University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

PROINSULIN-TRANSFERRIN RECOMBINANT FUSION PROTEIN: MECHANISM
OF ACTIVATION AND POTENTIAL APPLICATION IN DIABETES TREATMENT

by

Yan Wang

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(MOLECULAR PHARMACOLOGY AND TOXICOLOGY)

December 2012

Copyright 2012 Yan Wang

ii
DEDICATION

To my parents, Junpi Wang and Guomin Zhao,
thank you for your love and support.

iii
ACKNOWLEDGEMENTS

First and the most, I would like to express my sincere appreciation to Dr. Wei-
Chiang Shen, the best advisor and mentor that I could have wished for. His wisdom and
experience, his generosity and kindness, have been invaluable to my scientific and
personal growth for the past four and a half years. I am grateful for the research ideas he
imparted to me, for the time he spent in discussing about my project, for the flexibility he
provided that allowed me to try new stuffs, and for the support he gave to my career
development. Mrs. Daisy Shen has been a mother figure for students in the laboratory. I
am very thankful to her loving care and warm thoughtfulness that filled my loneliness
while studying abroad. It is my great fortune to be under Dr. Shen and Daisy's care over
the years.
Dr. Jennica Zaro has given me tremendous help and support to my research and
career development. I appreciate her advice and feedback on the research work, and her
suggestions on the manuscript revision.
I want to acknowledge my committee members including Dr. Ian Haworth, Dr.
Ralf Langen, Dr. Bangyan Stiles, and Dr. Clay Wang, for their time and advice regarding
my dissertation work since the qualifying exam. I am appreciative to Dr. Roberta Brinton
for her guidance and support at USC.
Furthermore, I am grateful for the friendship and daily-based assistance from my
labmates. I appreciate Yu-Sheng Chen (Sam) for his contribution to the fusion protein
purification and animal experiments. I am thankful to Xiaoying Chen and Hsin-Fang Lee
iv
for sharing their experiences of protein production when I started this project. I also
really enjoyed and appreciated the accompany from Sam and Chunmeng Sun for staying
late in the lab and giving me rides to the parking lot.
In addition to my labmates, I would also like to thank other faculty, students and
staff that I got to know at USC. Leana He, Ni Zeng, and Vivian Medina, for their
generousness in sharing their protocols and techniques. Liqin Zhao, Jennifer Mao, and
Shuhua Chen, for their kindness during the past years. Qing Cheng from Karolinska
Institutet, for his great help with literature searching and long life friendship.
Despite the geographical distance, my parents have always been nearby. They
have been a constant source of love for me. I could not have accomplished any of these
without their unreserved love. I especially appreciate them for enlightening me with their
life experiences, for understanding my passion, for encouraging me to pursue my dream
of becoming a Ph.D., and for supporting me when I was lost in the darkness.
Last, Robert Mo, my best friend and my loving fiancé. I am grateful for his
unconditional love and assistance with my dissertation work. I feel like this thesis project
belongs to both of us. He has always been there for me, every step of success or failure,
every piece of happiness or sadness. I am delighted that we have shared these wonderful
memories and will carry them on to the future.

v
TABLE OF CONTENTS

DEDICATION ii

ACKNOWLEDGEMENTS iii

LIST OF TABLES ix

LIST OF FIGURES x

LIST OF SCHEMES xiii

ABBREVIATIONS xiv

ABSTRACT xvii

CHAPTER 1: INTRODUCTION
1.1. Protein Therapeutics
1.1.1. Applications and Advantages of Protein Therapeutics 1
1.1.2. Challenges for Protein Therapeutics 2
1.1.3. Current Advances for Novel Protein Therapeutics 3
1.2. Insulin as a Protein Therapeutic
1.2.1. Diabetes 4
1.2.2. Human Insulin
1.2.2.1. Insulin Structure and Physiological Actions 5
1.2.2.2. Insulin Secretion and Insulin Therapy 6
1.2.2.3. Therapeutic Insulin Analogues in Current Use 7
1.2.3. Human Proinsulin 8
1.3. Transferrin
1.3.1. Physiological Properties and Functions 10
1.3.2. Receptor-mediated Endocytosis and Recycling Pathway 11
1.3.3. Therapeutic Application 12
1.4. Bifunctional Fusion Proteins
1.4.1. Fc-based Fusion Proteins 13
1.4.2. Albumin-based Fusion Proteins 14
1.4.3. Transferrin-based Fusion Proteins 14
vi
CHAPTER 2: DESIGN AND PRODUCTION OF PROINSULIN-
TRANSFERRIN RECOMBINANT FUSION PROTEIN
2.1. Background 16
2.2. Materials and Methods
2.2.1. Cell Cultures 17
2.2.2. Plasmid Construction 17
2.2.3. Fusion Protein Production 19
2.2.4. Fusion Protein Purification
2.2.4.1. His-tag Purification 20
2.2.4.2. Gel Filtration Chromatography 20
2.2.5. Fusion Protein Analysis
2.2.5.1. SDS-PAGE 21
2.2.5.2. Western Blot 21
2.2.5.3. Coomassie Blue Staining 22
2.2.5.4. Bicinchoninic Acid Protein Assay 22
2.3. Results
2.3.1. Design of Fusion Protein 22
2.3.2. Small-scale and Large-scale Production of Fusion Protein 23
2.3.3. Formulation Development of Fusion Protein for Long-term
Storage 26
2.3.4. Improvement of Fusion Protein Purity 27
2.3.5. Enhancement of Fusion Protein Productivity
2.3.5.1. Insertion of Peptide Linkers to Increase Protein
Productivity 32
2.3.5.2. Application of Fed-batch Strategy to Increase Protein
Productivity 34
2.4. Discussion 36
2.5. Summary 39

CHAPTER 3: CHARACTERIZATION OF IN VITRO BIOLOGICAL
ACTIVITIES OF PROINSULIN-TRANSFERRIN
FUSION PROTEIN
3.1. Background 40
3.2. Materials and Methods
3.2.1. Proteins 41
3.2.2. Cell Cultures 42
3.2.3. Insulin Receptor Competitive Binding Assay 42
3.2.4. Transferrin Receptor Competitive Binding Assay
3.2.4.1. Ion Saturation and Iodination of Transferrin 43
3.2.4.2. Transferrin Receptor Binding Assay 44
3.2.5. Adipocytes Differentiation 44
3.2.6. Glucose Uptake Assay in Differentiated Adipocytes 45
3.2.7. Glucose Production Assay in Heptatoma Cells 46
vii
3.2.8. Measurement of Akt Phosphorylation 47
3.2.9. Statistical Analysis 47
3.3. Results
3.3.1. Receptor Binding Affinity in Hepatoma Cells 47
3.3.2. Promotion Glucose Uptake in Adipocytes 50
3.3.3. Activation of Proinsulin-Transferrin by Trypsin Digestion 53
3.3.4. Inhibition of Glucose Production in Hepatoma Cells 56
3.3.5. Kinetic Stimulation of Akt Phosphorylation in Hepatoma Cells 58
3.4. Discussion 60
3.5. Summary 64

CHAPTER 4: TRANSFERRIN RECEPTOR MEDIATED INTRACELLULAR
CONVERSION OF PROINSULIN-TRANSFERRIN TO AN
ACTIVE FORM OF INSULIN IN HEPATOMA CELLS
4.1. Background 65
4.2. Materials and Methods
4.2.1. Chemicals 66
4.2.2. Cell Cultures 66
4.2.3. Enterocyte-like Caco-2 Cell Differentiation 67
4.2.4. Proinsulin-specific Radioimmunoassay 67
4.2.5. Insulin-specific Radioimmunoassay 67
4.2.6. Lactate Dehydrogenase Release Assay 68
4.2.7. Measurement of Alkaline Phosphate Activity 68
4.2.8. Statistical Analysis 69
4.3. Results
4.3.1. Conversion of Proinsulin-Transferrin to Immunoreactive
Insulin-Transferrin in Hepatoma Cells
4.3.1.1. Decrease in Detection of Proinsulin-Transferrin
Following Incubation with Hepatoma Cells 69
4.3.1.2. Increase in Detection of Immunoreactive Insulin-
Transferrin Following Incubation with Hepatoma Cells 71
4.3.1.3. Mechanism Studies of Proinsulin-Transferrin Conversion 74
4.3.2. Activation of Proinsulin-Transferrin in Hepatoma Cells 76
4.3.3. Conversion of Proinsulin-Transferrin to Immunoreactive
Insulin-Transferrin in Caco-2 Cells
4.3.3.1. Conversion in Caco-2 Cells Seeded in Regular Plates 80
4.3.3.2. Conversion in Caco-2 Cells Seeded in Trans-well Plates 82
4.4. Discussion 83
4.5. Summary 88

viii
CHAPTER 5: EVALUATION OF IN VIVO PHARMACOKINETICS AND
PHARMACODYNAMICS OF PROINSULIN-TRANSFERRIN
FUSION PROTEIN
5.1. Background 89
5.2. Materials and Methods
5.2.1. Animals 90
5.2.2. Streptozotocin-induced Diabetes Mice Model 91
5.2.3. Proteins 91
5.2.4. Measurement of In Vivo Pharmacokinetics 92
5.2.5. Measurement of In Vivo Hypoglycemic Efficacy 92
5.2.6. Measurement of Liver-associated Enzyme Expression 93
5.2.7. Statistical Analysis 93
5.3. Results
5.3.1. Pharmacokinetics 94
5.3.2. Pharmacodynamics
5.3.2.1. Hypoglycemic Efficacy of Subcutaneously Administered
Proinsulin-Transferrin in Fasted Diabetic Mice 97
5.3.2.2. Hypoglycemic Efficacy of Intravenously Administered
Proinsulin-Transferrin in Fasted Diabetic Mice 100
5.3.2.3. Hypoglycemic Efficacy of Subcutaneously Administered
Proinsulin-Transferrin in Free-fed Diabetic Mice 102
5.3.3 Inhibition of Hepatic Glucose Production by Proinsulin-
Transferrin 103
5.4. Discussion 105
5.5. Summary 108

CHAPTER 6: SUMMARY AND FUTURE PERSPECTIVES
6.1. Summary
6.1.1. Overall Goal of this Dissertation 109
6.1.2. Proinsulin-Transferrin is Produced using Recombinant DNA
DNA Technology 110
6.1.3. Proinsulin-Transferrin is Converted to an Active Form of Insulin
through Transferrin Receptor Mediated Endocytosis and
Recycling Pathway 111
6.1.4. Proinsulin-Transferrin Elicits Prolonged Pharmacokinetics and
Extended Pharmacodynamics with Liver-preferential Effects
In Vivo 112
6.1.5. Conclusion 113
6.2. Future Perspectives 115

REFERENCES 118

ALPHABETIZED REFERENCES 129
ix
LIST OF TABLES

Table 1: Large-scale production yield of ProINS-Tf fusion protein. 26

Table 2: Purification of his-tagged ProINS-Tf from TFF-concentrated
protein. 29

Table 3: IC
50
values for various proteins in inhibiting glucose production
in H4IIE hepatoma cells. 57

Table 4: PK analyses of subcutaneously administered ProINS and
ProINS-Tf in STZ-induced diabetic mice. 97

x
LIST OF FIGURES

Figure 1: Domain structures of preproinsulin, ProINS and INS. 5

Figure 2: Biosynthetic conversion of ProINS to INS. 9

Figure 3: Design of ProINS-Tf recombinant fusion protein. 23

Figure 4: Small-scale expression of ProINS-Tf in HEK 293 cells. 24

Figure 5: Expression and quantification of ProINS-Tf using anti-Tf
Western blot. 25

Figure 6: Formulation and stability analysis for ProINS-Tf. 27

Figure 7: Purification of ProINS-Tf by Sephacryl S-200. 28

Figure 8: Ni-NTA purification of ProINS-Tf. 30

Figure 9: Detection of concentrated and purified ProINS-Tf by
Western blot. 31

Figure 10: Increase of ProINS-Tf production by H
4
and AQ
3
peptide linkers. 33

Figure 11: Optimization of ProINS-Tf production using fed-batch strategy. 35

Figure 12: Low binding affinity of ProINS-Tf to IR in H4IIE hepatoma
cells. 48

Figure 13: Slightly decreased binding of ProINS-Tf to TfR in H4IIE
hepatoma cells. 50

Figure 14: Method validation for glucose uptake assays in adipocytes. 51

Figure 15: Dose-dependent response of ProINS and INS in promotion
of glucose uptake in adipocytes. 52

Figure 16: ProINS-Tf exhibited low activity in promotion of
glucose uptake in adipocytes. 53

Figure 17: ProINS-Tf was converted to INS-like-Tf by trypsin digestion. 54

Figure 18: ProINS-Tf was activated by trypsin digestion. 55
xi
Figure 19: Enhanced inhibition of glucose production by ProINS-Tf
in H4IIE hepatoma cells was TfR-mediated. 56

Figure 20: ProINS-Tf elicited a delayed but prolonged stimulation of
Akt phosphorylation in H4IIE hepatoma cells. 59

Figure 21: ProINS-Tf was progressively decreased during incubation with
H4IIE hepatoma cells. 70

Figure 22: Cell membrane integrity remained intact following ProINS-Tf
treatment in the presence or absence of excess Tf or BSA. 71

Figure 23: ProINS-Tf was converted to irINS-Tf in H4IIE hepatoma cells. 72

Figure 24: Conversion of ProINS-Tf to irINS-Tf was a TfR-mediated
process. 73

Figure 25: Conversion of ProINS-Tf was a pH-sensitive, microtubule-
and temperature-dependent process. 75

Figure 26: H4IIE-pretreated ProINS-Tf exhibited an increased binding
to IR. 77

Figure 27: H4IIE-pretreated ProINS-Tf elicited a significant increase
in the 5 min stimulation of Akt phosphorylation. 78

Figure 28: H4IIE-pretreated ProINS-Tf exhibited a significant increase
in promoting glucose uptake in adipocytes. 79

Figure 29: ProINS-Tf was converted to irINS-Tf in differentiated
Caco-2 cells. 81

Figure 30: Conversion of ProINS-Tf in differentiated Caco-2 cells
was a TfR-mediated process. 81

Figure 31: ProINS-Tf was slightly converted to irINS-Tf during
transcytosis in differentiated Caco-2 cells. 83

Figure 32: ProINS-Tf exhibited a longer half-life than ProINS through
intravenous injection to CF-1 mice. 94

Figure 33: ProINS-Tf exhibited prolonged PK profiles through
subcutaneous injection to STZ-induced diabetic mice. 96

xii
Figure 34: ProINS-Tf showed extended hypoglycemic efficacy following
subcutaneous injection in fasted STZ-diabetic mice. 99

Figure 35: ProINS-Tf exhibited a delayed but prolonged hypoglycemic
response following intravenous injection to diabetic mice. 101

Figure 36: ProINS-Tf-TD showed a decreased molecular weight 102
compared to ProINS-Tf.

