In the present study we have developed a high quality, rapid and cost effective CYP450 inhibition assay that does have the ability to detect both reversible and CYP3A4 metabolism-dependent inhibition (MDI), using recombinantly expressed P450 isoforms and fluorogenic P450 substrates. CYP3A4 isoform is screened with diethoxyflourescein (DEF) as probe substrate. The IC50 values can then be calculated for test compounds against the CYP3A4 isoform, based on the rate of metabolism of the probe substrate, measured for 10 minutes. In addition, the CYP3A4 metabolism-dependent inhibitory potential of test compounds is determined by extending for 30 minutes the determination of the rate of metabolism of diethoxyflourescein and calculating IC50 values every 5 minutes of the incubation period. An estimate of the CYP3A4 metabolism-dependent inhibitory potential of the test compounds can be determined comparing IC50 values, measured following 10 and 30 minutes incubation. The incubation was performed using the selective CYP inhibitors miconazole, for direct P450 inhibition, and troleandromycin, for metabolism-dependent inhibition, as positive controls. The entire screening process was fully-automated in 96-well plate format with the use of Hamilton liquid-handling robot technology coupled with two fluorimeters (Tecan) and a custom laboratory-information management. This assay is currently applied to screen compounds early in the lead optimization process and identify those compounds that cause reversible and/or metabolism-based CYP450 inhibition and therefore progress those molecules or chemical series with the lowest DDI potential possible. The high number of data generated through this assay can also be used to build an informative database and improve predictive models.
Drug-drug interactions (DDIs) following drug therapy, which may result in either reduced efficacy or increased toxicity [1,2] and may culminate in serious (and even occasionally life-threatening) adverse reactions [3,4] are primarily caused by macromolecule binding of reactive species or drug co-therapy resulting in plasma concentrations of one of the co-administered drugs being elevated to toxic levels . Since “Polypharmacy” or the simultaneous prescription of more than one drug to treat one or more conditions in a single patient is a very common practice, the possibility of DDI therefore exists in the majority of patients and effects can be particularly critical in drugs which have a narrow therapeutic index. Several drugs (e.g., terfenadine, mibefradil, astemizole, cisapride and sorivudine) have either been withdrawn from the market or suffer from restrictions in their prescription for this reason . Inhibition is a high profile issue for drug discovery and development programs and great importance is now placed on in vitro studies as tools for predicting in vivo DDIs, particularly those resulting from cytochrome P450 (CYP450) inhibition , since the metabolic elimination of a large number of drugs is dependent on the CYP450 family of enzymes [8-10]. CYP1A2, CYP2C9, CYP2C19, CYP2D6 and CYP3A4 are the major CYP450 enzymes present in the human liver that are responsible for oxidative metabolism of most drugs , with CYP3A  accounting for 30% and 80% of total CYP protein in human liver and gut, respectively [13,14]. CYP3A4 has broad substrate specificity and is estimated to be involved in the metabolism of approximately 50% of drugs used in humans. Broad substrate specificity makes CYP3A4 more susceptible to be reversible and/or irreversible inhibited by a variety of drugs [15,16]. Examples of cytochrome P450 inhibition include ketoconazole, itraconazole, erythromycin, clarithromycin, diltiazem, and nefazodone (CYP3A); quinidine, paroxetine, and terbinafine (CYP2D6); ticlopidine (CYP2C19); enoxacin (CYP1A2); and sulfaphenazole (CYP2C9); with some drugs possessing the potential to inhibit more than one CYP450 enzyme: fluconazole (CYP2C9 and CYP2C19) and fluvoxamine (CYP1A2 and CYP2C19). Inhibition of CYP450-dependent metabolism can generally be classified into three categories: reversible (mostly competitive), quasi-irreversible (when compounds complex the heme prosthetic group and leave the CYP450 functionally inactive), and irreversible or mechanism-based (MBI, compounds covalently bind to the heme or the surrounding protein) . The last two type of inhibition can also called MDI (metabolism-dependent inhibition) or TDI (time-dependent inhibition), which is a collective term for a change (most often an increase) in potency of CYP inhibitors during an in vitro incubation or dosing period in vivo . Assays to screen for competitive interactions are now a routine part of drug discovery screening cascades [19-21], but reports of metabolism-dependent inhibition (MDI) are increasing in prevalence [22-24]. Moreover, irreversible and quasi-irreversible inhibitions are often viewed as more serious than reversible inhibition, since