Engineering Saccharomyces cerevisiae for efficient anaerobic

Full text

Biotechnology
Journal
DOI 10.1002/biot.201100053 Biotechnol. J. 2012, 7, 34–46
34 © 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1 Introduction
Ethanol production from lignocellulosic biomass
represents an attractive way for alternative biofuel
production due to the rising concerns over the de-
pletion of fossil resources and damage to the envi-
ronment caused by fossil fuel usage. Lignocellu-
losic biomass must first be hydrolyzed to glucose,
xylose, and other sugars, before they can be fer-
mented. Saccharomyces cerevisiae is the most
promising candidate for lignocellulosic ethanol
production due to its excellent glucose fermenta-
tion capability, high ethanol tolerance, and resist-
ance to inhibitors presented in lignocellulosic hy-
drolysates [1]. However, native S. cerevisiae strains
cannot utilize xylose for either growth or ethanol
production, leaving a significant fractions of ligno-
cellulosic hydrolysates (20–30%) unusable [2, 3].
Despite of the inability to utilize xylose, S. cere-
visiae has the metabolic pathway to convert xylu-
lose, an isomerized product of xylose, into ethanol
(Fig.1). Xylulose is phosphorylated by xylulokinase
(XK), metabolized through the pentose phosphate
pathway (ppp), and then channeled into glycolysis.
Thus, numerous studies have been carried out to
link extracellular xylose with intracellular xylulose
so as to further increase the efficiency of xylose
fermentation (for reviews,see [4, 5]). Efforts on en-
abling S. cerevisiae to ferment xylose can be divid-
ed into six aspects according to the different engi-
neering targets along the entire xylose metabolic
pathway, as shown by various colors in Fig. 1:
Review
Engineering Saccharomyces cerevisiae for efficient anaerobic
xylose fermentation: Reflections and perspectives
Zhen Cai, Bo Zhang and Yin Li
Institute of Microbiology, Chinese Academy of Sciences, Beijing, China
Conversion of the abundant lignocellulosic biomass into ethanol is an environmentally sustain-
able solution to the energy crisis. Fermentation of lignocellulosic hydrolysates by Saccharomyces
cerevisiae is not cost-effective yet as substantial amounts of xylose in the hydrolysates cannot be
utilized by native S. cerevisiae strains. Extensive studies including both metabolic and evolutionary
engineering have been carried out to develop an efficient xylose-fermenting S. cerevisiae strain, yet
the ethanol yield and productivity from xylose fermentation of the best one are still far below ex-
pectation. This review compares the engineering approaches and resulted anaerobic xylose fer-
mentation performance of recently reported xylose-utilizing S. cerevisiae strains, with the aim to
understand the intrinsic reason for their low xylose fermentation capabilities. These comparative
analyses revealed that some of the current engineering targets and the so-called “hot issues” might
be overrated. Our opinions on the underrated parts and future efforts in this field are also pre-
sented. Overall, this review serves as a comprehensive reference to understanding xylose fermen-
tation by S. cerevisiae.
Keywords: Anaerobic xylose metabolism · Evolutionary engineering · Lignocellulosic ethanol · Metabolic engineering ·
Saccharomyces cerevisiae
Correspondence: Prof. Yin Li, Institute of Microbiology, Chinese Academy
of Sciences, Beijing 100101, China
E-mail: yli@im.ac.cn
Abbreviations: ADH, alcohol dehydrogenase; ALD, aldehyde dehydroge-
nase; AR, aldose reductase; DW, dry weight; GAPD, glyceraldehyde-3-phos-
phate dehydrogenase; GPD6, glucose-6-phosphate dehydrogenase; HXT,
hexose transporters; ppp, pentose phosphate pathway; RPE, ribulose-5-
phosphate 3-epimerase; RPI, ribulose-5-phosphate isomerase; TAL, transal-
dolase; TKL, transketolase; XDH, xylitol dehydrogenase; XI, xylose iso-
merase; XK, xylulokinase; XR, xylose reductase
Received 29 May 2011
Revised 20 September 2011
Accepted 6 October 2011

© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 35
Biotechnol. J. 2012, 7, 34–46 www.biotechnology-journal.com
(i) Introducing an efficient xylose transporter to
bring more xylose into the cell;
(ii) Bridging the intracellular xylose and xylulose
either through heterologous expression of
both xylose reductase (XR) and xylitol dehy-
drogenase (XDH) genes, or through xylose iso-
merase (XI) gene alone;
(iii) Relieving the redox imbalance of the intro-
duced XR and XDH by changing their cofactor
specificities, or by adjusting other NADH- or
NADPH-producing steps;
(iv) Strengthening the step from xylulose to xylu-
lose-5-P by expression of the endogenous XK
gene;
(v) Enlarging the flux of ppp by over-expression of
its enzymes, such as ribulose-5-phosphate 3-
epimerase (RPE), ribulose-5-phosphate iso-
merase (RPI), transketolase (TKL), and trans-
aldolase (TAL);
(vi) Enhancing the xylose utilization efficiency by
iterative cycles of variation followed by selec-
tion.
The ethanol yield and productivity of xylose fer-
mentation by S. cerevisiae strains engineered
through one or several approaches described
above are much inferior compared to that of glu-
cose fermentation [4–7]. This indicates that these
engineered S. cerevisiae strains are still far away
from an economically viable lignocellulosic ethanol
production.
In order to find out the reason hampering the
development of efficient xylose-fermenting S. cere-
visiae strains, the engineering approaches of re-
cently reported recombinant yeast strains and their
xylose fermentation performance under anaerobic
conditions were analyzed and summarized in Fig.2.
Although some engineered S. cerevisiae strains
were reported to be capable of fermenting xylose
aerobically or micro-aerobically [8–10], this review
only focuses on the anaerobic xylose fermentation
data as aerobic fermentation is energy intensive
and thus not a preferred option for industrial pro-
duction of ethanol. Moreover, excessive aeration
would lead to a competition for glycolytic NADH
between mitochondrial respiration and alcoholic
Figure 1. Xylose and glucose metabolic pathways in engineered xylose-utilizing S. cerevisiae. Reported efforts in developing xylose-utilizing S. cerevisiae
strains include increasing xylose transport (shaded in orange), bridging xylose and xylulose (shaded in green), relieving the redox imbalance (shaded in yel-
low), strengthening the flux of xylulose to xylulose-5-P (shaded in blue), increasing the expression of enzymes in the pentose phosphate pathway (shaded
in purple), and engineering the whole cell (shaded in gray). ADH, alcohol dehydrogenase; ALD, aldehyde dehydrogenase; AR, aldose reductase; ENO, eno-
lase; FBA, fructose 1,6-bisphosphate aldolase; FBP, fructose-1,6-bisphosphatase; GAPD, glyceraldehyde-3-phosphate dehydrogenase; GK, glycerol kinase;
GLK, glucokinase; GPD3, glycerol-3-phosphate dehydrogenase; GPD6, glucose-6-phosphate dehydrogenase; GPP, glycerol-3-phosphatase; HXK, hexo-
kinase; HXT, hexose transporters; PDC, pyruvate decarboxylase; PFK, phosphofructokinase; PGI, phosphoglucose isomerase; PGK, 3-phosphoglycerate
kinase; PGM, phosphoglyceromutase; PGND 6-phosphogluconate dehydrogenase; PYK, pyruvate kinase; RPE, ribulose-5-phosphate 3-epimerase; RPI, ribu-
lose-5-phosphate isomerase; TAL, transaldolase; TKL, transketolase; TPI, triose phosphate isomerase; XDH, xylitol dehydrogenase; XI, xylose isomerase;
XK, xylulokinase; XR, xylose reductase.

