Cytochrome P450, Subfamily Iia, Polypeptide 6

Description

The CYP2A6 gene encodes an enzyme (EC 1.14.14.1) that plays a major role in the oxidation of nicotine and coumarin in human liver microsomes. Polymorphisms in the CYP2A6 gene that affect enzyme activity and susceptibility to lung cancer have been identified.

Cloning and Expression

Phillips et al. (1985) used a cDNA clone coding for a phenobarbital-inducible cytochrome P-450 variant of rat liver microsomal membranes as a probe to screen a human cDNA library. Restriction mapping showed that 2 of the colonies isolated contained plasmids coding for overlapping regions of the same cDNA sequence. The sequence showed considerable homology to that of cytochrome P-450 isozymes isolated from other species. The phenobarbital-inducible P-450 gene is a member of a multigene family coded by human chromosome 19. Induction by phenobarbital is mediated almost entirely at the level of transcription.

Yamano et al. (1989) isolated a human CYP2A3 cDNA from a liver cDNA library. The human gene encodes a 448-amino acid polypeptide that is 85% identical to the rat protein.

Yamano et al. (1990) isolated 2 cDNAs coding for P450s in the CYP2A gene subfamily from a lambda-gt11 library prepared from human hepatic mRNA. The 2 cDNAs differed by only 1 amino acid, leu160 to his (L160H; 122720.0001), and presumably represented alleles. Designated CYP2A3, the gene was shown to be primarily responsible for coumarin 7-hydroxylase activity in human liver. The level of expression of this activity varied up to 40-fold among livers. Levels of mRNA also varied significantly, and 3 specimens had no detectable mRNA. When the human CYP2A3 gene was cloned, it was shown to encode IIA3, the enzyme for coumarin 7-hydroxylase (Yamano et al., 1990). Unfortunately, the CYP2A3 designation had already been taken for the rat gene, and it was uncertain that the human gene was orthologous to the rat gene. Therefore, the human IIA3 gene product is encoded by a gene designated CYP2A6 (Nebert, 1994).

Fernandez-Salguero et al. (1995) cloned 3 complete CYP2A genes, CYP2A6, CYP2A7 (608054), and CYP2A13 (608055), in addition to 2 pseudogenes truncated after exon 5, located on 19q13.2.

Gene Structure

Fernandez-Salguero et al. (1995) determined that the CYP2A6 gene contains 9 exons.

Mapping

By in situ hybridization, Davis et al. (1986) assigned P450PB to 19q13.1-q13.3. Shephard et al. (1985) used the same probe in Southern analysis of DNA from human-rodent cell hybrids and likewise concluded that the gene is located on chromosome 19. Mitchell et al. (1989) concluded that the cluster of CYP2A, CYP2B, and CYP2F genes is distal to the secretor locus (182100) on chromosome 19. By fluorescence in situ hybridization, Trask et al. (1993) localized the CYP2A gene to 19q13.2. By pulsed field gel electrophoresis, Miles et al. (1989) demonstrated that the CYP2A and CYP2B genes lie within the same 350-kb genomic DNA fragment.

Fernandez-Salguero et al. (1995) determined the structural organization of the CYP2 gene cluster, which spans 350 kb on chromosome 19q13.2. They also determined the directions of transcription.

The genes that correspond to PEPD (613230), GPI (172400), and P450PB on human chromosome 19 are on mouse chromosome 7 (Matsunaga et al., 1990). In the mouse, Coh, the gene for coumarin hydroxylase (a P-450 enzyme), is closely linked to Gpi1, on proximal chromosome 7. Using a RFLP of the CYP1 locus (CYP2 in the new nomenclature), Davis (1987) studied linkage with the PEPD polymorphism. In males, a maximum lod score of 2.69 at theta = 0.01 was observed. In mice, the homologous 2 loci are within 10 cM of each other on chromosome 7. Miles et al. (1989) discussed the possibility that CYP2A rather than CYP2B is responsible for coumarin hydroxylase activity in the mouse. Miles et al. (1990) demonstrated that in the mouse Cyp2a is closely linked to Cyp2b (as are the homologous genes in the human) and also to Coh, which, by biochemical evidence (Negishi et al., 1989), is encoded by a member of the P450IIA gene subfamily.

Molecular Genetics

Wood and Conney (1974) found that basal and phenobarbital-induced rates of hepatic metabolism of coumarin to 7-hydroxycoumarin were markedly higher in DBA-2J mice than in other strains. Intermediate activities in hybrids indicated codominant inheritance. They predicted similar variability in man. Kratz (1976) studied coumarin 7-hydroxylase activity in liver obtained by needle biopsy. A 4-fold range was observed and interpreted as genetic. Persons taking drugs that might induce enzyme activity were excluded from the study.

