- Email: [email protected]
Cloning and Functional Expression of a First Inducible Avian Cytochrome P450 of the CYP3A Subfamily (CYP3A37)
Archives of Biochemistry and Biophysics Vol. 373, No. 2, January 15, pp. 375–384, 2000 doi:10.1006/abbi.1999.1566, available online at http://www.idealibrary.com on
Cloning and Functional Expression of a First Inducible Avian Cytochrome P450 of the CYP3A Subfamily (CYP3A37) 1,2 Jean-Claude Ourlin, 3 Manuel Baader, 3 David Fraser, James R. Halpert,* and Urs A. Meyer 4 Pharmacology/Neurobiology, Biozentrum, University of Basel, Klingelbergstrasse 70, 4056 Basel, Switzerland; and *Department of Pharmacology and Toxicology, University of Texas Medical Branch, 301 University Boulevard, Galveston, Texas 77555-1031
Received June 28, 1999, and in revised form October 14, 1999
CYP3As represent a family of cytochromes P450 involved in the metabolism of both endogenous and exogenous natural and synthetic compounds. Well described in mammals, none have yet been cloned and characterized in avian species. In this paper, we report the cloning and analysis of an avian CYP3A (CYP3A37). Using an RNA differential display approach, an 80-bp phenobarbital-inducible cDNA fragment was amplified from chicken embryo liver. Based on its homology with mammalian CYP3As, this fragment was used to clone a full-length cDNA consisting of 1638 bp encoding a putative protein of 509 amino acids. The sequence shares between 57.4 and 62% identity at the amino acid level with CYP3As of other species. This cDNA was designated CYP3A37 according to the current cytochrome P450 nomenclature. When expressed in COS1 cells, the CYP3A37 cDNA produced a protein of >55 kDa, which was recognized by polyclonal anti-rat CYP3A1 antiserum. In a bacterial expression system, the CYP3A37 cDNA produced a protein capable of steroid 6b-hydroxylation. At a substrate concentration of 100 mM, progesterone, testosterone, and androstenedione were found to be 6b-hydroxylated at a rate of 15.4, 11.7, 12.2 nmol/min/ nmol P450, respectively. Used as control, the human CYP3A4 gave similar hydroxylation rates. Finally, in 1 The nucleotide sequence data reported in this paper have been submitted to EMBL nucleotide sequence data base under Accession No. AJ250337. 2 This research was supported in part by Swiss National Science Foundation (U.A.M.) and NIH Grant GM54995 (J.R.H.). 3 These two authors contributed equally to the work presented here. 4 To whom correspondence and reprint requests should be addressed at Biozentrum, Klingelbergstrasse 70, CH-4056 Basel, Switzerland. Fax: 41 61 267 22 08. E-mail: [email protected].
0003-9861/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
both chicken embryo liver and chicken hepatoma cells (LMH), CYP3A37 mRNA was increased after treatment with typical CYP3A inducers, such as metyrapone, phenobarbital, dexamethasone, and pregnenolone 16a-carbonitrile, but not rifampicin. CYP2H1, a wellcharacterized inducible chicken cytochrome P450, also was induced by the same compounds, suggesting similar regulation of CYP3 and CYP2 genes in this species. © 2000 Academic Press Key Words: cytochrome P450; chicken; induction; LMH cells; steroid 6b-hydroxylation; CYP3A37.
Cytochrome P450 (CYP) 5 enzymes are involved in the oxidative metabolism of numerous endogenous and exogenous compounds, including steroid hormones, drugs, carcinogens, and environmental pollutants. To fulfill their detoxifying role, these enzymes have three major characteristics. First, they catalyze the metabolism of a wide and overlapping spectrum of structurally unrelated substrates (1, 2). Second, they represent a superfamily of genes with multiple enzymes (1). Third, xenobiotic-metabolizing P450s are usually inducible by their own substrates, allowing dynamic adaptation to xenobiotic exposure (2). CYP3As are the predominant P450s expressed in mammalian liver (3). In man, CYP3A4 catalyzes the metabolism of 40 to 60% of all clinically used drugs. These include erythromycin, midazolam, cyclosporin A, FK506, nifedipine, and others (4). Numerous substrates of CYP3A enzymes induce the expression of the 5
Abbreviations used: CYP, cytochrome P450; DEX, dexamethasone; PB, phenobarbital; MET, metyrapone; RIF, rifampicin; PCN, pregnenolone 16a-carbonitrile; LMH, leghorn male hepatoma. 375
376
OURLIN ET AL.
corresponding genes at the transcriptional level. Four major classes of CYP3A inducers can be defined. These are represented by steroids, including glucocorticoids and antiglucocorticoids (e.g., RU-486), barbiturates (e.g., phenobarbital), macrolide antibiotics (e.g., rifampicin), and antifungal agents (e.g., azoles and imidazoles). Marked species differences in the response to inducers are observed in some compounds, exemplified by rifampicin or pregnenolone 16a carbonitrile (PCN) (5). Because of their inducibility and their wide substrate spectrum, CYP3As are often involved in drug– drug interactions in man (6). Moreover, they are known to activate potential carcinogens, such as benzo(a)pyrene and aflatoxin B1 (7). Recently, major advances in the understanding of the molecular mechanisms of CYP3A induction have been achieved. The pregnane X receptor (PXR) has been identified as a CYP3A regulator in mouse and human tissue (8 –11). This receptor is activated by typical CYP3A inducers, including rifampicin, phenobarbital, glucocorticoids, and antiglucocorticoids. Moreover, the differences in magnitude of activation of the mouse and human PXR by different inducers reflect the known differences in CYP3A induction potency in these two species. Finally, PXR specifically binds to cis-acting elements in the 59 flanking regions of the induced gene, DNA sequences previously identified by independent laboratories to be involved in the regulation of these genes (12). Although CYP3A enzymes are well described in many mammalian species, little is known about the CYP3A proteins and genes in birds, despite the presence of enzymatic activities known to be CYP3A-associated in other species. For example, Lorr et al. (13) described a protein in chicken liver microsomes that cross-reacted with a monoclonal antibody raised against pregnenolone-16a-carbonitrile-inducible cytochrome P450 of rat liver. The amount of the immunodetected protein correlated with the developmental profile and induction of erythromycin demethylase. Machala et al. (14) also described the phenobarbitalinducible N-demethylation of erythromycin in chicken embryos. Kimmett et al. (15) reported inducible steroid 6-b hydroxylation in chicken embryo livers. These observations and others (16, 17) strongly suggested the existence of CYP3A isoforms in chicken. In a previous report, we used RT-PCR differential display to determine dynamic change in the spectrum of gene expression caused by phenobarbital in chicken embryo livers (18). These data suggest that more than 50 genes may be up- or down-regulated by PB in this system. Among them, an up-regulated CYP3A-like fragment was detected. Here we describe the cloning, expression, and regulation of the corresponding CYP3A gene.
MATERIALS AND METHODS Materials. The chicken liver cDNA library (broiler breeder, male, 7 weeks old) maintained in the lZAP vector was obtained from Stratagene (La Jolla, CA). BA85 nitrocellulose filters were purchased from Schleicher & Schuell (Riehen, Switzerland). The random primed DNA labeling kit was obtained from Boehringer Mannheim (Rotkreuz, Switzerland). [ 32P]dATP was obtained from the Radiochemical Center, Amersham Switzerland (Zu¨rich, Switzerland). Oligonucleotides were designed using the Oligo Primer Analysis Software Version 5.0 and synthesized by Intron A.G. (Kaltbrunn, Switzerland). The pSE380 expression vector used for heterologous expression studies was purchased from Pharmacia (Alameda, CA). Chaps, progesterone, testosterone, androstenedione, NADPH, and DOPC were purchased from Sigma Chemical Co. (St. Louis, MO). [4- 14C]progesterone and [4- 14C]androstenedione were obtained from Dupont-New England Nuclear (Boston, MA). [4- 14C]testosterone was obtained from Amersham Life Sciences (Arlington Heights, IL). Thin-layer chromatography plates (silica gel, 250 mm, Si 250 PA (19C)) were purchased from J. T. Baker (Phillipsburg, NJ). Animals and cell culture. For induction studies in vivo, direct injections of inducers in 18-day-old chicken eggs were performed as previously described (18); 24- or 48-h induction experiments were done for mRNA and protein preparations, respectively. For induction studies in vitro, a chicken hepatoma cell line (LMH, ATCC, CRL-2117) was used. Cells were grown in William’s E medium (Amimed/Bioconcept, Allschwil, Switzerland) complemented with 10% fetal calf serum (Biological Industries, Israel), 2 mM glutamine, and 100 IU/ml penicillin and 100 mg/ml streptomycin (Gibco-BRL, Basel, Switzerland) on 0.1% gelatin-coated dishes (Falcon, Becton–Dickinson, Basel, Switzerland). For induction experiments, confluent cells were incubated in the presence of inducers for 16 or 48 h for mRNA or protein preparations, respectively. Screening of the chicken liver lZAP cDNA library. Screening of a chicken liver lZAP cDNA library was performed according to the protocols of the supplier; 10 6 phages were plated onto 15-cm petri dishes. Replicas were made using nitrocellulose filters. The probe for screening of the library consisted of a 1.4-kb fragment amplified from a cDNA library on the basis of the sequence of the amplified 80-bp fragment depicted in Figs. 1 and 2. The primers for amplification were CYP3A37 dn (59-GAACTCTTCTGGGTTTGGC-39) and KS (59CGAGGTCGACGGTATCG-39). The amplified