Values (ng Trp-P-2/mg) represent the mean 9 SD from three experiments performed in triplicate or duplicate. Nonspecific background values of DNA and protein adducts were 0.4 and 3, respectively. Except for late rEp, the adduct levels were higher in TCDD-treated cultures than in cultures not treated with TCDD (t-test, PB0.05). DNA adducts among SIK, hEp and early rEp were not significantly different (Duncan’s multiple range test PB0.05), but protein adducts were significantly higher for rEp than for SIK or hEp. DNA and protein adducts for rEp were significantly lower than for the others except for protein adducts in cultures not treated with TCDD (not different from SIK or hEp).
cultures treated with TCDD. The measured values of protein adducts in late passage rEp were not distinguishably lower than those from human epidermal cells.
4. Discussion To study responses of potential target cell types, keratinocytes are of particular interest in view of their ready cultivability, their expression of cytochrome P450
248
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
activities (CYP1A1, CYP1B1) found in many epithelial cell types, and the ability of these cells from different anatomic origins to display intrinsic differences in culture [21,22]. Both human and rat epidermal keratinocytes were sensitive to toxicity from Trp-P-1, but they exhibited strongly contrasting sensitivities to toxicity from Trp-P-2 and, in the rat, early passage epidermal cells were much more sensitive than keratinocytes from esophagus or fore-stomach. Although previous work has shown human epidermal and esophageal cells to have similar sensitivity to toxicity from polycyclic aromatic hydrocarbons [39], present experiments reveal a striking sensitivity of the esophageal cells to Trp-P-2 that merits further study, particularly if differential cytochrome P450 isozyme expression in human esophagus [40] and epidermis can be substantiated. Because the toxicity (including mutagenicity and carcinogenicity) of Trp-P-2, like many chemicals, is mediated by production of reactive metabolites, this process is a prime candidate for helping explain differences in response. In hepatocytes, for example, the cytotoxic response toward Trp-P-2 is correlated with cytochrome P450 activity [41], which is responsible for production of the cytotoxic and genotoxic 3-hydroxyamino metabolite, N – OH–Trp-P-2 [42]. Considerable differences in hepatic cytochrome P450-mediated metabolic activation of Trp-P-2 among species have been seen [43]. In the present work, the loss of sensitivity to toxicity of the late passage rEp cells is readily attributable to the loss of P450-mediated metabolism. Unlike other P450 activities, whose mechanisms of induction are less well understood, Ah receptor-induced CYP1A1 and 1B1 are commonly expressed in culture. The unusual suppression of their inducibility in rEp is known to prevent the toxic response of these cells to polycyclic aromatic hydrocarbons [39] and can occur in CYP1A1 by activation of a negative regulatory element in the gene 5%-flanking DNA [34]. That glutathione S-transferase activity was still inducible by TCDD indicates the silencing is gene-specific and not a general suppression of Ah receptor function. What has not been appreciated before is the rate with which the silencing occurs, largely within the first half dozen passages. Apparently starting almost immediately upon cultivation, and thus probably a physiological adaptation, the process appears nearly complete before the greatly elevated colony forming efficiencies characteristic of spontaneous immortalization become evident. As previously found [21], these were seen later, typically after a half dozen or more passages (data not shown). Differences in metabolite production could explain the relative insensitivity of rEs and rFst cells to Trp-P-2 toxicity, but not the contrast in sensitivity between early passage rEp and the human epidermal cells. The highest rate of N-hydroxylation reported for Trp-P-2 and other heterocyclic aromatic amines occurs with cytochrome P4501A2, but activation is catalyzed by cytochromes P4501A1 and 1B1 [16,44,45]. Although rEp and hEp could make different metabolites to which the respective cells have qualitatively different responses, this possibility appears remote. For example, expression by rEp of CYP1A2 could produce more toxic metabolites, but this isozyme seems poorly expressed or inducible outside the liver [16] and has not been found in mouse skin [46]. Moreover, no clear distinction was seen in mutagenicity of Trp-P-2 using microsomes from the human and rat sources.
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
249
On the other hand, unlike the TCDD stimulated toxicity of Trp-P-1 to hEp and SIK [17], the observed toxicity in rEp was independent of TCDD treatment, raising the possibility that metabolism was not required for this response. (Trp-P-1 and Trp-P-2 have been proposed to be toxic without metabolism to rat hepatocytes in primary culture, although at much higher concentrations than employed presently [47]. Other possibilities are that critical metabolites were produced by a constitutive activity to which the EROD assay is insensitive or that Trp-P-2 is a good inducer in the rEp. If cytochrome P450-mediated reactions cannot account for differential toxicity in these cultures, then conjugation reactions may be responsible. Glutathione conjugation is well known to provide a protective function toward aflatoxin B1, and benzo(a)pyrene toxicity, for example [19,48]. Glutathione S-transferase activity was considerably higher in the cytosolic fraction of the human than in the rat keratinocytes, possibly conferring upon the human cells a substantially higher detoxification ability. Glutathione S-transferase activities measured in pilot studies of TCDD-treated rEs and rFst cultures were also low, though 2–3 times higher than in rEp. Glutathione S-transferase also reportedly catalyzes the conjugation of N– OH – Trp-P-2 with glutathione to yield at least three conjugates, one of which is more mutagenic than N – OH – Trp-P-2 in vitro [49]. Whether this finding has relevance in vivo is unclear, but sulfhydryl compounds have been observed to reduce Trp-P-2 DNA adducts in mammalian cells [50]. Acetylation can be a key activation or deactivation reaction in the biotransformation of heterocyclic amines [51]. Activation involves O-acetylation of the N– OH-compound or N,O-acetyl transfer in the N–OH–N-acetylated compound. The result is an unstable N-acetoxyarylamine that decomposes spontaneously to give a nitrenium ion that binds covalently to DNA and protein [52]. Differences among human hepatoma cell lines in heterocyclic amine genotoxicity have been attributed to differences in cytosolic acetyltransferase activity [53]. However, Trp-P-2 (unlike Trp-P-1) appears not to be dependent upon O-acetylation for its genotoxic activation [51,54,55]. Moroever, since in bacterial mutagenesis assays N-acetyl-Trp-P-2 exhibits only a small fraction of the activity of Trp-P-2 [21], N-acetylation could be a protective step. In humans and several animal species, the acetyltransferases are encoded by two genetic loci, designated NAT1 and NAT2 [56]. Marked species variations, capable of modulating carcinogen susceptibilities, have been noted in N-acetylation of heterocyclic aromatic amines [57]. In the present experiments, although not definitive, the lack of detectable NAT2 activity suggests this isozyme does not contribute to differential toxicity. Other studies have detected little NAT2 activity in the rat [58]. NAT1 activity was slightly higher in the cytosol from human epidermal cells than from rEp, but not enough to constitute compelling evidence of much protection by N-acetylation in this case. In general, toxic effects of agents such as Trp-P-2 reflect a balance between metabolic activation and detoxification reactions. Since differences in sensitivity among human and rat keratinocyte cultures toward Trp-P-2 toxicity cannot clearly be attributed to differences in generation of activated metabolites, suspicion rests more heavily upon detoxification reactions, especially glutathione conjugation. If
250
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
Trp-P-2, unlike Trp-P-1 [17], is not mutagenic in human epidermal cells, this could be indicative of selective metabolite detoxification. The lack of significant difference over four days of treatment in DNA adduct accumulation between hEp and early passage rEp suggests DNA repair differences are not distinctive. However, differences in comparative radical production or scavenging [59], arachidonic acid-dependent peroxidation [60], bioactivation of N–OH-Trp-P-2 by esterification pathways other than O-acetylation [61], or enzymatic reduction of the N-hydroxy derivative [62] could be important factors in the relative toxicity of Trp-P-2.
Acknowledgements This work was supported by USPHS Grants AR27130, ES05705, ES08079 and ES04699 and a fellowship (to H.-S. C.) from the Republic of Korea.
