acute sodium arsenite administration induces pulmonary cyp1a1 mrna, protein and activity in the rat

J BIOCHEM MOLECULAR TOXICOLOGYVolume 16, Number 2, 2002

Acute Sodium Arsenite Administration InducesPulmonary CYP1A1 mRNA, Protein and Activityin the RatJohn M. Seubert, Christopher J. Sinal, and John R. BendDepartment of Pharmacology and Toxicology, University of Western Ontario, London, Ontario, Canada N6A 5C1; E-mail: [email protected]

Received 29 November 2001; revised 18 February 2002; accepted 20 February 2002

ABSTRACT: Modulation of the cytochrome P450 (CYP)monooxygenase system (P450) by arsenite was inves-tigated in male, adult Sprague-Dawley rats treatedwith a single dose (75 mmol/kg, sc) of sodium ar-senite (As3+). Total CYP content and P450-dependent7-pentoxyresorufin O-pentylation (PROD) and 7-ethoxyresorufin O-deethylation (EROD) activities ofliver microsomes decreased maximally (33, 35, and 50%of control, respectively) 1 day after As3+ treatment.Maximum decreases of CYP content and P450 catalyticactivities corresponded with maximum increases of mi-crosomal heme oxygenase (HO) activity and with in-creased total plasma bilirubin concentrations. ERODactivity increased maximally in lung (300%) 5 days aftera single dose of As3+. Lung CYP1A1 mRNA and proteinlevels also increased maximally 5 days after treatment.A small but significant increase in EROD activity (65%)was observed in lung microsomes 24 h following a 1 hinfusion of bilirubin (7.5 mg/kg) into rats. However, ad-ministration of bilirubin to the lung via intratrachealinjection (0.25 and 2.5 mg/kg) did not increase CYP1A1monooxygenase activity or mRNA. This study demon-strates that P450 is modulated in an isozyme (CYP1A1vs CYP2B1/2) selective manner in rat lung after acuteAs3+ administration. Administration of bilirubin, apotential aryl hydrocarbon receptor (AHR) ligand, byinfusion or intratracheal instillation did not upreg-ulate pulmonary CYP1A1 at the mRNA level underour treatment conditions. C© 2002 Wiley Periodicals, Inc.J Biochem Mol Toxicol 16:84–95, 2002; Published on-line in Wiley Interscience (www.interscience.wiley.com).DOI 10.1002/jbt.10022

KEYWORDS: Cytochrome P450; Arsenite; Oxidativestress; Heme oxygenase; Lung; Bilirubin

Correspondence to: John R. Bend.Present address of Christopher J. Sinal: Department of Pharma-

cology, Dalhousie University, Sir Charles Tupper Medical Building,5859 University Avenue, Halifax, Nova Scotia, Canada B3H 4H7.

Contract Grant Sponsor: Canadian Institutes of Health Research.Contract Grant Number: FRN 9972.

c© 2002 Wiley Periodicals, Inc.

INTRODUCTION

The cytochrome P450 (CYP) monooxygenasesystem (P450) is critical for the metabolism of bothendogenous and exogenous lipophilic substrates. CYPis a gene superfamily of heme-containing isozymes [1].The transcriptional regulation of the CYP subfamilies1A and 1B is mediated by the aryl hydrocarbon receptor(AHR), a ligand-activated transcription factor belong-ing to the basic helix loop helix (bHLH)/Per, ARNT,and SIM (PAS) family. The unliganded AHR residesin the cytoplasm complexed with heat shock protein90 (Hsp90) and AH receptor-interacting protein (AIP),an immunophilin type chaperone protein [2,3]. Thebinding of a ligand to the AHR initiates a transforma-tion, triggering the release of Hsp90 and AIP, therebyallowing the ligand/AHR complex to translocate intothe nucleus where it forms a heterodimer with theAHR nuclear translocator (ARNT) protein [4,5]. TheAHR/ARNT heterodimer has a high-affinity for specificDNA recognition sequences, known as dioxin respon-sive elements (DRE) [3,6]. These enhancer sequences,located upstream of the CYP1A or 1B transcriptionstart sites, promote gene transcription upon bindingof the AHR/ARNT heterodimer [6]. The interaction ofplanar polyaromatic ligands, such as benzo[a]pyreneor TCDD with the AHR has been extensively studied,however analogous roles for endogenous compoundsare poorly understood. We recently reported thatbilirubin upregulates CYP1A1 gene expression andCYP1A1 enzymatic activity in hepatoma Hepa 1c1c7cells in an AHR-dependent manner [7]. Subsequently,Phelan et al. [8] reported that bilirubin and biliverdinare AHR ligands which activate CYP1A1 expression inintact cells from several different species.

The mechanism(s) by which pathophysiologicalstressors such as oxidant stress following sodium ar-senite (As3+) administration [9–11] or orthotopic livertransplantation [12], selectively upregulate pulmonary

84

Volume 16, Number 2, 2002 MODULATION OF CYP1A1 BY SODIUM ARSENITE 85

CYP1A1-dependent catalytic activity in rats is currentlyunknown. As3+ is a naturally occurring metalloid asso-ciated with an increased risk of cancer in certain highlyexposed populations [13]. Induction of various hep-atic chemical detoxication enzymes such as NAD(P)H:quinone oxidoreductase [9] occurs rapidly followingAs3+ treatment. Similarly, induction of stress proteinsincluding heme oxygenase-1 (HO-1), a heat shock pro-tein, by As3+ is well known [14]. In contrast, acuteAs3+ treatment is known to decrease hepatic microso-mal CYP content and catalytic activity [10]. However,comparatively little is known regarding the extrahep-atic effects of this metalloid on the P450 system. Thisis of potential interest with respect to the pivotal roleof these enzymes in the metabolic activation of pro-carcinogens [15] and the fact that extrahepatic organs,such as the lung, are known sites of As3+-mediatedcarcinogenesis [13].

Previously, we reported that As3+ treatment poten-tiates the induction of guinea pig pulmonary CYP1A1-dependent 7-ethoxyresorufin O-deethylation (EROD)activity by b-naphthoflavone (bNF) [9]. This resultis relevant with respect to the known role for thisisozyme in the bioactivation of polyaromatic carcino-gens present in complex mixtures of organic com-pounds, metals, and particulates [15]. Similarly, wefound that As3+, in the absence of known CYP inducers,selectively increases CYP1A1-dependent EROD activ-ity in rat lung versus liver [11]. Thus, evidence exists in-dicating that acute As3+ administration, perhaps via anoxidative stress response, can increase CYP monooxy-genase activity in a tissue (lung) and isozyme (CYP1A1)selective manner in rats.

