Genetic engineering of candida utilis yeast for efficient production of L-lactic acid

Article abstract of DOI:10.1271/bbb.90186

Polylactic acid is receiving increasing attention as a renewable alternative for conventional petroleum-based plastics. In the present study, we constructed a meta-bolically-engineered Candida utilis strain that produces L-lactic acid with the highest efficiency yet reported in yeasts. Initially, the gene encoding pyruvate decarbox-ylase (CuPDCl) was identified, followed by four CuPDCl disruption events in order to obtain a null mutant that produced little ethanol (a by-product of L-lactic acid). Two copies of the L-lactate dehydrogenase (l-LDH) gene derived from Bos taurus under the control of the CuPDCl promoter were then integrated into the genome of the CuPdcl -null deletant. The resulting strain produced 103.3 g/1 of L-lactic acid from 108.7 g/1 of glucose in 33 h, representing a 95.1% conversion. The maximum production rate of L-lactic acid was 4.9 g/l/h. The optical purity of the L-lactic acid was found to be more than 99.9% e.e.

Full text of DOI:10.1271/bbb.90186

Biosci. Biotechnol. Biochem., 73 (8), 1818–1824, 2009
Genetic Engineering of Candida utilis Yeast for Efficient Production
of L-Lactic Acid
y
Shigehito IKUSHIMA, Toshio FUJII, Osamu KOBAYASHI, Satoshi YOSHIDA, and Aruto YOSHIDA
Central Laboratories for Frontier Technology, KIRIN Holdings Co., Ltd.,
1
-13-5 Fukuura, Kanazawa-ku, Yokohama, Kanagawa 236-0004, Japan
Received March 13, 2009; Accepted May 9, 2009; Online Publication, August 7, 2009
[doi:10.1271/bbb.90186]
Polylactic acid is receiving increasing attention as a
boost the production of L-lactic acid in Saccharomyces
4
–8)
9,10)
renewable alternative for conventional petroleum-based
plastics. In the present study, we constructed a meta-
bolically-engineered Candida utilis strain that produces
L-lactic acid with the highest efficiency yet reported in
yeasts. Initially, the gene encoding pyruvate decarbox-
ylase (CuPDC1) was identified, followed by four
CuPDC1 disruption events in order to obtain a null
mutant that produced little ethanol (a by-product of
L-lactic acid). Two copies of the L-lactate dehydrogenase
cerevisiae
and Kluyveromyces lactis.
In the case of S. cerevisiae, which is known as a
Crabtree-positive yeast, and which has three PDC genes
(ScPDC1, ScPDC5, and ScPDC6), reduction/inactiva-
tion of PDC activity causes a severe growth defect.
Increasing the copy number of L-LDH and the use of
EMS-generated mutants in which the growth defect is
suppressed has resulted in better strain performance.
For example, a yield of 122 g/l of L-lactic acid from
a 61% cane-juice sugar medium after 48 h of fermenta-
tion under neutral conditions using NaOH has been
(
L-LDH) gene derived from Bos taurus under the control
of the CuPDC1 promoter were then integrated into the
genome of the CuPdc1-null deletant. The resulting strain
produced 103.3 g/l of L-lactic acid from 108.7 g/l of
glucose in 33 h, representing a 95.1% conversion. The
maximum production rate of L-lactic acid was 4.9 g/l/h.
The optical purity of the L-lactic acid was found to be
more than 99.9% e.e.
1
1,12)
reported.
representing an 81.5% conversion of D-glucose within
A yield of 82.3 g/l of L-lactic acid,
1
3)
168–216 h, has also been achieved. While further
attempts have been reported, e.g., in which ScADH1
(alcohol dehydrogenase 1; EC 1.1.1.1) was disrupted in
1
4)
addition to the deletion of ScPDC1, and in which
ScJEN1, encoding a lactate transporter, was overex-
1
5)
Key words: Candida utilis; pyruvate decarboxylase;
L-lactate dehydrogenase; L-lactic acid
pressed,
fermentation are still needed.
improvements in yield and duration of
In the case of K. lactis, a Crabtree-negative yeast, the
growth rate of a PDC-null mutant was comparable to
Lactic acid, a monomer of polylactic acid (a material
of plastic), can be produced by fermentation from plant
biomass, and this should reduce the use of petroleum and
generate no net increase in carbon dioxide emissions.
L-Lactic acid is commonly produced by fermentation
using lactic acid bacteria, such as Lactobacillus species,
which have numerous growth requirements and produce
D-lactic acid as well, resulting in an optical purity of
1
6)
that of a wild-type strain, but the yield of a strain
expressing L-LDH and lacking KlPDC1, which was
1
0)
reported by Porro et al.,
was still far from the
maximum theoretical yield: 0.58 lactic acid produced/
g-glucose consumed rather than 1 g/g. Bianchi et al.
modified the strain by additional disruption of KlPDA1,
encoding the pyruvate dehydrogenase (PDH; EC 1.2.4.1)
E1ꢀ subunit, which converts pyruvic acid into acetyl-
1
–3)
about 95% L-lactic acid.
Since optical purity affects
9
)
physical characteristics, including thermostability and
crystallization of polylactic acid, it is desirable to have
the highest purity of L-lactic acid possible.
coenzyme A. The resulting strain produced L-lactic acid
with a higher yield, 0.85 g/g, under neutralizing con-
ditions, but the glucose consumption rate was extremely
low, less than 0.3 g/l/h even with adequate cell density.
