Optimizing bacterial expression of catalytically active human cytochromes P450: comparison of CYP2C8 and CYP2C9.

1. Methods for the co-expression in Escherichia coli of human cytochrome P450 (CYP) 2C8 and CYP2C9 with NADPH-cytochrome P450 reductase (OxR) to produce a catalytically active system were compared. 2. Approaches assessed were expression of a CYP:OxR fusion construct, bicistronic plasmids, simultaneous transformation with CYP and OxR plasmids, and separate expression of CYP and OxR with reconstitution of activity by mixing the bacterial membranes. Two N-terminal modifications (Delta3-20 and 17alpha-leader) of the individual P450s were additionally investigated. 3. Each approach gave efficient expression of CYP2C8 and CYP2C9, but the bicistronic constructs under the expression conditions used gave low OxR expression and low catalytic activity. CYP expression was higher with the Delta3-20 construct for CYP2C9 and with the 17alpha-presequence construct for CYP2C8. 4. Using torsemide as substrate, all methods gave catalytically active systems with K(m) values similar to human liver microsomes. Mixing bacterial membranes containing separately expressed CYP and OxR reconstituted a catalytically active system with the Delta3-20 construct for CYP2C9 but not for CYP2C8, and with neither of the 17alpha- presequence constructs. OxR co-expressed with CYP in the same membrane interacted with CYP to reconstitute activity more effectively than addition of exogenous OxR membranes. 5. Expression construct and OxR co-expression strategy should be individualized for CYP isoforms.

Polymorphic hydroxylation of perhexiline in vitro.

AIMS: The aims of this study were to examine the in vitro enzyme kinetics and CYP isoform selectivity of perhexiline monohydroxylation using human liver microsomes. METHODS: Conversion of rac-perhexiline to monohydroxyperhexiline by human liver microsomes was assessed using a high-performance liquid chromatography assay with precolumn derivatization to measure the formation rate of the product. Isoform selective inhibitors were used to define the CYP isoform profile of perhexiline monohydroxylation. RESULTS: The rate of perhexiline monohydroxylation with microsomes from 20 livers varied 50-fold. The activity in 18 phenotypic perhexiline extensive metabolizer (PEM) livers varied about five-fold. The apparent Km was 3.3 +/- 1.5 micro m, the Vmax was 9.1 +/- 3.1 pmol min-1 mg-1 microsomal protein and the in vitro intrinsic clearance (Vmax/Km) was 2.9 +/- 0.5 micro l min-1 mg-1 microsomal protein in the extensive metabolizer livers. The corresponding values in the poor metabolizer livers were: apparent Km 124 +/- 141 micro m; Vmax 1.4 +/- 0.6 pmol min-1 mg-1 microsomal protein; and intrinsic clearance 0.026 micro l min-1 mg-1 microsomal protein. Quinidine almost completely inhibited perhexiline monohydroxylation activity, but inhibitors selective for other CYP isoforms had little effect. CONCLUSIONS: Perhexiline monohydroxylation is almost exclusively catalysed by CYP2D6 with activities being about 100-fold lower in CYP2D6 poor metabolizers than in extensive metabolizers. The in vitro data predict the in vivo saturable metabolism and pharmacogenetics of perhexiline.

A cost-effectiveness approach to drug subsidy and pricing in Australia.

The Australian government offers its citizens subsidies on a select list of pharmaceuticals. For a drug to qualify for inclusion on this list, its manufacturer must demonstrate that the drug is both clinically effective and cost-effective. In part, this measure, along with others, was introduced to improve clinical and economic outcomes. Although this evidence-based system has provided transparency and consistency in decision making about which drugs will be covered, it may not have contained the rate of increase in drug costs.

Impact of the Minimum Pricing Policy and introduction of brand (generic) substitution into the Pharmaceutical Benefits Scheme in Australia.

