Characterization of the in vitro CYP450 mediated metabolism of the polymorphic CYP2D6 probe drug codeine in horses

Article abstract of DOI:10.1016/j.bcp.2019.07.005

Despite their widespread popularity as sport and companion animals and published and anecdotal reports of vast difference in drug disposition and pharmacokinetics between individuals, studies describing equine drug metabolism are limited. It has been theorized that similar to humans, members of the CYP2D family in horses may be polymorphic in nature leading to differences in metabolism of substrates. This study aims to build on the limited current knowledge regarding P450 mediated metabolism in horses by describing the metabolism of the polymorphic CYP2D6 probe drug codeine in vitro. Codeine, at varying substrate concentrations, was incubated with equine liver microsomes (±UDPGA) and a panel of baculovirus expressed recombinant equine P450s. Parent drug and metabolite concentrations were determined using LC-MS/MS. Incubation of codeine in equine liver microsomes generated norcodeine, morphine, codeine glucuronide and morphine 3- and 6- glucuronide. In recombinant P450 assays, the newly described CYP2D82 was responsible for catalyzing the biotransformation of codeine to morphine (K_m of 247.4 μM and a V_max of 1.6 pmol/min/pmol P450). CYP2D82 is 80% homologous to the highly polymorphic CYP2D6 enzyme, which is responsible for biotransformation of codeine to morphine in humans. CYP3A95, which shares 79% sequence homology with human CYP3A4 and CYP2D50 catalyzed the conversion of codeine to norcodeine (K_m of 104.1 and 526.9 μM, V_max of 2.8 and 2.6 pmol/min/pmol P450). In addition to describing the P450 mediated metabolism of codeine, the current study offers a candidate probe drug that could be used in vivo to study the functional implications of polymorphisms in the CYP2D gene in horses.

Full text of DOI:10.1016/j.bcp.2019.07.005

BiochemicalPharmacology168(2019)184–192
Contents lists available at ScienceDirect
Biochemical Pharmacology
journal homepage: www.elsevier.com/locate/biochempharm
Characterization of the in vitro CYP450 mediated metabolism of the
polymorphic CYP2D6 probe drug codeine in horses
⁎
Heather K. Knycha,b, , Russell W. Badena, Sophie R. Gretlera, Daniel S. McKemiea
a K.L. Maddy Analytical Chemistry Laboratory, School of Veterinary Medicine, University of California, Davis, CA, United States
b Department of Veterinary Molecular Biosciences, School of Veterinary Medicine, University of California, Davis, CA, United States
A R T I C L E I N F O
A B S T R A C T
Keywords:
Equine
Cytochrome P450
Codeine
Metabolism
Polymorphism
Despite their widespread popularity as sport and companion animals and published and anecdotal reports of vast
difference in drug disposition and pharmacokinetics between individuals, studies describing equine drug me-
tabolism are limited. It has been theorized that similar to humans, members of the CYP2D family in horses may
be polymorphic in nature leading to differences in metabolism of substrates. This study aims to build on the
limited current knowledge regarding P450 mediated metabolism in horses by describing the metabolism of the
polymorphic CYP2D6 probe drug codeine in vitro. Codeine, at varying substrate concentrations, was incubated
with equine liver microsomes ( UDPGA) and a panel of baculovirus expressed recombinant equine P450s.
Parent drug and metabolite concentrations were determined using LC-MS/MS. Incubation of codeine in equine
liver microsomes generated norcodeine, morphine, codeine glucuronide and morphine 3- and 6- glucuronide. In
recombinant P450 assays, the newly described CYP2D82 was responsible for catalyzing the biotransformation of
codeine to morphine (Km of 247.4 μM and a Vmax of 1.6 pmol/min/pmol P450). CYP2D82 is 80% homologous to
the highly polymorphic CYP2D6 enzyme, which is responsible for biotransformation of codeine to morphine in
humans. CYP3A95, which shares 79% sequence homology with human CYP3A4 and CYP2D50 catalyzed the
conversion of codeine to norcodeine (Km of 104.1 and 526.9 μM, Vmax of 2.8 and 2.6 pmol/min/pmol P450). In
addition to describing the P450 mediated metabolism of codeine, the current study offers a candidate probe drug
that could be used in vivo to study the functional implications of polymorphisms in the CYP2D gene in horses.
1. Introduction
identified as poor and ultra-rapid metabolizers have been reported [2].
