E N V I R O N M E N T A L RESEARCH 5 5 , 64--78

(1991)

Cytochrome P4501A mRNA Expression in Feral Hudson River Tomcod G . - L . KREAMER, K . SQUIBB, D . G I O E L I , S. J. GARTE, AND I. W I R G I N

New York University Medical Center, Institute of Environmental Medicine, Long Meadow Road, Tuxedo, New York 10987 Received October 18, 1990 We sought to determine if levels of cytochrome P450IA gene expression are environmentally induced in feral populations of Hudson River tomcod, a cancer prone fish, and whether laboratory exposure of tomcod to artificially spiked and naturally contaminated Hudson sediments can elicit a significant response. Using Northern blot analysis, we found levels of P450IA mRNA in tomcod collected from two Hudson River sites higher than those in tomcod from a river in Maine. Deputation of environmentally induced Hudson tomcod P450IA mRNA was rapid, with an initial detectable decline in P450 gene expression by 8 hr and basal levels reached by 5 days. Intraperitoneal injection of [3-napthoflavone in depurated Hudson tomcod resulted in a 15-fold induction of P450 gene expression within 26 hr. Exposure of depurated Hudson tomcod to natural sediment spiked with two PAHs resulted in a 7-fold induction of P450 gene expression. Exposure of depurated tomcod to sediment from a contaminated Hudson site also resulted in a 7- to 15-fold induction of P450IA mRNA expression. Northern blot analysis revealed a second polymorphic cytochrome P450IA mRNA band in some tomcod which was also detected by Southern blot analysis. Induction of cytochrome P450IA mRNA in Atlantic tomcod may provide a sensitive biomarker of environmentally relevant concentrations of some pollutants in the Hudson and other northeastern tidal rivers. © 1991AcademicPress, Inc.

INTRODUCTION The Hudson River is situated in geographic proximity to a large urban population, on occasion it serves as a domestic water supply, it has supported moderatesized commercial fisheries, and it receives extensive recreational use. The Hudson River has historically been polluted with a suite of different organic (O'Connor and Rachlin, 1982) and inorganic (Williams et al., 1978; O'Connor and Rachlin, 1982) pollutants. Organic xenobiotics have included dioxins, various PAHs, and Aroclor PCB mixtures (for a listing see Rohmann and Lilienthal, 1987). The point source release of PCBs in the upper Hudson, their physical dispersion throughout the system, and their bioaccumulation throughout the food chain have caused particular concern (Horn et al., 1979; Limburg, 1988) and have resulted in the closure of commercial fisheries for important resident Hudson River species such as striped bass. The Atlantic tomcod (Microgadus tomcod) is a very common bottom-dwelling fish which probably spends its entire life cycle within the confines of the lower Hudson River (Klauda et al., 1988; McLaren et al., 1988). The overall distribution of tomcod extends from Virginia to Labrador (Bigelow and Schroeder, 1953); however, it is believed that the Hudson River supports their southernmost spawning population. Hudson tomcod have unusually high liver lipid levels which can 64

00t3-9351/91 $3.00 C0py~igbt © 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.

CYTOCHROME P450IA mRNA IN T O M C O D

65

potentially accentuate the bioaccumulation of nonpolar xenobiotics (Cormier et al., 1989). Hudson tomcod demonstrate a rapid onset and a remarkably high incidence of liver tumors, such that more than 50% of spawning l-year-old and more than 90% of 2-year-old fish exhibit hepatocellular carcinomas (Dey et al., 1986). In comparison, tomcod from more pristine environments show little evidence of this condition (Cormier et al., 1989; Cormier and Raccine, 1990). Tomcod liver tumor DNA can transform mouse fibroblast cells in culture, is tumorigenic in the nude mouse assay, and contains an activated K-ras oncogene (Wirgin et al., 1989). A high rate of mutational change at another oncogene locus, c-abl, in Hudson tomcod has also been reported (Wirgin et al., 1990). Analysis of tissue loads of some common Hudson River organic and inorganic contaminants revealed high PCB concentrations (Klauda et al., 1981) in tomcod livers, whereas PAHs were nondetectable (Dey et al., 1986) perhaps resulting from their rapid metabolism (O'Connor et al., 1988). In total, this evidence suggests that tomcod may prove an unusually sensitive sentinel species of Hudson River perturbation. Cytochrome P450 genes are highly conserved throughout the animal kingdom (Nebert and Gonzalez, 1987) and encode for an array of enzymes which both detoxify and activate a variety of endogeneous and exogenous substrates. In some cases, these P450 enzymes convert xenobiotic substrates to their carcinogenic penultimate metabolite which may react to generate DNA lesions (Smolarek et al., 1987; Varanasi et al., 1989). The inducibility of the P450IA gene in fishes by xenobiotic agents such as PAHs (Stegeman and Kloepper-Sams, 1987; Van Veld et al., 1990), PCBs (Gooch et al., 1989), and dioxins (Cooper et al., 1990) has been demonstrated by quantification of P450 proteins by immunoreactivity with species-specific monoclonal antibodies (Stegeman and Kloepper-Sams, 1987), catalytic activity, or both. As a result, it has been suggested that quantification of P450IA gene products be used as biomarkers of biologically relevant exposure of fishes to some organic xenobiotics. The nomenclature, structure, and function of the cytochrome P450IA gene in fishes has been recently reviewed by Stegeman (1989). In mammals, two P450IA genes have been described, whereas a single 3-methylcholanthrene (3MC)-induced gene, P450IA1, has been detected to date in the fishes. P450IA1 cDNA from 3-MC-induced rainbow trout has been cloned (Heilmann et al., I988) and the rapid inducibility of P450IA1 mRNA by [3-napthoflavone (~-NF) in rainbow trout and estuarine killifish has been demonstrated by immunoprecipitation of translatable mRNA and by Northern blot analysis (Haasch et al., 1988 and 1989; Kloepper-Sams and Stegeman, 1989). We were interested in testing the applicability of induced cytochrome P450IA mRNA expression in Atlantic tomcod as a potentially broad biomarker of exposure to common organic Hudson River pollutants. Unlike most previous studies, we studied a population of feral fish from a known impacted aquatic system as our model organisms. We initially sought to determine if levels of P450IA1 mRNA expression are naturally induced in Hudson River tomcod and whether mRNA expression could he induced by laboratory exposure of tomcod to Hudson River sediments of concern.

