AZD6738

CYP-mediated sulfoximine de-imination of AZD6738

Abstract

In hepatic S9 and human liver microsomes (HLM) the sulfoximine moiety of the ATR inhibitor AZD6738 is metabolised to its corresponding sulfoxide (AZ8982) and sulfone (AZ0002). The initial de- imination to AZ8982 is nominally a reductive reaction but in HLM required both NADPH and oxygen and was also inhibited by 1-aminobenzotriazole (ABT) at 1mM. Studies in a panel of 11 recombinant cytochrome P450s (CYPs) – CYP1A2, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1 CYP2J2, CYP3A4 and CYP3A5 – confirmed that the de-imination was an oxidative process, mediated largely by CYP2C8 with some CYP2J2 involvement, whilst the subsequent oxidation to the sulfone was carried out largely by CYP2J2, CYP3A4 and CYP3A5. There was no measureable metabolism in flavin-containing monooxygenase (FMO) enzymes FMO3, FMO5 or NADPH cytochrome C reductase. Studies using SilensomesTM, a commercially available HLM where specific CYPs have been inhibited by selective mechanism-based inhibitors, showed that when CYP2C8 was inhibited the rate of de- imination was reduced by 95% suggesting that CYP2J2 is only playing a minor role in HLM. When CYP3A4 was inhibited the rate increased by 58% due to the inhibition of the subsequent sulfone formation. Correlation studies in HLM samples from different individuals confirmed the role of CYP2C8 in the de-imination over CYP1A2, CYP2C9, CYP2C19, CYP2D6 and CYP3A. Hence, although nominally a reduction, the de-imination of AZD6738 to its sulfoxide metabolite AZ8982 is an oxidation mediated by CYP2C8 and that this metabolite is subsequently oxidised to the sulfone (AZ0002) largely by CYP3A.

Introduction

Ataxia telangiectasia mutated and RAD3-related protein (ATR) is a serine/threonine protein kinase that forms part of the DNA-damage response (DDR) coordinating the cellular response to DNA damage, stress, and cell-cycle perturbation (Cimprich and Cortez, 2008) and has become an attractive therapeutic target in cancer therapy (Foote et al., 2015; Weber and Ryan, 2015). AZD6738 (fig 1) is a selective and potent inhibitor of ATR, is orally active and bioavailable (Guichard et al., 2013; Jones et al., 2013) and has been taken into clinical development (ClinicalTrials.gov Identifier: NCT02223923; NCT02630199; NCT01955668; NCT02264678).

From a structural and metabolic perspective, the key functional group of interest is the sulfoximine moiety. A recent review by Sirvent and Lucking (2017) has highlighted that sulfoximines are being increasingly used in drug discovery as well as in clinical candidates. Beyond AZD6738, roniciclib and BAY 1143572 both contain sulfoximine moieties. The authors point out that sulfoximines have a unique set of physico-chemical and hydrogen bonding properties that make it a moiety of utility for medicinal chemistry and show this by replacing either secondary or tertiary amines with sulfoximines into a set of clinical compounds and comparing pharmacological, physico-chemical and drug metabolism properties. The sulfoximine analogues had similar pharmacological potencies and lipophilicities despite an increased degree of hydrogen bonding. This work highlights that although the drug analogues that were made showed generally better metabolic stabilities, there is little known about the metabolism of sulfoximines. Roniciclib is in clinical trials as a pan-CDK inhibitor with reportedly favourable pharmacokinetic properties in preclinical species but to date there is no published metabolism information on this compound (Bahleda et al., 2017).

Buthionine sulfoximine is an inhibitor of glutamylcysteine synthetase and is used in cancer chemotherapy to deplete glutathione and so make cells vulnerable to reactive oxygen species (Glasauer and Chandel, 2014). Despite many years of clinical use there is very little information on its metabolism. The information that does exist shows that the compound is excreted largely unchanged in the urine with a small amount of an acylated metabolite also being formed (Ahluwalia et al., 1990). Thus there is no information on the metabolic fate of the sulfoximine.

