The Novel UDP Glycosyltransferase 3A2: Cloning, catalytic properties, and tissue distribution

Article abstract of DOI:10.1124/mol.110.069336

The human UDP glycosyltransferase (UGT) 3A family is one of three families involved in the metabolism of small lipophilic compounds. Members of these families catalyze the addition of sugar residues to chemicals, which enhances their excretion from the body. The UGT1 and UGT2 family members primarily use UDP glucuronic acid to glucuronidate numerous compounds, such as steroids, bile acids, and therapeutic drugs. We showed recently that UGT3A1, the first member of the UGT3 family to be characterized, is unusual in using UDP N-acetylglucosamine as sugar donor, rather than UDP glucuronic acid or other UDP sugar nucleotides (J Biol Chem 283:36205-36210, 2008). Here, we report the cloning, expression, and characterization of UGT3A2, the second member of the UGT3 family. Like UGT3A1, UGT3A2 is inactive with UDP glucuronic acid as sugar donor. However, in contrast to UGT3A1, UGT3A2 uses both UDP glucose and UDP xylose but not UDP N-acetylglucosamine to glycosidate a broad range of substrates including 4-methylumbelliferone, 1-hydroxypyrene, bioflavones, and estrogens. It has low activity toward bile acids and androgens. UGT3A2 transcripts are found in the thymus, testis, and kidney but are barely detectable in the liver and gastrointestinal tract. The low expression of UGT3A2 in the latter, which are the main organs of drug metabolism, suggests that UGT3A2 has a more selective role in protecting the organs in which it is expressed against toxic insult rather than a more generalized role in drug metabolism. The broad substrate and novel UDP sugar specificity of UGT3A2 would be advantageous for such a function. Copyright

Full text of DOI:10.1124/mol.110.069336

0
026-895X/11/7903-472–478$20.00
MOLECULAR PHARMACOLOGY
Copyright © 2011 The American Society for Pharmacology and Experimental Therapeutics
Mol Pharmacol 79:472–478, 2011
Vol. 79, No. 3
69336/3663716
Printed in U.S.A.
The Novel UDP Glycosyltransferase 3A2: Cloning, Catalytic
Properties, and Tissue Distribution
Peter I. MacKenzie, Anne Rogers, David J. Elliot, Nuy Chau, Julie-Ann Hulin,
John O. Miners, and Robyn Meech
The Department of Clinical Pharmacology, Flinders University School of Medicine, Flinders Medical Centre, Bedford
Park, Australia
Received October 12, 2010; accepted November 18, 2010
ABSTRACT
The human UDP glycosyltransferase (UGT) 3A family is one of trast to UGT3A1, UGT3A2 uses both UDP glucose and UDP
three families involved in the metabolism of small lipophilic com- xylose but not UDP N-acetylglucosamine to glycosidate a broad
pounds. Members of these families catalyze the addition of sugar range of substrates including 4-methylumbelliferone, 1-hydroxy-
residues to chemicals, which enhances their excretion from the pyrene, bioflavones, and estrogens. It has low activity toward bile
body. The UGT1 and UGT2 family members primarily use UDP acids and androgens. UGT3A2 transcripts are found in the thy-
glucuronic acid to glucuronidate numerous compounds, such as mus, testis, and kidney but are barely detectable in the liver and
steroids, bile acids, and therapeutic drugs. We showed recently gastrointestinal tract. The low expression of UGT3A2 in the latter,
that UGT3A1, the first member of the UGT3 family to be charac- which are the main organs of drug metabolism, suggests that
terized, is unusual in using UDP N-acetylglucosamine as sugar UGT3A2 has a more selective role in protecting the organs in
donor, rather than UDP glucuronic acid or other UDP sugar nu- which it is expressed against toxic insult rather than a more
cleotides (J Biol Chem 283:36205–36210, 2008). Here, we report generalized role in drug metabolism. The broad substrate and
the cloning, expression, and characterization of UGT3A2, the sec- novel UDP sugar specificity of UGT3A2 would be advantageous
ond member of the UGT3 family. Like UGT3A1, UGT3A2 is inac- for such a function.
tive with UDP glucuronic acid as sugar donor. However, in con-
wanted products of metabolism and in modulating cell sig-
Introduction
naling pathways controlled by chemical ligands.
Many lipophilic chemicals are metabolized to water-soluble
UDP glycosyltransferases are present in animals, plants,
products via ␤-linkage with hexose groups such as glucose,
and microorganisms (http://www.flinders.edu.au/medicine/
glucuronic acid, galactose, and xylose (Mackenzie et al.,
sites/clinical-pharmacology/ugt-homepage.cfm). In general,
2
005). The UDP glycosyltransferases (UGT) that catalyze
it seems that UDP glucuronic acid is the preferred sugar
donor for conjugation of lipophilic chemicals in vertebrates,
whereas UDP glucose is the preferred sugar donor in inver-
tebrates, plants, and microorganisms. The human genome
contains four UGT families. The UGT1 and UGT2 families
contain 9 and 10 members, respectively. Characterization of
the catalytic properties of all members of these two families
these reactions use UDP sugars as the donor, and functional
groups including hydroxyl, carboxyl, amine, thiol, and carbon
groups on the lipophilic chemical as acceptor. Conjugation
with hexose groups generally reduces the biological activity
of the aglycone and facilitates its removal from cells and from
the body in urine and bile (Miners and Mackenzie, 1991;
Meech and Mackenzie, 1997; Radominska-Pandya et al.,
999; Tukey and Strassburg, 2000; Miners et al., 2004). This
function of UGTs is central to their pivotal roles in protecting
cells against the accumulation of lipophilic toxins and un-
(
Mackenzie et al., 1997) shows that, although the substrate
1
profile for each member is unique, some substrates are al-
most exclusively metabolized by one UGT, whereas other
substrates are metabolized by several UGTs (Miners et al.,
2
010). This broad, overlapping substrate selectivity is well
This work was supported by the National Health and Medical Research
Council (NHMRC) of Australia [NHMRC Senior Principal Research Fellow-
ship (to P.I.M.).