PURPOSE: To describe the effects of introducing the Minimum Pricing Policy (MPP) and generic (brand) substitution in 1990 and 1994 respectively on the dispensing of Pharmaceutical Benefits Scheme (PBS) prescriptions both at the aggregate and individual patient level. METHODS: The relative proportion of prescriptions with a brand premium and those at benchmark was examined 4 years after introduction of the MPP and again 5 years later after generic substitution by pharmacists was permitted. To determine the impact of a price signal at the individual level, case studies involving a patient tracking methodology were conducted on two drugs (fluoxetine and ranitidine) that received a brand premium. RESULTS: From a zero base when the MPP was introduced in 1990, there were 5.4 million prescriptions (17%) dispensed for benchmark products 4 years later in 1994. At this stage generic (brand) substitution by pharmacists was then permitted and the market share of benchmark brands increased to 45% (25.2 million) by 1999. In the patient tracking studies, a significantly lower proportion of patients was still taking the premium brand of fluoxetine 3 months after the introduction of a price signal compared with patients taking paroxetine which did not have a generic competitor. This was also the case for the premium brand of ranitidine when compared to famotidine. The size of the price signal also had a marked effect on dispensing behaviour with the drug with the larger premium (fluoxetine) showing a significantly greater switch away from the premium brand to the benchmark product. CONCLUSIONS: The introduction in 1990 of the Minimum Pricing Policy without allowing generic substitution had a relatively small impact on the selection of medicines within the Pharmaceutical Benefits Scheme. However the effect of generic substitution at the pharmacist level, which was introduced in December 1994, resulted in a marked increase in the percentage of eligible PBS items dispensed at benchmark. Case studies showed a larger premium resulted in a greater shift of patients from drugs with a brand premium to the benchmark alternative.

Nonspecific binding of drugs to human liver microsomes.

AIMS: To characterize the nonspecific binding to human liver microsomes of drugs with varying physicochemical characteristics, and to develop a model for the effect of nonspecific binding on the in vitro kinetics of drug metabolism enzymes. METHODS: The extent of nonspecific binding to human liver microsomes of the acidic drugs caffeine, naproxen, tolbutamide and phenytoin, and of the basic drugs amiodarone, amitriptyline and nortriptyline was investigated. These drugs were chosen for study on the basis of their lipophilicity, charge, and extent of ionization at pH 7.4. The fraction of drug unbound in the microsomal mixture, fu(mic), was determined by equilibrium dialysis against 0.1 M phosphate buffer, pH 7.4. The data were fitted to a standard saturable binding model defined by the binding affinity KD, and the maximum binding capacity Bmax. The derived binding parameters, KD and Bmax, were used to simulate the effects of saturable nonspecific binding on in vitro enzyme kinetics. RESULTS: The acidic drugs caffeine, tolbutamide and naproxen did not bind appreciably to the microsomal membrane. Phenytoin, a lipophilic weak acid which is mainly unionized at pH 7. 4, was bound to a small extent (fu(mic) = 0.88) and the binding did not depend on drug concentration over the range used. The three weak bases amiodarone, amitriptyline and nortriptyline all bound extensively to the microsomal membrane. The binding was saturable for nortriptyline and amitriptyline. Bmax and KD values for nortriptyline at 1 mg ml-1 microsomal protein were 382 +/- 54 microM and 147 +/- 44 microM, respectively, and for amitriptyline were 375 +/- 23 microM and 178 +/- 33 microM, respectively. Bmax, but not KD, varied approximately proportionately with the microsome concentration. When KD is much less than the Km for a reaction, the apparent Km based on total drug can be corrected by multiplying by fu(mic). When the substrate concentration used in a kinetic study is similar to or greater than the KD (Km >/= KD), simulations predict complex effects on the reaction kinetics. When expressed in terms of total drug concentrations, sigmoidal reaction velocity vs substrate concentration plots and curved Eadie Hofstee plots are predicted. CONCLUSIONS: Nonspecific drug binding in microsomal incubation mixtures can be qualitatively predicted from the physicochemical characteristics of the drug substrate. The binding of lipophilic weak bases is saturable and can be described by a standard binding model. If the substrate concentrations used for in vitro kinetic studies are in the saturable binding range, complex effects are predicted on the reaction kinetics when expressed in terms of total (added) drug concentration. Sigmoidal reaction curves result which are similar to the Hill plots seen with cooperative substrate binding.

