FEMS Microbiology Letters 186 (2000) 245^250

www.fems-microbiology.org

Biosynthesis of L-ascorbic acid (vitamin C) by

Saccharomyces cerevisiae
Robert D. Hancock a , John R. Galpin b , Roberto Viola a;
*

Downloaded from https://academic.oup.com/femsle/article-abstract/186/2/245/689187 by guest on 04 May 2020

a
Scottish Crop Research Institute, Division of Biochemistry and Cell Biology, Unit of Plant Biochemistry, Invergowrie, Dundee DD2 5DA, UK
b
Nutritional Healthcare RpD, SmithKline Beecham Consumer Healthcare, 11 Stoke Poges Lane, Slough, Berkshire SL1 3NW, UK

Received 20 February 2000; received in revised form 24 March 2000 ; accepted 27 March 2000

Abstract

Saccharomyces cerevisiae cells incubated with D-glucose (D-Glc), D-galactose or D-mannose (D-Man) synthesised D-erythroascorbic acid
(D-EAA) but not L-ascorbic acid (L-AA). Accumulation of D-EAA was observed in cells incubated with D-arabinose (D-Ara) whilst
accumulation of L-AA occurred in cells incubated with L-galactose (L-Gal), L-galactono-1,4-lactone and L-gulono-1,4-lactone. When
S. cerevisiae cells were incubated with D-[U-14 C]Glc, D-[U-14 C]Man or L-[1-14 C]Gal, incorporation of radioactivity into L-AA was observed
only with L-[1-14 C]Gal. Pre-incubation of yeast cells with D-Ara substantially reduced the incorporation of L-[1-14 C]Gal into L-AA. Our
results indicate that, under appropriate conditions, yeast cells can synthesise L-AA via the pathway naturally used for D-EAA
biosynthesis. ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.

Keywords : Vitamin C; L-Ascorbic acid biosynthesis ; L-Galactose ; D-Erythroascorbic acid; Saccharomyces cerevisiae

1. Introduction catalyse the conversion of L-GL into L-AA in vitro [9,10],

There has been debate for some time concerning the
capacity of yeasts to synthesise the essential human nu-
trient L-ascorbic acid (L-AA) [1,2]. Early reports suggested
the presence of L-AA in certain yeasts including Saccha-
romyces cerevisiae [3,4]. However, these ¢ndings have been
questioned by more recent investigations using improved
analytical methods [5,6]. Some of the confusion seems to
be due to the presence in yeasts and other fungi of D-
erythroascorbic acid (D-EAA), a ¢ve carbon L-AA ana-
logue with similar redox properties [7] and which is also
oxidised by ascorbate oxidase (E.C. 1.10.3.3) ([2] and
references therein). D-EAA is thought to perform similar
antioxidant functions in yeast to those performed by L-AA
in other eukaryotes [8]. Further confusion arises from the
isolation of two L-galactono-1,4-lactone (L-GL) oxidases
(E.C. 1.1.3.24) in S. cerevisiae mitochondria both of which
* Corresponding author. Tel. : +44 (1382) 562-731;
Fax: +44 (1382) 562-426; E-mail : rviola@scri.sari.ac.uk
Abbreviations : D-Ara,
D-arabinose; L-AA, L-ascorbic acid; D-EAA,
D-erythroascorbicacid ; DTT, dithiothreitol ; DW, double distilled H2 O;
L-Gal, L-galactose; D-Glc, D-glucose ; L-GL, L-galactono-1,4-lactone;
L-GuL, L-gulono-1,4-lactone ; D-Man, D-mannose
0378-1097 / 00 / $20.00 ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.
PII: S 0 3 7 8 - 1 0 9 7 ( 0 0 ) 0 0 1 5 5 - 5
and one of which also converts L-gulono-1,4-lactone (L-
GuL) into L-AA [9]. L-AA biosynthesis in plant and ani-
mal cells is performed by mitochondrial enzymes catalys-
ing similar reactions. Plant L-GL dehydrogenase (E.C.
1.3.2.3) is very speci¢c for L-GL and catalyses its oxidation
into L-AA in the presence of cytC [11,12]. Animals syn-
thesise L-AA via L-GuL oxidase (E.C. 1.1.3.8) which can
use both L-GuL and L-GL, although L-GuL is thought to
be the sole physiological substrate [13]. Signi¢cantly, a
number of yeast strains accumulate L-AA if incubated
with L-GL [14], by analogy with what is generally observed
with plant or animal cells [15^18].
The capacity of S. cerevisiae to synthesise L-AA has
been re-examined here in the light of the recent resolution
of the complete pathway of L-AA biosynthesis in plants
[19]. In this work, we have used cells of the green micro-
alga Chlorella pyrenoidosa as a model system for the plant
pathway [20]. L-AA synthesis occurred in cells supplied
with L-galactose (L-Gal), L-GL or L-GuL, by analogy
with what has been observed with plant cells [19,15] and
animal cells (where only the 1,4-lactones have been tested)
[15]. However, yeast cells did not synthesise L-AA from D-
aldoses (D-glucose (D-Glc), D-galactose (D-Gal), D-mannose
(D-Man) or D-arabinose (D-Ara)). Experiments with radio-
tracers con¢rmed the absence of a pathway of L-AA bio-