Slow consumption was probably related to disruption of
KlPDA1, because this disruption in the wild-type strain
Yeast normally produces ethanol via alcoholic fer-
mentation, in which pyruvic acid (the final product of
glycolysis) is converted to acetaldehyde by decarbox-
ylation via pyruvate decarboxylase (PDC; EC 4.1.1.1),
followed by reduction to ethanol via alcohol dehydro-
genase. L-Lactic acid is not a product of this process
in yeast. However, pyruvic acid can be converted into
L-lactic acid by heterologous expression of L-lactate
dehydrogenase (L-LDH; EC 1.1.1.27). When both
L-LDH and PDC are expressed simultaneously, the
production or yield of L-lactic acid is low, presumably
due to the competition for pyruvic acid by the two
enzymes.4 Decreasing PDC activity has been found to
1
7)
resulted in poor growth in glucose medium.
The industrially important yeast Candida utilis,
classified as a Crabtree-negative yeast, is currently used
to produce several valuable chemicals, such as gluta-
1
8–20)
thione and RNA.
substrates, e.g., pulping-waste liquors from the paper
C. utilis can grow on inexpensive
2
1)
industry, while most other yeasts cannot. Since the
development of an efficient electroportation-based
method for transforming C. utilis, this yeast has
been used in the heterologous production of monellin,
2
2)
)
y
To whom correspondence should be addressed. Fax: +81-45-788-4042; E-mail: Shigehito Ikushima@kirin.co.jp
Abbreviations: kb, kilo-base pair; HygB, hygromycin B; PDC, pyruvate decarboxylase; L-LDH , L-lactate dehydrogenase; AcH, acetaldehyde

Breeding of C. utilis for Efficient Production of L-Lactic Acid
1819
ꢀ
-amylase, and carotenoids such as lycopene.23–26)
Recently, we developed new strategies for performing
efficient multiple transformations of C. utilis. One is
2
7)
co-transformation, and the other is based on the Cre-
2
8)
loxP recombination system. Using the latter method,
a four-round deletion of the CuURA3 gene, encoding
0
orotidine-5 -phosphate decarboxylase, yielded an uracil
8)
auxotroph.2
In the present study, we describe a genetically-
engineered strain of C. utilis which to our knowledge
produces L-lactic acid more efficiently than previously
reported in yeasts.
Fig. 1. Map of Constructed Plasmid pCU681 (see ‘‘Materials and
Methods’’).
Materials and Methods
Strains and media. Candida utilis NBRC0988, its derivatives
described below), and Saccharomyces cerevisiae S288C were used.
(
Cells were cultured at 30 C in YPD medium (1% Bacto yeast extract,
a 34-bp loxP sequence, the approximately 800-bp PGK promoter
(CuPGKpr) and the HPT gene, which is responsible for HygB-
resistance. Plasmid pGAPPT1023) was used as a template in a PCR
with primers IM-54 and IM-55 to amplify a DNA fragment containing
a GAP terminator (CuGAPtr) and loxP. The two PCR products were
mixed, and, due to the complementarity of primers IM-57 and -54,
were used as a template in the next round of PCR with primers IM-1
and IM-2, resulting in a 2.5-kb DNA fragment (the LHL module)
consisting of loxP, CuPGKpr, HPT, GAPtr, and loxP, which was
cloned into the pCR2.1-TOPO vector, yielding plasmid pCU621.
Two individual PCRs were performed, as follows: one fragment
was amplified using primers IM-347 and -348 and template DNA
pPGKPT2;32) the other was done with primers IM-349 and -350, and
pCU621. Because primers IM-348 and IM-349 have complementary
ꢀ
2% Bacto peptone, and 2% glucose), unless otherwise specified. Solid
media were made with 2% agar. G418 (Geneticin, Sigma-Aldrich, St.
Louis, MO) was added to YPD to a final concentration of 200 mg/ml to
select transformants expressing the APT gene, which encodes an
aminoglycoside phosphotransferase. Hygromycin B (HygB, Wako
Pure Chemical Industries, Osaka, Japan) was added to YPD to a final
concentration of 600 mg/ml to select for transformants carrying the
HPT gene, encoding HygB phosphotransferase.
E. coli DH5ꢀ (Toyobo, Osaka, Japan) served as a plasmid host.
E. coli was grown in LB (1% Bacto tryptone, 0.5% Bacto yeast extract,
and 1% NaCl) containing 50 mg/ml of ampicillin, and was transformed
by standard methods.2
9)
0
regions at their 5 -ends, the two resulting fragments were mixed and
used as a template in the next PCR with primers IM-347 and IM-350.
The amplified fragment (approximately 3 kb) was digested with BamHI
and ClaI, and ligated into the BamHI-ClaI gap of pCU670 to construct
plasmid pCU675.
Recombinant DNA techniques. Standard recombinant DNA tech-
niques were used,29,30) or the procedures recommended by the
suppliers. PCR (polymerase chain reaction) was performed with
LA-Taq polymerase (Takara Bio, Shiga, Japan) according to the
manufacturer’s instructions. The primers used are listed in Table 1.
Nucleotide sequences were determined using an ABI 3130 xl DNA
analyzer (Applied Biosystems, Foster City, CA).
