Kinetics of the time-dependent inactivation of CYP2D6 in cryopreserved human hepatocytes by methylenedioxymethamphetamine (MDMA)

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 1 ( 2 0 0 7 ) 53–61

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/ejps

Kinetics of the time-dependent inactivation of CYP2D6 in cryopreserved human hepatocytes by methylenedioxymethamphetamine (MDMA) Linh M. Van a , John Swales b , Clare Hammond b , Claire Wilson b , Judith A. Hargreaves b , Amin Rostami-Hodjegan a,c,∗ a b c

University of Sheffield, Academic Unit of Clinical Pharmacology, School of Medicine, Sheffield S10 2JF, UK AstraZeneca, DMPK, Alderley Park, Macclesfield, Cheshire SK10 4TG, UK Simcyp Limited, Blades Enterprise Centre, Sheffield S2 4SU, UK

a r t i c l e

i n f o

a b s t r a c t

Article history:

Methylenedioxymethamphetamine (MDMA) was investigated in cryopreserved human hep-

Received 8 November 2006

atocytes as a time-dependent inactivator (TDI) of CYP2D6 using dextromethorphan (DEX) as

Received in revised form

a probe substrate. Inhibition kinetic parameters kinact , the maximal rate of inactivation, and

4 February 2007

KI , the inhibitor concentration at half the maximal activation rate, were determined. Time-

Accepted 15 February 2007

and concentration-dependent inhibition were confirmed, and the influence of different

Published on line 20 February 2007

elements of study design (e.g. cell number, stability of hepatocytes, dilution after preincubation) on estimated kinetic parameters were evaluated. Dilution factors (DF) of 1.2, 5 or total

Keywords:

removal of inhibitor (by washing cells after preincubation, WR) resulted in kinact and KI (±S.E.)

Metabolic drug–drug interactions

values of 0.02 ± 0.002 min−1 and 0.88 ± 0.31 ␮M, 0.01 ± 0.001 min−1 and 1.23 ± 0.70 ␮M, and

Mechanism-based inhibition

0.01 ± 0.001 min−1 and 2.10 ± 1.32 ␮M, respectively; indicating that insufficient dilution may

CYP2D6 inhibition

lead to overestimation of CYP2D6 inactivation. Accounting for MDMA depletion during the

MDMA

preincubation, corrected KI values were significantly lower (0.11 ± 0.05 ␮M, 0.15 ± 0.09 ␮M, 0.24 ± 0.16 ␮M for DF of 1.2, 5, and WR, respectively). Inactivation efficiency in hepatocytes, as measured by kinact /KI , was 10-fold less than that previously reported in human liver microsomes or recombinantly expressed systems. Possible causes for the observed differences between in vitro systems warrant further investigation. These may include differences in metabolic consumption of MDMA in each system, non-specific binding and presence of active efflux in hepatocytes. © 2007 Published by Elsevier B.V.

∗ Corresponding author at: Academic Unit of Clinical Pharmacology, University of Sheffield, Floor M, Royal Hallamshire Hospital, Sheffield S10 2JF, UK. Tel.: +44 114 271 2156; fax: +44 114 226 8986. E-mail address: A.Rostami@sheffield.ac.uk (A. Rostami-Hodjegan). Abbreviations: CYP2D6, cytochrome P450 2D6; DEX, dextromethorphan; DF, dilution factor; DOR, dextrorphan; HLM, human liver microsome; kinact , maximum inactivation rate; KI , apparent inactivation constant; kobs , observed rate; LC–MS/MS, tandem mass spectrometry; LEV, levallorphan; MBI, mechanism-based inhibition; MDMA, 3,4-methylenedioxymethamphetamine; MRM, multiple reaction monitoring; rCYP, recombinant CYP450; TDI, time-dependent inactivation; WR, wash and removal of inhibitor after centrifugation 0928-0987/$ – see front matter © 2007 Published by Elsevier B.V. doi:10.1016/j.ejps.2007.02.005

54

1.