66

KREAMER ET AL. METHODS

Sample Collections In total, 183 tomcod were examined in this study (Table 1). Tomcod were collected from the Hudson River at two sites (n = 167). Spawning adults were sampled with unbaited boxtraps set off a bulkhead at Garrison, New York, located approximately 50 miles north of New York City, in January and February 1990. Additional sets of Hudson samples were collected in midchannel off midtown Manhattan, New York City, in March and April 1990. Spawning tomcod were also sampled with boxtraps set in the Saco River, Maine, in December 1987 (n = 16). Where possible, sex, age, and external and overall liver condition were characterized for each sample. Age was determined by otolith analysis. Mean total length of Hudson tomcod collected at Garrison and NYC were 170 and 165 mm, respectively. The vast majority of Hudson tomcod were 1-year-old fish. When possible, tomcod of similar size and age were grouped for experimental treatments. Hudson River white perch (Morone americana) and striped bass (Morone saxatilis) were collected by hook and line at Croton, New York, while hogchokers (Trinectes maculatus) were caught by otter trawling off Peekskill, New York. Rainbow trout (Onchorynchus mykiss) were obtained from a commercial vendor.

Depuration of Hudson Tomcod To determine the levels of P450IA expression in freshly caught Hudson tomcod, a subset of these samples was sacrificed in the field, and the livers were flash frozen in liquid nitrogen. All fish to be experimentally exposed or depurated were TABLE 1 COLLECTIONS AND TREATMENTS OF ATLANTIC TOMCOD

Treatment Field exposure

Collection site Garrison NYC

Depuration Reintroduction to Hudson

No. fish 14 6

Saco River, Maine

16

Garrison NYC Garrison NYC

44 21 5

10 10

10

13-NF injected

Control, undepurated

Garrison

Test, undepurated

Garrison

Control, depurated Test, depurated Sediment exposure Control sediment Spiked sediment

Garrison

Hudson contaminated sediment

Garrison

14 5

NYC Garrison NYC NYC

6 7 4 11

CYTOCHROME P4501A mRNA IN TOMCOD

67

returned to our laboratory in ambient Hudson River water. All fish were transferred to 5 ppt instant ocean in 150-gallon aquaria at 4°C and were maintained under these conditions until further treatment or sacrifice. Tomcod were fed chopped clams ad libitum. To determine rates of depuration, groups of tomcod were sacrificed 4, 8, 19, 120, and 480 hr post-transfer. Exposure and Treatments Reintroduction to the Hudson. Two groups of depurated tomcod (>20 days) were reintroduced into the Hudson at two sites; Garrison, New York, and lower Manhattan, New York City. Both groups of fish were suspended off bulkheads immediately above the substrate in 2 × 2 x 3-ft nylon mesh cages. The Garrison fish were exposed in March 1990 for 4 days and the New York City fish were exposed in June 1990 for 7 days. Both groups of fish were immediately sacrificed at the conclusion of the experiment. f3-Napthoflavone injection. A group of tomcod was injected ip with 100 ~g of f3-NF/kg of fish suspended in 0.1-0.2 ml of corn oil (stock concentration was 25 mg/ml) 8 hr post-transfer to clean laboratory water. Fish were maintained at 4°C and sacrificed after 19 or 26 hr. A second set of experimental fish was depurated >30 days, injected with 100 p~g of [3-NF/kg of fish, and sacrificed after 26 hr. Control fish were ip injected with 0.1 ml of corn oil and also sacrificed after 26 hr. Sediment exposure. A group of depurated Hudson tomcod (>30 days) was exposed for 3-5 days to approximately 1 kg of Long Island Sound sediment spiked with 16.3 ~g/g (dry weight) phenanthrene and 34.2 ~g/g fluoranthene in static 50-gallon aquaria filled with 5 ppt instant ocean at 4°C. A second group of depurated Hudson tomcod was exposed for 7 days to 1 kg of contaminated Hudson sediment under the conditions described above. This sediment was obtained from a cove off the main river in Cold Spring, New York. Control fish for both experiments consisted of depurated Hudson tomcod which were maintained under identical conditions in aquaria with 1 kg of unspiked Long Island Sound sediment and sacrificed at the end of 5 days. A summary of all treatments and exposures is presented in Table 1. R N A isolation. All RNA analysis was conducted on liver tissue. Livers were excised from freshly sacrificed fish, gall bladders were avoided, and the remaining tissue was flash frozen in liquid nitrogen. Tissues were maintained at - 80°C prior to RNA isolation. Total RNA was purified by the RNAzol method (Chomczynski and Sacchi, 1987) and levels of P450IA mRNA expression were determined by Northern blot analysis. Twenty to 80 mg of liver were homogenized by hand in a glass pestle with a Teflon-coated mortar in 1.2 ml of RNAzol reagent (Biotecx Laboratories, Houston, TX). One-tenth volume of chloroform was added, samples were vortexed and incubated at 4°C for 15 rain, and centrifuged at 14,000 RPM for 15 min. The aqueous layer was withdrawn, and 600 FI of isopropanol was added. Samples were frozen for at least 3 hr at -80°C, centrifuged, and the resulting pellet was resuspended in 100-200 p.l of diethylpryocarbonate (DEPC) treated water. One-tenth volume of 5 M NaC1 was added followed by 100% ethanol to a final ethanol concentration of 75%. Samples were incubated at - 80°C overnight and centrifuged, and the pellet was resuspended in DEPC-treated water