The Bos taurus (bovine) L-LDH gene sequence (DDBJ/EMBL/
GenBank accession no. AAI46211.1)33) was modified based on major
codon usage in Pichia jadinii (C. utilis is an anamorph of P. jadinii;
Codon Usage Database: http://www.kazusa.or.jp/codon/) and synthe-
sized (Takara Bio) in such a way that the original amino acid sequence
was unchanged. Using the L-LDH gene as template, a fragment of
about 1-kb DNA was amplified with primers IM-343 and IM-379. The
promoter region of the CuPDC1 gene was amplified using oligonu-
cleotides IM-341 and IM-342 as primers and C. utilis genomic DNA as
template. Due to the complementarity of primers IM-342 and -343, the
two resulting products were mixed and used as a template in the next
PCR with primers IM-341 and IM-379, after which the amplified
fragment (approximately 2.2 kb) was digested with NotI and BglII, and
ligated into the NotI-BamHI gap of pCU675 to construct pCU681
Southern blot hybridization analysis. DNA was digested with
selected restriction enzymes and separated in 1% agarose gels. After
electrophoresis, the gels were incubated in 0.25-M HCl for 15 min with
shaking, followed by denaturation and capillary blotting of DNA onto
a Hybond N+ nylon membrane (GE Healthcare UK, Buckingham-
shire, UK). Probe DNA labeled with a random primer labelling kit
(
hybridization solution, and hybridization was performed at 60 C for
Takara Bio) and [ꢀ-32P] dCTP (110 TBq/mmol) was added to the
ꢀ
1
6 h. After hybridization, the membranes were washed in 1 ꢁ SSC
containing 0.1% SDS, followed by a wash in 0:5 ꢁ SSC containing
(
Fig. 1).
ꢀ
0.1% SDS at 60 C for 2 h. The membranes were then subjected to
autoradiography.
Breeding of yeast strains. Transformations of C. utilis were carried
out by electroporation as previously described.28) In order to construct
a null disruptant of the CuPDC1 gene, designated Cupdc1ꢁ4, a fourth
deletion was carried out using the Cre-loxP system essentially as
Molecular cloning. Construction of a C. utilis genomic DNA library
has been described.22) In order to obtain fragments containing the PDC
gene, PCR was performed using C. utilis genomic DNA as template.
Oligonucleotides IKSM-29 and IKSM-30 were used as primers. The
amplified DNA fragments (207 bp) were cloned into pCR2.1-TOPO
vector (TOPO TA Cloning Kit; Invitrogen, Carlsbad, CA), followed by
sequencing. The PCR product was used as a probe in colony
hybridization and Southern blot hybridization in order to clone the
PDC gene from C. utilis. Colony hybridization was conducted as
previously described.29)
described for constructing the CuUra3-null deletant.28) Two different
gene disruption cassettes consisting of the aforementioned LHL
0
0
module flanked by the 5 - and 3 -regions adjacent to the CuPDC1
gene for targeted replacement were generated in two-step PCRs.
Initially, the fragments forming these cassettes were amplified
separately. The LHL module was constructed using primers IM-1
and 2 and pCU56328) as template. A fragment upstream of CuPDC1,
the targeted region, was amplified with primers IM-277 and -278, with
C. utilis genomic DNA as template. Two different downstream regions
for CuPDC1, one for the first and second deletions, and the other for
the third and fourth deletions, were generated with primer pairs IM-
279/-280 and IM-185/-168 respectively, with C. utilis genomic DNA
as template. Plasmid pCU595,28) which harbors the Cre-expressing
module with the APT gene for G418-resistance, was introduced after
every deletion in order to excise the HPT gene. The resulting strains
were then grown in non-selective YPD to allow loss of pCU563.
Construction of plasmids. The downstream region of the CuPDC1
gene was amplified with primer set IM-345 and IM-346, followed by
digestion with BssHII. The DNA fragment was ligated into the BssHII
site of pBluescriptIISK+ (Stratagene Products Division, Agilent
Technologies, La Jolla, CA) to construct plasmid pCU670.
Plasmid pGKHPT131) was used as a template in a PCR with primers
IM-283 and IM-57 in order to amplify a DNA fragment that contained

1820
S. IKUSHIMA et al.
Table 1. PCR Primers Used in This Study
0 0
Sequence (5 to 3 )
Name
IKSM-29 CARGTYTTRTGGGGTTCYATYGGTTT
IKSM-30 TTCAATRGTGTARCCMYYGTTGTTCAA
IM-345
IM-346
IM-283
ACTCGCGCGCAAGATCTAAGCGGCCGCTAATGGATCCAATAATCGATGCTGTCTTTCTTCTTCATGGG
ACTCGCGCGCAAGATCTGAACTTCTCCAACAGGTAGC
CGGCCGCCAGCTGAAGCTTCGTACGCTGCAGGTCGACAACCCTTAATATAACTTCGTATAATGTATGCTATACGAAGTT-
ATCCTTTGCTGTGTTCTACC
IM-57
IM-54
IM-55
CATGAGGATCATAATTTATAACGTAATCCCATAAATAAAAGTCATACAATCATTCCTTTGCCCTCGGA
ATTGTATGACTTTTATTTATGGGATTACGTTATAAATTATGATCCTCATG
TAGGCCACTAGTGGATCTGATATCACCTAATAACTTCGTATAGCATACATTATACGAAGTTATTCATTCATCCCTCACT-
ATCG
IM-1
IM-2
GGCCGCCAGCTGAAGCTTCG
AGGCCACTAGTGGATCTGAT
IM-347
IM-348
IM-349
ACTCGGATCCCTGCAAGCTACTTTGTAATTAAACAAATAACGGG
GGAGACTCTTCACACTGTTGGCGTCTATGATTCAAGATTGTCAGTTTCCATCGTGGATTGGAATAGTTGTGGTGACCTTG
ACAACTATTCCAATCCACGATGGAAACTGACAATCTTGAATCATAGACGCCAACAGTGTGAAGAGTCTCCAGCTGAAG-
CTTCGTACGCTG
IM-350
IM-343
IM-379
IM-341
IM-342
IM-277
IM-278
IM-279
IM-280
IM-185
IM-168
ACTCGGCCGGCCATCGATCACTAGTGGATCTGATATCACC
CCCGTTATACACAACAAACAAACAAAACTAAAACAATCGATACCATGGCTACCTTGAAGGACCA
ACTCAGATCTTCATCAGAACTGCAATTCCTTCTGG
ACTCGCGGCCGCTCTAGACACCAACTTTGAAGATAGGG
TGGTCCTTCAAGGTAGCCATGGTATCGATTGTTTTAGTTTTGTTTGTTTGTTGTGTATAACGGG
AGTTGGACTCGGATCATCTC
CTGCAGCGTACGAAGCTTCAGCTGGCGGCCAACATTCCTACGCTCAGAGC
ATTAGGTGATATCAGATCCACTAGTGGCCTACAGAACGCTCCTAAACACG
GAACTTCTCCAACAGGTAGC
ATTAGGTGATATCAGATCCACTAGTGGCCTTGCAATTGACCGTCCAAGAG
TCTTGTACGTGTTTAGGAGC
In order to construct strains harboring the L-LDH gene, the strains
EDTA, and 20-mM Bis-Tris was applied as the reaction buffer. The
ꢀ
were transformed with BglII-digested pCU681. The CuPdc1-null
disruptant Cupdc1ꢁ4 was transformed with the plasmid to construct a
strain harboring a single copy of L-LDH (named Cupdc1ꢁ4-LDH1).
