Gene Structure of a Major Form of Phenobarbital-inducible Cytochrome P-450 in Rat Liver*

The gene structure of cytochrome P-450b, a major form of phenobarbital-inducible cytochrome P-450 in rat livers was elucidated by sequence analysis of the cloned genomic DNAs and was compared with the pre- viously determined gene structures of cytochrome P-450e, a minor form of phenobarbital-inducible cyto- chrome P-450 and two forms of 3-methylcholan-threne-inducible cytochrome P-450 (P-45Oc and -d). The gene for cytochrome P-450b is 23 kilobase pairs (kb) long and is separated into 9 exons by 8 intervening sequences. This gene structure is very similar to that of cytochrome P-450e except for the first intron, the first intron being much longer in cytochrome P-450b gene (approximately 12 kb) than in cytochrome P-450e gene (3.2 kb), but differs greatly from the gene struc- tures of two 3-methylcholanthrene-inducible cytochrome P-450s as pointed out previously (Sogawa, K., Gotoh, O., Kawajiri, K. & Fujii-Kuriyama, Y. (1984) Proc. Natl. Acad. Sci. U. S. A. 81, 5066-5070). The nucleotide sequences in all 9 exons and their flanking regions in introns show very close homology between the two phenobarbital-inducible cytochrome

The gene structure of cytochrome P-450b, a major form of phenobarbital-inducible cytochrome P-450 in rat livers was elucidated by sequence analysis of the cloned genomic DNAs and was compared with the previously determined gene structures of cytochrome P-450e, a minor form of phenobarbital-inducible cytochrome P-450 and two forms of 3-methylcholanthrene-inducible cytochrome P-450 (P-45Oc and -d).
The gene for cytochrome P-450b is 23 kilobase pairs (kb) long and is separated into 9 exons by 8 intervening sequences. This gene structure is very similar to that of cytochrome P-450e except for the first intron, the first intron being much longer in cytochrome P-450b gene (approximately 12 kb) than in cytochrome P-450e gene (3.2 kb), but differs greatly from the gene structures of two 3-methylcholanthrene-inducible cytochrome P-450s as pointed out previously (Sogawa, K., Gotoh, O., Kawajiri . The close sequence homology between the two phenobarbital-inducible cytochrome P-450 genes is also found to extend to the promoter region with one notable exception. The simple repeated sequences of (CA), which is present at -254 position in cytochrome P-450e gene is also observed at the equivalent position in cytochrome P-450b gene, but the repetitiveness is greatly reduced in cytochrome P-450b gene ((CA), for P-450b uersm (CA)Is for P-450e), and this may somehow be related to the difference in the level of cyto- ll To whom reprint requests should be addressed. chrome P-450b and P-450e in the inductive phase of phenobarbital administration.
Cytochrome P-450 plays an important role in the oxidative metabolism of a wide variety of structurally unrelated compounds of endogenous as well as exogenous origin (1)(2)(3). It has been shown that this versatility of the cytochrome system results from the presence of multiple forms of the hemoprotein, each of which exhibits different but rather broad substrate specificities. The syntheses of some of them are induced in specific manners by the administration of various kinds of In livers of phenobarbital-treated rats, the amounts of at least two forms of cytochrome P-450 are remarkably increased, and they are purified to a homogenous state. These two forms of cytochrome P-450, designated P-450b and P-450e, show some differences in catalytic activities and electrophoretic mobility with each other, although they are immunologically indistinguishable (4).
Recent studies involving molecular cloning technology and protein chemistry have demonstrated that the primary structures of these two hemoproteins are very similar. There are only 14 amino acid replacements in the limited portion of the molecules which consist of 491 amino acids (5-7). Analyses of genetic cross experiments have indicated that the genes for cytochrome P-450b and P-450e are nonallelic and are closely linked to each other on the same autosome (8). Under variety of conditions, the syntheses of cytochrome P-450b and P-450e were always coinduced, and the level of the former was constantly 2 or more times greater than that of the latter (8,9). It would be of interest to study how these two forms of cytochrome P-450 are coinduced by the administration of phenobarbital and many other drugs such as y-chlordane, SKF525A (9, lo), and 1,1-di(P-chlorophenyl)-2,2-dichloroethylene (11) and how their levels in the hepatocytes are regulated.
