Abstract
The effect of binding of amitriptyline to human liver microsomes and to microsomes from human B-lymphoblastoid cells on the estimation of enzyme kinetic parameters describing N-demethylation to nortriptyline was investigated using a combination of microsomal binding and in vitro enzyme kinetic studies. Quantitative binding in both matrices increased with higher microsomal protein concentrations (free fractions 0.88–0.32 at 100–500 μg protein/ml in human liver microsomes and 0.82–0.26 at 250–1000 μg protein/ml in microsomes from B-lymphoblastoid cells) and was independent of amitriptyline concentration over a concentration range of 0.2 to 200 μM. Addition of heat-inactivated microsomal protein (50–450 μg/ml) to native human liver microsomes (50 μg/ml) reduced the amitriptylineN-demethylation rate in a protein concentration dependent manner. This effect was greater at lower substrate concentrations and was overcome by saturating concentrations of substrate, thereby decreasing the apparent affinities of the high- and low-affinity components of the N-demethylation process, with minimal effect on the net Vmax. Addition of metabolically inactive microsomes from untransfected human lymphoblastoid cells (750 μg/ml) to CYP2C19 (250 μg/ml protein) increased the apparent Km value for amitriptyline N-demethylation by 3.5-fold and increased the uncompetitive substrate inhibition constant (Ks) by 2.2-fold, making substrate inhibition essentially undetectable. A similar effect was seen with CYP3A4, with a 1.8-fold increase in the S50 (substrate concentration at which half-maximal velocity of a Hill enzyme is achieved). Microsomal binding did not alter theVmax of either CYP isoform to any appreciable extent. These findings emphasize the importance of incorporating microsomal binding in the estimation of enzyme kinetic parameters in vitro and making appropriate corrections for unbound drug concentrations.
Binding of drug substrates to in vitro incubation matrices results in an underprediction of in vivo drug clearance from apparent in vitro intrinsic clearance determined in enzyme kinetic studies. Correction for the fraction unbound in the in vitro incubation matrix has improved the prediction of pharmacokinetic clearance estimates in several studies, indirectly implicating this phenomenon as the cause of underprediction of intrinsic clearance (Obach, 1996, 1997, 1999; Obach et al., 1997; Carlile et al., 1999). However, the impact of nonspecific binding on the determination of Michaelis constants (Km values) describing the drug biotransformation process has not been directly investigated.
In the present study, we evaluated the effect of binding to human liver microsomes and microsomes from a human B-lymphoblastoid cell line (a widely used model for the production of heterologously expressed CYP isoforms) on the estimation of in vitro enzyme kinetic parameters, using the N-demethylation of the tricyclic antidepressant amitriptyline as the model pathway. The results show that amitriptyline is bound to both human liver and B-lymphoblastoid cell microsomes. The extent of binding increases with increasing microsomal protein concentration and is drug concentration-independent over the range of amitriptyline concentrations generally used in vitro. The predominant effect of microsomal binding on enzyme kinetic parameters was a decrease in the apparent affinity for substrate (i.e., an increase in apparent Km or S50), without any appreciable effect on the reaction velocity at saturating substrate concentrations (Vmax), consistent with an underprediction of intrinsic clearance.
Experimental Procedures
Materials.
Amitriptyline, nortriptyline, clomipramine, desipramine, NADP+, (±)-isocitric acid, MgCl2, and isocitrate dehydrogenase were purchased from Sigma Chemical Co. (St. Louis, MO).
Liver samples obtained from the International Institute for the Advancement of Medicine (Exton, PA) or the Liver Tissue Procurement and Distribution service (University of Minnesota, Minneapolis, MN) were from transplant donors with no known liver disease. The tissue was partitioned and kept at −80°C until the time of microsome preparation as described previously (von Moltke et al., 1993, 1994). One of four livers used in enzyme kinetic studies of amitriptylineN-demethylation was of poor metabolizer phenotype with respect to CYP2C19. All livers were normal metabolizers with respect to CYP2D6 activity. Enzymatically inactive microsomes were prepared by heat-inactivation of native liver microsomes in a boiling water bath for 4 min, with subsequent confirmation of metabolic inactivity at an amitriptyline concentration of 500 μM.