the inhibitory effect remains after elimination of the parent drug from the body. In contrast, with reversible inhibition, the effects of MBI are also more profound after multiple-dosing and the recovery period, typically several days, is independent of continued exposure to the drug [25,26]. MDI is an unusual occurrence with most enzymes, but it is observed at a higher frequency in CYP450-catalyzed reactions, perhaps due to the reactivity of the oxygenated species formed during the course of the oxygenation reactions . There are examples of irreversible or quasi-irreversible CYP450 inhibition across many classes of therapeutic drugs, recreational drugs, and herbal medicines and all of the major drug-metabolizing CYP450s have been implicated [27,28]. Many approaches have been developed to predict drug-drug interactions using various in vitro methodologies, nevertheless in early drug research efforts, focus has been on the development of high-throughput assays for major drug-metabolizing enzymes to avoid progression of new chemical entities that will possess a high potential to cause drugdrug interactions and to develop structure-activity relationships useful in the design of alternate agents that will lack this potential. In this research phase, speed is important, and high-throughput approaches that use fluorogenic substrates to measure CYP450 activities have been described [19-24], as well as “cocktail” experiments that simultaneously measure more than one CYP450 activity [29-34]. These high throughput assays can only determine reversible CYP450 inhibition and does not have the ability to detect mechanism-based inhibition, which is usually determined later in the process; however microtiter plate methods with real-time detection of fluorescent metabolites, commonly employed for studying CYP inhibition, can be simply modified for the screening of CYP MDI [22-24,33,34]. This approach is subject to some limitations, such as interference from fluorescent test inhibitors and, because the substrates for these assays are not CYP selective and necessitate the use of recombinant single enzyme systems, the inhibitory effect of metabolites generated by one CYP on other CYPs cannot be tested. In addition, MDI may also arise from the generation of potent reversible inhibitory metabolites. However, despite these problems, the shift in IC50 observed does allow compounds to be ranked with respect to potency of MDI .
In the present study we have developed a high quality, rapid and cost effective CYP450 inhibition assay which screen for both reversible and CYP3A4 metabolism-dependent inhibition, using recombinantly expressed P450 isoforms and fluorogenic P450 substrates. The reversible CYP450 screen utilizes a range of test compound concentrations to determine the IC50 value, with CYP3A4 isoform screened and diethoxyflourescein as substrate. The IC50 values can then be calculated for test compounds against the CYP3A4 isoform, based on the rate of metabolism of the substrates, measured for 10 minutes. The CYP3A4 metabolism-dependent inhibitory potential of test compounds is determined by extending for 30 minutes the determination of the rate of metabolism of diethoxyflourescein. An estimate of the CYP3A4 metabolism-dependent inhibitory potential of the test compounds can be determined comparing IC50 values, measured following 10 and 30 minutes incubation. The validation was carried out comparing, for a determined set of compounds, the fluorescence method with the widely used ‘IC50 shift’ method [22-24,36,37], where compound being evaluated are co-incubated in human liver microsomes with a known substrate for a specific CYP and a metabolism dependent inactivators are expected to cause a left-shift of the IC50 curve. This method provides some information on metabolism-dependency and has been used to rank order compounds based on the extent of the ‘shift’. To the best of our knowledge this is one of the first examples of assay with the ability to detect both reversible and CYP3A4 metabolism-dependent inhibition, using recombinantly expressed P450 isoforms and fluorogenic CYP450 substrates, that, along with the use of high capacity automated pipetting systems, may enable the rapid screening of compounds in a cost effective manner. By reducing the cost and increasing the speed of the assays, a comprehensive CYP450 inhibition screening package can be performed at an earlier stage in drug discovery thereby reducing the possibility of costly late stage failures.
Miconazole, Troleandromycin, Midazolam maleate salt and Ethylene diamine tetra acetic acid (EDTA) were purchased from Sigma-Aldrich (Italy). Diethoxyflourescein (DEF) was gift from GlaxoSmithKline, UK. Pooled human liver microsomes (HLM, XenoTech), and recombinant cytochrome CYP3A4 (CYP3A4 low reductase, Cypex®, Du ndee) was purchased from Trimital S.R.L, Milan, Italy. The other reagents and solvents used were of analytical or HPLC grade.
