Expression of heparan sulfate sulfotransferases in Kluyveromyces lactis and preparation of 3′-phosphoadenosine-5′-phosphosulfate

Article abstract of DOI:10.1093/glycob/cwr001

Heparan sulfate (HS) belongs to a major class of glycans that perform central physiological functions. Heparin is a specialized form of HS and is a clinically used anticoagulant drug. Heparin is a natural product isolated from pig intestine. There is a strong demand to replace natural heparin with a synthetic counterpart. Although a chemoenzymatic approach has been employed to prepare synthetic heparin, the scale of the synthesis is limited by the availability of sulfotransferases and the cofactor, 3′-phosphoadenosine- 5′-phosphosulfate (PAPS). Here, we present a novel method to produce secreted forms of sulfotransferases in the yeast cells, Kluyveromyces lactis. Five sulfotransferases including N-sulfotransferase, 2-O-sulfotransferase, 3-O-sulfotransferase 1 and 6-O-sulfotransferases 1 and 3 were expressed using this method. Unlike bacterial-expressed sulfotransferases, the yeast proteins can be directly used to modify polysaccharides without laborious purification. The yeast-expressed sulfotransferases also tend to have higher specific activity and thermostability. Furthermore, we demonstrated the possibility for the gram-scale synthesis of PAPS from adenosine 5'-triphosphate at only 1/5000th of the price purchased from a commercial source. Our results pave the way to conduct the enzymatic synthesis of heparin in large quantities.

Full text of DOI:10.1093/glycob/cwr001

Glycobiology vol. 21 no. 6 pp. 771–780, 2011
doi:10.1093/glycob/cwr001
Advance Access publication on January 11, 2011
Expression of heparan sulfate sulfotransferases
in Kluyveromyces lactis and preparation
of 3′-phosphoadenosine-5′-phosphosulfate
Xianxuan Zhou, Kasemsiri Chandarajoti,
in the extracellular matrix. HS interacts with a variety of pro-
Truong Quang Pham, Renpeng Liu, and Jian Liu1
teins to regulate many different biological systems, including
blood coagulation, viral/bacterial infection, tumor metastasis
and inflammatory response (Bernfield et al. 1999; Esko and
Selleck 2002). Heparin, the specialized form of HS, is a com-
monly used anticoagulant drug with the worldwide production
exceeding 100 t year−1 (Aquino et al. 2010). Pharmaceutical
heparin is primarily isolated from porcine intestinal mucosa
through a long supply chain that posts a potential risk of con-
taminants and adulteration (Liu et al. 2009). Recent worldwide
outbreak of contaminated heparin underscores the vulnerability
of heparin supply chain. Therefore, a method that can prepare
synthetic heparin in a confined facility is highly desirable
(Peterson et al. 2009). Chemical synthesis of short heparin
fragments is accomplished via a complex and inefficient
scheme. Attempts to synthesize any fragments larger than an
octasaccharide have been proved to be extremely challenging
(Petitou and van Boeckel 2004). Given the fact that the average
size for a full-length heparin is about 50 saccharide units, prep-
aration of heparin from a total synthetic approach is nearly
impossible. Preparation of heparin-like polysaccharide from a
bacteria capsular polysaccharide known as heparosan using a
chemical method has been reported. However, the method
yields 3-O-sulfated glucuronic acid unit, a residue that is not
believed to be present in natural heparin (Lindahl et al. 2005).
An enzymatic approach to prepare synthetic heparin has
recently been developed to overcome the difficulties encoun-
tered by chemical synthesis.
The HS biosynthetic pathway involves multiple enzymes,
including HS polymerase, epimerase and sulfotransferases
(Figure 1). HS polymerase is responsible for building the
polysaccharide backbone, containing the repeating unit of
-GlcUA-GlcNAc-. The backbone is then modified by
N-deacetylase/N-sulfotransferase (NDST, having two separate
domains displaying the activity of N-deacetylase and
N-sulfotransferase (NST), respectively), C5-epimerase (con-
verting GlcUA to IdoUA), 2-O-sulfotransferase (2-OST),
6-O-sulfotransferase (6-OST) and 3-O-sulfotransferase
(3-OST) to produce the fully elaborated HS. These enzymes
can be used to prepare synthetic heparin. The enzymatic syn-
thesis of HS from heparosan was first carried out at the micro-
gram scale (Kuberan et al. 2003). The milligram-scale
synthesis of HS was reported by our lab (Chen et al. 2005,
2007; Copeland et al. 2008). The success of a larger-scale
Division of Medicinal Chemistry and Natural Products, Eshelman School of
Pharmacy, University of North Carolina, Rm 303, Beard Hall, Chapel Hill,
NC 27599, USA
Received on October 25, 2010; revised on December 14, 2010; accepted on
December 31, 2010
Heparan sulfate (HS) belongs to a major class of glycans
that perform central physiological functions. Heparin is a
specialized form of HS and is a clinically used anticoagu-
lant drug. Heparin is a natural product isolated from pig
intestine. There is a strong demand to replace natural
heparin with a synthetic counterpart. Although a che-
moenzymatic approach has been employed to prepare
synthetic heparin, the scale of the synthesis is limited
by the availability of sulfotransferases and the cofactor,
3′-phosphoadenosine-5′-phosphosulfate (PAPS). Here, we
present a novel method to produce secreted forms of sulfo-
transferases in the yeast cells, Kluyveromyces lactis. Five
sulfotransferases including N-sulfotransferase, 2-O-sulfo-
transferase, 3-O-sulfotransferase 1 and 6-O-sulfotrans-
ferases 1 and 3 were expressed using this method. Unlike
bacterial-expressed sulfotransferases, the yeast proteins
can be directly used to modify polysaccharides without
laborious purification. The yeast-expressed sulfotrans-
ferases also tend to have higher specific activity and thermo-
stability. Furthermore, we demonstrated the possibility for
the gram-scale synthesis of PAPS from adenosine 5'-tripho-
sphate at only 1/5000th of the price purchased from a com-
mercial source. Our results pave the way to conduct the
enzymatic synthesis of heparin in large quantities.
Keywords: heparan sulfate / heparin / K. lactis / PAPS /
sulfotransferase
Introduction
Heparan sulfate (HS) is a highly sulfated polysaccharide repre-
senting a major component on the mammalian cell surface and
1To whom correspondence should be addressed: Tel: +1-919-843-6511;
Fax: +1-919-843-5432; e-mail: jian_liu@unc.edu
© The Author 2011. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
771

X Zhou et al.
