Inhibition of UDP-Glucuronosyltransferase Enzymes by Major Cannabinoids and Their Metabolites

Abstract

The UDP-glucuronosyltransferase (UGT) family of enzymes play a central role in the metabolism and detoxification of a wide range of endogenous and exogenous compounds. UGTs exhibit a high degree of structural similarity and display overlapping substrate specificity, often making estimations of potential drug-drug interactions difficult to fully elucidate. One such interaction yet to be examined may be occurring between UGTs and cannabinoids, as the legalization of recreational and medicinal cannabis and subsequent co-usage of cannabis and therapeutic drugs increases in the U.S. and internationally. In the present study, the inhibition potential of the major cannabinoids Δ9-tetrahydrocannabinol (THC), cannabidiol (CBD), and cannabinol (CBN), as well as their major metabolites, was determined in microsomes isolated from HEK293 cells over-expressing individual recombinant UGTs and in microsomes from human liver and kidney specimens. The highest inhibition was seen by CBD against the glucuronidation activity of UGTs 1A9, 2B4,1A6 and 2B7, with binding- corrected IC50,u values of 0.12 {plus minus} 0.020 µM, 0.22 {plus minus} 0.045 µM, 0.40 {plus minus} 0.10 µM and 0.82 {plus minus} 0.15 µM, respectively. Strong inhibition of UGT1A9 was also demonstrated by THC and CBN, with IC50,u values of 0.45 {plus minus} 0.12 µM and 0.51 {plus minus} 0.063 µM, respectively. Strong inhibition of UGT2B7 was also observed for THC and CBN; no or weak inhibition was observed with cannabinoid metabolites. This inhibition of UGT activity suggests that in addition to playing an important role in drug-drug interactions, cannabinoid exposure may have important implications in patients with impaired hepatic or kidney function.

Full text

Inhibition of UDP-Glucuronosyltransferase Enzymes by
Major Cannabinoids and Their Metabolites
S
Shamema Nasrin,
2
Christy J. W. Watson,
2
Keti Bardhi, Gabriela Fort,
1
Gang Chen, and Philip Lazarus
Department of Pharmaceutical Sciences, College of Pharmacy and Pharmaceutical Sciences, Washington State University,
Spokane, Washington
Received May 2, 2021; accepted September 01, 2021
ABSTRACT
The UDP-glucuronosyltransferase (UGT) family of enzymes play a
central role in the metabolism and detoxification of a wide range of
endogenous and exogenous compounds. UGTs exhibit a high
degree of structural similarity and display overlapping substrate
specificity, often making estimations of potential drug-drug inter-
actions difficult to fully elucidate. One such interaction yet to be
examined may be occurring between UGTs and cannabinoids, as
the legalization of recreational and medicinal cannabis and subse-
quent co-usage of cannabis and therapeutic drugs increases in the
United States and internationally. In the present study, the inhibi-
tion potential of the major cannabinoids D
9
-tetrahydrocannabinol
(THC), cannabidiol (CBD), and cannabinol (CBN), as well as their
major metabolites, was determined in microsomes isolated from
HEK293 cells overexpressing individual recombinant UGTs and in
microsomes from human liver and kidney specimens. The highest
inhibition was seen by CBD against the glucuronidation activity of
UGTs 1A9, 2B4, 1A6, and 2B7, with binding-corrected IC
50
values of
0.12 ± 0.020 mM, 0.22 ± 0.045 mM, 0.40 ± 0.10 mM, and 0.82 ± 0.15 mM,
respectively. Strong inhibition of UGT1A9 was also demonstrated
by THC and CBN, with binding-corrected IC
50
values of 0.45 ±
0.12 lMand0.51±0.063lM, respectively. Strong inhibition of
UGT2B7 was also observed for THC and CBN; no or weak inhibi-
tion was observed with cannabinoid metabolites. This inhibition
of UGT activity suggests that in addition to playing an important
role in drug-drug interactions, cannabinoid exposure may have
important implications in patients with impaired hepatic or kid-
ney function.
SIGNIFICANCE STATEMENT
Major cannabinoids found in the plasma of cannabis users inhibit
several UDP-glucuronosyltransferase (UGT) enzymes, including
UGT1A6, UGT1A9, UGT2B4, and UGT2B7. This study is the first to
show the potential of cannabinoids and their metabolites to inhibit
all the major kidney UGTs as well as the two most abundant UGTs
present in liver. This study suggests that as all three major kidney
UGTs are inhibited by cannabinoids, greater drug-drug interaction
effects might be observed from co-use of cannabinods and thera-
peutics that are cleared renally.
Introduction
UDP-glucuronosyltransferases (UGTs) are an important family of
phase II metabolizing enzymes that facilitate the detoxification of a
wide variety of endogenous and exogenous compounds, including ste-
roid hormones, drugs, and environmental carcinogens (Meech et al.,
2019). Mammalian UGTs are classified based on structural and amino
acid sequence homology into two main families, the UGT1 and UGT2
families, which are further divided into three subfamilies UGT1A,
UGT2A, and UGT2B, and catalyze the transfer of glucuronic acid from
UDP glucuronic acid (UDPGA) to an electrophilic moiety of a
given substrate, resulting in a more polar conjugate that is more
easily excreted from the body in the urine or bile (Bushey and
Lazarus, 2012). An additional subfamily, the UGT3A subfamily,
contains two members, UGT3A1 and UGT3A2, which use the
alternative sugar donors UDP-N-acetylglucosamine, UDP-glucose,
and UDP-xylose as co-substrates (MacKenzie et al., 2011). Mam-
malian UGTs are membrane-bound enzymes localized in the endo-
plasmic reticulum and expressed with a high degree of tissue
specificity (Meech et al., 2019). Although many UGTs are highly
expressed in the liver, some are also expressed in extrahepatic tis-
sues, including kidney and tissues of the aerodigestive tract (Meech
et al., 2019; Vergara et al., 2020). The UGTs that exhibit the high-
est level of hepatic expression are UGTs 2B7 (17% of total hepatic
UGT expression), 2B4 (16.1%), 2B15 (11.2%), and 1A1 (11%)
(Kasteel et al., 2020). A number of UGTs are also expressed in the
1
Current affiliation: Department of Oncological Science, Huntsman Cancer
Institute, University of Utah, Salt Lake City, Utah.
2
S.N. and C.J.W.W. contributed equally to this work.
This work was supported by the National Institutes of Health National Insti-
tutes of Environmental Health Sciences [Grant R01-ES025460] to P.L., the
Health Sciences and Services Authority of Spokane, WA [Grant WSU002292],
and funds provided by the State of Washington Initiative Measure No. 502.
dx.doi.org/10.1124/dmd.121.000530.
S
This article has supplemental material available at dmd.aspetjournals.org.
ABBREVIATIONS: AZT, azithromycin; BSA, bovine serum albumin; CBD, cannabidiol; CBN, cannabinol; DDI, drug-drug interaction; HEK,
human embryonic kidney; HKM, human kidney microsome; HLM, human liver microsome; IC
50,u
, binding-corrected IC
50
; 7-OH-CBD, 7-hydroxy-
cannabidiol; 11-OH-THC, 11-hydroxy-D
9
-tetrahydrocannabinol; P450, cytochrome P450; rUGT, recombinant UGT; THC, ()-trans-D
9
-tetrahydro-
cannabinol; THC-COO-Gluc, 11-nor-9-carboxy-D9-tetrahydrocannabinol glucuronide; THC-COOH, 11-nor-9-carboxy-D
9
- tetrahydrocannabinol;
UDPGA, UDP glucuronic acid; UGT, UDP-glucuronosyltransferase; UPLC, ultra-high-performance liquid chromatography; UPLC-MS/MS,
UPLC–tandem mass spectrometry.
1081
1521-009X/49/12/1081–1089$35.00 https://doi.org/10.1124/dmd.121.000530
DRUG METABOLISM AND DISPOSITION Drug Metab Dispos 49:1081–1089, December 2021
Copyright ©2021 by The Author(s)
This is an open access article distributed under the CC BY-NC Attribution 4.0 International license.
http://dmd.aspetjournals.org/content/suppl/2021/09/07/dmd.121.000530.DC1
Supplemental material to this article can be found at:
at ASPET Journals on January 20, 2023dmd.aspetjournals.orgDownloaded from

kidney, including UGT1A9 (45% of total renal UGT expression),
UGT2B7 (41%), and UGT1A6 (7%) (Rowland et al., 2013).