Automated CYP450 inhibition assay using individually expressed human cytochrome P450 enzymes to determine IC50 values and fold-change decrease for CYP3A4 isoform
All the incubation steps and IC50 determinations were performed using a robot sample processor (RSP) Microlab STARlet (Hamilton) equipped with two termomixers Comfort (Eppendorf) and two Spectraflour fluorimeters (Tecan). Stock solutions of test compounds and positive controls were prepared at 5mM and 0.5 mM, respectively, in methanol. Then, test inhibitors and troleandromycin (positive control for CYP3A4 MDI) stocks were serially diluted to give nine secondary solutions for each compound over 3.3-2000 μM and 0.33-200 μM, respectively. Miconazole (p ositive control for direct inhibition) stock solution was serially diluted to give eight secondary solutions over 0.23- 167 μM concentration range.
Incubation was performed in polystyrene 96-well plates, flat bottom with low evaporation lid. Incubation mixtures containing probe substrate (final concentration 1 μM DEF), Cypex rhCYP3A4 low reductase (final protein concentration of 0.1 mg/ml), Nine concentrations (0.065, 0.164, 0.41, 1.02, 2.56, 6.4, 16, 40, 100 μM) of tes t compound, EDTA (0.5 mM) and 100 mM Tris buffer, pH 7.4, were preincubated in 96-well plates for 5 min at 37°C. The enzymatic reaction was initiated by adding 50 μl of pre-warmed NADPH regenerating system (7.8 mg glucose 6-phosphate, 1.7 mg NADP, 6 units glucose 6-phosphate dehydrogenase per mL of 2% sodium bicarbonate) to 200 μL/well of incubation mixture. Incubations were carried out into the fluorimeter, using plates with lid, for 30 minutes. Either positive (miconazole and troleandromycin) or negative controls (methanol) were included in each assay to ensure the integrity of the enzyme incubation system.
Fluorescence measurements were made with excitation and emission wavelengths of 486 and 530 nm respectively, employing 80 Gain.
Automated MDI assay using pooled human liver microsomes
The fully automated MDI assay using human liver microsomes (HLM) was performed using a robot sample processor (RSP) Microlab STARlet (Hamilton) equipped with two termomixers Comfort (Eppendorf). Test compounds and positive control inhibitor (troleandomycin) were diluted as described above. The incubation procedure was carried out as previously described (23), HLM (final concentration 0.1 mg protein/ ml) were incubated without and with 25 μl of NADPH regenerating system (7.8 mg glucose 6-phosphate, 1.7 mg NADP, 6 units glucose 6-phosphate dehydrogenase per mL of 2% sodium bicarbonate) in the presence of various concentrations of test compound or troleandomycin (final concenrations 0, 0.065, 0.164, 0.41, 1.02, 2.56, 6.4, 16, 40, 100 μM) for 10 min. Then incubation mixtures were in cubated with substrate (midazolam, 2.5 μM final concentration, Km = 5 μM) in the presence of NADPH generating system for a further 5 min.
An aliquot (100 μl) was removed from the incubation and pipetted into the tubes in the quench rack (containing 100 μl of acetonitrile). Quenched samples were centrifuged at 3000 g for 15 min at 4°C and the supernatants were transferred to microtiter plates for LC/MS-MS analysis.
Liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS)
All measurements were performed using an Acquity UPLC system equipped with thermostated autosampler and column compartment (Waters, Milford, MA, USA) and coupled to a 4000QTrap mass spectrometer from AB Sciex (Concord, ON, Canada) and equipped with turbo ion-spray (TIS) ion source. Chromatographic separation was performed on a Waters BEH C18 column (2.1 mm x 30 mm, 1.7 μm particle sizes) at 60°C using a flow rate of 0.8 mL/min. The mobile phase consisted of phase A: water 0.1% (v/v) formic acid and phase B: acetonitrile containing 0.1% (v/v) formic acid. The HPLC gradient started at 95% mobile phase A and was held for 0.2 min. Mobile phase B was increased linearly to 95% over 1.3 min and was held at 95% for 1.55 min. The total LC/MS/MS run time was 2 min. Auto sampler temperature was kept at 4°C. The injection volume was 5 μl. The ion source temperature was set to 650°C with an ion spray voltage of 3000 V. The transitions monitored in MRM were m/z 343—>203 for 1-hydroxymidazolam and m/z 359—>337 for α-hydroxytrizolam (internal standard) and the collision energy was set at 37eV. The Peak area ratio between 1-hydroxymidazolam and alpha-hydroxytriazolam (internal standard) was used for calculation of IC50.