fragment was labeled with [ 32P]dATP using the random labeling kit of Boehringer Mannheim (see Materials and Methods). Filters were hybridized overnight at 60°C under the stringency conditions described in the Stratagene protocol. The filters were exposed to X-ray film using intensifying screens for 12 to 24 h. Positive clones were picked, diluted, and reincubated on 10-cm dishes for further purification. Four screening steps were performed to isolate and purify positive clones. Sequencing of positive clones. Positive pBluescript SK(2) phagemids were in vivo excised from the lZAP vector according to the suppliers protocol using the ExAssist/SOLR system provided with the library. Sequencing of positive clones was performed with an ABI 373A automated sequencer using the ABI Prism Dye terminater cycle sequencing kit (Perkin–Elmer Europe B.V., Rotkreuz, Switzerland). The sequencing reactions were carried out using either T3 or T7 primers contained in the polylinker of the pBluescript SK(2) vector. Sequence analysis. For computer-assisted analysis of sequences, the Wisconsin Package Version 9.0, from Genetics Computer Group (GCG) (Madison, WI), was applied. RNA purification and cDNA synthesis. Isolation of total RNA from frozen chicken embryo livers was performed using a guanidium thiocyanate-based method (19). For in vitro studies, the Trizol reagent (Gibco-BRL) was used according to the manufacturer’s protocol. Induction experiments were performed in six-well plates and 1 ml Trizol reagent was used per well. In both preparations, total RNA
AVIAN CYP3A37 was quantified and diluted to appropriate concentration for cDNA synthesis. One microgram of total RNA was then reverse-transcribed in a total volume of 30 ml in the presence of 200 mM free nucleotides, 300 U MMLV reverse transcriptase (Gibco-BRL), 10 mM DTT, 30 U of Rnasin, ribonuclease inhibitor (Promega, Madison, WI), and 2 mM oligo(dT14(A/G/C)) primer. The samples were incubated at 40°C for 1 h and then stored at 220°C before use. Semiquantitative PCR. Primers for PCR amplification were as follows: for CYP3A37—CYP3A37up, 59-GAATACCGCAAAGGCTTCTTGG-39; CYP3A37dn, see above; and for CYP2H1—CYP2H1up, 59-GACACTTGACATCTCTTCCTC-3; CYP2H1dn, 59-CTGGGCATTGACTATCATT-39. As controls, chicken b-actin was amplified with primers CAup, 59-CCCTGAACCCCAAAGCCAAC-39 and CAdn, 59GACTCCATACCCAAGAAAGA-39. Diluted cDNA was amplified in a 30-ml reaction containing 50 mM free nucleotides, 0.2 mM primer, and 1U of AmpliTaq DNA polymerase (Perkin–Elmer Europe B.V.) in the provided buffer. The appropiate cDNA dilutions were empirically determined for each gene to ensure that signals were derived only from the exponential phase of amplification. PCR was performed with a DNA Thermal Cycler (Perkin–Elmer Europe B.V.) programmed for initial denaturation of 1 min at 94°C, followed by 30 cycles of 45 s at 94°C, 45 s at 60°C, and 1 min 30 s at 72°C, and a final extension time of 5 min at 72°C. Aliquots (20 ml) of the PCR were subjected to electrophoresis in 1% agarose ethidium bromide containing gel. Intensity of the bands was quantified with the use of ImageQuant software (Molecular Dynamics, Sunnyvale, CA). Normalization of PCR results was done by analysis of chicken b-actin from the same cDNA dilution series. Northern blot. Twenty micrograms of total RNA was subjected to electrophoresis on a formamide-containing 1% agarose gel. RNA was then transferred to a nylon membrane (GeneScreen; New England Nuclear Research Products) by blotting overnight in 203 SSC (13 5 150 mM NaCl, 15 mM sodium citrate). The membrane was baked in a vacuum oven for 2 h at 80°C. Hybridization was carried out in buffer containing 50% deionized formamide, 53 SSC, 53 Denhardt’s solution, 1% SDS, and 10% (w/v) dextransulfate. Before being added to the hybridization solution, the random labeled probe was boiled for 5 min in 500 ml of 10 mg/ml salmon sperm DNA and quickly chilled on ice. Hybridization was carried out for 16 –20 h. Washes were performed in 23 SSC/1% SDS at room temperature for 30 min and 23 SSC/1% SDS at 65°C for 20 min. Membranes were exposed to X-ray film using intensifying screens or to PhosphorImager screens for 12– 48 h. Immunoblot analysis. Expression of CYP3A37 was also examined by immunoblot analysis. Microsomes were prepared as previously described (20). Samples were analyzed on 10% SDS–PAGE and proteins were transferred to nitrocellulose membranes (Schleicher & Schuell) using a Pharmacia-LKB semidry blotter, at 0.8 mA/cm 2 for 1 h at room temperature. Transfer was carried out in 2.5 mM Tris, 19.2 mM glycine, 0.01% SDS, and 20% methanol. The nitrocellulose filter was then treated with 2% nonfat powdered milk in phosphatebuffered saline (PBS) with 0.05% Tween 20 for 2 h at room temperature. Expressed avian CYP3A was detected with a rabbit antibody against rat CYP3A1 provided by Dr. F. J. Gonzalez (NIH, Bethesda, MD) and horseradish peroxidase-conjugated goat antirabbit IgG (ECL Western blotting detection; Amersham Switzerland). The antibody against CYP3A1 has been described as anti-P-4506b-2 by Nagata et al. (21). Heterologous expression of CYP3A37 in COS1 cells. For heterologous expression in COS1 cells, CYP3A37 cDNA was subcloned into the pCDNA1-1 expression vector (Invitrogen, Leek, The Netherlands). Cos1 cell transfection and analysis was performed as previously described (22). N-terminal modification and heterologous expression of 3A37. Modifications to the N-termini of human 3A4 and canine 3A12 have previously been described (23). Restriction endonucleases, PCR, and
377
subcloning were used in the modification of 3A37 for expression in Escherichia coli. Chicken 3A37 was modified via PCR to convert the second amino acid from an asparagine to an alanine and to remove 10 hydrophobic residues within the N-terminal signal anchor sequence, changes that have been shown to facilitate expression in E. coli (24, 25). The N-terminal 59-GGGCCCATGGCTCTGTTATTAGCAGTTTTTGTAGTCCTCCTG-39 (corresponding to amino acids M,A,L,L,L,A,V,F,V,V,L,L) and C-terminal 59-GGTTGTGACGGTTCCGTATCCTTAAGGGCCC-39 primers used in these reactions incorporated NcoI and EcoRI sites, respectively, for cloning purposes. The resulting PCR product was subsequently cloned into the pSE380 expression vector. DNA sequencing confirmed the fidelity of the PCRs. Protein expression and preparation of solubilized E. coli membranes of 3A37, 3A4, and 3A12 were done essentially as described previously (25, 26). All constructs were maintained in DH5a cells and grown to mid log phase at 37°C with 240 rpm shaking in 250 ml liquid TB media (12 g Bacto tryptone, 24 g Bacto yeast extract, 4 ml glycerol/L). IPTG (final concentration 1.0 mM) and 80 mg/L d-aminolevulinic acid (ALA) were added, and cells were harvested after incubation at 30°C with 190 rpm shaking. Maximal expression of 3A37 was observed at 48 h after IPTG/ALA addition, and typical recovery of 3A37 protein ranged from 12 to 48 nmol/liter of culture. Maximal expression of 3A4 and 3A12 was observed at 72 h after IPTG/ALA addition and yields ranging from 40 to 60 nmol/liter of culture were routine. 3-((3-Cholamidpropyl)-dimethylammonio)-1-propanesulfonate (Chaps)-solubilized E. coli membrane preparations were used directly in steroid hydroxylase assays as described previously (23, 25); 10 pmol P450 were reconstituted with 40 pmol E. coli-expressed rat NADPH-P450 reductase, 10 pmol rat cytochrome b 5 , and 0.1 mg/ml dioleoylphosphatidylcholine (DOPC) and 0.06% Chaps in a minimal volume. Assays were performed for 10 min at 37°C in 15 mM MgCl 2, 50 mM Hepes buffer (pH 7.6), 0.06% Chaps, and 1 mM NADPH. Reactions were stopped with the addition of 50 ml tetrahydrofuran (THF) to each reaction tube. [ 14C]Steroid stock solutions were made in 100% methanol. Care was taken so that methanol concentrations in the reaction mixture were equivalent and did not exceed 1% of the total reaction volume. Individual assays were performed using 100 mM concentrations of testosterone, progesterone, and androstenedione. Hydroxysteroid metabolites were identified by relative mobility on thin-layer chromatography and comparison with authentic standards.
RESULTS
Cloning of CYP3A37 Using the conditions previously described (18), a RTPCR differential display experiment was performed with in vivo-treated or untreated chicken embryo livers. A PB-inducible 80-bp fragment was detected using a specific pair of primers (Fig. 1). More precisely, as depicted in Fig. 1, two bands differing by 1 nucleotide in length were amplified. The independent subcloning of both fragments resulted in an identical sequence with the expected additional nucleotide in the 39 region of the longer fragment. The generation of two PCR bands is most likely due to oligo(dT) mispriming. The sequencing of the 80-bp fragment indicated 60% homology with most mammalian CYP3As. We therefore designed a specific primer (CYP3A37dn) based on the sequence of this 80-bp fragment. The combination of CYP3A37dn and a KS primer located in the lZAP
378
OURLIN ET AL.
FIG. 1. RT-PCR differential display of chicken embryo livers in the presence or absence of phenobarbital. Differential display experiments were performed as described in (18). Oligo(dT) CC and the 10-mer primer (59-CCT GGA CCG C-39) were used to amplify cDNAs from livers treated with or without PB. Each lane represents amplicons derived from mRNA isolated from two different chickens. Lanes 1 and 3 are the results of two independent experiments from chicken injected with vehicle alone. Lane 2 and 4 are results of two independent experiments from chicken induced for 48 h with PB.