References [1] T. Sugimura, T. Kawachi, M. Nagao, et al., Mutagenic principle(s) in tryptophan and phenylalanine pyrolysis products, Proc. Jpn. Acad. 53 (1977) 58 – 61. [2] M. Nagao, Y. Takahashi, T. Yahagi, T. Sugimura, K. Takeda, K. Shudo, T. Okamoto, Mutagenicities of g-carboline derivatives related to potent mutagens found in tryptophan pyrolysates, Carcinogenesis 1 (1980) 451–454. [3] K.M. Brookman, E.P. Salazar, L.H. Thompson, Comparative mutagenic efficiencies of the DNA adducts from the cooked-food-related mutagens Trp-P-2 and IQ in CHO cells, Mut. Res. 149 (1985) 249–255. [4] Y.F. Sasaki, A. Saga, M. Akasaki, et al., In vivo genotoxicity of heterocyclic amines detected by a modified alkaline single cell gel electrophoresis assay in a multiple organ study in the mouse, Mut. Res. 395 (1997) 57–73. [5] Y.F. Sasaki, A. Saga, K. Yoshida, et al., Colon specific genotoxicity of heterocyclic amines detected by the modified alkaline single cell gel electrophoresis assay of multiple mouse organs, Mut. Res. 414 (1998) 9–14. [6] T. Sugimura, Carcinogenicity of mutagenic heterocyclic amines formed during the cooking process, Mut. Res. 150 (1985) 33–41. [7] K. Fujii, Evaluation of the newborn mouse model for chemical tumorigenesis, Carcinogenesis 12 (1991) 1409–1415. [8] Z. Yamaizumi, T. Shiomi, H. Kasai, S. Nishimura, Y. Takahashi, M. Nagao, T. Sugimura, Detection of potent mutagens, Trp-P-1 and Trp-P-2, in broiled fish, Cancer Lett. 9 (1980) 75 – 83. [9] K. Skog, A. Solyakov, P. Arvidsson, M. Jagerstad, Analysis of nonpolar heterocyclic amines in cooked foods and meat extracts using gas chromatography-mass spectrometry, J. Chromatog. 803 (1998) 227–233. [10] S. Manabe, O. Wada, Carcinogenic tryptophan pyrolysis products in cigarette smoke condensate and cigarette smoke-polluted indoor air, Environ. Poll. 64 (1990) 121 – 132. [11] K. Wakabayashi, I.S. Kim, R. Kurosaka, et al., Identification of new mutagenic heterocyclic amines and quantification of known heterocyclic amines, Princess Takamatsu Sym. 23 (1995) 39 – 49. [12] T. Ohe, Quantification of mutagenic/carcinogenic heterocyclic amines, MeIQx, Trp-P-1, Trp-P-2 and PhIP, contributing highly to genotoxicity of river water, Mut. Res. 393 (1997) 73 – 79. [13] S. Manabe, E. Uchino, O. Wada, Carcinogenic tryptophan pyrolysis products in airborne particles and rain water, Mut. Res. 226 (1989) 215 – 221.
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
251
[14] R. Hasegawa, T. Kato, M. Hirose, S. Takahashi, T. Shirai, N. Ito, Enhancement of hepatocarcinogenesis by combined administration of food-derived heterocyclic amines at low doses in the rat, Food Chem. Toxicol. 34 (1996) 1097 – 1110. [15] J.A. Beamond, P.T. Barton, J.M. Tredger, R.J. Proce, B.G. Lake, Effect of some cooked food mutagens on unscheduled DNA synthesis in cultured precision-cut rat, mouse and human liver slices, Food Chem. Toxicol. 36 (1998) 455 – 466. [16] T. Shimada, C.L. Hayes, H. Yamazaki, S. Amin, S.S. Hecht, F.P. Guengerich, T.R. Sutter, Activation of chemically diverse procarcinogen by human cytochrome P-450 1B1, Cancer Res. 56 (1996) 2979–2984. [17] A.A. Walsh, L.A. deGraffenried, R.H. Rice, 2,3,7,8-Tetrachlorodibenzo-p-dioxin sensitization of cultured human epidermal cells to toxicity from a carcinogenic heterocyclic amine, Carcinogenesis 16 (1995) 2187–2191. [18] H.S. Ramsdell, D.L. Eaton, Species susceptibility to aflatoxin B1 carcinogenesis: comparative kinetics of microsomal biotransformation, Cancer Res. 50 (1990) 615 – 620. [19] G.H. Degen, H.G. Neumann, Differences in aflatoxin B1-susceptibility of rat and mouse are correlated with the capability in vitro to inactivate aflatoxin B1-epoxide, Carcinogenesis 2 (1981) 299–306. [20] B.L. Allen-Hoffman, L.G. Rheinwald, Polycyclic aromatic hydrocarbon mutagenesis of human epidermal keratinocytes in culture, Proc. Nad. Acad. Sci. USA 81 (1984) 7802 – 7806. [21] R. Heimann, R.H. Rice, Rat esophageal and epidermal keratinocytes: intrinsic differences in culture and derivation of continuous lines, J. Cell. Physiol. 117 (1983) 362 – 367. [22] M.A. Phillips, R.H. Rice, Convergent differentiation in cultured rat cells from nonkeratinizing epithelia: keratinocyte character and intrinsic differences, J. Cell Biol. 97 (1983) 686 – 691. [23] J.G. Rheinwald, H. Green, Formation of a keratinizing epithelium by a cloned cell line derived from a teratoma, Cell 6 (1975) 317– 330. [24] T.-T. Sun, H. Green, Differentiation of the epidermal keratinocyte in cell culture: formation of the cornified envelope, Cell 9 (1976) 511 – 521. [25] F.H. Ausubel, R. Brent, R. Kingston, et al., DNA protein interactions, Current Protocols in Molecular Biology, 2, John Wiley & Sons, New York, 1993, p. 12. [26] P.K. Smith, R.I. Krohn, G.T. Hermanson, et al., Measurement of protein using bicinchoninic acid, Analyt. Biochem. 150 (1985) 76–85. [27] J.J. Reiners Jr., A.R. Cantu, A. Pavone, Modulation of constitutive cytochrome P450 expression in vivo and in vitro in murine keratinocytes as a function of differentiation and extracellular Ca2 + concentration, Proc. Nad. Acad. Sci. USA 87 (1990) 1825 – 1829. [28] H.H. Andres, A.J. Klem, S.M. Szabo, W.W. Weber, New spectrophotometric and radiochemical assays for acetyl-CoA:arylamine N-acetyltransferase applicable to a variety of arylamines, Analyt. Biochem. 145 (1985) 367–375. [29] W.H. Habig, M.J. Pabst, W.B. Jacoby, Glutathione-S-transferase: the first enzymatic step in mercapturic acid formation, J. Biol. Chem. 249 (1974) 7130 – 7139. [30] A.P. Feinberg, B. Vogelstein, A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity, Analyt. Biochem. 132 (1983) 6 – 13. [31] N.Y. Kado, D. Langley, E. Eisenstadt, A simple modification of the Salmonella liquid incubation assay: increased sensitivity for detecting mutagens in human urine, Mut. Res. 121 (1983) 25 – 32. [32] B.N. Ames, J. McCann, E. Yamasaki, Methods for detecting carcinogens and mutagens with the Salmonella/mammalian-microsome mutagenicity test, Mut. Res. 31 (1975) 347 – 364. [33] D.J. Loury, N.Y. Kado, L.L. Byard, Enhancement of hepatocellular genotoxicity of several mutagens from amino acid pyrolysates and broiled foods following ethanol pretreatment, Food Chem. Toxicol. 23 (1985) 661–667. [34] A.A. Walsh, K. Tullis, R.H. Rice, M.S. Denison, Identification of a novel cis-acting negative regulatory element affecting transcription of the CYP1A1 gene in rat epidermal cells, J. Biol. Chem. 271 (1996) 22746–22753. [35] H. Raza, Y.C. Awasthi, M.T. Zaim, R.L. Eckert, H. Mukhtar, Glutathione S-transferases in human and rodent skin: multiple forms and species-specific expression, J. Invest. Dermatol. 96 (1991) 463–467.