68

KREAMER ET AL.

to a final concentration of 1.5-4.0 I-~gtotal RNA/~l H20. RNA concentrations and purity were determined spectrophotometrically at 260 and 280 nm. Total RNA yields ranged from 4 to l0 mg RNA/g of liver tissue. Each preparation generally provided sufficient RNA to run on >5 gels. Total RNA from l0 samples was prepared daily. R N A analysis. Cytochrome P450IA mRNA levels were quantified by Northern blot analysis. Ten micrograms of total RNA was denatured in glyoxal and electrophoresed through 1.0% agarose gels. Each gel contained at least one internal standard which allowed for intergel comparisons. To ensure equal loading of total RNA in each lane and to evaluate the integrity of the RNA, gels were stained with ethidium bromide (5 ixg/ml) for 45 min, destained for 15 min in H 2 0 , u v illuminated, and photographed. RNA was transferred after electrophoresis to ZetaBind filters (Cuno, Meriden, CT) by capillary action (Southern, 1975), washed in 2x SSC, uv fixed (Stratalinker, Stratagene), and stored under vacuum until use. Filters were then washed in agitating 0.1× SSC/0.5% SDS at 65°C for 1 hr and then incubated in 6x SSC for 5 min at room temperature. Filters were prehybridized and hybridized (in the presence of dextran sulfate) at 65°C (Wahl et al., 1979) to the nick-translated (Rigby et al., 1977) 32p-labeled plasmid, pfP1450-3 ', containing a 1.4-kb rainbow trout 3-MC-induced P450IA1 cDNA insert (a gift to K.S. from D. W. Nebert). Filters were washed initially in agitating 2.0x SSC/0.1% SDS three times at room temperature and then to a final stringency of 0.5x SSC/0.1% SDS for 30 min at 65°C. Filters were then exposed to Kodak XAR-5 film at -80°C with intensifying screens for 1-3 days. P450IA mRNA concentrations were quantified from autoradiographs by laser densitometry (LKB Ultroscan: LKB Instruments, Gaithersburg, MD) (Maniatis et al., 1982). To correct for potential unequal loading of mRNA in each lane, the P450IA1 probe was stripped off the filters by their immersion in boiling H 2 0 , and the H20 was allowed to cool to room temperature. Filters were then prehybridized as described above and rehybridized to a nick-translated carp [3-actin probe (a gift from Perry Hackett) as described above. Final wash conditions for S-actin were 0.4x SSC/0.1% SDS at 65°C for 30 min. Autoradiography was performed for 1-2 days with intensifying screens at -80°C. Levels of [3-actin mRNA expression were also determined from autoradiographs by laser densitometry.

Quantification of P45OIA mRNA Expression and Statistical Analysis. A single internal standard, one tomcod sample with a moderate signal for both P450IA and [3-actin expression, was electrophoresed on all gels to allow for intergel comparisons of mRNA expression. A correction factor was calculated for each filter based on the absorbance of the internal standard on that filter in comparison to its absorbance determined on a single master gel. Absorbance readings for each sample were then multiplied by the correction factor for its filter. To correct for unequal loading of total mRNA in each lane, the OD for P450IA mRNA for each sample was divided by the OD value for [3-actin obtained for that same sample on the same filter. Therefore, levels of P450IA mRNA expression for each sample were expressed per unit of total mRNA expression. Statistical analysis were performed according to standard methods. Single fac-

CYTOCHROME P450IA mRNA IN T O M C O D

69

tor analysis of variance was done with Statview (BrainPower). In those cases where ANOVA analysis rejected the hypothesis of equal means, Dunnett's test was used to determine which means were significantly different.

DNA Analysis Genomic DNA was isolated from tomcod, white perch, striped bass, hogchoker, and rainbow trout livers as described by Wirgin et al. (1990). Briefly, approximately 0.5 g of liver was pulverized under liquid nitrogen in a stainlesssteel mortar and pestle and then incubated in 5 vol of AT buffer (1 N NH4OH and 0.2% Triton X-100) for 60 rain at 37°C. DNA was then purified by standard procedures of phenol-chloroform extraction; however, two rounds of phenol extraction were performed. DNA was then precipitated in 100% EtOH and spoolable DNA was transferred to TE buffer. DNA concentrations and purity were determined spectrophotometrically at 260 and 280 nm. Ten micrograms of DNA from each species was digested with HindIII, BamHI, and EcoRI following the manufacturer's (Boehringer-Mannheim) recommendations and electrophoresed on 0.9% agarose gels, Southern blotted (Southern, 1975) to nitrocellulose filters (BA 85, Schleicher & Schuell, Keene, NH), and hybridized in the presence of dextran sulfate at 65°C (Wahl et al., 1979) to a nick-translated (Rigby et al., 1977) radiolabeled (32p) undigested plasmid, pfP1450-3', containing the P450IA1 rainbow trout cDNA insert. Filters were initially washed three times in agitating 2.0x SSC/0.1% SDS at room temperature, and final wash conditions were 0.5x SSC/0.1% SDS for 30 min at 65°C. Autoradiographs were generally exposed for 1-2 days at - 8 0 ° C with intensifying screens. RESULTS