vector allowed the PCR amplification of a 1.4-kb fragment from a chicken liver cDNA library. This large fragment was then used as a probe for screening fulllength cDNAs. Among six different clones, one (clone 6) contained the full-length cDNA in which the initial 80-bp fragment sequence was present (underlined in Fig. 2). This clone was fully sequenced in both directions, using a set of primers derived from the sequence of the 1.4-kb probe. The cDNA consists of 1638 bp coding for a putative protein of 509 amino acids. Fortyseven nucleotides define the 59 untranslated region and 64 nucleotides are found in the 39 untranslated region. The deduced protein sequence contains a putative heme-binding domain around the cysteine in position 448 that is required for functional cytochromes P450s (Fig. 2). No classical poly(A) signal or poly(A) tail is present, indicating that this clone does not contain all of the 39 untranslated region. The CYP3A37 cDNA shares equivalent strong homology with CYP3As of other species (Table I and Fig. 3). At the amino acid level, between 57 and 62% identity with mammalian and fish CYP3A proteins is seen. Lower identity values are obtained when comparisons are made with other P450 subfamilies. It is interesting to note that a 6-amino acid insertion is found in the chicken enzyme beginning at amino acid 285. This insertion causes a shift in the alignment and subsequent residue numbering between CYP3A4 and the chicken CYP3A enzyme. Based on these observations, this cDNA was assigned the name CYP3A37 by the cytochrome P450’s Nomenclature Committee (1). Induction of CYP3A37 mRNA To investigate how the cloned cDNA is regulated, we treated chicken embryos or a hepatoma cell line (LMH cells, recently characterized in this laboratory, manu-
script in preparation) with classical CYP3A inducers. To focus on CYP3A37, semiquantitative PCR was performed using a pair of highly specific primers. Results are presented in Fig. 4. CYP3A37 is strongly induced by metyrapone (MET) both in ovo and in cell culture (Figs. 4A and 4B). Phenobarbital, dexamethasone (DEX), and pregnenolone 16-a carbonitrile are less potent inducers (Fig. 4B). Interestingly, rifampicin (RIF) (a strong inducer in human and rabbit liver, but not in mice and rats) is a weak inducer even at 50 mM final concentration. This observation suggests that CYP3A37 may be more closely related to rodents than to human CYP3As in regard to its regulation. Used as an internal PB-inducible control, CYP2H1 revealed a similar pattern of regulation (Fig. 4). In particular, CYP2H1 is induced by PCN, an unexpected feature for an ortholog of the rodent CYP2B1/2 genes. These results in chick embryo livers in ovo and LMH cells were confirmed by Northern blotting using an EcoRI–PstI CYP3A37 probe of 450 bp located in the 59 part of the cDNA. The probe revealed an inducible mRNA of approximately 2.2 kb, indicating that our cDNA clone does not reflect the entire mRNA transcript. Analysis of the density of the signals revealed that the spectrum and potencies of the different inducing compounds were similar or identical to those in the semi-quantitative PCR experiments (Figs. 5A and 5B). Metyrapone apparently is the most potent inducer in both chick embryo liver and LMH cells, followed by PB, DEX, PCN, and, to a lower extent, RIF. Interestingly, PB induces CYP3A37 with a delay in vivo as compared to MET. Induction experiments were initially performed for 24 h in chicken embryos and PB induction was barely dectectable at this time, when metyrapone exposure already gave a strong signal (Fig. 5A, 24 h). However, when chicken embryos were treated for 48 h with PB, as in the RT-PCR differential display experiments, CYP3A37 was strongly and consistently upregulated (Fig. 5A, 48 h). Thus, not only do metyrapone and PB have different potencies for induction, but they also show different time-course profiles. The reasons for these differences are unknown at this time. Expression and Regulation of CYP3A37 at the Protein Level To investigate whether this regulatory pattern also occurs at the protein level, immunoblotting was performed using an anti-rat CYP3A1 polyclonal antibody. Transient transfection of CYP3A37 cDNA in COS1 cells led to a modest but significant overexpression of a protein of an estimated molecular mass of 55 kDa only in cells transfected with the sense plasmid (Fig. 6A). This result indicates that our clone represents a fulllength cDNA directing the expression of a protein of the expected size that is recognized by an anti-CYP3A
AVIAN CYP3A37
379
FIG. 2. Complete cDNA sequence of CYP3A37. Clone 6 was sequenced in both directions using a set of primers designed for overlapping amplifications. The complete sequence is the result of at least two independent experiments. Upper lane: nucleotide sequence of CYP3A37. Lower lane: putative protein sequence derived from the cDNA translation. A 6-amino acid insertion (GSSDAK) relative to CYP3A4 exists at amino acids 285–290. The putative conserved heme-binding domain is indicated in boldface. The 80-bp fragment initially discovered by the differential display technique to have homologies to CYP3A enzymes is underlined.
380
OURLIN ET AL. TABLE I
Sequence Alignment of CYP3A37 with Other CYP3As in Various Species Species
CYP
Percentage aa identity
Pig Rat Sheep Rabbit Dog Mouse Human Human Human Rat Trout Guinea pig Hamster
3A29 3A9 3A24 3A6 3A12 3a13 3A4 3A7 3A5 3A1 3A27 3A20 3A10
62.0 61.7 61.6 60.8 60.5 60.1 59.7 59.4 59.2 59.0 58.5 57.6 57.4
Note. The sequence comparison was performed using the GCG package software (GCG, Madison, WI). Chicken CYP3A37 shares up to 62% amino acid identity with CYP3A isoenzymes in different species.
antibody. It also indicates that this antibody is useful in measuring CYP3A37 levels in chicken. Figure 6B summarizes these observations made at the protein level in the LMH cells. It indicates that the mRNA increase of CYP3A37 is followed by a corresponding increase in immunoreactive protein. This suggests a prominent regulation at the mRNA level, as observed for most regulated mammalian P450s.
subtypes (23). Additionally, some variation in product ratios between the avian 3A37 and the mammalian 3A4 and 3A12 enzymes are evident. For example, the ratio of 6b-hydroxyprogesterone product to 16a-hydroxyprogesterone for the avian enzyme is 14.1, whereas human and canine ratios are 6.2 and 7.9, respectively. In a similar fashion, the product ratio of chicken 6b-hydroxytestosterone to 2b-hydroxytestosterone is 13.1, in comparison with 4.8 and 12.0 for the canine and human models. These findings provide evidence for potential structural differences that may be of interest regarding structure–function relationships of this subclass of enzymes. DISCUSSION
In this report we describe the cloning and characterization of a first chicken CYP3A isoform. The evidence that this gene belongs to the 3A family can be derived from the following results: (i) there is high similarity in amino acid sequence only with 3A subfamily members, (ii) the expressed cDNA in COS cells produces a protein of around 55 kDa recognized by a CYP3A1 antibody, (iii) the expressed cDNA in E. coli produces a protein with high steroid 6b-hydroxylase activity, and (iv) the corresponding gene is regulated by a typical set of CYP3A inducers. The presence of CYP3As in avian species is not surprising, given the functional importance of this sub-
CYP3A37 Activity Because the expression level of CYP3A37 in COS1 cells was too low for activity measurements and because an unknown CYP3A-like enzyme is detected in untransfected COS1 cells, we expressed the CYP3A37 cDNA in a bacterial expression system. Steroid hydroxylase assays were performed using solubilized E. coli membrane preparations containing 3A37 and were compared to preparations containing human 3A4 or canine 3A12, as previously described (25, 26). The major metabolite formed from steroids by mammalian 3A enzymes is the 6b-hydroxylated product (25, 27, 28). The studies performed here (Table II) indicate a high degree of similarity between activities of the avian 3A37 and its mammalian counterparts. Three different steroid substrates, progesterone, testosterone, and androstenedione, were examined to determine the relative activities of heterologously expressed P450s 3A4, 3A12, and 3A37. P450 3A37 exhibited high rates of steroid hydroxylase activity for all of the steroids employed, similar to that of human 3A4. As previously reported, the canine 3A12 displayed hydroxylation rates considerably lower than that of 3A4, ranging from 25 to 50% of the activity of the human and avian
FIG. 3. Phylogeny of CYP3A amino acid sequences. The phylogenic tree was created using the “distances” and “growtree” programs of the GCG sequence analysis package (GCG, Madison, WI).
AVIAN CYP3A37
381
FIG. 4. Semi-quantitative RT-PCR analysis of CYP3A37 mRNA regulation. Specific primer pairs were used to amplify CYP3A37 and chicken b-actin mRNA from reverse transcribed mRNAs from treated and untreated samples. (A) Chicken embryos treated with PB (6 mg/egg), MET (6 mg/egg), or vehicle alone for 24 h. (B) LMH cells treated with PB (1 mM), MET (1 mM), RIF (50 mM), DEX (50 mM), PCN (50 mM), or vehicle alone (DMSO 0.1%) for 16 h. Values are presented as fold induction in comparison to untreated samples and corrected with values obtained for chicken beta actin amplifications.
family of enzymes in mammals both in detoxication of numerous chemicals and in steroid metabolism (3). Moreover, basal or inducible 6b-hydroxylation of steroids, a CYP3A-associated activity, has been previously described by several groups (16, 17). Furthermore, PB-inducible erythromycin N-demethylation, another CYP3A-specific activity, has also been described (14). Immunodetection by a monoclonal antibody of dexamethasone or pregnenolone 16a-carbonitrile induced CYP3A-like proteins in chicken embryo liver was reported by Kimmet et al. (15) and Lorr et al. (13). These observations suggested the existence of an inducible CYP3A-like enzyme (or enzymes) in chicken. The present study now identifies a gene that encodes an enzyme exhibiting these previously reported activities. CYP3A37 was cloned from an adult chicken liver cDNA library. On the basis of the present results we cannot exclude the possibility that other less inducible 3A isoforms are also present in chicken.
Regulation CYP3A37 was found to be regulated by the classical inducers of CYP3As in mammals. PB, metyrapone, dexamethasone, and pregnenolone 16a-carbonitrile, but not rifampicin, induce CYP3A37. Taken together, these results suggest that CYP3A37 is regulated in a fashion similar to that of rodent 3As. In particular, the PCN versus RIF pattern of induction is a strong argument in favor of this hypothesis (3). Metyrapone is the most potent inducer in our in vitro model. In man, metyrapone is clinically used as a test for hypothalamic–pituitary–adrenal axis function because of its potent inhibition of 11b-hydroxylation of deoxycortisol and deoxycorticosterone by CYP11B in adrenals (29). In vitro, metyrapone also inhibits a wide spectrum of liver-specific P450s. The observation that this compound maintains the level of immunoreactive P450s in primary cultures led to the discovery of its inducing properties (30). Initially observed in rat primary cul-
382
OURLIN ET AL.
to CYP3A4 (34). However, because of its high sensitivity to barbiturate induction and because other 2B-like sequences were not detected in chicken, it was assumed that CYP2H1 was the orthologue of rodent 2B genes in chicken. In contrast, our present findings lead us to propose that CYP2H1 is a CYP2C orthologue and that other 2B-like genes may also exist in chicken. Indeed, we have strong recent evidence for the presence of such genes in chicken. This could have many implications, in particular on how the PB regulation of mammalian genes may have evolved. CYP3A37 Activity In addition to these common regulatory pathways, the steroid hydroxylase profile of CYP3A37 exhibits a high degree of similarity to mammalian 3A enzymes, as indicated in Table II. Chicken 3A37 catalyzes the hydroxylation of progesterone, testosterone, and androstenedione at rates comparable to those of 3A4, whereas at 100 mM substrate, canine 3A12 is less ro-
FIG. 5. Northern blot analysis of CYP3A37 mRNA regulation. Twenty micrograms total RNA were loaded onto a 1% agarose gel and processed with CYP3A37 and chicken GAPDH hybridization probes as described. (A) Chicken embryos (lanes 1–5) treated with PB (6 mg/egg), MET (6 mg/egg), or vehicle alone (UT) for 24 or 48 h. (B) LMH cells treated with PB (1 mM), MET (1 mM), DEX (50 mM), RIF (50 mM), or vehicle alone (DMSO 0.1%, UT) for 16 h. A typical experiment from several independent studies is shown.
tures, its effect on CYP2B and CYP3A mRNA maintenance, together with the stabilization of 2B and 3A immunoreactive proteins led, to its classification as a PB-like inducer (31). Transcriptional activation by metyrapone has clearly been established for CYP3A1 in rat (32). Metyrapone was also able to induce CYP3A immmunoreactive protein in primary culture of human hepatocytes. Finally, in our laboratory, previous work has demonstrated the potency of metyrapone in inducing P450s and 5-aminolevulinate synthase in chicken hepatocyte culture (20). Barbiturates at high dose are well-known inducers of CYP3A genes in different species. They are more potent inducers of members of the CYP2 family, such as 2B1/2 in the rat or 2b10 in the mouse (reviewed by Kemper (33)). Interestingly, CYP2H1 behaves, in regard to regulation, very similarly to CYP3A37. Together with its barbiturate sensitivity, CYP2H1 also resembles CYP2C genes in humans, which are regulated similar
FIG. 6. Western blot analysis of CYP3A37 expression and regulation. Using an anti-rat CYP3A1 antibody, immunoblots were performed with 20 mg microsomal protein loaded on a 10% polyacrylamide gel and subsequently transferred to nitrocellulose membranes. (A) COS1 cells were transfected with empty vector (lane 2), vector 1 CYP3A37 cDNA in sense orientation (lanes 3 and 5), and vector 1 CYP3A37 cDNA in antisense orientation (lane 4). Human liver microsomes with high levels of CYP3As were used as internal control (lane 1). (B) LMH cells were treated with PB (1 mM), MET (1 mM), DEX (50 mM), or vehicle alone (DMSO 0.1%, lane 2) for 48 h. Human liver microsomes with high levels of CYP3A protein were used as internal control. Typical results from three independent experiments are shown.