Biotechnology
Journal
Biotechnol. J. 2012, 7, 34–46
36 © 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
fermentation, and thus are not suitable for maximal
ethanol production [11]. After a thorough compar-
ison and analysis of the methods and anaerobic
fermentation results of 30 engineered xylose-uti-
lizing strains,we found that some of the current en-
gineering targets, such as xylose transport, are ac-
tually not the rate-limiting steps. Moreover, some
well-known, so-called “hot issues” such as redox
imbalance in anaerobic xylose fermentation might
be overrated. Further efforts to improve the anaer-
obic xylose fermentation capability of S. cerevisiae
were also proposed based on these comparative
analyses.
2 Industrial strains may be a better start for
engineering than laboratory ones
Among the engineered xylose-fermenting S. cere-
visiae strains listed in Table 1, only 1400(pLNH32),
TMB3099/3400, MA-R4/R5, and BWY10Xyl origi-
nated from industrial Saccharomyces strains (1400,
USM21, IR-2,and BarraGrande, respectively).They
gained the ability to metabolize xylose to ethanol
either through plasmid-expression or chromoso-
mal integration of XR and XDH genes from Pichia
stipitis, and XK gene from S. cerevisiae (PsXR,
PsXDH, ScXK), or through plasmid-expression of
xylose isomerase from Clostridium phytofermen-
tans (CpXI). Although the ethanol yield for the
same strain varied a lot with respect to different
fermentation conditions (mostly affected by the
cell mass, fermentation time, and medium compo-
sition), the calculated specific ethanol productivi-
ties per gram of cell mass under different condi-
tions seem to be at the same level (Fig. 2, bar 15
compared with bar 16, bar 24 compared with bar
25), and thus can be served as a better index for
strain comparison. For strains with similar genetic
modifications, the industrial strains TMB3399 and
Figure 2. Comparison of anaerobic fermentation performance of engineered xylose-utilizing S. cerevisiae strains listed in Table 1. Cells were grown in medi-
um containing xylose or glucose under aerobic condition. Certain amounts of the pre-culture were inoculated into different types of medium containing
various concentration of xylose as the sole carbon source. Cells were cultivated in sealed bottles with no head space, or in vessels continuously flushed
with nitrogen gas, to ensure an anaerobic environment. (A) The yields of ethanol and xylitol. (B) The productivities of ethanol and xylitol as well as xylose
consumption rate per gram of cell mass. Lab. strain, laboratory strain; Ind. strain, industrial strain; XR, xylose reductase; XDH, xylitol dehydrogenase; XI,
xylose isomerase; YEthanol, ethanol yield; YXylitol, xylitol yield; PEthanol, specific ethanol productivity; PXylitol, specific xylitol productivity; CXylose, specific xylose
consumption rate; DW, dry weight.

© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 37
Biotechnol. J. 2012, 7, 34–46 www.biotechnology-journal.com
Table 1. Genetic modifications and anaerobic fermentation conditions of engineered xylose-utilizing S. cerevisiae strains
No. Strain Original strain Genotype Anaerobic fermentation conditions Ref.
Medium Xylose Cell massa) Time
(g/L) (g DW/L) (h)
1 H1691 CEN.PK 2 PsXRG, PsXDHG, ScXKPMinimal 50 5 88 [8]
2 H2674 CEN.PK 2 PsXRG, PsXDHG, ScXKGMinimal 50 3.8 120 [46]
3 H2673 CEN.PK 2 PsXRG, PsXDHG, ScXKG, GDP1PMinimal 50 3 120 [46]
4 H2723 CEN.PK 2 PsXRG, PsXDHG, ScXKG, ΔZWF1 Minimal 50 4.2 120 [46]
5 H2684 CEN.PK 2 PsXRG, PsXDHG, ScXKG, GDP1P, ΔZWF1 Minimal 50 1.9 120 [46]
6 TMB3001 CEN.PK 113-7A PsXRG, PsXDHG, ScXKGMineral 50 5 70 [45]
7 TMB3255 TMB3001 PsXRG, PsXDHG, ScXKG, ΔZWF1 Mineral 50 5 70 [45]
8 TMB3008 CEN.PK 2-1C PsXRG, PsXDHG, ScXKG, ΔGND1 Mineral 50 5 70 [45]
9 TMB3057 CEN.PK 2-1C PsXRP, PsXDHP, ScXKG, TAL1G, TKL1G, Mineral 50 5 67 [26]
RKI1G,RPE1G, ΔGRE3
10 TMB3413 CEN.PK 2-1C PsXRP, PsXDHP, ScXKG, TAL1G, TKL1G, Mineral 10 2 118 [41]
RKI1G, RPE1G, ΔGRE3
11 TMB3411 CEN.PK 2-1C PsXRP, PsXDHP, ScXKG, TAL1G, TKL1G, Mineral 10 2 118 [41]
RKI1G, RPE1G, GXF1G, ΔGRE3
12 TMB3001 C1 TMB3001 PsXRG, PsXDHG, ScXKG, mutation-selection Minimal 10 N. R. N. R. [49]
13 1400 Saccharomyces PsXRP, PsXDHP, ScXKPComplex 50 30 47 [59]
(pLNH32) sp. 1400
14 TMB3399 USM21 PsXRG, PsXDHG, ScXKGMineral 20 3.7 18 [60]
15 TMB3400 TMB3399 PsXRG, PsXDHG, ScXKG, mutation-selection Mineral 20 3 17 [60]
16 TMB3400 TMB3399 PsXRG, PsXDHG, ScXKG, mutation-selection Mineral 50 5 96 [26]
17 MA-R4 IR-2 PsXRG, PsXDHG, ScXKGComplex 45 2.8 48 [13,14]
18 MA-R5 IR-2 PsXRG, PsXDH (D207A/I208R/F209S/ Complex 45 2.8 48 [14]
N211R)G, ScXKG
19 TMB3102 CEN.PK 2-1C TtXIP, ΔGRE3 Mineral 50 10 70 [61]
20 TMB3104 CEN.PK 2-1C TtXIP, ScXKP, ΔGRE3 Mineral 50 10 70 [61]
21 INVSc1/ INVSc1 OsXIPMinimal 50 5 140 [62]
pRS406/
pWOXYLA
22 INVSc1/ INVSc1 OsXIP, ScXKGMinimal 50 5 140 [62]
pRS406XKS/
pRS405/
pWOXYLA
23 INVSc1/ INVSc1 OsXIP, ScXKG, SUT1GMinimal 50 5 140 [62]
pRS406XKS/
pILSUT1/
pWOXYLA
24 ADAP8 INVSc1/pRS406XKS/ OsXIP, ScXKG, SUT1G, xylose adaptation Minimal 20 5 188 [63]
pILSUT1/pWOXYLA
25 ADAP8 INVSc1/pRS406XKS/ OsXIP, ScXKG, SUT1G, xylose adaptation Complex 20 5 116 [63]
pILSUT1/pWOXYLA
26 TMB3066 CEN.PK 2-1C PisXIP, ScXKG, TAL1G, TKL1G, RKI1G,RPE1G, Mineral 50 5 67 [26]
ΔGRE3
27 RW202-AFX CEN.PK 113-5D PisXIP, xylose adaptation Mineral 20 1.8 114 [11]
28 RWB217 CEN.PK 102-3A PisXIP, ScXKP, TAL1G, TKL1G, RKI1G,RPE1G, Mineral 20 1.7 44 [64]
ΔGRE3
29 RWB218 RWB217 PisXIP, ScXKP, TAL1G, TKL1G, RKI1G, RPE1G, Mineral 20 1.85 24 [65]
ΔGRE3, glucose and xylose adaptation
30 BWY10Xyl BarraGrande CpXIP, xylose adaptation Mineral 25 3.2 138 [23]
a) For srtains No. 2–5, 14, 15, 27–30, the final cell mass after anaerobic cultivation were listed. For the rest of the strains, the initial cell mass were listed as their final
values were not reported in the literature. XR, xylose reductase; XDH, xylitol dehydrogenase; XI, xylose isomerase. The superscript letters P or G after a gene name
indicate that the gene was inserted into the plasmid or genome, respectively. CpXI, xylose isomerase from Clostridium phytofermentans; GDP1, glyceraldehyde-3-phos-
phate dehydrogenase from Kluyveromyces lactis; GFX1, hexose transporter from Candida intermedia; GND1, 6-phosphogluconate dehydrogenase from S. cerevisiae;
GRE3, aldose reductase from S. cerevisiae; OsXI, xylose isomerase from Orpinomyces sp.; PisXI, xylose isomerase from Piromyces sp.; PsXDH, xylitol dehydrogenase
from P. stipitis; PsXR, xylose reductase from P. stipitis; RKI1, RPI from S. cerevisiae; RPE1, ribulose-5-phosphate 3-epimerase from S. cerevisiae; ScXK, xylulokinase from
S. cerevisiae; SUT1, hexose transporter from P. stipitis; TAL1, transaldolase from S. cerevisiae; TKL1, transketolase from S. cerevisiae; TtXI, xylose isomerase from
Thermus thermophilus; ZWF1, glucose-6-phosphate dehydrogenase from S. cerevisiae; DW, dry weight. Minimal media [66]. N. R., not reported.