252
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
[36] J.D. Hayes, D.J. Pulford, The glutathione S-transferase supergene family: regulation of GST and the contribution of the isozymes to cancer chemoprotection and drug resistance, Crit. Rev. Biochem. Mol. Biol. 30 (1995) 445– 446. [37] D.L. Eaton, T.K. Bamniler, Concise review of the glutathione S-transferases and their significance to toxicology, Toxicol. Sci. 49 (1999) 156 – 164. [38] S.H. Safe, Modulation of gene expression and endocrine response pathways by 2,3,7,8 tetrachlorodibeno-p-dioxin and related compounds, Pharmacol. Therap. 67 (1995) 247 – 281. [39] R. Heimann, R.H. Rice, Polycyclic aromatic hydrocarbon toxicity and induction of metabolism in cultivated esophageal and epidermal keratinocytes, Cancer Res. 43 (1983) 4856 – 4862. [40] M. Lechevrel, A.G. Casson, C.R. Wolf, et al., Characterization of cytochrome P450 expression in human esophageal mucosa, Carcinogensis 20 (1999) 243 – 248. [41] T. Segawa, H. Ishiga, H. Ueno, K. Nakamuro, Y. Sayato, Genotoxicity and cytotoxicity of 3-amino-1-methyl-5H-pyrido[4,3-b] lindole (Trp-P-2), Chemosphere 28 (1994) 853 – 861. [42] Y. Yamazoe, K. Ishii, T. Kamataki, R. Kato, T. Sugimura, Isolation and characterization of active metabolites of tryptophan-pyrolysate mutagen, Trp-P-2, formed by rat liver microsomes, Chem.Biol. Interact. 30 (1980) 125–138. [43] Y. Yamazoe, T. Kamataki, R. Kato, Species difference in N-hydroxylation of a tryptophan pyrolysis product in relation to mutagenic activation, Cancer Res. 41 (1981) 4518 – 4522. [44] T. Aoyama, F.J. Gonzalez, H.V. Gelboin, Human cDNA-expressed cytochrome P4501A2: mutagen activation and substrate specificity, Mol. Carcinogen. 2 (1989) 192 – 198. [45] M.E. McManus, W.M. Burgess, M.E. Veronese, A. Huggett, L.C. Quattrochi, R.H. Tukey, Metabolism of 2-acetylaminofluorine and benzo(a)pyrene and activation of food-derived heterocyclic amine mutagens by human cytochromes, P-450, Cancer Res. 50 (1990) 3367 – 3376. [46] Y. Shimizu, Y. Nakatsuru, M. Ichinose, et al., Benzo(a)pyrene carcinogenicity is lost in mice lacking the aryl hydrocarbon receptor, Proc. Nad. Acad. Sci. USA 97 (2000) 779 – 782. [47] H. Ashida, B. Shiotani, H. Adachi, T. Hashimoto, K. Kanazawa, G. Danno, Tryptophan pyrolysis products, Trp-P-1 and Trp-P-2 induce apoptosis in primary cultured rat hepatocytes, Biosci. Biotechnol. Biochem. 62 (1998) 2283 – 2287. [48] D. Ho, W.E. Fahi, Modification of glutathione levels in C3H/10T1/2 cells and its relationship to benzo(a)pyrene anti-7,8-dihydrodiol-9,10-epoxide-induced cytotoxicity, J. Biol. Chem. 259 (1984) 11231–11235. [49] K. Saito, Y. Yamazoe, T. Kamataki, R. Kato, Glutathione transferase-mediated and non enzymatic activation and detoxification of the N-hydroxy derivative of Trp-P-2, a potent pyrolysate proMutagen, Xenobiotica 14 (1984) 545–548. [50] S. Mita, Y. Yarnazoe, T. Kamataki, R. Kato, Metabolic activation of Trp-P-2, a tryptophan pyrolysis mutagen, by isolated rat hepatocytes, Biochem. Pharmacol. 32 (1983) 1179 – 1182. [51] D. Wild, W. Feser, S. Michel, H.L. Lord, P.D. Josephy, Metabolic activation of heterocyclic aromatic amines catalyzed by human arylamine N-acetyltransferase isoenzymes (NAT1 and NAT2) expressed in Salmonella typhimurium, Carcinogenesis 16 (1995) 643 – 648. [52] W.W. Weber, D.W. Hein, N-Acetylation pharmacogenetics, Pharmacol. Rev. 37 (1985) 25 – 79. [53] S. Knasmuller, C. Schwab, S. Land, et al., Genotoxic effects of heterocyclic aromatic amines in human derived hepatorna (HepG2) cells, Mutagenesis 14 (1999) 533 – 539. [54] Y. Hashimoto, K. Shudo, T. OkaMoto, Activation of a mutagen, 3-amino-1-methyl-5H pyrido[4,3b]indole. Identification of 3-hydroxyamino-1-methyl-5H-pyrido[4,3-b]indole and its reaction with DNA, Biochem. Biophys. Res. Commun. 96 (1980) 355 – 362. [55] Y. Oda, H. Yamazaki, M. Watanabe, T. Nohmi, T. Shimada, Development of high sensitive umu test system: rapid detection of genotoxicity of promutagenic aromatic amines by Salmonella typhimurium strain NM2009 possessing high O-acetyltransferase activity, Mut. Res. 334 (1995) 145–156. [56] M.A. Doll, D.W. Hein, Cloning, sequencing and expression of NAT1 and NAT2 encoding genes from rapid and slow acetylator inbred rats, Pharmacogenetics 5 (1995) 247 – 251. [57] A. Shinohara, Y. Yamazoe, K. Saito, T. Kamataki, R. Kato, Species differences in the N-acetylation by liver cytosol of mutagenic heterocyclic aromatic amines in protein pyrolysates, Carcinogenesis 5 (1984) 683–686.
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
253
[58] R.F. Jones, S.J. Land, C.M. King, Recombinant rat and hamster N-acetyltransferase-l and-2: relative rates of N-acetylation of arylamines and N,O-acyltransfer with arylhydroxamic acids, Carcinogenesis 17 (1996) 1729–1733. [59] J.J. Reiners Jr., T. Rupp, Conversion of xanthine dehydrogenase to xanthine oxidase occurs during keratinocyte differentiation: modulation by 12-O-tetradecanoylphorbol-13-acetate, J. Invest. Dermatol. 93 (1989) 132–135. [60] T. Flammang, Y. Yamazoe, R.W. Benison, et al., Arachidonic acid-dependent peroxidative activation of carcinogenic arylamines by extrahepatic human tissue microsomes, Cancer Res. 49 (1989) 1977–1982. [61] H.A.J. Schut, E.G. Snyderwine, DNA adducts of heterocyclic amine food mutagens: implications for mutagenesis and carcinogenesis, Carcinogenesis 20 (1999) 353 – 368. [62] R.S. King, C.H. Teitel, L.G. Shaddock, D.A. Casciano, F.F. Kadlubar, Detoxification of carcinogenic aromatic amines and heterocyclic amines by enzymatic reduction of the N-hydroxy derivative, Cancer Lett. 143 (1999) 167–171.
.
www.elsevier.com/locate/chembiont
Toxicity of Trp-P-2 to cultured human and rat keratinocytes Hyang-Sook Chun 1, Paul A. Kuzmicky, Norman Y. Kado, Robert H. Rice * Department of En6ironmental Toxicology, One Shields A6enue, Uni6ersity of California, Da6is, CA 95616 -8588, USA Received 14 February 2000; received in revised form 22 May 2000; accepted 1 June 2000
Abstract Keratinocytes cultured from human and rat epidermis exhibited strongly divergent sensitivities to toxicity from the heterocyclic amine food mutagen Trp-P-2. To find a biochemical basis for this difference, the cultured cells were compared in their expression of phase 1 and 2 biotransformation activities, mutagenic activation and macromolecular adducts. The human and early passage rat cells expressed similar levels of ethoxyresorufin O-deethylase and N-acetyl transferase activities, their microsomes were similarly active in inducing bacterial mutagenesis when incubated with Trp-P-2, and the keratinocytes accumulated similar levels of DNA adducts over a 4-day treatment period. However, the human cells expressed an order of magnitude higher cytosolic glutathione S-transferase activity than the rat cells, likely providing enhanced protection. Late passage rat epidermal cells were insensitive to Trp-P-2 toxicity, attributable to their rapid loss of measured cytochrome P450 activity. Rat esophageal and fore-stomach epithelial cells resembled late passage rat epidermal cells in their lack of sensitivity to Trp-P-2 toxicity and lack of P450 activity. Human esophageal epithelial cells expressed substantial P450 activity but, in contrast to human epidermal cells, were sensitive to Trp-P-2 toxicity. Thus keratinocytes provide a valuable
Abbre6iations: EROD, ethoxyresorufin-O-deethylase; hEp, human epidermal cells; hEs, human esophageal epithelial cells; NAT, N-acetyl transferase; rEp, rat epidermal cells; rEs, rat esophageal epithelial cells; rFst, rat fore-stomach epithelial cells; SIK, spontaneously immortalized keratinocytes; TCDD, 2,3,7,8-tetrachlorodibenzo-P-dioxin; Trp-P-1, 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole; Trp-P-2, 3-amino-1-methyl-5H-pyrido[4,3-b]indole. * Corresponding author. Tel.: + 1-530-7525176; fax: +1-530-7523394. E-mail address: [email protected] (R.H. Rice). 1 Present address: Korea Food Research Institute, San 46-1, Songnam-si, 463-420, South Korea. 0009-2797/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S0009-2797(00)00182-4
238
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
system in which to examine the basis for species- and tissue-specific differences in toxicity from this carcinogenic heterocyclic amine. © 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cytochrome P450; EROD; Heterocyclic amines; Bacterial mutagenicity; TCDD
1. Introduction 3-Amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2*) was first isolated in 1977 from tryptophan pyrolysates [1]. This heterocyclic amine is a potent bacterial mutagen [2,3] and, consistent with its genotoxicity in a variety of organs in the mouse [4,5], is a multiorgan animal carcinogen [6,7]. Along with Trp-P-1 and related heterocyclic amines, Trp-P-2 has been found in cooked foods [8,9], tobacco smoke and smoke-polluted indoor air [10,11], river water containing discharges from wastewater treatment plants [12], airborne particulates and rain [13]. Simultaneous exposure to such heterocyclic amines in combination appears additive in producing rat liver cancer [14]. Considerable species differences in genotoxicity from these agents are evident [15], however, emphasizing the importance of elucidating the mechanisms by which they act. Like many other toxic chemicals, Trp-P-2 requires metabolic activation to exert its deleterious biological effects. Although the liver is the major site for metabolism of Trp-P-2, extrahepatic metabolism contributes to its overall elimination from the body. The epidermis provides a major barrier against exposure to toxic environmental pollutants. Keratinocytes are the primary constituent of epidermis and of epithelia lining the esophagus, oropharynx, exocervix, vagina and conjunctiva and are potential peripheral targets of exposure to Trp-P-2 directly through ingestion or indirectly from the circulation. The sensitivity of this cell type to the toxic and mutagenic action of Trp-P-2 has received little attention despite the high levels of cytochromes P4501A1 and 1B1 this cell type can express. Since these isozymes are responsible for the activation of most procarcinogens in extrahepatic: tissues [16], the keratinocyte provides a useful model system for examining the cellular response to procarcinogens which these isozymes activate. Previous work has shown that a line of spontaneously immortalized keratinocytes (SIK) derived from normal human epidermis is sensitive to Trp-P-1 toxicity (stimulated by TCDD) but insensitive to Trp-P-2 [17]. The molecular basis for the difference in toxicity from these compounds, which differ by a single methyl group, is not known. Although a number of attempts have been made to link the toxicity of chemicals (especially aflatoxins) to levels of cytochrome P450 activity, the observed P450 differences alone often cannot account for species differences in toxic response; the influence of other activities and their concentration dependencies must be considered [18]. In some cases, species susceptibility has been correlated with detoxification by conjugating enzymes [19]. The present study documents differences in toxic response to Trp-P-2 between human and rat keratinocytes in culture and explores the relation of toxicity to biotransformation activities, macromolecular adducts and bacterial mutagenicity using microsomal preparations.