After removal of HPT using pCU563, the resulting clone was
transformed with the same plasmid to obtain a strain carrying two
copies of L-LDH (named Cupdc1ꢁ4-LDH2).
column was operated at 45 C at a 0.8-ml/min flow rate.
Nucleotide sequence accession numbers. The DNA sequences of
the CuPDC1 gene have been deposited in the DDBJ/EMBL/GenBank
database under accession no. AB489119. The nucleotide sequence
of the bovine L-LDH gene, whose codon-usage was modified for
expression in C. utilis, has been given accession no. AB489120.
Fermentation. In pre-cultivation, cells from a 2-d-old colony taken
ꢀ
from a YPD plate at 30 C were transferred to flasks containing 50 ml
of YPD medium in 500-ml Sakaguchi-flasks and incubated for 15–22 h
Results
ꢀ
at 30 C on a shaker (130 rpm). The cells were harvested by
ꢀ
Cloning of PDC genes in C. utilis
centrifugation at 3;000 ꢁ g for 5 min at 4 C, washed twice with fresh
While enhancement of the production of pyruvate
decarboxylase in C. utilis has been well studied in terms
of culture conditions, e.g., oxygen supplied and medium
fermentation medium, and then inoculated into the fermentation
ꢀ
medium. Fermentations were performed at 25/30/35 C in 100-ml
spherical flat-bottom flasks containing 15 ml (unless specified other-
wise) of YPD10 medium (1% Bacto yeast extract, 2% Bacto peptone,
3
5–38)
pH,
the PDC genes of C. utilis (CuPDC) have not
1
0–11% glucose) with gentle shaking (80 rpm). They were performed
with an initial OD600 of 10, otherwise described. Calcium carbonate
45 g/l) was added to neutralize the acidic products (e.g., lactic acid
previously been identified. In the present study, we
cloned a CuPDC gene. Genes encoding PDC have been
cloned and sequenced from various yeast species,
including S. cerevisiae, K. lactis, Pichia stipitis, and
(
and pyruvic acid), otherwise described. The yield of L-lactic acid was
calculated according to the following equation: (amount of L-lactic
acid at sampling time/amount of glucose at 0 h) ꢁ 100. Optical purity
1
6,39–41)
Candida glabrata.
Comparison of their amino
(
% e.e.) was calculated as follows: optical purity = (L-lactic acid ꢂ
acid sequences indicates that their carboxyl termini are
highly conserved (data not shown). Based on the
conserved region, mixed oligonucleotide primers
D-lactic acid)/(L-lactic acid þ D-lactic acid) ꢁ 100.
Analytical methods. Glucose and L-lactic acid were measured with
an immobilized enzymatic membrane method using the Biochemistry
Analyzer YSI-2700 System (YSI, YellowSpring, OH). D-Lactic acid
was determined using a diagnostic kit (Roche Diagnostics, Mannheim,
Germany) according to the instructions provided.
(
IKSM-29 and IKSM-30; see ‘‘Materials and Methods’’)
were synthesized and used in PCRs. The nucleotide
sequences of the two primers were deduced from the
conserved catalytic domains in S. cerevisiae PDC1
(
ScPDC1) and K. lactis PDC1 (KlPDC1), located at
Ethanol and acetaldehyde (AcH) were analyzed as described
previously34) with slight modifications. The supernatant solution was
diluted and subjected to high-pressure liquid chromatography (HPLC)
analysis using an ICSep-ION-300 column (CETAC Technologies,
the carboxy terminus. Sequence analysis of the ampli-
fied fragment (207 bp) indicated significant similarity to
the known ScPDC1 gene (approximately 80% identity in
deduced amino acid sequence). One of these DNA
fragments was used as a probe in Southern and colony
hybridization.
ꢀ
Omaha, NE). The column was operated at 60 C at a 0.4 ml/min flow
rate, with in 0.005 M sulphuric acid as the solvent.
The concentration of pyruvic acid was also assayed by HPLC.
Separation was achieved using a Shim-pack SPR-H column (Shimazu,
Kyoto, Japan), with 5-mM p-toluenesulfonic acid as the mobility
buffer. A solution consisting of 5-mM p-toluenesulfonic acid, 0.08-mM
As shown in Fig. 2, this probe hybridized to three
different bands in HindIII digests of S. cerevisiae

Breeding of C. utilis for Efficient Production of L-Lactic Acid
1821
Table 2. Fermentation Analysis of the Wild-Type and the Cupdc1-
Null Strain (Cupdc1ꢁ4)
Sc
H
Cu
X
H
Bg E Ba P
Medium
without cells
Strain
NBRC0988
Cupdc1ꢁ4
Size (kbp)
OD600
59:9 ꢃ 2:0
<0:1
3:96 ꢃ 0:04
26:2 ꢃ 9:0
462:4 ꢃ 5:5
2:8 ꢃ 0:2
80:0 ꢃ 0:6
<0:01
Not determined
100:2 ꢃ 0:3
<0:01
Glucose (g/l)
Ethanol (g/l)
AcH (mg/l)
Pyruvic acid
1
0
5
1:14 ꢃ 0:06
3659:9 ꢃ 25:6
0:30 ꢃ 0:14
<0:1
3
(
mg/l)
ꢀ
Fermentation was performed at 30 C for 48 h in 100-ml spherical flat-
bottom flasks with an initial OD600 of 0.5. The medium used was 50 ml of
YPD10 without neutralization reagent. Values are means for independent
triplicate experiments.
