Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The use of tobacco products is associated with a variety of medical disorders including increased risk of cancer and cardiovascular and respiratory diseases (Hecht and Hoffmann, 1988). Nicotine is the primary psychoactive substance in tobacco responsible for establishing and maintaining tobacco dependence (Henningfield et al., 1985). It is metabolized primarily (~70%) to its inactive metabolite cotinine by the genetically variable enzyme CYP2A6 (Nakajima et al., 1996; Messina et al., 1997). CYP2A6 also mediates the activation of several procarcinogens, such as tobacco-related nitrosamine 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone, aflatoxin B1, and hexamethyl-phosphoramide (Crespi et al., 1991; Yamazaki et al., 1992).

In addition to the wild-type CYP2A6*1 gene, a gene deletion CYP2A6*4 and point mutation CYP2A6*2 have been isolated and cloned (Yamano et al., 1990; Oscarson et al., 1999). Some studies have shown that individuals with a CYP2A6 null allele or gene deletion appear to smoke fewer cigarettes (Pianezza et al., 1998; Rao et al., 2000), while others have failed to demonstrate such a correlation (Sabol and Hamer, 1999; Tiihonen et al., 2000). Imitating the gene defect by chemical inhibition of CYP2A6 activity in vivo may decrease smoking and activation of carcinogens and could provide the basis for novel therapeutic approaches to smoking reduction and cessation (Sellers et al., 2000). In fact, our research group has recently demonstrated that 30 mg of oral methoxsalen plus 4 mg of nicotine attenuated nicotine clearance and increased its bioavailability, while decreasing cigarette smoking by 24% (p < 0.01), as compared with placebo plus placebo (Sellers et al., 2000). Methoxsalen, or any ideal CYP2A6 inhibitor used for the treatment of tobacco dependence, should be specific and selective for CYP2A6 to minimize potential adverse drug-drug interactions.

Tranylcypromine (TCP¹) is a monoamine oxidase inhibitor used in the treatment of depression. Methoxsalen is a pigmentation agent used in the treatment of psoriasis cutaneous, T-cell lymphoma, and vitiligo. Both are potent inhibitors of CYP2A6 in vitro (Maenpaa et al., 1993; Draper et al., 1997). Moreover, methoxsalen has been shown to significantly inhibit coumarin and nicotine metabolism in vivo in humans, as discussed (Maenpaa et al., 1994; Kharasch et al., 2000; Sellers et al., 2000). The indolealkaloid, tryptamine, is a trace amine found in the brain and is a structural congener of 5-hydroxytryptamine. Tryptamine is a substrate of monoamine oxidase (Sullivan et al., 1986) and has also proved to be a potent inhibitor of CYP2A6 in our laboratory. The human hepatic cytochrome P450 (P450) enzymes 1A2, 2A6, 2B6, 2C9, 2C19, 2D6, 2E1, and 3A4 are the most important forms of drug-metabolizing P450s in humans (Shimada et al., 1994). The metabolism of many therapeutically important drugs and endogenous compounds is mediated primarily by these enzymes.

The present study has identified the specificity and selectivity of these drugs toward CYP2A6 and the major P450s in cDNA-expressing (human lymphoblast cells) and human liver microsomes.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals and Reagents. Phenacetin, acetaminophen, coumarin, 7-hydroxycoumarin, 7-ethoxy-4-trifluoromethylcoumarin, 7-hydroxy-4-trifluoromethylcoumarin, diclofenac sodium, dextromethorphan hydrobromide, dextrorphan, 4-nitrophenol, 4-nitrocatechol, budipine, S-()-nicotine, S-()-cotinine, methoxsalen, pilocarpine, orphenadrine, sulfaphenazole, ketoconazole, -naphthoflavone, and NADPH were purchased from Sigma Chemical Co. (St. Louis, MO). TCP and trioxsalen were purchased from Aldrich Chemical Co. (Milwaukee, WI). S-()-TCP and R-(+)-TCP were kindly provided by Röhm Pharma (Weiterstadt, Germany). Omeprazole, 5-hydroxyomeprazole, omeprazole sulfone, and 5-methoxy-2-[[(3,4-dimethoxy-5-methyl-2-pyridinyl)methyl]sulfinyl]-1H-benzimid-azole (H168/24) were generously donated by Astra Hässle (Mölndal, Sweden). S-(+)-Mephenytoin and 4'-hydroxydiclofenac sodium were purchased from GENTEST (Woburn, MA). All other chemical reagents used were of the highest commercially available quality.

cDNA-Expressing and Human Liver Microsomes. cDNA-expressing cytochrome P450 microsomes (P450 1A2, 2A6, 2B6, 2C9, 2C19, 2D6, and 3A4) from human lymphoblast cells were purchased from GENTEST. The human livers were kindly provided by Dr. T. Inaba. The characteristics and sources of the K series livers in this study have been previously described (Tyndale et al., 1989). Microsomes from these tissues were prepared and stored according to established techniques (Tyndale et al., 1989). Protein concentrations were determined using bicinchoninic acid protein assay kit (Pierce Chemical CO., Rockford, IL). Cytosolic fractions from the livers of four male Wistar rats were used as a source of aldehyde oxidases for the CYP2A6 nicotine assay.