FEMSLE 9388 27-4-00

246 R.D. Hancock et al. / FEMS Microbiology Letters 186 (2000) 245^250
synthesis from D-Glc or D-Man which was detected in
C. pyrenoidosa. D-EAA was detected in S. cerevisiae under
all experimental conditions and it markedly accumulated
upon incubation with D-Ara. We conclude that L-AA is
not a natural product of S. cerevisiae metabolism but its
synthesis can be induced by the supply of non-physiolog-
ical substrates which are converted into L-AA, presum-
ably, via the enzymes of the D-EAA biosynthetic pathway.
2. Materials and methods
2.1. Materials
14
D-[U- C]Glc (speci¢c activity (S.A.) 11.47 GBq
mmol ) and D-[U-14 C]Man (S.A.
31
10.58 GBq mmol31 )
were obtained from Amersham, Buckinghamshire, UK.
14 31
L-[1- C]Gal (S.A. 2.04 GBq mmol ) was obtained from
American Radiolabeled Chemicals, St. Louis, MO, USA.
2.2. Micro-organisms and growth conditions
S. cerevisiae ATTC 287 was obtained from the Ameri-
can Type Culture Collection and stocks maintained on
malt extract peptone agar. Liquid cultures were grown in
YN base (Difco) supplemented with 1.5% (w/v) sucrose at
25³C with rotary shaking (150 rpm). C. pyrenoidosa
UTEX 1663 was obtained from the University of Texas
culture collection and stocks maintained on solid volvox^
dextrose medium [21] containing 1.5% agar. Cultures were
grown in the same medium, without agar, at 35³C in the
dark with rotary shaking (150 rpm).
2.3. Incubation with precursors
S. cerevisiae cultures were grown to mid-exponential
phase and harvested by centrifugation (4³C, 10 min,
3000Ug). Prior to incubation with precursors, the cells
were washed twice in YN base to remove any sucrose.
For experiments with unlabelled precursors, the cells
were resuspended in YN base to the original cell density,
the appropriate precursor was added to 0.5% (w/v) and
cultures incubated for 24 h. For experiments with labelled
precursors, cells were resuspended in YN base at a density
of 3.2U109 cells ml31 and incubated for 30 min at 25³C
prior to addition of labelled substrates. This treatment was
done in order to increase uptake of radioactivity. Finally,
aliquots (1 ml) of the cell suspension were transferred to
glass vials containing the appropriate labelled precursor
(111 kBq) and incubated for 4 h at 25³C. The vials were
sealed with rubber caps ¢tted with paper ¢lters impreg-
nated with 10% (w/v) KOH to absorb 14 CO2 .
For radiolabelling experiments with C. pyrenoidosa, the
conditions were as described for S. cerevisiae except that
cells were washed in dextrose-free volvox medium, resus-
pended at a density of 5U109 cells ml31 and 2-ml aliquots
FEMSLE 9388 27-4-00
were incubated with the appropriate precursor (111 kBq)
at 35³C.
2.4. Extraction and quanti¢cation of L-AA and D-EAA
S. cerevisiae cells were harvested, washed twice in YN
base and resuspended in 7% metaphosphoric acid contain-
ing 5 mM dithiothreitol (DTT) (¢nal volume 1/60 of the
original culture volume). Cells were subjected to three
cycles of freeze-thawing (liquid N2 /30³C) and subsequently
held on ice for 30 min. Cell debris was removed by cen-
trifugation (4³C, 16 000Ug, 5 min) and the supernatant
passed through a 0.22-Wm ¢lter prior to high performance
Downloaded from https://academic.oup.com/femsle/article-abstract/186/2/245/689187 by guest on 04 May 2020