Fernandez-Salguero et al. (1995) identified 3 different CYP2A6 alleles: the functional CYP2A6 allele; variant-1 (v1), which has a single base mutation (T to A) leading to a leu-to-his change in exon 3 (CYP2A6*2; 122720.0001); and v2, which is formed by gene conversion between the wildtype CYP2A6 and CYP2A7 genes in exons 3, 6, and 8 (CYP2A6*3; 122720.0004). Nakajima et al. (1996) also identified 3 CYP2A6 alleles: wildtype (CYP2A6*1) and 2 null, or inactive, alleles, CYP2A6*2 and CYP2A6*3.

Nicotine is the primary compound in tobacco that establishes and maintains tobacco dependence. Most of this nicotine is metabolized to cotinine by the CYP2A6 enzyme. Pianezza et al. (1998) showed that individuals lacking full functional CYP2A6 due to possession of null alleles CYP2A6*2 (122720.0001) and/or CYP2A6*3 (122720.0004), who therefore have impaired nicotine metabolism, are significantly protected against becoming tobacco-dependent smokers. In addition, smokers whose nicotine metabolism is thus impaired smoke significantly fewer cigarettes than those with normal nicotine metabolism. Individuals carrying CYP2A6-null alleles should have a decreased risk of developing tobacco-related cancers and other medical complications because they have a decreased risk of becoming a smoker and, if they do become dependent, they smoke less than those with normal nicotine metabolism. Since tobacco smoke contains nitrosamines that can be activated to carcinogens by CYP2A6, individuals who carry CYP2A6-null alleles may also be less efficient at activating tobacco smoke procarcinogens. These 3 factors may explain why there could be a reduction in tobacco-related cancers for carriers of CYP2A6-null alleles. Pianezza et al. (1998) found that among dependent smokers, the frequency of individuals with impaired nicotine metabolism (carriers of 1 or 2 CYP2A6-null alleles) was lower than in the control group (12.3% vs 19.6%). Even heterozygotes for a null allele showed significant reduction in the risk of tobacco dependence. CYP2A6 genotype may significantly affect nicotine levels from sources other than cigarettes, e.g., nicotine-replacement therapies for long-term maintenance against tobacco dependence and for treatment of other syndromes such as Alzheimer disease (104300), and Tourette syndrome (137580). The protective effect of CYP2A6-null alleles against the risk of becoming tobacco-dependent and in decreasing consumption suggests that inhibiting this enzyme may be a new way to help prevent and treat tobacco smoking.

Sabol and Hamer (1999) attempted to replicate the findings of Pianezza et al. (1998) by analyzing the CYP2A6 gene in a population of 385 individuals, using the same 2-step PCR assay described by Pianezza et al. (1998). They found no association between genotype and either smoking status or cigarette consumption. They then developed a single-step PCR method that is specific for the CYP2A6 locus and eliminated a high rate of false-positive mutations detected by the 2-step assay. Although this assay gave a much lower frequency of mutant alleles, there was again no association of the CYP2A6 genotype with smoking behavior.

Yokoi and Kamataki (1998) identified new mutations in the CYP2A6 and CYP2D6 (124030) genes in Japanese subjects. Oscarson et al. (1999) identified a deletion allele of CYP2A6 (122720.0002) which was rare in Europeans but had a frequency of 15.1% among 96 Chinese subjects. In the Chinese population, they detected no CYP2A6*2 alleles (122720.0001), in contrast to the frequency of 11 to 20% previously reported by Fernandez-Salguero et al. (1995).

In a study of 463 French adults, Gambier et al. (2005) found that subjects homozygous for CYP2A6*1B, a allele characterized by gene conversion in the 3-prime flanking region, smoked significantly more cigarettes per day as compared to those homozygous for CYP2A6*1A (wildtype), with a larger increase in their daily cigarette consumption over a 5-year period. No significant difference of smoking versus nonsmoking status was observed according to the CYP2A6 genotype.

Saito et al. (2003) provided a catalog of 680 variants among 8 CYP450 genes, 9 esterase genes, and 2 other genes in the Japanese population.

Mwenifumbo et al. (2008) characterized nonsynonymous CYP2A6 sequence variants among 281 individuals of black African descent with respect to their haplotype, allele frequency, and with in vivo CYP2A6 activity. The cohort could be categorized into normal, intermediate, and slow nicotine metabolism groups. In addition, alleles of individuals with unusual phenotype-genotype relationships were sequenced, resulting in the discovery of 5 novel uncharacterized alleles and at least 1 novel duplication allele. A total of 7% of this population of black African descent had at least one of the 8 novel characterized alleles, and 29% had at least 1 previously established allele. The findings could aid in the accuracy of association studies between CYP2A6 genotype and behavioral, disease, or pharmacologic phenotypes.

Animal Model

Paolini et al. (1999) found significant increases in the carcinogen-metabolizing enzymes CYP1A1 (108330), CYP1A2 (124060), CYP3A (124010), CYP2B (123930), and CYP2A in the lungs of rats supplemented with high doses of beta-carotene. The authors suggested that correspondingly high levels of CYPs in humans would predispose an individual to cancer risk from the widely bioactivated tobacco-smoke procarcinogens, thus explaining the cocarcinogenic effect of beta-carotene in smokers.

