Biosynthesis and identification of metabolites of maraviroc and their use in experiments to delineate the relative contributions of cytochrome P4503A4 versus 3A5

Article abstract of DOI:10.1124/dmd.117.079855

Maraviroc (MVC) is a CCR5 coreceptor antagonist indicated in combination with other antiretroviral agents for the treatment of CCR5-tropic human immunodefinciency virus-1 infection. In this study, the metabolism of MVC was investigated in human liver micr

Full text of DOI:10.1124/dmd.117.079855

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
BIOSYNTHESIS AND IDENTIFICATION OF METABOLITES OF MARAVIROC AND THEIR USE
IN EXPERIMENTS TO DELINEATE THE RELATIVE CONTRIBUTIONS OF CYTOCHROME
P4503A4 VS. 3A5
Elaine Tseng, Gwendolyn D. Fate, Gregory S. Walker, Theunis C. Goosen, and R. Scott Obach
Pfizer Inc., Groton, CT, USA
1

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
Running Title: Metabolism of Maraviroc by CYP3A4 and CYP3A5
Addresses for Correspondence
Gwendolyn D. Fate
Athena DMPK LLC
2
9 Willow Lane
East Lyme, CT, 06333
gwenfate@gmail.com
or
R. Scott Obach
Pfizer Inc.
Eastern Point Rd.
Groton, CT, USA, 06340
r.scott.obach@pfizer.com
Number of Words in:
Abstract: 211
Introduction 352
Discussion 925
Number of References: 21
Number of Tables: 2
Number of Figures: 8
List of Abbreviations: NADPH: nicotinamide adenine dinucleotide phosphate, reduced form; EM:
extensive metabolizer; HLM: human liver microsomes; MVC: maraviroc; PM: poor metabolizer
2

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
Abstract
Maraviroc (MVC) is an CCR5 co-receptor antagonist indicated in combination with other
antiretroviral agents for the treatment of CCR5-tropic human immunodefinciency virus-1 (HIV-
1
) infection. In this study, the metabolism of MVC was investigated in human liver microsomes
to delineate the relative roles of CYP3A4 and CYP3A5. MVC is metabolized to five
hydroxylated metabolites, all of which were biosynthesized and identified using mass and NMR
spectroscopy. The sites of metabolism were the 2- and 3-positions of the 4,4-difluorocyclohexyl
moiety and the methyl of the triazole moiety. Absolute configurations were ultimately
ascertained by comparsion to authentic standards. The biosynthesized metabolites were utilized
for quantitative in vitro experiments in liver microsomes using cyp3cide, a selective inactivator
of CYP3A4. (1S,2S)-2-OH-MVC was the main metabolite representing approximately half of
total metabolism, and CYP3A5 contributed approximately 40% to that pathway in microsomes
from CYP3A5 *1/*1 donors. The other four metabolites were almost exclusively metabolized by
CYP3A4. (1S,2S)-2-hydroxylation also correlated to T-5 N-oxidation, a CYP3A5 specific
activity. These data are consistent with clinical pharmacokinetic data wherein CYP3A5
extensive metabolizer subjects showed a modestly lower exposure to MVC.
3

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
Introduction
Maraviroc (MVC) is a human chemokine (C-C motif) receptor 5 (CCR5) antagonist indicated in
combination with other antiretroviral agents for the treatment of CCR5-tropic human
immunodefinciency virus-1 (HIV-1) infection (Dorr et.al. 2005). HIV-1 strains use CCR5 as a
coreceptor during the transmission stage and in the early stages of HIV disease (Michael et al.,
1997; Philpott, 2003). Blocking the CCR5 receptor prevents entry of HIV into host cells and is
shown to reduce viral load to undetectable levels in both treatment naïve and treatment
experienced populations and is sustained out to five years of follow up. MVC is rapidly
absorbed and extensively metabolized, although unchanged MVC is the major circulating
component in plasma and is the major excreted component after oral dosing. Previous in vitro
experiments showed that MVC undergoes oxidative metabolism mediated by CYP3A4 (Hyland
et al., 2008). Renal clearance contributes 23% of total clearance and the absolute bioavailability
of 100 mg oral MVC dose is 23% (estimated at 33% for 300 mg dose). (Abel et al., 2008, 2009;
Vourvahis, et al., 2013). A recent publication by Lu et al. (2014) described the effects of the
CYP3A5 genotype on MVC plasma concentrations. Those results suggested that CYP3A5 may
play a prominent role in the metabolism of MVC in subjects with functional CYP3A5 alleles.
This was in line with data in human liver microsomes (HLMs) showing that the estimated
CYP3A5 contribution to MVC metabolism in HLMs from wild-type CYP3A5 *1/*1 donors
(
extensive metabolizers; EMs) was 32% compared with only 2% in HLMs from CYP3A5 *3/*3
donors (poor metabolizers; PM) (Tseng et al., 2014).
In this paper we describe the biosynthesis and identification of four metabolites that are
hydroxylated on the difluoro cyclohexyl moiety, out of eight possible regio- and stereoisomers,
along with a fifth wherein hydroxylation is on the methyltriazole moiety. The contribution of
4

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
CYP3A5 to their formation was quantitatively delineated through the use of pooled human liver
microsomes from CYP3A5 *1/*1 genotype donors, the CYP3A4 selective inactivator cyp3cide
(
Walsky, et al., 2012), the CYP3A5 selective reaction T-5 N-oxidation (Li, et al., 2014), and
recombinant heterologously expressed CYP3A4 and CYP3A5.
5