Biotechnology
Journal
Biotechnol. J. 2012, 7, 34–46
38 © 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
MA-R4 produced 0.26–15.3 g/L ethanol within
18–48 h after inoculation with 3.7−2.8 g/L cells,with
ethanol productivities of 0.004–0.114 g/h/g dry
weight (DW) of cell (Fig. 2, bars 14 and 17), while
the laboratory strains H2674 and TMB3001 exhib-
ited ethanol productivities of 0.009–0.041 g/h/g DW
of cell (Fig. 2, bars 2 and 6). From these data, it is
difficult to conclude whether the industrial strains
are superior over the laboratory ones or vice versa.
Despite the incomparable anaerobic xylose-fer-
menting capabilities, many industrial strains ex-
hibited better tolerance to the inhibitors present in
lignocellulosic hydrolysates than the laboratory
ones. Strains TMB3099/3400 could tolerate up to
15% of lignocellulosic hydrolysates, while the max-
imal hydrolysates concentration for several labora-
tory strains was only 10% [12]. Fermentation of lig-
nocellulosic hydrolysates by strains MA-R4/R5
could proceed without any obvious inhibition, at an
ethanol yield of 0.48 g/g total sugar (93% of the the-
oretical yield) [13, 14].As the required inhibitor tol-
erance is often possessed by industrial strains but
hard to be achieved by laboratory strains [12, 15,
16], the industrial S. cerevisiae strains would theo-
retically be a better start-point for engineering for
lignocellulosic ethanol production [5, 17].
The anaerobic fermentation capabilities of all
the engineered industrial S. cerevisiae strains list-
ed in Table 1 are quite uneven, with a maximal
30-fold difference in specific ethanol productivity
(Fig. 2, bars 13–18 and 30), demonstrating the im-
portance of selecting a good start strain. Some
studies in comparison and evaluation of various in-
dustrial and commercial S. cerevisiae strains have
been reported, but most of them pertained to ap-
plication in baking industry [18, 19]. Recently,
Almeida et al. [20] used a facile microplate screen-
ing method to assess anaerobic growth of 12 S. cere-
visiae strains (four industrial isolates, seven com-
mercial baker’s yeasts, and one commercial brew-
er’s yeast) in three different lignocellulose hy-
drolysates, and found significant differences in
inhibitor tolerance among the strains. This work
provided insights into selecting a good candidate
for engineering xylose fermentation capability;
further efforts on collecting and evaluating more
industrial S. cerevisiae strains should thus be em-
phasized.
Limited genetic tools available for industrial
S. cerevisiae strains also hamper their engineering.
Because of the non-auxotrophic background and
plasmid instability in industrial S. cerevisiae strains
[21, 22], approaches to introduce and delete genes
are restricted to chromosomal integration and
deletion. Reported usable selection markers for in-
dustrial S. cerevisiae strains include G418 [23],
zeocin [24], aureobasidin A [13], and cycloheximide
[25]; however, only G418 and elevated concentra-
tions of cycloheximide (100 mg/L) can be used for
an industrial S. cerevisiae strain in our lab (Bo
Zhang and Yin Li, unpublished data), suggesting
that suitable selection markers and their concen-
trations should be tested for each industrial strain.
Thus,it is of great importance to develop new func-
tional selection markers for industrial S. cerevisiae
strains so as to expand their accessibility for genet-
ic manipulation.
3 Which pathway is more promising,
XR-XDH or XI?
For the two isogenic laboratory strains carrying
XR-XDH- or XI-pathway under identical fermen-
tation conditions, the XI-carrying strain TMB3066
showed a lower xylitol yield (0.2-fold) and a higher
ethanol yield (1.3-fold) than those of XR-XDH-car-
rying strain TMB3057,but the specific ethanol pro-
ductivity of TMB3066 was only a half of TMB3057
due to its low specific xylose consumption rate
(Fig. 2, bar 26 compared with bar 9). The lower xyl-
itol yield observed in the XI-expressing strain is
reasonable as XI directly converts xylose to xylu-
lose, therefore bypasses xylitol production. Thus,
the ethanol yield in the XI-expressing strain in-
creased since less xylitol is produced and more car-
bon flow is directed to ethanol. Moreover, an over-
all lower xylitol yield and higher ethanol yield
could be found for all the XI-expressing strains
(Fig. 2A, bars 19–30 compared with bars 1–18).
However, it is hard to explain why the XI-express-
ing strain TMB3066 has a lower xylose fermenta-
tion capability in terms of specific xylose consump-
tion rate and ethanol productivity, since the XI ac-
tivity in TMB3066 crude cell extract (0.82 U/mg to-
tal protein) is even higher than the XR activity in
TMB3057 cell extract (0.35 U/mg total protein) [26].
Therefore, a more systematic investigation should
be conducted to identify the intrinsic reasons for
the observed weak xylose consumption and
ethanol production in this XI-expressing strain.
Theoretically, introduction of XI and deletion of
endogenous aldose reductase (AR, encoded by
GRE3 gene) can completely skip xylitol production
and convert all xylose to xylulose (Fig. 1). In con-
trast, strains TMB3102 and TMB3104 produced xyl-
itol at a yield of 0.31 g/g xylose, which was almost
the same level as all the XR-XDH-expressing lab-
oratory strains did (Fig. 2A, bars 19–20 compared
with bars 1–13). The intrinsic reasons are still un-
known. A thorough comparison of the ethanol and
xylitol yields of all the XI-expressing laboratory

© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 39
Biotechnol. J. 2012, 7, 34–46 www.biotechnology-journal.com
strains listed in Fig. 2A reveals an interesting phe-
nomenon: strains using Thermus thermophilus XI
showed the lowest ethanol yield of 0–0.21 g/g xy-
lose and the highest xylitol yield of 0.31 g/g xylose
(Fig. 2A, bars 19–20), both the ethanol and xylitol
yields for strains expressing Orpinomyces sp. XI
belonged to the middle range (0.31–0.4 g ethanol/g
xylose, 0.07–0.26 g xylitol/g xylose, Fig. 2A, bars
21–25), while strains carrying the XI from Piro-
myces sp. exhibited the highest ethanol yield and
lowest xylitol yield (0.41–0.43 g ethanol/g xylose,
0.002–0.04 g/g xylose, Fig. 2A, bars 26–29). The ac-
tivities for XIs from Piromyces sp. [26] and T. ther-
mophilus [5] are reported to be relatively high and
low, suggesting that the lower xylitol yield and
higher ethanol yield, which are remarkable virtues
of the XI pathway, are only true for highly active
XIs.