Fig. 1. The biosynthetic/chemoenzymatic synthesis of heparin and HS-like polysaccharides and PAPS. (A) HS biosynthetic and chemoenzymatic synthesis
pathway. Polysaccharide backbone is a copolymer of GlcUA-GlcNAc repeating disaccharide unit. The straight arrows indicate the reactions catalyzed by
enzymes. The curved arrows indicate the conversion using a chemoenzymatic approach. The reaction sites at each modification step are cycled. The names and
abbreviations for the saccharide units are displayed underneath the sugar unit. (B) The steps involved in PAPS synthesis. Sodium sulfate and adenosine 5'-
triphosphate (ATP) are first converted to adenosine 5'-phosphosulfate (APS), and then the second phospho group is added to produce PAPS. PPi represents
pyrophosphate, and Pi represents phosphate.
synthesis was based on using the recombinant biosynthetic
enzymes expressed in Escherichia coli and a cofactor recycling
system. The expression of enzymes in E. coli allowed us to
have the easy access to the proteins. Coupling the cofactor
recycling system with the sulfotransferases reduced the cost for
the use of sulfo donor, 3′-phosphoadenosine 5′-phosphosulfate
(PAPS). Despite the limited success, using the recombinant
proteins from bacterial to prepare heparin raises the concerns
of the potential bacteria endotoxin contamination. To this end,
a protocol that is devoid of bacterial endotoxin and cost-
effective for the large-scale production of HS biosynthetic
enzymes is ideal for preparing synthetic heparin.
The yeast Kluyveromyces lactis expression system has been
successfully used for the industrial-scale production of pro-
teins over decades. Most of the proteins can be produced in
the secreted forms, significantly simplifying the down-stream
purification process (Van Ooyen et al. 2006). K. lactis has
been affirmed as a “Generally Recognized As Safe” organism
by the US FDA (Bonekamp and Oosterom 1994). Numerous
pharmaceutical grade proteins have been produced in K.
lactis, such as interleukin 1 − β, macrophage colony stimulat-
ing factor and insulin precursor (Van Ooyen et al. 2006).
Additionally, the expression in K. lactis does not require the
induction by methanol, and thus the explosion-proof equip-
ment for methylotrophic yeasts such as Pichia pastoris is not
required (Van Ooyen et al. 2006). These advantages make
K. lactis well suited for the large-scale preparation of HS bio-
synthetic enzymes.
In this work, we constructed several K. lactis strains expressing
the secreted form of recombinant sulfotransferases, including
NST, 2-OST, 3-OST-1, 6-OST-1 or 6-OST-3. The crude sulfo-
transferases can be directly used to prepare sulfated polysacchar-
ides. To compare the K. lactis enzymes with the bacterial
counterparts, the proteins were purified with a heparin-Sepharose
column and a 3′-phosphoadenosine-5′-phosphate (PAP)-Agarose
column. All these five yeast-expressed enzymes show higher
specific sulfotransferase activity, ranging from 2- to 10-fold.
Furthermore, we demonstrate that 3-OST-1 and 6-OST-3 have
significantly higher thermostability than that of the E.
coli-expressed enzymes. In addition, we also show the results of
PAPS synthesis and purification at gram scale through a simple
process, and the product can be used to prepare sulfated polysac-
charides. Our results demonstrated clear advancement for synthe-
sizing heparin in large scale.
Results
Optimization of the expression of HS sulfotransferases
in K. lactis
Our goal was to prepare a secreted form of recombinant pro-
teins, which will eliminate the purification steps. To this end,
772

Preparation of HS sulfotransferases
the N-terminal transmembrane domain of sulfotransferases
was replaced by K. lactis α-mating factor (α-MF) domain, a
signal peptide to allow the target proteins to be secreted into
the medium (Colussi et al. 2005; Klabunde et al. 2007). The
initial expression levels were low, especially for 6-OST-3,
which was barely above the detection limit.
that in the commercially available protease-deficient strains
K. lactis YCT389, YCT390, YCT569 and YCT598 (New
England Biolabs) as it is believed that the proteases could
degrade the target proteins. Unfortunately, neither of the
tested cell lines displayed an improved expression level
(data not shown).
We enhanced the expression of 6-OST-3 by three
approaches, altering the media composition, decreasing the
culture temperature and increasing the pH of media
(Figure 2). It was observed that the addition of ethylenediami-
netetraacetic acid (EDTA) or casamino acids increased the
antiserpin single-chain antibody production by 3–5-fold
(Borgenstrom et al. 2003). We observed that the addition of
2% casamino acids into the medium increased the expression
level by about 2-fold (Figure 2A). Temperature also played a
significant role in the expression of 6-OST-3. The activity of
6-OST-3 in the spent medium was increased about 10-fold
by dropping the growth temperature to 16°C from 28°C
(Figure 2B). We also found that the expression of 6-OST-3 is
sensitive to the pH. Increasing the pH value from 5.0 to 7.0
elevated the expression level by 3-fold (Figure 2C). It is
important to note that the growth rate after induction was
similar under different assayed conditions. By combining the
effects of medium composition, temperature and pH, we
improved the expression of 6-OST-3 by more than 30-fold.
Using the conditions for 6-OST-3 as a starting point, we
briefly optimized the expression of 3-OST-1, 6-OST-1, NST
and 2-OST.
The specific activity of sulfotransferases expressed
in K. lactis is higher than that expressed in E. coli
Totally, five sulfotransferases, NST, 2-OST, 3-OST-1,
6-OST-1 and 6-OST-3, were successfully expressed under
the optimal conditions (Figure 3). Our measurement
suggests that NST, 2-OST, 3-OST-1 and 6-OST-3 have
similar levels of sulfotransferase activities and the level of
6-OST-1 is almost 4-fold higher than that of others. We esti-
mated that the amount of crude enzymes in 1 L of media
adequately sulfate 15–100 mg of polysaccharide substrates.
The proteins were then purified by heparin-Sepharose and
PAP-Agarose chromatography, which method was used to
purify the proteins from natural sources (Orellana et al.
1994; Habuchi et al. 1995; Kobayashi et al. 1996; Liu et al.
1996). The recombinant proteins expressed in E. coli were
obtained by following the procedures described previously
(Chen et al. 2005, 2007). Next, we compared the specific
activity of the enzymes expressed in K. lactis with their
counterparts expressed in E. coli (Figure 4). Overall, the
specific activity of the enzymes expressed in K. lactis was
higher than those expressed in bacteria although with differ-
ent extents. Among them, the specific activity of NST and
2-OST expressed in K. lactis was increased by 2-fold. The
specific activity of 6-OST-1 and 6-OST-3 expressed in K.
lactis showed the highest value, increased by 8–10-fold. The
specific activity of 3-OST-1 expressed in K. lactis was
increased by 4-fold.