Article, publication date, and citation information can be found at
http://molpharm.aspetjournals.org.
doi:10.1124/mol.110.069336.
suited to a role in modulating the concentrations of chemicals
in cells and hence in protecting organs and tissues against
the toxic effects of chemical overload. The UGT1 and UGT2
families preferentially use UDP glucuronic acid as a sugar
ABBREVIATIONS: UGT, UDP glycosyltransferase; HEK, human embryonic kidney; 4-MU, 4-methylumbelliferone; GAPDH, glyceraldehyde-3-
phosphate dehydrogenase; PCR, polymerase chain reaction; HPLC, high-performance liquid chromatography; GAG, glycosaminoglycan.
472

UGT3A2 Glycosidates Xenobiotics with Glucose and Xylose
473
donor to aid in the removal of a myriad of endogenous and were from GE Healthcare (Chalfont St. Giles, Buckinghamshire,
xenobiotic compounds. In contrast, relatively little is known UK); UDP-glucose, -glucuronic acid, -galactose, and -N-acetylgluco-
samine were from Sigma-Aldrich (St. Louis, MO); and UDP-xylose
about the function of the UGT3 family, which consists of two
was from Carbosource Services (Athens, GA). 4-MU-glucoside, -gluc-
uronide, -galactoside, and -xyloside were purchased from Sigma-
Aldrich. All other reagents and solvents were of analytical reagent
grade.
cDNA Cloning and Expression. RNA from human embryonic
kidney (HEK) 293 cells was used as template to synthesize first
strand cDNA with the Superscript First Strand Synthesis System
members, UGT3A1 and UGT3A2. We have shown that
UGT3A1 preferentially uses UDP N-acetylglucosamine in
conjugation reactions with several substrates, including ur-
sodeoxycholic acid and 17-estradiol (Mackenzie et al., 2008).
However, the second member, UGT3A2 has not been charac-
terized, although its similarity in sequence to UGT3A1 (78%)
suggested that it was also a UDP N-acetylglucosaminyltrans- (Invitrogen, Carlsbad, CA). The UGT3A2 coding region (GenBank
ferase (Meech and Mackenzie, 2010). The fourth UGT family, accession number NM_174914) was amplified from this cDNA using
UGT8, contains only one member that uses UDP galactose to the forward primer 5Ј-AGCATGGCTGGGCAGCGAGTGCTT-3Ј and
the reverse primer 5Ј-TGGCCTTATGTCTCCTTCACCTTT-3Ј. The
galactosidate ceramide, a key step in the synthesis of brain
UGT3A2 initiation and stop codons in the forward and reverse prim-
sphingolipids (Bosio et al., 1996). This UGT seems not to be
ers, respectively, are underlined. PCR was performed in a volume of
involved in xenobiotic metabolism, because other substrates
have not been identified.
In mammals, UGTs are located in the membranes of the
endoplasmic reticulum and nuclear envelope. Each cell, tis-
sue, or organ expresses a specific subset of UGTs, which is
50 ␮l with 200 ng of cDNA, and 0.5 ␮M concentration of the forward
and reverse primers and the DNA polymerase Pfu Turbo (Strat-
agene, La Jolla, CA). The cycling parameters consisted of one cycle at
9
7
5°C for 2 min, 35 cycles of 95°C for 0.5 min, 60°C for 0.5 min, and
2°C for 2 min followed by a single 5-min cycle at 72°C. After
appropriate for its role in regulating the concentrations of electrophoresis on a 1% agarose gel, PCR products were excised and
chemical toxins and/or ligands involved in cell signaling purified from the gel using the QIAquick gel extraction kit (QIAGEN,
pathways. Commensurate with their dominant role in the Valencia, CA) and subcloned into the pCR2.1 shuttle vector (Invit-
metabolism and elimination of lipophilic compounds from the rogen) for sequencing. Sequencing revealed the presence of two spe-
cies of UGT3A2 cDNA, a full-length sequence corresponding to
body, the liver, kidney, and gastrointestinal tract contain
NM_174914, and a sequence missing exon 2, corresponding to Gen-
most members of the UGT1 and UGT2 families. However,
Bank accession number NM_001168316. Both cDNAs were cloned
into the pEF-IRESpuro6 expression vector, which contains a puro-
mycin resistance gene (Hobbs et al., 1998). Expression vectors con-
taining UGT3A2 in either the forward or reverse direction were
transfected into HEK293T cells, and cell lines stably expressing
UGT3A2 proteins were selected with puromycin (2 ␮g/ml). Ex-
there are differences in UGT content between these organs,
as exemplified by the presence of UGT1A7, UGT1A8, and
UGT1A10 in the gastrointestinal tract and their absence
from the liver and kidney (Mojarrabi and Mackenzie, 1998).
Other organs and tissues generally contain lower amounts of
UGT and/or have a more restricted complement of UGTs. pressed UGT3A2 was analyzed by Western blotting and enzyme
They may also contain UGTs that are expressed poorly in the activity assays.
liver or gastrointestinal tract, for example, UGT2A1 and
UGT2A2, which are mainly found in nasal mucosa (Sneitz et
al., 2009). The selective expression of UGTs in a tissue or
organ, compared with the liver, kidney, and gastrointestinal
tract, may reflect a special need for selective glucuronidating
capacity in that tissue or organ. For example, UGT2A1 and
Western Blotting. Proteins in lysates from HEK293T cells stably
expressing UGT3A cDNA were separated on SDS-polyacrylamide
gels and transferred to nitrocellulose membranes as described pre-
viously (Udomuksorn et al., 2007; Mackenzie et al., 2008). UGT3A2
protein was detected with UGT3A2 antibody and a secondary goat
anti-rabbit antibody conjugated with peroxidase (Neomarkers;
ThermoFisher Scientific, Fremont, CA). Immunocomplexes were
visualized with the Supersignal West Pico chemiluminescent kit
2
A2 glucuronidate phenolic compounds (Sneitz et al., 2009),
which may aid in odorant signal termination, and steroid-
(
ThermoFisher Scientific). The UGT3A2 antibody was prepared
responsive breast, prostate, and adipose tissues express the using amino acids 71 to 124 as antigen. This region was amplified
steroid-metabolizing UGT2B15 or UGT2B17 (Barbier and from the UGT3A2 expression vector by PCR with 5Ј- GTAGGATCC-
B e´ langer, 2008; Ohno and Nakajin, 2009), which have been GAAAAATCATATCAAGTTATC-3Ј and 5Ј-GTACTCGAGTCTTCCTT-
CGATAAAATGACTGCACTGCAACGC-3Ј as the forward and reverse
shown to be involved in steroid-signal termination (Choui-
primers, respectively. The BamH1 and Xho1 sites of the forward and
nard et al., 2008). Hence, characterization of both the cata-
reverse primers, respectively (underlined), were used to clone the PCR
lytic properties and tissue distribution of all UGTs is impor-
product into the pET23a bacterial expression vector, which introduces a
tant in elucidating their role in endogenous biochemistry and
six-histidine C-terminal tag to proteins (Novagen, Madison, WI). Esch-
in drug metabolism and toxicity.