Figure 37: ProINS-Tf demonstrated a weak hypoglycemic response
following subcutaneous injection to free-fed diabetic mice. 103

Figure 38: Liver G6Pase and PEPCK mRNA expression was inhibited
by ProINS-Tf at 12 h post-injection to fasted diabetic mice. 104

xiii
LIST OF SCHEMES

Scheme 1: The biological potency of protein drugs depends on both
receptor binding affinity and intracellular processing fate. 63

Scheme 2: A hypothetical model for the intracellular activation of
ProINS-Tf to irINS-Tf in hepatoma cells. 87

xiv
ABBREVIATIONS

ALP Alkaline phosphatase
Apo-Tf Iron-free transferrin
AUC Area under the curve
BBI Bowman-Birk soybean protease inhibitors
BCA Bicinchoninic acid
BG Blood glucose
BSA Bovine serum albumin
CHO Chinese hamster ovary
CPM Count per minute
DMEM Dulbecco’s Modified Eagle Medium
ECL Enhanced chemiluminescence
ER Endoplasmic reticulum
FBG Fasting blood glucose
FBS Fetal bovine serum
Fc Fragment crystallizable region of human IgG
FDA Food and drug administration
FcRn Neonatal Fc receptor
G6Pase Glucose-6-phosphatase
GAPDH Glyceraldehydate-3-phosphate dehydrogenase
G-CSF Granulocyte colony-stimulating factor
GLP-1 Glucagons-like peptide-1
xv
GI Gastrointestinal
HEK 293 Human embryonic kidney 293
hGH Human growth hormone
HGP Hepatic glucose production
HRP Horseradish peroxidase
INS Insulin
INS-like-Tf Insulin-like-transferrin
INS-Tf Insulin-transferrin
IR Insulin receptor
irINS-Tf Immunoreactive insulin-transferrin
KRP Kreb-Ringer phosphate
LDH Lactate dehydrogenase
MWCO Molecular weight cutoff
Ni-NTA Nickel-nitrilotriacetic acid
PBS Phosphate buffered saline
PC Prohormone convertase
PD Pharmacodynamics
PEI Polyethylenimine
PEPCK Phosphoenolpyruvate carboxykinase
PGD Peripheral glucose disposal
PK Pharmacokinetics
ProINS Proinsulin
xvi
ProINS-Tf Proinsulin-transferrin
ProINS-Tf-TD Trypsin-digested proinsulin-transferrin
RIA Radioimmunoassay
STZ Streptozotocin
TBS-T Tris-buffered saline with Tween-20
TEER Transepithelial electrical resistance
Tf Transferrin
TFF Tangential flow filtration
TfR Transferrin receptor
TGN Trans-Golgi network
TNF- α Tumor necrosis factor- α
TNFR Tumor necrosis factor- α receptor
USC University of Southern California
VCD Viable cell density

xvii
ABSTRACT

Long-acting insulin (INS) analogues that exhibit prolonged time-action profiles
and liver-specificity are currently in great demand for diabetes treatment. Native INS and
its protein precursor proinsulin (ProINS) are both small peptides with short in vivo half-
life and efficacy. Human transferrin (Tf) is a stable and large-sized plasma protein, and it
has been demonstrated to prolong the half-life of small proteins. With the purposes of
improving the therapeutic application of INS, a proinsulin-transferrin (ProINS-Tf)
recombinant fusion protein has been designed and developed. This fusion protein is
produced using recombinant fusion technology combined with his-tag purification
method. ProINS-Tf exhibits a low activity in the 30 min promotion of glucose uptake in
adipocytes, which corresponds with a low binding affinity to insulin receptor.
Additionally, ProINS-Tf can be activated by ex vivo trypsin digestion. These results
suggest that, similar to ProINS, the intrinsic potency of ProINS-Tf is low and an
activation is required to achieve biological activity. On the other hand, ProINS-Tf elicits
a Tf receptor (TfR) dependent enhanced activity in the 24 h inhibition of glucose
production in hepatoma cells. Radioimmunoassays clearly demonstrate a TfR-mediated
conversion and activation of ProINS-Tf to an immunoreactive insulin-transferrin fusion
protein during the 24 h incubation with hepatoma cells. Therefore, we have proposed an
intracellular ProINS-Tf activation mechanism that is mediated through TfR-mediated
endocytosis and recycling pathway. Furthermore, compared to ProINS and INS,
subcutaneously injected ProINS-Tf exerts an extended hypoglycemic efficacy with a
prolonged half-life in fasted diabetic mice. The correlation of hypoglycemic efficacy with
xviii
the suppression of liver-associated enzyme expression suggests a liver-preferential effect
by ProINS-Tf. In addition, intravenously injection of trypsin-digested ProINS-Tf shows
an immediate hypoglycemic response, whereas ProINS-Tf exhibits a delayed and long-
lasting hypoglycemic efficacy. These observations imply that an activation of ProINS-Tf
may also occur in vivo. Taken together, results from this dissertation have presented three
contributions to the therapeutic protein discovery and development of. First, ProINS-Tf is
a novel fusion protein and the first proprotein within the Tf-based fusion protein family.
Second, a novel receptor-mediated intracellular proprotein activation mechanism is
discovered. Lastly, ProINS-Tf shows great promise as a potential long-acting INS
analogue for diabetes treatment.
1
CHAPTER 1: INTRODUCTION

1.1. Protein Therapeutics
1.1.1. Applications and Advantages of Protein Therapeutics
Protein-based biological products have increasingly been used as therapeutics due
to emerging discoveries in biomedicine and the rapid development in biotechnology. As
of 2011, 135 recombinant proteins or monoclonal antibodies have been approved by the
Food and Drug Administration (FDA) as biopharmaceutical products and marketed for
use [1]. Many more are under clinical trials or early development. Leader et al have
classified protein therapeutics into three functional categories. The first group consists of
proteins with enzymatic or regulatory activity (e.g., cytokines, growth factors, enzymes,
and immunomodulators). This group of protein therapeutics is used to supplement
endogenous proteins that are deficient or abnormal in disease conditions, to augment an
existing pathway, or to provide a novel function or activity. The second group regards
protein therapeutics with special targeting activity, such as monoclonal antibodies that
interfere with targeted molecules. The last group relates to protein vaccines used for
immunotherapy or cancer treatment [2].
Protein therapeutics have many advantages over small-molecule drugs, since (1)
they possess complex set of functions that can hardly be imitated by small-molecules, (2)
their actions are highly specific to target sites, and (3) they are naturally produced and
thus more tolerated and safer for the patients [3]. These favorable attributes of protein
drugs make protein therapy an attractive field in drug design and delivery.
2
1.1.2. Challenges for Protein Therapeutics
The successful application of proteins as therapeutics is growing but limited due
to many challenges encountered with protein drug development. The first challenge is
that most therapeutic proteins show undesirable pharmacokinetic (PK) and
pharmacodynamic (PD) profiles [3]. The expression (or secretion) levels and distribution
patterns of the naturally produced proteins are elaborately controlled in vivo as to
optimize their biological activities, whereas it is challenging for exogenously
administered proteins to exactly mimic the endogenous counterparts.
The second challenge is that protein drugs usually require exquisite formulation
and delivery design. Proteins are susceptible to chemical and enzymatic degradation after
administration, so a stable formulation that maintains the in vivo stability is necessary [4].
Protein drugs have short plasma half-life and are rapidly cleared by the body, so they
usually require frequent dosing to achieve desired efficacy [5]. Moreover, they also have
low bioavailability due to the poor permeability across natural barriers such as epithelium
and endothelium [6].
The third challenge for protein therapeutics is the successfully production of an
active protein form. Endogenous proteins require post-translation modification, such as
glycosylation or proteolytic cleavage, in order to become biologically active [7].
Therefore, the development of therapeutic proteins requires a complex design and a cost-
ineffective large-scale production system [8].
Last, the immunogenicity of therapeutic proteins is a major concern. The body
elicits an immune response against the exogenously administered proteins by generating
3
anti-drug antibodies or neutralizing antibodies. As a result, the biological activities of
therapeutic proteins are neutralized or diminished. This scenario occurs more commonly
for antibody drugs. Therefore, research has been focused to develop humanized
antibodies to evade the immune surveillance and maintain therapeutic efficacy [9, 10].

1.1.3. Current Advances for Novel Protein Therapeutics
The challenges of protein therapeutics have been a driving force in the
advancement of protein design and maturation of novel technology. A variety of
strategies have emerged for the rational design and engineering of protein therapeutics as
summarized by Marshall et al. Many of the following protein therapeutics have been
marketed for clinical application today [11].
One approach to overcome the aforementioned obstacles is the well-known
PEGylation technology, where proteins are covalently bound to a flexible strand of
polyethylene glycol. The PEGylated version of proteins has addressed many challenges
of protein drugs, including reducing protein degradation, improving protein half-life, and
lowering potential immunogenicity. The best examples of PEGylated proteins are
Schering-Plough's PEG-Intron® (peginterferon α-2b) and Roche's PEGasys®
(peginterferon α-2a) [12].
Another method regards chemical modification or genetic mutation of specific
amino acids on proteins. For example, chemically modified insulin (INS) analogues exert
significantly prolonged half-life [13]. In addition, mutation of a cysteine that causes the
formation of unwanted intermolecular or intramolecular disulfide bonds on granulocyte
4
colony-stimulating factor (G-CSF) and human fibrablast growth factor 1 results in
increased stability for these two proteins [14, 15]. Similarly, aldesleukin, where a free
cysteine of interlukin-2 is mutated, demonstrates improved protein stability and
decreased aggregation. Moreover, multiple histidine mutations of G-CSF reduce its
receptor binding affinity under lower pH but not physiological pH, which leads to a more
effective ligand release after endocytosis and thus prolonged half-life [16].
Additional strategies include genetic fusion (discussed in Section 1.4),
engineering of host systems to boost proteins post-translation [4], as well as protein
glycosylation by adding oligosaccharides to proteins to increase their size and stability
(such as Amgen's darbepoetin α for treatment of anemia [17]).

1.2. Insulin as a Protein Therapeutic
1.2.1. Diabetes
Diabetes is a disease condition characterized by elevated blood glucose (BG)
levels due to defects in INS production and/or INS action. The latest National Diabetes
Statistics estimates indicate that by 2011 a total of 25.8 million people in the US (8.3
percent of the population) have diabetes, and among that population, 18.8 million people
are diagnosed. Patients with diabetes can result in serious complications and premature
death. Type I diabetes is caused by the inability to produce sufficient INS owing to
pancreatic beta-cell dysfunction. Therefore, these patients need INS to survive. Type II
diabetes initially results from INS resistance, a disorder in which cells do not respond to
5
INS. However, pancreatic cells gradually lose the INS-producing capability, and
eventually an INS replacement is required.

1.2.2. Human Insulin
1.2.2.1. Insulin Structure and Physiological Actions
INS is naturally synthesized within the beta-cells of the islet of Langerhans in the
pancreas. Following the cleavage of the signal peptide in the endoplasmic reticulum
(ER), proinsulin (ProINS) (9 kDa), consisting of B-chain, C-peptide and A-chain, is
folded into its native structure with the formation of two inter-chain and one intra-chain
disulfide bonds. ProINS is then transported through the Golgi, and packed into secretory
vesicles. Inside the vesicles, prohormone convertases (PC1/3 and PC2) and
carboxypeptidase E specifically remove the C-peptide, producing mature INS (5.8 kDa)
with B- and A-chain connected by two disulfide bonds (Fig. 1) [18].