UGTs account for the metabolism of 15% of pharmaceuticals, and
one-seventh of the drugs prescribed in the United States in 2002 are
cleared by the UGTs (Williams et al., 2004). Although studies of drug-
drug interactions (DDIs) are a major emphasis of research for phase I
metabolizing enzymes including the cytochrome P450 enzyme family,
UGT enzymes have historically received less scrutiny for their DDI
potential, even though drug interactions via the inhibition of glucuroni-
dation have been increasingly identified. Impaired glucuronidation activ-
ity can cause undesired effects resulting from the slow elimination of
endogenous substances such as bilirubin (Sun et al., 2017) as well as
the buildup of toxic drug metabolites, as has been documented in stud-
ies correlating individuals with UGT1A1-deficient phenotypes and iri-
notecan toxicity (Iyer et al., 1998; Tallman et al., 2007). DDIs between
therapeutics and UGT inhibitors have also been observed in the case of
UGT2B7 inhibition by both valproic acid and probenecid (Cimoch
et al., 1998; Rowland et al., 2006).
The recent legalization of cannabis has caused a dramatic increase in
the use of cannabis-derived products in both recreational and medicinal
situations, where cannabis is frequently used or targeted for more
chronic diseases like cancer, arthritis, and depression and often concur-
rently used with important groups of conventional medications includ-
ing anticancer agents, antidepressants, and pain medications
(Bridgeman and Abazia, 2017). Situations in which polypharmacy is
occurring within a patient could result in deleterious DDIs between can-
nabinoids and any number of therapeutic agents. D
9
-tetrahydrocannabi-
nol (THC) is the best described psychoactive constituent of cannabis,
and plasma concentrations of THC and its active metabolite, 11-
hydroxy (OH)-THC, quickly peak after usage and decrease rapidly over
a short duration, dependent on the specific mode of consumption (Fig.
1) (Sharma et al., 2012). In contrast, the inactive metabolites, 11-nor-9-
carboxy-D
9
-tetrahydrocannabinol (THC-COOH) and 11-COO-D
9
-tetra-
hydrocannabinol-glucuronide (THC-COO-Gluc), peak much more
slowly, to a lower level than the active cannabinoids, and remain pre-
sent in plasma over a much longer duration of time (Huestis, 2007).
Actual plasma levels of active and inactive cannabinoids are highly var-
iable (in the micromolar to submicromolar range) and will vary widely
depending on the user, dose, and method of ingestion. Cannabinol
(CBN) appears to be a degradation product of THC within the Cannabis
plant (Russo and Marcu, 2017) and has been shown to be only weakly
psychoactive. Cannabidiol (CBD) is often termed as medical marijuana
and interacts with the CB
1
and CB
2
receptors in the brain with a much
lower affinity as compared with THC and 11-OH-THC, resulting in
extremely low psychoactive effects (Pertwee, 2008). However, CBD
usage is rapidly expanding among many patient populations due in part
to its good safety profile (Larsen and Shahinas, 2020). Recent clinical
and preclinical trials have shown that CBD has a broad range of poten-
tial applications, displaying anti-inflammatory properties, antipsychotic,
and antiepileptic effects, as well as modulation of the immune system
and the central nervous system (Esposito et al., 2013; Boychuk et al.,
2015; Campos et al., 2016; Devinsky et al., 2016). Similarly, CBD and
its metabolites, 7-hydroxy-cannabidiol (7-OH-CBD) and 7-carboxy-
cannabidiol (CBD-COOH), are present in the plasma after cannabis
inhalation, with unchanged CBD and glucuronidated CBD (CBD-COO-
Gluc), as the main excretion products in urine (Harvey and Mechoulam,
1990; Huestis, 2007). All cannabinoids are highly lipophilic and con-
centrate in tissues with slow release back into the bloodstream (Huestis,
2007). This leads to varying plasma concentrations of active and inac-
tive cannabinoids that persist in the bloodstream, potentially incurring
deleterious DDIs over a much wider time frame than that of the initial
cannabis consumption.
Previous studies have shown that THC, CBD, and CBN can strongly
inhibit several major hepatic cytochrome P450s (P450s) (Yamaori et
al., 2010; Yamaori et al., 2011a; Yamaori et al., 2011b; Yamaori et al.,
2011c; Jiang et al., 2013; Cox et al., 2019; Nasrin et al., 2021). In addi-
tion, the major active metabolite of THC, 11-OH-THC, and two major
inactive metabolites, THC-COOH and THC-COO-Gluc, also exhibited
strong inhibition of a number of hepatic P450 enzymes (Nasrin et al.,
2021). In the present study, the inhibition potential of major cannabi-
noids and their metabolites against major hepatic and renal human UGT
enzymes were evaluated.
Fig. 1. Metabolic pathways and structures of major cannabinoids and their metabolites.
1082 Nasrin et al.
at ASPET Journals on January 20, 2023dmd.aspetjournals.orgDownloaded from

Material and Methods
Chemicals and Reagents. THC, 11-OH-THC, THC-COOH, THC-COO-
Gluc, CBD, 7-OH-CBD, and CBN were purchased from Cayman Chemicals
(Ann Arbor, MI) or Sigma-Aldrich (St. Louis, MO). Pooled human liver micro-
somes (HLMs) [n550, mixed gender (21 female and 29 male), race (42 Cauca-
sian, 4 Hispanic, 2 African American, and 2 Asian), and age (5–77 years)] and
pooled human kidney microsomes (HKMs) [n58, mixed gender (50% each),
race (3 African American, 3 Caucasian, and 2 Hispanic), and age (42–70 years)]
were obtained from Sekisui Xenotech, LLC (Lenexa, KS). b-estradiol, cheno-
deoxycholic acid, trifluoperazine, serotonin, propofol, codeine, zidovudine
(AZT), nicotine, oxazepam, dihydroexemestane, ketoconazole, diclofenac, acet-
aminophen, and furosemide were all purchased from Sigma-Aldrich. Optima
grade methanol, acetonitrile, and formic acid were obtained from Fisher Scien-
tific (Waltham, MA). Ultra-low-binding microcentrifuge tubes, Dulbecco’smodi-
fied Eagle’s medium, Dulbecco’s phosphate-buffered saline, UDPGA,
alamethacin, MgCl
2
, and geneticin (G418) were purchased from VWR (Radnor,
PA). BCA protein assays were purchased from Pierce (Rockford, IL), premium
grade FBS was purchased from Seradigm (Radnor, PA), and ChromatoPur
bovine serum albumin (BSA) was purchased from MB Biomedicals (Santa Ana,
CA).
Inhibition Assays. Human embryonic kidney (HEK) 293 cells individually
overexpressing recombinant UGTs 1A1, 1A3, 1A4, 1A6, 1A9, 2B4, 2B7, 2B10,
2B15, and 2B17 were developed and described previously (Dellinger et al.,
2006). Microsomal membrane fractions of UGT-overexpressing cell lines were
prepared by differential centrifugation as previously described, with total micro-
somal protein concentrations determined using the BCA assay as per the manu-
facturer’s recommendations. An initial screen, performed in duplicate, of the
inhibition potential of individual cannabinoids and their metabolites against
UGTs 1A1, 1A3, 1A4, 1A6, 1A9, 2B4, 2B7, 2B10, 2B15, and 2B17 were deter-
mined using microsomes (50–100 mg) from UGT-overexpressing HEK293 cell
lines in reactions containing either 10 mMor100mM of cannabinoid or metabo-
lite, probe substrate (Supplemental Table 1), 50 mM Tris-HCl buffer (pH 7.4),
MgCl
2
(5 mM), 2% BSA, and 4 mM UDPGA in a final reaction volume of 25
ml. All substrates were used at concentrations near their respective Michaelis--
Menten constant (K
m
; Supplemental Table 1). As cannabinoids exhibit extensive
nonspecificbinding(70–90%) to protein and labware (Garrett and Hunt, 1974),
microsomal incubation conditions were optimized to prevent underestimation of
inhibitory potency (IC
50
). To reduce nonspecific binding and adsorption to lab-
ware, low-binding microcentrifuge tubes were used for all reactions, with BSA
added to increase the solubility of cannabinoids as well as to sequester inhibitory
long-chain unsaturated fatty acids (Rowland et al., 2008; Patilea-Vrana et al.,
2019).
