Mycobacterium tuberculosis possesses a functional
enzyme for the synthesis of vitamin C,
L-gulono-1,4-lactone dehydrogenase
Beata A. Wolucka
1
and David Communi
2
1 Laboratory of Mycobacterial Biochemistry, Pasteur Institute of Brussels, Institute of Public Health, Belgium
2 Institute of Interdisciplinary Research, IRIBHM, Faculty of Medicine, Free University of Brussels, Belgium
Vitamin C (l-ascorbic acid; L-AA) is an important
metabolite of plants and animals. It functions as an
antioxidant (or pro-oxidant), an enzyme cofactor, an
effector of gene expression, and a modulator of react-
ive oxygen species (ROS)-mediated cell signaling.
L-AA is therefore involved in a wide array of crucial
physiologic processes, including: biosynthesis of colla-
gen and other hydroxyproline ⁄ hydroxylysine-containing
proteins ⁄ peptides; synthesis of secondary metabolites,
hormones and cytokines [1]; oxidative protein folding
and endoplasmic reticulum stress [2]; cell proliferation
and apoptosis [3]; activation of the epithelial cystic
fibrosis transmembrane conductance regulator chloride
channel [4] and of surfactant production in human
lungs [5]; macrophage function [6]; immune homeosta-
sis [5]; and stress resistance.
Plants synthesize ascorbic acid via de novo and sal-
vage pathways [7], whereas a de novo pathway invol-
ving UDP- d-glucuronic acid operates in animals [8].
l-Gulono-1,4-lactone is a direct precursor of vitamin C
in animals [8], but also in plants [9] and in some pro-
tists [10]. In plants, L-AA can be formed additionally

affinity-purified and characterized. The FAD-binding motif-containing
Rv1771 protein is a metalloenzyme that oxidizes l-gulono-1,4-lactone
(K
m
5.5 mm) but not l-galactono-1,4-lactone. The enzyme has a dehydroge-
nase activity and can use both cytochrome c (K
m
4.7 lm) and phenazine
methosulfate as exogenous electron acceptors. Molecular oxygen does not
serve as a substrate for the Rv1771 protein. Dehydrogenase activity was
measured in cellular extracts of a Mycobacterium bovis BCG strain. In con-
clusion, M. tuberculosis produces a novel, highly specific l-gulono-1,4-lac-
tone dehydrogenase (Rv1771) and has the capacity to synthesize vitamin C.
Abbreviations
GST, glutathione-S-transferase; IPTG, isopropyl thio-b-
D-galactoside; L-AA, L-ascorbic acid; MALDI Q-TOF, MALDI quadrupole TOF;
ROS, reactive oxygen species.
FEBS Journal 273 (2006) 4435–4445 ª 2006 The Authors Journal compilation ª 2006 FEBS 4435
oxidation of l-gulono-1,4-lactone to L-AA in animals
is catalyzed by an oxygen-dependent enzyme, l-gul-
ono-1,4-lactone oxidase (EC 1.1.3.8) [13]. In plants [9]
and in Euglena [10], the oxidation involves ill-defined
l-gulono-1,4-lactone dehydrogenases that use cyto-
chrome c and phenazine methosulfate respectively, as a
direct electron acceptor. The animal and plant l-gul-
onolactone oxidoreductases are also active towards the
l-galactono-1,4-lactone substrate.
Only scarce data are available on the presence of
ascorbic acid in lower eukaryotes. Fungi do not contain
L-AA but rather its 5-carbon homolog, d-erythroascor-

