Download The role of xylulokinase in Saccharomyces cerevisiae xylulose

FEMS Microbiology Letters 190 (2000) 39^43
www.fems-microbiology.org
The role of xylulokinase in Saccharomyces cerevisiae
xylulose catabolism
Peter Richard *, Mervi H. Toivari, Merja Penttila«
VTT Biotechnology, Tietotie 2, P.O. Box 1500, FIN-02044 Espoo, Finland
Received 9 May 2000; received in revised form 26 June 2000; accepted 2 July 2000
Abstract
Many yeast species have growth rates on D-xylulose of 25^130% of those on glucose, but for Saccharomyces cerevisiae this ratio is only
about 6%. The xylulokinase reaction has been proposed to be the rate-limiting step in the D-xylulose fermentation with S. cerevisiae. Overexpression of xylulokinase encoding XKS1 stimulated growth on D-xylulose in a S. cerevisiae strain to about 20% of the growth rate on
glucose and deletion of the gene prevented growth on D-xylulose and D-xylulose metabolism. We have partially purified the xylulokinase and
characterised its kinetic properties. It is reversible and will also accept D-ribulose as a substrate. ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.
Keywords : Xylulokinase ; D-Xylulose ; D-Ribulose; Xylose fermentation; Saccharomyces cerevisiae
1. Introduction
Xylulokinase converts the ¢ve-carbon sugar D-xylulose
and ATP to xylulose 5-phosphate and ADP. Xylulose 5phosphate can then be metabolised through the pentose
phosphate pathway. D-Xylulose is not abundant in nature
but can be formed from D-xylose, a major sugar of wood
¢bre hydrolysates, by the action of xylose isomerase in
bacteria or a combined action of xylose reductase and
xylitol dehydrogenase in yeast and ¢lamentous fungi.
Although cells of Saccharomyces cerevisiae cannot naturally utilise D-xylose, they can grow on D-xylulose as a sole
carbon source [1^3]. Ethanol is formed from D-xylulose
anaerobically in the same yields as from D-glucose [4],
which makes S. cerevisiae potentially relevant in the context of fermenting hemicellulose hydrolysates. However,
growth on D-xylulose is so slow that improved growth
on D-xylulose was used as a screen to identify a xylulokinase gene from S. cerevisiae [5,6]. The sequence of this
gene was ¢rst published by Ho and Tsao [7]. Recently,
an almost identical sequence of the open reading frame
YGR194c in the S. cerevisiae genome [8] was reported to
encode xylulokinase and renamed as XKS1 [9]. A signi¢cant increase in growth on xylulose was found when XKS1
* Corresponding author. Tel. : +358 (9) 456-7190;
Fax: +358 (9) 455-2103; E-mail: peter.richard@vtt.¢
was over-expressed [10]. However, other researchers found
the over-expression lethal [9]. In this study, we determine
the kinetic properties of the S. cerevisiae xylulokinase and
examine the role of the expression level of XKS1 for
growth on xylulose. We show that this is the only pathway
for D-xylulose metabolism and compare the growth rates
on xylulose to rates on glucose with and without XKS1
over-expression.
2. Materials and methods
2.1. Cloning of the xylulokinase gene and over-expression in
S. cerevisiae
The xylulokinase gene (XKS1) was ampli¢ed from the
total DNA of the S. cerevisiae strain W303-1B [11] by
PCR, using the forward primer 5P-CCA GTG ATA
TCG AGG ATG AGA TTA GTA C-3P and the reverse
primer 5P-CCA GTG ATA TCT GTA CTT GTC AGG
GCA T-3P. The PCR product was cloned between the
modi¢ed, constitutive ADH1 promoter and terminator
[12], in the Bluescript M13 vector (Stratagene). A BamHI
fragment, containing the XKS1 gene with ADH1 promoter
and terminator sequences was subsequently cloned to the
BamHI site in YEplac195 yeast expression vector [13]. The
plasmid was transformed to the S. cerevisiae strains W3031B and CENPK.2 [14] with the lithium acetate procedure
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 3 1 2 - 8
FEMSLE 9536 10-8-00
40
P. Richard et al. / FEMS Microbiology Letters 190 (2000) 39^43
[15]. The same BamHI fragment was also cloned into a
Bluescript vector (Stratagene) containing part of the S.
