BAHAGIAN A – Pengesahan Kerjasama* Adalah disahkan bahawa ...
Transcript of BAHAGIAN A – Pengesahan Kerjasama* Adalah disahkan bahawa ...
BAHAGIAN A – Pengesahan Kerjasama*
Adalah disahkan bahawa projek penyelidikan tesis ini telah dilaksanakan melalui
kerjasama antara _______________________ dengan _______________________
Disahkan oleh:
Tandatangan : Tarikh :
Nama :
Jawatan :
(Cop rasmi)
* Jika penyediaan tesis/projek melibatkan kerjasama.
BAHAGIAN B – Untuk Kegunaan Pejabat Sekolah Pengajian Siswazah
Tesis ini telah diperiksa dan diakui oleh:
Nama dan Alamat Pemeriksa Luar : Prof. Dr. Arbakariya Ariff
Department of Bioprocess Technology,
Faculty of Biotechnology & Biomolecular
Sciences, UPM 43400 Serdang, Selangor.
Nama dan Alamat Pemeriksa Dalam : Prof. Madya Dr. Firdausi Razali
Department of Bioprocess Engineering,
Faculty of Chemical & Natural Resources
Engineering (FKKKSA), UTM, 81310
UTM, Skudai, Johor.
Nama Penyelia Lain (jika ada) :
Disahkan oleh Timbalan Pendaftar di SPS:
Tandatangan : Tarikh :
Nama :
CHROMATOGRAPHIC PURIFICATION STRATEGIES FOR RECOMBINANT
HUMAN TRANSFERRIN FROM SPODOPTERA FRUGIPERDA
WEE CHEN CHEN
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
JUNE 2008
iii
To my beloved grandparents, parents and brothers
iv
ACKNOWLEDGEMENT
In this long research journey, I receive all kind of guidance and support,
technically, financially and also spiritually. Thanks to everyone and the institution
for making this work possible. Special thanks are due to my supervisor, PM. Dr
Azila Abdul Aziz and co-supervisor, Dr. Badarulhisam Abdul Rahman for the
opportunity to be involved in this interesting project, their professional advice and
their encouragement in the effort to complete this research. I appreciate the given
opportunity to have a closer insight into biomanufacturing industry. Thanks also to
Prof. Dr. Michael J. Betenbaugh of Johns Hopkins University, USA for providing
recombinant baculoviruses.
I would like to express gratitude to all Bioprocess Department laboratory staff
especially Puan Siti Zalita, Encik Muhammad, Encik Malek and Encik Yaakop. I
also would like to thank all of the staff of Research Manage Center and Faculty of
Chemical and Natural Resources Engineering especially Cik Yun and Pn Naza.
They have been very helpful. I am also feel gratitude to have a group of kind
labmates and friends. Thanks to Dr. Taher, Wei Ney, Clarence, Hafiz, Kamalesh and
Kian Mou for their knowledge sharing. “Fui Ling, Melissa, Lee Yu and Seat Yee,
thanks for your company and motivation through out this long run”. Not to forget,
thanks to all the teachers and lecturers who had taught me all the basic knowledge.
My deepest appreciation would be dedicated to my sweet family members.
Their patience, consideration, encouragement, consistent support, recognition and
invaluable love make me strong and proud. Thank you.
v
ABSTRACT
Insect cell-baculovirus system is an excellent artificial system for the
production of recombinant glycoprotein despite its glycosylation deficiencies. In this
study, laboratory scale production of recombinant human transferrin (rhTf) from
insect cell-BEVS was conducted and chromatographic purification strategies were
employed to obtain rhTf in high yield and high recovery. Research was started with
the amplification of recombinant baculovirus, using low multiplicity of infection
(MOI). Virus stock in a 1.2 x 109 pfu/ml infected suspension culture of Spodoptera
frugiperda (Sf9) at 15 MOI had produced 31µg/ml of rhTf. To purify the rhTf,
hydrophobic interaction chromatography, dialysis and ion exchange chromatography
were performed. For hydrophobic interaction chromatography, elution strategy,
flowrate and rhTf loading capacity of phenyl sepharose were optimized. By loading
38µg rhTf/ml of gel, employing step elution with 50% 1.2M (NH4)2SO4/0.4M
Na3C6H5O7, pH6 (buffer A) and 25% buffer A and flowrate at 1ml/min, 74.6% of
rhTf had been recovered from phenyl sepharose. For ion exchange chromatography,
batch purification in reduced size was used to select suitable anion exchange matrix,
suitable pH of equilibration buffer and concentration of equilibration buffer. 20mM
Tris/HCl buffer, pH8.5 and gradient elution with the increase of of 5mM NaCl/CV
succeeded in giving pure rhTf with 52.5% recovery from Q-sepharose. The overall
recovery of pure rhTf was 34% with 200 purification fold. A brief glycan
characterization of the recovered pure rhTf was performed for a better understanding
of the glycosylation feature of this protein expressed using optimized medium from
BEVS. The carbohydrate component of the purified rhTf was determined. The
purified rhTf was hydrolyzed and the release sugar was labeled with 1-Phenyl-3-
Methyl-5-Pyrazolone (PMP) before analysis with High performance Liquid
Chromatography (HPLC). The molar fractions of Man, GlcNAc and Gal of rhTf
were 3.78, 1.69 and 0.93, respectively.
vi
ABSTRAK
Sistem pengekspresan sel serangga-bakulovirus merupakan sistem pilihan
yang baik untuk menghasilkan rekombinan glikoprotein meskipun kekurangan
glikosilasi. Penghasilan produksi skala makmal mendapat rekombinan human
transferrin (rhTf) dari sistem sel serangga-bakulovirus dan strategi purifikasi jenis
kromatografi telah dijalankan untuk mendapatkan rhTf yang tulen dan perolehan
yang tinggi. Kajian bermula dengan peningkatan kuantiti rekombinan bakulovirus
dari gandaan jangkitan (MOI) yang rendah. Stok virus dalam 1.2 x 109 pfu/ml
menjangkiti kultur ampaian sel Spodoptera frugiperda (Sf9) dengan 15 MOI telah
menghasilkan 31µg/ml rhTf. Dalam proses purifikasi, kromatografi saling tindak
hidrofobik, dialisis dan kromatografi penukaran ion telah dijalankan. Bagi
kromatografi saling tindak hidrofobik, strategi elusi, kelajuan dan kapasiti muatan
rhTf ke atas phenyl sepharose telah dioptimumkan. Penggunaan muatan 38µg
rhTf/ml gel dengan elusi berperingkat menggunakan 50% 1.2M (NH4)2SO4/0.4M
Na3C6H5O7, pH6 (larutan penimbal A) and 25% larutan penimbal A dan kelajuan
pada 1ml/min berjaya memperoleh 74.6% rhTf daripada phenyl sepharose. Bagi
kromatografi penukaran ion, purifikasi dalam saiz kecil telah digunakan untuk
memilih matrik penukar ion, pH larutan penimbal pada fasa keseimbangan dan
kepekatan larutan penimbal pada fasa keseimbangan. 20mM Tris/HCl larutan
penimbal, pH8.5 and elusi cerun dengan peningkatan 5mM NaCl/CV berjaya
menghasilkan rhTf tulen dengan 52.5% perolehan daripada Q-sepharose. Perolehan
rhTf tulen secara keseluruhan ialah 34% dengan 200 lipat purifikasi. Pencirian
glikan secara kasar telah dijalankan ke atas rhTf tulen untuk mendapat pemahaman
tentang ciri-ciri glikosilasi bagi protein ini yang diekspresikan dengan sistem
pengekspresan sel serangga-bakulovirus dan media optimum. Komposisi
karbohidrat untuk rhTf tulen telah dikenalpasti. rhTf yang tulen telah dihidrolisis.
Gula telah dilepaskan, dan dilabelkan dengan 1-Phenyl-3-Methyl-5-Pyrazolone
(PMP) sebelum dianalisis dengan menggunakan kromatografi cecair prestasi tinggi
(HPLC). Nilai fraksi molar Man, GlcNAc and Gal daripada rhTf ialah 3.78, 1.69 and
0.93.
vii
TABLE OF CONTENTS
CHAPTER TITLE PAGE
DECLARATION ii
DEDICATION iii
ACKNOWLEDGEMENTS iv
ABSTRACT v
ABSTRAK vi
TABLE OF CONTENTS vii
LIST OF TABLES xii
LIST OF FIGURES xiii
LIST OF SYMBOLS/ ABBREVIATIONS xvi
1 INTRODUCTION 1
1.1 Preface 1
1.2 Objectives 6
1.2 Scopes of Research 6
viii
2 LITERATURE REVIEW 7
2.1 Recombinant Protein Expression System 7
2.2 Insect Cell Baculovirus Expression System 11
2.2.1 Insect cell 11
2.2.2 Baculoviruses 11
2.2.2.1 Invivo and Invitro Replication 13
2.2.2.2 Recombination 15
2.3 Glycosylation 17
2.3.1 N-Glycosylation and O-Glycosylation 17
2.3.2 Glyscosylation Pathway 20
2.3.2.1 Glycosylation Pathway in Insect Cell 21
2.3.3 Model Protein- Transferrin 23
2.3.3.1 Recombinant Human Transferrin 27
2.4 Analysis Method 28
2.4.1 Bicinchoninic Acid (BCA) Assay 28
2.4.2 Enzyme Linked Immunosorbent Assay (ELISA) 29
2.4.3 Sodium Dodecyl Sulfate -Polyacrylamide Gel
Electrophoresis (SDS-PAGE) 31
2.4.4 Western Blot 32
2.4.5 Glucose, Lactic Acid and Glutamine Analyzer 33
2.4.6 Carbohydrate Analysis Using High
Performance Liquid Chromatography (HPLC) 34
2.4.6.1 Hydrolysis 34
2.4.6.2 1-Phenyl-3-Methyl-5-Pyrazolone (PMP)
Derivative of Sugar 34
2.4.6.3 Reverse Phase-HPLC 35
2.5 Purification of Transferrin 36
2.5.1 Hydrophobic Interaction Chromatography
(HIC) 37
2.5.1.1 Factors Affecting HIC 39
2.5.2 Ion Exchange Chromatography 43
2.5.2.1 Factor Affecting IEX 44
2.5.3 Optimization Method in Process
Chromatography 48
ix
2.6 Summary of Literature Review 49
3 MATERIALS AND METHODS 51
3.1 Materials 51
3.1.1 Cell lines and Recombinant Baculovirus 51
3.1.2 Equipments 51
3.1.3 Chemicals 52
3.2 Spodoptera frugiperda (Sf-9) Cells Culture 53
3.2.1 Cells Thawing 53
3.2.2 Cells Count 54
3.2.3 Adapting Serum Contain Culture to Serum Free
Culture 55
3.2.4 Adapting Monolayer Cells to Suspension
Culture 55
3.2.5 Maintaining Suspension Culture 56
3.2.6 Preparation of Optimized Medium 56
3.2.7 Adapting Suspension Culture in SFM900II to
Optimized Medium 57
3.2.8 Cells Freezing 57
3.3 Recombinant Baculovirus 58
3.3.1 Generating Pure Recombinant Virus Stock 58
3.3.2 Amplification of Virus Stock 58
3.3.3 Optimization of rhTf Expression 59
3.3.4 Virus Titration (End-Point Dilution) 59
3.4 Recombinant Human Transferrin Detection 60
3.4.1 Enzyme Linked Immunosorbent Assay
(ELISA) 60
3.4.2 Sodium Dodecyl Sulfate -Polyacrylamide Gel
Electrophoresis 62
3.4.2.1 Silver Staining 63
A
x
3.4.2.2 Coomassie Blue Staining 63
3.4.3 Western Blot 64
3.5 Characterization of Nutrient Consumption and
Substances Release 65
3.5.1 Analysis of Glucose, Lactic Acid and
Glutamine 65
3.5.2 Ammonia Test 66
3.6 Protein Assay 67
3.6.1 Bicinchoninic Acid (BCA) Assay 67
3.7 Purification 68
3.7.1 Hydropbobic Interaction Chromatography 68
3.7.2 Dialysis 69
3.7.3 Initial Screening Step of IEX using Batch
Purification in Reduced Volume 70
3.7.4 Ion Exchange Chromatography 71
3.8 Monosaccharide Composition Analysis of rhTf by
HPLC 72
3.8.1 Preparation of Apotransferrin, rhTf, Standard
Monosccharides 72
3.8.2 Hydrolysis 73
3.8.3 Pre-column Derivatization 73
3.8.4 HPLC Analysis 74
4 RESULTS AND DISCUSSION 75
4.1 Expression of rhTf 75
4.1.1 Growth Profile of Infected Virus 75
4.1.2 Time Course Expression Profile of rhTf 80
4.2 Purification 83
4.2.1 Profile of Sample Elution from Hydrophobic
Interaction Chromatography 85
xi
4.2.2 Optimization of Hydrophobic Interaction
Chromatography 86
4.2.2.1 Optimization of Elution Method 86
4.2.2.2 Optimization of Elution Flowrate 89
4.2.2.3 Optimization of rhTf Loading Capacity 92
4.2.3 Initial Screening Step of IEX Using Batch
Purification in Reduced Volume 95
4.2.4 Anion Exchange Chromatography 98
4.2.4.1 Maximizing The Selectivity of Anion
Exchange Chromatography 98
4.2.5 Characterization of rhTf Purification 100
4.3 Characterization of The Carbohydrate Composition of
rhTf 104
5 CONCLUSIONS 108
5.1 Conclusions 108
5.2 Recommendations 110
REFERENCES 112
APENDICES 137
xii
LIST OF TABLES
TABLE NO. TITLE PAGE
1.1 Comparison of pharmaceutical expression system
(Elbehri, 2005) 3
2.1 Characterization of selected host systems for protein
production from recombinant DNA (Shuler and Kargi, 2002) 10
2.2 Posttranslational processing and yield of the protein product in
various expression systems (cited from Luckow and Summers,
1988) 10
2.3 Selected private company with the protein engineering
platform 26
2.4 Functional groups used on ion exchangers 46
2.5 Capacity data for sepharose fast flow ion exchangers 47
2.6 Characteristics of Q, SP, DEAE and CM Sepharose Fast Flow 47
3.1 Culture volume for different flask size 56
3.2 Specification of YSI calibrator 65
3.3 Applied Condition for different study factors 71
4.1 Summary of the characteristic of small scale production of rhTf 83
4.2 Optimization of step-wise elution method for achieving higher
recovery of rhTf 87
4.3 Optimization of elution flowrate 90
4.4 Optimization of rhTf loading capacity 93
4.5 Summary of the characteristic of purification of rhTf 101
4.6 Carbohydrate Composition Analysis of Glycoprotein 107
xiii
LIST OF FIGURES
FIGURE NO. TITLE PAGE
1.1 Worldwide sales forecast for protein drugs, 2006 and 2011
(Talukder, 2007) 2
1.2 Strength and weaknesses of various expression systems
(Cox, 2004) 4
2.1 Electron micrographs and schematic of baculoviruses 12
2.2 Structural compositions of the two baculovirus phenotypes,
budded virus (BV), and the polyhedron derived virus (PDV) 12
2.3 The baculovirus life cycle in vivo and in vitro 14
2.4 Construction of baculovirus expression vectors 16
2.5 Structure of the N-glycosidic bond and O-glycosidic bond
found in glycoproteins. 18
2.6 Structure of the different types of oligosccharidic chains of
N-glycoproteins 19
2.7 Pathway for generation of the dolichol-linked oligosaccharide donor
for protein N-glycosylation 21
2.8 Protein N-glycosylation pathways in insect and mammalian
cells 22
2.9 A ribbon diagram of a diferric rabbit serum transferrin
molecule 25
2.10 Reaction schematic for BCA assay 29
2.11 Schematic represents the a) Direct Sandwich ELISA; b)
Indirect ELISA; c) Sandwich ELISA; d) Competition ELISA 30
xiv
2.12 SDS-PAGE 31
2.13 Immobilized enzyme biosensor of YSI 33
2.14 Hydrolysis time course of bovine fetuin 34
2.15 Derivatization with pyrazolone derivatives 35
2.16 Different hydrophobic ligands coupled to cross-linked
agarose matrices 40
2.17 The Hofmeister series on the effect of some anions and
cations in precipitating proteins 41
2.18 Relative effects of some salts on the molal surface tension of
water 41
2.19 Effect of pH on protein at different net charge 44
2.20 Ion exchanger types 45
3.1 Schematic representative of the procedures employed for
virus titer-end point dilution 60
3.2 Schematic representative of the procedures used in ELISA
method 61
3.3 Schematic representation of the BCA protein assay 67
3.4 Schematic diagram of the dialysis procedure 70
3.5 Schematic diagram of the set up of the chromatography
equipment. 72
4.1 Photography of control and infected culture 76
4.2 Growth Characteristics of sf9 during rhTf virus propagation 78
4.3 Growth Characteristic of sf9 during rhTf production in
optimized suspension culture 79
4.4 The profile of glucose, glutamine consumption and lactate
formation in supernatant post infection 80
4.5 rhTf production profile in supernatant 81
4.6 Characterization of the rhTf production profile of infected
Sf9, using 9%, Coomassie blue staining, SDS-PAGE 82
4.7 Characterization of the rhTf production profile of infected
Sf9, using Western Blot 82
4.8 Steps and gradient elutions of rhTf from HIC column 85
4.9 HIC chromatograms for the optimization of elution method 88
4.10 HIC chromatograms for the optimization of elution flowrate 91
xv
4.11 The relationship between recovery percentage and loading
capacity 93
4.12 HIC chromatograms for the optimization of rhTf loading
capacity 94
4.13 SDS-PAGE characterizing the elution profile of rhTf 95
4.14 Binding capacity of two anion exchange matrix with Tris and
phosphate buffer used as equilibration buffer 96
4.15 Binding capacity of Q-Sepharose with equilibration buffer of
different pH 97
4.16 Binding capacity of Q-Sepharose with different concentration
of equilibration buffer 97
4.17 Anion exchange chromatograms for the optimization of
selectivity 99
4.18 SDS-PAGE characterizing the elution profile of rhTf 100
4.19 HIC chromatogram characterizing the separation and elution
profile of sample 102
4.20 SDS-PAGE characterizing the separated protein from phenyl
sepharose 6 fast flow column 102
4.21 Anion exchange chromatogram characterizing the separation
and elution profile of sample of after HIC and after dialysis 103
4.22 SDS-PAGE characterizing the separated protein from
Q-Sepharose column 103
4.23 SDS-PAGE characterizing the sample pooled from each
purification step 104
4.24 Chromatogram shows HPLC separation of PMP-labeled
transferrin 106
4.25 Standard calibration graph of monosaccharides 107
xvi
LIST OF SYMBOLS/ ABBREVIATIONS
% Percentage
α Alpha
β Beta
µm Micro meter
°C Degree Celsius
µg Micro gram
µg/ml Micro gram per milliliter
µl Microliter.
µm Micrometer
µmol/ml Micro mol per milliliter
AAGR Average annual growth rate
Ablank Absorbance for blank
AcMNPV Autographa californica multiple nuclear polyhedrosis virus
ACN Acetonitrile
AcNPV Autographa californica nuclear polyhedrolysis
Asample Absorbance for sample
Asn-X-Ser Asparagine-X-Serine
Asn-X-Thr Asparagine-X-Threonine
Astandard Absorbance for standard
ATCC American Tissue Culture Collection
BEVS Baculovirus expression vector system
BHK Baby hamster kidney cells
Bm Bombyx mori
BmNPV Bombyx mori nuclear polyhedrosis virus.
xvii
BV Budded virus
BmNPV Bombyx mori nuclear polyhedrosis virus.
BV Budded virus
BVs Budded viruses
cDNA Complementary deoxyribonucleic acid
cells/ml Cells per milliliter
CHO Chinese Hamster Ovary
CM Carboxymethyl
cm/hr Centimeter per hour
cm2 Centimeter square
CMP-NeuAc Cytidine-5’-monophospho N-acetylneuraminic acid
Cu1+
Cuprous ion
CuSO4•5H2O Copper (II) sulfate pentahydrate
CV Column Volume
DEAE Diethylaminoethyl
DMSO Dimethyl sulphoxide
DNA Deoxyribonucleic Acid
DO Dissolved oxygen
DPA Dipicolylamine
e- Electron
E.coli Escherichia coli
ELISA Enzyme linked immunorsorbent assay
ER Endoplasmic recticulum
FBS Fetal bovine serum
FDA Food and Drugs Administration
Fe3+
Ferric ion
Fuc Fucose
g Gravitational
g/l Gram per liter
Gal Galactose
GalNAc N-Acetylgalactosamine
GDP-mannose Guanosine diphoshate mannose
GlcN Glucosamine
GlcNAc N-Acetylglucosamine
xviii
GLDH Glutamate dehydrogenase
GLP-1 Glucagons-like peptide 1
GLP-1-R Glucagons-like peptide 1-receptor
GMP Good manufacturing practice
gp Glycoprotein
GV Granuloviruses (GV)
H+ Hydrogen cation
H2O2 Hydrogen peroxide
H3PO4 Phosphoric acid
HIC Hydrophobic interaction chromatography
His6 Hexahistidine
HPLC High performance Liquid Chromatografi
HRP Horseradish peroxidase
Hrs Hours
hTf Human transferrin
IEX Ion exchange chromatography
IgG Immunoglobulin G
IMAC Metal affinity chromatography
k constant
Kb/kbp Kilo base pair
kDa Kilo Dalton
M Molar
Man Mannose
Man3–1GlcNAc2 3(Mannose)-2(N-Acetyl Glucosamine)
Man3GlcNAc2 3(Mannose)-2(N-Acetylglucosamine)
Man8–GlcNAc2 8(Mannose)-2(N-Acetylglucosamine)
Man9GlcNAc2 9(Mannose)-2(N-Acetylglucosamine)
MeOH Methanol
mg Milligram
mg/ml Milligram per milliliter
min Minutes
ml/min Milliliter per minutes
mmol/L milli mol per liter
MOI Low multiplicity of infection
xix
MPa Mega Pascal
MW Molecular weight
MWCO Molecular Weight Cut Off
N Normal
N.D Not defined
NaCl Sodium Chloride
NADP+/NADPH Nicotinamide adenine dinucleotide phosphate
Na3C6H5O7 Sodium citrate
NaOH Sodium hydroxide
ng/ml Nanogram per milliliter
NH3 Ammonia
(NH4)2SO4 Ammonium Sulphate
Ni2+
Nickel ion
nm Nano meter
NPV Nucleopolyhedoviruses
O2 Oxygen
OB Occlusion bodies
ODS Octadecyl silica
ODV Occlusion derived virus
OV Occluded virus
p10 Phage-encoded protein-10
PBS Phosphate buffered saline
pfu/ml Plug performing unit per milliliter
pH Potential hydrogen
pI Isoelectric point
PIBs Polyhedral inclusion bodies
pmol Pico mol
PMP 1-Phenyl-3-Methyl-5-Pyrazolone
QAE Quaternary Aminoethyl
Q-sepharose Quaternary ammonium
rhTf Recombinant human transferrin
RP-HPLC Reversed phase HPLC
rpm Rotation per minutes
RT Retention time
xx
S Methyl sulphonate
S. cerevisiae Saccharomyces cerevisiae
SDS Sodium dodecyl sulfate
SDS-PAGE Sodium dodecyl sulfate-polyacrylamide gel electrophoresis
SFM Serum Free Medium
SP Sulphopropyl
T.ni Trichoplusia ni
TBS Tris buffered saline
TCID50 50 % Tissue Culture Infectious Dose
TCID50/ml 50 % Tissue Culture Infectious Dose per milliliter
TEMED N,N,N',N'-tetramethylethylenediamine
TFA Trifluoroacetic acid
TM Trademark
TMB 3,3’,5,5’-tetramethylbenzidene
TN5B1-4 High 5
TOI Time of Infection
Tris-HCl Tromethamine and Hydrochloric Acid
UDP Uridine-5’-diphophate
UDP-Gal Uridine-diphosphate galactose
UDP-Glc Uridine-diphosphate glucose
UDP-GlcNAc Uridine-diphophate N-acetylglucosamine
V Volts
W.R Working reagent
xxi
LIST OF APPENDICES
APPENDIX NO. TITLE PAGE
A-1 Stock Solution for SDS-PAGE 134
A-2 Working Solution for SDS-PAGE 135
A-3 Separating and Stacking Gel Preparation 136
B Coomassie Blue Staining 137
C Preparation of Optimized Medium 138
D Example of TCID50 Calculation (spreadsheet) 139
E Working Solution for ELISA 141
F Working Solution for Western Blot 143
G Mobile Phase for Purification 144
H Glycan Analysis 145
CHAPTER 1
INTRODUCTION
1.1 Preface
The biopharmaceutical industry has experienced a significant transformation
based on the development of recombinant DNA and hybridoma technologies in the
1970s. The industry has moved beyond simple replication of human proteins (such
as insulin or growth hormones) and played a key role in the development of large-
molecule drugs such as any protein, virus, therapeutic serum, vaccine, and blood
component. These genetically engineered therapeutic drugs are targeting some of the
major illnesses such as cancer, cardiovascular, and infectious diseases and they have
the full potential to tackle a whole array of new diseases effectively and safely.
By mid 2003, 148 biopharmaceuticals proteins were approved in the United
States and Europe compared to 84 in 2000 (Birch and Onakunle, 2005). The total
global market for protein drugs was $47.4 billion in 2006 and the market is presumed
to reach $55.7 billion by the end of 2011 with an average annual growth rate
(AAGR) of 3.3% (Figure 1.1). It is expected that current cell culture facilities are
unlikely to meet expected demand. The imbalance of supply-demand is
2
expected to get worse in the future, as more biotech therapeutics proteins are
approved. 20–50% of potential therapeutics could be delayed due to the lack of
manufacturing capacity (Fernandez et al., 2002). Hence, the ability in expanding the
existing capacity and producing a larger variety of products are crucial in order to
meet future demand. Drug companies and biotech firms are considering alternative
manufacturing platforms, besides increasing fermentation capacity (Table 1.1)
(Elbehri, 2005).
Figure 1.1: Worldwide sales forecast for protein drugs, 2006 and 2011 (Talukder,
2007).
Generally, recombinant therapeutic protein can be generated and produced in
various prokaryotic and eukaryotic expression systems. Until the early 1990s, the
majority of recombinant proteins were expressed in either microbial or mammalian
cell culture systems. The first approved recombinant therapeutic glycoproteins,
insulin is produced from Escherichia coli. Today, the manufacturing of
biotechnology products relies heavily on the use of mammalian cells, chiefly on
Chinese Hamster Ovary (CHO) cells. The well-known drugs Avonex (interferon
beta 1-a, Biogen, Inc) and EPOGEN/EPREX (epoetin alfa, Amgen Inc/ Ortho
Biotech) are produced in CHO. Insect, transgenic plant, transgenic animal and yeast
3
cells are also attractive as hosts for the production of recombinant proteins, as they
represent potentially inexpensive and versatile expression systems. Optimal
expression system can be varied, based on different critical parameters of the protein
of interest. Selecting an appropriate expression system for the protein of interest will
affect factors such as time to market, cost of goods, product characteristics,
regulatory hurdles, and intellectual property (Figure 1.2).
Table 1.1: Comparison of pharmaceutical expression system (Elbehri, 2005).
Expression System Advantages Disadvantages Applications
Cost per gram
Bacteria
Established regulatory track; well-understood genetics; cheap and easy to grow
Proteins not usually secreted; contain endotoxins; no posttranslational modifications
Insulin (E. coli; Eli Lilly); growth hormone (Genentech); growth factor; interferon
N.R
Yeast
Recognized as “safe;” long history of use; fast; inexpensive; posttranslational modifications
Overglycosylation can ruin bioactivity; safety; potency; clearance; contains immunogens/antigens
Beer fermentation; recombinant vaccines; hepatitis B viral vaccine; human insulin
$50-100
Insect cells
Posttranslational modifications; properly folded proteins; fairly high expression levels
Minimal regulatory track; slow growth; expensive media; baculovirus infection (extra step); mammalian virus can infect cells
Relatively new medium; Novavax produces virus-like particles
N.R
Mammalian cells
Usually fold proteins properly; correct posttranslation modifications; good regulatory track record; only choice for largest proteins
Expensive media; slow growth; may contain allergens/ contaminants; complicated purification
Tissue plasminogen activator; factor VIII (glycoprotein); monoclonal antibodies (Hercepin)
$500–5,000
Transgenic animals
Complex protein processing; very high expression levels; easy scale up; low-cost production
Little regulatory experience; potential for viral contamination; long time scales; isolation/GMPs on the farm
Lipase (sheep, rabbits; PPL Therapeutics); growth hormone (goats; Genzyme); factor VIII (cattle)
$20–50
Transgenic plants
Shorter development cycles; easy seed storage/scaling; good expression levels; no plant viruses known to infect humans
Potential for new contaminants (soil fungi, bacteria, pesticides); posttranslational modifications; contains possible allergens
Cholera vaccine (tobacco; Chlorogen, Inc.); gastric lipase (corn; Meristem); hepatitis B (potatoes; Boyce Thompson)
$10–20
N.R- Not Reported
4
The baculovirus expression vector system (BEVS) has a number of
significant advantages over other methods of recombinant protein production. It is
best known as providing quick access to biologically active proteins and used as a
research tool (Cox, 2004). The major advantages of BEVS over bacterial and
mammalian expression system is the very high expression of recombinant proteins
which in many cases are antigenically, immunogenically and functionally similar to
their native counterparts (Goosen, 1993). Lack of adventitious viral agents that
could replicate in mammalian cells (John Morrow, 2007), make BEVS a powerful
manufacturing platform for health care solutions to pandemic, biodefense, and
emergency scenarios (Cox, 2004). However, BEVS also has its limitation in
producing authentic mammalian proteins and glycoproteins. An absence of complex
sugars in BEVS-produced proteins may result in poor pharmacological activity in
vivo due to the rapid clearance from the circulatory system of glycoproteins with
non-human glycans (Betenbaugh et al., 2004)
Figure 1.2: Strength and weaknesses of various expression systems (Cox, 2004).