Heme oxygenase (HO) has the potential to reg-ulate P450 activity. For example, control of the levelof intracellular heme via degradation by HO maybe an important regulatory mechanism which limitsheme availability for incorporation into apoprotein toform functional CYP. Reactive oxygen species (ROS)resulting from As3+ administration that upregulateHO-1 might denature CYP isozymes and cause therelease of heme. AHR-dependent induction of CYP1A1gene expression and enzyme activity by bilirubin [7,8]formed from HO-1 might modulate CYP expression.Thus, increased production of bilirubin from hememetabolism in various tissues, including liver andkidney, subsequent to HO-1 induction could play arole in AHR-mediated toxicity. In addition, bilirubin isknown to be an efficient in vitro scavenger of ROS, so,increased production of biliverdin and bilirubin fromheme metabolism may also have a beneficial antiox-idant property in serum of an organism subjected tooxidative stress [16].

The objective of this study was to investigate thepotential mechanisms for the induction of pulmonary

CYP1A1 activity subsequent to As3+ administration.More specifically, we determined the time course ofthe effects of As3+ treatment on CYP1A1 and 2B1 con-tent, catalytic activity, mRNA, and protein levels in lungover a period of several days. In addition, we mea-sured the catalytic activity and mRNA levels of HO-1, the rate-limiting enzyme of heme degradation, andserum bilirubin levels in these rats. To further investi-gate the possible involvement of bilirubin in the induc-tion of pulmonary CYP1A1 subsequent to As3+ admin-istration, we infused this compound via the tail vein ortrachea into rats. This mimics at least one of the effectsassociated with As3+ treatment, the marked increase inserum bilirubin levels.

MATERIALS AND METHODS

Chemicals

As3+, NAD(P)H, hemin, and n-lauroylsarcosinewere purchased from Sigma (St. Louis, MO); bilirubinIX and mesobilirubin were from Porphyrin Products(Logan, UT); 7-pentoxyresorufin, 7-ethoxyresorufin,and resorufin were from Molecular Probes (Eugene,OR); [g -32P]dCTP (>3000 Ci/mmol) from ICN Biomed-icals Canada (Montreal, PQ); Prime-a-Gene randomprimer DNA labelling kit from Promega (Madison,WI); Hybond-N nylon filters from Amersham Canada(Oakville, ON); restriction enzymes from Pharma-cia Canada (Baie d’Urfe, PQ); Taq DNA polymeraseand Superscript II RNase H-reverse transcriptase werefrom GIBCO BRL Canada (Burlington, ON); PCRprimers were synthesized and purified by high per-formance liquid chromatography by General Synthesisand Diagnostics (Toronto, ON); and all other chemicals(reagent grade or better) were purchased from BDH(Toronto, ON) or Sigma (St. Louis, MO).

Animal Treatment

Male Sprague-Dawley rats (250–300 g; CharlesRiver, St. Constant, PQ) were injected subcutaneously(sc) with As3+ in saline at 75 mmol/kg. Control ani-mals were injected with saline alone. Standard rat chowand water were provided ad libitum. The animals werekilled by CO2 asphyxiation at the specified intervals af-ter a single injection. This experimental protocol wasreviewed and approved by the animal use committeeat the University of Western Ontario.

Preparation of Microsomes

Lung and liver microsomes were prepared bydifferential centrifugation of homogenized tissue as

86 SEUBERT, SINAL, AND BEND Volume 16, Number 2, 2002

previously described [12]. Briefly, individual lungs andlivers were rapidly removed and placed on ice. Afterweighing, the tissues were rapidly homogenized in 3volumes of ice-cold 0.1 M potassium phosphate buffercontaining 0.15 M KCl (pH 7.4), and centrifuged at10,000× g for 10 min. The supernatant fractions werethen centrifuged at 105,000× g for 60 min. The re-sulting microsomal pellet was resuspended in 0.1 Mpotassium phosphate buffer (pH 7.4), and stored at−80◦C until use. Microsomal protein concentrationswere determined by the method of Lowry et al. [17]using bovine serum albumin as the standard. Livermicrosomal CYP content was determined from thecarbon monoxide difference absorption spectrum ofdithionite reduced microsomes with ε = 91 mM−1 cm−1

[18].

Assays of Microsomal MonooxygenaseActivity

Rates of 7-pentoxyresorufin O-depentylation(PROD) and EROD were determined as describedpreviously [12]. Measurements were performed influorimeter cuvettes maintained at 37◦C with stirringin a Perkin-Elmer LS-5B fluorescence spectropho-tometer. Incubations contained 0.3 mg (liver andlung) of microsomal protein in 0.1 M potassiumphosphate buffer (pH 7.8) and 7-pentoxyresorufin or7-ethoxyresorufin (in Me2SO) with final concentrationsof 4 and 1 mM, respectively. Reactions were startedby the addition of NADPH to a final concentration of1 mM. Changes in fluorescence (excitation= 525 nm;emission= 585 nm) were monitored for 10 min, andthe rate of formation of resorufin was calculated bycomparison with known amounts (5 pmol) of resorufinadded to the incubations.

Assay of Microsomal Heme Oxygenase(HO) Activity

The rate of microsomal HO activity was deter-mined essentially as described previously [12]. Briefly,incubations contained 1 mg/mL microsomal protein,1.5 mg/mL 105,000× g supernatant protein from con-trol liver homogenate as a source of biliverdin reduc-tase, and 25 mM hemin. Incubations were performed at37◦C using 0.1 M potassium phosphate buffer (pH 7.4)and were initiated by adding NADPH to a final con-centration of 400 mM. The amount of bilirubin formedwas quantitated from the absorbance change at 470 nmrelative to 530 nm, using an extinction coefficient= 40mM−1 cm−1 [19]. Product formation was verified to belinear with time and protein concentration under thesereaction conditions.

cDNA Probes

The 1406 bp cDNA probe for rat CYP1A1 mRNAwas obtained by EcoRI/PstI digestion of plasmid pA8[20], which was generously provided by Dr. RonaldHines (Medical College of Wisconsin). The 484 bpcDNA probe for rat CYP2B1/2 mRNA was obtainedby HindIII/NcoI digestion of plasmid pB7 [21] whichwas generously provided by Dr. Alan Anderson (LavalUniversity). A cDNA probe for rat HO-1 mRNA wasprepared in our laboratory as described previously[7]. All probes were 32P-labeled by the random primermethod according to the manufacturer’s (Promega)instructions.

