University of Southern California Dissertations and Theses

Human growth hormone-transferrin recombinant fusion protein for oral delivery

(USC Thesis Other)

Human growth hormone-transferrin recombinant fusion protein for oral delivery

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
HUMAN GROWTH HORMONE – TRANSFERRIN RECOMBINANT FUSION
PROTEIN FOR ORAL DELIVERY

by
Nurmamet Amet

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PHARMACEUTICAL SCIENCES)

August 2008

Copyright 2008 Nurmamet Amet

ii
Dedication

For my parents who devoted much of their life providing the best
opportunities for their eight sons. Also I like to dedicate this work to
the memory of my immediate younger brother, best friend –
Memet’eli.

iii
Acknowledgements

I wish to thank my advisor Dr. Shen for providing me outstanding
opportunities, advices, and insightful tips that helped me grow
professionally and personally, and for guiding me throughout the
project. I am also grateful for Dr. Shen and Mrs. Daisy Shen’s kind,
pleasant attitude to all of us in the lab.

I want to express my gratitude to Dr. Ian S. Haworth and Dr. Vijay K.
Kalra who as part of my dissertation committee gave me exceptional
support and help.

Great appreciation also goes to my lab-mates who are very supportive,
helpful and kind.

Lastly, I like to thank my wife Nurnisa for her full support, trust,
motivation and encouragement that are essential for my success.
Finally, my beautiful daughters – Berna and Rena – are very patient
and unusually understanding of Daddy, and I like to thank them.

iv
Table of Contents
Dedication ii
Acknowledgements iii
List of Tables viii
List of Figures x
Abbreviations xiv
Abstract xv
Part 1. Human growth hormone-transferrin
fusion protein 1

Chapter 1. Introduction 1
Chapter 2. Materials and Methods 13
2.1. Cloning in pCR-Blunt II-TOPO vector
2.2. Fusion constructs in mammalian
expression vector 15
2.2.1. hGH-Tf expression construct 15
2.2.2. Tf-hGH expression construct 16
2.2.3. hGH expression construct 19
2.3. Small-scale transfection 23
2.4. SDS-PAGE 24
2.5. Silver staining 25
2.6. Western Blot 26
2.7. Characterizing for in vitro bioactivities 26
2.7.1. TfR binding 26
2.7.2. hGH bioactivity 28
2.7.2.1. JAK2/STAT5 activation 29
2.7.2.2. Nb2 cell proliferation 31

v
Chapter 3. Results 33
3.1. Plasmid construction 33
3.2. Fusion protein expression 44
3.3. Fusion protein detection 47
3.4. In vitro biological activity 50
3.4.1. TfR binding 50
3.4.2. Nb2 cell proliferation and
JAK2/STAT5 activation 52

Chapter 4. Discussions 57
4.1. Cloning strategy 57
4.2. Fusion protein with different orientation 58
4.3. Fusion protein with high expression 59
4.4. TfR binding capability 60
4.5. hGH bioactivity 62

Part 2. Insertion of the designed helical linker
led to increased expression of the Tf-
based fusion proteins 65

Chapter 5. Introduction 65

Chapter 6. Materials and Methods 68
6.1. Helical linker insertion 68
6.2. Cleavable linker insertion 69
6.3. Western blots 69

Chapter 7. Results 71
7.1. Linker insertion 71
7.2. Comparing hGH-Tf to hGH-(H4)
2
-Tf 76
7.3. Comparing Tf-hGH to Tf-(H4)
2
-hGH 83
7.4. In vitro biological activities 84
7.4.1. TfR binding 84
7.4.2. Nb2 cell proliferation 86

Chapter 8. Discussions 88

vi
Part 3. Producing fusion proteins at large-scale 92
Chapter 9. Introduction 92
Chapter 10. Materials and Methods 97
10.1. Large-scale plasmid preparation 97
10.2. Large-scale transient transfection 98
10.3. Formulation strategy 100
10.4. Radioimmunoassay 103
10.5. Stable transfection 105

Chapter 11. Results and Discussions 107
11.1. Large-scale expression 107
11.2. Generation of stable cell lines 117

Part 4. In vivo bioactivity of the fusion proteins 121
Chapter 12. Introduction 121
Chapter 13. Materials and Methods 124
13.1. Hypophysectomized rats and dosing 124
13.2. Trypsin/chymotrypsin proteolysis and
soybean trypsin inhibitor 126
13.3. Statistical analysis 127

Chapter 14. Results 128
14.1. s.c. administration of GT fusion protein
promoted weight gain while p.o.
did not have impact on weight increase 128
14.2. s.c. administration of GHT fusion protein
promoted significant weight gain meanwhile
p.o. led to modest weight gain 133
14.3. s.c administration of THG fusion protein
promoted significant weight gain meanwhile
p.o. did not produce weight gain 137
14.4. Choosing GHT fusion protein for further
in vivo evaluation 140
14.5. Dose-escalation study with GHT 140
14.6. Chymotrypsin led to degradation of GHT 145
14.7. Chymotrypsin led to degradation of THG 149

vii
14.8. Trypsin led to degradation of GHT 151
14.9. Trypsin led to degradation of THG 153
14.10. Transferrin was resistant to both trypsin
and chymotrypsin 155
14.11. Soybean trypsin inhibitor blocked
tryptic and chymotryptic cleavage of
GHT and THG 155
14.12. GT fusion protein is susceptible to both
tryptic and chymotryptic degradation 158
14.13. Oral coadministration of GHT with BBI
led to weight gain with improved
consistency 160
14.14. High dose of GHT fusion protein
coadministered with BBI did not promote
body weight gain 165

Chapter 15. Discussions 170
Chapter 16. Summary/Conclusion/
Future perspective 175

16.1. Summary 175
16.2. Conclusion 179
16.3. Future perspective 180

Bibliography 182

viii
List of Tables

Table.1. Primer sequences that were designed to generate
coding sequences for hGH-Tf and Tf-hGH fusion
constructs 21

Table.2. PCR conditions that were used to amplify coding
sequences for hGH and Tf 22

Table.3. Time course of JAK2 phosphorylation 55

Table.4. The ratio of expression with or without
helical linker 78

Table.5a. Oral and subcutaneous administration of GT
fusion protein 131

Table.5b. Oral and subcutaneous administration of GT
fusion protein 132

Table.6a. Oral as well as subcutaneous administration of
GHT fusion protein 135

Table.6b. Subcutaneous administration of hGH as positive
control, and buffer negative control 136

Table 7. Oral and subcutaneous administration of THG
fusion protein 139

Table 8a. Dose-escalation study with GHT fusion protein
and hGH 143

Table 8b. Subcutaneous administration of GHT fusion
protein as positive control, along with buffer
negative control 144

Table 9a. Oral administration of GHT fusion protein with/
without BBI 162

Table 9b. Oral administration of hGH with/without BBI 163

ix
Table 9c. Subcutaneous administration of GHT as positive
control, along with no treatment negative control 164

Table 10a. GHT was administered subcutaneously,
along with no treatment control 167

Table 10b. Coadministration of GHT with BBI orally at
high doses 168

Table 10c. Coadministration of hGH with BBI at high dose
and subcutaneous administration of hGH as
positive control 169

x
List of Figures
Figure 1. Structural highlights of transferrin 4

Figure 2. The schematic representation of hGH and
receptor binding 8

Figure 3. Human serum albumin-based fusion protein
technology 10

Figure 4. Computer generated molecular model for
hGH-Tf fusion protein with short linker 12

Figure 5. Schematic diagram illustrating direct ligation 18

Figure 6. Cloning hGH with signal sequences to
pCR-TOPO Vector 34

Figure 7. Tf without signal sequences was cloned into
pCR-TOPO vector 35

Figure 8. Subcloning of Tf to expression vector pcDNA3. 36

Figure 9. Subcloning of hGH upstream to Tf in pcDNA3.1 37

Figure 10. Creating Tf-hGH expression construct 38

Figure 11. Expression construct for hGH 39

Figure 12. Analyzing Tf with DNA sequencing 40

Figure 13. Assembled DNA-sequence for Tf 41

Figure 14. Comparing Tf sequence with published Tf 42

Figure 15. Enzymatic analysis of expression constructs 43

Figure 16. SDS-PAGE with Silver stain 45

Figure 17. SDS-PAGE with Coomassie stain 46

xi
Figure 18. Western blot with anti-hGH antibody 48

Figure 19. Western blot with anti-Tf antibody 49

Figure 20. Both hGH-Tf and Tf-hGH had bioactivity for Tf 51

Figure 21. Both hGH-Tf and Tf-hGH led to JAK2 activation 53

Figure 22. Both hGH-Tf and Tf-hGH led to STAT5 activation 54

Figure 23. hGH-Tf led to Nb2 cell proliferation 56

Figure 24. The isolation of helical linker 72

Figure 25. Insertion of helical linker between hGH-Tf 73

Figure 26. Insertion of helical linker between Tf-hGH 74

Figure 27. Introducing cleavable cyclic linker 75

Figure 28. Computer model of helical linker connecting
2 domains 77

Figure 29. High-level expression by the insertion of helical
linker in Tf-fusion proteins was analyzed by
anti-hGH Western blot 79

Figure 30. High-level expression by the insertion of helical
linker in Tf-fusion proteins was analyzed by anti-Tf
Western blot 80

Figure 31. SDS-PAGE revealing high-level expression
resulting from helical linker insertion 82

Figure 32. Fusion proteins containing the helical linker
were tested for their ability to bind TfR in
CaCo-2 cells 85

Figure 33. Fusion proteins containing the helical linker
led to Nb2 cell proliferation 87

xii
Figure 34. Agarose gel of high quality plasmids 108

Figure 35. Large-scale expression of hGH:
1
st
collection 109

Figure 36. Large-scale expression of hGH:
2
nd
collection 110

Figure 37. Large-scale expression of GHT fusion protein:
1
st
collection 111

Figure 38. Large-scale expression of GHT fusion protein:
2
nd
collection 112

Figure 39. Large-scale expression of GHT fusion protein:
3
rd
collection 113

Figure 40. GHT fusion protein expressed at large-scale
in 40 T225 flasks 114

Figure 41. Production of THG fusion protein in large-scale 115

Figure 42. Stable clone selection with anti-hGH Western blot 119

Figure 43. Confirming stable cell line GHT-14 for expression 120

Figure 44. Evaluating GT fusion protein for both oral and
subcutaneous bioactivity 130

Figure 45. Evaluating GHT fusion protein for both oral
and subcutaneous bioactivity 134

Figure 46. Evaluating THG fusion protein for both oral and
subcutaneous bioactivity 138

Figure 47. Dose-escalation of GHT with oral administration 142

Figure 48. Chymotrypsin dependent cleavage of GHT and
inhibition by Bowman Birk Inhibitor 147

xiii
Figure 49. Both GHT and THG were intact after 2 hr
Incubation at 37
o
C 148

Figure 50. Chymotrypsin dependent cleavage of THG and
inhibition by Bowman-Birk Inhibitor 150

Figure 51. Trypsin led to a time-dependent proteolysis of
GHT fusion protein but Bowman Birk Inhibitor
blocked trypsin activity 152

Figure 52. Trypsin led to a time-dependent proteolysis of
THG fusion protein but Bowman Birk Inhibitor
Blocked trypsin activity 154

Figure 53. Transferrin demonstrated strong resistance to
trypsin 156

Figure 54. Transferrin demonstrated strong resistance to
chymotrypsin 157

Figure 55. Both trypsin and chymotrypsin led to a time
dependent proteolysis of GT fusion protein 159

Figure 56. Oral administration of GHT fusion protein
together with protease inhibitor BBI led to a
small weight increase 161

Figure 57. Coadministration of BBI with GHT fusion protein
at high dose did not promote weight gain 166

xiv
Abbreviations
hGH Human growth hormone
hGH-R hGH receptor
Tf Human serum transferrin
TfR Tf receptor
(H4)
2
2 copies of helical linker [A(EAAAK)
4
A]
2

H4 1 copy of helical linker A(EAAAK)
4
A
GT hGH-Tf fusion protein consisting of hGH and Tf
GHT hGH-(H4)
2
-Tf fusion protein with 2 copies of helical linker
TG Tf-hGH fusion protein with Tf at N-terminus
THG Tf-(H4)
2
-hGH fusion protein with 2 copies of helical linker
GCT GT fusion protein with cleavable linker
TCG TG fusion protein with cleavable linker
JAK2 Janus kinase 2
Stat5 Signal transducer and activator of transcription
IM-9 Human B-lymphoma cells
Nb2 Rat T-lymphoma cells
s.c. Subcutaneous administration
p.o. Oral administration
BBI Soybean trypsin inhibitor/Bowman-Birk Inhibitor

xv
Abstract

Protein drug delivery is limited to needle injection associating with
pain, inconvenience, and non- compliance. Therefore, protein drug
with oral dosage form is preferred. We used transferrin (Tf) receptor
mediated transcytosis and Tf-based recombinant fusion protein
approach to achieve oral delivery of human growth hormone (hGH).
Plasmid constructs expressing the fusion proteins were established by
fusing coding sequences of both hGH and Tf in frame. Fusion proteins
were produced in serum free media by transcient transfection of
human embryonic kidney cells. The SDS-PAGE of conditioned media
showed that fusion proteins expressed at ~ 90% abundance and 100
kDa molecular weight; the Western blot analysis with anti-hGH and
anti-Tf antibodies verified the identity. The Nb2 cell proliferation and
TfR binding assays demonstrated that fusion proteins retained
bioactivity for both hGH and Tf, respectively. To minimize inter-domain
interference, we inserted 2 copies of helical linker as a spacer between
hGH- and Tf-domain, and discovered that it not only enhanced the
bioactivity but also led to high-level expression (30-50 mg/L, in select
fusion protein). Three fusion proteins (GT, GHT and THG) were
assessed in hGH-deficient hypophysectomized rats for in vivo

xvi
biological activity, with both subcutaneous and oral administration. 7-
day s.c. administration in rats showed both GHT and THG fusion
proteins promoted ~ 16 g body weight gain, whereas GT fusion protein
led to ~ 12 g increase. However, when administered orally, only GHT
fusion protein led to a small weight gain. We also investigated tryptic
and chymotryptic stability of fusion proteins and found that these
fusion proteins were highly susceptible to both enzymes, while Tf was
resistant. 7-day oral coadministration of GHT fusion protein with
soybean trypsin inhibitor resulted in modest body weight gain, with
better consistency. In conclusion, Tf-based recombinant fusion protein
approach has a potential for oral delivery of protein drugs.

1
Part 1. Human growth hormone-transferrin fusion
protein (hGH-Tf)

Chapter 1. Introduction

The success of recombinant DNA technology has led to the
development and approval of 165 biopharmaceutical drugs as of 2006,
and more than 400 of such drugs are currently in various stages of
clinical trials (56). These biological drugs help treat millions of patients
with indications such as cancer, infectious disease, heart failure and
metabolic disorders. Biological drugs are highly specific and less toxic
compared to the traditional small molecule drugs, and are only
treatment option for conditions such as insulin-dependent diabetes and
other indications resulting from deficiency of a functional protein such
as the human growth hormone. The demand for protein drugs is
growing with increased numbers as new indications and symptoms are
discovered and diagnosed, and these drugs are establishing
themselves as major player by accounting for more than 30 % of total
market share in pharmaceutical industry (53).

2
Protein drug administration is mainly limited to invasive, needle
injection, which is associated with pain, inconvenience, non-
compliance and reduced-accessibility to children and elderly patients.
As an alternative to needle injection, the pulmonary, nasal and trans-
dermal routes have been explored for protein drug delivery with
limited success including poor acceptance by patients and medical
community (10). Therefore, protein drugs with oral administration are
highly desirable but development of oral dosage forms faces significant
challenges owing to the physicochemical and biological properties of
the protein molecules such as poor absorption across the
gastrointestinal (GI) tract, large molecular size, less hydrophobicity
and high sensitivity to proteases that digest proteins into small
peptides with no bioactivity. To overcome those problems associated
with oral absorption of protein drugs, different strategies including
enteric coating, penetration enhancers and protease inhibitors have
been attempted as part of formulations (9, 11, 19). However, the
safety and efficacy of such strategies are not well characterized for
common, long-term application, as a result, are not accepted by the
Food and Drug Administration (FDA) (9)

3
Therefore, we sought to use a natural, physiological process to
investigate the feasibility to design an oral dosage form that delivers
protein drugs orally. This approach exploits receptor mediated
transcytosis mechanism coupled with transferrin-based recombinant
fusion protein to achieve oral delivery of protein drugs.

Transferrin (Tf) is a protein present in serum that transports iron to
cells expressing Tf-receptor (TfR) on the surface of plasma membrane
through Tf-TfR mediated-endocytosis. Tf binds to two atoms of iron
and becomes diferric-Tf, which then forms Tf-TfR complex at
physiological pH of 7.2 (Fig.1). Next, this complex is endocytosed to
cytoplasm as endosome, and iron is released under the low pH (~ 5.5)
environment in endosome (12, 24, 28, 32). Majority of Tf-TfR
complexes recycle back to the plasma membrane where Tf
disassociates from TfR, and the whole cycle starts again. To some
extent, the Tf-TfR complexes travels across the basolateral epithelial
cell membrane, thus enabling the transport of Tf-conjugated drug
moieties to the blood stream (22, 33, 65, 66). This process is known
as Tf-TfR mediated transcytosis (36) (31).

4

Fig. 1. The structural highlights of rabbit transferrin (r-Tf), which has over
80% of sequence identity with human Tf. Tf is composed of N and C-lobes,
each having binding site for one ferric iron (Fe
3+
). Residues in iron binding as
well as receptor binding sites are highly conserved among mammalian
species and they are important in the function of transferrin. Some of the
essential residues, along with two atoms of iron (large yellow ball), were
highlighted. Therefore, these sites should be free of any obstructions in order
for Tf to have optimal biological function, when fusion proteins are generated
using Tf as a partner, such as Tf-hGH fusion protein. WebLab Viewer program
was applied to generate the structure; the PDB accession number is: 1JNF

5

Tf has been applied as a vehicle to deliver small molecule drugs,
peptides, proteins, and genes to the target tissues including blood
brain barriers that express TfR abundantly (22) (43) (24). However,
most of the approaches used in these delivery schemes are limited to
the cell lines. Additionally, as endogenous protein abundantly present
at blood serum, Tf is resistant to enzymatic cleavage resulting from
both trypsin and chymotrypsin (unpublished data), which represent
major proteases in the GI. Moreover, human GI epithelial cells express
high number of TfR, a major player in Tf-TfR mediated transcytosis
(7). Therefore, Tf is a rationale choice as a carrier for oral delivery of
protein drugs.

For example, the chemical conjugates of Tf and protein drugs such as
granulocyte colony stimulating factor (G-CSF) and insulin (In) show
that both Tf and its conjugates not only transported through epithelial
or alveolar cells, but also demonstrated in vitro as well as in vivo
bioactivities as demonstrated in cell cultures and animal models (65).
However, the high manufacturing cost, less product homogeneity and
less reproducibility, associated with the chemical approach generally
limits its application as a viable choice for oral delivery (70).

6

Human growth hormone (hGH), as a biotechnology drug, is an
important regulator of metabolism and stimulates growth and
differentiation of target tissues such as muscle, bone, cartilage, and
liver (54, 62, 63). hGH exerts its effect directly, or indirectly via the
insulin-like growth factor-I (IGF-I) by binding two molecules of human
growth hormone receptors (hGH-R) on plasma membrane of the target
cells (Fig. 2), which then activates down-stream signaling molecules
(3, 18, 47). Subsequently, hGH brings about up-regulation of genes
that promote growth and differentiation in tissues such as liver and
muscle. Conceivably, the deficiency of hGH is associated with several
clinical indications including short stature, turner syndrome, chronic
kidney disease, HIV-associated wasting, and abnormal metabolism
(48). If left untreated, these indications pose significant health risk to
society.

Current treatment regimen for patients with growth hormone
deficiency is limited to needle injection of recombinant hGH. Prior to
the market approval of recombinant hGH by the United States Food
and Drug Administration (FDA), patients with hGH-deficiency were
treated with hGH extracted from the pituitary gland of human

7
cadavers. This was changed when Genentech (South San Francisco,
CA, USA) developed and obtained approval for recombinant hGH (r-
hGH) in 1985. Since then, many patients with hGH deficiency such as
dwarfism have benefited from this recombinant protein drug. But like
other peptide and protein drugs, the administration of hGH is limited
to invasive-injection several times a week (51, 52, 56), thereby
suggesting that there is a pressing need for oral hGH, preferably with
prolonged plasma half-life.

8

Fig. 2. The schematic representation of hGH and receptor binding. hGH first
binds to one molecule of receptor with high affinity, followed by subsequent
binding of second molecule of receptor, to which it has weaker affinity. Thus
binding of hGH results in dimerization and subsequent activation of its
receptor. More importantly, residues from C- terminal end of hGH play major
role in receptor binding, and thus this should be considered in fusion protein
construction, as in the case of Tf-hGH fusion protein. The image was
generated by WebLab Viewer molecular modeling program; the coordinates
for hGH were obtained from Protein Data Bank (PDB), with the accession
number of: 1HWG.
hGH
hGH-R
hGH-R

9

Recombinant fusion protein approach has been used in
biopharmaceutical industry as means to improve both plasma half-life
and targeting of a protein drug with poor pharmacokinetics and poor
efficacy (Fig. 3). Albumin-Interferon-2αb and IgG-TNFα fusion proteins
are few examples demonstrating the acceptance of this approach (49).
Recently, our lab has produced bioactive G-CSF-Tf recombinant fusion
protein in mammalian cells and demonstrated oral mylopoeitic activity
in BDF1 mice (4, 5). This finding then led to the hypothesis that
recombinant fusion protein consisting of hGH-Tf could be constructed
and produced in bioactive form in mammalian cells and that such
fusion proteins could then demonstrate oral activity by promoting
weight gain in animal model of hGH when administered orally.

10

Fig.3. Human serum albumin-based fusion protein technology. It permits the
creation of recombinant, long-acting protein drugs by fusing the cDNA (gene
X) encoding human albumin (HA) to cDNAs encoding therapeutically active
proteins. Fusion can occur at either the N- or C-terminal of HA (arrows
indicate transcription initiation). Molecular models depicting HA fusion to a
large cytokine (IFN--2b) or a small peptide (glucagon-like peptide 1)
illustrate the adaptability of the albumin fusion platform to biologically active
compounds of varying sizes. (b) Reduced dosing frequency and sustained
exposure. Nature Biotechnology (2007), 25, 1411-1419.

11

Therefore, in this dissertation we investigated the feasibility of
producing bioactive hGH-Tf and Tf-hGH fusion proteins in mammalian
HEK293 cells and evaluated in vivo growth promoting activity in
hypophysectomized rats as animal model for hGH, focusing on oral
absorption. Gene fusion constructs were established by fusing coding
sequences of hGH and Tf in frame in mammalian expression vector.
The resulting hGH-Tf and Tf-hGH fusion proteins (Fig. 4) were
produced in mammalian HEK293/293T cells and characterized for dual
biological activity, including both Tf and hGH. After confirming in vitro
biological activity, selected fusion proteins were administered both
subcutaneously and orally in hypophysectomized rats for 7-days to
evaluate in vivo bioactivity.

12

hGH
Fig. 4. Computer generated molecular model for hGH-Tf fusion
protein, connected by a short linker. Highlighted are amino acids
known to play important role in receptor binding, and two iron atoms
(in red) positioned in iron-binding pocket. The structure was
generated using WebLab Viewer and Insight II molecular modeling
programs. The coordinates used in this model are HWG for hGH and
1JNF for rat-Tf, both were obtained from protein data bank (PDB).