Although most of the XR-XDH-carrying S. cere-
visiae strains produced considerable amount of
xylitol at a high yield (Fig. 2A, bars 1–6 and 8–16),
some recent studies pointed out that xylitol forma-
tion varied with different fermentation media. In-
terestingly, fermentation with lignocellulose hy-
drolysates by XR-XDH-carrying strains always led
to no xylitol formation [26–29]. For example, the in-
dustrial strain TMB3400 carrying XR-XDH showed
a xylitol yield of 0.29 g/g xylose in mineral medium
(Fig. 2A, bar 16), whereas all the xylose was con-
sumed to ethanol without detection of xylitol and
other metabolites when lignocellulose hydrolys-
ates was used as medium [26]. One possible expla-
nation is that some external electron acceptors
such as furfural presented in the undetoxified hy-
drolysates may reduce xylitol production, because
such reagents can oxidize NADH to NAD+and thus
increase the supply of NAD+required for XDH
[30–32]. Therefore, the observed xylitol accumula-
tion in XR-XDH-expressing strains in laboratory
medium might not be so serious when they are
grown in lignocellulose hydrolysates.
The XI pathway to convert xylose to xylulose
has been considered as a promising approach for
xylose utilization in S. cerevisiae due to its one-step
conversion, cofactor-independent catalysis, and a
lack of xylitol production. However, only few het-
erologous XIs can be functionally expressed in
S. cerevisiae [5], and all the currently reported
XI-carrying strains employed strong promoters
and multicopy plasmids for its expression due to
the low activity of XI (Fig. 2, bars 19–30). Moreover,
when the XI from Piromyces sp., which has been
considered to have the highest activity among the
presently used XIs, was chromosomally integrated
into the parent strain TMB3066, the resulted strain
was unable to grow on xylose aerobically [26].
However, chromosomally integrated XR-XDH
pathway was sufficient to enable ethanol produc-
tion from xylose under anaerobic conditions (Fig.2,
bars 1–8, 12,14–18). Since chromosomal integration
is a prerequisite for strains to be used in industrial
production, XR-XDH pathway is still a better
choice unless the chromosomally integrated XI
could exhibit sufficient activity for xylose utiliza-
tion.
4 Is transport really a rate-limiting step
for xylose utilization?
Since the native S. cerevisiae strains take up xylose
through the family of hexose transporters (HXT),
which have a much higher affinity for glucose than
xylose [33, 34], much efforts have been made to
over-express new xylose transporters in S. cerevisi-
ae, with the expectation that higher xylose-uptake
activity will enhance the intracellular xylose me-
tabolism flux and eventually increase the ethanol
production. To date, both homologous (e.g.,
Hxt1/2/4/5/7and Gal2) and heterologous trans-
porters (e.g., SUT1 from P. stipitis,GXS1 and GXF1
from Candida intermedia, At5g59250 from Ara-
bidopsis thaliana) have been shown to enable the
HXT-null S. cerevisiae strains to transport xylose
into the cell [34–38]. Unlike other transporters that
transport sugars via facilitated diffusion, GXS1 is a
xylose/glucose–H+symporter. It may be less appli-
cable for anaerobic xylose fermentation as proton
symport requires more energy than facilitated dif-
fusion does, while limited amount of ATP is pro-
duced under anaerobic circumstance. The kinetic
parameters of xylose- and glucose-uptake for some
transporters described above were summarized in
Table 2. Compared with the homologous HXT, het-
erologous transporters showed lower maximal up-
take activities, but higher affinities to xylose. How-
ever, the trait of preference for glucose over xylose
has not been changed among all the heterologous
transporters.
Over-expression of xylose transporter indeed
enabled the recipient S. cerevisiae to transport xy-
lose, but no significant effects on improving cell
growth on xylose or specific ethanol productivity in
xylose-utilizing strains were observed [34, 35, 39].
For the two heterologous transporters GXF1 and
SUT1, their over-expression in xylose-utilizing
strains slightly increased the xylitol yield and re-
duced the ethanol yield, while the specific ethanol
productivity was nearly unchanged upon GXF1 ex-
pression, and increased by 0.5-fold upon SUT1 ex-
pression (Fig. 2, bar 11 compared with bar 10 for
GXF1; bar 23 compared with bar 22 for SUT1). The

Biotechnology
Journal
Biotechnol. J. 2012, 7, 34–46
40 © 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
negative effects of expressing a strong xylose
transporter on the xylitol and ethanol yields can be
ascribed to the insufficient xylose metabolism,
leading to the accumulation of xylitol. Whether ex-
pressing such a strong xylose transporter is of ben-
efit for ethanol production is dependent on the rel-
ative rates between xylose uptake and metabolism.
Only when xylose-uptake is the rate-determining
step during the entire xylose utilization process,
will the expression of a xylose transporter increase
the ethanol production, otherwise the portion of
excessively transported xylose will be converted to
xylitol rather than ethanol, as shown in the GFX1
example described above.
For the currently constructed xylose-utilizing
S. cerevisiae strains, which usually have poor xy-
lose-metabolizing ability, transport-controlled xy-
lose utilization only occurs at very low xylose con-
centration (<0.5–4 g/L) [40, 41]. Moreover, even the
HXT in the native S. cerevisiae themselves can take
up xylose at a rate of 0.14 g/h/g DW in the presence
of 50 mM xylose (7.5 g/L) [35], which exceeds most
of the specific xylose productivity listed in Fig. 2.
Therefore, the importance of increasing xylose
transportation in developing an efficient xylose-
utilization S. cerevisiae strain might be overrated,
and currently, the bottleneck remains the low effi-
ciency of intracellular xylose metabolism.
5 Is redox imbalance a stumbling block
for anaerobic xylose fermentation?
Because the P. stipitis XR prefers NADPH while the
xylose dehydrogenase from the same strain only
produces NADH, the notorious redox imbalance in
XR-XDH-expression S. cerevisiae strains is be-
lieved to be the reason for xylitol excretion and low
ethanol yield. So far, extensive studies have been
conducted to address this issue, and the schematic
diagrams of several different strategies are as
shown in Fig. 3.
Strategy I and II aim to alter the cofactor speci-
ficities of XR and XDH from P. stipitis. In Strategy
I, the preference of XR for NADPH was changed to
NADH. Then, the cofactors needed by the intro-
duced XR-XDH pathway could be self-supplied
and recycled. Numerous XRs with higher affinities
to NADH than NADPH (i.e., 6- to 146-fold in-
creased ratio of NADH/NADPH in catalytic effi-
ciency [(kcat/Km)NADH/(kcat/Km)NADPH]) have been
created by protein engineering, but these increased
ratios of NADH/NADPH catalytic efficiency main-
ly came from the decrease in catalytic efficiency
with NADPH, and the highest increase in catalytic
efficiency with NADH was just 1.8-fold [42, 43].
When the wild-type XR in xylose-utilizing strains
was replaced with these mutants, only the strain
harboring the XR mutant with both increased
NADH catalytic efficiency and NADH/NADPH cat-
alytic efficiency ratio exhibited increased ethanol
production and decreased xylitol excretion, while
the strain harboring the XR mutant with decreased
NADH catalytic efficiency but increased NADH/
NADPH catalytic efficiency ratio showed reduced
ethanol production [43]. In Strategy II, the coen-
zyme specificity of XDH from NAD+to NADP+is
engineered. Engineering XDH is more difficult as it
almost uses NAD+exclusively. However, an XDH
mutant with complete reversal of coenzyme speci-
ficity toward NADP+and higher thermostability
has been generated by multiple site-directed mu-
tagenesis [44]. The industrial S. cerevisiae strain
MA-R5 with this XDH mutant expressed in the
chromosome, exhibited both higher ethanol and
xylitol yield and productivity than the parent strain
MA-R4 with the wild-type XDH (Fig.2, bar 18 com-
pared with bar 17).
Table 2. Kinetic parameters of xylose transporters toward xylose and glucose
Transporter Strain/plasmid Genotype of the strain Xylose Glucose Ref.