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
Materials and Methods
Materials. MVC, d -MVC, T-5 (methyl 2-(4-aminophenyl)-1-oxo-7-(pyridin-2-ylmethoxy)-4-
3,4,5-trimethoxyphenyl)-1,2-dihydroisoquinoline-3-carboxylate), T-5 N-oxide, and cyp3cide
5
(
were prepared at Pfizer (Groton, CT). OH-MVC and deuterated OH-MVC isomers were
biosynthesized as described below; chemical synthesis of all eight 2- and 3-OH-MVC isomers
are described in the Supplemental Information. Ketoconazole, terfenadine, and NADPH was
obtained from Sigma-Aldrich (St. Louis, MO). A 50-donor pool of human liver microsome
(
HLM-102; equal females and males) was prepared under contract by BD Gentest (Woburn,
MA). Single genotyped human liver microsomes (HLM) used in this study are similar to those
from Walsky et al. (2012). [CYP3A5 *1/*1: HH47, BD HH785, BD HH867, BD HH860, HH86,
HH103, HH104, and HH107; CYP3A5 *1/*3: HH1, HH2, HH8, HH9, HH80, HH89, HH90,
HH91, HH92, HH108, HH117, HH48, and HH100; CYP3A5 *3/*3: BDHH189, HH25, HH116,
HH118, HH74, HH75, HH98] were obtained from BD Biosciences (San Jose, CA). The BD
designation in front of some microsomal donor samples indicates the sample has been genotyped
and immunoquantified for CYP3A4 and CYP3A5 by BD Biosciences; otherwise, the donor
characterizations were conducted by Pfizer, Inc. Pooled lots of HLM CYP3A5 *1/*1 (from 5
donors; 2 females, 3 males) and HLM CYP3A5 *3/*3 (from 6 donors; 1 female, five males)
were prepared by pooling equal volumes of some of the individual lots above such that the
resulting enzymatic activity (measured by 6β-OH testosterone formation) is similar between the
two lots. Recombinant heterologously expressed CYP3A4 and CYP3A5 were prepared under
contract by Panvera Corp. (Madison, WI). All other chemicals and reagents were from standard
suppliers.
6

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
Metabolism of Maraviroc by Recombinant CYP3A4 and 3A5. MVC (10 µM) was incubated
with recombinant human CYP3A4 or CYP3A5 both at 100 pmol/mL, in 100 mM potassium
o
phosphate buffer (pH 7.4) with MgCl (3.3 mM) and NADPH (1.3 mM) at 37 C in a shaking
2
water bath for 1 hr. The incubation was commenced with the addition of NADPH and
terminated with the addition of acetonitrile (5 mL). The mixture was spun in a centrifuge at
1
700g for 5 min, the supernatant was transferred to a vacuum centrifuge, and the liquid removed
in vacuo. The residue was reconstituted in 0.2 mL water containing methanol (5%) and formic
acid (1%) and analyzed by UHPLC-UV-MS.
Separation of OH-MVC Isomers by UHPLC-UV-MS. Samples (10 µL) were injected onto a
Thermo Orbitrap Elite mass spectrometer with an Accela UHPLC-UV system. Separation was
o
effected on an Acquity HSS T3 column (2.1 x 100 mm 1.8µm; Waters) maintained at 45 C. The
mobile phase was composed of 0.1% formic acid in water (A) and methanol (B) at a flow rate of
0
2
.4 mL/min. The mobile phase gradient was as follows: 0-15 min linear gradient from 15% B to
5% B, linear gradient to 90% B at 19 min, held at 90% B for 1 min, and re-equilibration to
initial conditions for 1 min. The UV detector was monitored from 200 to 400 nm. The mass
spectrometer was operated in the positive mode, with source conditions and potentials adjusted
to optimize the signal and fragmentation of MVC, and a resolution setting of 30000.
Biosynthesis of OH-MVC Metabolites. MVC (20 µM) was incubated with recombinant
CYP3A5 (5.04 nmoles) in 43 mL KH
2
PO
4
2
(0.1 M) containing MgCl (3.3 mM) and NADPH
o
(
1.3 mM). The incubation was carried out at 37 C for 1.75 hr in a shaking water bath. The
reaction was terminated with the addition of acetonitrile (50 mL), the mixture was centrifuged at
700g for 5 min, and the supernatant was reduced in vacuo to ~10 mL. To the remaining
1
material was added formic acid (0.5 mL), acetonitrile (0.5 mL), and water to a total volume of 50
7

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
mL, and the mixture was spun in a centrifuge for 30 min at 40000g. The resulting supernatant
was applied through a Jasco pump onto a Polaris C18 column (4.6 x 250 mm; 5µm) at a flow
rate of 0.8 mL/min. After the entire volume was applied, the column was moved to a Thermo-
Finnigan Surveyor HPLC-UV in line with an LTQ MS, and a mobile phase gradient was applied
to elute the metabolites. The mobile phase consisted of 0.1% formic acid in water (A) and
acetonitrile (B) at a flow rate of 0.8 mL/min. The composition began at 1% B followed by an
immediate increase to 5% B held for 4 min, followed by a linear gradient to 40% B at 100 min,
an immediate increase to 95% B to wash the column for 10 min, and reequilibration to initial
conditions for 10 min. The eluent passed through the UV detector and was then split
approximately 10:1 between a fraction collector and the mass spectrometer. Fractions were
collected every 20 sec. Fractions potentially containing metabolites of interest were injected on
the UHPLC-MS system described above to test for identity and apparent purity. Fractions were
combined as appropriate, and the solvent was removed in vacuo.
NMR Sample Analysis. Isolated samples and standards were reconstituted in 0.04 mL of
methanol-d “100%” (Cambridge Isotope Laboratories, Andover, MA) and placed in a 1.7 mm
4
1
13
NMR tube under a dry argon atmosphere. The H and C spectra were referenced using residual
1
13
3
methanol-d ( H δ=3.35 ppm relative to TMS, δ=0.00, C δ=49.5 ppm relative to TMS, δ=0.00).
NMR spectra were recorded on a Bruker Avance 600 MHz (Bruker BioSpin Corporation,
Billerica, MA) controlled by Topspin V3.2 and equipped with a 1.7 mm TCI Cryo probe. 1D
spectra were recorded using an approximate sweep width of 8400 Hz and a total recycle time of
approximately 7 s. The resulting time-averaged free induction decays were transformed using an
exponential line broadening of 1.0 Hz to enhance signal to noise. The 2D data were recorded
using the standard pulse sequences provided by Bruker. At minimum a 1K x 128 data matrix
8