13

Chapter 2. Materials and Methods

2.1. Cloning in pCR-Blunt II-TOPO vector

The DNA sequences coding (20, 23, 46) for hGH was amplified in a
PCR reaction containing cDNA from human pituitary gland (Clontech),
platinum pfx DNA polymerase (Invitrogen), and hGH-specific primers
with EcoRV and XhoI sites (USC Norris Cancer Center). In addition, the
amplification primers incorporated both the Kozak and signal
sequences upstream to coding sequences. The PCR amplified hGH
fragment (668 bp) was ligated to pCR-Blunt II-TOPO vector
(Invitrogen) and subsequently transformed to chemically competent
TOP10 cells (Invitrogen). Plasmids prepared from overnight E. coli
cultures were analyzed by restriction enzymes, and selected plasmids
were sequenced (USC Norris Cancer Center) to confirm that the hGH
coding sequences were correctly amplified without mutations.

Likewise, the DNA sequences coding for Tf (71) was amplified from
TFR27 plasmid (ATCC) using platinum pfx DNA polymerase with
proofreading and Tf-specific primers incorporating XhoI and XbaI

14
restriction enzyme sites and stop codon. The resulting 2052 bp Tf-
fragment was then cloned into pCR-Blunt II-TOPO vector, and
subsequently transformed to chemically competent TOP10 cells. The
plasmids prepared from overnight E. coli culture were analyzed with
restriction enzymes, followed by DNA sequencing to confirm that the
Tf coding sequences were correctly amplified without mutations.

Additionally, the Tf coding region including the signal sequences was
amplified using primers with incorporated-EcoRV and XhoI restriction
enzyme sites. The PCR amplified fragment was cloned into pCR-Blunt
II-TOPO vector and transformed to chemically competent TOP10 cells.
The stop codon from Tf was deleted to avoid interruption of
translation. Similarly, the hGH coding region without the signal
sequences was PCR amplified, cloned into pCR-Blunt II-TOPO vector,
and transformed to chemically competent TOP10 cells. Mini-plasmid
preps from overnight E. coli cultures were analyzed using DNA
sequencing and restriction enzyme analysis to ensure that both Tf and
hGH sequences were correctly amplified and cloned.

Furthermore, hGH coding region including signal sequences and stop
codon was amplified by primers that had EcoRV and XhoI restriction

15
enzyme sites at 5’-end. The amplicon was cloned into pCR-Blunt II-
TOPO vector, and subsequently, transformed into TOP10 competent
cells. Miniplasmids prepared from overnight E. coli cultures were
analyzed with restriction enzyme and DNA sequencing.

2.2. Fusion constructs in mammalian expression vector

2.2.1. hGH-Tf expression construct

The pCR-Blunt II-TOPO plasmid harboring the Tf fragment was double
digested using XhoI and XbaI restriction enzymes to release the Tf
coding sequences with cohesive ends, followed by gel extraction and
purification (Invitrogen). Likewise, the pcDNA3.1(+) expression vector
was digested using same restriction enzymes (XhoI and XbaI),
dephosphorylated with calf intestinal phosphatase (Invitrogen) to
prevent self-ligation, and prepared by gel extraction and purification.
Both fragments were quantified using agarose gel analysis and A
260

absorbance to determine insert:vector ratio for effective cloning.
Subsequently, the gel-purified Tf fragment was ligated to gel- purified
pcDNA3.1(+), and transformed into chemically competent TOP10 cells.
The plasmid mini-preps from the overnight culture of E. coli were

16
analyzed with restriction enzymes to identify positive colonies with the
correct insert, and the selected plasmids were analyzed by DNA
sequencing.

Next, the hGH coding region was released from pCR-Blunt II-TOPO
plasmid by EcoRV and XhoI digestion, gel purified, and fused upstream
to Tf coding region in pcDNA3.1(+) plasmid with EcoRV and XhoI
cohesive ends. The resulting plasmid construct, hGH-Tf, was
transformed into chemically competent TOP10 cells, and the selected
plasmids isolated from overnight cultures were analyzed for positive
colonies with the correct sequences.

2.2.2. Tf- hGH expression construct

The pCR-Blunt II-TOPO plasmid harboring the Tf fragment with signal
sequences was double digested using EcoRV and XhoI restriction
enzymes (Fig.5) to release the Tf coding sequences with cohesive
ends, followed by gel extraction and purification (Invitrogen). Likewise,
the pcDNA3.1(+) expression vector was digested using same
restriction enzymes (EcoRV and XhoI), dephosphorylated with calf
intestinal phosphatase (Invitrogen) to prevent self-ligation, and

17
prepared by gel extraction and purification. Both fragments were
quantified using agarose gel analysis and A
260
absorbance to
determine insert:vector ratio for effective cloning.

18

Fig. 5. Schematic diagram illustrates the steps involved in constructing Tf-
hGH expression plasmid using direct ligation approach. There are several
approaches available and all of which are well-established and work well.
Shown here is one of several approaches that could be used for the cloning
work.
Tf gene + sig
PCR, followed by
RE
EcoRV
Xho I
XhoI
Ligation
Xho I EcoR V Xba I
Xba I
hGH gene
EcoRV
Xba I
Xho I
pcDNA3.1

19

Subsequently, the gel-purified Tf fragment containing the signal
sequences was ligated to gel- purified pcDNA3.1(+), and transformed
into chemically competent TOP10 cells.

The plasmid mini-preps from the overnight culture of E. coli were
analyzed with restriction enzymes to identify positive colonies with the
correct insert, and the selected plasmids were analyzed by DNA
sequencing. Next, the hGH coding region without signal sequences was
released from pCR-Blunt II-TOPO plasmid by XhoI and XbaI digestion,
gel purified, and fused downstream to Tf coding region in pcDNA3.1(+)
plasmid with XhoI and XbaI cohesive ends. The resulting plasmid
construct, Tf-hGH, was transformed into chemically competent TOP10
cells, and the selected plasmids isolated from overnight cultures were
analyzed for positive colonies with the correct sequences.

2.2.3. hGH expression construct

hGH coding region containing both signal sequences and stop codon
was obtained from cloning vector pCR-Blunt II-TOPO by EcoRV and
XhoI digestion, followed by gel purification. Similarly, the cloning

20
vector pcDNA3.1(+) as well was digested with the same restriction
enzymes and dephosphorylated. And then they were ligated and
transformed into competent TOP10 cells. Miniplasmids were prepared
from overnight E.coli cultures, and analyzed for positive colonies with
the correct sequence. Glycerol stocks of the positive colony were
prepared with 20 % glycerol in LB medium using aseptic technique,
and stored at – 80 °C. This construct would be used to express hGH as
control for fusion protein.

21

Constructs Primer name Primer sequences
Tf_NF GATATC CCACC ATG AGG CTC GCC GTG
Tf-hGH Tf_NF CTCGAG AGG TCT ACG GAA AGT GCA GC
hGH_CF CTCGAG TTC CCA ACC ATT CCC TTA TC
hGH_CR TCTAGA CTA GAA GCC ACA GCT GCC CT
GH_NF GATATC CCACC ATG GCT ACA GGC TCC
hGH-Tf GH_NR CTCGAG GAA GCC ACA GCT GCC CTC CA
Tf_CF CTCGAG GTC CCT GAT AAA ACT GTG AG
Tf_CR TCTAGA CTA AGG TCT ACG GAA AGT GC
Table. 1 Primer sequences that were designed to generate
coding sequences for hGH-Tf and Tf-hGH fusion constructs.

22

Coding sequence PCR Condition Product size (bp)
denature
94 C – 2 min
# of cycles: 35
hGH 94 – 30 sec 668/588
55 – 30 sec
72 – 1 min
extention
72 – 7 min
denature
94 C – 2 min
# of cycles: 35
Tf 94 – 30 sec 2111/2052
63 – 30 sec
72 – 2 min
extention
72 – 7 min
Table. 2. PCR conditions that were used to amplify coding sequences for
hGH and Tf.

23
2.3. Small-scale transfection

Fusion proteins were expressed by the transient-transfected HEK293/
HEK293T cells in a protein free CD293 media. Prior to transfection,
high-quality plasmid constructs were prepared using Endo-toxin free
reagents (Qiagen), and quantified using both agarose gel analysis and
UV absorbance (@ 260 nm). The human embryonic kidney 293 cells
(HEK293) (ATCC) were cultured in DMEM media (Mediatech)
containing 10% FBS, 50 units penicillin/50 µg streptomycin in a
humidified incubator at 37 deg with 5% CO2. Cells were seeded in 6-
well plates (Costar) with ~ one million cells per well one day before
transfection.

On the day of transfection, the regular growth (DMEM with 10 % FBS)
media was replaced with Opti-Mem reduced serum media (Invitrogen).
2 µg expression construct and 5.5 µl Lipofectamine 2000 (Invitrogen)
were diluted in 250 µl Opti-Mem reduced serum media, respectively,
and mixed, followed by 15 minutes of incubation at room temperature.
Next, cells were transiently transfected with the complexes (lipid-DNA)
by dropwise and gentle addition into the wells of 6-well plates. These
cells incubated for about 5 hrs, and then the media containing

24
transfection agent and plasmid DNA was removed and replaced with
CD293 media. Then, cells were incubated for five days to allow them
to express fusion protein in CD293 media.

2.4. SDS-PAGE

The conditioned media from 5-day post-transfection was harvested,
clarified by centrifugation for 15 minutes at 4 deg, and then
concentrated using Amicon Ultra-4 or Ultra-15 filtering devices
(Millipore) by centrifugation at 4 °C. Small aliquots from this enriched
fusion protein was boiled for 5 minutes in sample loading buffer, and
then fractionated on a 10% SDS-PAGE or precast gel (Thermo
Scientific). After fractionation, the gel was stained with 0.1%
Coommassie blue in 10% acidic acid and 40% methanol for 1 hour
with constant shaking, and destained until background became clear.
The quantity of fusion protein was estimated using SDS-PAGE or
Western blot with known Tf-standards. Protein bands were detected
and analyzed using Discovery Imaging System and Quantity One
software (Bio-Rad).

25
2.5. Silver staining

50 µl-conditioned media (without further concentration or enrichment)
from 5-day post-transfection was fractionated on a 10 % SDS-PAGE to
evaluate the level of expression and its abundance amongst other
proteins. For protein visualization, the silver staining as a sensitive
method was carried out according to the instructions provided by the
Vendor (Invitrogen). Once protein bands revealed, the silver staining
was terminated by the addition of stop solution, and Discovery
Imaging System was used to image the gel.

2.6. Western Blot

After fractionation with SDS-PAGE, the fusion proteins were
transferred to PVDF membrane (GE healthcare) using wet transfer
technique for one hour at 400 mA. After blocking for 1 hr at room
temperature with 5% non-fat dry milk in PBST buffer, the membrane
was probed with either goat anti-human hGH (1:2000) (Santa Cruz),
or goat anti-human Tf (1:10000) (Sigma) primary antibodies for one
hour. After washing 3x with PBST, the primary antibody was detected
by donkey anti-goat secondary antibody (1:10000) conjugated with

26
HRP. The chemiluminescent signal was detected using ECL plus
reagents (GE Healthcare), and recorded by Discovery Imaging System.

2.7. Characterizing fusion proteins for in vitro bioactivity
2.7.1. TfR binding

After confirming of expression and identity, the fusion protein should
be characterized for in vitro bioactivity using suitable cell models.
CaCo-2 cells (ATCC) closely resembling the cells lining the
gastrointestinal tract express large numbers of TfR, and hence can be
used to evaluate the biological activity of Tf moiety of the fusion
protein. TfR competition binding assay therefore was used to
determine whether the fusion protein could compete with native Tf.
The native Tf was labeled with radioactive iodine (ICN, Irvine, CA)
using chloramines-T catalyzed iodination reaction (4) to obtain iodine-
labeled Tf (
125
I-Tf) that the fusion protein compete to bind TfR on the
surface of CaCo-2 cells.

The
125
I-Tf was then purified using Sephadex-G50 column
chromatography, and the fraction with highest radioactivity was used
for TfR competition binding. Briefly, CaCo-2 cells cultured in DMEM

27
supplemented with 20% FBS, 50 units penicillin/50 µg streptomycin in
a humidified incubator at 37 °C with 5% CO2, were seeded in 12-well
plates and permitted to grow as monolayer, indicative of full
differentiation.

Before assay, CaCo-2 cells were washed 3x with PBS and incubated for
1 hr in serum free DMEM media containing 0.1% BSA to deplete
endogenous Tf. After aspirating the media, 5 µg
125
I-Tf prepared in
serum free DMEM with 0.1% BSA was added to all wells as baseline
radioactivity, followed by unlabeled endogenous Tf with doses of 0, 3,
10, 30, 100 µg/well, fusion protein with doses of 6, 20, and 60 µg/well.
The 12-well plates were then incubated for half an hour to allow
binding reaction to take place at 4 °C. Next, the media was aspirated,
washed with cold PBS for 3x, and cells were dissolved by incubating
for 15 minutes in 0.5 ml 1M NaoH at room temperature.

The dissolved cells were transferred to 5 ml glass tubes and counted
for radioactivity using gamma counter. The specific binding was
obtained for each group of samples by subtracting nonspecific binding
from total binding, and % of surface bound
125
I-Tf was calculated
dividing specific binding from each sample by overall specific binding

28
(or total specific binding of unlabeled Tf). Lastly, the IC
50
value – the
concentration of fusion protein that competes out 50% of surface
bound
125
I-Tf – was assessed from the plot.

2.7.2. hGH in vitro biological activity

hGH regulates metabolism, differentiation and growth of mammalian
cells in various tissues including muscle, bone, cartilage and inner
organs such as liver. hGH accomplishes these functions by binding to
specific hGH-receptors (hGH-R) on target cells, which then cause hGH-
R activation, Janus kinase 2 (JAK2) and STAT5 phosphorylation (13),
ultimately resulting in up-regulation of genes associated with the
function of hGH. To be bioactive for hGH, the fusion protein should
produce hGH-like effect after stimulation of cells expressing hGH
receptor.

IM-9 cells (ATCC), human B-lymphoma cells, express high number of
hGH receptors and are known to respond to hGH stimulation, therefore
they were chosen for hGH bioactivity assay that used tyrosine
phosphorylation of either JAK2 or STAT5. Since fusion protein has Tf-
domain in addition to hGH-domain, it is likely that the kinetics of

29
binding to hGH-R could be different from hGH alone. Therefore, it is of
interest to find out optimal time for stimulation that produce
phosphorylation as a measure of biological activity at detectable level.
Based on the optimal stimulation time, optimal dose for such
stimulation then can be established.

2.7.2.1. JAK2/STAT5 activation in IM-9 cells

Experiments were performed in 6-well plates with about 3 million IM-9
cells cultured as suspension in RPMI-1640 (Mediatech) supplemented
with 10% FBS, 50 units penicillin/50 µg streptomycin in a humidified
incubator at 37 deg with 5% CO2. Before treatment, the IM-9 cells
were washed 3x with RPMI media + 0.1 % BSA, followed by pre-
incubation in the same media for 1 hour at 37 °C. IM-9 cells were
treated with 5 nM fusion proteins and 5 nM hGH and incubated for 0,
5, 15, 30, 45, and 60 minutes. After treatment with above pre-
determined time points, the cells in 6-well plates were harvested
rapidly to stop further stimulation, centrifuged, and pellets kept on ice
until all treatments were completed.

30
The cell pellets were re-suspended and lysed in 50 µl RIPA buffer
(Santa Cruz) containing freshly added protease and tyrosine
phosphates inhibitor on ice for half an hour, centrifuged for 10 min at
13000 rpm, supernatants transferred to new tubes, followed by adding
equal volume of 2x sample buffer. Protein concentration was measured
and 50 µg of total protein was loaded on to 7.5 % SDS-PAGE. The gel
was run at 15 mA/per gel with constant current until the marker pass
the stacking gel, then the current was changed to 20 mA. After gel
run, the proteins were transferred onto PVDF membrane for about 1
hour 20 minutes at 400 mA, followed by blocking in 5 % dry milk in
TBS-T for 1 hour and incubating with primary antibody diluted in
blocking buffer for overnight. HRP-conjugated secondary antibody
against primary antibody was used and signals recorded using
Discovery Series Imaging System after ECL plus detection.

The dose-response experiments were performed using the same
method described as above. IM-9 cells were stimulated for 30 minutes,
which was determined by time-course experiments, using fusion
proteins as well as hGH with doses of 0, 0.5, 1.5, 5, 15, and 50 nM.
After 30 minutes of stimulation, the cells were lysed and analyzed
using Western blot to detect phosphorylation of JAK2 and STAT5.

31
2.7.2.2. Nb2 cell proliferation

The phosphorylation of JAK2 and STAT5 indicate that fusion proteins
are capable of recognizing, binding and triggering signal transduction
in a manner similar to native hGH. However, fusion protein dependent
cell proliferation is important to demonstrate bioactivity. Nb2 cells
(Sigma) derived from rat T lymphoma cells can respond to exogenous
hGH stimulation by proliferation in dose-dependent manner. Nb2 cells
were cultured as suspension in RPMI 1640 (Mediatech) media
supplemented with 2 mM glutamine, 10% FBS, 10% horse serum (HS)
(Mediatech), 50 units of penicillin/50 µg streptomycin, and 50 µM 2-
mercaptoethanol in a humidified incubator at 37 deg with 5% CO2.

For proliferation assay, Nb2 cells growing at log-phase were pelleted,
washed 3x with excess volume of serum free RPMI 1640, re-
suspended in assay media excluding 10% FBS but including 10% HS,
and counted with Z1 Coulter particle counter (Beckman Coulter). Next,
5000-7000 Nb2 cells were seeded into 96-well plates in 200 µl assay
media and allowed to starve for 24 hrs. This was followed by adding
varying doses of hGH and fusion protein in 10 µl or 50 µl assay media,
respectively.

32
Because the molecular weight of fusion protein (100 kDa) is different
from the hGH (22 kDa), the dose for fusion protein was normalized to
that of hGH. After four-day incubation, 20 µl of Alamar Blue dye
(Biosource) was added to all wells and incubated overnight for Alamar
Blue dye reduction and color change, followed by absorbance
measurement (Phoenix gene) at 570 nm. Absorbance values were
then corrected by subtracting the values from the blanks containing
only the assay media.

33
Chapter 3. Results

3.1. Plasmid construction

Mammalian expression vector pcDNA3.1(+) was used to engineer the
final constructs for the expression of fusion proteins and hGH,
respectively. The final constructs, including hGH-Tf, Tf-hGH and hGH,
were confirmed for accuracy by both restriction enzyme digestion and
DNA sequencing. Data from restriction enzyme and DNA sequencing
analysis showed that both hGH and Tf were fused in frame without
mutations, and connected by a short di-peptide linker (Fig. 6-15).
Upon confirming of successful construction, high quality, endotoxin
free plasmids were prepared at large-scale for transfection.

34

668
bp
668
bp
3500
bp
A B
Fig.6. Cloning hGH with signal sequences to pCR-TOPO vector.
Coding region for hGH was PCR amplified, and fractionated on
agarose gel (A). This PCR amplified fragment was cloned into PCR
blunt II vector, and confirmed by both EcoRV and XhoI restriction
enzyme cleavage that released hGH fragment from vector (B).

35

Fig.7. Tf without signal sequence was cloned into pCR-TOPO vector.
Tf coding region was PCR amplified along with negative control (A).
Schematic map for cloning vector containing Tf insert (B). The
successful cloning of Tf was verified by XbaI enzyme digestion (C)
that released Tf and vector itself. This cloned Tf in the cloning vector
was used as a source for future manipulation, including the creation
of expression construct.
pCR-Blunt II-TOPO-Tf
5571 bp
Kan(R)
Zeo(R)
Tf insert
M13 (-20) forward primer
M13 reverse primer
TOPO binding site
TOPO binding site
pUC origin
XbaI (2048)
XbaI (2105)
XhoI (2)
XhoI (2093)
EcoRI (848)
EcoRI (2060)
EcoRI (5561)
2052
1650
2000
Tf
Tf
Vector
Vector+
Tf

36

pcDNA3.1(+)-Tf-Xho-Xba
7468 bp
Amp(R)
Neo(R)
Tf insert
CMV promoter
pUC origin
EcoRV (965)
XbaI (3032)
XhoI (986)
EcoRI (953)
EcoRI (1832)
Tf,
released
by
Xh+Xba
cut
pcDNA3.1
M 1 2
Fig. 8. Subcloning of Tf to expression vector pcDNA3.1(+). Tf coding
region obtained from cloning vector by XhoI and XbaI digestion was
subcloned into expression vector. Panel A show undigested and
digested expression vector with Tf cloned (lane 1 and 2). Panel B
shows schematic model for subcloned Tf in expression vector.

37

hGH-
Tf
270
0 bp
pcDNA
3.1
pcDNA3.1(+)-hGH-Tf
8107 bp
Amp(R)
Neo(R)
hGH insert
Tf insert
CMV promoter
pUC origin
Eco RV
Xba I
Xho I
Eco RI
Eco RI
M 1 2 3 4
A
Fig. 9. Subcloning hGH upstream to Tf in expression vector. Both
hGH and expression vector were cleaved with EcoRV and XhoI (Panel
A: lanes 1-2 hGH; lanes 3-4 pcDNA(3.1)), and subcloned upstream to
Tf in pcDNA(3.1) vector (Panel B). Finally, successful cloning was
confirmed digesting with EcoRV and XbaI that flanked the whole
coding region including hGH and Tf.
B C

38

Fig. 10. Creating Tf-hGH expression construct. The steps and procedures
needed to establish this construct similar to the creation of hGH-Tf, which
were shown in preceding figures. This final construct was confirmed by
enzymatic digestion using EcoRV + XbaI, which released a 2685 bp fragment
as expected, and DNA sequencing.
Tf-hGH
2685
pcDNA3.1
pcDNA3.1(+)-Tf-hGH
8089 bp
Amp(R)
Neo(R)
Tf insert
hGH
CMV promoter
pUC origin
EcoRV (965)
XbaI (3650)
XhoI (3068)

39

Fig. 11. Expression construct for hGH. This final construct was confirmed by
EcoRV + XhoI digestion, which released a fragment with about 668 bp, as
expected. Lane 1: undigested plasmid as a control; lane 2: same plasmid as
lane 1 with EcoRV and XhoI digestion.
pcDNA3.1(+)
pcDNA3.1(+)-hGH
6067 bp
Amp(R)
Neo(R)
hGH insert
CMV promoter
pUC origin
Eco RI (953)
Eco RV (965)
Xba I (1631)
Xho I (1625)
1 2
hGH
(668
bp)

40

Fig. 12. Tf coding region was analyzed by DNA sequencing. DNA
sequences from Tf coding region were verified with DNA sequencing
in both cloning and expression vector. Shown here was a snapshot
from a sequencing data used to confirm the accuracy of Tf coding
sequences.

41

M 13F _T F : 1 » 1082
T f_seq2_cod267: 940 » 2019
M 13R _T F : 1141 « 2245 (com plem entary)
1 5 0 0 1 0 0 0 1 5 0 0 2 0 0 0
Fig. 13. Assembled DNA-sequence for Tf coding region. Coding
region for Tf includes more than 2000 bp and thus requires three
sequencing primers to cover the whole region, assuming that one
primer generates only about 700-800 bp of high quality sequencing
data with the current technology. This diagram shows the assembly of
Tf sequences from three separate sequencing data, then this
assembled data can be used to check sequence accuracy.