Preparation and Analysis of Total RNA

Total cell RNA was isolated by the acid/guanidinium thiocyanate/phenol/chloroform extrac-tion method of Chomczynski and Sacchi [22]. Northernblot analysis of total RNA was performed as describedelsewhere [23]. Briefly, aliquots of RNA were sepa-rated in a denaturing (2.2 M formaldehyde) agarose(1.2%) gel and transferred to Hybond-N nylon filters.The RNA was fixed to the filters by baking at 80◦Cfor 2 h. Prehybridization of the filters was carried outin a solution containing 6× SSC (0.9 M NaCl, 0.09 Msodium citrate), 50% deionized formamide, 5× Den-hardt’s solution (0.1% polyvinylpyrrolidone, 0.1% Fi-coll, 0.1% bovine serum albumin), 0.5% sodium dodecylsulfate (SDS), and 100 mg/mL sheared salmon spermDNA for 4 h at 42◦C. Hybridization with the 32P-labeledcDNA probes was carried out in the same solution, mi-nus Denhardt’s reagent, for 16–24 h at 42◦C. The fil-ters were then washed twice at room temperature in2× SSC, 0.5% SDS for 15 min. This was followed by a30 min wash in 0.1× SSC, 0.5% SDS at 42◦C and a fi-nal 30 min wash in 0.1× SSC, 0.5% SDS at 65◦C for 30min. The washed filters were sealed in plastic wrap andexposed to Kodak X-OMAT AR film in the presence ofan intensifier screen at −80◦C. Hybridization signalswere quantitated relative to the signals obtained forglyceraldehyde-3-phosphate dehydrogenase (GAPDH)mRNA.

Western Immunoblotting Analysis

Proteins (10 mg) from lung microsomes preparedfrom rats killed 1, 3, 5, or 7 days following treat-ment with As3+ were resolved by denaturing elec-trophoresis on discontinuous polyacrylamide slab gels(SDS-PAGE) and were electrophoretically transferredto nitrocellulose (Amersham Arlington Heights, IL.)[23]. Protein blots were blocked for 2 h at 25◦C in buffer(0.15 M sodium chloride, 3 mM potassium chloride,


25 mM tris(hydroxymethyl)methylamine, pH 7.4 (Tris-saline buffer) with 5% skim milk powder). The blockingsolution was removed and blots were rinsed three timesin wash buffer (Tris-saline buffer containing 0.1% tween20). Immunoblot analysis was performed by incubat-ing with a primary polyclonal antirat CYP1A1/2 anti-body (Oxford Biomedical Research, Oxford, MI) for 12 hat 25◦C in Tris-saline buffer containing 0.02% sodiumazide. The primary antibody solution was removedand blots were rinsed three times with wash buffer.Blots were incubated with the monoclonal antibody toCYP1A1/2 (Amersham, Arlington Heights, IL) for 1 hand washed as previously described. Antibody detec-tion was performed using the enhanced chemilumines-cence method (Amersham, Arlington Heights, IL). Apositive control of lung microsomes from a bNF-treated(80 mg/kg for 4 days) rat was included for comparison.

Assay of Plasma Bilirubin

Whole blood was collected from saline orAs3+-treated rats under phenobarbital anesthesia(200 mg/kg) by cardiac puncture using a heparinizedbutterfly cannula apparatus. Blood was collected intoheparinized tubes and immediately centrifuged at10,000× g for 10 min at 4◦C. The plasma supernatantwas aliquoted to fresh tubes and stored at −80◦C un-til analysis (<1 week later). Colorimetric determinationof total plasma bilirubin was performed after couplingwith diazotized sulfanilic acid using a commerciallyavailable assay kit (Total Bilirubin Kit, Sigma, St. LouisMO). The concentration of total bilirubin in plasma wasquantitated by comparison with known amounts ofbilirubin standard prepared in a 1 mg/mL solution ofbovine serum albumin. All of the procedures describedabove were performed under subdued light conditionsto minimize photooxidation of bilirubin.

Bilirubin Tail Vein Infusion

Bilirubin was infused into male Sprague-Dawleyrats (∼300 g) following the method of Rhodes andPatterson [24]. Tails were immersed for about 10 minin a warm water bath (∼40◦C) to dilate the lateral cau-dal veins prior to insertion of the catheter. Rat tail veinswere catheterized with a 0.7× 19 mm intravenous (i.v.)catheter placement unit (Baxter, Becton Dickinson, UT)attached to PE 60 polyethylene tubing (Clay Adams,NJ, i.d. 0.28 mm, o.d. 0.61 mm) following applicationof a small amount of methyl salicylate to the surface ofthe tail. A sc injection of buprenorphrine (0.03 mg/kg)was administered prior to infusion as an analgesic, ac-cording to the University of Western Ontario’s animalcare policy. The catheter was secured to the tail with

medical adhesive tape and animals were restrained incages with their tails exposed through the rear slit.Bilirubin was infused in a solution (0.5 M NaOH, 0.055M sodium dihydrogen phosphate, 5% BSA, pH 7.4,2 U/mL heparin) for 1 h (0, 7.5, and 40 mg/kg/h).Bilirubin was absent from the infusion solution in con-trol animals. Because of the photolability of bilirubin,catheters were covered with aluminium foil and experi-ments were performed under conditions of dim light.Animals were sacrificed 24 h after the completion ofthe infusion and tissues were sampled as previouslydescribed.

Bilirubin Intratracheal Injection

Intratracheal injections of bilirubin into the lungsof male rats were performed using previously pub-lished procedures [25]. Briefly, rats were anesthesizedwith halothane (1–2%) and a balance of pure oxygen.A small midline incision was made to expose the tra-chea, after which a catheter was advanced through a tra-cheotomy into the left distal bronchus. A single 150 mLinjection of bilirubin (dissolved in 0.5 M NaOH, 0.055M sodium dihydrogen phosphate, 5% BSA, pH 7.4)at either 0.25 mg/kg or 2.5 mg/kg was administered.Two control groups included animals who had a tra-cheotomy but did not have an intratracheal injectionand animals who received an injection of vehicle so-lution (150 mL). These groups were included to con-trol for any adverse affects that might interfere withpulmonary CYP, such as febrile response from a possi-ble infection resulting from the bronchoalveolar lavage[26]. Animals were sacrificed 48 h after the completionof the injection and tissues were sampled as previouslydescribed.

