APPLIED AND ENVIRONMENTAL MICROBIOLOGY, July 2007, p. 4446–4454
Copyright © 2007, American Society for Microbiology. All Rights Reserved.
Vol. 73, No. 14
Engineering of the Yeast Yarrowia lipolytica for the Production
of Glycoproteins Lacking the Outer-Chain Mannose
Residues of N-Glycans?
Yunkyoung Song,1Min Hee Choi,1Jeong-Nam Park,1,2Moo Woong Kim,1,2
Eun Jung Kim,1,2Hyun Ah Kang,2* and Jeong-Yoon Kim1*
School of Bioscience and Biotechnology, Chungnam National University, Daejeon 305-764,1and
Korea Research Institute of Bioscience and Biotechnology, Daejeon 305-333,2Korea
Received 31 August 2006/Accepted 8 May 2007
In an attempt to engineer a Yarrowia lipolytica strain to produce glycoproteins lacking the outer-chain
mannose residues of N-linked oligosaccharides, we investigated the functions of the OCH1 gene encoding a
putative ?-1,6-mannosyltransferase in Y. lipolytica. The complementation of the Saccharomyces cerevisiae och1
mutation by the expression of YlOCH1 and the lack of in vitro ?-1,6-mannosyltransferase activity in the Yloch1
null mutant indicated that YlOCH1 is a functional ortholog of S. cerevisiae OCH1. The oligosaccharides
assembled on two secretory glycoproteins, the Trichoderma reesei endoglucanase I and the endogenous Y.
lipolytica lipase, from the Yloch1 null mutant contained a single predominant species, the core oligosaccharide
Man8GlcNAc2, whereas those from the wild-type strain consisted of oligosaccharides with heterogeneous sizes,
Man8GlcNAc2to Man12GlcNAc2. Digestion with ?-1,2- and ?-1,6-mannosidase of the oligosaccharides from
the wild-type and Yloch1 mutant strains strongly supported the possibility that the Yloch1 mutant strain has
a defect in adding the first ?-1,6-linked mannose to the core oligosaccharide. Taken together, these results
indicate that YlOCH1 plays a key role in the outer-chain mannosylation of N-linked oligosaccharides in Y.
lipolytica. Therefore, the Yloch1 mutant strain can be used as a host to produce glycoproteins lacking the
outer-chain mannoses and further developed for the production of therapeutic glycoproteins containing
Yeast can secrete a variety of proteins in much the same way
that mammalian cells do. The presence of yeast-specific outer-
chain mannosylation, however, has been a primary hindrance
to the exploitation of yeast for therapeutic glycoprotein pro-
duction, because glycoproteins decorated with yeast-specific
glycans are immunogenic and show poor pharmacokinetic
properties in humans (1, 24). In the budding yeast Saccharo-
myces cerevisiae, the N-linked oligosaccharides assembled on
glycoproteins include hypermannose structures with outer
chains that may contain up to 200 mannose units (6). Elonga-
tion of the outer chain is initiated by the Och1 protein, which
adds the first ?-1,6-linked mannose to the core N-linked oli-
gosaccharides upon their arrival in the Golgi apparatus in S.
cerevisiae (17). Following the addition of the first ?-1,6-man-
nose by Och1p, the core oligosaccharide is elongated by addi-
tional ?-1,6-mannosyltransferases, Mnn9p and Van1p, which
extend the ?-1,6-linked polymannose backbone, and the core
oligosaccharide is further branched by the addition of ?-1,2-
and ?-1,3-linked mannoses (5, 8). Other yeast species, including
Pichia pastoris, Hansenula polymorpha, and Schizosaccha-
romyces pombe, also use the Och1 protein to extend the man-
nose outer chain of N-glycans (14, 24, 26, 28). Therefore, the
elimination of the Och1 protein was performed to block the
yeast-specific outer-chain mannosylation, followed by further
engineering of yeast N-glycosylation pathways for the produc-
tion of glycoproteins with human-compatible oligosaccharides
(14, 17, 26).
Yarrowia lipolytica, a heterothallic yeast, is presently consid-
ered to be a good potential host for heterologous gene expres-
sion due to its ability to secrete large amounts of extracellular
proteins (4, 12, 21) and the simplicity with which it is cultivated
to a high cell density (13). Despite the potential of Y. lipolytica
as a valuable host, little information on the structural charac-
teristics of N-linked oligosaccharides of Y. lipolytica glycopro-
teins is available. It was recently reported that Y. lipolytica
OCH1 (YlOCH1), the homolog of S. cerevisiae OCH1 encoding
an ?-1,6-mannosyltransferase, may have a minor role in N
glycosylation (3). In our study, however, glycoproteins secreted
from a Yloch1 mutant strain showed more homogeneous band
patterns upon sodium dodecyl sulfate-polyacrylamide gel elec-
trophoresis (SDS-PAGE) than the wild-type strain. Therefore,
we have investigated the function of YlOCH1 by systematically
analyzing the structures of N-linked oligosaccharides assem-
bled on recombinant proteins secreted from the Yloch1 null
mutant strain, and we present several lines of evidence that
YlOch1p is a key enzyme responsible for adding the first
?-1,6-mannose residue onto the core oligosaccharide,
Man8GlcNAc2, in Y. lipolytica. The Yloch1 mutant strain,
which secretes glycoproteins lacking the outer-chain manno-
ses, can be used as a starting host for further glycoengineering
to produce human-compatible oligosaccharides.