1
.5
Lane
1
2
3
4
5
6
7
initial glucose provided, using just 20 g/l (Table 2).
Cupdc1ꢁ4 produced little AcH and ethanol, while a
significant amount of pyruvic acid, a precursor of AcH,
accumulated as observed in S. cerevisiae strains with
Fig. 2. Detection of PDC Genes by Southern-Blot Hybridization
Analysis.
Lane 1, HindIII-digested genomic DNA from S. cerevisiae
S288C. Lanes 2 to 7, digested genomic DNA from C. utilis
NBRC0988: XbaI (lane 2), HindIII (lane 3), BglII (lane 4), EcoRI
4
5,46)
low PDC activity.
This profile apparently differed
(
lane 5), BamHI (lane 6) and PstI (Lane 7). A 0.2-kb fragment
amplified with primers IKSM-29 and -30 was used as a probe
see ‘‘Materials and Methods’’). Abbreviations: Sc, genomic DNA
from that of the wild-type strain, which produced more
ethanol and AcH. These data suggest that CuPdc1p
provides the highest activity among any potentially
functional PDCs in C. utilis.
(
of Saccharomyces cerevisiae; Cu, genomic DNA of Candida utilis;
Ba, BamHI; Bg, BglII; E, EcoRI; H, HindIII; P, PstI; X, XbaI
Fermentation
genomic DNA, and these were deduced to be ScPDC1,
ScPDC5, and ScPDC6. C. utilis DNA digests with XbaI,
HindIII, BamHI, or PstI yielded a single strong band,
and the BglII digest gave two clear bands, in good
agreement with a BglII site in the probe. Two bands
were also observed in the EcoRI digest, but sequence
analysis indicated no EcoRI site in the probe. These
results suggest that C. utilis possesses one or two types
of PDC genes, or that it has two or more copies due to
its polyploidy.
Next we constructed strains Cupdc1ꢁ4-LDH1 and
Cupdc1ꢁ4-LDH2, which harbor one and two copies of
bovine L-LDH under the control of the CuPDC1
promoter and integrated at the CuPDC1 locus, respec-
tively (see ‘‘Materials and Methods’’). In aerobic
culture, their growth was poorer than that of the wild-
type strain, but was comparable to that of Cupdc1ꢁ4
(Fig. 3B). The maximum production rate of L-lactic
acid was calculated in order to investigate the effects of
L-LDH copy number and temperature (Table 3). When
ꢀ
Colony hybridization yielded several clones harboring
CuPDC genes, and their sequences were determined.
The C. utilis PDC gene, designated CuPDC1, consists
of a 1,692-bp open reading frame (ORF) encoding a
polypeptide of 563 amino acids. The deduced amino
acid sequence shared 76 and 77% identity with that of
ScPdc1p and KlPdc1p respectively. In the 1.2-kb region
upstream of the CuPDC1 ORF, a presumptive promoter
region was found containing putative cis-elements
including a TATA element and a UAS-PDC sequence
reported to be necessary for transcriptional activation in
fermentation was carried out at 30 C, production of L-
lactic acid by Cupdc1ꢁ4-LDH1 was lower than by
Cupdc1ꢁ4-LDH2. Subsequently, Cupdc1ꢁ4-LDH2 was
ꢀ
ꢀ
tested at 25 and 35 C, and the production at 35 C was
found to be the highest among the three temperatures.
ꢀ
Similarly, the yield of Cupdc1ꢁ4-LDH2 at 35 C was
the highest compared to that under the other conditions.
Based on these data, we analyzed fermentation con-
ꢀ
ducted by Cupdc1ꢁ4-LDH2 at 35 C in YPD10 media
over a 36-h time course (Fig. 4). Glucose was com-
pletely consumed at 36 h, and the concentration of
L-lactic acid produced was equivalent to the concen-
tration of glucose at 0 h. Ethanol was scarcely produced
(<0:01 g/l), and the concentration of pyruvate was
291 ꢃ 20 mg/l. Although the fermentation conditions
were not identical, the pyruvate concentration was
much lower than that of Cupdc1ꢁ4 (Table 2), suggest-
ing that the introduced L-LDH effectively metabolized
pyruvate into L-lactic acid. The yield of L-lactic acid
was highest at 33 h, 95:1 ꢃ 1:1%, and its optical purity
was more than 99.9% (L-lactic acid, 103:3 ꢃ 0:94 g/l;
D-lactic acid, 0:0214 ꢃ 0:004 g/l). We also analyzed
production in another medium (0.1% yeast extract,
0.2% peptone, and 10% glucose), and obtained a yield
of 91.0% at 33 h. We conclude that Cupdc1ꢁ4-LDH2
can produce L-lactic acid efficiently even in a relatively
nutrient-poor medium, which would simplify subse-
quent purification.
4
2,43)
S. cerevisiae and K. lactis.
In S. cerevisiae, reduc-
tion of PDC activity by disruption of both ScPDC1 and
ScPDC5 has been reported to cause a severe growth
defect in YPD,4 but expression of CuPDC1 under the
control of a ScPDC1 promoter complemented the
lethality (data not shown), strongly suggesting that
CuPdc1p has PDC activity.
