Isolation and characterisation of a novel rabbit sulfotransferase isoform belonging to the SULT1A subfamily

Sulfotransferases (SULTs) catalyse the sulfonation of both endogenous and exogenous compounds including hormones, catecholamines, drugs and xenobiotics. While in most occasions, sulfonation is a detoxication pathway, in the case of certain drugs and carcinogens, it leads to metabolic activation. Since, the rabbit has been extensively used for both pharmacological and toxicological studies, the purpose of this study was to further characterise the sulfotransferase system of this animal. In the present study, a novel sulfotransferase isoform (GenBank Accession no. AF360872) was isolated from a rabbit liver cDNA lambdaZAP II library. The full-length sequence of the clone was 1138 bp long and contained a coding region of 888 bp encoding a cytosolic protein of 295 amino acids (deduced molecular weight 34,193 Da). The amino acid sequence of this novel SULT isoform showed >70% identity with members of the SULT1A subfamily of sulfotransferases from other species. Upon expression of the encoded rabbit sulfotransferase in Escherchia coli (E. coli), it was shown that the enzyme was capable of sulfonating both p-nitrophenol (K(m) and Vmax values of 0.15 microM and 897.5 nmol/min/mg protein, respectively) and dopamine (K(m) and V(max) values of 175.3 microM and 151.1 nmol/min/mg protein, respectively). Based on the sequence data obtained and substrate specificity, this new rabbit sulfotransferase was named rabSULT1A1. Immunoblotting was used to demonstrate that rabSULT1A1 protein is expressed in liver, duodenum, jejunum, ileum, colon and rectum.

The serotonin transporter gene and Parkinson's disease

Dysfunction in the serotonin (5-hydroxytryptamine) system and reduced serotonin concentrations have been reported in patients with Parkinson's disease (PD). Serotonin concentrations in neural tissue are controlled by a presynaptic serotonin transporter protein that is encoded by a single gene. Therefore, we investigated whether a polymorphic region in the serotonin transporter gene is associated with PD. Three variable-number tandem repeat (VNTR) elements of the serotonin transporter gene were detected by polymerase chain reaction, those with 9, 10, 11 and 12 copies of the repeat element. The 10-copy VNTR element was significantly less common in patients with PD than controls in the univariate analysis (p < 0.05). Logistic regression analysis revealed no significant differences between patients (n = 198) and controls (n = 200) in the distribution frequencies of 9- and 12-copy alleles and combined genotypes (odds ratio = 1.20; p = 1.71). A positive family history of PD was a strong predictor of disease risk (odds ratio = 2.98; 95% confidence interval 1.51-5.87; p = 0.001). Although slight differences were observed between patient and control groups, these data suggest that defects in serotonin concentrations in patients with PD are unlikely to be due to polymorphisms in the serotonin transporter gene in this large Australian cohort; however, the inverse association observed with the 10-copy allele warrants further investigation.

Site-directed mutagenesis of the substrate-binding cleft of human estrogen sulfotransferase

The sulfonation of estrogens by human estrogen sulfotransferase (humSULT1E1) plays a vital role in controlling the active levels of these hormones in the body. To understand more fully the structural and functional characteristics of humSULT1E1, we have carried out site-directed mutagenesis of critical amino acids found in the substrate-binding cleft. Three single amino acid mutations of humSULT1E1 (V145E, H107A, and K85A) were created in this study. Kinetic studies were used to provide information about the importance of these residues in substrate specificity and catalysis, using a variety of substrates. Lysine at position 85 has been proposed to be within hydrogen bonding distance to the 3alpha-phenol group of beta-estradiol, thereby stabilising the substrate in the active site. However, substitution to a neutral alanine at this position improved substrate specificity of humSULT1E1 for beta-estradiol, estrone, and dehydroepiandrosterone (DHEA). The exchange of valine 145 for negatively charged glutamic acid markedly improved the ability of humSULT1E1 to sulfonate dopamine, but caused a reduction in specificity constants toward steroids tested, in particular DHEA. The presence of a histidine residue at position 107 was shown to be essential for the production of a functional protein, as substitution of this amino acid to alanine resulted in complete loss of activity of humSULT1E1 towards all substrates tested.

The epidemiology of Parkinson's disease in an Australian population

A prevalence study of Parkinson's disease (PD) was conducted in the rural town of Nambour, Australia. There were 5 cases of PD in a study population of 1207, yielding a crude prevalence ratio of 414 per 100,000 (95% confidence interval; 53-775). We performed a separate case-control study involving 224 patients with PD and 310 controls from South East Queensland and Central West New South Wales, to determine which factors increase the risk for PD in Australia. A positive family history of PD was the strongest risk factor for the development of the disease (odds ratio = 3.4; p < 0.001). In addition, rural residency was a significant risk factor for PD (odds ratio = 1.8, p < 0.001). Hypertension, stroke and well water ingestion were inversely correlated with the development of PD. There was no significant difference between patients and controls for exposure to herbicides and pesticides, head injury, smoking or depression. The high prevalence of PD in Nambour may be explained by rural residency. However, the most significant risk factor for PD was a positive family hisotry. This demonstrates the need for improved understanding of the genetic nature of the disease.

Localization of N-acetyltransferases NAT1 and NAT2 in human tissues

Human acetyl coenzyme A-dependent N-acetyltransferase (EC 2.3.1.5) (NAT) catalyzes the biotransformation of a number of arylamine and hydrazine compounds. NAT isozymes are encoded at 2 loci; one encodes NAT1, formerly known as the monomorphic form of the enzyme, while the other encodes the polymorphic NAT2, which is responsible for individual differences in the ability to acetylate certain compounds. Human epidemiological studies have suggested an association between the "acetylator phenotype" and particular cancers such as those of the bladder and colon. In the present study, NAT1- and NAT2-specific riboprobes were used in hybridization histochemistry studies to localize NAT1 and NAT2 mRNA sequences in formalin-fixed, paraffin-embedded human tissue sections. Expression of both NAT1 and NAT2 mRNA was observed in liver, gastrointestinal tract tissues (esophagus, stomach, small intestine, and colon), ureter, bladder, and lung. In extrahepatic tissues, NAT1 and NAT2 mRNA expression was localized to intestinal epithelial cells, urothelial cells, and the epithelial cells of the respiratory bronchioles. The observed heterogeneity of NAT1 and NAT2 mRNA expression between human tissue types may be of significance in assessing their contribution to known organ-specific toxicities of various arylamine drugs and carcinogens.

Genetically modified Chinese hamster ovary cells for investigating sulfotransferase-mediated cytotoxicity and mutation by 2-amino-1-methyl-6- phenylimidazo[4,5-b]pyridine

To test the hypothesis that the sulfotransferase gene plays a role in the phase II bioactivation of PhIP, a heterocyclic amine found in cooked meats, we transfected the UV5P3 cell line with cDNA plasmids of human aryl sulfotransferases (HAST1 and HAST3). UV5P3 is a nucleotide excision repair-deficient and P4501A2-expressing CHO cell line that we have previously developed. Functionally transformed clones were identified by the differential cytotoxicity (DC) assay that used PhIP as the cytotoxic agent. Two clones designated 5P3H1 and 5P3H3, expressing HAST1 and HAST3, respectively, were chosen for further characterization. Correct fragment sizes of the sulfotransferase cDNAs were identified in both cell lines by polymerase chain reaction. Immunoblot analysis confirmed the expression of the sulfotransferase proteins. The addition of the sulfotransferase inhibitor DCNP decreased the cytotoxic effects of PhIP in a dose-dependent manner. The increase in cell growth was 6. 5-fold for 5P3H1 and 2.4-fold for 5P3H3, relative to values obtained without DCNP. Based on D(50) values, the dose that reduced the survival to 50% relative to untreated controls, the cytotoxic effect of PhIP was increased threefold for 5P3H1 and 1.87-fold for 5P3H3 cell lines over the parental UV5P3 line. There was also a small increase in the mutation response at the aprt locus. These newly established 5P3H1 and 5P3H3 sulfotransferase-expressing cells provide valuable mechanistic information of the bioactivation of PhIP and related compounds. Environ. Mol. Mutagen. 35:57-65, 2000. Published 2000 Wiley-Liss, Inc.