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
239
2. Materials and methods
2.1. Chemicals 2,3,7,8-TCDD (99% purity) was purchased from the NCI Chemical Carcinogen Repository, Midwest Research Institute (Kansas City, MO). Trp-P-2 and [14C]TrpP-2 (10 Ci/mol, \ 98% purity) were purchased from Toronto Research Chemicals (Ontario, Canada). Ethoxyresorufin, resorufin, glucose-6-phosphate, glucose-6phosphate dehydrogenase, p-aminobenzoic acid, sulfamethazine, phosphotransacetylase, acetyl phosphate, dithiothreitol, acetyl-CoA, dimethylaminobenaldehyde, 1-chloro-2,4-dinitrobenzene and dimethylsulfoxide were obtained from the Sigma Chemical Co (St Louis, MO). All other chemicals were of analytical grade.
2.2. Cell culture Spontaneously immortalized human keratinocytes (SIK, passages 33–40), normal human epidermal or esophageal epithelial cells (passages 3–5) and Sprague–Dawley rEp (passages 2 – 5), rEs and rFst epithelial cells (passages 2–4) were grown with lethally irradiated 3T3 feeder layer support in a 3:1 mixture of Dulbecco–Vogt Eagle’s and Ham’s F-12 media supplemented with 5% fetal bovine serum, 0.4 mg/ml hydrocortisone, 10 ng/ml epidermal growth factor, 10 ng/ml cholera toxin, 5 mg/ml insulin, 5 mg/ml transferrin, 20 pM triiiodothyronine, 0.18 mM adenine and antibiotics [20]. Late passages (19–25) of rEp were cultured in the same fashion without 3T3 feeder layer support. Primary cultures were established from tissue disaggregated by trypsinization or from tissue explant outgrowths and passaged by routine trypsinization [21,22]. Cultures were treated with the indicated agent or solvent beginning with the first medium change and terminated either by trypsinization and passaging or by fixing and staining with rhodanile blue [23]. Except as noted, experiments were performed a minimum of three times.
2.3. Preparation of subcellular fractions Cells near confluence were harvested after removal of remaining 3T3 feeder cells with isotonic 0.5 mM EDTA [24] and washed with ice-cold phosphate-buffered saline (pH 7.4). After centrifugation for 10 min at 1850 × g the cells were resuspended in ice-cold homogenization buffer containing 10 mM Hepes (pH 7.9), 1.5 mM MgCl2, 10 mM KCl and 0.5 mM DTT. After homogenization using a loose fitting Potter – Elvehjem homogenizer, the homogenate was centrifuged successively at 3300×g for 15 min, 12 000× g for 5 min and finally 100 000× g for 1 h. The resulting supernatant was used as a cytosolic fraction after addition of 0.11 volume of cytoplasmic buffer containing 0.3 M Hepes (pH 7.9), 1.4 M KCl and 30 mM MgC12. The microsomal pellet was resuspended in homogenization buffer containing 0. 11 volume of cytoplasmic buffer and 20% (v/v) glycerol [25].
240
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
The S-9 fraction was obtained by centrifugation of the cell homogenate at 9000 × g for 1 h. Rat liver microsomes were obtained by further centrifugation of the S-9 fraction at 100 000 ×g for 1 h and were suspended in the original volume of buffer solution. The cytosolic and microsomal protein concentrations were estimated using bicinchoninic acid with bovine serum albumin as a protein standard [26].
2.4. Enzyme acti6ity 7-Ethoxyresorufin-O-deethylase (EROD), a marker for P4501A1, 1A2 and 1B1 activity, was determined spectrofluorometrically as described [27] with slight modification. Briefly, each reaction mixture had a total volume of 1.25 ml and contained 0.5 ml of the reconstituted system (5 mM glucose-6-phosphate, 2 units of glucose-6phosphate dehydrogenase, 0.8 mg bovine serum albumin, 5 mM MgSO4), 0.15 ml of microsomal suspension, 0.58 ml of 0.1 M Hepes buffer and 10 ml of ethoxyresorufin (dissolved in methanol) giving a final concentration of 1.5 mM. The solution was allowed to equilibrate at 37°C for 5 min. The reaction was initiated by addition of NADPH to 0.6 mM and stopped by addition of 2.5 ml of methanol after 30 min. After centrifugation at 2000 rpm for 15 min, the fluorescence of the supernatant was measured with an excitation wavelength of 530 rim and an emission wavelength of 585 nm. The EROD activity was normalized to microsomal protein content. N-acetyltransferuse (NAT) activity was measured spectrophotometrically using an acetyl-CoA regenerating system [28]. Sulfamethazine and p-aminobenzoic acid were used at concentrations of 0.8 mM. The incubation mixture had a total volume of 0.09 ml and contained 0.05 ml of cytosol fraction, 0.02 ml of a solution containing 25 mM Tris –Cl (pH 7.5 at 37°C), 4.5 mM dithioerythritol, 4.5 mM EDTA, 22.5 mM acetyl phosphate, 2.25 units/ml phosphotransacetylase and 0.8 mM arylamine. After preincubation at 37°C for 5 min, the reaction was started by addition of 0.02 ml of 1 mM acetyl-CoA. The reaction was terminated by addition of 0.05 ml of 20% (w/v) trichloroacetic acid and then 0.5 ml of 5% (w/v) dimethylaminobenzaldehyde in acetonitrile was added. The samples were recentrifuged and incubated for at least 10 min at room temperature. Blank values were obtained by substituting water for acetyl-CoA in the incubation mixture. Samples were prepared in triplicate and the absorbance at 450 mm was recorded. The units of NAT activity are expressed as nm of amine acetylated/ min. Glutathione S-transferase activity was measured spectrophotometrically [29] using 1-chloro-2,4-dinitrobenzene as a substrate. The specific activity was expressed as nmol of glutathione conjugate formed/min per mg cytosolic protein.
2.5. Northern blot Confluent cultures were incubated with 5 nM TCDD or solvent alone (0.1% DMSO final concentration) for 24 h. Cultures were rinsed twice with isotonic
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
241
phosphate buffer, lysed by addition of Trizol reagent, and the total cellular RNA was then isolated. The RNA obtained was electrophoresed (30 mg/lane) in a 1.5% agarose gel in 0.66 M formaldehyde and transferred to a nitrocellulose membrane. The membrane was pre-hybridized for 4 h at 42°C in a solution containing 50% formamide, 5 × SSC, 5× Denhardt’s solution, 50 mM sodium phosphate (pH 6.5), 0.1% SDS and 0.1 mg/ml salmon sperm DNA. cDNA probes were labeled with [32P]dCTP by random priming using Klenow polymerase [30] and then added to the hybridization mixture for 24 h at 42°C. The membrane was washed at 62°C twice with 2 ×SSC-0.1% SDS and once with 0.1× SSC-0.1% SDS. Specific hybridization signals were visualized by autoradiography and quantified by scanning densitometry. Rat CYP1A1 and 1B1 cDNA probes were prepared by RTPCR of full length coding regions using template RNA from early passage rEp. Glyceraldehyde-3-phosphate dehydrogenase mRNA was used for normalization.