liquid chromatography (HPLC) analyses. Under these
conditions, the recovery of authentic L-AA added to the
cells during extraction was 80 þ 5%.
For analysis by HPLC, aliquots (20 Wl) of the super-
natant were injected onto a Coregel 64H ion exclusion
column (Interaction Chromatography, San Jose, CA,
USA; column maintained at 50³C) using a Gynkotech
Gina 50 autoinjector equipped with a cooled sample
rack (4³C). The mobile phase was 8 mM H2 SO4 pumped
at a £ow rate of 0.6 ml min31 and compounds of interest
in the eluant were detected, initially by their absorbance at
245 nm, using a diode array detector (Gynkotech UVD
340S). The identi¢cation of the peaks corresponding to
L-AA or D-EAA in chromatograms were deduced by their
co-elution with standards. Further con¢rmation was
through analysis of peak absorption spectra and peak re-
moval after treatment of the sample with ascorbate oxi-
dase (E.C. 1.10.3.3) which is capable of utilising both com-
pounds as substrate [7]. Under these conditions, the
retention times of L-AA and D-EAA were 12.6 min and
14.4 min, respectively.
14
2.5. Incorporation and quanti¢cation of C-labelled
compounds
At the end of incubation, cells were harvested by cen-
trifugation (4³C, 5000Ug, 5 min), washed twice in YN
base (S. cerevisiae) or dextrose-free volvox medium (C.
pyrenoidosa) and resuspended in 700 Wl 5% HClO4 con-
taining 10 mM L-AA. The inclusion of L-AA in the ex-
traction medium was essential to minimise losses of L-
[14 C]AA during extraction. Under these extraction condi-
tions, recoveries of authentic L-[1-14 C]AA added to the
tissue during extraction were in excess of 85%. Cells
were lysed by freeze-thawing (liquid N2 /30³C) followed
by sonication in an ice/ethanol bath (4U15 s, 18 W ampli-
tude). Cell debris was removed by centrifugation (4³C,
16 000Ug, 15 min) and the supernatant neutralised by
addition of 5 M K2 CO3 . The solution was cooled on ice
and KClO4 removed by centrifugation (4³C, 16 000Ug, 15
min). An aliquot (500 Wl) of the neutralised supernatant
was applied to a SAX cartridge (100 mg; HPLC technol-
ogy, Maccles¢eld, UK) to bind anionic compounds. The
 R.D. Hancock et al. / FEMS Microbiology Letters 186 (2000) 245^250 247

resin was washed with 4 ml double distilled H2 O (DW) to Table 1

The e¡ect of sugars and aldonolactones on L-AA and D-EAA content
remove basic and neutral compounds (SAX unbound frac-
of S. cerevisiae
tion). L-AA was eluted from the cartridge using 60 mM
formic acid (formic acid fraction) essentially as described Precursor L-AA concentration D-EAA concentration
(Wg g DW31 )a (Wg g DW31 )a
by Conklin et al. [22]. Both fractions were lyophilised and
resuspended in 100 Wl distilled water. L-AA in the formic D-Glc ndb 1.67 þ 0.26
D-Man ndb 3.21 þ 0.38
acid fraction was further puri¢ed by HPLC (see above) D-Gal ndb 0.20 þ 0.06
and radioactivity in L-AA was quanti¢ed by using a radio- L-Gal 658.0 þ 94.5 0.18 þ 0.08
detector (Reeve 9201) connected in series with the diode D-Ara ndb 200.8 þ 21.2
array detector. For determination of non-metabolised sub- L-GL ndb 0.60 þ 0.14
L-GL 865.5 þ 113.6 0.19 þ 0.04
strate, the lyophilised SAX unbound fraction was ¢ltered
L-GuL 138.2 þ 10.3 0.22 þ 0.05
(0.22 mm) and injected onto a Styrex Car-Ca column
a
(300U7.8 mm; HPLC technology, Maccles¢eld, UK) Values are presented as mean þ S.D. (n = 5).

