1
PSZ 19:16 (Pind. 1/97)
UNIVERSITI TEKNOLOGI MALAYSIA
BORANG PENGESAHAN STATUS TESIS♦
JUDUL :
DEVELOPMENT OF NOVEL GALACTOSYLATION METHOD
FOR THE EXPRESSION OF RECOMBINANT HUMAN TRANSFERRIN
IN INSECT CELL CULTURE
SESI PENGAJIAN : 2004 / 2005
YAP WEI NEY
Saya
(HURUF BESAR)
mengaku membenarkan tesis (PSM/Sarjana/Doktor Falsafah)* ini disimpan di Perpustakaan
Universiti Teknologi Malaysia dengan syarat-syarat kegunaan seperti berikut:
1.
2.
3.
4.
Tesis adalah hakmilik Universiti Teknologi Malaysia.
Perpustakaan Universiti Teknologi Malaysia dibenarkan membuat salinan untuk tujuan
pengajian sahaja.
Perpustakaan dibenarkan membuat salinan tesis ini sebagai bahan pertukaran antara institusi
pengajian tinggi.
** Sila tanda ( √ )
√
SULIT
(Mengandungi maklumat yang berdarjah keselamatan atau
kepentingan Malaysia seperti yang termaktud di dalam AKTA
RAHSIA RASMI 1972)
TERHAD
(Mengandungi maklumat yang TERHAD yang telah ditentukan
oleh organisasi/badan di mana penyelidikan dijalankan)
TIDAK TERHAD
Disahkan oleh
(TANDATANGAN PENULIS)
(TANDATANGAN PENYELIA)
Alamat Tetap:
6, JLN TEMBAGA KUNING 14,
TAMAN SRI SKUDAI,
81300 SKUDAI,
JOHOR DARUL TAKZIM.
27th MARCH 2006
Tarikh:
CATATAN: *
**
♦
DR AZILA ABDUL AZIZ
Nama Penyelia
Tarikh:
27th MARCH 2006
Potong yang tidak berkenaan.
Jika tesis ini SULIT atau TERHAD, sila lampirkan surat daripada pihak berkuasa/organisasi
berkenaan dengan menyatakan sekali sebab dan tempoh tesis ini perlu dikelaskan sebagai
SULIT atau TERHAD.
Tesis dimaksudkan sebagai tesis bagi ijazah Doktor Falsafah dan Sarjana secara penyelidikan,
atau disertai bagi pengajian secara kerja kursus dan penyelidikan, atau Laporan Projek
Sarjana Muda (PSM).
2
" We hereby declared that we have read this thesis and in our
opinion this thesis is sufficient in terms of scope and quality for the
award of the degree of Master of Engineering (Bioprocess)".
Signature
Name of Supervisor I
: ……………………………………..
DR AZILA ABDUL AZIZ
:
Date
: 27th MARCH 2006
Signature
Name of Supervisor II
: ……………………………………..
DR BADARULHISAM ABDUL RAHMAN
:
Date
: 27th MARCH 2006
i
DEVELOPMENT OF NOVEL GALACTOSYLATION METHOD
FOR THE EXPRESSION OF RECOMBINANT HUMAN TRANSFERRIN
IN INSECT CELL CULTURE
YAP WEI NEY
A thesis submitted in fulfilment of the
requirements for the award of the degree of
Master of Engineering (Bioprocess)
Faculty of Chemical and Natural Resources Engineering
Universiti Teknologi Malaysia
MARCH 2006
ii
I declare that this thesis entitled “Development of Novel Galactosylation Method for
the Expression of Recombinant Human Transferrin in Insect Cell Culture” is the
result of my own research except as cited in the references. The thesis has not been
accepted for any degree and is not concurrently submitted in candidature of any other
degree.
Signature
:
Name
:
Date
:
……………………………………..
YAP WEI NEY
27th MARCH 2006
iii
To my beloved parents, sister and brother
iv
ACKNOWLEDGEMENT
In carrying out this thesis, I wish to express my deepest gratitude and thanks
to my main thesis supervisor, Dr. Azila Abdul Aziz for her encouragement,guidance,
critics and support. I am also very thankful to my co-supervisor, Dr. Badarulhisam
Abdul Rahman for his continual professional advice, dedicated guidance especially
he is the Vice President in Inno Biologics, Cyberjaya but he is still willing to take his
precious time to guide me in the effort to complete my research works.
I am also indebted to Universiti Teknologi Malaysia for funding my Master
study. Prof. Dr. Michael J. Betenbaugh of Johns Hopkins University, USA also
deserved special thanks for the recombinant baculoviruses as well as Prof. Dr. Mohd
Sanusi Jangi of Univesiti Kebangsaan Malaysia for his wild type baculoviruses.
I would like to extend my gratitude to all the laboratory staffs including Puan
Siti Zalita Abdul Talib, Encik Yaakop Sabudin and Encik Malek Yusof. They had
been very helpful and cooperative in providing assistance throughout the work. I
would to convey my sincere appreciations to my labmates and friends, Wong Fui
Ling, Wee Chen Chen, Clarence M. Ongkudon, Mohd Hafiz, Melissa Loh, Dr. Tahir,
Kamalesh, Tan Khooi Yeei and Lau Seat Yee for their friendship and understanding.
My special thanks to my boyfriend, Ng Ping for his endless encouragement.
Last but not least, I would like to thank the most important person in my life,
my mother, for her prayers and undivided love to me during this study.
v
ABSTRACT
The objective of this research is to develop a novel galactosylation method
for the expression of recombinant human transferrin (hTf) with better N-glycan
quality. The baculovirus-insect cell system, consisting of hTf as the model protein,
β1,4-galactosyltransferase (β1,4-GalT) as the enzyme, and uridine-diphosphogalactose (UDP-Gal) as the sugar nucleotide, has been successfully established. In
the early part of the study, fundamental works were carried out to optimize
Spodoptera frugiperda (Sf-9) cells growth and mock infection. Serum concentration,
different type of media, cell subculturing condition, initial cell density and spent
medium carry over had been found to significantly influence the growth kinetics of
Sf-9 cells. Multiplicity of infection (MOI) and spent medium carry over were found
to have direct impact on viral infectivity. The optimized parameters were then used
to evaluate the expression of recombinant hTf and β1,4-GalT in Sf-9 cells.
Subsequently, native UDP-Gal levels at normal and upon baculovirus infection
produced in Sf-9 cells were monitored using Reverse Phase High Performance Liquid
Chromatography.
UDP-Gal concentration was discovered to decrease gradually
once infected with the recombinant baculovirus. Finally, baculovirus coinfection
study was carried out to evaluate the recombinant glycoprotein quality. However,
lectin binding analysis using Ricinus communis agglutinin-I, revealed that coexpression between rhTf and β-1,4GalT (in vivo) did not show encouraging result
due to the reduction of UDP-Gal upon baculovirus infection. This finding suggested
that the introduction of β-1,4GalT alone was not sufficient for successful
galactosylation. However, another strategy was used to overcome the problem.
Commercial GalT and UDP-Gal were introduced artificially to the rhTf after it was
secreted from cell culture. It was found that the in vitro strategy promoted better Nglycan quality in insect cells.
vi
ABSTRAK
Kajian ini bermatlamat untuk mengkaji proses galaktosilasi yang baru bagi
penghasilan rekombinasi human transferrin (hTf) dengan kualiti N-glikan yang lebih
baik. Sistem bakulovirus-sel serangga yang terdiri daripada hTf sebagai protein
model,
β1,4-galaktositransferasa
(β1,4-GalT)
sebagai
enzim,
dan
uridina-
diphosphogalaktosa (UDP-Gal) sebagai gula nukleotida telah dibentuk dengan
berjaya. Dalam kajian awal, kerja asas mengenai pengoptimuman telah dilakukan
bagi pertumbuhan sel dan jangkitan bakulovirus tanpa membawa gen tertentu dalam
sel Spodoptera frugiperda (Sf-9). Kepekatan serum, medium yang berbeza, keadaan
sel subkultur, ketumpatan sel awal dan medium telah-guna telah memberi kesan yang
ketara terhadap kinetik pertumbuhan sel Sf-9. Gandaan Jangkitan dan medium telahguna menunjukkan kesan terus terhadap infektiviti virus. Semua parameter yang
telah dioptimumkan telah digunakan untuk menilai ekpresi bagi rekombinasi hTf and
β1,4-GalT dalam sel Sf-9. Seterusnya, tahap UDP-Gal semulajadi pada normal dan
atas jangkitan bakulovirus yang dihasilkan dianalisis dengan menggunakan Fasa
Terbalik Kromatografi Cecair Pertunjukkan Tinggi.
Didapati bahawa kepekatan
UDP-Gal menurun secara perlahan sebaik sahaja dijangkiti dengan rekombinasi
bakulovirus. Akhirnya, jangkitan serentak bakulovirus telah dilakukan bagi menilai
kualiti glikoprotein rekombinasi. Tetapi, analisis lektin perlekatan dengan
menggunakan Ricinus communis agglutinin-I, menunjukkan in vivo galaktosilasi
tidak cukup berkesan disebabkan kekurangan UDP-Gal semasa jangkitan
bakulovirus. Keputusan yang menarik ini mencadangkan bahawa penambahan β1,4GalT sahaja tidak cukup untuk menjayakan galaktosilasi. Oleh itu, strategi lain telah
digunakan untuk mengatasi kelemahan ini.
GalT mamalia dan UDP-Gal yang
diperolehi secara komersil diperkenalkan kepada supernatan hTf yang dikumpul.
Didapati bahawa kaedah ini berjaya meningkatkan kualiti N-glikan dengan baik.
vii
TABLE OF CONTENTS
CHAPTER
1
2
TITLE
PAGE
INTRODUCTION
1
1.1
Introduction
1
1.2
Objectives
3
1.3
Scopes of Research
3
LITERATURE REVIEW
5
2.1
Recombinant Protein Manufacturing Technologies
6
2.2
Glycosylation
8
2.2.1
N-Linked Glycosylation
9
2.2.2
O-Linked Glycosylation
9
2.3
Glycoprotein
10
2.4
Insect Cell- Baculovirus Expression System
11
2.4.1
Insect Cell Lines
11
2.4.2
Baculoviruses
12
2.4.2.1 Baculoviruses Replication
14
2.4.2.1.1 In Vivo Replication
14
2.4.2.1.2 In Vitro Replication
15
2.5
Advantages of BEVS Technology
20
2.6
Glycosylation in Insect Cells
22
viii
2.7
Glycosyltransferases and Glycosidases Involved
24
in N-glycan Processing in Insect Cells
2.7.1
α-Glucosidase I, II and α-Mannosidase I
24
2.7.2
N-Acetylglucosaminyltransferase I
25
(GlcNAcT-I) and α-mannosidase II
2.7.3
N-Acetylglucosaminyltransferase II
25
(GlcNAcT-II)
2.7.4
β-1,4-Galactosyltransferase (β-1,4GalT)
26
2.7.5
Core α-1,3- and α-1,6-Fucosyltransferases
26
(FucT)
2.8
2.7.6
β-N-Acetylglucosaminidase
27
2.7.7
Sialyltransferase (SiaT)
27
Sugar Nucleotides Involved in N-glycan
28
Processing in Insect Cells
2.8.1
Endogenous Sugar Nucleotide Levels
28
in Lepidopteran Insect Cells
2.8.2
Enzymes Involved in Sialic Acid and
28
CMP-Sialic Acid Synthesis
2.9
Engineering of N-glycan Processing Pathway
30
2.9.1
32
Improvement of N-Acetylglucosaminylation
of the Manα(1,3)- Branch
2.9.2
Improvement of Galactosylation
32
2.9.3
Production of Biantennary Complex-Type
33
N-glycans
2.10
2.9.4
Formation of Sialylated N-glycans
33
2.9.5
Synthesis of CMP-NeuNAc
33
Galactosylation in N-Glycan Processing in
34
Insect Cells
2.10.1 Sugar acceptor
35
2.10.2 Substrate Donor
35
2.10.3 Enzyme
38
ix
3
MATERIALS AND METHODS
41
3.1
Materials
41
3.2
Equipments
41
3.3
Chemicals
42
3.4
Spodoptera frugiperda (Sf-9) Insect Cells
43
3.4.1
Cells Thawing
43
3.4.2
Cells Maintaining
43
3.4.3
Cells Freezing
44
3.5
Wild Type and Recombinant Baculovirus
44
3.5.1
Virus Propagation
44
3.5.2
Virus Titrationr (End-Point Dilution)
45
3.5.3
Generating Pure Recombinant
46
Virus Stocks (End Point Dilution)
3.6
Recombinant Human Transferrin Detection
46
3.6.1
46
A
3.7
3.8
Sodium Dodecyl Sulfate – Polyacrylamide
Gel Electrophoresis
3.6.1.1 Silver Staining
47
3.6.2
Western Blot
48
3.6.3
Enzyme Linked Immunosorbent Assay
49
Recombinant β1,4-Galactosyltransferase Detection
49
3.7.1
Thin Layer Chromatography
49
3.7.2
Lectin Binding Assay
50
Native Uridine-5′-diphosphogalactose
51
(UDP-Gal) Level
3.8.1
UDP-Gal Extraction
51
3.8.2
Reverse Phase High Performance Liquid
52
Chromatography (RP-HPLC) Analysis
3.9
Coexpression of Recombinant Human Transferrin and
β1,4-Galactosyltransferase
52
x
4
RESULTS AND DISCUSSION
54
4.1
Sf-9 Cells Growth Optimization
55
4.2
Establishment of Baculovirus Expression
63
Vectors System (BEVS)
4.2.1
Mock Infection Optimization
63
4.2.2
Recombinant Human Transferrin
71
Expression
4.2.2.1 Time Course Expression of Recombinant
71
Human Transferrin
4.2.3
Recombinant β1,4-Galactosyltransferase
75
Expression
4.2.3.1 Time Course Expression of β1,4-
75
Galactosyltransferase
4.2.3.2 The Development of β1,4-
79
Galactosyltransferase Assay
4.2.4
Native Uridine-diphosphogalactose (UDP-Gal)
82
Monitoring at Normal and Upon Baculovirus
Infection
4.2.5
5
Baculovirus Coinfection Study
94
CONCLUSIONS
102
5.1
Conclusion
102
5.2
Further studies
104
REFERENCES
106
Appendices A-H
122
xi
LIST OF TABLES
TABLE NO.
4.1
TITLE
Growth Kinetics of Sf-9 Cells at Different
Parameters
PAGE
68
xii
LIST OF FIGURES
FIGURE NO.
2.1
TITLE
(a) N-linked protein glycosylation; (b) O-linked protein
PAGE
8
glycosylation
2.2
(a) High Mannose, (b) Complex and (c) Hybrid structures
10
of carbohydrates on the 3 major classes of glycoprotein
2.3
A few insect species used for glycoprotein production
11
2.4
Autographa californica multiple nuclear polyhedrosis virus
13
(AcMNPV)
2.5
A) Baculovirus particles, or polyhedra; B) Cross-section of
13
a polyhedron; C) Diagram of polyhedron cross-section.
(Jean Adams and V. D'Amico.)
2.6
In vivo baculovirus infection and replication
14
2.7
In vitro baculovirus infection and replication
17
2.8
Structural compositions of the two baculovirus phenotypes,
18
budded virus (BV), and the occlusion derived virus (ODV)
(Blissard, 1996)
2.9
a) A typical infected Sf-9 cells ( Steven Howard);
19
(b) Electron micrograph of AcMNPV infected Sf-9 Cell
(Greg V.Williams); (c) A portion of the nucleus containing
enveloped virions in the process of being occluded into a
developing polyhedron (Queen's University)
2.10
Protein N-glycosylation pathways in insect and mammalian
cells
23
xiii
2.11
CMP-Neuraminic acid synthesis pathway
29
2.12
General strategy for humanization of glycoprotein produced
31
by lepidopteran cell-baculovirus expression system
2.13
Structure of a nucleotide sugar that can serve as a sugar
37
donor in a glycosyltransferase reaction
2.14
Transporters for sugar nucleotides, PAPS, and ATP are
37
located in the Golgi membranes of mammals, yeast,
protozoa, and plants
3.1
Virus Titer Procedures – End Point Dilution
45
4.1
Sf-9 insect cells growth in monolayer culture at 3 different
56
serum concentrations. (a) TC-100 and (b) SF-900 II SFM
4.2
Sf-9 insect cells growth in monolayer culture for 2 types
58
of media. (a) without serum;(b) with 5% serum and
(c) with 10% serum
4.3
Sf-9 insect cells growth in monolayer culture for 3 different
59
initial cell density, i.e. 0.2, 1.2 and 2.33 x 105 cells/ml
4.4
Sf-9 insect cells growth in monolayer culture at 3 different
61
subculturing conditions, i.e. early exponential, late exponential
and stationary phase
4.5
Sf-9 insect cells growth in monolayer culture at 3 different
63
spent medium carry over percentage, i.e 100%, 50% and 0%
4.6
The effect of initial cell density on Sf-9 insect cells infected
64
with wild type AcMNPV viruses at MOI 10
4.7
The effect of spent medium carry over on Sf-9 insect cells
65
infected with wild type AcMNPV viruses at MOI 10
4.8
The effect of MOI on Sf-9 insect cells infected in the
67
stationary phase with wild type AcMNPV Viruses
4.9
(a) 9% SDS-PAGE analysis with silver stained; (b) Western
72
blot analysis indicated the rhTf protein synthesized in Sf-9
cells supernatant at hour 120
4.10
Time Course of rhTf protein production in supernatants
73
were resolved on 9% SDS-PAGE and stained with silver
4.11
Time course of rhTf protein production in (a) Lysates;
(b) Supernatants were detected using ELISA.
74
xiv
4.12
Detection of β1,4-GalT by using chromatogram of
76
thin layer chromatography
4.13
Time course of chromatogram of thin layer chromatography.
77
4.14
SDS-PAGE (9%) Time Course of β1,4-GalT Production
78
4.15
Standard curve for the determination of β1,4-GalT
80
activity from the Lectin Binding Assay values
4.16
Time course of β1,4-GalT enzyme production in
81
supernatants were detected using lectin binding assay.
4.17
RP-HPLC chromatogram for UDP-Gal standard at different
84
concentration
4.18
Standard curve for UDP-Gal
85
4.19
RP-HPLC chromatogram for native UDP-Gal sample with
87
spiking and without spiking
4.20
RP-HPLC Chromatogram for the time course of native
88
UDP-Gal level upon infection time at (a) 0h (Normal);
(b) 24h; (c) 48h; (d) 72h; (e) 96h and (f) 120h (Set Data 1)
4.21
RP-HPLC chromatogram for time course of native
89
UDP-Gal level upon infection in 3D diagram (Set Data 1)
4.22
RP-HPLC Chromatogram for the time course of native
90
UDP-Gal level upon infection time at (a) 0h (Normal);
(b) 24h; (c) 48h; (d)72h; (e) 96h and (f) 120h (Set Data 2)
4.23
RP-HPLC chromatogram for time course of native
91
UDP-Gal level upon infection in 3D diagram (Set Data 2)
4.24
Native UDP-Gal concentration in µM at normal and upon time
92
of infection
4.25
Verification of UDP-Gal fractions from RP-HPLC
93
analysis by using chromatogram of TLC
4.26
Galβ1→4GlcNAc linkage binding values at 450nm
for the time course upon coinfection between recombinant
baculovirus hTf and β1,4-GalT
96
xv
4.27
Effect of the mammalian galactosyltransferase on the
97
rate of in vitro galactosylation process
4.28
Galβ1→4GlcNAc linkage binding values at 450nm for the
100
different level of galactosylation process
4.29
Relationships among the three main elements in in vivo
galactosylation process
101
xvi
LIST OF SYMBOLS/ ABBREVIATIONS
2-ADN
-
2-acetamide-1,2-dideoxynojirimycin
AcMNPV
-
Autographa californica multicapsid nucleopolyhedrovirus
Asp
-
Asparagine
Ba(OH)2
-
barium hydroxide
BEVS
-
baculovirus expression vectors system
bIFN- γ
-
bovine interferon-γ
Bm
-
Bombyx mori
BSA
-
bovine serum albumin
BVs
-
budded viruses
CaCl2
-
calcium chloride
CHO
-
chinese hamster ovary
CMP
-
cytidine-5’-monophosphate
CMP-NeuNAc-
cytidine-5’-monophospho N-acetylneuraminic acid
CMP-SAS
-
CMP-NeuNAc synthase
DMSO
-
dimethyl sulphoxide
DNA
-
deoxyribonucleic acid
E.Coli
-
Escherichia coli
Ea
-
Estigmene acrea
xvii
EDTA
-
ethylenediamine tetraacetic acid disodium salt dehydrate
ELISA
-
Enzyme Linked Immunosorbent Assay
ER
-
endoplasmic reticulum
FBS
-
fetal bovine serum
Fuc
-
fucose
FucT
-
Fucosyltransferases
Gal
-
galactose
GalNAc
-
N-Acetylgalactosamine
GDP-Fuc
-
guanosine 5’-diphoshate-β-L-fucose
GDP-Man
-
guanosine 5’-diphoshate-D-mannose
Glc
-
glucose
GlcNAc
-
N-Acetylglucosamine
GlcNAcT II
-
N-Acetylglucosaminyltransferase II
GlcNAcT-I
-
N-Acetylglucosaminyltransferase I
H2O2
-
peroxidase
H3PO4
-
phosphoric acid
HCl
-
hydrochloric acid
HRP
-
horseradish peroxidase
hTf
-
human serum transferrin
IgG
-
immunoglobulin G
kbp
-
kilobasepairs
kDa
-
kilodalton
LacNAc
-
N-Acetyllactosamine
M
-
molar
Man
-
mannose
ManNAc
-
N-Acetylmannosamine
ManNAc
-
N-acetylmannosamine
ManNAc-6-P -
N-acetylmannosamine-6-phosphate
MB
-
Mamestra brassicae
Mg
-
magnesium
min
-
minute
mm
-
mililiter
MnCl2
-
manganese chloride
MOI
-
Multiplicities of Infection
xviii
MOPS
-
4-Morpholinepropanesulfonic acid
MWCO
-
molecular weight cut off
NaCl
-
sodium chloride
NAG
-
N-acetylglucosamine
NAL
-
N-Acetyllactosamine
NeuNAc
-
N-acetylneuraminic acid
NeuNAc-9-P -
N-Acetylneuraminic acid-9-phosphate
nm
-
nanometer
NOV
-
non-occluded virus particles
NPV
-
nucleocapsid nuclear polyhedrovirus
OBV
-
occlusion body-derived virus particles
PBS
-
Phosphate Buffer Saline
PBST
-
PBS containing 0.05% Tween 20
PI
-
Post infection
RCA I
-
Ricinus communis agglutinin 1
RP-HPLC
-
Reverse Phase High Performance Liquid Chromatography
rpm
-
rotation per minutes
SAS
-
N-Acetylneuraminate-9-phosphate synthase
SDS
-
sodium dodecyl sulfate
SDS-PAGE
-
sodium dodecyl sulfate-polyacrylamide gel electrophoresis
Sf
-
Spodoptera frugiperda
SiaT
-
sialyltransferase
TBAS
-
tetrabutylammonium hydrogen sulfate
TBS
-
Tris-buffered Saline
TCID50
-
Tissue Culture Infectious Dose 50
TEMED
-
N,N,N',N'-tetramethylethylenediamine
TLC
-
Thin Layer Chromatography
TMB
-
3,3’,5,5’-tetramethylbenzidene
Tn
-
Trichoplusia ni
TOI
-
Time of infection
UDP-Gal
-
uridine-diphosphogalactose
UDP-GlcNAc -
uridine-5’-diphopho-N-acetylglucosamine
UDP-hexose -
uridine-5’-diphopho-D-hexose
UF
ultrafiltration
-
xix
UTP
-
uridine 5’-triphosphate sodium
UV
-
ultraviolet
ZnSO4.7H2O -
zinc sulfate 7-hydrate
α2,6-ST
-
α2,6-sialytransferase
β1,4-GalT
-
β1,4-galactosyltransferase
µl
-
microliter
µm
-
micrometer
0
-
degree Celcius
C
xx
LIST OF APPENDICES
APPENDIX
TITLE
PAGE
A-1
Monosaccharide Mass and Structure
122
A-2
Common N-Linked Glycan Simplified
123
Structures and Masses
A-3
Cell Culture Glossary
124
B-1
Stock Solution for SDS-PAGE
126
B-2
Working Solution for SDS-PAGE
127
B-3
Separating and Stacking Gel Preparation
128
C
Working Solution for ELISA
129
D
Working Solution for Western Blot
130
E
Virus Calculation
131
F
Reaction Mixture for Lactose Synthetase
132
Assay
G
Reaction Mixture for Lectin Binding Assay
135
H
Publications
136
xxi
CHAPTER 1
INTRODUCTION
1.1
Introduction
Many glycoproteins have been produced by a variety of expression systems
including cell cultures of mammalian or insect cell lines. Of particular interest has
been the baculovirus expression system that generates high levels of recombinant
proteins from insect cells such as Spodoptera frugiperda (Sf-9).