The IC50 value was calculated using the excel macro XC50v2. The control rate of fluorescent metabolite production was established from no compound vehicle control incubations (uninhibited), assigned as 100%. The extent of inhibition at each compound concentration was calculated relative to the control rate and the pIC50 value was determined from these results. Data were fitted to a 4 parameter equation (Eq. A), where the lower data limit was 0.
Where: A = minimum y, D = slope factor, B = maximum y, x = log10 compound concentration [M], C = log10xC50, pxC50 = -C.
For the assessment of CYP3A4 MDI, pIC50 values were calculated every 5-min and MDI fold decrease in IC50 values was calculated between 10 and 30 minutes. The criteria to determine positive MDI were based on a twofold value of the fold-change of IC50 of test inhibitors measured at 10 and 30 minutes.
IC50 determination of CYP inhibitors
The purpose of this study was to create a high throughput assay in order to give results for reversible inhibition associated with preliminary information regarding a potential for MDI for the compounds tested. To speed-up the process two assays in one were implemented. A fluorescence-based assay using recombinant enzymes was chosen as starting reversible assay because it is very easy and cheap to use. In a single assay, it is possible to measure the IC50 and also distinguish MDI compounds from non MDI compounds. During this validation a correct solvent was required to solubilize the substrate that can give a minimum effect on CYP metabolism. Many organic solvent (methanol, DMSO, and ethanol) significantly inhibit CYP metabolism, even at low concentration.
The method was validated by incubating known reversible and irreversible (or quasi-irreversible) CYP inhibitors: miconazole (reversible inhibition) and troleandromycin (irreversible or quasiirreversible inhibition) with known specific substrates (DEF). Another 27 compounds were tested (10 marketed compounds and 17 in house compounds). IC50 values were determined transforming a pIC50 value calculated plotting the percent control activity versus concentration of the test compounds. The potential for MDI was calculated as a fold decrease in IC50 values between 10 minutes and 30 minutes of incubation. Figure 1 shows a representative metabolism dependent curve of troleandromycin, a well-known MDI inhibitor and miconazole as reversible inhibitor . The IC50 values of troleandromycin decreased from 1.35 μM at 10 minutes to 0.1 μM at 30 minutes of incubation (Table 1, Figure 1). In metabolism dependent inhibition, CYP enzymes lose activity progressively when more reactive metabolites are generated during incubation. Since more active CYP 3A4 enzyme existed in the incubation mixture, in the early stage of incubation, therefore, the IC50 value is high. Whereas in late stage of incubation, less active CYP3A4 enzyme remained due to increasing concentration of reactive metabolite(s) during incubation period. Although enzyme activity can decrease during incubation time because of CYP instability, but the activity of recombinant CYP3A4 were stable up to 30 minutes using DEF as probe base on the data obtained from optimization study (data not shown) [23-37].
|MDI RE||MDI HLM||literature|
|IC50 (µM)||IC50 (µM)||Fold||First IC50||Second IC50||Fold-|
|(0-10 min)||(25-30 min)||change||(µM)||(µM)||change|
Table 1: Results obtained for potential for MDI using recombinant enzyme (RE) and human liver microsomes (HLM).