Based on an analysis of the human genome, UGT3A2 is the
last member of the UGT superfamily whose function and
erichia coli (BL21-DE3) was transformed with this construct and
UGT3A2 antigen was purified on a nickel-nitrilotriacetic acid
column (QIAGEN). The purified antigen was used to prepare
tissue distribution have not been characterized. Here we antibody in rabbits.
report the cloning and expression of UGT3A2 and demon-
Quantitative PCR. The FirstChoice Human Total RNA Survey
strate that it is a xenobiotic-conjugating enzyme with a broad Panel (Ambion, Austin, TX) was used as template to quantify levels of
UGT3A2 transcripts in various human tissues with the Rotor-Gene 300
substrate selectivity but a unique UDP sugar selectivity and
tissue distribution.
(
QIAGEN) thermal cycler. The forward and reverse primers specific for
UGT3A2 were 5Ј-CATATCAAGTTATCAGTTGGCTTG-3 and 5Ј-ACT-
GCACTGCAACGCCAAGTA-3Ј, which correspond to nucleotides 218 to
2
41 and 346 to 366 of UGT3A2, respectively. The cycling parameters
consisted of one cycle at 95°C for 15 min and then 40 cycles of 95°C for
Materials. Radioactive and nonradioactive UDP sugars were ob- 10 s, 59°C for 15 s, and 72°C for 20 s. UGT3A2 plasmid was used as
Materials and Methods
14
tained from the following sources: [ C]UDP-galactose, -glucuronic standard to determine transcript copy number.
acid, and -xylose were from PelkinElmer Life and Analytical Sci- Enzyme Assays. For assays to assess substrate preference, gly-
ences (Waltham, MA); [ C]UDP-glucose and -N-acetylglucosamine cosidation reactions were performed as described previously (Mack-
14

4
74
MacKenzie et al.
enzie et al., 2008). In brief, incubations at 37°C for 1 h contained 100 grated with an apparent molecular mass of approximately
mM phosphate buffer, pH 7.5, 4 mM magnesium chloride, 100 ␮g of 49 kDa (Fig. 1, lane 4).
1
4
HEK293T cell lysate, 200 ␮M aglycone, and 2 mM [ C]UDP-sugar
(
Catalytic Properties of UGT3A2. The two members of
the UGT3 family are 78% identical in sequence. Because
UGT3A1 uses UDP-N-acetylglucosamine as sugar donor, we
initially explored the hypothesis that UGT3A2 may also
function as a UDP N-acetylglucosaminyltransferase. How-
ever, preliminary screens with a variety of potential sub-
strates and UDP N-acetylglucosamine did not reveal an ac-
tivity for UGT3A2 (data not shown). Both proteins contain a
0.1 ␮Ci/mmol). Radioactive products were separated by thin-layer
chromatography (Mackenzie et al., 2008) and quantified by expo-
sure to a Phosphor Screen (GE Healthcare), which was scanned
with a Typhoon 9400 scanner (GE Healthcare). Standard curves
with known amounts of [ C]UDP-sugar were constructed to
quantify product formation. All reactions were carried out under
conditions to give linear rates with respect to incubation time and
protein concentration.
1
4
Kinetic studies with 4-MU and each of the UDP sugars used the signal peptide, signature sequence, transmembrane domain
same incubation conditions as those described for the activity screen- and putative ER retention signal (Fig. 2). A more detailed
ing experiments, except the incubation time was 15 min, and non- comparison between the amino acid sequences of the two
radiolabeled UDP sugars were used. Rates of 4-MU glycoside forma-
UGT3A proteins revealed that the putative substrate-bind-
tion were determined at 9 or 10 4-MU concentrations over the ranges
ing domain (residues 23–250) and UDP sugar-binding do-
10 to 250 ␮M (UDP-glucose as cofactor) or 30 to 1000 ␮M (UDP-
main (residues 251–486) of the two proteins vary by 22 and
0%, respectively. This contrasts to members of the UGT1A
galactose and UDP-xylose as cofactors). UDP-glucose and UDP-xy-
lose kinetics were also characterized with 4-MU (2 mM) as the fixed
substrate; incubations included nine UDP-sugar concentrations in
the range of 25 to 5000 ␮M. All incubations were performed in
duplicate (Ͻ10% variance between duplicate samples). 4-MU glyco-
2
and UGT2 families, in which the putative substrate-binding
domain is much less conserved than the putative UDP sugar-
binding domain. In the UGT1A family, the putative UDP
side concentrations in incubation samples from kinetic experiments sugar-binding domain is identical between all nine members
were analyzed by HPLC. Chromatography was performed with an of this family, whereas their putative substrate-binding do-
Agilent 1100 HPLC system fitted with a NovaPak C18 column (3.9 ϫ mains often vary by Ͼ50%. Likewise, members of the UGT2B
1
50 mm, 4 ␮m particle size; Waters Corporation, Milford, MA) with
family have much less variation in their UDP sugar-binding
domains, as exemplified by a comparison between UGT2B4
and UGT2B7, in which their putative substrate-binding and
UDP sugar-binding domains vary by 18 and 8%, respectively
UV detection at 316 nm. The mobile phase consisted of an aqueous
component of 10 mM triethylamine (adjusted to pH 2.5 with perchlo-
ric acid), and 10% acetonitrile (A), and acetonitrile (B). Mobile phase
was delivered at a flow rate of 1 ml/min. For 4-MU glucoside and
(
data not shown). This unexpected large variation in the
4
-MU galactoside, initial conditions were 100% A/0% B held for 5.5
putative UDP sugar-binding domains of UGT3A1 and
UGT3A2 compared with the UGT1A and UGT2 families
prompted us to investigate whether UGT3A2 may have the
min, after which time the composition was changed to 65% A/35% B
for 1 min before returning to the initial conditions. Retention times
for 4-MU galactoside and 4-MU glucoside were 5.1 and 6.3 min,
respectively. For 4-MU xyloside, initial conditions were 90% A/10% B capacity to use other UDP sugars in conjugation reactions.