The xenobiotic inhibitor profile of cytochrome P4502C8.

AIMS: To investigate inhibition of recombinant CYP2C8 by: (i) prototypic CYP isoform selective inhibitors (ii) imidazole/triazole antifungal agents (known inhibitors of CYP), and (iii) certain CYP3A substrates (given the apparent overlapping substrate specificity of CYP2C8 and CYP3A). METHODS: CYP2C8 and NADPH-cytochrome P450 oxidoreductase were coexpressed in Spodoptera frugiperda (Sf21) cells using the baculovirus expression system. CYP isoform selective inhibitors, imidazole/triazole antifungal agents and CYP3A substrates were screened for their inhibitory effects on CYP2C8-catalysed torsemide tolylmethylhydroxylation and, where appropriate, the kinetics of inhibition were characterized. The conversion of torsemide to its tolylmethylhydroxy metabolite was measured using an h.p.l.c. procedure. RESULTS: At concentrations of the CYP inhibitor 'probes' employed for isoform selectivity, only diethyldithiocarbamate and ketoconazole inhibited CYP2C8 by > 10%. Ketoconazole, at an added concentration of 10 microM, inhibited CYP2C8 by 89%. Another imidazole, clotrimazole, also potently inhibited CYP2C8. Ketoconazole and clotrimazole were both noncompetitive inhibitors of CYP2C8 with apparent Ki values of 2.5 microM. The CYP3A substrates amitriptyline, quinine, terfenadine and triazolam caused near complete inhibition (82-91% of control activity) of CYP2C8 at concentrations five-fold higher than the known CYP3A Km. Kinetic studies with selected CYP3A substrates demonstrated that most inhibited CYP2C8 noncompetitively. Apparent Ki values for midazolam, quinine, terfenadine and triazolam ranged from 5 to 25 microM. CONCLUSIONS: Inhibition of CYP2C8 occurred at concentrations of ketoconazole and diethyldithiocarbamate normally employed for selective inhibition of CYP3A and CYP2E1, respectively. Some CYP3A substrates have the capacity to inhibit CYP2C8 activity and this may have implications for inhibitory drug interactions in vivo.

Compliance with restrictions on the subsidized use of proton pump inhibitors in Australia.

AIMS: To determine among a cohort of patients newly dispensed a prescription for a proton pump inhibitor (PPI) the extent of prior use of other less expensive agents such as antacids and H2-receptor antagonists as evidence of a 'stepped care' approach to peptic ulcer and oesophageal disease. METHODS: A retrospective drug utilization study was conducted within the Pharmaceutical Benefits Scheme (PBS) claims database in Australia. A cohort of social security recipients, who received approval for PPI supply in the month of October 1996, had no prior PPI approval in the previous 18 months and went on to have the drug dispensed, was assembled. This group of 'new PPI starters' was then examined for supply of less expensive prescription medicines to treat peptic ulcer and oesophageal disease in the 12 months prior to obtaining their PPI approval. RESULTS: In a cohort of 4554 defined new PPI users, 1205 (26.5%) showed no use of H2-receptor antagonists, antacids, cisapride, cytoprotectants or antiregurgitants in the 12 month period prior to commencing the PPI. The major reason for use given by prescribers for PBS supply was 'severe refractory ulcerating oesophagitis'. CONCLUSIONS: Subsidized supply is currently restricted on cost-effectiveness grounds to refractory peptic ulcer disease or severe oesophageal disease. Despite this, utilization and epidemiological data suggest that there is widespread leakage of use outside these indications particularly to less severe forms of oesophageal disease. This patient tracking study has shown within the PBS database that around a quarter of the patients are treated directly with a PPI without being prescribed less expensive agents at least in the preceding 12 months.

Cytochrome P4502C9: an enzyme of major importance in human drug metabolism.