Statistics

All experimental data were analyzed by the un-paired, Student’s t test for significant differences (p <0.05) between the As3+-treated or bilirubin-treated andthe corresponding control groups.

RESULTS

CYP Content and Modulation of CYPActivities Following As3+ Treatment

Treatment of rats with a single dose (75 mmol/kg,sc) of As3+ resulted in a significant decrease (33%) intotal hepatic CYP content after 1 day when comparedwith saline-treated controls (Figure 1). Following this,a time-dependent recovery of CYP content occurred,with no significant differences apparent between con-trol and treated animals from 2 days onward.


FIGURE 1. Time course for changes in liver microsomal CYP con-tent, EROD, PROD, and HO activities of rats 1, 2, 5, 7, or 10 daysafter treatment with saline (controls; ¤) or As3+ (¥). Each data pointrepresents the mean± standard deviation of experiments performedwith the livers of three individual animals. ∗ p < 0.05 and ∗∗ p < 0.01compared with the respective saline control group.

PROD activity is primarily catalyzed by CYP2Bisozymes in rat liver and lung, while EROD activity ismainly due to CYP1A1/1A2 isozymes in rat liver andCYP1A1 in rat lung. Thus, these activities are usefulmarkers for differential effects of acute As3+ adminis-tration on CYP-dependent catalytic activity. Consistentwith the total CYP data, significant decreases of liverPROD and EROD activity, compared to saline-treatedcontrols, were observed 1 day after a single dose of As3+

(Figure 1). There were also suggestions of decreasedlung PROD activity, however these changes were notsignificant (Figure 2). Liver PROD activity recoveredfully by 5 days after As3+ treatment and remained sim-ilar to control levels up to 10 days afterward.

FIGURE 2. Time course for changes in lung microsomal PROD,EROD, and HO activities of rats 1, 2, 5, 7, or 10 days after treatmentwith saline (controls; ¤) or As3+ (¥). Each data point represents themean± standard deviation of experiments performed with the lungsof three individual animals. ∗ p < 0.05 and ∗∗ p < 0.01 compared withthe respective saline control group.

Similar to PROD, EROD activity displayed sub-stantial, but differing degrees of recovery from 2 daysonward. For example, 2 days after treatment, livermicrosomal EROD activity of As3+-treated rats wassignificantly lower than that of saline-treated con-trols (Figure 1). Lung EROD was significantly in-creased to approximately 160% of control values 2 daysafter treatment. A marked increase of microsomalEROD activity occurred in these tissues 5 days af-ter As3+ treatment. Although liver EROD activitieswere increased to 200% of control values on day 5,only the increase in lung EROD activity (400% ofcontrol value) was significant because of individualvariation. Following this, a decline to control EROD


values was observed by 7 or 10 days after As3+

treatment.

Modulation of HO Activities FollowingAs3+ Treatment

In contrast to total CYP content, microsomal HOactivity was elevated dramatically (approximately 10-fold) by 1 day after As3+ treatment in liver (Figure 1)and to a lesser extent (approximately two fold) in lung(Figure 2). This order of induction was paralleled bythe relative order of decreases of CYP content and/ormonooxygenase activity observed for these tissues. Thetime required for the return of HO activity to basal lev-els in each of the tissues appears to be related to theinitial level of induction. For example, liver HO activitywas approximately 1100% of control values 1 day afterAs3+ treatment and did not return to control levels un-til 10 days posttreatment. In contrast, lung HO activitywas only approximately 160% of control values 1 dayafter As3+ treatment, and was not significantly differ-ent from control after 2 days. Similar to the decreasesin CYP-dependent catalytic activity 1 day after As3+

FIGURE 3. Northern blot analysis of total RNA isolated from the liver or lung of rats 1, 2, 5, 7, or 10 days after treatment with saline (C) or As3+

(A). Total RNA (25 mg) was separated on a 1.2% denaturing (formaldehyde) agarose gel, transferred to nylon membranes and hybridized with32P-labeled cDNA probes specific for CYP1A1, CYP2B1/2, HO-1, or GAPDH mRNA.

treatment, the magnitude of increase of EROD activityappears to be inversely related to the level of HO induc-tion in the three tissues studied. Specifically, in lung,HO activity was increased the least after 1 day, and re-turned to control values more rapidly than the liver, andpulmonary EROD activity was increased more dramat-ically than in the other tissues studied.

Modulation of CYP and HO-1 mRNAFollowing As3+ Treatment

To determine if the modulation of CYP-dependentEROD and PROD activity is mediated, at least in part, atthe pretranslational level, we carried out Northern blothybridization analysis for CYP1A1, CYP2B1/2, HO-1,and GAPDH mRNAs (Figure 3). Hybridization of totallung RNA blots with a cDNA probe for CYP1A1 mRNAshowed a time-dependent increase in the accumulationof this message species 1, 2, and 5 days after As3+ treat-ment when compared with saline-treated controls. Incontrast, no significant changes in the mRNA levelsfor CYP2B1/2, HO-1 or the housekeeping gene GAPDHwere apparent at any time point. Transcripts from


CYP1A1 or CYP2B1/2 were not detected in blots pre-pared from liver or kidney RNA 5 days (Figure 3) orat any other time point after saline or As3+ treatment(data not shown). Weak signals were obtained for HO-1mRNA on liver blots prepared from As3+ but not saline-treated rats 1 day after treatment (data not shown).

To ascertain whether the increase in mRNA levelsof CYP1A1 observed in the lung after As3+ treatment ledto an increase in CYP1A1 protein, Western blot analysiswas performed. Incubation of lung microsomal proteinwith anti-CYP1A1 antibody showed a time-dependentincrease in the accumulation of protein 1, 2, 5, or 7 daysafter As3+ treatment, with the greatest increases in pro-tein detected on days 2 and 5 (Figure 4).