Fig. 1 Domain structures of preproinsulin, ProINS and INS.

The key targets for INS to exert significant metabolic actions are skeletal muscle,
fat tissues, and liver. In muscle and adipose tissues, INS increases glucose disposal by
stimulating glucose uptake into the cells. In the liver, INS primarily restrains glucose
production through promoting glycogen synthesis (glucose storage) and inhibiting
6
glycogenolysis and gluconeogenesis. In addition to mediating glucose metabolism, INS
also promotes amino acid uptake, protein synthesis, and fatty acid synthesis, and thus
supports cellular growth and proliferation [18, 19].
The action of INS is initiated by binding to insulin receptor (IR) located on the
membrane of target cells. IR is a heterotetramer with two α-subunits and two β-subunits
connected by disulfide bonds. The α-subunits contain INS binding sites and β-subunits
contain the tyrosine kinase domain [20]. The IR binding of INS is complicated. Two
binding sites with high- and low-affinity coexist on IR and, correspondingly, two binding
domains are also present on INS [21, 22]. The activated IR phosphorylates insulin
receptor substrate 1 which transmits the downstream signaling including (1) translocation
of glucose transporter from vesicles to the membrane for glucose uptake, (2) activation of
PI3 kinase-Akt pathway to mediate glucose metabolism and fatty acid synthesis, (3)
activation of mTOR pathway for protein synthesis, and (4) activation of MAPK pathway
for cell growth [23].

1.2.2.2. Insulin Secretion and Insulin Therapy
Normal INS secretion consists of prandial (meal-stimulated) and basal
components. Prandial INS promotes peripheral glucose disposal (PGD) which is
exquisitely synchronized with the increase in BG levels. It is characterized by a prompt
rise to serum concentrations of 80-120 U/mL (equivalent to 480-720 pM) followed by a
quick return to basal levels. On the other hand, basal INS is continuously secreted
between meals and overnight to suppress hepatic glucose production (HGP), which is
7
usually in the range of 5-10 U/mL (equivalent to 30-60 pM) [24, 25]. The endogenous
INS produced by the pancreas goes through portal veins to the liver first where around
50% of the INS is cleared, and then flows to peripheral tissues (muscles and adipose
tissues). Thus, in the physiological pathway, the liver is exposed to higher levels of INS
compared to peripheral tissues.
To improve glycemic control and simulate the physiological INS secretion, a
basal-prandial INS therapy has currently been recommended for both types of diabetes. A
fast-acting INS replaces prandially-released INS to decrease BG elevation after meals,
and a long-acting INS replaces basal INS to control fasting blood glucose (FBG) levels
by suppressing HGP [26, 27]. It is well accepted that modulating FBG levels to normal
values is critical in glycemic control. Thus the earlier use of basal INS to control FBG
levels is highly recommended, especially to patients with type II diabetes [24].

1.2.2.3. Therapeutic Insulin Analogues in Current Use
Humulin R and Neutral Protamine Hagedorn have been used as short-acting and
intermediate-acting INS, respectively [28]. However, they are not ideal replacements.
With the advance of new technology, native INS structure is further optimized for
improved PK characteristics by chemical modification, such as changing the solubility,
altering the oligomerization state, or conjugating with additional moieties. Insulin lispro,
insulin aspart and insulin glulisine are classified as fast-acting INS analogues with rapid
onset and short duration of action in response to meals, and their PK profiles are closer to
postprandial INS response. On the other hand, insulin glargine, insulin detemir, and the
8
most recent insulin degludec are designed as long-acting bedtime INS analogues with
approximately 24 h long-lasting effect and no pronounced peak, where are used as basal
INS supplements [13, 28-30].

1.2.3. Human Proinsulin
Human ProINS is the single-chain protein precursor to INS. The biosynthetic
conversion of ProINS to INS is a complex process mediated through several conversion
intermediates including split (32,33) ProINS, split (65,66) ProINS, des-(31, 32) ProINS,
and des-(64,65) ProINS (Fig. 2) [31]. Due to the complexity of the conversion and the
lack of access to different endoproteases (not clearly identified in cells other than
pancreatic cells), processing of ProINS to mature INS and the conversion intermediates is
minimal in vitro and in vivo [32, 33].
ProINS has demonstrated ~1-10% potency of INS when tested in various in vitro
models including IR binding or INS-related activities [34]. It has also been reported that
the intracellular survival and recycling of radiolabeled ProINS are greater than that of
INS in a fibroblast cell line. ProINS has also shown a longer hepatic half-life than INS in
rats. When examined during clinical trials on normal volunteers, equal dosage of ProINS
has demonstrated a longer elimination half-life and a slower metabolic clearance rate than
INS [35]. Due to its low receptor affinity, its in vivo potency is lower than INS (4U/mg
for ProINS vs. 28 U/mg for INS). In addition, the effect of ProINS on PGD and HGP has
been found to be 8% and 12% that of INS, respectively, indicating a relatively
hepatospecific effect for ProINS [36].
9
Based on clinical pharmacology studies in patients with both types of diabetes,
ProINS has been shown to produce equivalent but not superior metabolic efficacy
compared with INS, and its hepatospecificity has been demonstrated to be marginal.
Subsequently, six myocardial infarctions, including two cases of death, have been
observed in patients treated for more than 1 year of ProINS. Therefore, lacking a unique
efficacy and the occurrence of severe side effects have led to suspension of ProINS for
further development [36, 37].

Fig. 2. Biosynthetic conversion of ProINS to INS [31]. The letters T and C
indicate reactions suggested to be catalyzed by trypsin-like and carboxypeptidase-like
enzymes, respectively.

10
1.3. Transferrin
1.3.1. Physiological Properties and Functions
Human transferrins (Tf) are a class of structurally homologues metal-binding
glycoproteins including serum Tf, lactoferrin and ovotransferrin. Serum Tf functions as
an iron carrier by binding and transporting iron through body fluids [38]. Tf contains 679
amino acids and has a molecular weight of ~80 kDa [39]. The molecule is stabilized
through 19 intramolecular disulfide bonds and is glycosylated with three heterosaccharide
chains. The single Tf polypeptide is composed of two homologous globular domains,
known as N-lobe (336 amino acids) and C-lobe (343 amino acids), which are linked by a
short spacer sequence. Both lobes of Tf molecule are able to bind to one iron molecule
individually. Tf is predominantly synthesized in the liver and is secreted into the blood
and other biological fluids such as lymph, cerebrospinal and bile [40]. In human, plasma
concentration of Tf is stable, ranging from 2 g/L to 3 g/L, and the in vivo half-life of Tf is
eight days [41]. Normally, essentially all the iron in the circulation exists as Tf-bound
form (except for the iron-overloaded conditions), and about one-third of Tf is occupied
by iron [42].
The major roles of Tf are iron binding and transport. It has been well
acknowledged that one principle pathway of cellular iron uptake is through Tf receptor
(TfR)-mediated internalization of iron-Tf complex [43]. TfR is a disulfide-linked
homodimer of 95 kDa subunits (760 amino acids). Each receptor subunit is able to bind
to one molecule of Tf. TfR is constitutively expressed in various tissues including
erythroid cells, hepatocytes, intestinal epithelial cells, endothelial cells and neurons. It is
11
also highly expressed in actively dividing cells such as tumor cells. Virtually all cells can
take up iron through TfR-mediated endocytosis [40].

1.3.2. Receptor-mediated Endocytosis and Recycling Pathway
The Tf-TfR endocytotic pathway has been extensively studied for decades. A
distinctive feature of Tf-TfR pathway is that, unlike most ligands that are sorted to the
lysosomes for degradation, it is released at the cell surface intact. After binding and
subsequent receptor-mediated endocytosis of the Tf-TfR complex, Tf unloads iron in the
acidic endosomal compartments. Iron-free Tf (apo-Tf) remains bound to TfR
intracellularly and is recycled back to the cell surface for release [44]. Within the Tf-TfR
pathway, a fast recycling and a slow recycling route have been identified. The fast
recycling route goes through early endosomes only and the recycling occurs rapidly
within 10 min. On the other hand, the slow recycling route reaches deeper inside the cells
to recycling vesicles and thus usually take more than 30 min for each cycle [45]. These
recycling endosomes have been characterized to be biochemically different from early
endosomes and are usually observed at the perinuclear location near the trans-Golgi
network (TGN) [46, 47]. Correspondingly, many studies of TfR intracellular trafficking
have indicated that that TfR is found inside or around the TGN region during its
intracellular sorting process [48-50].

12
1.3.3. Therapeutic Application
Beyond iron binding and transport, Tf has been increasingly explored as a drug
delivery and targeting carrier [51]. There are several advantages using the Tf system as
an protein drug delivery vehicle. First, Tf is an endogenous protein, and binding of Tf to
TfR does not elicit significant changes in other physiological pathways. Second, Tf is an
enzymatically inert serum protein with a long plasma half-life. Third, high density of TfR
has been found in various tissues. Various forms of Tf have been applied for caner
targeting delivery of proteins, toxins or chemotherapeutics. This target-specific delivery
to cancer cells by Tf is due to the overexpression of TfR in these cell types. Tf-
conjugated diphtheria toxin has shown increased cytotoxicity relative to non-conjugated
toxin in cancer cells [52]. Tf-bearing dendrimers have also been utilized to deliver tumor
necrosis factor- α (TNF- α) to tumor tissues to elicit anti-tumor activities [53]. The
application of Tf-based fusion proteins will be discussed in Section 1.4.3.

1.4. Bifunctional Fusion Proteins
The advancement of recombinant technology has enabled genetic fusion of any
two proteins to create bifunctional therapeutics. Fusion technology provides vast
opportunities to generate novel combinations of distinct functions. This include
improving PK/PD properties, facilitating protein transport, achieving a particular delivery
route, carrying proteins to target sites for local release, enhancing protein potency
through synergetic effects, increasing protein stability, or adding purification tags [3, 11,
54]. The widely applied fusion partners include the fragment crystallizable region (Fc)
13
domain of human IgG, albumin, and transferrin [55]. Examples of their applications are
summarized below.

1.4.1. Fc-based Fusion Proteins
Fc fusion proteins have become an alternative to monoclonal antibodies in current
biotherapeutics development. Fusion with the Fc domain of human IgG extends the half-
life of proteins, due to the pH-dependent binding to the neonatal Fc receptor (FcRn)
which prevents the fusion proteins from being delivered to lysosomes for degradation
[56, 57], and the large size of Fc domain owing to its homo-dimeric nature [58]. In
addition, Fc region in many cases improves the stability of the proteins. Furthermore, the
Fc-dependent effector functions may also offer some additional application for cancer
treatment [59].
Four soluble Fc-based fusion proteins have been approved by FDA and many
more are under clinical trials [60]. The first successful case is the Amgen's Enbrel®
(etanercept) marketed for treatment of rheumatoid arthritis. Etanercept is composed of the
necrosis factor receptor (TNFR) and the Fc domain. Due to the Fc fusion, it exerts
increased half-life and 50- to 1000-fold higher affinity for TNF- α than the monomeric
TNFR [61]. Other Fc fusion proteins include Amevive® (alefacept) where Fc is fused
with lymphocyte function-associated antigen 3, Orencia® (abatacept) where Fc is fused
with cytotoxic T lymphocyte associated molecule-4 [60]. Most of these fusion proteins
are used for the treatment of autoimmune or cancer-related diseases.

14
1.4.2. Albumin-based Fusion Proteins
Albumin-fusion is a well-established technology in current biotherapeutics
development. Fusing with serum albumin reduces the renal elimination of proteins and
improves their half-life, thereby achieving desirable PK/PD profiles. The bases for
albumin-fusion are the increased molecular weight, improved stability and the prolonged
in vivo circulation [62].
Albuferon® (albinterferon α-2b) is a fusion protein consisting of interferon- α and
human serum albumin. It has shown prolonged half-life in both rodents and cynomolgus
monkeys. Despite the lower in vitro potency due to the albumin fusion, albinterferon α-2b
exhibits an improved and extended in vivo PD efficacy which reflects a favorable PK
profiles [63]. In addition, albumin has also been fused with other proteins or peptides,
such as, glucagons-like peptide-1 (GLP-1), G-CSF, interleukin-2, single-chain INS, and
B-type natriuretic peptide [64-66]. Some of these fusion proteins are currently being
investigated in clinical trials.

1.4.3. Transferrin-based Fusion Proteins
In addition to the Fc domain and albumin, Tf is another molecule that has been
examined as a fusion partner for modulating PK/PD of proteins [51]. Currently, either
mammalian-based (fully glycosylated) or yeast-based (partially or non-glycosylated)
human Tf has been fused with four proteins or peptides including G-CSF, human growth
hormone (hGH), GLP-1, and exendin-4 [67-69]. These studies have demonstrated the
15
application of Tf for either prolonging plasma half-life or facilitating oral absorption of
protein drugs.
Based on the TfR-mediated transcytosis across the gastrointestinal (GI)
epithelium, Tf fusion proteins have exerted their respective pharmacological efficacy
following oral administration to rodent models. A G-CSF-Tf fusion protein induces an
increase in neutrophil number after oral administration to BDF1 mice [68]. Similarly, the
body weight of rats has gained significantly by oral dosing with hGH-Tf fusion proteins
[67]. On the other hand, GLP-1 fused with the N-terminus of Tf has exerted significantly
increased half-life (14 h in mice compared to less than 1-2 min for native GLP-1 alone).
Although having a decreased activity in vitro, both GLP-1-Tf and exendin-1-Tf fusion
protein have demonstrated similar in vivo effects with long-lasting PD response [69].