2.6. Adduct measurement After treatment with [14C]Trp-P-2 (1 mg/ml, 10 Ci/mol) for 4 days in the presence or absence of 5 nM TCDD, cultures were rinsed twice with serum free medium. Residual 3T3 cells were removed from near confluent cultures using isotonic EDTA and adducts were measured [17]. To this end, cells were lysed in 0.5% SDS, treated with 20 mg/ml pancreatic ribonuclease A at 37°C for 15 min, and extracted with buffer saturated phenol (pH 8.0). After phase separation, the aqueous and phenol phases were adjusted to 0.02 M sodium acetate and precipitated separately with 2.5 vol of 100% ice-cold ethanol. The protein was rinsed in 70% ethanol, redissolved in 0.1 N NaOH quantitated using bicinchoninic acid and counted by liquid scintillation. DNA was precipitated from the aqueous phase, redissolved in water and reprecipitated with ethanol. After estimation of the DNA concentration based on absorbance at 260 nm and confirmation of purity by A260/A280, aliquots were scintillation counted.
2.7. Mutagenicity testing A simple microsuspension modification [31] of the standard Salmonella assay [32] with increased sensitivity was used as described [33]. Keratinocyte cultures were treated with TCDD prior to assay. Briefly 0.1 ml of microsomes (25 mg of protein) containing enzyme cofactors, 0.1 ml of Salmonella typhimurium TA98 (1010 cells/ml) and 5 mg of Trp-P-2 dissolved in DMSO were preincubated at 37°C with shaking. After 90 min, top agar was added and the mixture was poured onto minimal glucose plates. After incubation at 37°C for 48 h, histidine revertants on each plate were counted. All experiments were performed in duplicate using duplicate plates except as noted.
242
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
3. Results
3.1. Toxicity 6ersus phase 1 biotransformation Human and rat epidermal keratinocytes were compared in sensitivity to the toxic effects of Trp-P-2 according to expansion of colonies at low density. Compared to control cultures treated with solvent alone, observed effects were judged to range from marginal to highly toxic as illustrated in Table 1. Trp-P-2 at 0.5 mg/ml was highly toxic to early passage rEp but virtually nontoxic to human epidermal cells (SIK and hEp). In contrast to the early passage rEp, those at late passage exhibited no toxic response to Trp-P-2. Human epidermal cultures treated with 5 nM TCDD (not toxic alone) showed a slight increase in sensitivity to Trp-P-2 toxicity, while there was no stimulation of Trp-P-2 toxicity by TCDD in any of the rat cultures. Keratinocytes, from the same rat but of different epithelial origin were also treated with Trp-P-2. In contrast to early passage rEp cells, rEs and rFst were affected little if at all. On the other hand, hEs were very sensitive, and the toxicity was markedly greater in the presence of TCDD. EROD activities were measured in microsomes from human and rat keratinocytes following treatment with or without TCDD. As shown in Fig. 1, similar Table 1 Trp-P-2 toxicity toward human and rat keratinocytesa Cells
Trp-P-2 (mg/ml) 0.05
0.1
0.5
−TCDD SIK hEp hEs Early rEp Late rEp rEs rFst
– – – – – – –
0.05
0.1
0.5
+TCDD – + + + – – –
+ 9 ++ +++ – 9 –
9 9 + – – – –
9 9 ++ + – – –
+ + +++ +++ – 9 –
a Cultures were treated for 10 days with Trp-P-2 at the indicated concentrations. The degree of toxicity (inhibition of colony growth) was judged in comparison with control cultures treated with solvent alone (0.1% dimethylsulfoxide) according to the scale shown below.
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
243
Fig. 1. EROD activities using microsomes from cultured human and rat keratinocytes treated with 5 nM TCDD. Each value represents the mean and standard deviation of triplicate determinations.
activities were seen in microsomes, from TCDD-treated SIK, hEp, hEs and early passage rEp cultures, indicating that CYP1A1 or 1B1-associated EROD activity was not well correlated with species differences in toxic response to Trp-P-2. In contrast, microsomes, prepared from late passage rEp treated with TCDD had little or no inducible EROD activity. The activities measured in rEs and rFst cultures were readily detected but similar to the low levels in the rEp cultures. In parallel experiments with microsomes from cultures not treated with TCDD, EROD activity was not detectable (data not shown).
3.2. Loss of toxicity and EROD with passage The effect of serial subcultivation of rEp on Trp-P-2 toxicity was examined further. As illustrated in Fig. 2, the epidermal cells lost their initial sensitivity after approximately five passages. Passages 6–14 exhibited slight to marginal toxicity, and by passage 15 the sensitivity was judged completely lost. Similar results were obtained in numerous trials using rat epidermal cells from three different animals. TCDD-inducible EROD activities were examined in rEp cultures upon serial subcultivation. As shown in Fig. 3, the activities decreased markedly through passage 5 (:50 generations) and continued to decline to background levels over
244
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
Fig. 2. Decreasing Trp-P-2 toxicity with increasing passage of rat epidermal keratinocytes. Cultures were inoculated with 200–500 cells at the indicated passages, treated with 5 ml of dimethylsulfoxide containing 0 (− ) or 2.5 mg (+ ) of Trp-P-2 at each medium change, and fixed and stained after 10 days of treatment.
Fig. 3. Diminished microsomal enzyme expression with increasing passage of rat epidermal cells. Shown are the means of duplicate or triplicate EROD determinations using TCDD-treated cultures. Cells from different animals are represented by the different symbols (circle, square, triangle). Three independent lines were initiated and passaged from one rat (circles). Inset: diminished expression of CYP1A1 (hatched bars) and CYPIB1 (solid bars) in a representative experiment upon passage of one line of rat epidermal cells. TCDD-treated cells at the indicated passages were examined by Northern blotting.
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
245
Table 2 N-acetyl transferase activities in human and rat keratinocytes Cells
SIK hEp Early rEp Late rEp
N-Acetyltransferasea (nmol/min mg−1) −TCDD
+TCDD
2.0 90.1 3.2 90.4 1.6 90.2 0.5 90.02
2.1 9 0.1 3.2 90.3 1.7 90.1 0.7 9 0.1
Each value is the mean 9 SD of three determinations. For each type of cell, the values 9TCDD were not significantly different (t-test, a= 0.05). Values for the four types of cell examined were significantly different from each other at PB0.05 by Duncan’s multiple range test. a
the next dozen passages. The rapid initial decrease was apparent using independently derived epidermal samples from three different rats. The results of the EROD assays were consistent with CYP1A1 and 1B1 expression in the rat and human epidermal cultures. As observed previously [17,34], the mRNAs for both P450s were clearly detectable in SIK, hEp and early passage rEp, but they were not detected in the late passage rEp or in any of the cultures in the absence of TCDD stimulation. As shown for a representative experiment in Fig. 3 (inset), rEp mRNA levels decreased essentially in parallel with measured EROD activities upon serial subcultivation.
3.3. Phase 2 biotransformation NAT activities were determined in the cytosolic fractions of human and rat keratinocytes with and without TCDD treatment. Using the NAT1-specific substrate p-dimethylaminobenzaldehyde, the measured activities were slightly higher in human than rat epidermal cells (Table 2). In contrast to the stimulation of EROD activity in some cells, NAT activity was not affected in any of the samples by TCDD treatment. Using the NAT2-selective substrate sulfamethazine with or without TCDD treatment, no activity was detected, suggesting this isozyme was expressed at low or negligible levels and did not contribute to the differential sensitivity of the keratinocytes to Trp-P-2. Cytosolic glutathione S-transferase activities were determined using the substrate 1-chloro-2,4-dinitrobenzene. Although the P-isozyme is reportedly the most prevalent form in human and rat skin and cultured keratinocytes [35], this substrate reacts at least moderately well with isoforms of the A, M and P classes [36,37]. As shown in Fig. 4, measured specific activites were an order of magnitude higher in the human than in the rat cells. Consistent with other reports that Ah receptor agonists increase glutathione S-transferase expression [38], the specific activities in cultures treated with TCDD were 1.5–2 (human) or 3–4 (rat) times those in untreated cultures. Thus, although TCDD was unable to induce certain P450 isozymes in late passage rEp, induction of other activities was still possible.