er oxydans and of Acetobacter suboxydans [21], which
are also active towards d-xylose and some hexoses.
Although interesting from a biotechnological point of
view, these enzymes are not related to the known
l
-gulono-1,4-lactone oxidase proteins and their physio-
logic role is unknown.
Surprisingly, the genome of Mycobacterium tubercu-
losis, the causative agent of tuberculosis, encodes a
protein (Rv1771) that is similar to the rat l-gulono-
1,4-lactone oxidase. In the present work, we cloned
and expressed the Rv1771 gene, and showed that it
encodes a novel l-gulono-1,4-lactone dehydrogenase of
the M. tuberculosis complex.
Results
Heterologous expression and purification of the
recombinant
L-gulono-1,4-lactone dehydrogenase
(Rv1771) of M. tuberculosis
The Rv1771 DNA was cloned into the pDEST15 vec-
tor by using the Gateway system, and the obtained
pDEST15_Rv1771 plasmid was used for expression
of the recombinant glutathione-S-transferase (GST)
fusion protein in Escherichia coli. The recombinant
protein contained an engineered enterokinase cleavage
site in the junction between the GST tag and the
Rv1771 sequence. Upon 3 h of induction of the
pDEST15_Rv1771 E. coli strain with isopropyl-b-d-
thiogalactopyranoside (IPTG) at 37 °C, the yield of
the recombinant Rv1771 protein was very low (0.1 mg

EYAIPR, SLPIMFPIEVR, and FSAPDDSFLSTA
YGR), which was accompanied by a host-derived
Hsp60 chaperone protein (the identified tryptic peptide
M. tuberculosis L-gulono-1,4-lactone dehydrogenase B. A. Wolucka and D. Communi
4436 FEBS Journal 273 (2006) 4435–4445 ª 2006 The Authors Journal compilation ª 2006 FEBS
was AAVEEGVVAGGGVALIR), as determined by
combined MALDI-TOF MS of trypsin-digested pro-
tein bands (Fig. 1, lane 4; Fig. 2) and western analysis
with anti-GST IgG (Fig. 1, lane 5). Copurification of
the mycobacterial l-gulono-1,4-lactone dehydrogenase
with the Hsp60 heat-shock protein might reflect physi-
ologic protein–protein interactions, as proposed for the
plant Hsc70.3 cognate heat-shock protein and another
vitamin C-related enzyme, the GDP-mannose-3¢,5¢-
epimerase [9]. The presence of multiple GST-contain-
ing bands of about 30 kDa (Fig. 1, lane 5) suggests
that an important portion of the fusion protein was
degraded by the host proteases. On the other hand,
attempts to produce a His-tagged version of the
Rv1771 protein by using the pRSETc or the Gateway
pDEST17 expression vectors were unsuccessful.
Characterization of the recombinant
L-gulono-1,4-lactone dehydrogenase of
M. tuberculosis
The Rv1771 gene of M. tuberculosis encodes a 428
amino acid protein (Fig. 2) that shows 32% identity
with the rat l-gulono-1,4-lactone oxidase and 22–24%
identity with the putative plant l-gulono-1,4-lactone
dehydrogenases At2g46740, At2g46750, At2g46760,
At5g56490, At5g11540, and At1g32300 [9]. The Myco-

or riboflavin (1 mm) had no effect on the enzyme
Fig. 2. Sequence analysis of the Rv1771
L-gulono-1,4-lactone dehydrogenase of
M. tuberculosis. The amino acids (16–168)
that form the FAD-binding domain (pfam
designation PF01565) are highlighted in
black. The
D-arabinono-1,4-lactone oxidase
domain (pfam designation PF04030) (amino
acids 172–427) is highlighted in gray. The
position of a potential transmembrane helice
(amino acids 205–227) is indicated by bold
italics. Tryptic peptides identified by MALDI
Q-TOF MS are underlined.
Fig. 1. Heterologous expression and purification of the recombinant
L-gulono-1,4-lactone dehydrogenase (Rv1771) of M. tuberculosis.
SDS ⁄ PAGE of the affinity-purified GST-tagged dehydrogenase (con-
taining an engineered enterokinase cleavage site) obtained from
the E. coli host after long (3 h) (lanes 2 and 3) and short (1 h) (lanes
4 and 5) periods of induction with IPTG. Fractions obtained with a
long period of induction before (lane 2) and after (lane 3) enterokin-
ase treatment were concentrated on Strataclean beads, as des-
cribed in Experimental procedures. Proteins were visualized by
Coomassie blue staining (lanes 1–4) and by western analysis (lane
5) using anti-GST IgG. Protein bands (lane 4) were identified by
MALDI-TOF MS of tryptic in-gel digests. Lane 1, molecular mass
standards.
B. A. Wolucka and D. Communi M. tuberculosis
L-gulono-1,4-lactone dehydrogenase
FEBS Journal 273 (2006) 4435–4445 ª 2006 The Authors Journal compilation ª 2006 FEBS 4437