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
was acquired using a minimum of 2 scans and 16 dummy scans with a spectral width of 10000
Hz in the f2 dimension. The 2D data sets were zero-filled to at least 1k data point. Post-
acquisition data processing was performed with either Topspin V3.2 or MestReNova Mnova 9.1.
1
Quantitation of NMR isolates was performed by external calibration against the H NMR
spectrum of either a 1 mM maleic acid or a 5 mM benzoic acid standard using the ERETIC2
function within Topspin V3.2.
Metabolism of 2- and 3-OH-MVC Diastereomers to 2,3-DiOH-MVC by Cytochrome
P4503A4. OH-MVC isomers (10 µM) were each individually incubated with recombinant
CYP3A4 (100 pmol/mL) in 0.5 mL potassium phosphate buffer (100 mM, pH 7.4) containing
o
MgCl (3.3 mM) and NADPH (1.3 mM) at 37 C for 1 hr. Incubations were terminated by
2
addition of acetonitrile (2.5 mL) and spun in a centrifuge at 1700g for 5 min. The liquid was
removed by vacuum centrifugation in a Genevac and the residue was reconstituted in water
containing formic acid (1%) and methanol (5%). The samples (5 µL) were injected onto a
Thermo Orbitrap Elite mass spectrometer with an Accela UHPLC-UV system. Separation was
effected on an Acquity HSS T3 column was a (2.1 x 100 mm 1.8µm) and two different mobile
phase gradients were used. (Two conditions were used to better ensure that dihydroxy MVC
metabolites were indeed the same from incubations of different starting materials.) One mobile
o
phase was 0.1% formic acid in water (A) and methanol (B), at a temperature of 45 C. The flow
rate was 0.4 mL/min. The mobile phase gradient was as follows: 0-15 min linear gradient from
5
% B to 25% B, linear gradient to 90% B at 19 min, held at 90% B for 1 min, and re-
equilibration to initial conditions for 1 min. The UV detector was monitored from 200 to 400
nm. The second gradient was comprised of mobile phase A of 10 mM ammonium acetate and
mobile phase B was methanol. The gradient started at 30% B and increased linearly to 60% B at
9

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
1
5 min, raised to 90% B at 19 min, held for 1 min, then re-equilibrated for 1 min at the initial
conditions. The mass spectrometer was operated in the positive mode, and m/z 546
corresponding to the protonated molecular ion of diOH-MVC) and its fragment ions were
(
monitored.
Substrate Saturation Incubation in Human Liver Microsomes and Human Recombinant
CYP3A4 and CYP3A5. MVC (1-200 µM) was incubated in 100 mM potassium phosphate
buffer (pH 7.4), 3.3 mM MgCl , and 0.05 mg/mL microsomal protein (HLM-102, HLM
2
CYP3A5*1/*1, and HLM CYP3A5 *3/*3) or 10 pmol/mL recombinant P450 enzymes
(
rhCYP3A4 and rhCYP3A5). Incubations were conducted in a 37°C heat block for 15 min.
These incubation conditions were selected after initial exploration of protein/enzyme
(
concentrations and incubation times that provided linear reaction velocities.) Reactions were
terminated by transferring 50 μL aliquots of the incubation mixture to 100 μL of 50/50
5
acetonitrile/methanol containing internal standard (mixture of d -OH-MVC metabolites),
followed by vortex mixing. Samples were centrifuged (1439 g) for 10 minutes and 120 μL of
supernatant was transferred to clean 96-well plates. The supernatants were dried down under N
and reconstituted in 100 μL 20/80 methanol/0.1% formic acid in water. Samples were assayed
using LC-MS/MS. Incubations for enzyme kinetic determination was conducted with NADPH,
in triplicate.
2
Selective Inhibition of CYP3A4 and CYP3A5 Activity in Human Liver Microsomes. MVC
(
8 µM) was incubated with HLM-102, HLM CYP3A5*1/*1, or HLM CYP3A5 *3/*3 in the
presence of selective chemical inhibitors ketoconazole (1μM; for CYP3A) and CYP3cide (1μM;
for CYP3A4), for the determination of CYP3A4 and CYP3A5 isoform contribution. General
1
0

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
incubation conditions (protein concentration and incubation time) and sample preparation were
as described for substrate saturation experiments (vide supra).
For incubations with ketoconazole, a mixture containing microsomes, inhibitors, and NADPH or
buffer (-NADPH control) was warmed for 5 min prior to initiating the reactions by the addition
of substrate. Incubations containing CYP3cide were preincubated with microsomes, inactivator,
and NADPH or buffer (-NADPH control) for 10 minutes to achieve complete inactivation of
CYP3A4 prior to the addition of MVC to initiate the reactions. Chemical inhibition experiments
were conducted in triplicate.
LC-MS/MS Method. Supernatants from the terminated incubation mixtures were analyzed
using a liquid chromatography mass spectrometry system (LC-MS/MS) consisting of an
API6500 mass spectrometer (AB Sciex, Framingham, MA) equipped with an electrospray
source, Agilent 1290 Binary UPLC system (Agilent Technologies, Santa Clara, CA) and a Leap
CTC HTS PAL autosampler. Chromatography was carried out on a Waters Acquity UPLC HSS
T3 column (2.1 x 100 mm; 1.8µ particle size) with a mobile phase comprised of water (A) and
methanol (B) containing 0.1% formic acid at a flow rate of 0.4 mL/min. Samples (0.01 mL)
were injected and the mobile phase composition was maintained at 20% B for 6.5 min followed
by a linear gradient to 35% B over the next 6.5 min, and followed by an increase to 90% B over
0
.5 min. It was held at 90% B for another 0.5 min before returning to initial conditions and re-
equilibrating to initial conditions for 2 min. Optimized transitions and parameters of MVC,
metabolites, and stable label internal standards used for mass spectrometric detection on the AB
Sciex Triple Quad 6500 are listed in Supplemental Table S1. Integration and quantitation of
metabolites and IS molecule peak areas were performed using Analyst v1.6.2 (AB Sciex,
Framingham, MA) and to derive the analyte to IS peak area ratios. Standard curves for the
1
1