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 1 ( 2 0 0 7 ) 53–61

Introduction

Mechanism-based inhibition (MBI) is associated with timedependent, irreversible or quasi-irreversible loss of enzyme activity, requiring de novo synthesis to replenish the enzyme. This quasi-irreversible or irreversible inhibition is considered more serious than reversible inhibition, since the inhibitory effect remains after elimination of the parent drug from the body (Ito et al., 1998; Silverman, 1998). MBI (also known as time-dependent inactivation) is an unusual occurrence with most enzymes, but is observed at a higher frequency in CYP450-catalyzed reactions due to the reactivity of the oxygenation reactions (Hollenberg, 2002). Human liver microsomes (HLM) and cDNA-expressed recombinant CYP450 (rCYP) enzymes are established model systems for characterizing and obtaining the kinetic parameters defining time-dependent inactivation (TDI), namely kinact (the maximum inactivation rate) and KI (the apparent inactivation constants). These kinetic parameters are used to extrapolate the extent and duration of enzyme inactivation in vivo (Mayhew et al., 2000; Van et al., 2006). Only limited studies have utilized hepatocytes as an alternate model for defining kinetic estimates for TDI. Zhao et al. (2005) evaluated TDI of CYP3A in cryopreserved hepatocytes for six therapeutic compounds. However, kinact and KI were not determined. TDI has also been examined in another recent study in cultured primary human hepatocytes (McGinnity et al., 2006) where full kinetic parameters of TDI were determined. Hepatocytes are considered to be more representative of the liver functionality than liver slices or cell free systems such as S9, microsomal and cytosolic fractions by some researchers (McGinnity et al., 2004; Takashima et al., 2005). The main reason for the latter view has been the fact that hepatocytes contain a full complement of enzymes (transporters, phase I/II enzymes) and cofactors at physiological levels in their appropriate natural orientations. However, as Rowland et al., 2007 have recently indicated, inhibitory fatty acids released during incubation with hepatocytes may alter kinetic values for drug metabolism, at least in the case of substrates undergoing glucuronidation. Therefore, extrapolating the results of in vitro studies on drug metabolism and metabolic drug interactions from HLM or rCYP to hepatocytes, and ultimately to in vivo studies, may not be straight forward. Following on from previous studies on TDI of CYP2D6 by MDMA using HLM and rCYP (Heydari et al., 2004; Van et al., 2006), investigations were undertaken to determine if the kinetic estimates of TDI can also be obtained in hepatocytes. MDMA is a ring-substituted amphetamine, structurally similar to methamphetamine and mescaline, possessing entactogenic properties (Nichols, 1986). In vitro studies have reported that MDMA is a potent competitive inhibitor of CYP2D6 in HLM and that the inhibitory effect of MDMA increased following preincubation (Wu et al., 1997) due to the opening of the methylenedioxyphenyl ring by CYP2D6 where the methylenedioxyphenyl substituents forms a metabolite intermediate complex with CYP2D6 (Heydari et al., 2004). The kinetic values describing TDI reactions for MDMA in HLM and rCYP have been derived and shown to be similar albeit

using limited number of HLM samples (Heydari et al., 2004). The current report extends the investigations to cryopreserved hepatocytes where kinetic estimates (kinact and KI ) of TDI by MDMA are defined. Previous work using yeast-expressed CYP2D6 enzymes and MDMA has demonstrated the impact of experimental conditions on derived TDI kinetic values (Van et al., 2006). Thus, the effect of study design on TDI estimates when using cryopreserved hepatocytes are also investigated and reported here. The importance of variation in estimated kinetic parameter values on in vitro–in vivo extrapolation has been emphasised in our previous report (Van et al., 2006).

2.

Materials and methods

2.1.

Materials

MDMA hydrochloride, dextromethorphan hydrobromide (DEX), dextrorphan tartrate (DOR), quinidine, and levallorphan (LEV, assay internal standard) were purchased from Sigma–Aldrich (Poole, Dorset, UK). Leibovitz (L15) medium supplemented with glutamate was purchased from Invitrogen (Loughborough, UK). Ammonium acetate, acetronitrile and methanol were obtained from Fisher Scientific Co. (Loughborough, UK). All reagents were of high-performance liquid chromatography (HPLC) grade. Cryopreserved hepatocytes (female individual) were supplied from BD Gentest (Oxford, UK).

2.2.

Preparation of cryopreserved hepatocytes

Protocol adapted from BD Gentest (Oxford, UK). Briefly, hepatocytes previously stored under liquid nitrogen were quickly thawed in a 37 ◦ C water bath and rinsed with L15 medium. The suspension was then centrifuged for 3 min (50 × g) and the supernatant decanted followed by resuspension of the pellet in fresh L15 medium. Cell viability was measured using trypan blue exclusion method. Cell viabilities were between 80 and 90%. Cells were allowed to recover for 15 min in 37 ◦ C incubator in a 95% air/5% CO2 atmosphere before conducting experiments.

2.3.

General experimental conditions

All reactions were done in duplicates at 37 ◦ C in an incubator containing 95%/5% CO2 atmosphere. Reactions terminated by quenching with equal volume of acetronitrile (containing 0.1 ␮g/␮L of LEV). Samples were centrifuged at 14,000 rpm for 10 min and the supernatant diluted with equal volume of HPLC-grade water for LC–MS/MS analysis.