held for 3 min, followed by 70% A/30% B for 1 min before returning Indeed, preliminary screens with other UDP sugars revealed
to the initial conditions. The retention time for 4-MU xyloside was that UDP-glucose and UDP-xylose were effective sugar do-
2
.5 min. The concentration of 4-MU conjugate was determined by
nors in UGT3A2-catalyzed glycosidations (see below).
UGT3A2 was active in the glucosylation of several hydroxy-
lated xenobiotics including 4-MU, 1-hydroxypyrene, 7-hydroxy-
coumarin, and 1-naphthol, and the bioflavones, naringenin,
genistein, and chrysin. It also glucosidated estrogens such
as 17␣-ethinylestradiol, 17␤-estradiol, and diethylstilbestrol.
However, UGT3A2 was inactive toward androgens and bile
comparison of peak areas with those of a calibration curve con-
structed under the same conditions using authentic individual 4-MU
conjugates. Because the formation of each 4-MU glycoside exhibited
hyperbolic kinetics, kinetic constants were derived by fitting the
Michaelis-Menten equation to experimental data (EnzFitter; Biosoft,
Ferguson, MO).
Results
Cloning and Expression of UGT3A2. The UGT3A fam-
ily, consisting of two adjacent genes of seven exons on chro-
mosome 5p13.2, was first identified in databases of the Hu-
man Genome Project in 2000 (●Authors et al., 2001●).
Because initial studies using PCR screening of various cell
lines with UGT3A2-specific primers revealed low levels of
UGT3A2 mRNA in HEK293 cells (data not shown), this cell
line was used to clone the UGT3A2 cDNA. Recombinant
UGT3A2 protein, when overexpressed in HEK293 cells, has
an apparent molecular mass of 53 kDa (Fig. 1, lane 3). De-
spite low but detectable levels of endogenous UGT3A2 tran-
scripts in HEK293 cells, UGT3A2 protein was not detected in
untransfected cells (Fig. 1, lane 1) or in cells ectopically
expressing UGT3A1 (used as a control to demonstrate the
kDa
1
00
7
5
0
5
37
specificity of the UGT3A2 antibody) (Fig. 1, lane 2). The Fig. 1. Expression of UGT3A in HEK293T cells. Lysates of HEK293 cells
transfected with UGT3A1, UGT3A2, and UGT3A2del-exon2 cDNAs were
recombinant UGT3A2del-exon2 protein, which is missing
the 34 amino acids of exon 2, was also overexpressed by
examined by Western blotting with an antibody specific for UGT3A2. The
presence of UGT3A2 and UGT3A2del-exon2 protein is denoted by arrows.
transfection of HEK293 cells. This shorter protein mi- The molecular weight markers are indicated on the left.

UGT3A2 Glycosidates Xenobiotics with Glucose and Xylose
475
acids with UDP-glucose as the cofactor (Table 1). UGT3A2 eled by the Hill equation. Derived values of S50, Vmax, and n
could also use UDP-xylose to conjugate the above substrates (the Hill coefficient) were 357 Ϯ 11 ␮M, 25,480 Ϯ 242 pmol/
(Table 1) but was inactive when UDP glucuronic acid and UDP min ⅐ mg, and 0.86 Ϯ 0.01, respectively, for UDP-glucose, and
N-acetylglucosamine were used as sugar donors (Fig. 3). Al- 631 Ϯ 21 ␮M, 3100 Ϯ 42 pmol/min ⅐ mg, and 0.90 Ϯ 0.01,
though UGT3A2 was active with UDP-galactose as cofactor, its respectively for UDP-xylose. UDP-galactose kinetics were
activity with this sugar donor was substantially less than with not characterized because of the presumed contamination
UDP-glucose and UDP-xylose. For example, the rates of 4-MU with the alternate sugar donor UDP-glucose. It should be
xylosidation and galactosidation were 60 and 2%, respectively, noted that UDP-glucuronic acid also exhibits negative coop-
of that for glucosidation, when incubations were conducted with erative kinetics with UGT1A9 and UGT2B7 as the enzyme
2
mM UDP-sugar for 15 min at 37°C.
To further examine the enzymatic characteristics of 2004).
UGT3A2, its relative glycosidation activity with the three In contrast to UGT3A2, cells transfected with UGT3A2del-
sources and 4-MU as the fixed substrate (Tsoutsikos et al.,
separate cofactors (UDP-galactose, UDP-glucose, and UDP- exon2 cDNA synthesized a 49-kDa protein (Fig. 1, lane 4)
xylose) was characterized kinetically with 4-MU as the agly- that was devoid of glycosidation activity. The UGT3A2del-
cone using an HPLC method that measured formation of the exon2 variant is missing all 34 amino acids of exon 2, includ-
individual galactoside, glucoside, and xyloside conjugates. ing His35, the presumptive catalytic base required for catal-
Formation of 4-MU-galactoside, 4-MU-glucoside, and 4-MU- ysis (Kubota et al., 2007; Kerdpin et al., 2009).
xyloside followed hyperbolic kinetics that were well described
Distribution of UGT3A2. Transcripts encoding UGT3A2
by the single-enzyme Michaelis-Menten equation (Fig. 4). were detected in human thymus, testis, and kidney, as as-
Derived kinetic constants are given in Table 2. 4-MU-gluco- sessed by quantitative PCR using 20 normal human tissue
side formation exhibited both the lowest apparent Km and samples, each of which contained a pool of RNA from 3 donors
highest Vmax values. As a consequence, the intrinsic clear- (Fig. 5). Only traces of transcript were detected in the liver,
ance (CLint, calculated as Vmax/K ) for 4-MU-glucoside was gastrointestinal tract, and other tissue samples.
