Environ. Sci. Technol. 2001, 35, 54-62
Cytochrome P450 1A Expression in Midwater Fishes: Potential Effects of Chemical Contaminants in Remote Oceanic Zones J O H N J . S T E G E M A N , * ,† JENNIFER J. SCHLEZINGER,† JAMES E. CRADDOCK,† AND DONALD E. TILLITT‡ Biology Department, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543, and Columbia Environmental Research Center, U.S. Geological Survey, Columbia, Missouri 65201
Cytochrome P450 1A (CYP1A) induction is a robust marker for exposure to polynuclear aromatic hydrocarbons and planar halogenated aromatic hydrocarbons that are aryl hydrocarbon receptor agonists. We examined CYP1A expression in mesopelagic fishes from the western North Atlantic. Individuals in 22 species were obtained from slope water and the Sargasso Sea in 1977, 1978, and 1993. Aryl hydrocarbon hydroxylase (AHH), a CYP1A activity, was detected in liver from all species in 1977/78. In some, including Gonostoma elongatum, AHH was inhibited by the CYP1A inhibitor R-naphthoflavone. CYP1A-dependent ethoxyresorufin O-deethylase (EROD) was detected in liver microsomes of all species in 1993; rates were highest in G. elongatum and Argyropelecus aculeatus. Immunoblot analysis with the CYP1A-specific monoclonal antibody 1-12-3 detected a single microsomal protein band in most 1993 samples; the highest content was in G. elongatum. Immunohistochemical analysis showed CYP1A staining in gill, heart, kidney, and/or liver of several species. Extracts of the 1993 G. elongatum and A. aculeatus, when applied to fish hepatoma cells (PLHC-1) in culture, elicited a significant induction of EROD in those cells. The capacity of the extracts to induce CYP1A correlated with the content of PCBs measured in the same fish (2-4.6 ng/g total body weight). Mesopelagic fish in the western North Atlantic, which experience no direct exposure to surface waters or sediments, are exposed chronically to inducers of CYP1A at levels that appear to be biochemically active in those fish.
Introduction Polynuclear aromatic hydrocarbons (PAH), pesticides, halogenated aromatic hydrocarbons (HAH), and other persistent organic pollutants are ubiquitous in the global environment. PAH and planar HAH (PHAH), such as 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and polychlorinated biphenyls (PCB) congeners bearing meta- and para- but no orthochlorine substitutions, are of particular concern. High-level exposure to such compounds in the environment has been * Corresponding author telephone: (508)289-2320; fax: (508)4572169; e-mail:
[email protected]. † Woods Hole Oceanographic Institution. ‡ Columbia Environmental Research Center. 54
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 1, 2001
associated with overt pathologies including carcinomas (1-3), reproductive toxicity, and/or developmental defects (4) in fish and other wildlife, and similar pathologies occur in animals experimentally treated with PAH or PHAH (2, 5). In many regions of the global aquatic ecosystem, the concentrations of such contaminants are orders of magnitude less than in highly industrialized bays and harbors. A major question is whether contaminants in such regions are present at levels able to affect biota. One approach to that question is to determine whether organisms in remote regions show molecular changes known to be associated with chemical exposure and that may be early indicators of effect. PAH and PHAH induce cytochrome P450 1A (CYP1A) gene expression by binding to the aryl hydrocarbon receptor (AHR) transcription factor (6). AHR and CYP1A have been described in all vertebrate groups from fish to humans (7, 8). CYP1As metabolize many PAH and PHAH, and both the AHR and CYP1A may mediate various toxic effects of these chemicals (9, 10). Treatment of animals with known inducers results in dose-dependent increases of CYP1A expression in liver and other organs (11, 12), reflecting AHR activation. Vertebrate animals from environments largely free of pollutant inducers show little or no detectable CYP1A in liver, while animals from contaminated sites express elevated levels of CYP1A (13-16). The levels of PAHs and/or PHAH in coastal marine environments or in tissues of fish often are correlated closely with the content of CYP1A in the liver (17, 18). Elevated levels of hepatic CYP1A have been correlated with PCB content in fish also in the deep ocean (19) in Coryphaenoides armatus (rattail), a deep demersal species living close to and feeding on the bottom. Contaminants in gaseous and particulate phases are delivered through the atmosphere to the open ocean from continental and coastal environments or may come directly from ship traffic (20-22). The mesopelagic zone of the water column, between 250 and 1500 m depth, receives a transient flux of contaminants associated with particulate matter from the surface (23). However, most particles pass through the midwater, and sorption properties favor retention of hydrophobic contaminants on particles (24, 25). Thus, contaminant concentrations in the mesopelagic zone might be among the lowest on earth, and mesopelagic fishes in regions distant from continental sources of contaminants may experience less exposure to toxic chemicals than that occurring in other regions of the world ocean. We addressed the question of whether fish in the midwater express detectable levels of CYP1A and whether this might be associated with exposure to environmental pollutants. We examined numerous species of midwater fish from the western North Atlantic, sampled on both sides of the Gulf Stream and at times 15-16 years apart. CYP1A expression was detected in many of the fish examined, and the data indicate that in some species the expression was induced by exogenous chemicals.