DNA Analysis Southern blot analysis of tomcod, white perch, striped bass, hogchoker, and rainbow trout genomic DNA digested with three restriction enzymes, BamHI, EcoRI, and HindIII, and hybridized to the 3-MC-induced rainbow trout cytochrome P450IA1 cDNA probe was performed. Genomic DNA from all four Hudson River species and rainbow trout provided a strong signal when washed under conditions of moderate stringency (0.5 x SSC/0.1% SDS) (see Fig. 1). Based on these data, the maximum molecular size for the P450IA gene in these species are tomcod, 6500 bp; white perch, 10,325 bp; striped bass, 5900 bp; hogchoker, 1750 bp; and rainbow trout, 6100 bp. Both BamHI and HindIII digests of tomcod DNA from most individuals generated a single DNA fragment. BamHI digests of striped bass and rainbow trout genomic DNA and a HindIII digest of hogchoker genomic DNA also provided a single band. These findings indicate a single chromosomal locus for the P450IA gene in these species and suggest the absence of allelic polymorphisms in these tested individuals. All three restriction enzymes generated more than one fragment from white perch DNA. Levels o f P45OIA mRNA in freshly caught Hudson River and Maine tomcod.

70

KREAMER ET AL. a

b

e

d

e

f

g

h

i

j

k

I

m

n

o

FIr. 1. Southern blot of tomcod (TC), rainbow trout (RT), hogchoker (HC), striped bass (SB), and white perch (WP) genomic DNA digested with HindIII and BamHI and hybridized to the plasmid pfP1450-3' containing the P450IA1 trout cDNA insert. Final wash stringencies were 0.5 × SSC/0.1% SDS at 65°C for 30 min. Lanes a-e are HindIII digests and i-m are BamHI digests. Lane to sample designations are as follows; (a) RT, (b) WP, (c) HC, (d) TC, (e) SB, (i) RT, (j) WP, (k) HC, (1) TC, (m) SB. Lanes g and o contain a molecular weight standard, k DNA digested with EcoRI and HindIII.

Levels of P450IA mRNA were measured from the livers of freshly caught tomcod from two Hudson River sites and from a single location on the Saco River in Maine. Relatively high levels of P450IA mRNA expression were detected in tomcod from all three sites in comparison to levels measured in depurated Hudson tomcod, with a significantly higher value (mean = 1.048 + 0.319) recorded at the Garrison site on the Hudson River (P = 0.0001). It is interesting to note that levels of expression were higher in tomcod collected from the winter than in the spring Hudson collection, mRNA levels at the New York City site (mean = 0.438 +0.194) and in the Saco River (mean --- 0.427 +- 0.149) were comparable and were approximately two times lower than at Garrison. Variation in mRNA levels within fish from single collection sites was quite high, with a 50-fold difference noted between maximum and minimum values at Garrison and a 4- to 5-fold difference at the New York site. However, I0 of I 1 fish from Garrison and 15 of 15 fish from New York City exhibited gene expression levels above the calculated depurated level for that site. Significant differences were not detected in levels of mRNA between male and female tomcod collected off Garrison in January 1990, nor was any relationship observed between total length of the fish and level of mRNA. Rate and extent of depuration of environmentally induced Hudson River tomcod. Two collections of tomcod were used for this segment of the study. One group consisted of 44 fish sampled from Garrison in January-February 1990 and the second group consisted of 21 fish which were collected in March-April 1990 from the New York City site. Subsets of the Garrison tomcod were sacrificed at 0, 4, 8, 19, 120, and 480 hr post-transfer to clean laboratory water. Tomcod from New York City were sacrificed at 0 and 480 hr post-transfer. As described above,

71

C Y T O C H R O M E P450IA m R N A IN T O M C O D

levels of expression in fish sacrificed at zero time from both sites were high. The rate of P450IA mRNA depuration was quite rapid in the Garrison sample. A slight decline in expression (15%) was observed by 4 hr, by 8 hr a pronounced 75% reduction in expression was observed, and by 1-5 days basal levels were reached. Tomcod from the New York City site exhibited a significant 5-fold reduction in P450IA mRNA expression over the 20-day span (0.438 + 0.050 to 0.090 -+ 0.021) (see Fig. 2). Induction of P4501A mRNA in depurated tomcod reintroduced to the Hudson. Depurated Hudson tomcod (>20 days) were reintroduced to two Hudson River sites and maintained in cages suspended immediately above the sediment for 4 or 7 days. Although mRNA expression increased in tomcod reintroduced at a pier off New York City (mean = 0.215 -+ 0.032), this difference did not prove significant in comparison to our composite Hudson River sample of depurated tomcod (mean = 0.139 -+ 0.025). Tomcod reintroduced to Garrison did display a significant (P ~< 0.05) 3-fold increase in P450IA mRNA expression (mean = 0.403 -+ 0.256). At both sites, the levels of mRNA expression in reintroduced tomcod were about 50% of that observed in freshly caught tomcod from that site. Induction of P45OIA mRNA in tomcod injected with ~-napthoflavone. A group of depurated (>20 days) Garrison tomcod were ip injected with [3-NF and sacrificed after 26 hr. An approximate 15-fold induction in P450IA mRNA expression (mean = 2.414 -+ 0.296) was observed in comparison to corn oil-injected depu1.6"

(11)

•

Garrison

[]

N.Y.C.