Modulation of Plasma Bilirubin Levels

Heme metabolism and elimination of bile pigmentsare profoundly altered by As3+ treatment [26]. As anindicator of the effect of As3+ treatment on heme andbile pigment metabolism of the rats used in this study,total plasma bilirubin was measured. As shown inFigure 5, As3+ treatment caused a rapid increase inplasma bilirubin concentration, which was maximal(5.4± 1.1 mM; mean± SD, N = 3) 1 day after treat-ment. The highest concentrations of plasma bilirubincorresponded well with the time (1 day) of the great-est HO-1 induction after As3+ treatment in the tissuesstudied (Figures 1 and 2). Furthermore, a temporal cor-respondence with maximum depression of hepatic CYPcontent, PROD, and EROD activity in response toAs3+ treatment was also observed (Figure 1). Consis-tent with the modulation of hepatic HO-1 activity, to-tal plasma bilirubin remained significantly elevated inAs3+-treated vs saline-treated rats up to 7 days aftertreatment and had returned to normal levels 10 daysposttreatment.

FIGURE 4. Western blot analysis for CYP1A1 in lung microsomes 1, 2, 5, or 7 days after treatment with saline (C), As3+ (A), or bNF (80 mg/kgfor 4 days). Protein (10 mg) was resolved on 10% SDS-PAGE gels, transferred to nitrocellulose membranes and incubated with antibodies specificfor CYP1A1.

Infusion of bilirubin resulted in a maximal level(8.1± 2.0 mM; mean± SD, N = 5) of total serum biliru-bin 24 h post 1 h infusion of 40 mg/kg/h but nosignificant difference at 7.5 mg/kg/h as compared tothe control animals (Figure 6). Total serum bilirubinwas higher in the controls from the infused animal ex-periment than in controls in the As3+-treated experi-ment, which is reflected in the higher hepatic HO ac-tivity observed in the infused animals. HO activities inhepatic microsomes from bilirubin-infused rats were65± 10, 104± 14, and 81± 7 (mean± SD, N = 5), andin lung microsomes were 13± 2, 26± 7, and 20± 6pmol/min/mg protein (mean± SD, N = 5) for 0, 7.5,and 40 mg/kg, respectively.

Modulation of CYP Activities FollowingTail Vein Bilirubin Infusion

To determine whether elevated levels of serumbilirubin, such as those observed in the As3+-treated an-imals might be involved in the induction of pulmonaryCYP1A1 activity, animals were infused with differentdoses of bilirubin. Treatment of rats with an infusionof bilirubin (0, 7.5, or 40 mg/kg/h) for 1 h resulted ina nondose-dependent increase in CYP1A1-dependentEROD activity in lung microsomes of rats killed 24 hlater (Figure 5). At a dose of 7.5 mg/kg a statistically sig-nificant increase of EROD activity (65%) occurred com-pared to buffer-infused controls. However, a smallerincrease at 40 mg/kg (25%) compared to control val-ues was observed. In contrast, a decrease in ERODactivity was observed in liver microsomes (Figure 5),with a 30% reduction occurring at 7.5 mg/kg and 20% at40 mg/kg compared to buffer-infused controls. As well,there were also apparent decreases in hepatic CYP2B1-dependent PROD activity, that mirrored the decreasesof hepatic EROD in terms of bilirubin dose-response.


FIGURE 5. Lung and liver microsomal EROD and PROD activitiesof rats infused with bilirubin (7.5 or 40 mg/kg/h) (¥) or buffer (con-trols;¤) for 1 h, assayed in microsomes prepared from rats killed 24 hafter infusion. Each data point represents the mean± standard devi-ation, N = 5. ∗ p < 0.05 compared with the respective buffer controlgroup.

No changes were observed in mRNA levels for CYP1A1or 2B1 in the lungs of rats infused with bilirubin at eitherdose (data not shown).

Modulation of CYP Activities FollowingTracheal Bilirubin Injection

To further evaluate whether elevated levels ofbilirubin might be involved in the induction ofpulmonary CYP1A1 activity by As3+ treatment, ratswere injected with different doses of bilirubin directlyinto the lung. Microsomes were prepared from indi-vidual lung lobes to determine any difference result-ing from direct bilirubin administration to the left dis-tal bronchus and the nontreated right side. Changes inCYP monooxygenase activities were compared to vehi-

FIGURE 6. Time course for changes in (A) serum total bilirubin lev-els of rats administered As3+ and (B) serum total bilirubin levels ofrats infused with bilirubin (7.5 or 40 mg/kg/h) (¥) for 1 h or buffer(controls; ¤). Each data point represents the mean± standard devi-ation; N = 3 in (A) and N = 5 in (B). ∗ p < 0.05 compared with therespective saline control group.

cle control (C2) animals. Treatment of rats with a singleinjection of bilirubin (0.25 or 2.5 mg/kg) resulted in nosignificant change in CYP1A1-dependent EROD activ-ity in lung microsomes from any rats killed 48 h later(data not shown).

DISCUSSION

We previously reported increased rat lungCYP1A1-dependent EROD activity after acute As3+

treatment [11] in the rat. To our knowledge the find-ings presented in this study [11] demonstrated, for thefirst time, a selective increase of the catalytic activ-ity of a CYP isozyme after As3+ treatment. We subse-quently reported that bilirubin induces Cyp1a1 by anAHR-dependent mechanism in cultured Hepa 1c1c7cells [7]. We report here that the upregulation of pul-monary CYP1A1 by As3+ is mediated, at least in part,at the pretranslational level. This suggests the possi-bility of increased transcription due to the release ofan endogenous AHR ligand. The delayed maximum(5 days) for the pulmonary response is consistent withan indirect mechanism of induction as might occur if


formation and translocation of an endogenous com-pound(s) is responsible.