Materials and Methods Chemicals. Benzo[a]pyrene (Gold Label) was purchased from the Aldrich Chemical Co. (Milwaukee, WI), and 7-ethoxyresorufin was from Molecular Probes (Eugene, OR). All other reagents were purchased from Sigma Chemical Co. (St. Louis, MO). Sample Collection. Members of 22 species of fish were collected from the North Atlantic on three cruises; R/V Knorr cruise 65 in April/May 1977, R/V Oceanus cruise 43 in April 1978, and R/V Columbus Iselin cruise 9307 in July/August 1993 (Figure 1). On Knorr and Iselin, fish were collected using 10.1021/es0012265 CCC: $20.00
2001 American Chemical Society Published on Web 11/23/2000
FIGURE 1. Midwater fish sampling sites. Knorr 65 sites were in the vicinity of 38°20′ N, 67°30′ W (slope), 36°38′ N, 69°27′ W (cold core ring Bob), and 34°20′ N, 71°23′ W (Sargasso Sea). Details on Knorr locations are in ref 27. On Oceanus 43, samples were obtained along a line from 36°50′ N, 70°36′ W to 39°46′ N, 71°00′ W. On Iselin cruise 9307, station 1 was in the vicinity of 38°22′-38°29′ N and 71°58′-72°18′ W in the slope water, and station 2 was at 36°8′ N, 73°26′ W in the Sargasso Sea. For reference, the Deep Water Dump Site 106 is in the vicinity of 38°49′ N and 71°55′ W. a MOCNESS-10 net system, which allows an accurate determination of the depth of capture and time between capture and retrieval of organisms. Fish were captured between 300 and 1500 m below the surface and 2000-3500 m above the bottom. On the Oceanus cruise, midwater fish were captured incidental to bottom trawling for deep benthic fish (26). In 1977, 87 fish representing 16 species were obtained from the slope water, from the Sargasso Sea, and from within Cold-Core Gulf Stream Ring Bob (27). Surface temperatures in the ring and slope water were between 15 and 17° C, and the 10° isotherm was shallow, between 200 and 300 m. Surface temperatures in the Sargasso Sea were between 20 and 25° C, and the 10° isotherm was at 500-1000 m. Fish taken inside ring Bob or in the slope water were alive on retrieval, indicated by active movement and/or beating heart. Fewer fish retrieved through warmer Sargasso Sea waters showed those signs of life, although retrieval times were similar. In 1978, 51 midwater fish of 10 species were obtained, all in the slope water (Figure 1). Surface waters were 6-7 °C, and fish were alive at retrieval. Samples obtained in 1993 included 121 fish in 9 species from two stations, station 1 in the slope water and station 2 in the Sargasso Sea (Figure 1); most were alive. After life signs and identity were determined, fish were placed on ice and within 1 h were further dissected and samples were prepared or archived. In 1977-1978, some livers were prepared and assayed immediately for enzyme activity. Liver mass was estimated by volume displacement adjusted for liver density determined with tissue that was frozen and later weighed on shore. Other liver samples were frozen whole for lipid analysis or were fixed in formalin or glutaraldehyde for later histology and electron microscopy. Gonads and carcasses were frozen at -20 °C for weight measurement on shore. Fish collected in 1993 were dissected, and livers were archived immediately in liquid N2. Further preparation and analysis took place on land within 2 months. Carcasses of
fish from 1993 were placed in solvent-cleaned foil, frozen in liquid N2, and archived at -80 °C for measurement of body weight and, in some cases, analysis of chemical residues. Smaller individuals were fixed whole in 10% neutral-buffered formalin for immunohistochemistry. Sample Preparation. Subcellular fractions were prepared from samples in 1977 and 1978 using modifications necessary due to the inability of centrifuges aboard ship to operate at speeds sufficient to prepare microsomes. Tissue from smaller individuals of the same species from a single trawl were pooled. Liver was homogenized in 4 vol of ice-cold buffer (0.15 M KCl and 0.05 M Tris, pH 7.4) and centrifuged at 3000 rpm for 25 min, sedimenting a nuclear fraction (nuclei and larger particles). The supernatant was removed carefully from under the lipid pellicle and then filtered to 30 µm with a Millipore syringe filter assembly. Assay of the mitochondrial enzyme succinate dehydrogenase and the microsomal enzymes indicated that the filtration effectively removed mitochondrial but not microsomal particles, resulting in a filtrate that we refer to as a “post-mitochondrial filtrate” or PMF, equivalent to a post-mitochondrial supernatant (or PMS). Freshly prepared PMF was analyzed immediately for enzyme activity. Any remaining PMF was archived in liquid N2. Upon return to Woods Hole, microsomes were prepared from some of these PMF and were analyzed for total P450 content. Livers returned to Woods Hole in 1993 were thawed and homogenized, and microsomal fractions were prepared by differential centrifugation (28). Microsomal pellets were resuspended in 50 mM Tris, pH 7.4, 1 mM dithiothreitol, 1 mM EDTA, and 20% glycerol, and held in liquid N2 until assay. Enzyme Assays. Rates of benzo[a]pyrene hydroxylase (aryl hydrocarbon hydroxylase, AHH) and aminopyrine N-demethylase (ApND) activities were measured on the 19771978 cruises as before (28). Samples were assayed in duplicate using an NADPH generating system and incubation at 27 °C for 30 (AHH) or 60 min (ApND). Reaction products were detected with a Turner Fluorometer or a Beckman DU spectrophotometer. Replicate blank reactions had no enzyme or no NADPH-generating system added. With samples having sufficient volume (i.e., from Gonostoma elongatum, Stomias boa, and Scopelogadus beanii), rates of AHH were determined to be linear with respect to PMF protein, incubation time, and substrate (B[a]P) concentration. AHH activity was examined for sensitivity to 100 µM R-naphthoflavone (ANF), an inhibitor of fish CYP1As (29). NADPH- and NADHdependent cytochrome c reductases (P450 and b5 reductase, respectively) and total P450 content were measured spectrophotometrically (28). The 1993 samples were assayed for 7-ethoxyresorufin O-deethylase (EROD) activity using a Cytofluor 2300 (Millipore) multiwell plate reader (30). Protein content was determined by either of two methods (31, 32) with bovine serum albumin as a standard. Lipid content of 1977 samples was measured by extracting liver with hexane:petroleum ether (1:1); the solvent was evaporated, and neutral lipid content was measured gravimetrically. Western Blotting for CYP1A. Immunoblotting procedures were modified from those previously described (33), using SDS-PAGE on 8-16% gradient gels and electrophoretic transfer to 0.05 µm of nitrocellulose. Antibodies were monoclonal antibody 1-12-3 to scup CYP1A (34) and alkaline phosphatase- or peroxidase-linked goat-anti-mouse IgG (Amersham). Immunoreactive bands were visualized by colorimetric staining or by enhanced chemiluminescence (Amersham), with staining quantified by densitometric analysis (NIH Image). CYP1A content was estimated by comparison to known amounts of CYP1A and is reported as scup CYP1A equiv. VOL. 35, NO. 1, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
55
TABLE 1. Charateristics of Midwater Fishes Examined species
common diet
vertical migrator?