Once a metabolic route has been defined, reaction phenotyping studies are now routinely carried out to define what proportion of the route is mediated by different enzymes. Generally these studies comprise a number of different parts which can include inhibition with specific inhibitors, metabolism by recombinant enzymes and correlation with specific probe substrate activities across a panel of different biological samples (Zientek and Youdim, 2015). This approach has been used to understand the contribution of CYP and non CYP enzymes to metabolic transformations (Argikar et al., 2016).

The aim of the current study was to characterise the in vitro metabolism of AZD6738, particularly as it relates to the sulfoximine moiety and to conduct reaction phenotyping studies to understand the fate of this group and the enzymes involved.

Materials and methods

Materials

Compounds AZD6738, AZ8982 (sulfoxide) and AZ0002 (sulfone) were synthesised and developed at AstraZeneca UK Ltd. Pooled (150 donors; equal gender mix) human liver microsomes (HLM), individual donor human liver microsomes and recombinant flavin-containing monooxygenases (FMO3 and FMO5) were obtained from BD Biosciences (UK). Human liver S9 (mixed gender) were obtained from BioreclamationIVT (Baltimore, MD). Recombinant cytochrome P450s (CYP1A2, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1, CYP2J2, CYP3A4 and CYP3A5) and
human reductase bactosomes (human NADPH CYP-reductase expressed in Escherichia coli) were obtained from CYPEX (Dundee, UK). Silensomes™ were obtained from Biopredic International (Saint Grégoire, France). β-nicotinamide adenine dinucleotide 2′-phosphate reduced tetrasodium salt (NADPH), dimethylsulfoxide (DMSO), ketoconazole, 1-Aminobenzotriazole (ABT), N-desethylamodiaquine, amodiaquine, midazolam, 1’-OH midazolam, phenacetin, paracetamol,diclofenac, 4’-OH diclofenac, bufuralol, hydroxybufuralol, S-mephenytoin and 4-OH mephenytoin were purchased from Sigma Aldrich (Poole, UK). Methanol (MeOH), water and acetonitrile (ACN) were HPLC grade from Sigma-Aldrich (Poole, UK).

Microsomal, S9, cytosol and recombinant enzymes incubation conditions: Metabolic studies in human liver S9 (1mg/ml), cytosol (1mg/ml), HLM (0.5mg/ml); recombinant CYPs (100pmol/ml), recombinant FMOs (0.25mg/ml) and NADPH CYP-reductase bactosomes (100pmol/ml) were conducted in the presence of 1mM NADPH, mixed with 0.1M phosphate buffer pH7.4, and warmed to 37oC on a temperature controlled heater block. Experiments were initiated by addition of test compounds to give 1µM nominal test compound concentration in the incubation. 1mM ABT was used in addition to the above conditions in the HLM incubations for determination of CYP contribution.

The data were analyzed using a linear fit of the natural logarithm of the ratio of the compound peak area to the internal standard peak area against time. Intrinsic clearance (CLint) values were then calculated from the negative slope of the linear fit divided by the protein concentration. All the incubations were carried out in duplicate.

HLM incubations under anaerobic conditions: 0.1M phosphate buffer in capped vials on ice was bubbled with nitrogen for 30 minutes. Next 495µL of incubation mixture containing HLM (0.5mg/ml) and NADPH (1mM) was added to the vials and the vials capped and bubbled gently with nitrogen for a further 5 mins on ice. The reaction was initiated by addition of 5ul of substrate (final concentration 1µM), the vials were capped, placed on a 37ºC heater block, and bubbled with nitrogen to maintain an anaerobic atmosphere. Control incubations under aerobic conditions were carried out alongside. All the incubations were carried out in duplicate.