2.4. Linearity of dextromethorphan metabolism with respect to cell number, and hepatocytes stability in the presence of MDMA or quinidine To investigate the linearity of dextromethorphan metabolism, different volumes of the cell stock (1.0 × 106 cells/mL) were co-incubated for 60 min with either MDMA (0, 5, 50 ␮M) or quinidine (0, 0.5, 5 ␮M) in the presence of DEX (100 ␮M) in a

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 1 ( 2 0 0 7 ) 53–61

reaction volume of 200 ␮L made up with L15 medium to obtain cell counts as follows: 1.25 × 105 cells/mL, 2.5 × 105 cells/mL and 5.0 × 105 cells/mL. To test the relative stability of the hepatocytes, cells were left out at room temperature for 1 h before adding substrate/inhibitors as above and repeating experiments as described before, using 2.5 × 105 cells/mL as the final cell number in a total reaction volume of 200 ␮L.

2.5.

Effect of inhibitor concentrations on cell viability

Cell viability using trypan blue exclusion method was carried out to explore for the cytotoxic effects of the inhibitors. Hepatocytes were resupsended in L15 medium containing MDMA (0, 2, 5, 10, 20, 50 ␮M) or quinidine (0, 1, 2, 5, 10 ␮M) in a total reaction volume of 200 ␮L and a resulting cell count of 2.5 × 105 cells/mL. After 15, 30 and 60 min, 10 ␮L aliquot was mixed with an equivalent volume of trypan blue dye (0.2%, w/v) and cells counted under × 10 objective.

2.6. Effect of preincubation of inhibitor on dextrorphan formation rate Cells (1.25 × 105 ) were co-incubated for either 30 or 60 min with MDMA (0, 2, 5, 10, 20 and 50 ␮M) and DEX (100 ␮M) in a total reaction of 200 ␮L yielding cell count of 6.25 × 105 cells/mL. For preincubation experiments, hepatocytes (1.25 × 105 cells) were incubated for 60 min with MDMA (0, 2, 5, 10, 20 and 50 ␮M) in total volume of 100 ␮L then, 100 ␮L of DEX (200 ␮M) was added, resulting in a cell number of 6.25 × 105 cells/mL and final DEX concentration of 100 ␮M. Reactions continued for a further 30 min before being terminated. This experiment was repeated using quinidine (0, 1, 2, 5 and 10 ␮M), a competitive reversible inhibitor, acting as a negative control.

2.7.

Metabolism of MDMA

To monitor the metabolism of MDMA by hepatocytes, cells (1.25 × 105 ) were preincubated with MDMA (0, 2, 5, 10, 20 and 50 ␮M) for 60 min in a total reaction volume of 100 ␮L, leading to a cell count of 1.25 × 106 cells/mL. 20 ␮L of DEX (600 ␮M) was added and the reaction continued for 30 min. MDMA concentrations were measured at the end of each experiment and compared with those of control tubes. Control tubes contained no active enzyme as the metabolic reaction had been quenched with acetronitrile prior to addition of the MDMA (similar studies were also carried out for quinidine (0, 1, 2, 5 and 10 ␮M)).

2.8. Time-dependent inhibition of cryopreserved hepatocytes Two different protocols were utilized to carry out TDI experiments, one comprising of a dilution step after preincubation and the other adapted from Zhao et al. (2005) which involved a washing and centrifugation process to remove the inhibitor (WR). Irrespective of the methodology undertaken, the final cell count and DEX concentration in the incubation well after the preincubation stage was 5.0 × 105 cells/mL and 100 ␮M.

55

Reactions proceeded for an additional 30 min before termination by acetronitrile (containing 0.1 ␮g/␮L of LEV). Briefly, for total removal of inhibitor (‘Wash and Removal’ Method): 100 ␮L of the cell stock (5.0 × 105 cells/mL) was added into L15 medium containing MDMA (0, 2, 5, 10, 20 and 50) or quinidine (0, 0.5, 1, 2, 5 and 10) resulting in total volume of 200 ␮L and cell counts of 2.5 × 105 cells/mL. Cells were preincubated with the inhibitor for 0, 30 or 60 min then centrifuged at 50 × g for 5 min at room temperature. The pellet was then washed with fresh L15 medium (150 ␮L) and spun again at 50 × g for 5 min. The supernatant was decanted and the post wash pellet resuspended with 100 ␮L of L15 containing DEX. In the ‘Dilution Method’, 100 ␮L of cell stock (5.0 × 106 cells/mL) was added to L15 medium containing MDMA (0, 2, 5, 10, 20 and 50 ␮M) or quinidine (0, 1, 2, 5 and 10 ␮M) to a total volume of 200 ␮L yielding 2.5 × 106 cells/mL. After 0, 15, 30 or 60 min, 20 ␮L of the preincubation mixture (five-fold dilution) was transferred to an incubation well containing 80 ␮L of fresh L15 containing DEX. Direct addition of the probe substrate in a small volume into the preincubation mixture led to essentially no dilution. Briefly 50 ␮L of cell stock (1.0 × 106 cells/mL) was transferred to L15 medium containing MDMA (0, 2, 5, 10, 20 and 50) or quinidine (0, 1, 2, 5 and 10) resulting in a total volume of 100 ␮L and cell count of 5.0 × 105 cells/mL. At 0, 15, 30 and 60 min, 20 ␮L of DEX (600 ␮M) was directly added to preincubation well leading to total volume of 120 ␮L and DEX concentration of 100 ␮M.