cerevisiae HIS3 gene and a geneticin G418 resistance
gene KMX2 [16], which was constructed as follows. A
1500-bp DrdI fragment from a yeast expression vector
pRS423 was cloned into the EcoRV site of the Bluescript
as blunt end ligation. A NheI^BclI fragment of the HIS3
gene was replaced with the G418 resistance gene KMX2
digested with PvuII and SpeI and blunt-ended. The BamHI fragment containing the XKS1 gene between the
ADH1 promoter and terminator was cloned to the BglII
site in the HIS3 gene. The BglII digestion removes a 50-bp
fragment from the HIS3 gene. A cassette containing approximately 500 bp of the HIS3 sequence in both ends and
the XKS1 and KMX2 genes in the middle was released
with BstBI and BssHII and integrated into HIS3 locus
of the CENPK.2 strain. Correct integration was identi¢ed
in the geneticin resistant transformants by Southern blot.
2.2. Deletion of the xylulokinase gene
A deletion cassette was constructed by ¢rst ligating a 1kb EaeI fragment of XKS1 to the compatible NotI site in
the pZErO-1 vector (Invitrogen) followed by ScaI digestion to remove a 500-bp fragment from the middle of the
XKS1 sequence. The XKS1 fragment with the ScaI deletion was moved from this vector as a NsiI fragment to the
PstI site of Bluescript SK (Stratagene). The KMX2 cassette encoding G418 resistance [16] was then cloned by
blunt end ligation to the BstEII site in the XKS1 sequence.
This disruption cassette on the resulting plasmid was then
ampli¢ed by PCR with oligos 5P-TCA GGG TCC TGC
CAG CA-3P and 5P-CGC TAG GAA CGA TCT CC-3P.
The fragment was transformed into the strain W303-1B
and the xylulokinase disruptants were screened on YPD
plates containing 200 mg l31 of G418. The disruption was
con¢rmed by PCR and Southern blot.
2.3. Growth on D-xylulose
Cells were grown in shake £asks at 30³C in synthetic
complete medium, leaving out uracil when required for
selection [17], containing either a mixture of D-xylulose
10 (g l31 ) and D-xylose 30 (g l31 ) prepared by isomerisation of D-xylose [18] or 5 (g l31 ) pure D-xylulose (Sigma).
The plasmid stability was tested by plating equal amounts
of cells to selective and non-selective medium. In the experiment to monitor the D-xylulose consumption in the
wild-type and xks1 deletion strains, the initial biomass
used corresponded to an OD600 of 5.
2.4.
D-Xylulose
assay
The prediluted supernatants from the D-xylulose cultivations were assayed in a bu¡er containing 100 mM
NaPO4 pH 7.0 and 200 WM NADH. The reaction was
started with 0.15 U ml31 sorbitol dehydrogenase (Boehringer). The concentration was calculated from the NADH
absorbance change at 340 nm after the reaction was completed after 15 min.
2.5. Xylulokinase activity assay
Cells were grown in synthetic complete medium on 20 g
l31 glucose and extracted with glass beads. Xylulokinase
activity was assayed in a medium containing 50 mM PipesK pH 7.0, 100 mM KCl, 5 mM MgCl2 , 5 mM ATP,
0.2 mM NADH, 1 mM phosphoenol pyruvate, pyruvate
kinase 5 U ml31 (Boehringer) and lactate dehydrogenase
7 U ml31 (Boehringer). The reaction was started by addition of D-xylulose (Sigma) to a ¢nal concentration of
17 mM. The background activity, ATPase activity in the
absence of D-xylulose, was subtracted in each case.