m
2
.4- and 80-fold higher than the intrinsic clearances of 4-MU-
xyloside and 4-MU-galactoside, respectively. It should be
noted, however, that 4-MU-glucoside was also detected as a
product in incubations of 4-MU with UDP-galactose, presum-
Discussion
Although the substrate profile, catalytic properties, and
ably because of the presence of UDP-glucose as a contami- tissue distribution of most human UGTs have been exten-
nant in the commercial source of UDP-galactose. Thus, Km sively characterized, this is the first report describing the
and Vmax values for 4-MU-galactoside may be over- and un- function and distribution of UGT3A2. In common with UGT1
derestimated, respectively, as a consequence of competition and UGT2 enzymes, UGT3A2 has a molecular mass in the
by the contaminating cofactor.
50- to 60-kDa range and broad substrate specificity, with the
UDP-glucose and UDP-xylose kinetics were additionally capacity to glycosidate a range of xenobiotics, including
characterized with 4-MU as the fixed substrate. Both fol- 4-MU, the classic “universal” substrate of most UGT1 and
lowed weak negative cooperative kinetics, which were mod- UGT2 forms (Uchaipichat et al., 2004). The kinetic parame-
3
3
A2
A1
1 MAGQRVLLLVGFLLPGVLLSEAAKILTISTVGGSHYLLMDRVSQILQDHGHNVTMLNHKR
1 MVGQRVLLLVAFLLSGVLLSEAAKILTISTLGGSHYLLLDRVSQILQEHGHNVTMLHQSG
*
******** *** *************** ******* ******** ********
3
3
A2
A1
61 GPFMPDFKKEEKSYQVISWLAPEDHQREFKKSFDFFLEETLGGRGKFENLLNVLEYLALQ
61 KFLIPDIKEEEKSYQVIRWFSPEDHQKRIKKHFDSYIETALDGRKESEALVKLMEIFGTQ
*
* * ******** * *****
** **
*
* **
* *
*
*
3
3
A2
A1
121 CSHFLNRKDIMDSLKNENFDMVIVETFDYCPFLIAEKLGKPFVAILSTSFGSLEFGLPIP
121 CSYLLSRKDIMDSLKNENYDLVFVEAFDFCSFLIAEKLVKPFVAILPTTFGSLDFGLPSP
*
*
* ************ * * ** ** * ******* ******* * **** **** *
3
3
A2
A1
181 LSYVPVFRSLLTDHMDFWGRVKNFLMFFSFCRRQQHMQSTFDNTIKEHFTEGSRPVLSHL
181 LSYVPVFPSLLTDHMDFWGRVKNFLMFFSFSRSQWDMQSTFDNTIKEHFPEGSRPVLSHL
Fig. 2. Comparison of UGT3A1 and UGT3A2 protein se-
quences. UGT3A2 is aligned above UGT3A1, and identical
residues are indicated by an asterisk. The signal peptide
and putative transmembrane regions are highlighted in
gray. The lysine residues of the C-terminal dilysine motif
are in boldface type and underlined. The signature se-
quence, which defines the UGT superfamily, is highlighted
in boldface type.
*
****** ********************** * * ************* **********
3
3
A2
A1
241 LLKAELWFINSDFAFDFARPLLPNTVYVGGLMEKPIKPVPQDLENFIAKFEDSGFVLVTL
241 LLKAELWFVNSDFAFDFARPLLPNTVYIGGLMEKPIKPVPQDLDNFIANFGDAGFVLVAF
*
******* ****************** *************** **** * * *****
3
3
A2
A1
301 GSMVNTCQNPEIFKEMNNAFAHLPQGVIWKCQCSHWPKDVHLAANVKIVDWLPQSDLLAH
301 GSMLNTHQSQEVLKKMHNAFAHLPQGVIWTCQSSHWPRDVHLATNVKIVDWLPQSDLLAH
*
** ** *
*
* * ************ ** **** ***** ****************
3
3
A2
A1
361 PSIRLFVTHGGQNSIMEAIQHGVPMVGIPLFGDQPENMVRVEAKKFGVSIQLKKLKAETL
361 PSIRLFVTHGGQNSVMEAIRHGVPMVGLPVNGDQHGNMVRVVAKNYGVSIRLNQVTADTL
*
************* **** ******* * *** ***** ** **** *
* **
3
3
A2
A1
421 ALKMKQIMEDKRYKSAAVAASVILRSHPLSPTQRLVGWIDHVLQTGGATHLKPYVFQQPW
421 TLTMKQVIEDKRYKSAVVAASVILHSQPLSPAQRLVGWIDHILQTGGATHLKPYAFQQPW
*
*** ******** ******* * **** ********* ************ *****
3
3
A2
A1
481 HEQYLLDVFVFLLGLTLGTLWLCGKLLGMAVWWLRGARKVKET
481 HEQYLIDVFVFLLGLTLGTMWLCGKLLGVVARWLRGARKVKKT
*
**** ************* ********
********* *

4
76
MacKenzie et al.
ters for UGT3A2 catalyzed 4-MU-glucoside formation (Km
ϭ
UGT3A2 is also active toward estrogens but has little activity
8
2 ␮M and Vmax ϭ 3.9 nmol/min ⅐ mg) are comparable with toward androgens and bile acids. However, unlike the other
those of UGT1 and UGT2 enzymes for 4-MU glucuronidation, human UGTs, UGT3A2 preferentially uses UDP-glucose and
measured under similar assay conditions (e.g., Km or S50 UDP-xylose as sugar donors in glycosidation reactions. The
values of 59, 78, 13, and 462 ␮M and Vmax values of 0.4, 82, preference for these UDP sugars is unique to this enzyme,
8
.3, and 1.0 nmol/min ⅐ mg for UGT1A1, UGT1A6, UGT1A9, because the major sugar donors for the other human UGTs
and UGT2B7, respectively) (Uchaipichat et al., 2004). are UDP glucuronic acid (UGT1 and UGT2 forms), UDP
N-acetylglucosamine (UGT3A1), and UDP galactose (UGT8).