* Corresponding author. Mailing address for Jeong-Yoon Kim:
School of Bioscience and Biotechnology, Chungnam National Univer-
sity, Daejeon 305-764, Korea. Phone: 82-42-821-6419. Fax: 82-42-822-
7367. E-mail: email@example.com. Mailing address for Hyun Ah Kang:
Korea Research Institute of Bioscience and Biotechnology, Daejeon
305-333, Korea. Phone: 82-42-860-4378. Fax: 82-42-860-4594. E-mail:
?Published ahead of print on 18 May 2007.
MATERIALS AND METHODS
Strains, plasmids, and media. The Y. lipolytica and S. cerevisiae strains used in
this study are described in Table 1. The yeast strains were routinely grown in
YPD (1% yeast extract, 2% Bacto peptone, and 2% glucose) at 28°C (Y. lipoly-
tica) or 30°C (S. cerevisiae). Synthetic complete medium was composed of 0.67%
yeast nitrogen base without amino acids (Difco), 2% glucose, and a dropout
amino acid mixture including all of the amino acids required. When required,
0.625 mg of 5?-fluoroorotic acid/ml was added to solid medium for the selection
of Ura auxotrophic strains. Drug sensitivity was assayed by spotting serially
diluted yeast cultures onto YPD solid medium containing 30 ?g of hygromycin
B (Sigma)/ml, 20 ?g of calcofluor white (Sigma)/ml, 100 ?g of Congo red (Junsei,
Japan)/ml, 0.05% SDS (Sigma), or 7 mM sodium orthovanadate (Sigma).
The plasmid pYEp352GAPII, containing the S. cerevisiae glyceraldehyde-3-
phosphate dehydrogenase (GAPDH) gene promoter and terminator (18), was
used as a backbone vector for the expression of the Y. lipolytica OCH1 gene in S.
cerevisiae. The pIMR53-AUX vector (11), containing the XPR2 promoter and
terminator and the URA3 gene, was used for the secretory expression of
Trichoderma reesei endoglucanase I (EGI) and Y. lipolytica lipase tagged with six
histidine residues. To express EGI and lipase, the recombinant Y. lipolytica cells
were cultivated on YPDm medium (1% yeast extract, 1% proteose peptone, 1%
glucose, and 50 mM phosphate-buffered saline, pH 6.8) at 28°C.
Recombinant DNA techniques and gene disruption. Recombinant DNA tech-
niques, Southern blot hybridization, and transformation were carried out essen-
tially as described by Sambrook and Russell (23). PCRs were performed with
ExTaq polymerase (Takara, Japan) using the GeneAmp PCR system 2400
Two pairs of primer sets, YO1F1 (5?-ACTTTTTGCATCTGCGGAC-3?) and
and YO1F2 (5?-AGATCT ACGGATCCATGGGACCGACTCTGTCTTCGA-
3?) and YO1R2 (5?-CATCCTCCTGATATACGC-3?), were designed to amplify
the 5? and 3? flanking regions of the YlOCH1 gene. The amplified PCR frag-
ments were fused by performing PCR using the YO1F1 and YO1R2 primers, and
the product was subcloned into the pGEMT-Easy vector (Promega). The tc-
URA3-tc cassette (11) was inserted into the linker site of the fused PCR product.
This disruption cassette was made linear by digestion with NotI and used to
transform Y. lipolytica. Correct disruption was confirmed by PCR and Southern
blot analysis. URA3 of the integrated tc-URA3-tc cassette was popped out of the
Yloch1? (URA3) strain by growing the strain on synthetic complete medium
containing 0.0625% 5?-fluoroorotic acid at 28°C for 5 days. The resulting
Yloch1? (ura3) strain was again transformed with the EGI and lipase expression
vectors, pAUX-EGI and pAUX-YlLIP2.
Purification and analysis of EGI and lipase. Culture supernatants containing
His6-tagged EGI and lipase were concentrated by ultrafiltration (YM30 mem-
brane, Millipore). The 100-fold-concentrated culture supernatants were dialyzed
against 50 mM sodium phosphate (pH 6.0) and 300 mM NaCl, and His6-tagged
EGI and lipase were purified by a His6-tagged affinity column using the A ´KTA
Prime chromatography system (Amersham Pharmacia Biotech AB, Sweden).
Western blot analysis of the purified His6-tagged EGI was performed with the
polyclonal immunoglobulin G Penta-His antibodies (QIAGEN, Germany), and
the EGI was detected using the DIG kit (Roche, Germany). Deglycosylation with
endoglycosidase H (endo H) was performed in accordance with the instructions
of the manufacturer (New England Biolabs). Activity staining of T. reesei EGI
was carried out as follows. After the cells expressing EGI were grown in 3 ml of
YPDm medium for 30 h, the culture supernatant was collected by centrifugation.
Each sample (16 ?l of supernatant) was subjected to electrophoresis on a 10%
PAGE gel. After the completion of electrophoresis, the substrate gel, which
contained 50 mM sodium citrate (pH 5.6), 1% carboxyl methyl cellulose, and
1.5% agarose, was overlaid with the PAGE gel. Both gels were wrapped and
incubated at 30°C for 3 h. The substrate gel was stained with 1% Congo red
solution for 10 min and then washed several times with 1 M sodium chloride. For
invertase activity staining, cells were grown in 3 ml of YPD medium to an optical
density at 600 nm of 6, transferred into YPsuc medium (2% yeast extract, 2%
Bacto peptone, and 1% sucrose) in order to derepress invertase expression, and
then incubated at 25°C for 3 h. The cell pellets were homogenized with acid-
washed glass beads (425 to 600 ?m in diameter) in PAGE sample buffer (125
mM Tris-HCl [pH 6.8], 1% ?-mercaptoethanol, 15% glycerol, 3% SDS, 0.1%
bromophenol blue, 2 mM phenylmethylsulfonyl fluoride) and were then sub-
jected to 5% PAGE. The gel was incubated in a solution containing 0.1 M
sodium acetate (pH 5.1) and 0.1 M sucrose for 30 min at 37°C, washed with H2O,
and then boiled in a 0.5 N NaOH solution containing 2,3,5-triphenyltetrazolium
chloride for the visualization of proteins with invertase activity.