(Table 1). The steady-state parameters of the recombin-
ant l-gulono-1,4-lactone dehydrogenase of M. tuber-
culosis were determined. The dehydrogenase obeys
Michaelis–Menten kinetics with l-gulono-1,4-lactone
and cytochrome c as substrates (Fig. 3A,B). The appar-
ent K
m
values for l-gulono-1,4-lactone and cytochrome c
were determined to be 5.5 mm (Fig. 3A) and 4.7 lm
(Fig. 3B), respectively. The V
max
value was determined
to be 2.44 lmolÆh
)1
Æmg protein
)1
(Fig. 3A). The kinetic
parameters of the recombinant GST-tagged mycobacte-
rial l-gulono-1,4-lactone dehydrogenase are therefore
similar to those reported for the plant l-galactonolac-
tone dehydrogenase (K
m
values equal 3.3 mm and
3.6 lm for l-galactono-1,4-lactone and cytochrome c,
respectively) [12,25]. These results suggest that the
mycobacterial enzyme could operate efficiently in vivo.
Optimal conditions for the mycobacterial dehydroge-
nase activity were determined. The optimal pH for the
dehydrogenase reaction is between 7.5 and 8 (Fig. 4A).
At higher pH values, enzyme activity rapidly

was inactive in the presence of 1 mm EDTA.
Presence of
L-gulono-1,4-lactone dehydrogenase
activity in M. bovis BCG strain Copenhagen
Crude extracts of exponentially growing M. bovis BCG
were prepared as described in Experimental proce-
dures. The dehydrogenase activity could be measured
in the soluble extracts [0.17 mUÆmg protein
)1
], but not
in the insoluble fraction, because of interfering con-
taminants. The determined activity of the mycobacteri-
al enzyme was comparable with that reported for
crude preparations of plant l-galactono-1,4-lactone
dehydrogenase [12,27], Thus, in agreement with previ-
ous results [28,29], the Rv1771 protein is expressed in
the M. tuberculosis complex; it is probably loosely
associated with the cell membrane [29], and is enzy-
matically active. In spite of the presence of dehydroge-
nase activity, ascorbic acid could not be detected in
Table 1. Substrate specificity of the recombinant GST-tagged L-gul-
ono-1,4-lactone dehydrogenase of M. tuberculosis. ND, not deter-
mined. All measurements were made in triplicate. The limit of
detection was 0.3 mUÆmg protein
)1
. Mean values ± SD are given.
Substrate
(50 m
M)
Enzyme specific activity with different

1,4-lactone dehydrogenase (Rv1771) of M. tuberculosis
that catalyzes the reaction depicted in Fig. 5. The
Rv1771 gene was difficult to express in E. coli, and
only small quantities of the corresponding GST-tagged
protein could be obtained (Fig. 1). The enzyme has an
absolute specificity for the l-gulono-1,4-lactone sub-
strate (K
m
5.5 mm) (Fig. 3A) and shows no activity
with l-galactono-1,4-lactone (Table 1). Thus, the
mycobacterial enzyme differs from the known l-gul-
ono-1,4-lactone oxidases (EC 1.1.3.8), which oxidize
both l-gulono-lactone and l-galactono-1,4-lactone
[13,17], and also from plant [12], yeast [15] and trypan-
osomal [19] l-galactono-1,4-lactone oxidoreductases,
which are inactive towards l-gulono-1,4-lactone.
Because l-galactono-1,4-lactone is not a substrate for
the mycobacterial dehydrogenase, we presume that
d-arabinono-1,4-lactone, a five-carbon homolog of
l-galactono-1,4-lactone, is not a substrate either. Thus,
the mycobacterial dehydrogenase is unusual in its
selectivity for l-gulono-1,4-lactone. Our preparations
of the recombinant dehydrogenase of M. tuberculosis
y = 0,1371x + 28,762
0
5
10
15
20
25

ML-gulono-1,4-lactone. Double-reciprocal Lineweaver–Burke plots are shown. V
0
is lmol of L-gulono-1,4-lactone oxidized per min and per mg of the recombinant dehydrogenase (UÆmg protein
)1
). L-Gulono-1,4-lactone con-
centrations ranged from 5 to 25 m
M, whereas cytochrome c concentrations ranged from 24 to 145 lM (B). All measurements were made in
duplicate in three independent experiments; the values obtained in a representative experiment are shown.
0
0.02
0.04
0.06
0.08
0.1
5.5 6 6.5 7 7.5 8 8.5 9
pH
Enzyme activityEnzyme activity
(% of control)
A
B
0
100
200
300
400
500
20 30 40 50
Temperature (°C)
Fig. 4. Effects of pH (A) and temperature (B) on the activity of the
recombinant GST-tagged

B. A. Wolucka and D. Communi M. tuberculosis
L-gulono-1,4-lactone dehydrogenase
FEBS Journal 273 (2006) 4435–4445 ª 2006 The Authors Journal compilation ª 2006 FEBS 4439
had low specific activity, ranging from 40 to
66 mUÆmg protein
)1
under the nonoptimal tempera-
ture conditions of the enzyme assay (24 °C). However,
taking into account that the enzyme activity is about
three-fold higher at 39 °C (Fig. 4B) and that the GST-
tagged Rv1771 protein represented only a portion of
the GST affinity-purified fraction (Fig. 1, lane 4), the
specific activity of the recombinant dehydrogenase
could be at least 10-fold higher [400–660 mUÆmg pro-
tein
)1
]. The relatively low activity of the recombinant
M. tuberculosis enzyme could be due to impaired pro-
tein folding, proteolytic degradation and ⁄ or the lack of
a mycobacterial cofactor in the E. coli expression sys-
tem. Another possibility is that the mycobacterial
dehydrogenase might require a specific post-transla-
tional modification that occurs inefficiently in the
E. coli host. As far as we know, the specific activities
of related recombinant enzymes of plant origin have
not been reported. Moreover, huge differences in the
specific activities of purified native aldonolactone
oxidoreductases, ranging from 760 mUÆmg protein
)1
[17] up to 51 000 UÆmg protein

rat l-gulono-1,4-lactone oxidase was reported in the
early literature [26,31], but the activity was not studied
further. Nowadays, the l-gulono-1,4-lactone oxidase
enzymes are considered exclusively as oxidases, the
reaction products of which are, paradoxically, L-AA
and hydrogen peroxide [8]. Dehydrogenase-to-oxidase
conversion is well known for another antioxidant (uric
acid)-producing enzyme, xanthine oxidoreductase [32].
Perhaps a similar molecular mechanism might be
responsible for the dehydrogenase-to-oxidase switch of
mammalian l-gulono-1,4-lactone oxidase proteins and
play a role in the metabolism of L-AA.
We showed that in vitro both cytochrome c
(K
m
4.7 lm) (Fig. 3B) and phenazine methosulfate
(Table 1) can serve as electron acceptors for the
l-gulono-1,4-lactone dehydrogenase of M. tuberculosis.
Remarkably, the phenazine methosulfate acceptor was
even more efficient than cytochrome c at saturation
(Table 1). Phenazines, ‘secondary metabolites’ of cer-
tain soil and pathogenic bacteria, are redox-active,
flavin-like low-molecular-weight compounds that can
produce ROS and play a role in quorum sensing and
biofilm formation in Pseudomonas aeruginosa lung
infection [33]. It is possible, therefore, that an unknown
phenazine-like, low-molecular-weight compound might
serve as an endogenous electron acceptor for the
Rv1771 dehydrogenase of M. tuberculosis.
Despite the presence of l-gulono-1,4-lactone dehy-

specific conditions. Examples of differential regulation
of gene expression and metabolic reprogramming in
M. tuberculosis are known [36,37]. Some earlier steps in
a pathway for ascorbic acid might be inducible, e.g.
during the intracellular growth of the pathogen in its
host. This could explain the absence of the l-gulono-
1,4-lactone dehydrogenase reaction product in the
M. tuberculosis cells grown in vitro.
The Rv1771 l-gulono-1,4-lactone dehydrogenase of
M. tuberculosis is a specific enzyme for the biosynthesis
of L-AA. As far as we know, this is the first report of a
specific biosynthetic enzyme for vitamin C in bacteria.
To detect related aldonolactone dehydrogenases ⁄
oxidases in other bacterial genomes, we used the rat
l-gulono-1,4-lactone oxidase as a query sequence in
blast searches of the protein database. These searches
retrieved, for a limited number of bacterial species,
additional putative aldonolactone oxidase orthologs
that display significant sequence identity (about 30%)
with the rat l-gulono-1,4-lactone oxidase protein, and
are closely related to the known and predicted
l-gulono-1,4-lactone oxidase-like proteins of animals,
plants and fungi (Fig. 6). Surprisingly, an important
number of bacterial species that contain a vitamin C
biosynthetic gene belong to the Actinomycetales
[M. tuberculosis, M. bovis, Thermobifida fusca, Streptomy-
ces coelicolor and Streptomyces avermitilis (NP_ 823585)].
It is worth noting that members of the Actinomycetales
(Streptomyces verticillus and Saccharothrix mutabilis)
possess orthologs of another vitamin C-related enzyme,

[40]. The gene is well conserved within the M. tuber-
culosis complex, except for the M. bovis BCG Pasteur
(1173P2) strain, which lost the gene due to the dele-
tion of chromosomal region RD14 [41]. Ironically,
whereas the pathogen’s ortholog is well conserved,
the l-gulono-1,4-lactone oxidase gene of tuberculosis-
prone species, such as humans and guinea pigs, is
nonfunctional because of mutations accumulated dur-
ing evolution [8]. These facts strongly suggest that
the l-gulonolactone dehydrogenase of M. tuberculosis
could play a role in virulence, pathogenesis and ⁄ or
survival of the parasite within its host. In agreement
Fig. 6. Sequence relationship between L-gulono-1,4-lactone dehy-
drogenase of M. tuberculosis and previously identified or predicted
L-gulono-1,4-lactone oxidase-like proteins. The unrooted neighbor-
joining (N-J) tree was generated () on the
basis of the amino acid sequences of proteins that show at least
30% identity with the rat
L-gulono-1,4-lactone oxidase. The acces-
sion numbers of the sequences used were: M. tuberculosis
L-gulono-1,4-lactone dehydrogenase, NP_216287; Streptomyces
coelicolor, NP_629980; Thermobifida fusca, ZP_00059445; Oceano-
bacillus iheyensis, NP_692632; Bacillus cereus, NP_830486;
Burkholderia cepacia, ZP_00218082; Saccharomyces cerevisiae
D-arabinono-1,4-lactone oxidase (ALO), P54783; Candida albicans
D-arabinono-1,4-lactone oxidase (ALO), O93852; Neurospora crassa,
Q7SGY1; Gibberella zeae, XP_388870; Arabidopsis thaliana
L-galac-
tono-1,4-lactone dehydrogenase (GLDH), At3g47930; Arabidopsis
thaliana putative

nase of M. tuberculosis is a new and distinct member
of the family of l-gulono-1,4-lactone dehydrogenase ⁄
oxidases that have been characterized up to now, and
the first example of a specific, vitamin C biosynthetic
enzyme of bacterial origin. Further studies will be
necessary to elucidate the role of the Rv1771
l-gulono-1,4-lactone dehydrogenase in M. tuberculosis
infection.
Experimental procedures
Chemicals
GST affinity and StrataClean resins were obtained from
Stratagene (Madison, WI). l-Gulono-1,4-lactone, cyto-
chrome c from horse heart (oxidized form), phenazine
methosulfate and 2,6-dichloroindophenol were purchased
from Sigma-Aldrich (St Louis, MO). All reagents were of
analytical grade.
Plasmid construction
The ORF corresponding to the mycobacterial Rv1771
l-gulonolactone dehydrogenase (1287 bp) was PCR ampli-
fied from the genomic DNA of M. tuberculosis H37Rv.
Oligonucleotide primers were designed with attB1 or attB2
sites for insertion into the Gateway donor vector
pDONR201 (Invitrogen, Gaithersburg, MD) by homolog-
ous recombination. A sequence (GATGACGACGACAAG)
corresponding to the enterokinase cleavage site (DDDDK)
was included within the forward primer immediately
upstream of the start codon (ATG). Primers with the fol-
lowing sequences were synthesized by Proligo (Paris,
France): 5forGulox (forward), 5¢-GGGGACAAGTTT
GTACAAAAAAGCAGGCTTCGATGACGACGACAAG

Triton X-100, 1 m m phenylmethanesulfonyl fluoride and
20% glycerol (buffer A). The column was washed with 15
volumes of buffer A, and the recombinant dehydrogenase
was eluted with three volumes of 10 mm glutathione
(reduced form) in buffer A. Fractions containing the recom-
binant dehydrogenase were pooled. For measurements of
the l-gulono-1,4-lactone dehydrogenase activity, glutathi-
one and dithiothreitol were removed by gel filtration of
the pooled GST affinity fractions on a prepacked NAP-25
column (Amersham Pharmacia Biotech, Uppsala, Sweden).
Enterokinase cleavage of the GST-tagged
L-gulono-1,4-lactone dehydrogenase
In order to remove the GST tag, an aliquot of the affinity-
purified recombinant dehydrogenase was incubated for 24 h
M. tuberculosis L-gulono-1,4-lactone dehydrogenase B. A. Wolucka and D. Communi
4442 FEBS Journal 273 (2006) 4435–4445 ª 2006 The Authors Journal compilation ª 2006 FEBS
at 22 °C with 2 units of enterokinase (Stratagene) in
500 lL (final volume) of 20 mm Tris ⁄ HCl buffer (pH 8.0)
containing 50 mm NaCl and 2 mm CaCl
2
. One unit of
enterokinase is the amount of enzyme required to cleave
100 lg of the CBP-EK-JNK fusion protein (Stratagene) to
90% completion at 21 °C in 16 hours.
Concentration of protein fractions on
Strataclean beads
To pooled GST affinity fractions, 10 lL of Strataclean
resin suspension was added. After overnight incubation at
4 °C, samples were centrifuged at 13 000 g for 5 min at
room temperature on a Microfuge 18 centrifuge (Beckman-

prior to measurements. l-Gulono-1,4-lactone dehydro-
genase activity was measured spectrophotometrically at
550 nm by following the l-gulono-1,4-lactone-dependent
reduction of cytochrome c [46]. When indicated, 2.5 mm
phenazine methosulfate was used as a direct electron accep-
tor in the presence of 100 lm 2,6-dichloroindophenol, and
the decrease in absorbance at 610 nm due to the reduction
of 2,6-dichloroindophenol was measured, as described [47].
PAGE
Proteins were separated by SDS ⁄ PAGE, using 10% mini-
gels and the buffer system described by Laemmli [48]. Gels
were stained with Coomassie Brilliant Blue R-250.
Immunoblotting
Samples fractionated by SDS ⁄ PAGE were transferred to a
nitrocellulose membrane by electroblotting (Bio-Rad, Her-
cules, CA) according to the manufacturer’s protocol. Mem-
branes were incubated with goat anti-GST IgG (1 : 1000
dilution; Amersham Pharmacia Biotech), and antibody
binding was detected using anti-goat IgG conjugated to
alkaline phosphatase (1 : 5000 dilution; Sigma-Aldrich) and
the 5-bromo-4-chloro-3-indolyl-phosphate ⁄ nitroblue tetra-
zolium reagent (Promega, Madison, WI).
MS
MALDI quadrupole TOF (MALDI Q-TOF) MS analysis
of in-gel-digested protein bands was performed on a
Q-TOF Ultima Global mass spectrometer equipped with a
MALDI source (Micromass, Waters Corporation, Milford,
MA), as described [49].
Protein determination
Protein concentration was determined by the method of

(2005) Hypoxia–reoxygenation-induced mitochondrial
damage and apoptosis in human endothelial cells are
inhibited by vitamin C. Free Radic Biol Med 38, 1311–
1322.
4 Fischer H, Schwarzer C & Illek B (2004) Vitamin C
controls the cystic fibrosis transmembrane conductance
regulator chloride channel. Proc Natl Acad Sci USA
101, 3691–3696.
5 Brown LAS & Jones DP (1997) Antioxidant action of
vitamin C in the lung. In Vitamin C in Health and Dis-
ease (Packer L & Fuchs J, eds), pp. 265–278. Marcel
Dekker, New York.
6 May JM, Huang J & Qu ZC (2005) Macrophage uptake
and recycling of ascorbic acid: response to activation by
lipopolysaccharide. Free Radic Biol Med 39, 1449–1459.
7 Valpuesta V & Botella MA (2004) Biosynthesis of
l-ascorbic acid in plants: new pathways for an old
antioxidant. Trends Plant Sci 9 , 573–577.
8 Nishikimi M & Yagi K (1996) Biochemistry and mole-
cular biology of ascorbic acid biosynthesis. Subcell
Biochem 25, 17–39.
9 Wolucka BA & Van Montagu M (2003) GDP-mannose
3¢,5¢-epimerase forms GDP-l-gulose, a putative interme-
diate for the de novo biosynthesis of vitamin C in plants.
J Biol Chem 278, 47483–47490.
10 Shigeoka S, Nakano Y & Kitaoka S (1979) Some prop-
erties and subcellular localization of l-gulono-c-lactone
dehydrogenase in Euglena gracilis z. Agric Biol Chem
43, 2187–2188.
11 Mapson LW & Breslow E (1958) Biological synthesis of

YW, Hwang CS, Lee JS & Kang SO (1998) D-Ery-
throascorbic acid is an important antioxidant molecule
in Saccharomyces cerevisiae. Mol Microbiol 30, 895–
903.
19 Wilkinson SR, Prathalingam SR, Taylor MC, Horn D
& Kelly JM (2005) Vitamin C biosynthesis in trypano-
somes: a role for the glycosome. Proc Natl Acad Sci
USA 102, 11645–11650.
20 Koshizaka T, Nishikimi M, Ozawa T & Yagi K (1988)
Isolation and sequence analysis of a complementary
DNA encoding rat liver l-gulono-c-lactone oxidase, a
key enzyme for l-ascorbic acid biosynthesis. J Biol
Chem 263, 1619–1621.
21 Sugisawa T, Ojima S, Matzinger PK & Hoshino T
(1995) Microbial production of l-ascorbic acid from
d-sorbitol, l-sorbose, l-gulose, and l-sorbosone by
Ketogulonicigenium vulgare DSM 4025. Biosci Biotech-
nol Biochem 59, 190–196.
22 Nakagawa H, Asano A & Sato R (1975) Ascorbate-
synthesizing system in rat liver microsomes. II. A peptide-
bound flavin as the prosthetic group of l-gulono-c-lac-
tone oxidase. J Biochem (Tokyo) 77, 221–232.
23 Nishikimi M, Kiuchi K & Yagi K (1977) Detection of
l-gulono-c-lactone oxidase on SDS-polyacrylamide gels
by the fluorescence of its covalently bound flavin. FEBS
Lett 81, 323–325.
24 Kito M, Ohishi N & Yagi K (1991) Micro-determina-
tion of l-gulono-
c-lactone oxidase activity. Biochem Int
24, 131–135.

J Biochem (Tokyo) 117, 120–124.
31 Kar NC, Ghosh NC & Guha BC (1963) Microsomal
l-gulonolactone dehydrogenase. Nature 197, 494–495.
32 Kuwabara Y, Nishino T, Okamoto K, Matsumura T,
Eger BT & Pai EF (2003) Unique amino acids cluster
for switching from the dehydrogenase to oxidase form
of xanthine oxidoreductase. Proc Natl Acad Sci USA
100, 8170–8175.
33 Price-Whelan A, Dietrich LE & Newman DK (2006)
Rethinking ‘secondary’ metabolism: physiological roles
for phenazine antibiotics. Nat Chem Biol 2, 71–78.
34 Linster CL & Van Schaftingen E (2006) Glucuronate,
the precursor of vitamin C, is directly formed from
UDP-glucuronate in liver. FEBS J 273, 1516–1527.
35 Wolucka BA, McNeil MR, Kalbe L, Cocito C & Bren-
nan PJ (1993) Isolation and characterization of a novel
glucuronosyl diacylglycerol from Mycobacterium smeg-
matis. Biochim Biophys Acta 1170, 131–136.
36 Fenhalls G, Stevens L, Moses L, Bezuidenhout J, Betts
JC, Helden PP, Lukey PT & Duncan K (2002) In situ
detection of Mycobacterium tuberculosis transcripts in
human lung granulomas reveals differential gene expres-
sion in necrotic lesions. Infect Immun 70, 6330–6338.
37 McKinney JD, Honer zu Bentrup K, Munoz-Elias EJ,
Miczak A, Chen B, Chan WT, Swenson D, Sacchettini
JC, Jacobs WR Jr & Russell DG (2000) Persistence of
Mycobacterium tuberculosis in macrophages and mice
requires the glyoxylate shunt enzyme isocitrate lyase.
Nature 406, 735–738.
38 Wolucka BA, Persiau G, Van Doorsselaere J, Davey

45 Allaudeen HS & Ramakrishnan T (1970) Biosynthesis
of isoleucine and valine in Mycobacterium tuberculosis
H37Rv. II. Purification and properties of acetohydroxy
acid isomerase. Arch Biochem Biophys 140, 245–256.
46 O
ˆ
ba K, Fukui M, Imai Y, Iriyama S & Nogami K
(1994) l-Galactono-c-lactone dehydrogenase: partial
characterization, induction of activity and role in the
synthesis of ascorbic acid in wounded potato tuber tis-
sue. Plant Cell Physiol 35, 473–478.
47 Nakagawa H & Asano A (1970) Ascorbate synthesizing
system in rat liver microsomes. I. Gulonolactone-reduci-
ble pigment as a prosthetic group of gulonolactone oxi-
dase. J Biochem (Tokyo) 68 , 737–746.
48 Laemmli UK (1970) Cleavage of structural proteins
during the assembly of the head of bacteriophage T4.
Nature 227, 680–685.
49 Wittamer V, Franssen JD, Vulcano M, Mirjolet JF,
Le Poul E, Migeotte I, Brezillon S, Tyldesley R,
Blanpain C, Detheux M et al. (2003) Specific recruit-
ment of antigen-presenting cells by chemerin, a novel
processed ligand from human inflammatory fluids.
J Exp Med 198, 977–985.
50 Bradford MM (1976) A rapid and sensitive method for
the quantitation of microgram quantities of protein util-
izing the principle of protein-dye binding. Anal Biochem
72, 248–254.
51 Wolucka BA, Davey MW & Boerjan W (2001) A high-
performance liquid chromatography radio method for