Gene Family

The cytochromes P-450 are among the major constituent proteins of the liver mixed function monooxygenases. They play a central role in the metabolism of steroids, the detoxification of drugs and xenobiotics, and the activation of procarcinogens. Most phase I metabolism of drugs and environmental pollutants is performed by cytochrome P-450 enzymes. In this process 1 or more water-soluble groups (such as hydroxyl) are introduced into the fat-soluble parent molecule, thereby rendering it vulnerable to attack by the phase II conjugating enzymes. The increased water-solubility of phase I and especially phase II products permits ready excretion. Examples of drug-metabolizing processes that are catalyzed by P-450 enzymes and show genetic variation include 4-hydroxylation of debrisoquine and N-oxidation of sparteine (see 124030). See review of Nebert and Gonzalez (1987).

Since the P450 superfamily is very ancient (the ancestral gene having existed more than 3.5 billion years ago, at a time predating drugs, animal-plant interactions, and combustion of organic matter), Nebert (1991) proposed that the P450 enzymes, as well as other so called 'drug-metabolizing' enzymes, play an important role in maintaining the steady-state levels of endogenous ligands involved in ligand-modulated transcription of genes effecting homeostasis, growth, differentiation, and neuroendocrine functions.

By December 14, 1992, Nelson et al. (1993) had accumulated a list of 221 P450 genes and 12 putative pseudogenes, representing 31 eukaryotes (including 11 mammalian and 3 plant species) and 11 prokaryotes. Of 36 gene families described to that time, 12 families had been found in all mammals examined. These 12 families comprised 22 mammalian subfamilies, of which 17 and 15 had been mapped to specific chromosomal sites in the human and mouse genomes, respectively. Each subfamily tend to be a cluster of tightly linked genes; there are exceptions.

Nelson et al. (1996) gave an update on the sequences, gene mapping, and nomenclature of the cytochrome P450 gene family.

Hoffman et al. (2001) reported the discovery of genes from 3 more CYP2 subfamilies inside the cluster of genes on chromosome 19 and assembled a complete map of the region. They reviewed the organization, structure, and expression of genes from all 6 subfamilies, and presented a general hypothesis for the evolution of this complex gene cluster.

Evolution

Within the mouse 2A subfamily, 2 P450s specifically catalyze either steroid 15-alpha-hydroxylase or coumarin 7-hydroxylase activity; the genes are designated Cyp2a-4 and Cyp2a-5, respectively. Aida et al. (1994) found that, whereas Mus musculus domesticus strains contains both genes, the wild mouse strain Mus spretus contains only Cyp2a-5. Evolutionarily, therefore, Cyp2a-5 is ancestral to Cyp2a-4. Moreover, the line to Cyp2a-4 descended as recently as 3 million years ago in an ancestral mouse. The evidence implied a rapid evolution of the P450 gene superfamily. The 2 genes are closely linked on mouse chromosome 7 (Lush and Andrews, 1978). Lindberg et al. (1992) also found evidence that a recent duplication in an ancestral mouse established a line of descent from the ancestral coumarin 7-hydroxylase gene to the gene encoding steroid 15-alpha-hydroxylase activity.

Nomenclature

'Cytochrome' means literally 'colored substance in the cell.' The color is derived from the subatomic properties of the iron in this hemoprotein, and, indeed, cytochromes appear reddish when present in sufficient concentration in the test-tube. 'P-450' denotes the unusual property of having its major optical absorption peak (Soret maximum) at about 450 nm, when the material has been reduced and combined with carbon monoxide (Omura and Sato, 1964). The name P-450 was intended to be temporary until more was known about the substance, but it has persisted because an ever-increasing complexity has been found and no agreement on a better nomenclature can be reached. It was the recommendation of Nebert (1986) that the genes be symbolized CYP1, CYP2, etc. It was further suggested (Nebert, 1986) that the dioxin-inducible P450 coded by chromosome 15 be called CYP1 and that the P450 coded by chromosome 19 be called CYP2A.

Nelson et al. (1993) noted that the Nomenclature Committee of the International Union of Biochemistry prefers the term 'heme-thiolate protein' instead of 'cytochrome' for P450 (Palmer and Reedijk, 1991). The original term 'cytochrome P-450' is a holdover from the provisional name given a protein by Sato and Omura (1961). These proteins are, in fact, not cytochromes. For the gene and cDNA, Nelson et al. (1993) recommended, as in the earlier reports from Nebert and his colleagues (e.g., Nebert, 1991), that the root symbol be CYP for human and Cyp for mouse, followed by an arabic number denoting the family, a letter designating the subfamily (when 2 or more exist), and an arabic number representing the individual gene within the subfamily. A hyphen should precede the final number in mouse genes. 'P' ('p' in mouse) after the gene number denotes a pseudogene. If a gene is a sole member of a family, the subfamily letter and gene number need not be included. It was recommended that the human nomenclature system be used for all species other than the mouse.