Analysis of N-linked oligosaccharides. N-linked oligosaccharides were re-
leased from 200 ?g of the purified EGI and lipase by PNGase F (New England
Biolabs). The oligosaccharides were labeled at their reducing ends with 2-amino-
pyridine (PA) by using the PALSTATION pyridylamination reagent kit
(Takara Shuzo Co., Japan). After pyridylamination, the samples were purified by
Sephadex G-15 spin columns (Amersham Pharmacia Biotech AB, Sweden) to
remove residual PA. The digestion of PA-labeled oligosaccharides with ?-1,2-
mannosidase from Aspergillus saitoi (Glyko, Japan) or ?-1,6-mannosidase from
Xanthomonas manihotis (New England Biolabs) was carried out according to the
instructions of the manufacturer.
Size fractionation high-performance liquid chromatography (HPLC) was per-
TABLE 1. Yeast strains used in this study
MATA ade1 ura3 xpr2
MATA ade1 ura3 xpr2 och1::tc-URA3-tc
MATA ade1 ura3 xpr2 och1::tc
SMS397A harboring pAUX-EGI
Yloch1? harboring pAUX-EGI
SMS397A harboring pAUX-YlLIP2
Yloch1? harboring pAUX-YlLIP2
MATa leu2 ura3 trp1 ade2 his3
MATa leu2 ura3 trp1 ade2 his3 och1?::TRP1
MATa leu2 ura3 trp1 ade2 his3 och1?::hisG mnn1?::hisG
W303-1A harboring pYEp352GAPII
Scoch1? harboring pYEp352GAPII
TOY137 harboring pYEp352GAPII
W303-1A harboring pYEp352GAPII-YlOCH1
Scoch1? harboring pYEp352GAPII-YlOCH1This study
TOY137 harboring pYEp352GAPII-YlOCH1 This study
VOL. 73, 2007PRODUCTION OF GLYCOPROTEINS IN ENGINEERED Y. LIPOLYTICA 4447
formed with a Shodex Asahipak NH2P-50 column (0.46 by 25 cm; Showa Denko
K. K., Japan) at a rate of 1.0 ml/min. The column was equilibrated with a solution
comprising 80% solvent A (200 mM acetic acid-triethylamine [pH 7.3]–acetoni-
trile, 1:9) and 20% solvent B (200 mM acetic acid-triethylamine [pH 7.3]–
acetonitrile, 9:1). After sample injection, the proportion of solvent B was in-
creased linearly up to 95% for 52 min. PA oligosaccharides were detected by
fluorescence (excitation ?, 320 nm, and emission ?, 400 nm) with a Waters 2475
FIG. 1. Functional complementation of S. cerevisiae och1 mutation by the Y. lipolytica OCH1 gene. (A) Spotting analysis of growth phenotype.
The S. cerevisiae wild-type W303-1A, och1?, and TOY137 (och1? mnn1? mnn4?) strains were transformed with a YlOCH1 expression vector,
pYEp352GAPII-YlOCH1, and a control vector, pYEp352GAPII. Serial (1/10) dilutions of cells were spotted onto plates containing YPD, YPD
with 20 ?g of hygromycin B/ml, or YPD with 4 mM vanadate. The plates were incubated for 3 days at 25°C. (B) Activity staining of invertase. Cells
grown to an optical density at 600 nm of 6 in 3 ml of YPD medium were transferred onto YPsuc medium (2% yeast extract, 2% Bacto peptone,
and 1% sucrose) to derepress invertase expression and were then incubated at 25°C for 3 h. Lanes 1, 3, and 5 contain cell lysates of W303-1A,
Scoch1?, and TOY137 strains transformed with a control vector, respectively. Lanes 2, 4, and 6 contain cell lysates of W303-1A, Scoch1?, and
TOY137 strains transformed with pYEp352GAPII-YlOCH1, respectively.
FIG. 2. Disruption of YlOCH1. (A) Schematic representation of various genomic fragments containing YlOCH1 or mutated alleles. The 1.2-kb
fragment of YlOCH1 was replaced with the tc-YlURA3-tc cassettes flanked by regions homologous to YlOCH1 by in vivo DNA recombination.
(B) Southern blot analysis of the YlOCH1 disruption. Lane 1, SMS397A (wild type); lanes 2 and 3, Yloch1? (URA3) strain (Yloch1::tc-YlURA3-tc);
lanes 4 and 5, Yloch1? strain (ura3) (Yloch1::tc). Each genomic DNA strand was digested with PstI and XbaI and hybridized with the
digoxigenin-labeled 690-bp DNA fragment.
4448SONG ET AL.APPL. ENVIRON. MICROBIOL.