2.6. Assay for measuring the Michaelis^Menten constants
for xylulokinase
The Michaelis^Menten constants were measured using
the partially puri¢ed enzyme, where no background ATPase activity could be detected (see below). The constants
for D-xylulose and ATP were measured in a medium containing 50 mM PipesK pH 6.5, 100 mM KCl, 5 mM
MgCl2 , 0.2 mM NADH, 1 mM phosphoenol pyruvate,
pyruvate kinase 5 U ml31 , lactate dehydrogenase 7 U
ml31 and 5 mM ATP or 1 mM D-xylulose. The reaction
was started by addition of D-xylulose or ATP. The Michaelis^Menten constants for xylulose 5-phosphate and
ADP were measured in a medium containing 50 mM PipesK pH 6.5, 100 mM KCl, 5 mM MgCl2 , 0.2 mM
NADH, 0.1 U ml31 sorbitol dehydrogenase (Boehringer)
and either 4 mM ADP or 10 mM xylulose 5-phosphate.
The reaction was started by addition of ADP or xylulose
5-phosphate. All assays were performed at 30³C in a Cobas Mira+automated analyser (Roche).
2.7. Partial puri¢cation of the xylulokinase enzyme
XKS1 over-expressing cells of the strain W303-1B were
grown overnight in selective medium [17]. The yeast (15 g
fresh weight) was collected by centrifugation and resuspended into 15 ml of 10 mM potassium phosphate bu¡er
and homogenised with an equal volume of glass beads (0.5
mm diameter) in a Braun homogeniser. The mixture was
¢ltered through a glass funnel and the ¢ltrate centrifuged
at 40 000Ug (Sorvall SS34). This resulted in 17 ml of a
supernatant with 17 mg ml31 protein and 58 nkat mg31
xylulokinase activity. This crude extract was then applied
to a Sephadex PD10 gel ¢ltration column (Pharmacia).
Two ml per column was used and the columns were equilibrated with 10 mM NaPO4 pH 7.0. A total of 90 mg
protein was loaded to a 10 ml DEAE-cellulose column
(Pharmacia), washed with 40 ml of 10 mM NaPO4 and
FEMSLE 9536 10-8-00
P. Richard et al. / FEMS Microbiology Letters 190 (2000) 39^43
41
eluted with 200 ml of a gradient from 10 mM to 100 mM
of NaPO4 pH 7.0. The fraction with the highest xylulokinase activity had 0.25 mg ml31 protein and an activity of
520 nkat mg31 and no background ATPase activity.
3. Results and discussion
3.1. Xylulokinase kinetics
The xylulokinase was extracted from a XKS1 overproducing strain and about 10-fold puri¢ed by chromatography on DEAE-cellulose. This partially puri¢ed protein
was free of any background ATPase and xylitol dehydrogenase activity, and therefore suitable for kinetic measurements.
The forward reaction of xylulokinase was measured in a
coupled assay with pyruvate kinase and lactate dehydrogenase in order to measure the production rate of ADP.
The Km for D-xylulose was 310 þ 10 WM and for ATP
1550 þ 80 WM (Fig. 3A,B). The reverse reaction was measured in a coupled assay with sorbitol dehydrogenase to
measure the production rate of D-xylulose. The Km for
xylulose 5-phosphate was 1300 þ 100 WM and for ADP
280 þ 30 WM (Fig. 3C,D). The apparent Vmax values of
the forward reactions were about 600 nkat mg31 (640
with D-xylulose, 500 with ATP). Km and Vmax were obtained from Hanes^Woolf plots of the presented data. All
these data were obtained under similar conditions at the
same pH (pH 6.5). From these data, an equilibrium constant of 90 was calculated using the Haldane relationship
Keq = (Vmax forward/Vmax reverse)2 (Km values products/Km
values substrates). The corresponding free energy for the
kinase reaction is 311 kJ mol31 . It needs to be noticed,
however, that this is not an accurate method because it is
subject to cumulative errors in the six kinetic constants
applied, including the very pH dependent Km for ATP.
We also checked the speci¢city of the reaction. Using 10
mM substrate, the rate with D-ribulose was 60% of that
with D-xylulose. With D-glucose, D-fructose, D-xylose, ribitol (adonitol), xylitol, D-tagatose and D-sorbose, the activity was below 1% under similar conditions, i.e. below the
detection limit.