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
quantitation of metabolite concentration were prepared from plots of area ratio vs concentration
2
and analyzed using a linear regression with either 1/x or 1/x weighting.
Correlation Analysis with T5. MVC (8 µM) and T5 (5 µM) were incubated in HLMs from
individual donors in triplicate. For MVC, general incubation conditions (protein concentration
and incubation time) and sample preparation were as described for substrate saturation
experiments. For T5, experimental and analysis conditions were similar to that described by Li,
et al. (2014). In general, final incubation mixtures contained 100 mM potassium phosphate
buffer (pH 7.4), 3.3 mM MgCl , and 0.1 mg/mL microsomal protein. Incubations were
2
conducted in a 37°C heat block. Reactions were terminated after 20 min by transferring 50 μL
aliquots of the incubation mixture to 100 μL of acetonitrile containing internal standard
(
terfenadine; 2 ng/mL), followed by vortex mixing. Samples were centrifuged (1439 g) for
0 min and 120 μL of supernatant was transferred to clean 96-well plates. Supernatants from the
1
terminated incubation mixtures were analyzed using the aforementioned liquid chromatography
mass spectrometry system. Chromatography was carried out on a Waters Acquity UPLC HSS T3
column (2.1 x 50 mm; 1.8µ particle size) with a mobile phase comprised of water (A) and
acetonitrile (B) containing 0.1% formic acid at a flow rate of 0.4 mL/min. Samples (0.01 mL)
were injected and the mobile phase composition was maintained at 5% B for 0.5 min followed
by a linear gradient to 95% B over 2 min and held at 95% B for another 0.5 min before returning
to initial conditions and reequilibrating to initial conditions for 1 min. T5 N-oxide and
terfenadine were monitored using mass transitions of m/z 584.2-476.2 and m/z 472.1 -436.1,
respectively.
1
2

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
Results
Profile of OH-MVC Metabolites Generated by Cytochrome P450 3A4 and 3A5. MVC was
incubated with recombinant human P450 3A4 and 3A5 and the incubation extracts were
analyzed by HPLC-MS. The extracted ion chromatograms for hydroxylated MVC are shown in
Figure 1. Five metabolites were resolved, observed eluting between 9 and 18 min. The first four
are 2- and 3-hydroxylations on the difluorocyclohexyl moiety and the fifth one is the hydroxyl
on the triazole portion (see below). All five were generated by both enzymes, however the
intensity of the peak eluting at 9.6 min was greater in the CYP3A5 incubation than the 3A4
incubation. The peaks were shown to coelute with authentic and biosynthesized standards of
individual diastereomers of 2- and 3-OH-MVC (Supplemental Information; Figures S1-S5).
Also, the other four standards of 2- and 3-OH-MVC isomers were shown to elute at different
retention times (Supplemental Figure S6).
Identification of OH-MVC Metabolites. OH-MVC metabolites were biosynthesized using
CYP3A4 and CYP3A5, isolated by HPLC, and regiochemical sites of hydroxylation were
assigned using high resolution mass spectrometry and NMR spectroscopy. The NMR spectra of
metabolites were compared to that of MVC. Exact stereochemical configurations of the
hydroxyl isomers on the 4,4-difluorocyclohexyl moiety were ultimately assigned by comparison
to authentic standards of all eight possible 2- and 3-hydroxy diastereomers.
2
-OH-MVC Isomers. The biosynthesis of OH-MVC metabolites using CYP3A5 and CYP3A4
yielded five products. The ones eluting at ~37-40 min (Fraction 1) and ~51 min (Fraction 4) in
the preparative purification (Supplemental Figure S7) were shown to be 2-hydroxy isomers
present in CYP3A4 and 3A5 incubation extracts eluting at 9.7 and 17.4 min (Figure 1). The MS
n
1
3

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
data showed protonated molecular ions of m/z 530.3285 to 530.3294 (-3 to -1 ppm) and
diprotonated molecular ions of m/z 265.6677 to 265.6683 (-4 to -1 ppm), with fragment ions of
m/z 405, 296, 117, and 106 which are diagnostic of the site of hydroxylation as being on the 4,4-
difluorocyclohexyl moiety (Supplemental Figures S8 and S9). However, distinction between the
2
- and 3-positions required the application of NMR spectroscopy.
1
In fractions 1 and 4 there is a new resonance in the H spectrum of each isolate integrating to a
1
single hydrogen at δ 3.94 (td, J = 10.9, 4.7 Hz) not observed in the H spectrum of MVC, Figure
1
13
2
A. Multiplicity edited H – C HSQC data correlate these resonances to a carbon resonance
with a chemical shift of 67.6 ppm (Supplemental Figures S16 and S19). This is consistent with
the oxidation of one of the methylenes of the difluorocyclohexane. Initial NMR characterization
of isolates from clinical samples assigned these resonances as the HCOH methine of the trans 2-
hydroxy MVC isomer. When spectra from fractions 1 and 4 are compared with similarly
acquired spectra from a synthetic standard of the 2-hydroxy MVC trans isomers (fraction 1 = 1S,
1
13
2
S and fraction 4 = 1R, 2R) there are no appreciable differences in the H/ C chemical shifts,
coupling patterns or coupling constants of these resonances (Supplemental Figures S16 and S19).
Therefore the structure of the fractions 1 and 4 are assigned as 2-OH-MVC trans isomers. These
structures are further supported by the HPLC coelution of the two biosynthesized standards with
the authentic 1S, 2S and 1R, 2R diastereomers and not the 1R, 2S or 1S, 2R diastereomers
(
Supplemental Figures S1, S2, and S6). Finally, in a previous report (Abel, et al., 2008), these
metabolites were ambiguously designated as H7, H8, H10, and/or H11. The present data
identifies (1S,2S)-2-OHMVC as H8 and (1R,2R)-2-OH-MVC as H11 which represented 7 and
5
% of total dose in excreta, respectively (Supplemental Table S2).
1
4