Microsomes from cDNA-transfected human lymphoblastoid cells expressing CYP2C19 or CYP3A4, as well as metabolically inactive control microsomes from untransfected cells, were purchased from Gentest Corporation (Woburn, MA), aliquoted, and stored at −80°C. Microsomes from BTI-TN-5B1-4 insect cells (SUPERSOMES) were purchased from Gentest as well. Microsomal protein concentrations and CYP content were provided by the manufacturer.
Spectra/Por 2.1 Biotech membranes (molecular weight cutoff, 15,000; flat width, 4 mm; diameter, 3.8 mm) were purchased from Spectrum Laboratories (Rancho Dominguez, CA).
Equilibrium Dialysis.
The binding of amitriptyline to human liver microsomes or microsomes from human lymphoblastoid cells or insect cells was determined by equilibrium dialysis using a modification of a previously described procedure (Obach, 1997). Dialysis tubing was soaked overnight in deionized water at 4°C. The tubing was rinsed again in deionized water, followed by dialysis buffer (50 mM KH2PO4, 5 mM MgCl2, pH 7.4). Amitriptyline in a methanolic solution was evaporated to dryness in round-bottomed glass tubes, and spiked microsomal suspensions of desired drug and protein concentrations were prepared in dialysis buffer. Dialysis tubing was cut into 5- to 6-inch-long pieces, and one end was sealed with a knot. Spiked microsomal suspension (approximately 400 μl) was loaded into the bag, and the other end was sealed with another knot. The bag was quickly rinsed in buffer and submerged in 7 ml of dialysis buffer in a 15-ml polypropylene centrifuge tube. The membranes were kept wet throughout the procedure. The tubes were capped and placed in a shaking water bath at 37°C for 6 h. Preliminary time course studies suggested that equilibrium was attained within 4 h of dialysis. Retentates were collected using a 1-ml tuberculin syringe. Dialysates and retentates were frozen at −20°C until analysis.
Amitriptyline concentrations in dialysates and retentates were measured by HPLC using a modification of a previously described extraction procedure (Abernethy et al., 1981). One milliliter of dialysate or 0.15 to 0.25 ml of retentate was transferred to round-bottomed glass tubes containing internal standard (200 ng clomipramine, evaporated to dryness from a methanolic solution). Drug-free microsomal suspension (0.15–0.25 ml) or dialysis buffer (1 ml) was added to the dialysate or retentate tubes, respectively. Buffer and drug-free microsomal suspension were both added to standard curve tubes containing 0 and 10 to 50,000 ng amitriptyline and internal standard. All tubes were alkalinized with 1 ml of 0.25 N NaOH and extracted with 3 ml of hexane:isoamyl alcohol (98:2) by vortex mixing for 45 s. The tubes were centrifuged, and the organic phase was transferred to another set of glass tubes, evaporated to dryness, and reconstituted in 200 μl of HPLC mobile phase. Then, 25 to 125 μl was injected onto a 30-cm × 3.9-mm steel reverse-phase C18 μBondapack column. The mobile phase was a 60:40 mixture of 50 mM KH2PO4 and acetonitrile at a flow rate of 1.8 ml/min, and ultraviolet detection at 214 nm was used. Calibration curves were linear, and amitriptyline was quantified by measurement of peak height ratios with reference to the calibration curve.