4)
Characterization of the Cupdc1-null mutant
We evaluated the fermentation ability of the Cupdc1-
null mutant (Cupdc1ꢁ4), constructed by performing
four individual CuPDC1 disruptions (see ‘‘Materials and
Methods’’). In aerobic culture on YPD medium,
Cupdc1ꢁ4 did not show severe growth deficiency, but
the glucose consumption and growth of Cupdc1ꢁ4 were
poorer than those of the wild type (Fig. 3). In micro-
aerobic culture, Cupdc1ꢁ4 consumed only 20% of the

1822
S. IKUSHIMA et al.
120
A
20
15
10
5
0
100
8
6
4
0
0
0
20
0
0
10
20
30
40
0
10
20
30
Time (h)
Time (h)
Fig. 4. Time Course for Micro-Aerobic Yeast Culture.
B
Data are means for at least three independent experiments. Error-
bars (standard deviations) are hidden by the symbols. The concen-
tration of glucose at 0 h was 108:7 ꢃ 0:5 g/l. At 33 h, a maximum
concentration of L-lactic acid was obtained, 103:3 ꢃ 0:94 g/l. The
purity of the L-lactic acid in all samples was 99.9% e.e. or higher.
5
4
3
2
1
0
0
0
0
0
C. utilis strain, whose ploidy was suggested to be
tetraploid because four rounds of disruption were
required to inactivate CuPDC1. This observation
accords with the copy number of CuURA3, determined
2
8)
in a previous study. The strain constructed in the
present study, Cupdc1ꢁ4-LDH2, had extremely high
L-lactic acid productivity compared to other genetically-
modified yeasts.
0
0
10
20
30
Time (h)
When a CuPDC1 promoter-driven L-LDH gene was
introduced into a wild-type strain, glucose was converted
into both L-lactic acid and ethanol (data not shown). By
contrast, recombinants lacking a functional CuPDC1
Fig. 3. Time Course in Aerobic Yeast Culture.
A, Glucose-Concentration in the Media; B, Cell-Density (OD600).
Solid circles, open circles, open squares, and solid squares represent
the wild-type strain, Cupdc1ꢁ4, Cupdc1ꢁ4-LDH1, and Cupdc1ꢁ4-
LDH2 respectively. The glucose-concentration values of L-LDH-
carrying strains were too similar to be distinguished. Data are means
for two independent experiments. Error bars (mean deviations) are
(Cupdc1ꢁ4, Cupdc1ꢁ4-LDH1, and Cupdc1ꢁ4-LDH2)
produced little ethanol, indicating that disruption of
CuPDC1 effectively decreased ethanol production.
Although inactivation of PDC activity in S. cerevisiae
prevented growth in YPD, this severe growth defect was
not observed in the corresponding C. utilis construct,
similarly to what has been reported for the Crabtree-
ꢀ
too small to be seen. Cells from 2-d-old YPD plates grown at 30 C
were added to 50-ml of YPD pre-cultures in 500-ml Sakaguchi
ꢀ
flasks and incubated overnight at 30 C on a shaker (130 rpm). The
ꢀ
cells were harvested by centrifugation at 3;000 ꢁ g for 5 min at 4 C,
ꢀ
and then added to fresh medium to initiate the time course at 30 C
in 50 ml of YPD in 500-ml Sakaguchi flasks (130 rpm).
1
6)
negative yeast K. lactis. Additional inactivation of
PDH activity was required to increase the yield of
Table 3. Effect of L-LDH Copy Number and Temperature on the
Production Rate of L-Lactic Acid
9
)
L-lactic acid in K. lactis, although this was accompa-
nied by a significant growth defect, which is not
considered desirable, particularly in fermentation proc-
esses. In C. utilis, however, inactivation of PDC activity
alone was sufficient to obtain high yields of L-lactic acid,
the yield being 95.1% (103.3 g/l; >99:9% e.e.) from
L-Lactic acid Glucose
Ratio
ꢄꢄ
Strain name/Temperature
ꢄ
ꢄ
(
g/l/h)
(g/l/h)
(%)
ꢀ
ꢀ
ꢀ
ꢀ
Cupdc1ꢁ4-LDH1/30 C
Cupdc1ꢁ4-LDH2/30 C
3:5 ꢃ 0:1
4:2 ꢃ 0:1
2:9 ꢃ 0:1
4:9 ꢃ 0:2
3:8 ꢃ 0:3 86:2 ꢃ 4:3
5:2 ꢃ 0:2 88:1 ꢃ 2:9
3:1 ꢃ 0:1 77:3 ꢃ 2:5
5:3 ꢃ 0:1 93:9 ꢃ 2:3
Cupdc1ꢁ4-LDH2/25 C
Cupdc1ꢁ4-LDH2/35 C
108.7 g/l of glucose in a 33-h fermentation.
It is still not fully evident why Cupdc1ꢁ4-LDH2 can
Values are means and standard deviations for at least three independent
experiments.
produce L-lactic acid efficiently, but we suppose that the
principal reason is derived from the high ability to
convert glucose into ethanol in the wild-type strain
ꢄ
The maximum production rate of L-lactic acid/the maximum consumption
rate of glucose were calculated from concentrations between 4, 6, 8, 10 and
(Table 2). It was higher than that of K. lactis, as
reported by others.
1
2 h, using a linear regression method.
