| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Thiamylal, a chiral thiobarbiturate, is marketed as a racemic product. We studied the serum protein binding and microsomal metabolism of thiamylal enantiomers in vitro. The unbound fraction of R(+)-thiamylal was greater than that of S(-)-thiamylal. The analysis of binding data revealed that both enantiomers bound to human serum albumin through only one site. In displacement studies with site-specific probes, dansylsarcosine, but not warfarin, significantly decreased the binding of both enantiomers. The bindings of enantiomers were also decreased by octanoate and a large concentration of oleate. These findings suggest that both enantiomers bind to Site II of albumin with higher affinity for S(-)-enantiomer. R(+)-thiamylal was metabolized more rapidly than S(-)-enantiomer by human liver microsomes. An experiment with isoform-selective inhibitors and cytochrome P-450 (CYP) isoforms showed that CYP2C9 had the highest activity for the metabolism of both enantiomers, the activity being 7 to 10 times that of CYP2E1 and CYP3A4. CYP2C9 showed a significantly rapid metabolism of R(+)-enantiomer, suggesting that CYP2C9 is mainly involved in the enantioselective metabolism of thiamylal.
Implications: Because clinically marketed thiamylal is a racemic compound, a pharmacokinetic study of each enantiomer may be beneficial. We found that the enantioselectivity of thiamylal existed in protein binding and metabolism. This may be caused by the differences in the affinities of enantiomers for albumin and cytochrome P-450 isoform.
Thiamylal is an ultrashort-acting thiobarbiturate, with a chemical structure very similar to thiopental and pentobarbital. Thiamylal has a chiral carbon in the side chain and is marketed for clinical use as a racemic product, i.e., an equimolar mixture of two enantiomers, R(+) and S(-). The properties of the enantiomers generally differ in their qualitative and quantitative pharmacodynamic activities. S(-)-enantiomers of barbiturates containing a 5-(1-methylbutyl)alkyl group, such as pentobarbital, thiopental, and thiamylal, are more potent as anesthetics than the corresponding R(+)-enantiomers (1). Differences in pharmacokinetic properties can exist between enantiomers. With respect to thiamylal, we previously reported (2) that the concentration of the S(-)-enantiomer was consistently larger than that of the R(+)-enantiomer in humans. The total clearance and the distribution volume of R(+)-enantiomer were significantly greater than those of the antipode, and these differences were implicated as occurring mainly because of an enantioselective binding (2). However, those results were obtained from patients who underwent thiamylal treatment; thus, their complicated background had to be considered. For example, most of the patients were co-administered phenytoin and cimetidine, which might have affected the pharmacokinetics of both enantiomers. We designed our study to clarify the enantioselectivity in protein binding and microsomal metabolism of thiamylal in vitro.
Because we could not obtain thiamylal enantiomers, we used racemic thiamylal sodium (Isozol®; Yoshitomi Pharmaceutical Industries, Osaka, Japan). The in vitro protein binding of thiamylal enantiomers in a racemic mixture was determined by an ultrafiltration technique with sera from six healthy male volunteers. Written informed consent was obtained from all subjects. Their serum albumin concentrations determined with the Vision system (Abbot Laboratories; Abbot Park, IL) were within the normal range (3647 g/L). Racemic thiamylal sodium (18180 µM) was added to the serum and incubated for 30 min in a water bath at 37°C to reach the equilibrium of binding. After incubation, 1 mL of mixture was instilled into the sample reservoir of a Centrifree Micropartition System (Amicon, Beverly, MA) for centrifugation at 1000g for 5 min at 25°C. Under these conditions, approximately 150 µL of the filtrate was obtained. The adsorption of thiamylal enantiomers to the filtration system was negligible and no protein leakage was observed during the experiment. The concentrations of thiamylal enantiomers in the incubated mixture and filtrate were measured by a chiral high-pressure liquid chromatography (HPLC) with ultraviolet detection as described below and the unbound fractions were estimated directly from the ratio of drug concentration in filtrate to the incubated mixture.