Increasing the copy number of expression cassette may
increase the expression of protein (Macauley-Patrick et al.
2005). Since all of the expressing cells harbor multiple
copies of the expression cassette, the copy number was not
an issue in our study. We also compared the expression
level of 6-OST-3 in the wild-type strain K. lactis 699 vs.
Fig. 2. Optimization for the expression of 6-OST-3 in K. lactis. (A) Effects of casamino acids on 6-OST-3 expression. YPCG medium is similar to YPG medium
with additional 2% casamino acids. The host strain is K. lactis 699. The cultures (OD600 = 1.0) were induced for 48 h at 28°C, then sulfotransferase activity was
determined as described in Materials and methods. (B) Effects of temperature on 6-OST-3 expression. YPCG medium was used to induce the protein expression
with K. lactis 699 as the host strain, and the induction was carried out at different temperatures for 48 h. (C) Effects of pH on 6-OST-3 expression. YPCGN
medium was used to induce the protein expression with K. lactis YCT598 as the host strain, and the induction was carried out at 16°C for 48 h. The data
presented are the average of three determinations, and the error bars indicate the standard deviations.
773

X Zhou et al.
Fig. 3. Sulfotransferase activity in the supernatant. Sulfotransferase expression
was carried out in YPCGN medium at 16°C for 48 h. Sulfotransferase activity
assay was carried out by incubating 20 µL of filtered supernatant with
appropriate substrate in 100 µL of reaction buffer at 37°C overnight. The data
presented are the average of two determinations, and the error bars indicate
the ranges.
Fig. 4. Comparison of the specific activity between E. coli- and K.
lactis-expressed sulfotransferases. Same amount of purified protein was mixed
with appropriate substrate in reaction buffer, and the activity was determined
after 1 h of incubation at 37°C. The data presented are the average of two
determinations, and the error bars indicate the ranges.
Comparison of the thermostability between
E. coli- and K. lactis-expressed sulfotransferases
Fig. 5. Comparison of the thermostability between E. coli- and K.
In an effort to improve the enzymatic synthesis efficiency, a
thermally stable enzyme is desirable. We hypothesized that the
sulfotransferases expressed in K. lactis are glycosylated and
thus have higher thermal stability than bacterial enzymes. To
this end, equal amount of purified enzymes expressed in K.
lactis and E. coli was maintained in the reaction buffer at 37°C.
At different time points, the enzymes were removed and
assayed for the sulfotransferase activities by the addition of
35S-labeled PAPS into the reaction mixture. The results show
that the thermostability of sulfotransferases was increased or
unchanged with the exception of NST (Figure 5). We found that
the most profound effect on the stability was for 6-OST-3. The
activity of 6-OST-3 (K. lactis) remained unchanged up to 96 h,
whereas the activity of 6-OST-3 (E. coli) was nearly completely
lost after 48 h (Figure 5C). The 3-OST-1 also showed a
lactis-expressed sulfotransferases. Same amount of protein was mixed with
appropriate substrate in reaction buffer without PAPS and incubated at 37°C.
At different time points during incubation, 75 µL of aliquots was removed
and the sulfotransfer reactions were performed for 1 h at 37°C by adding
[
35S] PAPS into the aliquots. The data presented are the average of two
determinations, and the error bars indicate the ranges.
significant improvement on the thermostability as about 50% of
3-OST-1 activity remained at around 30 h, whereas the activity
of 3-OST-1 (E. coli) was dropped below the detection limit in
6 h (Figure 5B). Both the K. lactis version 2-OST and the E.
coli version 2-OST were very stable, their activity remained
100% after 96 h of incubation (data not shown). A similar
observation was made for 6-OST-1, namely their activity
774

Preparation of HS sulfotransferases
remained 100% after 48 h of incubation and only decreased to
30–40% after 96 h of incubation regardless of the protein
expressed in K. lactis or in E. coli (data not shown). We
observed that NST expressed in K. lactis was less stable than
the one expressed in E. coli. The activity of K. lactis version
NST dropped to 24% in 8 h, whereas that of E. coli version
NST only dropped to 52% in 48 h. The observation for the
decrease in its stability for NST is somewhat unexpected. It
should be noted that NST is only the C-terminal domain of
NDST, which is not a natural enzyme present in the mammalian
cells. Whether the expression of entire sequence of NDST
improves its thermal stability is currently unknown.
The unique properties of the sulfotransferases expressed in
K. lactis are likely attributed to the glycosylation introduced
during the expression. Indeed, N-glycosylation sites in the sul-
fotransferases are predicted with the NetNGlyc 1.0 Server (http
://www.cbs.dtu.dk/services/NetNGlyc/) using artificial neural
networks that examine the context of Asn-Xaa-Ser/Thr
sequences (Blom et al. 2004). The putative N-linked glycosyla-
tion sites are present within the catalytic domain of the sulfo-
transferases, including NST (667NTT), 2-OST (108NTT and
127NVT), 6-OST-1 (261NMS and 317NST), 6-OST-3
(127NFT, 323NLT and 379NIT) and 3-OST-1(52NGS,
141NPT, 196NRS and 253NKT). The sodium dodecyl sulfate–
polyacrylamide gel electrophoresis (SDS-PAGE) analysis of the
sulfotransferases supports this notion. Unlike their counterparts
expressed in E. coli, the sulfotransferases expressed in K. lactis
did not migrate as a discrete band rather as a smear.
Furthermore, the apparent molecular weights of the yeast-
expressed proteins are significantly larger than that of the antici-
pated molecular weight without glycosylation. These obser-
vations suggest that the proteins are glycosylated (Figure 6).
synthesis in an academic lab. Previously, we employed a
PAPS regeneration system to prepare heparin (Chen et al.