This dissertation is focused on developing a proinsulin-transferrin (ProINS-Tf)
fusion protein, the fifth Tf-based recombinant fusion protein. ProINS is selected because
INS has a two-chain structure and is not a suitable candidate for direct fusion. Compared
with INS or ProINS, ProINS-Tf is expected to elicit improved performance in the in vitro
biological activities and the in vivo PK/PD, by taking advantage of Tf such as large size,
good stability, and the unique receptor-mediated recycling pathway.

16
CHAPTER 2: DESIGN AND PRODUCTION OF PROINSULIN-TRANSFERRIN
RECOMBINANT FUSION PROTEIN

2.1. Background
Protein therapeutics with desirable efficacy have become an emerging treatment
for various diseases. Large-molecule proteins are produced using recombinant DNA
technology and biological systems. Advances in this platform have enabled a variety of
protein modifications including manipulation of protein sequences, fusion of distinct
protein moieties, protein truncation, and protein labeling [11, 54].
In addition to the protein design, protein expression and downstream product
processing are also two critical areas of study. The productivity, purity, concentration,
stability and other biophysical properties all contribute to the biological efficacy of
protein drugs [8, 70]. Therefore, all of these factors need to be taken into consideration
for protein therapeutic development.
The main goal of this dissertation is to improve the therapeutic application of INS.
Connecting INS with Tf may offer multiple benefits, since Tf can increase the size,
prolong the half-life and facilitate the transport of small peptides [46, 51, 69]. Thus, a
bifunctional fusion protein consisting of both INS and Tf are designed to be produced
using the recombinant fusion technology. However, INS is not suitable for direct fusion,
since it has a two-chain structure connected by three disulfide bonds, as opposed to a
single polypeptide [18]. As a result, its single-chain protein precursor ProINS is selected
to replace INS, and a ProINS-Tf fusion protein is designed. In order to mimic a natural
17
protein, the original sequences of full-length human ProINS and Tf are used. Two amino
acids are inserted between the two moieties due to the restriction enzyme cutting.
Following the design of ProINS-Tf, this fusion protein is produced using mammalian
systems because human Tf is a highly glycosylated protein [71]. Reproducible production
of ProINS-Tf fusion protein is achieved using a robust protein concentration instrument
and the his-tag purification technique.

2.2. Materials and Methods
2.2.1. Cell Cultures
Mammalian cells were maintained in humid 5% CO
2
incubators at 37°C. Human
embryonic kidney 293 (HEK 293) cells were cultured in Dulbecco’s Modified Eagle
Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 2 mM L-
glutamine, and 50 U/mL penicillin - 50 μg/mL streptomycin (Invitrogen). HEK 293 cells
were trypsinized with 0.05% trypsin-EDTA during regular passage. Chinese Hamster
Ovary (CHO) cells were cultured in DMEM/F12 medium supplemented with 10% FBS,
2 mM L-glutamine, and 50 U/mL penicillin - 50 μg/mL streptomycin.

2.2.2. Plasmid Construction
The full-length human Tf sequence (NM_001063) was prepared as previously
reported [67]. For the construct containing the carboxyl-terminal his-tagged Tf, a Gly-
Gly-Ser-hexa-His peptide sequence (-GGSHHHHHH-) was incorporated into the
carboxyl-terminal region of the full-length human Tf (residues 1-679) to make a his-
18
tagged Tf (Tf-GGSH
6
) using PCR-based mutagenesis methods. TFR27 plasmid (ATCC,
Manassas, VA) containing the full-length human Tf sequence was used as the PCR
template. The mutagenic forward and reverse primers were designed as 5'-
CCGCTCGAGGTCCCTGATAAAACTGTGAGATGGT -3' (the XhoI recognition site
is indicated in bold print) and 5'-
TGCTCTAGACTAATGATGATGATGATGATGGCTGCCCCCAGGTCTACGGAA
AGTGCAGGCTTCC -3' (the hexa-His sequence is indicated in bold print, the Gly-Gly-
Ser sequence is indicated in italics, and the TAG stop codon is indicated by underline).
Original Tf sequence or his-tagged Tf sequence was engineered into the pcDNA3.1 (+)
expression vector (Invitrogen) by restriction enzyme cutting and ligation at XhoI and
XbaI sites.
The cDNA sequence coding for human preproinsulin (NM_000207) was
amplified from commercial plasmid (SC120054, Origene, MD) by PCR using forward
and reverse primers designed as 5'- GAGGATATCATGGCCCTGTGGA -3' (the EcoRV
recognition site is indicated in bold print, and the ATG start codon is indicated by
underline) and 5'- CCGCTCGAGGTTGCAGTAGTTC -3' (the XhoI recognition site is
indicated in bold print). The purified PCR products were double-digested by EcoRV and
XhoI, and ligated in frame with either pcDNA3.1 (+) empty vector or pcDNA3.1 (+)
vectors containing Tf/his-tagged Tf.
Vectors containing the preproinsulin-Tf fusion gene were subsequently
transformed to Z-Competent™ E. Coli strain DH5a (Zymo Research, CA). The positive
clones were selected by colony-PCR method and stored at -80°C. Plasmids were
19
extracted using QIAprep Spin Miniprep Kit (Qiagen, Valencia, CA) and sent out for
sequencing (Laragen, Los Angeles, CA). Large-scale plasmid production was performed
using EndoFree Plasmid Mega kit (Qiagen).

2.2.3. Fusion Protein Production
Small-scale production was carried out in 6-well plates. Cells were seeded at
~80% confluence at the time of transfection. Plasmids were mixed with Lipofectamine
2000 reagent (Invitrogen) according to manufacturer's instructions. The complex was
diluted in Opti-MEM and added to cells for 6 h incubation. Following transfection, cells
were cultured in serum-free CD 293 medium (Invitrogen) supplemented with 4 mM L-
glutamine for 7-8 days. Media were collected twice every 4 days. The combined medium
was concentrated by ultrafiltration using Amicon Ultra-15 centrifugal filter unit with
molecular weight cutoff (MWCO) of 30,000 (Millipore, Billerica, MA). The retentate
was collected and stored at 4°C.
Large-scale production was performed in T-175 or T-225 flasks. Cells were
maintained at ~80% confluence prior to transfection. Plasmids were transiently
transfected to HEK 293 cells through polyethylenimine (PEI)-mediated DNA
transfection. The ratio of PEI to DNA was 2.5 to 1 (w/w). The complex was prepared in
150 mM NaCl and diluted in DMEM. After 6 h transfection, medium was replaced by
serum-free CD 293 medium supplemented with 4 mM L-glutamine. Conditioned CD 293
medium was collected twice every 4 days, followed by concentration using tangential
20
flow filtration (TFF) system (Millipore). For non-tagged ProINS-Tf, the TFF-
concentrated protein solution was applied to Amicon for further concentration.

2.2.4. Fusion Protein Purification
2.2.4.1. His-tag Purification
For his-tagged ProINS-Tf, the concentrates were first buffer-exchanged to batch
incubation buffer (50 mM sodium phosphate, 300 mM NaCl, 10 mM imidazole, pH 7.0),
and then incubated with pre-equilibrated nickel-nitrilotriacetic acid (Ni-NTA) agarose
(Qiagen) at 4°C for 1.5 h to allow specific binding of his-tagged ProINS-Tf to Ni-NTA.
After washing with wash buffer (50 mM sodium phosphate, 300 mM NaCl, 20 mM
imidazole, 0.05% Tween-20, pH 7.0) to remove the impurities, his-tagged ProINS-Tf was
eluted from Ni-NTA agarose by elution buffer (50 mM sodium phosphate, 300 mM
NaCl, 250 mM imidazole, 0.05% Tween-20, pH 7.0). The excess imidazole was removed
by overnight dialysis using Spectra/Por dialysis membrane (MWCO 12-14 kDa,
Spectrum Laboratory, Rancho Dominguez, CA). The dialysates was composed of 50 mM
sodium phosphate, 50 mg/mL mannitol and 0.01% Tween-20 at pH 7.0. Dialyzed protein
can be frozen at -80°C and stored for months.

2.2.4.2. Gel Filtration Chromatography
Sephacryl-200 gel (GE Healthcare, Piscataway, NJ) was loaded to a 40 mL
column and pre-equilibrated in phosphate buffered saline (PBS). One mg protein (1
mg/mL) was added to the column and eluted with PBS as the mobile phase. The flow rate
21
was 0.67 mL/min, and 1 mL was collected for each fraction using fraction collector
(Gilson, Middleton, WI). Each fraction was assayed for OD
280
using UV-visible
spectrophotometer (Shimadzu, Japan).

2.2.5. Fusion Protein Analysis
2.2.5.1. SDS-PAGE
ProINS-Tf fusion protein was mixed with non-reducing 6X Laemmli sample
buffer and boiled for 10 min. For ProINS-Tf quantification, various amounts of apo-Tf
(20, 40, 80, and 160 ng) was used as loading standards. The samples were resolved by
8% SDS-PAGE, and transferred to 0.45 μm polyvinylidene fluoride membrane.

2.2.5.2. Western Blot
Following 1 h blocking with 5% non-fat milk, the membrane was incubated with
anti-Tf antibody (T2027, Sigma, St. Louis, MO) with the dilution of 1:10,000 at 4°C
overnight. The next morning, the membrane was washed three times with Tris-buffered
saline containing 0.05% Tween-20 (0.05% TBS-T), and subsequently incubated with
horseradish peroxidase (HRP)-conjugated anti-goat secondary antibody (Bio-Rad,
Hercules, CA) with the dilution of 1:10,000 at room temperature for 1 h. After washing
with 0.05% TBS-T buffer, the immunoreactive bands were detected by enhanced
chemiluminescence (ECL, GE Healthcare) and analyzed by Quantity One 1-D Analysis
software (Bio-Rad).
22
For detection of ProINS moiety, the membrane was incubated with anti-(Pro)INS
antibody (ab8304, Abcam, Cambridge, MA) with the dilution of 1:1000 at 4°C overnight,
followed by anti-mouse HRP-conjugated secondary antibody (Bio-Rad) with the dilution
of 1:10,000 at room temperature for 1 hour. The immunoreactive bands were detected by
ECL-plus (GE Healthcare).

2.2.5.3. Coomassie Blue Staining
Gels were first stained with Coomassie staining buffer (0.1% Coomassie R250,
10% acetic acid, 40% methanol) for 45 min to 1 h, and washed with destaining buffer
(10% acetic acid, 40% methanol) until the background color became clear.

2.2.5.4. Bicinchoninic Acid Protein Assay
Total protein was quantified by Pierce bicinchoninic acid (BCA) Protein Assay
Kit (Thermo Fisher Scientific, Waltham, MA) according to manufacturer's instructions.
Bovine Serum albumin (BSA) was used to establish standard curves.

2.3. Results
2.3.1. Design of Fusion Protein
Our design was to directly fuse human preproinsulin upstream in frame with Tf or
his-tagged Tf (Fig. 3A) and then insert the fusion gene into the pcDNA3.1 (+) vector for
expression in mammalian cells. This design allowed the translation of the fusion protein
to start from ProINS using ProINS's natural signal peptide. During translation in
23
mammalian cells, the N-terminal signal peptide was cleaved in the ER, generating the full
ProINS-Tf fusion protein. A leucyl-glutamyl dipeptide sequence was incorporated
between ProINS and Tf due to the XhoI restriction enzyme recognition site (Fig. 3B).

Fig. 3. Design of ProINS-Tf recombinant fusion protein. (A) The pcDNA3.1 (+)
vector containing preproinsulin-Tf sequence with EcoRV, XhoI and XbaI restriction
enzyme sites. (B) Delineation of ProINS-Tf fusion protein expressed from HEK 293
cells. A dipeptide linker was introduced between ProINS and Tf due to XhoI recognition
site.

2.3.2. Small-scale and Large-scale Production of Fusion Protein
To initially detect the expression of ProINS-Tf fusion protein, small-scale
production was carried out in 6-well plates. Based on anti-Tf Western blot in non-
reducing condition, a band at ~89 kDa was shown in the medium harvested at the end of
24
3 days incubation (Fig. 4, Lane 1-2). This indicated that ProINS-Tf fusion protein was
expressed and secreted into the medium. A fainter band below the fusion protein was
demonstrated to be Tf (80 kDa), which was the carryover from Opti-MEM during
transfection. On the other hand, a large quantity of ProINS-Tf (89 kDa) was also detected
in HEK 293 cell lysates (Fig. 4, Lane 3-4), which suggested some of the fusion protein
was not secreted and retained inside the cells.