383
AVIAN CYP3A37 TABLE II
Steroid Hydroxylation Rates for Human CYP3A4, Canine CYP3A12, and Chicken CYP3A37 Progesterone CYP 3A4 3A12 3A37
Testosterone
Androstenedione
6b
2b
16a
6b
2b
6b
16b
10.7 11.6 5.17 5.78 15.4 15.6
0.75 0.97 0.122 0.341 0.799 0.783
1.56 2.00 0.525 0.822 1.10 1.19
14.6 16.5 3.87 4.06 11.7 10.5
1.14 1.45 0.627 1.02 0.907 0.776
12.9 9.84 4.17 3.13 12.2 12.3
0.65 0.522 0.541 0.447 0.569 0.605
Note. Solubilized membrane preparations of each enzyme were reconstituted with 40 pmol cytochrome P450 reductase and 10 pmol cytochrome b5 and analyzed for progesterone, testosterone, and androstenedione hydroxylase activity. Substrate concentrations of 100 mM were used throughout. Rates are given in nmol of product formed/min/nmol of P450 and represent the means of duplicate incubations for each of two separate experiments.
bust than its avian and human counterparts. These findings are in agreement with previous reports of high steroid hydroxylase activities in chicken liver microsomes and of relatively lower rates for the canine 3A enzymes in reconstituted systems (15, 17, 25). It is noteworthy that chicken enzyme, which has a Pro residue at position 315 instead of the highly conserved Thr found in most P450s, retains high steroid hydroxylase activity. This is consistent with the finding that conversion of the corresponding Thr-309 in 3A4 to Ala preserves steroid hydroxylase activity (35). However, the findings presented in Table II also indicate several distinctions in catalytic profiles in 3A37 and the human and canine enzymes. Most notable is the difference in the product ratio of 6b-hydroxyprogesterone to 16a-hydroxyprogesterone, which is about twice as high for the chicken as for the mammalian enzymes. These differences are due to an apparent augmentation of the 6b-OH product in conjunction with a reduced 16a-OH product in chicken relative to canine and human enzymes. Presumably, the differences are related to the subtle differences in the size and/or shape of specific amino acid residues that contact the substrate molecule(s). Structure–function and homology modeling studies have resulted in the identification of a number of amino acid residues that play critical roles in the regio- and stereospecificity in steriod hydroxylations catalyzed by CYP3A enzymes (35– 38). Of particular interest are the findings of He et al. (36) demonstrating that the conversion of an alanine residue to a larger valine at residue 370 in CYP3A4 resulted in enhanced progesterone 16a-hydroxylase activity. This finding might indicate that the smaller glycine residue found in CYP3A37 at the corresponding position (Gly-376) could have the opposite effect of decreasing the 16a-hydroxylation of progesterone. The data presented here, indicating both a relative decrease in 16a-hydroxylase activity and a concomitant
rise in the ratio of 6b- to 16a-hydroxyprogesterone products, lend support to this hypothesis. REFERENCES 1. Nelson, D. R., Koymans, L., Kamataki, T., Stegeman, J. J., Feyereisen, R., Waxman, D. J., Waterman, M. R., Gotoh, O., Coon, M. J., Estabrook, R. W., Gunsalus, I. C., and Nebert, D. W. (1996) Pharmacogenetics 6, 1– 42. 2. Okey, A. B. (1992) in Pharmacogenetics of Drug Metabolism, pp. 549 – 608, Pergamon Press, Elmsford, NY. 3. Maurel, P. (1996) in Cytochromes P450: Metabolic and Toxicological Aspects (Ioannides, C., Ed.), pp. 241–270, CRC Press, Boca Raton, FL. 4. Li, A. P., Kaminski, D. L., and Rasmussen, A. (1995) Toxicology 104, 1– 8. 5. Denison, M. S., and Whitlock, J. P. (1995) J. Biol. Chem. 270, 175–178. 6. Pichard, L., Fabre, I., Fabre, G., Domergue, J., Saint Aubert, B., Mourad, J., and Maurel, P. (1990) Drug Metab. Disp. 18, 595– 606. 7. Shimada, T., Martin, M. V., Pruess-Schwartz, D., Marnett, L. J., and Guengerich, F. P. (1989) Cancer Res. 49, 6304 – 6312. 8. Kliewer, S. A., Moore, J. T., Wade, L., Staudinger, J. L., Watson, M. A., Jones, S. A., McKee, D. D., Oliver, B. B., Wilson, T. M., Zetterstro¨m, R. H., Perlmann, T., and Lehmann, J. M. (1998) Cell 92, 73– 82. 9. Lehmann, J. M., McKee, D. D., Watson, M. A., Wilson, T. M., Moore, J. T., and Kliewer, S. A. (1998) J. Clin. Invest. 102, 1016 –1023. 10. Blumberg, B., Sabbagh, W. J., Juguilon, H., Bolado, J. J., Van Meter, C. M., Ong, E. S., and Evans, R. M. (1998) Gene Dev. 12, 3195–3205. 11. Bertilsson, G., Heirdrich, J., Svensson, K., Asman, M., Jendenberg, L., Sydow-Ba¨ckman, M., Ohlsson, R., Postlind, H., Blomquist, P., and Berkenstam, A. (1998) Proc. Natl. Acad. Sci. USA 95, 12208 –12213. 12. Quattrochi, L. C., Mills, A. S., Yockey, C. B., and Guzelian, P. S. (1995) J. Biol. Chem. 270, 28917–28923. 13. Lorr, N. A., Bloom, S. E., Park, S. S., Gelboin, H. V., Miller, H., and Friedman, F. K. (1989) Mol. Pharmacol. 35, 610 – 616. 14. Machala, M., Nezveda, K., Irizar, A., and Ioannides, C. (1996) Arch. Toxicol. 71, 57– 63.
384
OURLIN ET AL.
15. Kimmet, S. M., McNamee, J. P., and Marks, G. S. (1996) Can. J. Physiol. Pharmacol. 74, 97–103. 16. Sinclair, J. F., and Sinclair, P. R. (1993) in Cytochrome P450 (Schenkman, J. B., and Greim, H., Eds.), pp. 259 –277, SpringerVerlag, Berlin. 17. Paolini, M., Pozzetti, L., Sapone, A., Biagi, G. L., and CantelliForti, G. (1997) Br. J. Pharmacol. 122, 344 –350. 18. Fru¨h, F. W., Zanger, U. M., and Meyer, U. A. (1997) Mol. Pharmacol. 51, 363–369. 19. Chomczynski, P., and Sacchi, N. (1987) Anal. Biochem. 162, 156 –159. 20. Giger, U., and Meyer, U. A. (1982) Biochem. Pharmacol. 31, 1735–1741. 21. Nagata, K., Gonzalez, F. J., Yamazoe, Y., and Kato, R. (1990) J. Biochem. 107, 718 –725. 22. Hintermann, E., Grieder, N. C., Amherd, R., Brodbeck, D., and Meyer, U. A. (1996) Proc. Natl. Acad. Sci. USA 93, 12315–12320. 23. Born, S. L., Fraser, D. J., Harlow, G. R., and Halpert, J. R. (1996) J. Pharmacol. Exp. Ther. 278, 957–963. 24. Barnes, H. J., Arlotto, M. P., and Waterman, M. R. (1991) Proc. Natl. Acad. Sci. USA 88, 5597–5601. 25. Fraser, D. J., Feyereisen, R., Harlow, G. R., and Halpert, J. R. (1997) J. Pharmacol. Exp. Ther. 283, 1425–1432. 26. John, G. H., Hasler, J. A., He, Y.-A., and Halpert, J. R. (1994) Arch. Biochem. Biophys. 314, 367–375.
27. Waxman, D. J., Attisano, C., Guengerich, F. P., and Lapenson, D. P. (1988) Arch. Biochem. Biophys. 263, 424 – 436. 28. Ciaccio, P. J., and Halpert, J. R. (1989) Arch. Biochem. Biophys. 271, 284 –299. 29. Engelhardt, D., and Weber, M. M. (1994) J. Steroid Biochem. Mol. Biol. 49, 261–267. 30. Padgham, C. R., Paine, A. J., Philips, I. R., and Shephard, E. A. (1992) Biochem. J. 285, 929 –932. 31. Padgham, R. W., and Paine, A. J. (1993) Biochem. J. 289, 621– 624. 32. Wright, M. C., Wang, X. J., Pimenta, M., Ribeiro, V., Paine, A. J., and Lechner, M. C. (1996) Mol. Pharmacol. 50, 856 – 863. 33. Kemper, B. (1998) Prog. Nucleic Acid Res. Mol. Biol. 61, 23– 64. 34. Rendic, S., and Di Carlo, F. J. (1997) Drug Metab. Rev. 29, 413–580. 35. Domanski, T. L., Liu, J., Harlow, G. R., and Halpert, J. R. (1998) Arch. Biochem. Biophys. 350, 223–232. 36. He, Y. A., He, Y. Q., Szklarz, G. D., and Halpert, J. R. (1997) Biochemistry 36, 8831– 8839. 37. Szklarz, G. D., and Halpert, J. R. (1997) Life Sci. 61, 2507– 2520. 38. Szklarz, G. D., and Halpert, J. R. (1998) Drug Metab. Dispos. 26, 1179 –1194.