Vmax KmVmax Km
(g/h/g DW) (mM) (g/h/g DW) (mM)
HXT1 H2219/pYX212-HXT1 19.28 880 N. R. N. R. [34]
HXT2 H2219/pYX212-HXT2 MATa Δhxt1-7 Δgal2 leu2-3,112 MAL2 8.74 260 N. R. N. R. [34]
HXT4 H2219/pYX212-HXT4 SUC2 GAL MEL ura3::XR/XDH his3::XK 4.89 170 N. R. N. R. [34]
HXT7 H2219/pYX212-HXT7 2.82 130 N. R. N. R. [34]
SUT1 RE700/YEpSUT1 MATa ura3–52 his3–11,15 leu2–3112 MAL2 1.2 145 0.49 1.5 [36]
SUC2 GAL MEL, Δhxt1–7
GXS1 TMB3201/pHXT7-GXS1 MATa Δhxt1-17 Δgal2 Δstl1 Δagt1 Δmph2 0.06 0.4 0.05 0.012 [37]
GXF1 TMB3201/pGXF1 Δmph3 leu2-3,112 ura3-52 trp1-289 his3-Δ1: 1.65 48.7 0.27 2 [37]
:YIpXR/XDH/XK MAL2-8cSUC2
DW, dry weight; N. R., not reported.

© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 41
Biotechnol. J. 2012, 7, 34–46 www.biotechnology-journal.com
Unlike the above two strategies, Strategy III and
IV adjust the redox balance through regulating the
central pathways. Strategy III lowers the NADPH
supply for XR by disrupting genes in the NADPH-
producing oxidative ppp. Obvious higher ethanol
yields and lower xylitol yields were observed for
strains with deletion of either the glucose-6-phos-
phate dehydrogenase (GPD6, encoded by ZWF1
gene) or the 6-phosphogluconate dehydrogenase
(PGND,encoded by GND1 gene) gene (Fig.2A, bar 4
compared with bar 2; bars 7 and 8 compared with
bar 6). However, for ZWF1 or GND1 gene deletion in
strains TMB3225 and TMB3008, a remarkable re-
duction in specific ethanol productivity and a sig-
Figure 3. Strategies for relieving the redox imbalance in XR-XDH-expressing S. cerevisiae strains. ADH, alcohol dehydrogenase; ALD, aldehyde dehydroge-
nase; GAPD, glyceraldehyde-3-phosphate dehydrogenase; GAPD*, mutant of glyceraldehyde-3-phosphate dehydrogenase with NADP+preference; GPD6
(ZWF1), glucose-6-phosphate dehydrogenase encoded by ZWF1 gene; PGND (GND1), 6-phosphogluconate dehydrogenase encoded by GND1 gene; RPE,
ribulose-5-phosphate 3-epimerase; TKL, transketolase; XDH, xylitol dehydrogenase; XDH*, mutant of xylitol dehydrogenase with NADP+preference; XK,
xylulokinase; XR, xylose reductase; XR*, mutant of xylose reductase with NADH preference.

Biotechnology
Journal
Biotechnol. J. 2012, 7, 34–46
42 © 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
nificant increase in acetate formation were ob-
served (Fig. 2A, bars 7 and 8 compared with bar 6,
and [45]), which suggested that acetate production
from acetaldehyde became a new source of NADPH
supply for XR when the main NADPH-producing
oxidative ppp were blocked. In this case, however,
the NADH generated by XDH was still unbalanced.
Strategy IV constructed a completely new redox
balance by deletion of ZWF1 gene and over-expres-
sion of a NADP+-dependent glyceraldehyde-3-
phosphate dehydrogenase (GAPD, encoded by
GDP1 gene) [46, 47]. ZWF1 gene deletion cut off the
main NADPH supply from oxidative ppp, which
avoided wasteful CO2production. The NADPH re-
quired for XR were then supplied by GDP1 cat-
alyzed reaction, which converted 3-phosphoglycer-
aldehyde into 3-phosphoglycerate with the con-
comitant NADPH production. Besides, the NADH
produced by XDH and GAPD were consumed by al-
cohol dehydrogenase (ADH) in the last step of
ethanol formation. Thus, the intracellular NADPH
and NADH were balanced as a whole. Strain H2684
engineered in a similar manner to this showed a
significant higher ethanol yield and lower xylitol
yield (Fig. 2A, bar 5 compared with bar 2).
When reviewing all the efforts in relieving the
redox imbalance in XR-XDH-expressing strains,
we found that a balanced redox could not guaran-
tee the decrease of xylitol excretion, as seen by
strain MA-R5 compared with MA-R4 (Fig. 2, bar 18
compared with bar 17), and sometimes the strain
with a balanced redox still produced remarkable
amounts of xylitol, as seen by strain H2684 (Fig. 2,
bar 5). Moreover, among all of the engineered
strains with reduced redox imbalance, few have
significant increase in the eventual specific ethanol
productivity (Fig. 2B, bars 4 and 5 compared with
bar 2; bars 7 and 8 compared with bar 6;bar 18 com-
pared with bar 17). Therefore, some unknown rea-
sons other than redox imbalance might be respon-
sible for the high xylitol yield and low specific
ethanol productivity in the anaerobic xylose fer-
mentation by XR-XDH-expressing S. cerevisiae
strains. Other efforts such as developing a quanti-
tative metabolic and regulatory network and taking
into account the entire cellular redox metabolism
should be considered to address this issue.
6 Is evolutionary engineering effective in
promoting anaerobic xylose fermentation
capability?
Combined with the traditional metabolic engineer-
ing, the so-called “evolutionary engineering”, which
follows nature’s “engineering” principle by variation
and selection [48] has also been adopted to develop
an anaerobic xylose-utilizing S. cerevisiae strain.
Six strains in Fig. 2 were generated by evolution-
ary engineering approach, including the XR-XDH-
carrying laboratory and industrial strains
(TMB3001C1, TMB3400), as well as the XI-carrying
laboratory and industrial ones (ADAP8, RWB202-
AFX, RWB218, BWY10Xyl). Detailed evolution
processes of the six strains were summarized in
Table 3. Since direct selection of S. cerevisiae strains
capable of anaerobic growth on xylose as the sole
carbon source was unsuccessful [49], the evolution
strategies had to be changed to select the one which
can grow on xylose aerobically (TMB3400, ADAP8,
BWY10Xyl), or select the one with anaerobic
growth ability on xylose and glucose (RWB218), or
step-wisely select for aerobic, micro-aerobic, and
then anaerobic xylose growth (TMB3001C1,
RWB202-AFX). Judging by the anaerobic xylose
fermentation performances of these engineered
strains,only the evolution of TMB3400 and RWB218
were successful since they indeed exhibited higher
specific ethanol productivity than the parent strains
(Fig. 2B, bar 15 compared with bar 14 for TMB3400;
bar 29 compared with bar 28 for RWB218), whereas
the evolution of ADAP8 gave a negative effect on
the anaerobic specific ethanol productivity (Fig. 2B,
bar 24 compared with bar 23).
As shown in Table 3, all the evolved strains
showed a considerable improvement in terms of
anaerobic or aerobic xylose growth rate after cor-
responding anaerobic or aerobic growth selection,
which is in accordance with the fundamental rule
of evolutionary engineering: you get what you
screen for. Thus, selection from aerobic xylose
growth might not improve the anaerobic ethanol
productivity, similar to the example of strain
ADAP8. However, even for anaerobic growth selec-
tion, there is still not an inevitable link between a
better xylose growth and a higher ethanol produc-
tivity. Taking the anaerobically evolved strain
TMB3001C1 as an example, it consumed two-fold
more xylose, but produced the same amount of
ethanol as compared with its parent strain
TMB3001 in a steady-state chemostat with a dilu-
tion rate of 0.05 h–1 [50]. Therefore, better anaero-
bic growth on xylose cannot guarantee ethanol
production from the utilized xylose.
The success of evolutionary engineering de-
pends on the effectiveness of the selection method
as mutations are randomly created and one cannot
determine which mutations are beneficial before
selection. To date, a sensitive and high-throughput
method capable of directly selecting strains show-
ing higher ethanol productivity is not yet available.