Fig. 4. Small-scale expression of ProINS-Tf in HEK 293 cells. Fusion protein was
detected using 8% non-reducing SDS-PAGE followed by anti-Tf Western blot. Lane 1
and 2 showed the collected medium. Medium (2 mL) was collected at the end of day 3
following transfection and centrifuged to remove suspended cells. Lane 3 and 4 showed
cell lysates from the attached HEK 293 cells.

During the large-scale production and concentration process, an aliquot of protein
at each step was sampled and applied to anti-Tf Western blot for ProINS-Tf
quantification. Fusion protein was concentrated ~40-fold in volume after TFF
concentration alone, and ~330-fold in volume after both TFF concentration and Amicon
ultrafiltration (Table 1 and Fig. 5A). The final concentration of ProINS-Tf was 0.1-0.2
mg/mL, which was enough to conduct the in vitro studies.
25
Furthermore, similar profiles of ProINS-Tf on Western blot were shown among
original harvested medium (Fig. 5A, Lane 8), TFF-concentrated protein (Fig. 5A, Lane
7), and retentate after Amicon ultrafiltration (Fig. 5A, Lane 5-6). No carryover of Tf was
detected, since CD 293 medium was used instead of Opti-MEM during large-scale
transfection. In addition to the major band of ProINS-Tf at 89 kDa, the presence of a
band with less density at molecular weight of ~170-180 kDa indicated the dimerization of
ProINS-Tf. The bands at the top position of the separating gel demonstrated that the
samples also contained some protein aggregates, which seemed to be more apparent in
more concentrated samples. In addition, transfection with pcDNA3.1 (+) empty vector
did not produce any Tf-containing protein (Fig. 5B).

Fig. 5. Expression and quantification of ProINS-Tf using anti-Tf Western blot. (A)
Lane 1-4: Tf standards (5, 40, 80, and 160 ng, respectively). Lane 5 and 6: ProINS-Tf
after TFF and Amicon concentration loaded at 1 and 2 μL, respectively. Lane 7: ProINS-
Tf after TFF concentration alone loaded at 8.8 μL. Land 8: original harvested medium
loaded at 35 μL. (B) Lane 1: Tf control. Lane 2 and 3: concentrated samples from
pcDNA3.1 (+)/ProINS-Tf transfection. Lane 4: concentrated samples from pcDNA3.1(+)
empty vector transfection.

26
Table 1. Large-scale production yield of ProINS-Tf fusion protein (based on five T-225
flasks).
Samples
Final Volume
(mL)
Estimated
Concentration
( μg/mL)*
Yield (mg)
Original harvested medium 700 0.69 0.48
TFF-concentrated protein 17.5 36 0.63
Retentate after TFF and
Amicon ultrafiltration
2.1 153 0.32

* Quantified by Western blot shown in Fig. 5.

2.3.3. Formulation Development of Fusion Protein for Long-term Storage
The long-term storage condition for concentrated ProINS-Tf was evaluated. First,
two alternative storage buffer formulations besides CD 293 medium were tested. Lactose
buffer (20 mM phosphate buffer, 30 mM lactose, pH 7.0) and mannitol buffer (50 mM
phosphate buffer, 50 mg/mL mannitol, 0.1 mg/mL Tween-80, pH 7.0) are two
conventional lyophilization buffers for many proteins. As shown in Fig. 6, fusion protein
maintained fairly good stability for up to 43 days when frozen at -80°C in all three
buffers including CD 293 medium. Furthermore, compared with freezing at -80°C,
storing the fusion protein at 4°C for 7 days already caused some degree of degradation.
Therefore, freezing the fusion protein at -80°C was necessary for long-term storage.
Although CD 293 medium was not a favorable formulation for freezing, the presence of
junk proteins in the concentrated fusion protein might provide a good bulking
environment for ProINS-Tf. Based on this result, prior to the introduction of his-tag
purification, concentrated ProINS-Tf was directly frozen at -80°C without further buffer-
exchange process.
27

Fig. 6. Formulation and stability analysis for ProINS-Tf. ProINS-Tf was stored at
both 4°C and -80°C in various buffers. Aliquots of protein were sampled on day 1, 7 and
43 individually, and applied to anti-Tf Western blot for evaluation.

2.3.4. Improvement of Fusion Protein Purity
In order to improve the purity and quality of ProINS-Tf fusion protein, size
exclusion chromatography was initially applied. One milligram of total protein was
separated using Sephacryl S-200 gel. The elution profile (Fig. 7) showed a huge and
sharp peak from fraction 15 to 23, which presumably was the high molecular weight
aggregates. Following the aggregate peak, there was a wide peak which covered from
fraction 24 until fraction 50, suggesting that various proteins with a wide range of sizes
were present. On the other hand, some fractions were sampled for anti-Tf Western blot to
detect the distribution of ProINS-Tf, and ten fractions from 24 to 33 were ProINS-Tf
positive. This method can separate ProINS-Tf from aggregates and other junk proteins
based on molecular weight. However, due to the wide distribution of ProINS-Tf in ten
individual fractions, ProINS-Tf was too diluted and further concentration process may be
needed.
28

Fig. 7. Purification of ProINS-Tf by Sephacryl S-200. One mg of total protein
(measured by BCA assay) was loaded onto a 40 mL Sephacryl S-200 column and washed
with column PBS. OD
280
was measured for each fraction to obtain the elution profile (left
X-axis, closed square). The presence of ProINS-Tf was assessed in fractions 19, 21, 23,
25, 27, 29, 31, and 33 using anti-Tf Western blot. The relative amount among these
fractions was shown by volume (INT*mm
2
) according to Quantity One 1-D Analysis
software (right X-axis, open circle).

A hexa-histidine sequence was introduced to the carboxyl-terminus of Tf and
used as a purification tag. Prior to purification, lots of proteins were present in TFF-
concentrated samples, and the his-tagged ProINS-Tf can hardly be identified on
Coomassie blue stained gels (Fig. 8AB, Lane 3-4). Following incubation with Ni-NTA
agarose, more than 80% of the proteins were present in the flow-through (Fig. 8AB, Lane
5), and a small fraction of proteins which were either non-specifically or poorly bound to
Ni-NTA agarose were washed away using 20 mM or 30 mM imidazole (Fig. 8AB, Lane
6-7). After removing these junk proteins, the final elutes, which only constituted ~1% of
the total TFF-concentrated proteins, clearly demonstrated a major band of his-tagged
ProINS-Tf at 89 kDa (Fig. 8AB, Lane 8-10, and Table 2). With the optimization of total
29
protein/agarose ratio, the majority of his-tagged ProINS-Tf can be eluted and
concentrated into one fraction. The purity of the final elute ranged from 50% to 80%, as
calculated by the ratio of his-tagged ProINS-Tf amount quantified using Western blot to
total protein amount quantified using BCA assay (Fig. 8C).

Table 2. Purification of his-tagged ProINS-Tf from TFF-concentrated protein.

Concentration
(mg/mL)*
Volume
(mL)
Total Amount
(mg)
Percentage (Compared
to TFF-Concentrated
Protein)
TFF-Concentrated
Protein
1.28 ± 0.02 36 46.18 ± 0.69 100%
Flow-Though 0.99 ± 0.003 37.5 37.21 ± 0.11 80.58%
Wash (20 mM
Imidazole)
0.18 ± 0.01 5 0.92 ± 0.04 2.00%
Wash (30 mM
Imidazole)
0.12 ± 0.01 3 0.35 ± 0.04 0.75%
Elute 0.43 ± 0.03 1 0.43 ± 0.03 0.94%

*Quantified by BCA protein assay, n=2.

30

Fig. 8. Ni-NTA purification of ProINS-Tf. Panel (A) and (B) represented samples
in non-reducing and reducing buffer. Lane 1: Tf. Lane 2: marker. Lane 3 and 4: TFF-
concentrated protein samples before purification. Lane 5: flow-though. Lane 6: washing
by 20 mM imidazole. Lane 7: washing by 30 mM imidazole. Lane 8-10: elute by 250
mM imidazole. (C) Elution profile of his-tagged ProINS-Tf. Elute (1 mL) was collected,
and total protein of each fraction was measured using BCA assay. The amount of his-
tagged ProINS-Tf was quantified using anti-Tf Western blot. The purity was estimated to
be 70%.
31

To further removing contaminating proteins, a washing step with 50 mM
imidazole was also evaluated. However, a portion of ProINS-Tf can be washed out using
50 mM imidazole. Buffer containing 20 mM imidazole was chosen as the optimal
washing conditions, since no significant difference was observed between 20 mM
imidazole and 30 mM imidazole wash. Additionally, in order to test whether fusion
protein yield was largely affected by dialysis, the concentration of ProINS-Tf was also
compared before and after dialysis. It was observed that fusion protein concentration only
slightly increased after dialysis, and it was probably due to the decrease of volume.
The purified ProINS-Tf can be clearly detected by both anti-Tf and anti-(Pro)INS
antibody. The dimerized protein band was not visible on anti-(Pro)INS blot presumably
due to the low detection limit of this antibody (Fig. 9).

Fig. 9. Detection of concentrated and purified ProINS-Tf by Western blot. Lane 1
and 2: human Tf and ProINS-Tf in anti-Tf blot. Lane 3: ProINS-Tf in anti-(Pro)INS blot.

32
2.3.5. Enhancement of Fusion Protein Productivity
2.3.5.1. Insertion of Peptide Linkers to Increase Protein Productivity
Two repeats of A(EAAAK)
4
A helical linkers (designated as H
4
) were introduced
between ProINS and Tf moieties using cloning methods. H
4
sequence was obtained from
G-CSF-(H
4
)
2
-Tf fusion protein produced in the laboratory [72]. Another helical-like
sequence (AAQAA)
3
(designated as AQ
3
) was reported to exhibit a similar structure as
H
4
sequence. Thus, upstream and downstream AQ
3
sequences were synthesized and
annealed to form double-strand oligonucleotides. One or two repeats were inserted into
ProINS-Tf. As shown in Fig. 10, insertion of two repeats of H
4
linker exhibited 2.9-fold
increase in protein yield, while insertion of one repeat or two repeats of AQ
3
linkers
resulted in 1.9- and 1.5-fold increase, respectively.

33

Fig. 10. Increase of ProINS-Tf production by H
4
and AQ
3
peptide linkers. (H
4
)
2

indicated two repeats of H
4
linkers, (AQ
3
)
1
and (AQ
3
)
2
indicated one and two repeats of
AQ
3
linkers, respectively. Expression was quantified relative to ProINS-Tf fusion
protein. The left lane in the blot represented Tf. Data was the average of three individual
expression experiments.

34
2.3.5.2. Application of Fed-batch Strategy to Increase Protein Productivity
Feeding HEK 293 cells with 5 g/L of Bacto™ TC Yeastolate (BD Biosciences,
San Jose, CA) led to an 1.2-fold increase in ProINS-Tf production, but only when fed at
48 h post-transfection. In comparison, feeding with the same concentration of Bacto™
Peptone (BD Biosciences) resulted in similar improvement when fed at both 24 and 48 h
post-transfection. However, the improvement in yield was quite limited with only ~20%
increase (Fig. 11A). On the other hand, feeding with these two nutrients at different
timepoints also affected cell growth, which was an important factor relevant to protein
productivity. Feeding the nutrients at early timepoints, such as 6 h post-transfection,
devastated cell growth significantly. On the contrary, it favors cell growth when fed at 48
h (Fig. 11B). Therefore, feeding at 48 h post-transfection, which was halfway through a 4
days batch collection, seemed to be the optimal choice.