Cloning and Functional Expression of a First Inducible Avian Cytochrome P450 of the CYP3A Subfamily (CYP3A37) 1,2 Jean-Claude Ourlin, 3 Manuel Baader, 3 David Fraser, James R. Halpert,* and Urs A. Meyer 4 Pharmacology/Neurobiology, Biozentrum, University of Basel, Klingelbergstrasse 70, 4056 Basel, Switzerland; and *Department of Pharmacology and Toxicology, University of Texas Medical Branch, 301 University Boulevard, Galveston, Texas 77555-1031
Received June 28, 1999, and in revised form October 14, 1999
CYP3As represent a family of cytochromes P450 involved in the metabolism of both endogenous and exogenous natural and synthetic compounds. Well described in mammals, none have yet been cloned and characterized in avian species. In this paper, we report the cloning and analysis of an avian CYP3A (CYP3A37). Using an RNA differential display approach, an 80-bp phenobarbital-inducible cDNA fragment was amplified from chicken embryo liver. Based on its homology with mammalian CYP3As, this fragment was used to clone a full-length cDNA consisting of 1638 bp encoding a putative protein of 509 amino acids. The sequence shares between 57.4 and 62% identity at the amino acid level with CYP3As of other species. This cDNA was designated CYP3A37 according to the current cytochrome P450 nomenclature. When expressed in COS1 cells, the CYP3A37 cDNA produced a protein of >55 kDa, which was recognized by polyclonal anti-rat CYP3A1 antiserum. In a bacterial expression system, the CYP3A37 cDNA produced a protein capable of steroid 6b-hydroxylation. At a substrate concentration of 100 mM, progesterone, testosterone, and androstenedione were found to be 6b-hydroxylated at a rate of 15.4, 11.7, 12.2 nmol/min/ nmol P450, respectively. Used as control, the human CYP3A4 gave similar hydroxylation rates. Finally, in 1 The nucleotide sequence data reported in this paper have been submitted to EMBL nucleotide sequence data base under Accession No. AJ250337. 2 This research was supported in part by Swiss National Science Foundation (U.A.M.) and NIH Grant GM54995 (J.R.H.). 3 These two authors contributed equally to the work presented here. 4 To whom correspondence and reprint requests should be addressed at Biozentrum, Klingelbergstrasse 70, CH-4056 Basel, Switzerland. Fax: 41 61 267 22 08. E-mail: [email protected].
0003-9861/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
both chicken embryo liver and chicken hepatoma cells (LMH), CYP3A37 mRNA was increased after treatment with typical CYP3A inducers, such as metyrapone, phenobarbital, dexamethasone, and pregnenolone 16a-carbonitrile, but not rifampicin. CYP2H1, a wellcharacterized inducible chicken cytochrome P450, also was induced by the same compounds, suggesting similar regulation of CYP3 and CYP2 genes in this species. © 2000 Academic Press Key Words: cytochrome P450; chicken; induction; LMH cells; steroid 6b-hydroxylation; CYP3A37.
Cytochrome P450 (CYP) 5 enzymes are involved in the oxidative metabolism of numerous endogenous and exogenous compounds, including steroid hormones, drugs, carcinogens, and environmental pollutants. To fulfill their detoxifying role, these enzymes have three major characteristics. First, they catalyze the metabolism of a wide and overlapping spectrum of structurally unrelated substrates (1, 2). Second, they represent a superfamily of genes with multiple enzymes (1). Third, xenobiotic-metabolizing P450s are usually inducible by their own substrates, allowing dynamic adaptation to xenobiotic exposure (2). CYP3As are the predominant P450s expressed in mammalian liver (3). In man, CYP3A4 catalyzes the metabolism of 40 to 60% of all clinically used drugs. These include erythromycin, midazolam, cyclosporin A, FK506, nifedipine, and others (4). Numerous substrates of CYP3A enzymes induce the expression of the 5
Abbreviations used: CYP, cytochrome P450; DEX, dexamethasone; PB, phenobarbital; MET, metyrapone; RIF, rifampicin; PCN, pregnenolone 16a-carbonitrile; LMH, leghorn male hepatoma. 375
376
OURLIN ET AL.
corresponding genes at the transcriptional level. Four major classes of CYP3A inducers can be defined. These are represented by steroids, including glucocorticoids and antiglucocorticoids (e.g., RU-486), barbiturates (e.g., phenobarbital), macrolide antibiotics (e.g., rifampicin), and antifungal agents (e.g., azoles and imidazoles). Marked species differences in the response to inducers are observed in some compounds, exemplified by rifampicin or pregnenolone 16a carbonitrile (PCN) (5). Because of their inducibility and their wide substrate spectrum, CYP3As are often involved in drug– drug interactions in man (6). Moreover, they are known to activate potential carcinogens, such as benzo(a)pyrene and aflatoxin B1 (7). Recently, major advances in the understanding of the molecular mechanisms of CYP3A induction have been achieved. The pregnane X receptor (PXR) has been identified as a CYP3A regulator in mouse and human tissue (8 –11). This receptor is activated by typical CYP3A inducers, including rifampicin, phenobarbital, glucocorticoids, and antiglucocorticoids. Moreover, the differences in magnitude of activation of the mouse and human PXR by different inducers reflect the known differences in CYP3A induction potency in these two species. Finally, PXR specifically binds to cis-acting elements in the 59 flanking regions of the induced gene, DNA sequences previously identified by independent laboratories to be involved in the regulation of these genes (12). Although CYP3A enzymes are well described in many mammalian species, little is known about the CYP3A proteins and genes in birds, despite the presence of enzymatic activities known to be CYP3A-associated in other species. For example, Lorr et al. (13) described a protein in chicken liver microsomes that cross-reacted with a monoclonal antibody raised against pregnenolone-16a-carbonitrile-inducible cytochrome P450 of rat liver. The amount of the immunodetected protein correlated with the developmental profile and induction of erythromycin demethylase. Machala et al. (14) also described the phenobarbitalinducible N-demethylation of erythromycin in chicken embryos. Kimmett et al. (15) reported inducible steroid 6-b hydroxylation in chicken embryo livers. These observations and others (16, 17) strongly suggested the existence of CYP3A isoforms in chicken. In a previous report, we used RT-PCR differential display to determine dynamic change in the spectrum of gene expression caused by phenobarbital in chicken embryo livers (18). These data suggest that more than 50 genes may be up- or down-regulated by PB in this system. Among them, an up-regulated CYP3A-like fragment was detected. Here we describe the cloning, expression, and regulation of the corresponding CYP3A gene.
MATERIALS AND METHODS Materials. The chicken liver cDNA library (broiler breeder, male, 7 weeks old) maintained in the lZAP vector was obtained from Stratagene (La Jolla, CA). BA85 nitrocellulose filters were purchased from Schleicher & Schuell (Riehen, Switzerland). The random primed DNA labeling kit was obtained from Boehringer Mannheim (Rotkreuz, Switzerland). [ 32P]dATP was obtained from the Radiochemical Center, Amersham Switzerland (Zu¨rich, Switzerland). Oligonucleotides were designed using the Oligo Primer Analysis Software Version 5.0 and synthesized by Intron A.G. (Kaltbrunn, Switzerland). The pSE380 expression vector used for heterologous expression studies was purchased from Pharmacia (Alameda, CA). Chaps, progesterone, testosterone, androstenedione, NADPH, and DOPC were purchased from Sigma Chemical Co. (St. Louis, MO). [4- 14C]progesterone and [4- 14C]androstenedione were obtained from Dupont-New England Nuclear (Boston, MA). [4- 14C]testosterone was obtained from Amersham Life Sciences (Arlington Heights, IL). Thin-layer chromatography plates (silica gel, 250 mm, Si 250 PA (19C)) were purchased from J. T. Baker (Phillipsburg, NJ). Animals and cell culture. For induction studies in vivo, direct injections of inducers in 18-day-old chicken eggs were performed as previously described (18); 24- or 48-h induction experiments were done for mRNA and protein preparations, respectively. For induction studies in vitro, a chicken hepatoma cell line (LMH, ATCC, CRL-2117) was used. Cells were grown in William’s E medium (Amimed/Bioconcept, Allschwil, Switzerland) complemented with 10% fetal calf serum (Biological Industries, Israel), 2 mM glutamine, and 100 IU/ml penicillin and 100 mg/ml streptomycin (Gibco-BRL, Basel, Switzerland) on 0.1% gelatin-coated dishes (Falcon, Becton–Dickinson, Basel, Switzerland). For induction experiments, confluent cells were incubated in the presence of inducers for 16 or 48 h for mRNA or protein preparations, respectively. Screening of the chicken liver lZAP cDNA library. Screening of a chicken liver lZAP cDNA library was performed according to the protocols of the supplier; 10 6 phages were plated onto 15-cm petri dishes. Replicas were made using nitrocellulose filters. The probe for screening of the library consisted of a 1.4-kb fragment amplified from a cDNA library on the basis of the sequence of the amplified 80-bp fragment depicted in Figs. 1 and 2. The primers for amplification were CYP3A37 dn (59-GAACTCTTCTGGGTTTGGC-39) and KS (59CGAGGTCGACGGTATCG-39). The amplified fragment was labeled with [ 32P]dATP using the random labeling kit of Boehringer Mannheim (see Materials and Methods). Filters were hybridized overnight at 60°C under the stringency conditions described in the Stratagene protocol. The filters were exposed to X-ray film using intensifying screens for 12 to 24 h. Positive clones were picked, diluted, and reincubated on 10-cm dishes for further purification. Four screening steps were performed to isolate and purify positive clones. Sequencing of positive clones. Positive pBluescript SK(2) phagemids were in vivo excised from the lZAP vector according to the suppliers protocol using the ExAssist/SOLR system provided with the library. Sequencing of positive clones was performed with an ABI 373A automated sequencer using the ABI Prism Dye terminater cycle sequencing kit (Perkin–Elmer Europe B.V., Rotkreuz, Switzerland). The sequencing reactions were carried out using either T3 or T7 primers contained in the polylinker of the pBluescript SK(2) vector. Sequence analysis. For computer-assisted analysis of sequences, the Wisconsin Package Version 9.0, from Genetics Computer Group (GCG) (Madison, WI), was applied. RNA purification and cDNA synthesis. Isolation of total RNA from frozen chicken embryo livers was performed using a guanidium thiocyanate-based method (19). For in vitro studies, the Trizol reagent (Gibco-BRL) was used according to the manufacturer’s protocol. Induction experiments were performed in six-well plates and 1 ml Trizol reagent was used per well. In both preparations, total RNA
AVIAN CYP3A37 was quantified and diluted to appropriate concentration for cDNA synthesis. One microgram of total RNA was then reverse-transcribed in a total volume of 30 ml in the presence of 200 mM free nucleotides, 300 U MMLV reverse transcriptase (Gibco-BRL), 10 mM DTT, 30 U of Rnasin, ribonuclease inhibitor (Promega, Madison, WI), and 2 mM oligo(dT14(A/G/C)) primer. The samples were incubated at 40°C for 1 h and then stored at 220°C before use. Semiquantitative PCR. Primers for PCR amplification were as follows: for CYP3A37—CYP3A37up, 59-GAATACCGCAAAGGCTTCTTGG-39; CYP3A37dn, see above; and for CYP2H1—CYP2H1up, 59-GACACTTGACATCTCTTCCTC-3; CYP2H1dn, 59-CTGGGCATTGACTATCATT-39. As controls, chicken b-actin was amplified with primers CAup, 59-CCCTGAACCCCAAAGCCAAC-39 and CAdn, 59GACTCCATACCCAAGAAAGA-39. Diluted cDNA was amplified in a 30-ml reaction containing 50 mM free nucleotides, 0.2 mM primer, and 1U of AmpliTaq DNA polymerase (Perkin–Elmer Europe B.V.) in the provided buffer. The appropiate cDNA dilutions were empirically determined for each gene to ensure that signals were derived only from the exponential phase of amplification. PCR was performed with a DNA Thermal Cycler (Perkin–Elmer Europe B.V.) programmed for initial denaturation of 1 min at 94°C, followed by 30 cycles of 45 s at 94°C, 45 s at 60°C, and 1 min 30 s at 72°C, and a final extension time of 5 min at 72°C. Aliquots (20 ml) of the PCR were subjected to electrophoresis in 1% agarose ethidium bromide containing gel. Intensity of the bands was quantified with the use of ImageQuant software (Molecular Dynamics, Sunnyvale, CA). Normalization of PCR results was done by analysis of chicken b-actin from the same cDNA dilution series. Northern blot. Twenty micrograms of total RNA was subjected to electrophoresis on a formamide-containing 1% agarose gel. RNA was then transferred to a nylon membrane (GeneScreen; New England Nuclear Research Products) by blotting overnight in 203 SSC (13 5 150 mM NaCl, 15 mM sodium citrate). The membrane was baked in a vacuum oven for 2 h at 80°C. Hybridization was carried out in buffer containing 50% deionized formamide, 53 SSC, 53 Denhardt’s solution, 1% SDS, and 10% (w/v) dextransulfate. Before being added to the hybridization solution, the random labeled probe was boiled for 5 min in 500 ml of 10 mg/ml salmon sperm DNA and quickly chilled on ice. Hybridization was carried out for 16 –20 h. Washes were performed in 23 SSC/1% SDS at room temperature for 30 min and 23 SSC/1% SDS at 65°C for 20 min. Membranes were exposed to X-ray film using intensifying screens or to PhosphorImager screens for 12– 48 h. Immunoblot analysis. Expression of CYP3A37 was also examined by immunoblot analysis. Microsomes were prepared as previously described (20). Samples were analyzed on 10% SDS–PAGE and proteins were transferred to nitrocellulose membranes (Schleicher & Schuell) using a Pharmacia-LKB semidry blotter, at 0.8 mA/cm 2 for 1 h at room temperature. Transfer was carried out in 2.5 mM Tris, 19.2 mM glycine, 0.01% SDS, and 20% methanol. The nitrocellulose filter was then treated with 2% nonfat powdered milk in phosphatebuffered saline (PBS) with 0.05% Tween 20 for 2 h at room temperature. Expressed avian CYP3A was detected with a rabbit antibody against rat CYP3A1 provided by Dr. F. J. Gonzalez (NIH, Bethesda, MD) and horseradish peroxidase-conjugated goat antirabbit IgG (ECL Western blotting detection; Amersham Switzerland). The antibody against CYP3A1 has been described as anti-P-4506b-2 by Nagata et al. (21). Heterologous expression of CYP3A37 in COS1 cells. For heterologous expression in COS1 cells, CYP3A37 cDNA was subcloned into the pCDNA1-1 expression vector (Invitrogen, Leek, The Netherlands). Cos1 cell transfection and analysis was performed as previously described (22). N-terminal modification and heterologous expression of 3A37. Modifications to the N-termini of human 3A4 and canine 3A12 have previously been described (23). Restriction endonucleases, PCR, and
377
subcloning were used in the modification of 3A37 for expression in Escherichia coli. Chicken 3A37 was modified via PCR to convert the second amino acid from an asparagine to an alanine and to remove 10 hydrophobic residues within the N-terminal signal anchor sequence, changes that have been shown to facilitate expression in E. coli (24, 25). The N-terminal 59-GGGCCCATGGCTCTGTTATTAGCAGTTTTTGTAGTCCTCCTG-39 (corresponding to amino acids M,A,L,L,L,A,V,F,V,V,L,L) and C-terminal 59-GGTTGTGACGGTTCCGTATCCTTAAGGGCCC-39 primers used in these reactions incorporated NcoI and EcoRI sites, respectively, for cloning purposes. The resulting PCR product was subsequently cloned into the pSE380 expression vector. DNA sequencing confirmed the fidelity of the PCRs. Protein expression and preparation of solubilized E. coli membranes of 3A37, 3A4, and 3A12 were done essentially as described previously (25, 26). All constructs were maintained in DH5a cells and grown to mid log phase at 37°C with 240 rpm shaking in 250 ml liquid TB media (12 g Bacto tryptone, 24 g Bacto yeast extract, 4 ml glycerol/L). IPTG (final concentration 1.0 mM) and 80 mg/L d-aminolevulinic acid (ALA) were added, and cells were harvested after incubation at 30°C with 190 rpm shaking. Maximal expression of 3A37 was observed at 48 h after IPTG/ALA addition, and typical recovery of 3A37 protein ranged from 12 to 48 nmol/liter of culture. Maximal expression of 3A4 and 3A12 was observed at 72 h after IPTG/ALA addition and yields ranging from 40 to 60 nmol/liter of culture were routine. 3-((3-Cholamidpropyl)-dimethylammonio)-1-propanesulfonate (Chaps)-solubilized E. coli membrane preparations were used directly in steroid hydroxylase assays as described previously (23, 25); 10 pmol P450 were reconstituted with 40 pmol E. coli-expressed rat NADPH-P450 reductase, 10 pmol rat cytochrome b 5 , and 0.1 mg/ml dioleoylphosphatidylcholine (DOPC) and 0.06% Chaps in a minimal volume. Assays were performed for 10 min at 37°C in 15 mM MgCl 2, 50 mM Hepes buffer (pH 7.6), 0.06% Chaps, and 1 mM NADPH. Reactions were stopped with the addition of 50 ml tetrahydrofuran (THF) to each reaction tube. [ 14C]Steroid stock solutions were made in 100% methanol. Care was taken so that methanol concentrations in the reaction mixture were equivalent and did not exceed 1% of the total reaction volume. Individual assays were performed using 100 mM concentrations of testosterone, progesterone, and androstenedione. Hydroxysteroid metabolites were identified by relative mobility on thin-layer chromatography and comparison with authentic standards.
RESULTS
Cloning of CYP3A37 Using the conditions previously described (18), a RTPCR differential display experiment was performed with in vivo-treated or untreated chicken embryo livers. A PB-inducible 80-bp fragment was detected using a specific pair of primers (Fig. 1). More precisely, as depicted in Fig. 1, two bands differing by 1 nucleotide in length were amplified. The independent subcloning of both fragments resulted in an identical sequence with the expected additional nucleotide in the 39 region of the longer fragment. The generation of two PCR bands is most likely due to oligo(dT) mispriming. The sequencing of the 80-bp fragment indicated 60% homology with most mammalian CYP3As. We therefore designed a specific primer (CYP3A37dn) based on the sequence of this 80-bp fragment. The combination of CYP3A37dn and a KS primer located in the lZAP
378
OURLIN ET AL.
FIG. 1. RT-PCR differential display of chicken embryo livers in the presence or absence of phenobarbital. Differential display experiments were performed as described in (18). Oligo(dT) CC and the 10-mer primer (59-CCT GGA CCG C-39) were used to amplify cDNAs from livers treated with or without PB. Each lane represents amplicons derived from mRNA isolated from two different chickens. Lanes 1 and 3 are the results of two independent experiments from chicken injected with vehicle alone. Lane 2 and 4 are results of two independent experiments from chicken induced for 48 h with PB.
vector allowed the PCR amplification of a 1.4-kb fragment from a chicken liver cDNA library. This large fragment was then used as a probe for screening fulllength cDNAs. Among six different clones, one (clone 6) contained the full-length cDNA in which the initial 80-bp fragment sequence was present (underlined in Fig. 2). This clone was fully sequenced in both directions, using a set of primers derived from the sequence of the 1.4-kb probe. The cDNA consists of 1638 bp coding for a putative protein of 509 amino acids. Fortyseven nucleotides define the 59 untranslated region and 64 nucleotides are found in the 39 untranslated region. The deduced protein sequence contains a putative heme-binding domain around the cysteine in position 448 that is required for functional cytochromes P450s (Fig. 2). No classical poly(A) signal or poly(A) tail is present, indicating that this clone does not contain all of the 39 untranslated region. The CYP3A37 cDNA shares equivalent strong homology with CYP3As of other species (Table I and Fig. 3). At the amino acid level, between 57 and 62% identity with mammalian and fish CYP3A proteins is seen. Lower identity values are obtained when comparisons are made with other P450 subfamilies. It is interesting to note that a 6-amino acid insertion is found in the chicken enzyme beginning at amino acid 285. This insertion causes a shift in the alignment and subsequent residue numbering between CYP3A4 and the chicken CYP3A enzyme. Based on these observations, this cDNA was assigned the name CYP3A37 by the cytochrome P450’s Nomenclature Committee (1). Induction of CYP3A37 mRNA To investigate how the cloned cDNA is regulated, we treated chicken embryos or a hepatoma cell line (LMH cells, recently characterized in this laboratory, manu-
script in preparation) with classical CYP3A inducers. To focus on CYP3A37, semiquantitative PCR was performed using a pair of highly specific primers. Results are presented in Fig. 4. CYP3A37 is strongly induced by metyrapone (MET) both in ovo and in cell culture (Figs. 4A and 4B). Phenobarbital, dexamethasone (DEX), and pregnenolone 16-a carbonitrile are less potent inducers (Fig. 4B). Interestingly, rifampicin (RIF) (a strong inducer in human and rabbit liver, but not in mice and rats) is a weak inducer even at 50 mM final concentration. This observation suggests that CYP3A37 may be more closely related to rodents than to human CYP3As in regard to its regulation. Used as an internal PB-inducible control, CYP2H1 revealed a similar pattern of regulation (Fig. 4). In particular, CYP2H1 is induced by PCN, an unexpected feature for an ortholog of the rodent CYP2B1/2 genes. These results in chick embryo livers in ovo and LMH cells were confirmed by Northern blotting using an EcoRI–PstI CYP3A37 probe of 450 bp located in the 59 part of the cDNA. The probe revealed an inducible mRNA of approximately 2.2 kb, indicating that our cDNA clone does not reflect the entire mRNA transcript. Analysis of the density of the signals revealed that the spectrum and potencies of the different inducing compounds were similar or identical to those in the semi-quantitative PCR experiments (Figs. 5A and 5B). Metyrapone apparently is the most potent inducer in both chick embryo liver and LMH cells, followed by PB, DEX, PCN, and, to a lower extent, RIF. Interestingly, PB induces CYP3A37 with a delay in vivo as compared to MET. Induction experiments were initially performed for 24 h in chicken embryos and PB induction was barely dectectable at this time, when metyrapone exposure already gave a strong signal (Fig. 5A, 24 h). However, when chicken embryos were treated for 48 h with PB, as in the RT-PCR differential display experiments, CYP3A37 was strongly and consistently upregulated (Fig. 5A, 48 h). Thus, not only do metyrapone and PB have different potencies for induction, but they also show different time-course profiles. The reasons for these differences are unknown at this time. Expression and Regulation of CYP3A37 at the Protein Level To investigate whether this regulatory pattern also occurs at the protein level, immunoblotting was performed using an anti-rat CYP3A1 polyclonal antibody. Transient transfection of CYP3A37 cDNA in COS1 cells led to a modest but significant overexpression of a protein of an estimated molecular mass of 55 kDa only in cells transfected with the sense plasmid (Fig. 6A). This result indicates that our clone represents a fulllength cDNA directing the expression of a protein of the expected size that is recognized by an anti-CYP3A
AVIAN CYP3A37
379
FIG. 2. Complete cDNA sequence of CYP3A37. Clone 6 was sequenced in both directions using a set of primers designed for overlapping amplifications. The complete sequence is the result of at least two independent experiments. Upper lane: nucleotide sequence of CYP3A37. Lower lane: putative protein sequence derived from the cDNA translation. A 6-amino acid insertion (GSSDAK) relative to CYP3A4 exists at amino acids 285–290. The putative conserved heme-binding domain is indicated in boldface. The 80-bp fragment initially discovered by the differential display technique to have homologies to CYP3A enzymes is underlined.
380
OURLIN ET AL. TABLE I
Sequence Alignment of CYP3A37 with Other CYP3As in Various Species Species
CYP
Percentage aa identity
Pig Rat Sheep Rabbit Dog Mouse Human Human Human Rat Trout Guinea pig Hamster
3A29 3A9 3A24 3A6 3A12 3a13 3A4 3A7 3A5 3A1 3A27 3A20 3A10
62.0 61.7 61.6 60.8 60.5 60.1 59.7 59.4 59.2 59.0 58.5 57.6 57.4
Note. The sequence comparison was performed using the GCG package software (GCG, Madison, WI). Chicken CYP3A37 shares up to 62% amino acid identity with CYP3A isoenzymes in different species.
antibody. It also indicates that this antibody is useful in measuring CYP3A37 levels in chicken. Figure 6B summarizes these observations made at the protein level in the LMH cells. It indicates that the mRNA increase of CYP3A37 is followed by a corresponding increase in immunoreactive protein. This suggests a prominent regulation at the mRNA level, as observed for most regulated mammalian P450s.
subtypes (23). Additionally, some variation in product ratios between the avian 3A37 and the mammalian 3A4 and 3A12 enzymes are evident. For example, the ratio of 6b-hydroxyprogesterone product to 16a-hydroxyprogesterone for the avian enzyme is 14.1, whereas human and canine ratios are 6.2 and 7.9, respectively. In a similar fashion, the product ratio of chicken 6b-hydroxytestosterone to 2b-hydroxytestosterone is 13.1, in comparison with 4.8 and 12.0 for the canine and human models. These findings provide evidence for potential structural differences that may be of interest regarding structure–function relationships of this subclass of enzymes. DISCUSSION
In this report we describe the cloning and characterization of a first chicken CYP3A isoform. The evidence that this gene belongs to the 3A family can be derived from the following results: (i) there is high similarity in amino acid sequence only with 3A subfamily members, (ii) the expressed cDNA in COS cells produces a protein of around 55 kDa recognized by a CYP3A1 antibody, (iii) the expressed cDNA in E. coli produces a protein with high steroid 6b-hydroxylase activity, and (iv) the corresponding gene is regulated by a typical set of CYP3A inducers. The presence of CYP3As in avian species is not surprising, given the functional importance of this sub-
CYP3A37 Activity Because the expression level of CYP3A37 in COS1 cells was too low for activity measurements and because an unknown CYP3A-like enzyme is detected in untransfected COS1 cells, we expressed the CYP3A37 cDNA in a bacterial expression system. Steroid hydroxylase assays were performed using solubilized E. coli membrane preparations containing 3A37 and were compared to preparations containing human 3A4 or canine 3A12, as previously described (25, 26). The major metabolite formed from steroids by mammalian 3A enzymes is the 6b-hydroxylated product (25, 27, 28). The studies performed here (Table II) indicate a high degree of similarity between activities of the avian 3A37 and its mammalian counterparts. Three different steroid substrates, progesterone, testosterone, and androstenedione, were examined to determine the relative activities of heterologously expressed P450s 3A4, 3A12, and 3A37. P450 3A37 exhibited high rates of steroid hydroxylase activity for all of the steroids employed, similar to that of human 3A4. As previously reported, the canine 3A12 displayed hydroxylation rates considerably lower than that of 3A4, ranging from 25 to 50% of the activity of the human and avian
FIG. 3. Phylogeny of CYP3A amino acid sequences. The phylogenic tree was created using the “distances” and “growtree” programs of the GCG sequence analysis package (GCG, Madison, WI).
AVIAN CYP3A37
381
FIG. 4. Semi-quantitative RT-PCR analysis of CYP3A37 mRNA regulation. Specific primer pairs were used to amplify CYP3A37 and chicken b-actin mRNA from reverse transcribed mRNAs from treated and untreated samples. (A) Chicken embryos treated with PB (6 mg/egg), MET (6 mg/egg), or vehicle alone for 24 h. (B) LMH cells treated with PB (1 mM), MET (1 mM), RIF (50 mM), DEX (50 mM), PCN (50 mM), or vehicle alone (DMSO 0.1%) for 16 h. Values are presented as fold induction in comparison to untreated samples and corrected with values obtained for chicken beta actin amplifications.
family of enzymes in mammals both in detoxication of numerous chemicals and in steroid metabolism (3). Moreover, basal or inducible 6b-hydroxylation of steroids, a CYP3A-associated activity, has been previously described by several groups (16, 17). Furthermore, PB-inducible erythromycin N-demethylation, another CYP3A-specific activity, has also been described (14). Immunodetection by a monoclonal antibody of dexamethasone or pregnenolone 16a-carbonitrile induced CYP3A-like proteins in chicken embryo liver was reported by Kimmet et al. (15) and Lorr et al. (13). These observations suggested the existence of an inducible CYP3A-like enzyme (or enzymes) in chicken. The present study now identifies a gene that encodes an enzyme exhibiting these previously reported activities. CYP3A37 was cloned from an adult chicken liver cDNA library. On the basis of the present results we cannot exclude the possibility that other less inducible 3A isoforms are also present in chicken.
Regulation CYP3A37 was found to be regulated by the classical inducers of CYP3As in mammals. PB, metyrapone, dexamethasone, and pregnenolone 16a-carbonitrile, but not rifampicin, induce CYP3A37. Taken together, these results suggest that CYP3A37 is regulated in a fashion similar to that of rodent 3As. In particular, the PCN versus RIF pattern of induction is a strong argument in favor of this hypothesis (3). Metyrapone is the most potent inducer in our in vitro model. In man, metyrapone is clinically used as a test for hypothalamic–pituitary–adrenal axis function because of its potent inhibition of 11b-hydroxylation of deoxycortisol and deoxycorticosterone by CYP11B in adrenals (29). In vitro, metyrapone also inhibits a wide spectrum of liver-specific P450s. The observation that this compound maintains the level of immunoreactive P450s in primary cultures led to the discovery of its inducing properties (30). Initially observed in rat primary cul-
382
OURLIN ET AL.
to CYP3A4 (34). However, because of its high sensitivity to barbiturate induction and because other 2B-like sequences were not detected in chicken, it was assumed that CYP2H1 was the orthologue of rodent 2B genes in chicken. In contrast, our present findings lead us to propose that CYP2H1 is a CYP2C orthologue and that other 2B-like genes may also exist in chicken. Indeed, we have strong recent evidence for the presence of such genes in chicken. This could have many implications, in particular on how the PB regulation of mammalian genes may have evolved. CYP3A37 Activity In addition to these common regulatory pathways, the steroid hydroxylase profile of CYP3A37 exhibits a high degree of similarity to mammalian 3A enzymes, as indicated in Table II. Chicken 3A37 catalyzes the hydroxylation of progesterone, testosterone, and androstenedione at rates comparable to those of 3A4, whereas at 100 mM substrate, canine 3A12 is less ro-
FIG. 5. Northern blot analysis of CYP3A37 mRNA regulation. Twenty micrograms total RNA were loaded onto a 1% agarose gel and processed with CYP3A37 and chicken GAPDH hybridization probes as described. (A) Chicken embryos (lanes 1–5) treated with PB (6 mg/egg), MET (6 mg/egg), or vehicle alone (UT) for 24 or 48 h. (B) LMH cells treated with PB (1 mM), MET (1 mM), DEX (50 mM), RIF (50 mM), or vehicle alone (DMSO 0.1%, UT) for 16 h. A typical experiment from several independent studies is shown.
tures, its effect on CYP2B and CYP3A mRNA maintenance, together with the stabilization of 2B and 3A immunoreactive proteins led, to its classification as a PB-like inducer (31). Transcriptional activation by metyrapone has clearly been established for CYP3A1 in rat (32). Metyrapone was also able to induce CYP3A immmunoreactive protein in primary culture of human hepatocytes. Finally, in our laboratory, previous work has demonstrated the potency of metyrapone in inducing P450s and 5-aminolevulinate synthase in chicken hepatocyte culture (20). Barbiturates at high dose are well-known inducers of CYP3A genes in different species. They are more potent inducers of members of the CYP2 family, such as 2B1/2 in the rat or 2b10 in the mouse (reviewed by Kemper (33)). Interestingly, CYP2H1 behaves, in regard to regulation, very similarly to CYP3A37. Together with its barbiturate sensitivity, CYP2H1 also resembles CYP2C genes in humans, which are regulated similar
FIG. 6. Western blot analysis of CYP3A37 expression and regulation. Using an anti-rat CYP3A1 antibody, immunoblots were performed with 20 mg microsomal protein loaded on a 10% polyacrylamide gel and subsequently transferred to nitrocellulose membranes. (A) COS1 cells were transfected with empty vector (lane 2), vector 1 CYP3A37 cDNA in sense orientation (lanes 3 and 5), and vector 1 CYP3A37 cDNA in antisense orientation (lane 4). Human liver microsomes with high levels of CYP3As were used as internal control (lane 1). (B) LMH cells were treated with PB (1 mM), MET (1 mM), DEX (50 mM), or vehicle alone (DMSO 0.1%, lane 2) for 48 h. Human liver microsomes with high levels of CYP3A protein were used as internal control. Typical results from three independent experiments are shown.