2005). In this system, arylsulfotransferase IV was used to
convert PAP to PAPS using p-nitrophenol sulfate as a sulfo
donor. However, whether coupling PAPS regeneration system
with HS sulfotransferases yields unexpected byproducts is dif-
ficult to determine, especially the product is a mixture of
polysaccharides. Using PAPS as a sulfo source, we should
eliminate this concern. We chose to synthesize PAPS enzy-
matically using adenosine 5'-triphosphate (ATP) sulfurylase
(from K. lactis), adenosine 5'-phosphosulfate (APS) kinase
(from Penicillium chrysogenum) and pyrophosphatase (from
E. coli) as described in Figure 1B. All three enzymes involved
in the synthesis of PAPS were expressed as (His)6-tagged pro-
teins in E. coli in high yields. As expected, the preparation of
PAPS can be achieved with purified ATP sulfurylase, APS
kinase and pyrophosphatase (Figure 7A). The products con-
tained both adenosine 5'-diphosphate (ADP) and PAPS, which
were identified by PAMN-HPLC (high-pressure liquid chrom-
atography; Figure 7A). Alternatively, the preparation of PAPS
was also achieved using crude extracts from the bacteria
expressing the three enzymes. Using crude enzymes, ATP and
sulfate were converted to PAPS effectively, although ADP,
adenosine 5'-monophosphate (AMP) and trace amount of
unreacted ATP were present in the product (Figure 7B). The
product was further purified by a diethylaminoethyl (DEAE)
column. The purified PAPS exhibited a single peak with poly-
amine anion exchange (PAMN)-HPLC, suggesting that the
preparation was pure (Figure 7C). The molecular mass of
product was determined to be 506.6 0.2 Da by electrospray
ionization mass spectrometry (MS), very close to the calcu-
lated molecular mass of PAPS (507.0; Figure 7D). The puri-
fied PAPS was employed to prepare N-sulfated and
N,6-O-sulfated heparosan using NST and 6-OST-1, respect-
ively. The disaccharide analysis of N-sulfated heparosan con-
firmed the presence of a disaccharide of ΔUA-GlcNS
(Figure 7F); the disaccharide analysis of N,6-O-sulfated
Synthesis, purification and the application of purified PAPS
to sulfotransfer reaction
PAPS is a key cofactor for sulfotransferases to synthesize
heparin. PAPS is expensive even for the milligram-scale
Fig. 6. SDS–PAGE analysis of sulfotransferases. (A) Sulfotransferases expressed in K. lactis. Lane 1, NST; lane 2, 2-OST; lane 3, 6-OST-1; lane 4, 6-OST-3;
lane 5, 3-OST-1. The anticipated molecular weight of sulfotransferases expressed by K. lactis without glycosylation is: 35 kDa for NST, 36 kDa for 2-OST, 41
kDa for 6-OST-1, 41 kDa for 6-OST-3 and 34 kDa for 3-OST-1. (B) Sulfotransferases expressed by E. coli. Lane 6, NST (in glutathione S-transferase fusion
protein form); lane 7, 2-OST (in maltose-binding protein fusion protein form); lane 8, 6-OST-1 (in maltose-binding protein fusion protein form); lane 9, 6-OST-3
(in maltose-binding protein fusion protein form); lane 10, 3-OST-1 (in (His)6-fusion protein form). The anticipated MW for bacterial-expressed protein is: 61 kDa
for NST, 70 kDa for 2-OST, 83 kDa for 6-OST-1, 81 kDa for 6-OST-3 and 35 kDa for 3-OST-1.
775

X Zhou et al.
Fig. 7. PAPS synthesis, purification and the application of purified PAPS to sulfotransfer reaction. (A) HPLC analysis of the products from PAPS synthesis
reaction catalyzed by purified enzymes. The area of PAPS peak (Area = 701 units) and that of the adenosine 5'-diphosphate (ADP) peak (Area = 685 units) are
similar to each other. (B) HPLC analysis of the products from PAPS synthesis reaction catalyzed by crude enzymes. ATP and sulfate were converted to PAPS
(Area = 625 units), ADP (Area = 214 units) and adenosine 5'-monophosphate (AMP) (Area = 503 units). Our data suggest that close to 95% of ATP was
converted to the product in 6 h reaction. The elution positions of PAPS, ATP, ADP and AMP were determined by injecting authentic standard compounds. (C)
PAPS purity determination by HPLC analysis. (D) MS analysis of the purified PAPS. (E–G) Disaccharide analysis. The purified PAPS was applied to NST (F) or
6-OST-1 (G) catalyzed sulfotransfer reaction, respectively. (E) The eluted positions of authentic disaccharide standards, where 1 represents ΔUA-GlcNAc, 2
represents ΔUA-GlcNS, 3 represents ΔUA-GlcNAc6S, 4 represents ΔUA-GlcNS6S, 5 represents ΔUA2S-GlcNS and 6 represents ΔUA2S-GlcNS6S.
heparosan confirmed the presence of a disaccharide of
ΔUA-GlcNS6S (Figure 7G). The results of disaccharide
analysis demonstrated that the synthetic PAPS can be utilized
by HS sulfotransferases to prepare sulfated polysaccharides.
Figure 7E shows the eluted positions of authentic disaccharide
standards. It is worthwhile to mention that we obtained 5 g of
PAPS from 1 L reaction with a cost of about $50, estimated to
be at $0.01 mg−1. The price of PAPS on Sigma catalog of
2010 is $36–$85 mg−1, almost 3600–8500-fold higher than
the synthetic PAPS using our method.
Discussion
Heparin is an important anticoagulant drug, which is currently
isolated from the porcine intestine. Developing a method to
prepare synthetic heparin will not only eliminate the potential
sources of contamination and adulteration. The method will
also offer the possibility to alter the sulfation patterns to syn-
thesize the heparin without side effects. For example, a
common side effect that is associated with the use of heparin
is heparin-induced thrombocytopenia because heparin binds
776

Preparation of HS sulfotransferases
to platelet factor 4 (Arepally and Ortel 2006). Using a syn-
thetic approach, we could potentially separate the anticoagu-
lant activity and the binding affinity to platelet factor 4 to
reduce the risk of heparin-induced thrombocytopenia. The
high regioselectivity of HS biosynthetic enzymes provides a
unique advantage to prepare synthetic heparin by eliminating
protecting/unprotecting steps used in chemical synthesis.
Despite our initial success in preparing milligram-scale syn-
thesis, the availability of sulfotransferases and PAPS remains
two obstacles for the cost-effective synthesis in a large scale.
Here, we described the expression of HS sulfotransferases in
K. lactis, permitting us to harvest the enzymes directly from
conditioned medium. We notice that the genes encode the sul-
fotransferases we expressed from three different species. This
is because the genes were available in our lab when the study
was initiated. The purified enzymes expressed in K. lactis
have favorable properties for conducting the enzymatic syn-
thesis of heparin: higher specific activity and improved ther-
mostability. In addition, an enzymatic approach was
successfully employed to synthesize a gram scale of PAPS at
low cost.