Following codeine ingestion, poor metabolizers have notably lower
Codeine is an opioid and a naturally occurring alkaloid used in
human medicine for relief of mild to moderate pain, as a cough sup-
pressant and as an anti-diarrheal. In most species, codeine undergoes
extensive biotransformation with only 2–7% of the administered dose
being excreted as the parent compound [1]. In humans, 5 different
metabolites have been reported in vivo, including codeine 6-glucuronide
(81%), norcodeine (2.2%), morphine (0.56%), morphine 3-glucuronide
(2.1%) and morphine 6-glucuronide (0.8%) [2]. The analgesic effects of
codeine have been largely attributed to metabolism to morphine, with
morphine reportedly ten times more potent than the parent compound
[3].
In humans, codeine is metabolized by the CYP450 enzyme, CYP2D6
[4–6]. While constituting only about 2–4% of the total P450 content in
the liver, CYP2D6 is responsible for the metabolism of 25–30% of all
therapeutic drugs [7]. CYP2D6 is highly polymorphic and notable dif-
ferences in the biotransformation of codeine to morphine in individuals
concentrations of morphine while ultra-rapid metabolizers have mea-
surably higher concentrations that can ultimately result in serious
toxicity, including death [2]. While to the best of the authors’ knowl-
edge, the metabolism of codeine in the horse has not been studied, it
has been theorized, following administration of other proposed CYP2D
substrates in horses, that mutations in the equine CYP2D orthologue
may also lead to notable differences in the metabolism of CYP2D sub-
strates in horses [8]. Additional CYP2D probe drugs would allow for
further characterization of the effects of polymorphisms in the gene that
codes for this enzyme in in vivo studies. In the current study we hy-
pothesized that, similar to humans, codeine is metabolized to morphine
by equine CYP2D enzymes. To that end, the goals of the current study
were to (1) add to the limited knowledge of P450-mediated drug me-
tabolism in the horse by characterizing the in vitro metabolic profile of
codeine and (2) to determine the specific enzymes responsible for the
metabolism of codeine, thereby potentially providing data supporting
⁎ Corresponding author at: K.L. Maddy Analytical Chemistry Laboratory, School of Veterinary Medicine, University of California, Davis, CA 95616, United States.
E-mail address: hkknych@ucdavis.edu (H.K. Knych).
https://doi.org/10.1016/j.bcp.2019.07.005
Received 29 May 2019; Accepted 7 July 2019
Availableonline08July2019
0006-2952/©2019ElsevierInc.Allrightsreserved.

H.K. Knych, et al.
BiochemicalPharmacology168(2019)184–192
Fig. 1. Michaelis-Menten (top) and Eadie Hofstee (bottom) plots for the determination of the apparent Km and Vmax values for codeine metabolism to norcodeine and
morphine by equine liver microsomes. Values on the Michaelis-Menten plot represent the mean ( SD) data from 3 separate incubations at each substrate con-
centration.
the use of codeine as a probe drug for in vivo screening to identify horses
Table 1
Estimates of kinetic parameters for norcodeine formation in incubations with
equine liver microsomes and several recombinant equine P450s.
that are poor and ultra-rapid metabolizers.
2. Materials and methods
Km (μM) Vmax (pmol/min/pmol
CLint
P450)
2.1. Chemicals
Liver microsomes Enzyme #1
Enzyme #2
926.6
196.2
3.0
1.5
0.003
0.008
Saline (0.9%), Tris-HCl, potassium chloride, potassium phosphate,
EDTA, glycerol, dithiothreitol, PMSF, NADPH, CHAPS, cytochrome c,
cytochrome b5, codeine, quinidine, 5-aminolevulinic acid, ferric citrate
and UDPGA were obtained from Sigma-Aldrich (St. Louis, MO).
+ Quinidine
Enzyme #1
Enzyme #2
946.4
284.5
129.2
2.8
1.6
0.22
1.6
2.8
0.24
0.64
2.6
0.003
0.006
0.002
0.006
0.027
0.001
0.002
0.005
CYP2D82
+ Quinidine 247.4
Acetonitrile and water were purchased from Burdick
& Jackson
CYP3A95
CYP3A93
CYP3A94
CYP2D50
104.1
381.7
308.1
526.9
(Muskegon, MS). Methanol was purchased from Thermo Fisher
Scientific (Fairlawn, NJ). All solvents were HPLC grade or better.
2.2. Preparation of liver microsomes
The liver sample used in the preparation of microsomes was
185

H.K. Knych, et al.