cruisea
N
body wt range (g)
liver/body ratio (%)
mg of microsomal protein/g of liverb
liver lipid (%)c
Anoplogaster cornuta Argyropelecus aculeatus Argyropelecus hemigymnus Bathylagus sp. Benthosema glaciale Chauliodus sloani Dyssoma sp. Diaphus effulgens Gonostoma elongatum Gonichthys cocco Lampanyctus cuprarius Lampanyctus macdonaldi* Lampedema speculigera Lepidophanes guentheri Malacosteus niger Melanostomias sp. Nemichthys scolopaceus Scopelogadus beanii Serrivomer beani Sternoptyx diaphana Stomias boa
fish, crustacea crustacea crustacea jellies, crustacea crustacea mostly fish mostly fish crustacea crustacea crustacea crustacea crustacea crustacea crustacea jellies, crustacea fish, crustacea crustacea jellies, crustacea fish, crustacea squid, crustacea fish, crustacea
no yes yes no yes yes no yes yes yes yes yes yes yes no yes no no no no yes
K K,I I K O,I K,O,I O K K,O,I I K,O,I K K I K,O K O K,O K,O I K,O
4 17 2 2 64 13 1 2 32 1 5 3 2 6 3 2 2 11 6 6 3
36-71 1.9-12.6 0.7, 1.5 8.3, 29.0 1.5-2.4 15.3-99.2 7.3 15.0, 6.8 3.7-58.3 1.5 2.5-8.2 18.5-20.5 30.0, 35.0 1.7-3.1 24-48.5 3.2, 3.4 13.2-15.9 15.7-34.4 19-140 1.2-2.3 15.5-50.0
0.75 ( 0.05 2.67 ( 1.28
6.6 ( 2.4
18.6 5.9
7.4 ( 3.0 6.8 ( 1.4
22.5
0.9 0.75 ( 0.62 1.02 ( 0.40 1.2 1.0 0.75 ( 0.27 4.0 0.62 ( 0.23 1.05 ( 0.10 1.0 0.97 ( 0.33 1.05 ( 0.17 0.4 0.98 ( 0.34 0.75 ( 0.31 1.55 ( 0.96 1.65
8.1 ( 1.8 11.7 3.7 ( 0.6 8.4 ( 0.9
17.3 34.0 23
6.4 ( 4.1 20.6
a
K, O, and I refer to Knorr (1977), Oceanus (1978), and Iselin (1993) cruise collections, respectively. Additional individuals of some species were used for immunohistochemistry or for analytical chemical analysis and are not represented in this table. b Microsomal yield was determined only for samples from Iselin (1993). c Lipid content was determined only for samples from Knorr (1977).
Immunohistochemical Detection of CYP1A Expression. The procedures used were those of Smolowitz et al. (35). Whole fish were decalcified and embedded in paraffin, and sections (5 µm) were mounted on Fisher Superfrost Plus slides. One section was stained with H and E. Unstained sections were deparaffinated and rehydrated as before (36) and stained using an indirect peroxidase stain (Universal IP Staining Kit (Murine), Signet Lab, Dedham, MA). Primary antibodies were MAb 1-12-3 or the nonspecific UPC-10. Secondary antibody was peroxidase-linked goat anti-mouse IgG. Prevalence and intensity of cell staining were scored as before (36). Tissue Extracts for Bioassay and Chemical Analysis. Carcasses of A. aculeatus and G. elongatum from 1993 were homogenized and dried with anhydrous sodium sulfate at a weight four times the sample weight (37). Dried homogenates were extracted in an open glass column with methylene chloride. A portion (1%) of the extract was used for lipid determination. Portions of extracts destined for bioassay in PLHC-1 cells were purified by using a sulfuric acid silica gel/potassium silicate column and then a sulfuric acid silica gel/silica gel column. Extracts were transferred to isooctane and brought to a volume of 40 µL. PLHC-1 Bioassay Methods. The PLHC-1 bioassay procedures were a slight modification of the H4IIE bioassay methods (38). The PLHC-1 cells were seeded at 20 000 cells/ well in 300 µL of D-MEM culture media (37) in 96-well microtiter plates. After 24 h, sample extracts or standards in 5 µL of isooctane were added to each well. Cells were exposed to eight different doses of the samples in a 25% dilution series, with six replicates at each dose; responses were calibrated against responses to TCDD given at eight concentrations (0, 0.069, 0.206, 0.617, 1.85, 5.6, 16.7, and 50 pg/ well), each replicated four times. Plates were incubated for 72 h at 30 °C, the wells were washed 3× with ultrapure water at 2-s intervals to lyse the cells, and then 20 µL of Trissucrose (0.05-0.2 M) with dicumarol (20 µM final concentration) and 20 µL of 5 µM 7-ethoxyresorufin (0.5 µM final concentration) were added to each well. Reactions were initiated with 10 µL of 10 mM NADPH, and resorufin was quantified using a Cytofluor 2300 and a resorufin standard curve. Cellular protein was measured in each well, and the 56
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 1, 2001
doses of each sample (g equiv of sample/mg of protein) or TCDD standards (pg of TCDD/mg of protein) were plotted against EROD (pmol min-1 (mg of protein)-1) to develop dose-response curves indicating the potencies of the samples relative to TCDD. Six TCDD curves were analyzed, and the potency of extracts at inducing EROD in PLHC cells is defined as the TCDD equivalent for induction or TCDD-EQ. The determination of TCDD-equivalents (TCDD-EQs) was by slope ratio assay (39). Variance estimates were based on an additive model (39) and were calculated as before (37, 40). The limit of detection was approximately 0.5 pg of TCDDEQ/g wet weight basis. PCB Determinations. Portions of the organic extracts prepared above were used to determine PCB concentrations by enzyme-linked immunosorbent assay (ELISA) (Ohmicron Corp., Newton, PA) as previously described (41). The limit of detection for the PCB ELISA analysis was approximately 1 ng of total PCB/g of tissue wet weight, based on a calibration against an Aroclor 1254 technical standard. Differences in relative binding affinities of the antibodies toward various congeners can cause a differential response in the ELISA if the PCB patterns are substantially different. However, analysis of fish from across the United States provided a strong positive correlation (r 2 ) 0.75) of ELISA to total PCBs (by GC) when calibrated against Aroclor 1254 (42). The midwater fish samples presumably would be contaminated from long-range transport of chemicals, and the differences in PCB congener patterns are not expected to be great. Statistics were calculated using Microsoft Excel (Microsoft, Inc., Redmond, WA) and SuperAnova for Macintosh (Abacus Concepts, Inc., Berkeley, CA). A Student t-test was used to test differences between groups from the two stations in 1993. A one-factor ANOVA in combination with the Tukey-Kramer multiple-comparisons tests was used to analyze all other data.