We have previously reported the gene structure of cytochrome P-450e, a minor form of phenobarbital-inducible cytochrome P-450(7). In the present paper, we report the gene structure of a major form of phenobarbital-inducible cytochrome P-450, P-450b, and compare its gene structure with that of cytochrome P-450e. Screening of the Rat Genomic DNA-Charon 4A Library-The HaeIII rat genomic library kindly donated by Drs. Jagodzinsky and Bonner was screened with the 32P-labeled 2.5-kb' BamHI/EcoRI fragment from P-450pb-6, a genomic clone for cytochrome P-450e gene, according to the method of Benton and Davis (12). A genomic clone P-450pb-6 was isolated and identified as described previously (7,13). The 2.5-kb fragment was radiolabeled with [(u-~'P]~CTP to a specific radioactivity of about 5 X 10' cpmlpg by nick translation (14). Phage plaques were transferred to nitrocellulose filters, denatured with the alkaline treatment, and then neutralized for plaque hybridization as described (12). Hybridization was carried out at 65 "C under the conditions as described previously (12) using the 32Plabeled DNA (1 X lo' cpm/filter) as a probe. The filters were washed, air dried, and exposed to x-ray film with an intensifying screen (DuPont Lightning Plus) as described. From about lo6 recombinant phages of the rat genomic library, 26 positive clones were finally plaque purified after three or four rounds of rescreening.
Characterization of Recombinant Phage-Phages were grown and purified by CsCl density gradient centrifugation as described by Blattner et al. (15). DNA was prepared from the purified phages as described (15). The purified DNA was analyzed by restriction cleavage mapping and by DNA blot hybridization. Approximately 0.5 gg of each recombinant phage DNA was digested with various sorts of restriction endonucleases under the conditions recommended by vendors, and the digested products were subjected to electrophoresis in 0.8% agarose gels. DNA fragments were transferred to nitrocellulose filters by blotting (16), and then the DNA fragments containing cytochrome P-450 gene sequences were identified by hybridization to the 32P-labeled inserted DNA (5 X lo6 cpm/ml) of PCP-450 pb4, which was cloned and characterized previously (5).
Subcloning of DNA Fragments for Sequencing-In order to facilitate the sequencing of the isolated genomic DNA, appropriate DNA fragments which had been generated by digestion with restriction enzymes were subcloned into plasmid pBR322, using E. coli HBlOl as host cells as described previously (7).
DNA Sequencing-All sequencing was performed by the chemical degradation method devised by Maxam and Gilbert (17). DNA fragments used for sequencing were labeled by 3zP either at the 5' end or at the 3' end depending on the restriction used for cutting of DNA fragments. For labeling of 5' protruding ends or flush ends, T4 polynucleotide kinase and [cP~'P]ATP were used (17), and the 3'protruding ends were labeled with [a-32P]ddATP using terminal deoxynucleotidyltransferase as described previously (18). Base substitutions (marked by asterisks in Fig. 3) from the sequence of cytochrome P-450b cDNA were confirmed by sequencing from the different sites, and other regions whose sequences were consistent with those of cytochrome P-450h cDNA were sometimes sequenced in one direction as shown in Fig. 1.

RESULTS AND DISCUSSION
Isolation of Cytochrome P-450b Gene-When we cloned cytochrome P-450e gene (7), we isolated another clone (clone 14), which also hybridized strongly with the DNA probe (the whole insert of PCP-450-pb4, Ref. 5) from a rat HaeIII gene library. As shown in Fig. 1, the cloned genomic DNA contained the 2.5-kb BamHIIEcoRI and the 1.2-kb EcoRI/ HindIII DNA fragments which were commonly found with cytochrome P-450e gene, but its detailed restriction cleavage maps in the other region were clearly different from those of cytochrome P-450e gene. We considered that this cloned genomic DNA (clone 14) was similar but clearly different from cytochrome P-450e gene (clone 11).