The apparent affinities of UGT3A2, assessed as S50 values,
for UDP-glucose (S50 of 357 ␮M) and UDP-xylose (S50 of 631
TABLE 1
Substrates of UGT3A2 expressed in HEK293T cells
UGT3A2 activities were assayed as described under Materials and Methods.
␮M) are in the same range as those of UGT1 and UGT2 forms
Activity
toward UDP glucuronic acid (e.g., S50 of 88 ␮M for UGT1A9
and Km of 493 ␮M for UGT2B7) (Tsoutsikos et al., 2004).
Given that quite divergent UGTs (e.g., UGT1A and UGT2B
forms, which may differ in sequence by Ͼ50%) have the same
UDP-sugar preference, the finding that two closely related
members of the same UGT family have different UDP-sugar
specificities was unexpected. The functional significance of
this is obscure, but in the case of UGT3A2, it may relate to
the possibility of preserving UDP glucuronic acid for glycos-
aminoglycan (GAG) synthesis. GAGs are synthesized in most
body tissues and have important roles in controlling cell
growth and differentiation during embryogenesis and in the
adult (Prydz and Dalen, 2000). Because the synthesis of
these compounds is regulated by UDP glucuronic acid avail-
ability (Spicer et al., 1998; Lind et al., 1999), there is the
potential for competition between drug glucuronidation and
UDP Glucose
UDP Xylose
pmol ⅐ minϪ1 ⅐ mg lysate proteinϪ1
Xenobiotic
Drug
Mycophenolic acid
453
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
607
Paracetamol
Phenobarbital
Sulfamethoxazole
Ibuprofen
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
Ketoprofen
Tetracycline
Chloramphenicol
Bioflavones
Naringenin
Genistein
842
659
403
358
278
257
235
100
39
33
28
Ϫ
Ϫ
Ϫ
ϩ
ϩ
Ϫ
ϩ
Ϫ
ϩ
470
523
352
N.D.
246
Chrysin
Isorhamnetin
Biochanin A
Acacetin
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
Diadzein
Formononetin
4
Ј-Hydroxyflavanone
Ј-Hydroxyflavanone
2
Primuletin
Baicalein
Epigallocatechin gallate
Morin
4
2
3
Ј-Hydroxyflavone
Ј-Hydroxyflavone
-Hydroxyflavone
4MU
conjugates
Kaempferol
Quercetin
Luteolin
Epichatechin
Other
Ϫ
4
-Methylumbelliferone
959
865
760
594
570
225
21
686
462
311
222
205
82
1
7
8
1
4
-Hydroxypyrene
-Hydroxycoumarin
Ј-Hydroxyquinoline
-Naphthol
-Nitrophenol
Phenolthalein
N.D.
N.D.
(Ϫ)Borneol
Ϫ
Steroids
1
7␣-Ethinylestradiol
7␤-Estradiol
104
86
66
44
39
32
Ϫ
Ϫ
Ϫ
Ϫ
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
1
Diethylstilbestrol
Estrone
Estriol
_
_
_
_
17␣-Estradiol
+
+
+
+
Androsterone
Testosterone
Dihydrotestosterone
Etiocholanolone
Bile Acids
Ursodeoxycholic acid
Chenodeoxycholic acid
Lithocholic acid
Hyodeoxycholic acid
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
Fig. 3. UDP-sugar selectivity of UGT3A2. Various UDP sugars were used
as cosubstrates in the glycosidation of 4-MU by lysates of HEK293T cells
transfected with UGT3A2 cDNA in the forward (ϩ) and reverse (Ϫ)
orientations. An autoradiograph of the thin-layer chromatography plate
containing 4-MU conjugates and unreacted UDP sugar and/or its break-
␣
␤
-Muricholic acid
-Muricholic acid
14
down products from assays with 250 ␮M substrate and 0.5 mM [ C]UDP
ϩ, low activity detected but not quantifiable by the thin-layer chromatography
method; Ϫ, no activity; N.D., not determined.
sugars is shown. Equal amounts of radiolabeled UDP sugar (0.4 ␮Ci/
mmol) were used in each reaction.

UGT3A2 Glycosidates Xenobiotics with Glucose and Xylose
477
GAG synthesis for the intracellular pool of UDP glucuronic main UGT3A2-expressing organs, thymus, testis, and kid-
acid. Indeed, this principle underlies the use of 4-MU as a ney, have levels of UDP glucose dehydrogenase that are
potent inhibitor of GAG synthesis (Kakizaki et al., 2004; significantly lower (approximately 5-fold) than that of the
Rilla et al., 2005), because 4-MU glucuronidation diverts liver and gastrointestinal tract (Spicer et al., 1998). Hence,
UDP glucuronic acid from GAG biosynthetic pathways. This the use of UDP glucose, rather than UDP glucuronic acid, to
competition may be especially intense in tissues that have inactivate and eliminate lipophilic chemicals in the thymus,
low levels of UDP glucose dehydrogenase, the enzyme that testis, and kidney, would help to preserve GAG synthesis and
synthesizes UDP glucuronic acid from UDP glucose. In this alleviate any competition for the UDP glucuronic acid pool.
situation, UDP glucose may be abundant, but UDP glucu- Moreover, as the concentration of UDP glucose is usually
ronic acid may be limiting. It is interesting to note that the greater than UDP glucuronic acid in those cells in which it
UDP- glucose
UDP - xylose
UDP - galactose
3
2
2
1
500
800
100
400
2500
2000
1500
1000
500
180
120
60
7
00
0
0
0
0
50
100
150
200
250
0
250
500
750
1000
0
250
500
750
1000
[4-methylumbelliferone] (µM)
[4-methylumbelliferone] (µM)
[4-methylumbelliferone] (µM)
Fig. 4. Kinetic plots for 4-MU glycosidation by UGT3A2. Rates of conjugate formation verses substrate concentration plots with UDP-glucose,
UDP-xylose, and UDP-galactose as cofactor are shown. Points (mean of duplicate estimates) are experimentally derived values, whereas curves are
from model fitting.