Results Species and numbers of individuals sampled in the 1977, 1978, and 1993 collections are listed in Table 1, along with body weights, liver/body weight ratios, microsomal protein per gram of liver, and liver lipid content. Fish from different
TABLE 2. Rates of Monooxygenase Activities in Liver of Midwater Fish
species
N (n) a
Anoplogaster cornuta Bathylagus sp. Benthosema glaciale Chauliodus sloani Diaphus effulgens Dysomma sp. Gonostoma elongatum Lampanyctus macdonaldi Lampedena speculigera Malacosteus niger Melanostomias sp. Nemicthys scolopaceus Scopelogadus beanii Serrivomer beani Stomias boa
2 1 3 (38) 2 (4) 1 (2) 1 7 3 1 3 2 1 (2) 5 (11) 3 2
aryl hydrocarbon (B[a]P) aminopyrine hydroxylase N-demethylase (units/mg (units/mg of of protein)b protein)c 0.4, 1.5 33.9 3.3 ( 2.6 0, 0.4 3.0 2.9 10.0 ( 8.9 0.4 ( 0.4 5.3 1.9 ( 1.0 0.7, 3.1 1.9 6.1 ( 5.9 15.4 ( 12 2.1, 0.4
29 21 6 13.8 35 25-79
12-44 5 26
a N refers to the number of individual or pooled samples examined, and (n) indicates the total number of individuals from which tissue was pooled. b Units are pmol min-1 (mg of post-mitochondrial filtrate protein)-1. c Units are nmol h-1 (mg of post-mitochondrial supernatant protein)-1. Only one or two samples of the indicated species were analyzed for aminopyrine N-demethylase.
cruises had similar weight ranges and liver/body weight ratios, and data from all cruises are combined in Table 1. Liver lipid content determined with fish collected on Knorr 65 exceeded 10% for most species (Table 1). Histological examination revealed abundant lipid vacuoles in hepatocytes, consistent with high lipid content (not shown). Gonad/body weight ratios were less than 0.02 in all but four individuals, and the sex could be determined for only half of the individuals collected. There were no sex differences evident in any of the activities measured (below), and the data for individuals within a species were combined. 1977-1978 Samples. Hepatic post-mitochondrial filtrates (PMF) from all species examined in 1977 and 1978 had detectable AHH activity (Table 2). Assay of whole homogenates, nuclear fractions, and PMF showed that 95+% of AHH activity was in the PMF. AHH rates per milligram of PMF protein or per gram of liver were highest with Bathylagus, G. elongatum, and S. beani. Microsomal yields (mg of protein/g of liver) determined for some samples (Table 1) were used to convert AHH rates per gram of liver to rates per milligram of microsomal protein. The rates (pmol min-1 mg-1) were 80 for G. elongatum, 3-15 for different pools of B. glaciale, and 0.9 for C. sloani. Cytochrome c reductase rates, determined for seven species, ranged from 2 to 30 nmol min-1 (mg of PMF)-1 with NADPH (P450 reductase) and 7-60 with NADH (cyt b5 reductase). On the basis of microsomal yield (Table 1), the rates converted to 7-56 and 14-109 nmol min-1 (mg of microsomal protein)-1 for NADPH- and NADH-dependent activity, respectively. Total P450 content was 0.071 per mg of hepatic microsomal protein in G. elongatum and 0.090 for B. glaciale. In each, less than 10% of total P450 was present as cytochrome P420. Aminopyrine N-demethylase activity was measured in selected samples from Knorr 65, all of which had detectable activity (Table 2). The ApND rates were greater than those of AHH, and rates of the two activities did not correlate with one another; ApND is catalyzed by P450s other than CYP1A, while AHH activity is catalyzed predominantly by CYP1A proteins (43). The synthetic flavonoid ANF inhibits fish CYP1A but stimulates activity of other CYP that catalyze low rates of B[a]P oxidation, e.g., CYP3As (44). CYP3As are expressed in fish liver (45). The effect of 100 µM ANF on AHH activity was
FIGURE 2. Hepatic microsomal EROD activity in midwater fishes. CYP1A catalytic activity is compared in fish from two stations off the eastern coast of the United States on Iselin 9307. EROD activity was measured by the method of ref 30. Station location is as indicated in Figure 1. For A. aculeatus only, station 2 was significantly different from station 1 (P e 0.05). examined with a subset of samples from Oceanus 43 (1978) (Table 3). In samples with greater AHH rates, that activity was inhibited by ANF, while in samples with lower AHH rates the activity was stimulated by ANF (Table 3), possibly reflecting a CYP3A. This relationship between AHH rate and response to ANF is seen in the results with samples of B. glaciale that differed 5-fold in AHH rates (Table 3). 1993 Samples. Samples collected from the western North Atlantic in 1993 (Iselin) were examined using more sensitive and specific measures of CYP1A expression, the rates of EROD, and immunoassay with CYP1A-specific antibodies. EROD activity was detected in hepatic microsomes prepared from all species examined. For some species (e.g., A. aculeatus), EROD rates differed between samples collected on different sides of the Gulf Stream (Figure 2). In other species (G. elongatum and B. glaciale), rates differed little between those stations. Data for the two stations are combined in Table 5, which shows average rates ranging from 20 to 93 pmol min-1 mg-1; G. elongatum and A. aculeatus had the highest average rates of hepatic microsomal EROD activity. Immunoblot analysis of hepatic microsomes prepared from fish in the 1993 collection showed only single protein bands recognized by MAb 1-12-3, at about 54 kDa (Figure 3), typical of CYP1As. Cross-reacting protein was quantifiable in hepatic microsomes from most species (Table 4), and levels were highest in G. elongatum. EROD rates and CYP1A content did not appear to correlate with the species status as vertical migrators; however, most of the species examined are considered to be vertically migrate. Individuals of five species from 1993 were examined immunohistochemically with MAb 1-12-3. Staining was detected in one or more organs including liver, gill, heart, and kidney (Figure 4). Table 5 indicates the number of individuals of each species that had CYP1A staining in these organs. Hepatocytes and renal tubular epithelial cells showed strong CYP1A staining in some A. aculeatus and C. sloani. Staining was also detected in gill epithelial cells and gill pillar cells, and both species showed weak to moderate VOL. 35, NO. 1, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
57
TABLE 3. r-Naphthoflavone Inhibition of Midwater Fish Liver AHH Activitya
species
Nb
AHH acty without ANFd (pmol min-1 g-1)
AHH acty with 100 µM ANF (pmol min-1 g-1)
% inhibition or stimulation
Benthosema glacialec Benthosema glacialec Gonostoma elongatum Nemichthys scolopaceus
33 5 2 2
128 25 932 101
93 40 20 62
inhibited, 28 stimulated, 60 inhibited, 98 inhibited, 38
a Samples were obtained on Oceanus 43 (1978) and analyzed aboard ship. Data are expressed as pmol min-1 (g of liver)-1. b N refers to the number of individuals from which tissues were pooled. Assays were done in duplicate with less than 10% difference between replicates. c The two pools of Benthosema were from different trawls. d ANF was added in acetone. Reaction mixtures without ANF had an equivalent volume of acetone added.