The kinetic properties of the S. cerevisiae xylulokinase
are quite similar with the xylulokinase from Pichia stipitis,
an e¤cient D-xylose fermenting yeast. The Km values for
D-xylulose are similar, 300 WM and 285 WM for S. cerevisiae at pH 6.5 and pH 7.0, respectively, and 520 WM for P.
stipitis at pH 7.6 as reported by Flanagan and Waites [19].
The Km values for ATP were 1500 WM and 165 WM at pH
6.5 and pH 7.0, respectively, measured for S. cerevisiae in
this work and 280 WM for P. stipitis at pH 7.6 [19]. One
di¡erence between the enzymes of the two organisms is
that the P. stipitis enzyme is more speci¢c for D-xylulose,
i.e. it has a much lower activity with D-ribulose than the S.
cerevisiae enzyme [19].
Fig. 1. Growth on xylulose. Growth of the S. cerevisiae strain W303-1B
(W303) and the strain over-expressing xylulokinase (W303 XKS1) on a
D-xylose/D-xylulose mixture (see text).
3.2. Growth on D-xylulose
To study whether the xylulokinase expression level is
limiting the D-xylulose fermentation and whether an
over-expression is detrimental for growth on D-xylulose,
we over-expressed the XKS1 gene in the S. cerevisiae strain
W303-1B. The doubling time of the W303-1B parent strain
was 34 h. Due to the high price of pure D-xylulose, growth
was ¢rst tested on D-xylose/D-xylulose mixtures, which can
be obtained by incubating D-xylose with glucose isomerase
[18]. Since S. cerevisiae does not naturally utilise D-xylose,
growth should result only from D-xylulose utilisation.
The XKS1 was expressed from a multicopy vector under
the modi¢ed constitutive ADH1 promoter [12]. Xylulokinase activity was 50 nkat per mg of extracted protein. In
the parent strain, the activity was not detectable, i.e. it was
below 1 nkat mg31 .
Over-expression of the XKS1 led to an about 3-fold
increase in the growth rate with strain W303-1B to a doubling time of 12 h (Fig. 1). The doubling time of the parent strain W303-1B on glucose is approximately 2 h.
The XKS1 was also over-expressed in another laboratory strain, the CENPK.2. This is a naturally better D-xylulose consumer with a doubling time of about 15 h, however, the xylulokinase activity is under our detection limit.
The over-expression by integration into the genome did
not lead to a signi¢cant increase in growth rate on D-xylulose. With the multicopy vector construct, the results
varied from enhanced growth rate to similar growth rate
compared to the parent strain. However, the growth on
xylulose never exceeded a doubling time of about 12 h.
3.3. The expression of xylulokinase had no e¡ect on the
growth rate on glucose
Wang and Schneider [1] found ratios between growth
rates on D-glucose and D-xylulose ranging from 0.75 and
4 for various yeasts, but S. cerevisiae had an exceptionally
high ratio of 15. By over-expression of the xylulokinase,
FEMSLE 9536 10-8-00
42
P. Richard et al. / FEMS Microbiology Letters 190 (2000) 39^43
we could stimulate the growth rate on D-xylulose in a S.
cerevisiae strain which naturally grows slow on D-xylulose.
The ratio between growth on glucose, doubling time 2 h,
and growth on D-xylulose, doubling time 12 h, is now 6,
which is closer to the ratio observed for other yeasts.
However, despite over-expressing xylulokinase, the growth
rate on xylulose is still several fold lower than on glucose.
In S. pombe for example, the doubling times on xylulose
and glucose are 5 and 3 h, respectively [1]. This means that
in S. cerevisiae other factors like activities of other enzymes may be limiting, e.g. transaldolase, as sedoheptulose
7-phosphate has been shown to accumulate in D-xylulose
fermentation [20].
Previously, Deng and Ho [6] saw improved growth on
D-xylulose with over-expressed XKS1. In contrast, Rodrigues-Pena et al. [9] found that the over-expression of
xylulokinase gene is toxic for the cell. We mainly used a
mixture of D-xylose and D-xylulose as a source of D-xylulose, whereas Rodrigues-Pena et al. [9] used pure D-xylulose. It has been reported for Fusarium oxysporum that Dxylulose uptake is partially inhibited by D-xylose [21]. An
unlimited uptake combined with phosphorylation through
xylulokinase might lead to ATP depletion, similarly to the
toxicity caused by unregulated glucose uptake [22]. To
verify that there is no e¡ect of the D-xylose present in
the mixture, we also tested growth on pure D-xylulose.