HLM enzyme kinetics incubation: Time and protein linearity experiments: AZD6738 was incubated at 1µM and 5µM containing 1mM NADPH, 0.1, 0.25, 0.5 and 1mg/ml HLM in 0.1M phosphate buffer pH 7.4 for 0, 5, 10, 20, 40, 60 minutes at 37oC. Following this, AZD6738 was incubated at 200, 150, 100, 75, 50, 25, 10, 7.5, 5, 2.5, 1, 0.5µM with 1mM NADPH and 0.5mg/ml HLM in 0.1M phosphate buffer pH 7.4 for 30 min at 37oC.

This study was carried out in duplicate. Data were analyzed in Graphpad Prism to obtain the Km and Vmax.
Incubations with SilensomesTM: AZD6738 was incubated at 25µM with 1mM NADPH and 0.5mg/ml protein mixed in 0.1M phosphate buffer pH 7.4 for 30 min at 37oC. Positive controls for each CYP enzyme was assayed alongside AZD6738 to confirm the viability of the SilensomesTM product (data not shown). All incubations were carried out in triplicate.

Correlation analysis: AZD6738 (25µM), amodiaquine (20µM), midazolam (3µM), diclofenac (10µM), S-mepheytoin (35µM), phenacetin (30uM) and bufuralol (5µM) were incubated with HLM from 7 individual donors along with the pooled HLM all at 0.5mg/ml, 1mM NADPH in 0.1M phosphate buffer pH 7.4 for 30 minutes at 37oC. All incubations were carried out in duplicate.