Our results demonstrated that K. lactis system is an excel-
lent expression approach to prepare HS sulfotransferases. HS
sulfotransferases are membrane-bound sulfotransferases and
glycosylated (Esko and Selleck 2002; Saribas et al. 2004).
Expression of recombinant HS biosynthetic enzymes can be
achieved in insect cells using a baculovirus expression
approach and in bacteria (Kuberan et al. 2003; Chen et al.
2005). Given the complexity of using the baculovirus
expression approach, bacterial expression offers an efficient
method to obtain the recombinant enzymes. However, E. coli
has no capability to introduce glycosylation that could reduce
the specific enzymatic activity and decrease the thermostabil-
ity. Furthermore, the expression was carried out in cytosol and
the recombinant proteins are likely to contain endotoxin.
Procedures for purifying the proteins and to remove endotoxin
are laborious and costly. Therefore, the expression of recombi-
nant HS sulfotransferases in yeast should be an ideal alterna-
tive because yeast cells do not produce endotoxin and are
cheap to grow. Up to date, only NDST1 has been reportedly
expressed in Saccharomyces cerevisiae (Saribas et al. 2004).
However, the expression was achieved intracellularly, and the
level of expression is not sufficiently high. By constructing a
secreted form of recombinant protein, we can now harvest HS
sulfotransferases from the conditioned medium of K. lactis.
We should be able to use the enzymes to carry out the syn-
thesis without purification.
(Macauley-Patrick et al. 2005). A previous work shows that
the activity of laccase was increased about 4-fold by induction
at 20°C instead at 30°C (Hong et al. 2002). The mechanisms
behind the temperature effect may be due to the decrease in the
stability of the recombinant protein at higher temperatures,
release of more proteases from dead cells and folding problems
at higher temperature (Hong et al. 2002; Macauley-Patrick
et al. 2005). In our work, dropping the induction temperature
from 28°C to 16°C increased the expression of 6-OST-3 by
about 10-fold. The pH value is another key factor affecting the
yield of recombinant protein production, mainly due to the
influence on proteases activity by variable pH values (Flores
et al. 1999). In our work, when the pH value increased from
5.0 to 7.0, the activity of 6-OST-3 increased about 3-fold
(Figure 2C).
It has been recognized for decades that post-translational
modification, especially glycosylation, can improve proper
folding, biological activity and stability of recombinant
proteins (Stanley 1992). Comparing the sulfotransferases
expressed in two different sources, we observed unique prop-
erties of sulfotransferases due to the glycosylation of the sul-
fotransferases carried out in K. lactis. The specific activities
of the five sulfotransferases expressed in K. lactis were
increased by 2–10-fold higher than those expressed in E. coli.
We observed a range of effects on the thermostability of the
enzymes. For 3-OST-1 and 6-OST-3, the enzymes were
clearly more stable, while 2-OST and 6-OST-1 remained the
same. For NST-1, the thermostability is decreased. These
interesting findings also offer us a potential opportunity to
modulate the thermostability of the enzyme by controlling the
glycosylation.
In summary, we report new methods to prepare sulfotrans-
ferases and PAPS for the enzymatic synthesis of heparin. The
enzymes can be expressed in a secreted form in K. lactis in
high yields. The recombinant sulfotransferases can be directly
harvested from media, and some proteins display improved
thermostability and specific activity. Furthermore, an enzy-
matic approach was also developed to synthesize PAPS in one
pot reaction at very low cost. Our results could contribute sig-
nificantly toward the large-scale synthesis of heparin drugs.
Materials and methods
Yeast and bacterial strains and plasmid construction
Strains and plasmids used in this study are listed in Table I.
E. coli strains were grown in Luria-Bertani (LB) medium or
on LB plates containing 1.5% agar. Unless otherwise stated,
K. lactis GG799 or K. lactis YCT598 were grown in YPG
medium (1% yeast extract, 2% peptone and 2% galactose) for
propagation, YCB medium (30 mM Tris–HCl, pH 7.0, 11.7 g
L−1 of yeast carbon base and 5 mM acetamide) for selection
or YPCGN medium (1% yeast extract, 2% peptone, 2% casa-
mino acids, 2% galactose and 5 g L−1 of ammonium sulfate)
for expression. Carbenicillin and streptomycin were added at
50 µg mL−1 when necessary.
Our results suggest that optimization has significant impact
on improving the protein expression in K. lactis. Proteolytic
degradation is a serious problem for yeast expression. Addition
of amino acid rich supplements to the culture medium, such as
casamino acids, can enhance the stability of recombinant pro-
teins, possibly by acting as alternative and competing sub-
strates for the undesired proteases, or repressing protease
induction caused by nitrogen limitation (Macauley-Patrick
et al. 2005). In our study, the activity of 6-OST-3 in the spent
culture increased about 2-fold by addition of 2% casamino
acids in the induction medium. Lower cultivation temperature
can also influence yield of recombinant proteins
To construct the secreted forms of recombinant proteins,
NST domain of Rattus norvegicus NDST1 (H591-R882),
Gallus gallus 2-OST (D69-N356), Mus musculus 3-OST-1
(G48-H311), G. gallus 6-OST-1 (P46-H378) or M. musculus
777

X Zhou et al.
Table I. Strains and plasmids used in this study
Strain or plasmid
Characteristics
Source or reference
E. coli strains
DH5α
Cloning host
Expression host
Invitrogen
Invitrogen
BL21 Star (DE3)
Yeast Strains
K. lactis GG799
K. lactis YCT598
Plasmids
New England Biolab
New England Biolab
pKLAC2
pET-15b
Expression vector, ApR
New England Biolab
Novagen
Novagen
Novagen
MacRae et al. (1998)
This work
Expression vector, ApR
pCDFDuet-1
pETDuet-1
APSK-pET23a(+)
APSK-pETDuet-1
KAST-APSK-pETDuet-1
Expression vector, SmR
Expression vector, ApR
pET-23a(+) carrying APS kinase gene of P. chrysogenum, ApR
pETDuet-1 carrying APS kinase gene of P. chrysogenum, ApR
pETDuet-1 carrying His-tagged ATP sulfurylase gene of K. lactis and APS kinase
gene of P. chrysogenum, ApR
This work
APSK-pET-15b
PPA-pCDFDuet-1
pET-15b carrying His-tagged APS kinase gene
This work
This work
pCDFDuet-1 carrying His-tagged pyrophosphatase gene of E. coli, SmR
APS, adenosine 5'-phosphosulfate; ATP, adenosine 5'-triphosphate.