Accumulating evidence indicates that CYP2C9 ranks amongst the most important drug metabolizing enzymes in humans. Substrates for CYP2C9 include fluoxetine, losartan, phenytoin, tolbutamide, torsemide, S-warfarin, and numerous NSAIDs. CYP2C9 activity in vivo is inducible by rifampicin. Evidence suggests that CYP2C9 substrates may also be induced variably by carbamazepine, ethanol and phenobarbitone. Apart from the mutual competitive inhibition which may occur between alternate substrates, numerous other drugs have been shown to inhibit CYP2C9 activity in vivo and/or in vitro. Clinically significant inhibition may occur with coadministration of amiodarone, fluconazole, phenylbutazone, sulphinpyrazone, sulphaphenazole and certain other sulphonamides. Polymorphisms in the coding region of the CYP2C9 gene produce variants at amino acid residues 144 (Arg144Cys) and 359 (Ile359Leu) of the CYP2C9 protein. Individuals homozygous for Leu359 have markedly diminished metabolic capacities for most CYP2C9 substrates, although the frequency of this allele is relatively low. Consistent with the modulation of enzyme activity by genetic and other factors, wide interindividual variability occurs in the elimination and/or dosage requirements of prototypic CYP2C9 substrates. Individualisation of dose is essential for those CYP2C9 substrates with a narrow therapeutic index.

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.

Inhibition of human hepatic cytochrome P4502E1 by azole antifungals, CNS-active drugs and non-steroidal anti-inflammatory agents.

1. The capacity of a number of antifungal azoles, CNS-active drugs (anticonvulsants, antidepressants, antipsychotics and benzodiazepine hypnosedative-anxiolytics) and nonsteroidal anti-inflammatory agents (NSAIDs) to inhibit human liver microsomal 4-nitrophenol (4NP) hydroxylation, a marker of CYP2E1 activity, was investigated. 2. The imidazoles bifonazole, clotrimazole, econazole and miconazole were un- or non-competitive inhibitors of 4NP hydroxylation, with apparent Ki values ranging from 4 to 25 microM. Fluonazole, itraconazole and ketoconazole caused minor or negligible inhibition. 3. Of the CNS-active drugs screened, significant inhibition occurred only with tricyclic antidepressants, phenothiazine antipsychotics and two benzodiazepines (flurazepam and medazepam). Un- or non-competitive inhibition was similarly observed for the tricyclic antidepressants, phenothiazines, flurazepam and medazepam, with apparent Ki values ranging from 175 to 1000 microM. 4. Diclofenac and flufenamic acid were the only NSAIDs found to inhibit 4NP hydroxylation substantially; kinetic analysis was suggestive of activation-inhibition phenomena. 5. These data indicate that, although not substrates for CYP2E1, some clinically used drugs have the capacity to inhibit this enzyme and hence have the potential to modulate the toxicity of non-drug xenobiotics metabolized by CYP2E1.

Baculovirus-mediated expression of cytochrome P4502C8 and human NADPH-cytochrome P450 reductase: optimization of protein expression.

1. High expression levels of cytochrome P450 (CYP) 2C8 and NADPH-cytochrome P450 oxidoreductase (OxR) in Spodoptera frugiperda (Sf21) cells have been achieved using the baculovirus expression system. 2. The baculovirus dual expression plasmid, pAcUW31, was used to insert CYP2C8 and OxR cDNAs downstream of the polyhedrin (polh) or p10 promoters, either separately or together, generating four recombinant baculoviruses; two expressing single proteins (CYP2C8 driven by the p10 promoter, bVp10.2C8 or OxR driven by the polh promoter, bVpolh.OxR) with another two coexpressing both CYP2C8 and OxR under reciprocal control of the polh and p10 promoters (bVpolh.OxR-p10.2C8 and bVpolh.2C8-p10.OxR). 3. High levels of singly expressed CYP2C8 and OxR were achieved from bVp10.2C8 and bVpolh.OxR, with levels of 0.7-1.2 nmol CYP/mg protein and 400-500 nmol cytochrome c reduced/min/mg protein respectively. 4. The two dual gene clones (bVpolh.OxR-p10.2C8 and bVpolh.2C8-p10.OxR) showed, in general, greater variation in CYP content and OxR activity than single gene clones. Screening was therefore necessary for the selection of dual gene clones expressing both proteins optimally. 5. Sf21 microsomes infected by selected dual gene clones were, on average, 14 times more active in tolbutamide hydroxylase activity than those expressing CYP2C8 alone, with a mean spectral CYP content of 79 pmol/mg cell lysate protein and a mean OxR level of 600 nmol/min/mg cell lysate protein.