35

Fig. 11. Optimization of ProINS-Tf production using fed-batch strategy.
Transfection of ProINS-Tf was performed in T-25 flasks. For regular protocol, cell
cultures was changed to CD 293 medium at 6 h post-transfection. For fed-batch
protocols, in addition to CD 293 medium change, cells were fed with 5 g/L of yeastolate
or peptone at 6 h, 24 h, or 48 h post-transfection. Media were collected on day 4 after
transfection and concentrated by Amicon. Viable HEK 293 cells were counted and
harvested to measure cellular protein amount. (A) Protein expression yield quantified by
anti-Tf Western blot. (B) Cell viability assessed by either cellular protein quantification
or viable cell density (VCD) at each experimental condition.
36
2.4. Discussion
A recombinant fusion protein composed of human ProINS peptide sequence and
human Tf sequence has been designed and produced using recombinant DNA technology
in mammalian cells. Prior to investigating the activities and therapeutic application of this
fusion protein, it is essential to characterize this novel protein with regards to the
biophysical property (aggregation, stability), purity and productivity (yield).
ProINS was designed upstream of Tf in the fusion protein so that the folding of
the fusion protein inside HEK 293 cells would start from ProINS. This design is believed
to allow a better folding efficiency of ProINS [73]. A dipeptide linker was introduced
between ProINS and Tf because of the restriction enzyme cutting site during fusion gene
cloning (Fig. 3). The lack of free Tf band in Western blot suggested that this dipeptide
linker remained stable during protein expression and secretion process.
Mammalian cell system was selected for ProINS-Tf expression because the
activity of Tf largely relied on its glycosylation status [51, 71]. ProINS-Tf was
successfully secreted out of the HEK 293 cells, as it can be detected from the collected
cultural media broth. Interestingly, a band corresponding to the size of protein dimer
(~170-180 kDa) appeared in both concentrated proteins and in the non-Amicon processed
samples (Fig. 5 and 9). This suggested that the dimerization of fusion protein was not
simply due to aggregation caused by Amicon, but it might occur during protein
expression. Western blot demonstrated that this dimer band disappeared when protein
samples were prepared in the reducing buffer (data not shown), indicating that the
dimerization was due to the formation of intermolecular disulfide bonds. Since ProINS
37
itself contains six cysteines and forms three disulfide bonds, a mismatched intermolecular
formation of disulfide bonds may occur during protein expression. The non-reducing gel
gave a more accurate protein profile because the disulfide bonds remained intact (Fig. 8).
Based on Western blot quantification and BCA assay, ProINS-Tf constituted only
less than 1% of the total collected proteins (Table 2). Gel filtration study also
demonstrated a huge amount of impurities in the concentrated proteins (Fig. 7).
Therefore, adding an additional step of purification became necessary and important to
obtain ProINS-Tf fusion protein with a higher purity. The commonly used his-tag affinity
chromatography has been applied to ProINS-Tf purification, since the presence of his-tag
usually neither affects protein activity nor has safety concerns [74]. Additionally, it has
also been shown that a carboxyl-terminal hexa-histidine sequence did not interfere with
the binding ability of Tf to TfR [75], so the his-tag on ProINS-Tf was not designed to be
removed after purification. In addition, previous studies of the crystal structure of Tf
indicated that the carboxyl-terminal three residues were buried in the interior of the Tf C-
lobe [76]. Thus, a tripeptide sequence GGS was inserted between Tf and the hexa-
histidine sequence in order to make the his-tag accessible to the metal chelate column for
purification. As shown in Fig. 8 and Table 2, a dramatic improvement of ProINS-Tf
purity proved the success of his-tag application to this fusion protein.
During protein production process, a robust concentration system is mandatory to
obtain highly concentrated fusion protein for further activity assessment. The established
TFF system coupled with Amicon ultrafiltration was able to concentrate several liters of
collected media down to 1-2 mL in volume (a couple of thousand fold of concentration).
38
However, Amicon ultrafiltration was not desirable since it often caused protein
aggregation due to centrifugation (Fig. 5). It was found that the addition of his-tag
purification can avoid the use of Amicon, because the final elution step from the nickel
resin required very little volume (usually 1 mL for one fraction). As a result, the his-tag
purification in this case also functioned as a concentration process.
The productivity of ProINS-Tf fusion protein is ~0.5 mg/L (0.32 mg from 700
mL collected medium), so it belongs to the category of low-expressing proteins in
mammalian expression system [70]. Protein production in mammalian cells comprises a
chain of events, including plasmid uptake (transfection), mRNA transcription, protein
translation, folding, and secretion . Additionally, it involves various cellular machineries
including the nucleus, ER, Golgi, and the secretory pathway [77]. A low productivity can
be caused by inefficiencies at any one of these steps. Since the production of a previously
reported hGH-Tf fusion protein has showed very high expression yield [67], it is unlikely
that Tf is the limiting factor for the low expression of ProINS-Tf. Therefore, the problems
may be attributed to ProINS that may exhibit poor folding or insufficient secretion
capacity.
Based on this assumption, several approaches have been tested in order to
enhance the productivity of ProINS-Tf. Introducing a peptide linker between ProINS and
Tf appeared the most effective method. The more rigid H
4
linker provided better
improvement than the less rigid AQ
3
linkers (Fig. 10). It is believed that the peptide
linker increased the distance of the two moieties and thus allowed better protein folding,
which eventually led to higher expression yield [72]. On the other hand, applying fed-
39
batch method slightly helped the production but did not produce a significant
enhancement (Fig. 11). This indicated that cell viability might not be a major issue [77].
Furthermore, other strategies have also been examined, such as changing the orientation
of ProINS-Tf to make a Tf-ProINS, adding bovine INS during the expression to prevent
the potential degradation of ProINS-Tf, modifying the batch collection time, and
switching to CHO cell lines. However, none of these approaches have offered dramatic
improvement on the productivity of ProINS-Tf.

2.5. Summary
In this chapter, a ProINS-Tf recombinant fusion protein has been designed and
produced in HEK 293 mammalian systems. Human ProINS is designed to be directly
fused with human Tf through a dipeptide linker sequence. A small-scale production of
ProINS-Tf has been initially carried out to test the secretion and estimate the productivity
range of this fusion protein. Subsequently, a large-scale production process system along
with protein concentration, his-tag purification, and storage optimization has been
substantially characterized and established. The purity of ProINS-Tf fusion protein can be
improved to as high as 80% following purification. Presumably due to the poor folding of
ProINS, ProINS-Tf fusion protein has a very low productivity (~0.5 mg/L). Among all
the examined approaches, inserting rigid helical peptide linkers has currently been the
most effective method to improve protein yield.

40
CHAPTER 3: CHARACTERIZATION OF IN VITRO BIOLOGICAL
ACTIVITIES OF PROINSULIN-TRANSFERRIN FUSION PROTEIN

3.1. Background
ProINS-Tf is a novel bifunctional fusion protein where human ProINS (9 kDa) is
connected with human Tf (80 kDa) through a peptide bond linkage. Direct fusion of two
functional protein moieties may result in dramatic changes to each individual protein with
regards to its receptor binding (target binding) affinity, intracellular processing, and
consequently its biological potency.
Due to steric hindrance effects, the receptor binding affinity of fusion proteins is
reduced in most situations. The influence of a larger moiety on the smaller moiety will be
greater than vice versa. Conversely, an enhanced binding affinity may also occur when
both proteins bind to their respective receptors at once to achieve a bivalent binding or
synergetic binding effect [78, 79]. Receptor binding activates the downstream signaling
to generate biological efficacy. Following receptor binding, proteins will be endocytosed
into endosomes. Subsequently, they will either be delivered to lysosomes for degradation
(e.g., INS, ProINS, and most hormones) [80], or recycled back with the receptor to the
cell surface (e.g., Tf-TfR, Fc-FcRn) [46, 56]. Therefore, the intracellular metabolic fate
of a fusion protein may be different from the individual protein depending on the
interaction between the protein moieties and receptors.
Taking these factors into consideration, the in vitro biological potencies of
ProINS-Tf fusion protein have been investigated and are reported in this chapter.
Additionally, they have been compared with the potencies of ProINS or Tf alone. ProINS
41
binds to the IR and elicits INS-like metabolic effects [81], albeit one-to-two orders of
reduced binding affinity and biological potency [34, 36, 82]. INS and ProINS are
responsible for two major biological activities: the promotion of glucose uptake in
adipocytes and the inhibition of glucose production in hepatocytes [19, 83, 84].
Therefore, two assays have been developed to model these systems. The results presented
in this chapter comprise of IR/TfR binding abilities, INS-related biological activities, and
stimulation of IR downstream signaling pathway [85, 86].

3.2. Materials and Methods
3.2.1. Proteins
Recombinant human INS produced from E. coli (Sigma) and recombinant human
ProINS (R&D System, Minneapolis, MN) were used in the assay. ProINS-Tf fusion
protein was produced in HEK 293 cells as described in Chapter 2. Human Tf and bovine
BSA were both from Sigma.
The trypsin-digested ProINS-Tf, designated as ProINS-Tf-TD with fusion
protein/trypsin ratio (w/w) shown in the parentheses, was prepared as follows. ProINS-Tf
was mixed with trypsin in different w/w ratio, and was incubated at 37°C for 15 min to
allow trypsin digestion. The reaction was stopped by the addition of Bowman-Birk
soybean protease inhibitors (BBI, Sigma) in the equivalent amount to trypsin (w/w). The
final mixtures were either directly applied to SDS-PAGE for protein analysis or loaded
onto nickel column for protein purification.

42
3.2.2. Cell Cultures
NIH-3T3 cell line was purchased from ATCC (CRL-1658). Cells were cultured in
DMEM medium supplemented with 10% calf bovine serum (ATCC), 2 mM L-glutamine,
1 mM sodium pyruvate, and 50 U/mL penicillin - 50 μg/mL streptomycin. NIH-3T3 cells
were fed every other day and maintained below 70-80% confluence. It was important to
avoid over-confluence and too many passages for the cells, since both would affect cell
differentiation to adipocytes.
Rat hepatoma H4IIE cells from ATCC were cultured in DMEM medium
supplemented with 10% FBS, 2 mM L-glutamine, and 50 U/mL penicillin - 50 μg/mL
streptomycin. Cells were trypsinized with 0.5% trypsin-EDTA during regular passage.

3.2.3. Insulin Receptor Competitive Binding Assay
IR binding assay was performed in H4IIE hepatoma cells. Briefly, cells were
seeded on 24-well plates at a density of 0.5 x 10
6
cells/well. Cells reached 95%
confluence on the next day. Monolayers were first washed twice with 37°C PBS, and
equilibrated with DMEM/0.1% BSA for 15-30 min at 4°C. All procedures were
performed on ice. Dosing solutions containing [
125
I-Tyr14]-INS (Perkin Elmer, Waltham,
MA) and various concentrations of proteins were prepared in DMEM/0.1% BSA and
incubated with cells for 2 h at 4°C. Following incubation, cells were washed three times
with ice-cold PBS to remove the unbound radiolabeled INS. Monolayers were dissolved
in 0.1% NaOH, and counted for radioactivity by the Packard Cobra II gamma counter
(GMI, Ramsey, MN). Total protein was quantified by the BCA assay. The binding curves
43
were plotted by non-linear fit based on one-site competitive binding, and the IC
50
values
were calculated using GraphPad Prism software.

3.2.4. Transferrin Receptor Competitive Binding Assay
3.2.4.1. Ion Saturation and Iodination of Transferrin
Human apo-Tf (Sigma) was dissolved in PBS to a concentration of 5.5 mg/mL,
and 10 μL of 10 mg/mL ferric ammonium citrate (Sigma) was added and mixed with Tf.
The mixture was incubated at 37°C for 2 h, followed by dialysis against PBS at 4°C
overnight. The iron-saturated Tf was then filtered through 0.2 μm polyethersulfone
membrane before BCA protein quantification.
For Tf iodination reaction, 8 mg/mL chloramine-T, 4.8 mg/mL sodium
metabisulfite, and 10 mg/mL potassium iodine solutions were freshly prepared in PBS
right before the experiment. One mg of iron-saturated Tf was mixed well with 50 μl
(equivalent to about 0.25 mCi) of [
125
I]-Na (Perkin Elmer) in a glass tube. The reaction
was started by adding 50 μl of 8 mg/mL chloramine-T to the mixture followed by gentle
swirling. The mixture was then kept on ice for 5 min with frequent swirling during the
incubation. To stop the reaction, 50 μl of 4.8 mg/mL sodium metabisulfite solution was
added to the mixture for a 5 min incubation on ice. Subsequently, 100 μl of 10 mg/mL
potassium iodine solution was added to the glass tube and left on ice for another 2 to 5
min. Following the chemical reaction, the whole mixture was loaded onto a 10 mL PBS-
equilibrated G-50 size exclusion column. PBS was used as mobile phase for column
44
wash, and each fraction was collected and counted by the gamma counter. At the end, the
fractions that had the highest counts were combined and stored at -20°C.

3.2.4.2. Transferrin Receptor Binding Assay
TfR binding assay was performed in H4IIE hepatoma cells. Briefly, cells were
seeded on 6-well plates at a density of 1.5 x 10
6
cells/well and reached confluence by the
next day. Prior to dosing, cells were first incubated with DMEM/0.1% BSA at 37°C for
0.5 h to remove attached serum components. Cells were then washed with 37°C PBS
twice, and incubated at 4°C for 15 min to allow equilibrium due to temperature shift.
Dosing solutions containing [
125
I]-Tf and various concentrations of proteins were
prepared in DMEM/0.1% BSA and incubated with cells for 2 h at 4°C. After incubation,
cells were washed three times with ice-cold PBS to remove the unbound radiolabeled Tf,
and monolayers were dissolved in 1 M NaOH. Radioactivity was determined by the
Packard Cobra II gamma counter and total protein was quantified by the BCA assay. The
binding curves were non-linearly fitted based on one-site competitive binding, and the
IC
50
values were obtained from GraphPad Prism software.

3.2.5. Adipocytes Differentiation
NIH-3T3 cells were seeded into 6-well plates at a density of 1 x 10
6
cells/well on
day -2, and grown to confluence for 2 days. On day 0, preadipocytes were stimulated
with MDI Induction Media which contained DMEM plus 10% FBS, 1 mM sodium
pyruvate, 2 mM L-glutamine, 50 U/mL penicillin - 50 μg/mL streptomycin, 0.5 mM
45
isobutylmethylxanthine (Sigma), 1 μM dexamethasone (Sigma), and 5 μg/ml bovine INS
(Sigma). A distinct change in the morphology of the cells (spindle-like) could be noticed
in the next few days. Two days after MDI induction (day 2), cells were changed to INS
Media consisting of DMEM plus 10% FBS, 1 mM sodium pyruvate, 2 mM L-glutamine,
50 U/mL penicillin - 50 μg/mL streptomycin, and 1 μg/ml bovine INS. The media
became viscous during this period as free fatty acids were produced by the cells and
secreted into the media. On day 4, media were changed to regular DMEM medium
supplemented with 10% FBS, 2 mM L-glutamine, and 50 U/mL penicillin - 50 μg/mL
streptomycin. Cells were then fed with regular DMEM medium every two days (on day
6, day 8, and day 10). The accumulation of lipid droplets inside the cells was gradually
observed, and cells were fully differentiated to mature adipocytes after day 12 and ready
for experiments [87].