6-OST-3 (P120-P449) was cloned into pKLAC2 downstream
of the K. lactis α-MF domain. Both the DNA sequences of
NST and 3-OST-1 were subjected to yeast codon optimization,
and the optimized DNAs were synthesized by GenScript. The
constructs were linearized with SacII, recovered using the
QIAquick Gel Extraction Kit (Qiagen Inc., Valencia, CA) and
transformed into K. lactis competent cells according to the
manual of K. lactis Protein Expression Kit (New England
Biolabs Inc.). Colonies that were grown on YCB agar
medium were inoculated into 1 mL of YCB medium and
incubating at 28°C overnight with shaking at 250 rpm. The
overnight culture (50 µL) was expanded into 1 mL of
YPCGN medium and induced by galactose for 48 h at 28°C.
The supernatant was subjected to sulfotransferase activity
assay. Clones showing the highest activity were chosen for
further analysis.
Protein expression and purification
The expression was carried out in YPCGN medium at 16°C
for 48 h with K. lactis YCT598 as the host strain, unless
otherwise stated. K. lactis cells carrying the NST, 2-OST,
3-OST-1, 6-OST-1 or 6-OST-3 expression cassette were
grown in YCB medium and then inoculated into YPG
medium. The cultures were incubated at 28°C overnight with
shaking at 250 rpm until the OD600 reached to 10 U. To
induce protein expression and secretion, overnight cultures
were expanded into 500 mL YPCGN medium to a density of
1.0 at OD600 in a 2.8 L flask and induced for 48 h at 16°C,
where galactose served as an inducer. The culture was har-
vested by centrifugation, adjusted to pH 5.5, and the super-
natant was filtered through a Whatman glass microfiber filter
(1.2 µm, Fisher).
All subsequent protein purification procedures were per-
To construct APSK-pETDuet-1, APSK-pET23a(+) (MacRae
et al. 1998) was digested with restriction enzyme NdeI and
XhoI. Fragments carrying APS kinase gene were purified and
ligated into the NdeI–XhoI site of pETDuet-1 (Novagen). To
construct KAST-APSK-pETDuet-1 plasmid, the ATP sulfury-
lase was amplified from K. lactis GG799 genome and cloned
into the PstI–NotI site of APSK-pETDuet-1 using primers
with the sequences of 5′-primer AAAACTGCAGATGC
CTTCTCCTCATGGTGGT and 3′-primer ATAAGAAT
GCGGCCGCTCAGAATTGGAAAAAGCCTTGGT, where
characters in bold face represent the restriction site.
APSK-pET-15b was constructed by inserting the fragment
with APS kinase gene into the NdeI–XhoI site of pET-15b.
Pyrophosphatase gene was amplified from E. coli K12 genome
using primers (5′-primer AGCCGCGAATTCGATGAGCTT
ACTCAACGTCCCTG and 3′-primer GTACCCAAGCTTTT
ATTTATTCTTTGCGCGCTCG, where characters in bold face
represent the restriction site) and cloned into the EcoRI–
HindIII site of pCDFDuet-1 to form the PPA-pCDFDuet-1
plasmid. All constructs were sequenced at the UNC genomic
analysis facility to verify the reading frame and free from
mutations.
formed at 4°C. The supernatant was applied to
a
heparin-Sepharose (GE Healthcare) column and washed with
buffer W1 (50 mM sodium acetate, 150 mM NaCl, 10% gly-
cerol, pH 5.5). The protein was eluted with buffer E1 (50 mM
sodium acetate, 1 M NaCl, 10% glycerol, pH 5.5). Fractions
containing 3-OST-1, 6-OST-1 or 6-OST-3 were pooled and
dialyzed against buffer W2 (50 mM MOPS, 50 mM NaCl,
5% glycerol, pH 7.0). Fractions containing NST or 2-OST
were pooled and dialyzed against buffer W3 (50 mM MOPS,
5% glycerol, pH 7.0). The dialyzed samples were applied to a
PAP-Agarose (Sigma) column and washed with buffer W2
(for 3-OST-1, 6-OST-1 and 6-OST-3 purification) or W3 (for
NST and 2-OST purification), respectively, at a flow rate of
0.5 mL min−1. The protein was eluted with a linear gradient
of 50 mM–1 M NaCl in buffer W3 for 60 min. Fractions were
pooled and stored at −80°C until use. E. coli-expressed NST,
2-OST, 3-OST-1, 6-OST-1 and 6-OST-3 were purified as
described previously (Kakuta et al. 1999; Edavettal et al.
2004; Chen et al. 2005, 2007; Abdel-Banat et al. 2010). The
protein concentration was determined by Bradford Reagent
(Sigma) according to the manufacturer’s introduction.
778

Preparation of HS sulfotransferases
The enzymes involved in synthesizing PAPS, including ATP
sulfurylase, APS kinase and pyrophosphatase, were expressed
in E. coli BL21 Star(DE3) (Invitrogen). Briefly, the bacteria
were grown in LB to the OD600nm value of 0.6. After decreas-
ing the temperature to 22°C, the culture was induced with 0.2
mM IPTG overnight. His-tagged ATP sulfurylase (expressed
by KAST-APSK-pETDuet-1), APS kinase (expressed by
APSK-pET-15b) and pyrophosphatase (expressed by
PPA-pCDFDuet-1) were purified with Ni-Agarose (GE Health)
column. All purification procedures were performed at 4°C.
The cultures were harvested and resuspended in 50 mL of
buffer A (20 mM Tris–HCl, 500 mM NaCl, pH 7.6). The cells
were lysed by sonication, and then 1 mM phenylmethanesulfo-
nyl fluoride was added immediately to prevent protein degra-
dation. After centrifugation at 12,000 rpm for 20 min, the
supernatant was filtrated through a 0.45 µm membrane and
applied to a Ni-Agarose column at a flow rate of 3.0 mL
min−1. The column was eluted with a linear gradient of 50–
300 mM imidazole in buffer A for 30 min. Fractions contain-
ing the protein were pooled, dialyzed against dialysis buffer
(20 mM Tris–HCl, pH 7.6, at 4°C; 500 mM NaCl; 10% gly-
cerol) and stored at −80°C. The purity of protein was analyzed
by 10% precasted SDS–PAGE (BioRad).
V and 200°C, and the compound stability set to 30%.