Downloaded from https://academic.oup.com/femsle/article-abstract/186/2/245/689187 by guest on 04 May 2020

maintained at 70³C. Sugars were separated using H2 O as
a mobile phase at a £ow rate of 1 ml min31 . Radioactivity
in the eluate was monitored as for L-AA. The retention
times of D-[U-14 C]Glc, D-[U-14 C]Man or L-[1-14 C]Gal were
estimated by the use of radiolabelled standards and were
6.76 min, 7.65 min and 8.45 min, respectively.
For calculation of total incorporation of radioactivity,
the 14 C incorporated into the crude PCA extract, the in-
soluble fraction (i.e. cell debris) and CO2 was quanti¢ed
by liquid scintillation counting (Packard 2000 CA). Me-
tabolised radioactivity was de¢ned as total radioactivity
incorporated minus that recovered as the original sub-
strate.
b
nd = not detected, detection limit v 0.038 Wg g DW31 .
161 mM for D-Ara and 180 mM for L-Gal [23,24]. The
increase in absorbance (340 nm) in the assay medium was
monitored spectrophotometrically using a Hitachi U-3010.
The reaction was conducted at 30³C. Boiled extracts were
used as reference.
2.7. Protein determination
The protein concentration was determined according to
the method of Bradford [25] using bovine serum albumin
as a reference protein.

2.6. Enzyme assays

Cultured S. cerevisiae cells were harvested by centrifu-
gation at 4000Ug for 10 min and washed twice with 50
mM Tris^HCl (pH 7.5) containing 1 mM EDTA. The
harvested cells were resuspended in 20 ml of 50 mM
Tris^HCl (pH 7.5) containing 1 mM EDTA, 5 mM
DTT and 0.5 mM phenylmethylsulfonyl £uoride and
were homogenised by sonication on an ice/ethanol bath
(6U15 s, 24 W). Cell debris was removed by centrifugation
at 33 000Ug for 30 min. The supernatant was desalted on
Sephadex G25 (PD10, Pharmacia, Sweden) pre-equilibrat-
ed with the extraction bu¡er. Reaction mixtures consisted
of 0.1 M Tris^HCl pH 8.2, 0.5 mM NAD(P)‡ , 200 mM
L-Gal or D-Ara and aliquots of the desalted extract. High
sugar concentrations were used due to previous investiga-
tions which reported the Km of D-Ara dehydrogenase to be
3. Results
Table 1 shows the e¡ect of various substrates on L-AA
and D-EAA biosynthesis in cells of S. cerevisiae. D-EAA
was detected under all conditions and accumulated in the
presence of D-Ara. No L-AA was found in cells incubated
with D-aldoses whilst appreciable amounts were detected
in cells incubated with L-Gal, L-GL and, to a lesser extent,
L-GuL.
Table 2 shows the incorporation of 14 C-radiotracers into
L-AA by cells of S. cerevisiae and of the green microalga
C. pyrenoidosa. All substrates used were converted into
L-AA by Chlorella cells and the proportion of label incor-
porated was consistent with the position of the precursor
in the L-AA biosynthetic pathway [19]. With S. cerevisiae
cells, only L-[1-14 C]Gal was incorporated into L-AA. The

Table 2
14 14
Incorporation of C into L-AA from C precursors in S. cerevisiae and C. pyrenoidosa
Substrate Pretreatmenta Metabolised label recovered in Metabolised label recovered in
L-AA S. cerevisiae (%) L-AA C. pyrenoidosa (%)
14
D-[U- C]Glc none ndb 0.07 þ 0.01c
14
D-[U- C]Man none ndb 2.53 þ 0.79c
14
L-[1- C]Gal none 24.37 þ 3.19c 58.70 þ 10.10c
14
L-[1- C]Gal 24 h with 0.5% (w/v) D-Ara 3.54 þ 0.27c
a
No pretreatment, cultures grown with D-Glc as the carbon substrate.
b
nd = not detected.
c
Values presented as mean þ S.D. (n = 3).