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
3
4
3
-OH-MVC Isomers. In the aforementioned biosynthesis, two other isolated peaks eluting at ~43-
5 min (Supplemental Figure S7) were shown to be 3-hydroxy isomers present in CYP3A4 and
A5 incubation extracts eluting at 11.8 and 12.5 min (Figure 1). As with the 2-hydroxyisomers,
n
MS data showed protonated molecular ions of m/z 530.3287 to 530.3291 (-3 to -2 ppm) and
diprotonated molecular ions of m/z 265.6678 to 265.6682 (-3 to -2 ppm), with fragment ions of
m/z 405, 296, 117, and 106 which are diagnostic of the site of hydroxylation as being on the 4,4-
difluorocyclohexyl moiety (Supplemental Figures S10 and S11). NMR spectroscopy was
required to delineate the 3-position as the regiochemical site of hydroxylation. As with fractions
1
1
and 4, there are new resonances in the H spectra of fractions 2 and 3 integrating to a single
hydrogen at δ 3.90 (broad singlet, fraction 2) and δ 3.78 (ddt, J = 20.1, 11.2, 4.5 Hz, fraction 3)
1
1
13
not observed in the H spectrum of MVC (Figure 2B). Multiplicity edited H – C HSQC data
correlate these resonances to a carbon resonances with a chemical shifts of δ 67.1(fraction 2) and
δ 69.6 (fraction 3), (Supplemental Figures S17 and S18). Again this is consistent with the
oxidation of one of the methylenes of the difluorocyclohexane. When spectra from fractions 2
and 3 are compared with similarly acquired spectra from synthetic standards of the 3-hydroxy
1
13
MVC isomers there are no appreciable differences in the H/ C chemical shifts, coupling
patterns or coupling constants of these resonances (Figure 2B). Therefore the structure of the
fractions 2 and 3 are assigned as 3-OH-MVC isomers. Unlike the 2-OH-MVC diastereomers,
the chemical synthesis of the 3-hydroxy isomers yielded two sets of stereoisomers (cis and trans)
but absolute configurations were not known. Absolute stereochemistry of the 3-OH-MVC
metabolites was elucidated using a ‘cross-metabolism’ approach (vide infra).
HydroxymethylMVC. The fifth hydroxyl metabolite of MVC generated by CYP3A was
biosynthesized and eluted at ~63-65 min (Fraction 5; Supplemental Figure S7) and corresponded
1
5

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
to the metabolite eluting at 17.8 min in CYP3A4 and CYP3A5 incubation extracts (Figure 1 and
Supplemental Figure S5). The protonated and diprotonated ions of m/z 530.3290 and 265.6680
(
-3 to -2 ppm) show addition of a single oxygen. Unlike the other four metabolites above, the
fragmentation was different (Supplemental Figure S12). The fragment ion at m/z 389 limits the
site of hydroxylation to the triazole moiety, since this ion represents no alteration to the rest of
1
the molecule. The H spectrum of fraction 5 contains all of the aromatic and aliphatic
1
13
resonances of MVC except the methyl of the triazole (Figure 3). Additionally, in the H- C
multiplicity edited HSQC there is a new methylene cross peak (δ 1H 4.84/ δ 13C 54.1) not
observed in MVC (Supplemental Figure S20). These data are consistent with the oxidation of
the methyl of the triazole to an alcohol, 4,4-difluoro-N-((S)-3-((1R,3R,5S)-3-(3-
(
hydroxymethyl)-5-isopropyl-4H-1,2,4-triazol-4-yl)-8-azabicyclo[3.2.1]octan-8-yl)-1-
phenylpropyl)cyclohexane-1-carboxamide. This metabolite corresponds to the previously
described H13 metabolite from Abel, et al., (2008) which was excreted as 10% of the dose of
MVC (Supplemental Table S2).
Elucidation of the Absolute Stereochemical Configuration of the 3-OH-MVC Metabolites.
At the time of their synthesis, the specific configurations of the four 3-OH-MVC diastereomers
was not known, however the cis- and trans-configurations were known. One diastereomer in
each of these preparations matched a CYP3A generated metabolite. Since the absolute
configuration of all four 2-OH-MVC isomers was already known, an approach was taken
wherein all eight synthetic 2- and 3-OH-MVC isomers were subjected to further metabolism by
CYP3A4 with the intent that specific 2,3-diOH-MVC metabolites would be generated from the
individual 2- and 3-OH-MVC substrates and would have identical HPLC-MS properties. If this
were the case, then using the known configuration at the 1-position of the 2-OH-MVC which
1
6

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
would be unchanged in a 2,3-dihydroxy metabolite, the configuration of the 1 position of the 3-
hydroxy metabolite that yields the same 2,3-dihydroxy metabolite can be inferred. Since this
will be one of a pair of diastereomers known to be either cis or trans, the 3-position absolute
configuration can be concluded (see Figure 4).
The cis 3-hydroxy isomer eluting at 12.4 min (Figure 1) and (1S,2S)-2-OH-MVC yielded
the same diOH-MVC metabolite eluting at 8.9 min (Supplemental Figure S13) with a protonated
n
molecular ion at m/z 546.3259 (0.9 ppm). The MS spectra showed fragment ions of m/z 421,
3
12, 117, and 106, which narrows the site of hydroxylation to the 4,4-difluorocyclohexyl moiety
(
Supplemental Figure S14). To confirm that this dihydroxy metabolite generated by the 2- and 3-
OH-MVC substrates is the same, the sample was analyzed by HPLC using a different mobile
phase condition, and the peaks continued to coelute (Supplemental Figure S13). The known
configuration of the 2-OH-MVC substrate at the 1-position is S. The cis 3-OH-MVC substrate
that yields the same dihydroxy metabolite were either 1S,3S or 1R,3R, but since it yielded the
same dihydroxy metabolite as the (1S,2S)-2-OH-MVC isomer, it can be concluded that the
CYP3A generated 3-OH-MVC metabolite that elutes at 12.4 min is the 1S,3S isomer (Figure 8).
An identical logic and approach was applied to elucidate the absolute stereochemistry of the 3-
OH-MVC metabolite eluting at 11.8 min (Figure 1). In this case, this 3-hydroxy metabolite was
shown to be the 1S,3R isomer (Figure 8). The 1S,3S and 1S,3R metabolites correspond to those
designated as H7 and H10 in the human radiolabel mass balance study which represented 8%
and 9% of dose (Abel, et al., 2008; Supplemental Table S2)
Enzyme Kinetics of Maraviroc in Human Liver Microsomes and Recombinant Human
CYPs. Enzyme kinetic parameters and intrinsic clearance values, determined from the formation
1
7