The fraction unbound in the microsomal suspension was calculated as the ratio of amitriptyline concentration in the dialysate to that in the retentate (Obach, 1997), with the latter being equivalent to total drug concentration (Wright et al., 1996). In studies of microsomal protein concentration dependence, the spiked drug concentration was 50 μg/ml (180 μM). Protein concentrations of 50 to 500 μg/ml (human liver microsomes) or 250 to 1000 μg/ml (human lymphoblastoid cell microsomes) were used. When drug concentration dependence was investigated, the microsomal protein concentration was 200 μg/ml (human liver microsomes) or 250 μg/ml (human lymphoblastoid cell microsomes), and amitriptyline was spiked at 3 to 1250 μg/ml (resultant total concentrations in the range of 0.2–200 μM). Pooled human liver microsomes (from seven livers) were used in all equilibrium dialysis experiments.
In Vitro Amitriptyline Biotransformation and Enzyme Kinetic Analyses.
Incubations of amitriptyline with human liver microsomes and HPLC analysis of nortriptyline in incubates were performed as previously described (Schmider et al., 1995, 1996; Venkatakrishnan et al., 1998). A 17-point nortriptyline formation curve (0 and 2.5–500 μM amitriptyline) was used to characterize the kinetics of amitriptyline N-demethylation by four individual human liver microsomal samples at a microsomal protein concentration of 50 μg/ml. The impact of microsomal binding on reaction rates and enzyme kinetic parameters was assessed by generating kinetic curves with the addition of 50, 150, or 450 μg/ml heat-inactivated human liver microsomal protein.
A 12-point nortriptyline formation curve was used to characterize the kinetics of amitriptyline N-demethylation by lymphoblast-expressed CYP2C19 and CYP3A4 at a microsomal protein concentration of 250 μg/ml. These CYP isoforms were chosen based on previous studies that have demonstrated their importance as the major high- and low-affinity amitriptyline N-demethylases, respectively (Olesen and Linnet. 1997; Venkatakrishnan et al., 1998). Kinetic curves were also generated with the coaddition of 750 μg/ml control microsomes from untransfected lymphoblastoid cells to assess the impact of microsomal binding on the estimation of enzyme kinetic parameters.
The kinetics of amitriptyline N-demethylation by human liver microsomes or heterologously expressed CYP isoforms were described by one of the following models relating nortriptyline formation rate (v) to concentration of the substrate (S), amitriptyline:
A one-enzyme Michaelis-Menten model with uncompetitive substrate inhibition for lymphoblast-expressed CYP2C19:
Model selection was based on examination of Eadie-Hofstee transformed plots. Kinetic parameters were determined by nonlinear least-squares regression using SigmaPlot software (Jandel Scientific, Costa Madre, CA).
Results
Binding Studies.
The effect of microsomal protein concentration on the unbound fraction of amitriptyline is shown in Table 1. There was essentially no binding at a human liver microsomal concentration of 50 μg/ml, and the free fraction progressively decreased as the protein concentration was increased, with an unbound fraction of only 0.32 at a protein concentration of 500 μg/ml. A similar trend was noted for lymphoblastoid cell microsomes, although the extent of binding was lower than that in liver microsomes. Amitriptyline was also bound to insect cell microsomes with a free fraction of nearly 0.7 at a protein concentration of 250 μg/ml, a concentration typically used in incubations with SUPERMIX and SUPERSOME formulations manufactured by Gentest. The extent of human liver microsomal binding of amitriptyline was increased by heat inactivation to a small extent, with the bound fraction being 26% higher at a protein concentration of 200 μg/ml.
Amitriptyline binding to human liver microsomes and microsomes from human B-lymphoblastoid cells or insect cells
The free fraction of amitriptyline in human liver microsomes (200 μg/ml protein) and human lymphoblastoid cell microsomes (250 μg/ml protein) averaged 0.59 ± 0.06 and 0.83 ± 0.09 (mean ± S.D., n = 20 determinations) over an amitriptyline total concentration range of 0.2 to 200 μM, with no evidence of concentration dependence (Fig. 1, A and B).
Free fraction of amitriptyline (AMI) in human liver microsomes (200 μg/ml, A) and in microsomes from human B-lymphoblastoid cells (250 μg/ml, B) in relation to total drug concentrations. Dots are experimental data points, and the line denotes the overall mean free fraction. Note that the free fraction is essentially concentration-independent over the amitriptyline concentration range studied.