246
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
3.4. Mutagenic acti6ation and macromolecular adduct formation The number of revertants induced in the Salmonella tester strain TA98 was measured using microsomes from cultured keratinocytes to activate Trp-P-2 to mutagenic metabolites. Consistent with results of EROD assays, the highest number of revertants was detected in the presence of microsomes from SIK cultures, followed closely by those from early passage rEp and hEp (Table 3). Thus, the striking contrast in sensitivities of rEp and the human epidermal cells to Trp-P-2 toxicity could not be attributed to markedly different abilities to produce mutagenic metabolites. Microsomes from late passage rEp gave little or no mutagenic activity, consistent with the lack of EROD activity and toxicity seen in these cells. The amount of Trp-P-2 covalent binding to DNA and protein was examined in cultured human and rat keratinocytes. As shown in Table 4, the degree of DNA adduct formation was similar in the human epidermal and early passage rEp cells and was at least doubled by TCDD treatment. Values for late passage rEp were low and barely above the nonspecific background of the measurement (1 vs. 0.4 ng Trp-P-2/mg DNA). Protein adducts were highest in early passage rEp, 30–50% higher than in the human epidermal cells, and in each case were :50% higher in
Fig. 4. Glutathione S-transferase activities in human vs rat epidermal cells. Cultures grown in the presence (solid bars) or absence (clear bars) of TCDD were assayed.
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
247
Table 3 Mutagenic activation of Trp-P-2 by microsomes from human and rat keratinocytes TA98 Revertants/platea Trp-P-2 (ng/tube) Cells
1
10
100
SIK hEpb Early rEp Late rEp Rat liverc
769 48 43(99) 8 94 494 10549352
319 9228 142( 91) 859 55 12 9 6 2246 9520
490 9 284 195 958 460 9 158 26 9 4 2313 925
a Data are the means and standard deviations of duplicate samples from two independent experiments except hEp at 1 and 10 ng/ml (range of duplicate samples in one experiment shown in parentheses). Background rates of spontaneous reversion subtracted in different trials ranged from 12–18 per plate. b In a parallel experiment performed in duplicate, Trp-P-2 at 12.5, 25 and 50 ng/tube gave 94 ( 92), 122 (9 1) and 154 (9 11) revertants/per plate, respectively. c Positive control microsomes obtained from Arochlor 1254-treated rat liver.
Table 4 Trp-P-2 adducts of DNA and protein in human and rat keratinocytes Cells
SIK hEp Early rEp Late rEp
DNA adductsa
Protein adductsa
−TCDD
+TCDD
−TCDD
+TCDD
1.39 0.4 2.09 0.9 2.09 0.6 0.99 0.9
3.7 91.4 5.7 93.3 6.7 94.7 1.2 90.3
17 94.3 19 9 1.7 25 9 7.1 18 95.9
23 95.7 25 93.0 37 94.8 19 93.8
a Values (ng Trp-P-2/mg) represent the mean 9 SD from three experiments performed in triplicate or duplicate. Nonspecific background values of DNA and protein adducts were 0.4 and 3, respectively. Except for late rEp, the adduct levels were higher in TCDD-treated cultures than in cultures not treated with TCDD (t-test, PB0.05). DNA adducts among SIK, hEp and early rEp were not significantly different (Duncan’s multiple range test PB0.05), but protein adducts were significantly higher for rEp than for SIK or hEp. DNA and protein adducts for rEp were significantly lower than for the others except for protein adducts in cultures not treated with TCDD (not different from SIK or hEp).
cultures treated with TCDD. The measured values of protein adducts in late passage rEp were not distinguishably lower than those from human epidermal cells.
4. Discussion To study responses of potential target cell types, keratinocytes are of particular interest in view of their ready cultivability, their expression of cytochrome P450
248
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
activities (CYP1A1, CYP1B1) found in many epithelial cell types, and the ability of these cells from different anatomic origins to display intrinsic differences in culture [21,22]. Both human and rat epidermal keratinocytes were sensitive to toxicity from Trp-P-1, but they exhibited strongly contrasting sensitivities to toxicity from Trp-P-2 and, in the rat, early passage epidermal cells were much more sensitive than keratinocytes from esophagus or fore-stomach. Although previous work has shown human epidermal and esophageal cells to have similar sensitivity to toxicity from polycyclic aromatic hydrocarbons [39], present experiments reveal a striking sensitivity of the esophageal cells to Trp-P-2 that merits further study, particularly if differential cytochrome P450 isozyme expression in human esophagus [40] and epidermis can be substantiated. Because the toxicity (including mutagenicity and carcinogenicity) of Trp-P-2, like many chemicals, is mediated by production of reactive metabolites, this process is a prime candidate for helping explain differences in response. In hepatocytes, for example, the cytotoxic response toward Trp-P-2 is correlated with cytochrome P450 activity [41], which is responsible for production of the cytotoxic and genotoxic 3-hydroxyamino metabolite, N – OH–Trp-P-2 [42]. Considerable differences in hepatic cytochrome P450-mediated metabolic activation of Trp-P-2 among species have been seen [43]. In the present work, the loss of sensitivity to toxicity of the late passage rEp cells is readily attributable to the loss of P450-mediated metabolism. Unlike other P450 activities, whose mechanisms of induction are less well understood, Ah receptor-induced CYP1A1 and 1B1 are commonly expressed in culture. The unusual suppression of their inducibility in rEp is known to prevent the toxic response of these cells to polycyclic aromatic hydrocarbons [39] and can occur in CYP1A1 by activation of a negative regulatory element in the gene 5%-flanking DNA [34]. That glutathione S-transferase activity was still inducible by TCDD indicates the silencing is gene-specific and not a general suppression of Ah receptor function. What has not been appreciated before is the rate with which the silencing occurs, largely within the first half dozen passages. Apparently starting almost immediately upon cultivation, and thus probably a physiological adaptation, the process appears nearly complete before the greatly elevated colony forming efficiencies characteristic of spontaneous immortalization become evident. As previously found [21], these were seen later, typically after a half dozen or more passages (data not shown). Differences in metabolite production could explain the relative insensitivity of rEs and rFst cells to Trp-P-2 toxicity, but not the contrast in sensitivity between early passage rEp and the human epidermal cells. The highest rate of N-hydroxylation reported for Trp-P-2 and other heterocyclic aromatic amines occurs with cytochrome P4501A2, but activation is catalyzed by cytochromes P4501A1 and 1B1 [16,44,45]. Although rEp and hEp could make different metabolites to which the respective cells have qualitatively different responses, this possibility appears remote. For example, expression by rEp of CYP1A2 could produce more toxic metabolites, but this isozyme seems poorly expressed or inducible outside the liver [16] and has not been found in mouse skin [46]. Moreover, no clear distinction was seen in mutagenicity of Trp-P-2 using microsomes from the human and rat sources.