In vitro ?-1,6-mannosyltransferase activity assay. Membrane fractions were
obtained as described previously (14). Cells grown in YPD medium were har-
vested, washed with 1% KCl, and resuspended in 5 ml of PMS buffer (50 mM
Tris-HCl [pH 7.5], 10 mM MnCl2, 1 mM phenylmethylsulfonyl fluoride, 5%
glycerol, and 2 ?g of each protease inhibitor [antipain, chymostatin, leupeptin,
and pepstatin A]/ml). Glass beads (425 to 600 ?m in diameter) were added to
half of the cell suspension volume, and the mixture was homogenized four times
for 1 min at 4°C. Homogenates were centrifuged at 10,000 ? g for 20 min, and
the supernatant obtained was further centrifuged at 100,000 ? g for 1 h. High-
speed pellets were collected and resuspended in PMS buffer, and protein con-
centrations were determined using a protein assay agent (Bio-Rad). The ?-1,6-
mannosyltransferase activity assay was performed as described by Nakajima and
Ballou (16). One hundred micrograms of high-speed pellet proteins was incu-
bated in 100 ?l of 50 mM Tris-HCl (pH 7.5) buffer containing 10 mM MnCl2, 1
mM GDP-mannose, 0.5 mM 1-deoxymannojirimycin, and 100 pmol of the
Man8GlcNAc2-PA acceptor at 30°C for 2 h. The reaction was terminated by
boiling at 99°C for 5 min, and the reaction mixture was filtered through an
Ultrafree-MC membrane (10,000-Da cutoff; Millipore) and analyzed by HPLC.
Matrix-assisted laser desorption ionization–time of flight (MALDI-TOF)
analysis. N-linked oligosaccharides released from 100 ?g of the purified recom-
binant EGI were isolated by using a porous graphitized carbon column (Alltech).
To elute neutral and acidic glycans, 900 ?l of a solution consisting of 25%
(vol/vol) acetonitrile (Applied Biosystems) and 0.05% (vol/vol) trifluoroacetic
acid (Aldrich) was added onto the column. The effluent was collected, lyophi-
lized on a freeze-dry system (Ilshin, Korea), and dissolved in 10 ?l of HPLC-
grade water. The glycan sample (0.7 ?l) was mixed with 0.7 ?l of a mixture of
6-aza-2-thiothymine and 2,5-dihydroxybenzoic acid (vol/vol, 1:1) as a matrix, and
the mixture was loaded onto a ground steel MSP 96 target (microScout target;
Bruker Daltonics, Germany). The mass spectrum was analyzed with a microflex
mass spectrometer (Bruker Daltonics, Germany) in the positive reflector mode
(for the detection of neutral sugars) or in the negative linear mode (for the
detection of acidic sugars).
YlOCH1 is a functional homolog of ScOCH1 encoding an
YlOCH1 in outer-chain biosynthesis has not yet been com-
pletely defined despite the previous study by Barnay-Verdier
et al. (3), we wanted to investigate whether YlOCH1 is a
functional homolog of ScOCH1. We first examined the effect
of YlOCH1 expression on the growth phenotypes of S. cerevi-
siae och1 mutant strains, such as sensitivity to hygromycin B
and resistance to orthovanadate, which are general character-
istics observed in yeast mutant strains defective in N glycosyl-
ation (2, 7, 18). A complementation vector, pYEp352GAPII-
YlOCH1, which expresses YlOCH1 under the control of the S.
cerevisiae GAPDH gene promoter, was constructed and intro-
duced into the S. cerevisiae wild-type and och1? mutant strains.
In contrast to the strains transformed with a control plasmid,
pYEp352GAPII, the S. cerevisiae och1 mutant strains trans-
formed with pYEp352GAPII-YlOCH1 recovered the pheno-
types of the wild-type strain, including the normal growth rate,
resistance to hygromycin B, and sensitivity to sodium or-
thovanadate (Fig. 1A). Next, we performed an invertase gel
electrophoretic mobility assay in order to determine whether
YlOCH1 can eliminate the glycosylation defect in the Scoch1?
mutant strains. The invertases secreted from the Scoch1? mu-
tant cells migrated faster than those secreted from the wild-
type cells (Fig. 1B), which indicates the lack of heavy glycosyl-
ation in the invertases from the Scoch1? mutants. It is evident,
however, that the introduction of the YlOCH1 gene restored
the ability of the Scoch1? mutant strains to secrete fully gly-
cosylated invertase. Along with the recovery of the normal
growth phenotype, the elimination of the glycosylation defect
in the S. cerevisiae och1 mutant strains by the expression of
Since the role of
YlOCH1 demonstrates that YlOCH1 is a functional homolog
of S. cerevisiae OCH1.
Effect of YlOCH1 deletion on cell growth and N glycosyla-
tion. To study the in vivo function of the YlOCH1 gene in Y.
lipolytica, the Yloch1 null mutant (Yloch1?) strain was con-
structed and confirmed by Southern analysis (Fig. 2). Interest-
ingly, unlike the S. cerevisiae och1 mutant, the Yloch1? mutant
grew as well as the wild-type strain both under normal growth
conditions and at a high temperature, 34°C (data not shown).
To test whether the Yloch1? mutant has the phenotypes of
defective glycosylation, we grew the Yloch1? mutant on me-
FIG. 3. Phenotypic analysis of the Yloch1 mutant strain. (A) Van-
adate resistance and hygromycin B sensitivity. (B) Calcofluor white
and Congo red sensitivity. The wild-type and Yloch1 mutant cells were
grown in YPD, and 10-?l serial (1/10) dilutions of each strain were
spotted onto YPD plates containing 7 mM vanadate, 30 ?g of hygro-
mycin B/ml, 20 ?g of calcofluor white (CFW)/ml, or 100 ?g of Congo
red (CR)/ml. The plates were incubated for 2 days at 28°C. (C) West-
ern blot analysis of T. reesei EGI secreted from the wild-type and
Yloch1? cells. The secreted EGI from the wild-type and Yloch1? cells
was purified, separated by 10% SDS-PAGE, and detected with anti-
His antibody. Lane 1, EGI secreted from the wild type; lane 2, EGI
treated with endo H; lane 3, EGI secreted from Yloch1? cells. (D) The
supernatants (16 ?l) from cell cultures grown in 3 ml of YPD medium
for 30 h were subjected to 10% PAGE without SDS. The substrate gel
containing 1% carboxyl methyl cellulose was overlaid with the PAGE
gel. Both gels were wrapped and incubated at 30°C for 3 h. The
substrate gel was stained with 1% Congo red solution for 10 min. Lane
1, EGI secreted from the wild type; lane 2, EGI treated with endo H;
lane 3, EGI secreted from Yloch1? cells.