BiochemicalPharmacology168(2019)184–192
Fig. 2. Amino acid alignment of CYP2D82, CYP2D50, CYP2D14 and CYP2D6. Key: (*) denotes conserved amino acid residues among the CYP enzymes.
collected from an adult female Thoroughbred horse euthanized for a
separate study previously approved by the Institutional Animal Care
and Use Committee of the University of California at Davis. The horse
was determined to be healthy by physical exam prior to euthanasia and
had not received any therapeutic medications that had the potential to
affect the expression of metabolic enzymes. The liver sample
(100–200 g) was collected within 20 min of the horse being euthanized
and placed in ice cold 0.9% saline for transport back to the laboratory.
186

H.K. Knych, et al.
BiochemicalPharmacology168(2019)184–192
Fig. 3. Comparison of the rates of norcodeine and morphine formation (per pmol of P450) by baculovirus expressed P450 enzymes. The maximum velocity of the
reactions (Vmax) was calculated using the mean values of 3 incubations at each substrate concentration.
The liver sample was rinsed and flushed with ice cold 0.9% saline
using visible arteries, veins or ducts. The sample was blotted dry,
weighed, chopped into small pieces and placed in an ice-cold blender.
The sample was subsequently homogenized by pulse blending several
times, followed by continuous blending on the lowest setting for ap-
proximately 20 s, in 3 volumes of 50 mM Tris-HCl buffer, pH 7.4,
containing 150 mM KCl and 2 mM EDTA. The mixture was further
homogenized with a motor-driven Teflon/glass tissue grinder. The
sample was then subject to a series of differential centrifugation steps.
The mixture was first centrifuged at 9000×g for 20 min followed by
further centrifugation at 100,000×g for 60 min, following removal of
the supernatant and resuspension in the Tris-HCl buffer described
above. This centrifugation step was repeated one additional time and
the final pellet resuspended by hand, using a teflon/glass homogenizer,
in 100 mM phosphate buffer, pH 7.4, containing 1 mM EDTA, 20%
glycerol and 1 mM dithiothreitol. Protein content was determined using
the bicinchoninic acid assay (BCA). P450 levels were determined by
obtaining the difference spectra of sodium dithionate-reduced versus
CO-bubbled samples at 500–400 nm according to the methods of
Omura and Sato [9]. Microsomes were stored at −80 °C until used.
and Vmax values were calculated from the Eadie Hofstee plot.
Additional incubations were conducted in the presence of the
CYP2D6 inhibitor, quinidine (5 µM final concentration). Incubations
and kinetic parameter calculations were as described above.
2.4. cDNA expressed P450
The cDNA for equine CYP1A1, 1A2, 2B6, 2C92, 2D14, 2D50, 3A89,
3A93, 3A94, 3A95, 3A96 and 3A97 were obtained by amplification of
total RNA prepared from equine liver samples and using primers gen-
erated from sequences obtained from GenBank. RNA was isolated from
previously collected liver samples stored in RNA Later (Ambion, Foster
City, CA) and stored at −20 °C. For RNA isolation, liver samples were
homogenized in Trizol Reagent (Life Sciences, Carlsbad, CA) and total
RNA isolated. The quality and concentration of total RNA was assessed
by gel electrophoresis and absorbance of 260 nm, respectively. The PCR
products were cloned into an expression vector (pFastBac1; Invitrogen,
Carlsbad, CA) and sequences confirmed using a commercial sequencing
facility (University of California Davis College of Biological Sciences,
University of California DNA Sequencing Facility). An additional se-
quence designated as CYP2D14-like in GenBank was also amplified as
described above using the forward primer sequence (5′- GGTGAATTC
ATGGGACTGCTGACCGGG) and reverse (5′- GGTCTCGAGCTAGCGGG
GCTCAGCGC). Upon sequencing, the deduced sequence was submitted
to the P450 nomenclature committee for name designation and then
submitted to GenBank (accession nos. MK928327).
The cDNA encoding the pFastBac1/P450 was site specifically
transposed into the baculovirus bacmid as described in the manu-
facturer’s protocol (Bac-to-Bac Expression System; Invitrogen, Carlsbad,
CA) and the high molecular weight bacmid DNA purified from re-
combinant clones was used to transfect TriEx Sf9 cells (Millipore Sigma,
Burlington, MA) for the production of viral stocks.
Cells grown as suspension cultures in serum-free medium supple-
mented with L-glutamine were used to optimize expression and produce
active enzyme. Production of functional protein was performed using
the optimal MOI in the presence of 200 μM 5-aminolevulinic acid and
20 μM ferric citrate. Cell lysates were prepared by pelleting cells at
800×g for 5 min at 4 °C, washing with cold 0.1 M phosphate buffer, pH
7.4, followed by resuspension in 0.1 M phosphate buffer, pH 7.4, with
20% glycerol, 1 mM EDTA, 1 mM PMSF and 1:200 Protease Inhibitor
Cocktail Set III; EDTA-free (MilliporeSigma, Burlington, MA). The re-
2.3. Microsomal incubations
Microsomal incubations were run in triplicate both with and
without UDPGA so as to measure both P450 generated and glucur-
onidated metabolites. A set of “blank” incubations were run in parallel.