The XKS1 overproducing strains were grown in shake
£asks with 5 g l31 of pure D-xylulose as a sole carbon
source. The growth of the XKS1 overproducing strains
was not a¡ected whether pure D-xylulose or a D-xylose/
D-xylulose mixture was used (not shown). We also did
not observe any plasmid loss during growth on D-xylulose.
To check whether there are other metabolic routes to
metabolise xylulose besides the xylulokinase, we also deleted the XKS1 gene. In the strain W303-1B with a xks1
deletion, we could not detect any growth on D-xylulose.
This is in accordance with Stevis and Ho [23]. To exclude
the possibility that some D-xylulose was metabolised, but
growth was too slow to be detected, we monitored the Dxylulose utilisation by a dense cell suspension of the wild-
Fig. 3. A^D: Reaction rates of the puri¢ed xylulokinase. The Michaelis^Menten constants were calculated from a linear regression of a
Hanes^Woolf plot. The curves are calculated Michaelis^Menten kinetics
using the constants as indicated. A: ATP is 5 mM. The Km for D-xylulose is 310 þ 10 WM, the Vmax 640 nkat mg31 . B: D-Xylulose is 1 mM.
The Km for ATP is 1550 þ 80 WM, the Vmax 500 nkat mg31 . C: ADP is
4 mM. The Km for xylulose 5-phosphate is 1300 þ 100 WM, the Vmax 60
nkat mg31 . D: Xylulose 5-phosphate is 10 mM. The Km for ADP is
310 þ 10 WM, the Vmax 50 nkat mg31 .
type and the xks1 deletion strain. It was found that the
wild-type strain W303-1B was consuming D-xylulose but
the xks1 deletion strain was not (Fig. 2). This is indicating
that the xks1 is the principle mechanism by which D-xylulose is metabolised. There is also no other open reading
frame in the S. cerevisiae genome with a similar sequence.
We conclude that the kinetic properties of the S. cerevisiae xylulokinase resemble those of the xylulokinase
from xylose fermenting yeast P. stipitis except that S. cerevisiae enzyme is also an e¤cient ribulokinase. We also
conclude that the expression level of xylulokinase can limit
xylulose metabolism, although other factors are also likely
to have a role.
Acknowledgements
P.R. was supported through Grant no. BIO4CT965020
from the Commission of the EU. We thank Jenita Rauta
and Eila Leino for technical assistance and John Londesborough for useful discussions.
References
Fig. 2. Xylulose consumption. D-Xylulose consumption in a strain of S.
cerevisiae with a deletion in the xylulokinase gene (W303 xks1) and the
corresponding control (W303).
[1] Wang, P.Y. and Schneider, H. (1980) Growth of yeasts on D-xylulose.
Can. J. Microbiol. 26, 1165^1168.
[2] Ueng, P.P., Hunter, C.A., Gong, C.S. and Tsao, G.T. (1981) D-Xylulose fermentation in yeasts. Biotechnol. Lett. 3, 315^320.
[3] Chiang, L.C., Gong, C.S., Chen, L.F. and Tsao, G.T. (1981) D-Xylulose fermentation to ethanol by Saccharomyces cerevisiae. Appl.
Environ. Microbiol. 42, 284^289.
[4] Wang, P.Y., Shopsis, C. and Schneider, H. (1980) Fermentation of a
pentose by yeasts. Biochim. Biophys. Res. Commun. 94, 248^254.
FEMSLE 9536 10-8-00
P. Richard et al. / FEMS Microbiology Letters 190 (2000) 39^43
[5] Ho, N.W.Y. and Chang, S.-F. (1989) Cloning of yeast xylulokinase
gene by complementation of E. coli and yeast mutations. Enzyme
Microb. Technol. 11, 417^421.