The potential
production of therapeutic glycoproteins in these systems has stimulated the desire to
monitor the glycosylation pattern of specific insect-cell-produced glycoproteins and
the glycosylation potential of insect cells in general.
significantly
affect
a
protein’s
stability,
The glycan moieties can
biological
activity,
antigenicity,
immunogenicity, solubility, cellular processing, secretion and pharmacokinetic
behaviour such as in vivo metabolic clearance rate (Takeuchi et al., 1990, Takeuchi
and Kobata, 1991, Munk et al., 1992).
It is well documented that the N-glycans found in recombinant glycoproteins
expressed by lepidopteran cells using the baculovirus vector are predominantly high
mannose type glycans and short truncated glycans (paucimannose) with α1,3/ α1,6linked fucose residue on its asparagines-bound N-acetylglucosamine (GlcNAc)
2
(Jarvis and Summers, 1989; Wathen et al., 1991; Grabenhorst et al.,1993; Yeh et al.,
1993; Manneberg et al., 1994; Ogonah et al.,1995; Hsu et al.,1997; Opez et al.,
1997). In contrast, mammalian cells usually produce sialylated complex-type Nglycans. Generation of complete forms of sialylated complex-type N-glycans in
insect cells may increase the value of insect cell derived products as vaccines,
therapeutic and diagnostics.
The glycosylation process in the cultured cells can be controlled by various
factors, which are sugar acceptor as model protein, substrate donor also known as
sugar nucleotide and glycosyltransferase as enzyme.
Activity measurements of
several glycosyltransferases involved in the elongation of N-glycans have
demonstrated that insect cells contain α1,6-fucosyltransferase (Staudacher et
al.,1992) and a significant level of β1,2-N-acetylglucosaminyltransferase I activities
(Velardo et al.,1993, Altmann et al.,1993) but they lack significant β1,4galactosyltransferase (Butters et al.,1981; van Die et al., 1996) and sialytransferase
activities (Hooker et al., 1999).
In addition to glycosyltransferases, another
important factor in protein glycosylation is the sugar nucleotides essential in the
biosynthesis of glycoconjugates.
Since these are the substrate donor of
glycosyltransferases that construct the glycan chains, the intracellular levels of sugar
nucleotides can affect the glycosylation potential of the cultured cells as well.
In this study, we will focus on the galactosylation processing pathway rather
than the whole glycosylation process. In the galactosylation process, recombinant
human transferrin (hTf) is the substrate acceptor, β1,4-galactosyltransferase (β1,4GalT) is the glycosyltransferase, and uridine-5’-diphosphogalactose (UDP-Gal) is
the substrate donor. Recombinant hTf will be used as a model protein simply due to
its simple biantennary N-glycan structure. Ailor et al. (2000) revealed that the Nglycan structures of hTf produced in insect cells included high mannose,
paucimannosidic, and hybrid structures with over 50% these structures containing
one or two fucoses linked to the Asn-linked N-acetylglucosamine. Furthermore,
neither sialic acid nor galactose was detected on any of the N-glycan.
3
In this study, establishment of the native UDP-Gal level at normal and upon
baculovirus infection was performed. It is proposed that analysis of the intracellular
concentration of sugar nucleotides could provide important information on the
potential of galactosylation in Sf-9 insect cells as little is known about the level of
native UDP-Gal level especially upon baculovirus infection. To evaluate the quality
of the recombinant hTf, different levels of galactosylation were conducted to obtain
the galactosylated hTf, including in vivo and in vitro study. In vivo refers to the
baculovirus coinfection by coexpressing β-1,4GalT and hTf simultaneously in
cultured cell, meanwhile the introduction of commercial GalT and UDP-Gal to the
hTf after it was secreted from insect cell cultures called as in vitro.
1.2
Objectives
(1)
To develop a method for the expression of galactosylated recombinant hTf in
insect cells
(2)
To optimize the expression of the galactosylated recombinant hTf
1.3
Scopes of Research
(1)
Recloning of recombinant virus stock, virus propagation and virus titration
(2)
Establishment of an assay system for the detection of β1,4-GalT and UDP-Gal
(3)
Monitoring of native UDP-Gal level at normal and upon baculovirus infection
(4)
Evaluation of the quality of the glycoprotein obtained through baculovirus
coinfection study to coexpress β1,4-GalT and hTf (in vivo study) and the
4
artificial introduction of commercial GalT and UDP-Gal to secreted hTf (in
vitro study)
5
CHAPTER 2
LITERATURE REVIEW
The insect cell-baculovirus expression system is widely used for the
production of a large number of glycoproteins due to its ability to express high levels
of heterologous proteins. This protein expression system primarily uses cell lines
established from lepidopteran insects and the Autographa californica multicapsid
nucleopolyhedrovirus (AcMNPV) mediated expression vector (Summers and Smith,
1987).
Lepidopteran-derived insect cell lines such as Sf-9 (from Spodoptera
frugiperda) and Tn-5B1-4 (from Trichoplusia ni) perform many of the posttranslational modifications observed in eukaryotic cells, including N-glycosylation,
and these cells are readily cultured in suspension in commercially available medium.
Unfortunately, however, most insect cells examined so far are incapable of
synthesizing sialylated complex-type N-glycans often found in glycoproteins
obtained from mammalian cells.
The inability of lepidopteran insect cells to
synthesize sialylated complex-type N-glycans and the presence of α1,3-fucosylation
have limited the insect cells as host cells for production of pharmaceutical
glycoproteins. The limitations of currently used insect cell lines may potentially be
overcome by means of genetic manipulation to include the necessary processing
enzymes (Ailor et al., 2000; Aumiller and Jarvis, 2002; Breitbach and Jarvis, 2001;
Hollister et al., 1998, 2002; Hollister and Jarvis, 2001; Tomiya et al., 2003), or by
use of an alternative insect cell line that may contain mammalian-like N-glycan
processing capabilities.
6
2.1
Recombinant Protein Manufacturing Technologies
A variety of protein expression systems have been developed which are
currently focusing on the therapeutic purpose that is for human use.
In
biopharmaceutical, the recombinant protein produced must achieving and
maintaining several important criteria such as efficacy, safety, immunogenic reaction
and blood circulation time.
Common bacterial expression system such as Escherichia coli (E.Coli) is the
simplest recombinant protein manufacturing process.
However, the bacterial
fermentation is associated with several drawbacks. For example, the products are not
glycosylated like natural human proteins and are therefore likely to cause side effects
in the therapeutic use. In addition, bacterial proteins tend to have more sequence
translation errors and fold less consistently than glycosylated proteins, so biological
activity can be variable. Furthermore, bacteria cannot be used to manufacture very
large protein such as erythropoeitin or multiprotein assemblies such as antibodies.
Yeast expressions offer a simple production process with high yield, powerful
secretary pathways, and some limited post-translational modifications. However, the
glycosylation often has to adjust after purification to produce a closer match with
human glycosylation patterns.
Mammalian cell culture is a slow and expensive process. Chinese hamster
ovary (CHO), mouse-human hybridomas, myelomas, and human cell lines are some
examples. Currently, all commercial antibodies are produced via mammalian cell
culture. The main advantages of mammalian products are they can be engineered to
produce more than one protein simultaneously and the cells have near-correct human
glycosylation but do not maintain complete glycosylation under production lines.
The cell culture process takes 3 to 4 months and required a special serum-free,
chemically defined medium as well as temperatures maintained precisely in the range
7
of 37-42 0C. The acidity of the culture must be kept very close to neutral; this
process involves gentle bubbling of nitrogen or carbon dioxide gases through the
fermenter to make slight adjustments. Also, mammalian cells are more fragile than
bacterial or yeast cells and can shear or break open if subjected to rough mixing or
bubbling. Yields are lower than from either yeast or bacterial fermentation.
Transgenic animals are being studied as an alternative to traditional CHO cell
production processes.
Transgenic animals provide a potentially less expensive
source of production for proteins compared to traditional cell culture systems.
Although the transgenic expression systems may solve the problem of protein
production yields and may lower the cost, they do not solve the problem of protein
glycosylation.
Recently, Baculovirus-mediated expression in insect cells has become wellestablished for the production of recombinant glycoproteins. Its frequent use arises
from the relative ease and speed with which a heterologous protein can be expressed
on the laboratory scale and the high chance of obtaining a biologically active protein.
In addition to Spodoptera frugiperda Sf-9 cells, which are probably the most widely
used in insect cell line, other mainly lepidopteran cell lines are exploited for protein
expression.
Recombinant baculovirus is the usual vector for the expression of
foreign genes but stable transfection of – especially dipteran-insect cells presents an
interesting alternative. Insect cells can be grown on serum free media which is an
advantage in terms of cost as well as of biosafety. For large scale culture, conditions
have been developed which meet the special requirements of insect cells. With
regards to protein folding and post-translational processing, insect cells are second
only to mammalian cell lines. Evidence is presented that many processing events
known in mammalian systems do also occur in insects (Altmann et al., 1999).
However, on protein glycosylation, particularly N-glycosylation, which is insect,
differs in many respects from that in mammals.
8
2.2
Glycosylation
Glycosylation is the most common post-translational modifications made to
proteins by eukaryotic cells, and can significantly affect biological activity and is
particularly important for recombinant glycoproteins in human therapeutic
applications. Glycosylation is a process where oligosaccharides, or sugar chain are
covalently linked to proteins.
The predominant sugars found on human
glycoproteins, include galactose, mannose, fucose, N-acetylgalactosamine (GalNAc),
N-acetylglucosamine (GlcNAc) and N-acetylneuraminic acid (the human form of
sialic acid). There are two types of glycosylation, which are N-glycosylation and Oglycosylation (Figure 2.1).
(a)
(b)
Figure 2.1: (a) N-linked protein glycosylation. The N-linked amino acid
consensus sequence is Asn-any AA- Ser or Thr. The middle amino acid can not
be proline (Pro); (b) O-linked protein glycosylation. Most O-linked
carbohydrate covalent attachments to proteins involve a linkage between
the monosaccharide N- Acetylgalactosamine and the amino acids serine or
threonine (Adapted from Altmann, 1996).
9
2.2.1
N-Linked Glycosylation
N-linked glycosylation is the oligosaccharide which link to the amino group
of asparagine (N) and have a core of Asp-GlcNAc-GlcNAc-Man-(Man)2 derived
from dolichol. The processing of N-glycans occurs co-translationally in the lumen of
the endoplasmic reticulum (ER) and continues in the Golgi apparatus. It can be bi-,
tri- and tetraantennary or if it is a poly-N-acetyl lactosamine type, it can be branched
or unbranched. There are three types of N-linked glycosylation which are complex,
hybrid, and high mannose (Figure 2.2 (a), (b) and (c)). The major distinguishing
feature of the complex class is the presence of sialic acid, whereas the hybrid class
contains no sialic acid. In contrast to the step-wise addition of sugar groups to the Olinked class of glycoproteins, N-linked glycoprotein synthesis requires a lipid
intermediate that is dolichol phosphate.
Dolichols are polyprenols (C80-C100)
containing 16 to 20 isoprene units, in which the terminal unit is saturated.
2.2.2
O-Linked Glycosylation
O-linked glycosylation most commonly links GalNAc to the hydroxyl group
of serine or threonine and occurs post-translationally in the Golgi apparatus. There
are no consensus sequence, no preformed intermediate and starts in trans Golgi. In
comparison with the N-linked glycosylation, most other O-linked glycans are highly
branched. Sulphates can be on Gal, GalNAc and GlcNAc and phosphate can be on
Man or Xyl.
10
2.3
Glycoprotein
Glycoproteins are the proteins that include complex carbohydrates as part of
their structure.
The carbohydrate components of glycoproteins affect the
functionality of the molecule by determining protein folding, oligomer assembly and
secretion processes. Without the proper shape, the ability of the protein to interact
correctly with its receptor is affected, possibly affecting function. There are three
types of glycoproteins (Figure 2.2).
(a)
Figure 2.2:
(b)
(c)
(a) High Mannose, (b) Complex and (c) Hybrid Structures of
carbohydrates on the 3 major classes of glycoprotein. (Adapted from Altmann,
1996)
11
2.4
Insect Cell- Baculovirus Expression System
The baculovirus-insect cell expression system is a binary system consisting of
a recombinant baculovirus vector and its host, an insect cell. The virus delivers the
gene encoding a glycoprotein of interest to the cell, then the gene is expressed by
virus-encoded transcription factors and the protein is translated and glycosylated by
host cell machinery during viral infection.
2.4.1
Insect Cell Lines
Spodoptera
frugiperda
Figure 2.3:
Estigmene
acrea
Mamestra
brassicae
Bombyx
mori
A few insect species used for glycoprotein production (Adapted
from Tomiya et al., 2003)
Sf-9 and Sf-21 cells from the fall army worm Spodoptera frugiperda are the
most frequently used cell lines used in the heterologous expressions. However, quite
a number of other cell lines has been established, including cell lines stem from
Trichoplusia ni, i.e., TN-368 and BT1-TN-5B1-4, Bombyx mori (Bm-N), Mamestra
brassicae (e.g., MB0503) and Estigmene acrea (Ea). In general, stable cell lines are
usually obtained from embryonic cells and thus represent essentially undifferentiated
cells. An alternative strategy but there appears to be few data is to infect whole
larvae with recombinant baculovirus (Reis et al., 1992; Maeda et al., 1982; Korth et
al., 1993).
12
2.4.2
Baculoviruses
The most widely used vectors for the production of foreign proteins in insect
cells or larvae are recombinant baculoviruses, such as Autographa californica
multicapsid nucleopolyhedrovirus (AcMNPV) as shown in Figure 2.4, which infects
lepidopteran cells (David, 1994; Luckow, 1995). Baculoviruses are viral pathogens
that cause fatal disease in insects, mainly in members of the families Lepidoptera,
Diptera, Hymenoptera and Coleoptera. More than 600 baculovirus isolates have
been described, categorized in two subfamilies, (a) nucleopolyhedroviruses and (b)
granuloviruses (Murphy et al., 1995).
Baculoviruses are characterized by the presence of rod shape nucleocapsids,
that are enveloped singly or in bundles by a unit membrane (Figure 2.4 and 2.5). The
virus particles usually embedded into large protein capsules or occlusion bodies (OB),
also called polyhedra c.q. granula. These OBs, 0.1-10 µm in diameter, provide
protection of the virus particle and enhance the persistence of the virus in the
environment. The occlusion body-derived virus particles (OBV) are the infectious
entities of the OBs.
The major constituent of OBs is a single protein (polyhedron c.q granulin)
with a subunit molecular weight of approximately 30 kDa. The amino acid sequence
is
highly
preserved
among
baculoviruses
(Vlak
and
Rohrmann,
1985).
Baculoviruses contain a double stranded, circular DNA molecule as genetic element.
This DNA varies in size between 100 and 200 kbp and is able to code for more than
70 average-sized proteins. Physical maps of various baculovirus DNAs have been
established, the most detailed one being of the prototype baculovirus AcMNPV,
whose genome has been entirely sequenced (Ayres et al., 1994).
About forty
functional genes have been mapped on the AcMNPV genome, including polyhedrin.
Approximately thirty of these have also been sequenced transcriptionally and
analyzed (Blissard and Rohrmann, 1990; Kool and Vlak, 1993).
13
Figure 2.4:
Autographa californica multicapsid nucleopolyhedrovirus
(AcMNPV). This high magnification electron micrograph shows a negativelystained baculovirus virion. Note the asymetric capsid structure and the presence
of an envelope with surface projections (peplomers). (From the Carstens' Lab at
Queen's University, Canada)
Figure 2.5:
A) Baculovirus particles, or polyhedra; B) Cross-section of a
polyhedron; C) Diagram of polyhedron cross-section. (Electron micrographs
(A&B) by Jean Adams, graphic (C) by V. D'Amico)
14
2.4.2.1 Baculoviruses Replication
2.4.2.1.1 In Vivo Replication
Viruses are unable to reproduce without a host because they are obligate
parasites. Baculoviruses are no exception. The cells of the host's body are taken
over by the genetic message carried within each virion, and forced to produce more
virus particles until the cell, and ultimately the insect, dies. Most baculoviruses
cause the host insect to die in a way that will maximize the chance that other insects
will come in contact with the virus and become infected in turn. Figure 2.6 shows
the infection by baculovirus begins when an insect eats virus particles on a plant perhaps from a sprayed treatment. The infected insect dies and "melts" or falls apart
on foliage, releasing more virus. This additional infective material can infect more
insects, continuing the cycle.
Figure 2.6:
Vlak, 1997)
In vivo baculovirus infection and replication cycle. (Adapted from
15
2.4.2.1.2 In Vitro Replication
Baculovirus infection starts when a susceptible insect larva ingests
baculovirus occlusion bodies (Figure 2.7). The midgut lumen of lepidopteran larvae
constitutes a highly alkaline environment in which OBs dissolve and the occlusion
derived virions are released into the gut lumen. These virions pass through the
peritrophic membrane and fuse with the microvillar membrane of the midgut
epithelial cells whereafter they are transported to the nucleus, initiating the first
replication cycle. Baculoviruses have a biphasic replication cycle, in which two
genetically identical, but phenotypically distinct virus types are formed.
The newly formed budded viruses (BVs) are initially released by budding
through the plasma membrane of the infected cell. The insect tracheal system and
the hemolymph play a major role in the transport of the BVs to other organs and
tissues (Volkman, 1997; Barrett et al., 1998). Budded virions differ in several
aspects from ODVs (Figure 2.8) which are formed later in infection. (In cell culture
BVs are 1000-fold more infectious than ODVs.) Budded virions are responsible for
the systemic infection; ODVs facilitate viral spread from one individual insect to
others. Budded virions enter the cell by endocytosis, followed by the fusion of the
viral envelope and the endosome membrane. The fusion process is mediated by a
virus encoded essential glycoprotein, gp64, which is exclusively found in BVs
(Blissard, 1996). The ODVs are not released by budding, but acquire an envelope
inside the nucleus, followed by occlusion in polyhedra. Finally, the infected cell
ruptures and the lyses of both the nuclear and cellular membranes allow the release
of the newly formed, mature polyhedra.
The polyhedra are surrounded by an
envelope composed of carbohydrates and specific proteins (Zuidema et al., 1989).
Baculovirus diseases are primarily diseases of the larval stages, and the progression
and signs of disease depend on several factors including the instars initially infected,
infection dose, nutrition, temperature, degree of compatibility of the virus with its
host, and the physical characteristics of the larva.
16
In typical nucleocapsid nuclear polyhedrovirus (NPV) infections, there are
very limited signs of disease during the first 3 days post infection. At about the
fourth day of the infection, larvae show reduced motor functions. They also respond
more slowly to tactile stimuli than healthy larvae. Their feeding begins to slow and
virtually ceases by day 6 or 7. At day 4 or 5 the larva will begin to appear swollen,
the cuticle will take on a pale creamy coloration. This is due to the presence of
polyhedra accumulating in epidermal and fat body cell nuclei. The hemolymph of
infected larvae at this stage is cloudy owing to the circulation of large numbers of
infected hemocytes and polyhedra released into the hemolymph as a result of lysis of
cells in various tissues during advanced stages of disease. Following this, larvae will
die within one or two days. Larvae of many lepidopteran species will crawl up to the
top of the plant on which they were feeding, and then die. After death, the larvae
become black, lose their turgor and become flaccid. The cuticle ruptures, releasing
billions of polyhedra.
Figure 2.9 showed a typical infected Sf-9 cells photo
containing the presence of polyhedra.
17
Figure 2.7:
In vitro baculovirus infection and replication. (A) Ingestion of
polyhedra and solubilization by digestive juices in the insect gut. (B) Fusion of
the viral envelope of the released virus with the plasma membrane of a midgut
cell. (C) Entry of the nucleocapsid into the nucleus. (D) Formation of virogenic
stroma where virus replication and assembly of progeny nucleocapsids occurs.
(H) Departure of nucleocapsids from the nucleus and formation of non-occluded
virus particles (NOV) by acquisition of an envelope from the nuclear (I) or
cellular (J) membrane by budding. (K) Systemic infection of cells from other
tissues by adsorption endocytosis. (E) Envelopment of single or multiple
nucleocapsids in membrane de novo synthesized in the nucleus. (F) Occlusion of
singly and multiply enveloped virus particles into polyhedra. (L) Formation of
cytoplasmic and nuclear inclusions (fibrillar structures) with unknown function.
(G) Release of polyhedra from deceased insect larvae. (Adapated from Vlak,
1997)
18
Figure 2.8:
Structural compositions of the two baculovirus phenotypes,
budded virus (BV), and the occlusion derived virus (ODV). (Adapted from
Blissard, 1996). Proteins common to both virus types are indicated in the
middle of the Figure 2.6. Proteins specific to either BV or ODV are indicated on
the left and right respectively. The polar nature of the baculovirus capsid is
indicated in the diagram with the claw-like structure at the bottom and the ringlike structure at the top of the capsid. The possible location of p74 is indicated
by a dashed line. Lipid composition of the BV and ODV envelopes derived from
AcMNPV infected Sf-9 cells (Braunagel and Summers, 1994) are indicated.
(LPC, lysophosphatidylcholine; SPH, sphingomyelin; PC, phospahetidylcholine;
PI, phosphatidylinositol; PS, phosphatidylserine; PE, phosphatidylethonalamine)
19
(a)
(a)
Figure 2.9:
(c)
(b)
(a) A typical infected Sf-9 cells showing the presence of polyhedra is indicated by the arrow (Steven Howard); (b) Electron
micrograph of AcMNPV infected Sf-9 Cell (Greg V. Williams); (c) A portion of the nucleus containing enveloped virions in the process
of being occluded into a developing polyhedron is shown. (From the Carstens' Lab at Queen's University, Canada)
19
20
In Figure 2.9 (b), the polyhedra (P) containing occluded virus are visible in
the nucleus of an infected Sf-9 cell at 36 hour post infection. One occlusion body (*)
is sectioned in a plane where no occluded virus is evident. The occlusion body calyx
(C) is visible. Calyx precursors (CP) are present both in association with p10 fibrous
bodies (F) and free in the nucleoplasm. A calyx precursor is seen attaching to an
occlusion body. Fibrous bodies are visible in both the nucleoplasm (F) and the
cytoplasm (Fc).
The nuclear membrane (N) is indicated.
Short open-ended
membrane profiles (M) are present near the nuclear periphery as are the remnants of
the virogenic stroma (vs) assembled nucleocapsids are seen in association with the
membrane profiles in the process of being enveloped to form PDV(↑)
2.5
Advantages of BEVS Technology
Since 1983, when BEVS technology was introduced, the baculovirus system
has become one of the most versatile and powerful eukaryotic vector systems for
recombinant protein expression (Smith et al., 1983). More than 600 recombinant
genes have been expressed in baculoviruses to date. Since 1985, when the first
protein (IL-2) was produced in large scale from a recombinant baculovirus, use of
BEVS has increased dramatically (Smith et al., 1983).
Baculoviruses offer the
following advantages over other expression vector systems.
(a) Safety:
Baculoviruses are essentially nonpathogenic to mammals and plants
(Ignoffo, 1975). They have a restricted host range, which often is limited to
specific invertebrate species. Because the insect cell lines are not transformed
by pathogenic or infectious viruses, they can be cared for under minimal
containment conditions. Helper cell lines or helper viruses are not required
because the baculovirus genome contains all the genetic information.
21
(b) Ease of Scale Up:
Baculoviruses have been reproducibly scaled up for the
production of biologically active recombinant products.
(c) High Levels of Recombinant Gene Expression: In many cases, the
recombinant proteins are soluble and easily recovered from infected cells late in
infection when host protein synthesis is diminished.