These reaction mixes were as described below for the other incubations
without the addition of NADPH. The incubations consisted of 1 mg/ml
protein in 100 mM potassium phosphate buffer (pH 7.4) and 1 mM
CHAPS in a total volume of 250 µl. All reactions (buffer + varying
concentrations of codeine) were pre-incubated at 37 °C for 5 min prior
to initiation of the reaction by the addition of 1 mM NADPH. Reactions
were allowed to proceed for 20 min and were terminated by the addi-
tion of 250 µl of ice-cold acetonitrile. For reactions containing UDPGA,
a final concentration of 5 mM of UDPGA was added to incubations.
Codeine metabolism was measured under linear conditions using sub-
strate concentrations ranging from 0 to 800 µM. Triplicate incubations
were run at each substrate concentration. The velocity (V), calculated
as pmol product/min/pmol of CYP450, at each substrate concentration
was determined using the average value of the triplicate reactions. The
data was evaluated by plotting V vs V/[Substrate] (Eadie Hofstee plot)
and the intercepts calculated by linear regression analysis. Apparent Km
suspended pellet was then homogenized using
a teflon/glass
187

H.K. Knych, et al.
BiochemicalPharmacology168(2019)184–192
Fig. 4. Eadie Hofstee plot for the generation of norcodeine by baculovirus expressed CYP3A95 and CYP2D50 and morphine by baculovirus expressed CYP2D82
following incubation with codeine.
Table 2
homogenizer and centrifuged at 800×g for 5 min at 4 °C to pellet cel-
lular debris, followed by subsequent centrifugation at 100,000×g for
1 h at 4 °C of the resultant supernatant. The supernatant was discarded,
and the pellet resuspended in phosphate storage buffer (0.1 M KPO4, pH
7.4 containing 20% glycerol and 1 mM EDTA) and stored at −80 °C.
P450 activity was determined by obtaining the difference spectra of
sodium dithionate-reduced vs CO-bubbled samples at 500–400 nm [9].
Estimates of kinetic parameters for morphine formation in equine liver micro-
somes and CYP2D82.
Km (μM) Vmax (pmol/min/pmol
CLint
P450)
Liver microsomes Enzyme #1
Enzyme #2
236.7
25.0
1.4
0.6
0.006
0.02
+Quinidine
Enzyme #1
Enzyme #2
401.0
–
1.4
–
0.003
–
2.5. Reductase expression
CYP2D82
247.4
+ Quinidine 405.6
1.6
1.5
0.007
0.004
The cDNA for equine NADPH Cytochrome P450 reductase was ob-
tained by amplification of total RNA prepared from equine liver sam-
ples as described above for recombinant P450s. The cDNA transcripts
were generated by reverse transcription with the forward (5′-GGTGA
ATTCATGGGGGACTCCAACATGGAC) and reverse (5′-GGTCTCGAGCT
188

H.K. Knych, et al.
BiochemicalPharmacology168(2019)184–192
Fig. 5. Norcodeine and morphine formation kinetics in equine microsomal incubation studies in the in the absence (filled circles) and presence (open circles) of the
CYP2D6 inhibitor, quinidine. Values on the Michaelis-Menten plot represent the mean ( SD) data from 3 separate incubations at each substrate concentration.
AGCTCCACACGTCCAGCGA) primers designed against the equine re-
ductase sequence from GenBank (NM_001122655.1). The PCR products
were cloned, sequences confirmed, and protein expressed as described
above for P450s. Specific activity was determined by the cytochrome c
reductase assay with one unit being defined as 1 µmol of cytochrome c
reduced per minute.
acetonitrile following a 30-minute incubation.
2.7. Enzyme activity and kinetics
Codeine metabolism by recombinant P450s was determined in
250 µl reaction volumes, including NADPH CYP450 reductase (amount
determined in optimization reactions with individual P450s), cyto-
chrome b5, rHuman (where appropriate), 1 mM CHAPS, 1 mM NADPH,
100 mM potassium phosphate buffer (pH 7.4) and varying codeine
concentrations. All reactions were incubated at 37 °C for 5 min prior to
initiation of the reaction by the addition of 2.5 pmol of the recombinant
P450 and then allowed to proceed for 30 min. Blank samples consisted
of uninfected TriEx cell homogenate in place of recombinant P450.