Nitrogen was used for both nebulizer (5 L min−1) and drying
gas (15 psi). Helium was used for collision-induced dis-
sociation. The MS and MS/MS data were acquired and pro-
cessed using Bruker Trap Software 4.1.
Sulfotransferase activity assay
Sulfotransferase activity was determined by incubating
3-OST-1 with 10 µg of HS (from bovine kidney; ICN) in
100 µL of reaction buffer (50 mM 2-(N-Morpholino)ethane-
sulfonic acid, pH 7.0, 5 mM MgCl2, 50 µM PAPS and 1–
10 × 105 cpm of [35S] PAPS (ꢀ2 µM)). Assays for other sul-
fotransferases were almost the same, while different substrates
were used. Deacetylated heparosan (1 µg) was used for NST
(Lindahl et al. 2005), and completely desulfated N-sulfated
heparin (1 µg; designated as DE-OS-Heparin from Neoparin
Inc.) was used as a substrate for 6-OST-1, 6-OST-3 and
2-OST. The reaction was incubated at 37°C for 1 h and
quenched by the addition of UPAS buffer (50 mM sodium
acetate, pH 5.5, 150 mM NaCl, 6 M urea, 1 mM EDTA and
0.01% Triton X-100). The samples were then subjected to
200 µL DEAE-Sepharose chromatography to purify the [35S]
products.
PAPS synthesis and purification
Disaccharide analysis
The PAPS preparation was achieved by crude enzymes. The
bacterial cells expressing ATP sulfurylase (expressed using
KAST-APSK-pETDuet-1 plasmid), APS kinase (expressed
using KAST-APSK-pETDuet-1plasmid) and pyrophosphatase
(expressed using PPA-pCDFDuet-1plasmid) were lysed as
described in protein expression and purification. After centrifu-
gation, filtration and addition of 10% glycerol, the crude
enzyme solutions were stored at −80°C until use. With a modi-
fied method described previously (Burkart et al. 2000), the
reaction included 22.5 mM ATP disodium, 100 mM Na2SO4,
10 mM MgCl2, 10 mM LiCl, 0.6 mg mL−1 of crude enzymes
including ATP sulfurylase and APS kinase, 0.3 mg mL−1 of
crude enzyme including pyrophosphatase and 50 mM Tris–
HCl at pH 8.0 and incubated at 30°C for 6 h. The completion
of PAPS synthesis was checked by PAMN-HPLC (polyamine
II column, YMC America, Inc.) as follows: 100% water for 10
min, followed with a linear gradient of 0–100% of 1 M
KH2PO4 for 30 min, followed by 100% 1 M KH2PO4 for 15
min at a flow rate of 1 mL min−1 with UV 254 nm detection.
The polysaccharides (100 µg) were purified by a 200 µL
DEAE-Sepharose column, then dialyzed against deionized
water, dried by speed vacuum and degraded by a mixture of
heparin lyases including Flavobacterium heparinum heparin
lyases I, II and III. The resultant disaccharides were desalted
on BioGel P-2 column (0.5 × 200 cm) eluted with 0.1 M
ammonium bicarbonate with a flow rate of 3 mL h−1. The dis-
accharides were analyzed on reverse phase-ion pairing HPLC
by using a C18 reversed phase column (4.6 × 250 mm; Vydac)
with UV 232 nm detection as described previously
(Kobayashi et al. 1996). The column was eluted with aceto-
nitrile as follows: 7% for 30 min followed by 15% for 15
min, followed by 19.5% for 15 min, followed by 50% for 15
min and followed by 7% for 10 min, in a solution containing
38 mM ammonium phosphate monobasic, 2 mM phosphoric
acid and 1 mM tetrabutylammonium phosphate monobasic at
a flow rate of 0.5 mL min−1.
Large-scale purification of PAPS was achieved on
a
Funding
DEAE-Sepharose fast flow column (GE Health; 1.5 × 60 cm).
The DEAE column was washed with water, and PAPS was
eluted with a gradient of 0–500 mM NaCl at 5.0 mL min−1 for
200 min. Fractions containing PAPS as determined by PAMN-
HPLC were pooled and stored at −80°C.
This work is supported in part by National Institutes of
Health (AI50050, HL094463, HL096972 and AI074775).
Acknowledgements
Analysis of PAPS by MS
The authors thank Dr. Kai Li for providing the E. coli-
expressed and -purified 2-OST and thank Dr. Yongmei Xu for
providing the E. coli-expressed and -purified 3-OST-1 and
6-OST-3.
Purified PAPS was dialyzed using the Spectra/Por dialysis
membrane (MWCO 1000 Da) against 20 mM ammonium
acetate for 4 h. A syringe pump (Harvard Apparatus) was
used to introduce the sample via direct infusion (10 µL min−1)
into an Agilent 1100 MSD-Trap at the ADME Mass
Spectrometry Center. Experiments were performed in a nega-
tive ionization mode with the electrospray source set to 2500
Conflict of interest
None declared.
779

X Zhou et al.
is induced by binding to heparan sulfate. Biochemistry (Mosc).
43:4680–4688.
Abbreviations
Esko JD, Selleck SB. 2002. Order out of chaos: Assembly of ligand binding
sites in heparan sulfate. Annu Rev Biochem. 71:435–471.
Flores MV, Cuellas A, Voget CE. 1999. The proteolytic system of the yeast
Kluyveromyces lactis. Yeast. 15:1437–1448.
Habuchi H, Habuchi O, Kimata K. 1995. Purification and characterization of
heparan sulfate 6-sulfotransferase from the culture medium of Chinese
hamster ovary cells. J Biol Chem. 270:4172.
Hong F, Meinander NQ, Jonsson LJ. 2002. Fermentation strategies for
improved heterologous expression of laccase in Pichia pastoris. Biotechnol
Bioeng. 79:438–449.
Kakuta Y, Sueyoshi T, Negishi M, Pedersen LC. 1999. Crystal structure of the
sulfotransferase domain of human heparan sulfate N-deacetylase/
N-sulfotransferase 1. J Biol Chem. 274:10673–10676.
Klabunde J, Kleebank S, Piontek M, Hollenberg CP, Hellwig S, Degelmann
A. 2007. Increase of calnexin gene dosage boosts the secretion of heter-
ologous proteins by Hansenula polymorpha. FEMS Yeast Res.
7:1168–1180.
Kobayashi M, Habuchi H, Habuchi O, Saito M, Kimata K. 1996. Purification
and characterization of heparan sulfate 2-sulfotransferase from cultured
Chinese hamster ovary cells. J Biol Chem. 271:7645–7653.
