Reconstruction of de novo pathway for synthesis of
UDP-glucuronic acid and UDP-xylose from intrinsic
UDP-glucose in Saccharomyces cerevisiae
Takuji Oka and Yoshifumi Jigami
Research Center for Glycoscience, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Japan
The d-glucuronic acid and d-xylose monosaccharides
are critically important for plants, fungi, vertebrates
and mammals [1–4]. In plants, d-xylose is mainly pre-
sent in the form of cell wall polysaccharides and
N-glycan [5]. In mammals, d-xylose is involved in link-
ing proteoglycans to proteins, and d-glucuronic acid
is involved in the elongation of various types of
glycosaminoglycans [6]. Some of the O-linked glycans,
including the Xyl-a1,3-Xyl-a1,3-Glc-b1-O-Ser chain,
have also been identified as d-xylose-containing oligo-
saccharides in bovine factor IX [7].
Glycosyltransferases make use of UDP-d-glucuronic
acid (UDP-GlcA) and UDP-d-xylose (UDP-Xyl) in the
synthesis of cell wall polysaccharides and for attachment
Keywords
Saccharomyces cerevisiae; UDP-glucuronic
acid; UDP-glucuronic acid decarboxylase;
UDP-glucose dehydrogenase; UDP-xylose
Correspondence
Y. Jigami, Research Center for
Glycoscience, National Institute of Advanced
Industrial Science and Technology (AIST),
AIST Tsukuba Central 6, Higashi 1-1,
Tsukuba 305-8566, Japan
Fax: +81 29 861 6161
Tel: +81 29 861 6160

Abbreviations
AtXT1, xylosyltransferase 1; GDP-Fuc, GDP-
L-fucose; GDP-Man, GDP-D-mannose; TEAA, triethylamine acetate; UDP-Glc, UDP-D-glucose;
UDP-GlcA, UDP-
D-glucuronic acid; UDP-Xyl, UDP-D-xylose; UGD, UDP-glucose dehydrogenase; UXS, UDP-xylose synthase.
FEBS Journal 273 (2006) 2645–2657 ª 2006 The Authors Journal compilation ª 2006 FEBS 2645
of oligosaccharides to proteins in a variety of organisms.
For example, O-xylosyltransferase I can transfer d-xy-
lose to proteoglycan core proteins in humans, and
b1,2-xylosyltransferase can transfer d-xylose via a b1,2-
linkage to the b-linked mannose of N-linked oligosac-
charides in plants [5,8]. It has also been reported that in
plants, xylosyltransferase 1 (AtXT1) is involved in cell
wall a1,6-xyloglucan biosynthesis [9].
In higher eukaryotes, UDP-Xyl is synthesized from
UDP-d-glucose (UDP-Glc) by two enzymes. These are
UDP-glucose dehydrogenase (UGD, EC 1.1.1.22),
which catalyzes formation of UDP-GlcA from UDP-
Glc by a concomitant reduction of two molecules of
NAD
+
to NADH, and UDP-glucuronic acid decarb-
oxylase (UDP-xylose synthase; UXS, EC 4.1.1.35),
which catalyzes the formation of UDP-Xyl from UDP-
GlcA via decarboxylation of the C6-carboxylic acid
component of glucuronic acid (Fig. 1). Moreover,
UDP-GlcA can also be generated by oxidation of
myoinositol in plants; however, the biological signifi-
cance and quantitative contribution of this pathway
are not yet clear (Fig. 1) [10].

of the wall polysaccharide mass. However, the yeast cell
wall consists mostly of b-glucan, a-mannan and chitin,
and there are no synthetic or breakdown pathways for
UDP-GlcA and UDP-Xyl. In the budding yeast Sac-
charomyces cerevisiae, UDP-Glc, which is a substrate of
b1,3-glucan synthase and b1,6-glucan synthase in the
synthesis of cell wall b-glucan polysaccharides, is abun-
dant in the cytoplasm [19–21]. Therefore, yeast appears
to have the potential to produce large amounts of
UDP-GlcA and UDP-Xyl in the cytoplasm.
Recently, two groups, including our own, reported
the synthesis of GDP-l-fucose (GDP-Fuc) from
inherent cytoplasmic GDP-d-mannose (GDP-Man) by
expressing either Escherichia coli- or A. thaliana-
derived GDP-mannose-4,6-dehydratase and GDP-
4-keto-6-deoxymannose-3,5-epimerase-4-reductase in
UDP-glucose dehydrogenase
(UGD)
UDP-glucuronic acid decarboxylase
(UXS)
2 NAD
2 NADH
+
CO
2
UDP-D-glucose
UDP-
D-glucuronic acid
UDP-
D-xylose