1.4

1.2 ;[ O 1.0

0.8

O Z 0.6

(4)

(5) (14) 0.2

0.0

. 0

.

. 4

. 8

.

. 19

.

120

480

Time In clean water, h r $ ,

FIG. 2. Time course of decline of P450IA1 m R N A in H u d s o n River tomcod.

72

KREAMER ET AL.

rated controls (mean = 0.169 + 0.033) (Fig. 3). All fish exhibited an induced phenotype, with induction levels ranging from 10- to 20-fold in individual fish. Two other groups of freshly caught, nondepurated Garrison tomcod were ip injected with [3-NF, transferred to clean water, and sacrificed 8 and 19 hr posttreatment. A 2.5-fold (mean = 0.702 +-_ 0.418) induction was observed in fish sacrificed after 8 hr in comparison to a control group of untreated Garrison fish also sacrificed 8 hr after Hudson collection (mean = 0.278 -+ 0.048). However, only one [3-NF-injected fish in this group exhibited a highly induced phenotype (OD = 2.34). At 19 hr post-treatment, a 2.2-fold induction (mean = 0.416 +- 0.070) was observed in treated fish in comparison to nontreated Garrison fish sacrificed 19 hr after transfer to clean water (mean = 0.190 +- 0.081). Induction of P45OIA mRNA by exposure of tomcod to PAH and contaminated Hudson sediment. Depurated Hudson tomcod (>20 days) were exposed to Long Island Sound sediment spiked with two PAHs, fluoranthene and phenanthrene, in a noncirculating 50-gallon aquarium maintained at 4°C for 3 to 5 days. A control group of depurated tomcod was maintained under similar conditions over noncontaminated Long Island Sound sediment. Fish exposed to contaminated sediment for 3 days exhibited a 2-fold induction, while tomcod exposed for 5 days showed a 7-fold induction in comparison to tomcod maintained over control sediment for a similar length of time (Fig. 4). No difference in mRNA expression was observed between depurated tomcod (mean = 0.160 -+ 0.033) and depurated tomcod exposed to control sediment (mean = 0.145 -+ 0.062), and therefore we conclude that the control sediment did not elicit an induced response (Table 2). More interestingly, exposure of depurated Hudson tomcod to naturally contaminated Hudson sediment also elicited an overall mean 7-fold induction in P450IA mRNA expression in comparison to the response of the control group exposed to clean sediment (Fig. 4). However, only five fish in this group displayed the induced phenotype, with expression levels in this subset exhibiting a 15-fold induction. DISCUSSION We found high levels of P450IA mRNA expression in Hudson River tomcod sacrificed immediately after collection from the Hudson. This result shows that levels of hybridizable P450IA mRNA in freshly caught Hudson tomcod are nat-

a

b

c

d

e

f

g

h

i

j

k

I

rn

FIG. 3. Northern blot of tomcod RNA hybridized to the plasmid pfP1450-3'. Final wash stringencies were 0.5x SSC/0.1% SDS at 65°C. Lanes a--d contain P450IA hybridizable mRNA from depurated Hudson tomcod, lanes e-h have P450IA mRNA from p-NF ip injected tomcod sacrificed after 8 hr, lanes i-m contain P450IA mRNA from [3-NF ip injected tomcod sacrificed after 24 hr.

73

CYTOCHROME P450IA mRNA IN T O M C O D

A

a

b

c

d

B

a

b

c

d

e

f

e

g

f

g

h

i

j

k

I

m

h

i

j

k

1

m

FIG. 4. (A) Northern blot of tomcod P450IA m R N A from fish exposed to contaminated Hudson and clean sediment in aquaria and hybridized to the plasmid pfP1450-3'. Lanes a-g contain tomcod P450IA rnRNA from fish exposed to contaminated sediments while lanes h-m depict levels of P450IA m R N A from tomcod exposed to clean sediment. (B) Northern blot of tomcod RNA stripped of the P450IA probe and rehybridized to the carp [3-actin probe. Final wash stringencies were 0.5 × SSC/0.1% SDS at 65°C.

urally induced and sufficiently high to be detectable by Northern blot analysis. This confirms that measurement of P450IA mRNA expression in Atlantic tomcod is sufficiently sensitive to serve as a biomarker of exposure to some exogenous Hudson River agent(s). Tomcod collected from two Hudson River sites showed induced P450IA phenotypes in comparison to depurated tomcod (>20 days) from the same sites. For example, tomcod collected from Garrison, New York, and New York City showed a 6.5- and a 5-fold induction, respectively, in mRNA expression. The lack of differentiation in mRNA expression levels in tomcod from these two geographically distant (50 miles) and likely xenobiotically dissimilar sites can be explained based on the within-river movements of Hudson River tomcod. Several studies have shown that Hudson tomcod undergo seasonal upriver movements in the late fall and downriver migrations in the late winter (Klauda et al., 1988; McLaren et al., 1988). Therefore, it is likely that at both sites we were quantifying gene expression in the same contingent of tomcod. Tomcod collected from the Saco River, Maine, showed levels of cytochrome P450IA mRNA expression lower than those of the composite Hudson River sample. We were unable to determine if these values represent environmentally inTABLE 2 EFFECTS OF SEDIMENT EXPOSURE ON P4501A MRNA LEVELS P4501A

Treatment

No. fish

mRNA levels ~

Clean sediment Spiked sediment 3-day exposure 5-day exposure Hudson contaminated sediment

6

0.145 -+ 0.062

7 4 11

0.321 _+ 0.072 0.975 -+ 0.435* 1.037 _+ 0.304*

Optical density units for P450IA m R N A per unit of total mRNA corrected for intergel differences. * Significantly different from clean sediment control (P ~< 0.05).