Crystal structure of human catecholamine sulfotransferase

Sulfonation, like phosphorylation, can modify the activity of a variety of biological molecules. The sulfotransferase enzymes sulfonate neurotransmitters, drugs, steroid hormones, dietary carcinogens and proteins. SULT1A3 specifically sulfonates catecholamines such as dopamine, adrenaline and noradrenaline. The crystal structure of SULT1A3 with a sulfate bound at the active site, has been determined at 2.4 A resolution. Although the core alpha/beta fold is like that of estrogen and heparan sulfotransferases, major differences occur in and around the active site. Most notably, several regions surrounding the active site, including a section of 40 residues, are disordered in SULT1A3. Regions that are topologically equivalent to the disordered parts of SULT1A3 are involved in substrate and cofactor binding in estrogen and heparan sulfotransferase. Flexibility in these regions suggests that ligand binding elicits a disorder-order transition in and around the active site of sulfotransferases and might contribute to the broad substrate specificity of these enzymes.

Analysis of the substrate specificity of human sulfotransferases SULT1A1 and SULT1A3: site-directed mutagenesis and kinetic studies

Sulfonation is an important metabolic process involved in the excretion and in some cases activation of various endogenous compounds and xenobiotics. This reaction is catalyzed by a family of enzymes named sulfotransferases. The cytosolic human sulfotransferases SULT1A1 and SULT1A3 have overlapping yet distinct substrate specificities. SULT1A1 favors simple phenolic substrates such as p-nitrophenol, whereas SULT1A3 prefers monoamine substrates such as dopamine. In this study we have used a variety of phenolic substrates to functionally characterize the role of the amino acid at position 146 in SULT1A1 and SULT1A3. First, the mutation A146E in SULT1A1 yielded a SULT1A3-like protein with respect to the Michaelis constant for simple phenols. The mutation E146A in SULT1A3 resulted in a SULT1A1-like protein with respect to the Michaelis constant for both simple phenols and monoamine compounds. When comparing the specificity of SULT1A3 toward tyramine with that for p-ethylphenol (which differs from tyramine in having no amine group on the carbon side chain), we saw a 200-fold preference for tyramine. The kinetic data obtained with the E146A mutant of SULT1A3 for these two substrates clearly showed that this protein preferred substrates without an amine group attached. Second, changing the glutamic acid at position 146 of SULT1A3 to a glutamine, thereby neutralizing the negative charge at this position, resulted in a 360-fold decrease in the specificity constant for dopamine. The results provide strong evidence that residue 146 is crucial in determining the substrate specificity of both SULT1A1 and SULT1A3 and suggest that there is a direct interaction between glutamic acid 146 in SULT1A3 and monoamine substrates.

Structural characterization of human aryl sulphotransferases

Human aryl sulphotransferase (HAST) 1, HAST3, HAST4 and HAST4v share greater than 90% sequence identity, but vary markedly in their ability to catalyse the sulphonation of dopamine and p-nitrophenol. In order to investigate the amino acid(s) involved in determining differing substrate specificities of HASTs, a range of chimaeric HAST proteins were constructed. Analysis of chimaeric substrate specificities showed that enzyme affinities are mainly determined within the N-terminal end of each HAST protein, which includes two regions of high sequence divergence, termed Regions A (amino acids 44-107) and B (amino acids 132-164). To investigate the substrate-binding sites of HASTs further, site-directed mutagenesis was performed on HAST1 to change 13 individual residues within these two regions to the HAST3 equivalent. A single amino acid change in HAST1 (A146E) was able to change the specificity for p-nitrophenol to that of HAST3. The substrate specificity of HAST1 towards dopamine could not be converted into that of HAST3 with a single amino acid change. However, compared with wild-type HAST1, a number of the mutations resulted in interference with substrate binding, as shown by elevated Ki values towards the co-substrate 3'-phosphoadenosine 5'-phosphosulphate, and in some cases loss of activity towards dopamine. These findings suggest that a co-ordinated change of multiple amino acids in HAST proteins is needed to alter the substrate specificities of these enzymes towards dopamine, whereas a single amino acid at position 146 determines p-nitrophenol affinity. A HAST1 mutant was constructed to express a protein with four amino acids deleted (P87-P90). These amino acids were hypothesized to correspond to a loop region in close proximity to the substrate-binding pocket. Interestingly, the protein showed substrate specificities more similar to wild-type HAST3 than HAST1 and indicates an important role of these amino acids in substrate binding.

Structural and functional characterisation of human sulfotransferases

The human aryl sulfotransferases HAST4 and HAST4v vary by only two amino acids but exhibit markedly different affinity towards the sulfonate acceptor p-nitrophenol and the sulfonate donor 3'-phosphoadenosine-5'-phosphosulfate (PAPS). To determine the importance of each of these amino acid differences, chimeric constructs were made of HAST4 and HAST4v. By attaching the last 120 amino acids of HAST4v to HAST4 (changing Thr235 to Asn235) we have been able to produce a protein that has a Km for PAPS similar to HAST4v. The reverse construct, HAST4v/4 produces a protein with a Km for PAPS similar to HAST4. These data suggests that the COOH-terminal of sulfotransferases is involved in co-factor binding.

Localisation of aryl sulfotransferase expression in human tissues using hybridisation histochemistry and immunohistochemistry

To date, the laboratory has cloned seven unique human sulfotransferases; five aryl sulfotransferases (HAST1, HAST2, HAST3, HAST4 and HAST4v), an estrogen sulfotransferase and a dehydroepiandrosterone sulfotransferase. The cellular distribution of human aryl sulfotransferases in human hepatic and extrahepatic tissues has been determined using the techniques of hybridization histochemistry and immunohistochemistry. Human aryl sulfotransferase expression was detected in liver, epithelial cells of the gastrointestinal mucosal layer, epithelial cells lining bronchioles and in mammary duct epithelial cells.

Bacterial expression of two human aryl sulfotransferases

The effect of replacing a single codon in the N-terminal of human aryl sulfotransferase (HAST) 1 and 3 with one that is more commonly found in E. coli genes was assessed. The pKK233-2 E. coli expression vector was employed and the polymerase chain reaction (PCR) was used to introduce the 5' nucleotide substitution, at the same time maintaining the fidelity of the amino acid sequence. The data indicates that this change had a minimal effect on protein production, subcellular localization or, in the case of HAST3, catalytic activity. In general, the pKK233-2 E. coli vector has been less than optimal for expressing human sulfotransferase cDNAs.

The role of xenobiotic metabolizing enzymes in arylamine toxicity and carcinogenesis: functional and localization studies

In both animal models and humans, the first and obligatory step in the activation of arylamines is N-hydroxylation. This pathway is primarily mediated by the phase-I enzymes CYP1A1, CYP1A2 and CYP4B1. In the presence of flavonoids such as alpha-naphthoflavone and flavone, both CYP3A4 and CYP3A5 have also been shown to play a minor role in the activation of food-derived heterocyclic amines. The further activation of N-hydroxyarylamines by phase-II metabolism can involve both N, O-acetylation and N, O-sulfonation catalyzed by N-acetyltransferases (NAT1 and NAT2) and sulfotransferases, respectively. Using an array of techniques, we have been unable to detect constitutive CYP1A expression in any segments of the human gastrointestinal tract. This is in contrast to the rabbit where CYP1A1 protein was readily detectable on immunoblots in microsomes prepared from the small intestine. In humans, CYP3A3/3A4 expression was detectable in the esophagus and all segments of the small intestine. Northern blot analysis of eleven human colons showed considerable heterogeneity in CYP3A mRNA between individuals, with the presence of two mRNA species in some subjects. Employing the technique of hybridization histochemistry (also known as in situ hybridization), CYP4B1 expression was observed in some human colons but not in the liver or the small intestine. Hybridization histochemistry studies have also demonstrated variable NAT1 and NAT2 expression in the human gastrointestinal tract. NAT1 and NAT2 mRNA expression was detected in the human liver, small intestine, colon, esophagus, bladder, ureter, stomach and lung. Using a general aryl sulfotransferase riboprobe (HAST1), we have demonstrated marked sulfotransferase expression in the human colon, small intestine, lung, stomach and liver. These studies demonstrate that considerable variability exists in the expression of enzymes involved in the activation of aromatic amines in human tissues. The significance of these results in relation to a role for heterocyclic amines in colon cancer is discussed.