Sample processing and analysis: The experiments above were initiated by addition of test compounds. 25µl of incubates were removed into 100µl of ACN containing internal standard per time point: 0, 5, 10, 20, 40 and 60 minutes (unless stated otherwise). The samples were then centrifuged at 3000rpm for 10 minutes and 50µl of supernatant was diluted with 300µl of water before liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis. For all experiments, sulfoxide and sulfone calibration curves were generated in the relevant matrices with a limit of detection for both the sulfoxide and sulfone of 1nM. In all experiments, the final solvent concentration was always <0.1% DMSO and <1% ACN. Instrumentation: All LC-MS/MS data was acquired on a Waters Xevo® TQ-S mass spectrometer connected to a Waters Acquity UPLC (Ultra high-Performance Liquid Chromatography). The Waters Acquity system consisted of an autosampler, binary UPLC PUMP, column oven, autoinjector and a photodiode array detector. The analytes were separated by reverse-phase liquid chromatography using Kinetex C18 (50 x 2.1 mm, 2.6 µm, Phenomenex, Macclesfield, UK) preceded by a guard filter in a column oven at 50ºC. The mobile phase consisted of high performance liquid chromatography (HPLC) grade water with 0.1% formic acid (eluent A) and MeOH with 0.1% formic acid (eluent B). The elution profile was: linear gradient 95% A to 5% A, 0.00 to 2.2 minutes; isocratic hold, 5% A 2.2 minutes to 2.6 minutes; re-equilibration 95% A, 2.61 minutes to 2.8 minutes. The flow rate was 0.6 ml/min and the eluent was introduced into the mass spectrometer via the divert valve at 0.5 minutes. The injection volume was 3uL. The Xevo TQ-S was equipped with an electrospray ionisation (ESI) source, which was operated in positive mode. The mass spectrometer source settings were as following: the capillary voltage was 0.7kV and source offset was 59V. The desolvation temperature was set to 600°C. Nitrogen was used as the desolvation gas (1000L/Hr) and cone gas (150L/Hr). Argon was used as the collision gas at a flow rate of 0.15 ml/min. Detection of the ions was performed in the multiple reaction monitoring (MRM) mode using the transitions described in Table 1. Peak integration and calibrations were performed using TargetLynx software (Version 4.1, Waters, Milford, MA).The structures of AZD6738, the sulfoxide metabolite AZ8982 and the sulfone metabolite AZ0002 are shown in fig 1. Results Preliminary characterisation AZD6738 was incubated in different human hepatic S9 cytosol and microsomal fractions at a concentration of 1µM to determine the nature of enzyme mediating the de-imination (fig 2).In S9 AZD6738 was metabolised to the sulfoxide (AZ8982) and the sulfone (AZ0002) in the presence of NADPH. In the absence of NADPH there was no apparent formation of either of these metabolites. In the presence of NADPH the rate of formation of AZ133682982 was 0.3pmol/min/mg whilst the rate for AZ0002 was lower at 0.16pmol/min/mg (fig 2).There was no apparent formation of either metabolite in cytosol ruling out the involvement of soluble enzymes such as aldehyde and xanthine oxidase. When AZD6738 (1µM) was incubated with HLM in the presence of NADPH the formation of both AZ8982 and the corresponding sulfone AZ0002 was observed. In the absence of NADPH there was no measureable formation of either metabolite. Similarly under anaerobic conditions there was no measurable formation of either metabolite compared to control incubations carried out under aerobic conditions. Pre-incubation with 1mM aminobenzatriazole (ABT) under aerobic conditions significantly reduced the rate of formation of both AZ8982 and AZ0002 by more than 80% for both metabolites (fig 2). These preliminary studies strongly suggested that the de-imination was not a reductive process but rather an oxidative one mediated by enzymes in the microsomal fraction such as either CYPs or FMOs. HLM enzyme kinetics The enzyme kinetics of the de-imination of AZD6738 was investigated in pooled HLM by following the rate of formation of AZ8982 over an AZD6738 concentration range. Initial studies at 1µM AZD6738 showed that the rate of formation was linear with protein and time up to 0.5mg/ml and 30 minutes respectively. These conditions were used in the enzyme kinetic study over an AZD6738 concentration range of 0.5 to 200µM (fig 3).The de-imination showed saturable kinetics with a Km of 137µM and a Vmax of 18pmol/min/mg protein. This provides further support to the de-imination being an enzymatic