(d) Accuracy:
Baculoviruses can be propagated in insect hosts which post-
translationally modify peptides in a manner similar to that of mammalian cells.
(e) Use of Cell Lines Ideal for Suspension Culture: AcMNPV is usually
propagated in cell lines derived from the fall armyworm Spodoptera frugiperda
or from the cabbage looper Trichoplusia ni. Cell lines are available that grow
well in suspension cultures, allowing the production of recombinant proteins in
large-scale bioreactors.
(f) Very High Expression of Recombinant Proteins: In many cases, the
recombinant proteins produced are antigenically, immunogenically and
functionally similar to their native counterparts.
22
2.6
Glycosylation in Insect Cells
Studies on the N-glycan structures produced by mosquito cells provided
earliest views of the insect protein N-glycosylation pathway (reviewed by Marchal et
al., 2001; Marz et al., 1995). The results showed that there were indeed some
striking differences, as the structures of the N-glycans on the glycoproteins produced
by insect cells have the significant differences from those produced by mammalian
cells (Figure 2.10). They are (i) inability to synthesize sialylated complex-type Nglycans in contrast to mammalian cells (Marz et al., 1995; Altmann et al., 1999;
Marchal et al., 2001) and (ii) the presence of potentially allergenic structure,
Fucα(1,3)GlcNAc-Asn.
It is clear that the inability of most lepidopteran insect cells to produce
mammalian-type N-glycan are attributable to extremely low levels of Nacetylglucosaminyltransferase II (GlcNAcT II), β1,4-galactosyltransferase (β1,4GalT) activities and no detectable α2,6-sialytransferase (α2,6-ST) activities (Stollar
et al., 1976; Butters et al., 1981; Altmann et al., 1993; van Die et al., 1996; Hooker
et al., 1999).
Furthermore, some insect cells have an N-acetylglucosaminidase,
which removes the terminal GlcNAc residue from GlcNAcMan3GlcNAc2-Asn and
eliminates the intermediate required for complex N-glycan production (Licari et al.,
1993; Altmann et al., 1995; Wagner et al., 1996; Marchal et al., 1999). Finally, it
has been reported that there is no detectable CMP-sialic acid, which is the donor
substrate required for sialoglycoprotein synthesis, in one insect cell line (Hooker et
al., 1999). Consequently, the major processed N-glycan typically produced by insect
cells is the paucimannose structure, as shown in Figure 2.10.
23
Asn
α1,2-glucosidase I
α1,3-glucosidase II
Asn
α-mannosidase I (RER)
α-mannosidase I (Golgi)
Insect and mammalian
N-glycan processing
pathways share a
common intermediate
Asn
N-Acetylglucosaminyltransferase I (GlcNAcT-I)
--UDP-GlcNAc
Asn
α-mannosidase II
Fucosyltransferase (FucT)
--GDP-Fuc
Asn
INSECT
MAMMALIAN
β-N-Acetylglucosaminidase
N-Acetylglucosaminyltransferase II
Asn
“PAUCIMANNOSE”
Asn
N-Acetylgalactosyltransferase
Sialytransferase
Galactosyltransferase
Sialytransferase
Asn
Asn
Key to Symbols
N-Acetylglucosamine (GlcNAc)
Mannose (Man)
Fucose (Fuc)
Galactose (Gal)
N-Acetylgalactosamine (GalNAc)
“COMPLEX”
Sialic acid (Neu5Ac)
Glucose (Glc)
Figure 2.10: Protein N-glycosylation pathways in insect and mammalian cells.
Monosaccharides are indicated by their standard symbolic representations, as
defined in the key. The insect and mammalian N-glycan processing pathways
share a common intermediate, as shown. The major products derived from this
intermediate are paucimannose and complex N-glycans in insect and
mammalian cells, respectively. (Adapted from Jarvis, 2003)
24
2.7
Glycosyltransferases and Glycosidases Involved in N-glycan Processing
in Insect Cells
The processing pathway of N-glycans in lepidopteran insect and mammalian
cells is shown as Figure 2.10 (Jarvis, 2003). A number of studies have suggested
that initial processing of N-glycans in insect cells is similar or identical to that of
mammalian cells. However, insect cells appear to lack some of the processing
pathways of mammalian cells but contain additional glycosylation activities absent in
mammalian cells.
2.7.1 α-Glucosidase I, II and α-Mannosidase I
Glc3Man9GlcNAc2 is processed by α-glucosidase I, II and α-mannosidase I
to generate Man5GlcNAc2 structure. Many glycoproteins produced by lepidopteran
insect cells have high mannose type glycans. For example, N-glycans on human IgG
and hTf produced by Tn-5B1-4 cells included various high mannose type and
paucimannosidic glycans, with some incomplete complex-type glycans (Hsu et al.,
1997; Ailor et al., 2000).
Expression of α-glucosidase I and II in several
lepidopteran insect cells appears adequate (David et al., 1993). In addition, αmannosidase I has been purified from Sf-21 cells (Ren et al., 1995) and cloned from
Sf-9 cells (Kawar et al., 1997), and its substrate specificity has been characterized
(Kawar et al., 2000). These results suggest that lepidopteran insect cells contain
ample α-glucosidase I, II and α-mannosidase I.
25
2.7.2
N-Acetylglucosaminyltransferase I (GlcNAcT-I) and α-mannosidase II
First of all, GlcNAc is added to Manα(1,3) branch of Man5GlcNAc2 by NAcetylglucosaminyltransferase I (GlcNAcT-I). Thereafter, two Man residues are
removed by α-mannosidase II.
Substantial levels of GlcNAcT-I activities were
observed in several insect cell lines including Sf-9, Sf-21, Mb0503, and Bm-N
(Velardo et al., 1993). Like its counterpart from mammalian cells, α-mannosidase II
from insect cells requires GlcNAc on the Manα(1-3) branch for its activity (Altmann
et al., 1995). These studies suggest that lepidopteran insect cells have high levels of
α-mannosidase II and GlcNAcT-I in order to generate the precursor glycan required
for the formation of complex-type N-glycans.
2.7.3
N-Acetylglucosaminyltransferase II (GlcNAcT-II)
In mammalian cells, the product N-glycan of α-mannosidase II reaction
serves
as
an
acceptor
for
the
next
reaction
catalyzed
by
N-
Acetylglucosaminyltransferase II (GlcNAcT-II), which adds another GlcNAc to the
Manα(1,6) branch.
However, lepidopteran insect cells, including Sf-9, Sf-21,
Mb0503, and Bm-N cells, have been shown to have only 1% or less of the
endogeneous GlcNAcT-II activity present in mammalian cells.
26
2.7.4
β-1,4-Galactosyltransferase (β1,4-GalT)
A terminal GlcNAc on either Man-branch is usually galactosylated by β1,4galactosyltransferase (β1,4-GalT) in mammalian cells. In contrast, galactosylated Nglycans are rarely found in glycoproteins from lepidopteran cells. In fact, negligible
levels of β1,4-GalT activity were detected in Sf-9, Tn-5B1-4 and Mb0503 cells
(Hollister et al.,1998, van Die et al., 1996, Hollister et al., 2001).
β1,4-GalT
activities in Sf-9 and Tn-5B1-4 cells were reexamined using an Eu-fluorescence
assay method (Abdul Rahman et al., 2002). Sf-9 did not contain any detectable
levels of β1,4-GalT activity (Abdul Rahman et al., 2002).
2.7.5
Core α-1,3- and α-1,6-Fucosyltransferases (FucT)
N-glycans with one or two GlcNAc on Man3-core can be further modified by
core fucosyltransferases. Both core Fuc-T’s require the presence of GlcNAc β(1,2)
on the Man α(1,3) branch for its action (Staudacher et al., 1998).
N-glycans
containing either one or both of Fucα(1,3) and Fucα(1,6) attached to the Asn-linked
GlcNAc were identified on the membrane glycoproteins from Mb0503, Sf-21, and
Bm-N cells, in which glycoproteins from Mb0503 cells containing highest levels of
α-1,3-fucosylated N-glycans (Kubieka et al., 1994). Fuc-T C6, but not Fuc-T C3
were easily detected in Sf-9 cells.
27
2.7.6
β-N-Acetylglucosaminidase
A β-N-Acetylglucosaminidase specific for the terminal GlcNAc on the
Manα(1,3) branch was found in Sf-21, Bm-N and Mb0503 cells (Altman et al.,
1995), and it was suggested that this enzyme was localized in the microsome-like
membrane fraction in Sf-21 cells (Altman et al., 1995). Similar enzymatic activity
was also detected in the cell lysates and cell culture supernatant of insect cell derived
from Spodoptera frugiperda, Trichoplusia ni, Bombyx mori, or Malacosoma disstria
(Licari et al., 1993). Structural analysis of N-glycans from human IgG (Hsu et al.,
1997) and hTf (Ailor et al., 2000) expressed in Tn-5B1-4 cells suggested the
presence of such a β-N-Acetylglucosaminidase in Tn-5B1-4 cells.
The further
removal of additional Man residues by α-mannosidase(s) can lead to the generation
of structures with fewer than three Man residues, as has been observed in several
studies.
2.7.7
Sialyltransferase (SiaT)
Sialytransferase (SiaT) adds N-acetylneuraminic acid to the terminal Gal
residues on N-glycans in mammalian cells. However, SiaT activity has yet to be
detected in Sf-9 (Hollister, 2001; Lopez et al., 1999; Hooker et al., 1999), Sf-21
(Hooker et al., 1999), Tn-5B1-4 (Lopez et al., 1999), Mb0503 (Lopez et al., 1999),
and Ea4 (Hooker et al., 1999) cells, even using highly sensitive assays with
radiolabeled CMP-NeuNAc or fluorescent CMP-NeuNAc derivatives as the donor
substrate.
28
2.8
Sugar Nucleotides Involved in N-glycan Processing in Insect Cells
2.8.1
Endogenous Sugar Nucleotide Levels in Lepidopteran Insect Cells
All glycosyltransferases in the synthetic pathway for complex-type N-glycans
require respective sugar nucleotides as substrate donor. Examination of the sugar
nucleotide concentrations in lepidopteran insect cells demonstrated the presence of
substantial levels of UDP-hexose, UDP-N-acetylhexosamine, GDP-Fuc, and GDPMan in Sf-9, Mb0503, and Tn-5B1-4 cells (Lopez et al., 1999). However, no CMPNeuNAc was detected in the same study (Lopez et al., 1999). Similar results were
obtained on the sugar nucleotide levels in Sf-9 and Tn-5B1-4 cells (Tomiya et al.,
2001).
2.8.2
Enzymes Involved in Sialic Acid and CMP-Sialic Acid Synthesis
Of particular significance is the absence in lepidopteran insect cells of the
CMP-NeuNAc necessary for sialylation of N-glycans. In mammalian cells, sialic
acids are synthesized from UDP-GlcNAc through multiple enzymatic reactions as
shown as Figure 2.11.
The bifunctional enzyme, UDP-N-acetylglucosamine (UDP-GlcNAc) 2
epimerase / N-acetylmannosamine (ManNAc) kinase, is believed to be a key enzyme
in the biosynthesis of NeuNAc in rat liver (Hinderlich et al., 1997). This enzyme
converts UDP-GlcNAc-6P to ManNAc-6-P, which is further converted to Nacetylneuraminic acids (NeuNAc) by N-acetylneuraminate-9-phosphate synthase
29
(SAS) and N-acetylneuraminate-9 phosphate phosphatase.
NeuNAc is then
converted to CMP-NeuNAc by CMP-NeuNAc synthase (CMP-SAS).
Effertz et al. (1999) reported that the UDP-GlcNAc 2-epimerase activity in
Sf-9 cells was about 30 times less (in term of specificity activity) than that in rat liver
cytosol fraction.
Interestingly, Sf-9 cells had 50 times higher ManNAc kinase
activity compared with the 2-epimerase activity (Effertz et al., 1999).
It was
reported that Sf-9 cells contained negligible levels of neuraminic acids, and no
detectable N-acetylneuraminic-9-phosphate synthase activity was present in the
lysate of Sf-9 cells (Lawrence et al., 2000). It was found that Sf-9 cells do not have
detectable CMP-sialic acid synthase activity (Lawrence, 2001).
UDP-GlcNAc
ManNAc
Bifunctional UDP-GlcNAc 2-epimerase/ManNAc kinase
ManNAc-6-P
N-acetylmannosamine kinase
N-acetylneuraminic acid 9-phosphate synthase (SAS)
NeuNAc-9-P
N-acetylneuraminic acid 9-phosphate phosphatase
NeuNAc
CMP-neuraminic acid synthase (CMP-SAS)
CMP-NeuNAc
Abbreviations:
UDP-GlcNAc
ManNAc
ManNAc-6-P
NeuNAc-9-P
NeuNAc
-
CMP-NeuNAc
-
Uridine-5’-diphopho-N-acetylglucosamine
N-acetylmannosamine
N-acetylmannosamine-6-phosphate
N-acetylneuraminic acid-9-phosphate
5-acetamido-3,5-dideoxy-D-glycero-D-galacto-2-nonulosonic acid
(N-acetylneuraminic acid)
Cytidine-5’-monopho-N-acetylneuraminic acid
Figure 2.11: CMP-Neuraminic acid synthesis pathway. The dotted arrow
indicates pathways which are insufficient in lepidopteran insect cells. (Tomiya et
al., 2003)
30
2.9
Engineering of N-glycan Processing Pathway
The general strategy for humanizing glycoproteins produced by the insect
cell-baculovirus expression system is shown in Figure 2.12. The goal of engineering
N-glycan processing is to develop a new insect cell-baculovirus expression vector
system(s) that can express human-like sialylated multi-antennary complex-type Nglycans. As described in the earlier sections, several lines of evidence suggest that
majority of lepidopteran insect cells currently used for protein expression apparently
lack several enzymes for such a goal. Moreover, lepidopteran insect cells contain the
undesirable β-N-acetylglucosaminidase and Fu-T C3. The former diminishes the key
glycans containing GlcNAc β (1,2)Manα(1,3) which stunt the normal growth of
complex-type N-glycans, and the latter generates potentially allergenic N-glycans.
Therefore, the N-glycan processing pathways need to be altered in the insect cells by
enhancing or suppressing respective processing pathways.
Many lines of evidence have indicated that the inability of the vast majority
of lepidopteran cells to synthesize mammalian type N-glycans. The inability to
obtain such N-glycans in lepidopteran cells can be attributed to the insufficient levels
of β1,4-GalT, GlcNAcT-II, SiaT, UDP-GlcNAc 2 epimerase / ManNAc kinase ,
UDP-N-acetylneuraminate-9-phosphate synthase, CMP-NeuNAc synthase activities.
β-N-Acetylglucosaminidase, which removes GlcNAc on the Manα(1,3) branch, have
been detected in several lines of lepidopteran cells. This enzyme apparently prevents
synthesis of complex-type N-glycans by removing the key intermediate glycan
containing GlcNAcβ(1,2)-Man (1,3). In addition, FucT C3 generates the potentially
allergenic glycan structure, Fucα(1,3)GlcNAc-Asn, on glycoproteins expressed in
lepidopteran cells.
31
Lacking some enzymes:
Problem
1.
2.
3.
4.
5.
β1,4-GalT
GlcNAcT-II
SiaT
N-acetylneuraminate-9-phosphate
synthase
CMP-NeuNAc synthase
Genetic Engineering:
Solution
1.
2.
Metabolic
Engineering:
Recombinant
baculovirus
Transgenic
Insect cells
1.
2.
Sugar Feeding
Chemical
Inhibitor
Glycoproteins Expression:
Evaluation
1.
2.
3.
4.
Glycosyltransferases activity
Glycosidases activity
Sugar nucleotides
Glycan analysis
Figure 2.12: General strategy for humanization of glycoprotein produced by
lepidopteran cell-baculovirus expression system. (Tomiya et al., 2003)
32
2.9.1
Improvement of N-Acetylglucosaminylation of the Manα(1,3)-Branch
β-N-acetylglucosaminidase was implicated as a problem in N-glycan
elongation by its absence of Estigme acrea cells, which produced N-glycans
containing terminal N-acetylglucosamine residues (Wagner et al., 1996). Sf-9 cells
are known to contain high levels of β-N-acetylglucosaminidase (Wagner et al.,
1996).
Using Sf-9 cells, Wagner et al. succeeded in N-glycan elongation by
coexpression of human β-N-acetylglucosaminyltransferase I and fowl plague virus
hemagglutinin.
Watanabe et al. (2001) examined the effect of a β-N-
acetylglucosaminidase inhibitor, that is 2-acetamide-1,2-dideoxynojirimycin (2ADN) on bovine interferon-γ (bIFN- γ) on production in Tn-5B1-4 cells. Watanabe
et al. (2001) speculated that the inhibitor enhanced accumulation of substrates
processing a β(1,2)-linked GlcNAc, thereby leading to further elongation by β1,4GalT and SiaT to form sialylated N-glycans. However, the overall increase of Nglycan containing β(1,2)-linked GlcNAc was not determined.
2.9.2
Improvement of Galactosylation
Expression of a β1,4-GalT by a baculovirus vector increased galactosylation
of glycoprotein (Jarvis et al., 1996), indicating that the mammalian enzyme
expressed by baculovirus infection could function in the infected lepidopteran cells
and that it could compete with the β-N-acetylglucosaminidase activity insect cell
(Jarvis et al., 1996). Similar results were obtained when human serum transferrin
(hTf) was expressed by Tn-5B1-4 cells infected with two baculoviruses, one
encoding a gene for hTf and the other encoding a gene for a mammalian GalT (Ailor
et al., 2000). In this study, 13% of the total N-glycans were galactosylated, and
protection of GlcNAc on Manα(1,3) branch against β-N-acetylglucosaminidase by
galactosylation was confirmed (Tomiya et al., 2003).
33
2.9.3
Production of Biantennary Complex-Type N-glycans
Production of biantennary complex-type N-glycans was achieved recently by
expressing a mammalian N-acetylglucosaminyltransferase II (GlcNAcT-II) in
lepidopteran cells using a transgenic insect cell, SfSWT-1 (Hollister et al., 2002), or
using baculovirus expression vector system (Tomiya et al., 2003).
2.9.4
Formation of Sialylated N-glycans
Sialylation of N-glycans was in Tn-5B1-4 cells when the cells were cultured
in the presence of a hexosaminidase inhibitor (2-ADN) (Watanabe et al., 2001). This
result is particular intriguing since Tn-5B1-4 cells lack β1,4-GalT, SiaT and CMPNeuAc synthase.
Unfortunately, analysis was only by lectin blot and not by
quantitative chemical analysis of the exact structures. Sialylation was also detected
in virion glycoprotein, gp64, when Sf-9 cells were infected with a recombinant
baculovirus vector encoding mammalian β1,4-GalT and α2,6-sialyltransferase (α2,6SiaT), while no sialylation was detected in the absence of either β1,4-GalT or α2,6SiaT (Jarvis et al., 2001).
2.9.5
Synthesis of CMP-NeuNAc
The processing steps catalyzed by UDP-N-acetylglucosamine 2 epimerase /
N-acetylmannosamine kinase, N-acetylneuraminate-9-phosphate synthase, and CMPNeuNAc synthase represent bottlenecks in the CMP-NeuNAc synthesis pathway of
lepidopteran cells (Tomiya et al., 2003). To overcome this problem, Tomiya et al.
34
(2003) reported that they cloned mammalian N-acetylneuraminic-9-phosphate
synthase and CMP-NeuNAc synthase (Lawrence et al., 2001), and expressed these
enzymes in Sf-9 cells. When Sf-9 cells were infected with a recombinant baculovirus
expression vector encoding N-acetylneuraminate-9-phosphate synthase and were
cultured in a medium supplemented with N-acetylmannosamine (ManNAc), Sf-9
cells produced high levels of N-acetylneuraminic acid (NeuNAc) (Lawrence et al.,
2000).
2.10
Galactosylation in N-Glycan Processing in Insect Cells
Galactosylation is a process which links the galactose sugar to the end of the
GlcNAc (β1,2)Man (α1,3) chains. In the in vivo galactosylation, mammalian β1,4GalT is being introduced artificially to the cell culture which secretes the protein.
This is also known as coinfection, which is the simultaneous infection of a single
host cell by two types of different virus particles. Another new technology being
established is the in vitro galactosylation, which uses mammalian β1,4-GalT to add
the missing galactose sugar units to the carbohydrate chains of the protein via UDPGal , a donor sugar known as sugar nucleotides. The process is performed after the
protein is expressed and secreted by the host cell.
There are three main factors that are involved in galactosylation, which are
sugar acceptor, sugar donor and enzyme. In this study, human serum transferrin was
used as the sugar acceptor. Human serum transferrin was used as the model protein
due to its simplicity of biantennary N-glycan structure. Uridine-diphosphogalactose
(UDP-Gal) was used as the substrate donor and mammalian β1,4-GalT was used as
the enzyme.
35
2.10.1 Sugar acceptor
Human serum transferrin (hTf) is a serum glycoprotein found in the
physiological fluids of vertebrates (Aisen, 1989; Thorstensen and Romslo, 1990) and
insect larva (Bartfeld and Law, 1990) that is responsible for carrying Fe+3 to all cells
in the body. When bound to iron, the circulating transferrin is recognized by a
specific surface receptor on cells and internalized to release iron into the cytoplasm
(Trowbridge et al., 1984). Serum transferrin also plays a role in host defense by
depriving any circulating microorganism of essential iron (Bullen et al., 1990). HTf
is a single-chain glycoprotein of 679 amino acids containing two potential N-linked
glycosylation sites in its carboxy-terminal domain at Asn413 and Asn611
(MacGillivray et al., 1983), with a glycosylation-dependent molecular mass in the
range 76-81 kDa (MacGillivray et al., 1982). Previous studies have shown that the
transferrin glycoforms present in human serum are comprised of species having
terminally sialylated bi-, tri-, and tetraantennary oligosaccharides (Leger et al., 1989;
Fu and van Halbeek, 1992). The most dominant glycoform includes bianntenary
oligosaccharides located at both asparagines positions, although, changes in
physiological conditions can affect the N-glycan pattern observed in the host
(Montreuil et al., 1997).
2.10.2 Substrate Donor
UDP-Gal is a substrate also known as sugar nucleotide (Figure 2.13), used by
galactosyltransferase for extension of sugar chain of glycoproteins. Once the sugar
nucleotides are synthesized in the cytosol, they are topologically mislocalized, since
most glycosylation occurs in the ER and Golgi. Their negative charge prevents them
from simply diffusing across membranes into these compartments. To overcome this
problem, cells have devised a set of nonenergy-requiring sugar nucleotide
transporters, actually antiporter, that deliver sugar nucleotides into the lumen if these
36
organelles with simultaneous exit of nucleotide monophosphates which must first
derived from the nucleotide diphosphates (Figure 2.14).
Nucleotide sugar transporters are membrane proteins localized in the
endoplasmic reticulum and Golgi apparatus. They play an indispensable role in
constructing the sugar chains of glycoconjugates. The transporters carry sugars into
the endoplasmic reticulum and Golgi apparatus, in which they are used by specific
transferases as precursors of sugar chains (Kawakita et al., 1998; Hirachberg et al.,
1998; Berninsone et al., 2000; Gerardy-Schahn et al., 2001; Hirschberg, 2001).
More than simply functioning as a passive entrance route of nucleotide sugars into
the organelles, the transporters may regulate the amounts of nucleotide sugars
available in the lumen of the endoplasmic reticulum or Golgi apparatus and
consequently may affect the sugar chain composition of a cell (Kumamoto et al.,
2001).
For most glycosylation reaction, the sugar nucleotide donates the sugar,
resulting in the formation of nucleoside diphosphate, which must be converted into a
monophosphate by the nucleoside diphosphatase that occurs in the Golgi lumen.
Exchange through the antiporters is electroneutral, since the sugar nucleotide with
two negative charges (one on each phosphodiester) enters and the nucleoside with a
single phosphomonoester exits.
37
Uridine O
OH
HO CH2
O
O
HO
OH
N
O
O P O P O CH2
OO-
O
N
O
HO
OH
Ribose
Galactose
UDP
Figure 2.13: Structure of a nucleotide sugar that can serve as a sugar donor in
a glycosyltransferase reaction. UDP, uridine diphosphate.