2.9. Liquid chromatography–tandem mass spectrometry (LC–MS/MS) Samples were analysed by LC–MS/MS using Micromass® Quattro UltimaTM Pt (Waters Ltd., Hertsfordshire, UK) equipped with an HP1100 LC series (Agilent Technologies Inc., Germany). Separation of compounds were performed using SynergiTM maxRP, 50 mm × 2.0 mm, 5 ␮ column (Phenomenex, Macclesfield, UK) protected by a security guard column (Phenomenex, Macclesfield, UK). The mobile phase consisted of solvent A (10 mM ammonium acetate in HPLC-grade water) and solvent B (10 mM ammonium acetate in methanol), using a gradient of 5–95% of B (0–3 min), 95% B (3.1–4.0 min), 5% B (4.01 min), with a run time of 4.5 min and a flow rate of 0.75 mL/min. Multiple reaction monitoring (MRM) in positive-ion mode was used for all analytes, using electrospray ionization with the capillary voltage set at 0.78 kV. System control, data acquisition and the analytical parameters including the selection of ions for each compound were performed by the application software, MassLynx 4.1 (Waters Ltd., Hertfordshire, UK). The mass-to-charge transition (m/z) of precursor ions and product ions for each compound was identified as follows: m/z 272.5 → 171.1 for DEX, m/z 258.2 → 201.2 for DOR, m/z 194.2 → 105.1 for MDMA, m/z 325.4 → 184.2 for quinidine, m/z 284.4 → 201.2 for LEV. Standard curves using quality control standards (n = 5) were analysed with each batch of study samples, with the overall accuracy (percent relative error) of the method being lower than 20% and overall precision (percent coefficient of variation) being lower than 10%.

56

2.10.

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 1 ( 2 0 0 7 ) 53–61

of 0.002 ␮M (a signal-to-noise ratio of 3) was determined:

Data analysis

For all LC–MS/MS analyses, the peak area of the analyte was expressed as a ratio to the internal standard (LEV). Determination of the concentration of substrate, metabolite and inhibitors during the incubation was interpolated from their respective calibration curve. To determine the inactivation constants that defines time-dependent inhibition, the relative inhibition of DOR formation or active enzyme (CYP2D6) remaining was expressed as a percentage of the average data (duplicate samples) of the time-matched control samples without inhibitor. This procedure takes into account any loss of enzyme activity unrelated to the inhibitor. The natural logarithm of the mean relative inhibition values was plotted versus preincubation time for each concentration of inhibitor used. Following visual inspection of inactivation data (LN% activity compared to control) with preincubation time, the initial slope (kobs ) and its variance were determined using “regression analysis” within Excel® . Initially, the parameters kinact and KI were obtained from non-linear fitting of kobs (weighted inversely by variance) against the inhibitor concentration ([I]) (GraFit® ; Erithacus Software Ltd) according to Eq. (1) assuming the [I] remained constant: kobs =

kinact × [I] KI + [I]

(1)

The weighted averages of kinact and KI (± standard error, S.E.) for the different dilutions were compared using Z test implemented in Excel where a p-value < 0.05 was considered statistically significant. Subsequently, correction for depletion of MDMA in hepatocyte medium ([I]corr ) was considered by calculating an average exposure (using Area Under the Curve, AUC) between the initial time (t = 0) to 60 min preincubation (Eq. (2), see Appendix A for derivation). The kinetic parameters (kinact and KI ) were then recalculated as described previously according to Eq. (1) using [I]corr instead of the initial [I]. In cases where inhibitor was completely depleted from the medium, a conservative value of average exposure based on a limit of quantification

[I]corr =

[I]t=0 − [I]t=60 AUC = time Ln[I]t=0 − Ln[I]t=60

3.

Results

3.1.

Cell number and viability

(2)

DOR formation rate increased linearly with cell number (Fig. 1) however the percent inhibition of dextromethorphan O-demethylation by MDMA and quinidine remained relatively similar as shown in Fig. 2. The metabolic activity of hepatocytes, as measured by formation of DOR, was stable over the period of 1 h used for the experiments (97.6% activity remaining, leading to an estimated CYP2D6 degradation rate (kdeg ) of 0.024 h−1 (half life of around 28 h assuming first order decay) data not shown). Various concentrations of MDMA and quinidine in this study did not contribute to cell loss as measured by trypan blue exclusion method. Regardless of the inhibitor concentrations, the number of live cells at different time points was between 70 and 80% (ANOVA p > 0.577, data not shown).