TABLE 4. Hepatic Microsomal EROD and Cytochrome P450 1A in Midwater Fisha genus
migrator
N ( n) b
EROD acty (pmol min-1 mg-1)
CYP1A (pmol/mg)d
Argyropelecus hemigymnus Argyropelecus aculeatus Benthosema glaciale Chauliodus sloani Gonicthys cocco Gonostoma elongatum Lampanyctus cuprarius Lepidophanes guentheri Sternoptyx diaphana
yes yes yes yes yes yes yes yes no
1 (2) 10 (15) 6 (26) 7 1 9 (17) 3 (5) 3 (6) 3 (6)
15 60 ( 67c 25 ( 9 20 ( 12 59 93 ( 40 34 ( 38 22 ( 9 46 ( 36
nde 0.3 ( 0.2 0.9 ( 0.3 1.4 ( 0.4 nd 2.5 ( 0.5 0.1 ( 0.1 0.5 ( 0.4 1.8
a Samples were collected on Iselin 9307 in 1993. b N refers to the number of individual or pooled samples examined, and (n) indicates the total number of individuals from which tissues were pooled. c Values are means of results for all fish analyzed from stations 1 and 2 ( SD. Individual rates ranged from 2 to 184 pmol min-1 mg-1. d Analysis was with large gel format, with 80 µg of microsomal protein applied per lane and staining visualized by enhanced chemiluminescence. Date are expressed as scup CYP1A equiv/mg of microsomal protein. e nd, not determined.
TABLE 5. CYP1A Immunostaining in Tissues of Midwater Fish species
Argyropelecus aculeatus Benthosema glaciale Chauliodus sloani Gonostoma elongatum Sternoptyx diaphana
liver hepatocytes
liver endothelia
renal tubules
gill epithelia
gill pillar cells
heart endothelia
2/6a -b
1/6 0/7 0/2 1/2
4/4 4/5 0/2
1/5 1/1 4/5 0/2
4/5 3/5 0/2
1/3 5/5 0/2 0/2
2/2 2/2
a Data show number of animals showing positive signal relative to number of animals in which tissue was examined. of that type was evident on the slides.
FIGURE 3. Immunoblot of hepatic microsomal CYP1A. Samples from 1993 (Iselin) were immunoblotted with the CYP1A1-specific MAb 1-12-3, using alkaline phosphatase-linked colorimetric detection, with 10 µg of protein applied to each lane. Lane 1, Scup CYP1A standard; lanes 2 and 3, Benthosema glaciale pooled samples; lanes 4 and 5, individual Gonostoma elongatum. staining in endothelium in various organs. In Sternoptyx and G. elongatum, only hepatocytes showed staining, although kidney and gill from G. elongatum were not examined. The single sample of B. glaciale gill showed epithelial cell staining. Bioassay and Chemical Residues. In 1993, G. elongatum and A. aculeatus quantities were sufficient for bioassay and chemical residue analysis. Extracts of samples of both species were tested for the capacity to induce CYP1A (EROD activity) 58
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 1, 2001
b
Dash indicates no tissue
in the fish hepatoma cell line, PLHC-1. The extracts caused a 3-5-fold induction of EROD, relative to the basal rates observed in PLHC cells (0.5-1.0 pmol min-1 mg-1). The resultant TCDD-EQs calculated were 1-5 pg of TCDD-EQ/g wet weight (Table 6), based on a comparison with EROD activity induced by TCDD in the same cell line. Thus, some component(s) of the extracts tested had a TCDD-like potency. The samples used for TCDD-EQ determination also had detectable concentrations of PCBs, as measured by a quantitative ELISA calibrated against Aroclor 1254. The concentrations measured were between 2 and 5.7 ng of total PCB/g wet weight (Table 6). The content of PCBs in extracts was closely correlated with the TCDD-EQs (Figure 5). Neither PCB residues nor TCDD-EQ showed close correlation with EROD rates in the same fish.
Discussion Midwater fishes are unusual, difficult to obtain, and often fragile organisms that inhabit regions of the world oceans remote from direct sources of contaminants. We examined CYP1A in midwater fishes collected from the western North Atlantic to assess whether environmental chemicals might be eliciting molecular changes in these fishes. Hepatic AHH activity, microsomal EROD rates, immunoblot detection of proteins by MAb 1-12-3, and immunohistochemical staining
FIGURE 4. Immunohistochemical staining of CYP1A in midwater fish. Sections of whole fish were stained using the CYP1A1-specific MAb 1-12-3 or the nonspecific antibody UPC10, which did not stain any tissues or cells (not shown). Arrows show the following: (A) MAb 1-12-3-positive staining in endothelial cells in heart of Chauliodus sloani (400×). (B) Positive endothelial cells in liver of C. sloani (400×). (C) Positive renal tubular epithelium of Argyropelecus aculeatus (400×). (D) Positive liver endothelial cells of A. aculeatus (400×). (E) Positive gill pillar cells of A. aculeatus (400×). (F) Positive hepatocytes of A. hemigymnus (400×). results together suggest the presence of CYP1A in many of the species. In most species, AHH and EROD rates were low relative to those in some coastal fishes; AHH and EROD rates in fish from Bermuda ranged from 0.01 to about 1 nmol min-1 mg-1 (46). The low rates do not appear to have resulted from enzyme inactivation during capture and retrieval. Most fish were alive when retrieved, and there was little microsomal P420 in selected species. By comparison, benthic fish from greater depths and with longer retrieval times showed little P450 degradation and had hepatic AHH and EROD rates higher than those in the midwater fishes (26). Results from 1977-1978 and 1993 cannot be directly compared, as different enzyme reactions were measured, and sample volumes did not permit further analyses. Yet, the data are consistent in indicating higher levels of CYP1A expression in the same species in the different collections. Thus, in 1977-1978 G. elongatum had higher AHH rates than the other species, and
the activity was inhibited by ANF. In 1993, G. elongatum had the highest average EROD rates and levels of CYP1A protein. Species differences in levels of CYP1A expression might reflect different amounts of inducers accumulated. The content of PCBs reported by Harvey and colleagues (47) differed by as much as 70-fold between individuals and up to 10-fold between species of midwater fish. Different CYP1A levels among individuals within a species, such as that indicated by differences in AHH rates and effects of ANF seen in B. glaciale (Table 4), are consistent with differences in exposure to inducers. Because xenobiotic inducers of CYP1A are ubiquitous in the environment, it is difficult to distinguish between lowlevel induction caused by foreign chemicals and low-level expression possibly due to endogenous regulation. Increases in CYP1A1 mRNA have been detected in cells in culture (48) and in early developmental stages of mice (49), in the absence of treatment with exogenous AHR agonists. Endogenous VOL. 35, NO. 1, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
59
TABLE 6. TCDD Equivalents and PCB Content in Midwater Fish sample/species
Na
Argyropelecus aculeatus Argyropelecus aculeatus Argyropelecus aculeatus Argyropelecus aculeatusd Argyropelecus aculeatusd Gonostoma elongatum Gonostoma elongatum Gonostoma elongatum Gonostoma elongatumd procedural blank matrix blank, bluegill positive control material 1, Saginaw Bay carp positive control material 2, Saginaw Bay carp
(1) (2) (2) (1) (2) (1) (1) (2) (1)
TCDD-EQb (pg/g)
total PCBc (ng/g)
1.2 ( 0.3 1.9 ( 0.4 2.0 ( 0.3 1.1 ( 0.2 1.9 ( 0.5 1.5 ( 0.2 3.9 ( 1.8 1.3 ( 0.2 1.6 ( 0.4 0.7 ( 0.1 0.3 ( 0.1 96.1 ( 8.6
3.0 3.7 3.7 2.6 3.8 3.3 5.7 2.3 3.9
50.5 ( 3.7
a N indicates number of fish from which tissue (whole carcass minus the liver) was pooled for extraction. b TCDD-EQ refers to CYP1A induction in PLHC cells equivalent to the induction caused by TCDD, expressed as pg TCDD-EQ-1 (g of tissue wet weight)-1. c PCB content was determined by ELISA. Values in ng/g wet weight are the average of replicate assays. d Indicates samples from station 2 in the Sargasso. All others were from station 1 in the slope water. e Mean and SD of replicate analyses.