For the determination of binding variables, the same procedure was performed with a 580 µM (40 g/L) human serum albumin (HSA) solution prepared in a 0.067 M phosphate buffer (pH 7.4). Racemic thiamylal was added at the concentration range from 18 to 724 µM. The binding data were fitted to the Scatchard model for one or two classes of independent sites described by the following equation:
Displacement studies of thiamylal enantiomers using specific binding probes (i.e., dansylsarcosine and warfarin) or fatty acids (octanoate and oleate) were performed at 36 µM of the racemic thiamylal with a 580 µM HSA solution prepared in a 0.067 M phosphate buffer (pH 7.4). Specific probes were used at 290 and 580 µM, and fatty acids were added at 290, 580, and 1160 µM. The conditions of incubation and ultrafiltration were the same as previously described. The amounts of microsomes and the incubation time to metabolize each enantiomer were evaluated as the indices of the enantioselectivity of thiamylal metabolism by using pooled human liver microsomes. An incubation mixture (0.27 mL) containing microsomes, 0.1 M phosphate buffer (pH 7.4), 0.1 mM EDTA disodium, and 72 µM racemic thiamylal was placed in a microtube and preincubated at 37°C for 1 min, and then, enzyme reactions were initiated by adding 30 µL of a nicotinamide adenine dinucleotide phosphate (NADPH)-generating system which consisted of 20 mM glucose-6-phosphate, 5 mM NADP, 40 mM MgCl2, and 10 units/mL of glucose-6-phosphate dehydrogenase. Control experiments without the NADPH-generating system were performed simultaneously. For assessing the amounts of microsomes for thiamylal metabolism, mixtures containing 72 µM racemic thiamylal, and microsomes equivalent to 0.2, 0.4, 0.8, and 1 mg protein/mL were incubated for 1 h. With regard to incubation time for thiamylal metabolism, the microsomes (1 mg protein/mL) were incubated with 72 µM racemic thiamylal for 0.5, 1, 1.5, and 2 h. The chemical degradation of thiamylal within this time was negligible. After an incubation of the final mixture at 37°C in a shaking water bath, the enzyme reactions were terminated by placing a microtube in crushed ice and the residual amounts of thiamylal enantiomers were measured by chiral HPLC.
For the determination of cytochrome P-450 (CYP) isoforms involved in the metabolism of thiamylal enantiomers, the isoform-selective inhibitor probes or isoforms of human CYP expressed in baculovirus-infected insect cells were used. The selective probes were as follows: Experiments with human recombinant CYPs were performed according to the procedure used for the experiments with human liver microsomes. However, the mixture contained 7.2 µM racemic thiamylal and 50 pmol CYP and was incubated for 30 min. We used 100 mM Tris-HCl buffer (pH 7.4) and 100 mM phosphate buffer (pH 7.4) for the CYP2C9 incubation mixtures and other CYP reaction mixtures, respectively. Control experiments with microsomes isolated from the same cell line containing the vector, but without a cDNA insert, were performed in parallel. The concentration of thiamylal enantiomer in various samples was determined simultaneously by a chiral HPLC (11). Thiamylal enantiomers and n-propyl-p-hydroxybenzoate (internal standard) were extracted with 20% (vol/vol) diethylether in n-hexane, and then, the organic phase was evaporated to dryness under a gentle stream of nitrogen gas. The resulting residue was resolved with 50% (vol/vol) methanol water solution and injected into the HPLC system. The filtrates in protein binding experiments were mixed with IS methanol solution and directly injected into the HPLC system. The conditions for HPLC were as follows: a column consisting of Chiral-AGP (100 x 4.0 mm, ChromTech AB, Hägersten, Sweden) equipped with a guard column and a prefilter; mobile phase of 20 mM KH2PO4 containing 3% (vol/vol) 2-propanol; a flow rate of 0.9 mL/min; and detection by ultraviolet absorbance at 288 nm. Data are expressed as the mean ± SD. Students paired t-test was used to compare the R(+)- and S(-)-enantiomer group. Multiple comparisons were performed by using one-way analysis of variance with post hoc comparisons of Dunnets test. The significance of difference was calculated by statistical analysis software (StatView 5.0; SAS Institute, Cary, NC) on an Apple computer. Differences were regarded as significant at P < 0.05.