3.2. The effect of preincubation time on the activity of CYP2D6 as measured by the rate of dextrorphan formation Inhibition of dextromethorphan demethylation activity in hepatocytes by both MDMA and quinidine was concentration dependent (Fig. 3). However, 60 min preincubation of MDMA with hepatocytes before substrate addition led to greater inhibition effects compared to controls without preincubation (Fig. 3a). In contrast, preincubation of quinidine was not associated with any increased inhibition (Fig. 3b). Coincubation experiments (30 and 60 min) at saturating substrate concentration (100 ␮M DEX) did not show any additional loss of enzyme activity with time in the presence of either MDMA or quinidine (Fig. 3c and d).

Fig. 1 – The effect of different cell density on the dextrorphan formation rate (pmol/min) for MDMA (a) and quinidine (b). Values represent the mean ± S.D. of duplicates.

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 1 ( 2 0 0 7 ) 53–61

57

Fig. 2 – The effect of different cell density on the inhibition level (expressed as a percentage of the control without inhibitor) for MDMA (a) and quinidine (b). Values represent the mean ± S.D. of duplicates.

3.3.

Inhibitor level during 60 min preincubation

The metabolic turnover of MDMA (as measured by the disappearance of the parent compound) was high, with total consumption of the inhibitor at low concentrations, reducing to approximately 50% over 60 min for the highest concentration. Quinidine, particularly at higher concentrations, remained relatively stable (>70% remaining after 60 min) (Fig. 4). Measurement of the inhibitor levels during the 60 min

coincubation with DEX (100 ␮M) showed greater than 70% remaining at the highest concentration.

3.4.

TDI of MDMA in cryopreserved hepatocytes

Time-dependent inhibition experiments were done to obtain estimates for the kinetic parameters, kinact and KI , based on either dilution (DF = 1.2 or 5) or total removal (‘WR’ method) of the inhibitor. Fig. 5 represents the results of nonlinear regres-

Fig. 3 – Effect of preincubation and different coincubation times on the inhibition of dextromethorphan O-demethylation by MDMA (a and c) and quinidine (b and d). Values are expressed as the percentage of the control without inhibitor. Each point is the mean ± R.S.D. of duplicates.

58

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 1 ( 2 0 0 7 ) 53–61

Fig. 4 – % MDMA (a) and quinidine (b) remaining in the hepatocyte medium after 60 min preincubation or 60 min coincubation. Each point is the mean ± R.S.D. of duplicates.

sion fit to kobs versus MDMA concentrations to determine kinact and KI for DF = 1.2 or 5 and ‘WR’. Lower dilution factor (DF = 1.2) caused the highest rates of inactivation whereas increased dilution and removal of MDMA by washing the cells after the

preincubation period showed similar albeit lower rate of inactivation. Table 1 summarises the kinetic parameters for CYP2D6 inactivation by MDMA in cryopreserved human hepatocytes, using either the initial inhibitor concentration (MDMAini ) or the corrected level of MDMA (MDMAcorr ) remaining in the medium after 60 min preincubation. Reported kinetic parameters from other in vitro systems are also shown for comparison.

4.

Fig. 5 – Plot of the inactivation rate (kobs ) vs. MDMA concentration in hepatocytes diluted 1.2 (square) or five-fold (triangle) or by ‘wash and removal’ (WR, diamond) method. Resulting cell count in the incubation after preincubation for all three methods was 5.0 × 105 cells/mL in the presence of 100 ␮M DEX. Close symbols (a) refer to TDI kinetic estimates derived using the initial inhibitor concentration (MDMAini ) whereas the open symbols (b) account for the depletion of MDMA (MDMAcorr ) occurring during the preincubation. Values represent mean ± S.E. of duplicates. The inset figure expanded the image from 0 to 2 ␮M of MDMA.

Discussion

This study has confirmed that MDMA is a TDI of CYP2D6 using human hepatocytes. Previous studies have reported TDI of CYP2D6 by MDMA using HLM and rCYP system (Wu et al., 1997; Heydari et al., 2004; Van et al., 2006). The inactivation effect of MDMA on CYP2D6 in hepatocytes followed the typical characteristics expected from a TDI (Silverman, 1998) thus preincubation of MDMA increased the inhibitory potency while this was not evident for quinidine (a competitive, reversible inhibitor). Coincubation of MDMA in the presence of saturating level of DEX reduced the inhibitory effect, indicating that the inhibitor must be metabolised in order to cause TDI. The time- and -concentration dependent inhibition effect of MDMA allowed us to obtain the kinetic parameters for MBI in human hepatocytes and to investigate the impact of experimental design. For example, kinact increased and MDMA became apparently more potent (decreased in KI ) when dilution of the inhibitor after preincubation was lower. This can be partially explained by both ongoing inactivation and competition during the incubation period due to insufficient dilution of the inhibitor. Although the coincubation study suggested inactivation was prevented in the presence of saturating DEX, preincubation of MDMA followed by dilution may not lead to a similar level of protection against inactivation. This follows since MDMA is present within the cells (once the cells are in the incubation mixture) while the added DEX has to cross the membrane before arriving at the site of competition with MDMA. This may explain why different dilutions