rather than chemical process. Recombinant CYP studies The rate of metabolism was studied a panel of 11 recombinant CYPs – CYP1A2, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1 CYP2J2, CYP3A4 and CYP3A5 at an AZD6738 concentration of 1µM (fig 4A).Of the enzymes studied only CYP3A4 and CYP3A5 produced CLint values of 0.177µl/min/pmol CYP and 0.04µl/min/pmol CYP respectively. However, AZ8982 was only detected being formed by CYP2C8 and CYP2J2 at rates of 0.74 and 0.16pmol/min/pmol CYP respectively. This suggests that although CYP3A has the ability to metabolise AZD6738 only CYP2C8 and CYP2J2 are responsible for forming AZ8982.In addition, the rate of metabolism was also investigated in recombinant FMO3, FMO5 and NADPH cytochrome C reductase to see if either of these enzymes were also playing role. None of these enzymes showed any metabolic activity towards AZD6738 either in terms of disappearance or AZ8982 formation. AZ8982 was also incubated with the same panel of recombinant of CYPs with CYP3A4 and CYP3A5 again producing the highest CLint values of 2.7 and 0.5µl/min/pmol CYP respectively (fig 4B). These were followed by CYP2J2 and CYP2C8 with CLints of 0.4 and 0.05µl/min/pmol CYP respectively. However, in this case the enzymes were responsible for producing the sulfone were also CYP2J2, CYP2C8, CYP3A4 and CYP3A5 with rates of 0.34, 0.03, 0.17 and 0.16pmol/min/pmol CYP respectively indicating a stronger relationship between the CLint and formation of the sulfone than was seen for the initial de-imination.Hence the recombinant enzymes confirm the de-imination of AZD6738 as an oxidative process which is mediated by CYP2C8 and CYP2J2. In addition, the subsequent oxidation to the sulfone (AZ0002) is catalysed by CYP2J2, CYP2C8, CYP3A4 and CYP3A5. SilensomesTM studies To elucidate the enzymes involved in the de-imination of AZD6738 the formation of AZ8982 was studied in CYP1A2, CYP2C8, CYP2C9, CYP2C19, CYP2D6 and CYP3A4 SilensomesTM (fig 5). Given the high Km previously determined these studies were conducted at 25µM AZD6738.SilensomesTM are commercially available HLM where specific CYPs have been inhibited by selective mechanism-based inhibitors. Excess inhibitor is then removed during the preparation process yielding microsomes chemically knocked for a specific CYP (Parmentier et al., 2017).When AZD6738 was incubated at a concentration of 25µM there was no significant difference in the rate of formation of AZ8982 in CYP1A2, CYP2C9, CYP2C19 and CYP2D6 SilensomesTM compared to the activity seen in the control incubations. However, in CYP2C8 SilensomesTM the rate was reduced to 5% of the control value and in CYP3A4 SilensomesTM the rate of formation of AZ8982 increased by 58% (fig 5). The data from the previous HLM and rCYP studies suggest that the apparent increase in the rate of formation of AZ8982 formation in CYP3A4 SilensomesTM was due, in fact, to inhibition of the subsequent metabolic step (the formation of AZ0002) causing a build-up of the sulfoxide. This supports the hypothesis that the de-imination is the initial metabolic step which is followed by oxidation to the sulfone. This was confirmed by measuring the CLint for AZ8982 in SilensomesTM CYP3A4 and CYP2C8 and showing a 97% and 10% reduction respectively (data not shown). Correlation analysis The rate of formation of AZ8982 from AZD6738 at 25µM was studied in set of 7 HLM samples from individual samples along with the pooled HLM sample used in the previous studies. In addition, marker enzyme activities for CYP1A2 (phenacetine O-deethylation), CYP2C8 (amidioquine N- deethylation), CYP2C9 (diclofenac 4’-hydoxylation), CYP2C19 (S-mephenytoin 4-hydroxylation) and CYP3A (midazolam 1’-hydoxyaltion). All of the marker substrates were incubated on or around the Km for the specific reaction. A correlation analysis was conducted between the rate of AZ8982 and the marker CYP enzyme activities (table 2). There was a poor correlation with the majority of the enzyme activities with the correlation coefficients determined as being less than 0.5. The most significant correlation (correlation coefficient = 0.73) was seen between the rate of formation of AZ8982 and the CYP2C8-mediated rate of N-deethylation of amidiaquine (fig 6).This further supports to the previous data which shows that in HLM the de-imination of AZD6738 is an oxidative process mediated primarily by CYP2C8. Discussion Reaction phenotyping is now a relatively routine activity conducted during the discovery and development of drugs. It involves