Figure 2.14:
Transporters for sugar nucleotides, PAPS, and ATP are located
in the Golgi membranes of mammals, yeast, protozoa, and plants. These
proteins are actually antiporters, and the corresponding nucleoside
monophosphate is carried into the cytosol with sugar nucleotide transport. Since
most glycosylation reactions produce a nucleoside diphosphate, this requires
conversion to the nucleoside monophosphate. (Adapted from Hirschberg et al.,
1998)
38
2.10.3 Enzyme
The golgi or endoplasmic reticulum glycosyltransferases constitute a
functional family of approximately 300 membrane-bound enzymes that, in general,
synthesizes complex carbohydrates of glycoconjugates of cells by transferring a
sugar moiety of a sugar nucleotide to an acceptor sugar (Roseman, 2001; Hill, 1979).
The galactosyltransferase family is the subset of the glycosyltranferases, in the
presence of the metal ion, transfers galactose from UDP-Gal to an acceptor sugar
molecule.
To date, three subfamilies, β1,4-, β1,3-, and α1,3-, have been well
characterized (Amado et al., 1998) and they generate β1,4-, β1,3-, and α1,3- linkages
between galactose and the acceptor sugar, respectively. Cloning has identified the
presence in each family of several members that have sequence homology within the
family members (Amado et al., 1998).
The β1,4-galactosyltransferase (Gal-T)
family, which was the first one to be cloned (Narimatsu et al., 1986; D’Agostaro et
al., 1989; Shaper et al., 1986), consists of at least seven members, β1,4Gal-T1 to
β1,4Gal-T7 (Amado et al., 1998), with a 25 to 55% sequence homology. These
enzymes are expressed in different tissues and show differences in the
oligosaccharide acceptor specificity (Lo et al., 1998; Guo et al., 2001).
In the mammary gland, only β1,4Gal-T1 is expressed (Shaper et al., 1998)
and it interacts with the calcium binding protein, α-lactalbumin, that is expressed in
the mammary gland during lactation, to form the lactose synthase complex. The
formation of this complex alters the substrate specificity of β1,4Gal-T1 such that
glucose at physiological concentrations can serve as the acceptor sugar, resulting in
the synthesis of lactose (Galβ1,4Glc).
The protein domain structure of β1,4Gal-T1 consists of a short NH2-terminal
cytoplasmic domain, a single transmembrane domain, a stem region, and a large
lumenal, catalytic domain which contains the metal (Mn2+), UDP-Gal, and the
acceptor sugar binding sites (Paulson et al., 1989; Aoki et al., 1990; Yadav et al.,
1990).
39
Lactose synthetase (UDP-galactose: D-glucose 1-galactosyltransferase, EC
2.4.1.22) catalyses the biosynthesis of lactose as Reaction (1):
Galactosyltransferase
α-lactalbumin
UDP-galactose + glucose
lactose + UDP
(1)
The enzyme was first described by Watkins and Hassid (1962) as a
particulate enzyme in rat and guinea pig mammary glands.
Brodbeck and Ebner (1966) resolved the soluble bovine lactose synthetase
from milk into two protein fractions, designated as A and B, which were required for
catalytic activity. Brew et al. (1968) reported that the A protein is a
galactosyltransferase. Brodbeck et al. (1967) reported that B protein has been found
to be identical with the familiar milk protein, α-lactalbumin.
Protein A is a galactosyltransferase which catalyzes Reaction (2), known as
N-acetyllactosamine (LacNAc) reaction:
UDP-galactose + N-acetylglucosamine
Galactosyltransferase
α-lactalbumin
N-acetyllactosamine + UDP
(2)
Under normal assay conditions, this reaction is inhibited by α-lactalbumin,
the Protein B. α-lactalbumin modifies the substrate (acceptor) specificity of
galactosyltransferase from N-acetylglucosamine to glucose and allows synthesis of
lactose in the presence of glucose by Reaction (1). Thus, α-lactalbumin is termed a
“Specifier Protein”. It binds to the enzyme moiety, or to an enzyme-substrate
complex, and thereby changes the specificity of the system. The effect of αlactalbumin on the synthesis of N-acetyllactosamine (NAL) by the protein A appears
40
to be complex. At low N-acetylglucosamine (NAG) concentration, α-lactalbumin
stimulates NAL synthesis, and appears to be acting in this respect as a regulatory
protein. At higher NAG concentration, α-lactalbumin shows an inhibitor effect which
increases with increasing NAG concentration.
41
CHAPTER 3
MATERIALS AND METHODS
3.1
Materials
Spodoptera frugiperda (Sf-9) insect cells was purchased from ATCC cat. No.
1711 (Rockville, MD). The recombinant baculovirus containing the gene coding for
Human Transferrin and β1,4-galactosyltransferase were provided by Prof Dr Michael
J. Betenbaugh of Johns Hopkins University, USA. Wild type recombinant virus
AcMNPV was a gift from Prof Dr Mohd Sanusi Jangi, UKM, Malaysia.
3.2
Equipments
The High Performance Liquid Chromatography (HPLC) System used was
LC-10Dvp HPLC system and a NovaPac C18 column (3.9 x 150mm, 4 µm) with a
NovaPac C18 guard cartridge. The electrophoresis system used was Mini-Protean II
from Bio-Rad (California, USA). Western blot analysis was done using Trans-Blot
Electrophoretic Transfer Cell (Bio-Rad Laboratories, Melville, NY). Shimadzu UV-
42
160 spectrophotometer (Minnesota, USA) was used to measure absorbance at
450nm. The slow rotary shaker was purchased from Bellco Biotechnology (New
Jersey, USA). Laminar flow hood was from Telstar Bio-II-A (Germany). Inverted
phase contrast microscope and Compound microscope (optional) were from Zeiss
Instruments (Germany). Incubator was purchased from Memmert (Germany).
3.3
Chemicals
Sf-900 II Serum Free Media (SFM) and Fetal Bovine Serum (FBS) were
from GIBCO BRL (Gaithersburg, MD).
Goat anti-Human Transferrin-affinity
purified, Goat anti-human transferrin-HRP conjugate, Calibrator-Human Reference
Serum and TMB (3,3’,5,5’-tetramethylbenzidene) Peroxidase Substrate and
Peroxidase Solution B (water soluble) were obtained from Bethyl Laboratories Inc
(Texas). TMB Stabilized Substrate for Horseradish Peroxidase (water insoluble) was
purchased from Promega, (Madison, WI).
Asialofetuin, β-galactosidase (from
bovine), peroxidase-labeled RCA 1, uridine-5’-diphosphogalactose disodium salt
(UDP-Gal), uridine 5’-Triphosphate sodium (UTP), acrylamide, bis-acrylamide,
bovine serum albumin (BSA), ammonium persulfate, citric acid, 2-mercaptoethanol,
silver nitrate, triton X-100, dimethyl sulphoxide (DMSO), α-lactalbumin, N,N,N',N'tetramethylethylenediamine (TEMED), anisaldehyde, 4-Morpholinepropanesulfonic
acid (MOPS), tris, glycine, lactose, glucose, manganese chloride and ammonium
phosphate were purchased from Sigma (Missouri, USA). Trypan blue, ethanol,
acetic acid, ethylenediamine tetraacetic acid disodium salt dehydrate (EDTA), 38%
formaldehyde, sodium chloride, sodium hydroxide, hydrochloric acid, sodium
dodecyl sulfate (SDS), bromophenol blue, sodium bicarbonate, tween 20, glycerol,
phosphoric acid, methanol, skimmed milk, potassium chloride, potassium phosphate
dibasic, potassium dihydrogen phosphate, zink sulfate 7-hydrate, barium hydroxide,
tetrabutylammonium hydrogen sulfate (TBAS) and dichloromethane were from
Fluka (Missouri, USA). Ammonium hydroxide, butanol, diethyl ether and
glutaldehyde were purchased from Merck (New Jersey, USA).
43
3.4
Spodoptera frugiperda (Sf-9) Insect Cells
3.4.1
Cells Thawing
A vial of Spodoptera frugiperda (Sf-9) insect cells was taken out from the
liquid nitrogen (-196 0C). Then the vial was placed in a 37 0C water bath and gently
swirled until the cell was completely thawed. A bottle of Sf-900 II SFM medium
was removed from the cold room and placed in a 37 0C water and was allowed to
acclimate to room temperature before using. Laminar flow hood was turned on and
the working surface was wiped down with 70 % ethanol. Two 25 cm3 T-flasks were
pre-wetted by coating the adherent surface with 4 ml of fresh media. The 1 ml of
cell suspension was directly transferred into a centrifuge tube (containing 4ml of
media) and 100 µl was then taken out for viability determination. The suspension
was centrifuged at 1000 rpm for 5 min to remove DMSO. The pellet was collected
and resuspended in 1 ml of fresh media and divided into the two T-flasks. The Tflasks were transferred to a 27 0C incubator for cells attachment and propagation.
3.4.2
Cells Maintaining
Sf-9 insect cells were maintained in 25 cm3 tissue culture flasks in a
humidified 27 0C incubator. Regular passage of cells was performed every 2 days
with fresh medium by gently dislodging the confluent monolayer, transferring of a
fraction of the suspension to sterile culture flasks, and adding of fresh medium to a
final cell density 5x105 cells/ml with the viability above 90 %. Cell viability was
determined using the trypan blue exclusion test and cell counts were performed using
an inverted microscope.
44
3.4.3
Cells Freezing
The cells were counted using a hemacytometer. Cells should be 90 % viable
and 80-90% confluent. It was recommended to freeze down several vials as low a
passage number as possible at a cell density 1x107 cells. Sterile cryovials were set up
in ice and labeled. The cells were centrifuged at 1000 rpm for 10 min at room
temperature. The supernatant was removed. The cells were resuspended to a given
density in the freezing medium (90 % FBS and 10 % DMSO).
1 ml of cell
suspension was transferred to sterile cryovials. The vials were placed at 4 0C for
15min, -20 0C for 30 min and –180 0C for 60 min. The vials were stored in liquid
nitrogen.
3.5
Wild Type and Recombinant Baculovirus
3.5.1
Virus Propagation
A 25cm3 T-flask was seeded with cells with a density of 5x105 cells/ml and
higher than 90 % viability. The cells were inoculated with virus stock by simply
adding 20 µl inoculum to the cells. The infectious culture was incubated at 27 0C for
7 days, and was visually examined daily to ensure the cells were well infected. To
collect extracellular virus, the infected cells were transferred to a centrifuge tube and
spinned at 1000 x g for 15 min. The supernatant was transferred to a fresh centrifuge
tube and stored at 4 0C. For long term storage, virus inocula should be kept at -80
0
C. The virus stock concentration was determined by end-point dilution.
45
3.5.2
Virus Titration (End-Point Dilution)
Tenfold serial dilutions of the virus stock were prepared. Dilutions of 10-5,
10-6, 10-7, 10-8 should be appropriate in most cases. The cells with the viability
higher than 90% were diluted in a concentration of 1 x105 cells/ml with fresh
medium. 10 µl aliquots of each virus dilution was mixed with 100 µl aliquots of the
cell suspension, and seeded into 96 wells plate. 4 wells were seeded with 100 µl of
cells, as uninfected controls.
The plate was incubated at 27 0C.
To avoid
dehydration, the plate was incubated in a humidified environment. The plate was
sealed in a plastic bag with damp paper towel. The plate was incubated for one
week. Each well was examined for virus replication.
10-5, 10-6, 10-7, 10-8
dilution of virus
Cell Concentration
1 x 105 cells/ml
10ul virus
100ul cell suspension
1 2 3 4 5 6 7 8 9 10 11 12
-5
control
10 Dilution
10-6 Dilution
10-7 Dilution
10-8 Dilution
Replicate
10-5 Dilution
10-6 Dilution
10-7 Dilution
10-8 Dilution
Examine the virus replication
Calculate TCID50
The plate is sealed
in plastic bag.
Incubate for 7 days.
Figure 3.1:
270C
Virus Titer Procedures – End Point Dilution
46
3.5.3
Generating Pure Recombinant Virus Stocks (End Point Dilution)
Sf-9 insect cells with the viability of higher than 90 % were diluted with fresh
medium to a concentration of 5 x105 cells/ml. A tenfold serial dilutions of the virus
were prepared and the dilutions of 10-6 and 10-7 were appropriate for most stocks.
10µl of each dilution was mixed with 100 µl of the cell suspension and seeded into
each well of a 96 wells plate. For each dilution at least 46 replicates were tested.
Therefore 2 tenfold dilutions were tested in one plate including 4 wells for uninfected
controls. Plate was incubated at 27 0C in humidified environment. Each well was
examined daily for virus replication and progress of infection. All wells with sign of
infection were scored as positive and tested for product gene expression using
Enzyme Linked Immunosorbent Assay (ELISA). Samples that gave high levels of
recombinant protein production yield were then selected to undergo the purification
process twice further or until the recombinant protein level reached a constant yield
provided that other parameters remain unchanged for every purification round.
Finally the high purity recombinant virus was amplified to generate large stock.
3.6
Recombinant Human Transferrin Detection
3.6.1
Sodium Dodecyl Sulfate - Polyacrylamide Gel Electrophoresis
A
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)
was performed using a Mini-protean II apparatus (Bio-Rad Laboratories, Melville,
NY). Two phases of polyacrylamide gel were prepared in advance. The gel was
divided into two parts, a stacking gel for the concentration of the protein samples
47
before separation and a separating gel for the separation of the protein samples. The
working solutions for both phases are shown in Appendix B.
The separating gel was mixed well, poured between two plates and overlaid
with water to keep the gel surface flat and left to polymerize for 1 hour. After the gel
had polymerized, a distinct interface appeared between the separating gel and the
water. The water was rinsed off with fresh distilled water and the stacking gel was
prepared. The stacking gel was then poured on the top of the separating gel. A comb
was carefully inserted into the top of the stacking gel, so that no bubbles would be
trapped on the ends of the teeth. The gel was allowed to polymerize for 30 min.
Once polymerized, the gel was attached to the electrode assembly of Bio-Rad
Mini-Protean II Gel System and inserted into an electrophoresis tank that was filled
with 1 x Tris-glycine electrophoresis buffer, afterwards.
Then the comb was
removed. Subsequently, the sample solution (combine protein sample, 20 µl and 1x
sample buffer, 20 µl and heated at 100 0C for 5 min) as well as the protein marker
(consisting of 7 precisely sized recombinant proteins, ranging from 15 kDa to
220kDa), were introduced into the wells on the stacking gel using a Hamilton
syringe.
The electrophoresis was carried out on a vertical slab gel using 9%
acrylamide gel at a constant voltage of 100 V for 90 min at room temperature.
Following electrophoresis, protein bands were visualized using silver stain.
3.6.1.1 Silver Staining
The polyacrylamide gel was soaked in 5:1:4 ratios of methanol, acetic acid
and water for at least 1 hour with 2 or 3 changes of the solution with gentle shaking.
The gel was then soaked for 30 min with water with at least 3 changes of water.
Solution A (0.8 g silver nitrate in 4 ml distilled water), solution B (21 ml 0.36 %
48
NaOH mixed together with 1.4 ml of 14.8 M ammonium hydroxide) and solution C
(solution A was added to solution B with constant stirring and later water was added
to make up a total volume of 100 ml) were freshly prepared. The gel was stained in
solution C for 15 min with gentle and constant agitation. After rinsing the gel twice
in deionized water, the gel was placed in solution D that was freshly prepared (0.5 ml
1% citric acid was added to 50 µl 38 % formaldehyde and later water was added to
make up a total volume of 100 ml) and was shaken until the bands appear.
Development was stopped with 1 % acetic acid.
3.6.2
Western Blot
Proteins were separated on SDS-PAGE gels (9 % polyacryamide, 90 min,
100V) and electroblotted (4 0C, 1 hour and 100 V) onto 0.45 µm nitrocellulose
membrane using a Trans-Blot Electrophoretic Transfer Cell. The transfer buffer
consisted of 15.6 mM Tris and 120 mM glycine and 10 % methanol, pH 8.1 – 8.4.
Membranes were incubated in blocking buffer (Phosphate Buffer Saline (PBS),
pH7.4 containing 5 % skimmed milk) at 4 0C overnight.
The following day,
membranes were washed 3 x 10min with washing solution (PBS containing 0.05 %
Tween 20). Membranes were incubated with primary antibody - Goat anti-Human
Transferrin (Bethyl Laboratories Inc, Texas) in blocking solution with gentle
agitation for 2 hours. This was followed by 3 x 10 min washing using with washing
solution. Membranes were incubated as before with Horseradish Peroxidase (HRP)conjugate secondary antibody, that is Goat anti-human transferrin-HRP conjugate
(Bethyl Laboratories Inc, Texas) in blocking solution. This was followed by 3 x
10min washings with washing solution. Bound antibody was detected by using TMB
(3,3’,5,5’-tetramethylbenzidene) Stabilized Substrate for HRP (Promega, Madison,
WI).
49
3.6.3
Enzyme Linked Immunosorbent Assay
Direct Sandwich Enzyme Linked Immunosorbent Assay (ELISA) were
performed in 96 well microtiter plates (TPP, Switzerland) which were coated with
primary antibody and incubated at 4 0C overnight. The following day, the plate was
washed with washing solution (Tris-buffered Saline (TBS), pH 8.0 containing 0.05%
Tween 20) 3 times. The plate was blocked with blocking solution (TBS, pH 8.0
containing 1% BSA) for 30 min and washed with washing solution 3 times. The
plate was subsequently incubated with serial dilutions of standards - Human
Reference Serum (Bethyl Laboratories Inc, Texas) and samples in sample/conjugate
diluent (TBS, pH 8.0 containing 1% bovine serum albumin (BSA) and 0.05% Tween
20) for 60 min and washed with washing solution 5 times. Next, the plate was
incubated with HRP-conjugated secondary antibody in sample/conjugate diluent for
60 min at 37 0C and washed with washing solution 5 times. Color development by
the enzyme substrate reaction was performed by adding to each well 100 µl of equal
volumes of Trimethyl Benzene (TMB) Peroxidase Substrate and Peroxidase Solution
B (H2O2). After 5-30min, the reaction was stopped with 100 µl of 1 M Phosphoric
Acid (H3PO4). The absorbance at 450 nm was determined.
3.7
Recombinant β1,4-Galactosyltransferase Detection
3.7.1
Thin Layer Chromatography
A typical incorporation mixture contained the following in a final volume of
0.1 ml: 5 µmole of Tris-HCl, pH 7.4, 0.04 µmole of UDP-Gal, 2.0 µmole of glucose,
4 µmole of MnCl2, 0.5 µmole of UTP, 0.14 µmole of α-lactalbumin and sample β1,4-
50
galactosyltransferase. After 30 min of incubation at 37 0C, the reaction was stopped
by adding 0.2 ml of 0.3 N Ba(OH)2 to ice-cooled mixture. This was neutralized with
1.5 volumes of a 5% solution of ZnSO4.7H2O and the precipitate was removed by
centrifugation. The supernatant was analysed using thin layer chromatography.
Silica plate (Whatmann, 200 µm) was marked with a straight line lightly
parallel to the short dimension of the plate, about 1 cm from one end of the plate. A
few small marks were made lightly perpendicular to this line to serve as a guide for
placing the substance spots. The substances were loaded on the plate and developed
in a trough chamber containing mobile phase: n-butanol-acetic acid-diethyl etherwater (9:6:3:1) to a depth of about 5 mm. The migration time was about 120 min.
The chromatogram was freed from the mobile phase and dipped in the solution
containing 8 ml concentrated sulfuric acid, 0.5 ml anisaldehyde, 85 ml methanol and
10 ml glacial acetic acid for 2 seconds. After drying for several minutes in cold air,
the plate was heated to 120 0C for 15 min.
3.7.2
Lectin Binding Assay
25 mg of asialofetuin was dissolved in 1 ml of 0.2 M sodium phosphate
buffer (pH 4.5) containing 0.1 M citric acid.
0.4 units (2.2mg) of bovine β-
galactosidase was added to the mixture. The mixture was incubated for 72 h at 37 0C
to remove galactose residues. The sample was diluted at least 20 times with 0.1M
sodium phosphate-buffered saline (0.15 M NaCl, pH 7.2) containing 1 mM CaCl2
and 1 mM MnCl2 and concentrated using Amicon Model 8010 UF with MWCO of
10 000. This procedure was repeated three times to remove any remaining sugars
which had been released from protein by the enzyme treatments.
produced was asialoagalactofetuin.
The protein
51
Each well of the 96 wells plate was coated with 100 µl of the
asialoagalactofetuin (1 µg/ml in 0.05 M Sodium Carbonate, pH 9.6 containing 2%
glutaldehyde) at room temperature for 1 hour, washed 3 times with washing solution
(PBS containing 0.05% Tween 20, PBST) and then blocked with 1% BSA in PBST
at room temperature for 1 hour. The plate was washed and the enzyme reaction was
started by adding to each well 100 µl of enzyme-donor substrate mixture (10 mM
MnCl2, 0.1% BSA, 0.32 mM UDP-Gal and sample β1,4-GalT) in 30 mM Mops
(pH7.4). The plate was incubated at 37 0C for 1 hour, the reaction was stopped by
discarding the reaction mixture.
The plates were washed and incubated with
peroxidase-labeled RCA 1 (0.16 mg/ml in PBST containing 1% BSA) at 37 0C for 90
min. Color development by the enzyme substrate reaction was performed by adding
to each well 100 µl of equal volumes of TMB Peroxidase Substrate and Peroxidase
Solution B (H2O2). After 5-30 min, the reaction was stopped with 100 µl of 1M
Phosphoric Acid. The absorbance at 450 nm was determined.
3.8
Native Uridine-5′-diphosphogalactose (UDP-Gal) Level
3.8.1
UDP-Gal Extraction
Sf-9 cells or infected Sf-9 cells (about 1 x 106 cells) were collected by
centrifugation (1000 x g, 15 min at 4 0C). Pellets were washed with PBS buffer,
pH7.4. Cells were lysed in ice-cold 75% ethanol (300 µl) by freeze-thawing and
homogenizing. Soluble fractions were obtained by centrifugation (16 000 rpm x
10min at 4 0C). Supernatant was filtered through 10 000 MWCO membranes.
52
3.8.2
Reverse Phase High Performance Liquid Chromatography (RP-HPLC)
Analysis
RP-HPLC elution was carried out at 1ml/min and the column was kept at
0
30 C. UDP-Gal was detected by absorbance at 260 nm. The ion pair RP-HPLC was
carried out using a LC-10Dvp HPLC system and a NovaPac C18 column (3.9 x 150
mm, 4 µm) with a NovaPac C18 guard cartridge. The following two solvents were
used as eluents: 5 mM tetrabutylammonium sulfate (TBAS) – 50 mM ammonium
phosphate, pH 5.0 (E1) and 5 mM TBAS-methanol (E2). A portion of the cell
extract was injected into a NovaPac C18 column equilibrated with a mixture (98:2,
v/v) of E1 and E2.
3.9
Coexpression of Recombinant Human Transferrin and β1,4Galactosyltransferase
Two Sf-9 insect cell cultures were infected with AcMNPV-hTf. One of the
cultures was coinfected with recombinant baculovirus carrying the gene for β1,4GalT (in vivo). For the in vitro analysis, 2.0 mU/ml commercial mammalian GalT
and 0.32 mM commercial UDP-Gal were added to the harvested AcMNPV-hTf
supernatant. There were three negative controls in this experiment which were Sf-9
cell culture infected with AcMNPV-β1,4-GalT, uninfected Sf-9 insect cell culture
and harvested uninfected Sf-9 cell culture mixed with 2.0 mU/ml commercial
mammalian GalT and 0.32 mM commercial UDP-Gal. All samples were harvested
at time 24 hours PI. As for the time course upon coinfection between recombinant
baculovirus hTf and β1,4-GalT, the medium from each of the coexpressed cell
culture supernatants were collected at time intervals of every 24 hours PI until 120
hours PI.