Data for HLM comes from Heydari et al. (2004) using DEX concentration of 20 ␮M. Yeast-expressed CYP2D6 enzyme, data derived from Van et al. (2006) using DEX concentration of 30 ␮M. ∗ Statistical significance based on p-value < 0.05 using Z-test when comparing the different in vitro systems. ∗∗ Statistical significance of the experimental conditions (e.g. DF vs. WR) within each systems on kinetic estimations. b

a

Values for hepatocytes represent mean (±S.E.) of duplicates using 100 ␮M of DEX and 5.0 × 105 cell/mL as the final cell count in the incubation. DF stands for dilution factor and WR refers to hepatocytes that have been ‘washed’ and inhibitor ‘removed’ before DEX addition. [MDMA]ini refers to the initial concentration of MDMA (assumed to be constant during the experiment) whereas [MDMA]corr is the correction of the level of MDMA actually present during the preincubation.

0.31 (±0.06) 5.78 (±2.07) 0.054 0.21** (±0.06) 2.22** (±1.90) 0.094 0.15 (±0.01) 8.80 (±2.60) 0.017 0.008 (±0.001) 0.235 (±0.160) 0.034 0.007 (±0.001) 0.153 (±0.087) 0.046 0.018** (±0.002) 0.111 (±0.050) 0.162 0.009 (±0.001) 2.102 (±1.315) 0.004 0.008 (±0.001) 1.230 (±0.678) 0.006

WR [MDMA]ini

DF = 5 DF = 1.2

0.019** (±0.002) 0.878 (±0.314) 0.022 kinact (min−1 ) KI (mL/nmol/min) kinact /KI (mL/nmol min)

DF = 5 DF = 1.25 DF = 5 DF = 1.2

[MDMA]corr

WR

DF = 4

[MDMA]ini [MDMA]ini

HLMa,* Hepatocytes*

Table 1 – Comparison of the kinetic estimates for TDI in hepatocytes, HLM and yeast-expressed CYP2D6 recombinant system (rCYP)

rCYPb,*

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 1 ( 2 0 0 7 ) 53–61

59

(DF = 1.2 versus 5) of the preincubation lead to different TDI kinetic values. This effect could not be associated with different cell numbers in the incubation mixture as this was investigated and shown to have no significant effect (Fig. 1). When the inhibitor was removed by washing the cells, the inactivation efficiency was similar to that of a DF = 5. Although the methodology of removing the inhibitor from the incubation media as suggested by Zhao et al. (2005), is expected to provide the most accurate results, it may not prevent ongoing inactivation during the washing and centrifugation of the cells. Moreover, centrifugation may alter the concentration equilibrium and hence intracellular concentrations of the inhibitor. Since the five-fold dilution methodology produced similar kinetic parameters for TDI as the “WR” methodology, and because the washing techniques are more labour intensive than the dilution method, the former might be considered an optimal protocol on practical grounds. Moreover, the dilution method is compatible with automated high-throughput screening which makes it more appealing in drug discovery and early stages of drug development. The observed relationship between kinact and dilution in this study is contradictory to our previous findings using recombinant CYP2D6 (Van et al., 2006). In a previous report we indicated that correction for varying non-specific binding of MDMA under different dilutions could not account for the observed effects of dilution on kinact . However, using hepatocytes, the observation that lower dilution leads to higher kinact can be explained by ongoing inactivation and competition during the incubation stage; thus inactivation apparens to become more efficacious. MDMA is a potent competitor of CYP2D6 (Wu et al., 1997), thus formation rate of DOR (measure of active enzyme remaining) is likely to be less with insufficient dilution; resulting in an apparently higher rate of inactivation. The dilution did not effect different experimental protocols equally since the loss of MDMA during preincubation was concentration dependent (see Fig. 4(a) MDMA). Metabolic consumption of MDMA during the 1 h preincubation in hepatocytes will decrease the availability of inactivator to the enzyme. Other enzymes beside CYP2D6, or the presence of phase II enzyme not found in HLM, will also contribute to reducing the level of MDMA available in the cytosol. These may lead to minimizing the effect of inactivation (Ito et al., 1998). Non-specific binding of MDMA to cellular constitutents in hepatocytes will also decrease the level of MDMA available for interaction with CYP2D6 (Obach, 1997; Gibbs et al., 1999). Involvement of transporters will alter the concentration level of MDMA in the cells (Yamazaki et al., 1996). All of these contributing factors will lead to a greater depletion of unbound inhibitor available to the enzyme. Due to the high turnover of MDMA in the hepatocyte medium, the kinetic estimates of TDI were recalculated to take into account the change of MDMA level during the preincubation. As expected, this resulted in a substantial decrease (10-fold) in KI values but not kinact (Table 1). Constant exposure of MDMA level is critical in maintaining the inhibitory effect, however, significant depletion from medium will underestimate the true inhibitory potency particularly if these kinetic estimates are to be used for in vitro–in vivo extrapolation. Literature reports rarely account for depletion of the inhibitor