Localization of cytochromes P450 in human tissues: implications for chemical toxicity

Cytochromes P450 comprise a remarkably diverse superfamily of heme-thiolate proteins critical in the metabolism of numerous endogenous ligands and xenobiotics. Among the myriad of P450 substrates are many compounds of toxicological and pharmacological significance. The precise complement of cytochrome P450 isoforms in any given tissue may therefore be an important determinant of susceptibility to chemical-mediated toxicity. We have used a histological approach to study the distribution of individual P450s in human and rabbit gastro-intestinal tissues. We have focused primarily on P450 enzymes of importance in the metabolism of carcinogens, namely CYP1A1, CYP1A2, CYP2E1, CYP3A4/3A5 and CYP4B1. Here we give an overview of the distribution of these enzymes in human and rabbit tissues and discuss the possible toxicological implications of the results. In addition we will discuss the value of archival human tissue specimens for histological analysis of P450 distribution.

cDNA cloning and expression of a new form of human aryl sulfotransferase

To date, four human cytosolic sulfotransferases have been cloned and characterised. The aim of the present study was to identify new forms of these enzymes using molecular cloning techniques. Two full length human aryl sulfotransferase (HAST) cDNAs were cloned from a lambda gt10 liver cDNA library. The COS cell expression system was used to express the cDNAs and to determine the ability of the encoded proteins to metabolise the model substrates p-nitrophenol and dopamine. The two cDNAs were 1036 bp (HAST4) and 1060 bp (HAST4v) in length, and encoded proteins that differed by two amino acids (Thr-7 to Ile and Thr-235 to Asn). The coding domains of HAST4 and HAST4v were 97 and 94% homologous to previously reported phenol (HAST1) and monoamine (HAST3) sulfonating forms of sulfotransferase, respectively. On expression of these cDNAs in COS cells the encoded proteins were capable of sulfonating p-nitrophenol with markedly different affinities: the K(m)s for HAST4 and HAST4v being 73.7 and 7.75 microM, respectively. For the same reaction HAST1 and HAST3 have K(m)s of 0.7 and 2200 microM, respectively. Unlike HAST1 and HAST3, the expressed HAST4/4v proteins could not sulfonate dopamine. In addition to having markedly different K(m)s for p-nitrophenol as a substrate, the expressed HAST4/4 proteins also differed significantly in their affinity for the cofactor 3'-phosphoadenosine-5'-phosphosulfate. This report on the functional dissimilarity between two allelic variants of HAST4 highlights that substitution at two residues, Thr-7 and -235, markedly alters their substrate specificities and provides insight into the domains that determine these characteristics.

Photoaffinity labeling of human recombinant sulfotransferases with 2-azidoadenosine 3',5'-[5'-32P]bisphosphate

Photoaffinity labeling with 2-azidoadenosine 3', 5'-[5'-32P]bisphosphate was used to identify and characterize adenosine 3',5'-bisphosphate-binding proteins in human liver cytosol and recombinant sulfotransferase proteins. The sulfotransferases investigated in these studies were the human phenol sulfotransferases, HAST1, -3, and -4, dehydroepiandrosterone sulfotransferase, and estrogen sulfotransferase. The cDNAs for these enzymes have been previously cloned and expressed in COS-7 cells or Escherichia coli. Photoaffinity labeling of all proteins was highly dependent on UV irradiation, was protected by co-incubation with unlabeled adenosine 3',5'-bisphosphate and phosphoadenosine phosphosulfate, and reached saturation at concentrations above 10 microM. To verify that the 31 35-kDa photolabeled proteins were indeed sulfotransferases, specific antibodies known to recognize human sulfotransferases were used for Western blot analyses of photolabeled proteins. It was shown unequivocally that the proteins in the 31-35-kDa region recognized by the antibodies also photoincorporated 2-azidoadenosine 3',5'-[5'-32P]bisphosphate. This is the first application of photoaffinity labeling with 2-azidoadenosine 3',5'-[5'-32P]bisphosphate for the characterization of recombinant human sulfotransferases. Photoaffinity labeling will be also useful in the purification and functional identification of other adenosine 3',5'-bisphosphate-binding proteins and to determine amino acid sequences at or near their active sites.

Stereoselective and regioselective hydration of 7-methylbenz[c]acridine-5,6-oxide enantiomers by rodent and human microsomal epoxide hydrolases

In the present study, we studied the regioselectivity and stereoselectivity of human microsomal epoxide hydrolase-catalyzed hydration of the enantiomers of the polycyclic aza-aromatic hydrocarbon K-region oxide, 7-methylbenz[c]acridine-5,6-oxide. We used a human microsomal epoxide hydrolase cDNA amplified from a liver cDNA library and expressed in COS-7 cells. Comparisons were made with the activities of rat and HLM preparations. The determination of the apparent Michaelis-Menten kinetic constants revealed that microsomal epoxide hydrolase, regardless of the source, exhibited enantioselectivity, with the 5S,6R-oxide being the preferred substrate. Regioselectivity of hydration for each stereoisomer was determined. Expressed human microsomal epoxide hydrolase and HLM catalyzed the attack of water predominantly (approximately 96%) at C5 of the 5R,6S-oxide, whereas 5S,6R-oxide was attacked less selectivity (approximately 60% at C5). These results are discussed in the context of available literature on the regioselectivity and stereoselectivity of rat and rabbit microsomal epoxide hydrolase and represents the first examination of human microsomal epoxide hydrolase regarding its regioselectivity and stereoselectivity of hydration.

Metabolism of polycyclic aza-aromatic carcinogens catalyzed by four expressed human cytochromes P450

The role of human cytochromes P4501A1, -1A2, -3A4, and -3A5 in the metabolism of the polycyclic aza-aromatic hydrocarbons 7-methylbenz(c)acridine and dibenz(aj)acridine was investigated. The regioselectivity of the reactions was determined, as well as the associated stereoselectivity in the production of dihydrodiol metabolites and K-region oxides. Metabolite distributions were also examined in the presence of the epoxide hydrolase inhibitor 1,1,1-trichloropropylene-2,3-oxide and the P450 modulator alpha-naphthoflavone. P4501A2 was most regioselective for the production of the proximate carcinogen; the 3,4-dihydrodiol of 7-methylbenz(c)acridine and P4503A4 showed the highest regioselectivity for K-region oxidation. In contrast, the analogous putative proximate carcinogen of dibenz(aj)acridine was formed with the highest relative abundance by P4503A4, while P4501A2 was most regioselective for K-region oxidation. For both compounds the proximate carcinogens possessed predominantly the 3R,4R-absolute configuration, independent of the P450 catalyzing the reaction. The K-region dihydrodiols of 7-methylbenz(c)acridine were formed with no stereoselectivity, except with P4501A2 which favored production of the S,S isomer. In contrast the K-region dihydrodiol of dibenz(aj)acridine was formed by P4501A1 and P4501A2 as the R,R isomer with almost 100% optical purity. P4501A2 and 3A4 showed no stereoselectivity in the formation of the K-region oxide of 7-methylbenz(c)acridine, while P4501A1 produced the 5R,6S-oxide with low optical purity. For dibenz(aj)acridine 5,6-oxide, P4501A1 predominantly formed 5S,6R-oxide (80% pure). These results emphasize the importance of the composition and levels of expressed P450s of an individual in relation to the activation and detoxification of toxicants.

Characterisation of CYP3A gene subfamily expression in human gastrointestinal tissues

The human CYP3A subfamily is of interest due to its multiplicity, activity toward known carcinogens, and extrahepatic expression. In situ hybridisation analysis of formalin fixed, routinely processed biopsy specimens was used to localise CYP3A mRNA in human gastrointestinal tissues from several individuals. CYP3A mRNA is abundant in human liver and in mucosal epithelial cells of all segments of the human small intestine. RNA blot analyses showed that the mRNA species observed in most livers and in human small intestine represent CYP3A3/3A4 transcripts. This was confirmed at the protein level by immunoblot comparison of small intestine microsomes to in vitro expressed CYP3A4 and CYP3A5 proteins. In liver and small intestine, CYP3A mRNA is not uniformly distributed, with grain density highest in cells within the respective non-proliferative compartments. CYP3A mRNA was also observed in human oesophagus and colon. RNA blot analysis of multiple colons showed heterogeneity in the CYP3A mRNAs present. Two CYP3A mRNAs (CYP3A3/3A4 and CYP3A5) were detected in colon samples from several individuals. In addition to those localisation studies, the capacity of expressed CYP3A4 and CYP3A5 to activate the dietary heterocyclic amine MeIQ in the presence of alpha-naphthoflavone was shown. These results show that there is considerable heterogeneity in the expression of the CYP3A subfamily in human gastrointestinal tissues.