Kuberan B, Lech MZ, Beeler DL, Wu ZL, Rosenberg RD. 2003. Enzymatic
synthesis of antithrombin III-binding heparan sulfate pentasaccharide. Nat
Biotechnol. 21:1343–1346.
Lindahl U, Li JP, Kusche-Gullberg M, Salmivirta M, Alaranta S, Veromaa T,
Emeis J, Roberts I, Taylor C, Oreste P, et al. 2005. Generation of “neohe-
parin” from E. coli K5 capsular polysaccharide. J Med Chem. 48:349–352.
Liu J, Shworak NW, Fritze LMS, Edelberg JM, Rosenberg RD. 1996.
Purification of heparan sulfate D-glucosaminyl 3-O-sulfotransferase. J Biol
Chem. 271:27072–27082.
ADP, adenosine 5'-diphosphate; AMP, adenosine 5'-monophos-
phate; APS, adenosine 5'-phosphosulfate; ATP, adenosine 5'-
triphosphate; α-MF, α-mating factor; DEAE, diethylami-
noethyl; EDTA, ethylenediaminetetraacetic acid; HPLC, high-
pressure liquid chromatography; HS, heparan sulfate; LB,
Luria-Bertani;
NDST,
N-deacetylase/N-sulfotransferase;
NST, N-sulfotransferase; 2-OST, 2-O-sulfotransferase; 3-OST,
3-O-sulfotransferase; 6-OST, 6-O-sulfotransferase; PAP, 3′-
phosphoadenosine-5′-phosphate; PAMN, polyamine anion
exchange; PAPS, 3′-phosphoadenosine-5′-phosphosulfate;
SDS–PAGE, sodium dodecyl sulfate–polyacrylamide gel
electrophoresis.
References
Abdel-Banat BMA, Nonklang S, Hoshida H, Akada R. 2010. Random and
targeted gene integrations through the control of non-homologous end
joining in the yeast Kluyveromyces marxianus. Yeast. 27:29–39.
Aquino RS, Pereira MS, Vairo BC, Cinelli LP, Santos GR, Fonseca RJ,
Mourao PA. 2010. Heparins from procine and bovine intestinal mucosa:
Are they similar drugs? Thromb Haemost. 103:1005–1015.
Arepally G, Ortel TL. 2006. Heparin-induced thrombocytopenia. N Engl J
Med. 355:809–817.
Bernfield M, Gotte M, Park PW, Reizes O, Fitzgerald ML, Lincecum J, Zako
M. 1999. Functions of cell surface heparan sulfate proteoglycans. Annu
Rev Biochem. 68:729–777.
Blom N, Sicheritz-Pontén T, Gupta R, Gammeltoft S, Brunak S. 2004.
Prediction of post-translational glycosylation and phosphorylation of pro-
teins from the amino acid sequence. Proteomics. 4:1633–1649.
Bonekamp FJ, Oosterom J. 1994. On the safety of Kluyveromyces lactis— a
review. Appl Microbiol Biotechnol. 41:1–3.
Liu H, Zhang Z, Linhardt RJ. 2009. Lessons learned from the contamination
of heparin. Nat Prod Rep. 26:313–321.
Macauley-Patrick S, Fazenda ML, McNeil B, Harvey LM. 2005.
Heterologous protein production using the Pichia pastoris expression
system. Yeast. 22:249–270.
Borgenstrom M, Jalkanen M, Salmivirta M. 2003. Sulfated derivatives of
Escherichia coli K5 polysaccharides as modulators of fibroblast growth
factor signaling. J Biol Chem. 278:49882–49889.
Burkart MD, Izumi M, Chapman E, Lin CH, Wong CH. 2000. Regeneration
of PAPS for the enzymatic synthesis of sulfated oligosaccharides. J Org
Chem. 65:5565–5574.
MacRae IJ, Rose AB, Segel IH. 1998. Adenosine 5′-phosphosulfate kinase
from Penicillium chrysogenum. J Biol Chem. 273:28583–28589.
Orellana A, Hirschberg CB, Wei Z, Swiedler SJ, Ishihara M. 1994. Molecular
cloning and expression of a glycosaminoglycan N-acetylglucosaminyl
N-deacetylase/N-sulfotransferase from a heparin producing cell line. J Biol
Chem. 269:2270–2276.
Chen J, Avci FY, Munoz EM, McDowell LM, Chen M, Pedersen LC, Zhang
L, Linhardt RJ, Liu J. 2005. Enzymatic redesigning of biologically active
heparan sulfate. J Biol Chem. 280:42817–42825.
Peterson S, Frick A, Liu J. 2009. Design of biologically active heparan
sulfate and heparin using an enzyme-based approach. Nat Prod Rep.
26:610–627.
Chen J, Jones CL, Liu J. 2007. Using an enzymatic combinatorial approach
to identify anticoagulant heparan sulfate structures. Chem Biol.
14:986–993.
Petitou M, van Boeckel CAA. 2004. A synthetic antithrombin III binding
pentasaccharide is now a drug! What comes next? Angew Chem Int Ed
Engl. 43:3118–3133.
Saribas AS, Mobasseri A, Pristatsky P, Chen X, Barthelson R, Hakes D,
Wang J. 2004. Production of N-sulfated polysaccharides using yeast-
expressed N-deacetylase/N-sulfotransferase-1 (NDST-1). Glycobiology.
14:1217–1228.
Stanley P. 1992. Glycosylation engineering. Glycobiology. 2:99–107.
Van Ooyen AJJ, Dekker P, Huang M, Olsthoorn MMA, Jacobs DI, Colussi
PA, Taron CH. 2006. Heterologous protein production in the yeast
Kluyveromyces lactis. FEMS Yeast Res. 6:381–392.
Colussi PA, Specht CA, Taron CH. 2005. Characterization of
a
nucleus-encoded chitinase from the yeast Kluyveromyces lactis. Appl
Environ Microbiol. 71:2862–2869.
Copeland R, Balasubramaniam A, Tiwari V, Zhang F, Bridges A, Linhardt RJ,
Shukla D, Liu J. 2008. Using a 3-O-sulfated heparin octasaccharide to
inhibit the entry of herpes simplex virus type 1. Biochemistry (Mosc).
47:5774–5783.
Edavettal SC, Carrick K, Shah RR, Pedersen LC, Tropsha A, Pope RM, Liu J.
2004. A conformational change in heparan sulfate 3-O-sulfotransferase-1
780