FEMSLE 9388 27-4-00

248 R.D. Hancock et al. / FEMS Microbiology Letters 186 (2000) 245^250

Table 3
D-Ara and L-Gal dehydrogenase activity of cell-free S. cerevisiae extracts

Substrate Cofactor Speci¢c activity

(nmol min31 mg protein31 )
D-Ara NAD‡ 0.92 þ 0.14a
NADP‡ 7.16 þ 0.27a
L-Gal NAD‡ 1.95 þ 0.25a
NADP‡ 6.36 þ 0.34a
a
Data presented as mean þ S.D. (n = 3).

incorporation of L-[1-14 C]Gal into L-AA was substantially

reduced (85%) in yeast cells treated with D-Ara for 24 h.
Table 3 shows the reduction of NAD(P)‡ by cell-free

Downloaded from https://academic.oup.com/femsle/article-abstract/186/2/245/689187 by guest on 04 May 2020

extracts of S. cerevisiae cells in the presence of 200 mM
L-Gal or D-Ara.

4. Discussion

The results of the present investigation strongly indicate

that S. cerevisiae cells do not possess the ability to synthe-
sise L-AA from D-aldoses. Firstly, when the cells were in-
cubated with D-Glc, D-Man, D-Gal or D-Ara, they were
found to contain D-EAA but not L-AA, in agreement
with previous reports [1,5,8]. Secondly, unlike cells of the
green microalga C. pyrenoidosa, S. cerevisiae cells did not
convert D-[U-14 C]Glc or D-[U-14 C]Man into L-AA. The
capacity to convert D-[U-14 C]Glc into L-AA is a feature
of both plant and animal cells ([26] and references
therein). In addition, plant cells can convert exogenous
14
D-[U- C]Man into L-AA, as demonstrated in seminal ex-
periments that unveiled the pathway of L-AA biosynthesis
in plants [19]. It was previously demonstrated that C. pyre-
noidosa converts D-[14 C]Glc into L-[14 C]AA without inver-
sion of the carbon skeleton in a manner similar to that
observed with higher plants, leading to the suggestion that
C. pyrenoidosa was a suitable model system for the study
of L-AA biosynthesis in higher plants [20]. Our work dem-
onstrates further similarities, namely that this organism,
like higher plants [19], incorporates D-[U-14 C]Man into
14 14
L-[ C]AA to a greater extent than D-[U- C]Glc is incor-
porated into L-[14 C]AA.
In the present work, we observed substantial incorpora-
tion of L-[1-14 C]Gal into L-AA in both S. cerevisiae and
C. pyrenoidosa. Moreover, accumulation of L-AA was ob-
served in S. cerevisiae cells incubated with L-Gal or L-1,4-
lactones (L-GL and L-GuL) which act as direct precursors
Fig. 1. Similarities in the pathways of L-AA biosynthesis in plants and
for L-AA biosynthesis in plant and animal cells [15,19]. D-EAA biosynthesis in yeasts.
The conversion of L-Gal into L-AA in plant cells involves
L-Gal dehydrogenase, a novel enzyme which displays high
speci¢city for L-Gal and selectivity for NAD‡ as a cofac- L-Gal. These observations are consistent with the kinetic
tor [19,27]. In cell-free extracts of the S. cerevisiae cultures properties of D-Ara dehydrogenase (E.C. 1.1.1.117), an
used in the present experiments, oxidation of L-Gal was enzyme which displays a broad substrate speci¢city and
much more substantial in the presence of NADP‡ than oxidises D-Ara, and L-Gal at comparable rates preferen-
with NAD‡ . We also observed rates of D-Ara oxidation in tially utilising NADP‡ as a cofactor [23,24]. D-Ara dehy-
the presence of NADP‡ similar to those observed with drogenase is involved in the biosynthesis of D-EAA in