Species-specific expression of CYP4B1 in rabbit and human gastrointestinal tissues

CYP4B1 is a P450 enzyme displaying tissue and species specific regulation. The rabbit CYP4B1 enzyme exhibits activity towards procarcinogenic aromatic amines. In the present study, CYP4B1 expression has been characterized in rabbit tissues using histological techniques, Northern blotting and Western blotting. Similar analyses were attempted using available human tissues. CYP4B1 mRNA and protein was demonstrated throughout the rabbit small intestine and colon. A unique 1.8 kb transcript, that is smaller than the transcript found in other tissues, was detected in rabbit stomach with a CYP4B1 specific RNA probe. No CYP4B1 protein was detected in this tissue. In rabbit liver, CYP4B1 was induced by phenobarbital primarily in zone 1 hepatocytes (periportal). In humans, CYP4B1 expression was demonstrated at low levels in human colon using in situ hybridization but not in liver or the small intestine. All rabbit gastrointestinal tissues other than stomach possess a high capacity for the activation of 2-aminofluorene compatible with CYP4B1 expression. In contrast, no activity was observed in human gastrointestinal microsomes. The present study therefore shows that CYP4B1 is an abundant P450 in the rabbit gastrointestinal tract and identifies species-specific differences in CYP4B1 expression and function.

Functional characterization of two human sulphotransferase cDNAs that encode monoamine- and phenol-sulphating forms of phenol sulphotransferase: substrate kinetics, thermal-stability and inhibitor-sensitivity studies

The present paper describes the functional characterization of two human aryl sulphotransferase (HAST) cDNAs, HAST1 and HAST3, previously isolated by us from liver and brain, respectively [Zhu, Veronese, Sansom, and McManus (1993) Biochem. Biophys. Res. Commun. 192, 671-676; Zhu, Veronese, Bernard, Sansom and McManus (1993) Biochem. Biophys. Res. Commun. 195, 120-127]. These appear to encode the two major forms of phenol sulphotransferase (PST) characterized in a number of human tissue cytosols, these being the phenolsulphating (P-PST) and monoamine-sulphating (M-PST) forms of phenol sulphotransferase. HAST1 and HAST3 cDNAs were functionally expressed in COS-7 cells and kinetically characterized using the model substrates for P-PST and M-PST, p-nitrophenol and dopamine (3,4-dihydroxyphenethylamine) respectively. COS-expressed HAST1 was shown to be enzymatically active in sulphating p-nitrophenol with high affinity (Km 0.6 microM), whereas dopamine was the preferred substrate for HAST3 (Km 9.7 microM). HAST1 could also sulphate dopamine, as could HAST3 sulphate p-nitrophenol, but the Km for these reactions were at least two orders of magnitude greater than for the preferred substrates. COS-expressed HAST1 and HAST3 displayed inhibition profiles with the ST inhibitor 2,6-dichloro-4-nitrophenol (DCNP), identical with human liver cytosolic P-PST and M-PST activities respectively. Thermal-stability studies with the expressed enzymes showed that HAST1 was considerably more thermostable (TS) than HAST3, which is consistent with P-PST being termed the TS PST and M-PST being termed the thermolabile (TL) PST. Western immunoblot analyses of the expressed PST proteins using an antibody generated to a bacterially expressed rat liver aryl/phenol ST showed that HAST1 and HAST3 migrated as single proteins with different electrophoretic mobilities (32 versus 34 kDa). This is consistent with the differences in electrophoretic mobilities observed for P-PST and M-PST in a variety of tissues reported by other workers. This report on the functional characterization of P-PST and M-PST cDNAs provides important information on the structural as well as functional relationships of human PSTs, which sulphate a vast array of exogenous and endogenous compounds.

Heterologous systems for expression of mammalian sulfotransferases

This paper describes the use of both mammalian and bacterial expression systems as tools to study the structural and functional relationships of proteins encoded by cDNAs to both rat and human aryl sulfotransferases. In particular, we describe the use of the mammalian COS cell system for functional expression studies, and the use of Escherichia coli for the expression and purification of a sulfotransferase fusion protein suitable as an antigen for the generation of sulfotransferase antibodies.

Caffeine metabolism by human hepatic cytochromes P450: contributions of 1A2, 2E1 and 3A isoforms

Caffeine (CA) N1-, N3- and N7-demethylase, CA 8-hydroxylase and phenacetin O-deethylase activities were measured in microsomes from 18 separate human livers which had been characterized previously for a range of cytochrome P450 (CYP) isoform-specific activities and immunoreactive CYP protein contents. Correlations between the high affinity components of the three separate CA N-demethylations were highly significant (r = 0.77-0.91, P < 0.001) and each of the three high affinity CA N-demethylations correlated significantly (r = 0.64-0.93, P < 0.05-0.001) with the high affinity phenacetin O-deethylase, 2-acetylaminofluorene N-hydroxylation and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) and 2-amino-3-methylimidazo[4,5-f]quinoline (IQ) mutagenicity (all predominantly CYP1A2-mediated reactions). Consistent with these observations, cDNA-expressed human CYP1A2 catalyzed the N1-, N3- and N7-demethylation of CA and apparent Km values were similar (0.24-0.28 mM) for all three reactions and comparable to those observed previously with human liver microsomes. The low affinity components of CA N1- and N7-demethylation correlated significantly (r = 0.55-0.85, P < 0.05-0.001) with immunoreactive CYP2E1 content and the CYP2E1-specific activities 4-nitrophenol and chlorzoxazone hydroxylation. Diethyldithiocarbamate, a selective inhibitor of CYP2E1, inhibited the low affinity CA N1- and N7-demethylation, with IC50 values of 23 microM and 11 microM, respectively. The apparent Km values for CA N1- and N7-demethylation by cDNA-expressed CYP2E1 (namely 28 and 43 mM, respectively) were of a similar order to those calculated for the low affinity microsomal activities. Significant correlations (r = 0.87-0.97, P < 0.001) were observed between CA 8-hydroxylation and immunoreactive CYP3A content and the CYP3A-mediated reactions benzo(a)pyrene hydroxylation, omeprazole sulfoxidation and aflatoxin B1 mutagenesis. Effects of alpha-naphthoflavone, erythromycin, troleandomycin and nifedipine on microsomal CA 8-hydroxylation were generally consistent with CYP3A involvement. Taken together with previous data, the results indicate a major involvement of CYP1A2 in the high affinity component of all three human hepatic CA N-demethylations. In contrast, CYP2E1 appears to be the main enzyme involved in the low affinity components of CA N1- and N7-demethylation while CA 8-hydroxylation is catalysed predominantly by a CYP3A isoform(s).

The role of cytochrome P450 2C3 in rabbit liver microsomal metabolism of 1-nitropyrene and 3-nitrofluoranthene

The male rabbit liver microsomal cytochrome P450 metabolism of 1-nitropyrene and 3-nitrofluoranthene was investigated. In this study, we used inhibitory antibodies specific for rabbit P450 2C3 and determined that, in untreated male rabbit liver microsomes, the antibody inhibited approximately 75% of the cytochrome P450-mediated C-oxidation of both 1-nitropyrene and 3-nitrofluoranthene. These results verify our previous prediction that mainly one cytochrome P450 is responsible for microsomal metabolism of 1-nitropyrene in the untreated rabbit liver.

Metabolic differences in colon mucosal cells

The colonic expression of cytochromes P450 from the CYP1A, CYP3A and CYP4B subfamilies has been characterized in rabbit and human tissues using RNA blotting, immunoblotting, immunohistochemistry and hybridization histochemistry. These studies demonstrate negligible expression of the CYP1A subfamily in either rabbit or human colon. The CYP3A6 gene is expressed in rabbit colon although at markedly reduced levels relative to liver and small intestine. Whilst at least two CYP3A genes are expressed at the mRNA level in human colon tissue from some individuals, no expression was demonstrated in others. Where expression was observed, this expression was continuous throughout the length of the colon. In rabbits, CYP4B1 represents a major colon P450 enzyme, expressed at levels in colon comparable to liver and small intestine. In contrast, the human CYP4B1 gene is expressed at low levels in some individuals. These studies highlight individual differences in the expression of cytochrome P450 enzymes of importance in procarcinogen metabolism.