Future progress on designing such selection meth-

© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 43
Biotechnol. J. 2012, 7, 34–46 www.biotechnology-journal.com
ods would greatly speed up the construction of an
efficient anaerobic xylose-fermenting S. cerevisiae
strains. Moreover, how to avoid loss of other exist-
ing good capabilities during evolution is another
challenge for evolutionary engineering. For in-
stance, strain TMB3001C1 lost 50% of its aerobic
glucose growth rate when gaining the ability of
anaerobic growth on xylose after evolution [49].
7 Concluding remarks
In an effort to develop a S. cerevisiae strain capable
of anaerobically fermenting xylose with high
ethanol yield and productivity, extensive studies on
metabolic and evolutionary engineering have been
performed, and moderate improvements have been
obtained, as summarized in Fig. 2. Although the in-
tact xylose metabolic pathway has been construct-
Table 3. Summary of evolutionary engineering of S. cerevisiae strains for anaerobic xylose utilization
Evolved strain Start strain Mutagenesis Selection strategy Selection Xylose growth Ref.
time rate (evolved/
start strain)
TMB3001C1 TMB3001 EMS (1) Chemostat at a dilution rate of 0.05 h–1, 266 days 0.012/0 h–1 [49]
(Lab., XR-XDH) minimal medium with 5 g/L xylose and (anaerobic)
1 g/L glucose, aerobic, 90 generations
(2) Chemostat at a dilution rate of 0.05 h–1,
minimal medium with 5 g/L xylose, aerobic
for 60 generations, micro-aerobic for
120 generations, anaerobic for 40 generations,
and anaerobic for 150 generations
TMB3400 TMB3399 EMS Batch, one contained YPX medium with 10 days 0.14/0.0255 h–1 [60]
(Ind., XR-XDH) 50 g/L xylose, the other contained YPX medium (aerobic)
with step-wise increased xylose concentration
(from 2 g/L xylose and 20 g/L glucose at first to
50 g/L xylose at last, change every 48 h), aerobic
ADAP8 INVSc1/ Spontaneous Batch, synthetic minimal medium with 20 g/L 8 transfers 0.133/0.025 h–1 [63]
(Lab., XI) pRS406XKS/ mutation of xylose, aerobic, 8 transfers (aerobic)
pILSUT1/
pWOXYLA
RWB202-AFX RWB202 Spontaneous (1) Batch, synthetic medium with xylose, 158 days 0.03/0.005 h–1 [11]
(Lab., XI) mutation aerobic, 30 transfers (anaerobic)
(2) Batch, synthetic medium with xylose,
oxygen limitation, 1:10 transfer when xylose
was depleted, 10 cycles
(3) Batch, synthetic medium with xylose,
anaerobic, 10 cycles
RWB218 RWB217 Spontaneous (1) Chemostat at a dilution rate of 0.06 h–1, 2100 h 0.12/0.09 h–1 [65]
(Lab., XI) mutation defined mineral medium with 30 g/L xylose, (anaerobic)
anaerobic, 1000 h
(2) Batch, defined mineral medium with
20 g/L glucose and 20 g/L xylose, 1:200 transfer,
anaerobic, 20 h, 25 cycles
(3) Batch, defined mineral medium with
20 g/L glucose and 20 g/L xylose for the first 20 h,
add 70 g/L xylose for the rest 10 h, 1:200 transfer,
anaerobic, 20 cycles
BWY10Xyl BarraGrande/ Spontaneous (1) Batch, synthetic medium with 20 g/l D-xylose, 42 days 0.04/0 h–1 [23]
(Ind., XI) YEp-opt. mutation 1 g/L yeast extract, 2 g/L peptone, and (aerobic)
XI-Clos-K 200 mg/L G418, aerobic, 4 transfers
(2) Batch, synthetic medium with 20 g/L D-xylose
and 200 mg/L G418, aerobic, 2 transfers
Lab., laboratory strain; Ind., industrial strain; XR, xylose reductase; XDH, xylitol dehydrogenase; XI, xylose isomerase; EMS, ethyl methane sulfonate; YPX medium,
10 g/L yeast extract, 20 g/L peptone, and 20 g/L xylose.

Biotechnology
Journal
Biotechnol. J. 2012, 7, 34–46
44 © 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ed and some known defects (e.g., redox imbalance,
weak xylose transportation) have been partially
improved, the rate and yield of anaerobic ethanol
production from xylose are considerably lower
than those from glucose, which implies our current
knowledge on intracellular xylose metabolic and
regulatory network might be inadequate. It is note-
worthy that the desirable glucose fermentation ca-
pability of S. cerevisiae is the consequence of their
long-term association with human-induced fer-
mentation [51]. Keeping this in mind, it is not im-
possible to realize anaerobic xylose fermentation
as efficient as that of glucose, and what we need
probably is only time and further understanding.
Nowadays most of the engineered S. cerevisiae
strains employs the three-step xylose-utilizing
pathway: (i) conversion of xylose to xylulose via the
XR-XDH or XI pathway; (ii) conversion of xylulose
to glyceraldehydes-3-phosphate through the en-
dogenous XK and ppp; and (iii) conversion of glyc-
eraldehydes-3-phosphate to ethanol by glycolysis.
Inspired by the recent progress in butanol pro-
duction in engineered Escherichia coli, which dra-
matically increased the titer of butanol from
0.5–1.22 g/L [52, 53] to 4.7 g/L [54] by replacing the
native butanol synthetic pathway in C. aceto-
butylicum with a chimeric pathway assembled from
three different organisms, a new or combined
metabolic pathway for xylose may break the exist-
ing bottleneck for xylose fermentation.Thus, it is of
great importance to expand our knowledge on xy-
lose-utilizing pathways by isolation and character-
ization of novel xylose-fermenting strains, such as
the work done by Suh et al. [55–58].
This work was supported by the Knowledge Innova-
tion Program of the Chinese Academy of Sciences
(KSCX1-YW-11C3).Yin Li is supported by the Hun-
dreds of Talents Program of the Chinese Academy of
Sciences.
The authors declare no conflict of interest.
8 References
[1] Olsson, L., Nielsen, J., The role of metabolic engineering in
the improvement of Saccharomyces cerevisiae: Utilization
of industrial media. Enzyme Microb.Technol. 2000, 26, 785–
792.
[2] Pettersen, R.,The chemical composition of wood.Adv. Chem.
Ser. 1984, 57–126.
[3] Carroll, A.,Somerville, C.,Cellulosic biofuels. Ann. Rev. Plant
Biol. 2009, 60, 165–182.
[4] Jeffries, T. W., Engineering yeasts for xylose metabolism.
Curr. Opin. Biotechnol. 2006, 17, 320–326.
[5] Matsushika, A., Inoue, H., Kodaki, T., Sawayama, S., Ethanol
production from xylose in engineered Saccharomyces cere-
visiae strains: Current state and perspectives. Appl. Micro-
biol. Biotechnol. 2009, 84, 37–53.
[6] Jeffries, T. W., Jin,Y. S., Metabolic engineering for improved
fermentation of pentoses by yeasts.Appl. Microbiol. Biotech-
nol. 2004, 63, 495–509.
[7] Hahn-Hagerdal, B., Karhumaa, K., Jeppsson, M., Gorwa-
Grauslund, M. F., Metabolic engineering for pentose utiliza-
tion in Saccharomyces cerevisiae.Adv. Biochem. Eng. Bio-
technol. 2007, 108, 147–177.
[8] Toivari, M. H.,Aristidou, A., Ruohonen, L., Penttila, M., Con-
version of xylose to ethanol by recombinant Saccharomyces
cerevisiae: Importance of xylulokinase (XKS1) and oxygen
availability. Metab. Eng. 2001, 3, 236–249.
[9] Jin, Y. S., Jones, S.,Shi, N. Q., Jeffries, T. W., Molecular cloning
of XYL3 (D-XK) from Pichia stipitis and characterization of
its physiological function. Appl. Environ.Microbiol. 2002, 68,
1232–1239.
[10] Ni, H., Laplaza, J. M., Jeffries, T.W., Transposon mutagenesis
to improve the growth of recombinant Saccharomyces cere-
visiae on D-xylose. Appl. Environ. Microbiol. 2007, 73, 2061–
2066.