Microsomes were preincubated with alamethicin (50 mg/mg of microsomal
protein) on ice for 20 minutes prior to incubation. The reaction was initiated by
the addition of UDPGA and incubated for 60–120 minutes (Supplemental Table
1) at 37C. Reactions were terminated and proteins precipitated by the addition
of an equal volume (25 ml) of ice-cold stop solution (acetonitrile:methanol; 1:1).
Samples were mixed on a vortex mixer and centrifuged at 17,000gfor
15 minutes. The supernatant (50 ml) was transferred to an ultra-high-perform-
ance liquid chromatograph (UPLC) sample vial, and the probe metabolite was
detected using a UPLC (Waters Acquity; Waters Corp, Milford, MA) coupled to
a triple-quadrupole mass spectrometer (Waters Xevo TQD; Waters Corp) by
multiple reaction monitoring analysis. As a positive control for every inhibition
experiment, 10 mMor100mM probe inhibitors (ketoconazole/diclofenac) were
added instead of the cannabinoid compounds. Reactions containing only vehicle
(3% methanol) and without any inhibitor were used as an indicator of 100%
activity for each substrate/enzyme combination. All IC
50
analyses were per-
formed in triplicate. Incubation conditions were optimized for HLMs, HKMs,
and overexpressing cell lines for both microsomal protein and reaction time, with
optimal conditions chosen based on the following criteria: 1) metabolite forma-
tion was linear with time and enzyme concentration, 2) substrate consumption
was no more than 20% of the initial amount, and 3) metabolite formation was
reliably and reproducibly detected by the UPLC–tandem mass spectrometry
(UPLC-MS/MS) method used.
For UPLC-MS/MS, samples (2–5ml) were injected onto an Acquity UPLC
column (BEH C
18
,1.7mM, 2.1 × 100 mm; Waters Corp). A 9-minute gradient
elution was used with mobile phases A (0.1% formic acid in water) and B
(100% methanol) as follows: 1 minute at 95% A:5% B followed by a linear gra-
dient for 7 minutes to 5% A:95% B, 1 minute at 5% A:95% B, and re-equilibra-
tion for 1 minute at 95% A:5% B. The flow rate was 0.4 ml/min, and the
column temperature was 40C. Analytes were detected using a Waters Xevo
TQD tandem mass spectrometer equipped with a Zspray electrospray ionization
interface operated in the positive ion mode for all the UGT metabolites tested in
this study except furosemide glucuronide, which was analyzed in negative ion
mode, with the capillary voltage at 0.6 kV. Nitrogen was used as both the cone
and desolvation gas at 50 and 800 L/h, respectively. Ultrapure argon was used
for collision-induced dissociation. The desolvation temperature was 500C. For
detection of the metabolite peaks, the mass spectrometer was operated in multi-
ple reaction monitoring mode using the ion-related parameters for each transition.
The following transitions were used for the detection of each probe metabolite:
b-estradiol-3-glucuronide (m/z 447 >271), acyl chenodeoxycholic acid-24-glu-
curonide (m/z 567.5 >391.5), trifluoperazine N-glucuronide (m/z 584 >408.2),
serotonin-glucuronide (m/z 352 >160.02), propofol-O-glucuronide (m/z 354 >
177.02), codeine-6-glucoronide (m/z 476.2 >300.2), AZT-50-glucuronide
(m/z 442 >125.05), nicotine-N-glucuronide (m/z 339.15 >163.124), S-oxaze-
pam-glucuronide (m/z 463.3 >269.1), and exemestane-17-O-glucuronide
(m/z 475.23 >281.19).
Determination of IC
50
Values. For those cannabinoids or metabolites that
inhibited UGT activity $50% at cannabinoid concentrations #100 mM, IC
50
determinations were performed in HLMs, HKMs, and microsomes from
HEK293 UGT-overexpressing cell lines, using multiple concentrations of canna-
binoid inhibitor ranging between 0.5 and 120 mM.
Experiments were performed to determine nonspecific binding constants
(f
u,inc
) for the individual cannabinoids in HEK293 microsomes, HLMs, and
HKMs as previously described (Nasrin et al., 2021).
Statistical Analysis. Data were exported and analyzed using an Excel
spreadsheet (Microsoft). The amount of metabolite formed at each concentration
of inhibitor relative to the control (percent relative activity) was calculated as
Peak area of metabolite with inhibitor/Peak area of metabolite without
inhibitor × 100%.
IC
50
values were calculated by plotting the percent relative activity of UGT
enzymes versus the log concentration of the test inhibitors using GraphPad Prism
7.04 software (GraphPad Software Inc., San Diego, CA).
Results
Glucuronide metabolite peaks were detected by liquid chromatogra-
phy–tandem mass spectrometry in incubations of each probe substrate
analyzed in these studies (Fig. 2). Using recombinant UGT (rUGT)–
overexpressing cell microsomes and probe UGT substrates, preliminary
screening studies demonstrated that 100 mM THC decreased the relative
activity of microsomes from rUGT 1A9, 2B4, and 2B7 overexpressing
cells by 74%, 79%, and 69%, respectively, as compared with control
reactions without added cannabinoid (Fig. 3). A similar pattern was
observed for CBD, with 10 mM CBD exhibiting 25%, 91%, 66%,
and 58% inhibition and 100 mM CBD exhibiting 54%, 98%, 94%, and
96% inhibition, against microsomes from rUGTs 1A6, 1A9, 2B4, and
2B7 overexpressing cells, respectively, as compared with control reac-
tions without added cannabinoid (Fig. 3). Similar to that observed for
THC and CBD, CBN exhibited significant inhibition against rUGT1A9
and rUGT2B7 microsomes. Unlike that observed for THC and CBD,
significant inhibition was not observed for rUGT2B4 microsomes with
CBN. Although no significant inhibition was observed in rUGT1A1,
rUGT1A3, rUGT1A4, and rUGT2B15 microsomes by THC, CBD, and
CBN, marginal inhibition was observed for 100 mM CBD and CBN
against rUGT2B17 microsomes (43% and 34%, respectively). Marginal
inhibition (43% and 47%, respectively) was also observed for 100 mM
CBN against rUGT1A6 and rUGT2B10 microsomes.
For THC and CBD metabolites, no significant inhibition was
observed using up to 100 mM THC-COOH or THC-COO-Gluc against
any of the UGT enzymes tested. However, 100 mM 11-OH-THC
resulted in marginal inhibition of the activities of rUGT1A9 (41%),
Cannabinoids and Their Metabolites as UGT Enzyme Inhibitors 1083
at ASPET Journals on January 20, 2023dmd.aspetjournals.orgDownloaded from

rUGT2B4 (40%), and rUGT2B7 (53%) microsomes, whereas 100 mM
7-OH-CBD resulted in marginal decreases in the activities of rUGT1A9
(40%) and rUGT2B7 (45%) microsomes (Fig. 3).
The inhibitory effects of THC, 11-OH-THC, CBD and CBN were
extended to establish IC
50
values and binding-corrected IC
50
values
(IC
50,u
) for each cannabinoid against the UGT enzymes shown to be
inhibited by $50% using 100 mM cannabinoid in the rUGT screening
assays (described above). The unbound fraction in the incubation mix-
ture were 0.042 ± 0.003, 0.038 ± 0.002, and 0.085 ± 0.005 in overex-
pressing HEK cell lines for THC, CBD and CBN respectively. For
HLMs, the unbound fraction of THC, CBD and CBN in the incubation
mixture were 0.048 ± 0.002, 0.051± 0.008, and 0.092 ± 0.006, respec-
tively, and for HKMs the unbound fractions were 0.052 ± 0.005,
0.062 ± 0.009, and 0.12 ± 0.015, respectively.