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
of the five metabolites of MVC in rCYP3A4, rCYP3A5, HLM-102, HLM CYP3A5 *1/*1, and
HLM CYP3A5*3/*3, are listed in Table 1 and substrate saturation data are presented graphically
in Figure 5. The results for all metabolites fit to the Michaelis-Menten model. In general, the
estimated K for each metabolite was comparable to each other within each enzyme system and
M
generally ranged from 7-19 µM. However, the CLint,app provided an initial indication of which
recombinant CYP and genotype is preferential to the formation of each MVC metabolite. While
all the metabolites are formed by CYP3A4 and CYP3A5 to a certain extent, (1S,2S)-2-OH-MVC
is predominately formed by rCYP3A5 and HLM CYP3A5*1/*1 with values for CLint,app of 0.87
µL/min/pmol P450 and 15.6 µL/min/mg, respectively. HydroxymethylMVC is mainly formed by
rCYP3A4, HLM-102, and CYP3A5*3/*3 with CLint,app of 0.31 µL/min/pmol P450, 14.0
µL/min/mg and 10.7 µL/min/mg, respectively.
Selective Inhibition of CYP3A4 and CYP3A5. MVC was incubated in the presence and
absence of ketoconazole (2 µM) and CYP3cide (1 µM) in human liver microsomes from either a
50-donor pool or pooled CYP3A5 *1/*1 and CYP3A5*3/*3 donors. The impact on the rate of
formation of each metabolite was examined in order to estimate the relative contributions of
CYP3A4 and CYP3A5. Inhibition data are listed in Supplemental Table S3.
For HLM-102 and CYP3A5*3/*3 pools (CYP3A5 PM), the estimated CYP3A5 contributions
were similar across the five metabolites with less than 12% by CYP3A5. In the CYP3A5*1/*1
pool (CYP3A5 EM), the greatest CYP3A5 contribution was observed only for (1S,2S)-2-OH-
MVC (42%). Contribution by CYP3A5 was relatively low (<13%) for the other four metabolites.
Correlation Analysis. T-5 oxidation to an N-oxide metabolite by CYP3A5 was first reported by
Li, et al. (2014) as a reaction that is highly correlated with the CYP3A5 genotype and CYP3A5
expression levels in human liver microsome and hepatocytes. Formation of the five OH-MVC
1
8

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
metabolites were compared to T-5 N-oxide formation in individual HLM donors genotyped to be
CYP3A5 *1/*1, CYP3A5*1/*3, and CYP3A5*3/*3 expressers. A positive correlation with T-5
N-oxide was observed for the (1S,2S)-2-hydroxy metabolite (Pearson r value of the correlation
was 0.81, P<0.0001; Figure 7) while the other four metabolites had no correlation (Supplemental
Figure S15). This suggests that CYP3A5 genotype has involvement in the metabolic pathway of
the (1S,2S)-2-OH-MVC and not the other metabolites.
Summary of CYP3A Enzyme Contribution to Maraviroc Metabolism Assignment.
Incorporating the estimated apparent intrinsic clearance and fraction of CYP3A5 inhibition of
each OH-MVC metabolite in HLM, the total fraction of metabolism contributed by CYP3A5 can
be estimated in various batches of human liver microsomes. The in vitro apparent intrinsic
clearance (Clint,app) values in CYP3A5*1/*1 HLM were calculated as the ratio of Vmax to KM
M
where Vmax and K are kinetic parameters determined in Table 1. CLint,app is the intrinsic
clearance for metabolite, and CLint,app,all is the sum of in vitro intrinsic clearance values for all
metabolites. The fraction of clearance (fCL) by each metabolite is calculated as the apparent
intrinsic clearance of each metabolite divided by the total apparent intrinsic clearance for MVC.
The fraction of CYP3A5 inhibition was determined from the chemical inhibition study in HLM
incorporating ketoconazole and CYP3cide (Supplemental Table S3). The fraction metabolized
(
f
m
) by CY3A5 for each metabolite is calculated as fCL multiplied by the fraction of CYP3A5
inhibition. The total fraction metabolized by CYP3A5 in the *1/*1 genotype is estimated to be
.25 (Table 2).
Based on the combined results from enzyme kinetics, HLM chemical inhibition, and
correlation analysis assays, CYP3A5 was consistently predicted by both HLM and rhCYPs to
0
1
9

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
contribute the formation of (1S,2S)-2-OH-MVC while the other four hydroxymethylMVC
metabolites were predominantly generated by CYP3A4.
2
0

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
Discussion
The application of metabolite biosynthesis using P450 enzymes, followed by NMR spectroscopy
to identify regiochemical sites of metabolism and to quantitate the concentration of the material
so that it can be used as an analytical standard, is a powerful and facile approach to conducting
drug metabolism studies. In this work, we began by biosynthesizing and isolating the five major
CYP3A-generated hydroxyl metabolites of MVC. These had previously been identified, but
only reported as Markush structures (Abel, et al., 2008). These biosynthesized metabolites were
then used as analytical standards to carry out work to quantitatively assess the potential role of
CYP3A5 in the metabolism of MVC. They were also used to justify investment in a lengthy and
expensive chemical synthesis of all eight possible isomers of 2- and 3-OH-MVC so that the
absolute configuration of these metabolites could be ascertained.
Previous in vitro work reported by Lu, et al., (2012) proposed that CYP3A5 has a major role in
the metabolism of MVC. However, quantitation of the metabolites done in that work as well as
analysis of human plasma samples (Lu, et al., 2014) was based on mass spectral response using
the 4-hydroxyphenyl analogue of MVC that was not the actual metabolite. Because 4-
hydroxyphenyl MVC is structurally dissimilar to the identified (1S,2S)-2-OH-MVC metabolite
in terms of the electronic character of the phenolic versus the aliphatic hydroxyl moieties, it is
likely that the estimation of the M1 concentration reported by Lu et. al. would show some bias
compared to the actual concentrations present in clinical plasma samples. The claim of a
prominent role for CYP3A5 by Lu et.al. (2012) was based primarily on work in recombinant
enzymes without the benefit of a relative activity factor or intersystem extrapolation factor for
this enzyme that is necessary in order to quantitatively relate rate data from recombinant
heterologously expressed enzymes to activities in liver microsomes or in vivo (Chen, et al.,
2
1