Since it is known that the 7th and 8th exons in the 1.2-kb EcoRIIHindIII fragment of cytochrome P-450e contained most frequently base substitutions from the cytochrome P-450b cDNA sequence (7), we determined the nucleotide sequence of the corresponding EcoRIIHindIII region in the cloned genomic DNA. The result of the sequence analyses showed that the sequences of this part of the cloned DNA ' The abbreviation used is: kb, kilobase pair(s). HinfI; Hp, HpaII; K, KpnI; P, PstI; S, SstI; S96, Sau96I; X, XbaI; concurred with those of PCP-450pbl and -4 (cytochrome P-450b cDNA clones). We, therefore, concluded that the genomic DNA isolated here was the gene for cytochrome P-450b. " " " _ agacagttgttcctaggaacacagcctgcttcttgggatctgatagctccaggtggcc t C C a g g g t t C t t C t t

kb
" " " _ Restriction fragments containing the other exon sequences ing the missing parts of cytochrome P-45Ob gene in the were identified by Southern blot analysis, using the labeled isolated genomic DNA by screening rat gene library again and cDNA of cytochrome P-450b which appropriately fragmented isolated four such clones as shown in Fig. 1. The DNA by restriction enzymes, as probes. It was found that the cloned fragments at both ends of clone 14 which is shown in Fig. 1 genomic DNA lacked the sequences for the first and the last were used as probes in the screenings. Two of them (clones 9th exons. Then, we searched for the genomic clone contain-31 and 36) contained the first exon and the other two (clones  Detailed restriction cleavage of these cloned DNAs confirmed that they contained partially overlapping regions with one another and altogether covered a stretch of approximately 40 kb of the chromosomal DNA. These results are represented in Fig. 1. The closed boxes indicate the localization of the exons which were first assigned by DNA blot analysis (data not shown) using the labeled cDNA as a probe and then were finally determined by sequence analysis of the genomic DNA. The general structure of cytochrome P-450b gene is very similar to cytochrome P-450e gene except for the length of the first intron. The first intron of cytochrome P-450b gene is much longer than that of cytochrome P-450e gene. This could result either from occurrence of insertion or deletion of large DNA fragment(s) in the first intron of phenobarbitalinducible cytochrome P-450 genes. Accordingly, the total length of cytochrome P-450b gene which spans from the putative transcription initiation site to the polyadenylation site comes to 12 kb, 9 kb longer than that of cytochrome P-450e gene. The gene is split into 9 exons by 8 intervening sequences.

GTGATCTGTGCTAATGGACTcTGTATATGGTCTCAGTGCTATGTCTACAGACTTACATAGTATGTATGGTTCAGGT~CAGMTCACAGAGTGTG~gcttcg
Nucleotide Sequence of Cytochrome P-450b Gene-Then, we determined the nucleotide sequence of all 9 exons and their flanking sequences in the introns by using the method of Maxam and Gilbert (17). The strategy for sequencing is outlined in Fig. 2. The determined sequence is consistent with that of PCP-45Opbl and -4, cytochrome P-450b cDNA clones, except for the two neutral base substitutions at the third letter of the codons (codon 333 and 354). These substitutions could be more likely due to allelic polymorphism than sequencing errors, because they were confirmed by sequencing from different sites. Therefore, we concluded that the cloned genomic DNA, the sequence of which was determined here, is the gene for cytochrome P-45Ob. The result is shown in Fig. 3.
The sequence of cytochrome P-450b gene is very similar to that of cytochrome P-450e gene in the partial flanking sequences in introns as well as in the nine exonic sequences. The putative transcription-initiation site was assigned at an A, 30 base pairs upstream from the initiation codon ATG by analogy from cytochrome P-450e gene (7), and the polyadenylation site was determined by comparison with the nucleotide sequence of cytochrome P-450b cDNA (7). A typical poly(A) addition signal AATAAA was not found, but a possible equivalent sequence, GGTAAA, was present at the corresponding position of the gene, as is also the case with cytochrome P-450e gene (7). This unusual poly(A) addition signal is considered to function as such, because the identical sequence was actually present at the same position in the cDNA sequence.
The insertion site of all 8 introns in relation to the coding nucleotide is also rigidly conserved between the two genes, and all the splicing junctions satisfy the canonical GT/AG rule (19). All the results described above confirmed the previous conclusion that the genes for cytochrome P-450b and -e diverged in the relatively recent past, some 8 million years ago (20).