Following acute administration, As3+ is knownto affect the expression of a number of intracellularproteins. For example, it is a potent inducer of heatshock and other stress-related proteins including HO-1(HSP32), the rate-limiting enzyme in the degradation ofheme to bilirubin via biliverdin reductase [14]. The factthat we observed significant induction of HO-1 activity24 h after As3+ administration in lung (Figure 2) in theabsence of elevated HO-1 mRNA (Figure 3) is not sur-prising. Rather this reflects the fact that the first mRNAsample was not taken until 24 h following treatment.It has been demonstrated in both cell culture [27] andin vivo [28] that following acute exposure to a stressorincluding UV radiation, H2O2, stilbenes, or As3+, HO-1mRNA increases rapidly within 1 h, peaks about 12 hand returns to basal levels by 24 h. Whereas, HO cat-alytic activity continues to increase until approximately24 h when it begins to return to basal levels. The role ofHO-1 in the decrease of CYP catalytic activity after As3+

administration is not readily apparent from our study.One thing is clear, however, CYP-dependent catalyticactivity is almost always maximal or minimal whencorresponding tissue HO activity is minimal or maxi-mal, respectively. Whether this relationship is causal orcoincidental is unclear. Two possibilities are that HO-1contributed to the decrease of CYP and CYP-catalyzedactivity observed in this study, or alternatively, that HO-1 merely served as a good marker for the level of oxidantstress in a particular tissue. While ample evidence ex-ists for the latter view [29], demonstrations of the formerare limited. The possibility of direct, selective degrada-tion of CYP isozymes by HO-1 has been suggested byKutty et al. [30]. However, recent reports in primarycultured hepatocytes suggest HO is not involved in de-creased CYP activity following acute As3+ exposure tocells [31, 32]. Consequently, it is most likely that reactiveoxygen species formed subsequent to As3+ administra-tion both denature CYP, in an isozyme-selective man-ner, releasing heme and inducing HO-1, resulting inelevated serum bilirubin levels observed in the presentreport.

HO-1 activity appears to be a good indicator ofthe level of oxidant stress experienced by a tissue ata particular point in time. For example, 1 or 2 days af-ter treatment, the level of oxidative stress in liver isrelatively high as indicated by both microsomal HO-1activity and plasma bilirubin concentrations. Concomi-tant with this, total CYP and/or CYP catalytic activitywere decreased, indicating the domination of negativeregulatory influences. Subsequent to this, HO-1 activitydecreases, presumably indicating a further decrease inoxidative stress. After 5 days, it appears that positiveregulatory influences, selective for CYP1A1 expression

and activity in lung, are dominant although transientin nature.

Orthotopic liver transplantation, a procedure thatresults in profound pathobiological stress and in-creased plasma bilirubin levels [33], is also associatedwith increased pulmonary CYP1A1-dependent ERODactivity [12]. Consistent with this, the increase of ERODactivity and CYP1A1 mRNA after acute As3+ adminis-tration in this study was most prominent in or restrictedto lung. The largest increases of HO-1 activity in re-sponse to As3+ treatment were found in liver, and theleast, in lung (Figures 1 and 2). Furthermore, the highestHO activities occurred at the same time as the largestdecreases of hepatic microsomal CYP content, and CYP-dependent PROD and EROD activities. This suggeststhat the elevated levels of plasma bilirubin are derivedprimarily from degradation of CYP heme. Thus, it ispossible that the relative selectivity for CYP1A1 induc-tion in lung may be a consequence of exposure to biliru-bin produced hepatically and released into the plasma,in combination with lack of substantial HO inductionin the rat lung. The latter factor is of potential impor-tance as HO-1 induction in lung would limit the quan-tity of free heme available for incorporation into CYPapoprotein and would be indicative of oxidative stress,a condition that compromises CYP1A1 induction [34].

Acute As3+ administration results in profoundalterations of heme metabolism and elevations of thebiliary excretion of bilirubin [29]. Work in rats injectedwith radiolabelled heme showed that the majorityof the bilirubin formed from hepatic heme passes tothe plasma before its biliary excretion [35]. Consistentwith this, we found that acute As3+ treatment led toincreased HO-1 activity and large increases in totalbilirubin serum concentrations within 24 h in the rat.Catabolism of heme via HO results in the productionof bilirubin and the temporal change observed in itsplasma concentration correlates well with the increasedhepatic HO-1 activity observed. The large increase inbilirubin entering the circulation over a relatively shortperiod of time will almost certainly exceed the bindingcapacity of serum proteins in the circulation, resultingin increased absorption of bilirubin into a variety of tis-sues [36]. This would allow the opportunity for biliru-bin to induce CYP1A1 via interaction with the AHR.This response might be expected to occur preferentiallyin lung which receives total cardiac output.

The normal route for bilirubin excretion in-volves reactions catalyzed by the uridine diphos-phoglucuronosyltransferase (UGT) isozyme (UGT1A1)that converts bilirubin to both monoglucuronidesand diglucuronide conjugates [37]. The congenitallyjaundiced Gunn rat exhibits endogenous activation ofhepatic CYP1A1 and CYP1A2 gene expression [38], aswell as severe hyperbilirubinemia because of impaired


elimination resulting from a mutation in the bilirubinUGT gene [39]. Administration of the CYP1A inducerTCDD to Gunn rats significantly lowers plasma biliru-bin levels [38], whereas PB, an inducer of other CYPand UGT isoforms, does not [40]. A TCDD-inducible“bilirubin oxidase” activity has been identified in hep-atic microsomes prepared from Gunn rats or chickenembryos which requires both NADPH and O2 and is in-hibited by an antibody to CYP1A1/CYP1A2 [41]. Takentogether, these data indicate that bilirubin may serveas a substrate for CYP1A1 or CYP1A2 in a metabolicpathway for elimination separate from bilirubin UGT.Substrate-mediated regulation is common for manyCYP genes, particularly for CYP1A1 [2]. The present re-sults support such an alternate mechanism for bilirubinelimination, especially as the lung receives total cardiacoutput and is exposed to all blood-borne constituents.The deficient UGT bilirubin conjugation capabilitiesof lung (versus liver) are consistent with the conceptthat elevated circulating bilirubin levels in As3+-treatedrats contribute to the increase in pulmonary CYP1A1mRNA, catalytic activity and protein levels observedin this study.

Data from rats infused with bilirubin provided con-ditional support for the idea that bilirubin contributesto the regulation of pulmonary CYP1A1 subsequentto oxidant stress. A moderate increase in pulmonaryCYP1A1-dependent EROD activity was observed inbilirubin infused animals. Similar levels of total serumbilirubin were achieved in bilirubin infused animalscompared to As3+-treated (Figure 6). Pulmonary HO-1activity was increased at the time point used to de-termine CYP activity in bilirubin infused animals, ascompared to the time of maximal induction in As3+-treated animals, when HO activity was at lower con-trol levels. The reason(s) for the nondose-dependenteffects and lack of change in mRNA levels observed ineither the pulmonary increase or hepatic decrease inCYP1A1-dependent activity is (are) unclear from thepresent infusion data. Infusion of bilirubin increasesthe cytotoxicity of the bile, suppresses the biliary secre-tion of phospholipids and produces cholestasis in pigs[42]. Various experiments producing cholestasis via bileduct ligation in rats have shown the increased plasmabile acids result in inhibition of hepatic CYP activity[43]. Such adverse effects may have contributed to theobserved results in the liver and masked a marked in-crease in pulmonary CYP1A1 activity in the presentstudy. As well, results from infusion experiments un-doubtedly reflect the single early time point analyzed.In addition, lack of increased CYP1A1 mRNA mayreflect the low dose of bilirubin infused and/or theinduction of pulmonary HO-1 by 24 h after infusion.To further evaluate if bilirubin might induce CYP1A1in lung, different doses were injected directly into the

bronchi. However, the bilirubin administered did notresult in an induction of pulmonary CYP1A1 activity.