for production of large amounts of this nucleotide
sugar. Thus, we worked to develop a similar system
for in vivo production of UDP-Xyl by introducing
UDP-Xyl synthetic genes into yeast cells. Our system
facilitates efficient production of UDP-GlcA and
UDP-Xyl via conversion of a large precursor pool of
UDP-Glc into the derivative molecules. Here, we
report the generation of yeast strains capable of produ-
cing large amounts of the nucleotide sugars, and also
discuss the implications of our results in yeast for the
study of metabolic regulation in plants.
Results
Cloning and expression of AtUGD1 and AtUXS3
genes in yeast
In order to generate yeast strains capable of produ-
cing large amounts of the nucleotide sugars, we first
cloned and expressed UDP-Xyl synthetic genes in
yeast. The AtUGD1 gene, which encodes UGD, and
the AtUXS3 gene, which encodes UXS, were ampli-
fied by PCR from an A. thaliana cDNA library. Next,
we placed the AtUGD1 and AtUXS3 genes under the
control of the S. cerevisiae constitutive TDH3 promo-
ter (plasmid vectors pRS305-UGD1-VSV-G and
pRS304-UXS3-c-Myc, respectively). AtUGD1 was
N-terminally tagged with VSV-G and AtUXS3 was
C-terminally tagged with c-Myc. The constructs were
integrated into the leu2 locus of chromosome III of
S. cerevisiae strain W303a to yield the strain
TOY1, and into the trp1 locus of chromosome IV of
S. cerevisiae strain W303a to yield the strain TOY2.

6 7 8 9 10 11 12 13 14
Time (min)
Time (min)
W303a
TOY1
A260
B
67891011121314
UDP-Xyl
6 7 8 9 10 11 12 13 14
UDP-GlcA
UDP-GlcA
TOY2W303a
A260
C
1234 1234
AtUXS3
(40 kDa)
AtUGD1
A
(54 kDa)
Fig. 2. Expression of functional AtUGD1 and AtUXS3 in yeast. (A)
Immunoblotting analyses of AtUGD1 and AtUXS3. The presence of
the AtUGD1 enzyme was detected with VSV-G antibody. A 54-kDa
protein band was observed in the TOY1 and TOY3 strains (left
panel). The AtUXS3 enzyme was detectable as a 40-kDa band with
a c-Myc antibody in the TOY2 and TOY3 strains (right panel). Lane
1, Control W303a strain; lane 2, TOY1 strain; lane 3, TOY2 strain;
lane 4, TOY3 strain. (B) In vitro activities of UDP-glucose dehydrog-
enase. (C) In vitro activities of UDP-glucuronic acid decarboxylase.

centrifugation. Next, formic acid saturated with 1-buta-
nol was added to the cell pellets on ice, and sugar-
nucleotides were extracted from the cytoplasm (see
Experimental procedures). Finally, the extracts were
separated by C18 chromatography, and UDP-Glc,
UDP-GlcA and UDP-Xyl fractions were collected based
on the retention times with the standards. The purified
UDP-sugar fractions were analyzed by ESI-MS.
Our expectation was that the TOY1 cells would pro-
duce UDP-GlcA but be unable to convert it to UDP-
Xyl. Consistent with this, both UDP-Glc (m ⁄ z: 564.6)
and UDP-GlcA (m ⁄ z: 578.6) were detected in TOY1
cells (Fig. 3A). As TOY3 carries both enzymes, we
expected that not only UDP-Glc but also UDP-GlcA
and UDP-Xyl would be present in these cells. As expec-
ted, both UDP-Glc (m ⁄ z: 564.6) and UDP-Xyl (m ⁄ z:
534.6) were detected (Fig. 3A). However, UDP-GlcA
was not detected in the cytoplasm of the TOY3 strain.
This suggests that there was a complete conversion of
the UDP-GlcA intermediate generated by AtUGD1
activity to UDP-Xyl as a final product by AtUXS3
activity. To confirm the complete conversion of UDP-
GlcA to UDP-Xyl, we performed ESI-MS analyses on
cytoplasmic fractions of W303a, TOY1 and TOY3
cells. The assay revealed that UDP-GlcA was present
only in the cells of strain TOY1, whereas UDP-Xyl was
only present in the cells of strain TOY3, consistent with
the idea that the TOY1 and TOY3 cells are useful for
production of UDP-GlcA and UDP-Xyl, respectively.
In order to quantify the levels of UDP-sugars in the