VOL. 73, 2007PRODUCTION OF GLYCOPROTEINS IN ENGINEERED Y. LIPOLYTICA4449
dium containing vanadate or hygromycin B. The Yloch1? mu-
tant was resistant to vanadate and sensitive to hygromycin B
compared to the wild-type strain (Fig. 3A), indicating that the
Yloch1? mutant was defective in terms of glycosylation. The
Yloch1? mutant also showed poor growth on medium contain-
ing calcofluor white or Congo red. However, the addition of an
osmotic stabilizer, 1 M sorbitol, to the medium restored growth
to the level of wild-type growth (Fig. 3B), although supplemen-
tation with 0.5 M sorbitol was shown to be insufficient to
support the growth of the Yloch1? mutant (3). This result
suggested that the Yloch1? mutant has alterations in its cell
wall composition, probably in mannoproteins (7, 22). The phe-
notypes of the glycosylation defect and the alteration of cell
wall composition and structure in the Yloch1? mutant imply
that the function of YlOCH1 may be important for the N-
glycosylation process in Y. lipolytica.
To further examine protein N glycosylation in the Yloch1?
mutant, T. reesei EGI, which has nine potential N-linked gly-
cosylation sites (9), was expressed in Y. lipolytica. The Yloch1?
mutant produced relatively homogeneously sized EGI proteins
compared to the wild-type strain, which secreted rather heter-
ogenous forms of EGI, as judged by the smeared pattern on
Western blots; this smearing was eliminated upon treatment
with endo H, which removes oligomannosidic N-glycans (Fig.
3C and D). This result led us to believe that the Yloch1?
mutant certainly lacked hyperglycosylation activity and that
YlOch1p might be involved in the outer-chain elongation of
N-linked oligosaccharides in Y. lipolytica.
Membrane fraction of Yloch1? mutant lacks an initiating
?-1,6-mannosyltransferase activity. To determine whether the
Yloch1? mutant lacks an initiating ?-1,6-mannosyltransferase
that uses the core oligosaccharide as a substrate, solubilized
membrane fractions were prepared from the wild-type and
Yloch1? mutant strains and used as enzyme sources for an in
vitro ?-1,6-mannosyltransferase assay. Whereas, as expected,
the peak corresponding to Man9GlcNAc2-PA was detected as
a reaction product in the membrane fraction of the wild-type
strain, the acceptor, Man8GlcNAc2-PA, was not converted into
Man9GlcNAc2-PA by the membrane fraction of the Yloch1?
mutant (Fig. 4). Furthermore, the Man9GlcNAc2-PA product
core oligosaccharide was converted into Man5GlcNAc2-PA
when the reaction products were treated with ?-1,2-mannosi-
dase from A. saitoi, which specifically removes ?-1,2-linked
mannose residues at the nonreducing ends of manno-oligosac-
charides (Fig. 4). These results strongly indicate that YlOCH1
encodes an initiation-specific ?-1,6-mannosyltransferase acting
on the core oligosaccharide.
Structural analysis of N-linked oligosaccharides released
from EGI and lipase in the Yloch1? mutant. To obtain de-
tailed information on the in vivo function of YlOch1p in N
glycosylation, we analyzed the structures of N-glycans assem-
bled on EGI secreted from the Y. lipolytica wild-type and
Yloch1? strains. In the case of the wild-type strain, two major
peaks corresponding to Man8GlcNAc2and Man9GlcNAc2in
the chromatogram of the oligosaccharides released from EGI
were observed; some other minor peaks also appeared (Fig.
5A). In the case of the Yloch1? strain, however, we found a
single major peak for Man8GlcNAc2, indicating that the
Yloch1? mutant was defective in its ability to synthesize oli-
gosaccharides larger than Man8GlcNAc2(Fig. 5A, panels a
and c). To further investigate the structures of N-linked oligo-
saccharides synthesized from the wild-type and Yloch1? mu-
tant strains, we analyzed the oligosaccharides by treating them
with ?-1,2- and ?-1,6-mannosidases. The Man8GlcNAc2and
Man9GlcNAc2oligosaccharides generated from the wild-type
FIG. 4. In vitro activity analysis of Y. lipolytica OCH1 gene product. Solubilized membrane fractions from the Y. lipolytica wild type (a and b)
and Yloch1? (c and d) strains were prepared as enzyme sources, and Man8GlcNAc2-PA was used as an acceptor. The reaction products were
analyzed by HPLC. The reaction products treated with ?-1,2-mannosidase were also analyzed by HPLC. M5, Man5GlcNAc2-PA; M6,
Man6GlcNAc2-PA; M8, Man8GlcNAc2-PA; M9, Man9GlcNAc2-PA.
4450SONG ET AL.APPL. ENVIRON. MICROBIOL.
and Yloch1? strains were treated with ?-1,2-mannosidase
from A. saitoi. Man8GlcNAc2was converted completely into
Man5GlcNAc2, and Man9GlcNAc2 was converted com-
pletely into Man6GlcNAc2(Fig. 5B, panels a and c). The
Man5GlcNAc2and Man6GlcNAc2oligosaccharides were sub-
sequently digested with ?-1,6-mannosidase from X. manihotis,
a highly specific exoglycosidase that removes unbranched
rides. The Man6GlcNAc2oligosaccharides were converted into
D-mannopyranosyl residues from oligosaccha-
Man5GlcNAc2(Fig. 5B, panels b and d), which indicates that
the Man6GlcNAc2species has a single unbranched ?-1,6-
linked mannose. These results demonstrate that the Yloch1?
mutant strain has a defect in terms of the addition of the
first ?-1,6-mannose residue onto the core oligosaccharide,
In order to demonstrate that the function of YlOCH1 is
generally required for the addition of the first ?-1,6-mannose
residue onto the core oligosaccharide, Man8GlcNAc2, in Y.