The deficiency of BEVS in producing mammalian like-glycoproteins of
potential therapeutic is a hot topic among researchers in this field. BEVS had been
reported to produce sialylated complex type N-glycan through the modification of its
metabolic engineering pathway (Betenbaugh et al., 2004; Viswanathan et al., 2005;
5
Yun et al., 2005). Protein Sciences Corporation (PSC) had developed technology for
large-scale (600 L) production of proteins in insect cells using the BEVS (Cox,
2004). Although currently there are no FDA-approved therapeutic proteins
expressed using BEVS, a number of products are in advanced clinical trials and
several are about to get acceptance. Among these, three vaccines that are close to
market are Provenge™, a prostate cancer immunotherapy from Dendreon
(www.dendreon.com); Ceravix™, a papilloma virus vaccine from GlaxoSmithKline
(www.gsk.com); and FluBIOk™ from Protein Sciences, a non-egg based flu vaccine
(John Marrow, 2007).
BEVS have tremendous potential to become the next therapeutic
manufacturing system. In this study, recombinant human transferrin was used as a
model protein. Transferrin was chosen because of the simplicity of its structure and
its recent important role in protein engineering. Non-glycosylated transferrin had
been used as a scaffold to extend the half life of peptide and proteins. Various
chromatographic methods for purification of transferrin have been reported. Among
these reports, Ali et al. (1996) and Ailor et al. (2000) had purified rhTf from sf9 and
Tn cells using phenyl sepharose and Q-Sepharose. In this study, hydrophobic
interaction chromatography utilizing phenyl sepharose was used as the capture step
and IEX chromatography utilizing Q-sepharose was used for further purification of
rhTf. To obtain pure rtTf, optimization of both chromatographic techniques had
been carried out. Basic characterization of the carbohydrate content of the pure rhTf
had also been carried out to get a better understanding of the glycan.
6
1.2 Objectives
The objective of this work was to optimize the chromatographic purification
process of recombinant human transferrin expressed using BEVS to obtain pure rhTf
in improved yield and recovery.
.
1.3 Scopes of Research
The following are the scopes of this work:
1) Propagation of baculovirus.
2) Small scale production of rhTf using optimized medium.
3) Characterization of productivity profiles using SDS-PAGE, ELISA and Western
Blot.
4) Optimization of purification process of rhTf using HIC and IEX.
5) Characterization of the monosaccharide composition of the expressed rhTf.
CHAPTER 2
LITERATURE REVIEW
2.1 Recombinant Protein Expression System
Procaryotic has been employed in the protein manufacturing system and one
of the dominant workhorses for commercial production is E. coli. (Lee, 1996;
Makrides, 1996). Prokaryotic expression systems offer high production yields at
reduced cost (Table 2.1). However, the expressed-recombinant protein is
aglycosylated and lost of biological effector functions (Wright and Morrison, 1997).
E. coli cannot produce some proteins containing complex disulfide bonds or
mammalian proteins that required posttranslational modification for activity. The
system maybe best suited to production of antibody fragments, rather than complete
immunoglobolins because of the complexity of the protein folding pathway. Product
of E. coli was primarily in the form of inclusion bodies, and thus biologically
inactive, misfolded and insoluble. Biologically active proteins can only be recovered
by complicated and costly denatured and refolding processes. Another disadvantage
of the system is release of endotoxins from inclusion bodies which affect the
recovery and purification.
8
Recently, the manufacturing of therapeutic compound relies heavily on the
use of mammalian cells. Recombinant protein production using mammalian cells
offers several advantages over microbial systems (Table 2.1). Mammalian cells are
able to secrete the protein product and perform post-translational modifications
which are necessary for human therapeutics protein. Chinese Hamster Ovary (CHO)
and murine myeloma (NSO) cells are favored because they efficiently assemble
complex multi proteins (such as immunoglobulins) and are believed to synthesize
glycans similar to those found in human glycoproteins (Chu and Robinson, 2001). In
mammalian cells, protein N-glycosylation is carried out by an elaborate, but well-
characterized metabolic pathway (Kornfeld and Kornfeld, 1985; Montreuil et al.,
1995; Varki et al., 1999) and closest to its natural counterpart. However, mammalian
cells have significantly slower growth rates, lower protein expression level and are
much more complex in their nutritional requirements compared to microbes.
Insect and yeast cells are attractive as hosts for the production of recombinant
proteins too, as they represent potentially inexpensive and versatile expression
systems (Table 2.1). Yeast is an attractive host for the expression of heterogous
protein (Reiser, 1990; Romanos et al., 1992; Muneo et al., 1992). It offers the
advantage of both bacterial and mammalian system. Saccharomyces cerevisiae was
the first to be used for the production of recombinant protein such as interferon
(Tuite et. al., 1982) and hepatitis surface antigen (Valenzuela et. al., 1982). The
advantages of yeast expression system are capability in processing authentic and
bioactive mammalian protein, high level of secretion into protein free medium, rapid
growth rate, ease of high density fermentation, scale up without loss of yield, ease of
genetic manipulation, lower cost compare to mammalian expression systems, lack of
endotoxins, lytic viruses and no know panthogenic relationship with man (Li et al.,
2001). However, yeasts sometimes form hypermannosyl glycans and add 50 or even
more Man residues to Man8–GlcNAc2 (Betenbaugh et al., 2004).
Hypermannosylation can hamper downstream processing of recombinant
glycoproteins and may complicate complete molecular characterization of the
molecules (Vervecken et al., 2004).
9
Another important technology that has gained much ground is insect cell
culture (Tulsi. 2004). Insect cells used in conjunction with the baculovirus
expression vector system (BEVS) have been widely used for the mass production of
heterologous proteins (Possee, 1997). The Insect-BEVS has significant advantages
over other methods of recombinant protein production, such as ease of culture, ideal
for suspension culture, ease of scale up, high product expression, high gene
expression, higher tolerance to osmolality and the absence of harmful factors that
could replicate in mammalian cells. The method required for generating and
maintaining baculovirus recombinants and stable insect cell are simple and cost
effective, requiring incubation without the support of carbon dioxide. Glycosylation
was found to be stable and rather insensitive to variations in ammonia concentration,
temperature and dissolved oxygen concentration (Donaldson et al. 1999). In addition
to high gene expression, BEVS allows synthesis of proteins varying in size and in
complexity, posttranslational proteolytic processing, cleavage of signal peptides,
expression of nonspliced genes, and adequate compartmentation of recombinant
proteins (Beljelarskaya, 2002). BEVS also allows simultaneous expression of
several genes and production of heterodimeric proteins in one infected cell (An et al,
1999).
Insect cells-BEVS provide for protein maturation and modification typical of
eukaryotic systems, including glycosylation, phosphorylation, palmitylation
(acylation with fatty acid residues), amidation, and carboxymethylation (table 2.2).
The heterologous proteins are posttranslationally modified in a similar pattern to
those observed in mammalian cells (Ailor and Betenbaugh, 1999). They form
disulfide bonds and assume native secondary and tertiary structures. However,
unlike the multiantennary, sialylated complex N-glycans produced in mammalian
cells, recombinant protein produced in insect cells are typically paucimannosidic
Man3–1GlcNAc2 N-glycans (Betenbaugh et al., 2004). Proteolysis is another
problem of the BEVS system due to its lytic nature. Several cell or baculovirus
proteases are involved in degradation events during protein production by insect cells
which affect both quality and quantity of the product. The problem is exacerbated in
serum free culture where there is lack of protection by serum proteins such as
albumin and macroglobulin.
10
Table2.1: Characterization of selected host systems for protein production from
recombinant DNA (Shuler and Kargi, 2002).
Organism
Characteristic E. coli
Yeast
(S. cerevisiae) Insect Mammalian
High growth rate E VG P-F P-F
Availability of genetic
systems E G F-G F-G
Expression levels E VG G-E P-G
Low-cost media
available E E P P
Protein folding F F-G VG-E E
Simple glycosylation No Yes Yes Yes
Complex glycosylation No No Yesa Yes
Low Levels of
proteolytic degradation F-G G VG VG
Excretion or secretion
P normally
VG in special
cases
VG VG E
Safety VG E E G
E, excellent; VG, very good; G, good; F, fair; P, poor.
aGlycosylation patterns differ from mammalian cells.
Table 2.2: Posttranslational processing and yield of the protein product in various
expression systems (cited from Luckow and Summers, 1988).
Expression System E. coli Yeast cells Mammalian
cells Insect cell
Proteolytic cleavage +/- +/- + +
Glycosylation - + + +
Secretion +/- + + +
Secondary structure formation +/- +/- + +
Phosphorylation - + + +
Acrylation - + + + Amidation - - + +
Protein yield, % dry weight 1-5% 1% <1% 30%
11
2.2 Insect Cell Baculovirus Expression System
2.2.1 Insect cell
The three most popular insect cell lines used in the BEVS are Sf9 and Sf21
from the fall armyworm, Spodoptera frugiperda (S.f), and TN5B1-4 (High 5) from
the cabbage looper, Trichoplusia ni (T.ni). Sf are the most frequently used cell line
and the popularity is due to the effectiveness in making proteins and being the best
cell line for producing viruses (Tulsi, 2004). T.ni is excellent for protein production,
especially secreted protein. However, the high metabolic activity of this cell line
results in a higher proportion of by-product accumulation (Rachel et al., 1995).
Besides that, T.ni has transposons that can inhibit the efficient production of insect-
derived virus-like particles (VLPs) (Tulsi, B 2004). Other cell lines, Bombyx mori
(Bm-N), Mamestra brassicae (e.g., MB0503), and Estigmene acrea are also notable
because of its glycosylation potential. In general, all the cell lines are obtained from
embryonic (Altmann et al., 1999).
2.2.2 Baculoviruses
Baculoviruses (family Baculoviridae) are viral pathogens, which cause fatal
disease in insects, mainly in members of the families Lepidoptera, Diptera,
Hymenoptera and Coleoptera. More than 600 baculoviruses have been identified,
categorized in two subfamilies: the nucleopolyhedoviruses (NPV) and granuloviruses
(GV) (Murphy et al., 1995). Baculoviruses are highly specific, not known to
propagate in any non-invertebrate host. They can reduce the size of insect pests in
agriculture and forestry as alternative to chemical insecticides (Granados and
Federici, 1986; Payne, 1998). Baculovirus genome is replicated and transcribed in
the nuclei of infected host cells. The large Baculovirus DNA (between 80 and 200
12
kb) is double stranded, circular, supercoiled DNA molecules that packaged into rod-
shaped nucleocapsids (Summers and Anderson, 1972; Burgess, 1977), that are
enveloped singly or in bundles by a unit membrane (Figure 2.1). Nucleocapsids exist
in distinctive virion phenotypes: 1) occluded virus (OV) and 2) budded virus (BV)
(Figure 2.2). Size of these nucleocapsids is flexible and large amounts of foreign
DNA can be accommodated by recombinant baculovirus.
Figure 2.1: Electron micrographs and schematic of baculoviruses A) Baculovirus
particles, or polyhedra; B) Cross-section of a polyhedron; C) Diagram of polyhedron
cross-section. Electron micrographs (A&B) by Jean Adams, graphics (C) by V.
D'Amico.
Figure 2.2: Structural compositions of the two baculovirus phenotypes, budded virus
(BV), and the polyhedron derived virus (PDV). Graphics by Kalmakoff & Ward.
13
BVs generally contain a single nucleocapsid and are enclosed in an envelope
obtained as the nucleocapsids bud out through the cell wall. Prior to the budding of
the virus, the cell wall is modified by the addition of the viral protein glycoprotein
(gp) 64. This protein has been shown to be required for effective spread of the virus
within the host. The occlusion derived virus (ODV) is the form of the virus which is
produced in the latter stages of viral infection and is enclosed in a proteinaceous
occlusion body. They allow for horizontal spread of the virus from insect to insect
and allow the virus to persist for long periods in the environment.
2.2.2.1 Invivo and Invitro Replication
Wild-type baculoviruses in both in vivo and invitro conditions exhibit both
lytic and occluded life cycles that develop independently throughout the three phases
of virus replication (Figure 2.3a). In the early phase which is also known as the virus
synthesis phase, the virus prepares the infected cell for viral DNA replication. Steps
of infection include attachment, penetration, uncoating, early viral gene expression,
and shut off of host gene expression. Actual initial viral synthesis occurs 0.5 to 6
hours (hrs) after infection. Late genes that code for replication of viral DNA and
assembly of virus are expressed in the late phase which is also known as the viral
structural phase. Between 6 and 12 h after infection, the cell starts to produce BV,
also called non-occluded virus (NOV) or extracellular virus (EV). The BV contains
the plasma membrane envelope and gp64 necessary for virus entry by endocytosis.
Peak release of extracellular virus occurs, 18 to 36 hrs after infection. The BV is
responsible for cell to cell transmission within an infected insect and cell culture.
In the very late phase, the viral occlusion protein phase, the polyhedrin and
p10 genes are expressed, OV—also called occlusion bodies (OB) or polyhedral
inclusion bodies (PIBs)—are formed between 24 and 96 h after infection. Particles
of OV assemble inside the nucleus, contain nuclear membrane envelopes and are
14
embedded in a homogenous matrix made predominantly of polyhedrin protein
(Rohrmann, 1986; Summers and Smith, 1978). The polyhedrin protein is not
essential for the life cycle in invitro cell culture, but essential in invivo replication for
its dissemination into the environment and allowing primary infection in susceptible
larva. Multiple virions like gp41 and gp74 are produced and surrounded by a
crystalline polyhedra matrix. OV are released when the infected cells lyses.
Occluded virions are protected from desiccation in the environment. Once
ingested, the occlusion body is solubilized in the gut, releasing virions which fuse
with midgut cells. The virion nucleocapsid migrates through the cytoplasm to the
nucleus. The core is uncoated from the capsid structure in the nucleus and
replicated. Secondary infection is mediated by the budded form of the virus entering
adjacent cells via adsorptive endocytosis. In vitro, a polyhedron gene modified to
express a recombinant gene product is used. Recombination takes place within the
insect cells between the homologous regions in the transfer vector and the
baculovirus DNA. Recombinant virus produces recombinant protein and also infects
additional insect cells thereby resulting in additional recombinant virus (Figure 2.3b).
Figure 2.3: The baculovirus life cycle (A) in vivo and (B) in vitro (adapted from
Pharmigen, 1999)
15
2.2.2.2 Recombination
Two of the most common isolates members used in foreign genes expression
are Autographa californica nuclear polyhedrolysis (AcNPV; also written AcNMPV)
and Bombyx mori (silkworm) nuclear polyhedrosis virus (BmNPV). Entire genome
of AcNPV has been mapped and fully sequenced (Ayres et al., 1994; Kool and Vlak,
1993; Harrap, 1972). For the recombinant in vitro infection (Figure 2.1b), the
naturally occurring polyhedrin gene within the wild-type baculovirus genome is
replaced with a recombinant gene or cDNA. Deletional or insertional inactivation of
the polyhedrin gene in AcNPV does not affect virus propagation but results in the
production of occlusion body-negative viruses. Promoters of varying strength and
differential expression during the virus-like cycle like polyhedrin and p10 promoters
can be used to control the expression of foreign gene. The promoter of the
polyhedrin gene has been widely used for directing the high level production of
heterogolous proteins. During the very late phase of infection, the inserted
heterologous genes are placed under the transcriptional control of the strong AcNPV
polyhedron promoter. Thus, recombinant product is expressed in place of the
naturally occurring polyhedrin protein.
The baculovirus genome is generally too large to easily insert foreign genes.
Several procedures were proposed for constructing recombinant baculoviruses,
including direct enzymic ligation of a foreign DNA fragment into the virus genome,
employment of large bacterial plasmids and use of shuttle vectors for insect cells
(Davies, 1994; Peakman et al., 1992; Luckow et al., 1993; Patel et al., 1992). The
most common method is based on homologous recombination between a transfer
vector and a wild-type virus (Matsuura et al., 1987). The transfer vector contains an
appreciable viral DNA fragment and the cDNA to be expressed, which is controlled
by the promoter of a baculovirus gene. It is constructed and amplified in E. coli.
Co-transfection of the transfer vector and AcMNPV DNA into Sf cells allows
recombination between homologous sites, transferring the heterologous gene from
the vector to the AcMNPV DNA (Figure 2.3b, 2.4). AcMNPV infection of Sf cells
results in the shut-off of host gene expression allowing for a high rate of recombinant
16
mRNA and protein production. Recombinant viruses can be easily identified and
purified because they produce occlusion body-negative viruses that formed distinctly
different plaques from wild type virus. Recombinant proteins can be produced at
levels ranging between 0.1% and 50% of the total insect cell protein.
As shown in Figure 2.4, the polyhedron gene (dashed area) is replaced by
foreign gene or the gene of interest (strippled area). Virus DNA and transfer vector
are co-transfected into the host insect cell and homologous recombination between
the flanking sequences common to both DNA molecules occurs. This causes the
insertion of the gene of interest into the viral genome at the polyhedrin locus,
resulting in the production of a recombinant virus genome. Plaque assay used to
screen the wild type and recombinant baculovirus. The genome then undergoes
replication within the host nucleus, generating recombinant baculovirus vector
containing the foreign gene under the control of the strong, late viral polyhedrin
promoter.
Figure 2.4: Construction of baculovirus expression vectors.
17
2.3 Glycosylation
Glycosylation is the process of addition of carbohydrate moiety to proteins or
lipids in covalent chemical linkage. Glycosylation is one of the principal post-
translational modification steps in the synthesis of membrane and secreted proteins.
It is a site specific, enzymatic process which involves a sequential series of trimming
and elongation reactions carried out by enzymes localized along the cellular
secretory pathway. The products of glycosylation are glycoproteins or glycolipids.
Many of the high-value therapeutic proteins in the market and in clinical
development today are glycoproteins. The carbohydrate components of
glycoproteins or the glycan are critical in biologic functions such as immunogenicity,
solubility, receptor recognition, inflammation, pathogenicity, metastasis, and other
cellular processes (Olden et al., 1982). Besides that, the specific glycan structures
are also essential for their structure, stability and functionality (Varki, 1993; Traving
and Schauer, 1998) and affect a number of physiological properties including in vivo
half-life, bioavailability, and tissue targeting.
2.3.1 N-Glycosylation and O-Glycosylation
The two main types of glycosylation are N-linked glycosylation and O linked
glycosylation (Figure 2.5). N-linked glycoprotein consists of glucose, mannose and
N- acetylglucosamine molecules. The glycosylation begins with the addition of 14-
sugar precursor to an asparagine amino acid via an amide bond in an Asn-X-Ser or
Asn-X-Thr motif and X can be any amino acid other than Proline. This entity is then
transferred to the endoplasmic recticulum (ER) lumen and the oligosaccharyl
transferase enzymes continue the glycosylation by attaching the oligosaccharide
chain to asparagine. The oligosaccharide attached protein sequence now folds
correctly and is now translocated to the Golgi body where the mannose residue is
18
removed. N-linked glycosylation is important for the folding of some of eukaryotic
proteins. It occurs widely in archaea, but very rarely in bacteria.
O-linked glycosylation begins with an enzyme mediated addition of N-acetyl-
galactosamine followed by other carbohydrates to hydroxyl group of serine or
threonine residues. O-linked glycosylation occurs at a later stage in protein
processing probably in the golgi apparatus. O-glycosidic chain or O-glycan is
smaller than N-glycan. Termination of O-linked glycans usually includes Gal,
GlcNAc, GalNAc, Fuc, or sialic acid. This linkage is found in mucinous
glycoproteins and fibrillar collagens (Carson, 1992). It is also important to form
components of the extracellular matrix, adhering one cell to another by interactions
between the large sugar complexes.
Figure 2.5: Structure of the N-glycosidic bond and O-glycosidic bond found in
glycoproteins. (Whitaker, 1977)
N-glycosylproteins can be categorized into three forms which are high
mannose type, complex type and hybrid type. High mannose or oligomannosidic
type glycoproteins are uniquely composed of mannose residues (Figure. 2.6). They
are basically the precursors to hybrid and complex type chains and have been
identified in plants, animals and yeast (Cummings et al.. 1989; Montreuil et al.,
1986; Kimura et al., 1992). The complex type glycoprotein contain almost any
number of the other types of saccharides, including more than the original two N-
19
acetylglucosamines. These chains can be still classified as biantennary, triantennary
or tetraantennary based on their branching pattern (Figure. 2.6). Additionally, the
glycans contain galactose, fucose and sialic acids. Hybrid type chains of have
structural features of both the high mannose and complex types (Figure. 2.6).
Figure 2.6: Structure of the different types of oligosaccharidic chains of N-
glycoproteins (adapted from Cummings et al., 1989).
20
2.3.2 Glyscosylation Pathway
In eukaryotic cells, the glycosylation takes place in the membrane cellular
compartments: endoplasmic reticulum (ER), golgi apparatus, lysosomes (Cumming,
1992). The biosynthesis of N-glycoproteins is initiated by the formation of a
precursor, consisting of a lipid, a dolichol, linked to an oligosaccharide by a
pyrophosphate bond (Lennarz, 1975; Waechter and Lennarz, 1976; Parodi and
Leloir, 1979) and followed by the transfer of the oligosaccharide from the precursor
to the protein. Oligosaccharide intermediates destined for protein incorporation are
synthesized by a series of transferases on the cytoplasmic side of the ER while linked
to the dolichol lipid. Following the addition of a specific number of mannose and
glucose molecules, the orientation of the dolichol precursor and its attached glycan
translocate to the lumen of the ER where further enzymatic modification occurs
(Figure 2.7). The completed oligosaccharide is then transferred from the dolichol
precursor to the Asn of the target glycoprotein which catalyzed by a high specific
enzyme, the dolichol pyrophosphoryl oligosaccharide polypeptide
oligosaccharyltransferase (Kaplan et al., 1987). Three glucose and one mannose
residues are then removed by two glucosidases (glucosidases I and glucosidases II)
and by a mannosidase located in the membrane of endoplasmic reticulum. Further
processing includes trimming of residues such as glucose and mannose, and addition
of new residues via transferases in the ER and, to a great extent, in the golgi (Figure
2.7). In the golgi, high mannose N-glycans can be converted to a variety of complex
and hydrid forms which are unique.
21
Figure 2.7: Pathway for generation of the dolichol-linked oligosaccharide donor for
protein N-glycosylation. The first reaction is the transfer of an N-acetylglucosamine-
phosphate from UDP-N-acetylglucosamine to a dolichol phosphate. Then, one N-
acetylglucosamine and five mannose residues are added to this product from UDP-N-
acetylglucosamine and GDP-mannose, respectively. Finally, the complete molecule
is obtained by the addition of four mannosyl three glucose from dolichol-phosphate-
mannose and three glucoses from dolichol-phosphate-glucose (Abeijon and
Hirschberg, 1992).
2.3.2.1 Glycosylation Pathway in Insect Cell
The nature of N-linked glycosylation is dependent on the protein expressed
and the host cell line. Insect cells, like other eukaryotic cells, modify many of their
proteins by N-glycosylation. At the early stage, N-glycosylation in insect cells is
similar to that in mammalian in ER and form Man9GlcNAc2 moiety. Then, this
moiety is usually trimmed to shorter oligosaccharide structures of Man3GlcNAc2 by
exoglycosidases and a glycosyltransferase. Man3GlcNAc2 is the common
intermediate to both mammalian cells and insect cells. In mammalian cells, terminal
glycosyltransferases can elongate this common intermediate to produce hybrid and
complex N-glycans with terminal sialic acids. However, N-glycan processing
22
machinery of insect cell generally does not produce complex, terminally sialylated
N-glycans. In contrast, they have insufficient expression of processing enzymes
including glycosyltransferases responsible for generating complex-type structures
and metabolic enzymes involved in generating appropriate sugar nucleotides. In
some cases, insect cells have a competing exoglycosidase that can remove the
terminal N-acetylglucosamine residue from GlcNAcMan3GlcNAc2-N-Asn. Hence,
the majority of processed N-glycan produced by insect cells is usually the one with
paucimannosidic structure, Man3GlcNAc2-N-Asn (Hollister et al., 2002; Betenbaugh
et al., 2004; Figure 2.8).
Figure 2.8: 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).
23
A lot of efforts have been done to modify the glycosylation pathway in insect
cells. There are reports that mentioned the treatment of several established insect
cell lines with a β-N-acetylglucosaminidase inhibitor (Watanabe et. al., 2002) and
culture of the cells in the presence of the sialic acid precursor, N-acetylmannosamine
(Joshi et al., 2001) allowed the production of recombinant glycoproteins with
terminally sialylated N-glycans. Co-infection of recombinant baculovirus expressing
the mammalian β1,4-galactosyltransferase and α2,6- sialyltransferase genes (Jarvis
et al. 2001) and the genetically transform insect cell lines with the required-
glycosyltransferases (Breitbach and Jarvis, 2001; Hollister and Jarvis 2001; Joosten
and Shuler 2003; Aumiller et al., 2003; Yun et al., 2005) are able to express
recombinant glycoproteins containing sialic acid residues. CMP-NeuAc metabolic
pathway also has been engineered to produce CMP-NeuAc which is the crucial
substrate for sialylation of glycoproteins (Lawrence et al., 2001; Viswanathan et al.,
2005). In conclusion, engineered insect cell-BEVS is capable of producing complete
glycoprotein.
2.3.3 Model Protein- Transferrin
Transferrin is the major iron-carrier protein in human plasma and
extracellular space in tissues (von Bonsdorff, L et al., 2001). Transferrin is the most
important source of iron for red cells (Ponka, 1997) and erythroid progenitor cells in
the bone marrow. Transferrin can be divided into four main members: the serum
transferrins (STf) from blood stream, the lactoferrins, found in milk, tears and other
bodily secretions of numerous mammals, the ovotransferrins, found in avian egg
white, and the melanotransferrins, found on the surface of melanocytes (Bullen et al.,
1999). Serum transferrin has a role in iron transport around the body.
Ovotransferrin may help protect the developing embryo in the semi-permeable egg
by sequestering iron that microbes need to grow. Lactoferrin can act as a site-
specific DNA binding protein. Transferrin exists as an extracellular protein (He and
24
Furmanski, 1995) and all the members have similar polypeptide folding patterns
(Baker and Lindley, 1992).
Human serum transferrin is a single chain glycoprotein of 679 amino acid
residues, with 19 disulphide bridges, 2 homologous lobes, two asparagines linked
glycan chains and a glycosylation dependent molecular mass in the range of 76±81
kDa (MacGillivray et al., 1982; MacGillivray et al., 1983). The two homologous
lobes, N-lobe and C-lobe of about 330 amino acids which were linked by a short
flexible spacer peptide, contain two dissimilar domains divided by a cleft which is
the binding site for Fe3+
(Bailey et al., 1988; Wang et al., 1992; Figure 1). At the
iron binding site, four of the six Fe3+
co-ordination sites are occupied by the protein
ligands (2 tyrosine, 1 histidine and 1 aspartate residue) and two by the bidentate
carbonate anion (Bailey et al., 1988; Hirose, 2000). Two N-linked oligosaccharides
are found in the C-lobe at aspargine residues Asn413 and Asn611. The glycan
chains are mainly biantennary (85%) and triantennary (15%) complex-type glycans
(Fu and van Halbeek, 1992; Spik et al., 1985). There are 4-6 sialic acid residues per
transferrin molecule. Variation in microheterogeneity of transferrin occurs during
certain physiological and pathological conditions, such as pregnancy, rheumatoid
arthritis, malignancies, alcohol abuse and genetic polymorphism (van Eijk et al.,
1987; de Jong et al., 1990; de Jong et al., 1992; Léger et al., 1989; Yamashita et al.,
1989; Stibler et al., 1978). Anyway, this variation neither influences the secretion
rate of transferrin by hepatoma cells (Bauer et al., 1985) nor the binding of
transferrin to its receptor (Mason et al., 1993).
Transferrin binds iron avidly with a dissociation constant of approximately
1022
M-1
at pH 7.4 (Aisen and Listowsky, 1980). It also capable of binding several
other metals, but with a lower affinity (Harris and Aisen, 1989). Ferric iron couples
to transferrin only in the company of an anion (usually carbonate) that serves as a
bridging ligand between metal and protein (Aisen and Listowsky, 1980; Harris and
Aisen, 1989; Shongwe et al., 1992). Each molecule of transferrin can bind two Fe3+
ions. Upon binding of iron, the lobes undergo a conformational transition from the
apo-structure with an open interdomain cleft to a closed holo-structure (Hirose,
25
2000). Transferrin exists in four iron forms: iron-free apotransferrin, the monoferric
transferrins with iron in the C- or the N-lobe, respectively, and the diferric
holotransferrin (Harris and Aisen, 1989). Under normal condition, all circulating
plasma iron (0.1% of the body iron) is bound to transferrin, and only 20–35%
transferrin is saturated with iron. Transferrin-bound iron which is in redox-inactive
state does not catalyze hydroxyl radical formation (Baldwin et al., 1984).