42

1 104 10 20 30 40 50 60 70 80 90 (1)
GGNNNNNNTCTATAGGGCGATTGGGCCCTCTAGATGCATGCTCGAGCGGCCGCCAGTGTGATGGATATCTGCAGAATTCGCCCTTCTCGAGGTCCCTGATAAAA Contig_Tf (1)
-------------------------------------------------------------------------------------CTCGAGGTCCCTGATAAAA HUM_TF_Mature (1)
CTCGAGGTCCCTGATAAAA Consensus (1)
105 208 110 120 130 140 150 160 170 180 190 (105)
CTGTGAGATGGTGTGCAGTGTCGGAGCATGAGGCCACTAAGTGCCAGAGTTTCCGCGACCATATGAAAAGCGTCATTCCATCCGATGGTCCCAGTGTTGCTTGT Contig_Tf (105)
CTGTGAGATGGTGTGCAGTGTCGGAGCATGAGGCCACTAAGTGCCAGAGTTTCCGCGACCATATGAAAAGCGTCATTCCATCCGATGGTCCCAGTGTTGCTTGT HUM_TF_Mature (20)
CTGTGAGATGGTGTGCAGTGTCGGAGCATGAGGCCACTAAGTGCCAGAGTTTCCGCGACCATATGAAAAGCGTCATTCCATCCGATGGTCCCAGTGTTGCTTGT Consensus (105)
209 312 220 230 240 250 260 270 280 290 300 (209)
GTGAAGAAAGCCTCCTACCTTGATTGCATCAGGGCCATTGCGGCAAACGAAGCGGATGCTGTGACACTGGATGCAGGTTTGGTGTATGATGCTTACCTGGCTCC Contig_Tf (209)
GTGAAGAAAGCCTCCTACCTTGATTGCATCAGGGCCATTGCGGCAAACGAAGCGGATGCTGTGACACTGGATGCAGGTTTGGTGTATGATGCTTACTTGGCTCC HUM_TF_Mature (124)
GTGAAGAAAGCCTCCTACCTTGATTGCATCAGGGCCATTGCGGCAAACGAAGCGGATGCTGTGACACTGGATGCAGGTTTGGTGTATGATGCTTAC TGGCTCC Consensus (209)
313 416 320 330 340 350 360 370 380 390 400 (313)
CAATAACCTGAAGCCTGTGGTGGCAGAGTTCTATGGGTCAAAAGAGGATCCACAGACTTTCTATTATGCTGTTGCTGTGGTGAAGAAGGATAGTGGCTTCCAGA Contig_Tf (313)
CAATAACCTGAAGCCTGTGGTGGCAGAGTTCTATGGGTCAAAAGAGGATCCACAGACTTTCTATTATGCTGTTGCTGTGGTGAAGAAGGATAGTGGCTTCCAGA HUM_TF_Mature (228)
CAATAACCTGAAGCCTGTGGTGGCAGAGTTCTATGGGTCAAAAGAGGATCCACAGACTTTCTATTATGCTGTTGCTGTGGTGAAGAAGGATAGTGGCTTCCAGA Consensus (313)
Fig. 14. Sequences from cloned Tf compared with published Tf
sequences using Vector NTI program. Cloned Tf was subjected to DNA
sequence analysis, then assembled sequencing data compared to
original, published Tf sequences to make sure no nucleotide change
had introduced during cloning process.

43

Fig. 15. Enzymatic analysis of expression constructs. Endotoxin free
plasmids were prepared and confirmed by triple enzymatic digestion. Both
digested and undigested plasmids were fractionated on 1% agarose gel and
bands were revealed using ethidium bromide staining, followed by imaging
with Discovery Imaging System from Bio-Rad. Lane 1: hGH (uncut-control);
lane 2: hGH (EcoRV + XhoI); lane 3: GT (uncut-control); lane 4: GT (EcoRV
+ XhoI); lane 5: TG (uncut-control); Lane 6: TG (EcoRV + XbaI). Note: GT
(hGH-Tf); TG (Tf-hGH).
u
n
c
ut
c
ut
u
n
c
ut
u
n
c
ut
c
ut
c
ut
hGH-Tf or
Tf-hGH
(2700 bp)
hGH (688 bp)
MM 1 2 3 4 5 6
Base
pairs

3000

2000

650

44
3.2. Fusion protein expression

The mammalian HEK-293 cells are known to express high quality
recombinant proteins, therefore, are selected as a host for fusion
protein expression. To produce fusion proteins, HEK293 cells were
subjected to transcient transfection (4). The transfected cells were
incubated for 5 days in CD293 media, followed by harvesting the
conditioned media, which was analyzed for fusion protein expression.
The data from SDS-PAGE stained with both silver and coomassie dye
showed a single band with high abundance (90%) corresponding to an
approximately 100 kDa, which is equivalent to hGH (22 kDa) and Tf
(78 kDa) (Fig. 16 and 17). Together with Coomassie stained SDS-
PAGE, the Silver staining of SDS-PAGE demonstrated that fusion
proteins including hGH-Tf and Tf-hGH, as well as hGH expressed at
high levels and high abundance with expected molecular weight, 100
kDa.

45

kDa
100
75

25
Tf (80
kDa)
hGH-Tf or
Tf-hGH
MM 1 2 3 4 5 6 7
8
Fig. 16. SDS-PAGE (7.5 %) showing the expression of fusion proteins
and hGH in HEK293 cells after Silver stain. 50 µl conditioned media
without concentration and 50 ng pure Tf as a control were
fractionated on the gel, followed by silver staining. The gel image was
captured by Discovery Series System from Bio-Rad. MM: molecular
weight marker; lanes 1, 4: hGH (made in lab); lanes 2 and 5: hGH-
Tf; lane 3 and 6: Tf-hGH. Note: each construct was transfected into 2
wells of 6-well plate.

46

Fig. 17. Fusion proteins were expressed in mammalian HEK293 cells
and visualized with Coomassie stain. Samples from conditioned media
were separated on SDS-PAGE and stained with 0.1 % Coomassie blue
to reveal protein bands. The result shows that fusion proteins
expressed at high level and high abundance. Lane 1: hGH; lane 2:
hGH-Tf; lane 3: Tf-hGH; lane 4: Tf- control.
MM 1 2 3 4

47
3.3. Fusion protein detection

To confirm the identity of the expressed protein, suitable antibodies
against the target proteins such as hGH and Tf should be used in a
Western blot assay. The Western blot analysis performed with either
anti-Tf or anti-hGH detected a 100-kDa band confirming fusion protein
is composed of both hGH and Tf (Fig. 18 and 19). Therefore, the data
from Western blot analysis that used both anti-hGH and anti-Tf
antibodies showed the fusion proteins were produced by transient
transfection of HEK293 cells in CD293 media, with correct identity and
molecular weight.

48

MM 1 2 3 4 5 6 7 8
kDa
100
75

25
hGH-Tf or
Tf-hGH

hGH
Fig. 18. Western blot with anti-hGH antibody detected both hGH-Tf
and Tf-hGH fusion proteins. SDS-PAGE (10%) was used to fractionate
2µl (lanes 1-4) and 1µl (lanes 5-8) conditioned media containing the
expressed fusion proteins. After gel run, the proteins were transferred
to PVDF membrane, probed with goat anti-hGH antibody (1:2000),
followed by donkey anti-goat secondary antibody (1:10000). The
signal was detected by ECL Plus, visualized and captured by Discovery
Imaging System. Lanes 1, 5: transferrin (negative control); lanes 2,
6: hGH (produced in lab); lanes 3, 7: hGH-Tf; lanes: 4, 8: Tf-hGH.

49

MM 1 2 3 4 5 6 7 8
kDa
100

75

hGH-Tf
Tf-hGH

apo-Tf
diferric-Tf

Fig. 19. Western blot with anti-Tf antibody detected both hGH-Tf and
Tf-hGH fusion proteins. SDS-PAGE (7.5%) were used to fractionate
10 µl conditioned media containing the expressed fusion proteins, and
50 ng apo-and di-ferric Tf. After gel run, the proteins were
transferred to PVDF membrane, probed with goat anti-Tf antibody
(1:10000), followed by donkey anti-goat secondary antibody
(1:10000). The signal was detected by ECL Plus system, visualized
and captured by Discovery Imaging System. Lanes 1, 4: hGH
(negative control); lanes 2, 5: hGH-Tf; lanes: 3, 6: Tf-hGH; lanes 7-
8: apo- and di-ferric Tf, respectively. Note: two wells of 6-well plate
were transfected by a fusion construct that is why there are two lanes
for each fusion protein.

50
3.4. In vitro biological activity

3.4.1. TfR binding

To evaluate the biological function of the Tf-domain, the fusion protein
was assessed for its ability to bind TfR in CaCo-2 cells. To this end, TfR
binding competition assay was performed in presence of 5 µg/ml
radioactively labeled-Tf (
125
I-Tf). The addition of 10-fold excess fusion
protein (60 µg/ml) resulted in approximately 40% reduction of surface
bound
125
I-Tf (Fig. 20). When added at relatively low dose (6 µg/ml),
the fusion protein still decreased the surface bound
125
I-Tf by 20%
(Fig. 20). The reduction of surface bound Tf as a result of competition
binding suggested that the fusion protein maintained Tf activity.

51

Fig. 20. Fusion proteins hGH-Tf and Tf-hGH had bioactivity for Tf.
Fusion proteins were assessed for Tf bioactivity in TfR competition
binding assay in which fusion proteins were added to CaCo-2 cells in
presence of radioactively labeled iodine-Tf (I
125
-Tf). The decrease of
surface bound radioactive-Tf by the addition of fusion protein
indicates that fusion proteins can bind to TfR on CaCo-2 cells, and
thus displace radioactively labeled Tf. Data represent duplicate values
for each point. IC
50
of hGH-Tf is ~ 1/20
th
of native-Tf.

52
3.4.2. Nb2 cell proliferation and JAK2/STAT5 activation

Nb2 cells require exogenous hGH supplementation for growth and
proliferation when cultured in assay media containing 10% horse
serum but not FBS. Therefore, the Nb2 cells grown in assay media
were treated with different doses of fusion protein to evaluate whether
the fusion protein triggers hGH-specific Nb2 cell proliferation. The data
showed that the fusion protein resulted in a dose-dependent
proliferation of Nb2 cells with the ED
50
value of approximately 1ng/ml
(Fig. 23), which is slightly higher than the ED
50
of native hGH. In
contrast, the untreated cells did not proliferate under the same
condition. The dose-dependent proliferation of Nb2 cells by the fusion
protein indicated that the fusion protein preserved the biological
activity of the hGH-domain.

The in vitro hGH biological activity of hGH-Tf and Tf-hGH fusion protein
was also analyzed using a rapid, sensitive, signal transduction assay in
which JAK2 and STAT5 phosphorylations were used to determine if
hGH domain is active (Fig. 21-22, Table 3.).

53

54

hGH-Tf Tf-hGH
control 0.5 1.5 5 15 0.5 1.5 5 15 nm
Fig. 22. Both hGH-Tf and Tf-hGH led to STAT5 activation in IM-9
cells. STAT5 as a transcription factor becomes activated with hGH
stimulation. Therefore, STAT5 activation via phosphorylation was
used as a measure to assess fusion protein for in vitro hGH
bioactivity. IM-9 cells were treated with varying concentrations of
fusion protein or hGH, and incubated for 10 minutes. Upon
completion of treatment, IM-9 cells were lysed using RIPA buffer with
phosphotase inhibitors, and supernatants analyzed by Western blot
with anti-STAT5 antibody.
hGH-Tf Tf-hGH
control 0.5 1.5 5 15 0.5 1.5 5 15 nm

55

Time (min) hGH-Tf Tf-hGH hGH hGH*
5 5.87 7.86 6.47 0.83
15 13.45 ND 12.91 4.65
30 15.87 10.34 15.76 12.78
60 2.12 -2.02 -3.23 -3.4
ND: not determined. *: commercial hGH
Bands from Western blot w ith anti-JAK2 w ere scanned and analyzed using
Quantity One softw are from Bio-Rad. The density values w ere corrected by
subtracting band density from no treatment control.
Table 3. Time course of JAK2 phosphorylation
Fusion protein hGH controls

56

Fig. 23. Stimulation with hGH-Tf led to Nb2 cell proliferation. To
further assess fusion protein for hGH bioactivity, the Nb2 cells were
treated with different doses of fusion protein, incubated for 4-days,
stained with alamar-blue overnight, followed by absorbance
measurement. Absorbance was corrected by subtracting values from
no treatment groups. Data represent average of triplicate results.
EC
50
: hGH ~ 0.3 ng/ml; hGH-Tf ~ 2 ng/ml

57
Chapter 4. Discussion

4.1. Cloning strategy

Gene fusion constructs were engineered and prepared in mammalian
expression vector pcDNA3.1(+). Subsequently, hGH-Tf and Tf-hGH
fusion proteins were expressed in transfected-HEK293 cells. The
HEK293T cells were used as host for hGH-Tf fusion protein expression,
because they are known to express high quality recombinant proteins
with posttranslational modifications, thereby possibly improving
stability and bioactivity. The resulting fusion proteins maintained the
biological activity of both hGH and Tf.

In this project, we decided to use mammalian expression vector
pcDNA3.1 because it provides powerful promoter from CMV driving the
expression of gene of interest when transfected into well-established
HEK293 cells. Prior to cloning, three restriction enzymes – EcoRV,
XhoI, XbaI – were chosen to use as cloning sites, and the coding
sequences for both Tf and hGH were carefully examined for absence of
those enzymes to avoid potential cleavage and thus fragmentation.
The restriction enzyme recognition sites were incorporated in the 5’-

58
end of the amplification primers, including both forward and reverse
primers while the integrity of the coding sequences preserved. To
boost transcription and subsequent translation efficiency of the target
gene, the Kozak sequences were incorporated in the amplification
primers that amplify coding sequences for N-terminus domain.
Furthermore, the signal sequences from the coding sequences that fold
at the N-terminus of fusion polypeptide were incorporated.

4.2. Tf-based fusion protein with different orientation

Fusion proteins with different domain-orientation are desirable because
it provides flexibility in producing fusion proteins with improved
bioactivity. We constructed and expressed two fusion proteins with
domain switch (Tf is at both amino-and carboxy-terminus), and our
results showed that both fusion proteins are bioactive and expressed
at high levels (Fig. 16-19). It is noteworthy that the Tf-hGH fusion
protein in which Tf-domain is at the amino-terminus demonstrated
high affinity binding to TfR as compared to hGH-Tf fusion protein (Fig.
20).

59
This could be attributable to free-accessibility of key amino acids
involved in receptor binding without disturbance from the nearby
amino acids from the hGH-domain. The X-ray crystal structures from
Tf-TfR complex support our argument by showing the important role of
N-terminus of the Tf in high affinity receptor binding (12, 44).
However, a previous study on the expression of Tf-fusion protein
reported the difficulty in production of fusion protein with Tf-domain at
the N-terminus (37). Similarly, even though the effort to express Tf-G-
CSF fusion protein was successful, it expressed at very low level and
showed poor bioactivity compared to G-CSF-Tf fusion protein (4).

Therefore, the expression of Tf-hGH fusion protein at clinically relevant
level with comparable bioactivity to the hGH-Tf fusion protein
supported the feasibility in creating Tf-fusion proteins with different
orientation that has acceptable bioactivity. It should be noted, in some
cases, that it might require modifications such as inserting linkers that
effectively eliminates unfavorable interactions posed by amino acids
from the nearby C-terminus domain.

4.3. Tf-based fusion protein with high-level of expression

60
The expression data from the two fusion proteins revealed that hGH-Tf
fusion protein had high-level expression than the Tf-hGH fusion
protein. This improved expression in hGH-Tf fusion protein could be
the result of correct folding at higher rate, leading to improved
stability and accumulation. It is also possible that the signal sequences
from the hGH could increase promoter activity by recruiting
transcription factors, contributing favorably to the initiation of
transcription. In deed, the expression of Tf-based fusion with G-CSF
was lower than the Tf-fusion protein with hGH, even though both
fusion proteins were expressed in same host cells grown in serum free
culture media. This discrepancy in expression underlines the impact of
DNA sequences on Tf-based fusion protein expression, possibly via the
enhanced promoter activity, increased transcription, correct folding
and improved stability. Therefore, incorporating signal sequences from
the hGH might help increase expression of other Tf-based fusion
protein with lower expression.

4.4. TfR binding capability

Producing hGH-Tf fusion protein at the accepted level is a first step in
developing therapeutic analog of hGH, with improved

61
pharmacokinetics, efficacy and delivery. Equally, if not more,
important is to characterization of such fusion protein for bioactivity
using cellular and animal models with hGH-deficiency. Therefore, the
hGH-Tf fusion protein must be evaluated for both Tf and hGH activity
to be fully bioactive. The Tf activity of the fusion protein was
determined in TfR binding assay, because it can determine if the fusion
protein has the capacity to bind to TfR. TfR binding of fusion protein is
a crucial step in oral delivery that enables GI absorption via the Tf-TfR
mediated transcytosis.

The fusion protein-dependent displacement of radioactively labeled-Tf
from TfR on CaCo-2 cells indicated that fusion protein recognizes and
binds to TfR (Fig. 20). Conversely, the control with 1mg/ml BSA did
not have the same effect as fusion protein, supporting that the
competition binding is fusion protein specific. This result was in line
with the previous study in which G-CSF-Tf fusion protein maintained
TfR binding capacity (G-CSF-Tf) in a similar study. The IC
50
value for
fusion protein is about 20-fold higher than the endogenous, un-labeled
Tf, suggesting that fusion protein binds to TfR with reduced affinity.
One possible reason for this reduced binding is that the hGH located at
the N-terminus or upstream of Tf could influence optimal receptor

62
binding in negative fashion by introducing steric hindrance. This
presumption is supported by the fact that the fusion protein with Tf at
the N-terminus showed improved TfR binding, with about 20%
increase as compared to hGH-Tf fusion protein.

This result not only underscores the effectiveness of having flexibility
in constructing fusion protein with different orientation for better
bioactivity, it also shows the importance of having free N-terminus for
TfR binding with increased affinity. Yet, TfR binding by the Tf-hGH
fusion protein still lower than the native Tf, suggesting that the hGH-
domain in the fusion protein in general had negative effect on TfR
binding. On the other hand, this low affinity binding to TfR by the
fusion protein could be enough to elicit biological effect, including TfR-
dependent transcytosis. Studies showed that high affinity binding to
receptor by the ligand is not always necessary for full biological
activity (39, 55).

4.5. hGH bioactivity

After characterizing fusion protein for Tf bioactivity, we sought to
determine whether the fusion protein possess the hGH bioactivity

63
using two different approaches: 1) hGH-dependent activation of both
JAK2 kinase and STAT5 transcription factor, and 2) hGH-dependent
proliferation of rat Nb2 cells. The growth promoting activity of hGH is
achieved by binding to its cell surface receptors on the plasma
membrane of target cells, leading to receptor dimerization, JAK2 and
STAT5 phosphorylation, among other events. Therefore, we used
phosphorylation of JAK2 and STAT5 as a measure to assess fusion
protein for hGH activity. As shown in the Western blot analysis, both
fusion proteins were effective in triggering phosphorylation, indicating
that fusion proteins had the ability to bind to hGH-R and trigger
activation of JAK2 and STAT5. Even with low dose (0.5 ng/ml), fusion
proteins led to hGH-specific phosphorylation in a short period of time
(as little as 5 minutes).

The activation of hGH-dependent signal transduction is a good way to
rapidly determine if the hGH-Tf fusion protein retains the ability to
bind hGH-R and activates down-stream signaling molecules, including
JAK2 and STAT5. However, it would be helpful to test fusion protein for
end-point bioactivity, such as cell proliferation, because it results from
a series of events that occur in proper and precise order.

64
Accordingly, the hGH-specific in vitro activity of the fusion protein was
tested in Nb2 cell proliferation, a widely used assay determining the in
vitro activity of the recombinant hGH (14). The results from the Nb2
cell proliferation assay supported the findings from JAK2 and STAT5
phosphorylation assay in IM-9 cells. The dose of fusion protein
required to stimulate the activation of signaling molecules was similar
to the dose needed for effective proliferation of Nb2 cells. Thus these
two assays were consistent and correctly assessed the fusion protein
for hGH bioactivity.

It is worth to mention that Nb2 cells express prolactin (a hormone
closely related to hGH) receptor, but not hGH receptor (14). However,
hGH can bind both prolactin receptor and hGH-receptor, and hGH
binding to prolactin receptor leads to hGH-dependent proliferation in
Nb2 cells. One prerequisite for this assay to be valid is that the fusion
protein or hGH should be free from prolactin contamination, because it
induces Nb2 cell proliferation as well. HEK293 cells that we used as
host for fusion protein production neither express nor secrete prolactin
into the conditioned media, thus the Nb2 cell proliferation assay
accurately reflects in vitro cell proliferation activity of hGH-Tf fusion
protein.

65
Part 2. Insertion of the designed helical linker led to
increased expression of Tf- based fusion proteins

Chapter 5. Introduction

During the past 25 years, more than 165 biotechnological drugs have
been approved and currently there are several hundred of candidates
are in varies stages of clinical trials (29, 56, 69). High quantities of
recombinant proteins ranging from hundreds of milligram to grams
must be produced in order to carry out preclinical evaluations and
clinical trials (6, 68, 69).

Unfortunately, the potential therapeutic proteins with poor expression
yield face obstacle to make it through the clinical trials to the final
approval by the FDA. Mammalian cells are preferred choice for
producing therapeutic proteins with correct posttranslational
modifications and reduced immunogenicity (69). Protein drugs such as
hormones and growth factors express, intrinsically, at low levels that
not only increases the manufacturing cost but also delays the product
evaluation. Thus strategies that increase expression of target proteins
in mammalian cells are highly desirable and beneficial.

66

Typically, the problem of low solubility and expression was improved
by incorporating carbohydrate binding domain (CBD) and maltose-
binding protein (MBP) as fusion partners to the target protein (27). But
these fusion partners should be removed during or after purification by
introducing peptide linkers with cleavage sites for endopeptidases such
as thrombin and factor Xa (42). Conceivably, this approach is not
feasible for large-scale production of target proteins because it
requires numerous steps of column purification and enzymatic
processing, limiting the production capacity and possibly causing non-
specific cleavage.

The selection of peptide linker with the ability to maintain functions of
domains in the fusion protein is becoming important (1, 2, 34, 37).
Recently, we designed a helical linker with 50 amino acids using
EAAAK helix-forming motif based on a previous study (1), and inserted
between G-CSF and Tf moieties, leading to increased biological activity
(5). Most recently, we found that the insertion of the same helical
linker in Tf-fusion proteins resulted in a high-level expression as
compared to the same fusion proteins without the helical linker in
HEK-293 cells.

67

Here we report this “helical linker-dependent increase of expression” in
2 Tf-based fusion proteins and provide evidence that confirms same
effect of high-level expression in both cases regardless of the level of
original expression. This approach can easily be introduced and applied
to other fusion proteins from limited to no expression, greatly
improving the production yield for downstream application.

68
Chapter 6. Materials and Methods

6.1. Helical linker insertion

The helical linker, LEA(EAAAK)
4
ALEA(EAAAK)
4
ALE, was prepared and
inserted according to the previous study (5) with some modifications.
The G-CSF-(H4)
2
-Tf plasmid construct was used as a source for helical
linker, which was released from the construct by digesting with XhoI
restriction enzyme. The release of the helical linker was confirmed by
agarose gel analysis, and the helical linker was gel purified using gel
extraction kit (Qiagen). The helical linker with XhoI sticky-ends was
then ligated to both hGH-Tf and Tf-hGH constructs that were digested
with XhoI and treated with calf alkaline phosphates, respectively.