Table 1 shows the MDI results obtained for 10 marketed drugs and 17 test compounds using recombinant enzyme (RE) and human liver microsomes (HLM). The results obtained with the inhibition assay using recombinant enzymes are in good agreement with MDI results calculated using human liver microsomes. In addition, potential for MDI determined using recombinant enzymes also agreed with the literature values calculated using human liver microsomes. Table 1 shows direct comparison between percent of MDI and non-MDI compounds in these two systems. The number and percentage of compounds in each category (positive or negative) is summarized in Table 2. 74% of the test compounds in recombinant enzyme had the same inhibition categorization in the HLM. For example, negative in recombinant enzyme was negative in HLM (n = 5), positive in recombinant enzyme was positive in HLM (n = 15). Nonetheless, 7 compounds (26%) in the data set had negative in recombinant enzyme but positive in HLM. Interestingly, there were no compounds which were positive in recombinant enzyme but negative in HLM. The false positive or negative could be due to metabolism of compound with different CYPs enzymes or the activity in recombinant enzyme CYP3A4 enzyme might be higher than CYP3A4 in pooled human liver microsomes. This demonstrates that potential for MDI can be accurately determined using recombinant enzyme and using fluorescence P450 inhibition assay instead of a specific assay using human liver microsomes with LC/MS/MS analysis, saving time and money.
|n=27 Marketed & In-house||RE assay+||RE assay-|
|HLM assay+||15* (56%) **||7 (26%)|
|HLM assay-||0* (0%)||5 (18%)|
Table 2: Comparison between recombinant enzyme (RE) MDI assay versus Human liver microsomes (HLM) MDI assay.
Three replicates were included in the experiments to minimize experimental errors. The value of the first time point (5min) was not used in data analysis because of high experimental errors due to low signal-to-noise ratios, IC50 at 10 min was used as the first time point for MDI fold changes determination and also general IC50 value. So the IC50 at 10 min refers to the first reliable IC50 measurement when the amount of reactive metabolite(s) was minimal with all the compounds that were used in this study. The ratio of initial IC50 (10 min)/final IC50 (30 min) has been proposed as the fold change. Decrease in the IC50 during incubation depends to the inactivation of the enzyme which is depends upon the reactivity of the active metabolite(s). The strong active metabolite gives high fold change with low IC50 and vase versa e.g., troleandromycin from 1.35 μM to 0.1 μM or phenelzine from 90 μM to 7.9 μM. The fold change and IC50 of compound can be used as indicator of potency ranking purpose for selecting the best compound for development. Also MDI compounds can be identified by this assay since the IC50 value of a non MDI compound does not change significantly with incubation time, if the inhibitor strictly follows simple Michaelis- Menten kinetics. As IC50 of miconazole (Figure 1) was relatively stable, and the fluctuation was within the experimental error range. However, IC50 values of some compounds showed increased with incubation time this could be due to conversion of compound to metabolite(s) which are not inhibiting the CYPs enzymes e.g., Nifedipine (Table 1). If a test compound is a CYP substrate, extensive metabolism by the CYP enzyme can result in a significant decrease in the concentration of the test compound during the incubation and consequently the IC50 of the test compound will increase during incubation. While the IC50 values of the some compounds are stable during incubation time this could be due to reversible inhibition test compound and also its metabolite(s), i.e., ketoconazole (Table 1), whereas, decrease in IC50 with incubation time shows to be an indication of metabolism dependent inhibitor. Base on the results obtained in this study suggested that DEF is a reliable probe substrate for identification MDI and 11 non MDI compounds. Although two fluorescence substrates were used for inhibition study for CYP 3A4, but DEF was chosen in this work for identification of MDI compounds because of better signal-to-noise ratio. In general, a decrease of more than two fold in IC50 was seen as indication of MDI compound. The high-throughput inhibition assay using recombinant enzyme to calculate IC50 value for general inhibition associated a potential for MDI described here has become a valuable addition to our routine inhibition fluorescence assays using individual isoforms (CYPs enzymes) in support of drug discovery programs.
The potential for MDI calculated for each positive control using recombinant enzyme were in agreement with MDI positivity calculated for the same compound using human liver microsomes and also with the value reported in literature. False negative results using RE were found in a percentage of 27% (Table 2) depending on a difference between RE expression and HLM, e.g., other CYPs present in HLM (or in vivo) may metabolize compound to alternative metabolites, thereby preventing MDI.
In Summary, these results suggest that recombinant enzymes could provide an alternative way to identify MDI compound and possibly lead to aware of clinical risks. It can be used as a primary screening approach used to identify MDI CYP inhibitors. This assay can be used as screening approach in the early drug discovery and followed by human liver microsomes approach when the compound reach to the lead optimization process. Also with this assay it is possible to identify MDI from non MDI compounds in same run by using recombinant human CYP3A4 and DEF as probe substrate.
The authors have no conflicts of interest.
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.