FIGURE 5. Toxic equivalents (TCDD-EQ) vs PCB content in midwater fish. Data are from Table 6. Values are pg of TCDD-EQ/g of tissue tissue wet weight, and the values for PCBs are in ng of PCB/g of tissue wet weight, measured in the same extracts of midwater fishes. inducers of CYP1A in adult organisms include bilirubin and tryptophan oxidation products (50, 51). Disease or genetic conditions can lead to accumulation of bilirubin and induced CYP1A1, for example, in Gunn rats (52). However, CYP1A1 protein is not readily detected in organs of healthy adult vertebrates not exposed to exogenous inducers. Other aspects of our data also suggest that CYP1A expression resulted from chemical exposure, at least in some species. Some individuals had hepatic EROD rates similar to those in some coastal fish that had EROD or CYP1A levels correlated closely with tissue levels of inducers (53). CYP1A was detected in heart endothelium and gill epithelium, cell types where CYP1A typically occurs only after exposure to exogenous inducers (35). Moreover, extracts of two species contained residues of PCBs and were able to induce CYP1A in PLHC cells in culture. The bioassay results are an important finding, emphasizing that TCDD-EQ can be detected in biota from environments remote from known sources of contaminants. Increasingly, cell lines with a native CYP1A response or engineered with reporter genes linked to CYP1A promoters 60
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 1, 2001
(50) are used to assess the presence of AHR agonists in environmental samples. Yet, demonstrating the presence of TCDD-EQ by itself does not allow the conclusion that the chemicals responsible are present at concentrations active in the animals of concern. That must be determined from empirical knowledge of dose-response relationships or by examining the animals themselves with approaches such as those used here. Significantly, the bioassay results were obtained with extracts of the same species (G. elongatum and A. aculeatus) that had higher levels of EROD and CYP1A expression. Potential inducers in the midwater include PCBs. Harvey et al. (47) detected PCBs by GC-EC in midwater fishes from the North Atlantic, including A. aculeatus. Our data support Harvey’s results, although the levels we detected are 10-fold less than those they reported (47), which could result from differences in analytical technique. However, Gustafsson et al. (25) estimated that the flux of PCBs exported from surface was about 10-fold less in 1993, when the fish we analyzed were obtained, than in the 1970s, when Harvey et al. collected fish they analyzed. While the potency with which extracts of midwater fish induced EROD in PLHC cells was correlated with content of total PCBs in those extracts, the inducing potency of extracts was greater than expected based on the amounts of PCBs detected. The ELISA that we used to measure PCBs was calibrated to Aroclor 1254. Possibly, non-ortho-substituted PCB congeners, the most potent CYP1A inducers among the PCBs, were enriched in these fish relative to levels in Aroclor 1254. Other PHAH likely to survive the extraction process but that were not analyzed include chlorinated dibenzo-pdioxins, which often co-vary with the PCBs (22) and could contribute to the TCDD-EQ. Chemicals more susceptible to degradation during extraction (PAH, natural products) tend not to bioaccumulate in fish and might not contribute to the TCDD-EQ yet could elicit induction in vivo. PAH enter the oceans in petroleum from natural seeps, tanker and ship traffic, and atmospheric sources (54). PAH, perhaps associated with soot particles, have been measured in deep ocean samples (24, 25). Indoles and halogenated indoles (55) able to induce CYP1A (M. Hahn, personal communication) are synthesized by some algae and sessile invertebrates (56). Exposure to those sources is unlikely in carnivorous fishes at midwater depths in the open ocean, but other natural inducers may be available. Astaxanthin, a major carotenoid in crustaceans, can induce CYP1A1 in mammals (57) and could do so in fish. Many midwater fishes, including G. elongatum and A. aculeatus, feed on crustaceans (Table 1). We have found that astaxanthin at concentrations from 0.05 to 50 µM did not induce CYP1A in PLHC cells, although it did cause a dose-dependent potentiation of induction by 0.1 nM TCDD (unpublished data). Thus, astaxanthin in fish extracts alone would not likely cause induction in PLHC cells, although it might enhance induction by other compounds in those extracts. There were no consistent differences in levels of CYP1A, nor in PCB content or TCDD-EQ, in fish from the slope water as compared to those from the Sargasso Sea. One might expect a greater degree of contamination or response in animals from the slope water. Harvey and Steinhauer (20) found that PCB concentrations in atmospheric samples taken over the western North Atlantic decreased exponentially with distance from land. Moreover, except for water transported across the Gulf Stream in cold core rings (58), the Stream acts as a barrier between the slope water and the Sargasso Sea. Yet, Harvey and Steinhauer did not detect differences in PCB concentrations between surface waters of the slope region and those of the Sargasso Sea, in line with though not directly comparable to our CYP1A results. A greater particulate flux in slope waters might contribute to this.