Identification of two human brain aryl sulfotransferase cDNAs

A 1,179 bp and a 1,424 bp full-length aryl sulfotransferase cDNAs were isolated from a human brain cDNA library. Their coding domains are 93% identical, each encoding a cytosolic protein of 295 amino acids. Their deduced amino acid sequences of these cDNAs are also 93% identical. The 1179 bp brain cDNA has an identical coding domain to a previously reported human liver aryl sulfotransferase cDNA but it has a different 5' noncoding sequence. Northern blot analysis using a probe specific for the 1,424 bp cDNA identified a 1500 bp band in mRNA of human liver, colon, kidney and lung. In a human hepatocellular carcinoma the same band plus an extra larger band was also recognised. An intron of the gene encoding the 1424 bp cDNA was also identified.

The catalytic activity of four expressed human cytochrome P450s towards benzo[a]pyrene and the isomers of its proximate carcinogen

The catalytic properties of expressed human cytochromes P4501A1, 1A2, 3A4 and 3A5 with respect to the metabolite distribution for the carcinogen benzo[a]pyrene (BP) and the proximate carcinogen (-)-trans-7,8-dihydroxy-7,8-dihydrobenzo[a]pyrene ((-)-BP-7,8-DHD) were determined. P4501A1 formed a higher proportion of the proximate carcinogen from BP than the other isozymes. Both P4501A1 and 1A2 were more selective for oxidation at the benzo ring of BP and showed a greater production of dihydrodiol metabolites than either P4503A4 or 3A5. The formation of diolepoxides from the individual enantiomers of BP-7,8-DHD occurred in a highly stereoselective fashion.

Molecular characterisation of a human aryl sulfotransferase cDNA

A full-length aryl sulfotransferase cDNA was isolated from a human liver cDNA library. It was 1155 bp long containing a coding region of 885 basepairs encoding a cytosolic protein (M(r) 34178 Da) of 295 amino acids. This human cDNA shared 80% homology to the rat aryl sulfotransferase cDNA, 58% to the bovine and rat oestrogen sulfotransferase cDNAs, 53% to the rat hydroxysteroid sulfotransferase cDNA and 51% to the human liver dehydroepiandrosterone sulfotransferase cDNA over its whole 885 bp coding region. The deduced amino acid sequence of this human cDNA was 79% homologous to that of the rat aryl sulfotransferase cDNA and the putative common-substrate binding site motif GXXGXXK of the sulfotransferases has been conserved in this human amino acid sequence. At least two sizes of this human aryl sulfotransferase mRNA were detected in the human liver and lung.

Specificity of substrate and inhibitor probes for human cytochromes P450 1A1 and 1A2

Kinetic and inhibitor studies using cDNA-expressed enzymes and human liver microsomes have characterized the specificity of a range of cytochrome P450 (CYP) 1A substrate and inhibitor probes towards the two isoforms comprising this subfamily. Expressed CYP1A1 and CYP1A2 both catalyzed the O-deethylation of phenacetin, although the apparent Km was about 4-fold lower for CYP1A2 (25 vs. 108 microM). Phenacetin O-deethylation exhibited biphasic kinetics in human liver microsomes, and the apparent Km for the high-affinity component (9 +/- 6 microM) was consistent with the involvement of CYP1A2 in this reaction. The prototypic CYP1A xenobiotic inhibitor and substrate probes alpha-naphthoflavone, ellipticine, 7-ethoxycoumarin and 7-ethoxyresorufin all inhibited CYP1A1- and CYP1A2-mediated phenacetin O-deethylation as well as the high-affinity component of human liver phenacetin O-deethylase activity. alpha-Naphthoflavone and 7-ethoxycoumarin were, however, approximately 10-fold more potent as inhibitors of CYP1A2 than CYP1A1. Other putative human CYP1A xenobiotic substrates and inhibitors, including caffeine, 5- and 8-methoxypsoralen, nifedipine, paraxanthine, propranolol and theophylline similarly inhibited CYP1A1- and 1A2-catalyzed phenacetin O-deethylation and the high-affinity human liver phenacetin O-deethylase. In contrast, the monoclonal antibody MAb 1-7-1, raised against 3-methylcholanthrene-inducible rat cytochromes 450, almost abolished CYP1A1-mediated phenacetin O-deethylation, but had no effect on human liver microsomal- or CYP1A2-catalyzed phenacetin dealkylation. Together with previous data, the results indicate that the majority of human CYP1A xenobiotic inhibitor and substrate probes are nonspecific in their recognition of CYP1A1 and CYP1A2, although selectivity is apparent for some compounds.

cDNA expression studies of rat liver aryl sulphotransferase

A cDNA encoding an isoenzyme of rat liver aryl sulphotransferase was isolated from a rat liver bacteriophage Lambda gt 11 library by the polymerase chain reaction technique. The resulting cDNA was functionally expressed in COS-7 cells and characterised by determining the sulphating capacity of the cells with a range of substrates. The COS-expressed enzyme catalysed the sulphation of both phenol and dopamine with Kms of the same order as those obtained for the high affinity isozyme in rat liver cytosol, while low activity was observed with tyrosine methyl ester. The common food additive vanillin was also a good substrate for sulphate conjugation. The sulphation of vanillin catalysed by the COS-expressed enzyme was consistent with a single enzyme system, in contrast, the kinetics of the reaction catalysed by cytosolic sulphotransferase indicated that vanillin was sulphated by more than one isozyme.

Site-directed mutation studies of human liver cytochrome P-450 isoenzymes in the CYP2C subfamily

Evidence from human studies in vivo and in vitro strongly suggests that the methylhydroxylation of tolbutamide and the 4-hydroxylation of phenytoin, the major pathways in the elimination of these two drugs, are catalysed by the same cytochrome P-450 isoenzyme(s). In the present study we used site-directed mutagenesis and cDNA expression in COS cells to characterize in detail the kinetics of tolbutamide and phenytoin hydroxylations by seven CYP2C proteins (2C8, 2C9 and variants, and 2C10) in order to define the effects of small changes in amino acid sequences and the likely proteins responsible in the metabolism of these two drugs in man. Tolbutamide was hydroxylated to varying extents by all expressed cytochrome P-450 isoenzymes, although activity was much lower for the expressed 2C8 protein. While the apparent Km values for the 2C9/10 isoenzymes (71.6-131.7 microM) were comparable with the range of apparent Km values previously observed in human liver microsomes, the apparent Km for 2C8 (650.5 microM) was appreciably higher. The 2C8 enzyme also showed quite different sulphaphenazole inhibition characteristics. The 4-hydroxylation of phenytoin was also more efficiently catalysed by the 2C9/10 enzymes. These enzymes showed similarities in kinetics of phenytoin hydroxylation and sulphaphenazole inhibition compared with human liver phenytoin hydroxylase. Also of interest was the observation that, among the 2C9 variants, small differences in amino acid composition could appreciably affect both tolbutamide and phenytoin hydroxylations. The amino acid substitution Cys-144-->Arg increased both the rates of tolbutamide and phenytoin hydroxylations, while the Leu-359-->Ile change had a greater effect on phenytoin hydroxylation. We conclude that: (1) although 2C8 and 2C9/10 proteins metabolize tolbutamide. only 2C9/10 proteins play a major role in human liver; (2) 2C9/10 proteins also appear to be chiefly responsible for phenytoin hydroxylation; and (3) subtle differences in the amino acid composition of these 2C9/10 proteins can affect the functional specificities towards both tolbutamide and phenytoin.

Co-regulation of phenytoin and tolbutamide metabolism in humans

1. The disposition of phenytoin and tolbutamide was compared in eighteen healthy young adults separately administered single therapeutic doses (sodium phenytoin 300 mg, tolbutamide 500 mg) of the two drugs. 2. Within the group, ratios of ranges of total and unbound areas under the plasma concentration-time curves were similar for both drugs. 3. There were significant (P < 0.001) correlations between total (r = 0.88) and unbound (r = 0.86) areas under the plasma phenytoin and tolbutamide concentration-time curves. 4. The results are consistent with the involvement of the same cytochrome P-450 isoenzyme(s) in the metabolism of tolbutamide and phenytoin.