74

K R E A M E R ET A L.

duced mRNA expression due to our lack of a depurated Maine sample. However, if we compare the mean level of mRNA expression in the Maine sample to our composite depurated Hudson River sample we find a higher level of mRNA expression in the Maine fish. This result suggests that P450IA mRNA expression is induced in Saco River tomcod assuming that their basal level of expression is comparable to that observed in our depurated Hudson sample. Currently, chemical analyses for sediments are not available for the Saco River although levels of contaminants are believed to be much lower than those in the Hudson River (Cormier and Racine, 1990). Levels of P450IA mRNA expression rapidly declined when environmentally induced Hudson River tomcod were introduced into clean water. At approximately 20 hr post-transfer, levels of mRNA expression were less than 20% of their environmentally induced state, and near basal levels were achieved by 120 hr and remained at these low levels for >20 days. Similar trends were observed in tomcod from both Hudson River sites and during two seasons, mid-winter and early spring. These results suggest that the use of P450IA mRNA expression in fishes as a biomarker to some organic pollutants such as PAHs may provide information on only the very recent exposure history of the individual. The rate of P450IA mRNA depuration in environmentally induced tomcod may provide insight into the identification of the Hudson River inducing agent(s). We have found a rapid decline in P450IA mRNA levels in tomcod ip injected with [3-NF, and other workers have reported similar results in [3-NF-injected rainbow trout (Haasch et al., 1989) and killifish (Kloepper-Sams and Stegeman, 1989). For example, we noted maximum P450IA mRNA levels in ~-NF ip injected tomcod 24 48 hr post-treatment, and a rapid decline in mRNA expression was noted thereafter (Wirgin et al., unpublished data). Killifish and rainbow trout injected with [3-NF also exhibited peak P450IA mRNA levels 18-24 hr post-treatment with control levels reached within 5 days (Kloepper-Sams and Stegeman, 1989; Haasch et al., 1988). In contrast, Gooch et al. (1989) detected a 2-fold induction of P450IA1 mRNA in the saltwater fish, scup, 5 days after the injection of the PCB congener, 3,3',4,4'-tetrachlorobiphenyl (TCB). Furthermore, analysis of PAH (Dey et al., 1986) and PCB (Klauda et al., 1981) concentrations in naturally exposed Hudson River tomcod livers have shown high concentrations of PCBs. However, PAHs were not detectable. It has been speculated that the inability to detect PAHs in Hudson tomcod livers may reflect their rapid metabolism (Dey et al., 1986). As evidence, O'Connor and Squibb (1988) have shown the extremely rapid clearance of phenanthrene in another Hudson River finfish species, the striped bass, and similar results were reported in DMBA and B(a)P-exposed rainbow trout (O'Connor et al., 1988). Investigations on the temporal aspects of induction and depuration of P450IA mRNA expression in depurated Hudson tomcod by exposure to pure PCB congeners, Aroclor mixtures, PAHs, and dioxins should provide further insights. To date, most studies quantifying levels of cytochrome P450IA expression in fishes have measured catalytic activities or immunoreactivity of the protein products of this locus (Stegeman and Kloepper-Sams, 1987). Levels of protein induction generally ranged between 4- and 100-fold for [3-NF- or 3-MC-treated fish

CYTOCHROME P450IA mRNA IN T O M C O D

75

(Stegeman and Kloepper-Sams, 1987; Haasch et al., 1988; Gooch et al., 1989). Using the 3-MC-induced cDNA probe from rainbow trout, Heilmann et al. (1988) showed an approximate 10-fold induction of hepatic P450IA1 mRNA expression in 3-MC-treated rainbow trout. Induction levels of P450IA1 m R N A in [3-NF-injected scup, brook trout, and kiUifish also ranged from 5- to 15-fold. We found a very similar response in depurated Hudson tomcod which were ip injected with equivalent doses of [3-NF. We report an approximate 15-fold induction in P450IA mRNA at 26 hr post-treatment, a peak level of mRNA induction in tomcod which is quite similar to that reported in killifish and trout. We also investigated temporal aspects of P450IA induction in depurated Hudson tomcod ip injected with p-NF. We have seen an initial induction at 8 hr, peak expression at 24 hr post-treatment, a decline by 3 days, and basal levels achieved by 5-7 days (Wirgin et al., unpublished data). It is interesting to note the different response of depurated versus nondepurated Hudson tomcod to ip injection of [3-NF. All tomcod that had been depurated for >20 days exhibited a highly induced P450IA mRNA response to ip injection of p-NF (N I> 30 fish) (Wirgin et al., unpublished data). In contrast, the response of nondepurated Hudson tomcod has been much more variable. Some fish are highly induced while others show little or no response. It is likely that recent exposure history of the fish plays a major role in determining the sensitivity of the response of the P450IA locus to induction. The cytochrome P450IA1 cDNA probe isolated from 3-MC-induced rainbow trout (Heilmann et al., 1988) had sufficient homology to hybridize under conditions of moderate stringency (0.5x SSC at 65°C) to genomic DNA from all four species of Hudson River fish. Strong signals were obtained in all cases in Southern blot analysis, suggesting the potential universal use of this probe in a wide variety of fish species for the purpose of biomonitoring. These results extend the earlier findings of Haasch et al. (1989) who obtained a hybridization signal using this probe on snake and turtle mRNA. The fact that B a m H I and E c o R I digests of most tomcod and all striped bass DNA generated single DNA fragments indicates a single cytochrome P450IA locus in these species. Heilmann et al. (1988) reported that B a m H I and E c o R I digests of rainbow trout DNA also generated single DNA fragments. They took this as evidence of a single P450IA gene locus in rainbow trout. There have been suggestions of multiple P450s induced in some fishes (Stegeman, 1989) based on catalytic profiles of EROD and A H H activity (Leaver et al., 1988) or mRNA structure (Haasch et al., 1989). In both Northern and Southern blot analyses, we have found an RFLP in the structure of the cytochrome P450IA gene in some Hudson River tomcod (Wirgin et al., in press). In addition to exhibiting the normal 6500-bp fragment in B a m H I digests, these individuals also displayed a second 6000-bp DNA fragment. Absorbance of the two DNA fragments in polymorphic individuals was approximately equal. Mammals do have a second gene, P450IA2, which heretofore has gone undetected in fish (Stegeman, 1989). We believe that this is the initia! confirmation of the presence of a second form of the P450IA gene in fish, although its molecular basis has yet to be characterized.