Reactions were subsequently terminated by the addition of 250 μl of
ice-cold acetonitrile. The rate of the reaction was measured under linear
conditions, Eadie Hofstee plots created and Km and Vmax calculated as
described previously for liver microsomes. Additional incubations with
the CYP2D6 inhibitor, quinidine, and CYP2D82 were conducted as
described above for other recombinant P450s.
2.6. Optimization of incubation conditions
Linearity of the assay with respect to time for recombinant P450
enzymes was established and all subsequent kinetic studies were con-
ducted within the linear portion of the rate curve. Additionally, prior to
performing kinetic studies, the optimal ratio of recombinant P450 to
equine NADPH cytochrome P450 oxidoreductase and cytochrome b5,
rHuman (where appropriate; Life Sciences, Carlsbad, CA) was de-
termined. Optimization assays were conducted by adding increasing
quantities of equine NADPH cytochrome P450 oxidoreductase to in-
cubations containing recombinant P450, 1 mM CHAPS, 100 mM po-
tassium phosphate buffer (pH 7.4), 1 mM NADPH and 800 µM codeine
in a total volume of 250 μl. The mixture was pre-incubated for 5 min in
a 37 °C shaking water bath prior to the addition of recombinant P450.
Incubation reactions were terminated by the addition of ice-cold
189

H.K. Knych, et al.
BiochemicalPharmacology168(2019)184–192
Acquity UPLC (Milford, MA). The system was operated at a resolution
of 60,000 using positive electrospray ionization (ESI(+)). The spray
voltage was set at 4300 V, sheath gas and auxillary gas were 40 and 10
respectively (arbitrary units), and capillary temperature was 350 °C.
Chromatography employed an Eclipse-XDB-Phenyl 2x150mm, 5 μm
column (Agilent Technologies, Inc., Santa Clara, CA) and a linear gra-
dient of ACN in water with a constant 0.2% formic acid, at a flow rate of
0.4 ml min−1. The initial ACN concentration was held at 3% for
0.2 min, ramped to 35% over 7.8 min, ramped to 80% over 0.2 min
before re-equilibrating for 3.3 min at initial conditions.
Detection and quantification were conducted using full scan accu-
rate mass from 150 to 575 (m/z), with a resolution of 60,000. The re-
sponses were plotted using a 20 ppm data window for the ions: mor-
phine (mass to charge ratio 286.14384 (m/z)), M3G (462.17566 (m/
z)), M6G (462.17566 (m/z)), C6G (476.19147 (m/z)), codeine
(300.15917 (m/z)), and norcodeine (286.14377 (m/z)). Quanbrowser
software (Thermo Scientific) was used to generate calibration curves
and quantitate all samples by linear regression analysis. A weighting
factor of 1/X was used for all calibration curves.
The responses were linear and gave correlation coefficients (R2) of
0.99 or better. The technique was optimized to provide a limit of
quantitation of 0.5 ng/mL for all analytes except morphine-6β-D-glu-
curonide which was 1 ng/mL. The limit of detection was approximately
0.1 ng/mL for all analytes, and 0.5 ng/mL morphine-6β-D-glucuronide.
3. Results
3.1. Kinetics of codeine metabolism in equine liver microsomes
Incubation of codeine with equine liver microsomes generated the
metabolites norcodeine and morphine. In addition to the phase 1 me-
tabolites listed above, the glucuronidated metabolites, codeine 6-glu-
curonide and morphine 3-glucuronide were observed in microsomal
incubations containing UDPGA. The Michaelis-Menton and Eadie-
Hofstee plots for norcodeine and morphine generated in the microsomal
incubations are depicted in Fig. 1. Examination of Eadie-Hofstee plots
and subsequent calculation of a Km and Vmax (Table 1) for both nor-
codeine and morphine revealed biphasic plots suggesting that 2 en-
zymes (high affinity, low Km) and low affinity (high Km), are involved in
both reactions.
3.2. Cloning and sequencing of equine CYP2D82
The sequence amplified using primers generated based on the se-
quence for CYP2D14-like was submitted to the CYP450 Nomenclature
Committee and was given the designation CYP2D82. The sequence
showed a homology of 94% to CYP2D14, 89% to equine CYP2D50 and
80% to human CYP2D6 (Fig. 2).
Fig. 6. Morphine formation kinetics in equine CYP2D82 incubation studies in
the absence (filled circles) and presence (absence) of the CYP2D6 inhibitor,
quinidine. Values on the Michaelis-Menten plot represent the mean ( SD)
data from 3 separate incubations at each substrate concentration.