3.2.6. Glucose Uptake Assay in Differentiated Adipocytes
Fully differentiated adipocytes, as evidenced by more than 95% of the cells
showing adiposity phenotype, were used for glucose uptake assays. Sixteen hours prior to
experiment, medium was changed to serum-free DMEM supplemented with 0.5% BSA.
After serum starvation overnight, cells were washed with Kreb-Ringer phosphate (KRP)
buffer twice, and incubated with or without various proteins at 37°C for 30 min in KRP
buffer supplemented with 0.1% BSA. Following protein treatment, 0.5 μCi/mL or 0.1
mmol/L of 2-Deoxy-D-[2, 6-
3
H] glucose (GE Healthcare) was added to the cells for
another 10 min incubation. The reaction was stopped by aspirating the medium and
46
washing four times with ice-cold KRP buffer. Cells were then solubilized with 0.4
mL/well of 0.1 M NaOH/0.1% SDS in KRP. Cell lysates were mixed with optiphase
supermix cocktail scintillation fluid (Perkin Elmer) and the amount of internalized
radiolabeled glucose was measured using 1450 MicroBeta TriLux microplate scintillation
and luminescence counter (Perkin Elmer). Total protein amounts were also quantified by
the BCA assay for normalization. The amount of radiolabeled glucose uptake was
recorded as count per minute (cpm) normalized by protein quantification.

3.2.7. Glucose Production Assay in Hepatoma Cells
Glucose production was determined using a modified method from previous
reports [88]. H4IIE cells were seeded in 24-well plates at a density of 0.5 x 10
6
cells/well
in high-glucose DMEM containing 10% FBS. Serial dilutions of proteins were prepared
in serum-free DMEM. Upon confluence, cells were washed with PBS twice to remove
excess serum, and treated with different concentrations of proteins for 24 h at 37°C. Cells
were then washed with PBS three times and incubated for 3 h in glucose production
medium consisting of serum-, glucose- and phenol red-free DMEM supplemented with 2
mM sodium pyruvate and 40 mM sodium DL-lactate. The medium was harvested, and
glucose concentrations were measured using the Amplex Red Glucose/Glucose Oxidase
Kit (Invitrogen). Cells were lysed in 1 M NaOH and cellular proteins were measured
using BCA assay. The inhibitory curves was fitted non-linearly based on log(inhibitor)
versus response, and the IC
50
values for proteins were calculated using GraphPad Prism
software.
47
3.2.8. Measurement of Akt Phosphorylation
Quiescent H4IIE cells that were serum-starved for 18 h were treated with or
without proteins in serum-free conditions. After treatment, cells were immediately lysed
with cell extraction buffer (Invitrogen) supplemented with phenylmethanesulfonyl
fluoride (Sigma) and protease inhibitor cocktail (Sigma). After protein quantification,
equal amount of cellular proteins (60 μg) were subjected to Western blot analysis by anti-
phospho-Akt antibody (Ser 473, 4060, Cell Signaling Technology, Danvers, MA) and
anti-beta-actin antibody (AC-15, A5441, Sigma). Immunoreactive bands were detected
using ECL, and quantified using Quantity One 1-D Analysis software.

3.2.9. Statistical Analysis
Experiments were all performed with n=3 for each condition. Student's t-test was
used for statistical analysis, where differences with p values less than 0.05 were
considered significant.

3.3. Results
3.3.1. Receptor Binding Affinity in Hepatoma Cells
In order to evaluate whether fusing ProINS with Tf would affect the binding
ability of each moiety to its respective receptor, the competitive binding of ProINS-Tf to
IR and TfR was measured. Since hepatoma cells express similar levels of IR and TfR
and both receptors are of high abundance [89], the binding studies were carried out in
H4IIE hepatoma cells.
48

Fig. 12. Low binding affinity of ProINS-Tf to IR in H4IIE hepatoma cells.
Following binding at 4°C for 2 h, the cell-associated radioactivity was recorded as cpm.
Data were presented as an average of cpm normalized by protein amount (n=3).

The IR binding assay was based on the competitive binding of tested proteins
with radiolabeled tracer [
125
I-Tyr14]-INS. The IR binding ability of INS, ProINS, and
ProINS-Tf was compared as shown in Fig. 12. ProINS-Tf fusion protein exhibited very
weak IR binding at the concentration range from 1 pM to 10 nM (higher concentrations
were not determined). The binding curve of ProINS-Tf appeared very similar to that of
ProINS below the concentration of 10 nM. The calculated IC
50
values for INS and
ProINS were 11.3 nM and 664.2 nM, respectively. It was expected that ProINS had a
lower affinity than INS. However, the obtained IC
50
of INS was much higher than the
reported data which was within nanomolar range (the dissociation constant K
d
of 10
-9
M)
[22, 90]. Theoretically, the IC
50
value of a competitive binding ligand depends on its own
49
binding affinity (relative to the affinity of the tracer) as well as the concentration of the
radiolabeled tracer used in the assay [91]. Therefore, the high IC
50
for INS could possibly
be due to the different binding affinity between [
125
I-Tyr14]-INS and the INS that were
used in the experiment.
On the other hand, the binding of ProINS-Tf to TfR was also determined and
compared with the binding of Tf in H4IIE cells. Through curve fitting (Fig. 13), the IC
50

value for Tf was 13.8 nM, which was close to the dosing concentration of radiolabeled Tf
(10 nM). ProINS-Tf, with an IC
50
value of 30.3 nM, exhibited a 2.2-fold reduced binding
affinity compared to Tf. In addition, whether the IR binding of the ProINS moiety on
ProINS-Tf would affect the TfR binding of the Tf moiety was also studied by co-
incubating ProINS-Tf with excess amount of INS (300 nM, equivalent to the highest
protein concentration tested) to block the IR on the cell membrane. In the presence of
excess INS, the binding affinity of ProINS-Tf decreased 2.1-fold with an IC
50
of 65.9
nM. This suggested that binding of ProINS moiety to IR, although relatively weak, might
slightly assist the binding of the Tf moiety to TfR. However, no such effect was observed
for Tf alone, as the IC
50
values were similar with or without excess INS (14.1 nM versus
13.8 nM).
50

Fig. 13. Slightly decreased binding of ProINS-Tf to TfR in H4IIE hepatoma cells.
Various concentrations of ProINS-Tf or Tf was co-incubated with 300 nM of INS at 4°C
for 2 h. Cell-associated radioactivity was measured and total cell protein was quantified.
Data were presented as an average of percentage of binding relative to non-treated cells
(n=3).

3.3.2. Promotion Glucose Uptake in Adipocytes
The method of measuring the amount of glucose taken up into differentiated
adipocytes was established and optimized before testing the efficacy of fusion protein.
The protocol was normalized to a known radioactive tracer and the radioactivity detection
efficiency was estimated to be 58%. Different ratio of cell lysates to scintillation fluid
was first tested, and higher amount of cell lysates gave higher cpm values and better
response between control and INS-treated groups (Fig. 14). Thus, the ratio of 1:3 cell
lysates-to-scintillation fluid was chosen for the experiments. In addition, the uptake of
radiolabeled glucose in cells seeded in either 6-well or 12-well plates was also compared.
51
The results from 6-well plates showed more significant stimulating response by INS
compared to control (Fig. 14). Therefore, all the following experiments were performed
in 6-well plates.

Fig. 14. Method validation for glucose uptake assays in adipocytes. The amount
of radiolabeled glucose measured as cpm was normalized to protein amount. Different
ratio of cell lysates to liquid scintillation fluid (1:3, 1:6, 1:9, and 1:15) was compared, and
different density of cells seeded in either 6-well or 12-well plates was also compared.

Various concentrations of ProINS or INS were dosed to adipocytes for 30 min,
and the dose-dependent response curves of these two proteins were shown in Fig. 15.
Apparently, INS exhibited a strong stimulating effect with an EC
50
value of 1.95 nM,
whereas ProINS showed considerably less promotion of glucose uptake. The stimulating
response of ProINS at 100 nM appeared similar to that of INS at 1 nM. This indicated
that ProINS probably retained ~1% activity of INS in promotion of glucose uptake in
vitro, which was consistent with previously reported data [81].
52

Fig. 15. Dose-dependent response of ProINS and INS in promotion of glucose
uptake in adipocytes. Adipocytes were treated with 100 pM, 1 nM, 10 nM and 100 nM of
ProINS or INS, and the amount of radiolabeled glucose taken up into the cells was
measured. Data were presented as an average of cpm normalized by protein amount
(n=3). The stimulatory curves were fitted non-linearly based on log(stimulator) versus
response, and the EC
50
values for proteins were calculated using GraphPad Prism
software.

Furthermore, the activity of ProINS-Tf fusion protein was evaluated and
compared with that of ProINS and INS (Fig. 16). Relative to control, INS showed 6.2-
and 8.8-fold increase in glucose uptake at 10 and 100 nM, respectively. In contrast to
INS, there was 1.3- and 4.3-fold increase by ProINS at these two concentrations. When
ProINS was mixed with Tf in equimolar ratio, the activity was similar to ProINS alone,
suggesting that Tf did not affect the glucose uptake. On the other hand, ProINS-Tf
showed only 1.2-fold increase of glucose uptake at 10 and 100 nM, and both were less
than ProINS. These results indicated that ProINS-Tf had a weak effect in promoting
glucose uptake in adipocytes.
53

Fig. 16. ProINS-Tf exhibited low activity in promotion of glucose uptake in
adipocytes. Adipocytes were treated with 10 and 100 nM of INS, ProINS, ProINS mixed
with Tf in equimolar ratio (shown as "ProINS+Tf"), or ProINS-Tf. Non-treated (shown as
"NT") control group was treated with buffer only. Data were expressed as an average of
percentage of glucose uptake relative to NT group (n=3).

3.3.3. Activation of Proinsulin-Transferrin by Trypsin Digestion
Treating ProINS-Tf with trypsin resulted in a decrease in the molecular weight of
the fusion protein as demonstrated by both anti-Tf and anti-ProINS Western blot (Fig.
17). A shifted band was present between ProINS-Tf (89 kDa) and Tf (80 kDa), and it was
recognized by both antibodies. This indicated that the C peptide was released from the
ProINS moiety [92], and the Tf moiety remained intact and attached.
Furthermore, the ProINS-Tf-TD fusion protein were tested for the short-term
activity of promoting glucose uptake. As demonstrated by Fig. 18, ProINS-Tf-TD
exhibited a significant increased ability to promote glucose uptake compared to ProINS-
Tf alone. The activity increased as the concentration of trypsin increased (from 1:0.1 to
54
1:0.91 in the ratio of ProINS-Tf to trypsin). This result suggested that ProINS-Tf was
activated and became insulin-like-transferrin (INS-like-Tf) after trypsin digestion.

Fig. 17. ProINS-Tf was converted to INS-like-Tf by trypsin digestion. ProINS-Tf
was incubated with various concentrations of trypsin at 37°C for 15 min, followed by the
addition of BBI to deactivate the excess trypsin. The mixtures were applied to anti-Tf or
anti-ProINS Western blot for analysis. Lane 1: marker. Lane 2 and 3: Tf. Lane 4 and 5:
ProINS-Tf. Lane 6 to 10: ProINS-Tf treated with trypsin in the ratio of 1:0.47, 1:0.67,
1:1.34, 1: 2.01, and 1:3.35 (w/w), respectively.

55

Fig. 18. ProINS-Tf was activated by trypsin digestion. ProINS-Tf was treated
with various concentrations of trypsin to obtain ProINS-Tf-TD, which was then applied
to the adipocytes for promotion of glucose uptake. Control group was treated with buffer
only. Data were expressed as an average of percentage of glucose uptake relative to
control group (n=3). Statistical significance determined by Student's t-test was shown by
the asterisk with * indicating p<0.05 and ** indicating p<0.01 compared to ProINS-Tf.

56
3.3.4. Inhibition of Glucose Production in Hepatoma Cells
INS is able to inhibit glucose production in the liver. Therefore, the activity of
ProINS-Tf fusion protein in inhibiting glucose production was determined in H4IIE
hepatoma cells. INS (IC
50
of 308.9 pM) exhibited a 4-fold higher activity than ProINS
(IC
50
of 1291.3 pM). ProINS-Tf, with and IC
50
value of 52.1 pM, showed a 24- and 6-
fold stronger inhibition of glucose production compared to ProINS and INS, respectively.
Since mixing ProINS with Tf in equimolar ratio (IC
50
of 1428.5 pM) showed similar
inhibition as ProINS alone, the enhanced inhibitory activity by ProINS-Tf was due to the
fusion of the two moieties and may be attributed to Tf binding to TfR (Fig. 19, Table 3).