Small aliquots from ligation reaction were transformed into chemically
competent TOP10 cells. Plasmids prepared from overnight cultures
were digested with XhoI restriction enzyme, and analyzed on agarose
gel to identify positive colonies. Selected plasmids were DNA
sequenced to determine the orientation and number of copies. Finally,
glycerol stocks were prepared for the positive colony with two copies
of helical linker.

69
6.2. Cleavable linker insertion

Our lab designed cleavable linker as PCR primers, which were then
annealed to form double stranded sequences coding for cleavable,
cyclic linker. The cleavable linker, designed to have two- thrombin
recognition sites and two cytokine residues that form one disulfide
bond, was analyzed on agarose gel to estimate quantity. Suitable
amount of cleavable linker that has XhoI sticky ends was ligated to
hGH-Tf and Tf-hGH constructs with same compatible ends,
respectively. Plasmid maniples were prepared from overnight cultures
after transformation with ligation reaction, and analyzed with XhoI
digestion. Selected plasmids were DNA sequenced to identify positive
colony with correct sequence and orientation.

6.3. Western blots

The fusion proteins were fractionated on a 4-20% pre-cast SDS-PAGE
(Thermo Scientific) and subsequently transferred to PVDF membrane
(Millipore) for 90 minutes at 200 mA. The membrane was blocked with
5% dry milk for one hour, and incubated with goat monoclonal anti-
hGH antibody at 0.1 µg/ml for one hour (R&D Systems). After washing

70
3x with PBST for 5 minutes, the membrane was incubated for one hour
in rabbit anti-goat secondary antibody (1:1000 dilution) conjugated to
HRP, followed by washing 3x with PBST for 10 minutes. The membrane
was incubated in premixed ECL reagents (GE Healthcare) for 5
minutes, and signal was scanned and recorded using Chasidic XBR
(Bio-Rad).

Western blot for Tf was performed with same way as above. Briefly,
the membrane was probed with goat anti-human Tf (Sigma) at
1:10000 dilutions for one hour. The signal was detected by using same
secondary antibody as above at 1:1000 dilution, and ECL reagents.
Signal was recorded and further analyzed using Chasidic XBR.

71
Chapter 7. Results

7.1 Establishing gene fusion constructs with helical and
cleavable linker

To investigate whether the insertion of helical linker between the
protein domains in the Tf-based fusion protein improves the
expression, we constructed two pairs of gene fusion plasmids with or
without the inserted helical linker, and confirmed that the insertion,
orientation and number of copies were correct. Additionally, the
cleavable, cyclic linker was introduced successfully with expected copy
number and orientation. Subsequently, these plasmid constructs were
transfected to HEK293 or HEK293T cells to produce fusion proteins
(Fig. 24-27).

72

H4 linker
(~ 50 bp)
H4 linker
(~ 50 bp)
100
bp
Fig. 24. Isolation of helical linker. Helical linker, H4, was released
from parent plasmid, G-CSF-(H4)
2
-Tf, by digesting with Xho I
enzyme. The released H4 linker was gel-purified and quantified. Panel
A, Lane 1 and 2: maker and released H4 linker; panel B, lane 1 and
2: marker and gel extracted helical linker. Enzymatic digestion was
fractionated on agarose gel.
1 2 1 2
2
A B

73

Fig. 25. Insertion of helical linker between hGH-Tf. Two
copies of helical linker were introduced between
domains of hGH-Tf construct by Xho I site. Plasmid
minipreps from overnight cultures were prepared and
digested with Xho I to identify positive colonies
containing the helical linker. Lanes 3, 5, 8, 11 and 12
are positive colonies with the helical linker.
1 2 3 4 5
6 7 8 9 10 11 12
H4 linker (~
50 bp)

74

Fig. 26. Insertion of helical linker between Tf-hGH. 2 copies of helical
linker were Introduced between domains of Tf-hGH construct. Plasmid
minipreps were prepared and digested with Xho I to identify positive
colonies containing the helical linker. The black lines show colonies
with the inserted helical linker.
H4 linker (~
50 bp)
H4 linker

75

Fig. 27. Introducing cleavable, cyclic linker. The cyclic linker was
introduced between two domains in hGH-Tf (left panel) and Tf-
hGH (right panel) constructs by Xho I site. Plasmid minipreps
were prepared and digested with Xho I to identify positive colonies
containing the cyclic linker. The lines show colonies with inserted
cyclic linker.
Cyclic
linker
Cyclic
linker

76
7.2. Comparing hGH-Tf and hGH-(H
4
)
2
-Tf fusion proteins for
expression

The hGH-Tf fusion protein is consisted of human growth hormone and
Tf and linked by a short di-peptide linker; whereas the hGH-(H
4
)
2
-Tf
fusion protein is linked by a helical linker with ~ 50 amino acids (Fig.
28). To assess the level of expression, the fusion proteins with or
without the helical linker were analyzed by Western blot as well as
SDS-PAGE with Coomassie stain.

Both anti-hGH and anti-Tf Western blots detected a band
corresponding to approximately 100 kDa, which is the sum of
molecular weights from Tf (79 kDa) and hGH (22 kDa), and confirmed
that the fusion protein was composed of two moieties including Tf and
hGH. To evaluate and compare the level of expression, band-densities
for each fusion protein were quantified and analyzed. The density data
revealed that the fusion protein with helical linker expressed 1.7 fold
more than the original fusion protein without the helical linker (Fig. 29-
31, and Table. 4).

77

Fig. 28. Computer generated model for hGH-Tf fusion protein
with helical linker. The structure was generated using WebLab
Viewer and Insight II molecular modeling programs. The
coordinates used in this model are HWG for hGH and 1JNF for
rat-Tf, both were obtained from protein data bank (PDB).

78

x: Linker inserted between Tf-and hGH/G-CSF domains.
H4: Helical linker
(H4)
2
: 2 copies of helical linker
ND: Not determined

X Tf-x-G-CSF G-CSF-x-Tf Tf-x-hGH hGH-x-Tf
No linker 1 1 1 1
H4 7.8 ND ND ND
(H4)
2
11.2 1.44 2.39 1.66
Table 4. The ratio of expression with or without helical linker
G-CSF fusion protein hGH fusion protein

79

Fig. 29. High-level expression by the insertion of helical linker in Tf-
fusion proteins was analyzed by Anti-hGH Western blot. Four Fusion
proteins with or without the inserted helical linker, expressed in
serum free media, were analyzed by Western blot using goat anti-
hGH monoclonal antibody (1:1000). Signal was detected using HRP-
conjugated rabbit anti-goat secondary antibody (1:1000) and ECL
reagents. The image was recorded and analyzed by ChemiDoc XBR.
Lane-1: Tf (negative control); lane-2: hGH (10 ng); lane-3: hGH-Tf;
lane-4: hGH-(H4)
2
-Tf; lane-5: Tf-hGH; lane-6: Tf-(H4)
2
-hGH.
1 2 3 4 5 6

80

Fig. 30. High-level expression by the insertion of helical linker in Tf-
fusion proteins was analyzed by Anti-Tf Western blot. Four Fusion
proteins with or without the inserted helical linker, expressed in
serum free media, were analyzed by Western blot using goat anti-Tf
antibody (1:5000). The signal was detected by rabbit anti-goat
secondary antibody conjugated to HRP (1:1000) and ECL reagents.
The image was recorded and analyzed by ChemiDoc XBR. Lane-1: Tf
(50 ng); lane-2: hGH (as negative control); lane-3: hGH-Tf; lane-4:
hGH-(H4)
2
-Tf; lane-5: Tf-hGH; lane-6: Tf-(H4)
2
-Tf.
1 2 3 4 5 6
80
kDa
100

81
In addition, fusion proteins were analyzed by SDS-PAGE stained with
Coomassie blue to confirm results from band-density analysis from
Western blots, and to evaluate relative purity and abundance. The
results from SDS-PAGE demonstrated that the fusion proteins were
expressed with high purity (~ 90%) and high abundance (~ 95%) at a
molecular weight of 100 kDa (Fig.30). Furthermore, band-density
analysis from Coomassie blue-stained SDS-PAGE showed that the
expression of fusion protein with the helical linker was about 1.7 fold
higher than the original fusion protein without the helical linker,
confirming results from Western blot analysis (Fig.30, table 4).

82

75
100
00
kD
a
MM 1 2 3 4 5 6 7 8 9
Fig. 31. SDS-PAGE revealing high-level expression resulting from
helical linker insertion. The insertion of helical linker led to an
increased expression of both hGH-Tf and Tf-hGH recombinant fusion
proteins in mammalian HEK-293 cells. Same volume of conditioned
media (15 µl) from transfected HEK-293 cells were fractionated using
SDS-PAGE, stained with Coomassie blue, destained, and analyzed
with Discovery Imaging System. Lane 1: marker; lanes 2, 3, and 4:
standards with 1, 2, and 4 µg Tf, respectively; lane 4: hGH-Tf fusion
protein without linker; lane 5: hGH-Tf fusion protein with helical
linker; lane 6: hGH-Tf fusion protein with cleavable linker; lane 7: Tf-
hGH fusion protein; lane 8: Tf-hGH fusion protein with helical linker;
lane 9: Tf-hGH fusion protein with cleavable linker.

83
7.3. Comparing Tf-hGH and Tf-(H
4
)
2
-hGH fusion proteins for
expression

To investigate the effect of helical linker on the expression of fusion
proteins with different orientation, the Tf-and hGH-domains were
switched and the helical linker was introduced resulting in two new
fusion proteins: Tf-hGH and Tf-(H4)2-Tf. The two fusion proteins were
expressed and the yield of expression was compared. The Western blot
data with fusion protein specific antibodies including anti-hGH and
anti-Tf confirmed the identity and molecular weight (100 kDa) of the
fusion protein, as well as the insertion of helical linker as shown by
slightly increased molecular weight (Fig. 29-31).

The band-density analysis from Western blot data showed fusion
protein with the inserted helical linker expressed at higher level with ~
2.4 fold increase as compared to original fusion protein without the
helical linker (Fig.29-31, Table 4). The SDS-PAGE with Coomassie
stain revealed that both fusion proteins were expressed with high
purity (~ 90%) and high abundance (90%) but the fusion protein with
the helical linker expressed at high levels with ~ 2.4 fold increase than

84
the original fusion protein without the helical linker, further confirming
the Western blot data (Fig. 29-31, Table 4)

7.4. In vitro biological activity

7.4.1. TfR binding

To evaluate the biological function of the Tf-domain, the fusion protein
was assessed for its ability to bind to TfR in CaCo-2 cells. To this end,
TfR binding competition assay was performed in presence of 5 µg/ml
radioactively labeled-Tf (
125
I-Tf). The addition of 10-fold excess fusion
protein (60 µg/ml) resulted in approximately 40% reduction of surface
bound
125
I-Tf (Fig. 32). When added at relatively low dose (6 µg/ml),
the fusion protein still decreased the surface bound
125
I-Tf by 20%
(Fig. 32). The reduction of surface bound Tf as a result of competition
binding suggested that the fusion protein maintained Tf activity.

85

Fig. 32. Fusion proteins containing the helical linker were tested for
their ability to bind to TfR in monolayer grown CaCo-2 cells. Fully
differentiated CaCo-2 cells were treated with varying doses of fusion
proteins in presence of radioactively labeled Tf (I
125
-Tf), and the
decrease of I
125
-Tf as a result of fusion protein binding to TfR
indicated fusion protein has bioactive Tf-domain. Data are
represented as mean (± SEM).

86
7.4.2. hGH-(H4)
2
-Tf fusion protein led to Nb2 cell proliferation

Nb2 cells require exogenous hGH supplementation for growth and
proliferation when cultured in assay media containing 10% horse
serum but not FBS. Therefore, the Nb2 cells grown in assay media
were treated with different doses of fusion protein to evaluate whether
the fusion protein triggers hGH-specific Nb2 cell proliferation. The data
showed that the fusion protein resulted in a dose-dependent
proliferation of Nb2 cells with the ED
50
value of approximately 1ng/ml
(Fig. 33), which is slightly higher than the ED
50
of native hGH. In
contrast, the untreated cells did not proliferate under the same
condition. The dose-dependent proliferation of Nb2 cells by the fusion
protein indicated that the fusion protein preserved the biological
activity of the hGH-domain.

87

Fig. 33. Fusion proteins containing the helical linker led to Nb2 cell
proliferation. The rat T lymphoma cells, Nb2, were treated by two
fusion proteins with varying doses, and allowed to proliferate for 4-
days, which was followed by Alamar Blue dye staining and absorbance
reading at 570. Nb2 cells proliferated in dose-dependent manner by
fusion protein stimulation, as compared to controls. The concentration
of fusion protein was normalized to molar equivalent of hGH. Data are
represented as mean (± SEM).

88
Chapter 8. Discussion

Our results demonstrated that the helical linker introduced between
two protein domains in Tf-based fusion proteins enhanced the yield of
expression. And this improved expression was more prominent in
fusion proteins where the Tf moiety located at the N-terminus and
whose initial expression was low. Therapeutic proteins with poor
expression in mammalian cells could benefit from this finding.

In general, high level expression can be achieved: 1) by designing
expression vectors with strong promoters and epifocal origins; 2) by
transfecting host cells such as HEK293T and HEK293EBNA1 with
episomal antigen expression including SV40 large-T antigen and EBV
nuclear antigen, respectively; 3) by selecting transfection reagents
with highest gene transfer efficiency; 4) by developing or using culture
media with enhanced cell proliferation and survival but with minimum
cell death and apoptosis (6, 68, 69). Addition of sodium butyrate and
peptone into the culture media containing the transfected cells had
positive effect on the expression of target protein. However, long-term
consequences of using such components in culture media remain
controversial and unknown (17, 41).

89
The insertion of helical linker yielded more fusion protein over the
same fusion protein without the helical linker (Table. 4). The
expression of Tf-G-CSF fusion protein after insertion of the helical
linker was markedly improved (Hsin-Fang Lee, unpublished data),
especially considering that the same fusion protein without the linker
did not express in a previous attempt (Table 4). The success of
constructing biologically active recombinant G-CSF-Tf fusion protein
with the helical linker led us to pursue the feasibility of producing other
Tf fusion proteins. To achieve that goal, we constructed Tf fusion
protein with hGH and introduced both short linker with two amino
acids and long helical linker as spacers between two domains, and
found fusion protein with the helical linker expressed at high levels.

Next, we explored the effect of helical linker insertion on the
expression of fusion protein with Tf-domain switch (Tf-H4-hGH) that
created fusion protein with different orientation, and again discovered
that the expression of Tf-H4-hGH fusion protein was significantly
higher than the same fusion protein without the helical linker (Fig. 29-
31 and Table 4). This new fusion protein expressed with much
improved yield over the same fusion protein that lacked the helical
linker (Table 4).

90
The possible mechanisms responsible for this increased expression of
Tf-based fusion protein by the insertion of helical linker were not
known. However, the rigid and extended nature of the helical linker
could help improve mRNA stability and increase the rate of proper
folding and thus cause high-level expression and accumulation of
fusion proteins in the conditioned media (45, 67). The ability of the
helical linker to maintain two domains at certain distance could
eliminate unfavorable interactions that in turn would result in correct
folding and higher in vitro stability, thereby increasing the yield of
expression (34).

Our assumptions are indirectly supported by three lines of evidence:
1) insertion of flexible linker (GSSSS)
3
between G-CSF-and Tf-domains
produced G-CSF-(GSSSS)
3
-Tf fusion protein with poor expression (5),
2) another paper on the expression of Tf-fusion protein with Tf at the
N-terminus reported that the insertion of IgG hinge region as the
spacer was ineffective (37), 3) when coding sequence for the helical
linker reversed, the fusion protein failed to express (unpublished
data), suggesting the importance of rigid linker with helical structure
on increased expression. Remarkably, the Tf-G-CSF fusion protein

91
regained its ability to express after insertion of the helical linker
(unpublished data).

In addition, our linker with its secondary structure could resist
enzymatic cleavage by not exposing the target amino acids with
protease recognition, thereby increasing overall stability of the fusion
protein (35). Furthermore, it is plausible to assume that a large
molecule such as Tf-G-CSF fusion protein with 100 kDa may require
large conformational space to fold correctly driving the equilibrium
towards the increased expression and accumulation. Conceivably, our
linker with helical structure is likely to play that role of holding the
domains at a distance providing larger space and separation. Further
studies that evaluate the stability of the helical linker against common
proteolytic enzymes including trypsin, chymotrypsin and thrombin
would aid partially in our understanding of mechanism responsible for
the increased expression.

92
Part 3. Producing fusion proteins in large-scale

Chapter 9. Introduction

The capability of manufacturing protein therapeutics in large-scale is
one of the major steps that often play big role in determining whether
to pursue potential protein drugs into clinical trials (68, 69). Because
sufficient quantities of recombinant proteins are needed to conduct
various cellular and animal experiments that help characterize for in
vitro and in vivo efficacies. Several hosts are commonly employed to
produce such protein products for downstream testing and evaluation,
and these hosts include bacteria, yeast, insects, transgenic plants and
animals, and mammalian cells, each with certain advantageous and
disadvantageous (40, 48, 50). This chapter provides a brief description
of benefits and limitations of several host species for large-scale
production of therapeutic proteins, and importantly, illustrates the
development of a new technique that uses adherent mammalian cells
to produce large-scale recombinant fusion protein within a reasonable
time frame.

93
Therapeutics proteins that are intended to treat human diseases can
be divided into two categories depending on the status of
posttranslational modification – glycosylated and non-glycosylated (57,
61). Glycosylation is one of the major posttranslational modifications
that have significant impact on efficacy, immunogenicity, stability,
solubility, and in vivo plasma half-life of a given protein drug (58, 60).
Glycosylation refers to attachment of sugar molecules such as glycan
to asparagines (N-linked), or serine and heroine (O-linked). Therefore,
choosing a host for the production of protein therapeutics comes down
to its posttranslational modification status (57). If a protein drug
requires glycosylations for its biological activity, then it is best to be
expressed in mammalian expression system, with some exceptions.

Bacteria-based expression system is suitable for the production of
protein drugs that does not necessitate posttranslational modifications,
and it is simple, convenient, cost-effective, scalable and rapid as host
for recombinant protein expression. However, it lacks complex
enzymes and machineries required to produce fully functional
therapeutic proteins with all the proper posttranslational modifications,
including those are not immediately needed for the function of the
protein but could be important for long-term efficacy and safety (60).

94
The yeast expression system is preferred over bacterial system, owing
to its ability to produce recombinant proteins with posttranslational
modifications. However, such modifications provided by the yeast host
differ from mammalian host, especially, with regards to the pattern
and composition of glycosylation (57). And this variation in
glycosylation compared to mammalian host cells is still a concern over
its long-term effect on the quality of the produced protein. Efforts are
underway to humanize the key enzymes and components responsible
for glycosylation, which is similar to the one produced in human cells.
Still the culture condition is drastically different from that of human
cells, and this different culture condition may influences the quality of
recombinant protein produced in the yeast host. Therefore, maybe it is
sometime away to be established and accepted as reliable alternative
to human cell culture for recombinant protein production.

Insect cells, transgenic plants and transgenic animals have been used
as hosts for therapeutic protein production (48). However, as stated in
the preceding paragraphs, each has its limitation and faces significant
challenges as to produce recombinant proteins with quality matching
to the ones produced by human cells. Increased research activities are
underway to address complex issues surrounding these host systems

95
towards establishing a widely accepted systems for manufacturing high
quality of human therapeutic proteins.

On the other hand, the mammalian cells, especially the human cells
are most preferred platform for production of uncompromised, high
quality therapeutic proteins that are intended to treat indications
affecting humans. The deficiency or absence of functional protein in
human body is usually the cause for particular indications such as
diabetes and dwarfism as a result of insufficient growth hormone (21,
51, 52). These deficient proteins can only be produced with high
confidence by human cells that are identical to original cells making
the functional protein. The notable advantage of using human cells as
host is that they produce high quality proteins that are complete with
all posttranslational modifications. Such modifications are essential for
long-term safety and efficacy as they eliminate or reduce unexpected
immunogenicity.

Chinese hamster ovary (CHO) and human embryonic 293 (HEK293)
cells are common hosts for recombinant protein expression because
they are well characterized and have good track of record in producing
such products. Currently, about 60% of FDA approved

96
biopharmaceutical agents are manufactured using human cells (38).
And this number could rise in near future as more cost-effective,
innovative, scalable approaches are developed and accepted. Stable
transfected CHO and HEK293 cells are cultured as suspensions for
large-scale production of human therapeutics. However, there is no
report regarding the use of HEK293 cells that are grown as adherent-
culture and transiently transfected for large-scale production of
recombinant proteins. Here we describe a procedure that used
adherent-culture grown-HEK293T, a variant of HEK293 cells
transformed with SV40 large T antigen, to produce fusion proteins in
large-scale, up to 50 mg/L. In addition, described here was the
generation of stable 293T cells that express Tf-fusion proteins for long
period of time in serum free media.

97
Chapter 10. Methods

10.1. Large-scale plasmid preparation

To perform large-scale transfection, it is necessary to prepare high
quality plasmids with fusion protein expression. The cryo-frozen E.coli
stocks, which were prepared by aseptically adding 20% glycerol into
the culture, were thawed on ice and small aliquots were added into 3
ml sterile LB media. After overnight culture at 37 °C with 270 rpm
constant shake, 500 µl of this culture were then inoculated into 500 ml
sterile LB media and allowed to culture for overnight.

Bacteria cells in 50 ml canonical tubes were harvested by centrifuge
for 15 minutes at 4 °C, and excess media were drained by placing
tubes on paper towels in inverted position for about 30 minutes.
Plasmid was isolated using endotoxin-free Plasmid Mega Kit (Qiagen)
and following instructions provided by the Vendor. Plasmid DNA was
quantified using UV absorbance at 260 and agarose gel analysis. The
typical yield was about 2-3 mg/500 ml culture, with excellent quality.
Having high quality, endotoxin-free plasmid is particularly important as

98
this could influence the quality of the protein expressed after
transfection. Plasmids were stored in TE buffer at – 20 °C.

10.2. Large-scale transient transfection

HEK293T cells were maintained as described in previous chapter,
frozen vials were prepared from healthy, well-maintain cells with 10%
DMSO in 10 % FBS supplemented DMEM media. For each transfection,
fresh cells revived from frozen stocks were cultured in T25 flask for 2-
3 passages in order to allow cells for complete recovery. To prepare for
scale-up transfection, cells cultured in T25 flask were passed into T75
flasks and cultured near confluence, which were passed into large
T225 flasks. One day before transfection, cells were seeded in T225
flask and allowed to grow near confluence.

Large-scale transcient transfection was performed on approximately
100 million cells in one T225 flask with 50 mL serum free DMEM media
using 100 µg expression construct and 400 µl linear polyethylene-
imine (PEI) (Polymer Science) diluted in 0.15 M sodium chloride. For
scale up production, fusion protein was produced by transient
transfection of HEK293T cells in 40 flasks of T225 at once.

99
After five-hour incubation, the serum free DMEM was replaced with 2L
of CD293 media supplemented with GlutaMax (Invitrogen). Cells were
allowed to express fusion protein for three days, after that, the
conditioned media were collected and cells were fed with fresh CD293
supplemented with GlutaMax, and cultured for additional two days,
followed by harvesting of conditioned media. Cells once again were fed
with fresh CD293 supplemented with GlutaMax for another 2-3 days
depending on the condition of cells, and conditioned media were
harvested for final time.