Transferrin-bound iron is taken up by the cells by receptor mediated endocytosis
(Richardson and Ponka, 1997) whereafter apotransferrin is recycled back to
circulation (Huebers and Finch 1987). Decrease of pH or protonation of the iron
ligands release metal from transferrin. This can be accelerated by other chemical
compounds capable of complexing iron such as pyrophosphates (Morgan, 1979) and
citrate (Gumerov et al., 2003).
Figure 2.9: A ribbon diagram of a diferric rabbit serum transferrin molecule. The
arrow indicates the position of the Fe3+
molecule in the inter-domain cleft in the N-
lobe (Hall et al., 2002).
Transferrin acts as chelating agent, which renders iron soluble under
physiologic conditions and facilitates transport of iron into cells (Lee et al., 2006). It
also plays role as an antioxidant (Chauhan et al. 2004) and anti microbial protein
which prevents iron-mediated free radical toxicity by controlling the level of free
iron and keeps the iron inaccessible from most bacteria and fungi (Weinberg, 1984).
26
Iron binding capacity of transferrin of patients undergoing high dose chemotherapy
(Harrison et al., 1994; Beare and Steward, 1996), myeloablative therapy and bone
marrow stem cell transplantation (Bradley et al, 1997; Sahlstedt et al., 2001) is
exceeded. Patients with leukaemia and other malignancies typically have a low
serum transferrin concentration. Administration of iron-free apotransferrin would be
a better alternative (von Bonsdorff et al., 2001) over clinically used iron chelator,
deferoxamine which has limited efficacy in the binding of non-transferrin-bound iron
and displays dose-related toxicity (Porter et al., 1996).
Table 2.3: Selected private company with the protein engineering platform (Haan
and Maggos, 2004).
Company Plattform
Affibody Uses protein scaffold based on a domain in Protein A to develop antibody-like
molecules
Ambrx Adds non-encoded amino acids to proteins, enabling the synthesis of proteins with
chemical diversity
BioRexis Uses protein scaffold based on transferrin to develop antibody-like molecules, make
fusion proteins and receptor agonists
Borean Uses protein scaffold bases on a C-type lectin to develop antibody-like molecules,
protein trimerization technology
Catalyst Engineers proteases to degrade targeted molecules
Compound
Therapeutics
Uses a fibronectin domain to develop antibody-like molecules; creates bi- functional
proteins with target-binding domain linked to enzymatic domain
KaloBios Develops improved methods for antibody humanization
Pleris Uses protein scaffold based on lipocalin to develop antibody-like molecules
Scil Uses protein scaffold based on gamma-crystallin to develop antibody-like molecules
Selecore Uses protein scaffold based on cysteine knots to develop antibody-like molecules
Trubion Engineers desired effector function into its SMIP antibody-like proteins
Xencor Uses its PDA technology to engineer desired effector function into antibodies and
create dominant-negative proteins and proteins with enhanced properties,
Transferrin also plays an important role in protein engineering. Transferrin
molecules which have multiple surface loops have excellent stability profile and
show non immunogenic behaviour are suitable to be used as scaffold or carrier
protein. Non-glycosylated transferrin has a half life of 14-17 days. A transferrin
27
fusion protein will similarly have an extended half life, provides high bioavailability,
biodistribution and circulating stability (Haan and Maggos, 2004). An alternate
monoclonal antibody (MAbs) using transferrin as scaffold and was produced in a
yeast expression, Trans-bodyTM
had been developed by BioRexis. Glucagons-like
peptide 1 (GLP-1) was also fused to tranferrin as an alternative to Exenatide, the first
GLP-1-R agonist compound for treating diabetes. This product requires less
frequency of parenteral injection time.
2.3.3.1 Recombinant Human Transferrin
The expression of a wide range of human serum transferin (hSTf) variants:
recombinant full-length and the truncated protein are important for mutagenesis
studies (Ali et al., 1996) and for the study of factors affecting mechanism of
homeostasis (Mason et al. 2001). Different expression systems have been applied to
produce recombinant transferrin. The most common system, which uses baby
hamster kidney cells (BHK) (Mason et al. 2001), has been shown to be successful
although the yield is relatively low. The expressed recombinant hTf was comprised
of numerous glycoforms (Mason et al., 1993). Bacterial expression systems have
been reported (Ikeda et al., 1992; Steinlein and Ikeda, 1993; de Smit et al., 1995).
Escherichia coli-expressed hSTf is biologically inactive, largely due to incorrect
intramolecular disulphide bond formation (Ikeda et al. 1992; de Smit et al., 1995).
Functional hSTf N-lobe was efficiently produced using methylotrophic yeast,
Pischia pastoris with a satisfactory high yield but lack of the full length protein
(Steinlein et al. 1995).
Recombinant hSTf expressed from insect cell-BEVS has the same structural
conformation and biological activity as native hSTf (Ali et al., 1996). The
recombinant protein can bind two ferric ions in the presence of bicarbonate, and is
actively taken up by receptor-mediated endocytosis (Ali et al., 1996). Study of
28
Lopez, 1997 showed that, both human and bovine lactoferrin expressed in Mamestra
brassiase cells lack of complex or hybrid structures. The glycan structures of
recombinant hTf expressed in Tn-5B1-4 cells (Ailor et al. 2000) consists of 54%
paucimannosidic, 30.8% high-mannose and 13.9% hybrid glycans with over 50%
containing fucose.
2.4 Analysis Method
2.4.1 Bicinchoninic Acid (BCA) Assay
BCA is a protein quantitation method based on colorimetric detection. The
principles of total protein method can be divided into protein-dye binding chemistry
(coomassie/Bradford) and protein-copper chelation chemistry. A rapid method of
determining the existence of protein is absorbance at UV 280nm. A few assays like
Bradford assay, Lowry assay and Bicinchoninic acid (BCA) assay with different
specifications and sensitivities are the most popular protein methods
BCA assay is a protein-copper chelation assay. BCA Protein Assay
combines the reduction of Cu2+
to Cu1+
by protein in an alkaline medium with the
highly sensitive and selective colorimetric detection of the cuprous cation (Cu1+
) by
bicinchoninic acid (Figure 2.10). Protein chelates the copper in an alkaline
environment and forms a blue colored complex. This is also known as the biuret
reaction. The color development reaction is started when two molecules of BCA™
reagent chelate with one cuprous ion (Cu1+
) (Figure 2.10) and formed a purple
colored product. This water soluble BCA/Copper Complex exhibits a strong linear
absorbance at 562 nm with increasing protein concentrations. The presence of any of
four amino acid residues (cysteine or cystine, tyrosine, and tryptophan) in the amino
acid sequence of the protein strongly influenced the formation of color. The rate of
29
BCA Color Formation is dependent on the incubation temperature, the types of
protein present in the sample and the relative amounts of reactive amino acids
contained in the proteins. Linear working range for BCA assay at 37°C is 20µg/ml-
2000µg/ml and 60°C is 5µg/ml-250µg/ml.
Figure 2.10: Reaction schematic for BCA assay (Pierce Biotechnology, 2005).
2.4.2 Enzyme Linked Immunosorbent Assay (ELISA)
ELISA is a sensitive enzyme immunoassay which has been widely used for
diagnostic purpose. It is a simple and economic analytical method which allows the
use of small volume and avoided the troublesome of separation. ELISA can rapidly
analyze a large number of samples with high sensitivity and precise estimation of
biological parameters. It has been applied in detection and identification of disease
agents; discrimination of disease agents; quantification of agent to estimate parasite
or immunogenic protein loaded in vaccines. Basically, the mechanism of ELISA
involves the immunological reaction of antibodies and antigens, detection of enzyme
linked antibody/antigen and color change of soluble substrate by the enzyme activity.
ELISA can be classified as direct ELISA, indirect ELISA, sandwich ELISA and
competition ELISA (Figure 2.11). The result of ELISA is a color reaction that can
30
be observed visually and read rapidly by multichannel spectrophotometer or ELISA
plate reader.
Figure 2.11: Schematic represents the a) Direct ELISA; b) Indirect ELISA; c)
Sandwich ELISA; d) Competition ELISA. a) Antigen is attached to the solid phase
and detected by enzyme-labeled antibodies. After incubation period and washing,
the substrate system is added and the color is allowed to develop. b) Antibodies
from a particular species react with antigen attached to the solid phase. Any bound
antibodies are detected by addition of an antispecies antiserum labeled with enzyme.
This system is widely used in diagnosis. c) This system exploits the antibodies
attached to the solid phase to capture antigen. This is then detected using an
enzyme-labeled serum specific for the antigen. The detecting antibody can be the
same serum or from different sources. The antigen must have at least two different
antigenic sites. d) The test scheme involves the reaction of two antibodies with an
antigen attached to the solid phase. Competition implies simultaneous addition of
reagents. The degree of inhibition by binding of antibodies contained in sample for a
pretitrated enzyme labeled antibodies reaction is determined (Crowther, 1995).
31
2.4.3 Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-
PAGE)
SDS-PAGE is a technique used to separate proteins according to their
electrophoresis mobility. During eletrophoresis, protein molecules will generally
migrate in a direction and at a speed that reflects their size and net charge. The
folding pattern of protein molecules would not affect the mobility because they have
been linearized and form a complex with negatively charged molecules of sodium
dodecyl sulfate (SDS) (Figure 2.12). Therefore, protein molecules migrate as a
negatively charged SDS-protein complex through the porous polyacrylamide gel. A
reducing agent (mercaptoethanol) would break any –S–S– linkages in or between
proteins. Under these conditions, proteins migrate at a rate that reflects their
molecular weight. The migration is proportional to the molecular weight with
formula as below:
log (MASS) = k (Migration Distance) (2.1)
The bands of the separated proteins can only be visualized after stained. Different
staining method with different sensitivity like Coomassie blue staining, silver
staining, zinc staining is available.
(a) (b)
Figure 2.12: SDS-PAGE. (a) Folded single unit protein or protein with 2 subunits
will be denatured, linearized and became single strand negative charge SDS-protein
molecules after heated with SDS and mercaptoethanol. (b) Treated sample (SDS-
32
protein molecules) which is loaded into the well of SDS-Polyacramide gel will
mobilize from upper part of gel to lower part of the gel according to the molecular
weight. Light molecules will move faster than the heavy one.
2.4.4 Western Blot
Western blot or immunoblot is a method to detect a specific protein in a given
sample of tissue homogenate or extract by means of antigenicity and molecular
weight. Proteins are first separated by mass in the SDS-PAGE, and then specifically
detected in the step of immunoassay. The proteins are transferred from
polyacrylamide to a membrane (typically nitrocellulose or PVDF) prior detection.
The immobilization of protein on membranes matrix is preferred than
polyacrylamide gel because the proteins are more accessible, easier to be handled,
smaller amounts of required reagents and shorter processing time (Gershoni and
Palade, 1982). Principle of the immunoassay of Western blot is similar to ELISA
which involves immunological reaction and detection of enzyme linked
antibody/antigen. Various probes are available for the detection of antibody binding,
for example: conjugated anti-immunoglobulins, conjugated staphylococcal Protein
A, which binds IgG of various species of animal and biotinylated primary antibodies.
The applied chromogenic substrate is different from what is used in ELISA method
in which it involves immunoprecipitation instead of showing color change in
soluable solution. Besides chemiluminescent substrates, other possibilities for
probing are fluorescent and radioisotope labels.
33
2.4.5 Glucose, Lactic Acid and Glutamine Analyzer
The main principle of the analyzers is the application of enzyme sensor
technology. The technology is fast and gives accurate measurement. Enzyme sensor
technology employs specific enzyme to catalyze reactions to produce hydrogen
peroxidase. Hydrogen peroxidase is electrochemically oxidized at the anode to
produce signal current which would convert to concentration value base on single
point calibration. The membrane of the enzyme sensor contains three layers (Figure
2.13). The first layer is porous polycarbonate which limits the diffusion of the
analyte to enzyme layer to prevent enzyme-limited reaction. Oxidization of the
analyte takes place once the analytes enters the enzyme layers. Different type of
enzymes is applied for detection of different substance. The specific enzyme
reaction is show as below:
Dextrose + O2 →OxidaseGlu
H2O2 + D-Glucono-δ-Lactone…………………….(2.2)
L-lactate + O2 →− OxidaseLacL
H2O2 + Pyruvate…………………………………(2.3)
L-Glutamine + O2 →aseGluta min
L-Glutamate + NH3……………………………(2.4)
L-Glutamate + O2 →OxidaseGlut
H2O2 + α-Ketoglutarate + NH3……………….(2.5)
H2O2 + O2 →anodePlatinum
2H+ + O2 + 2e
-………………………………………(2.6)
The third layer, cellulose acetate is used to eliminate many electrochemically-active
compounds that could interfere with the measurement and permits only small
molecules, such as hydrogen peroxide, to reach the electrode.
Figure 2.13: Immobilized enzyme biosensor of YSI (adapted from YSI, 2001).
34
2.4.6 Carbohydrate Analysis Using High Performance Liquid
Chromatography (HPLC)
2.4.6.1 Hydrolysis
Accuracy in monosaccharide composition analysis of oligosaccharide and
glycoprotein relies to a large extent on effective hydrolysis. Fu and O’Neill (1995)
have studied in detail the hydrolysis of free N-linked oligosaccharides and intact
glycoproteins at 121°C under various conditions. Figure 2.14 shows that hydrolysis
of fetuin at 121°C with 4N TFA was completed after 3hrs of hydrolysis.
Figure 2.14: Hydrolysis time course of bovine fetuin. Fetuin sample, 0.5mg in 4N
TFA (5ml) was hydrolyzed at 121°C (Fu and O’Neill, 1995).
2.4.6.2 1-Phenyl-3-Methyl-5-Pyrazolone (PMP) Derivative of Sugar
A number of reactions have been reported as pre-column derivatization. The
condensation between the active hydrogen of PMP or 1-(p-methoxy)-phenyl-3-
methyl-5-pyrazolon (PMPMP) with the carbonyl group of the reducing
carbohydrates under slightly basic conditions, resulting in bis-PMP and bis-PMPMP
35
derivatives. The bis-PMP-sugars which have no stereoisomers are used for
component sugar analysis (Honda et al., 1989). The procedure requires slightly
alkaline conditions (pH8.3). PMP reacts with reducing carbohydrates almost
quantitatively under mild reaction conditions without epimerization to yield strongly
UV-absorbing (245 nm) and electrochemically sensitive derivatives. This method is
attractive for the sialylated oligosaccharides because no loss of sialic acid occurs.
The detection limit is 1-0.1 pmol (Honda et al., 1980).
Figure 2.15: Derivatization with pyrazolone derivatives (Hase, 1996)
2.4.6.3 Reverse Phase-HPLC
High-performance liquid chromatography (HPLC) is a form of column
chromatography which is used to separate components of a mixture by using a
variety of chemical interactions between analyte and the chromatography column. It
is also sometimes referred to as high-pressure liquid chromatography. Different type
of HPLC which include normal phase chromatography, reverse phase
chromatography, size exclusion chromatography, ion-exchange chromatography and
bioaffinity chromatography are available. Reversed phase HPLC (RP-HPLC)
consists of a non-polar stationary phase and an aqueous, moderately polar mobile
phase. The stationary phase is a silica bonded with straight chain alkyl group such as
C18H37 or C8H17. RPC operates on the principle of hydrophobic interactions, which
result from repulsive forces between a polar eluent, the relatively non-polar analyte,
and the non-polar stationary phase. Molecules which are more non-polar in nature
36
are retained longer than polar molecules. Retention Time (RT) is increased by the
addition of polar solvent to the mobile phase and decreased by the addition of more
hydrophobic solvent. The retention can be decreased by adding less-polar solvent
(MeOH, ACN) into the mobile phase to reduce the surface tension of water. Mobile
phase modifiers like inorganic salts can causes a moderate linear increase in the
surface tension of aqueous solutions and increase the retention time of analyte.
Another important component influence the retention time is pH. pH can change the
hydrophobicity of the analyte. Most methods use a buffering agent, such as sodium
phosphate, to control the pH. The buffers serve multiple purposes: they control pH,
neutralize the charge on any residual exposed silica on the stationary phase and act as
ion pairing agents to neutralize charge on the analyte.
2.5 Purification of Transferrin
The purity of a protein is a pre-requisite for its structure and function studies
or its potential application. For structure studies or therapeutic applications, protein
of high degree is required. A wide variety of protein purification techniques like gel
filtration chromatography, ion-exchange chromatography, affinity chromatography
and hydrophobic interaction chromatography (HIC), are available. Every separation
technique is important and the application is dependent on target proteins which vary
in biological and physico-chemical properties: molecular size, net charge, biospecific
characteristics and hydrophobicity (Kennedy, 1990; Garcia and Pires, 1993).
Phenyl Sepharose chromatography has been widely used in the purification of
transferrin. Vieira and Schneider, 1993; Choudhury et al., 2002 had used phenyl-
Sepharose CL 4B to purify avian serotransferrin. Testicular transferrin from rat
sertoli cells, rat serum transferrin (Skinner et al., 1984), rhTf from Sf9 (Ali et al.,
1996) and rhTf from Trichopulsia ni cells (Ailor et al., 2000) was also purified using
phenyl sepharose. Ion-exchange chromatography is very popular too. SP and Q-
37
sepharose had been used to purify apotransferrin from human plasma (von Bonsdorff
et al., 2001). Two steps anion chromatography: Q sepharose fast flow and mono Q
had been use to purify ovotransferrin produced from Pichia pastoris (Mizutani et al.,
2004). Steinlein et al., 1995 used Whatman DE52 column to purify N-terminal half
human serum transferrin from Pichia pastoris. Ali et al., 1996 used Q Sepharose as
a second column to purify recombinant human transferrin from Sf9.
Improved recombinant hTf with histidine tagged has employed immobilized
metal affinity chromatography (IMAC) as the main purification method.
Hexahistidine (His)-6 epitope tag hTf from transfected Drosophila melanogaster S2
cells (Lim et al., 2004), His-tagged hTf secreted from transfected BHK (Mason et
al., 2001) were purified using metal chelate column. Transferrin isolated from
Manduca sexta which possesses a large number of histidine residues was purified by
high capacity and low capacity Ni2+
-dipicolylamine (DPA)-Novarose gel
(Winzerling et al., 1995). Affinity colun, anti-hTf-IgG immobilized Sepharose 4
Fast Flow, had been use to purify recombinant human serum transferrin expressed by
Lymantria dispar 652Y cells (Choi et al., 2003) and recombinant His-tagged hTf
from a transformed insect cell line (Tn5b4GalT) (Tomiya et al.,2003).
2.5.1 Hydrophobic Interaction Chromatography (HIC)
Hydrophobic interactions have a great importance in biological systems.
They are the dominant force in protein folding and structure stabilization (Privalov
and Gill, 1988; Dill, 1990a; Murphy et al., 1990; Makhatafze and Privalov, 1995)
and the maintenance of the lipid bilayer structure of biological membranes (Tanford,
1973). Proteins comprise of a number of hydrophobic amino acids, with different
distribution and hydrophobicity. Hence, a specific separation can be possible with
hydrophobic supports or matrices (Ochoa, 1978; Vogel et al., 1983; Lindahl and
Vogel, 1984). Although HIC exploits nonspecific affinities, it has been successfully
38
used for separation purposes as it displays binding characteristics complementary to
other protein chromatographic techniques (Janson and Rydén, 1993).
Many theories about the principle of HIC have been proposed. Porath, 1986,
proposed ‘‘salt-promoted adsorption’’ and suggested a salting-out effect in
hydrophobic adsorption (Porath et al. 1973), which extended the earlier observations
of Tiselius, 1948. Hofstee, 1973 and later Shaltiel and Er-El, 1973 believed that the
mode of interaction between proteins and the immobilized hydrophobic ligands was
similar to the self association of small aliphatic organic molecules in water.
Melander and Horvath, 1977 suggested that hydrophobic interaction accounted for
the by increase in the surface tension of water arising from the structure – forming
salts dissolved in it. Srinivasan and Ruckenstein (1980); Van Oss et al. (1986)
proposed that HIC is due to van der Waals attraction forces between proteins and
immobilized ligands caused by the increase of the ordered structure of water in the
presence of salting out salts.
The commercial availability of well-characterized HIC adsorbents opened
new possibilities for purifying a variety of biomolecules such as serum proteins
(Janson and Låås, 1978; Hrkal and Rejnkova, 1982), membrane-bound proteins
(McNair and Kenny, 1979), nuclear proteins (Comings et al., 1979), receptors
(Kuehn et al., 1980), cells (Hjertén, 1981), and recombinant proteins (Lefort and
Ferrara, 1986; Belew et al., 1991 in research and industrial laboratories. The
principle for protein adsorption to HIC media is complementary to ion exchange
chromatography and gel filtration. HIC can separate the pure native protein from
other forms (Fausnaugh et al., 1984; Regnier, 1987). HIC has also found use as an
analytical tool to detect protein conformational changes.
39
2.5.1.1 Factors Affecting HIC
The main factors affecting HIC are: 1) Ligand type and degree of
substitution, 2) Type of base matrix, 3) Type and concentration of salt, 4) pH, 5)
Temperature and 6) Additives (Amersham Biosciences, 2000).
The type of immobilized ligand determines primarily the protein adsorption
selectivity of the HIC absorbent. HIC contain alkyl or aryl chains of any size, and in
practice, most separation employ phenyl and butyl group. Figure 2.14 showed the
glycidyl ether coupling HIC media, which produces charge free gels and only have
hydrophobic interactions with proteins. At constant substitution, the protein binding
capacities of HIC absorbents, hydrophobicity and the strength of interaction would
increase, but the adsorption selectivity would decrease with increased alkyl chain
length. Increased degree of substitution of immobilized ligand would also increase
the protein binding capacities. At sufficient high degree of ligand substitution or n-
alkyl chain length, the strength of interaction would increase although the apparent
binding capacity of the adsorbent remains constant and the bound solutes are more
difficult to elute due to multi-point attachment (Jennissen and Heilmeyer, 1975;
Rosengren et al., 1975; Lăăs, 1975; Maisano, et al., 1985). The selectivity of a
copolymer support can change even with same type of ligand. The two most widely
used types of support are strongly hydrophilic carbohydrates, e.g. cross-linked
agarose, or synthetic copolymer materials.
40
Figure 2.16: Different hydrophobic ligands coupled to cross-linked agarose matrices
(Amersham Biosciences, 2000).
According to Melander, et al. (1984), the most important parameters that
determine the effect of salt on the retention in HIC are the salt molality and the molal
surface increment of the salt. The effects of salts in HIC can be accounted for
referring to the Hofmeister series for the precipitation of proteins or for their positive
influence in increasing the molal surface tension of water (Figure 2.15, Figure 2.16).
The salts at the beginning of the series promote hydrophobic interactions and protein
precipitation (salting-out or antichaotropic), are considered to be water structuring;
whereas salts at the end of the series (salting-in or chaotropic ions) randomize the
structure of the liquid water and thus tend to decrease the strength of hydrophobic
interactions (Porath, 1987). Salts such as sodium, potassium or ammonium sulfates
are the most effective to promote ligand protein interactions. Magnesium sulphate
and magnesium chloride do not enhance the protein retention despite the fact that
they increase the surface tension of water. Type of salt in the eluent not only altered
the overall retention of the proteins, but also affects selectivity of the separations
(Rippel and Szepesy, 1994).
41
Increasing precipitation (“salting- out”) effect
Anions: PO43-
, SO42-
, CH3·COO-, Cl
-, Br
-, NO3
-, CLO4
-, I
-, SCN
-
Cations: NH4+, Rb
+, K
+, Na
+, Cs
+, Li
+, Mg
2+, Ca
2+, Ba
2+
Increasing chaotropic (“salting-in”) effect
Figure 2.17: The Hofmeister series on the effect of some anions and cations in
precipitating proteins (Amersham Biosciences, 2000b).
Na2SO4>K2SO4>(NH4)2SO4>Na2HPO4>NaCl>LiCl…>KSCN
Figure 2.18: Relative effects of some salts on the molal surface tension of water
(Amersham Biosciences, 2000b).
The concentration of salt strongly influences the selectivity in protein
adsorption and the influence is different and dependent both on the stationary phase
and the buffer salts (Oscarsson and Kårsnås, 1998). In HIC, the use of high salt
concentration on the equilibration buffer and sample solution promotes the ligand–
protein interactions and consequently the protein retention. As the concentration of
such salts is increased, the amount of bound proteins also increases almost linearly
up to a specific salt concentration and continues to increase in an exponential manner
at still higher concentrations. The adsorbed proteins are eluted by step wise or
gradient elution at decreasing salt concentration in the eluent. The viscosity, UV
transparency and stability at alkaline pH values are other important factors in
choosing the neutral salts (Narhi et al., 1989).
In general, an increase in pH weakens hydrophobic interactions (Porath et al.,
1973; Hjertén, S., 1973); a decrease in pH results in an apparent increase in
hydrophobic interactions. This is probably due to changing of charged groups at
different pH and thereby leading to an increase in the hydrophilicity or
42
hydrophobicity of the proteins. Proteins which do not bind to a HIC adsorbent at
neutral pH bind at acidic pH (Halperin et al., 1981). Hjertén et al. (1986) found that
the retention of proteins changed more drastically at pH values above 8.5 and/or
below 5 than in the range pH 5–8.5. These findings suggest that pH is an important
separation parameter in the optimizing the selectivity of hydrophobic interaction
chromatography.
In HIC, increasing the temperature enhances protein retention and lowering
the temperature generally promotes the protein elution (Hjertén et al., 1974). Van
der Waals attraction forces, which operate in hydrophobic interactions (Srinivasan
and Ruckenstein, 1980) increase with increase in temperature (Parsegian and
Ninham, 1970). However, an opposite effect was reported by Visser and Strating
(1975). This apparent discrepancy is probably due to the differential effects exerted
by temperature on the conformational state of different proteins and their solubilities
in aqueous solutions (Amersham Bioscience, 2000).
Additives can be used in HIC, not only to improve protein solubility or to
modify protein conformation, but also to promote the elution of the bound proteins.
The most widely used are water-miscible alcohols (e.g. ethanol and ethylene glycol)
and detergents. Additives decrease the surface tension of water thus weakening the
hydrophobic interactions to give a subsequent dissociation of the ligand-solute
complex. The non-polar parts of alcohols and detergents compete for the adsorption
site to displace the bound proteins. The separation mode involving charged group of
detergent is a mixed ion-exchange hydrophobic interaction process (Janson and
Rydén, 1993). Elution using additive could lead to denaturation of protein. Hence, it
is only applied for cleaning up HIC columns and when other milder conditions do
not promote protein recovery.
43
2.5.2 Ion Exchange Chromatography
Ion exchange is probably the most frequently used chromatographic
technique for the separation and purification of proteins, polypeptides, nucleic acids,
polynucleotides, and other charged biomolecules (Bonnerjera et al., 1986). The
reasons for the success of ion exchange are its widespread applicability, its high
resolving power, its high capacity, and the simplicity and controllability of the
method. Separation in ion exchange chromatography depends upon the reversible
adsorption of charged solute molecules to immobilized ion exchange groups of
opposite charge. Separation is obtained since different substances have different
degrees of interaction with the ion exchanger due to differences in their charges,
charge densities and distribution of charge on their surfaces. These interactions can
be controlled by varying conditions such as ionic strength and pH.
The separation using ion exchange is based primarily on differences in the
ionic properties of surface amino acids. Thus, at a given pH, proteins possess an
overall net charge. The relationship of the protein and the net charge can be
visualized as a titration curve (Figure 2.17). This curve reflects how the overall net
charge of the protein changes according to the pH of the surroundings. The
isoelectric point (pI) of each protein is the pH at which the protein has zero surface
charge. The net charge will be more positive at a pH lower than pI protein; more
negative at a higher pH. Proteins with different pI can be separated by being passed
through chromatofocusing. Selected working pH is 1 unit away from the pI of
protein.
44
Figure 2.19: Effect of pH on protein at different net charge (Amersham Biosciences,
2000b).
2.5.2.1 Factor Affecting IEX
Matrix of IEX may be based on inorganic compounds, synthetic resins or
polysaccharides. The characteristics of the matrix determine its chromatographic
properties such as efficiency, capacity and recovery as well as its chemical stability,
mechanical strength and flow properties. The nature of the matrix will also affect its
behaviour towards biological substances and the maintenance of biological activity.
The first ion exchangers designed for use with biological substances were the
cellulose ion exchangers developed by Peterson and Sober (1956), then Ion
exchangers based on dextran (Sephadex), followed by those based on agarose
(Sepharose) and cross-linked cellulose (Sephacel). Hydrophilic nature of cellulose
has little tendency to denature protein, but it has low capacities and has poor flow
properties due to its irregular shape.
45
An ion exchanger consists of covalently bound charged group to an insoluble
matrix. The charged groups are associated with mobile counter ions which can be
reversibly exchanged with other ions of the same charge without altering the matrix.
Positively charged exchangers have negatively charged counter-ions (anions)
available for exchange and are called anion exchangers; negatively charged
exchangers have positively charged counter-ions (cations) and are termed cation
exchangers (Figure 2.18).
Figure 2.20: Ion exchanger types (Amersham Biosciences, 2000b).