[14,24], whereas the UGDs from the virulent bacterial
strain Streptococcus pyogenes and the pathogenic yeast
strain Cryptococcus neoformans form dimers [25,26].
For plants, however, the active complex for UGD has
not previously been defined. To determine the oligo-
meric state of AtUGD1, we purified recombinant
AtUGD1 protein from yeast. To do this, we construc-
ted the TOY4 strain, which harbors an expression
plasmid that should produce a 6 · His-tagged form of
AtUGD1. Next, AtUGD1 was purified by FPLC
from a crude enzyme fraction. A single polypeptide of
approximately 54 kDa was visualized in the final
preparation following SDS ⁄ PAGE (Fig. 4A). The
molecular mass of the purified AtUGD1 was deter-
mined by HPLC analysis. The purified recombinant
AtUGD1 was eluted as a peak at a time of 26.9 min
from the gel filtration column (peak A). The inset
panel shows molecular mass determination for the
peaks with gel filtration standards. On the basis of
the elution of the molecular mass markers, this peak
corresponds to a molecular mass of approximately
328 kDa (Fig. 4B). Based on the predicted monomer
molecular mass of 53 925 Da, this indicates that
AtUGD1 is a hexamer protein. Since no peaks of
monomer AtUGD1 were detected in the purified
recombinant AtUGD1 fraction, the hexamer structure
of AtUGD1 is necessary to form an active enzyme.
AtUGD1 is involved in the maintenance of the
cytoplasmic pool of UDP-Glc in vivo
The quantitative analysis of UDP-sugars reveals that

578.6
534.6
564.6
500 510 520 530 540 550 560 570 580 590 600
(m/z)
intensity
A
UDP-Glc
UDP-GlcA
564.6
UDP-Glc
564.6
UDP-Glc
UDP-Xyl
W303a
TOY1
TOY3
A260
B
0255101520
Time (min)
UDP-Glc
UDP-Glc
UDP-Glc
UDP-Xyl
UDP-GlcA
Fig. 3. In vivo activities of UDP-glucose dehydrogenase and UDP-glucuronic acid decarboxylase. (A) ESI-MS analysis of UDP-sugars in yeast.
Nucleotide sugars were extracted as described in Experimental procedures. Ten A
600 nm
cells of UDPS-C18 fractions from W303a, TOY1

concomitantly converted UDP-GlcA to UDP-Xyl.
UDP-Xyl strongly inhibited AtUGD1 activity, and
then the UDP-Glc pool of TOY3 cells recovered to a
level comparable to that in W303a cells. This result
indicates that AtUGD1 maintains the pool of UDP-
Glc of the cell in cooperation with AtUXS3 via inhibit-
ion of UDP-Xyl by AtUGD1 in vivo.
Discussion
Bioinformatic analysis of several genome sequences
has revealed the presence of many glycosyltransferase-
like genes in the genomes of diverse species, including
the plant A. thaliana [27]. In order to study these glyc-
osyltransferase-like genes, it would be advantageous to
have access to a ready supply of sugar nucleotide sub-
strates to be used in functional analyses. However,
UDP-GlcA and UDP-Xyl have until now been pre-
cious materials. It has been difficult to produce UDP-
GlcA and UDP-Xyl in plants or bacteria, because
UDP-GlcA or UDP-Xyl synthesized in those organ-
isms are further converted to the other UDP-sugars or
used in the synthesis of oligosaccharides and polysac-
charides of the cell wall. In S. cerevisiae, however,
d-glucuronic acid and d-xylose are not components of
the cell wall and are not attached to proteins, which
gives every indication that the yeast S. cerevisiae lacks
consumptive pathways for UDP-GlcA and UDP-Xyl.
13
250
150
100