At all thiamylal concentrations examined, the unbound fraction of R(+)-thiamylal was significantly greater than that of the S(-)-enantiomer (P < 0.001) (Table 1). The unbound fraction of both enantiomers was raised by increasing the thiamylal concentration from 9 to 90 µM.
The binding data of thiamylal enantiomers as a transformed (i.e., Scatchard) representation were almost linear, and the AIC value obtained from the model of one class of site was smaller than that from the model of two classes. The numbers of binding sites per molecule of albumin of R(+)- and S(-)-thiamylal were 0.94 ± 0.03 and 0.88 ± 0.02, respectively, and the association constant (1/mol) were 34,657 ± 1497 and 87,282 ± 2404, respectively. R(+)- and S(-)-thiamylal had similar numbers of binding sites, whereas the affinity was higher for S(-)-thiamylal. The addition of warfarin had no effect on the binding of R(+)- and S(-)-thiamylal, whereas dansylsarcosine significantly increased the unbound fraction of both enantiomers in a concentration-dependent manner (Table 2). The influences of two fatty acids with different chain lengths, octanoic acid (medium-chain) and oleic acid (long-chain), were examined on the binding of each thiamylal enantiomer. The unbound fractions of the R(+)- and S(-)-enantiomer, respectively, were significantly increased by the addition of each fatty acid (Table 2). Octanoic acid markedly increased the unbound fractions of R(+)- and S(-)-enantiomer at the small concentration of 280 µM by 170% and 200%, respectively.
Thiamylal enantiomers were metabolized by human liver microsomes depending on the microsomal concentration and incubation time (Fig. 1). The residual amounts of S(-)-thiamylal tended to be higher than the antipode in the range of 0.41.0 mg protein/mL or more than that of human liver microsomes (Fig. 1A). At each period during the incubation (0.52 h), S(-)-thiamylal showed significantly larger residual amounts, when compared with R(+)-enantiomer (Fig. 1B).
Sulfaphenazole, DDC, and ketoconazole, probe for CYP2C9, CYP2E1, and CYP3A4, respectively, showed a potent inhibitory effect in a concentration-dependent manner (Fig. 2). Quinidine, a probe for CYP2D6, slightly stimulated the metabolism of R(+)-thiamylal. Other probes for the respective CYP isoforms tested in this study ( -naphthoflavone for CYP1A2, coumarin for CYP2A6, and omeprazole for CYP2C19) showed no inhibitory effects on the metabolism of thiamylal.
CYP2C9 had the highest activity for the metabolism of both enantiomers, at 7 to 10 times that of CYP2E1 and CYP3A4. CYP2C9 and CYP3A4, but not CYP2E1, had a significant enantioselectivity (Fig. 3).