Total of 6L of conditioned media from single transfection were
obtained, concentrated using TFF Labscale system (Millipore), and
buffer exchanged. The quantity of the fusion protein was assessed by
SDS-PAGE stained with Coomassie blue and NanoDrop
spectrophotometer (NanoDrop Technologies, Inc.). In the latter
approach, values from UV absorbance at 280 of fusion protein were
corrected dividing by 6, due to overestimation for unknown reasons.
These combined methods appeared very useful and were relatively
accurate in estimating final concentration after enrichment of total
fusion protein.

100
10.3. Formulation strategy

Fusion proteins, like other recombinant proteins, require special
formulation for short and long-term storage and for optimal biological
activity (19, 58, 60). They are susceptible to degradation resulted
from physical and chemical stresses. Physical stress including agitation
caused during handling, change in temperature, and absorption to
surface of the container could lead to loss of fusion protein activity as
a result of protein aggregation or unfolding. Chemical stress of
oxidation, reduction, and deamination, among other modifications, are
common factors that induce structural and conformational change in
fusion protein. Such changes are great concern with respect to folding,
stability, and biological activity.

Recombinant proteins, including the Tf-fusion protein, are typically
formulated either as lyophilized power (solid form), or aqueous
solution (liquid form). The purpose of both formulation strategies is to
maintain the native structure and activity of the protein of interest in
stable form so it can be stored and used without loss of activity. For
proteins formulated as lyophilized powder, selecting suitable stabilizers
that protect protein during freeze-drying process becomes important,

101
in addition to other excipients. On the other hand, for proteins with
liquid formulations, finding stabilizers that prevent damages caused by
freeze-and-thaw cycle is essential (59, 60).

A good formulation should provide recombinant proteins with optimal
pH, ionic strengths, and protection from both freeze-thaw and freeze-
dry cycles. Extensive literature search showed that mannitol-
phosphate was used as buffer for liquid formulation of human growth
hormone. Accordingly, this buffer was adopted as liquid formulation for
the hGH-Tf fusion protein. This formulation is composed of mannitol as
a stabilizer and phosphate as buffering agent keeping pH at desirable
range (58). Liquid formulation for Tf-fusion protein contains 88 mM
mannitol in 5 mM phosphate buffer, and was prepared with the
following ingredients:

Preparation of phosphate buffer (10x): 50 mM

Prepare 50 mM dibasic potassium phosphate in 200 ml ddH
2
O
By dissolving 1.8 g dibasic potassium phosphate in 200 ml ddH
2
O
(0.05 M x 174.2 g/mol x 0.2 L = 1.8 g)

102
Prepare 50 mM monobasic potassium phosphate in 200 ml ddH
2
O
By dissolving 1.36 g monobasic potassium phosphate in 200 ml
ddH
2
O (0.05 M x 136 g/mol x 0.2 L = 1.36 g)

Mix 50 mL of monobasic potassium phosphate with 200 mL dibasic
potassium phosphate.

4) Prepare 880 mM mannitol (10x) in 10x phosphate buffer
By dissolving 40.1 g of mannitol in 250 mL 10x phosphate buffer
(0.88 M x 0.25L x 182.17 g/mol = 40.1 g)

5) 10x buffer was diluted at 1:10 ratio with clean water and filter
sterilized with Stericup (Millipore) and kept at 4 °C. When stored at 4
°C, the 10x solution forms crystals in the bottom of bottle, which can
be dissolved into solution by slightly warming and mixing before
making 1x buffer. Finally, the pH of buffer was measured to be 7.3-
7.4, thus no pH adjustment was needed.

103
10.4. Development of Radioimmunoassay with I
125
-hGH

Radioimmunoassay is extremely sensitive method to detect trace
amounts of antigen in a given sample. Because it is sensitive but less
expensive compared to commercially available ELISA, we decided to
use it to select stable colonies with fusion protein expression for
further development of stable cell lines.

In this method, 10 µg hGH was labeled with I
125
, purified using G-50
column, and eluted with 0.1 % gelatin in PBS. 0.1 ml of Radioactively
labeled hGH with 20,000-30,000 counts per minute (cpm) was added
into test tubes that contain positive control, negative control and
conditioned media from 96-well plates in 0.1 ml buffer, followed by
mouse anti-hGH antibody (1:1000). After mix, brief centrifugation,
and incubation for 1 hr at 37 °C water bath, the samples were added
with normal mouse serum (1:100), rabbit anti-mouse secondary
antibody (1:8), followed by overnight incubation at 4 °C. Next, the
samples were centrifuged to precipitate the complex (I
125
-hGH +
primary antibody + secondary antibody), the supernatants with
unbound I
125
-hGH were decanted, and then the pellets were counted

104
using gamma counter for radioactivity. Low radioactivity means
increased expression of fusion protein by a specific colony that
competes out I
125
-hGH. The colonies with high fusion protein
expression were further evaluated by anti-hGH Western blot and
selected colonies were developed into stable cell lines.

105
10.5. Stable transfection

Several variants of HEK293 cells (HEK293, HEK293T, and HEK293F)
were explored for expression of fusion proteins, among them, the
HEK293F cells grow as suspension in an incubator with shaker. And
expression comparison between these three different cell lines showed
that 293T cells expressed fusion proteins at high levels; therefore,
HEK293T cells were selected as a host for stable transfection and
subsequent stable cell line development. Freshly revived cells were
allowed to grow for 3 passages, and cells were closely observed to
make sure they were in good condition. One day before transfection,
cells were plated in 10-cm culture dishes at ~ 25% confluence.

Stable transfections were performed using 12 µg plasmid construct
and 33 µL LipoFactimine2000 (LF2000) in 10-cm culture dish. Both
plasmid and LF2000 were diluted in 1.5 mL Opti-mem-1 media. Total
of 5 dishes were transfected by control without plasmid construct,
hGH, GT, GHT, and THG constructs. Cells were incubated for 24 hrs to
allow plasmid DNA to integrate to the host chromosome, and then
each dish containing transfected cells was diluted at 1:10, 1:50, and
1:100 ratio and cultured in selection media supplemented with 1mg/ml

106
G418 (Invitrogen) to select single colony. The dishes with 1:100
dilutions were chosen among other dilutions and continuously cultured
with selection media to allow colonies to develop.

The cells grown as colonies were trypsinized, counted and diluted to
adjust cell density to single cell/100 µl media with 1 mg/ml selection
agent. This was plated into 96-well plates, with one 96-well plate per
construct, and cultured until single colony was developed. Once visible,
single colonies were marked from bottom of 96-well plate with Sharpie
and cultured for several more days for fusion protein expression. The
marked colonies were screened for fusion protein expression using
I
125
-hGH radioimmunoassay, which we established in house.

The fusion protein expression by individual colonies was confirmed
with Western blot analysis using anti-hGH antibody. Selected colonies
from each construct with high-level expression were expanded to 24-
well plates for further amplification, which were then expanded into 6-
well plates. After confirming for fusion protein expression with Western
blot using anti-hGH, selected clones were cultured in T25 as well as
T75, and cryo-vials were established. These stable transfected cells
also were adapted to culture in serum free CD293 media.

107
Chapter 11. Results and Discussions

11.1. Large-scale expression

For large-scale production of fusion proteins, the high quality,
endotoxin-free hGH, THG and GHT plasmids (Fig. 34) were transfected
into HEK293T cells cultured in 20 or 40 T225 flasks, respectively. 2L of
conditioned media from 1
st
collection were harvested. hGH (Fig. 35-
36), THG (Fig. 41) and GHT fusion proteins were enriched by Labscale
TFF concentrator. SDS-PAGE with known Tf controls (Fig. 37-40)
showed that ~ 60 mg GHT fusion protein was extracted with 90-95 %
purity. After 3-day culture with fresh media, the conditioned media
were harvested for 2
nd
time, and GHT fusion proteins were isolated,
and the yield as determined by SDS-PAGE was ~ 70 mg. Cells were
fed with fresh media and cultured for another 2-3 days. The GHT
fusion protein was isolated for 3
rd
time from 2L of conditioned media,
and the yield was ~ 70 mg. Together, transiently transfected HEK293
cells cultured for 8-9 days in serum free CD293 media generated
about 200 mg of fusion protein, sufficient for one in vivo experiment.

108

Fig. 34. Agarose gel of high quality plasmids prepared at large-scale
for transfection. 2 µl plasmid samples were fractionated on 1%
agarose gel and stained with ethidium bromide to assess purity and
concentration. M: DNA marker; lane 1: GHT; lane 2: THG; lane 3:
GT; lane 4: GHT.
Note: GT (hGH-Tf); GHT (hGH-(H4)
2
-Tf ); THG (Tf-(H4)
2
-hGH).
M 1 2 M 3 4

109

1 2 3 4 5
Fig. 35. Large-scale expression of hGH: 1st collection. 20 flasks of
T225 were transfected with hGH construct and samples were
analyzed for purity and quantity by fractionating on SDS-PAGE.
Shown here were analyses of samples from 1st collection. The
concentration and volume were 1.6mg/mL and 27 ml, respectively.
Total yield from 2nd collection was 43 mg. Lanes 1-2:hGH standards
with 2, 4 µg; lanes 3-5: concentrated hGH with 1, 2, 3 µl volume.

110

Fig. 36. Large-scale expression of hGH: 2nd collection. 20 flasks of
T225 were transfected with hGH construct and samples were
analyzed for purity and quantity by fractionating on SDS-PAGE.
Shown here were analyses of samples from 2nd collection. The
concentration and volume were 1.5mg/mL and 25 ml, respectively.
Total yield from 2nd collection was 37.5 mg. Lane 1: marker; lanes 1-
3:hGH standards with 2, 4, 8 µg; lanes 4-6: concentrated hGH with 1,
2, 3 µl volume.
MM 1 2 3 4 5 6

111

Fig. 37. Large-scale expression of GHT fusion protein: 1st collection.
1L of conditioned media from 3-day posttransfection culture were
concentrated to ~ 25 mL using TFF system, followed by buffer
exchange. The concentration was ~ 2 mg/ml, therefore, the
production yield for 1st collection from 20 T225 flasks was ~ 50 mg.
MM: marker; lanes 1-3: Tf standards with 2,4,8 µg; lanes 4-6:
purified GHT with 1, 2, 3 µl volume, respectively.
MM 1 2 3 4 5 6

112

Fig. 38. Large-scale expression of GHT fusion protein: 2nd collection.
1L of conditioned media from 3-day posttransfection culture were
concentrated to ~ 22 mL using TFF system, followed by buffer
exchange. The concentration was ~ 1.5 mg/ml, therefore, the
production yield for 2nd collection from 20 T225 flasks was ~ 33 mg.
MM: marker; lanes 1-3: Tf standards with 2,4,8 µg; lanes 4-6:
purified GHT with 1, 2, 3 µl volume, respectively.
MM 1 2 3 4 5 6

113

Fig. 39. Large-scale expression of GHT fusion protein: 3rd collection.
1L of conditioned media from 3-day post-transfection culture were
concentrated to ~ 22 mL using TFF system, followed by buffer
exchange. The concentration was ~ 1 mg/ml, therefore, the
production yield for 3
rd
collection from 20 T225 flasks was ~ 22 mg.
MM: marker; lanes 1-3: Tf standards with 2,4,8 µg; lanes 4-6:
purified GHT with 1, 2, 3 µl volume, respectively.
MM 1 2 3 4 5 6

114

MM 1 2 3 4 5 6
Fig. 40. GHT fusion protein was expressed at large-scale in serum
free media by the transfected HEK293 cells in 40 T225 flasks. Total,
6L of conditioned medium were harvested, concentrated, and pulled
together using Labscale TFF system. Shown here was SDS-PAGE with
analysis of enriched samples (40 ml) from just 1st collection (2L
medium). MM: marker; lanes 1-3: Tf standards with 2, 4, 8 µg
protein, respectively; lanes 4-6: 1-3 µl aliquots from the enriched
GHT fusion protein, respectively. The gel result shows the
concentration of GHT fusion protein is more than 1 mg/ml.

115

Fig. 41. THG fusion protein was produced at large-scale in serum free
CD293 media by the transfected HEK293T cells. Total of four
collection were pulled together and concentrated using TFF system
and buffer exchanged, which then further concentrated with Amicon-
15 column to ~ 15 mL final volume. The concentration of THG as
estimated by SDS-PAGE and Quantity One software was ~5-6 mg/ml.
MM: marker; lanes 1-3: Tf standards with 2,4,8 µg; lanes 4-6:
purified GHT with 1, 2, 3 µl volume, respectively.
MM 1 2 3 4 5 6

116

Animal experiments with oral protein delivery demands high quality of
fusion protein in large quantity. Therefore, it is importance to develop
an expression system with mammalian cells that produce fusion
protein at high levels. Being a variant of HEK93, the HEK293T cells
express SV40 large T-antigen constitutively (50), combined with a
vector such as pcDNA3.1 with SV40 promoter, plasmid vectors can
replicate and generate many copies of same plasmids, which express
fusion proteins at increased level, as compared to HEK293 parent cells.
Surprisingly, fusion protein expression by these cells was in high
quality, high purity, and high abundance. And because the expressed
fusion protein was in such high purity, no tedious, extra purification
steps were required. Direct concentration of fusion protein with TFF
system greatly aided in product recovery, saved time, increased
productivity, and possibly enhanced or at least preserved biological
activity.

Since TFF system uses tangential flow for sample concentration,
potential damage to fusion protein can be reduced greatly. In addition,
the TFF system has relatively large capacity that concentrates 2-4L of
conditioned media in a day, including buffer exchange. In conclusion,

117
we established condition for a transient transfection of HEK293T cells
with capacity to produce up to 200-400 mg GHT fusion protein in
about one month. This system combined with TFF concentrator has
capability of processing up to 4L of conditioned media in a single day.

11.2. Generation of stable cell lines

Transcient transfection is rapid, convenient, cost-effective approach in
producing fusion proteins at a level that is enough for in vivo animal
experiments. However, it requires frequent transfection and expresses
fusion protein for only about 10-days. In contrast, stable transfection
is capable of expressing fusion protein for unlimited time. Therefore, it
is desirable to establish a stable cell line with permanent expression of
fusion protein. In this regard, HEK293T cells were stably transfected
and single-cell derived colonies were selected using radioactively
labeled hGH radioimmunoassay (RIA) for fusion protein expression.

The RIA assay showed that selected colonies expressed the fusion
protein, indicating the plasmid was integrated into host genome
permanently, because these colonies survived from selection agent,
which kills colonies without genome integration. Western blot analysis

118
performed with anti-hGH antibody showed fusion protein bands at
expected molecular weight (Fig. 42), ~ 100 kDa. Subsequently,
selected colonies were then developed into cell lines and then adapted
to grow in serum free culture. Stable cell line revived from cryo-frozen
vials were tested with Western blot for fusion protein expression.

119

Fig. 42. Anti-hGH Western blot was used to select stable clone with
high expression. 7.5 µl conditioned media from 96-well plates was
analyzed with antibody against hGH to select fusion protein
expressing colonies. Based on the hGH Western blot result, the
following colonies were selected, each derived from a single cell,for
the stable production of fusion proteins. THG-4 and 7; GT-11; GHT-13
and 14. C is control; M is marker.
GT GHT hGH THG
kDa
100
8 7 6 5 4 3 2 1 C M
THG
18 17 16 15 14 13 12 11 10 9

120

Result demonstrated that this newly established stable cell line
expressed fusion protein (Fig. 43), confirming stable integration into
host genome and adoption into serum free media. However, compared
to transient transfection, the stable cell lines generated extra protein
band other than fusion protein. This is a concern because extra-band
could minimize the expression of fusion protein and mandates further
purification for demanding applications such as oral delivery.
Nonetheless, our data showed stable cell line with fusion protein
expression can be established in reasonable time, and it can be
adapted to serum free media.

Fig. 43. Western blot with anti-hGH antibody Confirmed the stable
cell line GHT-14 for expression. Stable transfected cells from GHT-
14 clone were expanded to T25 followed by T75 flasks. DMEM
media with serum in T75 flask was replaced with CD293 media
and incubated for about 24 hrs, and then the conditioned media
was concentrated using Amicon-15 to 500 uL. In this gel about 30
µL of sample was loaded and shows expression of fusion protein
GHT. Lanes 1-4 (GT, GHT, TG, THG for expression study); lane 5:
stable GHT-14 cell line.

1 2 3 4 5

121
PART 4. In vivo bioactivity of the fusion proteins

Chapter 12. Introduction

Fusion proteins including GT, GHT, and THG all showed good biological
activity as determined by their ability to induce Nb2 cell proliferation
and TfR binding. However, our goal is to demonstrate oral absorption
of hGH-Tf fusion protein in animal disease model for human growth
hormone, and in vivo environment is different from that of in vitro.
Furthermore, in vivo model has so many variables and unknowns that
could affect the function of fusion protein, thus conducting in vivo
experiments for full evaluation of fusion protean is necessary.

Oral administration requires large quantities of biologically active
fusion proteins due to intrinsically low oral bioavailability. Fortunately,
our production scheme with adherent HEK293T cells has capacity to
produce GHT fusion protein up to several hundred milligram with
short-turnaround, allowing us to conduct in vivo animal experiments.

122
Selecting suitable animal model for human growth hormone is
essential step in designing in vivo experiments (8). There are two in
vivo assays that are currently used to detect biological activity of hGH.
One of which measures bone growth using tibia, and another
measures body weight gain, after treatment with hGH in
hypophysectomized rats (15, 16). Because the pituitary gland was
surgically removed, these hypophysectomized rats are not capable of
secreting their own growth hormone, and conceivably they have
trouble in weight-gain and bone-growth. Furthermore, it has been
known that the hypophysectomized rats respond to exogenously
added-hGH in dose-dependent manner. Therefore, we decided to use
the hypophysectomized rats, as an animal model, to study the effects
of recombinant fusion protein in vivo, by measuring the body weight
before and after treatment with fusion proteins.

123
We have constructed and subsequently expressed several variants of
hGH-Tf fusion proteins to evaluate them for in vitro and in vivo
activity, and thus to select the very best fusion protein with enhanced
production, stability, and bioactivity. In vitro assays including
activation of signal transduction pathways in IM-9 cell and cell
proliferation assay with Nb2 cells, however, did not differentiate
among various fusion proteins definitively as to which one is the most
suitable for further testing in animal model. Therefore, in this chapter
three fusion proteins of hGH and Tf will be assessed for their in vivo
biological activity when administered both orally and subcutaneously.
Based on the performance in initial animal experiments, one fusion
protein was selected and further experimented to achieve oral delivery
in hypophysectomized rats.

124
Chapter 13. Materials and Methods

13.1. Hypophysectomized rats and dosing strategy

The hypophysectomized rats cannot produce growth hormone
endogenously due to the pituitary gland removal surgery, and they
respond to exogenous hGH stimulation by weight gain. Therefore,
these rats were used as animal model evaluating in vivo biological
activity of the hGH–Tf fusion protein. Female Sprague Dawley rats with
hypophysectomy (Taconic, Germantown, NY), housed two per cage in
a temperature and light controlled room with free access to food and
water, were subjected to initial weight screening for 7-days. The rats
with average weight of 100 g and gaining no more than 4 g of
bodyweight were selected for the experiment. However, for rats with
average weight of 90 g or less, only the rats with no more than 1.5 g
body weight increase over 7-day weight-screening period were
selected for the in vivo experimentation.

The hypophysectomized rats were assigned to five or six experimental
groups each with five rats depending on the purpose of the
experiment. For example, to evaluate the effect of Bowman-Birk
Inhibitor (BBI), a soybean trypsin/chymotrypsin inhibitor, on oral

125
absorption of GHT fusion, the experiment had six groups including: 1)
GHT fusion protein with BBI, 2) GHT fusion protein without BBI, 3)
hGH with BBI, 4) hGH without BBI, 5) mannitol-phosphate buffer, 6)
no treatment as control. Rats were fasted for 4 hrs, weighed, and then
dosed for 7-consecutive days with 12.5 mg/kg fusion protein and 2.5
mg/kg hGH in 0.5 ml volume for oral administration using gavage
needle, and 1.25 mg/kg fusion protein, 0.25 mg/kg hGH in 0.2 ml
volume for subcutaneous administration. The dose of fusion protein
was normalized to that of hGH to ensure that the hGH in fusion protein
is equal in molar to endogenous hGH.

In dose escalation study with GHT fusion protein, the oral dose for
GHT fusion protein was 50 mg/kg. However, in another dose-
escalation study where GHT fusion protein coadministered with BBI
inhibitor, the oral dose for GHT was 25 mg/kg, which is 2-fold more
than the regular oral dose (12.5 mg/kg) for GHT. Oral dose
preparation also contained sodium bicarbonate (30 mg/ml) in order to
neutralize the acid in stomach. Likewise, the subcutaneous dose
contained 1x phosphate buffered saline as an isotonic agent. The
average daily weight change was calculated for each group of rats, and
plotted against time as day.

126
13.2. Trypsin/chymotrypsin proteolysis and Bowman Birk
protease inhibitor
Fusion proteins including GT, GHT, and THG were tested in vitro to
evaluate the extent of proteolysis by trypsin or chymotrypsin,
representing major intestinal enzymes that breakdown protein drugs.
Each fusion protein was digested with trypsin (Sigma) or chymotrypsin
(Sigma) in a reaction condition closely resembling the intestinal fluid.
The cleavage reaction was performed using 1.7 ml centrifuge tube in
PBS containing 25 µg fusion protein, 5 Units trypsin, and incubated at
37 °C water bath for predetermined time points including: 0, 15, 30,
45, 60, 120 minutes.

To demonstrate whether the Bowman Birk protease inhibitor (BBI), a
soybean trypsin inhibitor, can protect fusion protein from tryptic or
chymotryptic hydrolysis, 50 µg BBI (Sigma) was added to the cleavage
reaction as described above. When reaction complete, the samples
were removed immediately from water bath at the different time
points, placed on ice, mixed with 15 µl 2X sample buffer, boiled for 5
min, and kept at 4 °C. Samples then were fractionated on SDS-PAGE,

127
stained in Coomassie blue, destained, and analyzed using HighChem
XBR Imaging System with Quantity One software (Bio-Rad). In
addition, same amount of Tf (25 µg), as other fusion proteins, was
digested with same units of trypsin or chymotrypsin, in presence or
absence of BBI.
13.3. Statistical analysis
The-Two-sample-T-test (http://ccnmtl.columbia.edu/projects/qmss/t_two.html)
was used to evaluate weight gain in rats with and without oral dose
and to determine whether the gain in weight between these groups is
significant. The weight gain in rats with oral dosing is considered
significant, if p < 0.05, when compared to no treatment or buffer
control group.

128
Chapter 14. Results

14.1. s.c. administration of GT fusion protein promoted weight
gain, while oral administration did not have impact on weight
increase in hypophysectomized rats

The hypophysectomized rats that gained weight during 7-day weight-
screening period were excluded from the experiment. The data
obtained from 7-daily subcutaneous injection with 1.25 mg/kg dose
showed that the fusion protein was biologically active in vivo and
promoted 11.2 g of weight gain accounting for approximately 11 %
increase in body weight (Fig. 44). This effect on weight gain was
smaller than the hGH positive control, which caused about 16 g of
body weight gain with the same dose. Conversely, the
hypophysectomized rats injected with buffer control were unable to
gain weight (Fig. 44, Table 5b). The subcutaneous injection of fusion
protein therefore resulted in a strong growth response by promoting
weight gain in hypophysectomized rats.