FEMSLE 9388 27-4-00

 R.D. Hancock et al. / FEMS Microbiology Letters 186 (2000) 245^250
yeast [23,24], a pathway that shares many features with the
pathway for L-AA biosynthesis in plants [2] (Fig. 1). For
example, in both pathways, the ¢nal stages involve the
oxidation of a sugar at C1 to its corresponding aldonolac-
tone followed by oxidation of the aldonolactone at C2.
Two L-GL oxidases have been reported in S. cerevisiae.
One, reported by Nishikimi et al. [9], has similar properties
to the animal L-GuL oxidase and readily oxidises D-arabi-
nono-1,4-lactone, L-xylono-1,4-lactone, L-GuL and L-GL
in vitro. The second, puri¢ed by Bleeg and Christensen
[10], has distinct molecular and kinetic properties includ-
ing an inability to oxidise L-GuL. Our ¢nding that both L-
GL and L-GuL are converted into L-AA in S. cerevisiae
questions the involvement of the latter enzyme during L-
AA synthesis in yeast. The lower L-AA accumulation ob-
served in cells cultured with L-GuL compared with cells
cultured with L-GL is also consistent with the properties of
the L-GL oxidase puri¢ed by Nishikimi et al. [9] which
catalyses the oxidation of L-GuL at only 32% of the rate
of L-GL oxidation. Recently, Huh et al. [8] proposed the
re-classi¢cation of yeast L-GL oxidase as D-arabinono-1,4-
lactone oxidase, on the basis of the likely physiological
substrate of the enzyme. This claim has received support
from recent ¢ndings. The gene encoding D-arabinono-1,4-
lactone oxidase in S. cerevisiae has recently been cloned
and found to have a complete identity in the amino acid
sequence with an unknown open reading frame YL086C
[8]. Independently, Nishikimi et al. [28] observed that the
expression product of YML086C in Escherichia coli has L-
GL oxidase activity and covalently binds £avin as is the
case with the L-GL oxidase they puri¢ed earlier [9]. The
proposal is that this enzyme, together with D-Ara dehy-
drogenase, catalyses the synthesis of D-EAA from D-Ara
in yeast under physiological conditions [8]. This hypothesis
is supported by the ¢nding of D-EAA in yeast grown with
D-aldoses ([5], this paper) and its accumulation in the pres-
ence of exogenous D-Ara. Moreover, genetic manipulation
of D-Ara dehydrogenase results in changes in D-EAA
which are correlated with changes in enzyme activity
whilst no L-AA is detected even with up-regulated D-Ara
dehydrogenase activity [8]. Based on these considerations,
we propose that the conversion of L-Gal, L-GL and L-GuL
to L-AA in S. cerevisiae occurs via utilisation of these
unphysiological substrates by the enzymes of the D-EAA
pathway by virtue of their broad substrate speci¢city. This
hypothesis is supported by the ¢nding that pre-loading of
S. cerevisiae cells with D-Ara resulted in a very marked
reduction of L-[1-14 C]Gal incorporation into L-AA.
Our data demonstrate that synthesis of L-AA in S. ce-
revisiae can be induced by supply of unphysiological
substrates such as L-Gal and various L-1,4-lactones which
are converted into L-AA via the enzymes of the D-EAA
pathway. This ¢nding o¡ers opportunities for the bioen-
gineering of yeast to carry out one-step fermentation pro-
cesses for the conversion of simple precursors into L-AA,