The strongest inhibition was observed by CBD against rUGTs 1A9
and 2B4, with IC
50
values of 3.2 ± 0.52 mM and 5.8 ± 1.2 mM, and
IC
50,u
values of 0.12 ± 0.020 mM and 0.22 ± 0.045 mM, using propofol
and codeine as UGT1A9 and UGT2B4 probe substrates, respectively
(Table 1). CBD also exhibited significant inhibition of the glucuronida-
tion of serotonin (a probe substrate for rUGT1A6) in rUGT1A6 micro-
somes (IC
50
510 ± 2.6 mMandIC
50,u
50.40 ± 0.10 mM), and AZT
glucuronidation as a probe substrate in rUGT2B7 microsomes (IC
50
5
21 ± 3.9 mMandIC
50,u
50.82 ± 0.15 mM). The IC
50
values for CBD
for propofol glucuronidation were similar in HKMs (IC
50
55.5 ±
0.56 mMandIC
50,u
50.34 ± 0.035 mM) but higher in HLMs (IC
50
5
19 ± 4.6 mMandIC
50,u
51.0 ± 0.24 mM) as compared with that
observed for rUGT1A9 microsomes (Table 1), a pattern that was
reversed in HKMs (IC
50
539 ± 5.9 mMandIC
50,u
52.5 ± 0.37 mM)
versus HLMs (IC
50
58.0 ± 1.1 mMandIC
50,u
50.40 ± 0.058 mM) for
CBD inhibition of codeine glucuronidation. The decreased level of inhi-
bition of propofol glucuronidation by CBD in HLMs versus HKMs and
the similar inhibition pattern of HKMs and rUGT1A9 microsomes is
apparent when examining plots of percent glucuronidation activity ver-
sus CBD concentrations (Fig. 4). Similar to that observed for rUGT2B7
microsomes, more moderate inhibition was observed for CBD of AZT
glucuronidation in HLMs (IC
50
530 ± 4.1 mMandIC
50,u
51.5 ± 0.21
mM) and HKMs (IC
50
535 ± 3.5 mMandIC
50,u
52.2 ± 0.22 mM),
with IC
50
values that were only slightly higher than that observed for
rUGT2B7 microsomes (Table 1; Fig. 4). The IC
50
values for serotonin
glucuronidation of 28 ± 6.5 mM(IC
50,u
51.4 ± 0.33 mM), and 17 ±
3.7 mM(IC
50,u
51.0 ± 0.23 mM) in HLMs and HKMs, respectively,
were slightly higher than that observed for rUGT1A9 microsomes
(Table 1).
THC exhibited IC
50
values that were slightly higher than CBD for
propofol, codeine, and AZT glucuronidation in rUGT microsomes,
HLMs, and HKMs (Table 1). Similar to that observed for CBD, THC
exhibited similar IC
50
values for propofol in rUGT1A9 microsomes
(IC
50
511 ± 3.0 mMandIC
50,u
50.45 ± 0.12 mM) and codeine
glucuronidation in rUGT2B4 microsomes (IC
50
511 ± 2.7 mMand
IC
50,u
50.47 ± 0.11 mM), with a higher value observed for AZT
glucuronidation in rUGT2B7 microsomes (IC
50
533 ± 8.5 mMand
IC
50,u
51.4 ± 0.36 mM). Also similar to that observed for CBD, the
IC
50
values for THC for propofol glucuronidation was similar in HKMs
(IC
50
512 ± 3.4 mMandIC
50,u
50.64 ± 0.18 mM) but higher in
HLMs (IC
50
530 ± 6.4 mMandIC
50,u
51.4 ± 0.31 mM) as compared
with that observed for rUGT1A9 microsomes, a pattern that was
β-estradiol-3-glucuronide
447 > 271
acyl CDCA-24-glucuronide
567.5 > 391.5
trifluoperazine N-glucuronide
584 >408.2
serotonin-glucuronide
352 > 160.02
propofol-O-glucuronide
354 > 177.02
codeine-6-glucoronide
476.2 > 300.2
AZT-5’-glucuronide
442 > 125.05
nicotine-N-glucuronide
339.15 > 163.124
S-oxazepam-glucuronide
463.3 > 269.1
exemestane-17-O-glucuronide
475.23 > 281.19
Fig. 2. Chromatograms of probe metabolites in microsomes from UGT-overexpressing HEK293 cell lines. Probe substrates (at concentrations close to their known
K
m
; see Supplemental Table 1) were incubated in rUGTs for 60–120 minutes at 37C, and individual corresponding metabolites were analyzed by UPLC-MS/MS as
described in Supplemental Table 1 and in the Materials and Methods.
1084 Nasrin et al.
at ASPET Journals on January 20, 2023dmd.aspetjournals.orgDownloaded from

reversed for THC inhibition of codeine glucuronidation in HKMs
(IC
50
555 ± 5.2 mMandIC
50,u
52.9 ± 0.27 mM) versus HLMs
(IC
50
513 ± 2.6 mMandIC
50,u
50.61 ± 0.13 mM). Again similar to
that observed for CBD, more moderate inhibition was observed for
THC inhibition of AZT glucuronidation in HLMs and HKMs, with
IC
50
values that were only slightly higher than that observed for
rUGT2B7 microsomes [IC
50
values 559 ± 6.6 (IC
50,u
52.8 ±
0.32 mM)and51±12mM(IC
50,u
52.6 ± 0.65 mM), respectively].
The pattern of inhibition observed for CBN for propofol glucuronida-
tion was virtually identical to that observed for both THC and CBD,
with similar IC
50
values observed for rUGT1A9 microsomes (IC
50
5
6.0 ± 0.75 mMandIC
50 u
50.51 ± 0.063 mM) and HKMs (IC
50
5
7.5 ± 1.7 mMandIC
50 u
50.90 ± 0.20 mM) and a higher IC
50
value
observed for HLMs (IC
50
531 ± 4.1 mMandIC
50 u
52.9 ± 0.38 mM;
Table 1). Similar to that observed for both THC and CBD, CBN exhib-
ited more moderate inhibition of AZT glucurondation, with similar IC
50
values observed for rUGT2B7 microsomes (IC
50
549 ± 12 mMand
IC
50 u
54.2 ± 1.1 mM), HLMs (IC
50
559 ± 8.6 mMandIC
50 u
5
5.5 ± 0.79 mM) and HKMs (IC
50
557 ± 7.5 mMandIC
50 u
56.9 ±
0.090 mM). Since CBN did not exhibit inhibitory activity against
codeine glucuronidation in the screening assays, IC
50
values were not
determined for CBN against codeine glucuronidation in rUGT2B4
microsomes, HLMs, or HKMs.
The only THC metabolite that exhibited $50% inhibition for any
UGT in the rUGT microsomal screening assays was 11-OH-THC for
UGT2B7. This metabolite exhibited weak inhibition of rUGT2B7
10 µM
100 µM
0
50
100
1A1
1A3
1A4
1A6
1A9
2B4
2B7
2B10
2B15
2B17
yt
iv
itcA
%
0
50
100
1A1
1A3
1A4
1A6
1A9
2B4
2B7
2B10
2B15
2B17
% Acvity
0
50
100
1A1
1A3
1A4
1A6
1A9
2B4
2B7
2B10
2B15
2B17
% Acvity
0
50
100
1A1
1A3
1A4
1A6
1A9
2B4
2B7
2B10
2B15
2B17
% Acvity
0
50
100
1A1
1A3
1A4
1A6
1A9
2B4
2B7
2B10
2B15
2B17
% Acvity
0
50
100
1A1
1A3
1A4
1A6
1A9
2B4
2B7
2B10
2B15
2B17
% Acvity
0
50
100
1A1
1A3
1A4
1A6
1A9
2B4
2B7
2B10
2B15
2B17
% Acvity
THC 11- O H- T HC 11-COOH-THC THC-COO-Gluc
CBD 7-OH-CBD CBN
Fig. 3. Screening of cannabinoid inhibition of major hepatic UGTs in microsomes from UGT-overexpressing HEK293 cell lines. Probe substrates were b-estradiol for
UGT1A1, chenodeoxycholic acid for UGT1A3, trifluoperazine for UGT1A4, serotonin for UGT1A6, propofol for UGT1A9, codeine for UGT2B4, zidovudine for
UGT2B7, nicotine for UGT2B10, oxazepam for UGT2B15, and dihydroexemestane for UGT2B17. Incubations were performed using 10 or 100 mM of cannabinoid,
with probe substrate concentrations at or close to their known K
m
for their corresponding enzyme (see Supplemental Table 1). Shown are the mean inhibition of two
individual experiments performed for each probe substrate. Data are expressed as a percentage of metabolite formation formed in assays with cannabinoid compared
with assays without cannabinoid.