2.8. Determination of codeine and metabolite concentrations in liver
microsomal and recombinant P450 reactions
3.3. cDNA expressed equine P450s
The major phase 1 metabolites identified in equine microsomal in-
cubations with codeine were morphine and norcodeine. Recombinant
baculovirus-expressed equine P450s were used to examine which P450
isoforms were responsible for generating each metabolite. The oxida-
tive activity of individual equine P450s in forming morphine and nor-
codeine, normalized for P450 content, are shown in Fig. 3. CYP3A95
and CYP2D50 demonstrated the highest rates for conversion of codeine
to norcodeine. Several other P450s, including CYP2D82, CYP3A93 and
CYP3A94 were capable of generating norcodeine but the velocity of the
reaction was much slower. Of all the recombinant P450s tested,
CYP2D82 was the only P450 that generated morphine in incubations
with codeine. The Eadie-Hofstee plots for the major P450s are depicted
in Fig. 4 and kinetic parameters listed in Tables 1 and 2.
The analytical reference standards for codeine, codeine-6β-D-glu-
curonide, morphine, morphine-3β-D-glucuronide, morphine-6β-D-glu-
curonide, and norcodeine were obtained from Cerilliant (Round Rock,
TX). The six analytes were combined into one solution and and working
solutions prepared by dilution of the 1 mg/mL or 0.1 mg/mL stock so-
lutions with methanol to concentrations of 0.1, 1, and 10 ng/µL.
Calibrators were prepared by dilution of the working standard solutions
with 5% ACN in water, with 0.2% formic acid, to concentrations from
0.5 to 1000 ng/mL. Calibration curves were prepared fresh for each
quantitative assay.
Quantitative analysis was performed on an LTQ XL Orbitrap mass
spectrometer (Thermo Scientific, San Jose, CA) coupled with a Waters
190

H.K. Knych, et al.
BiochemicalPharmacology168(2019)184–192
3.4. Inhibition assays with the CYP2D6 inhibitor quinidine
catalyzing the reaction that generates morphine from codeine [4–6].
Despite being highly homologous to CYP2D82, CYP2D14 and CYP2D50
were not capable of generating morphine following incubation with
codeine, suggesting that a high degree of sequence homology is not
necessarily indicative of similar substrate specificity. Additionally, al-
though microsomal incubations demonstrated biphasic kinetics, sug-
gesting that 2 enzymes catalyze the codeine to morphine reaction, in-
cubations with recombinant P450s, in addition to members of the
CYP2D family described above, failed to identify the second enzyme
responsible for this reaction.
In order to ascertain which of the 2 kinetic components reflected
CYP2D mediated metabolism in microsomal incubations, additional
incubations with the CYP2D6 inhibitor, quinidine were included.
Biphasic kinetics were observed for norcodeine generation in micro-
somal incubations including quinidine (Fig. 5). The Km for the high
affinity (low Km) enzyme increased in the presence of quinidine while
Vmax remained similar to microsomal incubations without quinidine
(Table 1). The low affinity (high Km) enzyme appeared unaffected by
the addition of quinidine to the microsomal incubations.
P450 inhibition studies were subsequently conducted in an attempt
to determine whether CYP2D82 was the high or low affinity enzyme
identified in microsomal incubations. Selection of quinidine, an estab-
lished CYP2D6 inhibitor was based on the high degree of sequence
homology between the equine enzyme and CYP2D6. Additionally, it has
been suggested that quinidine decreases the metabolic activity of the
equine enzyme, CYP2D50, also highly homologous to CYP2D82, in
equine microsomal incubations [13]. Kinetic parameters for the low
affinity (high Km), high velocity enzyme in microsomal incubations
with quinidine were similar to those generated in CYP2D82 incubations
with quinidine suggesting that CYP2D82 was the low affinity enzyme
identified in microsomal incubations without quinidine. While the
identity of the high affinity, low velocity enzyme observed in micro-
somal incubations was not determined in recombinant P450 incuba-
tions, it is interesting to note that this enzyme did not appear when
quinidine was added to the microsomal incubations. Although a
number of P450s were tested, including 3 members of the CYP2D fa-
mily (CYP2D14, CYP2D50 and CYP2D82), it is possible that there are
additional, as of yet unidentified members of the CYP2D family in the
horse that are inhibited by the CYP2D6 inhibitor quinidine. It is also
possible that this second enzyme is a member of a different P450 family
altogether and that quinidine is not a specific CYP2D inhibitor.