This ratio was calculated according to the following equation: (g-L-lactic
ꢄꢄ
10,16)
Ethanol production was almost
acid produced for 12 h)/g-glucose consumed for 12 h) ꢁ 100.
completely abolished in Cupdc1ꢁ4, and its capacity
appeared to be efficiently shifted to L-lactic acid
production by the introduction of two copies of
exogenous L-LDH. In addition, a certain level of
metabolic capacity from pyruvate to the citric acid
cycle in the strain might also affect L-lactic acid
production, because ATP generation in the cells is
Discussion
The industrially-important yeast Candida utilis is
widely used to produce heterologous reagents, but to our
knowledge the elimination of a major metabolic end-
product and its replacement by another via recombinant
DNA technology has not yet been reported in this
species. Here we conducted genetic engineering of a
4
7)
important in increasing L-lactic acid. We observed
that Cupdc1ꢁ4-LDH2 showed higher production in the
micro-aerobic culture than in the static culture (data not

Breeding of C. utilis for Efficient Production of L-Lactic Acid
1823
shown), and this is consistent with recent reports that
lactic acid-producing S. cerevisiae strains require oxy-
4) Dequin S and Barre P, Biotechnology, 12, 173–177 (1994).
5)
6)
7)
Porro D, Brambilla L, Ranzi BM, Martegani E, and Alberghina
L, Biotechnol. Prog., 11, 294–298 (1995).
1
2,47)
gen for the generation of ATP.
Since the growth
Adachi E, Torigoe M, Sugiyama M, Nikawa J, and Shimizu K,
J. Ferment. Bioeng., 86, 284–289 (1998).
rate of Cupdc1ꢁ4-LDH2 was comparable to that of
Cupdc1ꢁ4 (Fig. 3B), it was assumed that Cupdc1ꢁ4-
LDH2 had the ability to generate enough respiration-
metabolism to obtain sufficient ATP under micro-
aerobic conditions. However, further study is needed
to confirm these hypotheses.
Ishida N, Saitoh S, Tokuhiro K, Nagamori E, Matsuyama T,
Kitamoto K, and Takahashi H, Appl. Environ. Microbiol., 71,
1964–1970 (2005).
8
9
)
)
Skory CD, J. Ind. Microbiol. Biotechnol., 30, 22–27 (2003).
Bianchi MM, Brambilla L, Protani F, Liu CL, Lievense J, and
Porro D, Appl. Environ. Microbiol., 67, 5621–5625 (2001).
Cupdc1ꢁ4-LDH2 had the highest ability to produce
ꢀ
1
0) Porro D, Bianchi MM, Brambilla L, Menghini R, Bolzani D,
Carrera V, Lievense J, Liu CL, Ranzi BM, Frontali L, and
Alberghina L, Appl. Environ. Microbiol., 65, 4211–4215 (1999).
L-lactic acid at 35 C, although it produced high
concentrations of L-lactic acid over a range of temper-
ꢀ
atures, spanning 25 to 35 C. This feature is similar to a
previous report that PDC in C. utilis was induced only at
11) Ishida N, Saitoh S, Ohnishi T, Tokuhiro K, Nagamori E,
Kitamoto K, and Takahashi H, Appl. Biochem. Biotechnol., 131,
795–807 (2006).
ꢀ
48)
permissive growth temperatures, 5 to 35 C, and that the
ꢀ
12) Saitoh S, Ishida N, Onishi T, Tokuhiro K, Nagamori E,
Kitamoto K, and Takahashi H, Appl. Environ. Microbiol., 71,
highest activity was detected at 35 C. Accordingly,
ꢀ
the high expression of CuPDC at 35 C might contribute
2
789–2792 (2005).
to the large production of L-lactic acid at that temper-
ature. This relatively high optimal temperature is also
advantageous because lactic acid at high concentrations
easily becomes solidified in the medium due to low
solubility. Usually, fermentation of S. cerevisiae and
K. lactis is limited at lower temperatures from 30 to
1
1
3) Ishida N, Saitoh S, Onishi T, Tokuhiro K, Nagamori E,
Kitamoto K, and Takahashi H, Biosci. Biotechnol. Biochem., 70,
1148–1153 (2006).
4) Tokuhiro K, Ishida N, Nagamori E, Saitoh S, Onishi T, Kondo
A, and Takahashi H, Appl. Microbiol. Biotechnol., 82, 883–890
(
2009).
1
1
5) Branduardi P, Sauer M, De Gioia L, Zampella G, Valli M,
Mattanovich D, and Porro D, Microb. Cell Fact., 5, 4 (2006).
6) Bianchi MM, Tizzani L, Destruelle M, Frontali L, and
Wesolowski-Louvel M, Mol. Microbiol., 19, 27–36 (1996).
ꢀ
5–7,9–14)
3
2 C.
The pH value of the medium also appeared to be
important for the production of L-lactic acid, since a
previous study showed that adjustment of pH was
17) Zeeman AM, Luttik MA, Thiele C, van Dijken JP, Pronk JT,
and Steensma HY, Microbiology, 144 (Pt 12), 3437–3446
3
5)
effective for the high expression of PDC in C. utilis.
(
1998).
8) Boze H, Moulin G, and Galzy P, Crit. Rev. Biotechnol., 12, 65–
6 (1992).
However, the limited research equipment in the present
study did not enable us to control this during the
fermentation shown in Fig. 3 (strain, Cupdc1ꢁ4-LDH2;
medium, YPD10 with CaCO3), and the initial pH of 6.6
finally dropped to 4.0 (data not shown). This decrease in
pH was more likely due to the accumulation of a major
acidic product, L-lactic acid (pKa = 3.86). Therefore,
exact control of pH might further improve the efficiency
of L-lactic acid production.
1
1
8
9) Ichii T, Takehara S, Konno H, Ishida T, Sato H, Suzuki A, and
Yamazumi K, J. Ferment. Bioeng., 75, 375–379 (1993).
20) Li Y, Wei G, and Chen J, Appl. Microbiol. Biotechnol., 66, 233–
242 (2004).