The presence of charged groups is a fundamental property of an ion
exchanger. The type of group determines the type and strength of the ion exchanger;
their total number and availability determines the capacity. Table 2.3 show some
functional groups which have been chosen for use in ion exchangers. Sulphonic and
quaternary amino groups are used to form strong ion exchangers; the other groups
form weak ion exchangers. Strong ion exchangers like sulfo group and quaternary
ammonium (Q) group are completely ionized over a wide pH range whereas with
weak ion exchangers, the degree of dissociation and thus exchange capacity varies
much more markedly with pH. Carboxymethyl (CM) group begin to protonated at
pH below 5, diethylaminoethyl (DEAE) groups become uncharged at pH above
pH8.5. DEAE- and Q- groups are highly charged at low pH, so they also suitable to
purify low pI protein.
46
Table 2.4: Functional groups used on ion exchangers (Amersham Biosciences,
2000b).
Anion Exchangers Functional Group
Diethylaminoethyl (DEAE) -O-CH2-CH2-N+H(CH2CH3)2
Quaternary aminoethyl (QAE) -O-CH2-CH2-N+(C2H5)2-CH2-CHOH-CH3
Quaternary ammonium (Q) -O-CH2-CHOH-CH2-O-CH2-CHOH-CH2-N+(CH3)3
Cation Exchangers Functional Group
Carboxymethyl (CM) -O-CH2-COO-
Sulphopropyl (SP) -O-CH2-CHOH-CH2-O-CH2-CH2-CH2SO3-
Methyl sulphonate (S) -O-CH2-CHOH-CH2-O-CH2-CHOH-CH2SO3-
The pH in the micro environment of an ion exchanger is not exactly the same
as eluting buffer because Donnan effect can repel or attract protons within the
adsorbent matrix. In general, pH in the matrix is up to 1 unit higher than that in the
surrounding buffer in anion exchanger and 1 unit lower in cation exchanger. The
lower the ionic strength of the buffer, the larger the Donnna effect. This
phenomenon is very important considering the stability of enzymes as a function of
pH. The Donnan effect limits the operational pH range of ion exchangers, especially
in the mildly acid range.
The charges, the nature of the matrix particles in terms of bead size, flow rate
required, capacity also determine the choice of adsorbent. Table 2.4 and Table 2.5
show the capacity data and the characteristics of 4 common commercial ion
exchange matrices.
47
Table 2.5: Capacity data for sepharose fast flow ion exchangers (Amersham
Bioscience, 2000b).
Ion Exchanger Q Sepharose
Fast Flow
SP Sepharose
Fast Flow
DEAE
Sepharose Fast
Flow
CM Sepharose
fast Flow
Total ionic capacity
(µmol/ml gel)
180-250 180-250 110-160 90-130
Dynamic binding capacity*
(mg/ml gel)
Thyroglobulin (MW
669000)
HAS (MW 68000)
α-lactalbumin (MW 14300) IgG (MW 160000)
Bovine COHb (MW 69000)
Ribonuclease (MW 13700)
3
120
110
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
50
50
70
3.1
110
100
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
15
30
30
N.D. = Not determined
*For anion exchangers (DEAE and Q) the starting buffer was 0.05 M Tris, pH 8.3
and for cation exchangers (CM and S) 0.1 M acetate buffer, pH 5.0. Limit buffers
were the respective start buffers containing 2.0 M NaCl.
Table 2.6: Characteristics of Q, SP, DEAE and CM Sepharose Fast Flow
(Amersham Biosciences, 2000b).
Product Q Sepharose
Fast Flow
SP Sepharose
Fast Flow
DEAE
Sepharose Fast Flow
CM Sepharose
fast Flow
Type of gel Strong Anion Strong Cnion Weak Anion Weak Cation
Total ionic capacity
(µmol/ml gel)
180-250 180-250 110-160 90-130
Recommended working
flowrate range (cm/hr)
100-300 100-300 100-300 100-300
Approx. mean particle size
(µm)
90 90 90 90
Particle Size Range (µm) 45-165 45-165 45-165 45-165
Working pH Range* 2-12 4-13 2-9 6-10
pH Stability**
Short Term
Long Term
1-14
2-12
3-14
4-13
1-14
2-13
2-14
4.13
* working pH range refers to the pH range over which the ion exchange groups
remain charged and maintain consistently high capacity.
** pH stability, long term refers to the pH interval where the gel is stable over a long
period of time without adverse effects on its subsequent chromatographic
performance. pH stability, short term refers to the pH interval for regeneration and
cleaning procedures
48
2.5.3 Optimization Method in Process Chromatography
Method development work has to focus on purifying the product of interest to
the highest yield and the required purity as quickly, cheaply and easily as possible.
Process chromatography can be divided into capture, intermediate purification and
polishing, based on chromatography purpose. The objective of the capture step is to
adsorb and isolate the protein of interest quickly from the crude sample from other
critical contaminants. It is designed to maximize capacity or speed at the expense of
some resolution. Intermediate purification is used to remove most of the significant
impurities. In this step, achieving resolution of similar components and good
capacity are both important, optional balance should be decided. Polishing step is
focus on achieving the highest possible resolution and removes most impurities
except for some trace amounts. In any chromatography steps, resolution, speed,
capacity and recovery are the important performance properties which need to be
adjusted. The relative priority of these properties is similar for HIC, ion exchange,
gel filtration and affinity chromatography but different depending on the particular
purposes.
In capture step, the prime consideration when optimizing capture step is to
find the highest possible sample load over the shortest possible sample application
time with acceptable loss in yield. The applied media should offer high speed and
high capacity. Selectivity during sample absorption would affect loading capacity.
Typically, binding conditions are selected to avoid binding of contaminating
substances and increase the availability of the protein of interest. Step wise elution is
often applied. Optimization of pH and elution condition is necessary to remove
critical contaminants, provided loading capacity is not affected. For IEX
chromatography, pH buffer is adjusted as near to value isoelectric point (pI) of
protein of interest without losing the capacity and yield. Dynamic capacity of a
chromatography adsorbent for HIC and IEX is a function of the linear flow rate.
Sample loading capacity must be checked at different flow rates to reveal the
optimum level that gives highest productivity. Significant increase in flow rate
during sample application will always give a decrease in dynamic binding capacity.
49
Besides that, the salt concentration during sample application for HIC should not be
too low since this will have a negative effect on dynamic binding capacity.
If a column is used as an intermediate or polishing step, the resolution is
maximized by working on the eluting conditions such as gradient shape, gradient
slope or concentration and volume of steps in a step-wise elution procedure to meet
the purity requirement. Shallow gradients or even isocratic elution is applied in
polishing steps. In many intermediate steps and always in polishing step, the sample
loading capacity is limited by the required resolution since certain bed height is
required to achieve separation between closely related substances. High efficiency
media with small bead size is important in a typical polishing situation because
working on the selectivity alone may not achieved required resolution. Besides that,
the flow rate during elution will affect resolution between compounds to be separated
and also the concentration of these compounds. In HIC, as in ion exchange
chromatography, sample load, flow rate and gradient volume are interrelated. Flow
rate and sample load are optimized to find highest possible productivity where
resolution is still high enough to meet the predefined purity requirements. Increased
flow rate will give a decrease in resolution, but this decrease will be negligible at
high sample loadings. In laboratory separations especially in polishing step, the best
possible separation is frequently a major consideration and the flow rate is frequently
traded off against improved resolution.
2.6 Summary of Literature Review
Literature review gives a brief insight into some closely related topics of the
research, which are expression system, characteristics of model protein, analytical
methods and chromatographic purification. Advantages and disadvantages of
various recombinant protein expression systems which include E. Coli, mammalian
cell, yeast and insect cell are discussed and compared. Review about insect cell-
50
BEVS which include insect cell, baculovirus, in vivo and in vitro replication and
recombination and knowledge about glycosylation which comprises of type of
glycosylation and glycosylation pathway have given a better understanding of the
recombinant protein expression pathway and subsequently lead to a better
manipulation of the system. The model protein, transferrin, is reviewed in terms of
the functionality, structure and physiology. Principle of the analysis methods which
include BCA, ELISA, SDS-PAGE, Western Blot, biochemical analyzer and
carbohydrate analysis using HPLC are illustrated. Chromatographic purification
methods of transferrin are summarized. HIC and IEX are discussed in detail.
Information about optimization methods in process chromatography provides a clear
guideline for the optimization of chromatography, which is a highlighted study in
this research. In summary, the literature review has fully supported the research by
giving both overall picture and sufficient technical knowledge for the
‘chromatographic purification strategies for recombinant human transferrin from
Spodoptera frugiperda’.
CHAPTER 3
MATERIALS AND METHODS
3.1 Materials
3.1.1 Cell lines and Recombinant Baculovirus
Spodoptera frugiperda (Sf9) insect cell was purchased from ATCC
(Rockville, MD). The recombinant baculovirus carrying human transferrin gene was
provided by Prof Dr Michael J. Betenbaugh of Johns Hopkins University, USA.
3.1.2 Equipments
The electrophoresis system used was Mini-Protean II from Bio-Rad
(California, USA). Western blot analysis was done using Mini Trans-Blot
Electrophoretic Transfer Cell from Bio-Rad (Melville, NY). Shimadzu UV-160
spectrophotometer (Minnesota, USA) was used to measure absorbance at 450 nm
52
and 562 nm. The Minitec bench top shaker was purchased from The Tec family
(Switzerland). Class II type biological safety cabinet was from Muaire (Minnesota,
USA). Inverted phase contrast microscope and transmitted light microscope were
from Zeiss Instruments (Germany). Hemacytometer was purchased from Fortuna
(Wertheim, Germany). Incubator was purchased from Memmert (Germany). Fixed
rotor centrifuge and swinging bucket centrifuge were purchased from Selecta
(Germany). Biochemical analyzer YSI 2700 SELECT from YSI Life Sciences
(Ohio, USA) was used to analyze glucose, lactate and glutamine contents. Econo
Column 1.5 x 15 cm and Econo-Column® Flow Adaptor and fraction collector from
Waters (Japan) and peristaltic pump from Pharmacia Biotech (Sweden) were used for
chromatography. Micro 22R centrifuge from Hettich (Germany) and Amicon Ultra-
4 centrifugal filter from Milipore (Ireland) was used to concentrate purified rhTf.
The high performance liquid chromatography (HPLC) systems used was Series 200
from Perkin Elmer (USA)
3.1.3 Chemicals
Fetal bovine serum (FBS), penicillin-streptomycin and Sf-900 II serum free
media (SFM) were from GIBCO BRL (Gaithersburg, MD). Calibrator-human
reference serum, goat anti-human transferrin-affinity purified, goat anti-human
transferrin-horseradish peroxidase (HRP) conjugate, 0.02% peroxidase in citric acid
buffer and 3,3’,5,5’-tetramethylbenzidene (TMB) peroxidase substrate were obtained
from Bethyl Laboratories Inc (Texas). Broad range protein molecular marker and
TMB stabilized substrate for HRP (water insoluble) were purchased from Promega,
(Madison, WI). Acrylamide, bis-acrylamide, bovine serum albumin (BSA),
ammonium persulfate, citric acid, dimethyl sulphoxide (DMSO), glycine, glucose,
glutamine, lipid 1000x, 2-mercaptoethanol, N,N,N',N'-tetramethylethylenediamine
(TEMED), tris, triton X-100 and trypan blue, glucose Sigma Ultra and ammonium
acetate were purchased from Sigma (Missouri, USA). Other products purchased
from Sigma include galactose Sigma Ultra and phenyl-3-methyl-5-pyrazolone (PMP)
53
(Germany), D-glucosamine hydrochloride (China) and mannose (India). Acetic acid,
bromophenol blue, ethanol, dichloromethane sodium chloride, 38% formaldehyde,
hydrochloric acid, glycerol, methanol, phosphoric acid, potassium chloride,
potassium dihydrogen phosphate, potassium phosphate dibasic, skimmed milk,
sodium bicarbonate, sodium dodecyl sulfate (SDS), sodium hydroxide, tween 20,
were from Fluka (Missouri, USA). Ammonium hydroxide, ammonium sulphate and
trifluoroacetic acid were purchased from Merck (New Jersey, USA). Silver nitrate
was purchased from Unilab (Sydney, Australia). Ammonia enzymatic-UV test kit
which consists of Nicotinamide adenine dinucleotide phosphate (NADPH), α-
ketaglutarate, triethanolamine, glutamate dehydrogenase (GLDH), and ammonia
were from Randox Laboratories (Antrim, UK). BCA protein assay kit which
consists of CuSO4•5H2O, BCA™ and tartrate in an alkaline carbonate buffer and
albumin standard were from Pierce (Illinois, USA). D-Glucose, L-Lactate, L-
Glutamine and L-Glutamate calibrator were from YSI laboratory (Ohio. USA). Q-
Sepharose Fast Flow and Phenyl Sepharose 6 Fast Flow (high sub) were purchased
from Amersham Biosciences (Sweden)
3.2 Spodoptera frugiperda (Sf-9) Cells Culture
3.2.1 Cells Thawing
A vial of cells was taken out from the liquid nitrogen facility. The cells were
thawed rapidly, by holding the vial in a water bath at 37°C. The vial was thoroughly
sprayed with ethanol 70% and opened in laminar hood. A small amount of growth
medium, SFM900II/10% FBS was dropped into the vial. Then, the cells were
pipetted out and gently added to 10ml of 10% FBS containing growth medium in a
centrifuge tube. Immediately, the cells viability was checked. A new vial would be
thawed, if there were more dead cells than healthy one. Then, the cells were
centrifuged at 1000rpm, 5min and resuspended in 10ml, 10% FBS & 1% antibiotic
54
containing growth medium. The cell suspension was equally split into two, 25cm2 T-
flask and 5ml each. The cells were incubated at 27°C for about 1 to 2 hours until
they became attached to the surface. The DMSO containing medium was removed
and replaced with fresh growth medium consisting of 10% FBS and 1% antibiotic.
The cells were incubated until they formed a confluent monolayer.
3.2.2 Cells Count
The T-flask was knocked gently to dislodge the adherent and semi adherent
cells into suspension. 100-200 µl of cell suspension was removed under sterile
condition. Then, equal volume of Trypan Blue was added and mixed to the cells
gently. Next, the hemacytometer was cleaned with tissue paper. The cover-slip was
moistened with exhaled breath. Then, the cover-slip was then slid over the chamber
back and forth using slight pressure until Newton’s refraction rings appear (Newton’s
refraction rings are seen as rainbow-like rings under the cover-slip). Both sides of the
chamber (approx. 5-10 µl) were filled with cell suspension. Then, the
hemacytometer was viewed under a light microscope using 20X magnification.
Viable cells were seen as bright cells and the non viable cells were stained blue. The
numbers of viable and non-viable cells within 8 major squares of the chambers were
counted. The concentration of viable and non-viable cells and the percentage of
viable cells were calculated using the equations below.
Viable cells concentration (cells/ml)
= (total non stained cells within 8 major squares) x (dilution factor) x 104/8 3.1
Percentage Viability
= 100 x numbers of viable cells / total numbers of cells. 3.2
55
3.2.3 Adapting Serum Contain Culture to Serum Free Culture
Cells were adapted to serum free medium after a few passages. Confluent
cells were split 1:1 during each passage of subculture. At the next routine passage,
the cells were transferred into a 75% growth medium/10% FBS and 25% serum-free
medium. The cells were allowed to become confluent. At the following passage, the
cells were transferred into an equal mixture of serum-containing medium and serum
free medium. The cells were allowed to grow to confluence. This took 2 to 3 days.
The previous step was repeated if growth was slow. For the next passage, the cells
were transferred into a 75% serum-free medium and 25% serum-consisting medium.
Finally, the cells were split and transferred into 100% serum free medium after
reaching confluence. The cells would take another two to three passages to grow to
optimum densities.
3.2.4 Adapting Monolayer Cells to Suspension Culture
Insect cells were dislodged from the bottom of flasks. Confluent cells from
two units of 75cm2 T-flask would be sufficient to initiate a 50ml suspension culture.
After cell count, cell suspension was diluted to 5x105cells/ml in serum free growth
medium. Suspension culture was maintained in a shaker flask or a spinner flask.
Stirring rate for shaker flask and spinner cultures was started at 100rpm and 75rpm
respectively. Culture condition was maintained at 27°C±5°C. The cells were
subcultured when viable cell density reached 1-2x106cells/ml. Stirring rate was
increased by 5-10rpm with subsequent passage until constant stirring speed reached
130-150rpm for shaker culture and 90-100rpm for spinner culture. If the viability
dropped below 75%, stirring speed would be decreased by 5rpm for one passage till
the culture viability recover to 80% and above.
56
3.2.5 Maintaining Suspension Culture
Insect cell culture was incubated at 27°C in a non CO2 aerated incubator for
both adherent and suspension cultures. Generally, suspension culture was
subcultured twice weekly, centrifuged at 150g (1000rpm) for 5min, and resuspended
in fresh medium once every 3 weeks. For each subculture, confluent cells (2-
3x106cells/ml) were diluted to 5x10
5cells/ml in serum free medium. Stirring rate
maintained at 130rpm-150rpm for shaker and 90rpm-100rpm for spinner flask.
Suitable volume for respective flask size was shown in Table: 3.1. The caps of flask
were loosen to about ¼ to ½ of a turn to maintain the aeration of cultures.
Table 3.1: Culture volume for different flask size.
Flask Size (ml) Shaker Flask Culture
Volume (ml)
Spinner Flask Culture
Volume (ml)
125
250
500
1000
3000
25-50
50-125
125-200
200-400
400-800
50-100
150-200
200-300
300-1000
2000-3000
3.2.6 Preparation of Optimized Medium
Optimized medium was SFM900II added with 2211.2mg/ml of glutamate,
1291.95mg/ml of glucose and 0.64% (v/v) of lipid mixture 1000x. Powder of
glutamate and glucose were dissolved in SFM900II and filtered with nitrocellulose
membrane, 0.22µm. Original stock of glutamate and glucose was prepared in 25g/l
respectively. A defined volume of optimized medium was prepared by mixing the
calculated volume of glucose solution, glutamate solution, lipid mixture and
SFM900II.
57
3.2.7 Adapting Suspension Culture in SFM900II to Optimized Medium
When the suspension culture has reached more than 2x106cells/ml, the culture
was split and optimized medium in equal volume was added in. Then, the cells were
allowed to become confluent again. The culture was split again and equal volume of
optimized medium was added in. This step was repeated for few passages. Finally,
the suspension culture was centrifuged then transferred into 100% optimized
medium. The culture was always seeded at ≈1.0 x 106 cells/ml when optimized
medium was used as the growth medium. Lower seeding densities in optimized
medium would cause the denaturation of cells.
3.2.8 Cells Freezing
Cell density and viability were determined to ensure only cells with density
of at least 1x106cells/ml and 90% viable were used for freezing. The cells were
centrifuged at 1000rpm for 5 min at room temperature, and resuspended in fresh
growth medium /10%FBS. The cell concentration was adjusted to 1-2 x107cells/ml
with fresh growth medium/10%FBS. A freezing mixture which consisting of 80%
(v/v) growth medium/10% FBS and 20% (v/v) DMSO was prepared and chilled to
4°C. An equal volume of freezing mixture was added to the Sf9 cell suspension and
was quickly mixed until homogeneous. 1ml aliquots were placed into each
cryogenic vial. Immediately, the cryogenic vials were placed upright in a 4°C
refrigerator for 15 minutes, -20°C freezer for 1-2 hours and then -70°C to -80°C
freezer for 4-6 hours, or overnight. Finally, the vials were transferred promptly to
the liquid nitrogen storage facility. All frozen cell stocks were recorded.
58
3.3 Recombinant Baculovirus
3.3.1 Generating Pure Recombinant Virus Stock
Purified rhTf virus stock was obtained from Ongkudon (2006). Method of
end point dilution was applied to generate pure virus. The virus stock has to be
diluted until only 10% or less of the total cultures are infected. Cells were diluted to
a concentration of 1-2.5x105cells/ml with growth medium. Virus stock of 10
-6 and
10-7
dilution were prepared. 10µl of each dilution was mixed with 100µl of cell
suspension and seeded into each well of a 96 well plate. 40 replicates were tested for
each dilution of virus. 4 wells were kept uninfected as controls. Plates were
incubated at 27°C in humidified environment for 7 days. Supernatants of all infected
cells and control were tested for product expression using ELISA. Samples that gave
high level of hTf yield were then selected to undergo the purification process twice
more or until the hTf level reached a constant yield provided that other parameters
remained unchanged for every purification round
3.3.2 Amplification of Virus Stock
Purified virus which had been kept in 4°C for few months was amplified
before use. Suspension culture at 0.5-2x106cells/ml was centrifuged and
resuspended in fresh growth medium. Cell was infected with low MOI (0.1-0.2
pfu/cells) by simply adding the required volume of seed virus stock to the suspension
culture. The infected cells were then left for 6 days with stirring. When the cell was
well infected, the medium was harvested by centrifuging at 150g (1000rpm) for 5
minutes. The virus stock was titrated using end point dilution method.
Amplification of virus was repeated until a high titer working stock was obtained.
59
Volume of Inoculum (ml) =(pfu/ml) inoculum viraloftiter
(pfu) cells ofnumber x totalMOI Desired.............(3.3)
3.3.3 Optimization of rhTf Expression
Suspension culture which has adapted to optimized medium was seeded at
1x106cells/ml. When the suspension culture reached 1.6x10
6cells/ml, the culture was
resuspended in fresh optimized medium. The culture was infected with amplified
rhTf baculovirus at day 2, at MOI 15. The infected culture was harvested at day 8 or
day 6 post infection. The product was harvested by centrifuging at 1000g (2600rpm)
for 5 minutes.
3.3.4 Virus Titration (End-Point Dilution)
Virus titration by end point dilution involves the estimation of the dilution of
virus that would infect 50% of the cultures. The quantity of virus is known as 50%
tissue-culture infection dose or TCID50. Virus titers may be expressed as TCID50/ml
or converted to pfu/ml. A serial of dilution of virus stock was prepared from 10-1
to
10-8
. Cells were diluted to a concentration of 1-2.5x105cells/ml with growth
medium. 100µl aliquots of each virus dilution were mixed with 900µl aliquots of the
cell suspension in appendorf tubes. Cell-virus suspension of the same dilution was
added to ten wells of 96 wells flat shaped micro titer plate, in same row and 100µ l
each. This was repeated every row for cell-virus suspension of different dilution
from 10-1
to 10-8
. The last two wells in the row were seeded with 100µl of cells, as
uninfected controls. The plate was sealed in a plastic bag with damp paper towel and
60
incubated at 27°C for 5 to 7 days. Signs of infection in the lower dilutions were
monitored daily.
Figure 3.1: Schematic representative of the procedures employed for virus titer-end
point dilution.
3.4 Recombinant Human Transferrin Detection
3.4.1 Enzyme Linked Immunosorbent Assay (ELISA)
Direct Sandwich ELISA was applied to determine hTf activity. A mixture of
capture antibody (goat anti-human transferrin) and coating buffer with ratio 1:100
was coated on a flat bottom, 96 wells plate or a ELISA plate for 1 hour at room
temperature. After incubation, the capture antibody solution will be aspirated from
each well and washed with washing solution (TBS-Tween-20) 3 times. 200 µl of
blocking solution (1% BSA in TBS) per well was incubated for 30 minutes to block
the plate. Then, the plate was washed three times. The standards (human serum
61
transferrin) and samples were diluted in samples diluents (1% BSA in TBS-Tween
20). The samples were diluted in sample diluents based on the expected
concentration so that they will fall within the concentration range of the standards.
100µl of standards and samples were transferred to assigned wells and incubated for
1hour.
After incubation, samples and standards were removed and the washing steps
were repeated 5 times. HRP-detection antibody (Goat anti-human transferrin-HRP
conjugate) was diluted in conjugate diluents (1% BSA in TBS-Tween 20) with a
range of 1:50,000. 100µl of the HRP-detection antibody was transferred to each well
and incubated for 60 minutes. Washing steps were repeated 5 times after incubation.
Equal volumes of TMB substrate and solution B from ELISA Started Accessory
package were mixed. 100µl of the mixture was incubated in each well for 15
minutes. Blue color developed immediately. The TMB reaction was stopped using
100µl of 1M phosphoric acid (H3PO4). The absorbance was read at 450nm
wavelengths.
Figure 3.2: Schematic representative of the procedures used in ELISA method.
62
3.4.2 Sodium Dodecyl Sulfate -Polyacrylamide Gel Electrophoresis (SDS-
PAGE)
SDS-Page separates proteins based on their molecular weight. SDS-PAGE
was used to check the existence of the desired product. Operation of SDS-PAGE can
basically be divided into 3 steps, (1) preparation of gel (2) electrophoresis of the
sample and (3) staining. The gel can divided into two parts: stacking gel for loading
of samples and separating gel for protein separation. Working solution for separating
and stacking gel was prepared as the methods listed in Appendix A. Plastic gloves
should be worn throughout the preparation because acrylamide is a neurotoxin.
A solution of the separating gel was added with 50µl of ammonium persulfate
and 10µl of TEMED. The solution was mixed well and poured between 2 plates
until about 0.5cm below the level where the wells will be formed by comb. Then,
water was layered on top of the separating gel solution to keep the gel surface flat.
The gel was allowed to polymerize and this took about 30 minutes. After the
separating gel polymerized, water that covered the separating gel was poured off.
Ammonium persulfate and TEMED were added to a solution of the stacking gel.
Stacking gel solution was pipetted onto a separating gel until the top of the front
plate. A comb was inserted carefully into the gel sandwich until the bottom of the
teeth reached the top of the front plate. Stacking gel took less than 30 minutes to
polymerize. After that, the polymerized gel in the plates was attached to the
electrode assembly and then was inserted into the electrophoresis chamber.
Electrophoresis buffer was added to the chamber and covered until the top of the gel.
The comb was removed carefully.
Protein sample was combined with 5x sample buffer at 4:1 in an appendorf
tube. Then, the protein sample was heated at 100ºC for 5 minutes. 20µl sample
solution, 5µl standard hTf and protein molecular weight marker were introduced into
the wells using a sample loading tips. Electrode plugs was attached to proper
electrodes. Power supply was turned on to 100V at a constant voltage for 90 minutes.
63
3.4.2.1 Silver Staining
Silver staining occurs mainly at gel surfaces, so thin gel (0.5-0.75mm) is
suggested. The gel was soaked in 50% methanol/10% acetic acid in a clean
container for at least 1 hour with 2-3 changes of methanol/acetic acid. The gel was
rinsed with water, soaked in water for 30 minutes, with constant stirring and at least
3 changes of water. Solution C was prepared by adding solution A (0.8g silver
nitrate in 4ml distilled water) to solution B (21ml 0.36 % NaOH mixed together with
1.4ml of 14.8M ammonium hydroxide) with constant stirring and then water was
added to make a total volume of 100ml. The gel was removed to a clean container
and stained in Solution C for 15 minutes with gentle, constant agitation. The gel was
rinsed twice and then soaked in deionized water for 2 minutes with gentle agitation.
Gel was removed to a clean container and developed by washing the gel in solution
D (0.05% (v/v) formaldehyde/0.5% (v/v) citric acid 0.1%). Bands appeared in less
than 15 minutes. Reaction was stopped by rinsing in 1% (v/v) acetic acid. The gel
was washed in water for at least 1 hour with at least 3 changes of water
3.4.2.2 Coomassie Blue Staining
.
Coomassie blue staining and destaining solution was prepared according to
the formulation mentioned in Appendix B. The gel was soaked in a staining solution
enough to cover the whole gel and agitated on an orbital shaker for 15 minutes.
Longer staining is required if recycled staining solution was used. After that, the
solution was discarded, and the gel was rinsed with distilled water. The gel was
destained overnight using destaining solution. Frequent changing of the destaining
solution would accelerate destaining of gel.
64
3.4.3 Western Blot
A mini Trans-Blot electrophoretic transfer cell was used to transfer
polyacrylamide gel to cellulose membrane. Bio-Ice cooling system unit was filled
with water and stored at -20°C. Transfer buffer and PBS buffer were prepared
according to the method mentioned in Appendix C. Transfer buffer was chilled to
4°C. The cellulose membrane and filter paper were cut to the dimensions of the gel.
The gel was equilibrated and the membrane, filter paper and fiber pads were soaked
in transfer buffer for 15min to 1hr. Gel holder cassette with gray side down, one
presoaked fiber pad, a sheet of filter paper, an equilibrated gel, a pre-wetted
membrane, another sheet of filter paper and the another fiber pad were placed layer
by layer. The gel sandwich was rolled by a glass roll gently to remove any bubbles.
The cassette was closed firmly and locked with the white latch. Then, the cassette
was placed in an electrode module. After that, the electrode module, frozen Bio-Ice
unit and a stirring bar were placed in buffer tank. The tank was filled with transfer
buffer. Finally, the blot was run on a stirring platform at a fixed voltage of 100V for
1 hour.
After transfer of protein, the membrane was incubated in blocking solution
(5% skimmed milk in PBS) for 1hour at room temperature or overnight at 4°C with
gentle agitation. The membrane was rinsed briefly with washing solution. Washing
step was repeated another 2 times for 10 minutes per each wash and the washing
solution was aspirated after each washing steps. HRP-detection antibody (Goat anti-
human transferrin-HRP conjugate) was diluted in blocking solution at 1:25,000 and
incubated the membrane with gentle agitation for 1 hour. Washing steps was
repeated as mentioned above. Then, the membrane was incubated with TMB
precipitating substrate for 15-30 minutes. A light green color stain developed on
supports bearing horseradish peroxide labeled conjugates.