Both G. elongatum and A. aculeatus appeared to have induced CYP1A levels; however, the TCDD-EQs derived from cell bioassay are lower than TCDD-EQs associated with threshold induction of CYP1A in all but the most sensitive species, e.g., lake trout (59). As suggested above, chemicals not contributing to TCDD-EQ could be acting in vivo in these fish. Midwater fishes also might be more sensitive to induction than coastal species that are chronically exposed to greater levels of contaminants. If so, this could be the converse of the resistance to toxic and CYP1A-inducing effects of PHAH in estuarine fishes exposed for generations to very high levels of inducers (60, 61). The sensitivity to inducers could be determined by evaluating CYP1A induction in primary cultures of hepatocytes from midwater species, an approach used successfully with birds (62). If midwater fish are more sensitive to induction, that could suggest that their AHR signal transduction systems differ from those in coastal fishes or perhaps that there is some novel induction mechanism in the midwater fishes. Our results from the western North Atlantic provide a foundation for evaluating CYP1A expression in midwater fishes from other oceans. Further studies of the identity and concentration of PCB congeners and other chemicals in mesopelagic fishes, and studies to establish whether lowlevel occupancy of Ah receptors and low-level expression of CYP1A are risk factors for toxic effects in sensitive fish species, could indicate whether mesopelagic fish are likely to experience further effects of globally distributed anthropogenic chemicals.
Acknowledgments We gratefully acknowledge the assistance of Albert Sherman and Heidi Kaplan in analyses of fish collected on the Knorr and Oceanus cruises and the ships and scientific crews on the R/V Knorr, R/V Oceanus, and R/V Columbus Iselin for assistance in obtaining fish. We also thank John Meadows, Diane Nicks, and James Zajicek from the CERC (Columbia, MO) for assistance in analysis of samples and Dr. Eli Hestermann for the assay of astaxanthin effect in PLHC cells. Dr. Bruce Robison provided input regarding diets of midwater fish. Drs. Jay Gooch, John Farrington, Mark Hahn, Jeff Whyte, and Hisato Iwata provided helpful comments on the manuscript. This work was supported in part by National Science Foundation Grant OCE 76-84415, U.S. EPA Grant R-823890-01, Sea Grant NA46RG0470-R/P-61, and the Lyons Graduate Fellowship at MIT to J.J.S. This is Contribution No. 10001 from the Woods Hole Oceanographic Institution.
Notation AHH
aryl hydrocarbon hydroxylase
AHR
aryl hydrocarbon receptor
ANF
R-naphthoflavone
B[a]P
benzo[a]pyrene
CYP1A
cytochrome P450 1A
PAH
polynuclear aromatic hydrocarbon
PHAH
planar halogenated aromatic hydrocarbon
PCB
polychlorinated biphenyl
TCDD
2,3,7,8-tetrachlorodibenzo-p-dioxin
Literature Cited (1) Moore, M. J.; Stegeman, J. J. Dis. Aquat. Org. 1994, 20, 33-48. (2) Malins, D. C.; McCain, B. B.; Myers, M. S.; Brown, D. W.; Krahn, M. M.; Roubal, W. T.; Schiewe, M. S.; Landahl, J. T.; Chan, S.-L. Environ. Health Perspect. 1987, 71, 5-16.
(3) Harshbarger, J. C.; Clark, J. B. Sci. Total Environ. 1990, 94. (4) Peterson, R. E.; Theobald, H. M.; Kimmel, G. L. Crit. Rev. Toxicol. 1993, 23, 283-335. (5) Walker, M. K.; Peterson, R. E. Aquat. Toxicol. 1991, 21, 219-238. (6) Hankinson, O. Annu. Rev. Pharmacol. Toxicol. 1995, 35, 307340. (7) Morrison, H. G.; Oleksiak, M. F.; Cornell, N. W.; Sogin, M. L.; Stegeman, J. J. Biochem. J. 1995, 308, 97-104. (8) Hahn, M. E.; Karchner, S. I.; Shapiro, M. A.; Perera, S. A. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 13743-13748. (9) Poland, A.; Knutson, J. C. Annu. Rev. Pharmacol. Toxicol. 1982, 22, 517-554. (10) Guiney, P. D.; Smolowitz, R. M.; Peterson, R. E.; Stegeman, J. J. Toxicol. Appl. Pharmacol. 1997, 143, 256-273. (11) Hahn, M. E.; Stegeman, J. J. Toxicol. Appl. Pharmacol. 1994, 127, 187-198. (12) Van Veld, P. A.; Vogelbein, W. K.; Cochran, M. K.; Goksoyr, A.; Stegeman, J. J. Toxicol. Appl. Pharmacol. 1997, 142, 348-359. (13) Burns, K. A. Comp. Biochem. Physiol. 1976, 53B, 443-446. (14) Spies, R. B.; Felton, J. S.; Dillard, L. Mar. Biol. 1982, 70, 117127. (15) Stegeman, J. J.; Teng, F. Y.; Snowberger, E. A. Can. J. Fish. Aquat. Sci. 1987, 44, 1270-1277. (16) Varanasi, U.; Collier, T. K.; Williams, D. E.; Buhler, D. R. Biochem. Pharmacol. 1986, 35, 2967-2971. (17) Van Veld, P. A.; Westbrook, D. J.; Woodin, B. R.; Hale, R. C.; Smith, C. L.; Huggett, R. J.; Stegeman, J. J. Aquat. Toxicol. 1990, 17, 119-132. (18) Stegeman, J. J.; Woodin, B. R.; Goksøyr, A. Mar. Ecol. Prog. Ser. 1988, 46, 55-60. (19) Stegeman, J. J.; Kloepper-Sams, P. J.; Farrington, J. W. Science 1986, 231, 1287-1289. (20) Harvey, G. R.; Steinhauer, W. G. Atmos. Environ. 