a well-defined set of studies usually including a combination of metabolite formation by recombinant enzymes, inhibition by specific inhibitors and correlation across different biological samples which have been characterised for different enzyme activities (Zientek and Youdim, 2015). In the context of this study, reaction phenotyping has been used to understand the de-imination of AZD6738. This study has investigated the de-imination of AZD6738 to its sulfoxide metabolite (AZ8982). It has shown that this is an oxidative process mediated by cytochrome P450. Following this reaction phenotyping studies including recombinant enzymes, inhibitor (SilensomesTM) and correlation studies identified CYP2C8 as the major CYP involved in this biotransformation. The sulfone metabolite, AZ0002, was also observed in these studies but the data suggest that this is a subsequent reaction following the de-imination and is mediated by CYP3A. CYP2J2 was also shown to have a role in the de-imination of AZD6738. However, there is a relatively low level of expression of this enzyme in human liver (around 1.2pmol/mg protein), which equates to 0.3% of the total hepatic cytochrome P450 content (Achour et al., 2014). When compared with CYP2C8 which has an expression level of around 22pmol/mg, which in turn equates to around 5% total hepatic cytochrome P450 content, it is clear that CYP2J2 would be a minor contributor to the hepatic de-imination of AZD6738. Interestingly CYP3A did not play a significant role in the de-imination of AZD6738. The recombinant CYP CLint data show that there was metabolism of AZD6738 by CYP3A but this did not include the formation of the sulfoxide. This finding was supported by the SilensomesTM and correlation data. However, CYP3A was the major enzyme involved in the formation of the sulfone (AZ0002). Based on the data from these studies it is possible to suggest that this is a subsequent step to the initial de- imination rather than being formed directly from AZD6738.CYP2C8 is increasingly recognised as a key human drug metabolising enzyme. AZD6738 can now be added to the increasing list of CYP2C8 substrates which include compounds such as cerivastatin, montelukast and rapaglinide (Backman et al., 2016). In addition, like AZD6738, there are a number of anti-cancer agents on the substrate list including paclitaxel, imatinib and tozasertib. Unlike CYP2C9 and CYP2D6, CYP2C8 does not appear to have a characteristic substrate recognition motif such as an acidic or basic function although pharmacophore models for CYP2C8 have been proposed that involve a combination of hydrophobic and polar functions at fixed distances from the site of metabolism (Melet et al., 2004). Whilst AZD6738 may have many of the structural features which align with this type of pharmacophore model, it is difficult to rationalise why the sulfoxide should not be metabolised by the same enzyme but is preferentially metabolised by CYP3A. There does seem to be a significant overlap between substrates of CYP2C8 and CYP3A (Backman et al., 2016). However, the x-ray crystal structures have revealed that the active sites have a similar size 1438Å3 for CYP2C8 (Schoch et al., 2004) and 1386Å3 for CYP3A (Yano et al., 2004) but have different shapes. The active site of CYP3A is generally thought of being a large open space whereas the CYP2C8 active site has been described as being T or V shaped (Backman et al., 2016). However, these high level structural differences do not explain why CYP2C8 is able to mediate the de-imination of AZD6738 when CYP3A4 cannot as it is likely that AZD6738 will fit into both these active sites. In the absence of co-crystal structures of AZD6738 with CYP2C8 and CYP3A4 it is not possible to shed any further light on the differences observed with the selectivity of the de-imination reaction. Having identified CYP2C8 as the major enzyme involved in the de-imination of AZD6738, the question becomes what is the mechanism for this reaction. Nominally going from a sulfoximine to a sulfoxide is a reductive reaction but the data detailed here shows it to actually be an oxidation. This is a kin to the well described deamination reaction mediated by cytochrome P450s seen with compounds such as diltiazem (Nakamura et al., 1990). However, the mechanism for this reaction is generally described as proceeding via oxidation of the alpha carbon atom adjacent to the nitrogen. In the case of AZD6738 there is no carbon atom adjacent to the imine nitrogen so this mechanism is unlikely. One other