The same fusion protein (GT) was also orally administered in
hypophysectomized rats for 7-days to evaluate oral biological activity.

129
But results from oral administration did not show weight increase
when compared with negative control groups (Fig. 44 and Table 5a
and b).

130

Fig. 44. Evaluating GT fusion protein for both oral (p.o.) and
subcutaneous (s.c.) bioactivity using weight gain as indicator in
hypophysectomized (hGH-deficient) rats. Rats were administered with
GT fusion protein either orally or subcutaneously at doses shown in
figure legends for 7-consecutive days. The change in weight,
compared to the weight on the day of dosing began (day 1), was
measured for each day and plotted against time as day. GT fusion
protein led to significant weight increase when administered
subcutaneously. However, when given orally, there was no change in
weight compared to buffer control. Data were represented as average
±SEM; (n=5), except GT p.o. where n=3.

131

Table 5a. Oral and subcutaneous administration of GT fusion protein in
hypophysectomized rats.
Day Rat 1 Rat 2 Rat 3 Rat 4 Rat 5 chnge1 chnge2 chnge3 chnge4 chnge5 Ave SD SEM
0 92.9 89.0 83.2 90.2 105.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
1 96.2 90.3 87.3 92.7 106.7 3.3 1.3 4.1 2.5 1.6 2.6 1.0 0.5
2 95.4 91.2 89.0 95.0 108.4 2.5 2.2 5.8 4.8 3.3 3.7 1.4 0.6
3 96.5 91.6 91.1 98.8 106.6 3.6 2.6 7.9 8.6 1.5 4.8 2.9 1.3
4 98.3 94.5 90.4 98.1 110.8 5.4 5.5 7.2 7.9 5.7 6.3 1.0 0.5
5 98.6 96.6 91.4 99.6 111.7 5.7 7.6 8.2 9.4 6.6 7.5 1.3 0.6
6 102.0 99.3 94.6 102.1 113.3 9.1 10.3 11.4 11.9 8.2 10.2 1.4 0.6
7 102.8 100.9 99.7 103.3 112.7 9.9 11.9 16.5 13.1 7.6 11.8 3.0 1.3
GT (1.25 mg/kg, s.c.)
Day Rat 1 Rat 2 Rat 3 Rat 4 Rat 5 chnge1 chnge2 chnge3 chnge4 chnge5 Ave SD SEM
0 100.0 92.1 98.7 91.9 92.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
1 98.5 93.1 101.0 92.4 93.8 -1.5 1.0 2.3 0.5 1.3 0.7 1.4 0.6
2 99.4 91.2 97.6 92.8 93.4 -0.6 -0.9 -1.1 0.9 0.9 -0.2 1.0 0.4
3 101.5 95.8 102.0 92.1 93.4 1.5 3.7 3.3 0.2 0.9 1.9 1.5 0.7
4 100.0 90.6 102.2 93.1 92.4 0.0 -1.5 3.5 1.2 -0.1 0.6 1.9 0.8
5 101.1 91.6 100.4 90.0 92.5 1.1 -0.5 1.7 -1.9 0.0 0.1 1.4 0.6
6 102.6 92.9 102.4 91.0 91.5 2.6 0.8 3.7 -0.9 -1.0 1.0 2.1 0.9
7 104.1 92.4 102.2 90.4 91.9 4.1 0.3 3.5 -1.5 -0.6 1.2 2.5 1.1
GT (12.5 mg/kg, p.o.)

132

Day Rat 1 Rat 2 Rat 3 Rat 4 Rat 5 chnge1 chnge2 chnge3 chnge4 chnge5 Ave SD SEM
0 90.0 84.3 100.3 90.1 90.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
1 92.9 86.4 103.2 93.2 93.2 2.9 2.1 2.9 3.1 3.2 2.8 0.4 0.2
2 96.3 90.2 106.2 94.8 95.9 6.3 5.9 5.9 4.7 5.9 5.7 0.5 0.2
3 98.4 92.6 106.3 99.3 97.0 8.4 8.3 6.0 9.2 7.0 7.8 1.1 0.5
4 98.5 94.2 109.1 98.1 98.6 8.5 9.9 8.8 8.0 8.6 8.8 0.6 0.3
5 101.0 96.0 110.3 101.0 99.4 11.0 11.7 10.0 10.9 9.4 10.6 0.8 0.4
6 105.9 98.9 112.5 103.9 102.3 15.9 14.6 12.2 13.8 12.3 13.8 1.4 0.6
7 107.7 100.0 114.0 106.1 103.0 17.7 15.7 13.7 16.0 13.0 15.2 1.7 0.8
hGH (0.25 mg/kg, s.c.)
Day Rat 1 Rat 2 Rat 3 Rat 4 Rat 5 chnge1 chnge2 chnge3 chnge4 chnge5 Ave SD SEM
0
87.3 87.6 92.1 99.3 106.6
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
1
88.1 88.5 93.1 101.1 108.7
0.8 0.9 1.0 1.8 2.1 1.3 0.6 0.3
2
87.2 89.8 93.5 102.8 108.8
-0.1 2.2 1.4 3.5 2.2 1.8 1.3 0.6
3
86.6 87.6 94.7 102.2 109.4
-0.7 0.0 2.6 2.9 2.8 1.5 1.7 0.8
4
85.0 88.1 94.5 103.2 109.9
-2.3 0.5 2.4 3.9 3.3 1.6 2.5 1.1
5
85.7 86.6 94.4 102.6 109.8
-1.6 -1.0 2.3 3.3 3.2 1.2 2.4 1.1
6
85.9 87.5 95.2 105.9 113.1
-1.4 -0.1 3.1 6.6 6.5 2.9 3.7 1.6
7
86.6 88.5 94.4 104.7 110.7
-0.7 0.9 2.3 5.4 4.1 2.4 2.4 1.1
hGH (2.5 mg/kg, p.o.)
Table 5b. Oral and subcutaneous administration of GT fusion protein
in hypophysectomized rats.

133

14.2. s.c. administration of GHT fusion protein promoted
significant weight gain meanwhile oral administration led to
modest weight gain in hypophysectomized rats

To demonstrate oral hGH activity of the fusion protein, the
hypophysectomized rats were orally dosed with 12.5 mg/kg of GHT
fusion protein for 7-consecutive days. The daily weight change was
analyzed as compared to 1
st
day of treatment. Results from 7-daily
treatment demonstrated that the orally dosed hypophysectomized rats
led to a modest 6 g gain in body weight. Conversely, the
hypophysectomized rats dosed with hGH orally did not gain weight
(Fig. 45).

On the other hand, when subcutaneously administered, the GHT fusion
protein resulted in strong body weight increase of 16 g (Fig. 45 and
Table 6a). This affect was comparable in magnitude to that of hGH
alone, which also yielded 16 g body weight gain in 7-day subcutaneous
administration (Fig. 45 and Table 6b). In contrast, the buffer
administered subcutaneously did not induce body weight gain. Taken
together, the GHT fusion protein showed strong growth promotion with

134
subcutaneous administration, and mild weight gain with oral
administration (Fig. 45 and Table 6a).

Fig. 45. Evaluating GHT fusion protein for both oral and
subcutaneous bioactivity. Rats without capability of producing
hGH were administered GHT fusion protein either orally or
subcutaneously at doses shown in figure legends for 7-
consecutive days. The change in weight compared to the
weight on the day of dosing was measured for each day and
plotted. GHT fusion protein led to significant weight increase
when administered subcutaneously. However, when given
orally, the weight increase was small but significant compared
to buffer control. Data were represented as average ±SEM;
n=3, except hGH (s.c.), where n=2.

135

Day Rat 1 Rat 2 Rat 3 chge 1 chge 2 chge 3 Ave. SD SEM
0 103 106 109 0 0 0 0.0 0.0 0.0
1 108 106 114 5 0 5 3.3 2.9 1.7
2 112 112 117 9 6 8 7.7 1.5 0.9
3 114 111 119 11 5 10 8.7 3.2 1.9
4 114 112 121 11 6 12 9.7 3.2 1.9
5 116 114 123 13 8 14 11.7 3.2 1.9
6 120 116 125 17 10 16 14.3 3.8 2.2
7 123 118 126 20 12 17 16.3 4.0 2.3
GHT (1.25 mg/kg, s.c.)
Day Rat 1 Rat 2 Rat 3 chge 1 chge 2 chge 3 Ave. SD SEM
0 115 108 114 0 0 0
0.0 0.00 0.0
1 117 110 113 2 2 -1
1.0 0.58 0.3
2 117 111 115 2 3 1
2.0 0.82 0.5
3 117 114 116 2 6 2
3.3 1.89 1.1
4 119 115 118 4 7 4
5.0 1.41 0.8
5 118 114 118 3 6 4
4.3 1.25 0.7
6 120 115 118 5 7 4
5.3 1.25 0.7
7 121 116 118 6 8 4
6.0 1.63 0.9
GHT (12.5 mg/mg, p.o.)
Table 6a. Oral and subcutaneous administration
of GHT fusion protein in hypophysectomized rats.

136

Day Rat 1 Rat 2 Rat 3 chge 1 chge 2 chge 3 Ave. SD SEM
0
107 109 0 0 0.0 0.0 0
1
110 110 3 1 2.0 1.4 1
2
113 116 6 7 6.5 0.7 0.5
3
114 118 7 9 8.0 1.4 1
4
116 119 9 10 9.5 0.7 0.5
5
118 121 11 12 11.5 0.7 0.5
6
120 125 13 16 14.5 2.1 1.5
7
123 128 16 19 17.5 2.1 1.5
hGH (0.25 mg/mg, s.c.)
Day Rat 1 Rat 2 Rat 3 chge 1 chge 2 chge 3 Ave. SD SEM
0 95 99 109 0 0 0 0.0 0.0 0.0
1 97 100 107 2 1 -2 0.3 0.7 0.4
2 98 100 110 3 1 1 1.7 1.2 0.7
3 97 98 111 2 -1 2 1.0 0.0 0.0
4 97 100 108 2 1 -1 0.7 0.7 0.4
5 97 98 108 2 -1 -1 0.0 0.7 0.4
6 99 101 109 4 2 0 2.0 2.0 1.2
7 98 101 112 3 2 3 2.7 0.6 0.3
Control (mannitol/phosphate buffer)
Table. 6b. Subcutaneous administration of hGH as positive
control, and buffer negative as control in hypophysectomized
rats.

137
14.3. s.c. administration of THG fusion protein promoted
significant weight gain meanwhile oral administration did not
produce weight gain in hypophysectomized rats

In a same set of experiment carried out at same time, the THG fusion
protein was evaluated for its in vivo biological activity when
administered both orally and subcutaneously. 7-day subcutaneous
administration of THG fusion protein showed that it strongly increased
body weight gain and this affect was identical to both of hGH and GHT
fusion protein, gaining ~ 16 of body weight (Fig. 46 and Table 7).

Also THG fusion protein was analyzed for oral growth promotion in 7-
day oral dosing, and results showed that THG fusion protein failed to
induce body weight gain (Fig. 46 and Table 7), which was different
from the effect of GHT fusion protein that led to modest (6 g) body
weight gain.

138

Fig. 46. Evaluating THG fusion protein for both oral and
subcutaneous bioactivity. Rats without capability of producing hGH
were administered THG fusion protein either orally or subcutaneously
at doses shown in figure legends for 7-consecutive days. The change
in weight compared to the weight on the day of dosing was measured
for each day and plotted. THG fusion protein led to significant weight
increase when administered subcutaneously. However, when given
orally, there was no change in weight compared to buffer control.
Data were represented as average ±SEM; n=3, except THG s.c., n=4.

139

Day Rat 1 Rat 2 Rat 3 chge 1 chge 2 chge 3 Ave. SD SEM
0 105 94 104 0 0 0 0.0 0.0 0.0
1 107 98 109 2 4 5 3.7 1.5 0.9
2 114 100 113 9 6 9 8.0 1.7 1.0
3 116 102 115 11 8 11 10.0 1.7 1.0
4 116 103 116 11 9 12 10.7 1.5 0.9
5 117 103 117 12 9 13 11.3 2.1 1.2
6 120 108 120 15 14 16 15.0 1.0 0.6
7 123 107 119 18 13 15 15.3 2.5 1.5
THG (1.25 mg/mg, s.c.)
Day Rat 1 Rat 2 Rat 3 Rat 4 chge 1 chge 2 chge 3 chge 4 Ave. SD SEM
0 100 104 99 105 0 0 0 0 0.0 0.0 0.0
1 101 104 103 107 1 0 4 2 1.8 1.7 0.9
2 101 103 100 105 1 -1 1 0 0.3 1.0 0.5
3 102 103 101 106 2 -1 2 1 1.0 1.4 0.7
4 104 104 100 106 4 0 1 1 1.5 1.7 0.9
5 103 104 101 104 3 0 2 -1 1.0 1.8 0.9
6 105 105 105 103 5 1 6 -2 2.5 3.7 1.8
7 107 105 101 101 7 1 2 -4 1.5 4.5 2.3
THG (12.5 mg/mg, p.o.)
Table 7. Oral and subcutaneous administration of
THG fusion protein in hypophysectomized rats.

140

14.4. Choosing GHT fusion protein for further in vivo evaluation

Three fusion proteins were evaluated for their in vivo biological activity
in hypophysectomized rats with both subcutaneous and oral
administration. Preliminary results indicated that GHT fusion protein
possessed best bioactivity, resulting in 16 g and 6 g body weight gain
with subcutaneous and oral administration, respectively. In addition,
GHT fusion protein express at high levels compared to other 2 fusion
proteins. Therefore, based on these results we selected GHT fusion
protein for further analysis in animal experiments.

14.5. Dose-escalation study with GHT fusion protein

As described above, GHT fusion protein demonstrated modest oral
activity, while other fusion proteins (GT and THG) did not promote
weight increase in oral administration. Next, we wanted to test GHT
fusion protein for oral activity when administered with elevated dose
(~ 50 mg/kg) to evaluate if the weight gain by oral GHT is dose-
dependent. Hypophysectomized rats were orally administered with
high dose of GHT fusion protein for 4-consecutive days, and weight

141
change in each day verses first day of dosing was plotted (Fig. 47,
Table 8a and b). Data from 4-day oral administration did not support
weight increase, while subcutaneous administration of GHT led to
strong increase in body weight (Fig. 47 and Table 8b). Therefore, oral
administration of GHT with high dose did not induce weight gain in
dose-dependent manner (Fig. 47 and Table 7a). Furthermore, since no
weight increase was observed after 4-day dosing, experiment was
halted to save and use fusion protein in other experiments.

142

Fig. 47. Dose-escalation study of GHT fusion protein with oral
administration. Hypophysectomized rats were administered GHT
fusion protein at high dose (50 mg/kg) orally for 4-consecutive days.
However, there was no weight-gain after 4-day’s of dosing, and
experiment discontinued. Subcutaneously administered GHT fusion
protein (1.25 mg/kg) led to significant weight increase in 4-day
dosing. Data were represented as average ±SEM; n=5.

143

Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD
0 94 94 99 101 95 0 0 0 0 0 0 0.0
1 95 94 101 100 93 1 0 2 -1 -2 0 1.0
2 97 95 99 102 94 3 1 0 1 -1 0.8 1.3
3 95 96 103 103 94 1 2 4 2 -1 1.6 1.3
4 96 97 101 104 95 2 3 2 3 0 2 1.2
GHT p.o. (50 mg/kg, or 5 mg/rat; 10 mg/kg molar equvivalent of hGH)
Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD
0 104 99 93 93 108 0 0 0 0 0 0 0.0
1 102 99 95 94 108 -2 0 2 1 0 0.2 1.0
2 105 104 95 93 111 1 5 2 0 3 2.2 1.9
3 105 102 95 95 111 1 3 2 2 3 2.2 0.8
4 106 105 95 95 108 2 6 2 2 0 2.4 2.2
hGH p.o. (10mg/kg, or 1 mg/rat)
Table 8a. Dose-escalation study with GHT fusion protein, along with
hGH in hypophysectomized rats.

144

Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD
0 101 93 94 104 98 0 0 0 0 0 0 0.0
1 107 97 95 108 102 6 4 1 4 4 3.8 1.8
2 108 101 97 113 104 7 8 3 9 6 6.6 2.3
3 111 102 99 114 107 10 9 5 10 9 8.6 2.1
4 112 103 100 115 106 11 10 6 11 8 9.2 2.2
GHT s.c. (1.25 mg/kg, or 0.125 mg/rat; 0.25 mg/kg molar equvivalent of hGH)
Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD
0 93 102 108 99 107 0 0 0 0 0 0 0.0
1 95 101 108 100 107 2 -1 0 1 0 0.4 1.0
2 95 100 108 99 107 2 -2 0 0 0 0 1.0
3 95 103 108 101 109 2 1 0 2 2 1.4 0.9
4 97 102 109 104 110 4 0 1 5 3 2.6 2.1
Control p.o. (mannitol/phosphate buffer)
Table 8b. Subcutaneous administration of GHT fusion protein
as positive control, along with buffer negative control in
hypophysectomized rats.

145

14.6. Chymotrypsin led to degradation of GHT fusion protein

Even though GHT fusion protein demonstrated mild increase in body
weight gain with 12.5 mg/kg oral dose, it failed to produce body
weight gain when administered orally at high-dose, 50 mg/kg. This
was a concern, and we wanted to address this inconsistency in oral
activity of GHT fusion protein. Therefore, we reasoned that proteases
such as trypsin and chymotrypsin could cleave the GHT fusion protein
before absorption occurs in small intestine. To determine the affect of
chymotrypsin on stability of GHT fusion protein, a cleavage reaction
was assembled with chymotrypsin and fusion protein and incubated at
37 ºC for different length of time (Fig. 48).

Chymotrypsin cleavage of GHT fusion protein showed that this fusion
protein is very sensitive and is cleaved extensively. The cleavage
starts as early as in 15 minutes, which cleaved about 25% of GHT. By
1 hr, 70-80% of GHT fusion protein was cleaved; by 2 hrs, almost all
of the fusion protein was degraded (Fig. 48). To rule out the possibility
of other factors including the temperature on stability of GHT fusion
protein, it was incubated at same pre-determined time points at 37 ºC

146
in absence of chymotrypsin. Result showed GHT fusion protein was
intact and stable during 2-hour incubation period without degradation,
indicating the degradation of GHT is chymotrypsin-dependent (Fig.
49).

147

GH
T
GHT + BBI + Chymotrypsin GHT + Chymotrypsin
MM 0 15 30 45 60 120 15 30 45 60 120
mmin
Tf
Fig. 48. Chymotrypsin led to a time-dependent proteolysis of GHT
fusion protein, but co-incubation with Bowman-Birk Inhibitor blocked
chymotrypsin activity. To assess chymotrypsin sensitivity, 25 µg GHT
fusion protein was digested with 0.5 µg chymotrypsin and incubated
at 37 °C at varying time points. Then, to investigate whether such
cleavage can be blocked, one set of reactions prepared with exactly
same conditions as above were added with 50 µg Bowman-Birk
Inhibitor. Samples were removed immediately after completion of
cleavage reaction at designated time points, put on ice, mixed with
sample buffer, boiled for 5 min, and kept at 4 deg. 20 µl of samples
(~ 5 µg) were analyzed by SDS-PAGE, stained in Coomassie blue, de-
stained, and imaged using Bio-Rad Imaging System.

148

GHT
15 30 45 60 120 MM 0 15 30 45 60 120
120
THG
Fig. 49. Fusion proteins GHT and THG were intact after 2 hr
incubation at 37 °C with absence of trypsin. As a control, both GHT
and THG fusion protein were incubated in exactly same reaction
conditions used for c enzymtic leavage. The SDS-PAGE confirmed that
both fusion proteins are stable at 37 °C for at least 2 hrs. Thus this
result indicated that degradation of GHT and THG fusion protein was
due to enzymatic cleavage by trypsin and chymotrypsin.

149

14.7. Chymotrypsin led to degradation of THG fusion protein

On the other hand THG fusion protein was ineffective in promoting of
body weight gain. But we wanted to compare chymotryptic stability of
GHT and THG fusion protein as this provides further information on
oral activity and stability. Therefore, THG fusion protein was
challenged with chymotrypsin in a reaction similar to the GHT
cleavage. The SDS-PAGE analysis showed that THG fusion protein was
also sensitive to tryptic cleavage, and became degraded in presence of
chymotrypsin. At 1 hr time point, most of the THG fusion protein was
cleaved (Fig. 50).

150

Fig. 50. Chymotrypsin led to a time-dependent proteolysis of THG
fusion protein, but co-incubation with Bowman-Birk Inhibitor blocked
chymotrypsin activity. To assess chymotryptic sensitivity, 25 µg THG
fusion protein was digested with 0.5 µg chymotrypsin and incubated
at 37 °C at varying time points. Then, to investigate whether such
cleavage can be blocked, one set of reactions prepared with exactly
same conditions as above were added with 50 µg Bowman-Birk
Inhibitor. Samples were removed immediately after completion of
cleavage reaction at designated time points, put on ice, mixed with
sample buffer, boiled for 5 min, and kept at 4 deg. 20 µl of samples
(~ 5 µg) were analyzed by SDS-PAGE, stained in Coomassie blue, de-
stained, and imaged using Bio-Rad Imaging System.
THG + BBI + Chymotrypsin THG + Chymotrypsin
MM 0 15 30 45 60 120 15 30 45 60 120

151

14.8. Trypsin led to degradation of GHT fusion protein

To determine the affect of trypsin on stability of GHT fusion protein, a
cleavage reaction was performed with trypsin, and incubated at 37 ºC
for different length of time. Trypsin cleavage of GHT fusion protein
showed that GHT fusion protein is very sensitive and cleavage starts
as early as in 15 minutes. By 1 hr, the majority of GHT fusion protein
was degraded by trypsin (Fig. 51).

152

GHT + BBI + Trypsin GHT + Trypsin
MM 0 15 30 45 60 120 15 30 45 60 120
Fig. 51. Trypsin led to a time-dependent proteolysis of GHT fusion
protein, but co-incubation with Bowman-Birk Inhibitor blocked trypsin
activity. To assess tryptic sensitivity, 25 µg GHT fusion protein was
digested with 0.5 µg trypsin and incubated at 37 °C at varying time
points. Then, to investigate whether such cleavage can be blocked,
one set of reactions prepared with exactly same conditions as above
were added with 50 µg Bowman-Birk Inhibitor. Samples were
removed immediately after completion of cleavage reaction at
designated time points, put on ice, mixed with sample buffer, boiled
for 5 min, and kept at 4 deg. 20 µl of samples (~ 5 µg) were analyzed
by SDS-PAGE, stained in Coomassie blue, de-stained, and imaged
using Bio-Rad Imaging System.

153

14.9. Trypsin led to degradation of THG fusion protein

For the same reason as GHT fusion protein, the THG fusion protein
was also tested for tryptic sensitivity in a reaction similar to GHT. The
SDS-PAGE analysis confirmed that the THG fusion protein was not
stable in presence of trypsin. At 1 hr post incubation with trypsin
almost all of the THG fusion protein was cleaved without intact fusion
protein (Fig. 52).