Effect of Nonspecific Microsomal Binding on Human Liver Microsomal Amitriptyline N-Demethylation.
The effect of microsomal binding on the kinetics of amitriptylineN-demethylation was studied in four human liver microsomal preparations. The active microsomal protein concentration was fixed at 50 μg/ml, and kinetic curves were generated with the addition of 0, 50, 150, or 450 μg/ml heat-inactivated microsomes from the same liver. Averaged data are shown in Figs. 2 (primary plots) and 3 (Eadie-Hofstee transformations), and a two-enzyme model (high-affinity Michaelis-Menten component plus low-affinity Hill component; eq. 3) was used in the determination of enzyme kinetic parameters. The choice of this model is consistent with previous enzyme kinetic studies of amitriptylineN-demethylation (Schmider et al., 1996; Ghahramani et al., 1997; Venkatakrishnan et al., 1998) and with the underlying profile of the biotransformation pathway (Olesen and Linnet. 1997; Venkatakrishnan et al., 1998). The apparent Km and S50 values of the high- and low-affinity components, respectively, increased with increasing amounts of heat-inactivated microsomal protein (i.e., with increasing extent of microsomal binding) without any consistent effect on the estimation of the Vmax value of the high- and low-affinity components (Table 2). Unbound Km and S50 values were calculated from the estimated apparent values of these parameters and estimated free fractions (1.0, 0.88, 0.63, and 0.32 at total microsomal protein concentrations of 50, 100, 200, and 500 μg/ml, respectively). These unboundKm and S50values are similar at all four protein concentrations (Table 2). Figure4 shows the effect of addition of heat-inactivated microsomal protein on amitriptylineN-demethylation rate as a function of substrate concentration. In all the livers, the fractional decrement in reaction rate decreased with increasing substrate concentration, with the effect being maximal at low (2.5–25 μM amitriptyline) substrate concentrations. The decrement of reaction rate also increased with increasing protein concentration, consistent with the results of binding studies. The effects of substrate concentration as well as microsomal protein concentration on the fraction of control rate were statistically significant (P < .001, two-way repeated measures ANOVA).
Kinetics of amitriptyline (AMI)N-demethylation by human liver microsomes (50 μg/ml protein), with varying concentrations of heat-inactivated microsomal protein added. Mean ± S.E. values of rates determined for four individual livers are presented. Lines are fitted functions using a two-enzyme model with a high-affinity Michaelis-Menten component and a low-affinity Hill enzyme component (eq. 3). See Table 2 for kinetic parameters.
Kinetic parameters for amitriptyline N-demethylation by human liver microsomes in the presence of varying amounts of heat-inactivated microsomal protein.
Protein concentration-dependent decrease in human liver microsomal amitriptyline (AMI) N-demethylation rates as a function of substrate concentration. Data are presented as the fraction of control rates measured in the absence of heat-inactivated microsomes. Mean ± S.E. (n = 4 livers) values are presented. Circles, squares, and diamonds represent added inactive protein concentrations of 50, 150, and 450 μg/ml, respectively. Note that the effect is overcome at saturating concentrations of substrate.
Effect of Microsomal Binding on AmitriptylineN-Demethylation by Lymphoblast-Expressed CYP Isoforms.
Lymphoblast-expressed CYP2C19 (250 μg/ml protein) N-demethylated amitriptyline with high affinity and displayed substrate inhibition (eq. 1). Addition of 750 μg/ml of metabolically inactive control microsomes from untransfected cells produced a 3.5-fold increase in the apparentKm and a 2.2-fold increase in the uncompetitive substrate inhibition constantKs (Table3 and Fig.5, A and B), without a significant effect on the Vmax value. Calculation ofKm values based on free drug concentrations using estimated free fractions yielded unboundKm estimates of 8.3 and 9.2 μM, without and with addition of inactive control microsomes, respectively.