AtUGD1 cDNA was expressed under the TDH3 promoter in W303a cells. The proteins extracted from the recombinant yeast cells were sep-
arated by 4–20% SDS ⁄ PAGE, and the gel was stained with Coomassie Brilliant Blue. Lane 2, crude protein fraction; lane 3, purified protein
fraction (54 kDa shown by His-AtUGD1); lanes 1 and 4, protein molecular mass markers. (B) Oligomeric form analysis of the 6 · His-tagged
AtUGD1 protein. Purified 6 · His-tagged AtUGD1 protein was fractionated by gel filtration HPLC at a flow rate of 0.20 mLÆmin
)1
. A single
AtUGD1 peak was detected (peak A). The molecular mass of the oligomeric form was estimated by comparison with molecular mass stand-
ards. Average retention time in the column was plotted versus the log of the molecular weight for each standard (inset panel). Standards
used were as follows: thyroglobulin, 669 kDa; ferritin, 440 kDa; catalase, 232 kDa; ovalbumin, 43 kDa.
Table 1. Quantification of UDP-sugar levels in yeast. The amounts
of UDP-glucose, UDP-glucuronic acid and UDP-xylose were calcula-
ted from peak areas shown in Fig. 3B. The amounts of sugar-nucle-
otides are expressed as the amounts of UDP-glucose equivalent,
obtained in the triplicate experiments. ND, not determined.
Strain
UDP-glucose
(lmolÆg
)1
dry weight)
UDP-glucuronic
acid (lmolÆg
)1
dry weight)
UDP-xylose
(lmolÆg
)1
dry weight)
W303a 2.21 ± 0.28 ND ND
TOY1 1.20 ± 0.23 5.71 ± 1.06 ND
TOY3 2.53 ± 0.11 ND 1.69 ± 0.40

expression of active UGD-converting enzymes. It is
known that eukaryotic proteins expressed in E. coli
often form protein inclusion bodies, due to a difference
in the protein-folding system between prokaryotes and
eukaryotes.
As, like A. thaliana, the yeast S. cerevisiae is eukary-
otic, it is perhaps not surprising that AtUGD1 was
successfully expressed in yeast and that the recombin-
ant protein was stable and active. However, one can-
not exclude the possibility that expression levels of
protein are dependent on codon usage of the host. For
example, arginine codons (AGA, 18.9%; AGG,
10.9%) are used at a high frequency in A. thaliana, but
both codons are rarely used in E. coli (AGA, 2.1%;
-100
00
100
200
300
400
500
-100 10 20 30
-100
0
100
200
300
400
500
600

1/UGD activity
(µ
M [UDP-Glc]/min)
-1
(Slope (v vs UDP-Glc
-1
)
-1
[UDP-Xyl] (µM)
[UDP-GlcA] (µ
M)
1/[UDP-Glc]
-1
1/[UDP-Glc] (µM)
-1
(Slope (v vs UDP-Glc
-1
)
-1
B
A
(µM)
Fig. 5. Inhibition of recombinant AtUGD1
protein. (A) Inhibition kinetics on UDP-
D-
xylose (UDP-Xyl) using UDP-
D-glucose (UDP-
Glc) as the variable substrate. (B) Inhibition
kinetics on UDP-
D-glucuronic acid (UDP-