76

KREAMER ET AL.

We have shown that levels of cytochrome P450IA mRNA expression in naturally exposed Hudson River tomcod are sufficiently above background depurated levels to indicate the potential environmental presence of exogenous inducing agents. Thus, unlike most laboratory studies, we have shown that this marker system is sufficiently sensitive to detect environmentally realistic concentrations of xenobiotic agents. Induction of this P450 phenotype by laboratory exposures of depurated tomcod to contaminated sediments and by the reintroduction of caged fish to Hudson River sites further highlights the potential application of this assay as a broad biomarker of biologically relevant recent exposure to some Hudson River organic contaminant(s).

ACKNOWLEDGMENTS This work was supported in part by NIH Grants ES 04895, ES 5003, ES 00260, and CA 13343. We thank Dr. Perry Hackett for supplying us with the carp 13-actin probe and Dr. Daniel W. Nebert for kindly providing the plasmid, pf~P450-3'. Paul Geoghegan and Mark Mattson of Normandeau Associates and Dr. Dennis Duning of the New York Power Authority generously made arrangements for us to obtain tomcod from New York City. We also thank Kathy Drew of the River Project for her assistance. Finally, we thank Dr. Susan Cormier for providing our Maine tomcod collection.

REFERENCES Bigelow, H., and Schroeder, W. (1953). Fishes of the Gulf of Maine. US Fish Wildl. Serv. Fish. Bull. 74, 1-577. Chomczynski, P., and Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156-159. Cooper, K. R., Prince, R., Haasch, M. L., Wejksnora, P. J., and Lech, J. J. (1990). Detection of cytochrome P450 induction in feral and caged fish. Toxicologist 10, 249. Cormier, S. M., Racine, R. N., Smith, C. E., Dey, W. P., and Peck, T. H. (1989). Hepatocellular carcinoma and fatty infiltration in the Atlantic tomcod, Microgadus tomcod (Walbaum). J. Fish Dis. 12, 105-116. Cormier, S. M., and Racine, R. N. (1990). Histopathology of Atlantic tomcod: A possible monitor of xenobiotics in northeast tidal rivers and estuaries. In "Biological Markers of Environmental Contamination" (J. F. McCarthy and L. R. Shugart, Eds.). Lewis, Chelsea, MI. Dey, W., Peck, T., Smith, C., Cormier, S., and Kreamer, G.-L. (1986). " A Study of the Occurrence of Liver Cancer in Atlantic Tomcod (Microgadus tomcod)," final report to the Hudson River Foundation, New York, NY. Gooch, J. W., Elskus, A. A., Kloepper-Sams, P. J., Hahn, M. E., and Stegeman, J. J. (1989). Effects of ortho- and non-ortho substituted polychlorinated biphenyl congeners on the hepatic monooxygenase system in scup (Stenotornus chrysops). ToxicoL Appl. Pharmacol. 98, 422--433. Haasch, M. L., Kleinow, K. M., and Lech, J. J. (1988). Induction of cytochrome P-450 mRNA in rainbow trout: In vitro translation and immunodetection. Toxicol. Appt. Pharmacol. 94, 246-253. Haasch, M. L., Wejksnora, P. J., Stegeman, J. J., and Lech, J. J. (1989). Cloned rainbow trout liver P1450 complementary DNA as a potential environmental monitor. Toxicol. Appl. Pharmacol. 98, 362-368. Heilmann, L. J., Sheen, Y.-Y., Bigelow, S. W., and Nebert, D. W. (1988). The trout P4501Al: cDNA and deduced protein sequence, expression in liver, and evolutionary significance. DNA 7, 379398. Horn, E. G., Hetling, L. J., and Tofflemire, T. J. (1979). The problem of PCB's in the Hudson River system. Ann. N.Y. Acad. Sci. 320, 591-609. Klauda, R. J., Moos, R. E., and Schmidt, R. E. (1988). Life history of Atlantic tomcod, Microgadus tomcod, in the Hudson River estuary, with emphasis on spatio-temporal distribution and move-