Kinetic parameters for amitriptyline N-demethylation by lymphoblast-expressed CYP2C19 and CYP3A4 without and with the addition of control microsomes from untransfected lymphoblast cells
Effect of nonspecific binding on the kinetics of amitriptyline (AMI) N-demethylation by lymphoblast-expressed CYP2C19 (A and B) and CYP3A4 (C and D). A and C, primary plots. B and D, Eadie-Hofstee transformations. Dots are experimental data points and lines are fitted functions (eq. 1 for CYP2C19 and eq. 2 for CYP3A4). Filled dots and lines represent the control experiment (with minimal nonspecific binding). Open dots and the dashed line represent kinetics in the presence of 750 μg/ml inactive control microsomes from untransfected B-lymphoblastoid cells. See Table 3 for kinetic parameters.
Lymphoblast-expressed CYP3A4 displayed Hill enzyme kinetics (eq. 2). Addition of 750 μg/ml control microsomes from untransfected cells produced a 1.8-fold increase in the apparent S50without affecting the Vmax or Hill coefficient A to any appreciable extent (Table 3 and Fig. 5, C and D). Calculation of S50 values based on free drug concentrations using estimated free fractions yielded unbound S50 estimates of 69 and 39 μM, without and with addition of inactive control microsomes, respectively. The reason for the difference in unbound S50 values estimated at the two different microsomal protein concentrations is not clear.
Figure 6 shows the effect of microsomal binding on the rate of amitriptyline N-demethylation by CYP2C19 and CYP3A4. As with human liver microsomes, the effect was maximal at low substrate concentrations and was overcome by increasing concentrations of amitriptyline.
Decrement in the rate of amitriptyline (AMI)N-demethylation by CYP2C19 and CYP3A4 by addition of 750 μg/ml control microsomes from untransfected B-lymphoblastoid cells. Data are presented as the fraction of control rates measured in the absence of inactive control microsomes. Note that the effect of nonspecific binding is overcome at high substrate concentrations.
Discussion
Amitriptyline is bound to human liver microsomes and microsomes from human B-lymphoblastoid cells. This binding results in an overestimation of Km and S50, without affecting the metabolic rate at saturating substrate concentrations.
Microsomal binding of amitriptyline is consistent with its high lipophilicity and plasma protein binding—mean free fraction in plasma 0.057 (Schulz et al., 1985). The amitriptyline free fraction was drug concentration-independent over the range of amitriptyline concentrations generally used in vitro (Fig. 1, A and B). Concentration-independent human liver microsomal binding of propranolol, imipramine (Obach, 1997), and phenytoin (Carlile et al., 1999) has also been described, although the free fraction of warfarin in human liver microsomes increased with increasing drug concentration (Obach, 1997).
At the highest total concentration of amitriptyline studied (approximately 200 μM), the bound concentrations of amitriptyline were 70 and 20 μM in human liver microsomes (200 μg/ml protein) and lymphoblast microsomes (250 μg/ml protein), respectively. The molar concentration of binding sites and the affinity constant for binding could not be determined due to the essentially linear relationship between free and bound amitriptyline concentrations over the concentration range studied. However, because the molar concentration of binding sites should in principle be greater than or equal to the bound drug concentrations (Wright et al., 1996), the measured bound drug concentrations suggest that human liver microsomes and human lymphoblastoid cell microsomes contain at least 350 and 80 nmol of binding sites for amitriptyline/mg of microsomal protein, respectively. These concentrations are far in excess of the average molar concentration of total CYP in human liver microsomes, 340 pmol/mg protein (Shimada et al., 1994), and in these lymphoblastoid cell microsomal preparations. This suggests that the observed binding is in fact mainly nonspecific and not reflective of specific interactions with the enzyme active site.
The amitriptyline free fraction progressively decreased with increasing microsomal protein concentration, in both human liver microsomes and microsomes from B-lymphoblastoid cells (Table 1). It is thus likely that microsomal binding may contribute in part to the nonlinear increase in amitriptyline N-demethylation rate with increasing microsomal protein concentrations at protein concentrations of more than 200 μg/ml (data not shown). Even at a relatively low human liver microsomal protein concentration of 200 μg/ml (the upper limit of the linear range), 40% of the drug is bound to the microsomal matrix.