P450 Index Reaction Assays. All incubations were carried out in 25 mM Tris-HCl buffer (pH 7.4) at 37°C in a shaking water bath with 1 mM NADPH. For each substrate (probe drug), preliminary experiments were performed to determine whether metabolite formation was linear with respect to time, NADPH, and microsomal protein concentrations. The percentage of conversion of all metabolites never exceeded 15% of total substrate added. Incubation conditions varied depending on the characteristics of the probe drug, which are outlined in Table 1. Conditions were the same for both human liver and cDNA-expressing microsomes.

Phenacetin O-Deethylation (CYP1A2). Extraction was essentially that of Venkatakrishnan et al. (1998). 7-Hydroxycoumarin was added as an internal standard, and the mixture was extracted with ethyl acetate. Samples were reconstituted into mobile phase (200 µl) prior to injection into the HPLC [HP Spherisorb ODS2 column; UV = 245 nm; 10 mM sodium acetate buffer and acetonitrile (ACN) (87.5:12.5, v/v, pH 4.5) at 1 ml/min].

Coumarin 7-Hydroxylation (CYP2A6). Extraction was essentially that of Li et al. (1997). 7-Aminocoumarin was added as the internal standard, and the mixture was extracted with ethyl acetate. Samples were reconstituted in 200 µl of 30% methanol prior to HPLC injection [HP ODS-2 column; UV = 320 nm; ACN/water/acetic acid (25:75:0.1, v/v/v) at 1 ml/min].

7-Ethoxy-4-Trifluoromethylcoumarin O-Deethylation (CYP2B6). Trioxsalen was added as internal standard, and the mixture was extracted with ethyl acetate. Samples were reconstituted in mobile phase (200 µl) prior to HPLC analysis [HP Spherisorb ODS2 column; UV = 360 nm; ACN/water/acetic acid (50:50:0.1, v/v/v) at 1 ml/min].

Diclofenac 4'-Hydroxylation (CYP2C9). Coumarin was added as an internal standard, and the mixture was extracted with ethyl acetate. Samples were reconstituted in 20% acetic acid (200 µl) prior to HPLC analysis [HP Spherisorb ODS2 column; UV = 280 nm; ACN/water/acetic acid (40:60:0.25, v/v/v) at 1 ml/min].

Omeprazole Assay (CYP2C19 and CYP3A4). Omeprazole and its metabolites were assayed as per the method of Andersson et al. (1994). H168/24 was used as the internal standard, and the mixture was extracted with dichloromethane. Samples were reconstituted in 200 µl of mobile phase prior to HPLC analysis [Waters Spherisorb C₆ column; UV = 304 nm; 10 mM KH₂PO₄ buffer (pH 7.0)/ ACN (75:25, v/v) at 1 ml/min; Waters Corp., Milford, MA].

Dextromethorphan O-Demethylation (CYP2D6). Extraction was essentially that of Otton et al. (1993). The enzymatic reaction was stopped by the addition of 200 µl of NaHCO₃-Na₂CO₃ buffer, and butorphanol was used as the internal standard. The mixture was extracted with hexane/ether (4:1, v/v), then back-extracted with 200 µl of 0.01 N HCl prior to HPLC analysis [CSC-Spherisorb-phenyl column; UV = 210 nm; 10 mM KH₂PO₄ buffer (pH 3.8)/ACN (73:27, v/v) at 1 ml/min].

4-Nitrophenol 4-Hydroxylation (CYP2E1). The reaction was stopped by the addition of 3 µM HCl (50 µl), and 7-hydroxycoumarin was added as the internal standard. The mixture was extracted with ether (2 ml), and the organic phase was evaporated to dryness and reconstituted into mobile phase (200 µl) prior to HPLC analysis [Spherisorb ODS-2 column; UV = 250 nm; KH₂PO₄ buffer containing 1 mM octanesulfonic acid and 0.5% triethylamine (pH 2.6)/ACN (81.5:18.5, v/v) at 1 ml/min].