CYTOCHROME P450IA mRNA IN TOMCOD

77

ments. In "Fisheries Research in the Hudson River" (C. L. Smith, Ed.), pp. 219-251. The Hudson River Environmental Society. Klauda, R. J., Peck, T. H., and Rice, G. K. (1981). Accumulation of polychlorinated biphenyls in Atlantic tomcod (Microgadus tomcod) collected from the Hudson River estuary, New York. Bull. Environ. Contamin. Toxicol. 27, 829-835. Kloepper-Sams, P. J., and Stegeman, J. J. (1989). The temporal relationships between P450E protein content, catalytic activity, and m R N A levels in the teleost Fundulas heteroclitus following treatment with [3-napthoflavone. Arch. Biochem. Biophys. 268, 525-535. Leaver, M. J., Burke, M. D., George, S. G., Davies, J. M., and Raffaelli, D. (1988). Induction of cytochrome P-450 monooxygenase activities in plaice by 'model' inducers and drilling muds. Mar. Environ. Res. 24, 27-30. Limburg, K. E. (1988). PCB's in the Hudson. In "The Hudson River Ecosystem" (K. E. Limburg, M. A. Moran, and W. H. McDuweI1, Eds.). Springer-Verlag, New York. McLaren, J. B., Peck, T. H., Dey, W. P., and Gardiner, M. (1988). Biology of the Atlantic tomcod in the Hudson River estuary. In "Science, Law, and Huson River Power Plants." (L. W. Barnthouse, R. J. Klauda, D. S. Vaughan, and R. L. Kendall, Eds.), Vol. 4, pp. 102-112. American Fisheries Society Monograph. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982). "Molecular Cloning: A Laboratory Manual." Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Nebert, D. W., and Gonzalez, F. J. (1987). P450 genes: Structure, evolution, and regulation. Annu. Rev. Biochem. 56, 945-993. O'Connor, J. M., Klotz, J. B., and Kneip, T. J. (1982). Sources, sinks and distribution of organic contaminants in the New York Bight ecosystem. In "Ecological Stress and the New York Bight: Science and Management" (G. F. Mayer, Ed.). Estuarine Research Federation, Columbia, SC. O'Connor, J. M., and Rachlin, J. W. (1982). Perspective of metals in New York Bight organisms: Factors controlling accumulation and body burdens. In "Ecological Stress and the New York Bight: Science and Management" (G. F. Mayer, Ed.). Estuarine Research Federation, Columbia, SC. O'Connor, J. M., Schnitz, A. R., and Squibb, K. S. (1988). In vivo kinetics of benzo(a)pyrene and 7,12-dimethylbenz(a)anthracene assimilation and metabolism in rainbow trout. Mar. Environ. Res. 24, 63-67. O'Connor, J. M., and Squibb, K. S. (1988). "Bioaccumulation of Polycyclic Aromatic Hydrocarbons and Metals in Estuarine Organisms," final report submitted to the American Petroleum Institute, Washington, DC. Rigby, P. W. J., Dieckmann, M., Rhodes, C., and Berg, P. (1977). Labeling dexyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase. I. J. Mol. Biol. 113, 237-251. Rohmann, S. O., and Lilienthal, N. (1987). "Tracing a River's Toxic Pollution." Inform, New York. Smolarek, T. A., Morgan, S. L., Moynihan, C. G., Lee, H., Harvey, R. G., and Baird, W. M. (1987). Metabolism and DNA adduct formation of benzo(a)pyrene and 7,12-dimethylbenz(a)anthracene in fish cell fines in culture. Carcinogenesis 8, I501-1509. Southern, E. M. (1975). Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mot. Biol. 98, 503-517. Stegeman, J. J. (1989). Cytochrome P-450 forms in fish: Catalytic, immunoIogical and sequence similarities. Xenobiotica 19, 1093-1110. Stegeman, J. J., and Kloepper-Sams, P. J. (1987). Cytochrome P-450 isozymes and monooxygenase activity in aquatic animals. Environ. Health Perspeet. 71, 87-95. Van Veld, P. A , Westbrook, D. J., Woodin, B. R., Hale, R. C., Smith, C. L., Huggett, R. J., and Stegeman, J . J . (1990). Induced cytochrome P-450 in intestine and liver of spot (Leiostomus xanthurus) from a polycyclic aromatic hydrocarbon contaminated environment. Aqua. Toxicol. I7, 11%132. Varanasi, U., Reichert, W. L., Eberhart, B.-T. L., and Stein, J. E. (1989). Formation and persistence of benzo(a)pyrene-diolepoxide-DNA adducts in liver of English sole (Parophrys vetulus). Chem. Biol. Interact. 69, 203-216.

78

KREAMER ET AL.

Wahl, G. M., Stern, M., and Stark, G. R. (1979). Efficient transfer of large DNA fragments from agarose gels to diazobenyloxymethyl-paper and rapid hybridization by using dextran sulfate. Proc. Natl. Acad. Sci. USA 76, 3683-3687. Williams, S. C., Simpson, H. J., Olsen, C. R., and Bopp, R. F. (1978). Sources of heavy metals in sediments of the Hudson River estuary. Mar. Chem. 6, 195-213. Wirgin, I., Currie, D., and Garte, S. J. (1989). Activation of the K-ras oncogene in liver tumors of Hudson River tomcod. Carcinogenesis 10, 2311-2315. Wirgin, I. I., D'Amore, M., Grunwald, C., Goldman, A., and Garte, S. J. (1990). Genetic diversity at an oncogene locus and in mitochondrial DNA between populations of cancer-prone Atlantic tomcod. Biochem. Genet. 28, 459-475. Wirgin, I., Kreamer, G.-L., and Garte, S. J. (in press). Genetic polymorphism of cytochrome P-450IA in cancer-prone Hudson River tomcod. Aqua. Toxicol.