Consistent with the binding studies, addition of heat-inactivated human liver microsomal protein caused a protein concentration-dependent decrement in reaction rate. The effect became pronounced at low amitriptyline concentrations (Fig. 4), resulting in an increase in apparent Km and S50 of the high- and low-affinity components of amitriptyline N-demethylation, without a consistent effect on the net Vmax (Figs. 2 and 3 and Table 2). Although heat-inactivated microsomes bound amitriptyline to a greater extent than native microsomes (Table 1), the binding was still comparable, thereby allowing their use as a source of nonspecific binding sites devoid of functional enzyme. UnboundKm and S50values were calculated for each total microsomal protein concentration, using the respective apparent Km and S50 values, and amitriptyline free fractions (Table 2). These unbound Km and S50 values are similar at all four protein concentrations. Therefore, correction for binding reduces much of the variability in Km and S50 attributable to microsomal protein concentration.
Eadie-Hofstee transformations of data presented in Fig. 2. The scaling of the V/S axis has been fixed in all panels to illustrate the impact of nonspecific binding on the apparent affinity.
As with human liver microsomes, addition of control lymphoblast microsomes to CYP2C19 or CYP3A4 resulted in an increase in apparentKm or S50, respectively, without any appreciable effect on theVmax (Table 3 and Fig. 5). UnboundKm values were calculated for CYP2C19 using the apparent Km values and amitriptyline free fractions. The unboundKm values determined without and with addition of inactive microsomes were similar (8.3 and 9.2 μM, respectively), suggesting that the alteration of apparentKm is explained by microsomal binding of amitriptyline. However, the unbound S50 values for CYP3A4 determined without and with addition of inactive control microsomes were not identical (69 and 39 μM, respectively). The reason for this difference in unbound S50 values determined at two different microsomal protein concentrations is not clear and may be related to the increased complexity due to cooperative binding. Exogenously added albumin, for example, decreased the unboundKm of human liver microsomal phenytoin hydroxylation (Ludden et al., 1997). Thus, addition of exogenous protein may in fact alter the true affinity of the enzyme for substrate, affecting the unbound Kmvalue. However, it is likely that addition of albumin to an in vitro reaction mixture is not comparable to addition of microsomal protein. In any case, the estimation of Kmvalues based on unbound drug concentrations relies on the assumption that the intrinsic affinity of enzyme for unbound substrate is independent of microsomal protein concentration and that free drug concentrations, rather than total concentrations, are better estimates of enzyme-available concentrations in vitro. Although the validity of this assumption remains unclear, correction for microsomal binding using the free fraction in incubation matrices clearly improves the prediction of in vivo clearance from in vitro estimates of intrinsic clearance for drugs that are extensively bound to microsomal matrices (Obach, 1996, 1997, 1999; Obach et al., 1997; Carlile et al., 1999). Thus, unbound Km values based on free drug concentrations rather than apparentKm values based on total drug concentrations are expected to be better estimates of the trueKm based on the existing data.
In addition to its effects on apparentKm and S50, microsomal binding also increased the estimated substrate inhibition constant Ks for CYP2C19-mediated amitriptyline N-demethylation, with negligible substrate inhibition in the presence of 750 μg/ml control (inactive) microsomes (Fig. 5A). Reaction rates in the presence of binding protein at 350 and 500 μM concentrations of amitriptyline were higher than control rates, whereas no such increase in rate was seen for CYP3A4 at saturating concentrations of substrate (Fig. 6). Although microsomal binding should in principle affect only the apparentKm or S50 and not alter the rates at saturating substrate concentrations, this may be true only if the kinetics is consistent with monotonic Michaelis-Menten or Hill functions.