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
2
011; Proctor, et al., 2004). The lack of a suitable activity factor makes translating the in vitro
results to in vivo contribution problematic.
In the present work, we have used the actual metabolite standards in calibration curves required
for quantitation by HPLC-MS. We have also employed the CYP3A4 selective inactivator
cyp3cide to quantitatively delineate the role of CYP3A4 vs CYP3A5 in pooled liver microsomes
from CYP3A5 *1/*1 donors to get a more reliable estimate of the contribution of CYP3A5 to all
five of the major in vitro metabolites. From these data, it can be clearly concluded that CYP3A5
has a contributing role in the generation of the (1S,2S)-2-OH-MVC metabolite (which is most
likely the “metabolite 1” described by Lu, et al, 2012), but not the other metabolites. By
defining the individual metabolic pathways, determining the intrinsic clearance of their
formation to gain knowledge of the fraction that each contributes to the whole, and then
delineating the role of CYP3A4 vs CYP3A5 to each the metabolites, we have reconstructed the
total picture of MVC metabolism to show that CYP3A5 contributes approximately 25% to the
metabolism of MVC in CYP3A5 EMs (Figure 8). This is similar to a previous estimate (32%)
made by monitoring depletion of MVC in HLM from a CYP3A5 *1/*1 donor (Tseng et.al.
2
014).
CYP3A5 is subject to a genetic polymorphism wherein a larger percentage of individuals of
African descent express the enzyme (Roy, et al., 2005; Lamba, et al., 2012). Thus, drugs that are
shown to be substrates of CYP3A5 have the possibility to show lower exposure in individuals
possessing one or two copies of the CYP3A5*1 allele (i.e. IMs and EMs) But the quantitative
impact of CYP3A5 must be weighed against the role that other enzymes, especially CYP3A4
have in the overall metabolism, as well as non-metabolic clearance pathways such as the
considerable portion of maraviroc excreted unchanged in urine (Abel, et al., 2008). To date, all
2
2

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
drugs that have been shown to be metabolized by CYP3A5 are also metabolized by CYP3A4,
therefore quantitatively assessing the role of CYP3A5 in the metabolism of a compound has been
challenging up until the introduction of the CYP3A4 selective inactivator cyp3cide (Walsky, et
al., 2012), which permits the activities of CYP3A4 and CYP3A5 to be separated in intact in vitro
models of drug metabolism (i.e. human liver microsomes or hepatocytes). The results from this
work suggest that CYP3A5 would play a minor but measurable role in the metabolic clearance of
MVC in CYP3A5 EM individuals. This has been shown in the clinic wherein CYP3A5 EM
subjects have been shown to have between 26-41% lower exposure to MVC and a higher
(
1S,2S)-2-OH-MVC to MVC exposure ratio, consistent with estimates made from the in vitro
data (Vourvahis, manuscript in preparation; Lu, et al., 2014). However it is also the observation
that these lower exposures are still in excess of exposures associated with near-maximal MVC
virologic efficacy, which may explain why there was no observation of decreased efficacy in
CYP3A5 EM patients in the phase 3 efficacy trials of MVC (McFadyen 2008, Vourvahis, et al.,
2
015). Additionally, it should be noted that in clinical practice, many HIV patients receive MVC
in the presence of CYP3A-inhibiting protease inhibitors. Such drugs are potent inhibitors of
both CYP3A4 and CYP3A5 (Ernest et.al. 2005, Granfors et.al. 2006) and therefore this could
potentially diminish the overall role of CYP3A (both 3A4 and 3A5) in the overall metabolism of
MVC in patients receiving these drugs.
Overall, these results demonstrate that CYP3A5 can contribute partially to the generation of the
(
1S,2S)-2-OH-MVC metabolite. The use of biosynthesized metabolites with quantitative NMR
spectroscopy served well to enable the subsequent P450 reaction phenotyping experiments
needed to quantitatively evaluate the relative roles of CYP3A4 and 3A5 in the metabolism of
MVC. Such an approach can be employed to tackle quantitative in vitro drug metabolism
2
3

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
experiments for other compounds, even when synthetic standards of metabolites are difficult or
impossible to obtain.
2
4

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
Acknowledgements
The authors thank the following individuals: Olivier Dirat of the Pharmaceutical Research and
Development Department at Pfizer, Sandwich, UK for coordination of the synthesis of 2- and 3-
OH-MVC standards conducted at Peakdale Molecular, Jan Szeliga of the Pharmaceutical
Research and Development Department, Pfizer, Groton, USA for preparative chromatography of
the four synthetic 3-OH-MVC isomers, Mark Savage, Torren Peakman, and Angus Nedderman,
formerly of Pfizer, Sandwich, UK who conducted excellent early work on maraviroc metabolism
that was used as a starting point for these investigations.
Authorship Contributions
Participated in research design: Tseng, Fate, Walker, Goosen, Obach
Conducted experiments: Tseng, Walker, Obach
Performed data analysis: Tseng, Fate, Walker, Obach
Wrote or contributed to the writing of the manuscript: Tseng, Fate, Walker, Goosen, Obach
2
5

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
References
Abel S, Russell D, Whitlock LA, Ridgway CE, Nedderman ANR, and Walker DK. (2008) Br J
Clin Pharmacol. 65(Suppl 1): 60-67.
Abel S, Back DJ, Vourvahis M. (2009) Review. Maraviroc: pharmacokinetics and drug
interactions. Antiviral Therapy 14:607-618.
Chen Y1, Liu L, Nguyen K, Fretland AJ (2011) Utility of intersystem extrapolation factors in
early reaction phenotyping and the quantitative extrapolation of human liver microsomal
intrinsic clearance using recombinant cytochromes P450. Drug Metab Dispos. 39: 373-82.
Dorr P, Westby M, Dobbs S, Griffin P, Irvine B, Macartney M, Mori J, Rickett G, Smith-
Burchnell C, and Napier C, et al. (2005) Maraviroc (UK-427,857), a potent, orally bioavailable,
and selective small-molecule inhibitor of chemokine receptor CCR5 with broad-spectrum anti-
human immunodeficiency virus type 1 activity. Antimicrob Agents Chemother 49:4721–4732.
Ernest CS, 2nd, Hall SD, and Jones DR (2005) Mechanism-based inactivation of CYP3A by
HIV protease inhibitors. J Pharmacol Exp Ther 312:583–591.
Granfors MT, Wang JS, Kajosaari LI, Laitila J, Neuvonen PJ, and Backman JT (2006) Differ-
2
6

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
ential inhibition of cytochrome P450 3A4, 3A5 and 3A7 by five human immunodeficiency virus
HIV) protease inhibitors in vitro. Basic Clin Pharmacol Toxicol 98:79–85.
(
Hyland R, Dickins M, Collins C, Jones H, Jones B (2008) Maraviroc: in vitro assessment of
drug-drug interaction potential. Br J Clin Pharmacol. 66:498-507.
Lamba J, Hebert JM, Schuetz EG, Klein TE and Altman RB (2012) PharmGKB summary: very
important pharmacogene information for CYP3A5. Pharmacogenet Genomics. 22: 555–558.
Li X, Jeso V, Heyward S, Walker GS, Sharma R, Micalizio GC, and Cameron MD (2014)
Characterization of T-5 N-Oxide Formation as the First Highly Selective Measure of CYP3A5
Activity. Drug Metab Dispos. 42: 334–342.
Lu Y, Hendrix CW, Bumpus NN (2012) Cytochrome P450 3A5 plays a prominent role in the
oxidative metabolism of the anti-human immunodeficiency virus drug maraviroc. Drug Metab
Dispos. 40: 2221-30.
Lu Y, Fuchs EJ, Hendrix CW, Bumpus NN. (2014) CYP3A5 genotype impacts maraviroc
concentrations in healthy volunteers. Drug Metab Dispos 42: 1796–1802.
2
7

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
McFadyen L, Jacqmin P, Wade JR, and Weatherley B (2008) Maraviroc exposure-efficacy
<50copies/mL) analysis in HIV-1–infected treatment-naive subjects — ITT population (MERIT
(
study), in: 17th International AIDS Conference, Mexico City, Mexico, August 3–8, 2008.
Poster TUPE0053.
Michael NL, Chang G, Louie LG, Mascola JR, Dondero D, Birx DL, Sheppard HW. (1997) The
role of viral phenotype and CCR-5 gene defects in HIV-1 transmission and disease progression.
Nat Med. 3:338-40.
Philpott SM. (2003) HIV-1 coreceptor usage, transmission, and disease progression.
Curr HIV Res. 2:217-27.
Proctor NJ, Tucker GT, Rostami-Hodjegan A.(2004) Predicting drug clearance from
recombinantly expressed CYPs: intersystem extrapolation factors. Xenobiotica. 34:151-78.
Roy JN, Lajoie J, Zijenah LS, Barama A, Poirier C, Ward BJ, Roger M. (2005)
CYP3A5 genetic polymorphisms in different ethnic populations. Drug Metab Dispos. 33: 884-7.
Tseng E, Walsky RL, Luzietti RA, Jr, Harris JJ, Kosa RE, Goosen TC, Zientek MA, Obach RS.
(2014) Relative contributions of cytochrome CYP3A4 versus CYP3A5 for CYP3A-cleared drugs
2
8

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
assessed in vitro using a CYP3A4-selective inactivator (CYP3cide). Drug Metab Dispos
2:1163–1173.
4
Vourvahis M, Fang J, Checchio T, Milton A, Weatherley B, McFadyen L, Heera J. (2013)
Pharmacokinetics, safety, and tolerability of maraviroc in HIV-negative subjects with impaired
renal function. HIV Clin Trials. 14: 99-109.
Vourvahis M, McFadyen L, Heera J, and Clark A (2015) Clinical Relevance of CYP3A5
Genotype on Maraviroc Exposures. Drug Metab Dispos 43: 771-772;
Walsky RL, Obach RS, Hyland R, Kang P, Zhou S, West M, Geoghegan KF, Helal CJ, Walker
GS, Goosen TC, and Zientek MA (2012) Selective Mechanism-Based Inactivation of CYP3A4
by CYP3cide (PF-04981517) and Its Utility as an In Vitro Tool for Delineating the Relative
Roles of CYP3A4 versus CYP3A5 in the Metabolism of Drugs. Drug Metab. Dispos. 40: 1686-
1
697.
2
9

DMD Fast Forward. Published on February 23, 2018 as DOI: 10.1124/dmd.117.079855
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #79855
Legends for Figures
Figure 1. HPLC-MS Chromatograms of Metabolites Generated in Incubations of MVC with
Cytochrome P450 3A4 and 3A5. The chromatograms are extracted ion traces for m/z 530.3278
(
5 ppm width). The peaks are annotated with the notation for 2- and 3-OH-MVC and
hydroxymethylMVC. Top panel: CYP3A4. Bottom panel: CYP3A5.
1
Figure 2. Abbreviated H NMR spectra of A) MVC (bottom), fraction 1: (1S,2S)-2-OH-MVC
(
middle), and fraction 4: (1R,2R)-2-OH-MVC (top); B) MVC (bottom), fraction 2: (1S,3R)-3-
OH-MVC (middle), and fraction 3: (1S,3S)-3-OH-MVC (top).
1
Figure 3. Abbreviated H NMR spectrum of hydroxymethylMVC and MVC.
Figure 4. Metabolism of 2- and 3-OH-MVC Diastereomers by Cytochrome P450 3A4 to 2,3-
DiOH-MVC.
Figure 5. Substrate saturation curves of (1S,2S), (1R,2R), (1S,3S), (1S,3R), and
hydroxymethylMVC formation in pooled human liver microsomes, pooled CYP3A5 *1/*1
human liver microsomes, pooled CYP3A5 *3/*3 liver microsomes, and recombinant human
CYP3A4 and CYP3A5. Plots on the left are in human liver microsomes and plots on the right are
in recombinant systems.
Figure 6. Formation of (1S,2S)-2-OH-MVC, (1R,2R)-2-OH-MVC, (1S,3S)-3-OH-MVC,
(
1S,3R)-3-OH-MVC, and hydroxymethylMVC in pooled human liver microsomes, pooled
CYP3A5 *1/*1 human liver microsomes, pooled CYP3A5 *3/*3 liver microsomes in the
presence and absence of selective CYP3A and CYP3A4 inhibitors, ketoconazole and CYP3cide,
respectively. Percent values represent the calculated contribution by CYP3A5.
3
0