TABLE 2
Derived kinetic parameters for 4-MU-galactoside, 4-MU-glucoside, and 4-MU-xyloside formation
Data presented as parameter Ϯ S.E. of the parameter fit.
UDP-Sugar
Product
K
m
V
max
CLint
␮
M
pmol ⅐ minϪ1 ⅐ mgϪ1
␮l ⅐ minϪ1 ⅐ mgϪ1
UDP-Galactose
UDP-Glucose
UDP-Xylose
4-MU-gal
4-MU-glu
4-MU-xyl
348 Ϯ 21
82 Ϯ 1
116 Ϯ 5
208 Ϯ 5
3918 Ϯ 26
2332 Ϯ 34
0.6
48
20
Testes
Trachea
Thyroid
Thymus
Spleen
Prostate
Placenta
Sm Intest
Sk Muscle
Ovary
Fig. 5. Tissue distribution of UGT3A2. Levels of UGT3A2
RNA in different tissues as represented by a human tissue
RNA panel were quantified as described under Materials
and Methods. The copies of UGT3A2 mRNA per copy of
GAPDH transcript in each tissue are shown.
Lung
Liver
Kidney
Heart
Esophagus
Colon
Cervix
Brain
Bladder
Adipose
0
50
100
150
200
250
4
copies 3A2/10 copies GAPDH

4
78
MacKenzie et al.
and utilization in various tissues actively synthesizing glycosaminoglycans.
Biochem J 128:215–227.
Hobbs S, Jitrapakdee S, and Wallace JC (1998) Development of a bicistronic vector
driven by the human polypeptide chain elongation factor 1alpha promoter for
creation of stable mammalian cell lines that express very high levels of recombi-
nant proteins. Biochem Biophys Res Commun 252:368–372.
Kakizaki I, Kojima K, Takagaki K, Endo M, Kannagi R, Ito M, Maruo Y, Sato H,
Yasuda T, Mita S, et al. (2004) A novel mechanism for the inhibition of hyaluronan
biosynthesis by 4-methylumbelliferone. J Biol Chem 279:33281–33289.
Kerdpin O, Mackenzie PI, Bowalgaha K, Finel M, and Miners JO (2009) Influence of
N-terminal domain histidine and proline residues on the substrate selectivities of
human UDP-glucuronosyltransferase 1A1, 1A6, 1A9, 2B7, and 2B10. Drug Metab
Dispos 37:1948–1955.
Kubota T, Lewis BC, Elliot DJ, Mackenzie PI, and Miners JO (2007) Critical roles of
residues 36 and 40 in the phenol and tertiary amine aglycone substrate selectivities of
UDP-glucuronosyltransferases 1A3 and 1A4. Mol Pharmacol 72:1054–1062.
Lind T, Falk E, Hjertson E, Kusche-Gullberg M, and Lidholt K (1999) cDNA cloning and
expression of UDP-glucose dehydrogenase from bovine kidney. Glycobiology 9:595–600.
Mackenzie PI (2000) UDP Glucuronosyltransferase: Nomenclature and the Human
Genome Project, 10th International Workshop on Glucuronidation and the UDP
Glucuronosyltransferases; 22–25 April 2001; Hyogo, Japan.
Mackenzie PI, Bock KW, Burchell B, Guillemette C, Ikushiro S, Iyanagi T, Miners JO,
Owens IS, and Nebert DW (2005) Nomenclature update for the mammalian UDP glyco-
syltransferase (UGT) gene superfamily. Pharmacogenet Genomics 15:677–685.
Mackenzie PI, Owens IS, Burchell B, Bock KW, Bairoch A, B e´ langer A, Fournel-
Gigleux S, Green M, Hum DW, Iyanagi T, et al. (1997) The UDP glycosyltrans-
ferase gene superfamily: recommended nomenclature update based on evolution-
ary divergence. Pharmacogenetics 7:255–269.
Mackenzie PI, Rogers A, Treloar J, Jorgensen BR, Miners JO, and Meech R (2008)
Identification of UDP glycosyltransferase 3A1 as a UDP N-acetylglucosaminyl-
transferase. J Biol Chem 283:36205–36210.
Meech R and Mackenzie PI (1997) Structure and function of uridine diphosphate
glucuronosyltransferases. Clin Exp Pharmacol Physiol 24:907–915.
Meech R and Mackenzie PI (2010) UGT3A: novel UDP-glycosyltransferases of the
UGT superfamily. Drug Metab Rev 42:43–52.
Miners JO and Mackenzie PI (1991) Drug glucuronidation in humans. Pharmacol
Ther 51:347–369.
Miners JO, Mackenzie PI, and Knights KM (2010) The prediction of drug-
glucuronidation parameters in humans: UDP-glucuronosyltransferase enzyme-
selective substrate and inhibitor probes for reaction phenotyping and in vitro-in
vivo extrapolation of drug clearance and drug-drug interaction potential. Drug
Metab Rev 42:189–201.
Miners JO, Smith PA, Sorich MJ, McKinnon RA, and Mackenzie PI (2004) Predicting
human drug glucuronidation parameters: application of in vitro and in silico
modeling approaches. Annu Rev Pharmacol Toxicol 44:1–25.
Mojarrabi B and Mackenzie PI (1998) Characterization of two UDP glucuronosyl-
transferases that are predominantly expressed in human colon. Biochem Biophys
Res Commun 247:704–709.
Ohno S and Nakajin S (2009) Determination of mRNA expression of human UDP-
glucuronosyltransferases and application for localization in various human tissues by
real-time reverse transcriptase-polymerase chain reaction. Drug Metab Dispos 37:32–40.
Prydz K and Dalen KT (2000) Synthesis and sorting of proteoglycans. J Cell Sci
has been measured (Gainey and Phelps, 1972), it is likely
that glucosylation may be less subject to fluctuating levels of
UDP sugars than glucuronidation in extrahepatic tissues.
Indeed, evidence that the supply of UDP glucuronic acid is
rate-limiting for androgen glucuronidation in the prostate
has been provided (Wei et al., 2009).