53
Each well of the ELISA plate was coated with 100 µl of the glycoprotein
(1µg/ml in 0.05M Sodium Carbonate, pH 9.6 containing 2% glutaldehyde) at room
temperature for 1 hour, washed 3 times with washing solution (PBST) and then
blocked with 1% BSA in PBST at room temperature for 1 hour. The plate was
washed and the enzyme reaction was started for the in vitro galactosylation samples
by adding to each well 100 µl of enzyme-donor substrate mixture (10 mM MnCl2,
0.1% BSA, 0.32 mM UDP-Gal and 2.0 mU/ml of β1,4-GalT) in 30mM Mops (pH
7.4). The plate was incubated at 37 0C for 1 hour, and then the reaction was stopped
by discarding the reaction mixture. The plates were washed and incubated with
peroxidase-labeled RCA 1 (0.16 mg/ml in PBST containing 1% BSA) at 37 0C for 90
min. Color development by the enzyme substrate reaction was performed by adding
to each well 100 µl of equal volumes of TMB Peroxidase Substrate and Peroxidase
Solution B (H2O2). After 5-30 min, the reaction was stopped with 100 µl of 1M
Phosphoric Acid. The absorbance at 450 nm was determined.
54
CHAPTER 4
RESULTS AND DISCUSSION
For the production of human therapeutic recombinant glycoproteins, it is
essential to determine if the expression system being used can synthesize the
necessary glycan structures required for full biological activity in vivo.
The
baculovirus-insect cell system has an excellent track record for high level expression
of biologically active eukaryotic proteins and this system is used routinely for
foreign glycoprotein production (O’Reilly et al., 1992; Jarvis et al., 1997). However,
glycoproteins produced by insect cells usually lack fully elaborated complex Nlinked glycans (Marz et al., 1995; Jarvis et al., 1997). In achieving and maintaining
proper glycosylation, there are three main factors to ensure the success of the posttranslational modification process, which are acceptor substrate, substrate donor and
enzyme. In the current study, fundamental works regarding the optimization of Sf-9
cells growth and mock infection were carried out.
This was followed by the
establishment of baculovirus-insect cell system consisting of the expressed model
protein, enzyme and sugar nucleotide. The recombinant baculovirus encoded with
hTf (acceptor substrate) was coinfected with recombinant baculovirus carrying the
gene for β1,4-GalT (enzyme). Meanwhile, the native UDP-Gal (substrate donor)
was monitored at normal and upon baculovirus infection.
Different levels of
galactosylation were performed and the relationships among the three elements were
been investigated.
55
4.1
Sf-9 Cells Growth Optimization
Insect cell growth can be significantly improved by paying close attention to
the conditions used in the inoculum stages. Serum concentration, different type of
media, cell subculturing conditions, initial cell density and spent medium carry over
significantly influenced the growth kinetics of Sf-9 cells. Efficient operation of
insect cell culture requires full assessment of these factors which are expected to
have significant influence on the cell metabolic activity.
During cell infection, as cell division stops, other cellular activities such as
respiration still continue as does the transcriptional and translational machinery that
are being switched to viral multiplication and expression of its genes. It is therefore
necessary that cells are held in a healthy physiological state, free of nutrient
limitation, if high recombinant protein yield are to be achieved. Consequently,
comprehensive data are needed on the effects of these important environmental and
physiological parameters that can influence growth, metabolism, cell infection, viral
multiplication and recombinant protein expression in insect cells.
In the early fundamental work, a few parameters which control the growth
rate of Sf-9 cells culture including initial cell density, effect of cell subculturing
conditions and spent medium were investigated. For the mock- and recombinant
baculovirus infection, the interaction of the infection parameters especially MOI and
spent medium with the above culture parameters were also examined.
As shown in Figure 4.1, higher viable cell numbers were attained in the
media (TC-100 and SF900 II SFM) containing higher FBS concentration. Higher
viable cell density in insect cell culture (Luis Maranga et al., 2002) because FBS
could replace insect cell hemolymph as the source of vitamins, growth factors and
other undefined compounds.
56
90.0
TC-100
5
Viable Cell Density (x 10 cells/ml)
80.0
0% serum
5% serum
10% serum
70.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0
0
2
4
6
8 10 12
Time (Day)
14
16
18
(a)
110.0
SF-900 II SFM
5
Viable Cell Density (x 10 cells/ml)
100.0
0% serum
5% serum
10% serum
90.0
80.0
70.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0
0
2
4
6
8 10 12
Time (Day)
14
16
18
(b)
Figure 4.1:
Sf-9 insect cells growth in monolayer culture at 3 different serum
concentrations. (a) TC-100 and (b) SF-900 II SFM. Error bars indicate ±S.D of
duplicates data.
57
Different types of media also resulted in different cell growth. In this study,
two types of media have been used, TC-100 insect medium and SF-900 II SFM. As
presented in Figure 4.2, Sf-900 II SFM support higher cell densities compared to TC100 regardless of whether the medium was serum enriched or not. Based on this
finding, Sf-900 II SFM was used for the rest of the experiments. However, even
though serum affected cell growth positively (Figure 4.2 (b) and (c)), serum free
media were used for the rest of the experiments as serum contained trace amount of
sugar nucleotides and enzymes which may interfere with hTf and β1,4-GalT assay.
The effect of seeding density was investigated at three different cell
concentrations i.e. at 0.20, 1.20 and 2.33 x 105 cells/ml respectively. As shown in
Figure 4.3, the lowest cell concentration resulted in the lowest maximum viable cell
number achieved. This observation is in contrast with Kioukia et al. (1995) which
found that the maximum cell number achieved was highest for the lowest density.
However, the maximum growth rate, µ, was similar in all three cell concentrations at
about 0.004 to 0.011 h-1 (Table 4.1).
58
(a)
60.0
TC-100
SF-900 II SFM
40.0
5
Viable Cell Number (x 10 cells/ml)
50.0
30.0
20.0
10.0
0.0
0
2
4
6
8
10 12
Time (Day)
14
16
18
90.0
TC-100
(b)
SF-900 II SFM
70.0
60.0
5
Viable Cell Number (x 10 cells/ml)
80.0
50.0
40.0
30.0
20.0
10.0
0.0
0
2
4
6
8 10 12
Time (Day)
14
16
18
110.0
(c)
TC-100
100.0
SF-900 II SFM
80.0
70.0
5
Viable Cell Number (x 10 cells/ml)
90.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0
0
Figure 4.2:
2
4
6
8
10 12
Time (Day)
14
16
18
Sf-9 insect cells growth in monolayer culture for 2 types of media.
(a) without serum; (b) with 5% serum and (c) with 10% serum. Error bars
indicate ±S.D of duplicates data.
59
24.0
Initial Cell Density
0.2 x 105
22.0
1.2 x 105
2.33 x 105
18.0
16.0
5
Viable Cell Density (x 10 cells/ml)
20.0
14.0
12.0
10.0
8.0
6.0
4.0
2.0
0.0
0
Figure 4.3:
2
4
6
8
10
Time (Day)
12
14
16
Sf-9 insect cells growth in monolayer culture for 3 different initial
cell density, i.e. 0.2, 1.2 and 2.33 x 105 cells/ml. Error bars indicate ±S.D of
duplicates data.
60
Cell subculturing condition was also investigated. Three inocula at fixed
density of 1.6 x 105 cells/ml were seeded at three different phases i.e. early
exponential, late exponential and stationary phase.
Early exponential phase
subculturing resulted in the fastest cell growth rate (0.014/h) compared to the other
two phases; while those from stationary phase were obviously unsatisfactory (growth
rate and maximum viable cell number were 0.006/h and 12.55 x 105 cells/ml) as
shown in Figure 4.4. In related work (Kiokia et al., 1995), it has been shown that
there were higher proportions of G1 and S phase cells in the early exponential than in
the other growth phases. It was reported that in insect cells, the resting phase is G2
where most cells are accumulated when nutrient is depleted (Fertig et al., 1990).
This could explain the observation that cells from the early exponential phase set off
faster and achieved the highest growth rate.
The effect of spent medium on cell growth was also investigated. Spent
medium carry over has also been considered as the factor that can affect cell growth.
Cells were inoculated at fixed density, 1.5 x 105 cells/ml as shown in Figure 4.5. The
effect on growth became more significant for the spent medium percentage of 100%
and 50% which resulted in reduction in growth rate and maximum cell number
compared to the negligible percentage (0%). The result is expected because cells
prefer to survive in a rich nutrient culture for their metabolism, kinetics, respiration,
viral multiplication and protein expression.
61
20.0
Early Exponential
Late Exponential
Stationary
18.0
5
Viable Cell Number (x 10 cells/ml)
16.0
14.0
12.0
10.0
8.0
6.0
4.0
2.0
0.0
0
Figure 4.4:
2
4
6
8
Time (Day)
10
12
14
Sf-9 insect cells growth in monolayer culture at 3 different
subculturing conditions, i.e. early exponential, late exponential and stationary
phase. Error bars indicate ±S.D of duplicates data.
62
16.0
Spent Medium
Percentage:
14.0
100 %
50 %
5
Viable Cell Density (x 10 cells/ml)
0%
12.0
10.0
8.0
6.0
4.0
2.0
0.0
0
Figure 4.5:
2
4
6
Time (Day)
8
10
12
Sf-9 insect cells growth in monolayer culture at 3 different spent
medium carry over percentage, i.e 100%, 50% and 0%. Error bars indicate
±S.D of duplicates data.
63
4.2
Establishment of Baculovirus Expression Vectors System (BEVS)
4.2.1
Mock Infection Optimization
Three factors were investigated in the mock infection optimization including
effect of initial cell density, spent medium and MOI. In all experiments, Sf-9 cells
were infected with wild type baculovirus (AcMNPV) during the early exponential
phase.
Three different initial cell densities, i.e. 0.95, 2.05 and 5.13 x 105 cells/ml
respectively were infected with AcMNPV at MOI 10 in fresh medium. The cell
densities used in this experiment were relatively low to ensure oxygen and nutrient
will not be the limiting factor. As shown in Figure 4.6, the rates of infection were
similar and reached a maximum infectivity of 98.5%, 99.5% and 100% for 0.95, 2.05
and 5.13 x 105 cells/ml respectively by 120 h post-infection (PI).
The results
indicated that initial cell density alone was not a critical parameter for determining
cell infectivity.
The finding on the effect of spent medium carry over on viral infectivity is
interesting. As shown in Figure 4.7, this observation suggests that the medium must
be replenished before viral infection in order to achieve highest protein expression.
.
64
100.0
(a)
Initial Cell Density
0.95 x 105
90.0
2.05 x 105
5.13 x 105
80.0
Infectivity (%)
70.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0
0
24
48
72
96
Time post-infection (h)
120
20.0
(b)
Initial Cell Density
0.95 x 105
2.05 x 105
5.13 x 105
16.0
14.0
5
Viable Cell number (x 10 cells/ml)
18.0
12.0
10.0
8.0
6.0
4.0
2.0
0.0
0
Figure 4.6:
24
48
72
96
Time post-infection (h)
120
The effect of initial cell density on Sf-9 insect cells infected with
wild type AcMNPV viruses at MOI 10. (a) Infectivity percentage versus time
post-infection (TPI) and (b) Viable cell number versus TPI. Error bars indicate
±S.D of duplicates data.
65
100.0
(a)
Spent Medium Percentage:
90.0
100%
80.0
50%
0%
Infectivity (%)
70.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0
0
24
48
72
96
120
96
120
Time post-infection (h)
10.0
(b)
9.0
100%
8.0
50%
0%
7.0
5
Viable Cell Number (x 10 cells/ml)
Spent Medium Percentage:
6.0
5.0
4.0
3.0
2.0
1.0
0.0
0
Figure 4.7:
24
48
72
Time post-infection (h)
The effect of spent medium carry over on Sf-9 insect cells infected
with wild type AcMNPV viruses at MOI 10. (a) Infectivity percentage versus
TPI and (b) Viable cell number versus TPI. Error bars indicate ±S.D of
duplicates data.
66
It is expected that increasing the added amount of virus in cultures can
intensify the process of cell infection. Therefore, by increasing the number of virus
per cell (MOI), a reduction in the time of cell infection can be achieved. To study
this phenomenon, different MOIs were used to infect the stationary phase cell culture
using fresh media as cell infectivity and viral yields in stationary phase have been
found to be strongly dependent on the MOI (Licari et al., 1991). The behavior is
different to that when cells were infected in the exponential phase (Maiorella et al.,
1988, Schorp et al., 1990). As shown in Figure 4.8, in the stationary phase, higher
MOI will enhance the rate of infection (rate of polyhedra development as observed
microscopically). However, final infectivity for each MOI used was similar and this
might probably be due to the availability of nutrient in the fresh media allowing the
cells to survive long enough to be infected by viruses released from primary infected
cells.
67
100.0
(a)
MOI :
1
15
50
100
90.0
80.0
Infectivity (%)
70.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0
0
24
48
72
96
Time post-infection (h)
120
40.0
(b)
MOI :
1
15
50
100
5
Viable Cell Density (x 10 cells/ml)
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
0
Figure 4.8:
24
48
72
96
Time post-infection (h)
120
The effect of MOI on Sf-9 insect cells infected in the stationary
phase with wild type AcMNPV Viruses. (a) Infectivity percentage versus TPI
and (b) Viable cell number versus TPI. Error bars indicate ±S.D of duplicates
data.
68
Table 4.1: Growth Kinetics of Sf-9 Cells at Different Parameters
1
Experiments
Maximum
viable cell
1
2
µ ( ) 3 t d (h)
density
h
x 105
cells/ml
INSECT CELL CULTURE
4
GI
5
t d ( h)
Final
infectivity
percentage
(%)
_
Fig. 4.1: Effect of serum
(a) TC-100 without serum
TC-100 with 5% serum
TC-100 with 10% serum
51.88
74.88
81.03
0.011
0.017
0.019
63.4
39.7
37.4
24.7
35.7
38.6
83.0
74.5
63.8
ND
ND
ND
(b) SFM without serum
SFM with 5% serum
SFM with 10% serum
57.65
86.63
102.43
0.010
0.010
0.013
66.5
70.2
51.4
27.5
41.3
48.8
70.3
53.7
51.4
ND
ND
ND
(a) TC-100 without serum
SFM without serum
51.88
57.65
0.011
0.010
63.4
66.5
24.7
27.5
83.0
70.3
ND
ND
(b) TC-100 with 5% serum
SFM with 5% serum
74.88
86.63
0.017
0.010
39.7
70.2
35.7
41.3
74.5
53.7
ND
ND
(c) TC-100 with 10% serum
SFM with 10% serum
81.03
102.43
0.019
0.013
37.4
51.4
38.6
48.8
63.8
51.4
ND
ND
15.00
17.43
23.00
0.004
0.011
0.011
165.3
62.8
64.8
75.0
14.5
9.9
46.2
62.2
58.1
ND
ND
ND
19.25
16.25
12.55
0.014
0.010
0.006
49.6
67.4
121.7
10.9
12.8
8.4
55.6
52.1
94.0
ND
ND
ND
Fig. 4.2: Two types of
media
comparison
Fig. 4.3: Effect of initial
density of seeding
inoculum
- 0.2 x 105 cells/ml
- 1.2 x 105 cells/ml
- 2.33 x 105 cells/ml
Fig. 4.4: Effect of cell
subculturing
conditions
- Early exponential
- Late exponential
- Stationary
69
Table 4.1: Growth Kinetics of Sf-9 Cells at Different Parameters (Continue)
1
Experiments
Maximum
viable cell
density
x 105 cells/ml
1
2
µ( )
h
5.45
12.10
15.35
0.006
0.013
0.017
3
t d (h)
t d ( h)
Final
infectivity
percentage
(%)
4.4
9.8
12.5
111.8
72.8
39.5
ND
ND
ND
4
GI
5
_
Fig. 4.5: Effect of
6
spent
medium
carry-over
Spent medium
percentage:
- 100 %
- 50%
- 0%
118.7
54.9
40.3
MOCK INFECTION
Fig. 4.6: Effect of
initial
density
- 0.95 x 105 cells/ml
- 2.05x 105 cells/ml
- 5.13 x 105 cells/ml
1.98
7.23
17.63
0.006
0.014
0.020
110.9
50.7
34.6
2.1
3.5
3.4
90.9
39.6
40.4
98.5
99.5
100.0
3.23
8.23
14.60
0.016
0.034
0.038
43.2
20.5
18.4
3.4
8.7
15.4
40.8
23.1
18.3
72.5
93.8
99.8
34.30
31.80
21.58
12.63
0.021
0.016
0.015
0.008
32.7
44.3
46.0
88.3
4.6
4.3
2.9
1.7
32.6
34.3
46.8
62.8
97.3
97.3
99.5
99.5
Fig. 4.7: Effect of
spent
medium
carry-over
Spent medium
percentage:
- 100 %
- 50%
- 0%
Fig. 4.8: Effect of
MOI
-
MOI = 1
MOI = 15
MOI = 50
MOI = 100
70
1
All experiments were carried out in 25cm3 T-flask.
2
Maximum growth rate,
1
µ ( ) = ln X 2 − ln X 1
t 2 − t1
h
where X2 = viable cell number at t2
X1 = viable cell number at t1
3
4
Doubling time,
t d (h) = ln 2
Growth index,
GI = max imum cell density
µ
initial cell density
5
_
Average doubling time,
t d ( h) =
ln 2
ln GI
(
)
tmax
,
where tmax = time at the maximum viable cell density
6
Spent medium was prepared by centrifugation of medium from a 12 day culture at
death phase. The low viability percentage, 18.5% indicated that most of the cells
were lysed and the breakage generated a lot of impurities in the suspension.
ND = Not Determined
71
4.2.2
Recombinant Human Transferrin Expression
Sf-9 cells infected with recombinant baculovirus encoded with hTf were
harvested at 120 hour PI.
Following clarification by low speed centrifugation,
recombinant protein from infected cells was analyzed using silver stained SDSPAGE and western blotting. A silver stained gel revealed a 70 kDa protein in the
supernatant (Figure 4.9 (a)). The protein was further examined for reactivity with
antibody directed against human transferrin. Supernatant of the infected cells at 120
hour PI (Figure 4.9 (b)) was separated by SDS-PAGE, transferred to nitrocellulose
membranes, and reacted with antibody directed against human transferrin. The
molecular weight of hTf produced in insect cells is lower than that of the native
protein (Apo transferrin-human) which is 75 kDa. This is expected as native hTf
contains two N-glycosylation sites that are predominantly occupied by terminally
sialylated biantennary complex oligosaccharides (Ailor et al., 2000), whereas the hTf
produced in insect cells contain neither galactose nor sialic acid.
4.2.2.1 Time Course Expression of Recombinant Human Transferrin
Sf-9 cells infected with recombinant baculovirus carrying hTf gene were
harvested at every 24 h intervals. Silver stained electrophoresis gel revealed a 70
kDA protein in the supernatants of the infected cells (Figure 4.10).
expression was detected as early as 24 hour PI.
Protein
The rhTf accumulation in
supernatants increased steadily until hour 120 (Figure 4.11 (b)). This pattern is
different compared to the rhTf detection in cell lysates which decreased with the
increase in harvest time (Figure 4.11 (a)). These data supported the notion that the
rhTf were being secreted out into the culture medium.
72
During cell infection, while cell division stops, other cellular activities such
as respiration still continue as does the transcriptional and translational machinery
that is being switched to viral multiplication and expression of its genes. The
recombinant protein produced was secreted into the environment. At the same time,
viral propagations occurred. This mechanism can be used to explain why with the
increase in time, rhTf in supernatant increased while rhTf in lysates decreased.
M
kDa
Apo
transferrinhuman
rhTf
rhTf
225
150
100
75
50
35
25
(a) SDS-PAGE
Figure 4.9:
(b) Western Blot
(a) 9% silver stained SDS-PAGE; (b) Western blot analysis of
the rhTf protein synthesized in Sf-9 cells supernatant at hour 120. M, molecular
weight standards. Arrows indicate the position of rhTf (Product) at molecular
weight 70 kDa.
73
kDa
M
1
2
3
4
5
6
225
150
100
75
50
35
25
Figure 4.10: Time Course of rhTf protein production in supernatants were
resolved on 9% SDS-PAGE and stained with silver. Lane 1 represents Apo
Transferrin Human. Lane 2 – 6 represent supernatants harvested at hour 24,
48, 72, 96 and 120 respectively. M, molecular weight standards. Arrows
indicate the position of rhTf (Product) at molecular weight 70 kDa
74
7.0
(a)
rhTf in cell lysates
6
hTf Prodiuction (ug/x10 cells)
6.0
5.0
4.0
3.0
2.0
1.0
0.0
0
24
48
72
96
Time Post-Infection (h)
120
144
45.0
(b)
rhTf in supernatants
6
hTf Production (ug/x10 cells)
40.0
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
0
24
48
72
96
Time Post-Infection (h)
120
144
Figure 4.11: Time course of rhTf protein production in (a) Lysates;
(b) Supernatants were detected using ELISA. Goat anti-human transferrinaffinity purified as a primary antibody whereas goat anti-human transferrinHRP conjugate as a secondary antibody. Error bars indicate ±S.D of duplicates
data.
75
4.2.3
Recombinant β1,4-Galactosyltransferase Expression
The supernatant of the Sf-9 cell culture infected with recombinant baculovirus
carrying the gene for β1,4-GalT was harvested at 120 hours PI. Following lactose
synthetase assay, recombinant protein from infected cells was analyzed using thin
layer chromatography (TLC). β1,4-GalT catalyzed UDP-Gal in the biosynthesis of
lactose in the presence of α-lactalbumin as the specifier protein and glucose as the
substrate (Brodbeck and Ebner, 1966; Brodbeck et al., 1967). α-lactalbumin is
termed the “Specifier Protein” because it modifies the substrate specificity of
galactosyltransferase from N-acetylglucosamine to glucose and allows the synthesis
of lactose in the presence of glucose (Brodbeck and Ebner, 1966; Brodbeck et al.,
1967). In the reaction, a typical mixture contained the following in a final volume of
0.1 ml: 5 µmole of Tris-HCl, pH 7.4; 0.04 µmole of MnCl2; 0.50 µmole of UTP; 0.14
µmole of α-lactalbumin and supernatant harvested at 120 hours PI. As shown in lane
3 of Figure 4.12, lactose was produced and glucose concentration decreased after the
reaction suggesting that β1,4-GalT was expressed in the supernatant.
4.2.3.1 Time Course Expression of β1,4-Galactosyltransferase
The supernatants of infected Sf-9 cell culture with recombinant baculovirus
encoded with β1,4-GalT, were harvested at 24 h intervals. Lactose synthetase assays
were performed and analyzed using TLC (Figure 4.13). Silver stained electrophoresis
gel of the supernatants revealed a 45 kDa protein (Figure 4.14) in infected cells.
Thus it can be concluded that protein expression was started as early as hour 24. The
enzyme β1,4-GalT accumulation in supernatants increased steadily until hour 144.
76
Glucose
Lactose
Start
1
2
3
Figure 4.12: Detection of β1,4-GalT by using chromatogram TLC. Layer:
Whatmann, 200µm. Solvent : n-butanol-acetic acid-diethyl ether-water
(9:6:3:1). Detection Method: Anisaldehydyde-sulfuric acid reagent. Lane 1
represents standard lactose; lane 2 represents standard glucose; lane 3
represents reaction mixture containing the following in a final volume of 0.1 ml:
5 µmole of Tris-HCl, pH 7.4; 0.04 µmole of MnCl2; 0.50 µmole of UTP; 0.14
µmole of α-lactalbumin and the sample was the source from 120 hours PI
culture supernatant.
77
Glucose
Lactose
Start
1
2
3
4
5
6
7
8
Figure 4.13: Time course of chromatogram of TLC. Layer: Whatmann,
200µm. Solvent : n-butanol-acetic acid-diethyl ether-water (9:6:3:1). Detection
Method: Anisaldehydyde-sulfuric acid reagent. Lane 1 represents standard
lactose; lane 2 represents standard glucose; lane 3 represents standard lactose
and glucose mixture; lane 4 - 8 represent supernatants harvested at hour 24, 48,
72, 96 and 120 respectively.
78
kDa
M
1
2
3
4
5
6
7
225
150
100
75
50
35
25
Figure 4.14: SDS-PAGE (9%) time course of β1,4-GalT production. A silver
stained gel revealed lane 1-7 represent cell culture supernatants harvested at
every 24 hours intervals from 0 hour to 144 hours, respectively. M, molecular
weight marker. Arrow indicate the position of β1,4-GalT at molecular weight
45 kDa (Product).