1974, 8, 777782. (21) Simonich, S. L.; Hites, R. A. Science 1995, 269, 1851-1854. (22) Iwata, H.; Tanabe, S.; Sakai, N.; Tatsukawa, R. Environ. Sci. Technol. 1993, 27, 1080-1098. (23) Brewer, P. O.; Glover, D. M. Rev. Geophys. 1987, 25, 1376-1386. (24) Takada, H.; Farrington, J. W.; Bothner, M. H.; Johnson, C. G.; Tripp, B. W. Environ. Sci. Technol. 1994, 28, 1062-1072. (25) Gustafsson, O.; Gschwend, P. M.; Buesseler, K. O. Environ. Sci. Technol. 1997, 31, 3544-3550. (26) Stegeman, J. J. Can. J. Fish. Aquat. Sci. 1983, 40, 78-85. (27) Backus, R. H.; Craddock, J. E. Midwater fish data report: Cold core ring time series cruises Knorr 62, 65, and 71 and Endeavor 11; WHOI-80-18; Woods Hole Oceanographic Institution: Woods Hole, MA, 1980. (28) Stegeman, J. J.; Binder, R. L.; Orren, A. Biochem. Pharmacol. 1979, 28, 3431-3439. (29) Klotz, A. V.; Stegeman, J. J.; Walsh, C. Anal. Biochem. 1984, 140, 138-145. (30) Hahn, M. E.; Woodward, B. L.; Stegeman, J. J.; Kennedy, S. W. Environ. Toxicol. Chem. 1996, 15, 582-591. (31) Lowry, O. H.; Rosebrough, N. J.; Farr, A. L.; Randall, R. J. J. Biol. Chem. 1951, 193, 265-275. (32) Smith, P. K.; Krohn, R. I.; Hermanson, G. T.; Mallia, A. K.; Gartner, F. H.; Provenzano, M. D.; Fujimoto, E. K.; Goeke, N. M.; Olson, B. J.; Klenk, D. C. Anal. Biochem. 1985, 150, 76-85. (33) Kloepper-Sams, P. J.; Park, S. S.; Gelboin, H. V.; Stegeman, J. J. Arch. Biochem. Biophys. 1987, 253, 268-278. (34) Park, S. S.; Miller, H.; Klotz, A. V.; Kloepper-Sams, P. J.; Stegeman, J. J.; Gelboin, H. V. Arch. Biochem. Biophys. 1986, 249, 339-350. (35) Smolowitz, R. M.; Schultz, M. E.; Stegeman, J. J. Carcinogenesis 1992, 13, 2395-2402. (36) Woodin, B. R.; Smolowitz, R. M.; Stegeman, J. J. Environ. Sci. Technol. 1997, 31, 1198-1205. (37) Tillitt, D. E.; Giesy, J. P.; Ankley, G. T. Environ. Sci. Technol. 1991, 25, 87-92. (38) Tysklind, M.; Tillitt, D.; Eriksson, L.; Lundgren, K.; Rappe, C. Fundam. Appl. Toxicol. 1994, 22, 277-285. (39) Finney, D. J. Statistical Methods in Biological Assay; Charles Griffin: London, 1980. (40) Ankley, G. T.; Tillitt, D. E.; Giesy, J. P.; Jones, P. D.; Verbrugge, D. A. Can. J. Fish. Aquat. Sci. 1991, 48, 1685-1690. (41) Zajicek, J. L.; Tillitt, D. E.; Huckins, J. N.; Petty, J. D.; Potts, M. E.; Nardone, D. A. In Environmental Immunochemical Methods; American Chemical Society: Washington, DC, 1996; pp 307325. (42) Zajicek, J. L.; Tillitt, D. E.; Schwartz, T. R.; Schmitt, C. J.; Harrison, R. O. Chemosphere 2000, 40, 539-548. (43) Stegeman, J. J.; Woodin, B. R.; Park, S. S.; Kloepper-Sams, P. J.; Gelboin, H. V. Mar. Environ. Res. 1985, 17, 83-86. VOL. 35, NO. 1, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
61
(44) Lee, C. A.; Kadwell, S. H.; Kost, T. A.; Serabjit-Singh, C. J. Arch. Biochem. Biophys. 1995, 319, 157-167. (45) Celander, M.; Buhler, D. R.; Forlin, L.; Goksoyr, A.; Miranda, C. L.; Woodin, B. R.; Stegeman, J. J. Fish Physiol. Biochem. 1996, 15, 323-332. (46) Stegeman, J.; Woodin, B.; Singh, H.; Oleksiak, M.; Celander, M. Comp. Biochem. Physiol. 1997, 116, 61-75. (47) Harvey, G. R.; Miklas, H. P.; Bowen, V. T.; Steinhauer, W. G. J. Mar. Res. 1974, 32, 103-118. (48) Sadek, C. M.; Allen Hoffmann, B. L. J. Biol. Chem. 1994, 269, 16067-16074. (49) Dey, A.; Nebert, D. Biochem. Biophys. Res. Commun. 1998, 251, 657-661. (50) Heath-Pagliuso, S.; Rogers, W.; Tullis, K.; Seidel, S.; Cenijn, P.; Brouwer, A.; Denison, M. Biochemistry 1998, 37, 11508-11515. (51) Sinal, C.; Bend, J. Mol. Pharmacol. 1997, 52, 590-599. (52) Kapitulnik, J.; Gonzalez, F. Mol. Pharmacol. 1993, 43, 722-725. (53) Spies, R.; Stegeman, J.; Rice, D.; Woodin, B.; Thomas, P.; Hose, J.; Cross, J.; Prieto, M. In Biomarkers; American Chemical Society: Washington, DC, 1991; pp 87-121. (54) Neff, J. W. Polycyclic Aromatic Hydrocarbons in the Aquatic Environment: Sources, Fate and Biological Effects; Applied Science Publishers: Barking, Essex, U.K., 1979.
62
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 1, 2001
(55) Wei, Y.; Helleberg, H.; Rannug, U.; Rannug, A. Chem. Biol. Interact. 1998, 110, 39-55. (56) Kondo, K.; Nishi, J.; Ishibashi, M.; Kobayashi, J. J. Nat. Prod. 1994, 57, 1008-1011. (57) Gradelet, S.; Astorg, P.; Leclerc, J.; Chevalier, J.; Vernevaut, M.; Siess, M. Xenobiotica 1996, 26, 49-63. (58) Ring-Group. Science 1981, 212, 1091-1100. (59) Whyte, J. J.; Van den Heuvel, M. R.; Clemons, J. H.; Huestis, S. Y.; Servos, M. R.; Dixon, D. G.; Bols, N. C. Environ. Toxicol. Chem. 1998, 17, 2214-2226. (60) Prince, R.; Cooper, K. R. Environ. Toxicol. Chem. 1995, 14, 579587. (61) Elskus, A.; Monosson, E.; McElroy, A.; Stegeman, J. J.; Woltering, D. Aquat. Toxicol. 1999, 45, 99-113. (62) Kennedy, S. W.; Lorenzen, A.; Jones, S. P.; Hahn, M. E.; Stegeman, J. J. Toxicol. Appl. Pharmacol. 1996, 141, 214-230.
Received for review April 28, 2000. Revised manuscript received September 26, 2000. Accepted September 29, 2000. ES0012265