Nicotine Metabolism Assay. The assay used was previously described (Messina et al., 1997). S-()-Nicotine (5-1000 µM) was incubated for 30 min with cDNA-expressing microsomes (final concentration, 1.5 mg/ml) in the presence of 1 mM NADPH (final concentration) and 25 µl of rat liver cytosol as the aldehyde oxidase source. Methyl cotinine (50 µl of 2.0 µg/ml) was added as the internal standard, and samples were extracted with 2 ml of dichloromethane, dried under nitrogen, and reconstituted in 0.01 M HCl (200 µl) prior to HPLC analysis [Supelcosil LC-8DB column; UV = 260 nm; KH₂PO₄ buffer containing 1 mM octanesulfonic acid and 0.5% triethylamine (pH 4.6)/ACN (90:10, v/v) at 1 ml/min].

Chemical Inhibition Studies of P450s with Inhibitors of CYP2A6. Initial screening experiments were carried out using two concentrations of inhibitor (20 and 200 µM), and known selective P450 inhibitors were selected as controls according to previously published reports (Bourrie et al., 1996; Eagling et al., 1998; Hichman et al., 1998). They were -naphthoflavone (CYP1A2), pilocarpine (CYP2A6), orphenadrine (CYP2B6), sulfaphenazole (CYP2C9), S-(+)-mephenytoin (CYP2C19), budipine (CYP2D), and ketoconazole (CYP3A4). Probe substrate concentrations used for each P450 were identical to K_m values for metabolism of these substrates in human liver microsomes. In the case of mechanism-based inhibitors (e.g., orphenadrine), the inhibitor was preincubated with microsomes and NADPH for 30 min prior to the addition of substrate. For determination of apparent K_i values, the substrate concentrations used for each index reaction were equal to 1/2 K_m, K_m, 2 K_m. At each of the substrate levels, metabolite formation was monitored in the absence and in the presence of methoxsalen, tryptamine, and TCP individually (e.g., 1/4 IC₅₀, 1/2 IC₅₀, IC₅₀, 2 IC₅₀). IC₅₀ refers to the concentration of inhibitor required to inhibit 50% of substrate activity (at K_m concentration).

Data Analysis. Metabolic constants (K_m, V_max) were determined by use of nonlinear regression analysis after Michaelis-Menten representation (rate of metabolite formation by substrate concentration) using Enzpack 3 software (Cambridge, UK). Inhibition patterns were determined by Dixon plots. Apparent inhibition constant (K_i) values for competitive inhibition were estimated by visualization of data through Dixon plots or by using Pharm/PCS software (Wynnewood, PA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Evaluation of TCP and Methoxsalen Inhibition of Cotinine Formation from Nicotine. Of about 200 drugs screened as CYP2A6 inhibitors, including various monoamine oxidase inhibitors, antifungals, antidepressants, antipsychotics, and coumarin analogs, TCP, methoxsalen, and tryptamine displayed the highest CYP2A6 inhibitory potency upon screening. To confirm the potency inhibition, apparent K_i values were determined in CYP2A6-expressing microsomes using nicotine metabolism to cotinine as the substrate reaction. Dixon plots indicate that TCP and tryptamine showed competitive inhibition (K_i values of 65 nM and 1.7 µM, respectively) in cDNA-expressing microsomes (Figs. 1 and 2). Methoxsalen showed noncompetitive inhibition of nicotine metabolism, with an apparent K_i value of 0.1 µM in cDNA-expressing microsomes (Fig. 3). Similar results were also observed in HLM, with apparent K_i values of 0.2, 1.9, and 0.2 µM, for TCP, tryptamine, and methoxsalen, respectively (data not shown). Previous studies indicate that methoxsalen is a mechanism-based inactivator of CYP2A6 in human liver and cDNA-expressing microsomes (Draper et al., 1997; Koenigs et al., 1997). The noncompetitive nature of methoxsalen inhibition of CYP2A6 observed presently suggests that mechanism-based inactivation may be taking place, although further studies are necessary to confirm these findings.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Dixon plot of tranylcypromine inhibition of nicotine metabolism in cDNA-expressing CYP2A6 microsomes. Velocity is expressed in nanomoles per milligram of protein per minute.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Dixon plot of tryptamine inhibition of nicotine metabolism in cDNA-expressing CYP2A6 microsomes. Velocity is expressed in nanomoles per milligram of protein per minute.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Dixon plot of methoxsalen inhibition of nicotine metabolism in cDNA-expressing CYP2A6 microsomes. Velocity is expressed in nanomoles per milligram of protein per minute.