Evidence for involvement of human CYP3A in the 3-hydroxylation of quinine.

AIMS: Our previous studies using in vitro hepatic microsomal preparations suggested that the hepatic metabolism of quinine to form the major metabolite 3-hydroxyquinine is most likely catalysed by human P450 3A (CYP3A). The present study was carried out to investigate the kinetics and to identify and further characterise the human liver CYP isoforms involved in the metabolism of quinine. METHODS: In vitro human microsomal techniques were employed. RESULTS: The mean apparent Km value for 3-hydroxyquinine formation was 83 +/- 19 (s.d.) microM, ranging from 57 microM to 123 microM in microsomes from ten human livers. There was a 6.7-fold variation in Vmax values (mean 547 +/- 416 pmol min-1 mg-1). Quinine 3-hydroxylation was inhibited by the specific CYP3A inhibitors, troleandomycin, midazolam and erythromycin. Inhibitors selective for CYP1A1/2, CYP2D6, CYP2E1, CYP2C9/10 or CYP2C19 had little or no effect on quinine 3-hydroxylation. Using microsomes from a panel of livers, significant correlations were found only between 3-hydroxyquinine activity and other CYP3A activities (caffeine 8-oxidation, omeprazole sulphoxidation, midazolam 1'-hydroxylation and midazolam 4-hydroxylation) and immunoreactive CYP3A content. There were no statistically significant correlation with activities selective for CYP1A2, CYP2C9 and CYP2E1. Competitive inhibition of quinine 3-hydroxylation was observed with a substrate known to be specifically metabolized by human CYP3A, i.e. midazolam, with an apparent Ki value of 11.0 microM. CONCLUSIONS: The present results strongly indicate that the conversion of quinine to 3-hydroxyquinine is the major metabolic pathway in human liver in vitro and that the reaction is catalysed by CYP3A isoforms.

1-Methylxanthine derived from caffeine as a pharmacodynamic probe of oxypurinol effect.

AIMS: In the present study we have investigated the use of caffeine, administered in the form of instant coffee, as a prodrug for 1MX to validate the use of the 1MU:1MX ratio following caffeine administration as a pharmacodynamic measure of oxypurinol effect on xanthine oxidase. METHODS: Five healthy volunteers took caffeine 75 mg 8 hourly administered as instant coffee over a 7 day period. They were given allopurinol 600 mg on day 4. Urine was collected in 8 h aliquots from day 1-day 7. The ratio of 1-methyluric acid (1MU) to 1-methylxanthuric (1MX) was determined. RESULTS: The relationship between the plasma oxypurinol (the active metabolite of allopurinol) concentration at the midpoint of each caffeine dosage interval and the decrement in the urinary 1MX to 1MU ratio fitted well by a sigmoid Emax model. Mean (+/-s.d.) values of the oxypurinol EC50(3.9 +/- 1.4 mg l-1), EC90(8.7 +/- 1.8 mgl-1) and the exponent, n (3.0 +/- 1.2) were similar to those obtained previously following either the direct administration of 1MX or the use of theophylline as a prodrug for 1MX. CONCLUSIONS: These data indicate that the use of caffeine as a source of 1MX could provide a simple and ethically acceptable method for monitoring oxypurinol effect in patients taking allopurinol for the treatment of gout.

Allelic and functional variability of cytochrome P4502C9.