TABLE 1
IC
50
values (mM) of cannabinoids against major hepatic UGT enzymes in microsomes from recombinant UGT–overexpressing cells, HLMs, or HKMs.
Probe substrate Microsomes THC CBD CBN
IC
50
IC
50,u
IC
50
IC
50,u
IC
50
IC
50,u
mMmMmMmMmMmM
Serotonin rUGT1A6 10 ± 2.6 0.40 ± 0.10
HKM NA 17 ± 3.7 1.0 ± 0.23 NA
HLM 28 ± 6.5 1.4 ± 0.33
Codeine rUGT2B4 11 ± 2.7 0.47 ± 0.11 5.8 ± 1.2 0.22 ± 0.045
HKM 55 ± 5.2 2.9 ± 0.27 39 ± 5.9 2.5 ± 0.37 NA
HLM 13 ± 2.6 0.61 ± 0.13 8.0 ± 1.1 0.40 ± 0.058
AZT rUGT2B7 33 ± 8.5 1.4 ± 0.36 21 ± 3.9 0.82 ± 0.15 49 ± 12 4.2 ± 1.1
HKM 51 ± 12 2.6 ± 0.65 35 ± 3.5 2.2 ± 0.22 57 ± 7.5 6.9 ± 0.90
HLM 59 ± 6.6 2.8 ± 0.32 30 ± 4.1 1.5 ± 0.21 59 ± 8.6 5.5 ± 0.79
Propofol rUGT1A9 11 ± 3.0 0.45 ± 0.12 3.2 ± 0.52 0.12 ± 0.020 6.0 ± 0.75 0.51 ± 0.063
HKM 12 ± 3.4 0.64 ± 0.18 5.5 ± 0.56 0.34 ± 0.035 7.5 ± 1.7 0.90 ± 0.20
HLM 30 ± 6.4 1.4 ± 0.31 19 ± 4.6 1.0 ± 0.24 31 ± 4.1 2.9 ± 0.38
Furosemide rUGT1A9 8.0 ± 0.47 0.33 ± 0.020 2.4 ± 0.66 0.090 ± 0.025 9.2 ± 2.1 0.78 ± 0.18
HKM 10 ± 4.1 0.54 ± 0.22 3.6 ± 0.80 0.22 ± 0.049 15 ± 0.77 1.9 ± 0.092
HLM 32 ± 6.3 1.5 ± 0.30 29 ± 4.0 1.5 ± 0.20 34 ± 6.3 3.1 ± 0.58
Acetaminophen rUGT1A9 12 ± 3.7 0.49 ± 0.15 1.9 ± 0.29 0.073 ± 0.011 6.9 ± 0.54 0.59 ± 0.046
HKM 15 ± 3.0 0.79 ± 0.16 3.8 ± 0.82 0.24 ± 0.05 21 ± 3.4 2.6 ± 0.41
HLM 29 ± 8.9 1.4 ± 0.43 12 ± 3.2 0.64 ± 0.16 30 ± 4.5 2.8 ± 0.96
IC
50
values are presented as means ± S.D. of three independent experiments. IC
50,u
, binding-corrected IC
50
; NA, not analyzed.
Cannabinoids and Their Metabolites as UGT Enzyme Inhibitors 1085
at ASPET Journals on January 20, 2023dmd.aspetjournals.orgDownloaded from

microsomal activity (IC
50
579 ± 11 mMandIC
50,u
54.9 ± .41 mM,
calculated using the 11-OH-THC f
u,inc
value from previous studies
(Nasrin et al., 2021). The IC
50
values for AZT glucuronidation in
HLMs and HKMs were not determined as inhibition did not occur at >
50% at the concentration range tested (up to 100 mMAZT).
To better validate the inhibitory effects of cannabinoids on UGT1A9-
mediated glucuronidation, two additional UGT1A9 probe substrates,
furosemide and acetaminophen, were examined. As shown in Table 1,
the glucuronidation of both agents was strongly inhibited by THC,
CBD and CBN at levels similar to those observed for propofol glucuro-
nidation in rUGT1A9 microsomes. The highest level of inhibition was
again observed with CBD, with IC
50
values in rUGT1A9 microsomes
of 2.4 ± 0.66 mM(IC
50,u
50.090 ± 0.025 mM) and 1.9 ± 0.29 mM
(IC
50,u
50.073 ± 0.011 mM) for furosemide and acetaminophen glu-
curonidation, respectively (Table 1). The IC
50
values observed in
HKMs were very similar to those determined in rUGT1A9 microsomes,
with CBD exhibiting the highest level of inhibition at 3.6 ± 0.80 mM
(IC
50,u
50.22 ± 0.049 mM) and 3.8 ± 0.82 mM(IC
50,u
50.24 ± 0.05
mM) for furosemide and acetaminophen glucuronidation, respectively,
and less inhibition in HLMs, with IC
50
values of 29 ± 4.0 mM
(IC
50,u
51.5 ± 0.20 mM) and 12 ± 3.2 mM(IC
50,u
50.64 ± 0.16 mM),
respectively. The decreased level of inhibition of furosemide and acet-
aminophen glucuronidation by CBD in HLMs versus HKMs and the
similar inhibition pattern of HKMs and rUGT1A9 microsomes is appar-
ent when examining plots of percent glucuronidation activity versus
CBD concentrations (Fig. 4). THC and CBN were slighly less potent
inhibitors in rUGT1A9 microsomes, with IC
50
values of 8.0 ± 0.47 mM
(IC
50,u
50.33 ± 0.020 mM) and 9.2 ± 2.1 mM(IC
50,u
50.78 ±
0.18 mM), respectively, against furosemide, and 12 ± 3.7 mM(IC
50,u
5
0.49± 0.15 mM) and 6.9 ± 0.54 mM(IC
50,u
50.59 ± 0.046 mM),
respectively, against acetaminophen (Table 1). However, the same trend
observed in the tissue microsomes was also observed for rUGT1A9
microsomes, with similar IC
50
values for HKMs against furosemide
(IC
50
510 ± 4.1 mMandIC
50,u
50.54 ± 0.22 mMforTHC;IC
50
5
15 ± 0.8 mMandIC
50,u
51.9 ± 0.092 mM for CBN) and acetamino-
phen (IC
50
515 ± 3.0 mMandIC
50,u
50.79 ± 0.16 mMforTHC;
IC
50
521 ± 3.4 mMandIC
50,u
52.6 ± 0.41 mM for CBN), but some-
what higher for HLMs against furosemide (IC
50
532 ± 6.3 mMand
IC
50,u
51.5 ± 0.30 mMforTHC;IC
50
530 ± 4.5 mMandIC
50,u
5
2.8 ± 0.96 mM for CBN) and acetaminophen (IC
50
529 ± 8.9 mMand
IC
50,u
51.4 ± 0.43 mMforTHC;IC
50
534 ± 6.3 mMandIC
50,u
5
3.1 ± 0.58 mM for CBN; Table 1). The decreased level of inhibition of
furosemide and acetaminophen glucuronidation by THC and CBN in
HLMs versus HKMs and the similar inhibition pattern with both THC
and CBN of HKMs and rUGT1A9 microsomes is apparent when exam-
ining plots of percent glucuronidation activity versus CBD concentra-
tions (Supplemental Fig. 1).
Discussion
The present study is the first to conduct a comprehensive examination
of the inhibitory effects of major cannabinoids (THC, CBD and CBN)
on the enzymatic activities of each of the primary hepatic UGT
enzymes (UGTs 1A1, 1A3, 1A4, 1A9, 2B4, 2B7, 2B10, 2B15, and
2B17). In addition, the major metabolites of THC and CBD (11-OH-
THC, THC-COOH, THC-COO-Gluc, and 7-OH-CBD) were also
screened as potential inhibitors. The results from the present study indi-
cate that the parent cannabinoids (THC, CBD and CBN) exhibit strong
inhibition of the glucuronidation activities of UGTs 1A6, 1A9, 2B4 and
2B7, and marginal inhibition of a number of additional UGTs including
Fig. 4. Inhibitory effects of CBD on the glucuronidation of UGT probe substrates in microsomes from UGT-overexpressing HEK293 cell lines (rUGT), HLMs, and
HKMs. Shown are representative plots comparing CBD concentration with the percent glucuronidation activity against probe substrates in rUGT microsomes, HLMs,
and HKMs. Incubations were performed for 60–120 minutes at 37C using 80–90 mg of rUGT microsomes or 90–200 mg HLMs or HKMs with the following probe
substrates: propofol, acetaminophen, and furosemide for UGT1A9; serotonin for UGT1A6; codeine for UGT2B4; and AZT for UGT2B7 (see Supplemental Table 1
for concentrations). Individual metabolites were analyzed by UPLC-MS/MS as described in the Materials and Methods.