Potency of Inhibition (Apparent K_i Values) of Major P450s by TCP, Tryptamine, and Methoxsalen in cDNA-Expressing Microsomes. Inhibition of major P450s by methoxsalen, TCP, and tryptamine was first evaluated in HLM and cDNA-expressing microsomes at 20 and 200 µM. Even higher concentrations of TCP, tryptamine, and methoxsalen demonstrated a relatively higher degree of selectivity and specificity in their inhibition of CYP2A6. As shown in Table 2, racemic TCP selectively inhibited CYP2A6-mediated metabolism of coumarin with an apparent K_i value of 0.08 µM. P450 2B6, 2C9, 2C19, and 2E1 were also inhibited by TCP, but with higher K_i values (5, 10, 15, and 12 µM, respectively). P450 1A2, 2D6, and 3A4 were inhibited with lower potency (K_i = 25, 30, and 450 µM, respectively) relative to CYP2A6. Tryptamine selectively inhibited CYP1A2 and CYP2A6, both characterized by apparent K_i values of 1.7 µM. P450 2B6, 2C9, 2C19, 2D6, 2E1, and 3A4 were also inhibited, but with higher K_i values (75, 75, 85, 58, 11, and 363 µM, respectively).

Potency of Inhibition (Apparent K_i Values) of Major P450s by TCP and Methoxsalen in HLM. In general, the degree and selectivity of inhibition followed the same pattern in HLM as in cDNA-expressing microsomes. The greatest differences between K_i values measured in cDNA-expressing and human liver microsomes were observed in the inhibition of metabolic reactions mediated by more than one enzyme. For example, O-deethylation of 7-ethoxy-4-trifluoromethylcoumarin in HLM is mediated by multiple enzymes including CYP2B6 (Ekins et al., 1997). However, 7-ethoxy-4-trifluoromethylcoumarin is the best available substrate because of high product turnover in cDNA-expressing microsomes. CYP1A2 has been shown to be the only enzyme involved in the high-affinity component of phenacetin O-deethylation in vitro (Venkatakrishnan et al., 1998); therefore, the lower K_i observed in cDNA-expressing microsomes is likely due to the absence of other P450s present in HLM that may interact with the inhibitor.

Stereoselective Inhibition of P450s by R-(+)-TCP and S-()-TCP. TCP is a mixture of ()- and (+)-trans-2-phenylcypropylamine, and previous studies have shown that the enantiomers of tranylcypromine have markedly different pharmacological properties (Hampson et al., 1986). To determine whether there was any stereoselective inhibition of CYP2A6 by TCP, we compared the formation of cotinine from nicotine in the presence of R-(+)-TCP and S-()-TCP (Fig. 4). S-()-TCP had a much lower inhibitory effect on the formation of cotinine relative to R-(+)-TCP. Apparent K_i values were determined for R-(+)- and S-()-TCP with cDNA-expressing P450s, which showed the lowest K_i values with racemic TCP (Table 3). As shown, R-(+)-TCP was a 20 times more potent CYP2C19 inhibitor than S-()-TCP, while S-()-TCP was 2 times more potent in the inhibition of CYP2B6 than R-(+)-tranylcypromine. Both enantiomers displayed low K_i values with CYP2A6 (using coumarin as the substrate), although R-(+)-TCP was a stronger inhibitor, with a 4-fold reduced K_i, as compared with S-()-TCP. The inhibitory potency of S-()-TCP and R-(+)-TCP for CYP2E1 was not significantly different.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Stereoselective inhibition of CYP2A6-mediated cotinine formation from nicotine by two isomers of tranylcypromine. , (±)-tranylcypromine; , R-(+)-tranylcypromine; , S-()-tranylcypromine.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In vitro kinetic and inhibition studies using cDNA-expressing and human liver microsomes, which can determine the selectivity and specificity of a particular drug in vitro, can be useful tools in assessing the potential for interactions among drugs in vivo. In this study, selective substrates and inhibitors for each of the principal drug-metabolizing P450s were used to determine the relative potency of inhibition of TCP, methoxsalen, and tryptamine for each enzyme.