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
249
On the other hand, unlike the TCDD stimulated toxicity of Trp-P-1 to hEp and SIK [17], the observed toxicity in rEp was independent of TCDD treatment, raising the possibility that metabolism was not required for this response. (Trp-P-1 and Trp-P-2 have been proposed to be toxic without metabolism to rat hepatocytes in primary culture, although at much higher concentrations than employed presently [47]. Other possibilities are that critical metabolites were produced by a constitutive activity to which the EROD assay is insensitive or that Trp-P-2 is a good inducer in the rEp. If cytochrome P450-mediated reactions cannot account for differential toxicity in these cultures, then conjugation reactions may be responsible. Glutathione conjugation is well known to provide a protective function toward aflatoxin B1, and benzo(a)pyrene toxicity, for example [19,48]. Glutathione S-transferase activity was considerably higher in the cytosolic fraction of the human than in the rat keratinocytes, possibly conferring upon the human cells a substantially higher detoxification ability. Glutathione S-transferase activities measured in pilot studies of TCDD-treated rEs and rFst cultures were also low, though 2–3 times higher than in rEp. Glutathione S-transferase also reportedly catalyzes the conjugation of N– OH – Trp-P-2 with glutathione to yield at least three conjugates, one of which is more mutagenic than N – OH – Trp-P-2 in vitro [49]. Whether this finding has relevance in vivo is unclear, but sulfhydryl compounds have been observed to reduce Trp-P-2 DNA adducts in mammalian cells [50]. Acetylation can be a key activation or deactivation reaction in the biotransformation of heterocyclic amines [51]. Activation involves O-acetylation of the N– OH-compound or N,O-acetyl transfer in the N–OH–N-acetylated compound. The result is an unstable N-acetoxyarylamine that decomposes spontaneously to give a nitrenium ion that binds covalently to DNA and protein [52]. Differences among human hepatoma cell lines in heterocyclic amine genotoxicity have been attributed to differences in cytosolic acetyltransferase activity [53]. However, Trp-P-2 (unlike Trp-P-1) appears not to be dependent upon O-acetylation for its genotoxic activation [51,54,55]. Moroever, since in bacterial mutagenesis assays N-acetyl-Trp-P-2 exhibits only a small fraction of the activity of Trp-P-2 [21], N-acetylation could be a protective step. In humans and several animal species, the acetyltransferases are encoded by two genetic loci, designated NAT1 and NAT2 [56]. Marked species variations, capable of modulating carcinogen susceptibilities, have been noted in N-acetylation of heterocyclic aromatic amines [57]. In the present experiments, although not definitive, the lack of detectable NAT2 activity suggests this isozyme does not contribute to differential toxicity. Other studies have detected little NAT2 activity in the rat [58]. NAT1 activity was slightly higher in the cytosol from human epidermal cells than from rEp, but not enough to constitute compelling evidence of much protection by N-acetylation in this case. In general, toxic effects of agents such as Trp-P-2 reflect a balance between metabolic activation and detoxification reactions. Since differences in sensitivity among human and rat keratinocyte cultures toward Trp-P-2 toxicity cannot clearly be attributed to differences in generation of activated metabolites, suspicion rests more heavily upon detoxification reactions, especially glutathione conjugation. If
250
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
Trp-P-2, unlike Trp-P-1 [17], is not mutagenic in human epidermal cells, this could be indicative of selective metabolite detoxification. The lack of significant difference over four days of treatment in DNA adduct accumulation between hEp and early passage rEp suggests DNA repair differences are not distinctive. However, differences in comparative radical production or scavenging [59], arachidonic acid-dependent peroxidation [60], bioactivation of N–OH-Trp-P-2 by esterification pathways other than O-acetylation [61], or enzymatic reduction of the N-hydroxy derivative [62] could be important factors in the relative toxicity of Trp-P-2.
Acknowledgements This work was supported by USPHS Grants AR27130, ES05705, ES08079 and ES04699 and a fellowship (to H.-S. C.) from the Republic of Korea.
References [1] T. Sugimura, T. Kawachi, M. Nagao, et al., Mutagenic principle(s) in tryptophan and phenylalanine pyrolysis products, Proc. Jpn. Acad. 53 (1977) 58 – 61. [2] M. Nagao, Y. Takahashi, T. Yahagi, T. Sugimura, K. Takeda, K. Shudo, T. Okamoto, Mutagenicities of g-carboline derivatives related to potent mutagens found in tryptophan pyrolysates, Carcinogenesis 1 (1980) 451–454. [3] K.M. Brookman, E.P. Salazar, L.H. Thompson, Comparative mutagenic efficiencies of the DNA adducts from the cooked-food-related mutagens Trp-P-2 and IQ in CHO cells, Mut. Res. 149 (1985) 249–255. [4] Y.F. Sasaki, A. Saga, M. Akasaki, et al., In vivo genotoxicity of heterocyclic amines detected by a modified alkaline single cell gel electrophoresis assay in a multiple organ study in the mouse, Mut. Res. 395 (1997) 57–73. [5] Y.F. Sasaki, A. Saga, K. Yoshida, et al., Colon specific genotoxicity of heterocyclic amines detected by the modified alkaline single cell gel electrophoresis assay of multiple mouse organs, Mut. Res. 414 (1998) 9–14. [6] T. Sugimura, Carcinogenicity of mutagenic heterocyclic amines formed during the cooking process, Mut. Res. 150 (1985) 33–41. [7] K. Fujii, Evaluation of the newborn mouse model for chemical tumorigenesis, Carcinogenesis 12 (1991) 1409–1415. [8] Z. Yamaizumi, T. Shiomi, H. Kasai, S. Nishimura, Y. Takahashi, M. Nagao, T. Sugimura, Detection of potent mutagens, Trp-P-1 and Trp-P-2, in broiled fish, Cancer Lett. 9 (1980) 75 – 83. [9] K. Skog, A. Solyakov, P. Arvidsson, M. Jagerstad, Analysis of nonpolar heterocyclic amines in cooked foods and meat extracts using gas chromatography-mass spectrometry, J. Chromatog. 803 (1998) 227–233. [10] S. Manabe, O. Wada, Carcinogenic tryptophan pyrolysis products in cigarette smoke condensate and cigarette smoke-polluted indoor air, Environ. Poll. 64 (1990) 121 – 132. [11] K. Wakabayashi, I.S. Kim, R. Kurosaka, et al., Identification of new mutagenic heterocyclic amines and quantification of known heterocyclic amines, Princess Takamatsu Sym. 23 (1995) 39 – 49. [12] T. Ohe, Quantification of mutagenic/carcinogenic heterocyclic amines, MeIQx, Trp-P-1, Trp-P-2 and PhIP, contributing highly to genotoxicity of river water, Mut. Res. 393 (1997) 73 – 79. [13] S. Manabe, E. Uchino, O. Wada, Carcinogenic tryptophan pyrolysis products in airborne particles and rain water, Mut. Res. 226 (1989) 215 – 221.
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
251
[14] R. Hasegawa, T. Kato, M. Hirose, S. Takahashi, T. Shirai, N. Ito, Enhancement of hepatocarcinogenesis by combined administration of food-derived heterocyclic amines at low doses in the rat, Food Chem. Toxicol. 34 (1996) 1097 – 1110. [15] J.A. Beamond, P.T. Barton, J.M. Tredger, R.J. Proce, B.G. Lake, Effect of some cooked food mutagens on unscheduled DNA synthesis in cultured precision-cut rat, mouse and human liver slices, Food Chem. Toxicol. 36 (1998) 455 – 466. [16] T. Shimada, C.L. Hayes, H. Yamazaki, S. Amin, S.S. Hecht, F.P. Guengerich, T.R. Sutter, Activation of chemically diverse procarcinogen by human cytochrome P-450 1B1, Cancer Res. 56 (1996) 2979–2984. [17] A.A. Walsh, L.A. deGraffenried, R.H. Rice, 2,3,7,8-Tetrachlorodibenzo-p-dioxin sensitization of cultured human epidermal cells to toxicity from a carcinogenic heterocyclic amine, Carcinogenesis 16 (1995) 2187–2191. [18] H.S. Ramsdell, D.L. Eaton, Species susceptibility to aflatoxin B1 carcinogenesis: comparative kinetics of microsomal biotransformation, Cancer Res. 50 (1990) 615 – 620. [19] G.H. Degen, H.G. Neumann, Differences in aflatoxin B1-susceptibility of rat and mouse are correlated with the capability in vitro to inactivate aflatoxin B1-epoxide, Carcinogenesis 2 (1981) 299–306. [20] B.L. Allen-Hoffman, L.G. Rheinwald, Polycyclic aromatic hydrocarbon mutagenesis of human epidermal keratinocytes in culture, Proc. Nad. Acad. Sci. USA 81 (1984) 7802 – 7806. [21] R. Heimann, R.H. Rice, Rat esophageal and epidermal keratinocytes: intrinsic differences in culture and derivation of continuous lines, J. Cell. Physiol. 117 (1983) 362 – 367. [22] M.A. Phillips, R.H. Rice, Convergent differentiation in cultured rat cells from nonkeratinizing epithelia: keratinocyte character and intrinsic differences, J. Cell Biol. 97 (1983) 686 – 691. [23] J.G. Rheinwald, H. Green, Formation of a keratinizing epithelium by a cloned cell line derived from a teratoma, Cell 6 (1975) 317– 330. [24] T.-T. Sun, H. Green, Differentiation of the epidermal keratinocyte in cell culture: formation of the cornified envelope, Cell 9 (1976) 511 – 521. [25] F.H. Ausubel, R. Brent, R. Kingston, et al., DNA protein interactions, Current Protocols in Molecular Biology, 2, John Wiley & Sons, New York, 1993, p. 12. [26] P.K. Smith, R.I. Krohn, G.T. Hermanson, et al., Measurement of protein using bicinchoninic acid, Analyt. Biochem. 150 (1985) 76–85. [27] J.J. Reiners Jr., A.R. Cantu, A. Pavone, Modulation of constitutive cytochrome P450 expression in vivo and in vitro in murine keratinocytes as a function of differentiation and extracellular Ca2 + concentration, Proc. Nad. Acad. Sci. USA 87 (1990) 1825 – 1829. [28] H.H. Andres, A.J. Klem, S.M. Szabo, W.W. Weber, New spectrophotometric and radiochemical assays for acetyl-CoA:arylamine N-acetyltransferase applicable to a variety of arylamines, Analyt. Biochem. 145 (1985) 367–375. [29] W.H. Habig, M.J. Pabst, W.B. Jacoby, Glutathione-S-transferase: the first enzymatic step in mercapturic acid formation, J. Biol. Chem. 249 (1974) 7130 – 7139. [30] A.P. Feinberg, B. Vogelstein, A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity, Analyt. Biochem. 132 (1983) 6 – 13. [31] N.Y. Kado, D. Langley, E. Eisenstadt, A simple modification of the Salmonella liquid incubation assay: increased sensitivity for detecting mutagens in human urine, Mut. Res. 121 (1983) 25 – 32. [32] B.N. Ames, J. McCann, E. Yamasaki, Methods for detecting carcinogens and mutagens with the Salmonella/mammalian-microsome mutagenicity test, Mut. Res. 31 (1975) 347 – 364. [33] D.J. Loury, N.Y. Kado, L.L. Byard, Enhancement of hepatocellular genotoxicity of several mutagens from amino acid pyrolysates and broiled foods following ethanol pretreatment, Food Chem. Toxicol. 23 (1985) 661–667. [34] A.A. Walsh, K. Tullis, R.H. Rice, M.S. Denison, Identification of a novel cis-acting negative regulatory element affecting transcription of the CYP1A1 gene in rat epidermal cells, J. Biol. Chem. 271 (1996) 22746–22753. [35] H. Raza, Y.C. Awasthi, M.T. Zaim, R.L. Eckert, H. Mukhtar, Glutathione S-transferases in human and rodent skin: multiple forms and species-specific expression, J. Invest. Dermatol. 96 (1991) 463–467.