1086 Nasrin et al.
at ASPET Journals on January 20, 2023dmd.aspetjournals.orgDownloaded from

UGT2B17 by both CBD and CBN, and UGT2B10 by CBN. In contrast
to that observed previously for major hepatic P450 enzymes (Yamaori
et al., 2011a; Yamaori et al., 2011b; Yamaori et al., 2011c; Yamaori et
al., 2012; Bansal et al., 2020; Nasrin et al., 2021), major THC metabo-
lites exhibited little inhibition of hepatic UGT enzymes, with only 11-
OH-THC exhibiting significant inhibition against a single UGT
(UGT2B7). Similar to that observed for THC metabolites, the major
CBD metabolite, 7-OH-CBD, exhibited no significant inhibition against
the UGTs tested, with only marginal inhibition observed for UGTs 1A9
and 2B7. In a pattern similar to that observed for P450 enzymes, THC-
COOH exhibited no significant inhibition against any of the UGT
enzymes tested in the present study.
Tissue and plasma concentrations of cannabinoids vary widely by
user and are dependent upon a number of factors including dose, canna-
bis strain, mode of consumption and expertise of the user (Sharma et
al., 2012). Average plasma concentrations of THC from a 10 mg dose
by inhalation are 110 ug/L (0.35 mM), which are about 3-fold higher
than those observed for oral dosing (360 ug/L; 1.1 mM). The average
CBD plasma level after a 400 mg oral dose is 181 ug/L (0.76 mM)
(Manini et al., 2015). The IC
50,u
values observed in the present study
for THC and CBD against several UGT enzymes are in the micromolar
to submicromolar range, suggesting that unwanted DDIs with xenobiot-
ics metabolized by the same UGT enzymes may occur in co-users of
cannabis.
Of the cannabinoids tested, the strongest inhibition in rUGT micro-
somes was observed by CBD against the glucuronidation of propofol
(UGT1A9), serotonin (UGT1A6), codeine (UGT2B4), and AZT
(UGT2B7), followed by THC, which also exhibited strong inhibition
against the same suite of enzymes. CBN was shown to be similarly
effective at inhibiting the glucuronidation of propofol and AZT in
rUGT1A9 and rUGT2B7 microsomes, respectively; however, unlike
that observed for CBD and THC, no inhibition of rUGT2B4 micro-
somal activity was observed by CBN using codeine as the probe
substrate.
Atlhough the liver is considered the most important organ for the
metabolism of drugs and other xenobiotics, the kidney also plays an
important role, especially when glucuronidation is a primary component
of a drug’s metabolism and elimination (Margaillan et al., 2015). UGT
protein expression in both the human liver and human kidney has large
interindividual variability; however, current literature estimates that 13
UGTs are expressed in significant amounts in liver, whereas only 3
UGTs are appreciably expressed in human kidney, including UGTs
1A9 and 2B7, which are expressed at similar levels, and UGT1A6,
which is expressed at a much lower level; UGT2B4 shows negligible
expression in human kidney (Margaillan et al., 2015; Basit et al., 2020).
Consistent with the relatively high expression pattern of UGT1A9 in
human kidney, the IC
50
values observed in HKMs for propofol, furose-
mide, and acetaminophen, all UGT1A9 substrates, were similar to that
observed for each agent in rUGT1A9 microsomes for CBD, THC and
CBN. This contrasts with HLMs, where the IC
50
values were higher
(approximately 3-fold) than those observed in rUGT1A9 microsomes in
all cases. In addition, the 6-fold lower IC
50
exhibited by UGT1A9 as
compared with UGT2B7 in rUGT microsomes by THC, CBD, and
CBN corresponds with the larger IC
50
values observed in HKMs using
a UGT2B7 probe substrate (AZT) versus that observed for UGT1A9
probe substrates, reflecting the relative inhibition of the two enzymes by
cannabinoids. These data support the possibility that the major cannabi-
noids in Cannabis, CBD, THC, and CBN, may all act to inhibit the two
highly expressed UGT enzymes in human kidney, UGT1A9, and
UGT2B7, in vivo.
The relative inhibition of codeine glucuronidation observed in HKMs
was approximately 5–7-fold higher for THC and CBD as compared
with rUGT2B4 microsomes, suggesting that UGT2B4 is likely not a
major glucuronidating enzyme in kidney. This pattern is consistent with
the low relative expression of UGT2B4 in this organ (Basit et al.,
2020). This activity pattern contrasts to the very similar IC
50
values
observed in HLMs for codeine glucuronidation as compared with those
observed in rUGT2B4 microsomes, a pattern consistent with the high
expression of UGT2B4 in human liver. Although codeine glucuronida-
tion is considered a probe substrate of UGT2B4 activity, UGT2B7 is
also likely a major contributor to the hepatic glucuronidation of this
agent (Court et al., 2003). When comparing codeine glucuronidation
activities for both THC and CBD in the present study, the IC
50
values
are nearly identical in HLMs to those determined for rUGT2B4 micro-
somes. In addition, the IC
50
values were approximately 3.6-fold lower
for rUGT2B4 microsomes than for rUGT2B7 microsomes for both can-
nabinoids. These data indicate that although these two UGT enzymes
are highly homologous, they may have unique binding interactions with
cannabinoids and suggest that CBD, THC, and CBN strongly inhibit
hepatic UGT2B4 activity.
UGT2B7 is arguably the most important UGT enzyme involved in
phase II metabolism, as it is expressed at high levels in the liver and is
the most commonly listed UGT involved in the biotransformation of
the top 200 drugs currently prescribed in the United States (Williams
et al., 2004). Inhibition of this enzyme has the potential to impact thou-
sands of patients through unwanted drug-drug interactions, toxicities,
and off-target effects. As seen from the IC
50
values of THC, CBD, and
CBN against the UGT2B7 probe substrate AZT in HLMs, the
UGT2B7-mediated glucuronidation of AZT is moderately inhibited by
these cannabinoids. In all three cases, the IC
50
determined in HLMs is
similar to the value determined in rUGT2B7, suggesting that UGT2B7
is inhibited by these cannabinoids and that this inhibition can be trans-
lated to the human liver, where the potential for unwanted DDIs may
occur. Indeed, one such interaction has been observed when Epidolex
(CBD) is prescribed as an antiseizure medication concurrently with the
sedative midazolam (Patsalos et al., 2020). Although midazolam itself
is not glucuronidated by UGT2B7, its active metabolite, 1-hydroxymi-
dazolam, is a well documented UGT2B7 substrate (Seo et al., 2010).
Administration of midazolam with steady state levels of Epidolex
results in increased plasma concentrations of active 1-hydroxymidazo-
lam (C
max
5"12%, area under the plasma concentration versus time
curve from 0 to t5"68%),aswellasadelayint
max
(difference in
median of 2.2 hours) and an increase in t
1/2
of 35%. Another study
examined the disposition kinetics of the opioid morphine with and with-
out concurrent inhaled vaporized cannabis (900 mg, 3.56% THC)
(Abrams et al., 2011). Morphine is a well studied substrate for
UGT2B7 (Osborne et al., 1990) and is glucuronidated to both the inac-
tive 3-glucuronide and the highly active 6-glucuronide. A statistically
significant decrease in steady state plasma levels of morphine was found
when administered with vaporized cannabis (which was attributed to a
decrease in the uptake of morphine), and a near significant decrease in
the C
max
of inactive metabolite 3-glucuronide was also observed, indi-
cating reduced UGT2B7 glucuronidation activity in the presence of the
inhaled vaporized THC.