FIG. 5. HPLC analysis of N-linked oligosaccharides assembled on glycoproteins. (A) Chromatograms of the N-linked oligosaccharides released
from EGI (a and c) and lipase (b and d) from the wild-type and Yloch1? strains. (B) Chromatograms of the N-linked oligosaccharides released
from EGI secreted by the wild-type (a and b) and Yloch1? strains (c and d) after treatment with ?-1,2-mannosidase and ?-1,6-mannosidase. The
elution times of authentic PA-sugar chains are indicated by arrows. M5, Man5GlcNAc2-PA; M6, Man6GlcNAc2-PA; M7, Man7GlcNAc2-PA; M8,
Man8GlcNAc2-PA; M9, Man9GlcNAc2-PA; M10, Man10GlcNAc2-PA.
VOL. 73, 2007 PRODUCTION OF GLYCOPROTEINS IN ENGINEERED Y. LIPOLYTICA4451
lipolytica, we analyzed the N-linked oligosaccharide structure
of an endogenous glycoprotein, lipase (YlLip2p), which is a
secreted protein with two potential N-linked glycosylation sites
(20). The HPLC profiles of the oligosaccharides assembled on
the lipases secreted from the wild-type and Yloch1? strains
were the same as those of the oligosaccharides assembled on
EGI (Fig. 5A, panels b and d). Therefore, it is clear that
YlOCH1 encodes an enzyme that initiates the outer-chain
elongation by adding the first ?-1,6-linked mannose to the core
oligosaccharide in Y. lipolytica.
There were two minor peaks appearing much later than the
major peaks in the HPLC profiles of the oligosaccharides from
the wild-type strain (Fig. 5). Judging from a comparison of the
HPLC profiles with the data from the study by Wang et al.
(25), in which the mannosylphosphate transferase activity of S.
cerevisiae Mnn6p was assayed, we suspected that the two minor
peaks might represent phosphate-containing oligosaccharides.
To determine whether N-linked oligosaccharides in Y. lipoly-
tica are phosphorylated, we analyzed N-glycans derived from
the recombinant EGI by using MALDI-TOF mass spectrom-
etry. The glycans were analyzed in the positive reflector mode
for the detection of neutral sugars (Fig. 6A) and in the negative
linear mode for the detection of acidic sugars (Fig. 6B). The
major peaks detected in the negative mode corresponded to
molecular masses higher than those of the neutral sugars,
which nearly equaled the molecular masses of mannosylphos-
phorylated Man7GlcNAc2to Man9GlcNAc2(Man8PGlcNAc2
to Man10PGlcNAc2) species. These results indicate that N-
linked oligosaccharides assembled on glycoproteins in Y. lipo-
lytica are modified by mannosylphosphate. In addition, the
observation that the overall glycan mass spectra are consistent
with the HPLC data (Fig. 5A and 6A) confirms that the
Yloch1? mutant is defective in synthesizing oligosaccharides
larger than Man8PGlcNAc2.
The Och1 protein has been proven to be an initiating ?-1,6-
mannosyltransferase that plays a key role in the addition of the
first mannose to the core oligosaccharide in several yeast spe-
cies, including S. cerevisiae, P. pastoris, Schizosaccharomyces
pombe, and H. polymorpha (14, 15, 17, 19, 28). However, Y.
lipolytica Och1p was suspected to have a minor role in the
outer-chain elongation of N glycosylation or to be important
only for specific proteins (3). In this study, we reexamined the
Y. lipolytica OCH1 gene and analyzed the structure of the
N-linked oligosaccharide from the Yloch1? mutant based on
our observation that the recombinant glycoprotein EGI se-
creted from the Yloch1? mutant appeared to be less heavily
glycosylated than the recombinant EGI from the wild-type
strain. Here, we present several lines of strong evidence indi-
cating that YlOCH1 encodes an ?-1,6-mannosyltransferase
that plays an important role in the addition of the first man-
nose to the core oligosaccharide in Y. lipolytica, as in other
yeasts. First, the ability of the YlOCH1 gene to eliminate the
defects of the S. cerevisiae och1 mutant, i.e., retarded growth
rate and a reduction in the sizes of N-linked oligosaccharides,
demonstrated that YlOCH1 is a functional homolog of
ScOCH1. Second, the membrane fraction of the Ylochl? mu-
tant lacked an initiation-specific ?-1,6-mannosyltransferase ac-
tivity. Third, the N-linked oligosaccharides attached to the
recombinant EGI and the endogenous lipase secreted from the
Ylochl? mutant cells were composed of a single predominating
species, Man8GlcNAc2, and some minor ones, which is consis-
tent with the rapid mobility of the glycoproteins from the
Ylochl? mutant cells. Last, the structural analysis of N-glycans
revealed that the N-linked oligosaccharides from Ylochl? did
not contain additional ?-1,6-linked mannoses. Taken together,
these data provide clear evidence that the major function of
YlOch1p is to initiate ?-1,6-linked mannose elongation of the
core oligosaccharide, Man8GlcNAc2.However, we cannot ex-
clude the possible presence of another ?-1,6-mannosyltrans-
ferase that may play a minor role in the outer-chain elongation
of certain glycoproteins in Y. lipolytica.
In S. cerevisiae, N-linked oligosaccharides are capped with
immunogenic ?-1,3-linked mannose residues (16, 27), and
?-1,2-linked mannoses of N-linked oligosaccharides are not
digested with ?-1,2-mannosidase if they are attached to ?-1,3-
linked mannose. However, in Y. lipolytica, Man8GlcNAc2was
converted completely into Man5GlcNAc2upon treatment with
?-1,2-mannosidase. This result, along with the finding that no
open reading frame that is homologous to S. cerevisiae MNN1
appears in the Y. lipolytica genome database (http://cbi.labri.fr
/Genolevures/), strongly suggests that the core oligosaccha-
rides from Y. lipolytica may not be terminally capped by ?-1,3-
linked mannose residues.
Analysis using MALDI-TOF mass spectrometry suggests
that N-linked glycans assembled on secreted glycoproteins
from Y. lipolytica contain phosphates and that the two minor
peaks appearing in the HPLC profiles of the oligosaccharides
from the wild-type strain (Fig. 5) may represent mannosylphos-
phorylated forms of Man7GlcNAc2to Man9GlcNAc2(Fig. 6).
The absence of the latter of the two minor peaks in the chro-
matograms corresponding to the oligosaccharides from the
Yloch1? mutant strain and the ?-1,6-mannosidase-treated oli-
gosaccharides from the wild-type strain (Fig. 5B) indicates that
the difference between the two peaks comes from an additional
?-1,6-linked mannose at the latter peak. Therefore, we think
that mannosylphosphates are transferred to the core forms of
N-glycans in Y. lipolytica. Further studies are presently in
progress to identify genes involved in the mannosylphosphor-
ylation of N-linked oligosaccharides in Y. lipolytica.
Recently, many studies have reported the engineering of
FIG. 6. MALDI-TOF mass spectrometry analysis of N-linked oligosaccharides assembled on EGI. (A) Mass spectra analyzed in the positive
reflector mode for the detection of neutral sugars released from the recombinant EGI secreted from the wild-type (top panels) and Yloch1?
(bottom panels) strains. The intermediate peaks for the wild type, designated a, b, and c, are assumed to represent the mannosylphosphorylated
forms of Man7GlcNAc2, Man8GlcNAc2, and Man9GlcNAc2, based on their m/z of 1,662.792, 1,808.617, and 1,970.866, respectively. (B) Mass
spectra analyzed in the negative reflector mode for the detection of acidic sugars released from the recombinant EGI secreted from the wild-type
(top panel) and Yloch1? (bottom panel) strains. The analyzed glycan samples are free, nonreduced forms without any labeling.
VOL. 73, 2007PRODUCTION OF GLYCOPROTEINS IN ENGINEERED Y. LIPOLYTICA 4453
novel yeast strains secreting therapeutic glycoproteins with hu- Download full-text
man-compatible types of N-linked oligosaccharides (10, 14, 24,
26). Y. lipolytica is considered to be one of the most suitable
strains for heterologous protein expression (12) and can be
developed for the production of glycoproteins. In this study, we
found that the Yloch1? mutant strain, despite the complete
block of the outer-chain elongation of N-linked oligosaccha-
rides, could grow as well as the wild-type strain under normal
growth conditions. This growth property of the Yloch1? mu-
tant strain will be very useful for the further development of Y.
lipolytica as a host for the production of glycoproteins. Along
with the possibility of a lack of the immunogenic terminal
?-1,3-mannose linkages in Y. lipolytica, the Yloch1? mutant
strain can be exploited as a platform strain for developing
another potential yeast system that can produce recombinant
glycoproteins with human-compatible oligosaccharides.
We thank Y. Jigami for providing S. cerevisiae och1 mutant strains
and Jeong Mi Lee for MALDI-TOF analysis.
This work was supported by grants from the 21C Frontier Microbial
Genomics and Application Center Program (MG05-0309-3-0) to J.-Y.
Kim and from the MOCIE of the Republic of Korea (Next Generation
New Technology Development Program) to H. A. Kang. J.-N. Park is
presently a postdoctoral fellow of BK21.
1. Ballou, C. E. 1990. Isolation, characterization, and properties of Saccharo-
myces cerevisiae mnn mutants with nonconditional protein glycosylation de-
fects. Methods Enzymol. 185:440–470.
2. Ballou, L., R. A. Hitzeman, M. S. Lewis, and C. E. Ballou. 1991. Vanadate-
resistant yeast mutants are defective in protein glycosylation. Proc. Natl.
Acad. Sci. USA 88:3209–3212.
3. Barnay-Verdier, S., A. Boisrame, and J. M. Beckerich. 2004. Identification
and characterization of two alpha-1,6-mannosyltransferases, Anl1p and
Och1p, in the yeast Yarrowia lipolytica. Microbiology 150:2185–2195.
4. Barth, G., and C. Gaillardin. 1997. Physiology and genetics of the dimorphic
fungus Yarrowia lipolytica. FEMS Microbiol. Rev. 19:219–237.
5. Cohen, R. E., L. Ballou, and C. E. Ballou. 1980. Saccharomyces cerevisiae
mannoprotein mutants. Isolation of the mnn5 mutant and comparison with
the mnn3 strain. J. Biol. Chem. 255:7700–7707.
6. Dean, N. 1999. Asparagine-linked glycosylation in the yeast Golgi. Biochim.
Biophys. Acta 1426:309–322.
7. Dean, N. 1995. Yeast glycosylation mutants are sensitive to aminoglycosides.
Proc. Natl. Acad. Sci. USA 92:1287–1291.