which are currently not available. One possibility may be
FEMSLE 9388 27-4-00
249
through overexpression of genes responsible for the syn-
thesis of L-Gal in plants, at least one of which has been
identi¢ed [29]. An alternative may be through direct epi-
merisation of D-aldoses into L-Gal. Such a mechanism has
been suggested for conversion of D-Glc into L-Gal during
the synthesis of the tunic of the marine ascidian Styela
plicata [30].
Acknowledgements
We wish to thank Drs. Won-Ki Huh and Sa-Ouk Kang
of the Seoul National University, Korea, for their kind gift
Downloaded from https://academic.oup.com/femsle/article-abstract/186/2/245/689187 by guest on 04 May 2020
of D-EAA. SCRI receives grant-aided support from the
Scottish Executive Rural A¡airs Department (R.V.). The
authors wish to acknowledge the ¢nancial support of
SmithKline Beecham.
References
[1] Boudrant, J. (1990) Microbial processes for ascorbic acid biosynthe-
sis : a review. Enzym. Microb. Technol. 12, 322^329.
[2] Loewus, F.A. (1999) Biosynthesis and metabolism of ascorbic acid in
plants and of analogues of ascorbic acid in fungi. Phytochemistry 52,
193^210.
[3] Heick, H.M.C., Stewart, H.B., Gra¡, G.L.A. and Humpers, J.E.C.
(1969) The occurrence of ascorbic acid in the yeast Lypomyces star-
keyi. Can. J. Biochem. 47, 751^752.
[4] Heick, H.M.C., Gra¡, G.L.A. and Humpers, J.E.C. (1972) The oc-
currenec of ascorbic acid among the yeasts. Can. J. Microbiol. 18,
597^600.
[5] Nick, J.A., Leung, C.T. and Loewus, F.A. (1986) Isolation and iden-
ti¢cation of erythroascorbic acid in Saccharomyces cerevisiae and
Lypomyces starkeyi. Plant Sci. 46, 181^187.
[6] Running, J.A., Huss, R.J. and Olsen, P.T. (1994) Heterotrophic pro-
duction of ascorbic acid by microalgae. J. Appl. Phycol. 6, 99^104.
[7] Shao, Y.-Y., Seib, P.A., Kramer, K.J. and Van Galen, D.A. (1993)
Synthesis and properties of D-erythroascorbic acid and its vitamin C
activity in tobacco hornworm (Manduca sexta). J. Agric. Food Chem.
41, 1391^1396.
[8] Huh, W.-K., Lee, B.-H., Kim, S.-T., Kim, Y.-R., Rhie, G.-E., Baek,
Y.-W., Hwang, C.-S., Lee, J.-S. and Kang, S.-O. (1998) D-Erythroas-
corbic acid is an important antioxidant molecule in Saccharomyces
cerevisiae. Mol. Microbiol. 30, 895^903.
[9] Nishikimi, M., Noguchi, E. and Yagi, K. (1978) Occurrence in yeast
of L-galactonolactone oxidase which is similar to a key enzyme for
ascorbic acid biosynthesis in animals, L-gulonolactone oxidase. Arch.
Biochem. Biophys. 191, 486^489.
[10] Bleeg, H.S. and Christensen, F. (1982) Biosynthesis of ascorbate in
yeast. Puri¢cation of L-galactono-1,4-lactone oxidase with properties
di¡erent from mammalian L-gulonolactone oxidase. Eur. J. Biochem.
127, 391^396.
[11] Oba, K., Ishikawa, S., Nishikawa, M., Mizuno, H. and Yamamoto,
T. (1995) Puri¢cation and properties of L-galactono-Q-lactone dehy-
drogenase, a key enzyme for ascorbic acid biosynthesis, from sweet
potato roots. J. Biochem. 117, 120^124.
[12] Òstergaard, J., Persiau, G., Davey, M.W., Bauw, G. and Van Mon-
tagu, M. (1997) Isolation of a cDNA coding for L-galactono-Q-lac-
tone dehydrogenase, an enzyme involved in the biosynthesis of ascor-
bic acid in plants. J. Biol. Chem. 272, 30009^30016.