Although the role of renal metabolism is still an underexplored area
compared with hepatic metabolism, mounting evidence from recent
publications indicates that the human kidney has significant metabolic
capacity. Renal metabolism by UGT enzymes plays a major role in
clearance of many drugs including acetaminophen and furosemide
(assayed in this study) as well as carbamazepine, codeine, gemfibrozil,
morphine, and the commonly used over the counter nonsteroidal anti-
inflammatory drugs ibuprofen, ketoprofen, and S-naproxen (Knights
et al., 2013). Preferential inhibition of the renal UGTs may have a larger
effect on drugs that are mainly excreted by renal glucuronidation, and
Cannabinoids and Their Metabolites as UGT Enzyme Inhibitors 1087
at ASPET Journals on January 20, 2023dmd.aspetjournals.orgDownloaded from

interestingly, the two most highly expressed UGTs in human kidney
(UGTs 1A9 and 2B7) were inhibited by CBD, THC, and CBN in the
present study. Therefore, cannabinoids, and especially CBD, may signif-
icantly and disproportionately affect the 1.5 million people in the United
States (Rein, 2020) who are diagnosed with chronic kidney disease and
acute kidney injury. One-quarter to one-half of those patients also expe-
rience chronic symptoms such as pain, nausea, anorexia, sleep distur-
bance, anxiety, and depression (Rein, 2020), several of which are
approved indications for medical cannabis (CBD). Additionally, chronic
kidney disease is associated with decreased activity of drug metaboliz-
ing enzymes and transporters (Dreisbach and Lertora, 2008). Moreover,
a recent study showed significant reduction in the glucuronidation
capacity of drugs metabolized by UGT1A9 and UGT2B7 in patients
with kidney tumors (Margaillan et al., 2015). AZT and propofol metab-
olism were decreased 96- and 7.6-fold, respectively, in a patient with
neoplastic kidney when compared with normal kidney, suggesting that
the use of Cannabis or CBD in these patients may be deleterious.
In conclusion, the present study is the first to demonstrate that the
major cannabinoids present in Cannabis are able to inhibit several of
the primary UGT enzymes involved in phase II metabolism. CBD was
shown to be the most potent cannabinoid inhibitor, exhibiting IC
50
val-
ues 2–3-fold lower than that observed for THC. Although this is the first
study to specifically address the inhibition of UGTs by CBD and other
cannabinoids, previous reports indicate that CBD, THC, and several
THC metabolites are potent inhibitors of several major P450 enzymes
(Yamaori et al., 2011a; Yamaori et al., 2011b; Yamaori et al., 2011c;
Bansal et al., 2020; Nasrin et al., 2021). Results from this study now
show that two major hepatic UGTs and three of the most highly
expressed UGTs present in kidney are strongly inhibited by these can-
nabinoids, suggesting that deleterious drug-drug interactions may be
more likely to occur in patients in whom reduced hepatic or kidney
function and cannabis use are occurring simultaneously. In light of the
rising acceptance of cannabis use in the United States and internation-
ally, further in vivo studies examining cannabinoid-drug interactions of
both phase I and phase II are warranted.
Acknowledgments
The authors would like to thank Shelby M. Coates for her helpful con-
tributions to the study. Figure 1 was created with BioRender.com.
Authorship Contributions
Participated in research design: Nasrin, Lazarus.
Conducted experiments: Nasrin, Watson, Bardhi, Fort.
Performed data analysis: Nasrin, Watson, Chen, Lazarus.
Wrote or contributed to the writing of the manuscript: Nasrin, Watson,
Lazarus.
References
Abrams DI, Couey P, Shade SB, Kelly ME, and Benowitz NL (2011) Cannabinoid-opioid interac-
tion in chronic pain. Clin Pharmacol Ther 90:844–851.
Bansal S, Maharao N, Paine MF, and Unadkat JD (2020) Predicting the potential for cannabinoids
to precipitate pharmacokinetic drug interactions via reversible inhibition or inactivation of major
cytochromes P450. Drug Metab Dispos 48:1008–1017.
Basit A, Neradugomma NK, Wolford C, Fan PW, Murray B, Takahashi RH, Khojasteh SC, Smith
BJ, Heyward S, Totah RA et al. (2020) Characterization of differential tissue abundance of
major non-CYP enzymes in human. Mol Pharm 17:4114–4124.
Boychuk DG, Goddard G, Mauro G, and Orellana MF (2015) The effectiveness of cannabinoids
in the management of chronic nonmalignant neuropathic pain: a systematic review. J Oral
Facial Pain Headache 29:7–14.
Bridgeman MB and Abazia DT (2017) Medicinal cannabis: history, pharmacology, and implica-
tions for the acute care setting. P&T 42:180–188.
Bushey RT and Lazarus P (2012) Identification and functional characterization of a novel UDP-
glucuronosyltransferase 2A1 splice variant: potential importance in tobacco-related cancer sus-
ceptibility. J Pharmacol Exp Ther 343:712–724.
Campos AC, Fogac¸a MV, Sonego AB, and Guimar~
aes FS (2016) Cannabi diol, neuroprotection
and neuropsychiatric disorders. Pharmacol Res 112:119–127.
Cimoch PJ, Lavelle J, Pollard R, Griffy KG, Wong R, Tarnowski TL, Casserella S, and Jung D
(1998) Pharmacokinetics of oral ganciclovir alone and in combination with zidovudine,
didanosine, and probenecid in HIV-infected subjects. J Acquir Immun e Defic Syndr H um Retro-
virol 17:227–234.
Court MH, Krishnaswamy S, Hao Q, Duan SX, Patten CJ, Von Moltke LL, and Greenblatt DJ
(2003) Evaluation of 30-azido-30-deoxythymidine, morphine, and codeine as probe substrates for
UDP-glucuronosyltransferase 2B7 (UGT2B7) in human liver microsomes: specificity and influ-
ence of the UGT2B7*2 polymorphism. Drug Metab Dispos 31:1125–1133.
Cox EJ, Maharao N, Patilea-Vrana G, Unadkat JD, Rettie AE, McCune JS, and Paine MF (2019)
A marijuana-drug interaction primer: precipitants, pharmacology, and pharmacokinetics. Phar-
macol Ther 201:25–38.
Dellinger RW, Fang J-L, Chen G, Weinberg R, and Lazarus P (2006) Importance of UDP-glucuro-
nosyltransferase 1A10 (UGT1A10) in t he detoxification of polycyclic aromatic hydrocarbons:
decreased glucuronidative activity of the UGT1A10139Lys isoform. Drug Metab Dispos
34:943–949.
Devinsky O, Marsh E, Friedman D, Thiele E, Laux L, Sullivan J, Miller I, Flamini R, Wilfong A,
Filloux F et al. (2016) Cannabidiol in patients with treatment-resistant epilepsy: an open-label
interventional trial. Lancet Neurol 15:270–278.
Dreisbach AW and Lertora JJL (2008) The effect of chronic renal failure on drug metabolism and
transport. Expert Opin Drug Metab Toxicol 4:1065–1074.
Esposito G, Filippis DD, Cirillo C, Iuvone T, Capoccia E, Scuderi C, Steardo A, Cuomo R, and
Steardo L (2013) Cannabidiol in inflammatory bowel diseases: a brief overview. Phytother Res
27:633–636.
Garrett ER and Hunt CA (1974) Physiochemical properties, solubility, and protein binding of
delta9-tetrahydrocannabinol. JPharmSci63:1056–1064.
Harvey DJ and Mechoulam R (1990) Metabolites of cannabidiol identified in human urine. Xeno-
biotica 20:303–320.
Huestis MA (2007) Human cannabinoid pharmacokinetics. Chem Biodivers 4:1770–1804.
Iyer L, King CD, Whitington PF, Green MD, Roy SK, Tephly TR, Coffman BL, and Ratain MJ
(1998) Genetic predisposition to the metabolism of irinotecan (CPT-11). Role of uridine diphos-
phate glucuronosyltransferase isoform 1A1 in the glucuronidation of its active metabolite (SN-
38) in human liver m icrosomes. J Clin Invest 101:847–854.
Jiang R, Yamaori S, Okamoto Y, Yamamoto I, and Watanabe K (2013) Cannabidiol is a potent
inhibitor of the catalytic activity of cytochrome P450 2C19. Drug Metab Pharmacokinet
28:332–338.
Kasteel EEJ, Darney K, Kramer NI, Dorne JLCM, and Lautz LS (2020) Human variability in iso-
form-specific UDP-glucuronosyltransferases: markers of acute and chronic exposure, polymor-
phisms and uncertainty factors. Arch Toxicol 94:2637–2661.