The apparent K_i values of tryptamine, methoxsalen, and TCP for each enzyme in cDNA-expressing microsomes showed that they are relatively selective and specific inhibitors of CYP2A6. Inhibition studies showed that methoxsalen inhibits CYP2A6 (0.2 and 0.8 µM) and CYP1A2 (K_i = 1.0 and 0.2 µM) in human liver and cDNA-expressing microsomes, respectively, which is in agreement with previously published reports (Maenpaa et al., 1993; Ono et al., 1996; Draper et al., 1997; Koenigs et al., 1997). However, TCP inhibited CYP2A6 with greater potency and selectivity than methoxsalen. TCP inhibition of CYP2A6-mediated coumarin 7-hydroxylation was characterized by apparent K_i values of 0.08 and 0.2 µM in cDNA-expressing and human liver microsomes, respectively. TCP inhibition of other P450 enzymes was generally characterized by higher apparent K_i values than those for methoxsalen in human liver microsomes, indicating that TCP has greater selectivity for CYP2A6 than does methoxsalen (Tables 2 and 3). These results are in agreement with a previous study showing that TCP (at 1000 µM) has higher CYP2A6 selectivity than methoxsalen (at 10 µM) (Ono et al., 1996). The present study, in which K_i values were determined using a range of pharmacologically relevant inhibitor concentrations, more clearly demonstrates the potency of CYP2A6 inhibition by TCP and methoxsalen. The differences in interactions of both drugs with CYP3A4 are such that methoxsalen inhibits the enzyme with approximately 6- to 25- fold greater affinity than TCP in human liver and cDNA-expressing microsomes, respectively. However, even though CYP3A4 is the most abundant P450 enzyme in the liver and mediates the metabolism of many therapeutically important drugs, the concentrations needed to produce inhibition far exceed those achieved by methoxsalen in vivo (Sellers et al., 2000).

R-(+)-TCP is a more potent inhibitor of CYP2A6 (K_i = 0.05 µM) than S-()-TCP (2.0 µM). As molecular modeling studies of CYP2A6 with coumarin have shown, Phe181, Gln74, and His437 residues at the CYP2A6 active site are mainly responsible for substrate-enzyme contact (Lewis et al., 1995). His437 and Gln74, in particular, form hydrogen bonds with the substrate. Phe181 forms - stacking interactions with the aromatic system of the substrate, ultimately positioning the substrate for the reaction (Lewis et al., 1995). The positioning of both the aromatic ring and amino groups of R-(+)-TCP in the CYP2A6 active site may be more favorable than that with the S-()-isomer, leading to a stronger enzyme-inhibitor interaction. Further studies are necessary to elucidate the stereoselectivity of TCP with P450 enzymes.

Some reports have suggested that individuals with a CYP2A6 null allele or gene deletion appear to smoke fewer cigarettes (Pianezza et al., 1998; Rao et al., 2000). In this study, TCP, methoxsalen, and tryptamine showed to be relatively selective inhibitors of CYP2A6, in particular, TCP. In vivo pharmacokinetic studies showed that people receiving a 20-mg oral dose of racemic TCP (Parnate) have peak plasma levels of 64.5 to 190 ng/ml (about 0.57-1.68 µM) at 1.5 h and a mean elimination half-life of 1.5 to 2.5 h (Mallinger and Edwards, 1986). The K_i values observed here in vitro with racemic TCP (0.2 and 0.08 µM in human liver and cDNA-expressing microsomes, respectively) suggest that a conventional, therapeutic dose of TCP will inhibit CYP2A6 in vivo. Indeed, it has been shown that 2.5 and 10 mg of oral racemic TCP significantly increased the oral bioavailability of nicotine in vivo (Sellers et al., 2000).

It was also demonstrated, in vitro, that R-(+)-TCP is a more potent CYP2A6 inhibitor than S-()-TCP. It has been shown that individuals receiving a standard 20-mg dose of racemic TCP showed significantly higher plasma levels of S-()-TCP, as compared with R-(+)-TCP (Spahn-Langguth et al., 1992). Moreover, the apparent oral clearance of R-(+)-TCP exceeded that of S-()-TCP (Spahn-Langguth et al., 1992). Therefore, it remains to be determined whether R-(+)-TCP will be a stronger CYP2A6 inhibitor in vivo as compared with S-()-TCP.

Tryptamine was shown to be a selective inhibitor of CYP2A6 in vitro. L-Tryptophan can be metabolized into tryptamine in vivo by tryptophan decarboxylase. The concentration of tryptamine is correlated with that of L-tryptophan (Wollmann et al., 1985). Therefore, it may be possible to apply L-tryptophan as an in vivo CYP2A6 inhibitor, although further in vivo studies are clearly necessary.

In conclusion, the data presented here demonstrate that tryptamine, TCP, and methoxsalen have high specificity and relative selectivity for CYP2A6. TCP (particularly the R-isomer), methoxsalen, and tryptamine have relative selectivity for CYP2A6 inhibition and therefore would not be expected to give rise to significant clinical drug-drug interactions. Imitating the gene defect by chemical inhibition with R-(+)-TCP may be an effective way to treat nicotine dependence (Sellers et al., 2000). However, further long-term in vivo studies are required to validate this hypothesis.