Single nucleotide substitutions are known to result in a different amino acid at one of four sites in cytochrome P4502C9 (CYP2C9) namely: residue 144: Arg/Cys; residue 358: Tyr/Cys; residue 359: Ile/Leu and residue 417: Gly/Asp. Polymerase chain reaction (PCR)-based amplification of the nucleotide fragments encompassing the four residues (144, 358-359 and 417) in 18 samples of human genomic DNA from a liver bank and one sample of DNA extracted from the blood of a known poor metabolizer of tolbutamide has been carried out. The products of PCR amplification were analysed by either allele-specific restriction endonucleases or probed with radioactively labelled allele-specific oligonucleotides in dot blot hybridizations. Fourteen individuals were homozygous for Arg144 and four were heterozygous Arg/Cys144. All individuals analysed were homozygous for Tyr358 (n = 17) and for Gly417 (n = 18). With the exception of one heterozygote the other 17 subjects were homozygous for Ile359. The genotype of the known poor metabolizer of tolbutamide was homozygous for Arg144, Leu359 and Gly417. The relative levels of expression of the Cys and Arg144 alleles was studied in the heterozygotes. A relative 5- to 10-fold greater expression of the Cys- over the Arg144 allele was noted in two heterozygotes. There was no apparent correlation of genotype to the hydroxylation of the known CYP2C9 substrates phenytoin, tolbutamide, torasemide and diclofenac. Apparent K(m) values for the cDNA-expressed Arg144/Ile359, Cys144/ Ile359 and Arg144/Leu359 variants towards tolbutamide were 91 microM, 62 microM and 229 microM, respectively. It is likely that functional changes occurring as a result of the Ile359Leu transition are responsible for the tolbutamide poor metabolizer phenotype.

Limited value of the urinary phenytoin metabolic ratio for the assessment of cytochrome P4502C9 activity in vivo.

Relationships between the ratio of p-hydroxyphenytoin (p-HPPH), the major metabolite of phenytoin, to unchanged phenytoin excreted in urine (the urinary metabolic ratio or MR) were compared with a number of indices of the metabolic clearances of phenytoin and tolbutamide published previously for seventeen subjects separately administered these known cytochrome P4502C9 (CYP2C9) substrates. Significant correlations (rs = 0.50-0.60, P < 0.05) were observed between the phenytoin MR, derived from either 0-24 or 24-48 h urine collections, and inverse areas under the plasma unbound concentration-time curves (measured over various time intervals) of phenytoin and with plasma unbound tolbutamide clearance. Significant correlations (rs = 0.59-0.74) were also observed between the phenytoin MRs and metabolic unbound clearances for p-hydroxyphenytoin formation. Despite the significant correlations, variability in tolbutamide and phenytoin metabolic clearance parameters tended to account for < 50% of the variability in phenytoin MR. Correlations between the renal clearance of phenytoin and the phenytoin MRs suggest that variability in the renal clearance of unchanged drug limits the usefulness of the phenytoin MR for the investigation of factors influencing CYP2C9 activity in vivo.

The role of the CYP2C9-Leu359 allelic variant in the tolbutamide polymorphism.

Tolbutamide undergoes hydroxylation in humans via a cytochrome P450-mediated pathway. The primary P450 isozyme responsible for this metabolism is thought to be CYP2C9. Population studies have indicated the existence of slow metabolizers of tolbutamide (approximately 1 in 500) suggesting a rare polymorphism associated with 2C9. Several allelic variants of 2C9 have been identified; however, the effect of these allelic variations on metabolism in vivo is not established. In the present study, the coding regions, intron-exon junctions, and upstream region of CYP2C9 were amplified by PCR and sequenced in two slow metabolizers. One individual was homozygous for Leu359/Leu359 and the other individual was heterozygous for Arg144/Cys144 and for Ile359/Leu359. No other genetic variations in 2C9 were detected in these individuals. PCR-RFLP tests showed that Arg144 Tyr358 Ile359 Gly417 is the principle CYP2C9 allele. Frequencies of the rarer Leu359 and Cys144 alleles were 0.06 and 0.08, respectively, in a Caucasian-American population and 0.005 and 0.01 respectively in African-Americans. The frequency of the Leu359 allele was 0.026 in Chinese-Taiwanese, but the Cys144 allele was not detected in this population. Studies in a recombinant yeast expression system showed that the Leu359 variant had the highest Km and the lowest Vmac for hydroxylation of tolbutamide of all the CYP2C9 allelic variants. This allelic variant also had the highest Km for the 7-hydroxylation of S-warfarin. The present data suggest that the incidence of the Leu359 allelic variant of CYP2C9 may account for the occurrence of poor metabolizers of tolbutamide.