79
4.2.3.2 The Development of β1,4-Galactosyltransferase Assay
A
number
glycosyltransferase.
of
methods
are
available
for
the
measurement
of
In radiochemical assays, the radioactivity incorporated into
substrate acceptor from radiolabeled sugar nucleotide donors will be proportional to
the amount of enzyme present. However, this approach has some drawbacks such as
high costs and inevitable disposal problem of the radiochemical wastes. In this
respect, many nonradioactive ELISA-based methods for glycosyltransferase
activities have been developed (Stult and Macher, 1990; Taki et al., 1990; Zatta et
al., 1991; Keshvara et al., 1992; Keusch et al., 1995).
All these methods are
essentially based on the same principle. First, either a glycolipid or glycoprotein is
used as an acceptor substrate on the solid surface. Second, the reaction products are
identified by either monoclonal antibodies or specific lectins labeled with enzymes or
fluorescent compounds.
In the study, a lectin binding assay similar to an ELISA-based method was
Asialofetuin was digested with bovine β-galactosidase and the
performed.
asialoagalactofetuin produced was used as an acceptor substrate.
The
asialoagalactofetuin was coated onto ELISA 96 wells plate, whereas peroxidase
labeled-Ricinus communis agglutinin-I (RCA-I) lectin, which recognized galactose
residues was used to recognize Galβ1,4→GlcNAc linkage on N-glycan of
glycoprotein.
Using the optimal conditions determined for the substrate donor as well as
Mn2+ concentration (Oubihi et al., 1998), the relationship between ELISA values as
the peroxidase labeled-RCA 1 binding signal and the enzyme reaction time was
assessed with different concentrations of β1,4-GalT. As illustrated in Figure 4.15,
sufficient linearity (R2 = 0.9935) was obtained for the β1,4-GalT activity between 0.5
to 2 mU/ml. As shown in Figure 4.16, enzyme expression was seen starting from
hour 24.
120.
β1,4-GalT accumulation in cell culture supernatants increased until hour
80
0.060
0.050
y = 0.025x + 0.0017
A 450nm
0.040
2
R = 0.9935
0.030
0.020
0.010
0.000
0.0
0.5
1.0
1.5
2.0
2.5
β 1,4-GalT (mU/mL)
Figure 4.15: Standard curve for the determination of β1,4-GalT activity from
the lectin binding assay values. Asialoagalactofetuin was used as the substrate
acceptor, UDP-Gal was used as the substrate donor and peroxidase-labeled
RCA 1 was used as the lectin to recognize the Gal β1,4-GlcNAc linkage. Error
bars indicate ±S.D of duplicates data.
81
1.8
1.6
β 1,4-GalT(mU/ml)
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
0
24
48
72
96
120
Time of Infection (hours)
144
Figure 4.16: Time course of β1,4-GalT enzyme accumulation in supernatants
detected using lectin binding assay. Asialoagalactofetuin was used as an
acceptor, UDP-Gal was used as the donor substrate and Ricinus communis
agglutinin 1 (RCA 1) was used as the lectin. Error bars indicate ±S.D of
duplicates data.
82
4.2.4
Native Uridine-diphosphogalactose (UDP-Gal) Monitoring at Normal
and Upon Baculovirus Infection
Three main elements to ensure successful protein galactosylation process, are
the presence of sufficient amount of hTf as the substrate acceptor, β1,4-GalT as the
enzyme and UDP-Gal as the substrate donor. In order to achieve galactosylation
effectively, one of the strategies is the introduction of β1,4-GalT artificially. Ailor et
al. (2000) revealed that the oligosaccharide structures of hTf produced in insect cells
infected with GalT baculovirus can alter the glycoforms of the expressed transferrin.
However, another issue is whether the use of native UDP-Gal is sufficient. Tomiya
et al. (2001) had applied High Performance Anion Exchange Chromatography
(HPAEC) method to determine the intracellular sugar nucleotide level of cultured Sf9 insect cells at normal level which was found to be around 1400 pmol/mg protein,
but not at the infection level. Thus, the introduction of β1,4-GalT artificially during
the expression of the hTf would not guarantee the success of the galactosylation
process.
As illustrated previously, after infection recombinant hTf and β1,4-GalT
expression increased over time . This can be explained by the nature of baculovirus
infection cycle.
Upon infection, the cells’ mechanism will be shifted to viral
multiplication and expression of its genes and thus the recombinant protein secretion
will increase upon time of infection and will be secreted into the environment.
However, the effect of sugar nucleotide content upon baculovirus infection has never
been reported.
This finding is very important to ensure the success of the
galactosylation process. This study is very interesting as it can unlock the potential of
BEVS for the production of recombinant biopharmaceuticals.
In order to establish and monitor the native UDP-Gal at normal and upon
infection, a RP-HPLC analysis was carried out as described in section 3.8.2. Ion pair
RP-HPLC is known to be the one of the most effective methods for separating sugar
nucleotide and nucleotide (Ryll and Wagner, 1991). In this study a RP-HPLC
83
column (NovaPac C18, Waters) with tetrabutylammonium sulfate as an ion-pairing
reagent was used. A series of different concentrations of standard UDP-Gal was
monitored by RP-HPLC at a flowrate 1 ml/min and UV 260 nm. From Figure 4.17
(a) and (b), it was observed that the UDP-Gal peaks were eluted at around 4.8 min.
The heights of the standard peaks are proportional to the substrate donor
concentrations. From the HPLC chromatogram, a standard curve which plotted the
area against the concentration was shown in Figure 4.18. A satisfactory linearity of
0.9994 was obtained.
84
0.095
0.00013 umole
0.085
0.00025 umole
0.075
0.00038 umole
0.00050 umole
0.065
0.00063 umole
A260
0.055
0.00075 umole
0.045
0.035
0.025
0.015
0.005
-0.005
0
1
2
3
4
5
6
7
Elution Time (min)
(a)
0.1
A260 nm
0.08
0.06
0.04
3
3.6 4.2
Elu tio
4.8
n Tim
5.3 5.9
e (Mi
n)
at io
ntr
e
c
n
Co
0.00075
0.00050
1.8 2.4
0.00025
1.2
0.00013
0 0.6
0.00038
0
0.00063
0.02
)
ole
um
(
n
(b)
Figure 4.17:
RP-HPLC chromatogram for UDP-Gal standard at different concentrations. 50
mM ammonium phosphate, pH 5.0 containing 5 mM tetrabutylammonium sulfate (E1) and
methanol containing 5 mM tetrabutylammonium sulfate (E2) were used as the eluents. Standard
UDP-Gal portion (10 µl) was injected into NovaPac C18 column (ø3.9 x 150mm) equilibrated with
the mixture of E1 and E2 (98:2, v/v). (a) 2D diagram; (b) 3D diagram.
85
4000000
3500000
3000000
y = 5E+09x
R2 = 0.9994
Area
2500000
2000000
1500000
1000000
500000
0
0
0.0002
0.0004
0.0006
0.0008
µmole
Figure 4.18: Standard curve for UDP-Gal. Area was plotted against the
corresponding UDP-Gal concentration in µmole.
86
In order to verify the reliability of the assumption that the eluted peak at 4.8
min was UDP-Gal, the native UDP-Gal sample was spiked with 10 µmole of
commercial UDP-Gal. As observed in Figure 4.19, at the elution time of 4.8 min the
peak of the native sample with the spiking was higher compared to the native sample
without the spiking thus confirming the assumption.
In order to monitor the level of native UDP-Gal at normal and upon infection,
extraction of Sf-9 cells and infected Sf-9 cells with AcMNPV-hTf were prepared as
described in section 3.8.1. From the RP-HPLC chromatograms as shown in Figure
4.20, 4.21, 4.22 and 4.23, the UDP-Gal peaks which eluted at around 4.8 min were
significantly becoming smaller starting from 0 hour PI until 120 hour PI at 24 hour
intervals. Using the standard curve generated in Figure 4.18, a graph of UDP-Gal
concentration in µM versus time of infection in hours is illustrated in Figure 4.24.
The UDP-Gal level was 15 µM at the beginning and dropped to almost zero upon
five days recombinant baculovirus infection.
To further confirm that the disappearing peak was UDP-Gal, the UDP-Gal
fractions from the RP-HPLC analysis were collected and verified with another assay,
TLC.
The methods are as described in section 3.7.1.
The time course assay
confirmed the reduction of UDP-Gal content upon baculovirus infection occurs as
observed in Figure 4.25.
87
0.50
Native
0.45
Native with Spiking
0.40
0.35
UDP-Gal
A260
0.30
0.25
0.20
0.15
0.10
0.05
0.00
-0.05
0
1
2
3
4
5
6
7
Elution Time (min)
Figure 4.19:
RP-HPLC chromatogram for native UDP-Gal sample with spiking and without
spiking. In spiking, 10 µmole commercial UDP-Gal was introduced into the native sample to
further confirm the elution time of UDP-Gal peak.
88
0.50
0.50
0.45
0.45
0.40
0.40
0.35
0.35
0.30
UDP-Gal
0.25
A260
A260
0.30
0.20
0.25
UDP-Gal
0.20
0.15
0.15
0.10
0.10
0.05
0.05
0.00
0.00
-0.05
-0.05
0
1
2
3
4
5
6
0
7
1
2
3
4
5
6
7
Elution Time (min)
Elution Time (min)
(b)
(a)
0.50
0.45
0.45
0.40
0.40
0.35
0.35
0.30
0.25
0.25
A260
A260
0.30
UDP-Gal
0.20
0.20
UDP-Gal
0.15
0.15
0.10
0.10
0.05
0.05
0.00
0.00
-0.05
-0.05
0
1
2
3
4
5
6
0
7
1
2
3
4
5
6
7
Elution Time (min)
Elution Time (min)
(d)
(c)
0.55
0.55
0.45
0.45
0.35
A260
A260
0.35
0.25
0.25
UDP-Gal
UDP-Gal
0.15
0.15
0.05
0.05
-0.05
-0.05
0
1
2
3
4
Elution Time (min)
5
(e)
6
7
0
1
2
3
4
Elution Time (min)
5
6
(f)
Figure 4.20: RP-HPLC Chromatogram for the time course of native UDP-Gal
level upon infection. Infection time at (a) 0h (Normal); (b) 24h; (c) 48h; (d) 72h;
(e) 96h and (f) 120h. (Set Data 1)
7
89
0.5
0.4
0.3
0.2
0.1
me
Ti
o
96 hrs
72 hrs
48 hrs
24 hrs
5.4
5
4.7
0 hr (Native)
UDP-Gal
3.7
3.1
Time
(Min)
4.3
Elu tio
n
2.5
2
1.6931
1.2
0.6
0
0
ect
nf
fI
120 hrs
Absorbance 260 nm
0.6
io n
Figure 4.21: RP-HPLC chromatogram for time course of native UDP-Gal level
upon infection in 3D diagram. Arrow indicated the UDP-Gal peak eluted at
time 4.8 min. (Set Data 1)
90
0.50
0.45
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.35
0.30
0.25
A260
UDP-Gal
UDP-Gal
0.20
0.15
0.10
0.05
0.00
-0.05
0.05
0.00
-0.05
0
1
2
3
4
5
6
0
7
1
2
3
4
5
6
7
Elution Time (min)
Elution Time (min)
(b)
(a)
0.50
0.55
0.45
0.50
0.45
0.40
0.40
0.35
0.35
0.30
0.30
A260
0.25
UDP-Gal
0.20
0.25
0.20
0.15
UDP-Gal
0.15
0.10
0.10
0.05
0.05
0.00
0.00
-0.05
0
1
2
3
4
5
6
7
-0.05
0
1
2
Elution Time (min)
3
0.50
0.45
0.40
0.35
0.30
A260
0.25
UDP-Gal
0.15
0.10
0.05
0.00
-0.05
0
1
2
5
6
7
(d)
0.55
0.20
4
Elution Time (min)
(c)
A260
A260
A260
0.50
0.45
0.40
3
4
Elution Time (min)
5
6
7
(e)
0.60
0.55
0.50
0.45
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
-0.05
UDP-Gal
0
1
2
3
4
Elution Time (min)
5
6
(f)
Figure 4.22: RP-HPLC Chromatogram for the time course of native UDP-Gal
level upon infection. Infection time at (a) 0h (Normal); (b) 24h; (c) 48h;
(d)72h; (e) 96h and (f) 120h. (Set Data 2)
7
91
0.5
0.4
0.3
0.2
0.1
me
Ti
o
96 hrs
72 hrs
48 hrs
24 hrs
5
5.4
0 hr (Native)
UDP-Gal
4.7
3.7
3.1
Time
(Min)
4.3
Elu tio
n
2.5
2
1.6931
1.2
0.6
0
0
ect
nf
fI
120 hrs
Absorbance 260 nm
0.6
io n
Figure 4.23: RP-HPLC chromatogram for time course of native UDP-Gal level
upon infection in 3D diagram. Arrow indicated the UDP-Gal peak eluted at
time 4.8 min. (Set Data 2)
92
16
UDP-Gal Concentration
(µ Molar)
14
12
10
8
6
4
2
0
0
24
48
72
96
120
Time of Infection (Hours)
Figure 4.24:
Native UDP-Gal concentration in µM at normal and upon time
of infection. The UDP-Gal concentrations at 0, 24, 48, 72, 96 and 120 h PI were
derived from the chromatograms from Figure 4.20 and 4.22. Error bars
indicate ±S.D of duplicates data.
93
Glucose
Lactose
Start
1
Figure 4.25:
2
3
4
5
6
7
8
Verification of UDP-Gal fractions from RP-HPLC analysis using
TLC. Layer: Whatmann, 200µm. Solvent : n-butanol-acetic acid-diethyl etherwater (9:6:3:1). Detection Method: Anisaldehydyde-sulfuric acid reagent. Lane
1: Lactose; Lane 2: Glucose; Lane 3 lane to 8: Time course infection of UDPGal level at 0 (normal), 24, 48, 72, 96, 120h PI respectively.
94
4.2.5 Baculovirus Coinfection Study
Baculovirus coinfection study was carried out in order to evaluate the
recombinant glycoprotein quality i.e whether the hTf oligosaccharides included a
terminal Galactose residue.
As mentioned in Chapter 2, a lot of experimental
evidences suggested that glycoproteins produced in insect cells comprised of
incomplete N-glycans structures. Ailor et al. (2000) analyzed the N-glycan of human
serum transferrin produced in insect cells using metabolic radiolabeling of the
intracellular and extracellular protein fractions, followed by three-dimensional
HPLC.
The attached oligosaccharides included high mannose, paucimannocidic
(Butters and Hughes,1981; Hsieh and Robbins, 1984; Kuroda et al., 1990; Chen and
Bahl, 1991; Kulakosky et al., 1998), and some hybrid structures (Hard et al., 1993;
Kubelka et al., 1994; Davidson et al., 1990; Ogonash et al., 1996) with over 50% of
these structures containing one fucose, α(1,6)- or two fucoses, α(1,6)- and α(1,3)-,
linked to the Asn-linked N-acetylglucosamine. Neither sialic acid nor galactose was
detected on any of the N-glycans.
One possible reason for the limitation in N-glycan processing of
glycoproteins in insect cells is the deficiency in the enzymes necessary for the
production of complex oligosaccharides.
In order to determine if altering the
intracellular level of an enzyme in the oligosaccharide processing pathway can
promote the elongation of hTf N-glycan, recombinant β1,4-GalT was overexpressed
in Sf-9 insect cells in conjuction with hTf.
To evaluate the extent of glycosylation by coinfection strategy, an assay was
established. In this binding assay, glycoprotein was absorbed onto the ELISA plate
surface. A lectin known as Ricinus communis agglutinin-I (RCA-I) labeled with
peroxidase, was added and allowed to recognize Galβ1→4GlcNAc group on Nglycan of glycoprotein. Unbound lectin was removed by washing and the bound
lectin determined by adding equal volumes of TMB and Peroxidase Solution B,
which can be measured by the appropriate color reaction.
95
In this study, the time course upon coinfection between recombinant
baculovirus hTf and β1,4-GalT was investigated. The medium from each of the
coexpressed cell cultures was collected every 24 hour post infection until 120 hour PI
as shown in Figure 4.26. The coinfection strategy was a success. However, as
shown in Figure 4.26, in vivo galactosylation efficiency decreased gradually upon
infection due to the limitation of the substrate donor concentration, UDP-Gal, to
construct the Galβ1→4GlcNAc linkage at the end of the N-glycan hTf. The trend of
Fig. 4.26 is obviously similar to that of Fig. 4.24.
Since the coexpression between the hTf and β1,4-GalT (in vivo) did not
achieved satisfactory results in improving glycoprotein quality due to the reduction
of UDP-Gal upon bacolovirus infection, another alternative, in vitro galactosylation
was proposed to overcome this problem. Commercial mammalian GalT and UDPGal were introduced artificially to the hTf after it was secreted from Sf-9 cell culture.
To determine the optimal conditions for the reaction, different concentrations of
GalT were added to the hTf in the presence of excess amount of sugar nucleotide
(0.32 mM UDP-Gal). Lectin RCA-1 was latter added to recognize the β1,4-linkage
on the N-glycan of hTf, followed by TMB color reaction measurement. As shown in
Figure 4.27, the galactosylation reaction reached a saturation point once the
concentration of the mammalian GalT achieved 2.0 mU/ml or above.
One
explaination is that the concentration of glycoprotein bound to the every solid surface
is now constant.
Furthermore, excess enzyme responsible for transferring the
galactose from UDP-Gal to the N-glycan chain was washed away during the washing
procedure of the assay. For the following works, 2.0 mU/ml mammalian GalT was
selected in order to ensure sufficient amount of the enzyme in the reaction. Also, the
substrate donor will always be in the excess amount for the in vitro galactosylation
process.
96
0.30
Binding(450 nm)
0.25
0.20
0.15
0.10
0.05
0.00
24
48
72
96
Time of Infection (hours)
120
Figure 4.26: Galβ1→4GlcNAc linkage binding values at 450nm for the time
course upon coinfection between recombinant baculovirus hTf and β1,4-GalT.
Supernatants from each of the coexpressed cell culture were collected at hour
24, 48, 72, 96 and 120 respectively. Error bars indicate ±S.D of duplicates data.
97
0.35
0.30
Binding(450 nm)
0.25
0.20
0.15
0.10
0.05
0.00
0.0
1.0
2.0
3.0
4.0
β 1,4-GalT (mU/ml)
Figure 4.27: Effect of the mammalian GalT on the rate of in vitro
galactosylation process. Commercial mammalian GalT and UDP-Gal were
introduced to the hTf after it was secreted from Sf-9 cell culture. Different
concentrations enzyme were added to the hTf in the presence of excess amount
of sugar nucleotide, 0.32 mM. Lectin RCA-1 was later added to recognize the
Galβ1,4→GlcNAc linkage on the N-glycan of hTf, followed by TMB color
reaction measurement. Error bars indicate ±S.D of duplicates data.
98
To represent different levels of galactosylation, several conditions were
investigated which include positive controls, negative controls, in vivo and in vitro
galactosylation as shown in Figure 4.28. The positive control in this experiment was
the commercial apo hTf with two N-glycosylation sites which include galactose
residues at each branch. As for the negative controls, there were the Sf-9 cell culture
infected with AcMNPV-β1,4-GalT, uninfected Sf-9 insect cell culture as well as 2.0
mU/ml commercial mammalian GalT and 0.32 mM commercial UDP-Gal which
were added to harvested uninfected Sf-9 cell culture. For the samples, two Sf-9
insect cell cultures were infected with AcMNPV-hTf. One of the cultures was
coinfected with AcMNPV-β1,4-GalT (in vivo). For the in vitro analysis, 2.0 mU/ml
commercial mammalian GalT and 0.32 mM commercial UDP-Gal were added to the
harvested AcMNPV-hTf supernatant. For this part of the study, all cell cultures were
harvested at time 24 hour PI. The concentration of glycoprotein absorbed onto the
ELISA plate was constant, which was 1 µg/ml. A lectin, peroxidase labeled-RCA 1,
was used to interact with the Galβ1,4GlcNAc-linkage on the N-glycan of
glycoprotein, which can be measured by the TMB color reaction.
As observed in Figure 4.28, the various data for the conditions described as
above represent different level of glycosylation. Apo hTf was used as a guide for the
comparison with other conditions due to its N-glycan oligosaccharide chains with
Galactose residues. AcMNPV-hTf produced in insect cell culture did not achieved
satisfactory galactosylation, as expected, because it was missing one important
element that was the enzyme to construct the N-glycan chain.
However, this
phenomenon took a positive turn once the AcMNPV-hTf was coinfected with the
AcMNPV-β1,4-GalT. The absorbance value for the in vivo coexpression was 12.5
times higher compared to the AcMNPV-hTf culture. This result showed that the
success of galactosylation did not depend on the presence of substrate acceptor and
substrate donor only, but strongly also on the enzyme needed to elongate the chains.
As mentioned in section 4.2.4 regarding the native UDP-Gal level upon baculovirus
infection, it was found that the sugar nucleotide content decreased upon infection.
Thus, it was important to consider whether the reduction of UDP-Gal will have any
effect on the galactosylation. Hence, in vitro galactosylation was studied, where
harvested AcMNPV-hTf supernatant at time 24 hour PI was introduced with GalT
99
and UDP-Gal artificially. Surprisingly, the absorbance for the in vitro culture (0.300.35) was higher compared to the in vivo culture (0.22-0.29). This observation
indicated that in addition to GalT, sufficient amount of sugar nucleotide was another
key factor in guaranteeing the success of the galactosylation process. As for the
negative controls, they were showed to be insignificantly galactosylated.
For the galactosylation process to successfully occur, it needs the cooperation
of the substrate acceptor, substrate donor and enzyme. The relationships among the
three main elements of the in vivo galactosylation process, which were hTf as the
substrate acceptor, β1,4-GalT as the enzyme and UDP-Gal as the substrate donor is
illustrated in Figure 4.29.
Sf-9 cell culture infected with AcMNPV-hTf and
coinfected with the AcMNPV-β1,4-GalT was used to demonstrate the relationship.
The supernatants and lysates were collected at every 24 h intervals. The hTf and
β1,4-GalT expression was determined using ELISA and lectin binding assay as
described in section 3.6.3 and 3.7.2. Meanwhile, the UDP-Gal monitoring was
performed using RP-HPLC analysis as described in section 3.8. As expected, UDPGal concentration decreased gradually once the Sf-9 cells were coinfected with hTf
and β1,4-GalT. The pattern of the curve showed the same trend as Figure 4.24, even
though the cell culture was coinfected with hTf and β1,4-GalT. Figure 4.29 shows
that β1,4-GalT and hTf accumulation rates increased proportional to the time of
infection. However, not all hTf was galactosylated due to the limitation of UDP-Gal
(refer to Fig. 4.26). Thus, it can be concluded that even though hTf and β1,4-GalT
accumulation increased upon the time of coinfection, the gradual decrease of sugar
nucleotide (UDP-Gal) still affect the effectiveness of the galactosylation process.
100
0.50
0.45
0.40
Binding (450nm)
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
A
B
C
D
Group
E
F
G
Figure 4.28: Galβ1→4GlcNAc linkage binding values at 450 nm for the
different levels of galactosylation process. A: Apo human transferrin
(Standard); B: AcMNPV-rhTf supernatant harvested at 24h PI; C:
Coexpression of rhTf and β1,4-GalT supernatant harvested at 24h PI (in vivo);
D: Introduction of artificial GalT and UDP-Gal to harvested AcMNPV-rhTf
supernatant at 24h PI (in vitro); E: Uninfected Sf-9 cell culture; F: Introduction
of artificial GalT and UDP-Gal to uninfected Sf-9 cell culture; G: AcMNPVβ1,4-GalT supernatant harvested at 24 h PI. Error bars indicate ±S.D of
duplicates data.
101
18
50
UDP-gal
rGalT
rhTf
16
45
UDP-Gal Conc. (µ Molar)
35
12
30
10
25
8
20
6
15
4
10
2
5
0
0
0
24
48
72
96
Recombinant Protein Production (ug/ml)
40
14
120
Time of Infection (Hours)
Figure 4.29: Relationships among the three main elements in in vivo
galactosylation process. hTf as the sugar acceptor, β1,4-GalT as the enzyme and
UDP-Gal as the substrate donor in the process. Error bars indicate ±S.D of
duplicates data.