60

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 1 ( 2 0 0 7 ) 53–61

during pre-incubation even this is known to influence the calculated parameters (Yang et al., 2005). Extrapolation of TDI depends on kinact , KI , the unbound inhibitor concentration and degradation rate (Ito et al., 1998; Mayhew et al., 2000). Comparison of TDI of MDMA in different in vitro systems, based on kinetic values which were not corrected for MDMA depletion during the preincubation, would have implied that inactivation efficiency of MDMA was in the order of rCYP > HLM > hepatocytes. For instance, kinact for MDMA in hepatocytes (0.008 ± 0.001 min−1 ) could be considered greater than 20-fold lower than the values determined in HLM (0.15 ± 0.01 min−1 ) and rCYP (0.31 ± 0.06 min−1 ) (Heydari et al., 2004; Van et al., 2006). However, after kinetic estimates were reanalysed by taking into account the effect of MDMA level remaining in the preincubation mixture at the end of preincubation period; hepatocytes seemed to be more efficiently inactivated than HLM. Corrections for MDMA level in rCYP (using 2D6 expressed in yeast) and HLM were not necessary in previous reports as there was no significant depletion of inhibitor during the preincubation; however our recent unpublished data shows that such depletion occurs in Bactosomes® and Supersomes® (Van et al., unpublished data). The use of hepatocytes has therefore been advocated by Li et al. (1999), however, as indicated recently by Grime and Riley (2006), the results obtained from hepatocytes are heavily dependent on the individual donor. Interindividual variability including genetic polymorphism, age, and diet will influence drug metabolism (van de Weide and Steijns, 1999). Since different individuals were used as donors for the hepatocytes and HLM, and the rCYP studied expresses only one functional form of P450 enzyme (CYP2D6*1A/*1A), kinetic estimates are likely to differ from each other. The in vivo implications of different kinetic parameters for MBI on MDMA’s pharmacokinetics might be limited since under repeated administration of MDMA (e.g. 24 h apart), or with adequately high doses (e.g. 100 mg), CYP2D6 is not longer the major enzyme contributing to metabolic inactivation of MDMA (Yang et al., 2006). However, the variations in estimated kinetic parameters from different in vitro systems may influence the predicted in vivo consequences of intermediate inhibitors as discussed by Ghanbari et al. (2006). In conclusion, this study shows that hepatocytes provide another alternative model that can be used to define the kinetic estimates of time-dependent inhibition. Although only one donor was used, this investigation was primarily concerned on the methodology and technical aspects of acquiring kinetic estimates of TDI in hepatocytes. Experiments that do not account for TDI will underestimate the inhibitory potency in vivo and under-predict the actual magnitude of pharmacokinetic drug interactions. Proper experimental design for conducting TDI experiments, the knowledge of the appropriate in vitro model to be used for data analysis, and better understanding the role of contributing factors are necessary to obtain accurate kinetic estimates for robust assessment and prediction of TDI in vivo.

Acknowledgement L.M.V. is supported by an AstraZeneca PhD studentship.

Appendix A Derivation of Eq. (2): [I]corr =

AUC t

(3)

where [I]corr refers to the corrected inhibitor concentration due to depletion from hepatocyte medium and AUC is the area under the curve during the exposure of time, t:



t

AUC = 0

[I]0− [I]t k

(4)

where k is the first order elimination constant and is described as follows: k=

Ln[I]0 − Ln[I]t t

(5)