Knights KM, Rowland A, and Miners JO (2013) Renal drug metabolism in humans: the potential
for drug-endobiotic interactions involving cytochrome P450 (CYP) and UDP-glucuronosyltrans-
ferase (UGT). Br J Clin Pharmacol 76:587–602.
Larsen C and Shahinas J (2020) Dosage, efficacy and safety of cannabidiol administration in
adults: a systematic review of human trials. JClinMedRes12:129–141.
MacKenzie PI, Rogers A, Elliot DJ, Chau N, Hulin J-A, Miners JO, and Meech R (2011) The
novel UDP glycosyltransferase 3A2: cloning, catalytic properties, and tissue distribution. Mol
Pharmacol 79:472–478.
Manini AF, Yiannoulos G, Bergamaschi MM, Hernandez S, Olmedo R, Barnes AJ, Winkel G,
Sinha R, Jutras-Aswad D, Huestis MA et al. (2015) Safety and pharmacokinetics of oral canna-
bidiol when administered concomitantly with intravenous fentanyl in humans. JAddictMed
9:204–210.
Margaillan G, Rouleau M, Fallon JK, Caron P, Villeneuve L, Turcotte V, Smith PC, Joy MS, and
Guillemette C (2015) Quantitative profiling of human renal UDP-glucuronosyltransferases and
glucuronidation activity: a comparison of normal and tumoral kidney tissues. Drug Metab Dis-
pos 43:611–619.
Meech R, Hu DG, McKinnon RA, Mubarokah SN, Haines AZ, Nair PC, Rowland A, and Mac-
kenzie PI (2019) The UDP-glycosyltransferase (UGT) superfamily: new members, new func-
tions, and novel paradigms. Physiol Rev 99:1153–1222.
Nasrin S, Watson CJW, Perez-Paramo YX, and Lazarus P (2021) Cannabinoid metabolites as
inhibitors of major hepatic CYP450 enzymes, with implications for cannabis-drug interactions.
Drug Metab Dispos 49:1070–1080.
Nasrin S, Watson CJ, Chen G, and Lazarus P (2021) Cannabinoid metabolites as potential inhibi-
tors of major CYP450 enzymes, with implications for cannabis-drug interactions. FASEB J
DOI: 10.1096/fasebj.2020.34.s1.05863 [published ahead of print].
Osborne R, Joel S, Trew D, and Slevin M (1990) Morphine and metabolite behavior after different
routes of morphine administration: demonstration of the importance of the active metabolite
morphine-6-glucuronide. Clin Pharmacol Ther 47:12–19.
Patilea-Vrana GI, Anoshchenko O, and Unadkat JD (2019) Hepatic enzymes relevant to the dispo-
sition of (-)-D
9
-tetrahydrocannabinol (THC) and its psychoactive metabolite, 1 1-OH-THC. Drug
Metab Dispos 47:249–256.
Patsalos PN, Szaflarski JP, Gidal B, VanLandingham K, Critchley D, and Morrison G (2020) Clin-
ical implications of trials investigating drug-drug interactions between cannabidiol and enzyme
inducers or inhibitors or common antiseizure drugs. Epilepsia 61:1854–1868.
Pertwee RG (2008) The diverse CB1 and CB2 receptor pharmacology of three plant cannabinoids:
delta9-tetrahydrocannabinol, cannabidiol and delta9-tetrahydrocannabivarin. Br J Pharmacol
153:199–215.
Rein JL (2020) The nephrologist’s guide to cannabis and cannabinoids. Curr Opin Nephrol Hyper-
tens 29:248–257.
Rowland A, Elliot DJ, Williams JA, Mackenzie PI, Dickinson RG, and Miners JO (2006) In vitro
characterization of lamotrigine N2-glucuronidation and the lamotrigine-valproic acid interaction.
Drug Metab Dispos 34:1055–1062.
Rowland A, Knights KM, Mackenzie PI, and Miners JO (2008) The “albumin effect”and drug
glucuronidation: bovine serum a lbumin and fatty acid-free h uman serum albumin enhance the
glucuronidation of UDP-glucuronosyltransferase (UGT) 1A9 substrates but not UGT1A1 and
UGT1A6 Activities. Drug Metab Dispos 36:1056–1062.
Rowland A, Miners JO, and Mackenzie PI (2013) The UDP-glucuronosyltransferases: their role in
drug metabolism and detoxification. Int J Biochem Cell Biol 45:1121–1132.
Russo EB and Marcu J (2017) Chapter Three - Cannabis Pharmacology: The Usual Suspects and a
Few Promising Leads, in Adv Pharmacol (Kendall D and Alexander SPH, eds) pp 67–134, Aca-
demic Press, New York .
Seo K-A, Bae SK, Choi Y-K, Choi CS, Liu K-H, and Shin J-G (2010) Metabolism of 10-and
4-hydroxymidazolam by glucuronide conjugation is largely mediated by UDP-glucuronosyl-
transferases 1A4, 2B4, and 2B7. Drug Metab Dispos 38:2007–2013.
1088 Nasrin et al.
at ASPET Journals on January 20, 2023dmd.aspetjournals.orgDownloaded from

Sharma P, Murthy P, and Bharath MMS (2012) Chemistry, metabolism, and toxicology of canna-
bis: clinical implications. Iran J Psychiatry 7:149–156.
Sun L, Li M, Zhang L, Teng X, Chen X, Zhou X, Ma Z, Qi L, and Wang P (2017) Differences in
UGT1A1 gene mutations and pathological liver changes between Chinese patients with Gilbert
syndrome and Crigler-Najjar syndrome type II. Medicine (Baltimore) 96:e8620.
Tallman MN, Miles KK, Kessler FK, Nielsen JN, Tian X, Ritter JK, and Smith PC (2007) The
contribution of intestinal UDP-glucuronosyltransferases in modulating 7-ethyl-10-hydroxy-
camptothecin (SN-38)-induced gastrointestinal toxicity in rats. J Pharmacol Exp Ther
320:29–37.
Vergara AG, Watson CJW, Chen G, and Lazarus P (2020) UDP-glycosyltransferase 3A metabo-
lism of polycyclic aromatic hydrocarbons: potential importance in Aerodigestive tract tissues.
Drug Metab Dispos 48:160–168.
Williams JA, Hyland R, Jones BC, Smith DA, Hurst S, Goosen TC, Peterkin V, Koup JR, and
Ball SE (2004) Drug-drug interactions for UDP-glucuronosyltransferase substrates: a pharmaco-
kinetic explanation for typically observed low exposure (AUCi/AUC) ratios. Drug Metab Dis-
pos 32:1201–1208.
Yamaori S, Ebisawa J, Okushima Y, Yamamoto I, and Watanabe K (2011a) Potent inhibition of
human cytochrome P450 3A isoforms by cannabi diol: role of phenolic hydroxyl groups in the
resorcinol moiety. Life Sci 88:730–73 6.
Yamaori S, Koeda K, Kushihara M, Hada Y, Yamamoto I, and Watanabe K (2012) Comparison
in the in vitro inhibitory effects of major phytocannabinoids and polycyclic aromatic hydrocar-
bons contained in marijuana smoke on cytochrome P450 2C9 activity. Drug Meta b Pharmac o-
kinet 27:294–300.
Yamaori S, Kushihara M, Yamamoto I, and Watanabe K (2010) Characterization of major phyto-
cannabinoids, cannabidiol and cannabinol, as isoform-selective and potent inhibitors of human
CYP1 enzymes. Biochem Pharmacol 79:1691–1698.
Yamaori S, Maeda C, Yamamoto I, and Watanabe K (2011b) Differential inhibition of human
cytochrome P450 2A6 and 2B6 by major phytocannabinoids. Forensic Toxicol 29:117–124.
Yamaori S, Okamoto Y, Yamamoto I, and Watanabe K (2011c) Cannabidiol, a major phytocanna-
binoid, as a potent atypica l inhibitor for CYP2D6. Drug Metab Dispos 39:2049–2056.
Address correspondence to: Dr. Philip Lazarus, Department of Pharmaceuti-
cal Sciences, College of Pharmacy and Pharmaceutical Sciences, Washington
State University, 412 E. Spokane Falls Blvd., Spokane, Washington 99202-
2131. E-mail: phil.lazarus@wsu.edu
Cannabinoids and Their Metabolites as UGT Enzyme Inhibitors 1089
at ASPET Journals on January 20, 2023dmd.aspetjournals.orgDownloaded from