154

THG + BBI + Trypsin THG + Trypsin
MM 0 15 30 45 60 120 15 30 45 60 120
min
Fig. 52. Trypsin led to a time-dependent proteolysis of THG fusion
protein, but co-incubation with Bowman-Birk Inhibitor blocked trypsin
activity. To assess tryptic sensitivity, 25 µg THG fusion protein was
digested with 0.5 µg trypsin and incubated at 37 °C at varying time
points. Then, to investigate whether such cleavage can be blocked,
one set of reactions prepared with exactly same conditions as above
were added with 50 µg Bowman-Birk Inhibitor. Samples were
removed immediately after completion of cleavage reaction at
designated time points, put on ice, mixed with sample buffer, boiled
for 5 min, and kept at 4 deg. 20 µl of samples (~ 5 µg) were analyzed
by SDS-PAGE, stained in Coomassie blue, de-stained, and imaged
using Bio-Rad Imaging System.

155
14.10. Transferrin was resistant to both trypsin and
chymotrypsin

It was reported that diferric-Tf was stable and resistant to trypsin-
induced cleavage. However, we wanted to test Tf for tryptic as well as
chymotryptic stability in a physiologically relevant reaction condition.
SDS-PAGE analysis of tryptic and chymotryptic cleavage reaction
verified that Tf in deed was strongly resistant to both trypsin and
chymotrypsin (Fig. 53 and 54).

14.11. Soybean trypsin inhibitor (BBI) blocked both tryptic and
chymotryptic cleavage of GHT and THG fusion proteins

Bowman Birk inhibitor, a soybean trypsin inhibitor, is known to inhibit
both trypsin and chymotrypsin. To demonstrate its inhibition effect on
GHT and THG fusion proteins, the inhibitor was coincubated with GHT
or THG fusion protein in cleavage reactions containing either
chymotrypsin or trypsin. The results as shown in SDS-PAGE analysis
revealed that the Bowman Birk inhibitor was highly effective in
blocking trypsin and chymotrypsin activity (Fig. 48-52).

156

Tf + BBI + Trypsin Tf + Trypsin
MM 0 15 30 45 60 120 15 30 45 60 120
Fig. 53. Transferrin (Tf) demonstrated strong resistance to tryptic
cleavage. To assess tryptic sensitivity, 25 µg Tf was digested with 0.5
µg trypsin and incubated at 37 °C at varying time points, in presence
or absence of 50 µg BBI. Samples were removed immediately after
completion of cleavage reaction at designated time points, put on ice,
mixed with sample buffer, boiled for 5 min, and kept at 4 deg. 20 µl
of samples (~ 5 µg) were analyzed by SDS-PAGE, stained in
Coomassie blue, de-stained, and imaged using Bio-Rad Imaging
System.

157

Tf + BBI + Chymotrypsin Tf + Chymotrypsin
MM 0 15 30 45 60 120 15 30 45 60 120
min
Tf
Fig. 54. Transferrin (Tf) demonstrated strong resistance to
chymotryptic cleavage. To assess chymotryptic sensitivity, 25 µg Tf
was digested with 0.5 µg chymotrypsin and incubated at 37 °C at
varying time points, in presence or absence of 50 µg BBI. Samples
were removed immediately after completion of cleavage reaction at
designated time points, put on ice, mixed with sample buffer, boiled
for 5 min, and kept at 4 deg. 20 µl of samples (~ 5 µg) were analyzed
by SDS-PAGE, stained in Coomassie blue, de-stained, and imaged
using Bio-Rad Imaging System.

158

14.12. GT fusion protein is susceptible to both tryptic and
chymotryptic degradation

Results from stability analysis of GHT and THG fusion proteins with
both trypsin and chymotrypsin showed that they could be cleaved.
This cleavage raises a question on stability of the helical linker in GHT
and THG fusion protein. Therefore, We investigated tryptic and
chymotryptic stability of GT fusion protein, with short di-peptide linker.
Our results revealed that GT fusion protein was also susceptible to
these two enzymes (Fig. 55), suggesting that the instability of the
fusion protein was not due to the inserted helical linker.

159

Typsin cleavage of GT fusion protein
MM 0 15 30 45 60 120 min
MM 0 15 30 45 60 120 min
Chymotypsin cleavage of GT fusion protein
Fig. 55. Both trypsin and Chymotrypsin led to a time-dependent
proteolysis of GT fusion protein. To assess tryptic sensitivity, ~ 25 µg
GT fusion protein was digested with 0.5 µg trypsin and incubated at
37 °C at varying time points. Samples were removed immediately
after completion of cleavage reaction at designated time points, put
on ice, mixed with sample buffer, boiled for 5 min, and kept at 4 deg.
20 µl of samples (~ 5 µg) were analyzed by SDS-PAGE, stained in
Coomassie blue, de-stained, and imaged using Bio-Rad Imaging
System.

160
14.13. Oral co-administration of GHT fusion protein with BBI
led to weight gain with improved consistency

To improve tryptic/chymotryptic stability and thus oral activity, GHT
fusion protein was coadministered with BBI in hypophysesctomized
rats. Data from 7-day oral administration illustrated that the rats co-
administered with fusion protein and BBI promoted modest ~ 2.5
gram weight gain (Fig. 56 and Table 9a), whereas the rats in control
groups not only failed to gain weight, but lost weight (Fig. 56 and
Table 9b). Modest but consistent weight gains in hypophysecotmized
rats that were orally administered with fusion protein mixed with BBI
suggested that soybean trypsin inhibitor could be effective in
improving oral growth promoting activity of the fusion protein.

161

Fig. 56. Oral administration of GHT fusion protein together with
protease inhibitor BBI led to a small weight increase. GHT fusion
protein was coadministered with BBI inhibitor orally for 7-days at
12.5 mg/kg. Data from daily weight measurement showed small but
consistent increase in weight in rats receiving both GHT and BBI, as
compared to negative control groups. Data were represented as
average ±SEM; n=3-5.

162

Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD SEM
0 92 92.4 91.2 92.2 93.6 0 0 0 0 0 0.0 0.0 0.0
1 92 92.2 93 93.5 93.7 0 -0.2 1.8 1.3 0.1 0.6 0.9 0.4
2 92.1 91.4 92.5 93.2 94.2 0.1 -1 1.3 1 0.6 0.4 0.5 0.2
3 93 94.7 92.7 93.8 94.1 1 2.3 1.5 1.6 0.5 1.4 0.7 0.3
4 91.5 94.1 92.1 94 94.7 -0.5 1.7 0.9 1.8 1.1 1.0 0.4 0.2
5 93.3 93.8 93 93.8 94.1 1.3 1.4 1.8 1.6 0.5 1.3 0.5 0.2
6 94.2 94.5 92.3 94.1 94.2 2.2 2.1 1.1 1.9 0.6 1.6 0.7 0.3
7 93.6 95.1 93.2 96.1 95.7 1.6 2.7 2 3.9 2.1 2.5 0.9 0.4
8
Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD SEM
0 89.7 95.6 94.2 84.8 0 0 0 0 0.0
1 85.7 91.4 94.8 85.1 -4 -4.2 0.6 0.3 -2.5
2 88 93.7 93.6 84.6 -1.7 -1.9 -0.6 -0.2 -1.4
3 89.7 95.2 94.6 85.7 0 -0.4 0.4 0.9 0.0
4 89.8 93.1 94.3 85.4 0.1 -2.5 0.1 0.6 -0.8
5 89.7 94.5 93.8 87.3 0 -1.1 -0.4 2.5 -0.5
6 89.3 93.4 93.2 86.1 -0.4 -2.2 -1 1.3 -1.2
7 89.5 92.6 95.1 88.9 -0.2 -3 0.9 4.1 -0.8
GHT + BBI (12.5 mg/kg, p.o.)
GHT (12.5 mg/kg, p.o.)
Table. 9a. Oral administration of GHT fusion protein
with/without BBI in hypophysectomized rats.

163

Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD SEM
0 96.4 96.3 91.1 91.7 90.8 0 0 0 0 0 0.0
1 91.5 94.1 89.3 90.5 94 -4.9 -2.2 -1.8 -1.2 3.2 -1.7
2 92.1 94.5 89.6 91.4 92 -4.3 -1.8 -1.5 -0.3 1.2 -1.2
3 95.8 97.1 91.1 89.4 91.9 -0.6 0.8 0 -2.3 1.1 -0.5
4 93.2 96.4 89.5 90.1 92 -3.2 0.1 -1.6 -1.6 1.2 -1.0
5 92.2 96.7 90.9 90.4 94.5 -4.2 0.4 -0.2 -1.3 3.7 -0.4
6 93.3 95.5 90.1 90.2 93.3 -3.1 -0.8 -1 -1.5 2.5 -1.1
7 93.7 97.9 90.6 90.4 94.5 -2.7 1.6 -0.5 -1.3 3.7 -0.1
8
Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD SEM
0 87.9 88.5 91.6 92.0 0.0 0.0 0.0 0.0 0.0
1 88.4 88.5 90.5 91.9 0.5 0.0 -1.1 -0.1 -0.2
2 87.3 88.7 91.3 91.8 -0.6 0.2 -0.3 -0.2 -0.2
3 87.0 88.5 93.0 92.5 -0.9 0.0 1.4 0.5 0.3
4 87.4 89.2 91.7 93.2 -0.5 0.7 0.1 1.2 0.4
5 88.4 88.2 91.2 91.2 0.5 -0.3 -0.4 -0.8 -0.3
6 87.7 88.7 92.4 92.5 -0.2 0.2 0.8 0.5 0.3
7 87.5 88.5 90.2 92.9 -0.4 0.0 -1.4 0.9 -0.9
hGH + BBI (2.5 mg/kg, p.o.)
hGH (2.5 mg/kg, p.o.)
Table 9b. Oral administration of hGH with/without
BBI in hypophysectomized rats.

164

Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD SEM
0 92.2 90.3 93.3 95 0 0 0 0 0.0 0.0 0.00
1 97.6 95 96 98.1 5.4 4.7 2.7 3.1 4.0 1.3 0.64
2 100.2 95.2 97.3 98.8 8 4.9 4 3.8 5.2 1.9 0.97
3 100.5 95.5 100.2 101.7 8.3 5.2 6.9 6.7 6.8 1.3 0.63
4 100 98 102.7 101.7 7.8 7.7 9.4 6.7 7.9 1.1 0.56
5 101 99.6 103.3 103.5 8.8 9.6 10 8.5 9.2 0.7 0.35
6 104.5 98.9 106.2 105.3 12.3 8.9 12.9 10.3 11.1 1.8 0.92
7 105.7 100.6 106.5 107.2 13.5 10.3 13.2 12.2 12.3 1.4 0.72
8
Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD SEM
0 89 95.4 86.3 0 0 0 0.0
1 90.8 95.7 86 1.8 0.3 -0.3 0.6
2 89.6 95.6 86.8 0.6 0.2 0.5 0.4
3 90.1 97.5 85.4 1.1 2.1 -0.9 0.8
4 89.6 95.7 86.5 0.6 0.3 0.2 0.4
5 90.2 96.9 87.4 1.2 1.5 1.1 1.3
6 90.1 96.5 86.3 1.1 1.1 0 0.7
7 90.7 96.5 87.3 1.7 1.1 1 1.3
No treatment (daily weight measurement only)
GHT (1.25 mg/kg, s.c.)
Table 9c. Subcutaneous administration of GHT fusion
protein as positive control, along with no treatment
negative control in hypophysectomized rats.

165

14.14. High dose of GHT fusion protein coadministered with
BBI did not promote body weight gain

In previous experiment, GHT fusion protein when orally
coadministered at 12.5 mg/kg with BBI led to body weight gain at
modest level (~ 2.5 g weight gain). In this experiment, we want to
confirm such effect with escalated dose in presence of BBI. Therefore,
rats were given GHT fusion protein at 25 mg/kg orally together with
soybean trypsin inhibitor for 7-consecutive days. Data from 5-day
results showed that at high-dose coadministration of GHT with BBI did
not cause body weight gain (Fig. 57 and Table 10b). However,
coadministration of GHT at normal dose (12.5 mg/kg) resulted in small
increase in weight, which was not statistically significant (Fig. 57).

166

Fig. 57. Coadministration of BBI with GHT fusion protein at
high- dose did not promote weight gain. Rats were dosed for 7-
consecutive days with high doses of GHT, and change in weight
with respect to weight in 1st day of dosing was plotted for each
day. Oral administration of GHT (12.5 mg/kg) with BBI,
resulted in small, insignificant weight increase. However, at
high dose, the same GHT fusion protein failed to promote
weight gain.

167

Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD SEM
0 80.4 88.3 82.9 81.8 92.5 0 0 0 0 0 0.0 0.0 0.0
1 82.2 88.4 85.9 82.9 92.8 1.8 0.1 3 1.1 0.3 1.3 1.2 0.5
2 83.9 90 88 84.3 96 3.5 1.7 5.1 2.5 3.5 3.3 1.3 0.6
3 84.3 90.9 89.1 86.2 96.5 3.9 2.6 6.2 4.4 4 4.2 1.3 0.6
4 87.7 92.4 93 88.2 99.4 7.3 4.1 10.1 6.4 6.9 7.0 2.1 1.0
5 90.2 95.4 94.9 89.8 100.5 9.8 7.1 12 8 8 9.0 2.0 0.9
6 90.6 97 96.3 90.3 102.5 10.2 8.7 13.4 8.5 10 10.2 2.0 0.9
7 90.7 98.6 96.6 91.8 103.6 10.3 10.3 13.7 10 11.1 11.1 1.5 0.7
GHT(1.25 mg/kg, s.c.)
Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD SEM
0 79.3 88.9 86.1 88.5 74.9 0 0 0 0 0 0.0
1 76.2 83.9 82.8 87 72.5 -3.1 -5 -3.3 -1.5 -2.4 -3.1
2 78.7 85.7 83.3 89.1 74.5 -0.6 -3.2 -2.8 0.6 -0.4 -1.3
3 76.2 84.6 83.4 89.7 75.1 -3.1 -4.3 -2.7 1.2 0.2 -1.7
4 77.4 85.6 83.7 88.5 75.5 -1.9 -3.3 -2.4 0 0.6 -1.4
5 77.9 85.6 83 88 74.5 -1.4 -3.3 -3.1 -0.5 -0.4 -1.7
6 78.3 85.7 84.4 90.5 76.6 -1 -3.3 -1.7 2 1.7 -0.5
7 78 84.6 84.7 89.5 74.4 -1.3 -4.3 -1.4 1 -0.5 -1.3
No treatment (daily weight measurement only)
Table 10a. GHT fusion protein was administered for 7-days
subcutaneously, along with no treatment control.

168

Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD SEM
0 83.4 93.6 88.6 84.3 84.1 0 0 0 0 0 0.0
1 82.5 92.7 90.5 82.3 81.2 -0.9 -0.9 1.9 -2 -2.9 -0.5
2 82.6 92 89.9 80.8 81.8 -0.8 -1.6 1.3 -3.5 -2.3 -1.2
3 84.2 93.2 90.6 81.4 81.2 0.8 -0.4 2 -2.9 -2.9 -0.1
4 84.9 96.2 90.9 83.1 81.9 1.5 2.6 2.3 -1.2 -2.2 1.3
5 86 94.1 90.3 84.7 82.8 2.6 0.5 1.7 0.4 -1.3 1.3
6 84.3 94.4 91.5 85.1 83.1 0.9 0.8 2.9 0.8 -1 1.4
7 85.5 94.6 91.7 83.1 82.5 2.1 1 3.1 -1.2 -1.6 1.3
8
Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD SEM
0 87.2 84.4 82.7 86.4 81.3 0 0 0 0 0 0.0
1 86 84 81.6 84.8 80.9 -1.2 -0.4 -1.1 -1.6 -0.4 -0.8
2 86.1 82.9 81.1 83.9 79.5 -1.1 -1.5 -1.6 -2.5 -1.8 -1.5
3 86.7 83.7 80.6 84.6 81.7 -0.5 -0.7 -2.1 -1.8 0.4 -0.7
4 87.4 84.9 82.2 84.9 81.8 0.2 0.5 -0.5 -1.5 0.5 0.2
5 88.4 85 83.4 85.4 82 1.2 0.6 0.7 -1 0.7 0.8
6 86.4 84.2 84.9 86.3 81.9 -0.8 -0.2 2.2 -0.1 0.6 0.5
7 87.3 86.7 84.7 86.7 82 0.1 2.3 2 0.3 0.7 1.3
GHT + BBI (12.5 mg/kg, p.o.)
GHT + BBI (25 mg/kg, p.o.)
Table 10b. High doses of GHT fusion protein coadministered with BBI
orally.

169

Day Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD SEM
0 91.8 83 84.5 86.5 88.6 0 0 0 0 0 0
1 90.9 81.7 83.9 84.3 84.8 -0.9 -1.3 -0.6 -2.2 -3.8 -1.3
2 89.4 80.7 83.1 84.6 86.9 -2.4 -2.3 -1.4 -1.9 -1.7 -2.0
3 91.5 82.9 83.5 86.4 87.1 -0.3 -0.1 -1 -0.1 -1.5 -0.4
4 91.3 83.4 85.3 86 89.1 -0.5 0.4 0.8 -0.5 0.5 0.1
5 91.3 84.8 86.2 85.9 89.1 -0.5 1.8 1.7 -0.6 0.5 0.6
6 90.7 83.7 85.4 87.9 89.3 -1.1 0.7 0.9 1.4 0.7 0.5
7 93.8 84.5 85.3 87.6 90 2 1.5 0.8 1.1 1.4 1.1
hGH + BBI (5 mg/kg, p.o.)
Rat-1 Rat-2 Rat-3 Rat-4 Rat-5 Chge-1 Chge-2 Chge-3 Chge-4 Chge-5 Ave. SD SEM
0 83.5 91.9 86.6 85.6 81 0 0 0 0 0 0.0 0.0
0.0
1 83 89.5 86.6 86 82.6 -0.5 -2.4 0 0.4 1.6 -0.2 1.5
0.7
2 84.5 91.2 91.2 88.8 86.7 1 -0.7 4.6 3.2 5.7 2.8 2.6
1.2
3 88.1 91.5 91.1 90.3 88.9 4.6 -0.4 4.5 4.7 7.9 4.3 3.0
1.3
4 89.5 94.2 93.3 94.5 93.1 6 2.3 6.7 8.9 12.1 7.2 3.6
1.6
5 93.5 97.1 97.1 96.7 95.8 10 5.2 10.5 11.1 14.8 10.3 3.4
1.5
6 94.9 100 98.5 98.8 99 11.4 8.1 11.9 13.2 18 12.5 3.6
1.6
7 94.3 100.4 99.5 99.7 98.7 10.8 8.5 12.9 14.1 17.7 12.8 3.5
1.6
hGH (0.25 mg/kg, s.c.)
Table 10c. High doses of hGH coadministered with BBI orally for 7-
days in hypophysectomized rats, along with subcutaneous hGH as
positive control.

170
Chapter 15. Discussions

We used Tf-based fusion protein approach to achieve oral absorption
of hGH through a Tf-TfR dependent transcytotic pathway. For fusion
protein to demonstrate oral biological activity via effective absorption
across intestinal cells, first it has to bind to TfR distributed on the GI,
followed by receptor mediated transcytosis and subsequent transport
to the systemic circulation. But fusion protein has to survive from the
acidic environment in stomach and be available at the absorption site
in the small intestine. We administered fusion protein directly to the
stomach of the rats to demonstrate oral activity of the fusion protein.

Our data showed (Fig. 45) small weight increase in the group of rats
receiving the fusion protein but not in groups receiving hGH or buffer.
However, we noted large variations in weight gain in rats in the same
group. For example, one rat did not gain as much weight as other rats
in the same group that were orally dosed with fusion protein. This
difference in weight gain following the oral dosing could be attributable
to the differences in fasting, eating habits, gastric empting behavior of
individual rats and thus altered absorption in the GI. Furthermore, we

171
used the average weight (100 g) for dose calculation, while the actual
weight was ranged from 90 - 110 g, which could contribute to some
variation in oral response. Importantly, the small intestine is the main
absorption site where the fusion protein has a chance to transport
across the epithelial cells in its journey to the systemic circulation, but
it also presents the greatest challenge to the stability of the fusion
protein owing to the existence of proteolytic enzymes like
chymotrypsin and trypsin that break down and hence render fusion
protein inactive. Therefore, it is conceivable to think that the variation
in oral response could be the result of difference in local proteoliytic
enzyme concentration in individual rats (25, 30, 33).

In deed our result from in vitro chymotrypsin/trypsin cleavage of
fusion protein validated our assumptions by revealing the extent of
fusion protein destruction that began within 15 minutes of incubation.
Next, we reasoned that oral co-administration of Bowman Birk
chymotrypsin Inhibitor (BBI) should protect the fusion protein from
tryptic/chymotryptic degradation, which then could lead to a reduced
variation in response and improved oral efficacy. As predicted, the co-
administration of BBI with the fusion protein orally in fact caused a
growth promoting activity with improved consistency and this effect

172
was significantly different from the group that received BBI+hGH as
control (Fig. 56).

GHT fusion protein proved to be superior to other fusion proteins that
include GT and THG in terms of expression level and bioactivity. This
finding indirectly suggests the importance of placing hGH domain at
amino terminus of fusion polypeptide and the helical linker insertion to
develop fusion protein with improved hGH bioactivity and increased
expression. When administered at high dose with or without soybean
trypsin inhibitor, the GHT fusion protein was not effective. This non-
responsiveness by large dose administration could be due to TfR
saturation. Even though GI epithelium express high numbers of TfR, it
is still possible that TfR could be overwhelmed by large amounts of Tf
in the fusion protein, which account for 80 % of the dose (50 mg/kg or
25 mg/kg +BBI). As a result, transcytosis of fusion protein likely to be
reduced, causing low to no bioactivity.

The ineffectiveness of soybean trypsin inhibitor when coadministered
with elevated dose (25 mg/kg) of GHT fusion protein further support
the assumption of TfR saturation in presence of large quantity of
transferrin. Because to know the effect of soybean trypsin inhibitor on

173
oral absorption, fusion protein has to transcytose across the intestinal
cells via Tf-TfR pathway and this process requires effective TfR binding
by the fusion protein. Therefore, TfR saturation by excess Tf in the
fusion protein likely to limit the effectiveness of fusion protein
transport across the intestinal cell layers, thus leading to low or no
hGH bioactivity.

Tf as a fusion partner offers two advantages including the mechanism
for oral absorption as well as the longer plasma half-life with improved
PK. Tf-based fusion protein technology therefore has the potential to
become an approach for the development of oral protein drugs (4, 5,
31, 64, 66, 70). Our study supported the applicability of Tf-fusion
protein approach by engineering, producing hGH-Tf fusion protein with
dual biological activity in mammalian cells, as well as demonstrating
oral growth promoting activity in the animal model of hGH. This study
can easily be applicable to other protein drug candidates such as
Insulin, IFN-α2b and EPO for potential oral delivery.

Tf-fusion protein technology in combination with appropriate
formulations has a potential to become a platform for oral dosage
formulation for protein drugs – which benefit millions of patients

174
worldwide by offering convenience, accessibility and improved
compliance. Formulations including enteric coding, soybean trypsin
inhibitor, microspheres and nanoparticles (9-11, 26) with controlled
release at absorption site should prove helpful in this endeavor of oral
protein delivery with increased efficiency.

175
Chapter 16. Summary/Conclusion/Future perspective

16.1. Summary

1. Constructed 7 expression vectors, six for fusion protein expression
and one for hGH expression in mammalian vector, pcDNA3.1. All 7
constructs expressed in Human Embryonic Kidney cells (HEK293) at
high level and purity, only the Tf-hGH fusion protein with cyclic linker
(TCG) expressed at a low level.

2. SDS-PAGE with Coomassie or Silver stain, as well as Western Blot
with anti-hGH and anti-Tf antibodies, identified and confirmed the
fusion protein expression in serum free CD293 media.