Cytochrome P-450 3A4 (nifedipine oxidase) is responsible for the C-oxidative metabolism of 1-nitropyrene in human liver microsomal samples

The nitrated polycyclic aromatic hydrocarbon 1-nitropyrene is a ubiquitous environmental pollutant. The role of cytochromes P-450 in the human metabolism of [3H]-1-nitropyrene was investigated using human liver microsomes. The range of microsomal metabolism from 16 individual liver specimens was 0.13 to 0.99 nmol/min/mg protein. Using 3 microsomal samples exhibiting different maximal velocities, the Km of 1-nitropyrene metabolism was 3.3 +/- 0.5 microM, indicating that perhaps a single or similar cytochromes P-450 was involved in the metabolism of 1-nitropyrene in these samples. The P-450 3A inhibitor triacetyloleandomycin inhibited 86 +/- 8% of the microsomal metabolism of 1-nitropyrene. Further evidence for the role of P-450 3A in human microsomal metabolism of 1-nitropyrene was gained using inhibitory anti-P-450 3A antibodies. Using 3 separate microsomal samples, antibody conditions that inhibited approximately 90% of the metabolism of the P-450 3A4-specific substrate nifedipine inhibited approximately 60-70% of the metabolism of 1-nitropyrene. Human liver microsomes demonstrated a preference for 1-nitropyren-3-ol formation over 1-nitropyren-6-ol or 1-nitropyren-8-ol, which is in contrast to that noted in rodents where the 6-ol and 8-ol are preferentially formed over the 3-ol, yet in agreement with earlier studies on the metabolism of 1-nitropyrene using Vaccinia-expressed human cytochromes P-450. These results indicate that the human hepatic metabolism of 1-nitropyrene is carried out by at least two or more P-450s including those in the P-450 3A subfamily. These studies also suggest that the metabolism of this compound by humans may differ from that in rodents in both the cytochromes that are involved and the specific metabolites that are formed.

The formation of proximate carcinogens from three polycyclic aromatic compounds by human liver microsomes

1. The metabolism of 3H-benzo[a]pyrene (BP), 3H-7-methylbenz[c]acridine (7MBAC) and 3H-dibenz[a,j]acridine (DBAJAC) have been studied in human liver microsomes from 13 subjects. 2. When the metabolism of these carcinogens to more polar ethyl acetate-soluble metabolites were compared, the activities towards the nitrogenous carcinogens were twice that determined for BP. 3. The specific rates of formation of the three proximate carcinogens, BP-7,8-dihydrodiol, 7MBAC-3,4-dihydrodiol and DBAJAC-3,4-dihydrodiol per nmol cytochrome P-450 for 12 subjects were positively correlated. 4. These dihydrodiols constituted 5.9 +/- 0.7% (mean +/- SEM), 57.8 +/- 2.6% and 3.0 +/- 0.4% of the total metabolites identified by cochromatography with standards, 7MBAC, DBAJAC and BP respectively.

Caffeine as a probe for human cytochromes P450: validation using cDNA-expression, immunoinhibition and microsomal kinetic and inhibitor techniques

The molecular basis for the use of caffeine (CA; 1,3,7-trimethylxanthine) as a probe for specific human cytochromes P450 has been investigated. The CA 1-, 3- and 7-demethylations (to form theobromine, paraxanthine and theophylline, respectively) all followed biphasic kinetics in human liver microsomes. Mean apparent Km values for the high- and low-affinity components of the demethylations ranged from 0.13-0.31 nM and 19.2-30.0 mM, respectively. cDNA-expressed CYP1A2 catalysed all three CA demethylations, and the apparent Km for CA 3-demethylation (the major metabolic pathway in humans) by the expressed enzyme was similar to the Km for the high-affinity liver microsomal CA 3-demethylase. IC50 values for inhibition of the CA demethylations by alpha-naphthoflavone were similar for both expressed CYP1A2 and the high-affinity microsomal demethylases. Moreover, CA was a competitive inhibitor of expressed CYP1A2 catalysed phenacetin O-deethylation, with the apparent Ki (0.080 mM) closely matching the apparent Km (0.082 mM) for CA 3-demethylation by the expressed enzyme. Expressed CYP1A1 was additionally shown to catalyse the 3-demethylation of CA, although activity was lower than that observed for CYP1A2. While these data indicate that CYP1A2 is responsible for the high-affinity component of human liver CA 3-demethylation, two limitations associated with the use of CA as an in vitro probe for CYP1A2 activity have been identified: (i) CA 3-demethylation reflects hepatic CYP1A2 activity only at appropriately low substrate concentrations; and (ii) CA is a non-specific CYP1A substrate and CYP1A1 may therefore contribute to CA 3-demethylase activity in tissues in which it is expressed. An anti-CYP3A antibody essentially abolished the 8-hydroxylation of CA to form trimethyluric acid, suggesting formation of this metabolite may potentially serve as a marker of CYP3A isozyme(s) activity.

NADPH-supported and arachidonic acid-supported metabolism of the enantiomers of trans-7,8-dihydrobenzo[a]pyrene-7,8-diol by human liver microsomal samples

Using a new sensitive reverse-phase HPLC assay relying on UV detection at 344 nm, the capacity of 18 human liver microsomal samples to support NADPH-dependent, cytochrome P450-mediated oxidation and arachidonic acid-dependent oxidation of the enantiomers of trans-7,8-dihydroxy-7,8-dihydrobenzo[a]pyrene (B[a]P-7,8-DHD) was determined. The (-)-7R,8R-enantiomer, the preferred substrate of cytochrome P450, formed 94% diolepoxide 2 (anti-isomer; 7R,8S-dihydroxy-9S,10R-epoxy-7,8,9,10-tetrahydrobenzo[a]-pyrene) measured as derived alcohols, and the (+)-7S,8S-enantiomer formed 67% diolepoxide 1 (syn-isomer; 7S,8R-dihydroxy-9S,10R-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene). Arachidonic acid-supported oxidations gave approximately 70% diolepoxide 2 from each enantiomer. The involvement of different sets of cytochrome P450 isozymes was supported by incubations in the presence of alpha-naphthoflavone (alpha-NF) (50 microM) and correlation studies. In the absence of alpha-NF, a positive correlation was found between the metabolism of the (-)-enantiomer but not the (+)-isomer of B[a]P-7,8,-DHD and the relative content of P450IA2. In the presence of alpha-NF, the P450IIIA3/4 content correlated positively with the metabolism of both the (+)-enantiomer and the (-)-enantiomer. Gestodene (100 microM) inhibited the alpha-NF-stimulated metabolism, confirming the involvement of cytochrome P450IIIA3/4. No difference was found between the extent of arachidonic acid-supported, peroxyl radical-mediated metabolism of the (+)- and (-)-enantiomers of B[a]P-7,8-DHD. The metabolism was almost completely abolished by 2 microM butylatedhydroxyanisole and 100 microM nordihydroguaiaretic acid, confirming the free radical nature of the reaction.

N-and O-acetylation of aromatic and heterocyclic amine carcinogens by human monomorphic and polymorphic acetyltransferases expressed in COS-1 cells

Human monomorphic and polymorphic arylamine acetyltransferases (EC 2.3.1.5) were expressed in monkey kidney COS-1 cells and used to study the N- and O-acetylation of a number of carcinogenic amines and their N-hydroxy metabolites. The monomorphic enzyme N-acetylated the aromatic amines, 2-aminofluorene and 4-aminobiphenyl, and also O-acetylated their N-hydroxy derivatives. None of the food-derived heterocyclic amines (Glu-P-1, PhIP, IQ, MeIQx) were substrates and their N-hydroxy metabolites were poorly O-acetylated by this isozyme. By contrast, the polymorphic acetyltransferase catalyzed the N-acetylation of both aromatic amines, and to a lesser extent, Glu-P-1 and PhIP. However, all six N-hydroxy amine substrates were readily O-acetylated to form DNA-bound adducts by the polymorphic isozyme. These data suggest that, for the heterocyclic amine carcinogens, rapid acetylator individuals will be predisposed to their genotoxicity.