potential mechanism would be similar to the proposed mechanism for N- oxidation (Guengerich, 2001) but in this case after the initial oxidation of the imine nitrogen there is potential for a subsequent oxidation and the liberation of nitrous acid to yield the sulfoxide (fig 7). In summary, these in vitro studies show that AZD6738 undergoes oxidative de-imination to the corresponding sulfoxide (AZ8982) mediated principally by CYP2C8. AZ8982 undergoes subsequent CYP3A-mediated oxidation to the sulfone (AZ0002). This represents one of the first studies into the metabolism of the sulfoximine moiety. References Achour B, Barber J, and Rostami-Hodjegan A (2014) Expression of hepatic drug-metabolizing cytochrome p450 enzymes and their intercorrelations: a meta-analysis. Drug metabolism and disposition: 42:1349-1356. Ahluwalia GS, Grem JL, Hao Z, and Cooney DA (1990) Metabolism and action of amino acid analog anti-cancer agents. Pharmacology & therapeutics 46:243-271. Argikar UA, Potter PM, Hutzler JM, and Marathe PH (2016) Challenges and Opportunities with Non- CYP Enzymes Aldehyde Oxidase, Carboxylesterase, and UDP-Glucuronosyltransferase: Focus on Reaction Phenotyping and Prediction of Human Clearance. The AAPS journal 18:1391- 1405. Backman JT, Filppula AM, Niemi M, and Neuvonen PJ (2016) Role of Cytochrome P450 2C8 in Drug Metabolism and Interactions. Pharmacological reviews 68:168-241. Bahleda R, Grilley-Olson JE, Govindan R, Barlesi F, Greillier L, Perol M, Ray-Coquard I, Strumberg D, Schultheis B, Dy GK, Zalcman G, Weiss GJ, Walter AO, Kornacker M, Rajagopalan P, Henderson D, Nogai H, Ocker M, and Soria JC (2017) Phase I dose-escalation studies of roniciclib, a pan-cyclin-dependent kinase inhibitor, in advanced malignancies. British journal of cancer 116:1505-1512. Cimprich KA and Cortez D (2008) ATR: an essential regulator of genome integrity. Nature reviews Molecular cell biology 9:616-627. Foote KM, Lau A, and Nissink JW (2015) Drugging ATR: progress in the development of specific inhibitors for the treatment of cancer. Future medicinal chemistry 7:873-891. Glasauer A and Chandel NS (2014) Targeting antioxidants for cancer therapy. Biochemical pharmacology 92:90-101. Guengerich FP (2001) Common and uncommon cytochrome P450 reactions related to metabolism and chemical toxicity. Chemical research in toxicology 14:611-650. Guichard SM, Brown E, Odedra R, Hughes A, Heathcote D, Barnes J, Lau A, Powell S, Jones CD, Nissink W, Foote KM, Jewsbury PJ, and Pass M (2013) Abstract 3343: The pre-clinical in vitro and in vivo activity of AZD6738: A potent and selective inhibitor of ATR kinase. Cancer Research 73:3343-3343. Jones CD, Blades K, Foote KM, Guichard SM, Jewsbury PJ, McGuire T, Nissink JW, Odedra R, Tam K, Thommes P, Turner P, Wilkinson G, Wood C, and Yates JW (2013) Abstract 2348: Discovery of AZD6738, a potent and selective inhibitor with the potential to test the clinical efficacy of ATR kinase inhibition in cancer patients. Cancer Research 73:2348-2348.
Melet A, Marques-Soares C, Schoch GA, Macherey AC, Jaouen M, Dansette PM, Sari MA, Johnson EF, and Mansuy D (2004) Analysis of human cytochrome P450 2C8 substrate specificity using a substrate pharmacophore and site-directed mutants. Biochemistry 43:15379-15392.
Nakamura S, Ito Y, Fukushima T, Sugawara Y, and Ohashi M (1990) Metabolism of diltiazem. III. Oxidative deamination of diltiazem in rat liver microsomes. Journal of pharmacobio- dynamics 13:612-621.
Parmentier Y, Pothier C, Delmas A, Caradec F, Trancart MM, Guillet F, Bouaita B, Chesne C, Brian Houston J, and Walther B (2017) Direct and quantitative evaluation of the human CYP3A4 contribution (fm) to drug clearance using the in vitro SILENSOMES model. Xenobiotica; 47:562-575.
Schoch GA, Yano JK, Wester MR, Griffin KJ, Stout CD, and Johnson EF (2004) Structure of human microsomal cytochrome P450 2C8. Evidence for a peripheral fatty acid binding site. The Journal of biological chemistry 279:9497-9503.
Weber AM and Ryan AJ (2015) ATM and ATR as therapeutic targets in cancer. Pharmacology & therapeutics 149:124-138.
Yano JK, Wester MR, Schoch GA, Griffin KJ, Stout CD, and Johnson EF (2004) The structure of human microsomal cytochrome P450 3A4 determined by X-ray crystallography to 2.05-A resolution. The Journal of biological chemistry 279:38091-38094.
Zientek MA and Youdim K (2015) Reaction phenotyping: advances in the experimental strategies used to characterize the contribution of drug-metabolizing enzymes. Drug metabolism and disposition: 43:163-181.