In contrast to most members of the UGT1 and UGT2 families,
UGT3A2 mRNA is not found in two of the major organs of drug
metabolism, the liver and gastrointestinal tract, but is found in
the kidney, thymus, and testis, as mentioned previously. Many
members of the UGT1 and UGT2 families are expressed in the
kidney at much higher levels than UGT3A2. These include
UGT1A6, UGT1A9, and UGT2B7, whose mRNA levels are ap-
proximately 20-, 150-, and 120-fold greater than those of
UGT3A2, respectively, based on data for these UGTs relative to
levels of glyceraldehyde-3-phosphate dehydrogenase (GAPDH)
(Ohno and Nakajin, 2009). Hence, it is unlikely that UGT3A2
contributes significantly to glycosidation of lipophilic chemicals
in this organ. However, UGT3A2 is expressed at higher
amounts than other UGTs in thymus and testis. Only UGT1A5,
UGT2B4, and UGT2B7 have been detected in thymus (Ohno
and Nakajin, 2009), but these are at levels approximately 31-,
9-, and 7-fold less then that of UGT3A2. Testis contains equiv-
alent amounts of UGT3A2 and UGT2B15 but 18- and 35-fold
less UGT2B4 and UGT2B17, respectively. The exact role of
UGT3A2 in the thymus and testis is unknown but is most likely
protective in nature, although a specific role in modulating ligand
concentrations in signaling pathways cannot be excluded.
Finally, the discovery that UGT3A1 and UGT3A2 have
divergent UDP sugar preferences provides a unique oppor-
tunity to identify the amino acids involved in UDP sugar
selection via site-directed mutagenesis and analyses of chi-
meras of these two closely related UGTs. This approach is
currently being pursued in our laboratory.
In summary, we demonstrate that UGT3A2 is a novel UDP
glycosyltransferase with a unique UDP sugar selectivity, in
that it has the capacity to covalently attach glucose and
xylose to many foreign chemicals and estrogen-like com-
pounds. Its high expression in the thymus and testis and its
poor expression in the liver and gastrointestinal tract sug-
gest a unique role for this enzyme in drug metabolism. Fur-
ther substrate characterization and a more detailed exami-
nation of the expression of UGT3A2 in cells within the
thymus and testis may help clarify this role.
113:193–205.
Radominska-Pandya A, Czernik PJ, Little JM, Battaglia E, and Mackenzie PI (1999)
Structural and functional studies of UDP-glucuronosyltransferases. Drug Metab
Rev 31:817–899.
Rilla K, Siiskonen H, Spicer AP, Hyttinen JM, Tammi MI, and Tammi RH (2005)
Plasma membrane residence of hyaluronan synthase is coupled to its enzymatic
activity. J Biol Chem 280:31890–31897.
Sneitz N, Court MH, Zhang X, Laajanen K, Yee KK, Dalton P, Ding X, and Finel M
(2009) Human UDP-glucuronosyltransferase UGT2A2: cDNA construction, ex-
pression, and functional characterization in comparison with UGT2A1 and
UGT2A3. Pharmacogenet Genomics 19:923–934.
Spicer AP, Kaback LA, Smith TJ, and Seldin MF (1998) Molecular cloning and
characterization of the human and mouse UDP-glucose dehydrogenase genes.
J Biol Chem 273:25117–25124.
Tsoutsikos P, Miners JO, Stapleton A, Thomas A, Sallustio BC, and Knights KM (2004)
Evidence that unsaturated fatty acids are potent inhibitors of renal UDP-
glucuronosyltransferases (UGT): kinetic studies using human kidney cortical micro-
somes and recombinant UGT1A9 and UGT2B7. Biochem Pharmacol 67:191–199.
Tukey RH and Strassburg CP (2000) Human UDP-glucuronosyltransferases: metab-
olism, expression, and disease. Annu Rev Pharmacol Toxicol 40:581–616.
Uchaipichat V, Mackenzie PI, Guo XH, Gardner-Stephen D, Galetin A, Houston JB,
and Miners JO (2004) Human udp-glucuronosyltransferases: isoform selectivity
and kinetics of 4-methylumbelliferone and 1-naphthol glucuronidation, effects of
organic solvents, and inhibition by diclofenac and probenecid. Drug Metab Dispos
Authorship Contributions
Participated in research design: Mackenzie, Miners, and Meech.
Conducted experiments: Mackenzie, Rogers, Elliot, Chau, Hulin,
and Meech.
Performed data analysis: Mackenzie, Rogers, Elliot, Chau, Hulin,
Miners, and Meech.
Wrote or contributed to the writing of the manuscript: Mackenzie,
Elliot, Miners, and Meech.
32:413–423.
Udomuksorn W, Elliot DJ, Lewis BC, Mackenzie PI, Yoovathaworn K, and Miners
JO (2007) Influence of mutations associated with Gilbert and Crigler-Najjar type
II syndromes on the glucuronidation kinetics of bilirubin and other UDP-
glucuronosyltransferase 1A substrates. Pharmacogenet Genomics 17:1017–1029.
Wei Q, Galbenus R, Raza A, Cerny RL, and Simpson MA (2009) Androgen-stimulated
UDP-glucose dehydrogenase expression limits prostate androgen availability
without impacting hyaluronan levels. Cancer Res 69:2332–2339.
References
Barbier O and B e´ langer A (2008) Inactivation of androgens by UDP-glucuronosyltrans-
ferases in the human prostate. Best Pract Res Clin Endocrinol Metab 22:259–270.
Bosio A, Binczek E, Le Beau MM, Fernald AA, and Stoffel W (1996) The human gene
CGT encoding the UDP-galactose ceramide galactosyl transferase (cerebroside
synthase): cloning, characterization, and assignment to human chromosome 4,
band q26. Genomics 34:69–75.
Chouinard S, Yueh MF, Tukey RH, Giton F, Fiet J, Pelletier G, Barbier O, and
B e´ langer A (2008) Inactivation by UDP-glucuronosyltransferase enzymes: the end
of androgen signaling. J Steroid Biochem Mol Biol 109:247–253.
Address correspondence to: Dr. Peter I. Mackenzie, Department of Clinical
Pharmacology, Flinders Medical Centre, Bedford Park, SA 5042, Australia.
E-mail: peter.mackenzie@flinders.edu.au
Gainey PA and Phelps CF (1972) Uridine diphosphate glucuronic acid production