Lymphoblast-expressed CYP isoforms are generally used at microsomal protein concentrations of 250 to 1000 μg/ml in the determination of enzyme kinetic parameters. These parameters are subsequently used to predict relative contributions of individual CYP isoforms to the overall rate of hepatic drug biotransformation in vivo, accounting for the relative abundance of each CYP isoform in human liver. Due to the differing turnover numbers of the CYP isoforms that catalyze a given drug biotransformation pathway and the relatively similar CYP content (on a per-milligram of protein basis) of the lymphoblast-expressed CYP preparations, it may be necessary to use different total microsomal protein concentrations for different isoforms. Thus, low microsomal protein concentrations (100–250 μg/ml) may be needed to minimize substrate consumption by CYP isoforms with a high intrinsic clearance, whereas higher protein concentrations (500–1000 μg/ml) may be required to identify and kinetically characterize low-affinity and low-capacity isoforms, due to limits of analytical assay sensitivity. For substrates extensively bound to microsomes, theKm or S50 values determined for the various CYP isoforms may be biased by the microsomal protein concentration used, resulting in misprediction of the relative contributions of the various CYP isoforms to net intrinsic clearance. For example, nortriptyline is biotransformed viaE-10-hydroxylation, a reaction that is catalyzed in vitro by both CYP2D6 (high affinity, high capacity) and CYP3A4 (low affinity, low capacity) (Venkatakrishnan et al., 1999). Kinetic characterization of CYP3A4 required the use of a microsomal protein concentration of 1000 μg/ml, whereas a protein concentration of 200 μg/ml had to be used for CYP2D6 to minimize substrate consumption. Given the structural similarity with amitriptyline, nortriptyline binding to microsomes may be significant. Thus, the Km value for CYP3A4 in relation to CYP2D6 may have been overpredicted.
In addition to its effects on in vitro drug biotransformation kinetics, microsomal binding of inhibitors can theoretically bias the estimation of in vitro IC50 orKi values in chemical inhibition studies, potentially causing underestimation of inhibitor potency. In fact, a microsomal protein concentration-dependent increase in apparent IC50 values of clotrimazole and ketoconazole versus human liver microsomal midazolam 1′-hydroxylation has been described, although this effect was attributed to inhibitor depletion by specific binding to the enzyme (Gibbs et al., 1999).
Although the physicochemical determinants of microsomal binding are not completely understood, significant binding has been described for lipophilic basic drugs like desipramine, imipramine, amitriptyline, chlorpromazine, and propranolol (Obach, 1997, 1999), whereas warfarin (Obach, 1997), tolbutamide (Carlile et al., 1999; Obach, 1999), and dextromethorphan (Witherow and Houston, 1999) are not significantly bound to human liver microsomes.
Microsomal binding should be measured and incorporated in enzyme kinetic analyses so unbiased kinetic parameters can be determined, based on unbound rather than added (total) drug concentrations. This phenomenon may explain in part the laboratory-to-laboratory variation in Km values reported for a given drug biotransformation pathway, which in turn may lead to incorrect predictions of the relative contributions of high- and low-affinity CYP isoforms to the overall metabolic rate, when varying microsomal concentrations of the heterologously expressed CYP isoforms are used in kinetic analyses.
Acknowledgments
We would like to thank Bart E. Laurijssens for assistance with development of the equilibrium dialysis method.
Footnotes
-
Send reprint requests to: David J. Greenblatt, M.D., Department of Pharmacology and Experimental Therapeutics, Tufts University School of Medicine, 136 Harrison Ave., Boston, MA 02111. E-mail: Dj.Greenblatt{at}tufts.edu
-
↵1 This work was supported by Grants MH-34223, DA-05258, MH-19924, and RR-00054 from the Department of Health and Human Services. L.L.v.M. is the recipient of a Scientist Development Award (K21-MH-01237) from the National Institute of Mental Health, National Institutes of Health.
- Received October 5, 1999.
- Accepted January 6, 2000.
- The American Society for Pharmacology and Experimental Therapeutics