3. TfR competition binding in CaCo-2 cells in presence of radioactively
labeled Tf (I
125
-Tf) by four fusion proteins, including GT, TG, GHT, and
THG, demonstrated that they maintained the biological activity of Tf.
But the affinity of binding is much less than that of native Tf. In
addition, both TG and THG fusion proteins had higher binding affinity
than GT and GHT fusion proteins, in which Tf is at C-terminus.

176
4. Both GT and TG fusion proteins demonstrated capacity to bind hGH-
R and to stimulate signal transduction via the activation of JAK2 and
STAT5 in human B-lymphoma cells (IM-9), indicating that these two
fusion proteins retained bioactivity of hGH.

5. Three fusion proteins including GT, GHT and THG were evaluated for
hGH-bioactivity in Nb2 cell proliferation assay, and results
demonstrated that these fusion proteins led to Nb2 cell proliferation,
thereby confirming the retention of hGH specific bioactivity.

6. Established an hGH radioimmunoassay (hGH-RIA) to select stable
clones that express fusion proteins and developed four stable cell lines
in HEK293T cells, expressing hGH, GT, GHT and THG fusion proteins.

7. Demonstrated that the insertion of helical linker between hGH and
Tf domain enhanced both expression and biological activity of the
fusion proteins, including both GHT and THG.

8. Developed a large-scale production scheme for fusion proteins by
scaled- up transcient transfection (up to 40 T225 flasks) and
harvesting conditioned media frequently (up to 3 times). The Labscale

177
TFF system proved very effective; it not only expedited the
concentration and buffer exchange of large quantities of fusion
proteins but also enhanced the quality of such protein products.

9. Proteolytic stability assay with GHT, THG and GT fusion proteins
showed that they were very sensitive to trypsin and chymotrypsin
cleavage in a reaction resembling physiological conditions in the GI
tract. However, Tf showed unusually strong resistance to both trypsin
and chymotrypsin cleavage.

10. Co-incubation of soybean trypsin inhibitor (BBI) with either GHT or
THG fusion protein blocked both tryptic and chymotryptic activity,
indicating BBI could be effective in increasing oral absorption.

11. All three-fusion proteins that include GT, GHT and THG were
effective in promoting weight gain in hGH-deficient rats
(hypophysectomized) with subcutaneous administration. But only GHT
fusion protein led to a small weight increase when administered orally.

12. Oral dose-escalation studies with GHT fusion protein did not
produce weight increase in hypophysectomized rats.

178

13. Oral co-administration of GHT fusion protein with BBI in
hypophysecstomized rats led to a small body weight gain with
improved consistency as compared to that of without BBI co-
administration.

179
16.2. Conclusion

We demonstrated the feasibility of producing Tf-based fusion protein
with hGH in mammalian cells that showed dual biological activity as
confirmed in cell culture and animal model for hGH. The Insertion of 2
copies of helical linker had positive effect on both expression and
biological activity and could be used as an approach to optimize other
Tf-based fusion proteins. Transcient transfection of HEK293T cells at
large-scale proved to be effective approach in producing several
hundred mg of Tf-based fusion protein with hGH in short period of
time. Oral administration of GHT fusion protein containing the helical
linker resulted in a small weight increase in hypophysectomized rats,
and co-administration of soybean trypsin inhibitor with GHT may
improve consistency of oral absorption. Our results confirmed the
feasibility of Tf-based approach on oral delivery of hGH. However, to
be an effective oral delivery approach, the proposed protein
moiety/domain for Tf-based fusion protein need to be studied
thoroughly with respect to its target tissues, stability, receptor binding
and subsequent processing.

180
16.3. Future perspectives

The low response and large variations in oral absorption warrants
further investigation on oral bioavailability of hGH-Tf fusion proteins in
healthy rats. Such study involves oral dosing of radioactively labeled
fusion protein in healthy rats and comparing oral bioavailability to
relative bioavailability after subcutaneous administration of the same
fusion protein with radioactivity. If oral bioavailability found to be low,
then it would be helpful to study biodistribution after oral
administration of radioactive fusion protein. This study requires organ
and tissue harvesting at predetermined time points, homogenizing,
and counting radioactivity from tissue extracts, and results will provide
information on whether fusion protein localizes only in liver or it
travels to key target tissues such as muscle and bone after oral
absorption.

One alternative, more advanced approach in study of fusion protein
biodistribution is to use Positron Emission Tomography (PET) combined
with Computerized Tomography (CT). PET/CT scan can trace and
monitor radioactively labeled fusion protein for tissue localization and
distribution in whole rat, providing relatively complete information as

181
to where the fusion protein travels and localizes. Results from this
study help assess the extent of fusion protein elimination and help
explain oral bioavailability data, which could potentially lead to
improved approaches for oral delivery of Tf-fusion protein.

Finally, the use of cleavable, cyclic linker could be a viable option to
improve oral bioavailability of fusion protein with hGH and Tf, because
this approach enables hGH to be separated from Tf during or after
absorption across GI epithelium, and likely to reduce elimination but
increase distribution of hGH to main target tissues other than the liver.

182
Bibliography

1. Arai, R., H. Ueda, A. Kitayama, N. Kamiya, and T. Nagamune.
2001. Design of the linkers which effectively separate domains of
a bifunctional fusion protein. Protein Engineering 14:529-532.

2. Arai, R., W. Wriggers, Y. Nishikawa, T. Nagamune, and T.
Fujisawa. 2004. Conformations of variably linked chimeric
proteins evaluated by synchrotron X-ray small-angle scattering.
Proteins Structure Function and Bioinformatics 57:829-838.

3. Argetsinger, L. S., G. S. Campbell, X. Yang, B. A. Witthuhn, O.
Silvennoinen, J. N. Ihle, and C. Carter-Su. 1993. Identification of
JAK2 as a growth hormone receptor-associated tyrosine kinase.
Cell 74:237-244.

4. Bai, Y., D. K. Ann, and W.-C. Shen. 2005. Recombinant
granulocyte colony-stimulating factor-transferrin fusion protein
as an oral myelopoietic agent. Proceedings of the National
Academy of Sciences 102:7292-7296.

5. Bai, Y., and W. C. Shen. 2006. Improving the oral efficacy of
recombinant granulocyte colony-stimulating factor and
transferrin fusion protein by spacer optimization. Pharmaceutical
Research 23:2116-2121.

6. Baldi, L., D. L. Hacker, M. Adam, and F. M. Wurm. 2007.
Recombinant protein production by large-scale transient gene
expression in mammalian cells: state of the art and future
perspectives. Biotechnol Lett 29:677-684.

7. Banerjee, D., P. R. Flanagan, J. Cluett, and L. S. Valberg. 1986.
Transferrin receptors in the human gastrointestinal tract.
Relationship to body iron stores. Gastroenterology 91:861-869.

8. Bellini, M. H. 1993. In vivo bioassay for the potency
determination of human growth hormone in dwarf" little" mice.
Endocrinology 132:2051-2055.

183

9. Bernkop-Schnürch, A. 1998. The use of inhibitory agents to
overcome the enzymatic barrier to perorally administered
therapeutic peptides and proteins. Journal of Controlled Release
52:1-16.

10. Bosquillon, C., V. Préat, and R. Vanbever. 2004. Pulmonary
delivery of growth hormone using dry powders and visualization
of its local fate in rats. Journal of Controlled Release 96:233-
244.

11. Carino, G. P., J. S. Jacob, and E. Mathiowitz. 2000. Nanosphere
based oral insulin delivery. Journal of Controlled Release 65:261-
269.

12. Cheng, Y., O. Zak, P. Aisen, S. C. Harrison, and T. Walz. 2004.
Structure of the Human Transferrin Receptor-Transferrin
Complex. Cell 116:565-576.

13. Choi, H. K., and D. J. Waxman. 2000. Plasma Growth Hormone
Pulse Activation of Hepatic JAK-STAT5 Signaling: Developmental
Regulation and Role in Male-Specific Liver Gene Expression.
Endocrinology 141:3245-3255.

14. Clark, R. 1997. The Somatogenic Hormones and Insulin-Like
Growth Factor-1: Stimulators of Lymphopoiesis and Immune
Function. Endocrine Reviews 18:157-179.

15. Clark, R. G. 1996. Recombinant human growth hormone (GH)-
binding protein enhances the growth-promoting activity of
human GH in the rat. Endocrinology 137:4308-4315.

16. Cox, G. N., M. S. Rosendahl, E. A. Chlipala, D. J. Smith, S. J.
Carlson, and D. H. Doherty. 2007. A Long-Acting, Mono-
PEGylated Human Growth Hormone Analog Is a Potent
Stimulator of Weight Gain and Bone Growth in
Hypophysectomized Rats. Endocrinology 148:1590.

184
17. Crowell, C. K., Q. Qin, G. E. Grampp, R. A. Radcliffe, G. N.
Rogers, and R. I. Scheinman. 2008. Sodium butyrate alters
erythropoietin glycosylation via multiple mechanisms. Biotechnol
Bioeng 99:201-213.

18. Cunningham, B. C., M. Ultsch, A. M. De Vos, M. G. Mulkerrin, K.
R. Clauser, and J. A. Wells. 1991. Dimerization of the
extracellular domain of the human growth hormone receptor by
a single hormone molecule. Science 254:821-825.

19. Davis, S. S. 2005. Formulation strategies for absorption
windows. Drug Discovery Today 10:249-257.

20. DeNoto, F. M., D. D. Moore, and H. M. Goodman. 1981. Human
growth hormone DNA sequence and mRNA structure: possible
alternative splicing. Nucleic Acids Res 9:3719-3730.

21. Duttaroy, A., P. Kanakaraj, B. L. Osborn, H. Schneider, O. K.
Pickeral, C. Chen, G. Zhang, S. Kaithamana, M. Singh, and R.
Schulingkamp. 2005. Development of a long-acting insulin
analog using albumin fusion technology. Diabetes 54:251-258.

22. Friden, P. M., T. S. Olson, R. Obar, L. R. Walus, and S. D.
Putney. 1996. Characterization, receptor mapping and blood-
brain barrier transcytosis of antibodies to the human transferrin
receptor. J Pharmacol Exp Ther 278:1491-1498.

23. Goeddel, D. V., H. L. Heyneker, T. Hozumi, R. Arentzen, K.
Itakura, D. G. Yansura, M. J. Ross, G. Miozzari, R. Crea, and P.
H. Seeburg. 1979. Direct expression in Escherichia coli of a DNA
sequence coding for human growth hormone. Nature 281:544-
548.

24. Gomme, P. T., K. B. McCann, and J. Bertolini. 2005. Transferrin:
structure, function and potential therapeutic actions. Drug
Discovery Today 10:267-273.

25. Hamman, J. H., G. M. Enslin, and A. F. Kotzé. 2005. Oral
Delivery of Peptide Drugs: Barriers and Developments. BioDrugs
19:165.

185
26. Hussain, A., J. J. Arnold, M. A. Khan, and F. Ahsan. 2004.
Absorption enhancers in pulmonary protein delivery. Journal of
Controlled Release 94:15-24.

27. Kavoosi, M., A. L. Creagh, D. G. Kilburn, and C. A. Haynes.
2007. Strategy for selecting and characterizing linker peptides
for CBM9-tagged fusion proteins expressed in Escherichia coli.
Biotechnology and Bioengineering 98:599-610.

28. Lawrence, C. M., S. Ray, M. Babyonyshev, R. Galluser, D. W.
Borhani, and S. C. Harrison. 1999. Crystal Structure of the
Ectodomain of Human Transferrin Receptor. Science 286:779.

29. Leader, B., Q. J. Baca, and D. E. Golan. 2008. Opinion: Protein
therapeutics: a summary and pharmacological classification.
Nature Reviews Drug Discovery 7:21-39.

30. Lee, V. H., A. Yamamoto, and U. B. Kompella. 1991. Mucosal
penetration enhancers for facilitation of peptide and protein drug
absorption. Crit Rev Ther Drug Carrier Syst 8:91-192.

31. Lim, C. J., and W. C. Shen. 2004. Transferrin-Oligomers as
Potential Carriers in Anticancer Drug Delivery. Pharmaceutical
Research 21:1985-1992.

32. Liu, R., J. Q. Guan, O. Zak, P. Aisen, and M. R. Chance. 2003.
Structural reorganization of the transferrin C-lobe and transferrin
receptor upon complex formation: the C-lobe binds to the
receptor helical domain. Biochemistry 42:12447-12454.

33. Low, S. C., S. L. Nunes, A. J. Bitonti, and J. A. Dumont. 2005.
Oral and pulmonary delivery of FSH-Fc fusion proteins via
neonatal Fc receptor-mediated transcytosis. Human
Reproduction 20:1805-1813.

34. Maeda, Y., H. Ueda, J. Kazami, G. Kawano, E. Suzuki, and T.
Nagamune. 1997. Engineering of Functional Chimeric Protein G–
VargulaLuciferase. Analytical Biochemistry 249:147-152.

186
35. Marqusee, S., and R. L. Baldwin. 1987. Helix Stabilization by
Glu-\ cdots Lys+ Salt Bridges in Short Peptides of De novo
Design. Proceedings of the National Academy of Sciences of the
United States of America 84:8898-8902.

36. Ng, P. P., D. Cruz, S. Jay, D. N. Sorour, J. M. Stinebaugh, S. U.
Shin, D. S. Shin, S. L. Morrison, and M. L. Penichet. 2002. An
anti-transferrin receptor-avidin fusion protein exhibits both
strong proapoptotic activity and the ability to deliver various
molecules into cancer cells. Proceedings of the National Academy
of Sciences 99:10706.

37. Park, E., R. M. Starzyk, J. P. McGrath, T. Lee, J. George, A. J.
Schutz, P. Lynch, and S. D. Putney. 1998. Production and
characterization of fusion proteins containing transferrin and
nerve growth factor. J Drug Target 6:53-64.

38. Pavlou, A. K., and J. M. Reichert. 2004. Recombinant protein
therapeutics—success rates, market trends and values to 2010.
Nature Biotechnology 22:1513-1519.

39. Pearce Jr, K. H., B. C. Cunningham, G. Fuh, T. Teeri, and J. A.
Wells. 1999. Growth hormone binding affinity for its receptor
surpasses the requirements for cellular activity. Biochemistry
38:81-89.

40. Pham, P. L., A. Kamen, and Y. Durocher. 2006. Large-scale
transfection of mammalian cells for the fast production of
recombinant protein. Mol Biotechnol 34:225-237.

41. Pham, P. L., S. Perret, B. Cass, E. Carpentier, G. St-Laurent, L.
Bisson, A. Kamen, and Y. Durocher. 2005. Transient gene
expression in HEK293 cells: Peptone addition posttransfection
improves recombinant protein synthesis. Biotechnology and
Bioengineering 90:332-344.

42. Pryor, K. D., and B. Leiting. 1997. High-Level Expression of
Soluble Protein inEscherichia coliUsing a His6-Tag and Maltose-
Binding-Protein Double-Affinity Fusion System. Protein
Expression and Purification 10:309-319.

187
43. Qian, Z. M., H. Li, H. Sun, and K. Ho. 2002. Targeted Drug
Delivery via the Transferrin Receptor-Mediated Endocytosis
Pathway. Pharmacological Reviews 54:561.

44. Richardson, D. R. 2004. Mysteries of the Transferrin-Transferrin
Receptor 1 Interaction Uncovered. Cell 116:483-485.

45. Robinson, C. R., and R. T. Sauer. 1998. Optimizing the stability
of single-chain proteins by linker length and composition
mutagenesis. Proceedings of the National Academy of Sciences
95:5929-5934.

46. Roskam, W. G., and F. Rougeon. 1979. Molecular cloning and
nucleotide sequence of the human growth hormone structural
gene. Nucleic Acids Research 7:305.
47. Silva, C. M. 1993. Stimulation of tyrosine phosphorylation in
human cells by activation of the growth hormone receptor.
Endocrinology 132:101-108.

48. Staub, J. M., B. Garcia, J. Graves, P. T. J. Hajdukiewicz, P.
Hunter, N. Nehra, V. Paradkar, M. Schlittler, J. A. Carroll, L.
Spatola, D. Ward, G. N. Ye, and D. A. Russell. 2000. High-yield
production of a human therapeutic protein in tobacco
chloroplasts. Nature Biotechnology 18:333-338.

49. Subramanian, G. M., M. Fiscella, A. Lamousé-Smith, S. Zeuzem,
and J. G. McHutchison. 2007. Albinterferon alpha-2b: a genetic
fusion protein for the treatment of chronic hepatitis C. Nature
Biotechnology 25:1411-1419.

50. Sun, X., P. E. Goh, K. T. K. Wong, T. Mori, and M. G. S. Yap.
2006. Enhancement of transient gene expression by fed-batch
culture of HEK 293 EBNA1 cells in suspension. Biotechnol Lett
28:843-848.

51. Takahashi, Y., H. Kaji, Y. Okimura, K. Goji, H. Abe, and K.
Chihara. 1996. Short Stature Caused by a Mutant Growth
Hormone. New England Journal of Medicine 334:432.

188
52. Takahashi, Y., H. Shirono, O. Arisaka, K. Takahashi, T. Yagi, J.
Koga, H. Kaji, Y. Okimura, H. Abe, and T. Tanaka. 1997.
Biologically Inactive Growth Hormone Caused by an Amino Acid
Substitution. Journal of Clinical Investigation 100:1159-1165.

53. Tang, L., A. M. Persky, G. Hochhaus, and B. Meibohm. 2004.
Pharmacokinetic aspects of biotechnology products. J Pharm Sci
93:2184-2204.

54. Thomas, M. J. 1998. The molecular basis of growthhormone
action. Growth Hormone & IGF Research 8:3-11.

55. Towns, R. 1992. Evidence that the N-terminus of human growth
hormone is involved in expression of its growth promoting,
diabetogenic, and insulin-like activities. Endocrinology 130:1225-
1230.

56. Walsh, G. 2006. Biopharmaceutical benchmarks 2006. Nature
Biotechnology 24:769-776.
57. Walsh, G., and R. Jefferis. 2006. Post-translational modifications
in the context of therapeutic proteins. Nat Biotechnol 24:1241-
1252.

58. Wang, W. 1999. Instability, stabilization, and formulation of
liquid protein pharmaceuticals. International Journal of
Pharmaceutics 185:129-188.

59. Wang, W. 2000. Lyophilization and development of solid protein
pharmaceuticals. International Journal of Pharmaceutics 203:1-
60.

60. Wang, W. 2005. Protein aggregation and its inhibition in
biopharmaceutics. International Journal of Pharmaceutics 289:1-
30.

61. Wang, W., K. Antonsen, Y. J. Wang, and D. Q. Wang. 2008. pH
dependent effect of glycosylation on protein stability. European
Journal of Pharmaceutical Sciences 33:120-127.

189
62. Waters, M. J., H. N. Hoang, D. P. Fairlie, R. A. Pelekanos, and R.
J. Brown. 2006. New insights into growth hormone action.
Journal of Molecular Endocrinology 36:1-7.

63. Wells, J. A. 1996. Binding in the growth hormone receptor
complex. Proceedings of the National Academy of Sciences of the
United States of America 93:1.

64. Widera, A., Y. Bai, and W. C. Shen. 2004. The Transepithelial
Transport of a G-CSF-Transferrin Conjugate in Caco-2 Cells and
Its Myelopoietic Effect in BDF1 Mice. Pharmaceutical Research
21:278-284.

65. Widera, A., K. J. J. Kim, E. D. Crandall, and W. C. Shen. 2003.
Transcytosis of GCSF-Transferrin Across Rat Alveolar Epithelial
Cell Monolayers. Pharmaceutical Research 20:1231-1238.

66. Widera, A., F. Norouziyan, and W. C. Shen. 2003. Mechanisms of
TfR-mediated transcytosis and sorting in epithelial cells and
applications toward drug delivery. Advanced Drug Delivery
Reviews 55:1439-1466.

67. Wriggers, W., and S. Chakravarty. 2005. Control of Protein
Functional Dynamics by Peptide Linkers. Biopolymers (Peptide
Science) 80:736-746.

68. Wurm, F., and A. Bernard. 1999. Large-scale transient
expression in mammalian cells for recombinant protein
production. Current Opinion in Biotechnology 10:156-159.

69. Wurm, F. M. 2004. Production of recombinant protein
therapeutics in cultivated mammalian cells. Nature
Biotechnology 22:1393-1398.

70. Xia, C. Q., J. Wang, and W. C. Shen. 2000. Hypoglycemic Effect
of Insulin-Transferrin Conjugate in Streptozotocin-Induced
Diabetic Rats. Journal of Pharmacology and Experimental
Therapeutics 295:594.

190
71. Yang, F., J. B. Lum, J. R. McGill, C. M. Moore, S. L. Naylor, P. H.
van Bragt, W. D. Baldwin, and B. H. Bowman. 1984. Human
transferrin: cDNA characterization and chromosomal localization.
Proceedings of the National Academy of Sciences of the United
States of America 81:2752.

Abstract (if available)

Abstract Protein drug delivery is limited to needle injection associating with pain, inconvenience, and non- compliance. Therefore, protein drug with oral dosage form is preferred. We used transferrin (Tf) receptor mediated transcytosis and Tf-based recombinant fusion protein approach to achieve oral delivery of human growth hormone (hGH). Plasmid constructs expressing the fusion proteins were established by fusing coding sequences of both hGH and Tf in frame. Fusion proteins were produced in serum free media by transcient transfection of human embryonic kidney cells. The SDS-PAGE of conditioned media showed that fusion proteins expressed at ~ 90% abundance and 100 kDa molecular weight

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Developing recombinant single chain Fc-dimer fusion proteins for improved protein drug delivery

PDF

Characterization of rice-derived proinsulin-transferrin fusion protein and its oral bioactivity as a basal insulin analogue

PDF

Proinsulin-transferrin recombinant fusion protein: mechanism of activation and potential application in diabetes treatment

PDF

Characterization of the transferrin oligomer and its potential application in drug delivery

PDF

The intracellular trafficking of transferrin in Caco-2 cells and its implication to transferrin-mediated oral drug delivery

PDF

Proinsulin-transferrin fusion protein as a liver targeting insulin therapeutic in Type 1 diabetes treatment

PDF

Sustained and targeted delivery of rapamycin using FKBP-elastin like polypeptide fusion proteins

PDF

Cell penetrating peptide-based drug delivery system for targeting mildly acidic pH

PDF

Characterization of actin based motility in mammalian cells through LIM and SH3 domain protein 1 (LASP1) and elastin like polypeptide (ELP) fusion protein

PDF

Cell penetrating peptide-based polyplexes for sirRNA delivery

Asset Metadata

Creator Amet, Nurmamet (author)

Core Title Human growth hormone-transferrin recombinant fusion protein for oral delivery

School School of Pharmacy

Degree Doctor of Philosophy

Degree Program Pharmaceutical Sciences

Publication Date 08/06/2008

Defense Date 05/23/2008

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag fusion protein,helical linker,human growth hormone,OAI-PMH Harvest,oral delivery,protein drug,transferrin

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Shen, Wei-Chiang (committee chair), Haworth, Ian S. (committee member), Kalra, Vijay K. (committee member)

Creator Email amet@usc.edu,nurmamet@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1548

Unique identifier UC1125080

Identifier etd-Amet-2316 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-109040 (legacy record id),usctheses-m1548 (legacy record id)

Legacy Identifier etd-Amet-2316.pdf

Dmrecord 109040

Document Type Dissertation

Rights Amet, Nurmamet

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu