Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The cytochromes P450 (CYPs)² are a superfamily of enzymes that play an important role in the oxidative metabolism of a wide variety of xenobiotics as well as endogenous compounds, such as steroids, arachidonic acid, and fatty acids. Members of this superfamily of enzymes are classified into families and subfamilies based on their amino acid sequence identities. With the development of molecular biological techniques over the past 20 years, the cDNAs of individual CYP members from numerous species, including humans, have been cloned and expressed in a variety of heterologous expression systems. Such studies have greatly increased our understanding concerning the role specific CYP enzymes play in the metabolism of specific substrates, the mechanisms involved in differences in drug metabolism between species, and the mechanisms underlying polymorphic differences in drug metabolism within a species.

In contrast to the large number of CYP cDNA clones that have been isolated, sequenced, and extensively studied from laboratory rodents, rabbits, and humans, the number of canine CYP cDNAs is limited. Because the dog is extensively used in pharmaceutical research as well as in drug safety assessment studies, information concerning specific canine CYP enzymes will be valuable in understanding their metabolic role in relation to that of other species. At present, the canine CYP enzymes involved in xenobiotic metabolism that have been cloned and sequenced include: 1) two members of the CYP1A subfamily (CYP1A1 and 1A2) (Uchida et al., 1990); 2) a member of the CYP2B subfamily (2B11) (Graves et al., 1990); 3) two members of the CYP2C subfamily (2C21 and 2C41) (Uchida et al., 1990; Blaisdell et al., 1998); 4) one member of the CYP2D subfamily (2D15) (Sakamoto et al., 1995); and 5) two members of the CYP3A subfamily (3A12 and 3A26) (Ciaccio et al., 1991; Fraser et al., 1997). Some of these canine cDNA clones have been heterologously expressed. One important CYP subfamily, CYP2E, has not been cloned from the dog, although it has been cloned and extensively studied from humans, laboratory rodents, and rabbits.

CYP2E is a constitutively expressed CYP that is induced by ethanol (Koop et al., 1982), by fasting (Hong et al., 1987), and during diabetes (Dong et al., 1988). These latter two conditions implicate CYP2E as having important roles in the intermediary metabolism of endogenous ketones and fatty acids. In most species studied, CYP2E is present as a single isoform, CYP2E1. In rabbits, however, a second highly similar isoform, CYP2E2, has also been identified (Khani et al., 1988). A number of therapeutic agents, including acetaminophen (Morgan et al., 1983), chlorzoxazone (Peter et al., 1990), theophylline (Zhang and Kaminsky, 1995), ethanol (Morgan et al., 1982), and the anesthetic agents enflurane (Thummel et al., 1993) and halothane (Gruenke et al., 1988), have been shown to be substrates for CYP2E. In addition to participating in the metabolism of therapeutic agents, CYP2E also plays a significant role in the metabolic activation of numerous precarcinogens such as benzene (Koop et al., 1989), nitrosamines (Yamazaki et al., 1992) and hepatotoxins such as acetaminophen (Hu et al., 1993), ethanol, and halogenated hydrocarbons (Olson et al., 1991), presumably through the production of free radicals.

The major objective of this study was to isolate, clone, and sequence members of the canine CYP2E subfamily from a canine liver cDNA library, and to compare the catalytic activities of heterologously expressed canine and human CYP2Es.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals and Reagents. All restriction enzymes were obtained from New England Biolabs (Beverly, MA). -Aminolevulinic acid, isopropyl -D-thiogalactopyranoside, NADPH, dilauroyl L--phosphatidylcholine, ampicillin, cholate, phenylmethylsufonyl fluoride, Nonidet P40, and 4-methylpyrazole (4-MP) were obtained from Sigma (St. Louis, MO). Hydroxyapatite was purchased from Clarkson Chemical Company (Williamsport, PA). Recombinant human NADPH-CYP reductase and cytochrome b₅ were obtained from Oxford Biomedical Research, Inc. (Oxford, MI). Chlorzoxazone was obtained from Aldrich Chemical Company (Milwaukee, WI), and 6-hydroxychlorzoxazone was purchased from Research Biochemicals International (Natick, MA). Other chemicals used were of the highest grade commercially available.

Tissue and Blood. Frozen liver tissue from an untreated female beagle dog, generously supplied by M. Faletto and C. J. Serabjit-Singh at Glaxo-Wellcome (Research Triangle Park, NC), was used for the cDNA library. Canine whole blood samples from mixed breed dogs were obtained from the College of Veterinary Medicine at North Carolina State University, from a local animal hospital, and from a local animal shelter (Institutional Animal Care and Use Committee approval number 98-100-B).

RNA Isolation, cDNA Library, and Screening. RNA was extracted from the female beagle liver by a guanidine isothiocyanate-phenol/chloroform method (Chomczynski and Sacchi, 1987). Poly (A+) RNA was isolated using an oligo(dT) cellulose column (Life Technologies, Gaithersburg, MD). A ZAP cDNA synthesis kit (Stratagene, La Jolla, CA) was used to construct an oligo(dT)-primed cDNA library from the mRNA. Approximately 2 × 10⁵ recombinant phage plaques from the unamplified library were screened on nitrocellulose filters by plaque hybridization. The probe was a CYP2E1 polymerase chain reaction (PCR) product made from a canine liver cDNA template using PCR primer pairs designed from two highly homologous regions in rat and human CYP2E1 cDNAs, in exons 2 and 6. The forward and reverse primers were 5'-CTTCGGGCCAGTGTTCAC-3' and 5'-CCCATATCTCAGAGTTGTGC-3', respectively. The PCR reaction mix contained 1× PCR buffer, 1.5 mM MgCl₂, 0.2 mM dNTPs, 0.25 µM each primer, and 2 U Taq DNA polymerase (Qiagen, Valencia, CA) in a final volume of 20 µl. PCR, performed on a Perkin-Elmer 4800 thermal cycler (PE Applied Biosystems, Foster City, CA), consisted of an initial 5-min denaturation at 95°C followed by 32 cycles of: 95°C for 1 min., 45°C for 1 min., and 72°C for 1 min., with a final 5 min. extension at 72°C. The resulting 744 base pair (bp) product was isolated on a 1.5% ethidium bromide-stained agarose gel, gel-purified with a Micropure 0.22/Microcon 30 kit (Amicon, Beverly, MA), and labeled with ³²P dCTP by random primer labeling (Life Technologies, Gaithersburg, MD). Hybridization was performed at 59°C, and filters were washed with 2× sodium chloride/sodium citrate buffer containing 0.1% SDS. The resulting clones were plaque-purified by three rounds of screening. pBluescript phagemids were excised from the Uni-ZAP XR vector using ExAssist helper phage and the SOLR strain of Escherichia coli.

Sequencing. Phagemid DNA was isolated using a S.N.A.P. miniprep kit (Invitrogen, Carlsbad, CA). The cloned cDNA was sequenced in both directions by primer walking starting with modified T3 and T7 primers. Cycle sequencing using dye-labeled dideoxynucleotides was performed using an ABI PRISM cycle sequencing kit with AmpliTaq Restriciton Fragment Length Polymorphism DNA polymerase, FS (PE Applied Biosystems, Foster City, CA). Samples were autosequenced on an ABI model 310 Genetic Analyzer (PE Applied Biosystems). The Genetic Computer Group program (Madison, WI) was used for cDNA and protein sequence analyses.

Genotyping. Genomic DNA for genotyping was extracted from 200 µl of whole blood with a Qiagen blood kit according to the manufacturer's directions.

Allele-Specific PCR Mismatch-Restriction Fragment Length Polymorphism Test. Primers were designed to amplify a 118-bp product of the CYP2E1 gene to differentiate between two CYP2E1 alleles found in this study. The forward and reverse primers were 5'-TAACCTGAAGTCTCTCGTCG-3' (anneals at bases 1374-1393) and 5'-GGGTCTTCAGCCCGAGCGGGGAACGACACAGAGTTTTT-3' (anneals at bases 1454-1491), respectively. The underlined nucleotide in the reverse primer represents a mismatch with the CYP2E1 sequence. Genomic DNA (1-5 µl) was added to a PCR reaction mix containing 1× PCR buffer, 1.5 mM MgCl₂, 0.2 mM dNTPs, 0.25 µM each primer, 5% dimethyl sulfoxide, and 2 U Taq DNA polymerase in a final volume of 20 µl. PCR, performed on a GeneAmp 2400 PCR System (PE Applied Biosystems), included an initial 5-min denaturation at 95°C, followed by 35 cycles of: 95°C for 20 s, 64°C for 10 s, and 72°C for 20 s, with a final 5-min extension at 72°C. The PCR product was digested overnight at 37°C with 8 U MseI in a 50-µl volume containing 1× NE buffer 2 and 100 µg/ml of BSA. After digestion, the reaction volume was reduced to 15 µl on a Speed Vac, loading buffer was added, and the entire sample was electrophoresed on a 4% agarose gel stained with ethidium bromide in 1× TBE buffer. Figure 1 includes a schematic of the PCR-based restriction enzyme test for the two CYP2E1 alleles.

View larger version (32K): [in this window] [in a new window]   Fig. 1.   PCR-based restriction enzyme test of CYP2E1 DNA. A, schematic of the PCR mismatch restriction fragment length polymorphism test used to genotype canine CYP2E1. A region of DNA, which aligns with a portion of exon 9 when compared with human and rat sequences, is amplified by PCR from genomic DNA using a 3' primer containing a mismatch. The resulting PCR product is digested with MseI. B, predicted sizes of the DNA fragments resulting from MseI digestion of the PCR products from the three possible genotypes. MseI digests the Tyr485 allele but does not digest the His485 allele. C, ethidium bromide-stained 4% agarose gel showing the actual fragmentation patterns resulting from MseI digestion of the PCR products amplified from the genomic DNA from dogs 9, 16, and 21. The sizes of the PCR-MseI fragments are shown on the left. The rightmost lane contains 50-bp DNA ladder. Genotypes: H/H represents the His485 homozygote, Y/Y represents the Tyr485 homozygote, and H/Y represents the heterozygote.

PCR Sequencing of Two CYP2E1 Alleles. PCR was performed to amplify a 309-bp product of the CYP2E1 gene. The forward and reverse primers were 5'-CTCGCATGGAGCTCTTCC-3' (anneals at bases 1328-1345) and 5'-AAAGCAGCATCTCAGGACC-3' (anneals at bases 1618-1636), respectively. Two microliters of genomic DNA was added to PCR reaction mix containing 1 × PCR buffer, 1.5 mM MgCl₂, 0.2 mM dNTPs, 0.25 µM each primer, and 2 U Taq DNA polymerase in a final volume of 20 µl. PCR, performed on a GeneAmp 2400 PCR System, included an initial 5-min denaturation at 95°C, followed by 35 cycles of: 95°C for 20 s, 64°C for 10 s, and 72°C for 20 s, with a final 5-min extension at 72°C. Fifteen microliters of the reaction mix was electrophoresed on an ethidium bromide-stained 2% agarose gel in 1 × TBE buffer. The 309-bp band was cut from the gel and the DNA was extracted with a QIAEX II Gel Extraction kit (Qiagen). The PCR product was sequenced, in both directions, using dye-labeled dideoxynucleotides in an ABI PRISM cycle sequencing kit with AmpliTaq DNA polymerase, FS. The forward and reverse sequencing primers were 5'-TAACCTGAAGTCTCTCGTCG-3' (anneals at bases 1374-1393) and 5'-TCCACCAGACTGTGCAGGG-3' (anneals at bases 1570-1588), respectively. Samples were autosequenced on a ABI model 373 Genetic Analyzer (PE Applied Biosystems).

Construction of Expression Plasmids. The N-termini of the canine CYP2E1 cDNAs were modified by replacing the first eight codons with the corresponding N-terminal codons of the bovine 17-hydroxylase sequence. A PCR product was amplified using a forward primer designed with a XbaI restriction site 5 bases from the 5'-end followed by a NdeI restriction site, the modified N-terminal sequence, and an additional 19 bases of CYP2E1 sequence at the 3'-end. The forward primer for the bovine 17-hydroxylase (B17-H) modification is shown in Fig. 2, the reverse primer was 5'-AGGTACAAGGTGAACACTGG-3'. The 229-bp PCR product, digested with XbaI and BglII, yielded a 83-bp fragment that was used to replace the corresponding N-terminal fragment of the CYP2E1s in Bluescript. Subsequently, the plasmid was digested with Eco0109I (a restriction site 13-bp after the stop codon), blunt ended, and then digested with NdeI. The resulting 1499-bp insert was ligated into NdeI and blunt ended XbaI sites of the pCW vector.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Forward primers and the ensuing N-terminal modification of canine CYP2E1 for expression in an E. coli cDNA expression system. The initiation codons within the PCR primers are underlined and aligned. Mutated amino acids are bold. Native: Unmodified canine CYP2E1 sequence; B-17-H: Bovine 17-hydroxylase modification; Pho Del: N-terminal hydrophobic deletion modification.

Protein Expression and Purification. E. coli strain DH5 was transformed with the modified CYP2E1s in the pCW vector. Single colonies, selected on Luria-Bertani (LB)-ampicillin plates, were used to seed 10 ml of LB broth containing 100 ug/ml ampicillin and then cultured overnight at 37°C. The overnight cultures were diluted 100-fold with Terrific broth containing 0.2% bacto-peptone, 200 ug/ml ampicillin, and 0.5 mM -aminolevulinic acid, and were cultured at 37°C with vigorous shaking for 2 h. The cultures were brought to room temperature, 1 mM isopropyl -D-thiogalactopyranoside was added, and the cultures were incubated at 30°C with shaking (150 rpm) for 48 h. CYP expression was monitored daily with a SLM Aminco DW-2000 split-beam spectrophotometer.

Enzymatic Assays. All samples were kept on ice except where otherwise noted. Recombinant CYP2E1s (50 pmol of canine CYP2E1 or 25 pmol of human CYP2E1) were preincubated with recombinant human NADPH-CYP reductase (reductase/CYP molar ratio, 4:1), rabbit cytochrome b₅ (cytochrome b₅/CYP molar ratio, 2:1), and dilauroylphosphatidylcholine (15 µg/ml, final concentration) for 3 min in a 37°C shaking water bath. After preincubation, 200 µl of reaction buffer (250 mM KPO₄, pH 7.4, 3.75 mM MgCl₂, and 0.25 mM EDTA) and chlorzoxazone (final concentration 0.0038-1 mM for human samples, 0.02-1 mM for canine samples) dissolved in fresh 60 mM KOH were added, and the total reaction volume was brought to 495 µl with water. After incubation in a 37°C shaking water bath for 4 min, reactions were initiated with the addition of NADPH (0.5 mM, final concentration). After a 5-min incubation, the reactions were quenched with the addition of 25 µl of 43% o-phosphoric acid.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

cDNA Sequencing. A number of CYP2E1 cDNA clones were isolated from a canine liver cDNA library, sized, and three were completely sequenced. The canine sequences were aligned with rat and human cDNA and gene sequences. Based on these alignments, the open reading frame was defined for the canine sequence. The assigned initiation ATG is in a strong context for translation initiation with an A residue at the 3 position and a G residue at the +4 position (Kozak, 1996). Two of the cloned inserts contained a complete 1485-bp open reading frame that codes for a polypeptide 494 amino acids in length. The two full-length clones differed from one another at position 1453, estimated to reside in exon 9 based on comparison with rat and human gene sequences, and at position 1719 in the poly-A tail. The T₁₄₅₃C change codes for a Tyr₄₈₅His amino acid substitution. To determine whether this discrepancy in the coding region was a result of the cloning process, an additional 33 clones from the unamplified library were sequenced across this region, and two populations of clones, which differed at this position, were found. A third cDNA insert had an early stop codon due to a 1-bp deletion at position 1157 of the coding region. Fourteen clones were sequenced over this region but no other clones with this deletion were detected. The canine CYP2E1 cDNA and deduced protein sequence are shown in Fig. 3. Amino acid sequence identity and similarity comparisons between canine, human, rodent, and rabbit forms of CYP2E1 are summarized in Table 1.

View larger version (63K): [in this window] [in a new window]   Fig. 3.   Nucleotide and deduced amino acid sequence of canine CYP2E1. The nucleotide sequence is shown in the top line and is numbered from the start codon. A star represents the stop codon. The conserved heme binding region is underlined and the cysteine residue providing the thiolate ligand to the heme iron is shown in the darkened box. The nucleotide base and resulting amino acid residue of the variant allele at bp 1453 (residue 485) are shown in the spaces above and below the sequence, respectively. The single bp difference at position 1719 in the poly-A tail is also shown above the sequence.

Allele Frequency Determination. We designed an allele-specific PCR-based restriction enzyme test so that individual dogs could be readily genotyped for the Tyr₄₈₅His alleles from genomic DNA samples. The restriction enzyme maps of the two alleles differ at a single BccI site, which is present in the His₄₈₅ allele but not in the Tyr₄₈₅ allele. Because BccI is not commercially available, we created a MseI restriction site in the Tyr₄₈₅ allele by designing a PCR primer with a 1-bp mismatch. A 118-bp product was amplified from exon 9 of CYP2E1 using the mismatch reverse primer and was then digested with MseI, which digests the PCR product of the Tyr₄₈₅ allele into two fragments, 39 and 79 bp, but does not digest the PCR product of the His₄₈₅ allele. The predicted sizes of the fragments are shown in Fig. 1, A and B. Genomic DNA from dogs with each of the three possible genotypes was subjected to the PCR-restriction enzyme test; representative results are shown in Fig. 1C. The enzyme test results were confirmed in a number of individuals by sequencing a PCR product amplified from exon 9. One hundred mixed breed dogs were genotyped. The Tyr₄₈₅ allele occurred at a frequency of 0.85 and the His₄₈₅ allele at a frequency of 0.15. Thirteen purebred beagles were genotyped; the frequencies of the Tyr₄₈₅ and His₄₈₅ alleles were determined to be 0.81 and 0.19, respectively.

Protein Expression and Characterization. To determine whether the genetic polymorphism in CYP2E1 altered catalytic activity of the enzyme, we expressed N-terminally modified canine CYP2E1 Tyr₄₈₅ and His₄₈₅ cDNAs in E. coli using the pCW vector. Replacement of the first eight codons of the CYP2E1 cDNA with those of modified bovine 17 -hydroxylase resulted in very low expression. In an attempt to increase expression levels, we subsequently made N-terminal modifications as described by Gilliam et al. (1994). The first 63 bases encoding the N-terminal hydrophobic region were removed and the next two codons were altered to encode a methionine and an alanine. Additionally, silent mutations were made in the next 24 bases to enrich adenine and thymine content. Expression levels of up to 300 nmol P450 per liter of culture were obtained from canine cDNAs with this modification. A similarly modified human CYP2E1 cDNA, obtained from Dr. F. P. Guengerich (Vanderbilt University) was also expressed. The expressed and purified canine and human recombinant CYP2E1s were catalytically active in the presence of cytochrome b₅ as demonstrated by their ability to 6-hydroxylate chlorzoxazone. In further experiments K_m and V_max values of the expressed and purified canine and human recombinant CYP2E1s were determined using the substrate chlorzoxazone. The kinetic parameters, determined by the direct linear plot method, of the canine Tyr₄₈₅ and His₄₈₅ variants and the human ortholog are compared in Table 2.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This is the first report of a canine CYP2E1 cDNA. The canine CYP2E1 amino acid sequence, deduced from the full-length cDNA, has a hydrophobic amino terminus as well as the conserved putative heme binding domain (Graham-Lorence and Peterson, 1996) (amino acid position 430-443) characteristic of the CYP proteins. The amino acid sequence of canine CYP2E1 has greater than 75% identity to human, rodent, and rabbit CYP2E1 sequences, and is one amino acid longer than the human, rodent, and rabbit orthologs. When the canine amino acid sequence is compared with sequences of the other CYP2 family members, identity is less than 54% and less than 34% when compared with the other CYP families. The canine CYP2E1 amino acid sequence has a higher identity with human CYP2E1 (77%) than with rodent or rabbit CYP2E1 (75-76%). However, the human CYP2E1 amino acid sequence has greater identity to rabbit and rodent CYP2E1 (78-79%) than to canine CYP2E1 (77%).

One hundred mixed breed dogs were genotyped and the results were analyzed. The Tyr₄₈₅ allele occurred at a higher frequency (0.85) than the His₄₈₅ allele (0.15). A test for Hardy-Weinberg equilibrium (HWE) shows a significant deviation from equilibrium (P = .031) with a higher proportion of heterozygotes than would be expected. This could be explained if some of the animals tested were first generation mixed breed and the parental breeds involved differed in allele frequency at the CYP2E1 locus. However, beagles, a breed used in drug testing studies, also varied in the incidence of these alleles.

The canine T₁₄₅₃C point mutation results in a Tyr₄₈₅His amino acid substitution. Although this amino acid residue is beyond the putative heme-binding region, it is part of a  sheet structure in close proximity to substrate recognition site 6 as predicted by Gotoh (1992). Interestingly, the amino acid at position 485 of CYP2E1 is occupied by either a tyrosine or a phenylalanine in all species sequenced to date (human, rat, mouse, rabbit, hamster, macaque, pig,³ and cow⁴) (Song et al., 1986; Khani et al., 1988; Freeman et al., 1992; Komori et al., 1992; Sakuma et al., 1994), which suggests that the presence of an aromatic residue at this position could have an important function.

Heterologous expression levels of the canine CYP2E1 alleles were highest when the cDNAs were modified by removing the first N-terminal hydrophobic segment (Nelson and Strobel, 1988) codons. Previous studies have shown that removal of this portion of the protein does not affect subcellular localization of CYP2E1 expressed in E. coli (Pernecky and Coon, 1996), nor does it affect catalytic activity of the heterologously expressed protein (Gilliam et al., 1994; Pernecky and Coon, 1996). The CYP2E1 ligand 4-MP was added to the buffers used for protein purification to stabilize the expressed enzyme. The expressed and purified canine and human recombinant CYP2E1s, in the presence of cytochrome b₅, were shown to be catalytically active toward the prototypic substrate chlorzoxazone. No significant difference in K_m or V_max was seen when the two allelic variants were compared by univariate ANOVAs. The V_max value of the canine and human CYP2E1 were not significantly different from each other; however, the K_m value of human CYP2E1 was significantly lower (P = .003) than that of canine CYP2E1. As a consequence, the intrinsic clearance (V_max/K_m ratio) of canine CYP2E1 is 4-fold lower than that of human CYP2E1, and the clearance of this substrate in vivo might be predicted to be lower in dogs than in humans. This study indicates that canine CYP2E1 has a lower affinity for chlorzoxazone than human CYP2E1.

Several polymorphisms of human CYP2E1 have also been identified, most of them occurring in noncoding regions of the gene. Single nucleotide polymorphisms in the 5'-flanking region include a PstI site (G_-1259C) (Watanabe et al., 1990; Hayashi et al., 1991) and an RsaI site (C_-1019T) (Watanabe et al., 1990; Hayashi et al., 1991), as well as polymorphisms at positions A_-316G (Fairbrother et al., 1998), T_-297A (Fairbrother et al., 1998), and G_-35T (Hu et al., 1997). The C_-1019T variant, located within a putative binding site for the transcription factor hepatocyte nuclear factor 1, has been shown to exhibit higher in vitro transcriptional activity in a CAT construct compared with wild type (Hayashi et al., 1991). However, in vivo studies suggest no relationship between this polymorphism and chlorzoxazone 6-hydroxylase activity in a Caucasian population (Kim et al., 1995). The G_-35T variant, located 9 bp upstream from the TATA box, showed increased transcriptional activity in vitro using luciferase constructs (Fairbrother et al., 1998). McCarver et al. (1998) identified a 100-bp insertion mutation in a region from 2270 to 1672 associated with increased in vivo chlorzoxazone metabolism only among individuals who were obese or had recently consumed alcohol. Polymorphisms in several introns have also been identified.

Four variant alleles have been identified in the coding region of human CYP2E1, G₁₁₆₈A (Arg₇₆His) in exon 2 (Hu et al., 1997), G₄₈₀₄A (Val₁₇₉Ile) in exon 4 (Fairbrother et al., 1998), G₁₀₀₅₉A (Val₃₈₉Ile) in exon 8 (Hu et al., 1997), and C₁₀₁₅₇T in exon 8 (silent) (Fairbrother et al., 1998). In vitro COS-1 cell expression studies indicate that the G₁₁₆₈A and the G₁₀₀₅₉A variants encode proteins with normal catalytic activity. However, the G₁₁₆₈A variant protein had decreased stability compared with wild type CYP2E1 (Hu et al., 1997) and might therefore be expressed at different levels in vivo. Lymphoblastoid cell expression studies have shown that catalytic activity of the protein encoded by the G₄₈₀₄A variant is similar to that of wild type CYP2E1 (Fairbrother et al., 1998).

This study reports the full-length sequence of canine CYP2E1 cDNA. A variant allele of canine CYP2E1 has also been identified, and allele frequencies were determined in a population of dogs. In addition, a genetic test for the alleles has been developed so that individual dogs may be genotyped from blood samples before their use in drug safety assessment studies or studies in dogs of the metabolism of newly developed drugs. The cDNAs of the two canine variants along with the human CYP2E1 cDNA were heterologously expressed in a bacterial expression system, and the recombinant proteins were purified. The canine CYP2E1 variants had similar catalytic activity toward the prototype substrate chlorzoxazone but their affinities and intrinsic clearances for chlorzoxazone were lower than that of human CYP2E1, suggesting that the canine CYP2E1 is less efficient in metabolizing this substrate than human CYP2E1. These differences in canine and human CYP2E1 may affect predictions of clearance of CYP2E1 substrates in the canine model.