[11] Kuyper, M., Winkler,A. A., van Dijken, J. P., Pronk, J.T., Min-
imal metabolic engineering of Saccharomyces cerevisiae
for efficient anaerobic xylose fermentation: A proof of prin-
ciple. FEMS Yeast Res. 2004, 4, 655–664.
[12] Sonderegger, M., Jeppsson, M., Larsson, C., Gorwa-Graus-
lund, M. F. et al., Fermentation performance of engineered
and evolved xylose-fermenting Saccharomyces cerevisiae
strains. Biotechnol. Bioeng. 2004, 87, 90–98.
[13] Matsushika, A.,Inoue, H., Murakami, K.,Takimura, O. et al.,
Bioethanol production performance of five recombinant
strains of laboratory and industrial xylose-fermenting Sac-
charomyces cerevisiae. Bioresour. Technol. 2009, 100, 2392–
2398.
[14] Matsushika, A., Inoue, H.,Watanabe, S., Kodaki, T. et al., Ef-
ficient bioethanol production by a recombinant flocculent
Saccharomyces cerevisiae strain with a genome-integrated
NADP+-dependent xylitol dehydrogenase gene. Appl. Envi-
ron. Microbiol. 2009, 75, 3818–3822.
[15] Garay-Arroyo,A., Covarrubias, A. A., Clark, I., Nino, I. et al.,
Response to different environmental stress conditions of
industrial and laboratory Saccharomyces cerevisiae strains.
Appl. Microbiol. Biotechnol. 2004, 63, 734–741.
[16] Nilsson, A., Gorwa-Grauslund, M. F., Hahn-Hagerdal, B., Li-
den, G., Cofactor dependence in furan reduction by Saccha-
romyces cerevisiae in fermentation of acid-hydrolyzed lig-
nocellulose. Appl. Environ. Microbiol. 2005, 71, 7866–7871.
[17] Hahn-Hagerdal, B., Karhumaa, K., Fonseca, C., Spencer-
Martins, I. et al., Towards industrial pentose-fermenting
yeast strains. Appl. Microbiol. Biotechnol. 2007, 74, 937–953.
[18] Bell, P. J., Higgins, V. J., Attfield, P. V., Comparison of fermen-
tative capacities of industrial baking and wild-type yeasts of
the species Saccharomyces cerevisiae in different sugar
media. Lett. Appl. Microbiol. 2001, 32, 224–229.
[19] Rollini, M., Casiraghi, E., Pagani, M. A., Manzoni, M., Tech-
nological performances of commercial yeast strains in dif-
ferent complex dough formulations. Eur. Food Res.Technol.
2007, 226, 19–24.
[20] Almeida, J. R., Karhumaa, K., Bengtsson, O., Gorwa-Graus-
lund, M. F., Screening of Saccharomyces cerevisiae strains
with respect to anaerobic growth in non-detoxified ligno-
cellulose hydrolysate. Bioresour. Technol. 2009, 100, 3674–
3677.

© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 45
Biotechnol. J. 2012, 7, 34–46 www.biotechnology-journal.com
[21] Futcher,A. B.,Cox, B. S., Copy number and the stability of 2-
micron circle-based artificial plasmids of Saccharomyces
cerevisiae. J. Bacteriol. 1984, 157, 283–290.
[22] Zhang, Z., Moo-Young, M.,Chisti,Y., Plasmid stability in re-
combinant Saccharomyces cerevisiae.Biotechnol. Adv. 1996,
14, 401–435.
[23] Brat, D., Boles, E.,Wiedemann, B., Functional expression of
a bacterial xylose isomerase in Saccharomyces cerevisiae.
Appl. Environ. Microbiol. 2009, 75, 2304–2311.
[24] Zaldivar, J., Borges, A., Johansson, B., Smits, H. P. et al., Fer-
mentation performance and intracellular metabolite pat-
terns in laboratory and industrial xylose-fermenting Sac-
charomyces cerevisiae.Appl. Microbiol. Biotechnol. 2002, 59,
436–442.
[25] Ganga, M.A., Pinaga, F., Valles, S., Ramon, D.,Querol, A.,Aro-
ma improving in microvinification processes by the use of a
recombinant wine yeast strain expressing the Aspergillus
nidulans xlnA gene. Int. J. Food Microbiol. 1999, 47, 171–178.
[26] Karhumaa, K., Garcia Sanchez, R., Hahn-Hagerdal, B., Gor-
wa-Grauslund, M. F., Comparison of the xylose reductase-
xylitol dehydrogenase and the xylose isomerase pathways
for xylose fermentation by recombinant Saccharomyces
cerevisiae. Microb. Cell Fact. 2007, 6,5.
[27] Ohgren, K., Bengtsson, O., Gorwa-Grauslund, M. F., Galbe,
M. et al., Simultaneous saccharification and co-fermenta-
tion of glucose and xylose in steam-pretreated corn stover
at high fiber content with Saccharomyces cerevisiae
TMB3400. J. Biotechnol. 2006, 126, 488–498.
[28] Sedlak, M., Ho, N.W., Production of ethanol from cellulosic
biomass hydrolysates using genetically engineered Saccha-
romyces yeast capable of cofermenting glucose and xylose.
Appl. Biochem. Biotechnol. 2004, 114, 403–416.
[29] Katahira, S.,Mizuike, A.,Fukuda, H., Kondo,A., Ethanol fer-
mentation from lignocellulosic hydrolysate by a recombi-
nant xylose- and cellooligosaccharide-assimilating yeast
strain. Appl. Microbiol. Biotechnol. 2006, 72, 1136–1143.
[30] Alexander, N. J., Acetone stimulation of ethanol production
from d-xylose by Pachysolen tannophilus. Appl. Microbiol.
Biotechnol. 1986, 25, 203–207.
[31] Ligthelm, M. E., Prior, B. A., du Preez, J. C., Effect of hydro-
gen acceptors on D-xylose fermentation by anaerobic cul-
ture of immobilized Pachysolen tannophilus cells. Biotech-
nol. Bioeng. 1989, 33, 839–944.
[32] Wahlbom, C. F., Hahn-Hagerdal, B., Furfural, 5-hydrox-
ymethyl furfural, and acetoin act as external electron ac-
ceptors during anaerobic fermentation of xylose in recom-
binant Saccharomyces cerevisiae. Biotechnol. Bioeng. 2002,
78, 172–178.
[33] Lee, W. J., Kim, M. D., Ryu, Y. W., Bisson, L. F., Seo, J. H., Ki-
netic studies on glucose and xylose transport in Saccha-
romyces cerevisiae. Appl. Microbiol. Biotechnol. 2002, 60,
186–191.
[34] Saloheimo,A., Rauta, J., Stasyk, O.V., Sibirny, A. A. et al., Xy-
lose transport studies with xylose-utilizing Saccharomyces
cerevisiae strains expressing heterologous and homologous
permeases. Appl. Microbiol. Biotechnol. 2007, 74, 1041–1052.
[35] Hamacher, T., Becker, J., Gardonyi, M., Hahn-Hagerdal, B.,
Boles, E.,Characterization of the xylose-transporting prop-
erties of yeast hexose transporters and their influence on
xylose utilization. Microbiology 2002, 148, 2783–2788.
[36] Weierstall,T., Hollenberg, C. P., Boles, E., Cloning and char-
acterization of three genes (SUT1-3) encoding glucose
transporters of the yeast Pichia stipitis.Mol. Microbiol. 1999,
31, 871–883.
[37] Leandro, M. J., Goncalves, P., Spencer-Martins, I., Two glu-
cose/xylose transporter genes from the yeast Candida in-
termedia: First molecular characterization of a yeast xylose-
H+ symporter. Biochem. J. 2006, 395, 543–549.
[38] Hector, R. E., Qureshi, N., Hughes, S. R., Cotta, M. A., Ex-
pression of a heterologous xylose transporter in a Saccha-
romyces cerevisiae strain engineered to utilize xylose im-
proves aerobic xylose consumption. Appl. Microbiol. Bio-
technol. 2008, 80, 675–684.