Metabolism of food-derived heterocyclic amines in human and rabbit tissues by P4503A proteins in the presence of flavonoids

The ability of human and rabbit gastrointestinal-tract microsomes to metabolize the heterocyclic amine 2-amino-3,4-dimethylimidazo[4,5-f]quinoline (MeIQ) to a mutagen was determined with the Ames test. When human jejunal and ileal microsomes were used as the metabolic activation source, MeIQ produced 1675 and 388 revertants/mg of microsomal protein, respectively, and this increased to 29,230 and 17,963 revertants/mg of microsomal protein, respectively, in the presence of 100 microM alpha-naphthoflavone. MeIQ in the presence of control rabbit duodenal, jejunal, and ileal microsomes produced 2304 +/- 1018, 988 +/- 386, and 444 +/- 134 (mean +/- SD, four samples) revertants/mg of microsomal protein, respectively. In the presence of alpha-naphthoflavone (100 microM), these activities increased greater than 7-fold. P4503A proteins were detectable on Western blots of microsomes prepared from both human and rabbit small intestine. Further, rifampicin-induced rabbit hepatic-microsomal activation of MeIQ was completely inhibited at low concentrations of alpha-naphthoflavone, but at higher concentrations (i.e., 100 microM) this returned to control levels. Flavone also caused a marked stimulation of MeIQ activation in human and rabbit gastrointestinal-tract microsomes. The aforementioned data suggest that flavonoids markedly increase the ability of P4503A isozymes to activate heterocyclic amines to mutagens in the Ames test.

Localization of CYP1A1 and CYP1A2 messenger RNA in normal human liver and in hepatocellular carcinoma by in situ hybridization

To better characterize the precise cellular distribution of CYP1A gene products in man, we have undertaken Northern-blot and in situ hybridization analyses of CYP1A expression in human liver. Using riboprobes transcribed from both CYP1A1 and CYP1A2 complementary DNAs to probe a series of Northern blots of 23 human liver messenger RNA samples, CYP1A1 expression was demonstrated in 11 samples and CYP1A2 expression was evident in 22 samples. The level of expression of both CYP1A enzymes in these livers demonstrated marked variability. The CYP1A1 and CYP1A2 riboprobes were then used for in situ hybridization localization of CYP1A1/1A2 messenger RNA sequences on paraffin-embedded, formalin-fixed human liver sections. These studies demonstrated that both CYP1A1 and CYP1A2 messenger RNAs are distributed nonuniformly across the human liver acinus, with levels highest in hepatocytes surrounding terminal hepatic venules and intercalated veins. Immunohistochemistry with an anti-rabbit CYP1A1 serum demonstrated a corresponding distribution for the translated CYP1A proteins. In situ hybridization analysis was also performed on sections of hepatocellular carcinoma, demonstrating a significant down-regulation in CYP1A expression. Functional studies using the activation of the food-derived heterocyclic amine MeIQ (2-amino-3,4-dimethylimadazo [4,5-f] quinoline) to a mutagen in the Ames test as an indicator of CYP1A expression confirmed this down-regulation. These results demonstrate heterogeneity of hepatic CYP1A expression both between individuals and in different acinar zones. This variation in expression may be of significance in assessing cell specific toxicities of various drugs and carcinogens.

Tolbutamide and phenytoin hydroxylations by cDNA-expressed human liver cytochrome P4502C9

A human cytochrome P4502C9 cDNA clone has been isolated from a human liver bacteriophage Lambda gt11 library using oligonucleotide probes. Expression of the 1762 base pair cDNA in COS cells demonstrated that the encoded enzyme has a molecular mass of 55 kDa as determined by sodium dodecyl sulphate-polyacrylamide gel electrophoresis. The expressed enzyme catalysed the methylhydroxylation of tolbutamide with an apparent Km of 131.7 microM, similar to that observed in human liver microsomes. P4502C9 also catalysed the 4-hydroylation of phenytoin, and inhibition experiments demonstrated that phenytoin was a competitive inhibitor of tolbutamide hydroxylation with an apparent Ki of 19.1 microM. Sulphaphenazole was a potent inhibitor of the expressed enzyme with respect to both tolbutamide and phenytoin hydroxylations. These data demonstrate that a single isozyme can catalyse the hydroxylations of both tolbutamide and phenytoin, and suggest that both reactions are mediated by the same isozyme(s) of cytochrome P450 in human liver.

Relationship between phenytoin and tolbutamide hydroxylations in human liver microsomes

1. The metabolic interaction of phenytoin and tolbutamide in human liver microsomes was investigated. 2. Phenytoin 4-hydroxylation (mean Km 29.6 microM, n = 3) was competitively inhibited by tolbutamide (mean Ki 106.2 microM, n = 3) and tolbutamide methylhydroxylation (mean Km 85.6 microM, n = 3) was competitively inhibited by phenytoin (mean Ki 22.6 microM, n = 3). 3. A significant correlation was obtained between phenytoin and tolbutamide hydroxylations in microsomes from 18 human livers (rs = 0.82, P less than 0.001). 4. Sulphaphenazole was a potent inhibitor of both phenytoin and tolbutamide hydroxylations with IC50 values of 0.4 microM and 0.6 microM, respectively. 5. Mephenytoin was a poor inhibitor of both phenytoin and tolbutamide hydroxylations with IC50 values greater than 400 microM for both reactions. 6. Anti-rabbit P450IIC3 IgG inhibited both phenytoin and tolbutamide hydroxylations in human liver microsomes by 62 and 68%, respectively. 7. These in vitro studies are consistent with phenytoin 4-hydroxylation and tolbutamide methylhydroxylation being catalysed by the same cytochrome P450 isozyme(s) in human liver microsomes.

Metabolism of 2-acetylaminofluorene and benzo(a)pyrene and activation of food-derived heterocyclic amine mutagens by human cytochromes P-450

The human P-450 CYP1A1 gene and a P450IA2 complementary DNA have been expressed in Cos-1 cells and the expressed proteins were assayed for their capacity to metabolize the carcinogens 2-acetylaminofluorene (AAF), benzo(a)pyrene, 2-amino-3-methylimidazo[4,5-f]quinoline (IQ), 2-amino-3,4-dimethylimidazo[4,5-f]quinoline, 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline, 2-amino-3,4,8-trimethylimidazo[4,5-f]quinoxaline, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) was determined. The expressed human P450IA1 and P450IA2 proteins, when run on a 7.5% sodium dodecyl sulfate-polyacrylamide gel, migrated with different mobilities, with the former displaying the lower molecular weight. In human liver microsomes from 18 subjects, only a protein band corresponding to P450IA2 was detectable. Cos-1 cell-expressed P450IA1 and P450IA2 were capable of N-hydroxylating AAF and these activities were inhibited by alpha-naphthoflavone. In human liver microsomes, a correlation of r = 0.76 (P less than 0.05; n = 18) was obtained between AAF N-hydroxylase activity and P450IA2 content. AAF N-hydroxylase activity of human liver microsomes was also strongly inhibited by alpha-naphthoflavone. Except in the case of PhIP, where both proteins exhibited similar activities, P450IA2 was at least an order of magnitude more efficient than P450IA1 in activating IQ, 2-amino-3,4-dimethylimidazo[4,5-f]quinoline, 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline, and 2-amino-3,4,8-trimethylimidazo[4,5-f]quinoxaline to mutagens as measured in the Ames test. Statistically significant correlations were obtained between IQ activation and P450IA2 content (r = 0.75, r2 = 0.56) and PhIP activation and P450IA2 content (r = 0.71, r2 = 0.5) in human liver microsomes. The activation of both IQ and PhIP by expressed proteins and human liver microsomes was strongly inhibited by alpha-naphthoflavone. The above data suggest a major role for P450IA2 in activation (N-hydroxylation) of aromatic amides and amines in human liver. When benzo(a)pyrene hydroxylase activity was determined, only Cos-1 cell-expressed P450IA1 exhibited appreciable activity. While alpha-naphthoflavone inhibited Cos-1 cell-expressed P450IA1 benzo(a)pyrene hydroxylase activity, it caused a marked stimulation of this activity in human liver microsomes, which lack P450IA1 protein. The lack of a role for P450IA proteins in benzo(a)pyrene metabolism is further supported by the poor correlation (r = 0.43, P greater than 0.05) between this activity and P450IA2 content of human liver microsomes. However, when P450IIIA3 content of the above human liver microsomes was determined by using the Western blot technique and correlated with benzo(a)pyrene metabolism, an r value of 0.70 (P less than 0.5) was obtained. These data suggest that human P450IIIA proteins are involved in benzo(a)pyrene metabolism.