252
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
[36] J.D. Hayes, D.J. Pulford, The glutathione S-transferase supergene family: regulation of GST and the contribution of the isozymes to cancer chemoprotection and drug resistance, Crit. Rev. Biochem. Mol. Biol. 30 (1995) 445– 446. [37] D.L. Eaton, T.K. Bamniler, Concise review of the glutathione S-transferases and their significance to toxicology, Toxicol. Sci. 49 (1999) 156 – 164. [38] S.H. Safe, Modulation of gene expression and endocrine response pathways by 2,3,7,8 tetrachlorodibeno-p-dioxin and related compounds, Pharmacol. Therap. 67 (1995) 247 – 281. [39] R. Heimann, R.H. Rice, Polycyclic aromatic hydrocarbon toxicity and induction of metabolism in cultivated esophageal and epidermal keratinocytes, Cancer Res. 43 (1983) 4856 – 4862. [40] M. Lechevrel, A.G. Casson, C.R. Wolf, et al., Characterization of cytochrome P450 expression in human esophageal mucosa, Carcinogensis 20 (1999) 243 – 248. [41] T. Segawa, H. Ishiga, H. Ueno, K. Nakamuro, Y. Sayato, Genotoxicity and cytotoxicity of 3-amino-1-methyl-5H-pyrido[4,3-b] lindole (Trp-P-2), Chemosphere 28 (1994) 853 – 861. [42] Y. Yamazoe, K. Ishii, T. Kamataki, R. Kato, T. Sugimura, Isolation and characterization of active metabolites of tryptophan-pyrolysate mutagen, Trp-P-2, formed by rat liver microsomes, Chem.Biol. Interact. 30 (1980) 125–138. [43] Y. Yamazoe, T. Kamataki, R. Kato, Species difference in N-hydroxylation of a tryptophan pyrolysis product in relation to mutagenic activation, Cancer Res. 41 (1981) 4518 – 4522. [44] T. Aoyama, F.J. Gonzalez, H.V. Gelboin, Human cDNA-expressed cytochrome P4501A2: mutagen activation and substrate specificity, Mol. Carcinogen. 2 (1989) 192 – 198. [45] M.E. McManus, W.M. Burgess, M.E. Veronese, A. Huggett, L.C. Quattrochi, R.H. Tukey, Metabolism of 2-acetylaminofluorine and benzo(a)pyrene and activation of food-derived heterocyclic amine mutagens by human cytochromes, P-450, Cancer Res. 50 (1990) 3367 – 3376. [46] Y. Shimizu, Y. Nakatsuru, M. Ichinose, et al., Benzo(a)pyrene carcinogenicity is lost in mice lacking the aryl hydrocarbon receptor, Proc. Nad. Acad. Sci. USA 97 (2000) 779 – 782. [47] H. Ashida, B. Shiotani, H. Adachi, T. Hashimoto, K. Kanazawa, G. Danno, Tryptophan pyrolysis products, Trp-P-1 and Trp-P-2 induce apoptosis in primary cultured rat hepatocytes, Biosci. Biotechnol. Biochem. 62 (1998) 2283 – 2287. [48] D. Ho, W.E. Fahi, Modification of glutathione levels in C3H/10T1/2 cells and its relationship to benzo(a)pyrene anti-7,8-dihydrodiol-9,10-epoxide-induced cytotoxicity, J. Biol. Chem. 259 (1984) 11231–11235. [49] K. Saito, Y. Yamazoe, T. Kamataki, R. Kato, Glutathione transferase-mediated and non enzymatic activation and detoxification of the N-hydroxy derivative of Trp-P-2, a potent pyrolysate proMutagen, Xenobiotica 14 (1984) 545–548. [50] S. Mita, Y. Yarnazoe, T. Kamataki, R. Kato, Metabolic activation of Trp-P-2, a tryptophan pyrolysis mutagen, by isolated rat hepatocytes, Biochem. Pharmacol. 32 (1983) 1179 – 1182. [51] D. Wild, W. Feser, S. Michel, H.L. Lord, P.D. Josephy, Metabolic activation of heterocyclic aromatic amines catalyzed by human arylamine N-acetyltransferase isoenzymes (NAT1 and NAT2) expressed in Salmonella typhimurium, Carcinogenesis 16 (1995) 643 – 648. [52] W.W. Weber, D.W. Hein, N-Acetylation pharmacogenetics, Pharmacol. Rev. 37 (1985) 25 – 79. [53] S. Knasmuller, C. Schwab, S. Land, et al., Genotoxic effects of heterocyclic aromatic amines in human derived hepatorna (HepG2) cells, Mutagenesis 14 (1999) 533 – 539. [54] Y. Hashimoto, K. Shudo, T. OkaMoto, Activation of a mutagen, 3-amino-1-methyl-5H pyrido[4,3b]indole. Identification of 3-hydroxyamino-1-methyl-5H-pyrido[4,3-b]indole and its reaction with DNA, Biochem. Biophys. Res. Commun. 96 (1980) 355 – 362. [55] Y. Oda, H. Yamazaki, M. Watanabe, T. Nohmi, T. Shimada, Development of high sensitive umu test system: rapid detection of genotoxicity of promutagenic aromatic amines by Salmonella typhimurium strain NM2009 possessing high O-acetyltransferase activity, Mut. Res. 334 (1995) 145–156. [56] M.A. Doll, D.W. Hein, Cloning, sequencing and expression of NAT1 and NAT2 encoding genes from rapid and slow acetylator inbred rats, Pharmacogenetics 5 (1995) 247 – 251. [57] A. Shinohara, Y. Yamazoe, K. Saito, T. Kamataki, R. Kato, Species differences in the N-acetylation by liver cytosol of mutagenic heterocyclic aromatic amines in protein pyrolysates, Carcinogenesis 5 (1984) 683–686.
H.-S. Chun et al. / Chemico-Biological Interactions 127 (2000) 237–253
253
[58] R.F. Jones, S.J. Land, C.M. King, Recombinant rat and hamster N-acetyltransferase-l and-2: relative rates of N-acetylation of arylamines and N,O-acyltransfer with arylhydroxamic acids, Carcinogenesis 17 (1996) 1729–1733. [59] J.J. Reiners Jr., T. Rupp, Conversion of xanthine dehydrogenase to xanthine oxidase occurs during keratinocyte differentiation: modulation by 12-O-tetradecanoylphorbol-13-acetate, J. Invest. Dermatol. 93 (1989) 132–135. [60] T. Flammang, Y. Yamazoe, R.W. Benison, et al., Arachidonic acid-dependent peroxidative activation of carcinogenic arylamines by extrahepatic human tissue microsomes, Cancer Res. 49 (1989) 1977–1982. [61] H.A.J. Schut, E.G. Snyderwine, DNA adducts of heterocyclic amine food mutagens: implications for mutagenesis and carcinogenesis, Carcinogenesis 20 (1999) 353 – 368. [62] R.S. King, C.H. Teitel, L.G. Shaddock, D.A. Casciano, F.F. Kadlubar, Detoxification of carcinogenic aromatic amines and heterocyclic amines by enzymatic reduction of the N-hydroxy derivative, Cancer Lett. 143 (1999) 167–171.
.