[39] Sedlak, M., Ho, N.W., Characterization of the effectiveness
of hexose transporters for transporting xylose during glu-
cose and xylose co-fermentation by a recombinant Saccha-
romyces yeast. Yeast 2004, 21, 671–684.
[40] Gardonyi, M., Jeppsson, M., Liden, G., Gorwa-Grauslund, M.
F., Hahn-Hagerdal,B., Control of xylose consumption by xy-
lose transport in recombinant Saccharomyces cerevisiae.
Biotechnol. Bioeng. 2003, 82, 818–824.
[41] Runquist, D., Fonseca, C., Radstrom, P., Spencer-Martins, I.,
Hahn-Hagerdal, B., Expression of the Gxf1 transporter
from Candida intermedia improves fermentation perform-
ance in recombinant xylose-utilizing Saccharomyces cere-
visiae. Appl. Microbiol. Biotechnol. 2009, 82, 123–130.
[42] Liang, L., Zhang, J., Lin, Z., Altering coenzyme specificity of
Pichia stipitis xylose reductase by the semi-rational ap-
proach CASTing. Microb. Cell Fact. 2007, 6, 36.
[43] Watanabe, S., Abu Saleh, A., Pack, S. P., Annaluru, N. et al.,
Ethanol production from xylose by recombinant Saccha-
romyces cerevisiae expressing protein-engineered NADH-
preferring xylose reductase from Pichia stipitis. Microbiol-
ogy 2007, 153, 3044–3054.
[44] Watanabe, S., Kodaki, T., Makino, K., Complete reversal of
coenzyme specificity of xylitol dehydrogenase and increase
of thermostability by the introduction of structural zinc.
J. Biol. Chem. 2005, 280, 10340–10349.
[45] Jeppsson, M., Johansson, B., Hahn-Hagerdal, B., Gorwa-
Grauslund, M. F., Reduced oxidative pentose phosphate
pathway flux in recombinant xylose-utilizing Saccha-
romyces cerevisiae strains improves the ethanol yield from
xylose. Appl. Environ. Microbiol. 2002, 68, 1604–1609.
[46] Verho, R., Londesborough, J., Penttila, M., Richard, P., Engi-
neering redox cofactor regeneration for improved pentose
fermentation in Saccharomyces cerevisiae. Appl. Environ.
Microbiol. 2003, 69, 5892–5897.
[47] Bro, C., Regenberg, B., Forster, J., Nielsen, J., In silico aided
metabolic engineering of Saccharomyces cerevisiae for im-
proved bioethanol production. Metab.Eng. 2006, 8, 102–111.
[48] Sauer, U., Evolutionary engineering of industrially impor-
tant microbial phenotypes. Adv. Biochem. Eng. Biotechnol.
2001, 73, 129–169.
[49] Sonderegger,M., Sauer, U.,Evolutionary engineering of Sac-
charomyces cerevisiae for anaerobic growth on xylose.
Appl. Environ. Microbiol. 2003, 69, 1990–1998.
[50] Sonderegger,M., Jeppsson, M., Hahn-Hagerdal, B.,Sauer, U.,
Molecular basis for anaerobic growth of Saccharomyces
cerevisiae on xylose,investigated by global gene expression
and metabolic flux analysis. Appl. Environ. Microbiol. 2004,
70, 2307–2317.
[51] Donalies, U. E., Nguyen, H. T., Stahl, U., Nevoigt, E., Im-
provement of Saccharomyces yeast strains used in brewing,
wine making and baking. Adv. Biochem. Eng. Biotechnol.
2008, 111, 67–98.
[52] Atsumi, S.,Cann, A. F.,Connor, M.R., Shen, C.R. et al.,Meta-
bolic engineering of Escherichia coli for 1-butanol produc-
tion. Metab. Eng. 2008, 10, 305–311.

Biotechnology
Journal
Biotechnol. J. 2012, 7, 34–46
46 © 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
[53] Inui, M., Suda, M., Kimura, S.,Yasuda, K. et al., Expression
of Clostridium acetobutylicum butanol synthetic genes in
Escherichia coli. Appl. Microbiol. Biotechnol. 2008, 77, 1305–
1316.
[54] Bond-Watts, B. B., Bellerose, R. J., Chang, M. C., Enzyme
mechanism as a kinetic control element for designing syn-
thetic biofuel pathways. Nat. Chem. Biol. 2011, 7, 222–227.
[55] Suh, S. O., McHugh, J. V., Pollock, D. D., Blackwell, M., The
beetle gut: A hyperdiverse source of novel yeasts. Mycol.
Res. 2005, 109, 261–265.
[56] Suh, S. O., White, M. M., Nguyen, N. H., Blackwell, M., The
status and characterization of Enteroramus dimorphus: A
xylose-fermenting yeast attached to the gut of beetles. My-
cologia 2004, 96, 756–760.
[57] Suh, S. O., McHugh, J. V., Blackwell, M., Expansion of the
Candida tanzawaensis yeast clade: 16 Novel Candida
species from basidiocarp-feeding beetles. Int. J. Syst. Evol.
Microbiol. 2004, 54, 2409–2429.
[58] Suh, S. O., Marshall, C. J., McHugh, J. V., Blackwell, M.,Wood
ingestion by passalid beetles in the presence of xylose-fer-
menting gut yeasts. Mol. Ecol. 2003, 12, 3137–3145.
[59] Ho, N. W., Chen, Z., Brainard, A. P., Genetically engineered
Saccharomyces yeast capable of effective cofermentation of
glucose and xylose. Appl. Environ. Microbiol. 1998, 64, 1852–
1859.
[60] Wahlbom,C. F., van Zyl,W.H., Jonsson, L.J., Hahn-Hagerdal,
B. et al., Generation of the improved recombinant xylose-
utilizing Saccharomyces cerevisiae TMB 3400 by random
mutagenesis and physiological comparison with Pichia
stipitis CBS 6054. FEMS Yeast Res. 2003, 3, 319–326.
[61] Traff, K. L., Otero Cordero, R. R., van Zyl, W. H., Hahn-
Hagerdal, B., Deletion of the GRE3 aldose reductase gene
and its influence on xylose metabolism in recombinant
strains of Saccharomyces cerevisiae expressing the xylA
and XKS1 genes. Appl. Environ. Microbiol. 2001, 67, 5668–
5674.
[62] Madhavan, A., Tamalampudi, S., Ushida, K., Kanai, D. et al.,
Xylose isomerase from polycentric fungus Orpinomyces:
Gene sequencing, cloning, and expression in Saccha-
romyces cerevisiae for bioconversion of xylose to ethanol.
Appl. Microbiol. Biotechnol. 2009, 82, 1067–1078.
[63] Madhavan,A., Tamalampudi,S., Srivastava,A., Fukuda, H. et
al., Alcoholic fermentation of xylose and mixed sugars us-
ing recombinant Saccharomyces cerevisiae engineered for
xylose utilization. Appl. Microbiol. Biotechnol. 2009, 82,
1037–1047.
[64] Kuyper, M., Hartog, M. M., Toirkens, M. J.,Almering, M. J. et
al., Metabolic engineering of a xylose-isomerase-express-
ing Saccharomyces cerevisiae strain for rapid anaerobic xy-
lose fermentation. FEMS Yeast Res. 2005, 5, 399–409.
[65] Kuyper, M.,Toirkens,M. J.,Diderich, J.A.,Winkler, A.A. et al.,
Evolutionary engineering of mixed-sugar utilization by a
xylose-fermenting Saccharomyces cerevisiae strain. FEMS
Yeast Res. 2005, 5, 925–934.
[66] Verduyn,C., Postma,E., Scheffers,W.A., Van Dijken, J. P., Ef-
fect of benzoic acid on metabolic fluxes in yeasts: A contin-
uous-culture study on the regulation of respiration and al-
coholic fermentation. Yeast 1992, 8, 501–517.