2
2
2
1) Kurtzman CP and Fell JW, ‘‘The Yeast, A Taxonomic Study’’
Fourth edition, Elsevier Science B.V., Amsterdam (1998).
2) Kondo K, Saito T, Kajiwara S, Takagi M, and Misawa N,
J. Bacteriol., 177, 7171–7177 (1995).
There are other advantages to fermentative production
by C. utilis. For example, it is known to produce
3) Kondo K, Miura Y, Sone H, Kobayashi K, and Iijima H, Nat.
Biotechnol., 15, 453–457 (1997).
24) Miura Y, Kettoku M, Kato M, Kobayashi K, and Kondo K,
4
9)
significant amounts of invertase, which is valuable
when sucrose-containing molasses is used as a growth
medium. Cupdc1ꢁ4-LDH2 was found to produce
L-lactic acid with a yield of 94.8% when grown in
YP-based media containing 100 g/l of sucrose for 33 h
J. Mol. Microbiol. Biotechnol., 1, 129–134 (1999).
2
2
2
5) Miura Y, Kondo K, Saito T, Shimada H, Fraser PD, and Misawa
N, Appl. Environ. Microbiol., 64, 1226–1229 (1998).
6) Miura Y, Kondo K, Shimada H, Saito T, Nakamura K, and
Misawa N, Biotechnol. Bioeng., 58, 306–308 (1998).
7) Ikushima S, Minato T, and Kondo K, Biosci. Biotechnol.
Biochem., 73, 152–159 (2009).
(
densities,
data not shown). Also, C. utilis can ferment at high cell
which should contribute to increasing
50,51)
production efficiency. Taking all this, the recombinant
C. utilis Cupdc1ꢁ4-LDH2 strain probably holds great
promise for the industrial production of L-lactic acid.
28) Ikushima S, Fujii T, and Kobayashi O, Biosci. Biotechnol.
Biochem., 73, 879–884 (2009).
2
9) Sambrook J, Fritsch EF, and Maniatis T, ‘‘Molecular Cloning, a
Laboratory Manual,’’ Cold Spring Harbor Laboratory Press,
Cold Spring Harbor (1989).
Acknowledgments
3
0) Rose MD, Winston F, and Hieter P, ‘‘Methods in Yeast
Genetics,’’ Cold Spring Harbor Laboratory Press, Cold Spring
Harbor (1990).
We wish to thank Hiroshi Ashigai, Hideyuki
Tamakawa, Fumi Osawa, and Maiko Nakamura for
valuable discussion and technical assistance throughout
the course of this study.
31) Shimada H, Kondo K, Fraser PD, Miura Y, Saito T, and Misawa
N, Appl. Environ. Microbiol., 64, 2676–2680 (1998).
3
3
2) Kondo K, Kajiwara S, and Misawa N, International Patent
WO95/32289 (Nov. 30, 1995).
3) Ishiguro N, Osame S, Kagiya R, Ichijo S, and Shinagawa M,
Gene, 91, 281–285 (1990).
References
1)
2)
3)
Hofvendahl K and Hahn-H a¨ gerdal B, Enzyme Microb. Technol.,
0, 301–307 (1997).
Hofvendahl K and Hahn-H a¨ gerdal B, Enzyme Microb. Technol.,
6, 87–107 (2000).
34) Shi NQ, Cruz J, Sherman F, and Jeffries TW, Yeast, 19, 1203–
1220 (2002).
2
35) Chen AK, Breuer M, Hauer B, Rogers PL, and Rosche B,
Biotechnol. Bioeng., 92, 183–188 (2005).
2
Olmos-Dichara A, Ampe F, Uribelarrea JL, Pareilleux A, and
Goma G, Biotechnol. Lett., 19, 709–714 (1997).
36) Kaliterna J, Weusthuis RA, Castrillo JI, Van Dijken JP, and
Pronk JT, Yeast, 11, 317–325 (1995).

1824
S. IKUSHIMA et al.
3
7) Shin HS and Rogers PL, Biotechnol. Bioeng., 49, 52–62 (1996).
8) Weusthuis RA, Visser W, Pronk JT, Scheffers WA, and van
Dijken JP, Microbiology, 140 (Pt 4), 703–715 (1994).
9) Kellermann E, Seeboth PG, and Hollenberg CP, Nucleic Acids
Res., 14, 8963–8977 (1986).
45) Hohmann S, Mol. Gen. Genet., 241, 657–666 (1993).
46) Schmitt HD and Zimmermann FK, J. Bacteriol., 151, 1146–
1152 (1982).
3
3
4
4
47) van Maris AJ, Winkler AA, Porro D, van Dijken JP, and Pronk
JT, Appl. Environ. Microbiol., 70, 2898–2905 (2004).
48) Franzblau SG and Sinclair NA, Mycopathologia, 83, 29–33
(1983).
0) Lu P, Davis BP, and Jeffries TW, Appl. Environ. Microbiol., 64,
9
4–97 (1998).
1) Wang Q, He P, Lu D, Shen A, and Jiang N, J. Biochem., 136,
47–455 (2004).
49) Dworschack RG and Wickerham LJ, Appl. Microbiol., 9, 291–
294 (1961).
4
4
2) Butler G and McConnell DJ, Curr. Genet., 14, 405–412 (1988).
3) Destruelle M, Menghini R, Frontali L, and Bianchi MM, Yeast,
50) Liang G, Liao X, Du G, and Chen J, J. Appl. Microbiol., 105,
1432–1440 (2008).
4
1
5, 361–370 (1999).
51) Zhang X, Ito T, Kondo K, Kobayashi T, and Honda H, J. Chem.
Eng. Jpn., 35, 654–659 (2002).
4
4) Hohmann S, J. Bacteriol., 173, 7963–7969 (1991).