In human medicine, there are countless reports describing poly-
morphisms, or sequence variations, in genes coding for CYP2D6 and the
functional implications with respect to drug metabolism [14]. In the
case of codeine and mutations in the CYP2D6 gene, increased produc-
tion of morphine has been reported in individuals classified as ultra-
rapid metabolizers which has led to toxicity and in some cases death.
Experimentally, understanding the functional implications of these
mutations requires administration of a probe drug (a known selective
substrate for a particular enzyme) followed by determination of parent
drug and metabolite concentrations at various time post administration
[15]. As codeine appears to be highly selective for CYP2D82, this would
appear to be a good probe drug candidate for identification/determi-
nation of the functional implications of mutations in the CYP2D82 gene
in horses.
While this study adds to knowledge regarding drug metabolism in
horses, it is important to note that there is much left to be investigated.
In the current study, a well-established inhibitor of CYP2D6 was uti-
lized and while data from a single previous study as well as the current
study suggests that quinidine may be an effective inhibitor of members
of the CYP2D6 family in horses, additional characterization of the in-
hibitory potential of this compound is necessary. Of equal importance
in metabolic capabilities is the relative amounts of the individual P450
enzymes. While kinetic parameters are important in determining the
metabolic capabilities of individual enzymes, the relative amount of
each enzyme is also an important determinant of the overall contribu-
tion of individual enzymes to the clearance of a particular compound.
This has yet to be determined in the horse as we are only beginning to
identify and characterize individual P450 enzymes in this species.
In the current study, we describe the metabolic pathway for codeine
in the horse, including identification of specific P450s involved. In
addition, a new member of the CYP2D family, CYP2D82, which appears
to be an orthologue to the highly polymorphic CYP2D6 enzyme in
humans, was identified and its contribution to codeine metabolism
elucidated. Based on the results of the presently reported study, much
Microsomal incubations including quinidine demonstrated uni-en-
zyme kinetics with respect to morphine generation (Fig. 5) with the
activity of the high affinity (low Km), lower velocity enzyme completely
inhibited. The activity of the low affinity (high Km), high velocity en-
zyme persisted; however, the Km for morphine formation in the pre-
sence of quinidine increased compared to incubations without the
CYP2D6 inhibitor while Vmax was similar between the 2 incubations
(Table 1). Since CYP2D82 appeared to be the primary enzyme re-
sponsible for generation of morphine in recombinant incubations, ad-
ditional CYP2D82 incubations in the presence of quinidine were con-
ducted (Fig. 6). In these incubations, the Km and Vmax were similar to
values generated in microsomal incubations with quinidine for the low
affinity (high Km), high velocity enzyme (Table 1).
4. Discussion
The current study describes the in vitro metabolism of codeine by
hepatic microsomal enzymes and expressed P450s from horses, in-
cluding identification of the specific P450 isoforms responsible for
biotransformation. Similar to studies reported in humans [10,11], 2
phase 1 metabolites, norcodeine and morphine, were generated in mi-
crosomal and recombinant P450 incubations. Norcodeine and mor-
phine are also the major phase 1 metabolites observed in in vivo
pharmacokinetic studies in horses (anecdotal report).
Knowledge of the specific enzyme(s) responsible for metabolizing
therapeutic compounds is important in the prediction of drug-drug in-
teractions. Concurrent administration of compounds that rely on the
same enzymes for metabolism and ultimately drug clearance has the
potential to alter the clearance of one or all concurrently administered
drugs. Tools for identifying specific enzymes include recombinantly
expressed enzymes, microsomal incubations with known inhibitors of
specific enzymes or microsomal incubations with inhibitory antibodies.
To the best of the authors’ knowledge, in contrast to the extensive
knowledge base in human 2D polymorphisms and inhibitors only lim-
ited data are available in veterinary medicine to characterize potential
inhibitors. Studies have been undertaken to determine the P450 in-
hibitory potential of different compounds in animal species.
Furthermore, there are no commercially available or published studies
describing the use of P450 inhibitory antibodies in animal species.
In the current study, baculovirus expressed recombinant equine
P450 enzymes were utilized in an attempt to identify the specific iso-
forms responsible for codeine metabolism in the horse. The selection of
specific P450s for expression and subsequent codeine incubations was
based on the P450s known to metabolize the vast majority of ther-
apeutic compounds in humans [12]. Similar to reports in humans,
whereby a member of the CYP3A family is responsible for catalyzing
the conversion of codeine to norcodeine, in the current study, a CYP3A
enzyme (CYP3A95) appears to be a major contributor to this reaction in
the horse. However, in contrast to reports in humans, a second P450, a
member of the CYP2D family (CYP2D50) appears to be as efficient in
carrying out this reaction in horses [2]. Of the recombinant equine
enzymes studied, the newly assigned CYP2D82 was the only enzyme
shown to generate morphine following incubation with codeine. Based
on sequence homology, CYP2D82 is an orthologue to the human en-
zyme, CYP2D6 (80% homology) which in humans is responsible for
191

H.K. Knych, et al.