8. Devlin, C., and C. E. Ballou. 1990. Identification and characterization of a
gene and protein required for glycosylation in the yeast Golgi. Mol. Micro-
9. Garcia, R., J. A. Cremata, O. Quintero, R. Montesino, K. Benkestock, and J.
Stahlberg. 2001. Characterization of protein glycoforms with N-linked neu-
tral and phosphorylated oligosaccharides: studies on the glycosylation of
endoglucanase 1 (Cel7B) from Trichoderma reesei. Biotechnol. Appl. Bio-
10. Gerngross, T. U. 2004. Advances in the production of human therapeutic
proteins in yeasts and filamentous fungi. Nat. Biotechnol. 22:1409–1414.
11. Hwang, J. S., D. H. Seo, and J.-Y. Kim. 2006. Soluble forms of YlCrh1p and
YlCrh2p, cell wall proteins of Yarrowia lipolytica, have beta-1,3-glycosidase
activity. Yeast 23:803–812.
12. Juretzek, T., M. Le Dall, S. Mauersberger, C. Gaillardin, G. Barth, and J.
Nicaud. 2001. Vectors for gene expression and amplification in the yeast
Yarrowia lipolytica. Yeast 18:97–113.
13. Kim, J. W., T. J. Park, D. D. Ryu, and J.-Y. Kim. 2000. High cell density
culture of Yarrowia lipolytica using a one-step feeding process. Biotechnol.
14. Kim, M. W., E. J. Kim, J.-Y. Kim, J. S. Park, D. B. Oh, Y. Shimma, Y. Chiba,
Y. Jigami, S. K. Rhee, and H. A. Kang. 2006. Functional characterization of
the Hansenula polymorpha HOC1, OCH1, and OCR1 genes as members of
the yeast OCH1 mannosyltransferase family involved in protein glycosyla-
tion. J. Biol. Chem. 281:6261–6272.
15. Murakami, K., and N. Sugio. January 1997. Production of glycoprotein.
Japanese patent JP 1997000261-A1.
16. Nakajima, T., and C. E. Ballou. 1975. Yeast manno-protein biosynthesis:
solubilization and selective assay of four mannosyltransferases. Proc. Natl.
Acad. Sci. USA 72:3912–3916.
17. Nakanishi-Shindo, Y., K. Nakayama, A. Tanaka, Y. Toda, and Y. Jigami.
1993. Structure of the N-linked oligosaccharides that show the complete loss
of alpha-1,6-polymannose outer chain from och1, och1 mnn1, and och1
mnn1 alg3 mutants of Saccharomyces cerevisiae. J. Biol. Chem. 268:26338–
18. Nakayama, K., T. Nagasu, Y. Shimma, J. Kuromitsu, and Y. Jigami. 1992.
OCH1 encodes a novel membrane bound mannosyltransferase: outer chain
elongation of asparagine-linked oligosaccharides. EMBO J. 11:2511–2519.
19. Nakayama, K., Y. Nakanishi-Shindo, A. Tanaka, Y. Haga-Toda, and Y.
Jigami. 1997. Substrate specificity of alpha-1,6-mannosyltransferase that ini-
tiates N-linked mannose outer chain elongation in Saccharomyces cerevisiae.
FEBS Lett. 412:547–550.
20. Pignede, G., H. Wang, F. Fudalej, C. Gaillardin, M. Seman, and J. M.
Nicaud. 2000. Characterization of an extracellular lipase encoded by LIP2 in
Yarrowia lipolytica. J. Bacteriol. 182:2802–2810.
21. Pignede, G., H. J. Wang, F. Fudalej, M. Seman, C. Gaillardin, and J. M.
Nicaud. 2000. Autocloning and amplification of LIP2 in Yarrowia lipolytica.
Appl. Environ. Microbiol. 66:3283–3289.
22. Ram, A. F., A. Wolters, R. Ten Hoopen, and F. M. Klis. 1994. A new
approach for isolating cell wall mutants in Saccharomyces cerevisiae by
screening for hypersensitivity to calcofluor white. Yeast 10:1019–1030.
23. Sambrook, J., and D. W. Russell. 2001. Molecular cloning: a laboratory
manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
24. Vervecken, W., V. Kaigorodov, N. Callewaert, S. Geysens, K. De Vusser, and
R. Contreras. 2004. In vivo synthesis of mammalian-like, hybrid-type N-
glycans in Pichia pastoris. Appl. Environ. Microbiol. 70:2639–2646.
25. Wang, X. H., K. Nakayama, Y. Shimma, A. Tanaka, and Y. Jigami. 1997.
MNN6, a member of the KRE2/MNT1 family, is the gene for mannosylphos-
phate transfer in Saccharomyces cerevisiae. J. Biol. Chem. 272:18117–18124.
26. Wildt, S., and T. U. Gerngross. 2005. The humanization of N-glycosylation
pathways in yeast. Nat. Rev. Microbiol. 3:119–128.
27. Yip, C. L., S. K. Welch, F. Klebl, T. Gilbert, P. Seidel, F. J. Grant, P. J.
O’Hara, and V. L. MacKay. 1994. Cloning and analysis of the Saccharomyces
cerevisiae MNN9 and MNN1 genes required for complex glycosylation of
secreted proteins. Proc. Natl. Acad. Sci. USA 91:2723–2727.
28. Yoko-o, T., K. Tsukahara, T. Watanabe, N. Hata-Sugi, K. Yoshimatsu, T.
Nagasu, and Y. Jigami. 2001. Schizosaccharomyces pombe och1(?) encodes
alpha-1,6-mannosyltransferase that is involved in outer chain elongation of
N-linked oligosaccharides. FEBS Lett. 489:75–80.
4454 SONG ET AL.APPL. ENVIRON. MICROBIOL.