[6] Deng, X.X. and Ho, N.W.Y. (1990) Xylulokinase activity in various
yeasts including Saccharomyces cerevisiae containing the cloned xylulokinase gene. Appl. Biochem. Biotech. 24/25, 193^199.
[7] Ho, N.W.Y. and Tsao, G.T. (1995) Recombinant yeasts for e¡ective
fermentation of glucose and xylose. International Application Published Under the Patent Cooperation Treaty, WO 95/13362.
[8] Go¡eau, A. et al. (1997) The yeast genome directory. Nature 387,
suppl.
[9] Rodrigez-Pena, J.M., Cid, V.J., Arroyo, J. and Nombela, C. (1998)
The YGR194c (XKS1) gene encodes the xylulokinase from the budding yeast Saccharomyces cerevisiae. FEMS Microb. Lett. 162, 155^
160.
[10] Ho, N.W.Y., Chen, Z. and Brainard, A.P. (1998) Genetically engineered Saccharomyces yeast capable of e¡ective cofermentation of
glucose and xylose. Appl. Env. Microbiol. 64, 152.
[11] Thomas, B.J. and Rothstein, R. (1989) Elevated recombination rates
in trascriptionally active DNA. Cell 56, 619^630.
[12] Ruohonen, L., Aalto, M. and Kera«nen, S. (1995) Modi¢cations to
the ADH1 promoter of Saccharomyces cerevisiae for e¤cient production of heterologous proteins. J. Biotechnol. 39, 193^203.
[13] Gietz, R.D. and Sugino, A. (1988) New yest^Escherichia coli shuttle
vectors constructed with in vitro mutagenized yeast genes lacking sixbase pair restriction sites. Gene 74, 527^534.
[14] Boles, E., Go«hlmann, H.W.H. and Zimmermann, F.K. (1996) Cloning of a second gene encoding 6-phospho-fructo-2-kinase in yeast,
and characterization of mutant strains without fructose-2,6-bisphosphate. Mol. Microbiol. 20, 65^76.
43
[15] Gietz, D., St Jean, A., Woods, R.A. and Schiesti, R.H. (1992) Improved method for high e¤ciency transformation of intact yeast cells.
Nucleic Acids Res. 20, 1425.
[16] Wach, A., Brachat, A., Po«hlmann, R. and Philippsen, P. (1994) New
heterologous modules for classical or PCR-based gene disruptions in
Saccharomyces cerevisiae. Yeast 10, 1793^1808.
[17] Sherman, F., Fink, G. and Hicks, J.B. (1983) Methods in Yeast
Genetics. A Laboratory Manual. Cold Spring Harbor Laboratory
Press, Cold Spring Harbor, NY.
[18] Olsson, L., Linden, T. and Hahn-Ha«gerdahl, B. (1994) A rapid chromatographic method for the production of preparative amounts of
xylulose. Enzyme Microb. Technol. 16, 388^394.
[19] Flanagan, T. and Waites, M.J. (1992) Puri¢cation and characterization of D-xylulokinase from the pentose-fermenting yeast Pichia stipidis NCYC 1541. Enzyme Microb. Technol. 14, 975^979.
[20] Senac, T. and Hahn-Ha«gerdal, B. (1990) Intermediary metabolite
concentrations in xylulose- and glucose-fermenting Saccharomyces
cerevisiae cells. Appl. Environ. Microbiol. 56, 120^126.
[21] Singh, A. and Schugerl, K. (1992) Induction and regulation of Dxylose catabolizing enzymes in Fusarium oxysporum. Biochem. Int.
28, 481^488.
[22] Hohmann, A., Neves, M.J., de Koning, W., Alijo, R., Ramos, J. and
Thevelein, J.M. (1993) The growth and signalling defects of the ggs1
(fdp1/byp1) deletion mutant on glucose are suppressed by a deletion
of the gene encoding hexokinase PII. Curr. Genet. 23, 281^289.
[23] Stevis, P.E. and Ho, N.W.Y. (1989) Construction of yeast xylulokinase mutant by recombinant DNA techniques. Appl. Biochem. Biotechnol. 20/21, 327^334.
FEMSLE 9536 10-8-00