protein was misfolded in bacteria, thus clouding inter-
pretation of the experimental result. A hexameric
structure has been observed for the native soybean
UGD in gel filtration studies [31]. We confirmed the
results of the above studies. In this work, in no case
did we observe any data consistent with a monomer.
Instead, the most probable interpretation of the
results presented here is that the AtUGD1 protein is
active as a hexamer, just as has been found for sev-
eral other eukaryotic proteins of this type. In addi-
tion, Sommer et al. indicated that the Lys279 residue
of human UGD is likely to have a role in maintain-
ing the hexameric structure [24], and the Lys279 resi-
due is conserved in the amino acid sequence of
AtUGD1, consistent with our results.
The quantification of UDP-sugars in yeast cell
extracts revealed that in the TOY1 strain expressing
only AtUGD1, UDP-GlcA accumulated to a large
extent in the cytoplasm. In addition, the TOY3 strain
coexpressing AtUGD1 and AtUXS3 could produce
UDP-Xyl but we did not observe significant accumulat-
ion of the UDP-GlcA intermediate (Table 1). Recently,
Ernst and Klaffke reported the chemical synthesis of
UDP-Xyl [32]; however, their method had some poten-
tial problems, including low yield and contamination of
the final compound with a ⁄ b anomers (a ⁄ b ¼ 5 : 3). In
the case of enzymatic conversion of UDP-sugar, it is
not necessary to consider contamination with a ⁄ b ano-
mers. Furthermore, as our system is based on yeast,
which is essentially a renewable factory for protein

it has been impossible to obtain direct confirmation of
the hypothesis on the basis of the size of the UDP
pool in vivo, because of the complexity of the UDP-
sugar regulation system in plant cells. Our yeast sys-
tem, by contrast, made it possible to quantify changes
in the UDP-sugar pool in vivo, as these cells lack
endogenous UDP-sugar-converting enzymes, with the
exception of enzymes used in the synthesis of UDP-
Glc, UDP-d-galactose and UDP-N-d-acetylglucosa-
mine.
We previously reported that MUR1 and GER1
tightly associate to form a functional complex required
for the stable enzymatic activity that can produce
GDP-Fuc from GDP-Man [23]. However, interaction
between AtUGD1 and AtUXS3 was not observed in
immunoprecipitation experiments (data not shown),
suggesting that the regulation of the UDP-Glc pool is
not the result of direct protein interaction but is
instead mediated by an intermediary inhibition mech-
anism of UDP-Xyl. Thus, the yeast reconstruction sys-
tem will be useful to further understand the regulation
and interaction of UDP-sugar-converting enzymes.
Yeast can be used as a host for the expression of
valuable proteins modified by artificial glycosylation
[38,39]. Kainuma et al. indicated that protein glycosy-
lation remodeling can be carried out using intrinsic
sugar nucleotides in yeast via the introduction of
heterologous genes required for artificial glycosylation
[38]. Here, we built on that success by constructing
recombinant yeast strains that produce the sugar

(5¢-AGAATTCATGTATACTGATATTGAAATGAATAG
ATTGGGTAAAATGGTGAAGATATGCTGCATAGGA
G-3¢) and UGD1-SalI-R (5¢ -AAAAAGTCGACTCATGCC
ACAGCAGGCATATCCTT-3¢); for UGD1-His, UGD1-
His-EcoRI-F (5¢-AGAATTCATGCATCACCATCACCAT
CACATGGTGAAGATATGCTGCATAG-3¢) and UGD1-
SalI-R, UXS3-EcoRI-F (5¢-AGATTCATGGCAGCTACA
AGTGAGAAACAGA-3¢); and for UXS-c-Myc, UXS3-c-
Myc-XhoI-R (5¢-TCTCGAGTTACAAATCTTCTTCAGAA
ATCAATTTTTGTTCGTTTCTTGGGACGTTAAGCCTT
AG-3¢). The PCR products were digested with the appro-
priate restriction enzymes and ligated into similarly digested
YEp352-GAP-II [23] to yield YEp352-GAP-II-UGD1-VSV-
G, Yep352-GAP-II-UGD1-His and YEp352-GAP-II-
UXS3-c-Myc. Next, BamHI fragments that included the
AtUGD1 and AtUXS3 gene expression cassettes from
YEp352-GAP-II-UGD1-VSV-G and YEp352-GAP-II-
UXS3-c-Myc were inserted into the BamHI sites of pRS305
and pRS304 to yield pRS305-UGD1-VSV-G and pRS304-
UXS3-c-Myc, respectively. The DNA sequence of the
expression constructs was confirmed using an ABI PRISM
3100 Genetic Analyzer (Applied Biosystems, Foster, CA).
Immunoblot analysis
Protein concentration was determined using the bicinchoni-
nic acid protein assay reagent (Pierce Biotechnology, Inc.,
Rockford, IL) with bovine serum albumin as a standard.
SDS ⁄ PAGE was performed on crude cell lysates
(D
600 nm
¼ 2.0). Proteins were then transferred to a polyv-

W303a MATa leu2-3, his3-11, trp1-1, can1-100, ade2-1, ura3-1 [40]
TOY1 As in W303a and leu2-3::pRS-305-UGD1-VSV-G This study
TOY2 As in W303a and trp1-1::pRS-304-UXS3-c-Myc This study
TOY3 As in W303a and leu2-3::pRS-305-UGD1-VSV-G, trp1–1::pRS-304-UXS3-c-Myc This study
TOY4 As in W303a harboring expression plasmid YEp352-GAP-II-UGD1-His This study
T. Oka and Y. Jigami Synthesis of UDP-glucuronic acid and UDP-xylose
FEBS Journal 273 (2006) 2645–2657 ª 2006 The Authors Journal compilation ª 2006 FEBS 2653
UGD assay
The in vitro assay for UGD was performed using the follow-
ing reaction mixture (total volume 100 lL): 5 mm UDP-
glucose; 0.5 mm NAD
+
; protease inhibitor (one tablet of
Complete ⁄ 50 mL; Roche, Mannheim, Germany); 50 mm
Tris ⁄ HCl (pH 8.6); and S. cerevisiae cell extract supernatant
(see above) at D
600 nm
¼ 10. Reaction mixtures were incuba-
ted at 30 °C for 60 min and the reaction was stopped by
vortex mixing with 100 lL of ice-cold phenol ⁄ chloroform ⁄
isoamyl alcohol (25 : 24 : 1). Next, 5 lL of the reacted cell
supernatants were analyzed by HPLC with cosmosil 5C
18
-
AR-II (250 · 4.6 mm; Nacalai Tesque, Kyoto, Japan). The
column was equilibrated with 20 mm triethylamine acetate
(TEAA) buffer (pH 7.0) at a flow rate of 1 mLÆmin
)1
. UDP-
sugars were detected by UV

5C
18
-AR-II column (Nacalai Tesque). The column was
equilibrated with 20 mm TEAA buffer (pH 7.0) at a flow
rate of 1 mLÆmin
)1
. UDP-sugars were detected by UV
260 nm
absorbance. The peaks of UDP-Glc, UDP-GlcA and UDP-
Xyl activity that were detected were collected based on the
retention times of the standards. UDP-Glc, UDP-Xyl and
UDP-GlcA fractions from the W303a, TOY1 and TOY3
strains were harvested and mixed, respectively. The mixed
fractions were designated ‘UDPS-C18’. The UDPS-C18
fractions were analyzed by ESI-MS. Mass spectra were
acquired on an Esquire 3000-plus instrument (Bruker
Daltonik GmbH, Bremen, Germany) in the negative-ion
mode. Conditions for ESI-MS were as follows: 68.95 kPa
nebulizer flow, 300 °C nozzle temperature, and 5.0 LÆmin
)1
flow of drying gas (N
2
). Negative-ion spectra were gener-
ated by scanning the m ⁄ z range 500–600.
Analysis and quantification of UDP-sugar
nucleotides
The UDPS-C18 fractions were reseparated on a Develosil
RPAQUEOUS column (250 · 4.6 mm; Nomura Chemical
Co., Ltd, Seto, Japan). The column was equilibrated with
20 mm TEAA buffer (pH 7.0) at a flow rate of 0.7 mLÆ

of protein had been eluted. The enzyme was then eluted by a
gradient up to 500 mm imidazole. The fractions containing
6 · His-tagged AtUGD1 protein were pooled and concen-
trated with a YM30 membrane (Millipore), applied to a Hi-
Load 16 ⁄ 60 Superdex 200-pg column (1.6 cm · 60.0 cm;
GE Healthcare Bio-Sciences Corp.), and equilibrated in buf-
fer B (10 mm Tris ⁄ HCl, pH 8.0, and 150 mm NaCl). The
sample was eluted at a rate of 1 mLÆmin
)1
in buffer B. Act-
ive fractions were concentrated to 1 mL by ultrafiltration
over a YM30 membrane (Millipore), and stored at 4 °C.
The purified enzymes were analyzed by SDS ⁄ PAGE. Protein
concentrations were determined with the bicinchoninic acid
protein assay reagent (Pierce Biotechnology, Inc., Rockford,
IL) using bovine serum albumin as a standard.
Synthesis of UDP-glucuronic acid and UDP-xylose T. Oka and Y. Jigami
2654 FEBS Journal 273 (2006) 2645–2657 ª 2006 The Authors Journal compilation ª 2006 FEBS
Kinetic studies
The level of activity of UGD was estimated by determining
the amount of UDP-GlcA. Reaction mixtures contained
Tris ⁄ HCl (50 mm, pH 8.6), NAD
+
(0.5 mm), UDP-Glc and
10.8 · 10
)6
Units of purified AtUGD1 in a total volume
of 50 lL. Variations in the reaction mixture are noted in
the text. Reactions were started by the addition of NAD
+

parameter was determined by replot analysis.
Molecular mass determination of protein
complex
The functional molecular mass of active 6 · His-tagged
AtUGD1 enzyme complex was determined on a PROTEIN
KW-803 column (Showa Denko K. K., Tokyo, Japan). The
column was equilibrated with buffer B (10 mm Tris ⁄ HCl,
pH 8.0, and 150 mm NaCl). The protein complex was
detected by UV
220 nm
absorbance. Purified recombinant
AtUGD1 was loaded onto the column with an HPLC sys-
tem (Shimadzu Co., Kyoto, Japan) at a flow rate of
0.2 mLÆmin
)1
. Size determination was performed by com-
parison with molecular mass standards (GE Healthcare
Bio-Sciences Corp.) loaded onto the column under the same
conditions. The molecular mass standards used were as fol-
lows: thyroglobulin, 669 kDa; ferritin, 440 kDa; catalase,
232 kDa; ovalbumin, 43 kDa.
Acknowledgements
This work was supported by grants from the New
Energy and Industrial Technology Development
Organization of Japan (NEDO). We thank Dr Shige-
yasu Ito and Minako Takashiba for ESI-MS analysis,
Toshihiko Kitajima for protein purification, and
Dr Takehiko Yoko-o for critical reading of the manu-
script. We are indebted to Drs Ken-ichi Nakayama,
Yasunori Chiba, Xiao-Dong Gao, Yoh-ichi Shimma

ine 53 found in the first epidermal growth factor-like
domain of bovine blood clotting factor IX. J Biol Chem
265, 1858–1861.
8 Gotting C, Kuhn J, Zahn R, Brinkmann T & Kleesiek
K (2001) Molecular cloning and expression of human
UDP-d-xylose: proteoglycan core protein beta-d-xylosyl-
transferase and its first isoform XT-II. J Mol Biol 304,
517–528.
9 Faik A, Price NJ, Raikhel NV & Keegstra K (2002) An
Arabidopsis gene encoding an alpha-xylosyltransferase
involved in xyloglucan biosynthesis. Proc Natl Acad Sci
USA 99, 7797–7802.
10 Kanter U, Usadel B, Guerineau F, Li Y, Pauly M &
Tenhaken R (2005) The inositol oxygenase gene family
of Arabidopsis is involved in the biosynthesis of nucleo-
tide sugar precursors for cell-wall matrix polysacchar-
ides. Planta 221, 243–254.
11 Sieberth V, Rigg GP, Roberts IS & Jann K (1995)
Expression and characterization of UDP-Glc dehydro-
genase (KfiD), which is encoded in the type-specific
region 2 of the Escherichia coli K5 capsule genes.
J Bacteriol 177, 4562–4565.
T. Oka and Y. Jigami Synthesis of UDP-glucuronic acid and UDP-xylose
FEBS Journal 273 (2006) 2645–2657 ª 2006 The Authors Journal compilation ª 2006 FEBS 2655
12 Griffith CL, Klutts JS, Zhang L, Levery SB &
Doering TL (2004) UDP-glucose dehydrogenase plays
multiple roles in the biology of the pathogenic fungus
Cryptococcus neoformans. J Biol Chem 279, 51669–
51676.
13 Tenhaken R & Thulke O (1996) Cloning of an enzyme

JA, Nielsen JB & Foor F (1995) Differential expression
and function of two homologous subunits of yeast 1,3-
beta-D-glucan synthase. Mol Cell Biol 15, 5671–5681.
21 Shahinian S & Bussey H (2000) beta-1,6-Glucan synthesis
in Saccharomyces cerevisiae. Mol Microbiol 35, 477–489.
22 Mattila P, Rabina J, Hortling S, Helin J & Renkonen R
(2000) Functional expression of Escherichia coli enzymes
synthesizing GDP-L-fucose from inherent GDP-D-man-
nose in Saccharomyces cerevisiae. Glycobiology 10,
1041–1047.
23 Nakayama K, Maeda Y & Jigami Y (2003) Interaction
of GDP-4-keto-6-deoxymannose-3,5-epimerase-4-reduc-
tase with GDP-mannose-4,6-dehydratase stabilizes the
enzyme activity for formation of GDP-fucose from
GDP-mannose. Glycobiology 13, 673–680.
24 Sommer BJ, Barycki JJ & Simpson MA (2004)
Characterization of human UDP-glucose dehydrogen-
ase. CYS-276 is required for the second of two succes-
sive oxidations. J Biol Chem 279, 23590–23596.
25 Campbell RE, Mosimann SC, van De Rijn I,
Tanner ME & Strynadka NC (2000) The first structure
of UDP-glucose dehydrogenase reveals the catalytic resi-
dues necessary for the two-fold oxidation. Biochemistry
39, 7012–7023.
26 Bar-Peled M, Griffith CL, Ory JJ & Doering TL (2004)
Biosynthesis of UDP-GlcA, a key metabolite for capsu-
lar polysaccharide synthesis in the pathogenic fungus
Cryptococcus neoformans. Biochem J 381, 131–136.
27 Scheible WR & Pauly M (2004) Glycosyltransferases
and cell wall biosynthesis: novel players and insights.

ing and characterization of a membrane-anchored
UDP-D-glucuronate 4-epimerase from Arabidopsis.
Plant Physiol 135, 1221–1230.
36 Molhoj M, Verma R & Reiter WD (2003) The biosynth-
esis of the branched-chain sugar d-apiose in plants: func-
tional cloning and characterization of a UDP-d-
apiose ⁄ UDP-d-xylose synthase from Arabidopsis. Plant J
35, 693–703.
37 Usadel B, Kuschinsky AM, Rosso MG, Eckermann N
& Pauly M (2004) RHM2 is involved in mucilage pectin
synthesis and is required for the development of the
seed coat in Arabidopsis. Plant Physiol 134, 286–295.
38 Kainuma M, Ishida N, Yoko-o T, Yoshioka S, Takeu-
chi M, Kawakita M & Jigami Y (1999) Coexpression of
Synthesis of UDP-glucuronic acid and UDP-xylose T. Oka and Y. Jigami
2656 FEBS Journal 273 (2006) 2645–2657 ª 2006 The Authors Journal compilation ª 2006 FEBS
alpha1,2 galactosyltransferase and UDP-galactose trans-
porter efficiently galactosylates N- and O-glycans in
Saccharomyces cerevisiae. Glycobiology 9, 133–141.
39 Chiba Y, Suzuki M, Yoshida S, Yoshida A, Ikenaga H,
Takeuchi M, Jigami Y & Ichishima E (1998) Production
of human compatible high mannose-type (Man5-
GlcNAc2) sugar chains in Saccharomyces cerevisiae.
J Biol Chem 273, 26298–26304.
40 Thomas BJ & Rothstein R (1989) The genetic control
of direct-repeat recombination in Saccharomyces: the
effect of rad52 and rad1 on mitotic recombination at
GAL10, a transcriptionally regulated gene. Genetics 123,
725–738.
41 Sherman F (1991) Getting started with yeast. Methods