383
AVIAN CYP3A37 TABLE II
Steroid Hydroxylation Rates for Human CYP3A4, Canine CYP3A12, and Chicken CYP3A37 Progesterone CYP 3A4 3A12 3A37
Testosterone
Androstenedione
6b
2b
16a
6b
2b
6b
16b
10.7 11.6 5.17 5.78 15.4 15.6
0.75 0.97 0.122 0.341 0.799 0.783
1.56 2.00 0.525 0.822 1.10 1.19
14.6 16.5 3.87 4.06 11.7 10.5
1.14 1.45 0.627 1.02 0.907 0.776
12.9 9.84 4.17 3.13 12.2 12.3
0.65 0.522 0.541 0.447 0.569 0.605
Note. Solubilized membrane preparations of each enzyme were reconstituted with 40 pmol cytochrome P450 reductase and 10 pmol cytochrome b5 and analyzed for progesterone, testosterone, and androstenedione hydroxylase activity. Substrate concentrations of 100 mM were used throughout. Rates are given in nmol of product formed/min/nmol of P450 and represent the means of duplicate incubations for each of two separate experiments.
bust than its avian and human counterparts. These findings are in agreement with previous reports of high steroid hydroxylase activities in chicken liver microsomes and of relatively lower rates for the canine 3A enzymes in reconstituted systems (15, 17, 25). It is noteworthy that chicken enzyme, which has a Pro residue at position 315 instead of the highly conserved Thr found in most P450s, retains high steroid hydroxylase activity. This is consistent with the finding that conversion of the corresponding Thr-309 in 3A4 to Ala preserves steroid hydroxylase activity (35). However, the findings presented in Table II also indicate several distinctions in catalytic profiles in 3A37 and the human and canine enzymes. Most notable is the difference in the product ratio of 6b-hydroxyprogesterone to 16a-hydroxyprogesterone, which is about twice as high for the chicken as for the mammalian enzymes. These differences are due to an apparent augmentation of the 6b-OH product in conjunction with a reduced 16a-OH product in chicken relative to canine and human enzymes. Presumably, the differences are related to the subtle differences in the size and/or shape of specific amino acid residues that contact the substrate molecule(s). Structure–function and homology modeling studies have resulted in the identification of a number of amino acid residues that play critical roles in the regio- and stereospecificity in steriod hydroxylations catalyzed by CYP3A enzymes (35– 38). Of particular interest are the findings of He et al. (36) demonstrating that the conversion of an alanine residue to a larger valine at residue 370 in CYP3A4 resulted in enhanced progesterone 16a-hydroxylase activity. This finding might indicate that the smaller glycine residue found in CYP3A37 at the corresponding position (Gly-376) could have the opposite effect of decreasing the 16a-hydroxylation of progesterone. The data presented here, indicating both a relative decrease in 16a-hydroxylase activity and a concomitant
rise in the ratio of 6b- to 16a-hydroxyprogesterone products, lend support to this hypothesis. REFERENCES 1. Nelson, D. R., Koymans, L., Kamataki, T., Stegeman, J. J., Feyereisen, R., Waxman, D. J., Waterman, M. R., Gotoh, O., Coon, M. J., Estabrook, R. W., Gunsalus, I. C., and Nebert, D. W. (1996) Pharmacogenetics 6, 1– 42. 2. Okey, A. B. (1992) in Pharmacogenetics of Drug Metabolism, pp. 549 – 608, Pergamon Press, Elmsford, NY. 3. Maurel, P. (1996) in Cytochromes P450: Metabolic and Toxicological Aspects (Ioannides, C., Ed.), pp. 241–270, CRC Press, Boca Raton, FL. 4. Li, A. P., Kaminski, D. L., and Rasmussen, A. (1995) Toxicology 104, 1– 8. 5. Denison, M. S., and Whitlock, J. P. (1995) J. Biol. Chem. 270, 175–178. 6. Pichard, L., Fabre, I., Fabre, G., Domergue, J., Saint Aubert, B., Mourad, J., and Maurel, P. (1990) Drug Metab. Disp. 18, 595– 606. 7. Shimada, T., Martin, M. V., Pruess-Schwartz, D., Marnett, L. J., and Guengerich, F. P. (1989) Cancer Res. 49, 6304 – 6312. 8. Kliewer, S. A., Moore, J. T., Wade, L., Staudinger, J. L., Watson, M. A., Jones, S. A., McKee, D. D., Oliver, B. B., Wilson, T. M., Zetterstro¨m, R. H., Perlmann, T., and Lehmann, J. M. (1998) Cell 92, 73– 82. 9. Lehmann, J. M., McKee, D. D., Watson, M. A., Wilson, T. M., Moore, J. T., and Kliewer, S. A. (1998) J. Clin. Invest. 102, 1016 –1023. 10. Blumberg, B., Sabbagh, W. J., Juguilon, H., Bolado, J. J., Van Meter, C. M., Ong, E. S., and Evans, R. M. (1998) Gene Dev. 12, 3195–3205. 11. Bertilsson, G., Heirdrich, J., Svensson, K., Asman, M., Jendenberg, L., Sydow-Ba¨ckman, M., Ohlsson, R., Postlind, H., Blomquist, P., and Berkenstam, A. (1998) Proc. Natl. Acad. Sci. USA 95, 12208 –12213. 12. Quattrochi, L. C., Mills, A. S., Yockey, C. B., and Guzelian, P. S. (1995) J. Biol. Chem. 270, 28917–28923. 13. Lorr, N. A., Bloom, S. E., Park, S. S., Gelboin, H. V., Miller, H., and Friedman, F. K. (1989) Mol. Pharmacol. 35, 610 – 616. 14. Machala, M., Nezveda, K., Irizar, A., and Ioannides, C. (1996) Arch. Toxicol. 71, 57– 63.
384
OURLIN ET AL.
15. Kimmet, S. M., McNamee, J. P., and Marks, G. S. (1996) Can. J. Physiol. Pharmacol. 74, 97–103. 16. Sinclair, J. F., and Sinclair, P. R. (1993) in Cytochrome P450 (Schenkman, J. B., and Greim, H., Eds.), pp. 259 –277, SpringerVerlag, Berlin. 17. Paolini, M., Pozzetti, L., Sapone, A., Biagi, G. L., and CantelliForti, G. (1997) Br. J. Pharmacol. 122, 344 –350. 18. Fru¨h, F. W., Zanger, U. M., and Meyer, U. A. (1997) Mol. Pharmacol. 51, 363–369. 19. Chomczynski, P., and Sacchi, N. (1987) Anal. Biochem. 162, 156 –159. 20. Giger, U., and Meyer, U. A. (1982) Biochem. Pharmacol. 31, 1735–1741. 21. Nagata, K., Gonzalez, F. J., Yamazoe, Y., and Kato, R. (1990) J. Biochem. 107, 718 –725. 22. Hintermann, E., Grieder, N. C., Amherd, R., Brodbeck, D., and Meyer, U. A. (1996) Proc. Natl. Acad. Sci. USA 93, 12315–12320. 23. Born, S. L., Fraser, D. J., Harlow, G. R., and Halpert, J. R. (1996) J. Pharmacol. Exp. Ther. 278, 957–963. 24. Barnes, H. J., Arlotto, M. P., and Waterman, M. R. (1991) Proc. Natl. Acad. Sci. USA 88, 5597–5601. 25. Fraser, D. J., Feyereisen, R., Harlow, G. R., and Halpert, J. R. (1997) J. Pharmacol. Exp. Ther. 283, 1425–1432. 26. John, G. H., Hasler, J. A., He, Y.-A., and Halpert, J. R. (1994) Arch. Biochem. Biophys. 314, 367–375.
27. Waxman, D. J., Attisano, C., Guengerich, F. P., and Lapenson, D. P. (1988) Arch. Biochem. Biophys. 263, 424 – 436. 28. Ciaccio, P. J., and Halpert, J. R. (1989) Arch. Biochem. Biophys. 271, 284 –299. 29. Engelhardt, D., and Weber, M. M. (1994) J. Steroid Biochem. Mol. Biol. 49, 261–267. 30. Padgham, C. R., Paine, A. J., Philips, I. R., and Shephard, E. A. (1992) Biochem. J. 285, 929 –932. 31. Padgham, R. W., and Paine, A. J. (1993) Biochem. J. 289, 621– 624. 32. Wright, M. C., Wang, X. J., Pimenta, M., Ribeiro, V., Paine, A. J., and Lechner, M. C. (1996) Mol. Pharmacol. 50, 856 – 863. 33. Kemper, B. (1998) Prog. Nucleic Acid Res. Mol. Biol. 61, 23– 64. 34. Rendic, S., and Di Carlo, F. J. (1997) Drug Metab. Rev. 29, 413–580. 35. Domanski, T. L., Liu, J., Harlow, G. R., and Halpert, J. R. (1998) Arch. Biochem. Biophys. 350, 223–232. 36. He, Y. A., He, Y. Q., Szklarz, G. D., and Halpert, J. R. (1997) Biochemistry 36, 8831– 8839. 37. Szklarz, G. D., and Halpert, J. R. (1997) Life Sci. 61, 2507– 2520. 38. Szklarz, G. D., and Halpert, J. R. (1998) Drug Metab. Dispos. 26, 1179 –1194.