therefore, substituting Eqs. (4) and (5) to Eq. (3) results in



[I]corr =

[I]0− [I]t t

  ×

t Ln[I]0 − Ln[I]t



=

[I]0 − [I]t Ln[I]0 − Ln[I]t

references

Ghanbari, F., Rowland-Yeo, K., Bloomer, J.C., Clarke, S.E., Lennard, M.S., Tucker, G.T., Rostami-Hodjegan, A., 2006. A critical evaluation of the experimental design of studies of mechanism based enzyme inhibition, with implications for in vitro–in vivo extrapolation. Curr. Drug Metab. 7, 315–334. Gibbs, M.A., Kunze, K.L., Howald, W.N., Thummel, K.E., 1999. Effect of inhibitor depletion on inhibitory potency, tight binding inhibition of CYP3A by clotrimazole. Drug Metab. Dispos. 27, 596–599. Grime, K., Riley, R.J., 2006. The impact of in vitro binding on in vitro–in vivo extrapolations, projections of metabolic clearance and clinical drug–drug interactions. Curr. Drug Metab. 7, 251–264. Heydari, A., Rowland-Yeo, K., Lennard, M.S., Ellis, S.W., Tucker, G.T., Rostami-Hodjegan, A., 2004. Mechanism-based inactivation of CYP2D6 by methylenedioxymethamphetamine. Drug Metab. Dispos. 32, 1213–1217. Hollenberg, P.F., 2002. Characteristics and common properties of inhibitors, inducers, and activators of CYP enzymes. Drug Metab. Rev. 34, 17–35. Ito, K., Iwatsubo, T., Kanamitsu, S., Nakajima, Y., Sugiyama, Y., 1998. Quantitative prediction of in vivo drug clearance and drug interactions from in vitro data on metabolism, together with binding and transport. Annu. Rev. Pharmacol. Toxicol. 38, 461–499. Li, A.P., Lu, C., Brent, J.A., Pham, C., Fackett, A., Ruegg, C.E., Silber, P.M., 1999. Cryopreserved human hepatocytes, characterization of drug-metabolizing enzyme activities and applications in higher throughput screening assays for hepatotoxicity, metabolic stability, and drug–drug interaction potential. Chemico-Biol. Interact. 121, 17–35. Mayhew, B.S., Jones, D.R., Hall, S.D., 2000. An in vitro model for predicting in vivo inhibition of cytochrome P450 3A4 by metabolic intermediate complex formation. Drug Metab. Dispos. 28, 1031–1037.

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 1 ( 2 0 0 7 ) 53–61

McGinnity, D.F., Berry, A.J., Kenny, J.R., Grime, K., Riley, R.J., 2006. Evaluation of time-dependent cytochrome P450 inhibition using cultured human hepatocytes. Drug Metab. Dispos. 34, 1291–1300. McGinnity, D.F., Soars, M.G., Urbanowicz, R.A., Riley, R.J., 2004. Evaluation of fresh and cryopreserved hepatocytes as in vitro drug metabolism tools for the prediction of metabolic clearance. Drug Metab. Dispos. 32, 1247–1253. Nichols, D.E., 1986. Differences between the mechanism of action of MDMA, MBDB, and the classic hallucinogens. Identification of a new therapeutic class, entactogens. J. Psychoactive Drugs 18, 305–313. Obach, R.S., 1997. Nonspecific binding to microsomes, impact on scale-up of in vitro intrinsic clearance to hepatic clearance as assessed through examination of warfarin, imipramine and propanolol. Drug Metab. Dispos. 25, 1359–1369. Rowland, A., Gaganis, P., Elliot, D.J., Mackenzie, P.I., Knights, K.M., Miners, J.O., 2007. Binding of inhibitory fatty acids is responsible for the enhancement of UDP-glucuronosyltransferase 2B7 (UGT2B7) activity by albumin: implications for in vitro–in vivo extrapolation. J. Pharmacol. Exp. Ther. Jan 19, [Epub ahead of print] (doi:10.1124/jpet.106.118216). Silverman, R.B., 1998. Mechanism-based enzyme inactivation. In: Chemistry and Enzymology. CRC Press Inc., Boca Raton, FL. Takashima, T., Murase, S., Iwasaki, K., Shimada, K., 2005. Evaluation of dextromethorphan metabolism using

61

hepatocytes from CYP2D6 poor and extensive metabolizers. Drug Metab. Pharmokinet. 20, 177–182. van de Weide, J., Steijns, L.S., 1999. Cytochrome P450 enzyme system, genetic polymorphisms and impact on clinical pharmacology. Ann. Clin. Biochem. 36, 722–729. Van, L.M., Heydari, A., Yang, J., Hargreaves, J., Rowland-Yeo, K., Lennard, M.S., Tucker, G.T., Rostami-Hodjegan, A., 2006. The impact of experimental design on assessing mechanism-based inactivation of CYP2D6 by MDMA (Ecstasy). J. Psychopharmacol. 20, 834–841. Wu, D., Otton, v., Inaba, T., Kalow, W., Sellers, E.M., 1997. Interactions of amphetamine analogs with human liver CYP2D6. Biochem. Pharmacol. 53, 1605–1612. Yamazaki, M., Suzuki, H., Sugiyama, Y., 1996. Recent advances in carrier-mediated hepatic uptake and biliary excretion of xenobiotics. Pharm. Res. 13, 497–513. Yang, J., Jamei, M., Heydari, A., Yeo, K.R., De La Torre, R., Farre, M., Tucker, G.T., Rostami-Hodjegan, A., 2006. Implications of mechanism-based inhibition of CYP2D6 for the pharmacokinetics and toxicity of MDMA. J. Psychopharmacol. 20, 842–849. Yang, J., Jamei, M., Yeo, K.R., Tucker, G.T., Rostami-Hodjegan, A., 2005. Kinetic values for mechanism-based enzyme inhibition: assessing the bias introduced by the conventional experimental protocol. Eur J Pharm Sci. 26, 334–340. Zhao, P., Kunze, K.L., Lee, C.A., 2005. Evaluation of time-dependent inactivation of CYP3A in cryopreserved human hepatocytes. Drug Metab. Dispos. 33, 853–861.