Metabolism of 2-amino-1-methyl-6-phenylimidazo[4,5-b] pyridine (PhIP) by liver microsomes and isolated rabbit cytochrome P450 isozymes

The cytochrome P450-dependent metabolism of the heterocyclic amine mutagen 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) has been determined. We investigated the in vitro metabolism of PhIP by polycyclic hydrocarbon-induced mouse and rabbit liver microsomes, and by purified rabbit liver P450 isozymes. Following a 60 min incubation, 3-methylcholanthrene-induced mouse microsomes converted 36% of the PhIP to two major metabolites, N-hydroxy-PhIP and 4'-hydroxy-PhIP, with 43% total metabolism. Rabbit P450 form 6 and form 4 produced the same two major metabolites (20 and 5% total metabolism respectively). Additional metabolites were produced in low yields and amounts varied depending on the isozyme used (1-5%). Metabolites were not detected in incubations of PhIP with P450 forms 2 and 3C. N-Hydroxy-PhIP was found to be directly mutagenic to Salmonella TA98, while the 4'-hydroxy-PhIP was not mutagenic either with or without additional metabolic activation. These data suggest that the cytochrome P450IA isozymes are involved in the metabolism of PhIP by rabbit liver and that formation of N-hydroxy-PhIP is involved in the mutagenicity of PhIP.

Tolbutamide hydroxylation by human, rabbit and rat liver microsomes and by purified forms of cytochrome P-450

Tolbutamide hydroxylation has been investigated in human, rabbit and rat liver microsomes and by six purified forms of hepatic rabbit cytochromes P-450. These studies were carried out to investigate whether an appropriate animal model could be developed for the human cytochrome(s) P-450 metabolizing tolbutamide. Selective induction was used in rats and rabbits to indicate the isozymes primarily responsible for tolbutamide hydroxylation in these species. Microsomal tolbutamide hydroxylase activity was significantly induced only by phenobarbital pretreatment in the rat which induces P-450 forms b (P-450IIB1) and/or e (P-450IIB2). Only pretreatment of rabbits with rifampicin, which induces cytochrome P-450 form 3c (P-450IIIA6), significantly increased the microsomal hydroxylation of tolbutamide. However, the increase in tolbutamide hydroxylase activity in rifampicin-induced microsomes (congruent to 50%) appears low compared to known levels of induction of P-450IIIA6 following rifampicin pretreatment (5-10-fold). These data suggest that P-450IIIA6 is at least partially involved in tolbutamide hydroxylation in rabbit liver but that other form(s) may be relatively more important. Reconstitution experiments with six purified forms of rabbit cytochrome P-450 indicated that the highest activity occurred with P-450IIIA6 (form 3c). As isozymes from different gene families or subfamilies appeared to metabolize tolbutamide in the three species studied, catalytic similarities between the P-450s with respect to inhibition was further investigated in microsomes using sulfaphenazole, alpha-naphthoflavone and mephenytoin. These studies showed that the catalytic characteristics in relation to inhibition differ markedly between species. Hence, it appears that the animal model approach is not likely to be successful in the identification and characterization of the cytochrome P-450 form(s) metabolizing tolbutamide in humans.

Phenytoin 4-hydroxylation by rabbit liver P450IIC3 and identification of orthologs in human liver microsomes

The ability of rabbit liver microsomes to 4-hydroxylate phenytoin to 5-(4-hydroxyphenyl)-5-phenylhydantoin was studied. No significant difference was observed between the capacity of control and rifampicin, phenobarbital, acetone, 2,3,7,8-tetrachlorodibenzo-p-dioxin and phenytoin induced rabbit liver microsomes to 4-hydroxylate phenytoin. In reconstitution experiments using six purified rabbit cytochromes P450 isozymes, only P450IIC3 was capable of 4-hydroxylating phenytoin whereas P450IA1, P450IA2, P450IIB4, P450IIIA6, and P450IVB1 were inactive. Further, anti-P450IIC3 IgG completely inhibited phenytoin 4-hydroxylase activity in rabbit liver microsomes. The above data suggest a major role for the constitutive isozyme P450IIC3 in phenytoin 4-hydroxylase activity in rabbit liver. In human liver microsomes P450IIC3 IgG inhibited phenytoin 4-hydroxylase activity by 66%, suggesting that an ortholog to rabbit P450IIC3 is in part responsible for this activity in man.

Structural and functional analysis of NADPH-cytochrome P-450 reductase from human liver: complete sequence of human enzyme and NADPH-binding sites

The complete amino acid sequence of human liver NADPH-cytochrome P-450 reductase has been determined by microsequence analysis and mass spectrometry. The total sequence consists of 676 amino acids initiated by an amino-terminal acetyl group. There is no evidence for posttranslational modifications, including Asn-linked glycosylation. The human enzyme exhibits sequence homology in the range of 92-95% with other mammalian enzymes. Sequence differences were mainly confined to several hydrophilic regions in the NH2-terminal and COOH-terminal domains. Since the human enzyme is immunochemically distinct from the rabbit enzyme despite similar enzymatic properties, it is likely that these variable hydrophilic regions are potential antigenic determinants. The NADPH-depleted enzyme is inactivated by either fluorescein isothiocyanate, a lysine-specific reagent, or 5-(iodoacetamido)fluorescein, a cysteine-specific reagent. In both cases, protection by NADP(H) prevents enzyme inactivation by the reagents. Isolation of fluorescent peptide from 5-(iodoacetamido)fluorescein-inactivated enzyme identified Cys 565 as the specifically NADPH-protected residue.

The role of cytochromes P-450 and N-acetyl transferase in the carcinogenicity of aromatic amines and amides

1. Epidemiological studies have clearly identified the aromatic amines 2-naphthylamine, benzidine and 4-aminobiphenyl as human carcinogens. Many other compounds of this class are suspected to be human carcinogens. 2. At least 12 heterocyclic aromatic amines have been isolated from cooked food, nine of which have been shown to be carcinogenic in vivo. The significance of these compounds in the aetiology of human cancer is yet to be established. 3. Specific forms of cytochrome P-450 have been shown to be involved in the initial step in the activation (N-hydroxylation) of aromatic amines and amides. A possible polymorphism exists for the N-hydroxylation of 2-acetylaminofluorene in human liver microsomes. 4. N-Acetyltransferase is involved in the further activation of arylhydroxamic acids and N-hydroxyarylamines to reactive intermediates. Polymorphisms in the expression of this enzyme activity have been associated with certain types of human cancer. 5. Polymorphisms in both cytochrome P-450 and N-acetyltransferase expression may be critical determinants in the susceptibility of individuals to the toxicity and carcinogenicity of aromatic amines and amides.

Immunohistochemical localization of NADPH-cytochrome P450 reductase in human tissues

In an attempt to elucidate the mechanism(s) behind the susceptibility of tissues to the toxic effects of chemicals, NADPH-cytochrome P450 reductase IgG has been used to map the distribution and localization of the cytochrome P450 system in hepatic and extrahepatic human tissues. Employing the Western blot procedure this antibody recognized a single band in human liver microsomes which corresponded in mol. wt to the purified reductase. Immunoreactive NADPH-cytochrome P450 reductase staining was detected in all zones of the liver acinus, with maximal staining in hepatocytes adjacent to the terminal hepatic venules (zone 3). Considerable variation in the intensity of reductase staining was observed in different segments of the gastrointestinal tract. Staining was most intense in the enterocytes of the small intestine, with maximal staining at the tips of the villi. Colonic epithelial cells were variably positive while the rectum was negative. Pancreatic ductal cells were positive whereas exocrine cells were negative. In the lung, reductase was detected in bronchiolar and bronchial epithelial cells, Clara cells and alveolar lining cells. In the kidney, the proximal and distal convoluted tubules, the loops of Henle, the collecting ducts in the medulla and the transitional epithelium all stained positively for reductase. The results demonstrate specific cellular localization of NADPH-cytochrome P450 reductase, and hence the cytochrome P450 system, in human tissues. The differential distribution of the reductase within human tissues may lead to a better understanding of the mechanism underlining site-specific carcinogenesis.