The present results from the tail vein infusion andintratracheal injection of bilirubin experiments indicateeither bilirubin in vivo is not the sole mediator pro-ducing the induction of pulmonary CYP1A1 followingAs3+ treatment, or that we were unable to satisfacto-rily mimic the supply of bilirubin to the lung in ourtreatment studies.

Based upon the changes observed in pulmonaryCYP1A1 activity after acute As3+ treatment, it is possi-ble that the ability of this tissue to oxidize exogenousand endogenous substrates, and to modulate toxic re-sponses may be altered subsequent to oxidant stress.Indeed, the increased accumulation of CYP1A1 mRNA,concomitant with increased CYP1A1-dependent cat-alytic activity and protein levels indicates that mean-ingful biological effects are possible. It is important torealize however, that the cell specific nature of manypulmonary toxicants is an important consideration instudies of this type. Thus, it is likely that modulationof the CYP monooxygenase system measured in wholelungs underestimates the changes typical of the mostresponsive cell type(s).

In conclusion, rat pulmonary CYP1A1 protein, cat-alytic activity and mRNA are increased maximally5 days after acute As3+ treatment, a mediator of oxidantstress. This interactive response exhibits both isozymeselectivity (CYP1A1 vs CYP2B1/2) and tissue selectiv-ity (lung vs liver) in rats. The results of this study sug-gest that pulmonary CYP1A1 catalytic activity, mRNAand protein levels are responsive to conditions of oxida-tive stress. This is consistent with our previous findingthat rat lung CYP1A1-dependent EROD activity is stillincreased 21 days following orthotopic liver transplan-tation [12]. Thus, if the increases of CYP1A1 mRNA andcatalytic activity occurred in response to a mediators(s)present within the circulation, it is not surprising thatthe greatest magnitude of response occurs in lung.

ACKNOWLEDGMENTS

JMS is a recipient of an Ontario Graduate Schol-arship. The authors are also grateful for the techni-cal contributions of Chris Webb to these experiments,Asma Yaghi who performed the intratracheal surgeriesand for instruments purchased with an award from theUWO Academic Development Fund.

REFERENCES

1. Nelson DR, Koymans L, Kamataki T, Stegeman JJ,Feyereisen R, Waxman DJ, Waterman MR, Gotoh O, Coon


MJ, Estabrook RW, Gunsalus IC, Nebert DW. P450 su-perfamily: Update on new sequences, gene mapping,accession numbers and nomeclature. Pharmacogenetics1996;6:1–42.

2. Carver LA, Bradfield CA. Ligand-dependent inter-action of the aryl hydrocarbon receptor with anovel immunophilin homolog in vivo. J Biol Chem1997;272:11452–11456.

3. Okey AB, Riddick DS, Harper PA. Molecular biology ofthe aromatic hydrocarbon (dioxin) receptor. Trends Phar-macol Sci 1994;15:226–232.

4. Whitlock JP. Induction of cytochrome P4501A1. AnnuRev Toxicol 1999;39:103–125.

5. Ma Q. Induction of CYP1A1. The Ahr/DRE paradigm:Transcription, receptor regulation, and expanding biolog-ical roles. Curr Drug Metab 2001;2:149–164.

6. Nebert DW, Puga A, Vasiliou V. Role of the Ah recep-tor and the dioxin-inducible [Ah] gene battery in toxi-city, cancer, and signal transduction. Ann NY Acad Sci1993;685:624–640.

7. Sinal CJ, Bend JR. Aryl hydrocarbon receptor-dependentinduction of Cyp1a1 by bilirubin in mouse hepatomahepa 1c1c7 cells. Mol Pharmacol 1997;52:590–599.

8. Phelan D, Winter GM, Rogers WJ, Lam JC, Denison MS.Activation of the Ah receptor signal transduction path-way by bilirubin and biliverdin. Arch Biochem Biophys1998;357:155–163.

9. Falkner KC, McCallum GP, Bend JR. Effects of arsenitetreatment on NAD(P)H: Quinone acceptor oxidoreduc-tase activity in liver, lung, kidney and heart of rat. DrugMetab Dispos 1993;21:334–337.

10. Albores A, Cebrian ME, Connelly JC, Bach PH, BridgesJW. Effects of arsenite on hepatic mixed-function oxidaseactivity in rats. Xenobiotica 1992;22:591–597.

11. Albores A, Sinal CJ, Cherian MG, Bend JR. Selectiveincrease of rat lung cytochrome P450 1A1 dependentmonooxygenase activity after acute sodium arsenite ad-ministration. Can J Physiol Pharmacol 1995;73:153–158.

12. Sinal CJ, Zhu L-F, Zhong R, Cherian MG, Bend JR. Livertransplantation induces cytochrome P450 1A1 dependentmonooxygenase activity in rat lung and kidney. Can JPhysiol Pharmacol 1995;73:146–152.

13. Chen CJ, Chen CW, Wu MM, Kuo TL. Cancer-potential inliver, lung, bladder and kidney due to ingested inorganicarsenic in drinking water. Br J Cancer 1992;66:888–892.

14. Ryter SW, Tyrell RM. The heme synthesis and degra-dation pathways: Role in oxidant sensitivity. Hemeoxygenase has both pro- and antioxidant properties. FreeRadical Biol Med 2000;28:289–309.

15. Shimada T, Yun C-H, Yamazaki H, Gautier J-C, BeaunePH, Guengerich FP. Characterization of human lungmicrosomal cytochrome P450 1A1 and its role inthe oxidation of chemical carcinogens. Mol Pharmacol1992;41:856–864.

16. Stocker R, Yamamoto Y, McDonagh AF, Glazer A, AmesBN. Bilirubin is an antioxidant of possible physiologicimportance. Science 1987;235:1043–1046.

17. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Proteinmeasurement with the Folin phenol reagent. J Biol Chem1951;193:265–275.

18. Omura T, Sato R. The carbon monooxide-binding pig-ment of liver microsomes. I. Evidence for its hemoproteinnature. J Biol Chem 1964;239:2370–2378.

19. Trakshel GM, Kutty RK, Maines MD. Purificationand characterization of the major constitutive form of

testicular heme oxygenase. J Biol Chem 1986;261:11131–11137.

20. Hines RH, Levy JB, Conrad RD, Iversen PL, Shen ML,Renli A, Bresnick E. Gene structure and nucleotide se-quence for rat cytochrome P-450c. Arch Biochem Biophys1985;237:465–476.

21. Affolter M, Labbe D, Jean A, Raymond M, Noel D,Labelle Y, Parent-Vaugeois C, Lambert M, Bojanowski R,Anderson A. cDNA clones for liver cytochrome P-450sfrom individual Aroclor-treated rats: Constitutive ex-pression of a new P-450 gene related to phenobarbital-inducible forms. DNA 1986;5:209–218.

22. Chomczynski P, Sacchi N. Single-step method ofRNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:156–159.

23. Sambrook J, Fritsch EF, Maniatatis T. In: Ford N editor.Molecular Cloning. A Laboratory Manual. Plainview, NY:Cold Spring Harbour Laboratory Press; 1989.

24. Rhodes ML, Patterson CE. Chronic intravenous infu-sion in the rat: A nonsurgical approach. Lab Anim Sci1979;29:82–84.

25. Yaghi A, McCormack DG. Pulmonary artery contractil-ity in pneumonia: Role of cyclooxygenase products andnitric oxide. J Cardiovasc Pharmacol 1999;34:468–474.

26. Reynolds HY. Bronchoalveolar lavage. Am Rev RespirDis 1987;135:250–263.

27. Keyse SM, Tyrrell RM. Heme oxygenase is the major 32-kDa stress protein induced in human skin fibroblasts byUVA radiation, hydrogen peroxide and sodium arsenite.Proc Natl Acad Sci USA 1989;86:99–103.

28. Oguro T, Kanekok E, Numazawa S, Imaoka S, FunaeY, Yoshida T. Induction of hepatic heme oxygenase andchanges in cytochrome P-450s in response to oxidativestress produced by stilbenes and stilbene oxides in rats.J Pharmacol Exp Ther 1997;280:1455–1462.

29. Albores A, Cebrian ME, Bach PH, Connelly JC, HintonRH, Bridges JW. Sodium arsenite induced alterations inbilirubin excretion and heme metabolism. J Biochem Tox-icol 1989;4:73–78.

30. Kutty RK, Daniel RF, Ryan DE, Levin W, Maines MD. Ratliver cytochrome P450b, P420b and P420c are degradedto biliverdin by heme oxygenase. Arch Biochem Biophys1988;260:638–644.

31. Jacobs J, Roussel R, Roberts M, Marek D, Wood S, WaltonH, Dwyer B, Sinclair P, Sinclair J. Effect of arsenite oninduction of CYP1A and CYP2H in primary cultures ofchick hepatocytes. Toxicol Appl Pharmacol 1998;150:376–382.

32. Jacobs JM, Nichols CE, Andrew AS, Marek DE, WoodSG, Sinclair PR, Wrighton SA, Kostrubsky VE, SinclairJF. Effect of arsenite on induction of CYP1A, CYP2B andCYP3A in primary cultures of rat hepatocytes. ToxicolAppl Pharmacol 1999;157:51–59.

33. Forster J, Yan ZY, Payne KM, Wood JG, Eisenbach JB,Delcore R. Bilirubin secretion as an early indicator of pa-tient survival following orthotopic liver transplantation.Trans Proc 1993;25:1889–1890.

34. Barker CW, Fagan JB, Pasco DS. Down-regulation ofP4501A1 and P4501A2 mRNA expression in isolatedhepatocytes by oxidative stress. J Biol Chem 1994;269:3985–3990.

35. Farrell GC, Gollan JL, Schmid R. Efflux of bilirubininto plasma following hepatic degradation of exogenousheme. Proc Soc Exp Biol Med 1980;163:504–509.


36. Stevenson DK, Vreman HJ. Carbon monoxide and biliru-bin production in neonates. Pediatrics 1997;100:252–254.

37. Burchell B, Coughtrie MWH, Jansen PLM. Functionand regulation of UDP-glucuronosyltransferase genesin health and liver disease: Report of the seventhinternational workshop on glucuronidation, Septem-ber 1993, Pitlochry, Scotland. Hepatology 1994;20:1622–1630.

38. Kapitulnik J, Gonzalez FJ. Marked endogenous activa-tion of the CYP1A1 and CYP1A2 genes in the congene-tially jaundiced Gunn rat. Mol Pharmacol 1993;43:722–725.

39. Iyanagi T, Watanabe T, Uchiyama Y. The 3-methylcholan-threne-inducible UDP-glucuronosyltransferase defi-ciency in the hyperbilirubinemic rat (Gunn rat) iscaused by a− 1 frameshift mutation. J Biol Chem1989;264:21302–21307.

40. Cohen AN, Kapitulnik J, Ostrrow JD, Zenone EA,Cochrane C, Celic L, Cheney H. Effects of phenobarbitalon bilirubin metabolism and its response to phototherapyin the jaundiced Gunn rat. Hepatology 1985;5:310–316.

41. De Matteis F, Dawson SJ, Boobis AR, Comoglio A. In-ducible bilirubin-degrading system of rat liver micro-somes: Role of cytochrome P450IA1. Mol Pharmacol1991;40:686–691.

42. Bjørnbeth BA, Enderesen M, Hvattum E, Lyberg T,Villanger O, Ræder MG. Large intravenous bilirubinloads increase the cytotoxicity of bile and lower theresistance of canalicular membrane to cytotoxic injuryand cause cholestasis in pigs. Scand J Gastroenterol1998;33:201–208.

43. Chen J, Robertson G, Field J, Liddle C, Farrell GC. Ef-fects of bile duct ligation on hepatic expression of female-specific CYP2C12 in male and female rats. Hepatology1998;28:624–630.

acute sodium arsenite administration induces pulmonary cyp1a1 mrna, protein and activity in the rat

Documents