[13] Touster, O. (1962) Carbohydrate metabolism. Ann. Rev. Biochem.
31, 407^450.
250 R.D. Hancock et al. / FEMS Microbiology Letters 186 (2000) 245^250
[14] Onofri, S., Poerio, E., Serangeli, P., Tosi, S., Garuccio, I. and Arri-
goni, O. (1997) In£uence of L-galactonoic acid Q-lactone on ascorbate
production in some yeasts. Antonie van Leeuwenhoek 71, 277^280.
[15] Isherwood, F.A., Chen, Y.T. and Mapson, L.W. (1954) Synthesis of
L-ascorbic acid in plants and animals. Biochemistry 56, 1^15.
[16] Baig, M.M., Kelly, S. and Loewus, F. (1970) L-Ascorbic acid biosyn-
thesis in higher plants from L-gulono-1,4-lactone and L-galactono-1,4-
lactone. Plant Physiol. 46, 277^280.
[17] Arrigoni, O. (1994) Ascorbate system in plant development. J. Bio-
energ. Biomembr. 26, 407^419.
[18] Viola, R., Vreugdenhil, D., Davies, H.V. and Sommerville, L. (1998)
Accumulation of L-ascorbic acid in tuberising stolon tips of potato
(Solanum tuberosum L.). J. Plant Physiol. 152, 58^63.
[19] Wheeler, G.L., Jones, M.A. and Smirno¡, N. (1998) The biosynthetic
pathway of vitamin C in higher plants. Nature 393, 365^369.
[20] RenstrY è m, B., Gro©n, M. and Loewus, F.A. (1982) Biosynthesis of
L-ascorbic acid in Chlorella pyrenoidosa. Plant Sci. Lett. 28, 299^305.
[21] Provasoli, L. and Pintner, I.J. (1959) In: The Ecology of Algae.
Special Publication No. 2 Pymatuning Laboratory of Field Biology
(Tyron, C.A. and Hartman, R.T., Eds.), pp. 84^96. University of
Pittsburgh.
[22] Conklin, P.L., Pallanca, J.E., Last, R.L. and Smirno¡, N. (1997)
L-Ascorbic acid metabolism in the ascorbate-de¢cient Arabidopsis
mutant vtc1. Plant Physiol. 115, 1277^1285.
FEMSLE 9388 27-4-00
[23] Kim, S.-T., Huh, W.-K., Kim, J.-Y., Hwang, S.-W. and Kang, S.-O.
(1996) D-Arabinose dehydrogenase and biosynthesis of erythroascor-
bic acid in Candida albicans. Biochim. Biophys. Acta 1297, 1^8.
[24] Kim, S.-T., Huh, W.-K., Lee, B.-H. and Kang, S.-O. (1998) D-Ara-
binose dehydrogenase and its gene from Saccharomyces cerevisiae.
Biochim. Biophys. Acta 1429, 29^39.
[25] Bradford, M.M. (1976) A rapid and sensitive method for the quanti-
tation of microgram quantities of protein utilizing the principle of
protein^dye binding. Anal. Biochem. 72, 248^254.
[26] Loewus, F. (1971) Carbohydrate interconversions. Annu. Rev. Plant
Physiol. 22, 337^364.
[27] Smirno¡, N. and Wheeler, G.L. (1999) Plant galactose dehydrogen-
ase. UK Patent 33995.
[28] Nishikimi, M., Ohta, Y. and Ishikawa, T. (1998) Identi¢cation by
bacterial expression of the yeast genomic sequence encoding L-galac-
Downloaded from https://academic.oup.com/femsle/article-abstract/186/2/245/689187 by guest on 04 May 2020
tono-Q-lactone oxidase, the homologue of L-ascorbic acid-synthesizing
enzyme of higher animals. Biochem. Mol. Biol. Int. 44, 907^913.
[29] Conklin, P.L., Norris, S.R., Wheeler, G.L., Williams, E.H., Smirno¡,
N. and Last, R.L. (1999) Genetic evidence for the role of GDP-
mannose in plant ascorbic acid (vitamin C) biosynthesis. Proc.
Natl. Acad. Sci. USA 96, 4198^4203.
[30] Moura¬o, P.A.S. (1991) Epimerization of D-glucose to L-galactose
during the biosynthesis of a sulfated L-galactan in the ascidian tunic.
Biochemistry 30, 3458^3464.