102
CHAPTER 5
CONCLUSIONS
5.1
Conclusion
The baculovirus expression vector system (BEVS) is widely used for the
production of recombinant glycoproteins, but it is not ideal for pharmaceutical
glycoprotein production due to the characteristics of the N-glycans in the expressed
products. Insect cells lack several enzymes required for mammalian-type N-glycan
synthesis and contain a specific N-acetylglucosaminidase that stunts the growth of
chains and a core α-1,3-fucosyltransferase that yields potentially allergenic
glycoforms.
In this study, we developed a method to produce galactosylated
recombinant human transferrin and optimized the expression of recombinant human
transferrin with better N-glycan quality.
A good inoculum is a prerequisite for successful cell growth and cell
infection with wild type and recombinant baculoviruses. Thus, in the first part of this
study, parameters which optimize the growth rate of Sf-9 cells culture were
investigated. The parameters investigated were the effects of serum, different types
of media, initial cell density, cell subculturing conditions as well as spent medium
103
carry-over. Serum affected viable cell numbers positively. However, since serum
contained trace amount of sugar nucleotides and enzymes which may interfere with
protein assay, serum free media was used for the rest of of the experiments. In this
study, SF-900II SFM was found to support cell growth better than TC-100. In
addition, high concentration of inoculum, subculturing at early exponential phase and
fresh medium without spent medium carry-over resulted in an insect culture with
high viable cell numbers and fast growth rate. For the mock baculovirus infection,
the interaction of the infection factors especially multiplicities of infection (MOI)
and spent medium carry-over with the above parameters were also investigated. In
order to achieve higher viral infectivity, the MOI range should be within the range of
1 to 15. Furthermore, the medium must also be replenished during the exponential
phase before viral infection.
Three main elements to ensure successful protein galactosylation are the
presence of sufficient amount of hTf as the substrate acceptor, β1,4-GalT as the
enzyme and UDP-Gal as the substrate donor. Unfortunately, the limitation in the
elongation of the N-glycan processing of hTf in insect cells occurs due to the lack of
β1,4-GalT needed to produce the galactosylated hTf. Thus, in this current study, a
proposed strategy for the production of galactosylated hTf is the introduction of GalT
artificially to the cell cultures infected with AcMNPV-hTf.
This can be
accomplished through in vivo or in vitro manners.
Analysis of secreted AcMNPV-hTf and AcMNPV-β1,4-GalT expression had
showed that their production rates increased over the time of infection. These were
confirmed by numerous analyses including SDS-PAGE, western blot, TLC, ELISA
and lectin binding assay. A simple explaination is the nature of baculovirus infection
cycle itself.
Upon infection, the cells’ mechanism will be shifted to viral
multiplication and expression of its genes. Hence, the recombinant protein secretion
will increase upon time of infection and will be secreted into the environment. To
examine another element involved in galactosylation processing, native UDP-Gal
level at normal and upon AcMNPV-hTf infection had been monitored using RP-
104
HPLC.
It revealed that substrate donor concentration decreased upon time of
infection.
After the three elements’ expression and monitoring analyses were
successfully established, the next step was to perform different levels of
galactosylation. Apo hTf containing two N-glycosylation sites that included Gal
residues was used as a standard for comparison with others. Since AcMNPV-hTf
produced in insect cell culture was not satisfactorily galactosylated due to the
deficiency of the enzyme to construct the N-glycan chain, a strategy involving the
introduction of artificial enzyme was investigated. To examine this strategy, in vivo
galactosylation was conducted using coexpression of AcMNPV-hTf and AcMNPVβ1,4-GalT in cultured cell. Also, in vitro study was carried out by the introduction of
commercial GalT and UDP-Gal to the harvested AcMNPV-hTf supernatant.
Although coexpression of AcMNPV-hTf and AcMNPV-β1,4-GalT resulted in
galactosylated recombinant hTf, the reduction of UDP-Gal upon infection still limit
the extent of galactosylation process. However, this is different for the in vitro
galactosylation. The commercial UDP-Gal was able to provide sufficient amount of
sugar nucleotide in the processing pathway.
The relationships among the three main elements in in vivo galactosylation
process are found to be very interesting.
As expected, UDP-Gal concentration
decreased gradually once the Sf-9 cells were coinfected with the baculovirus coding
the genes for hTf and β1,4-GalT. The hTf accumulation rate increased proportional
to the time of infection, but not all of the hTf were galactosylated due to the
limitation of UDP-Gal. This was proven by the time course analysis of UDP-Gal
upon coexpression of hTf and β1,4-GalT. The conclusion from this study is that
even though the model protein hTf and enzyme β1,4-GalT accumulation increased
upon the time of coinfection, the gradual decrease of sugar nucleotide still affect the
effectiveness of the galactosylation process.
105
5.1
Further Studies
In order to produce recombinant glycoprotein in insect cells with a more
“humanised” form, there are several recommendations for further studies:
(a)
N-glycans of the insect cells may be improved using in vitro
glycosylation, which utilizes the specific enzyme to transfer the sugar
to the protein after it is secreted from cell culture.
(b)
More sensitive, high-throughput, and detailed analytical techniques
for the detection of enzyme activities, glycan structures, donor sugar
nucleotides, intracellular metabolites and extracellular structures need
to be established.
(c)
Engineering of N-glycan processing pathway by means of genetic
manipulation to include the necessary processing enzymes.
(d)
Use of an alternative insect cell line that may contain mammalian-like
N-glycan processing capabilities.
106
REFERENCES
Abdul-Rahman B., Ailor E., Jarvis D. L., Betenbaugh M. J., Lee Y. C. (2002). β-1,4galactosyltransferase activity in native and engineered insect cells measured
with time-resolved Eu-fluorescence. Carbohydrate Research. 337: 21812186.
Ailor E., Takahashi N., Tsukamoto Y., Masuda K., Rahman, B.A., Jarvis D.L., Lee
Y.C., and Betenbaugh M. J. (2000). N-glycan patterns of human transferrin
produced
in
Trichoplusia
ni
insect
cells:
effects
of
mammalian
galactosyltransferase. Glycobiology. 10: 837-847.
Aisen P. (1989). Physical biochemistry of the transferrin: update, 1984-1988. In
Lochr, T.M. (ed.), Iron Carriers and Iron Proteins. VCH Publishers, New
York, pp. 353-371.
Altman F., Schwihla H., Staudacher E., Glossl J., Marz L. (1995). Insect cells
contain an unusual, membrane-bound beta-N-acetylglucosaminidase probably
involved in the processing of protein N-glycans. J. Biol. Chem. 270: 1734417349.
Altmann F., Kornfeld G., Dalik T., Staudacher E., and Glossl J. (1993). Processing of
asparagine-linked
oligosaccharides
in
insect
cells.
N-
acetylglucosaminyltransferase and II activities in cultured lepidopteran cells.
Glycobiology. 3: 619-625.
Altmann F., Marz L. (1995). Processing of asparagines-linked oligosacharides in
insect cells: evidence for α-mannosidase II. Glycoconjugate. 12: 150-155.
107
Altmann F., Staudacher E., Wilson I. B., Marz L. (1999). Insect cells as hosts for the
expression of recombinant glycoproteins. Glycoconjugate J. 16: 109-123.
Altmann F. (1996). N-glycosylation in insects revisited. Trends Glycosci. Glycotech.
8: 101-114.
Amado M., Almeida R., Schwientek T., and Clausen H. (1998). Identification and
characterization
of
large
galactosyltransferase
gene
families:
galactosyltransferases for all functions. Biochem. Biophys. Acta. 1473: 35-53.
Aoki D., Appert H. E., Johnson D., Wong S. S., and Fukuda M. N. (1990). Analysis
of the substrate binding sites of human galactosyltransferase by protein
engineering. EMBO. J. 9: 3171-3178.
Aumiller J. J., and Jarvis D.L. (2002). Expression and functional characterization of
a nucleotide sugar transporter from Drosophila melanogaster: relevance to
protein glycosylation in insect cell expression systems. Protein Expr. Purif.
26: 436-448.
Ayres M. D., Howard S. C., Kuzio J., Lopez-ferber M. and R. D. Possee. (1994). The
complete DNA sequences of Autographa californica nuclear polyhedrosis
virus. Virology. 202: 586-605.
Barrett A. J., Mavroudis D., Tisdale J., Molldrem J., Clave E., Dunbar C., CottlerFox M., Phang S., Carter C., Okunnieff P., Young N.S., Read E.J. (1998). T
cell-depleted bone marrow transplantation and delayed T cell add-back to
control acute GVHD and conserve a graft-versus-leukemia effect. Bone
Marrow Transplant. 21(6): 543-551.
Bartfeld N.S. and Law J.H. (1990). Isolation and molecular cloning of transferrin
from the tobacco hornworm, Manduca sexta. J.Biol. Chem. 265: 2168421691.
108
Berninsone P. M., and Hirachberg C.B. (2000). Nucleotide sugar transporters of the
Golgi apparatus. Curr. Opin. Struct. Biol. 10: 542-547.
Blissard G. W., and G. F. Rohrmann. (1990). Baculovirus diversity and molecular
biology. Annual review of Entomology. 53: 127-155.
Blissard G.W. (1996). Baculovirus--insect cell interactions. Cytotechnology. 20:7393.
Braunagel and Summers. (1994). Autographa californica nuclear polyhedrosis virus,
PDV, and ECV viral envelopes and nucleocapsids: structural proteins,
antigens, lipid and fatty acid profiles. Virology. 202(1) : 315-328.
Breitbach K., and Jarvis D.L. (2001). Improved glycosylation of a foreign protein by
Tn-5B1-4 cells engineered to express mammalian glycosyltransferases. J.
Biotechnol. Bioeng. 74: 230-239.
Brew Vanaman T.C., Hill R.L. (1968). The role of alpha-lactalbumin and the A
protein in lactose synthetase: a unique mechanism for the control of a
biological reaction. Proc Natl Acad Sci U S A. 59(2): 491-497.
Brodbeck U., Denton W. L., Tanahashi N., Ebner K. E. (1967). The isolation and
identification of the B protein of lactose synthetase as alpha-lactalbumin. J.
Biol Chem. 242(7): 1391-1397.
Brodbeck U. and Ebner K. E. (1966). The subcellular distribution of the A and B
proteins of lactose synthetase in bovine and rat mammary tissue. The Journal
of Biological Chemistry. 241: 5526-5532.
Bullen J. J., Ward C. G., and Rogers H. J. (1990). Iron, infection and the role of
bicarbonate. FEMS Microbiol. Lett. 71: 27-30.
Butters T. D. and Hughes R. C. (1981). Isolation and characterization of mosquito
cell membrane glycoproteins. Biochim, Biophys Acta. 640: 655-671.
109
Butters T. D., Hughes R. C., and Vischer P. (1981). Steps in the biosynthesis of
mosquito cell membrane glycoproteins and the effects of tumicamycin.
Biochim. Biophys. Acta. 640: 672-686.
Chen W., and Bahl O. P. (1991). Recombinant carbohydrate and selenomethionyl
variants on human choriogonadotropin. J. Biol. Chem. 266: 8192-8197.
D’Agostaro G., Bendiak B., and Tropak M. (1989). Eur. J.Biochem. 183: 211-217.
David R. O’Reilly, Lois K.Miller, Verne A. Luckow. (1994). Baculovirus expression
vectors: a laboratory manual. New York, Oxford, Oxford University Press.
David R. O’Reilly, Rosemary S.Hails, Thomas J. Kelly. (1998). The impact of host
developmental status on baculovirus replication. Journal of Invertebrate
Pathology. 72: 269-275.
Davidson D. J., Fraser M. J., and Castillino F. J. (1990). Oligosaccharide processing
in the expression of human plaminogen cDNA by lepidopteran insect
(Spodoptera frugiperda) cells. Biochemistry. 29: 5584-5590.
David T. R., Schuler M. L., Granados R. R., Wood H. A. (1993). Comparison of
oligosaccharide processing among various insect cell lines expressing a
secreted glycoprotein. In Vitro Cell Dev Biol Anim. 29A: 842-846.
Effertz K., Hinderlich S., Reutter W. (1999). Selective loss of either the epimerase or
kinase
activity
of
UDP-N-acetylglucosamine
2-epimerase/N-
acetylmannosamine kinase due to site-directed mutagenesis based on
sequence alighments. J. Biol Chem. 274: 28771-28778.
Fu D. and van Halbeek, H. (1992). N-glycosylation site mapping of human
serotransferrin by serial lectin affinity chromatography, fast atom
bombardment-mass spectrometry and
spectroscopy. Anal. Biochem. 206: 53-63.
1
H nuclear magnetic resonance
110
Fertig G., Kloppinger M. and Miltenburger H. G.. (1990). Cell cycle kinetics in
insect cell cultures compared to mammalian cell cultures. Exp. Cell. Res. 189:
208-212.
Gerardy-Schahn R., Oelmann S., and Bakker H. (2001). Biochimic (Paris). 83: 775762.
Grabenhorst E., Hofer B., Nimtz M., Jager V. and Conradt H. S. (1993). Biosynthesis
and secretion of human interleukin 2 glycoprotein variants from baculovirusinfected Sf-21 cells. Characterization of polypeptides and posttranslational
modifications. Eur J Biochem. 32: 11087-11099.
Guo S., Sato T., Shirane K., and Furukawa K. (2001). Galactosylation of N-linked
oligosaccharides by human ß-1,4-galactosyltransferases I, II, III, IV, V, and
VI expressed in Sf-9 cells. Glycobiology. 11: 813-820.
Hard K., van Dorn J. M., Thomas-Oates J. E., Kamerling J. P. and van der Horst D.J.
(1993). Structure of the ASN-linked oligosaccharides of apolipoprotein III
from the insect Locusta migraforia. Chemistry. 32: 766-775.
Hill R.L. (1979). Glycosyltransferases in analysis of oligosaccharide structure and
function. UCLA Forum Med. Sci. 21: 63-86.
Hinderlich S., Stasche R., Zeitier R., Reutter W. (1997). A bifunctional enzyme
catalyzes the first two steps in N-acetylneuraminic acid biosynthesis of rat
liver. Purification and characterization of UDP-N-acetylglucosamine 2epimerase/N acetylmannosamine kinase. J. Biol Chem. 272: 24313-24318.
Hirschberg C. B., Robbins P. W., and Abeijon C. (1998). Transporters of nucleotide
sugars, ATP and nucleotide sulfate in the endoplasmic reticulum and Golgi
apparatus. Annu. Rev. Biochem. 67: 49-69.
111
Hirschberg C. B. (2001). Golgi nucleotide sugar transport and leukocyte adhesion
deficiency II. J.Clin. Invest. 108: 3-6.
Hollister J. R., Shaper J. H., Jarvis D. L. (1998). Stable expression of mammalian
beta 1,4-galactosyltransferase extends the N-glycosylation pathway in insect
cells. Glycobiology. 8: 473-480.
Hollister J. R., Grabenhorst E., Nimtz M., Conradt H., and Jarvis D. L. (2002).
Engineering the protein N-glycosylation pathway in insect cells for
production of biantennary, complex N-glycans. Biochemistry. 41: 1509315104.
Hollister J. R. and Jarvis D. L. (2001). Engineering lepidopteran insect cells for
sialoglycoprotein production by genetic transformation with mammalian beta
1,4-galactosyltransferase and alpha 2,6-sialytransferase genes. Glycobiology.
11: 1-9.
Hooker A. D., Green N. H., Baines A. J., Bull A. T., Jenkins N., Strange P. G., James
D. C. (1999). Constraints on the transport and glycosylation of recombinant
IFN-gamma in Chinese hamster ovary and insect cells. Biotechnol. Bioeng.
63: 559-572.
Hsieh P. and Robbins P. W. (1984). Regulation of asparagines-linked
oligosaccharide processing. Oligosaccharide processing in Aedes albopitus
mosquito cells. J. Biol. Chem. 259: 2975-2382.
Hsu T. A., Takahashi N., Tsukamoto Y., Kato K., Shimada I., Masuda K., Whiteley
E. M., Fan J. Q., Lee Y. C., Betenbaugh M. J. (1997). Differential N-glycan
patterns of secreted and intracellular IgG produced in Trichoplusia ni cells. J
Biol Chem. 272(14): 9062-9070.
Ignoffo C. M. (1975). Entomopathogens as insecticides. Environ Lettert. 8(1): 23-40.
112
Jarvis D. L. and Summers M. D. (1989). Glycosylation and secretion of human tissue
plasminogen activator in recombinant baculovirus-infected insect cells. Mol
Cell Biol. 9: 214-223.
Jarvis D. L., Finn E. E. (1996). Modifying the insect cell N-glycosylation pathway
with immediate early baculovirus expression vectors. Nat. Biotechno.l. 14:
1288-1292.
Jarvis D. L. (2003). Developing baculovirus-insect cell expression systems for
humanized recombinant glycoprotein production. Virology. 310: 1-7.
Jarvis D. L. (1997). Baculovirus Expression Vectors. In Miller, L.K. (ed.). The
Baculoviruses. Plenum, New York, pp. 389-431.
Kawakita M., Ishida N., Miura N., Sun Wada G. H., and Yoshioka S. (1998). OAcetylserine and O-acetylhomoserine sulfhydrylase of yeast. Further
purification and characterization as a pyridoxal enzyme. J.Biochem. 123:
777-785.
Kawar Z., Herscovics A., Jarvis D. L. (1997). Isolation and characterization of an
alpha 1,2-mannosidase cDNA from the lepidopteran insect cell line Sf-9.
Glycobiology. 7: 433-443.
Kawar Z., Romero P. A., Herscovics A., Jarvis D. L. (2000). N-glycan processing by
a lepidopteran insect alpha 1,2-mannosidase. Glycobiology. 10: 347-355.
Kioukia N., Nienow A. W., Emery A.N., Al-Rubeai M.. (1995). Physiological and
environmental factors affecting the growth of insect cells and infection with
baculovirus. Journal of Biotechnology. 38: 243-251.
Keshvara L. M., Newton E. M., Good A. H., Hindsgaul O., and Palcic M. M. (1992).
Enzyme-Linked Immunosorbent Assays for the Measurement of Blood
Group A and B Glycosyltransferase Activities. Glycoconjugate J. 9: 16-20.
113
Keusch J., Lydyard P. M., Isenberg D. A., and Delves P.J. (1995). β1,4galactosyltransferase activity in B cells detected using a simple ELISA-based
assay Glycobiology. 5: 365-370.
Korth K. L., Levings C. S. (1993). Baculovirus expression of the maize
mitochondrial protein URF13 confers insecticidal activity in cell cultures and
larvae. Proc Natl Acad Sci USA. 90: 3388-3392.
Kubelka V., Altmann F., Kornfeld G. and Marz L. (1994). Structures of the N-linked
oligosaccharides of the membrane glycoproteins from three lepidoptran cell
lines (Sf-21, IZD-Mb-0503, Bm-N). Arch. Biochem. Biophysc. 308: 148-157.
Kulakosky P. C., Hughes P. R. and Wood H. A. (1998). N-linked glycosylation of a
baculovirus-expressed recombinant glycoprotein in insect larvae and tissue
culture cells. Glycobiology. 8: 741-745.
Kumamoto K., Goto Y., Sekikawa K., Takenoshita S., Ishida N., Kawashita M., and
Kannagi R. (2001). Increased Expression of UDP-Galactose Transporter
Messenger RNA in Human Colon Cancer Tissues and Its Implication in
Synthesis of Thomsen-Friedenreich Antigen and Sialyl Lewis A/X
Determinants. Cancer Res. 61: 4620-4627.
Kuroda K., Geyer R., Doergler W. and Klenk H. D. (1990). The oligosaccharides of
influenza virus hemmaglutinin expressed in insect cells by a baculovirus
vector. Virology. 174: 418-428.
Kool M., Vlak J. M. (1993). The structural and functional organization of the
Autographa californica nuclear polyhedrosis virus genome. Arch Virol.
130(1-2): 1-16.
Lawrence S. M., Huddleston K. A., Tomiya N., Nguyen N., Lee Y. C., Vann W. F.,
Coleman T. A., Betenbaugh M. J. (2001). Cloning and expression of human
sialic acid pathway genes to generate CMP-sialic acids in insect cells.
Glycoconjugate J. 18: 205-213.
114
Lawrence S. M., Huddleston K. A., Pitts L. R., Nguyen N., Lee Y. C., Vann W. F.,
Coleman T. A., Betenbaugh M. J. (2000).
Cloning and expression of the
human N- acetylneuraminic acid phosphate synthase gene with 2-keto-3deoxy-D-glycero-D galacto-nononic acid biosynthetic ability. J Biol Chem.
275: 17869-17877.
Leger D., Campion B., Decottignies J. P., Montreuil J. and Spik G. (1989).
Physiological significance of the marked increased branching of the glycans
of human serotransferrin during pregnancy. Biochem J. 257: 231-238.
Licari P. J., Jarvis D. L., Bailey J. E. (1993). Insect cell hosts for the baculovirus
expression vectors contain endogeneous exoglycosidase activity. Biotechnol
Prog. 9: 146-152.
Licari R. and Bailey J. E. (1991). Factors influencing recombinant protein yields in
an insect cell-baculovirus expression system: multiplicity of infection and
intracellular protein degradation. Biotech. Bioeng. 37: 238-246.
Lo N. W., Shaper J. H., Pevsner J., and Shaper N. (1998). The expanding beta 4galactosyltransferase
gene
family:
messages
from
the
databanks.
Glycobiology. 8: 517-526
Lokesh J., Michael L. S., and H.Alan Wood. (2001). Production of a sialylated Nlinked glycoprotein in insect cells. Biotechnol Prog. 17: 822-827.
Lopez M., Coddeville B., Langridge J., Plancke Y., Sautiere P., Chaabihi H., Chirat
F., Harduin-Lepers A., Cerutti M., Verbert A. and Delannoy P. (1997).
Microheterogeneity of the oligosaccharides carried by the recombinant
bovine lactoferrin expressed in Mamestra brassicae cells. Glycobiology.
7:635-651.
Lopez M., Tetaert D., Juliant S., Gazon M., Cerutti M., Verbert A., Delannony P.
(1999). O-glycosylation potential of lepidopteran insect cell lines. Biochim
Biophys. Acta. 1427: 49-61.
115
Luckow V. A. (1995). In Baculovirus Expression Systems and Biopesticides (Shuler
ML, Wood HA, Granados RR, Hammer DA, eds) pp 51-90. New York: John
Wiley Sons.
Luis Maranga, Ana S. Coroadinha, and Manuel J. T. Carrondo. (2002). Insect cell
culture medium supplementation with fetal bovine serum and bovine serum
albumin: effects on baculovirus adsorption and infection kinetics. Biotechnol
Prog. 18: 855-861.
MacGillivray R. T. A., Mendez E., Sinha S. K., Shewale G., Lineback-Zins J. and
Brew, K. (1983). The primary structure of human serum transferrin. J.Biol.
Chem. 258: 3543-3553.
MacGillivray R. T. A., Mendez E., Sinha S. K., Sutton M. R., Lineback-Zins J. and
Brew, K. (1982). The complete amino acid sequence of human serum
transferrin. Proc. Natl. Acad. Sci. U.S.A. 79: 2504-2508.
Maeda S., Kawai T., Obinata M., Fujiwara H., Horiuchi T., Saeki Y., Sato Y.,
Furusawa M. (1985). Production of human alpha-interferon in silkworm
using a baculovirus. Nature. 315: 592-594.
Maiorella B., Inlow D., Shauger A. and Harano D. (1988). Large scale insect cell
culture for recombinant protein production. Biotechnology. 6: 1406-1410.
Manneberg M., Friedlein A., Kurth H., Lahm H. W. and Fountoulakis M. (1994).
Structural analysis and localization of the carbohydrate moieties of a soluble
human interferon gamma receptor produced in baculovirus-infected insect
cells. Protein Sci. 3: 30-38.
Marchal I., Jarvis D. L., Cacan R., Verbert A. (2001). Glycoproteins from insect
cells: sialylated or not? Biol. Chem. 382: 151-159.
Marchal I., Mir A. M., Kmiecik D., Verbert A., and Cacan R. (1999). Use of
inhibitors to characterize intermediates in the processing of N-glycans
116
syntherized by insect cells: a metabolic study with Sf-9 cell line.
Glycobiology. 9: 645-654.
Marz L., Altmann F., Staudacher E., Kubelka V. (1995). Protein glycosylation in
insects, in Montreuil J, Vliegenthart J F G, Schachter H (Eds.),
Glycoproteins, Vol 29a. Elsevier, Amsterdam, pp 543-563.