BiochemicalPharmacology168(2019)184–192
like in humans, codeine may prove effective as an in vivo probe drug in
identifying polymorphisms in genes that code for members of the
CYP2D family.
practice and evidence, Postgrad. Med. J. 77 (2001) 371–378, https://doi.org/10.
1136/pmj.77.908.371.
[4] S.H. Sindrup, K. Brøsen, The pharmacogenetics of codeine hypoalgesia,
Pharmacogenetics 5 (1995) 335–346.
[5] D.G. Williams, D.J. Hatch, R.F. Howard, Codeine phosphate in paediatric medicine,
Br. J. Anaesth. 86 (2001) 413–421, https://doi.org/10.1093/bja/86.3.413.
[6] H.S. Smith, Opioid metabolism, Mayo Clin. Proc. 84 (2009) 613–624, https://doi.
org/10.1016/S0025-6196(11)60750-7.
[7] M. Ingelmansundberg, The human genome project and novel aspects of cytochrome
450 research, Toxicol. Appl. Pharmacol. 207 (2005) 52–56, https://doi.org/10.
1016/j.taap.2005.01.030.
[8] C.R. Corado, D.S. McKemie, H.K. Knych, Dextromethorphan and debrisoquine
metabolism and polymorphism of the gene for cytochrome P450 isozyme 2D50 in
thoroughbreds, Am. J. Vet. Res. (2016;77.), https://doi.org/10.2460/ajvr.77.9.
1029.
[9] T. Omura, R. Sato, The carbon monoxide-binding pigment of liver microsomes. I.
Evidence for its hemoprotein nature, J. Biol. Chem. 239 (1964) 2370–2378.
[10] Q. Yue, C. von Bahr, I. Odar-Cederlöf, J. Säwe, Glucuronidation of codeine and
morphine in human liver and kidney microsomes: effect of inhibitors, Pharmacol.
Toxicol. 66 (1990) 221–226.
Funding source
Financial support for this study was provided by the Center for
Equine Health at the University of California, Davis and funds provided
by the California Horse Racing Board.
Declaration of Competing Interest
The authors have no conflicts of interest to declare.
Acknowledgements
[11] Q.Y. Yue, J. Säwe, Different effects of inhibitors on the O- and N-demethylation of
codeine in human liver microsomes, Eur. J. Clin. Pharmacol. 52 (1997) 41–47.
[12] U.M. Zanger, M. Schwab, Cytochrome P450 enzymes in drug metabolism: regula-
tion of gene expression, enzyme activities, and impact of genetic variation,
Pharmacol. Ther. 138 (2013) 103–141, https://doi.org/10.1016/j.pharmthera.
2012.12.007.
The authors would like to thank Dr Alan Buckpitt for sharing his
drug metabolism expertise and review of the manuscript and Sandy Yim
and Briana Hamamoto-Hardman for technical assistance.
[13] C.R. Corado, D.S. Mckemie, A. Young, H.K. Knych, Evidence for polymorphism in
the cytochrome P450 2D50 gene in horses, J. Vet. Pharmacol. Ther. 39 (2016),
https://doi.org/10.1111/jvp.12269.
[14] C.L. Ventola, Role of pharmacogenomic biomarkers in predicting and improving
drug response: part 1: the clinical significance of pharmacogenetic variants, P T 38
(2013) 545–560.
[15] D. Frank, U. Jaehde, U. Fuhr, Evaluation of probe drugs and pharmacokinetic
metrics for CYP2D6 phenotyping, Eur. J. Clin. Pharmacol. 63 (2007) 321–333,
https://doi.org/10.1007/s00228-006-0250-8.
References
[1] K. Oguri, N. Hanioka, H. Yoshimura, Species differences in metabolism of codeine:
urinary excretion of codeine glucuronide, morphine-3-glucuronide and morphine-6-
glucuronide in mice, rats, guinea pigs and rabbits, Xenobiotica 20 (1990) 683–688,
https://doi.org/10.3109/00498259009046884.
[2] Dean L. Codeine, Therapy and CYP2D6 Genotype, National Center for
Biotechnology Information (US), 2012.
[3] M. Barnett, Alternative opioids to morphine in palliative care: a review of current
192