65
3.5 Characterization of Nutrient Consumption and Substances Release
3.5.1 Analysis of Glucose, Lactic Acid and Glutamine
Glucose, lactic acid and glutamine were analyzed using biochemical analyzer
from YSI utilizing YSI immobilized enzyme membrane. A packet of buffer
concentrate from YSI was reconstituted in 500ml distilled water. Electrical leads
from the sensor were assembly into buffer and the YSI calibrations standard
respectively. YSI immobilized enzyme membrane was gently assembled onto the
probe face. Type of the enzyme membranes that had been used includes glucose
oxidase, lactate oxidase, glutaminase, glutamate oxidase membranes. The instrument
initialized the baseline current and auto calibrated every 15 minutes when it was in
run mode. Sample of about 500µl in appendorf tube was started analyzed once a
stable calibration was established. The samples were diluted with distilled water if
the sample was out of the detection range. The detection range was varied with
different standards.
Table 3.2: Specification of YSI calibrator.
Standards Calibration Point Detection Range
D-Glucose 2.5g/L 0-9g/L
L-Lactate 0.5g/L 0-2.67g/L
L-Glutamate 5.00mmol/L 0-10mmol/L
L-Glutamine 5mmol/L 0-8mmol/L
66
3.5.2 Ammonia Test
Randox’s kit was used to check ammonia content. Enzymatic UV method
was applied. Ammonia combines with α-ketoglutarate and NADPH in the presence
of glutamate dehydrogenase (GLDH) to yield glutamate and NADP+. The
corresponding decrease in absorbance at 340nm is proportional to the plasma
ammonia concentration.
α-ketoglutarate + NH3+
NADPH →GLDH
glutamate + NADP+……..............(3.4)
Each vial of reagent 1 of the kits (0.26mM NADPH/3.88mM α-ketoglutarate)
was reconstituted with 5ml of 0.15M triethanolamine buffer, pH8.6. 0.1ml of water
as blank, standard and samples were pipetted into different cuvettes. Duplicate set
was prepared. Then, 1ml of reagent 1 was added to the cuvettes. The mixture was
mixed and allowed to stand for 5 minutes. The absorbance of the mixture was read
at 340nm. Then, 10µl of GLDH was added to each cuvette. The solution was mixed
and left to stand for 5 minutes. Finally, the absorbance at 340nm was read once
again.
Concentration of ammonia = 294tan
xAA
AA
blankdards
blanksample
−
− µmol/l..……………………(3.5)
blankA = Absorbance (1) for Blank – Absorbance (2) for Blank …………………(3.6)
dardsA tan = Absorbance (1) for Standard – Absorbance (2) for Standard……..…..(3.7)
sampleA = Absorbance (1) for Sample – Absorbance (2) for Sample.……………..(3.8)
67
3.6 Protein Assay
3.6.1 Bicinchoninic Acid (BCA) Assay
BCA protein assay kit was used to quantify total protein. BCA working
reagent was prepared by mixing reagent A with reagent B at a ratio of 50:1.
Sufficient volume of working reagent was prepared for duplicate set of standards and
samples. Serials dilution method was used to prepare the standards. 0.05ml of each
standard and unknown sample replicate was placed into an appropriately labeled test
tube. 1.0 ml of the working reagent was added to each tube and well mixed. For
working ranges between 20-2,000 µg/ml, test tubes were incubated in water bath at
37°C for 30 minutes. For working ranges between 5-250 µg/ml, test tubes were
incubated in water bath at 60°C for 30 minutes. After that, all tubes were cool to
room temperature. With the spectrophotometer set to 562 nm, the reading was auto-
zeroed with cuvettes filled only with water. Subsequently, the absorbances of all the
samples was measured within 10 minutes.
Figure3.3: Schematic representation of the BCA protein assay.
68
3.7 Purification
3.7.1 Hydropbobic interaction Chromatography
Slurry of Phenyl Sepharose 6 fast flow was prepared by decanting 20%
ethanol solution and replacing it with water or other low ionic strength buffer in a
ratio of 50–70% settled gel to 50–30% packing solution. The gel was de-gassed
using a vacuum pump filter system. Econo-column 1.5 x 20 cm and the flow adaptor
from Bio-Rad were used to pack the matrix. The column was flushed with distilled
water to eliminate air from the column dead spaces. Then, the column was closed,
with some water remaining in the column. The slurry was poured into the column in
one continuous motion using a glass rod held against the wall of the column. The
rest of column was filled with distilled water until an upward meniscus was formed
at the top.
The flow adaptor which was connected to the pump was flushed and fully
filled with distilled water. After removing all bubbles, the pump was stopped and the
adaptor was inserted into the top of the column at an angle until it reached the gel
slurry. The adaptor o-ring was kept tight to give a sliding seal on the column wall.
The bottom outlet of the column was opened and the pump was set at the desired
flow rate. Ideally, Phenyl Sepharose 6 Fast Flow matrix should be packed at a
constant pressure of 0.15 MPa (1.5 bar) or flow rate less than 400 cm/hr. The
packing flow rate was maintained for 3 bed volumes till a constant bed height was
reached. The pump was closed, the bottom outlet was closed and the adaptor was
repositioned and locked on the surface of the matrix. The column was ready for used
when the bed medium was stable.
The column was equilibrated with starting buffer (1.2M Ammonium
Sulphate/ 0.4M Sodium citrate buffer, pH6) for 3 column volumes. Sample was
filtered through 0.45 µm membrane, mixed with 2x starting buffer (2.4M
69
Ammonium Sulphate/ 0.8M Sodium citrate buffer, pH6), and loaded into column
using pump. Then, 3 column volumes of starting buffer were used to wash away
unbound proteins. Elution buffer is the mixture of starting buffer and deionized
water. Percentage of the elution buffer was the percentage of starting buffer of the
mixture. For each step elution, three to four column volumes of elution buffer was
applied. Gradient elution was monitored using 2 pumps which drew deionized water
to starting buffer and to the column after homogenously mixing the solution. All the
equilibrating and operating flowrates were the same. Various flowrates, steps elution
and loading capacity were applied and studied for column optimization.
The column was washed with three column volumes of deionized water at
flowrates of 4ml/minute, and re-equilibrate with starting buffer after each run. For
the cleaning in place, precipitated proteins was removed by washing the column with
4 column volume (CV) of 1M NaOH solution at a flow rate of 1.2-1.4ml/min,
followed immediately with 2 to 3 CV of deionized water and re-equilibrated with 5
CV of starting buffer. Strongly hydrophobically bound proteins was removed by
washing the column with 4 CV of 70% ethanol, followed by with water and re-
equilibrated with starting buffer. The column was stored in 20% ethanol in distilled
water at 4°C when not in used.
3.7.2 Dialysis
Dialysis was used for desalting, buffer exchange and removal of small
molecular weight contaminants in samples. Snake SkinTM
pleated dialysis tubing
with 10,000 molecular weight cut off was used. The already-open tubing was pulled
from the stick to the required length. The amount of tubing can be calculated using
3.7ml sample per cm of dry tubing. 2-3 inches of one end of the tubing was briefly
dipped into water and tied tightly in the wetted end of the tubing. Sample was added
into the open end of the tubing. Then, another knot was tied securely in the other
70
open end. Finally, the tubing was immersed in 2 liters 20mM Tris/HCl buffer, pH
8.5, with constant stirring for 24 hours. The buffer was changed every 12 hours.
stirrer
Dialysis
tubing
Schoot
bottle
Buffer
Stirring
Platform
4oC
Figure 3.4: Schematic diagram of the dialysis procedure
3.7.3 Initial Screening Step of IEX using Batch Purification in Reduced
Volume
300µl anion exchange matrix was transferred into appendorf tube. Appendorf
tube was centrifuged at 500g (3000rpm) for 3–5 min to sediment the matrix. The
supernatant was discarded carefully. The matrix was washed five times with 3
matrix volumes of equilibration buffer. For each time, the slurry was centrifuged at
500 × g for 3–5 min and the equilibration buffer was discarded carefully. 500µl of
sample in equilibration buffer was added to the matrix. It was estimated that 1ml of
matrix could bind approximately 30mg of protein. Sample after HIC and after
dialysis was incubated in the matrix and agitated gently on a shaker for 2 hours at
room temperature. After that, the appendorf tube was centrifuged at 500g (3000rpm)
for 3–5 min to sediment the matrix. The supernatant was collected, and the rhTf in
the supernatant was determined using ELISA. Binding capacity was calculated by
minusing the total rhTf in supernatant from the total loaded rhTf per volume of
matrix.
71
Type of matrix, type of equilibration buffer, a range of pH of the selected
equilibration buffer and varied concentration of the selected equilibration buffer was
screened through to look for optimum conditions which give best binding capacity.
All the screened through parameters was listed in Table 3.3.
Table 3.3: Applied condition for different study factors.
Studied Factors Types/Parameters
Matrices Q-Sepharose and DEAE Sephadex A-25
Equilibration buffers Phosphate buffer and Tris/HCl buffer
pH of the equilibration buffer pH 7, pH 7.5, pH 8, pH 8.5, pH 9.5
Concentration of the equilibration buffer 10mM, 20mM, 30mM, 40mM, 50mM,
100mM
3.7.4 Ion Exchange Chromatography
Matrix Q-Sepharose fast flow was settled in starting buffer (20mM Tris
Buffer, pH8.5) and packed as mentioned in 3.7.1. The column was equilibrated with
starting buffer for 3 column volumes. Sample after dialysis in starting buffer was
loaded into the column using a pump. Then, 2 column volumes (CV) of starting
buffer were used to wash away unbound protein. The elution method was a
combination of gradient and steps elution. It was started with a linear gradient
elution where the percentage of buffer B (0.5M NaCl/ Tris Buffer, pH8.5) was
increased from 10% to 20% in nine column volumes and followed by step elution
with 20% of buffer B and lastly 100% of buffer B. Regeneration of Q-sepharose fast
flow was performed by washing 1M NaCl and followed by reequilibrating in 100ml
of starting buffer at flow rates of 4–5 ml/min. Matrix was stored in 20% ethanol in
distilled water at 4°C.
72
Fraction
CollectorTest tube
column
matrix
adaptor
Peristaltic
pumps
Buffer A
stirrer
Stirring
platform
Buffer B
Figure 3.5: Schematic diagram of the set up of the chromatography equipment.
3.8 Monosaccharide Composition Analysis of rhTf by HPLC
3.8.1 Preparation of Apotransferrin, rhTf, Standard Monosccharides
After expression and purification, rhTf was almost ready for monosaccharide
composition analysis. Pure rhTf at 6.69µg/ml was concentrated using Amicon
ultracentrifuge tube, 10000 MWCO, by centrifuging at 5000rpm for 20 minutes,
twice. Standards apo transferrin was prepared in same concentration, 383.08µg/ml.
5mg/ml of stock solution of mannose (Man), glucosamine (GlcN), galactose (Gal)
and glucose (Glc) were prepared and diluted to 50µg/ml. Set of monosaccharides
standards, standard apo-transferrin and sample rhTf in total volume of 100µl were
mixed and prepared as in Appendix H.
73
3.8.2 Hydrolysis
1ml of 4.4M trifluoroacetid acid (TFA) was added to all monosaccharides
standards, apo transferrin standard and rhTf sample in hydrolysis tubes. The
hydrolysis tube was sealed and incubated at 1210C for 4hrs in oven. After being
cooled to room temperature, the hydrolysis tubes were opened. Each reaction
mixture was transferred to a microcentrifuge tube and then evaporated to dryness by
concentration under reduced pressure in a desiccator. The residue was then dissolved
in 0.5ml of 2-propanol and again evaporated to dryness to remove residue TFA.
3.8.3 Pre-column Derivatization
The dried hydrolyzed glycoprotein or oligosaccharides samples,
monosaccharides standards, and the neuraminidase-/ aldolase-treated samples were
directly labeled with PMP by adding 20µl of PMP solution (0.5M in methanol) and
20µl of sodium hydroxide solution (0.3M). Then, the microcentrifuges tubes were
vortexed, and incubated at 70°C for 2hr. After that, the mixture was neutralized by
adding 20µl hydrochloric acid solution (0.3M). Butyl ether (0.5m) was added and
mixed thoroughly by vortexing for at least 5 seconds. Phase separation was
enhanced by brief centrifugation. The organic phase (upper layer) was carefully
removed and discarded. This extraction process was repeated two additional times.
The resulting phase was mixed with 250µl water before being analyzed by HPLC.
74
3.8.4 HPLC Analysis
A Inertsil ODS-2 column (150x3mm) was used to separate PMP-labeled
carbohydrates. The flowrates was set to 200µl/min and the wavelength for UV
detection was 245nm. For neutral and amino sugar separation, buffer A and B were
100mM ammonium acetate (pH5.5) with 10% and 25% acentonitrile, respectively.
A combined gradient of 45% to 55% buffer B in 30 minutes and 25 minutes elution
of 55% buffer B was used for separation. Column was equilibrated to 45% buffer B
before the next run.
CHAPTER 4
RESULTS AND DISCUSSION
Purification is important in the production of human therapeutic protein and
also in structural study. The focus of this research work was to produce rhTf from
Sf9-BEVS and to optimize the purification of rhTf in high yield and recovery.
Hydrophobic interaction chromatography and ion exchange chromatography were
used to purify the rhTf. Optimizations of these two columns were done to improve
the recovery and selectivity of rhTf. Carbohydrate component of the expressed rhTf
was characterized by HPLC.
4.1 Expression of rhTf
4.1.1 Growth Profile of Infected Virus
Sf9 cells were infected with AcMNPV. Success of infection was identified
by visual checking using inverted microscope. Virus infectivity was determined
using the end point dilution method. The physical appearances of rhTf-
76
-AcMNPV-infected sf9 were different from normal healthy cells. As shown in
Figure 4.1, the infected cells were swollen, enlarged and showed rough surfaces
around the cells.
Figure 4.1: Photography of control and infected culture. (a) is image of normal sf9
and (b) is image of infected sf9, observed using a x40 objective on an inverted
microscope.
Purified rhTf-AcMNPV stock which showed very low infectivity response
(<<106 pfu/ml) was amplified to increase the infectivity and virus stock volume for
rhTf production. Amplification of virus was done twice, both at low multiplicity of
infection (MOI), <0.5 pfu/cells to minimize the reproduction of defective interfering
particles (DIPs). These defective particles have extensive mutations in their genome
which resulted in the reduction of infectious virus yield. Infection at high MOI will
lead to a rapid increase in proportion of defective particles (Kool et al., 1991).
For rhTf production, the culture was seeded at 1.6x106cells/ml and infected
with rhTf virus at day 2 and MOI 15. Optimized medium (Ongkudon, 206) was used
to grow the cells. Cell culture was infected at high MOI for the production of
recombinant proteins to ensure synchronous infection of the majority of cells. Cell
was adapted to culture environment before infection. The spent medium might
contain secreted growth promoting factors with a positive effect on protein
production (Jesionowski and Ataai, 1997). Supernatant was harvested at day 6 post
(b) (a)
77
infection, the optimum harvest time reported by Ongkudon (2006). The harvested
rhTf was 31µg/ml.
Figure 4.2 shows the growth characteristic of Sf9 during rhTf virus
propagation. After infection at low MOI, cell viability dropped but cell cultures
continued to grow post infection and reached a plateau at day 6 (day 4 post infection).
Infectivity of virus stock was amplified to 5.6x107pfu/ml after 1st amplification and
1.2x109pfu/ml at final amplification. Virus stock from the second amplification
which showed high infectivity was used for rhTf production. Figure 4.3 shows the
growth characteristics of Sf9 during rhTf production. Sign of infection during rhTf
production was more significant. Cell viability dropped and cells stopped growing at
day 4 (day 2 post infection). Net growth rate and doubling time of cells after
infection at high MOI were 0.006hr-1 and 115hrs respectively and maximum cell
density was 2.99x106cells/ml (Table 4.1).
78
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
0 2 4 6
Times (day)
Via
ble
cell
Den
sity
(cel
ls/m
l x105 )
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
100.0
Via
bilit
y,%
Viable Cell Density (control)
Viable Cell Density (1st Amplification)Viable Cell Density (2nd Amplification)
Viability,% (control)Viability,% (1st Amplification)
Viability,% (2nd Amplification)
Figure 4.2: Growth Characteristics of sf9 during rhTf virus propagation. Opened
and closed circles refer to viability percentages (%) and viable cell density (cells/ml)
of noninfected Sf9 as control; square refers to growth profile during first
amplification; triangle refers to second amplification. Arrows shows the day of
infection. Baculovirus was amplified at low MOI and harvested at day 6. Non-
optimized medium (SFM900II) was used. Results are means ±SE for 2 replicates.
79
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
0 2 4 6 8Time (Day)
Via
ble
Cell
Den
sity
x 10
5 (ce
lls/m
l)
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
100.0
Viability Percentage (%
)
Viable Cell DensityViability, %
Figure 4.3: Growth Characteristic of sf9 during rhTf production in optimized
suspension culture. Sf9 was seeded at 1.6x106cells/ml, infected with rhTf virus at
day 2 and MOI 15. Optimized medium was used. Results are means ±SE for 2
replicates.
Glucose was the most important single sources of organic C for insect cells in
all cultures (Bedard et al., 1993). Glutamine was the second most rapidly consumed
amino acid (Bedard et al, 1993). Lipid mixtures, glutamine and glucose were found
to have the most positive effect on rhTf production with more than 95% significance
(Ongkudon, 2006). Optimized medium with the addition of 2211.2mg/L of
glutamine, 1291.95mg/L of glucose and 0.64% (v/v) lipids mixtures 1000x in
SFM900II (Ongkudon, 2006) was used for rhTf production. Time course profiles of
glucose, glutamine consumption and lactate formation in supernatant post infection is
shown in Figure 4.4. Glucose and glutamine was consumed post infection and the
consumption stopped at day 5 and day 4 respectively. Concentration of glucose,
glutamine and lactate in harvested supernatant were 7.41g/L, 15.1mM, 0.45g/L
80
respectively. Concentration of ammonia remaining in the culture was lower than
2mM, which would not affect the growth of sf9 (Bedard and Tom, 1993). Lactate
content was oxidized to carbon dioxide (Chiou et al. 2000) and lactate concentration
drop post infection. Low lactate level maintained pH level and thus improves
productivity (Gorfien et al. 2003). Lactate started to accumulate at day 4 post
infection. This maybe due to the decrease in dissolved oxygen (DO) (Palomares and
Ramirez, 1996).
0
2
4
6
8
10
12
0 1 2 3 4 5 6Day Post infection
Conc
entra
tion
of G
luco
se (g
/L)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
2.2
Conc
entra
tion
of L
acta
te (g
/L) &
Co
ncen
tratio
n of
Glu
tam
ine
(x10
mM
)
glucose lactate glutamine
Figure 4.4: The profile of glucose, glutamine consumption and lactate formation in
supernatant post infection. Results are means ±SE for 2 replicates.
4.1.2 Time Course Expression Profile of rhTf
Time course expression of rhTf post infection was quantified using ELISA.
As illustrated in Figure 4.5, concentration of rhTf in supernatant increased post
infection. Expression of rhTf started to decrease after day 5 post infection (Figure
81
4.5), however, the expression of rhTf, as reported by Ongkudon (2006), only started
to decrease after day 6. To ensure complete expression of rhTf, rhTf was harvested
at day 6 post infection. The extracellular protein from the expression system was
qualified using SDS-PAGE, stained in Coomassie blue. Protein expression from the
infected cell culture increased daily (Figure 4.6). The most significant protein band
was almost at the same row as standard transferrin and was assumed to be the
expressed rhTf. Expression of rhTf was reconfirmed using Western blot. Protein
was trans-blotted to nitrocellulose membrane from SDS-polyacrylamide gel and
detected by HRP-anti-transferrin antibody and TMB. As shown in figure 4.7, rhTf
was detected. The molecular weight of the expressed rhTf was slightly lower than
that of native human transferrin (76kDa). This might be due to the lack of complex
type oligosaccharides which were attached to the polypeptide (Ailor et al., 2000).
0
5
10
15
20
25
30
35
0 1 2 3 4 5 6 7
Day Post-Infection (Day)
Con
cent
ratio
n of
rhTf
( µg/
ml)
Figure 4.5: rhTf production profile in supernatant. Supernatant rhTf was collected
daily post infection and the rhTf content was checked using ELISA. Miniature graph
in the figure is the rhTf standard curve of ELISA which shows absorbance at 450nm
over logarithm concentration of hTf in ng/ml. Results are means ±SE for 2 replicates.
0 1 2 30.0
0.1
0.2
0.3
0.4
Log[concentration of hTf] ng/ml
Abs
at 4
50nm
82
Figure 4.6: Characterization of the rhTf production profile of infected Sf9, using 9%,
Coomassie blue staining, SDS-PAGE. Lanes d0 to d6 represent supernatants harvest
at day 0 to day 6 post infection. m, molecular weight standards and s, commercial
human transferrin.
Figure 4.7: Characterization of the rhTf production profile of infected Sf9, using
Western Blot. Protein in SDS-PAGE was trans-blotted to nitrocellulose membrane
and detected by HRP-anti-transferrin antibody and TMB. Lanes d0 to d6 represent
supernatants harvested at day 0 to day 6 post infection. m, molecular weight
standards and s, commercial human transferrin.
225kDa 150kDa 100kDa 75kDa 50kDa 35kDa 25kDa
s m d0 d1 d2 d3 d4 d5 d6
s d0 d1 d2 d3 d4 d5 d6
76kDa
Produced rhTf (73kDa)
Produced rhTf
83
Table 4.1: Summary of the characteristic of small scale production of rhTf.
Parameter Small Scale Production of rhTf
Type of culture Suspension
Type of Flask 250ml Shaker Flask
Speed 130rpm
Culture volume per flask 50ml/flask
Type of Medium
Optimized Medium
SFM900II with 2211.2mg/ml of Glutamate;
1291.95mg/ml of Glucose and 0.64% (v/v) of
lipid mixture
Seeding Density 1.6 x 106 cells/ml
MOI 15
TOI 48hrs
Length of cultivation Period 192hrs
Maximum Cell Density 2.99 x 106 cells/ml
Doubling Time (Post Infection) 115hrs
Net Growth Rate (Post Infection) 0.006hr-1
Protein Density 6200 µg/ml
rhTtf Density 31 µg/ml
4.2 Purification
Purification of all kinds of transferrin had been reported in a number of
papers. Hydrophobic interaction chromatography (Skinner et al., 1984; Ali et al.,
1996; Ailor et al., 2000), anion and cation exchange chromatography (Steinlein et al.,
1995; Mizutani et al., 2004), immunosorbent affinity chromatography (Tomiya et al.,
2003; Choi et al., 2003), metal ion affinity chromatography (Mason et al., 2001; Lim
et al., 2004) had been used to purify recombinant transferrin. Among these, Ali et al.
84
(1996) and Ailor et al. (2000) had purified rhTf from sf9 and Tn cells using phenyl
sepharose and Q-Sepharose and recovered rhTf up to 95% purity. Phenyl sepharose
together with affinity gel had been used to purify testicular transferrin from rat sertoli
cells and the 100% pure transferrin gave 28% overall recovery (Skinner et al., 1984).
Choi et al. (2003) purified rhTf from Lymantria dispar 652Y cells using column anti-
hTf-lgG immobilized Sepharose 4 Fast flow. Using immobilized metal affinity
chromatography (IMAC) Hexahistidine (His6) epitope tag hTf with 95.5% purity had
been obtained from transfected Drosophila melanogaster S2 cells with a recovery of
32% (Lim et al., 2004).
Recombinant transferrin from different expression systems needs different
extent of purifying. Some need simple purification, whereas others need more
complicated purification methods. Baculoviruses have lytic infection mode. Large
proportions of the host cells are lysed and degradative enzymes are released when the
products are harvested. Hence, the insect cell-baculovirus system is not considered a
“clean” secretion system (Altmann et al., 1999). It had been reported that neither
apo nor diferric hTf was bound to metal chelate matrix to any appreciable extent.
(Mason et al., 2001). Thus, metal affinity chromatography is not applicable to our
non-histidine tag rhTf. Expression of rhTf, as reported by Ali et al. (1996) and Ailor
et al. (2000), was similar to our expression system. Hence, the reported purification
methods which involving phenyl sepharose and Q-sepharose chromatography were
used as main references. Hydrophobic interaction chromatography utilizing phenyl
sepharose was used as the capture step and IEX chromatography utilizing Q-
sepharose was used for further purification of rhTf.
85
4.2.1 Profile of Sample Elution from hydrophobic Interaction
Chromatography
A screening step which involved a combination of step and gradient elutions
of transferrin from phenyl sepharose was done to understand the elution profile of
sample from HIC column. Sample was loaded at 15µg rhTf/ml of gel. Column was
equilibrated with 100% buffer 1.2M ammonium sulphate/0.4M sodium citrate (buffer
A), pH6.0. Proteins was eluted with 3 column volumes of 50% buffer A, followed
by gradient elution from 50% to 25% buffer A in 10 column volume (CV) and lastly
3 CV of water. Figure 4.8 shows that a large amount of unwanted protein was
hydrophilic and was washed out by the equilibration buffer. The third and second
large peaks in figure 4.8 showed that 50% buffer A eluted protein without rhTf and
0
0.2
0.4
0.6
0 20 40 60 80 100 120 140 160Fraction
Abs
at U
V 2
80nm
0
10
20
30
40
50
60
70
80
90
100
Perc
enta
ge o
f Buf
fer A
(%) &
Co
ncen
tratio
n of
r-hT
f in µg
/ml
AbsSteps & Gradient ElutionrhTf
Figure 4.8: Step and gradient elution of rhTf from HIC column. Sample was loaded
at 15µg rhTf /ml of gel. Unbound protein was washed out when the column was
equilibrated with 100% buffer A. Elution started with 3 CV of 50% buffer A and
followed by gradient elution using 50% to 25% buffer A in 10 CV. rhTf was pooled
from fraction 80 to fraction 111 and the recovery was 34%. Lane of SDS-PAGE, (a)
is sample loaded to column; (b) is sample recovered from fraction 80-111; (c) is
impurity eluted at 0% buffer A; m is marker, s is standard hTf.
a s b c m
hTf
86
0% buffer A or water (buffer B) eluted strong bound proteins which contained a
small amount of rhTf, respectively. Elution of a broad rhTf peak started when the
buffer contained less than 35% buffer A and stopped before 25% buffer A was
applied. rhTf pooled from fractions 80 to fraction 111 gave 34% recovery and did
not give satisfactory selectivity. Although fractions eluted using 0% buffer A
slightly contained rhTf, they were neglected because of the significant impurities
content (Figure 4.8).
4.2.2 Optimization of Hydrophobic Interaction Chromatography
HIC was used as the capture step, where recovery of target proteins is more
important compared to resolution. Type of gel, type of buffer and pH of buffer
which affects resolution was maintained as in 4.2.1 (Ali et al., 1996; Ailor et al.,
2000). Flowrates, loading capacity and elution method were optimized in order to
improve capacity, recovery, resolution and ease of use.
4.2.2.1 Optimization of Elution Method
Adsorption coefficient, α value of protein, in HIC changes only slowly with
changes in buffer condition. Protein-protein interaction brings similar proteins to
interact with each other and the adsorbent. Sharp separation is usually not achieved
in HIC (Scopes, 1994). Since gradient elution could not give satisfactory resolution
(Figure 4.8), step-wise elution which is technically simpler, reproducible and also
able to elute interest protein in a more concentrated form was applied in this study.
A few approaches using step-wise elutions had been done in order to increase the
resolution of the area where the peak of interest eluted without affecting the recovery
87
of rhTf. Strength of elution buffer was optimized to elute all less strongly bound
compounds, but must not exceed the level where peak of interest start to co-elute. As
mentioned in section 4.2.1, rhTf was eluted in between 35% to 25% of buffer A,
which mean 1st elution step can be optimized using buffer containing 50%-35%
buffer A. The second step elution was fixed at 25% buffer A because 25% buffer A
was expected to give complete elution of rhTf with minimum unwanted compound
(Figure 4.8).
HIC using 3 different step wise elutions involving 50% buffer A, 45% buffer
A and 35% buffer A was studied as the 1st elution buffer at a fixed flowrate of
0.5ml/min. 32±2µg rhTf per bed volume was loaded. Recovery of rhTf was the
highest when 50% buffer A was used as 1st elution buffer, 25% buffer A as 2nd
elution buffer (Table 4.2). Chromatograms and SDS-PAGE in Figure 4.9
characterizes the elution profile of the 3 different step wise elution methods.
Although 1st step elution with lower percentage of buffer A increased the elution of
unwanted compound (Figure 4.9) but the recovery of rhTf was low (Table 4.2).
Eluted rhTf did not show significant differences in resolution for the various step
elutions applied (Figure 4.9). Hence, step wise elution using 50% buffer A as 1st
elution buffer and 25% buffer A as 2nd elution buffer which gave 64% recovery was
chosen as the best elution method.
Table 4.2: Optimization of step-wise elution method for achieving higher recovery
of rhTf.
No Step elution Flowrates
(ml/min)
Recovery of
rhTf (%)
A 50% Buffer A and 25% Buffer A, (50/25) 0.5 64
B 45% Buffer A and 25% Buffer A, (45/25) 0.5 42
C 35% Buffer A and 25% Buffer A, (35/25) 0.5 29
Note: Different step elutions were optimized at a fixed flowrate, 0.5ml/min. Loaded
rhTf per bed volume was fixed at 32±2µg/ml. Recovery of rhTf was percentage of
pooled rhTf over total loaded rhTf.
88
0
0.1
0.2
0.3
0.4
0.5
0 20 40 60 80Fraction
Abs
at U
V 2
80nm
0
20
40
60
80
100
120
Step
s(Pe
rcen
tage
of B
uffe
r A(%
) &
Conc
netra
tion
of rh
Tf in
µg/
ml
AbsSteps ElutionrhTf
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 20 40 60 80 100 120 140Fraction
Abs
at U
V 2
80nm
0
20
40
60
80
100
120Pe
rcen
tage
of B
uffe
r A(%
) &
Conc
entra
tion
of r-
hTf i
n µg
/ml
AbsSteps ElutionrhTf
0
0.1
0.2
0.3
0.4
0.5
0.6
0 20 40 60 80 100 120Fraction
Abs
at U
V 2
80nm
0
20
40
60
80
100
120
Perc
enta
ge o
f Buf
fer A
(%)
&
Conc
entra
tion
of r-
hTf i
n µg
/ml
AbsSteps ElutionrhTf
*Volume for the first 24 fractions collected during equilibration was double compared to other
fractions
Figure 4.9: HIC chromatograms for the optimization of elution method. (a) 50/25
elution profile, (b) 45/25 elution profile and (c) 35/25 elution profile. Lane of SDS-
PAGE, (a) to (c) characterize the sample, which was pooled from experiment (a) to
(c); m is marker, s is standard hTf.
m a s b c *a
b c
89
4.2.2.2 Optimization of Elution Flowrate
Flow rate and sample load are interrelated. Flow rate and sample load are
optimized to find highest productivity where resolution is still high enough to meet
the predefined purity requirement. In this study, flowrate had not effect upon the
purity because step wise elution was applied. For the optimization of elution
flowrate, an average loading capacity of rhTf of 33±5µg/ml of gel and the optimized
step elution method as reported in section 4.2.2.1 were used. The prime
consideration when optimizing for highest possible productivity is to find the highest
possible sample load over the shortest possible sample application time with
acceptable loss in yield.
The elution flowrate which gave the highest recovery was 1ml/min and
followed by 0.5ml/min and 2ml/min (Table 4.3). Figure 4.10 shows the elution
profile of rhTf at different elution flowrates. High elution flowrate will always give
a decrease in dynamic binding capacity (Scopes, 1994) which will affect elution
profile and recovery of step elution. This may be the reason for the low recovery of
rhTf obtained at the elution flowrate of 2ml/min. Components of samples would
tend to ‘saturate’ non-specific binding sites of a column (Liljedah, 2000). A very
low flow rate will increase the sample holding time and thus increases the
opportunity of the nonspecific binding and result in loss of recovery. This may be
the reason the elution flowrate at 0.5ml/min showed lower recovery of rhTf than
1ml/min. In this study, optimized elution flowrate was 1ml/min which gave 74.6%
recovery.
90
Table 4.3: Optimization of elution flowrate.
No Flowrates Step Elution Recovery of
rhTf (%)
A 0.5ml/min 50/25 64
B 1ml/min 50/25 74.6
C 2ml/min 50/25 62.1
Note: Elution flowrate at 0.5ml/min, 1ml/min and 2 ml/min were studied at fix steps
elution. Loaded rhTf per bed volume was fixed at 33±5µg/ml. Recovery of rhTf
was percentage of pooled rhTf over total loaded rhTf.
91
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0 20 40 60 80Fraction
Abs
at U
V 2
80nm
0
10
20
30
40
50
60
70
80
90
100
Perc
enta
ge o
f Buf
fer A
(%) &
Co
ncen
tratio
n of
rhTf
in µ
g/m
)
AbsSteps ElutionrhTf
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0 20 40 60Fraction
Abs
at U
V 2
80nm
0
10
20
30
40
50
60
70
80
90
100
Perc
enta
ge o
f Buf
fer A
(%) &
Co
ncen
tratio
n of
rhTf
in µ
g/m
l)
AbsSteps ElutionrhTf
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0 20 40 60Fraction
Abs
at U
V 2
80nm
0
10
20
30
40
50
60
70
80
90
100
110
Perc
enta
ge o
f Buf
fer A
(%) &
Co
ncen
tratio
n of
rhTf
in m
g/m
l
AbsSteps ElutionrhT f
All the fraction is 5ml/fraction except fraction 25-63 of figure 4.10a are 2.5ml/fraction.
Figure 4.10: HIC chromatograms for the optimization of elution flowrate. Elution
profiles at (a) 0.5ml/min flowrate, (b) 1ml/min flowrate and (c) 2ml/min flowrate.
The peaks characterizing the elution of rhTf at 25% Buffer A were pooled.
a
c
b
92
4.2.2.3 Optimization of rhTf Loading Capacity
After fixing the elution mode and the elution flowrates, rhTf loading capacity
was also optimized to get the optimum level that gives highest recovery. 15, 38, 55
and 74µg rhTf/ml of gel were loaded respectively. The optimized elution method and
flow rates were as mentioned in section 4.2.2.1 & section 4.2.2.2 respectively. The
loading capacity of rhTf at optimized flowrates and steps elution which gave best
recovery was 55µg rhTf/ml gel (Table 4.4, Figure 4.11). The relationship between
loading capacity and recovery percentage in figure 4.11 shows that loading of rhTf
between 38-58µg/ml gel was expected to result in the recovery of 75% and above.
Loading of 38-55µg rhTf/ml gel was selected as the optimal range of loading
capacity. Figures 4.12 and 4.13 show the elution profile of rhTf.
As mentioned in 4.2.2.2, we lost some amount of sample due to the
nonspecific binding of each matrix. The percentage of loss is more significant when
the loading is small. This is the reason why the lowest loading of rhTf gave lower
recovery (Figure 4.11) although the loading also not exceed the dynamic binding
capacity. Loading capacity could also affect the elution profile of rhTf. Binding
strength of rhTf became weaker and it was eluted earlier when using higher
percentage of buffer A. As shown in Figure 4.12d, rhTf was eluted when 50% buffer
was applied. Hence, the recovery of rhTf was decreased even though all the rhTf
was bound to the gel during equilibration stage (Figure 4.12d). Figure 4.13 shows
that the resolution was not significantly affected by the loading capacity. Higher
loading increased both the concentration of rhTf and impurities in collected fraction.
Anyhow, this conclusion is not applicable to the case of overloading since the
binding strength and the elution profile were affected. Figure 4.13d, shows more
unwanted impurity compare to the others. It was predicted that early elution of
unwanted protein, which supposed to be eluted at 0% buffer A, was due to high
loading.
93
Table 4.4: Optimization of rhTf loading capacity.
No Flowrates Step Elution Total rhTf/ml of gel Recovery
A 1ml/min 50/25 15 µg/ml 67.5%
B 1ml/min 50/25 38 µg/ml 74.6%
C 1ml/min 50/25 55 µg/ml 79.7%
D 1ml/min 50/25 74 µg/ml 40%
Note: Loading capacity of rhTf was studied at optimized flowrates and steps elution.
Recovery of rhTf was percentage of pooled rhTf over total loaded rhTf.
30
40
50
60
70
80
90
10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0
Loaded rhTf per bed volume (µg/ml)
Rec
over
y pe
rcen
tage
(%
)
Figure 4.11: The relationship between recovery percentage and loading capacity.
94
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0 20 40Fraction
Abs
at U
V 2
80nm
0
10
20
30
40
50
60
70
80
90
100
110
Perc
enta
ge o
f Buf
fer A
(%) &
Co
ncen
tratio
n of
rhTf
in µ
g/m
l)
AbsSteps ElutionrhTf
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0 20 40 60Fraction
Abs
at U
V 2
80nm
0
10
20
30
40
50
60
70
80
90
100
Perc
enta
ge o
f Buf
fer A
(%) &
Co
ncen
tratio
n of
rhTf
in µ
g/m
l)
AbsSteps ElutionrhTf
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0 20 40 60Fraction
Abs
at U
V 2
80nm
0
10
20
30
40
50
60
70
80
90
100
110
Perc
enta
ge o
f Buf
fer A
(%) &
Co
ncen
tratio
n of
rhTf
in µ
g/m
l)
AbsSteps ElutionrhTf
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0 20 40 60 80Fraction
Abs
at U
V 2
80nm
0
10
20
30
40
50
60
70
80
90
100
Perc
enta
ge o
f Buf
fer A
(%) &
Co
ncen
tratio
n of
rhTf
in µ
g/m
l)
AbsSteps ElutionrhTf
Figure 4.12: HIC chromatograms for the optimization of rhTf loading capacity.
Flowrate was fixed at 1ml/min and the step elution method was fixed at 50/25.
Loading capacity at (a) 15µg rhTf/ml of gel, (b) 38µg rhTf/ml of gel, (c) 55µg
rhTf/ml of gel and (d) 74µg rhTf/ml of gel. The peak characterizing the elution of
rhTf at 25% buffer A was pooled.
a b
c d
95
(a) (b)
(c) (d)
Figure 4.13: SDS-PAGE characterizing the elution profile of rhTf. (a), (b), (c) & (d)
show the eluted fractions at 25% buffer A of chromatograms (a) to (d) of Figure 4.12;
m is protein marker.
4.2.3 Initial Screening Step of IEX using Batch Purification in Reduced
Volume
Batch purification was carried out using appendorf tube to screen the rhTf
binding capacity when using different anion exchange matrices, different types of
equilibration buffer, different pH of the equilibration buffer and different
concentration of equilibration buffer. Binding capacity was calculated by
centrifuging the appendorf tube that had been loaded with rhTf and subtracting the
m m
m m
rhTfrhTf
rhTf rhTf
96
amount of rhTf in the supernatant from the total rhTf loading. This step was taken as
a preliminary study of the ion exchange column to gain insight into the relationship
between buffer, matrix and binding capacity. Tris buffer gave higher binding
strength to anion exchange matrix compared to phosphate buffer (Figure 4.14).
Binding Strength of Q-Sepharose was stronger than DEAE Sephadex-25. Binding
capacity of Q-Sepharose was optimun when pH was increased to at least 8.5 and
concentration of Tris buffer was less than 30mM (Figure 4.15, Figure 4.16). The
choice of pH for anion exchanger should be the lowest pH that gives high binding
capacity. Ionic strength of the counter ion should not be too low to avoid pH
fluctuation and dilution of sample. Q-Sepharose with 20mM Tris/HCl buffer, pH8.5
which gave the highest binding capacity (figure 4.14, 4.15, 4.16) was identified as a
suitable parameter for the second stage separation.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
DEAE Sephadex A-25 Q-Sepharose
Type of Buffer
Bind
ing
Capa
city
of A
nion
Exc
hang
e M
atrix
/( µg
hTf/m
l) 20mM Tris-HCl buffer, pH7.5
20mM Sodium Phosphate
Figure 4.14: Binding capacity of two anion exchange matrix with Tris and
phosphate buffer used as equilibration buffer. Matrices in the study were DEAE
Sephadex A-25 and Q-Sepharose; buffers were 20mM Tris-HCl buffer, pH7.5
and 20mM sodium phosphate buffer, pH7.5. Results are means ±SE for 2
replicates.
97
2.40
2.60
2.80
3.00
3.20
3.40
3.60
6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0pH Buffer
Bind
ing
Capa
city
( µg
hTf/m
l of m
atrix
)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
Conc
entra
tion
of h
Tf in
Sup
erna
tant
(µg
/ml)
Binding capacity/(ug ofbound hTf/ml of matrix)Concentration of hTf inSupernatant/(ug/ml)
Figure 4.15: Binding capacity of Q-Sepharose with equilibration buffer of
different pH. Optimization buffer was carried out to achieve higher recovery of
rhTf. 20mM Tris-HCl buffer was used as the equilibration buffer and Q-
Sepharose as the matrix. Results are means ±SE for 2 replicates.
2.60
2.70
2.80
2.90
3.00
3.10
3.20
0 20 40 60 80 100Concentration of Buffer/mM
Bind
ing
Capa
city
( µg
hTf/m
l of m
atrix
)
-0.05
0.00
0.05
0.10
0.15
0.20
Conc
entra
tion
of h
Tf in
Sup
erna
tant
(µg
/ml)
Binding capacity/(ug ofbound hTf/ml of matrix)
Concentration of hTf inSupernatant/(ug/ml)
Figure 4.16: Binding capacity of Q-Sepharose with different concentration of
equilibration buffer. Buffer Tris-HCl buffer, pH8.5 was used as the equilibration
buffer and Q- sepharose as the binding matrix. Results are means ±SE for 2
replicates.
98
4.2.4 Anion Exchange Chromatography
According to Ali et al. (1996), anion exchange chromatography with Q
sepharose was used as the polishing step in rhTf purification in which 20mM
Tris/HCl buffer, pH8.0 was used as the equilibration buffer and a gradient elution of
0-100% KCl was employed. 95% purity of rhTf was obtained. In this work, the
relationship between matrix and buffer with rhTf binding capacity was obtained
(Section 4.2.3). Q-sepharose matrix and 20mM Tris buffer, pH8.5 would be used for
subsequent anion exchange chromatography. pH 8.5 which are slightly different
from the pH mentioned in Ali et al. (1996) but gave good binding capacity was
considered. The purity of the rhTf was the prime concern as anion exchange
chromatography was taken as the final polishing step. To improve selectivity, the
focus was on the gradient elution profile.
4.2.4.1 Maximizing the Selectivity of Anion Exchange Chromatography
Flowrates, pH and concentration of elution buffer would affect the resolution
of chromatography. In maximizing the selectivity of anion exchange column,
gradient elution was optimized by varying the shallowness of the gradient. Flowrate
of elution was not optimized but was fixed at a very slow, 0.5ml/min to maximize the
selectivity. Step elution was combined to simplify the elution where 20mM Tris-HCl,
pH8.5 was used to wash out unbound protein during equilibration and 100% 0.5M
NaCl/20mM Tris-HCl, pH8.5 (buffer B) was used to remove strongly bound protein.
rhTf was predicted to be eluted at around 15% of buffer B, so gradient elution was
started from 10% buffer B to 20% or 30% of Buffer B (Figure 4.17).
Three experiments using different gradient elutions were carried out with
different percentage of buffer B per CV at 3% per CV, 2% per CV and 1% per CV.
99
Figure 4.17 and 4.18 show that decreasing of the gradient slope from 3% buffer
B/CV (10%-30% Buffer B in 65 ml) (a) to 2% buffer B/CV (10-20% Buffer B in
53ml) (b) did not improved selectivity. Decreasing the slope of the gradient elution
to 1% buffer B/CV (10%-20% buffer B in 147 ml) (c) gave 100% pure rhTf. This
showed that by increasing the total gradient volume (decreasing gradient slope) of a
linear gradient, resolution could be improved (Figure 4.17).
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0 40 80 120 160 200 240 280 320time/minutes
Abs
at U
V 2
80nm
0
5
10
15
20
25
30
35
40
45
50
Perc
enta
ge o
f Buf
fer B
/% &
C
once
ntra
tion
of rh
Tf in
µg/m
l
AbsSteps & Gradient ElutionrhTf
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0 40 80 120 160 200 240 280time/minutes
Abs
at U
V 2
80nm
0
5
10
15
20
25
30
35
40
45
50
Perc
enta
ge o
f Buf
fer B
/% &
C
once
ntra
tion
of rh
Tf in
µg/m
l
AbsSteps & Gradient ElutionrhTf
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
0 50 100 150 200 250 300 350 400 450 500 550 600 650 700Time/Minutes
Abs
at 2
80nm
and
% o
f Buf
fer B
0
5
10
15
20
25
30
35
40
45
50Pe
rcen
tage
of B
uffe
r B/%
&
Con
cent
ratio
n of
rhTf
in µg
/ml
AbsSteps & Gradient ElutionrhTf
Figure 4.17: Anion exchange chromatograms for the optimization of selectivity.
Each chromatogram, (a) to (c), shows the elution profile of rhTf, with different
gradient elution. Gradient elutions of (a) to (c) were elutions with increasing of
buffer B, 3%per CV, 2%perCV and 1%perCV. Column volumes of (a) and (b) were
fixed at 10.6ml; column volume (c) was 16.8ml.
a b
c
100
(a) (b)
(c)
Figure 4.18: SDS-PAGE characterizing the elution profile of rhTf. (a), (b) & (c)
show the eluted fractions consisted of rhTf as shown in chromatogram (a), (b) & (c)
of Figure 4.17, during gradient elution; (m) is protein marker. (a) is fractions eluted
in between 156 to 192 minutes, (b) is fractions eluted in between 136-172 minutes
and (c) is fractions eluted in between 310 to 390 minutes.
4.2.5 Characterization of rhTf Purification
After optimization, a compete sequence of rhTf purification was carried out
using the optimized parameters. Crude sample with 0.5% yield of rhTf, was
harvested at day 6 post infection. Sample was loaded to column Phenyl Sepharose 6
fast flow at 38µg rhTf/ml of gel and at a flowate of 1ml/min. The column was
equilibrated with 100 buffer A. The sample was eluted with a step wise sequence
profile of 50% buffer A, 25% buffer A and water. 74.56% of rhTf was recovered.
After dialysis, the sample was loaded to Q-Sepharose fast flow. 20mM Tris-HCl, pH
m m
m
rhTfrhTf
rhTf
101
8.5 was used as the equilibration buffer. Elution flowrate was fixed at 0.5ml/min.
Gradient elution was initiated by increasing 50mM NaCl to100mM NaCl in 5
column volumes. 100% pure rhTf with 34% overall recovery was achieved (Table
4.5). Due to the difference in the characteristics of the tools in analyzing rhTf and
total protein, the purity percentage was actually slightly more than 100%, but was
taken to be 100%. Figure 4.19 and Figure 4.20 characterize the elution profile of
rhTf from Phenyl Sepharose column. Figures 4.21 and 4.22 characterize the elution
profile of rhTf from Q-Sepharose. Figure 4.23 qualifies the purity of the sample
from crude to final purification step. The total purification fold is 200.
Table 4.5: Summary of the characteristic of purification of rhTf.
Parameter Sample Phenyl
SepharoseDialysis Q Sepharose
Volume (ml) 23.00 59.00 79.50 36.20
rhTf Concentration (µg/ml) 31.00 9.01 5.81 6.69
Protein Concentration (µg/ml) 6200.00 - 24.47 6.62
Total hTf (µg) 713.00 531.63 461.50 242.08
Total Protein (µg) 142600.00 - 1945.66 239.70
Purity (%): 0.5 23.72 100.00
Recovery (%) 74.56 86.81 52.46
Overall Recovery (%) 74.56 64.73 33.95
102
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0 20 40 60Fraction
Abs
at U
V 2
80nm
0
10
20
30
40
50
60
70
80
90
100
Perc
enta
ge o
f Buf
fer A
(%) &
Co
ncen
tratio
n of
rhTf
in µ
g/m
l)
AbsSteps ElutionrhTf
Figure 4.19: HIC chromatogram characterizing the separation and elution profile of
sample. Chromatography was carried out using optimized flowrate, step elution and
loading.
m. marker
a. Fraction-40
b. Fraction-42
c. Fraction-43
d. Fraction-45
e. Fraction-47
f. Fraction-49
g. Fraction-51
h. Fraction-53
i. Fraction-55
Figure 4.20: SDS-PAGE characterizing the separated protein from phenyl sepharose
6 fast flow column. Protein in fractions 40, 42, 43, 45, 47, 49, 51, 53, 55 were shown
in 9%, silver staining, SDS-PAGE. m is molecular weight standards.
a b c m d e f g h i
rhTf
103
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
0 20 40 60 80 100 120 140Fraction
Abs
at U
V 2
80nm
0
10
20
30
40
50
60
Perc
enta
ge o
f Buf
fer B
/% &
Con
cent
ratio
n of
rhTf
in µ
g/m
l
AbsSteps & Gradient ElutionrhTf
Figure 4.21: Anion exchange chromatogram characterizing the separation and elution
profile of sample of after HIC and after dialysis. Q-Sepharose Chromatography was
carried out with 20mM Tris HCl, pH 8.5 as the equilibration buffer.
m- marker
a. Fraction-69
b. Fraction-71
c. Fraction-73
d. Fraction-75
e. Fraction-77
f. Fraction-79
g. Fraction-81
h. Fraction-83
i. Fraction-85
Figure 4.22: SDS-PAGE characterizing the separated protein from Q-Sepharose
column. Protein in fractions 69, 71, 73, 75, 77, 79, 81, 83, 85 were shown in 9%,
silver staining, SDS-PAGE. m is molecular weight standards.
a m b c d e f g h i
rhTf
104
Figure 4.23: SDS-PAGE characterizing the sample pooled from each purification
step. m is molecular weight standards, (a) is supernatant sample harvest at day 6 post
infection, (b) is sample after hydrophobic interaction chromatography and after
dialysis and (c) is pure rhTf after anion exchange chromatography.
4.3 Characterization of The Carbohydrate Composition of rhTf
Characterization of the carbohydrate composition of purified rhTf had been
carried out. 36.5µg standard apo-transferrin (76kDa) and purified rhTf (73kDa) were
hydrolyzed, labeled with PMP and analysed using HPLC. Separation profile of the
standard human transferrin and rhTf from insect cells are shown in Figure 4.24.
Mannose was eluted at t= 20.5±0.5min; Glucosamine was eluted at t= 28.5±0.5min;
Glucose was eluted at t= 51.5±0.5min; Galactose was eluted at t= 53.0±0.5min. The
carbohydrate content was quantified using the standard calibration curve of each
monosaccharide (Figure 4.25). During acid hydrolysis, N-acetylglucosamine would
loss the acetyl group to yield glucosamine. So, glucosamine was used as standard to
quantify N-acetylglucosamine.
m a b c
rhTf
225kDa150kDa 100kDa75kDa 50kDa 35kDa 25kDa
105
Results of the standard apo-transferrin analysis were compared with reported
values. The ratios of Man:GlcNAc:Gal of the apo-transferrin were 3:4.9:2. These
values were similar to the ratio reported by Spik, 1985 and Fu, 1995 (Table 4.6). For
the expressed rhTf, molar fractions of Man, GlcNAc and Gal over rhTf were 3.78,
1.69 and 0.93; the determined ratios of Man:GlcNAc:Gal were 3:1.34:0.74 (Table
4.6). These values were similar to the ratios reported by Salmon (1997) (Table 4.6).
Glucose was detected in both standard apo transferrin and the purified rhTf
(Figure 4.24). Glucose attached structure which was found early in oligosaccharide
processing pathway, was identified by Ailor et al. (2000). Alonzi et al. (2007)
mentioned that inhibition of endoplasmic reticulum (ER) alpha-glucosidases I and II
by imino sugars caused the retention of glucose residues on N-linked
oligosaccharides. Re-glucosylation by insect glucosyltransferase had been
recognized (Parker et al., 1995).
According to van Die (1996), galactosyltransferase activity was detected in
Spodoptera frugiperda. UDP-GlcNAc, UDP-Gal, UDP-Glc also existed in cultured
Sf9 (Tomiya, 2001). However, galactosylated or sialylated complex-type glycans are
rarely found in glycoprotein produced by insect cells (Jarvis and Summers, 1989;
Wathen et al., 1991; Grabenhorst et al., 1993; Yeh et al., 1996; Ogonah et al., 1996).
Despite above, minor galactose over rhTf as reported by Salmon et al., 1997 was
found in this study.
Total weight percent of Man, GlcNAc and Gal over standard human
transferrin in this study was 3.5%, which was similar to 3.9% reported by Zdebska
and Kościelak (1999). The total weight percent of Man, GlcNAc and Gal was only
1.61 % for the transferrin expressed from Sf9, which was about half of standard
human transferrin. This result was similar to that obtained by Salmon et al. (1997)
which was 1.88%. The difference with the native protein is related to the nature of
the expression system. Glycoproteins expressed in insect cells often have incomplete
glycan structure compared to those expressed in mammalian cells (Kuroda, 1990).
106
There are also other factors that can alter the glycosylation pathway. Castro et al.
(1995) found that supplementation of lipid affected the glycosylation pattern.
Gawlitzek et al. (1995), Hayte et al. (1992) and Hayte et al. (1993), also found that
glycosylation was dependent on media components. The optimized culture medium
used in this work which consisted of high glucose, glutamine and lipid content was
believed to affect the glycosylation pattern.
a)
b)
Figure 4.24: Chromatograms show HPLC separation of PMP-labeled transferrin. (a)
36.5µg of standard apo-transferrin and (b) purified sample rhTf which were
hydrolyzed, labeled and analyzed by HPLC as described under material and methods.
Mannose eluted at t= 20.5±0.5min; Glucosamine eluted at t= 28.5±0.5min; Glucose
eluted at t= 51.5±0.5min; Galactose eluted at t= 53.0±0.5min.
Man
nose
Glu
cosa
min
e
Glu
cose
Gal
acto
se
0
0
55
55
Retention time (min)
Retention time (min)
Man
nose
Glu
cosa
min
e
Glu
cose
Gal
acto
se
107
Standard for Monosaccharide Analysis using HPLC
R2 = 0.9916
R2 = 0.9933
R2 = 0.9907R2 = 0.9973
0.00E+00
2.00E+05
4.00E+05
6.00E+05
8.00E+05
1.00E+06
1.20E+06
1.40E+06
1.60E+06
1.80E+06
0.00 20.00 40.00 60.00 80.00 100.00 120.00
Concentration, µM
Are
a
Mannose Glucose Galactose Glucosamine
Figure 4.25: Standard calibration graph of monosaccharides. Hydrolyzed and
labeled monosaccharides standards in different concentration were analyzed using
HPLC. The graph shows the linearity of UV response for PMP labeled Glucosamine,
Mannose, Galactose and Glucose.
Table 4.6: Carbohydrate Composition Analysis of Glycoprotein.
Sample Man GlcNAc Gal
Literature: Human Serotransferrin (Fu, 1995) Determined ratio Human Serotransferrin (Spik, 1985) Determined ratio Recombinant Human Lactoferrin from Sf9 (Salmon, 1997) Determined ratio
3 3 3
4.44 4.7 1.5
1.88 2.4 0.5
Apotransferrin (Sigma T4382) Amount, mol/ mol protein Determined ratio
4.06 3
6.68 4.9
2.73 2
Sample, rhTf from Sf9 Amount, mol/ mol protein Determined ratio
3.78 3
1.69 1.34
0.93 0.74
CHAPTER 5
CONCLUSIONS
5.1 Conclusions
Using the optimized expression method reported by Ongkudon (2006),
31.1µg/ml of rhTf had been produced from Sf9. Expression of transferrin was
initiated by infecting Sf9 suspension culture in optimized medium at the seeding
density of 1.6x106 cells/ml with 15 MOI recombinant transferrin baculovirus at day 2
post culture. Maximum cell density, 2.99x106 cells/ml was achieved at day 2 post
infection. rhTf in 6200µg/ml of protein was harvested at day 6 post infection. Cells
growth profile, nutrient consumption and rhTf expression after infection were studied
to give more insight into the behavior of this insect cells-BEVS culture which were
done in suspension and using optimized medium. Concentration of glucose,
glutamine and lactate in harvested supernatant were 7.41g/L, 15.1mM and 0.45g/L
respectively. Concentration of ammonia remaining in the culture was lower than
2mM.
Purification was performed after rhTf expression. Few purification strategies
were studied to improve the purification yield and recovery. As reported in Ali et al.
(1996) and Ailor et al. (2000), phenyl sepharose and Q-sepharose chromatography
109
were used as the main tools in this separation. A screening step involving the
combination of step and gradient elution of transferrin from phenyl sepharose
provides a better understanding of the elution profile of the sample from HIC
column. For the column optimization, step wise elution which was technically
simpler, reproducible and also able to elute interest protein in a more concentrated
forms was used instead of gradient elution. For the optimization of elution method,
step elution at 50% buffer A and 25% buffer A gave rhTf in good recovery within a
predefined resolution.
The prime consideration when optimizing for highest possible productivity is
to find the highest possible sample load over the shortest possible sample application
time with acceptable loss in yield. Flow rates and loading capacity were optimized
in order to improve capacity, recovery and ease of use. The optimized flow rate was
1ml/min. Maximum loading capacity of rhTf at the optimal elution method and flow
rate was 55µg/ml. The selected range of loading of rhTf from 38-58µg/ml gel was
expected to result in the recovery of 75% and above.
Preliminary study of the second column, the ion exchange column, gave
better knowledge about the relationship between buffer, matrix and the binding
capacity. Batch purification method in reduced volume was applied to screen the
rhTf binding capacity of different anion exchange matrices, different types of
equilibration buffer, different concentrations of equilibration buffer and different pH
of the equilibration buffer. Q-Sepharose with 20mM Tris/HCl buffer, pH8.5 gave
the highest binding capacity. These parameters were applied for rhTf purification in
the second column. In maximizing the selectivity of the anion exchange column,
gradient elution was optimized by varying the slope of the gradient. Gradient elution
affects selectivity significantly. Gradient elution with 1% buffer B/CV (10%-20%
buffer B in 147 ml) succeeded in giving pure rhTf.
Crude sample with 0.5% yield of rhTf was loaded to Phenyl Sepharose 6 fast
flow column at 38µg rhTf/ml of gel using the optimized parameters gave 74.56%
110
recovery. The partially purified sample was loaded to Q-Sepharose fast flow column
at 0.5ml/min and pure rhTf was obtained. 34% overall recovery with 200
purification fold were achieved.
Lastly, a brief glycan characterization of the recovered pure rhTf was done
for a better understanding of the glycosylation feature of the expressed protein. The
carbohydrate component of the purified rhTf was determined. The ratios of
Man:GlcNAc:Gal for rhTf were 3:1.34:0.74. Minor galactose was found in this
study. Glucose attached structure was also obtained. Re-glucosylation had probably
occurred. The total weight percent of Man, GlcNAc and Gal was only half of the
standard apo-transferrin. Applied optimized culture condition had probably affected
the glycosylation pathway.
5.2 Recommendations
In this research, the whole process train for the production of rhTf, which
include expression, purification and basic characterization of the carbohydrate
content had been completed. There are a lot more improvement and interesting study
that should be performed in the future. A few recommendations are given to
improve the production, purification and the glycosylation of the system.
(a) Detail glycan characterization using nuclear magnetic resonance (NMR) or
HPLC mass spectrometry (HPLC-MS) to have a better understanding of the
glycan alteration.
(b) Detail study about the relationship between culture condition and
glycosylation in order to get optimized expression with improved
glycosylation.
(c) Improvement of glycosylation using in-vitro galactosylation and in-vitro
sialylation.
111
(d) Apply statistical analysis instrument such as Response Surface Method
(RSM) in the optimization of chromatographic process.
(e) Scaling up of the expression of rhTf to a large scale bioreactor for
commercial purpose. Study the alteration of glycosylation pathway after
scaling up and create a model to adjust the optimized culture condition and
minimize the deviation.
(f) Scaling up of the purification of rhTf to pilot scale.
REFERENCES
Abeijon, C. and Hirschberg, C. (1992) Topography of glycosylation reactions in the
endoplasmic reticulum. Trends in Biochemical Sciences. 17, 33-36.
Ailor, E., Betenbaugh, M.J. (1999). Modifying Secretion and Post-Translational
Processing in Insect Cells. Current Opinion in Biotechnology. 10, 142–145.
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., and Listowsky, I. (1980). Iron transport and storage proteins. Annual
Reviews of Biochemistry 49, 357-93.
Ali, S.A., Joao, H.C., Csonga, R., Hammerschmid, F. and Steinkasserer, A. (1996).
High-yield production of functionally active human serum transferrin using a
baculovirus expression system, and its structural characterization.
Biochemical Journal. 319, 191-195.
Alonzi, D., Neville, D., Lachmann, R., Dwek, R. and Butters, T. (2007).
Glucosylated Free Oligosaccharides Are Biomarkers of Endoplasmic
Reticulum Alpha-glucosidase Inhibition. Biochemical Journal. 409, 571-580.
113
Altmann, F., Staudacher, E., Wilson, I.B.H. and März, L. (1999). Insect Cells as
Hosts for The Expression of Recombinant Glycoproteins. Glycoconjugate
Journal. 16, 109–123.
Amersham Biosciences (2000b). Ion Exchange Chromatography: Principles and
Methods. (Edition AA). Amersham Biosciences.
Amersham Biosciences. (2000). Hydrophobic Interaction Chromatography:
Principles and Methods. (Edition AB). Amersham Biosciences.
An Ke, Gillock, E.T., Sweat, J.A., Reeves, W.M. and Consigli, R.A. (1999). Use of
The Baculovirus System to Assemble Polyomavirus Capsid-like Particles
with Different Polyomavirus Structural Proteins: Analysis of The
Recombinant Assembled Capsid-like Particles. The Journal of General
Virology. 80, 1009–1016.
Aumiller, J.J., Hollister, J.R. and Jarvis, D.L. (2003). A Transgenic Insect Cell Line
Engineered to Produce CMP-sialic Acid and Sialylated Glycoproteins.
Glycobiology. 13, 497-507.
Ayres, M.D., Howard, S.C., Kuzio, J., Lopez-Ferber, M. and Possee, R.D. (1994).
The Complete DNA Sequence of Autographa Californica Nuclear
Polyhedrosis Virus. Virology. 202, 586-605.
Bailey, S., Evans, R.W., Garratt, R.C., Gorinsky, B., Hasnain, S., Horsburgh, C.,
Jhoti, H., Lindley, P.F., Mydin, A. and Sarra, R. (1988). Molecular Structure
of Serum Transferrin at 3.3-A Resolution. Biochemistry. 27, 5804-5812.
Baker, E. N. and Lindley, P. F. (1992). New Perspectives on The Structure and
Function of Transferrins. Journal of Inorganic Biochemistry. 47, 147-160.
Baldwin, D.A., Jenny, E.R. and Aisen, P. (1984). The Effect of Human Serum
Transferrin and Milk Lactoferrin on Hydroxyl Radical Formation from
114
Superoxide and Hydrogen Peroxide. The Journal of Biological Chemistry.
259, 13391-13394.
Bauer, H. C., Parent, J. B. and Olden, K. (1985). Role of Carbohydrate in
Glycoprotein Secretion by Human Hepatoma Cells. Biochemical and
Biophysical Research Communications. 128, 368-375.
Beare, S. and Steward, W.P. (1996). Plasma Free Iron and Chemotherapy Toxicity.
The Lancet. 347, 342–343.
Bedard, C., Tom, R. and Kamen, A. (1993). Growth, Nutrient Consumption and End-
Product ccumulation in Sf9 and BTI-EAA Insect Cell Cultures: Insights into
Growth Limitation and Metabolism. Biotechnology Progress. 9, 615-624.
Belew, M., Yafang, M., Bin, L., Berglöf, J. and Janson, J-C. (1991). Purification of
Recombinant Hepatitis B Surface Antigen Produced by Transformed Chinese
Hamster Ovary (CHO) Cell Line Grown in Culture. Bioseparation. 1, 397–
408.
Beljelarskaya, S.N. (2002). A Baculovirus Expression System for Insect Cells.
Molecular Biology. 36(3), 281–292.
Betenbaugh, M.J, Tomiya, N., Narang, S., Hsu, J.T.A. and Lee, Y.C. (2004).
Biosynthesis of Human-Type N-Glycans in Heterologous Systems. Current
Opinion in Structural Biology. 14,601–606.
Birch, J.R. and Onakunle, Y. (2005). Biopharmaceutical Proteins: Opportunities and
Challenges. In: Mark Smales, C. and James, D.C. Therapeutic Proteins:
Methods and Protocols. New Jersey: Humana Press.
Bonnerjera, J., Oh, S., Hoare, M. and Dunhill, P. (1986). The Right Step at The Right
Time. Bio/Technology. 4, 954-958.
115
Bradley, S.J., Gosriwatana, I., Srichairantankool, S., Hider, R.C. and Porter, J.B.
(1997). Non-transferrin-bound Iron Induced by Myeloablative Chemotherapy.
British Journal of Haematology. 99, 337–343.
Breitbach, K., Jarvis, D.L. (2001). Improved Glycosylation of A Foreign Protein by
Tn-5B1-4 Cells Engineered to Express Mammalian Glycosyltransferases.
Biotechnology And Bioengineering. 74(3), 230-239.
Bullen, J.J., Griffiths, E. (1999). Iron and Infection, Molecular, Physiological and
Clinical Aspects. (2nd Edition). Chichester: John Wiley and Sons.
Burgess, S. (1977). Molecular Weights of Lepidopteran Baculovirus DNAs:
Derivation by Electron Microscopy. The Journal of General Virology. 37,
501-510.
Carson, D.D. (1992) Proteoglycans in Development. In: Fukuda, M. (Ed.). Cell
Surface Carbohydrates and Cell Development. (pp. 257-274). London: CRC
Press.
Castro, P.M.L., Ison, A.P., Hayte, P.M. and Bull, A.T. (1995). The
Macroheterogeneity of Recombinant Human Interferon-γ Produced by
Chinese Hamster Ovary Cells is Affected by The Protein and Lipid Content
of The Culture Medium. Biotechnology and Applied Biochemistry. 21, 87-
100.
Chauhan, A., Chauhan, V., Brown, W.T. and Cohen, I. (2004). Oxidative Stress in
Autism: Increased Lipid Peroxidation and Reduced Serum Levels of
Ceruloplasmin and Transferrin – The Antioxidant Proteins. Life Sciences. 75,
2539–2549.
Chiou, T.W., Hsieh, Y.C. and Ho, C.S. (2000). High Density Culture of Insect Cells
using Rational Medium Design and Feeding Strategy. Bioprocess
Engineering. 22, 483-491.
116
Choi, O., Tomiya, N., Kim, J.H., Slavicek, J.M., Betenbaugh, M.J. and Lee, Y.C.
(2003). N-Glycan Structures of Human Transferrin Produced by Lymantria
dispar (Gypsy moth) Cells using The LdMNPV Expression System.
Glycobiology. 13(7), 539-548.
Choudhury, D., Thakurta, P.G., Dasgupta, R., Sen, U., Biswas, S., Chakrabarti, C.
and Dattagupta, J.K. (2002). Purification and Preliminary X-ray Studies on
Hen Serotransferrin in Apo- and Holo-Forms. Biochemical and Biophysical
Research Communications. 295,125–128.
Chu, L. and Robinson, D.K. (2001). Industrial Choices for Protein Production by
Large Scale Cell Culture. Current Opinion in Biotechnology. 12: 180-187
Comings, D.E., Miguel, A.G. and Lesser, H.H. (1979). Nuclear proteins. VI.
Fractionation of Chromosomal Non-Histone Proteins using Hydrophobic
Chromatography. Biochimica et Biophysica. Acta. 563, 253–260.
Cox, M.M.J (2004). Chapter 3: Commercial Production in Insect Cells-One
Company’s Perspective. Bioprocess International. 2(2), 34-38.
Crowther, J.R. (1995). Elisa: Theory and Practice. Totowa, New Jersey: Human
Press.
Cumming, D.A. (1992). Physiological Relevance of Protein Glycosylation.
Development in Biological. Standardization. 76, 83-94
Cummings, R.D., Merkle, R.K. and Stults, N.L. (1989). Separation and analysis of
glycoprotein oligosaccharides. Methods in Cell Biology. 32, 141-183.
Davies, A.H. (1994). Current Methods for Manipulating Baculoviruses.
Bio/Technology. 12, 47–50.
117
de Jong G., van Noort W. L., Feelders R. A., de Jeu-Jaspars C. M. H. and van Eijk H.
G. (1992). Adaptation of Transferrin Protein and Glycan Synthesis. Clinical
Chimica Acta. 212, 27-45.
de Jong, G., van Dijk, J.P. and van Eijk, H.G. (1990) The Biology of Transferrin.
Clinical Chimica Acta. 190, l-46.
de Smit, M.H., Hoefkens, P., de Jong, G., van Duin, J., van Knippenberg, P.H. and
van Eijk, H.G. (1995). Optimized Bacterial Production of Nonglycosylated
Human Transferrin and Its Half-Molecules. The International Journal of
Biochemistry and Cell Biology. 27, 839–850.
Dill, K.A. (1990a). Dominant Forces in Protein Folding. Biochemistry. 29, 7133–
7155.
Donaldson, M., Wood, H.A., Kulakosky, P.C., Shuler, M.L. (1999). Glycosylation of
A Recombinant Protein in The Tn5B1-4 Insect Cell Line: Influence of
Ammonia, Time of Harvest, Temperature, and Dissolved Oxygen.
Biotechnology and Bioengineering. 63, 255–262.
Elbehri, A. (2005). Biopharming and The Food System: Examining The Potential
Benefits and Risks. AgBioForum. 8(1), 18-25.
Fausnaugh, J.L., Kennedy, L.A. and Regnier, F.E. (1984). Comparison of
Hydrophobic Interaction and Reversed-Phase Chromatography of Proteins.
Journal of Chromatography. 317, 141–155.
Fernandez, M., Crawford, L. and Hefferan, C. (2002). Pharming the Field: A Look at
The Benefits and Risks of Bioengineering Plants to Produce Pharmaceuticals.
Proceedings from A Workshop Sponsored by The Pew Initiative on Food and
Biotechnology, FDA, and USDA. July. Washington D.C., 4-12.
118
Fu, D. and O’Neill, R.A. (1995). Monosaccharide Composition Analysis of
Oligosaccharides and Glycoprotein by High-Performance Liquid
Chromatography. Analytical Biochemistry. 227, 377-384.
Fu, D. and O’Neill, R.A. (1995). Monosaccharide Composition Analysis of
Oligosaccharides and Glycoprotein by High-Performance Liquid
Chromatography. Analytical Biochemistry. 227, 377-384.
Fu, D., van Halbeek H. (1992). N-glycosylation Site Mapping of Human
Serotransferrin by Serial Lectin Affinity Chromatography, Fast Atom
Bombardment-Mass Spectrometry, and 1H Nuclear Magnetic Resonance
Spectroscopy. Analytical Biochemistry. 206, 53-63
Garcia, F.A.P. and Pires, E.M.V. (1993). Chromatography. In: Kennedy, J.F. and
Cabral, J.M.S. (Eds.) Recovery Processes for Biological Materials. (pp. 415-
451). London: Wiley.
Gawlitzek, M., Condradt, H.S. and Wagner, R. (1995). Effect of Different Cell
Culture Conditions on The Polypeptide Integrity and N-Glycosylation of A
Recombinant Model Glycoprotein. Biotechnology and Bioengineering. 46,
536-544.
Gershoni, J.M. and Palade, G.E. (1982). Electrophoretic Transfer of Proteins from
Sodium Dodecyl Sulfate-Polyacrylamide Gels to a Positively Charged
Membrane Filter. Analytical Biochemistry. 124, 396-405.
Goosen, M.F.A. (1993). Insect Cell Culture Engineering: An Overview. In: Goosen,
M.F.A., Gaugulis, A.J. and Faulkner, P. (Eds.) Insect Cell Culture
Engineering. New York: Marcel Dekker.
Gorfien, S.F., Paul, W., Judd,D., Tescione, L. and Jayme, D.W. (2003). Optimized
Nutrient Additives for Fed-Batch Cultures. Biopharm. International. 34-40.
119
Grabenhorst, E., Hofer, B., Nimtz, M., Jager, V. and Conradt, H.S. (1993).
Biosynthesis and Secretion of Human Interleukin-2 Glycoprotein Variants
from Baculovirus Infected Sf21 Cells: Characterization of Polypeptides and
Post-Translational Modifications. European Journal of Biochemistry. 215,
189−197.
Granados, R.R. and Federici, B.A. (Eds.). (1986). The Biology of Baculoviruses.
(2nd volume). Boca Raton, Florida: CRC Press.
Gumerov, D.R., Mason, A.B. and Kaltashov, I.A. (2003). Interlobe Communication
in Human Serum Transferrin: Metal Binding and Conformational Dynamics
Investigated by Electrospray Ionization Mass Spectrometry. Biochemistry. 42,
5421-5428.
Haan, K. and Maggos, C. (2004). The Protein Engineer. BioCentury.12(5), A1-A7.
Hall, D.R., Hadden, J.M., Leonard, G.A., Bailey, S., Neu, M., Winn, M. and Lindley,
P.F. (2002). The Crystal and Molecular Structures of Diferric Porcine and
Rabbit Serum Transferrins at Resolutions of 2.15 and 2.60 A, Respectively.
Acta Crystallografica. Section D, Biological Crystallography. 58, 70-80.
Halperin, G., Breitenbach, M., Tauber-Finkelstein, M. and Shaltiel, S. (1981).
Hydrophobic Chromatography on Homologous Series of Alkyl Agaroses. A
comparison of Charged and Electrically Neutral Column Materials. Journal
of Chromatography. 215, 211–228.
Harrap, K.A. (1972). The Structure of Nuclear Polyhedrosis Viruses. The Inclusion
Body. Virology. 50, 114-123.
Harris, D.C., and Aisen, P. (1989). Physical Biochemistry of The Transferrins. In
Loehr, T.M., Gray, H.B. and Lever, A.B.P. (Eds.). Iron Carriers and Iron
Proteins. (pp. 239-351). Weinheim: VCH Publishers.
120
Harrison, P., Marwah, S.S., Hughes, R.T. and Bareford, D. (1994). Nontransferrin
Bound Iron and Neutropenia after Cytotoxic Chemotherapy. Journal of
Clinical Pathology. 47, 350–352.
Hase, S. (1996). Precolumn Derivatization for Chromatographic and Electrophoretic
Analyses of Carbohydrates. Journal of Chromatography A. 720, 173-182.
Hayter, P.M., Curling, E.M.A., Baines, A.J., Jenkins, N., Salmon, I., Strange, P.G.,
Tong, M.M. and Bull, A.T., (1992). Glucose-Limited Chemostat Culture of
Chinese Hamster Ovary Cells Producing Recombinant Human Interferon-γ.
Biotechnology and Bioengineering. 39, 327-335.
Hayter, P.M., Curling, E.M.A., Gould, M.L., Baines, A.J., Jenkins, N., Salmon, I.,
Strange, P.G. and Bull, A.T., (1993). The Effect of The Dilution Rate on
CHO Cell Physiology and Recombinant Interferon-γ Production in Glucose-
Limited Chemistat Culture. Biotechnology and Bioengienering. 42, 1007-
1085.
He, J. and Furmanski, P. (1995). Sequence Specificity and Transcriptional Activation
in the Binding of Lactoferrin to DNA. Nature. 373, 721-724.
Hirose M. (2000). The Structural Mechanism for Iron Uptake and Release by
Transferrins. Bioscience Biotechnology and Biochemistry. 64, 1328-1336.
Hjertén, S. (1973). Some General Aspects of Hydrophobic Interaction
Chromatography. Journal of Chromatography. 87, 325–331.
Hjertén, S. (1981). Hydrophobic Interaction Chromatography of Proteins, Nucleic
Acids, Viruses and Cells on Non-Charged Amphiphilic Gels. In: Glick, D.
(Ed.). Methods of Biochemical Analysis. (pp. 89-108). Canada: John Wiley
and Sons.
121
Hjertén, S., Rosengren, J. and Påhlman, S. (1974). Hydrophobic Interaction
Chromatography. The Synthesis and The Use of Some Alkyl and Aryl
Derivatives of Agarose. Journal of Chromatography. 101, 281–288.
Hjertén, S., Yao, K., Eriksson, K.-O. and Johansson, B. (1986). Gradient and
Isocratic High Performance Hydrophobic Interaction Chromatography of
Proteins on Agarose Columns. Journal of Chromatography. 359, 99–109.
Hofstee, B.H.J. (1973). Hydrophobic Affinity Chromatography of Proteins.
Analytical Biochemistry. 52, 430–448.
Hollister, J., 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. 15093-15104.
Hollister, J.R. and Jarvis, D.L. (2001). Engineering Lepidopteran Insect cell Cells for
Sialoglycoprotein Production by Genetic Transformation with Mammalian
β1,4-Galactosyltransferase and α2,6-Sialytransferase Genes. Glycobiology.
11(1), 1-9.
Honda, S., Akao, E., Suzuki, S., Okuda, M., Kakehi, K. and Nakamura, J. (1989).
High-Performance Liquid Chromatography of Reducing Carbohydrates as
Strongly Ultraviolet-Absorbing and Electrochemically Sensitive 1-Phenyl-3-
Methyl-5-Pyrazolone Derivatives. Analytical Biochemistry. 180, 351-357.
Hrkal, Z. and Rejnkova, J. (1982). Hydrophobic Interaction Chromatography of
Serum Proteins on Phenyl-Sepharose CL-4B. Journal of Chromatography.
242, 385–388.
Huebers, H.A. and Finch, C.A. (1987). The physiology of Transferrin and
Transferrin Receptors. Physiological Reviews. 67, 520–582.
Ikeda, R.A., Bowman, B.H., Yang, F.M. and Lokey, L.K. (1992) Production of
Human Serum Transferrin in Escherichia coli. Gene. 117(2), 265-269.
122
Janson, J.-C. and Rydén, L. (1993). Protein Separation and Purification. In: Rehm,
H.-J. and Reed, G. (Eds.). Biotechnology, Vol. 3. (pp. 617-642). Weinheim:
VCH.
Janson, J-C. and Låås, T. (1978). Hydrophobic Interaction Chromatography on
Phenyl- and Octyl-Sepharose CL-4B. In: Roger, E. (Ed.). Chromatography of
synthetic and biological macromolecules. Chichester, England: Ellis
Horwood.
Jarvis, D.L. (2003). Developing Baculovirus-Insect Cell Expression Systems for
Humanized Recombinant Glycoprotein Production. Virology. 310, 1–7.
Jarvis, D.L. and Summers, M.D. (1989). Glycosylation and Secretion of Human
Tissue Plasminogen Activator in Recombinant Baculovirus-Infected Insect
Cells. Molecular Cell Biology. 9, 214–223.
Jarvis, D.L., Howe, D. and Aumiller, J.J. (2001) Novel Baculovirus Expression
Vectors That Provide Sialylation of Recombinant Glycoproteins in
Lepidopteran Insect Cells. Journal of Virology. 75, 6223–6227.
Jennissen, H.P. and Heilmeyer, I.M.G. (1975). General Aspects of Hydrophobic
Chromatography. Adsorption and Elution Characteristics of Some Skeletal
Muscle Enzymes. Biochemistry. 14, 754–760.
Jesionowski, G.A. and Ataai, M.M. (1997). An Efficient Medium for High Protein
Production in The Insect Cell/Baculovirus Expression System. Biotechnology
Progress. 13, 355-360.
John Morrow, K. (2007, March 1). Strategic Protein Production: Revamping and
Reengineering Existing Technology Can Increase Protein Yield and Efficacy.
Genetic Engineering and Biotechnology News. Retrieved September 6, 2007,
from http://www.genengnews.com
123
Joosten, C.E., Shuler, M.L. (2003). Production of A Sialylated N-Linked
Glycoprotein in Insect Cells: Role of Glycosidases and Effect of Harvest
Time on Glycosylation. Biotechnology Progress. 19, 193–201.
Joshi, L., Shuler, M.L. and Wood, H.A. (2001). Production of A Sialylated N-Linked
Glycoprotein in Insect Cells. Biotechnology Progress. 17, 822–827.
Kaplan, H.A., Welply, J.K. and Lennarz, W.L. (1987). Oligosacchary1 Transferase:
The Central Enzyme in The Pathway of Glycoprote in Asssembly.
Biochimica et Biophysica. Acta. 906, 161-173.
Kennedy, R.M. (1990). Hydrophobic Chromatography. In: Deutscher, M.P. (Ed.)
Guide to Protein Purification, Methods Enzymology, Vol. 182. (pp. 339-343).
London: Academic Press Inc.
Kimura, Y., Nakagawa, Y., Tokuda, T., Yamai, M.. Nakajima, S., Higashide. E. and
Takagi, S. (1992) Structures of N-linked Oligosaccharides of Microsomal
Glycoproteins from Developing Castor Bean Endosperms. Bioscience,
Biotechnology, and Biochemistry. 56, 215-222.
Kool, M. and Vlak, J.M. (1993). The Structural and Functional Organization of The
Autographa Californica Nuclear Polyhedrosis Virus Genome. Archives of
Virology. 130, 1-16.
Kool, M., Voncken, J.W., Vanlier, F.L.J., Tramper, J. and Vlak, J.M. (1991).
Detection and Analysis of Autographa Californica Polyhedrosis Virus
Mutants with Defective Interfering Properties. Virology. 183, 739-746.
Kornfeld, R. and Kornfeld, S. (1985). Assembly of Asparagine-Linked
Oligosaccharides. Annual Review of Biochemstry. 54, 631–66.
Kuehn, L., Meyer, H. and Reinauer, H. (1980). Hydrophobic Interaction
Chromatography as A Tool in Insulin Receptor Study In: Brandenburg, D.
124
and Wollmer, A. (Eds.) Proceedings of the 2nd. International Insulin
Symposium. Aachen, 243–250.
Kuroda, K., Geyer, H., Geyer, R., Doerfler, W. and Klenk, H.D. (1990). The
Oligosaccharides of Influenza Virus Hemagglutinin Expressed in Insect Cells
by A Baculovirus Vector. Virology. 174, 418–429.
Lăăs, T. (1975). Agar Derivatives for Chromatography, Electrophoresis & Gel-
Bound Enzymes. IV. Benzylated Dibromopropanol Cross-Linked Sepharose
as An Amphophilic Gel for Hydrophobic Salting Out Chromatography of
Enzymes with Special Emphasis on Denaturing Risks. Journal of
Chromatography. 111, 373–387.
Lawrence, S.M., Huddleston, K.A., Tomiya, N., Nguyen, N., Lee, Y.C., Vann, W.F.,
Coleman, T.A. and Betenbaugh, M.J. (2001). Cloning and Expression of
Human Sialic Acid Pathway Genes to Generate CMP-Sialic Acids in Insect
Cells. Glycoconjugate Journal. 18, 205-213.
Lee, K.S., Kim, B.Y., Kim, H.J., Seo, S.J., Yoon, H.J., Choi, Y.S., Kim, I., Han, Y.S.,
Je, Y.H., Lee, S.M., Kim, D.H., Sohn, H.D. and Jin, B.R. (2006). Transferrin
Inhibits Stress-Induced Apoptosis in A Beetle. Free Radical Biology and
Medicine. 41, 1151–1161.
Lee, S.Y. (1996). High Cell Density Cultivation of Escherical coli. Trends
Biotechnology. 14, 98-105.
Lefort, S. and Ferrara, P. (1986). Hydrophobic Adsorbants for The Isolation and
Purification of Biosynthetic Human Growth Hormone from Crude
Fermentation Mixtures. Journal of Chromatography. 361, 209–216.
Léger 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. Biochemical Journal.
257, 231-238.
125
Lennarz, W.J. (1975). Lipid Linked Sugars in Glycoprotein Synthesis. Science. 188,
986-991.
Li, P.Z., Gao, X.G., Arellano, R.O. and Renugopalakrishnan, V. (2001).
Glycosylated and Phosphorylated Proteins—Expression in Yeast and Oocytes
of Xenopus: Prospects and Challenges—Relevance to Expression of
Thermostable Protein. Protein Expression and Purification. 22, 369–380.
Lim, H.J., Kim, Y,K., Hwang, D.S. and Cha, H.J. (2004). Expression of Functional
Human Transferrin in Stably Transfected Drosophila S2 Cells. Biotechnology
Progress. 20, 1192-1197.
Lindah, L. and Vogel, H.J. (1984). Metal-Ion-Dependent Hydrophobic Interaction
Chromatography of α-Lactalbumins. Analytical Biochemistry. 140, 394–402.
Luckow, V. L., and Summers, M.D. (1988). Trends in The Development of
Baculovirus Expression Vectors. Bio/Technology. 6, 47-55.
Luckow, V.A., Lee, S.C., Barry, G.F. and Olins, P.O. (1993). Efficient Generation of
Infectious Recombinant Baculviruses by Site-Specific Transposon-Mediated
Insertion of Foreign Genes into A Baculovirus Genome Propagated in
Escherischia coli. Journal of Virology. 67, 4566−4579.
MacGillivray, R. T. A., Mendez, E., Sinha, S. K., Sutton, M. R., Lineback-Zins, J.
and Brew, K. (1982). The Complete Sequence of Human Serum Transferrin.
Proceedings of The National Academic of Sciences of the United States of
America.79(8), 2504-2508.
MacGillivray, R. T. A., Mendez, E., Shewale, G., Sinha, S. K., Lineback-Zins, J. and
Brew, K. (1983). The Primary Structure of HumanSerum Transferrin. The
Structures of Seven Cyanogen Bromide Fragments and The Assembly of The
Complete Structure. The Journal of Biological Chemistry. 258, 3543-3553.
126
Maisano, F., Belew, M. and Porath, J. (1985). Synthesis of New Hydrophobic
Adsorbents Based on Homologous Series of Uncharged Alkyl Sulphide
Agarose Derivatives. Journal of Chromatography. 321, 305–317.
Makhatafze, G.I. and Privalov, P.L. (1995). Energetics of Protein Structure.
Advances in Protein Chemistry. 47, 307–325.
Makrides, S.C. (1996). Strategies for Achieving High Level Expression of Gene in
Escherichia coli. Microbiological Reviews. 60, 512-538.
Mason, A.B., He, Q-Y., Adams, T.E., Gumerov, D.R., Kaltashov, I.A., Nguyen, V
and MacGillivray, R.T.A. (2001). Expression, Purification, and
Characterisation of Recombinant Non-Glycosylated Human Serum
Transferrin Containing A C-Terminal Hexahistidine Tag. Protein Expression
and Purification. 23, 142–150.
Mason, A.B., Miller, K.M., Funk, W.D., Banfield, D.K., Savage, K.J., Oliver,
R.W.A., Green, B.N., MacGillivray, R.T.A. and Woodworth, R.C. (1993).
Expression of Glycosylated and Nonglycosylated Human Transferrin in
Mammalian Cells. Characterization of the recombinant proteins with
comparison to three commercially available transferrins. Biochemistry. 32,
5472-5479.
Matsuura, Y., Possee, R.D., Overton, H.A. and Bishop, D.H.L. (1987). Baculovirus
Expression Vectors: the Requirements for High Level Expression of Proteins,
Including Glycoproteins. Journal of General Virology. 68, 1233–1250.
McNair, R.D. and Kenny, A.J. (1979). Proteins of The Kidney Microvillar
Membrane. The Amphipathic Form of Dipeptidyl Peptidase IV. Biochemical
Journal. 179, 379-395.
Melander, W. and Horvath, C. (1977). Salt Effects on Hydrophobic Interactions in
Precipitation and Chromatography of Proteins: An Interpretation of The
Lyotropic Series. Archives of Biochemistry and Biophysics. 183, 200–215.
127
Melander, W., Corradini, D. and Horváth, C. (1984). Salt Mediated Retention of
Proteins in Hydrophobic Interaction Chromatography. Application of
Solvophobic Theory. Journal of Chromatography. 317, 67–85.
Mizutani, K., Okamoto, I., Fujita, K., Yamamoto,K. and Hirose, M. (2004).
Structural and Functional Characterization of Ovotransferrin Produced by
Pischia pastoris, Bioscience Biotechnology and Bichemistry. 68(2), 376-383.
Montreuil, J., Vliegenthart, J.F.G. and Schachter, H. (1995). Glycoproteins. In:
Neuberger, A. and Van Deenen, L.L.M. (Eds.). New Comprehensive
Biochemistry. (pp. 29a). Amsterdam: Elsevier.
Montreuil. J., Bouquelet. S., Dcbray, H., Foumet. B., Spik. G. and Strecker, G.
(1986). Glycoproteins. In Chaplin, M.F. and Kennedy, J.F. (Eds.)
Carbohydrate Analysis. (pp. 143-204). Britain: Oxford University Press.
Morgan, E.H. (1979). Studies on The Mechanism of Iron Release from Transferrin.
Biochimica et Biophysica Acta. 580, 312-326.
Muneo, T., Ken, O., Massanori, M. And Toshizumi, T. (1996). Secretion of A
Variant of Human Single-Chain-Urokinase-Type Plasminogen Activator
Without An N-Glycosylation Site in The Methyiotrophic Yeast, Pischia
pastoris and Characterization of The Secreted Product. Yeast. 12, 541-553.
Murphy, F. A., Fauquet, C. M., Bishop, D. H. L., Ghabrial, S. A., Jarvis, A. W.,
Martelli, G. P., Mayo, M. A. and Summers, M. D. (Eds.). (1995). Virus
Taxonomy. Sixth Report of the International Committee on Taxonomy of
Viruses. Berlin: Springer Verlag.
Murphy, K.P., Privalov, P.L. and Gill, S.J. (1990). Common Features of Protein
Unfolding and Dissolution of Hydrophobic Compounds. Science. 247, 559–
561.
128
Narhi, L.O., Kita, Y. and Arakawa, T. (1989). Hydrophobic Interaction
Chromatography in Alkaline pH. Analytical Biochemistry.182, 266–270.
Ochoa, J.L. (1978). Hydrophobic Interaction Chromatography. Biochimie. 60, 1–15.
Ogonah, O.W., Freedman, R.B., Jenkins, N., Patel, K., and Rooney, B.C. (1996).
Isolation and Characterization of An Insect-Cell Line Able to Perform
Complex N-Linked Glycosylation on Recombinant Proteins. Bio/Technology.
14, 197−202.
Olden, K., Parent, J.B. and White, S.L. (1982). Carbohydrate Moieties of
Glycoproteins A Reevaluation of Their Function. Biochimica. et Biophysica.
Act. 650, 209-232.
Ongkudon, C.M. (2006). Optimization of Recombinant Human Transferin
Expression in Insect Cells Baculovirus System. Master, Universiti Teknologi
Malaysia, Skudai
Oscarsson, S. and Kårsnås, P. (1998). Salt-Promoted Adsorption of Proteins Onto
Amphiphilic Agarose-Base Adsorbents II. Effects of Salt and Salt
Concentration. Journal of Chromatography A. 803, 83–93.
Parker, C.G., Fessler, L.L., Nelson, R.E. and Fessler, J.H. (1995). Drosophila UDP-
Glucose: Glycoprotein Glycosyltransferase: Sequence and Characterization
of The Enzyme That Distinguishes Between Denatured and Native Proteins.
The EMBO Journal. 14, 1294-1303.
Parodi, A.J. and Leloir, L.L.. (1979) The Role of Lipid Intermediates in The
Glycosylation of Proteins in The Eukaryotic Cell. Biochimica et Biophysica
Acta. 559, l-37.
Parsegian, V.A. and Ninham, B.W. (1970). Temperature-Dependent van der Waals
Forces. Biophysical Journal. 10, 664–674.
129
Patel, G., Nasmyth, K., and Jones, N. (1992). A New Method for The Isolation of
Recombinant Baculovirus. Nucleic Acids Research. 20, 97–104.
Patterson, R.M., Selkirk, J.K. and Merrick, B.A. (1995). Baculovirus and Insect Cell
Gene Expression: Review of Baculovirus Biotechnology. Enviromental
Health Perspectives. 103, 7-8.
Payne, S.L., Qi, X.M., Dwyer, A. and Fuller, F. (1998). Disease Induction by Virus
Derived from Molecular Clones of Equine Infectious Anemia Virus. Journal
of Virology. 72, 483-487.
Peakman, T.C., Harris, R.A. and Gewert, D.R. (1992). Highly Efficient Generation
of Recombinant Baculoviruses by Enzymatically Medicated Site-Specific in
Vitro Recombination. Nucleic Acids Research. 20, 495–500.
Peterson, E.A. and Sober, H.A. (1956). Chromatography of Proteins I. Cellulose Ion
Exchange Adsorbents. Journal of The American Chemical Society. 78, 751-
755.
Pharmigen (1999). Baculovirus Expression Vector System Manual. (6th Edition). San
Diego, USA: Pharmigen.
Pierce Biotechnology. (2005). Protein Assay: Technical Handbook. United States:
Pierce Biotechnology.
Ponka, P. (1997). Tissue-Specific Regulation of Iron Metabolism and Heme
Synthesis: Distinct Control Mechanisms in Erythroid Cells. Blood. 89(1), 1-
25.
Porath, J. (1986). Salt-Promoted Adsorption: Recent Developments. Journal of
Chromatography. 376, 331–341.
130
Porath, J. (1987). Metal Ion-Hydrophobic, Thiophilic and II-Electron Governed
Interactions and Their Application to Salt-Promoted Protein Adsorption
Chromatography. Biotechnology Progress. 3, 14–21.
Porath, J., Sundberg, L., Fornstedt, N. and Olson, I. (1973). Salting-Out in
Amphiphilic Gels as A New Approach to Hydrophobic Adsorption. Nature.
245, 465–466.
Porter, J.B., Abeysinghe, R.D., Marshall, L., Hider, R.C. and Singh, S. (1996).
Kinetics of Removal and Reappearance of Non-Transferrin Bound Plasma
Iron with Deferoxamine Therapy. Blood. 88, 705–713.
Possee, R.D. (1997). Baculoviruses as Expression Vectors. Current Opinion in
Biotechnology. 8, 569–572.
Privalov, P.L., Gill, S.J. (1988). Stability of Protein Structure and Hydrophobic
Interaction. Advances in Protein Chemistry. 39, 191–234.
Regnier, F.E. (1987). The Role of Protein Structure in Chromatographic Behavior.
Science. 238, 319–323.
Reiser, J. (1990). Advance in Biochemical Engineering/Biotechnology. (pp. 75-102).
Berlin, Germany: Spriner-Verlag.
Richardson, D.R. and Ponka, P. (1997). The Molecular Mechanisms of The
Metabolism and Transport of Iron in Normal and Neoplastic Cells.
Biochimica et Biophysica Acta. 1331, 1-40.
Rippel, G. and Szepesy, L. (1994). Hydrophobic Interaction Chromatography of
Proteins on An Alkyl-Superose Column. Journal of Chromatography A. 664,
27–32.
Rohrmann, G.F. (1986). Polyhedrin Structure. Journal of General Virology. 67,
1499-1513.
131
Rosengren, J., Påhlman, S., Glad, M. and Hjertén, S. (1975). Hydrophobic
Interaction Chromatography on Non-Charged Sepharose Derivatives. Binding
of A Model Protein, Related to Ionic Strength, Hydrophobicity of The
Substituent, and Degree of Substitution (Determined by NMR). Biochimica.
et Biophysica. Acta. 412, 51–61.
Sahlstedt, L., Ebeling, F., von Bonsdor, L., Parkkinen, J., Ruutu, N. (2001). Non-
Transferrin-Bound Iron During Allergenic stem cell transplantation. British
Journal of Haematology. 113, 836–838.
Salmon, V., Legrand, D., Georges, B., Slomianny, M-C., Coddeville, B. and Spik, G.
(1997). Characterization of Human Lactoferrin Produced in the Baculovirus
Expression System. Protein Expression and Purification. 9, 203-210.
Shaltiel, S. and Er-El, Z. (1973). Hydrophobic chromatography: Use for Purification
of Glycogen Synthetase. Proceedings of The National Academic of Sciences
of the United States of America. 70, 778–781.
Scopes, R.K. (1994). Protein Purification. (3rd Edition). New York: Springer-Verlag.
Shongwe, M.S., Smith, C.A., Ainscough, E.W., Baker, H.M., Brodie, A.M., and
Baker, E.N. (1992). Anion Binding by Human Lactoferrin: Results from
Crystallographic and Physicochemical Studies. Biochemistry. 31(18), 4451-
4458.
Shuler, M.L and Kargi, F. (2002). Bioprocess Engineering. (2nd edition). Upper
Saddle River, N.J.: Prentice Hall PTR.
Skinner, M.K., Cosand, W.L. and Griswold, M.D. (1984). Purification and
Characterization of Testicular Transferrin Secreted by Rat Sertoli Cells.
Biochemical Journal. 218, 313-320.
132
Spik, G., Debruyne, V., Montreuil, J., van Halbeek, H. and Vliegenthart J.F.G.
(1985). Primary Structure of Two Sialylated Triantennary Glycans from
Human Serotransferrin. FEBS Letters. 183(1), 65-69.
Srinivasan, R. and Ruckenstein, E. (1980). Role of Physical Forces in Hydrophobic
Interaction Chromatography. Separation and Purification Methods. 9, 267–
370.
Steinlein, L.M. and Ikeda, R.A. (1993). Production of N-Terminal and C-Terminal
Human Serum Transferrin in Escherichia coli. Enzyme and Microbial
Technology. 15, 193–199.
Steinlein, L.M., Graf, T.N. and Ikeda, R.A. (1995). Production and Purification of N-
Terminal Half-Transferrin in Pichia pastoris. Protein Expression and
Purification. 6, 619-624.
Stibler, H., Allgulander, C., Borg, S. and Kjellin, K.G. (1978). Abnormal
Microheterogeneity of Transferrin in Serum and Cerebrospinal Fluid in
Alcoholism. Acta Medica Scandinavica. 204(1-2), 49-56.
Summers, M.D. and Smith, G.E. (1978). Baculovirus Structural Polypeptides.
Journal of Virology. 84, 390-402.
Summers, M.D., Anderson, D.L. (1972). Granulosis Virus Deoxyribonucleic Acid: A
Closed, Double-Stranded Molecule. Journal of Virology. 9(4), 710-713.
Talukder, G. (2007, January). The Market for Bioengineered Protein Drugs. BCC
Research. Retrieved September 6, 2007, from http://www.bccresearch.com
Tanford, C. (1973). The Hydrophobic Effect: Formation of Micelles and Biological
Membranes. New York: Wiley.
Tiselius, A. (1948). Adsorption separation by salting out. Arkiv för Kemi, Mineralogi
Geologi. 26B, 1–5.
133
Tomiya, N., Howe, D., Aumiller, J.J., Pathak, M., Park, J., Palter, K.B., Jarvis, D.L.,
Betenbaugh, M.J. and Lee, Y.C. (2003). Complex-Type Biantennary N-
Glycans of Recombinant Human Transferrin from Trichoplusia ni Insect
Cells Expressing Mammalian β-1,4-Galactosyltransferase and β-1,2-N
Acetylglucosaminyltransferase II. Glycobiology. 13(1), 23-34.
Traving, C. and Schauer, R. (1998). Structure, Function and Metabolism of Sialic
Acids. Cellular and Molecular Life Sciences. 54, 1330-1349.
Tuite, M.F., Dobson, M.J., Roberts, N.A., King, R.M., Burke, D.C., Kingsman, S.M.,
and Kingsman, A.J. (1982). Regulated High Efficiency Expression of Human
Interferon-γ in Saccharomycas cerevisiae. The EMBO Journal. 1, 603-608.
Tulsi, B. (2004). Bugs Punch the Clock as Next Protein Manufacturers: Interest in
The Capabilities of Competing Insect Cell-Based Technologies Heightens as
Demand for Therapeutic Protein Production Capacity Escalates. Drug
Discovery & Development. 7, 56-62.
Valenzuela, P., Medina, A., Rutter, W.J., Ammerer, G. and Hall, B.D. (1982).
Synthesis and Assembly of Hepatitis-B Virus Surface Antigen Particles in
Yeast. Nature. 298, 347-350.
van Die, I., van Tetering, A., Bakker, H., van den Eijnden, D.H. and HJoziasse, D.
(1996). Glycosylation in Lepidopteran Insect Cells: Identification of A β1-4-
N-Acetylgalactosaminyltransferase Involved in The Synthesis of Complex-
Type Oligosaccharide Chains. Glycobiology. 6(2), 157-164.
van Eijk, H.G., van Noort, W.L., de Jong, G. and Koster, J.F. (1987) Human Serum
Sialo Transfertins in Diseases. Clinical Chimica Acta. 165, 141-145.
Van Oss, C.J., Good, R.J. and Chaudhury, M.K. (1986). Nature of The Antigen-
Antibody Interaction. Primary and Secondary Bonds: Optimal Conditions for
Association and Dissociation. Journal of Chromatography. 376, 111–119.
134
Varki, A. (1993). Biological Roles of Oligosaccharides: All of The Theories are
Correct. Glycobiology. 3, 97-130.
Varki, A., Cummings, R., Esko, J., Freeze, H., Hart, G. and Marth, J. (1999).
Essentials of Glycobiology. Cold Spring Harbor, NY: Cold Spring Harbor
Press.
Vervecken, W., Kaigorodov, V., Callewaert, N., Geysens, S., De Vusser, K., and
Contreras, R. (2004). In Vivo Synthesis of Mammalian-Like, Hybrid-Type N-
Glycans in Pichia pastoris. Applied And Environmental Microbiology. 70(5),
2639–2646.
Vieira, A.V. and Schneider, W.J. (1993). One-step chromatographic method for the
puification of avian serotransferrin. Protein Expression and Purification. 4,
110–113.
Visser, J. and Strating, M. (1975). Separation of Lipoamide Dehydrogenase
Isoenzymes by Affinity Chromatography. Biochimica et Biophysica Acta.
384, 69–80.
Viswanathan, K., Narang, S., Hinderlich, S., Lee, Y.C.| and Betenbaugh, M.J. (2005).
Engineering Intracellular CMP-Sialic Acid Metabolism into Insect Cells and
Methods to Enhance Its Generation. Biochemistry. 44, 7526-7534.
Vogel, H.J., Lindahl, L. and Thulin, E. (1983). Calcium-Dependent Hydrophobic
Interaction Chromatography of Calmodulin, Troponin C and Their
Proteolytic Fragments. FEBS Letters. 157, 241–246.
von Bonsdorff, L., Tölö, H., Lindeberg, E., Nyman, T., Harju, A. and Parkkinen, J.
(2001). Development of a Pharmaceutical Apotransferrin Product for Iron
Binding Therapy. Biologicals. 29, 27–37.
Waechter, C.J. and Lennarz, W.J. (1976) The Role of Polyprenol Linked Sugars in
Glycoprotein Synthesis? Annual Review of Biochemistry. 45, 95- 113.
135
Wang, Y., Chen, J., Luo, Y., Funk, W.D., Mason, A.B., Woodworth, R.C.,
MacGillivray, R.T. and Brayer, G.D. (1992). Preliminary Crystallographic
Analyses of The N-Terminal Lobe of Recombinant Human Serum
Transferrin. Journal of Molecular Biology. 227, 575-576.
Watanabe, S., Kokuho, T., Takahashi, H., Takahashi, M., Kubota, T. and Inumaru, S.
(2002). Sialylation of N-Glycans on The Recombinant Proteins Expressed by
A Baculovirus-Insect Cell System Under β-N-Acetylglucosaminidase
Inhibition. The Journal Of Biological Chemistry. 277(7), 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 Sf9. Biochemistry. 30, 2863−2868.
Weinberg, E. (1984). Iron Withholding: A Defense Against Infection and Neoplasia.
Physiological Reviews. 64, 65–102.
Whitaker, J.R. (I 977). Enzymatic Modification of Proteins Applicable to Foods. In:
Feeney, R.E. and Whitaker, J.R. (Eds.). Food Proteins. (pp. 126-134).
Advance in Chemistry Series 160. Washington D.C: American Chemical
Society.
Winzerling, J.J., Pham, D.Q.-D., Kunz, S., Samaraweera, P., Law, J.H. and Porath, J.
(1996). Purification of Recombinant Insect Transferrin from Large Volumes
of Cell Culture Medium Using High Capacity Ni 2+ -Dipicolylamine Gel
Protein Expression and Purification. 7, 137–142.
Wright, A. and Morrison, S. (1997). Effect of Glycosylation on The Antibody
Function: Implications for Genetic Engineering. Trends biotechnology. 15,
26-32.
Yamashita, K., Koide, N., Endo, T., Iwaki, Y. and Kobata, A. (1989). Altered
Glycosylation of Serum Transferrin of Patients with Hepatocellular
Carcinoma. The Journal of Biological Chemistry. 264, 2415-2423.
136
Yap, W.N. (2006). Development of Novel Galactosylation Method for the
Expression of Recombinant Human Transferrin in Insect Cell Culture. Master,
Universiti Teknologi Malaysia, Skudai.
Yeh, J.C., Seals, J.R, Murphy, C.I., van Halbeek, H. and Cummings, R.D. (1996).
Site-Specific N-glycosylation and Oligosaccharide Structures of
Recombinant HIV-1 gp120 Derived from A Baculovirus Expression System.
Biochemistry. 32(41), 11087–11099.
YSI. (2001). Immobilized Enzyme Biosensor. YSI. Retrieved October 13, 2007, from
http://www.ysilifesciences.com.
Yun, E.Y., Goo, T.W., Kim, S.W., Choi, K.H, Hwang, J.S., Kang, S.W and Kwon,
O.Y (2005). Galatosylation and Sialylation of Mammalian Glycoproteins
Produced by Baculovirus-Mediated Gene Expression in Insect Cells.
Biotechnology Letters. 27, 1035-1039.
Zdebska, E. and Kościelak, J. (1999). A Single-Sample Method for Determination of
Carbohydrate and Protein Contents Glycoprotein Bands Separated by Sodium
Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis. Analytical
Biochemistry. 275, 171-179.
APPENDIX A
APPENDIX A-1 Stock Solution for SDS-PAGE
1. 2M Tris-HCl (pH8.8), 100ml
Weight out 24.2g Tris-base and add to 80ml distilled water.
Add 1M HCl slowly to adjust pH buffer to pH 8.8.
Add distilled water to total volume 100ml.
2. 1M Tris-HCl (pH 6.8), 100ml
Weight out 12.1g Tris base and add to 80ml distilled water.
Add 1M HCl slowly to adjust pH buffer to pH 6.8.
Add distilled water to total volume 100ml.
3. 10% SDS(w/v), 50ml
Weight out 5g SDS.
Dissolve the SDS with a total volume of 50ml distilled water.
4. 50% glycerol (v/v), 100ml
Add 50ml distilled water to 50ml 100% glycerol.
5. 1% bromophenol blue (w/v),10ml
Weight out 100mg bromophenol blue.
Dissolve the powder with a total volume of 10ml distilled water.
138
APPENDIX A-2 Working Solution for SDS-PAGE
1. 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. Solution B (4x separating gel buffer), 100ml
75ml 2M Tris-HCl (pH8.8)
4ml 10% SDS
21ml distilled water
3. Solution C (4x stacking gel buffer), 100ml
50ml 1M Tris-HCl (pH6.8)
4ml 10% SDS
46ml distilled water
4. 10% ammonium persulfate
0.05 g in 0.5ml distilled water
5. Electrophoresis buffer, 1L
3g Tris
14.4g glycine
1g SDS
Add distilled water to make 1L.
6. 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
139
APPENDIX A-3 Separating and Stacking Gel Preparation
1. Separating Gel X% Preparation
Chemical Volume
Solution A x/3 ml
Solution B 2.5 ml
H2O (7.5-x/3) ml
10% ammonium persulfate 50ul
TEMED 10ul
2. Stacking Gel Preparation
Chemical Volume
Solution A 0.67ml
Solution C 1.0ml
H2O 2.3ml
10% ammonium persulfate 30ul
TEMED 10ul
140
APPENDIX-B
Coomassie Blue Staining
1. Coomassie Gel Stain, 1 liter
1g Coomassie Blue R-250
450ml methanol
450ml distilled water
100ml glacial acetic acid
2. Coomassie Gel Destain, 1 liter
100ml methanol
100ml glacial acetic acid
800ml distilled water
141
APPENDIX C
Preparation of Optimized Medium
1) Stock Solution
Note: *Stock solution was purchased from Sigma
**Powder glutamine dissolved in medium SFM 900 II
2) Working Solutions:
According to this the equation: M1V1 = M2V2
Note: All the stock solutions, V1 were mixed and diluted to a total volume of 500ml
medium SFM900II
M1 V1 M2 V2
Chemicals Stock Solutions
(mg/L)
Volume
(ml)
Working Solutions
(mg/L)
Volume
(ml)
Glucose 100000 6.46 1291.95 500
Glutamine 25000 44.22 2211.2 500
Lipids 1000xc 100% 3.2 0.64 500
Chemicals Concentration g/L/v or v/v%
*Glucose 100
**Glutamine 25
Lipids 1000xc 100%
142
Example of TCID50 Calculation (spreadsheet)
(1)
(2)
Number Number Total Total % % % Log of of Number Number Total Above Above Below Dilution Infected Uninfected Infected Uninfected Infected 0.5 0.5 0.5 Above
Dilution
Wells Wells 0.5 0.0001 7.000 3.000 22.000 3.000 0.880 TRUE 0.000 0.000 0.000
0.00001 7.000 3.000 15.000 6.000 0.714 TRUE 0.714 0.000 -5.0000.000001 4.000 6.000 8.000 12.000 0.400 FALSE 0.000 0.400 0.000
0.0000001 4.000 6.000 4.000 18.000 0.182 FALSE 0.000 0.000 0.000 71.429 40.000 -5.000
Number Number Total Total % % % Log of of Number Number Total Above Above Below Dilution Infected Uninfected Infected Uninfected Infected 0.5 0.5 0.5 Above
Dilution
Wells Wells 0.5 0.0001 9.000 1.000 26.000 1.000 0.963 TRUE 0.000 0.000 0.000 0.00001 7.000 3.000 17.000 4.000 0.810 TRUE 0.000 0.000 0.000 0.000001 5.000 5.000 10.000 9.000 0.526 TRUE 0.526 0.000 -6.000 0.0000001 5.000 5.000 5.000 58.000 0.079 FALSE 0.000 0.079 0.000 52.632 7.937 -6.000
APPE
ND
IX D
142
143
APENDIX D: (Continued)
Spread sheet (1) Spread Sheet (2)
Number of Wells 10.000 10.000
ml/well 0.010 0.010
Prop. Dist. 0.682 0.059
Log TCID -5.682 -6.059
TCID50 2.081E-06 8.732E-07
1/TCID50 4.806E+05 1.145E+06
TCID50/ml 4.806E+07 1.145E+08
pfu/ml 3.32E+07 7.90E+07
Average pfu/ml: (3.32E+07)+(7.9E+07)/2= 5.61E+07
144
APPENDIX E
Working Solution for ELISA
1. Coating Buffer-Carbonate-bicarbonate buffer (50ml)
0.05M carbonate sodium carbonate, pH 9.6.
Dissolve 1 sodium carbonate-bicarbonate capsule (Sigma) in 100ml distilled
water.
Add 1M HCl to adjust pH to 9.6
Add distilled water to a total volume of 50ml
2. Tris Buffered Saline (TBS)
50mM Tris/0.14M NaCl, pH 8.0
Dissolve 12.11g Tris, 58.44g NaCl in 1800 ml distilled water.
Add 1M HCl to adjust pH buffer to pH8.0.
Add distilled water to a total volume of 2000ml
3. Wash Solution
50mM Tris buffered saline, pH 8.0; 0.14M NaCl,; 0.05% Tween 20
Add 0.5ml Tween 20 to 1000ml TBS
4. Blocking Solution (freshly prepared)
50mM Tris, 0.14 M NaCl, 1%BSA, pH8.0.
Dissolve 0.5g BSA with 50ml TBS.
5. Sample/Conjugate Diluent (freshly prepared)
50mM Tris, 0.14 M NaCl, 1%BSA, 0.05% Tween 20, pH 8.0
Dissolve 1g BSA with 100ml washing solution
145
6. Stopping Solution
1M Phosphoric Acid
Diluted 125ml 4M phosphoric acid to a total volume of 500ml 1M phosphoric
acid
7. Preparation of Standard
Step Ng/ml Calibrator Sample diluents
0 1000 5ul 1ml
1 250 1ml from step 0 1ml
2 125 1ml from step 1 1ml
3 62.5 1ml from step 2 1ml
4 31.25 1ml from step 3 1ml
5 15.625 1ml from step 4 1ml
6 7.8 1ml from step 5 1ml
7 3.9 1ml from step 6 1ml
146
APPENDIX F
Working Solution for Western Blot
1. Towbin Buffer
24 mM Tris, 192 mM glycine and 10% methanol
2. 1X Phosphate Buffer Saline (PBS), pH 7.4
8g NaCl, 0.2g K2HPO4 and 0.24g KH2PO4
Adjust to pH7.4 and total up volume to 1L.
Sterilize by autoclaving
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)
147
APPENDIX G
Mobile Phase for Purification
Buffer for HIC
(1) 2.4M Ammonium Sulphate/0.8M Sodium Citrate Buffer, pH6.0 (2x Buffer A)
Dissolved 79.28g ammonium sulphate, 50.5g sodium citrate and 5.38g citric acid
in 250ml distilled water.
(2) 1.2M Ammonium Sulphate/0.4M Sodium Citrate Buffer, pH6.0 (Buffer A)
Dissolved 158.568g ammonium sulphate, 107.052g sodium citrate and 6.917g
citric acid in 1000ml distilled water
Buffer for IEX
(1) 20mM Tris-HCl, pH8.5
Dissolved 4.846g Tris in 1800ml distilled water.
Adjust pH with 1M HCl to pH8.5
Total up volume to 2000ml with distilled water.
(2) 20mM Tris-HCl/ 0.5M NaCl, pH8.5
Dissolved 2.4223g Tris and 29.22g NaCl in 800ml distilled water
Adjust pH with 1M HCl to pH8.5
Total up volume to 1000ml with distilled water.
148
APPENDIX H
Glycan Analysis
Stock Solution
Chemical Molecular Weight
(g/mol) Concentration
Mannose 180.16 5mg/ml
Glucosamine 215.64 5mg/ml
D-Galactose 180.16 5mg/ml
D-Glucose 180.16 5mg/ml
Arabinose 150.13 5mg/ml
TFA 114.02 4.4M
NaOH 40.00 0.3M
HCl 36.45 0.3M
Apo-Transferrin 76000 383.8µg/ml
Sample 73000 383.8µg/ml
149
Example for the Preparation for sample and standards for carbohydrate hydrolysis
Volume
Apo Transferrin
(365µg/ml)
Sample/rhTf
(365µg/ml)
Std 1
(10µg/ml )
Std 2
(5µg/ml)
Std 3
(2.5µg/ml)
Std 4
(1.25µg/ml)
Apo-Transferrin
(383.8µg/ml)
95µl - - - - -
Sample
(383.8µg/ml) - 95µl - - - -
Mannose
(50µg/ml) - - 20µl 10µl 5µl 2.5µl
Glucosamine
(50µg/ml) - - 20µl 10µl 5µl 2.5µl
Galactose
(50µg/ml) - - 20µl 10µl 5µl 2.5µl
Glucose
(50µg/ml) - - 20µl 10µl 5µl 2.5µl
Water 5µl 5µl 20µl 60µl 80µl 90µl
Total Volume 100µl 100µl 100µl 100µl 100µl 100µl
*Each sample/standard was mixed in hydrolysis tube and added with 1ml TFA.
APPE
ND
IX H
: Continued
M2 M1
149
150
APPENDIX H: Continued
Mobile phase of HPLC
(1) Mobile Phase A
0.1M Ammonium Acetate/ 10% Acetonitrile, pH5.5
Dissolved 7.708g Ammonium Acetate in 250ml acetonitrile and about 600ml
deionized water
Adjust pH with 1M HCl to pH5.5
Total up volume to 1000ml with deionized water.
(3) Mobile Phase B
0.1M Ammonium Acetate/ 25% Acetonitrile, pH5.5
Dissolved 7.708g Ammonium Acetate in 250ml acetonitrile and about 600ml
deionized water
Adjust pH with 1M HCl to pH5.5
Total up volume to 1000ml with deionized water.