Montreuil. J., Spik G., and Mazurier J. (1997). Transferrin superfamily. In Montreuil,
J., Vliegenthart, J.F.G. and Schacter H. (eds.) Glycoproteins II. Amsterdam:
Elsevier. pp 203-242
Munk K., Pritzer E., Kretzschmar E., Gutte B., Garten W. and Klenk H. D. (1992).
Carbohydrate masking of an antigenic epitope of influenza virus
haemagglutinin independent of oligosaccharide size. Glycobiology. 2: 233240.
Narimatsu H., Sinha S., Brew K., Okayama H., and Qasba P.K. (1986). Cloning and
sequencing
of
cDNA
of
bovine
N-acetylglucosamine
(ß1-4)
galactosyltransferase. Proc Natl. Acad. Sci. USA. 83: 4720-4724.
Ogonah O. W., Freedman R. B., Jenkins N. and Rooney B. C. (1995). Analysis of
human interferon-gamma glycoforms produced in baculovirus infected insect
cells by matrix assisted laser desorption spectrometry. Biochem Soc Trans.
23: 100S.
Ogonash O. W., Freedman R. B., Jenkins N., Kirit P., and Rooney B. C. (1996).
Isolation and characterization of an insect cell line to perform complex Nlinked glycosylation on recombinant proteins. Bio/techniques. 14: 197-202.
Oubihi M., Kitajima K., Kobayashi K., Adachi T., Aoki N., Matsuda T. (1998).
Development of an enzyme-linked immunosorbent assay-based method for
measuring galactosyltransferase activity using a synthetic glycopolymer
acceptor substrate. Anal Biochem. 257(2):169-175.
117
Paulson J. C., and Colley K. J. (1989). Glycosyltransferases. Structure, localization,
and control of cell type-specific glycosylation. J. Biol Chem. 264: 1761517618.
Reis U., Blum B., Von Specht B. U., Domdey H., Collins J. (1992). Antibody
production in silkworm cells and silkworm larvae infected with a dual
recombinant Bombyx mori nuclear polyhedrosis virus. Biotechnology. 10:
910-912.
Ren J., Bretthuer R. K., and Castellino F. J. (1995). Purification and properties of a
Golgi-derived
(alpha
1,2)-mannosidase
I
from
baculovirus-infected
lepidopteran insect cells (IPLB-SF21AE) with preferential activity toward
mannose 6-N-acetylglucosamine 2. Biochemistry. 34: 2489-2495.
Roseman S. (2001). Reflections on glycobiology. J.Biol. Chem. 276: 41527-41542.
Ryll T., and Wagner R. (1991). Improved ion-pair high-performance liquid
chromatographic method for the quantification of a wide variety of
nucleotides and sugar-nucleotides in animal cells. J. Chromatography. 570:
77-88.
Schorp B, Howaldt M. W. and Bailey J. E. (1990). DNA distribution and respiratory
activity of Spodoptera frugiperda population infected with wild-type and
recombinant AcMNPV. J.Biotechnol. 15: 169-185.
Shaper N., Shaper J. H., Meuth J. L., Fox J. L., Chang H., Kirsch I. R., and Hollis G.
F. (1986). Bovine Galactosyltransferase: Identification of a Clone by Direct
Immunological Screening of a cDNA Expression Library. Proc Natl. Acad.
Sci. USA. 83: 1573-1577.
Smith G. E., Summers M. D., Fraser M. J. (1983). Production of human beta
interferon in insect cells infected with a baculovirus expression vector. Mol
Cell Biol. 3(12): 2156-2165.
118
Staudacher E. and Marz L. (1998). Strict order of (Fuc to Asn-linked GlcNAc)
fucosyltransferases forming core-difucosylated structures. Glycoconjugate J.
15: 355-360.
Stollar V., Stollar B. D., Koo R., Harrap K. A., and Schlesinger R. W. (1976). Sialic
acid contents of sindbis virus from vertebrate and mosquito cells.
Equivalence of biological and immunological viral properties. Virology. 69:
104-115.
Stult C. L. M., and Macher B.A. (1990). Measurement of beta-Galactosyltransferase
activity in Cell extracts with an ELISA-based assay. Arch. Biochem. Biophys.
280: 20-26.
Summers M. D., Smith G. E. (1987). A manual method for baculovirus vectors and
insect cell culture procedures. Texas Agricultural Experiment Station Bulletin
No. 1555. Texas A & M University, College Station, Texas.
Takeuchi M. and Kobata A. (1991). Structures and functional roles of the sugar
chains of human erythropoietins. Glycobiology. 1: 337-346.
Takeuchi M., Takasaki S., Shimada M. and Kobata A. (1990). Role of sugar chains
in the in vitro biological activity of human erythropoietin produced in
recombinant Chinese hamster ovary cells. Journal of Biological Chemistry.
265: 12127-12130.
Taki T., Nishikawa S., Ishii K., and Handa. (1990). A simple and specific assay of
glycosyltransferase and
immunosorbent
assay
glycosidase activities by an enzyme-linked
method,
and
its
application
to
assay
of
galactosyltransferase activity in sera from patients with cancer. J. Biochem.
107: 493-498.
Thorstensen K. and Romslo I. (1990). The role of transferrin in the mechanism of
cellular iron uptake. Biochem. J. 271: 1-10.
119
Tomiya N., Ailor E., Shawn M. Lawrence, Betenbaugh M. J., Yuan C. L. (2001).
Determination of nucleotides and sugar nucleotides involded in protein
glycosylation by high-performance anion-exchange chromatography: sugar
nucleotide contents in cultured insect cells and mammalian cells. Analytical
Biochemistry. 293: 129-137.
Tomiya N., Betenbaugh M. J. and Yuan C. L. (2003). Humanization of Lepidopteran
insect cell produced glycoproteins. Acc. Chem. Res. 36: 613-620.
Tomiya N., Howe D., Aumiller J. J., Pathak M., Park J., Palter K., Jarvis D. L.,
Betenbaugh M. J., Yuan C. L. (2003). Complex type biantennary N-glycans
of recombinant human transferrin from Trichoplusia ni insect cells expression
mammalian
beta1,4-galactosyltransferase
and
beta1,2-N
acetylglucosaminyltransferase II. Glycobiology. 13: 23-34.
Trowbridge I. S., Newman R. A., Domingo D. L. and Sauvage C. (1984). Transferrin
receptors: structure and function. Biochem. Pharmacol. 33: 925-932.
van Die I., van Tetering A., Bakker H., van den Eijnden D. H., and Joziasse D. H.
(1996). Glycosylation in lepidopteran insect cell: identification of a β1,4-Nacetylgalactosaminyltransferase involved in the synthesis of complex-type
oligosaccharide chains. Glycobiology. 6: 157-164.
Velardo M. A., Bretthauer R. K., Boutaud A., Rinhold B., Reinhold V. N., Castellino
F. J. (1993). The presence of UDP-N-acetylglucosamine: Alpha-3-Dmannoside beta 1,2-N-acetylglucosaminyltransferase I activity in Spodoptera
frugiperda cells (IPLB-Sf-21AE) and its enhancement as a result of
baculovirus infection. J.Bio Chem. 268: 17902-17907.
Vlak J. M. and G. F. Rohrmann. (1985). Baculovirus polyhedrins. In: viral and
microbial pesticides (KE Sherman & K Maramorosch) Academic Press, N Y.
489-542.
120
Vlak J. M. (1997). Baculoviruses and insect cells: a versatile system for the
production of recombinant proteins including antibodies. Science &
Technology International Courses, Leiden, The Netherlands. 1-12.
Volkman L. E. (1997). Nucleopolyhedrovirus interactions with their insect hosts.
Adv Virus Res. 48: 313-348.
Wagner R., Geyer H., Geyer R., Klenk H. D. (1996). N-acetyl-beta-glucosaminidase
accounts for differences in glycosylation of influenza virus hemagglutinin
expressed in insect cells from a baculovirus vector. J. Virology. 70: 41034109.
Watanabe A., Takehiro Kokuho, Hitomi Takahashi, Masashi Takahashi, Takayuki
Kubota and Shigeki Inumaru. (2001). Sialylation of N-glycans on the
recombinant proeins expressed by a baculovirus-insect cell system under β-N
-acetylglucosaminidase inhibition. The Journal of Biological Chemistry. 277:
5090 -5093.
Wathen M. W., Aeed P. A. and Elhammer A. P. (1991). Characterization of
oligosaccharide structures on a chimeric respiratory syncytial virus protein
expressed in insect cell line Sf-9. Biochemistry. 30. 2863-2868.
Yadav S. P., and Brew K. (1990). Identification of a region of UDP-galactose: Nacetylglucosamine beta 4-galactosyltransferase involved in UDP-galactose
binding by differential labeling. J.Biol. Chem. 265: 14163-14169.
Yeh J. C., Seals J. R., Murphy C. I., Van Halbeek H. and Cummings R. D. (1993).
Site-specific N-glycosylation and oligosaccharide structures of recombinant
HIV-1 gp120 derived from a baculovirus expression system. Biochemistry.
32: 11087-11099.
Zatta P. F., Nyame K., Cormier M. J., Mattox S. A., Smith D. F., and Cummings R.
D. (1991). A Solid-Phase Assay for b1,4 Galactosyltransferase in Human
Serum Using Recombinant Aequorin. Anal. Biochem. 194: 185-191.
121
Zuidema D., Klinge-Roode E. C., van Lent J. W., Vlak J. M. (1989). Construction
and analysis of an Autographa californica nuclear polyhedrosis virus mutant
lacking the polyhedral envelope. Virology. 173(1): 98-108.
122
APPENDIX A
APPENDIX A-1
Monosaccharide Mass and Structure
Fucose, Fuc
C6O5H12
164.0685 / 164.2
Galactose, Gal
C6O6H12
146.0579 / 146.1
180.0634 / 180.2
Mannose, Man
C6O6H12
180.0634 / 180.2
162.0528 / 162.1
162.0528 / 162.1
Sialic Acid,
NANA
C11O9NH19
N-Acetylgalactosamine, N-Acetylglucosamine,
GalNAc
GlcNAc
C8O6NH15
C8O6NH15
309.1060 / 309.3
221.0899 / 221.2
221.0899 / 221.2
291.0954 / 291.3
203.0794 / 203.2
203.0794 / 203.2
Key to masses shown above
Fucose monoisotopic average
intact
residue
164.0685
164.2
146.0579
146.1
123 123
APPENDIX A-2
.
Common N-Linked Glycan Simplified Structures and Masses
complex
sialylated
fucosylated
diantennary
complex
sialylated
fucosylated
triantennary
complex
high mannose
sialylated
(man 7)
fucosylated
tetraantennary
hybrid
common
core str.
subunit
sialylated
branch
subunit
branch
subunit
N-Linked
Structures
Composition C90O65N6H146 C115O83N8H186 C140O101N10H226 C64O49N2H106
Mono. Mass
2350.8303
3007.0579
3663.2856
1686.5864
Av. Mass
2352.13
3008.72
3665.30
1687.51
C83O62N4H136 C34O25N2H56 C25O18N2H40 C14O10N1H23
2180.7612
892.3172
656.2276
365.1322
2181.96
892.81
656.59
365.33
Symbol key
Residue
Symbol Residue Composition Mono Isotopic Mass Average Mass
Sialic Acid
C11O8NH17
291.0954
291.26
Galactose
C6O5H10
162.0528
162.14
N-Acetylglucosamine
Mannose
Fucose
C8O5NH13
C6O5H10
C6O4H10
203.0794
162.0528
146.0579
203.19
162.14
146.14
124
APPENDIX A-3
Cell Culture Glossary
Floaters
Definition:
Note:
Cells that are either loosely attached or suspended in the medium.
Floaters are normal occurrence and are often seen in older cultures
and cultures, which have, overgrown.
If floaters constitute more than 5% of the culture, remove the old
medium containing the floaters and replace with fresh medium before
subculturing.
Sloughing
Definition:
Note:
To dislodge cells from a surface by streaming medium over them.
This subculturing method is very gentle and results in high cell
viabilities. We typically use this method to dislodge adherent cell
cultures.
Doubling Time
Definition:
Note:
Time using for cells for double replicating.
Population doubling times for insect cells will vary depending on
growth conditions.
Healthy Doubling Times: Sf9 cells double every 72 hours
Viability
Definition:
Note:
Cell viability refers to the percent of cells in a culture that are living.
Cell viability is determined by treating the cells with Trypan blue.
Trypan blue dye molecules are excluded from viable cell membranes,
but readily enter non-viable cells. Cells that are blue are dead.
Cell viability should be at least 95% for healthy log-phase cultures.
Cells below 95% viability are not growing under optimal conditions
and should not used in experiments.
125
Passaging/ Subculturing
Definition:
Note:
Diluting cells back to a density that maintains log phase growth and
maximum viability.
Adherent Cultures:
Adherent cultures should be passaged at confluency and are typically
diluted at a 1:5 dilution (volume of cells: final volume of medium) in
order to maintain log phase growth.
Suspension Cultures:
Suspension cultures should be passaged before they reach a density of
2.0 to 2.5x 106 cells/ml and diluted back to 0.7 to 1.0 x 106 cells/ml.
Confluency
Definition:
A confluent monolayer is an adherent cell culture (dish, plate, flask)
in which the cells have formed a single layer over the entire surface
area available for growth. Once the cells have started to form clusters
above the first layer or have started to lift up from surface, the culture
is past confluency.
Note:
Passaging past confluency:
Cells that are repeatedly passaged at densities past confluency display
decreased doubling times, decreased viabilities, and a decreased
ability to attach. The culture is considered to be unhealthy.
Passaging before confluency:
Cell cultures that have not reached confluency are move difficult to
dislodge, and require more mechanical force to dislodge them from
the monolayer. When repeatedly subcultured before confluency, cells
display decreased doubling times and decreased viabilities. The
culture is considered to be unhealthy.
126
APPENDIX B
APPENDIX B-1
1.
2.
3.
4.
5.
Stock Solution for SDS-PAGE
2M Tris-HCl (pH8.8), 100ml
Weight out 24.2g Tris-base and add to 50ml distilled water.
Add HCl slowly to pH 8.8
Add distilled water to total volume 100ml.
1M Tris-HCl (pH 6.8), 100ml
Weight out 12.1g Tris base and add to 50ml distilled water.
Add HCl slowly to pH 6.8.
Add distilled water to total volume 100ml.
10% SDS(w/v), 100ml
Weight out 10g SDS
Add distilled water to a total volume 100ml.
50% glycerol (v/v), 100ml
Pour 50ml 100% glycerol
Add 50ml distilled water.
1% bromophenol blue (w/v),10ml
Weight out 100mg bromophenol blue
Bring to 10ml with distilled water, stir until dissolved.
127
APPENDIX B-2
1.
Working Solution for SDS-PAGE
Solution A (Acrylamide Stock Solution), 100ml
30% (w,v) acrylamide, 0.8% (w/v) bis-acrylamide
Weight out 29.2g acrylamide and 0.8g bis-acrylamide and make total
volume to 100ml.
2.
3.
4.
Solution B (4x separating gel buffer), 100ml
75ml 2M Tris-HCl (pH8.8)
4ml 10% SDS
21ml distilled water
Solution C (4x stacking gel buffer), 100ml
50ml 1M Tris-HCl (pH6.8)
4ml 10% SDS
46ml distilled water
10% ammonium persulfate
5.
6.
0.05 g in 0.5ml distilled water
Electrophoresis buffer, 1L
3g Tris
14.4g glycine
1g SDS
Add distilled water to make 1L.
5x sample buffer, 10ml
0.6ml 1M Tris-HCl (pH6.8)
5ml 50% glycerol
2ml 10% SDS
0.5ml 2-mercaptoethanol
1ml 1% bromophenol blue
0.9ml distilled water
128
APPENDIX B-3
1.
2.
Separating and Stacking Gel Preparation
Separating Gel X% Preparation (see note)
Solution A
x/3 ml
Solution B
2.5 ml
H2O
(7.5-x/3) ml
10% ammonium persulfate
50ul
TEMED
10ul
Stacking Gel Preparation
Solution A
0.67 ml
Solution C
1.0 ml
H2O
2.3 ml
10% ammonium persulfate
30ul
TEMED
10ul
Note: Optimal Resolution Ranges (adapted from Hames, B.D. pp 1-91 in Hames,
B. D. and D. rickwood, eds. 1981. Gel Electrophoresis of Proteins: a Practical
Approach. 290 pages. IRL Press, Oxford and Washington, D.C.)
Acrylamide Percentage
Separating Resolution
15 % Gel
15 – 45 kDa
12.5% Gel
15 – 60 kDa
10% Gel
18 – 75 kDa
7.5% Gel
30 – 120 kDa
5% Gel
60 – 212 kDa
129
APPENDIX C
Working Solution for ELISA
1.
Coating Buffer
0.05 M Sodium Carbonate, pH 9.6
2.
Washing Solution
50 mM Tris, 0.14 M NaCl, 0.05% Tween 20, pH 8.0
3.
Blocking (Postcoat) Solution
50 mM Tris, 0.14 M NaCl, 1% BSA, pH 8.0
4.
Sample/Conjugate Diluent
50 mM Tris, 0.14 M NaCl, 1% BSA, 0.05% Tween 20, pH 8.0
5.
Enzyme Substrate
TMB (3,3’,5,5’-tetramethylbenzidene)
6.
Stopping Solution
1 M Phosphoric acid
Note: All the working solution were purchased from Bethyl Laboratories Inc,
Texas, USA.
130
APPENDIX D
Working Solution for Western Blot
1.
Towbin Buffer
24 mM Tris, 192 mM glycine and 10% methanol
2.
Phosphate Buffer Saline 1X., pH 7.4
3.
Blocking Solution
5% skimmed milk in 1X PBS buffer
4.
Washing Solution
0.05% Tween 20 in 1X PBS buffer
5.
TMB (3,3’,5,5’-tetramethylbenzidene) Stabilized Substrate for HRP
(Promega, Madison, WI)
131
APPENDIX E
Virus Calculation
(i)
Initial Density = A cells/ml
From virus titer, pfu/ml = B
pfu/ml
(M1)
From definition, MOI = C = C pfu/cell
Concentration of virus needed = C pfu x A cells = D pfu / ml (M2)
cell
From equation M1V1=M2V2
M1
=B
pfu/ml
M2
=D
pfu / ml
V2
= E ml
V1
= F (volume of virus stock)
V1
= F µl of virus stock
ml
132
APPENDIX F
Reaction Mixture for Lactose Synthetase Assay
Mixture Volume, µl
100
Buffer
5 µmole of Tris-HCl, pH 7.4
UDP-Gal, (µmol)
0.040
Glucose, (µmol)
2.00
MnCl2, (µmol)
4.00
UTP, (µmol)
0.50
α-lactalbumin, (µmol)
0.14
GalT
5 µl
Reaction
Time (min)
20
Temperature (0C)
37
Stopping Solution
Add 0.2ml of 0.3 N Ba(OH)2, to the ice-cooled
mixture. This was neutralized with 1.5 volumes of a
5% solution of ZnSO4.7H2O and precipitate was
removed by centrifugation
Stock Solutions:
Chemicals
Molarity (mM)
Tris
50
UDP-Gal
4
Glucose
200
MnCl2
400
UTP
50
α-lactalbumin
1.4
ZnSO4.7H2O
5% solution
Ba(OH)2
0.3 N
Mass (gram) MW (g/mol) Volume (ml)
0.06060
121.14
10
0.00244
610.3
1
0.03604
180.2
1
0.07916
197.9
1
0.02407
484.1
1
0.01985
14176
1
0.50000
287.54
10
----
133
Working Solutions:
According to this the equation: M1V1 = M2V2
Chemicals
UDP-Gal
Glucose
MnCl2
UTP
α-lactalbumin
M1
Stock
Solutions
Molarity
(mM)
4
200
400
50
1.4
V1
M2
V2
Volume (µl)
Working
Solutions
Molarity (mM)
Volume (µl)
10
10
10
10
10
0.4
20
40
5
0.14
100
100
100
100
100
Example Calculation:
Concentration of UDP-GalT in the reaction mixture is
0.04µmol
100µl
= 0.0004 M
To prepare stock solution, M1V1 = M2V2
Where
M1 = Concentration of stock solution
V1 = Volume of stock solution to be added to the mixture (Decide by
own)
M2 = Concentration of working solution = 0.0004 M
V2 = Volume of the mixture = 100 µl
M1 (10 µl) = 0.0004 M (100 µl)
M1 = 0.004 M
Mass of UDP-Gal
=
0.004 mol
g
1L
× 610.3
× 1ml ×
L
mol
1000 ml
=
0.00244 gram in 1 ml of 0.05 M Tris-HCl, pH 7.4
134
Stock Solutions Preparations:
1. 0.05 M Tris-HCl, pH 7.4
Weigh out 0.0606g Tris base
Add 5ml distilled water
Add concentrated HCl slowly to pH 7.4.
Add distilled water to a total volume of 10ml
2. 4mM UDP-Gal
Weigh out 0.0024412 g of UDP-Gal
Add 0.05 M Tris-HCl, pH 7.4 to a total volume of 1 ml
3. 200 mM of Glucose (Freshly Prepare)
Weigh out 0.03604 g of glucose
Add 0.05 M Tris-HCl, pH 7.4 to a total volume of 1 ml
4. 400 mM of MnCl2
Weigh out 0.07916 g of MnCl2
Add 0.05 M Tris-HCl, pH 7.4 to a total volume of 1 ml
5. 50 mM of UTP
Weigh out 0.02407 g of UTP
Add 0.05 M Tris-HCl, pH 7.4 to a total volume of 1 ml
6. 1.4 mM of α-lactalbumin
Weigh out 0.019846 g of α-lactalbumin
Add 0.05 M Tris-HCl, pH 7.4 to a total volume of 1 ml
7. 5% solution of ZnSO4.7H2O
Weight out 0.50 g of ZnSO4.7H2O
Add distilled water to a total volume of 10 ml
135
APPENDIX G
Reaction Mixture for Lectin Binding Assay
(GalT, 10mM MnCl2, 0.1%BSA and 0.32 mM UDP-Gal) in 30mM MOPS (pH7.4))
Chemicals
UDP-Gal
BSA
MnCl2
M1
Stock
Solutions
Molarity (mM)
3.2
1%
100
V1
Volume (µl)
10
10
10
M2
Working
Solutions
Molarity (mM)
0.32
0.1%
10
V2
Volume (µl)
100
100
100
Dilution for 11.18 units/ml (5.2mg/ml) dengan 1000 x
→
10 µl
10ml MOPS buffer
(11.18 units/ml
GalT
Concentration
(mU/mL)
0
0.5
1.0
1.5
2.0
3.0
4.0
11.18 munits/ml)
VGalT
(µl)
VUDP-Gal
(µl)
VBSA
(µl)
VMnCl2
(µl)
VMOPS Total
(µl)
volume
0
4.5
9.0
13.0
18.0
27.0
36.0
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
70
66
61
57
52
43
34
100
100
100
100
100
100
100
136
APPENDIX H
Publication
Yap Wei Ney, Muhd Nazrul Hisam Zainal Alam, Siti Zalita Abdul Talib, Wong Fui
Ling, Badarulhisam Abdul Rahman, Azila Abdul Aziz and Firdausi Razali (2003).
Case Study: Large Scale Expression and Purification of His-tagged Recombinant
Protein in E.Coli Fermentation. Proceedings: The 17th Symposium of Malaysia
Chemical Engineers. 436-443.
Yap Wei Ney, Badarulhisam Abdul Rahman, Azila Abdul Aziz (2004). Cell Culture
Optimization for the baculovirus Expression Vector System (BEVS). 1st National
Postgraduate Colloquium. 295-299.
Yap Wei Ney, Badarulhisam Abdul Rahman, Azila Abdul Aziz (2005). Baculovirus
Infection Reduced UDP-Galactose Level in Spodoptera frugiperda Insect Cells.
Proceedings: 15th Scientic Meeting of Malaysian Society for Molecular Biology and
Biotechnology. 10-11.
Yap Wei Ney, Badarulhisam Abdul Rahman, Azila Abdul Aziz (2005). Baculovirus
Coinfection Strategy for Improved Galactosylation of Recombinant Glycoprotein
Produced By Insect Cell Culture. Proceedings: 2nd International Conference on
Chemical and Bioprocess Engineering. 137-142.