Only 40 base substitutions were found in all exon sequences of the two genes, and 15 of them resulted in 14 amino acids replacements. As summarized in Table I, c a g a c a a a g a a a t a t a c a t c t g t a c g t t t a t a t c a g a a a t g a t c t t t c a c a t a g a a a  catgccctag taagtaaaca gagctgacaa aactgagctg acaagtgcac acccatcccc t t t a -1 50 a t a a a a c a a g a g g c c t a a g t c c c a g t g c c c t t t t g t c c t g t g t a t c t g t t t c g t g g t g t c . From their genetic cross experiments using inbred and outbred rat strains/colonies, Rampersaud and Waltz (8) have reported that there are two closely linked loci encoding cytochrome P-450b and P-450e with at least 6 alleles at the cytochrome P-450b locus and at least 2 alleles at the P-450e locus. Simmons and Kasper (22) have also demonstrated that phenobarbital-inducible cytochrome P-450 genes are located in a cluster near the Coh locus on mouse chromosome 7. The close proximity of phenobarbital-inducible cytochrome P-450 genes to one another may enhance nonreciprocal gene conversion events between nonallelic genes and this may also explain the high degree of apparent polymorphism observed at the cytochrome P-450e and P-450b loci.
When the gene organization of the two phenobarbitalinducible cytochrome P-450s is compared with that of two 3methylcholanthrene-inducible cytochromes P-450, P-450c, and P-450d, they look very different from each other as pointed out previously (23,24). None of the insertion sites of 8 introns for phenobarbital-inducible cytochrome P-450 genes occur at the equivalent positions of 6 introns of 3-methylcholanthrene-inducible ones, although on the basis of the similarity of the amino acid sequence, all these forms of cytochrome P-450 are reasonably assumed to be evolved from a common ancestor (20,23,24). The implication of this different gene organization in the evolution of cytochrome P-450 gene family has been previously discussed (24,25).
Nucleotide Sequence of 5' Flanking Region of Cytochrome P-450b and P-450e Genes-Many structurally unrelated compounds such as phenobarbital, y-chlordane, SKF-525A, and trans-stilbene are known to induce remarkably the synthesis of cytochrome P-450b and P-450e at the same time (9,lO).
It has been demonstrated by using the cloned cDNA of cytochrome P-450b as a probe (25)(26)(27) that his induced synthesis of these two cytochrome P-450s by drug administration results from the increased level of transcription of the mRNAs. This suggests that the levels of these enzymes are coordinately regulated at the transcriptional level.
The elevated level of the synthesis of cytochrome P-450b has been found to be constantly higher (2-fold or more) than that of cytochrome P-450e (8,9). In this regard, it is interesting to elucidate and compare the nucleotide sequences responsible for gene regulation of the two cytochrome P-450s.
The nucleotide sequences in 5' flanking region of the two genes are shown in Fig. 4. The transcription-initiation site was determined previously for cytochrome P-450e gene by S1 nuclease mapping experiment (7). For cytochrome P-450b gene, it was putatively assigned at the equivalent position to cytochrome P-450e gene as described. Both genes have the same modified TATA sequence, CATAAAA, 20 bases upstream from the transcription-initiation site. The nucleotide sequences in the other promoter region are also very homologous between cytochrome P-450b and P-450e genes, with one notable exception. An alternating purine-pyrimidine sequence CACACA---was found approximately 260 bases upstream from the transcription-initiation site in the two genes.
The chain length of the simple repeated sequence, however, is much longer in cytochrome P-450e gene than in cytochrome P-450b gene. Since the alternating purine-pyrimidine sequence has a potential to form the Z-DNA structure (28-32) which has been suggested to play a role in the regulation of gene expression (33), the difference in chain length of the repeated sequence between cytochrome P-450b and P-450e genes may somehow be related to the increased levels of the two proteins by the drug administration.
Recently, Guarente and Mason (33,34) have reported that iso-1-cytochrome c gene was controlled by the 5' flanking region of the TATA box and was subjected to the downregulation possibly due to the sequence containing an extensive run of alternating purine-pyrimidine residues. Deletion of this sequence increased the basal level of transcription to 2-to %fold (33,34). These phenomena appear to be analogous to the expression of cytochrome P-450b and P-450e genes, in that the cytochrome P-450b gene with the very short alternating purine-pyrimidine stretch in its 5' flanking region was expressed at a constantly higher level than cytochrome P-450e gene with the much longer repeated sequence. But it remains to be seen whether the Z conformation of DNA potentiated by the simple repeated sequence plays a role in down-regulation in the case of phenobarbital-inducible cytochrome P-450 genes as suggested with iso-1-cytochrome c gene. This problem will be tested experimentally.