The extent to which a particular enantiomer of a chiral drug is bound to plasma proteins is an important factor in the overall pharmacokinetic profile of that drug. We found that thiamylal had an enantioselective binding activity for HSA, where the S(-)-enantiomer bound more than the R(+)-enantiomer. Within the range of serum concentrations found in the patients, the bound fraction in human serum was almost constant at 87% and 93% for R(+)- and S(-)-thiamylal, respectively. Both the Scatchard analysis of binding data and the comparison of AIC values demonstrated that thiamylal enantiomers occupied only one class of binding site in HSA, a site characterized by enantioselectivity with a higher affinity for S(-)-thiamylal. HSA has two major, distinct drug-binding sites, termed Sites I and II. We used site-specific probes and fatty acids to characterize the domains of HSA involved in the binding of thiamylal enantiomers. Warfarin and dansylsarcosine were specific binding probes of Sites I and II, respectively (12), and both R(+)- and S(-)-thiamylal were displaced by dansylsarcosine in a concentration-dependent manner, but not by warfarin, indicating that the binding of both enantiomers to HSA occurs only at Site II. Fatty acids affect the binding of drug to albumin differently because of the length of the fatty acid chain. Octanoic acid (medium-chain) selectively decreases the binding of diazepam and dansylsarcosine to Site II (13). On the other hand, long-chain fatty acids, such as oleic acid and palmitic acid, increase the binding of warfarin to Site I and decrease the binding of L-tryptophan to Site II (14,15). In our study, octanoic acid significantly decreased the bindings of R(+)- and S(-)-thiamylal to HSA, and oleic acid also decreased these bindings. These results were consistent with those obtained by using site-specific probes. Because little thiamylal is excreted in urine in the unchanged form (16), this compound is eliminated from the body mostly by hepatic metabolism. In some chiral drugs, enantiomers have a different metabolic rate. When thiamylal was incubated with human liver microsomes, there was less residual R(+)- than S(-)-thiamylal. This suggests an enantioselectivity with rapid metabolism of the R(+)-enantiomer in microsomal metabolism. The difference in systemic clearance between the enantiomers observed in the previous report (2) may be caused by, not only the enantioselective protein binding, but also by the enantioselective microsomal metabolism. In our study, the residual amounts of thiamylal enantiomers increased on simultaneous incubation with the isoform-selective inhibitors, sulfaphenazole (CYP2C9), DDC (CYP2E1), and ketoconazole (CYP3A4) in human liver microsomes. Although the specificity of DDC or ketoconazole decreases at large concentrations (17,18), CYP2C9, CYP2E1, and CYP3A4 may be responsible for the metabolism of both R(+)- and S(-)-thiamylal. It was confirmed, in the experiments by using human recombinant CYPs, that the same CYP isoforms participate in the metabolism of both enantiomers. CYP2C9 was dominant in the metabolism of thiamylal; its activity was 7 to 10 times that of CYP2E1 and of CYP3A4. CYP2C9, CYP2E1, and CYP3A4 have been reported to account for 20%, 7%, and 30%, respectively, of total P-450 in human liver microsomes (19,20). When each of the activities is corrected for the respective P-450, CYP2C9 is the most active in thiamylal metabolism followed by CYP3A4 and CYP2E1. It is, therefore, highly likely that CYP2C9 is the major isoform responsible for the thiamylal metabolism, and that it significantly contributes to the enantioselective metabolism of thiamylal. Phenytoin, S-warfarin, and tolbutamide are well known substrates for CYP2C9. The metabolism of these drugs is induced by phenobarbital, carbamazepine, and rifampicin but inhibited by various drugs including cimetidine, imidazole derivatives, and sulfonamides (21). These drugs should have similar effects on the pharmacokinetics of both R(+)- and S(-)-thiamylal, because both enantiomers are mainly metabolized by CYP2C9. Additionally, CYP2C9 has genetic polymorphism (21). An allelic variant of CYP2C9 that causes a substitution of Leu359 for Ile359 is associated with a decreased metabolic clearance of many of the therapeutic drugs previously mentioned (22,23). Although the frequency of this variant allele is relatively small, 6% to 9% in Caucasians (23,24) and 2% in Japanese (25), this polymorphic metabolism has to be taken into account when monitoring patients treated with thiamylal. In summary, R(+)- and S(-)-thiamylal differ from each other in fraction unbound and metabolic rate. This enantioselectivity is because of the different affinities for HSA and CYP, because both enantiomers bind to same binding site on HSA (Site II) and are metabolized by the same CYP isoform (CYP2C9).
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|