Fig. 19. Enhanced inhibition of glucose production by ProINS-Tf in H4IIE
hepatoma cells was TfR-mediated. Different concentrations of INS, ProINS, ProINS
mixed with Tf in the equimolar ratio (shown as "ProINS+Tf"), ProINS-Tf fusion protein
was incubated with hepatoma cells for 24 h followed by the measurement of glucose
production levels. To determine the blocking effects of excess Tf or BSA, ProINS-Tf was
co-incubated with 1000-fold of Tf or BSA shown as "ProINS-Tf+1000Tf" or "ProINS-
Tf+1000BSA". All data were expressed as average glucose output levels normalized by
protein amount, with error bars indicating the standard deviation (n=3).
57

In order to demonstrate the contribution of Tf fusion to the potent activity of
ProINS-Tf, a co-incubation experiment with excess Tf was performed. Cells were first
treated with excess Tf (1000-fold) for 30 min to pre-occupy the TfR on the cell surface,
and then ProINS-Tf mixed with 1000-fold excess of Tf was incubated with cells for 24 h.
Following the incubation, the glucose production levels were measured. As shown in Fig.
19, co-incubating with 1000-fold excess of Tf resulted in a significant reduction in the
inhibitory ability of ProINS-Tf, whereas co-incubation with 1000-fold excess of BSA did
not significantly changed the IC
50
value (33.6 pM versus 52.1 pM in the presence or
absence of excess BSA, Table 3).

Table 3. IC
50
values for various proteins in inhibiting glucose production in H4IIE
hepatoma cells.

Protein IC
50
(pM)
INS 308.9±19.7 *
ProINS 1291.3±148.5 **
ProINS+Tf (equimolar) 1428.5±677.3 *
ProINS-Tf 52.1±8.48
ProINS-Tf+1000Tf > 1000
ProINS-Tf+1000BSA 33.6±6.13 *

Data were presented as averages ± standard deviation (n=3). Statistical
significance determined by Student's t-test was shown by the asterisk with * indicating
p<0.05 and ** indicating p<0.01 compared to ProINS-Tf.

58
3.3.5. Kinetic Stimulation of Akt Phosphorylation in Hepatoma Cells
Upon binding to IR on the cell membrane, INS activates the IR signaling by
stimulating the PI3 kinase pathway, during which Akt protein is phosphorylated [86].
Therefore, the kinetics of Akt phosphorylation stimulated by ProINS-Tf were measured
and compared to that stimulated by ProINS or INS.
As shown in Fig. 20, INS showed a very fast onset with the phospho-Akt reaching
the maximal level following 5 min of incubation, and the Akt phosphorylation
subsequently decreased in a time-dependent manner as recorded at 4 and 12 h. In contrast
to INS, ProINS exhibited very minimal stimulating effect. In spite of the slightly
increased phospho-Akt level stimulated by ProINS after 24 h of incubation, it was
nevertheless remarkably lower than that by INS. On the other hand, ProINS-Tf fusion
protein elicited a distinct kinetic pattern compared to both INS and ProINS. Similar to the
ProINS-treated cells, ProINS-Tf-treated cells exhibited a very low phospho-Akt level at 5
min of short-term incubation. However, the Akt phosphorylation significantly enhanced
at 4 and 24 h of incubation, both of which were higher than the respective levels induced
by ProINS alone.
59

Fig. 20. ProINS-Tf elicited a delayed but prolonged stimulation of Akt
phosphorylation in H4IIE hepatoma cells. Serum-starved cells were treated with 100 pM
of ProINS-Tf, ProINS, or INS at 37°C for 5 min, 4 h, and 24 h, respectively. Cells were
lysed in the presence of protease and phosphatase inhibitors. Cell lysates were resolved in
10% SDS-PAGE and immunoblotted against phospho-Akt and beta-actin antibodies.
Relative signal intensities were quantified by densitometry. Results were presented as the
relative intensity of phospho-Akt normalized by beta-actin. Data were expressed as the
average of relative intensity, with error bars indicating the standard deviation (n=3).

60
3.4. Discussion
ProINS-Tf fusion protein showed a very low activity in promoting glucose uptake
in differentiated adipocytes where, at the concentration of 10 nM, its activity was only
90% and 19% of ProINS and INS, respectively (Fig. 16). On the other hand, ProINS-Tf
exhibited a notably strong activity in inhibiting glucose production in hepatoma cells
where it was 24-fold more potent than ProINS and even 6-fold stronger than INS. In
addition, this strong activity of ProINS-Tf was TfR-mediated, since it was considerably
blocked by the presence of excess Tf (Fig. 19 and Table 3). Short-term activities usually
reflect the intrinsic potencies which mainly depend on the receptor binding affinities. It is
reasonable that ProINS-Tf is as weak as, or even weaker than, ProINS, because ProINS-
Tf is a ProINS-based fusion protein. However, the TfR-mediated enhanced potency of
ProINS-Tf in hepatoma cells, which was dramatically distinct from the activity in
adipocytes, is a very interesting observation that deserved in-depth investigation.
Measurements of these two biological activities relate to responses from different
cell types (adipocytes vs. hepatoma cells). Promotion of glucose uptake assay was carried
out in differentiated adipocytes where much higher abundance (20-40 fold) of IR is
present on cell surface than TfR [93, 94]. On the other hand, inhibition of glucose
production assay was performed in hepatoma cells where comparable abundance of IR
and TfR has been demonstrated [95]. In addition, the incubation duration of these two
assays is distinct. Promoting glucose uptake only requires a 30 min incubation, whereas
inhibiting glucose production needs a 24 h treatment.
61
First of all, it is well-accepted that bivalent binding in fusion proteins may
promote the binding affinity of one moiety due to the binding of the other. The binding
affinity of Tf to TfR is within the nanomolar range (K
d
of 10
-9
), which is similar to that of
INS to IR but about two orders of magnitude stronger than that of ProINS to IR (K
d
of
10
-7
) [22, 82]. Thus, the observed TfR-mediated enhanced potency of ProINS-Tf in
hepatoma cells may be attributed to an enhanced IR binding of ProINS-Tf due to Tf-TfR
binding. It would occur in hepatoma cells but not adipocytes, because the former shows
similar abundance of IR and TfR. Our competitive binding assay demonstrated that the
binding of ProINS-Tf to IR appeared to slightly facilitate its binding to TfR (Fig. 13).
However, no improved binding of ProINS-Tf to IR was demonstrated, since ProINS-Tf
behaved similar to ProINS based on the binding curves below 10 nM concentration (Fig.
12). Therefore, the IR binding results of ProINS-Tf ruled out the possibility of the
enhanced bivalent binding hypothesis in hepatoma cells.
Another interpretation for the enhanced activity of ProINS-Tf in hepatoma cells
regards to the relationship of biological potency with the protein intracellular processing
following receptor binding, as illustrated in Scheme 1. Several studies consistently
reported that, compared to INS, the reduced intracellular degradation of ProINS
accounted for its slower cellular extraction and thus the prolonged effects on IR
endocytosis and IR phosphorylation in vitro and in vivo [80, 96, 97]. This explains only
6-fold difference between ProINS and INS in inhibiting glucose production whereas
about 100-fold difference in promoting glucose uptake. Accordingly, it is likely that the
strong binding of Tf to TfR directs ProINS-Tf to the Tf-TfR recycling pathway, thereby
62
bypassing the degradation machinery (endosomes and lysosomes) for ProINS moiety.
Therefore, it may allow a prolonged exposure of ProINS-Tf to the cells and consequently
improve its potency during the 24 h long-term incubation with hepatoma cells.
Contrarily, this prolonging effect fails to become evident in promoting glucose uptake,
since the uptake assay is measured as a short-term response. As a result, the ability of
ProINS-Tf to promote glucose uptake has to solely rely on its IR binding capability, as
demonstrated by the good correlation of these two results and the trypsin digestion
studies (Fig. 12 and Fig. 16). Taken together, the prolonged exposure of ProINS-Tf may
compensate for its low IR binding affinity, and this can only be demonstrated during a
long-term but not short-term response.
On the other hand, the delayed onset of stimulation of Akt phosphorylation by
ProINS-Tf suggests that the TfR-mediated enhanced potency may result from an
activation of ProINS-Tf to INS-like-Tf. The confirmation of the activation mechanism is
discussed in Chapter 4.

63

Scheme 1. The biological potency of protein drugs depends on both receptor
binding affinity and intracellular processing fate. (1) Higher receptor binding ability
accounts for more potent biological activity. (2) Receptor binding results in endocytosis
to endosomes where proteins are subjected to proteolytic degradation (cellular
extraction). (3) This increased cellular extraction of proteins reduces total drug exposure
(equivalent to shortened in vivo half-life), which eventually weakens the biological
potency.

64
3.5. Summary
In this chapter, ProINS-Tf fusion protein have been evaluated in vitro for the INS-
related biological activities using four independent cell-based assays in two types of cell
models. Compared to INS, ProINS-Tf exhibited a very low binding affinity to IR, which
was similar to ProINS. Correspondingly, both ProINS-Tf and ProINS had little effect in
promoting glucose uptake into adipocytes. In addition, trypsin digestion was able to
convert ProINS-Tf to an INS-like-Tf, and the latter exhibited an increased promotion of
glucose uptake. These results indicated that the intrinsic potency of ProINS-Tf is very
low, since it primarily depended on its receptor binding affinity. On the contrary, ProINS-
Tf elicited a significantly potent activity in inhibiting glucose production following a
long-term incubation with hepatoma cells, which was 24- and 6-fold stronger than
ProINS and INS, respectively. More importantly, this potent activity was specifically
mediated through Tf-TfR interaction in hepatoma cells. In addition, stimulation of Akt
phosphorylation by ProINS-Tf demonstrated a delayed but enhanced kinetic pattern.
Therefore, this TfR-mediated enhanced activity in hepatoma cells can potentially be
interpreted as either a TfR-related prolonged exposure of ProINS-Tf to hepatoma cells or
a TfR-mediated activation of ProINS-Tf in hepatoma cells. These potential causes are
further evaluated in the rest of this dissertation.

65
CHAPTER 4: TRANSFERRIN RECEPTOR MEDIATED INTRACELLULAR
CONVERSION OF PROINSULIN-TRANSFERRIN TO AN ACTIVE FORM OF
INSULIN IN HEPATOMA CELLS

4.1. Background
With the intention of developing ProINS as an alternative INS analogue, scientists
have executed numerous preclinical and clinical investigations to explore ProINS's
therapeutic potential back in the 1990s [36]. PK studies have demonstrated that both the
distribution and elimination half-life of ProINS were longer than INS [35, 98]. In
addition, ProINS is found marginally hepatospecific, and this selective action in the liver
is recognized as an important benefit in the treatment of diabetes [99, 100]. However,
ProINS has ~1-10% potency compared to INS, and it has demonstrated low conversion to
the active form INS [34]. Therefore, high doses of ProINS are required to achieve a
biological effect, which has led to detrimental side effects and subsequent withdrawal
from clinical trials [36].
In this dissertation, a ProINS-Tf fusion protein is designed and developed to take
advantage of Tf properties to overcome some of the challenges encountered with ProINS.
A unique characteristic of Tf-TfR endocytic pathway is the recycling of Tf with TfR.
After endocytosis into acidic endosomal compartments, apo-Tf remains bound to the TfR
and is recycled back to the cell surface intact, instead of being routed to lysosomes for
degradation [101, 102]. In addition, the recycling process can occur via either a slow or a
fast recycling route [42, 101, 102]. Through the slow recycling route, Tf can be delivered
to intracellular compartments such as the TGN [46, 103]. Many studies have found that
endocytosed Tf also merges with vesicles of the protein secretory pathway located at the
66
TGN [46, 49, 50, 103]. The close proximity would potentially allow access of the
endocytosed Tf to secretory proteases that are responsible for prohormones conversion.
To our knowledge, this intracellular trafficking route and the final release step of Tf from
TfR following recycling have not yet been taken advantage of in the development of Tf
fusion proteins.
Results from Chapter 3 have clearly shown a TfR-mediated enhanced potency of
ProINS-Tf fusion protein in hepatoma cells. One hypothesis is the activation of ProINS-
Tf through the Tf-TfR pathway. Therefore, in this chapter, a conversion and activation of
ProINS-Tf are evaluated and a hypothetical model for the TfR-mediated intracellular
activation mechanism is proposed [104].

4.2. Materials and Methods
4.2.1. Chemicals
Ammonium chloride, chloroquine, nocozadole, and brefeldin A were all in
chemical grade and purchased from Sigma.

4.2.2. Cell Cultures
Caco-2 cells were cultured in DMEM medium supplemented with 20% FBS, 2
mM L-glutamine, 0.1 mM non-essential amino acids, and 50 U/mL penicillin - 50 μg/mL
streptomycin.

67
4.2.3. Enterocyte-like Caco-2 Cell Differentiation
Differentiation of Caco-2 cells was carried out in either regular or Trans-well 6-
well plates. Cells were seeded in plates and waited for 1-2 days to reach confluence.
Upon 100% confluence, cells were further cultured for another 2-3 weeks for
differentiation, during which cells were fed with fresh medium every other day. The
transepithelial electrical resistance (TEER) was measured with an epithelial voltage ohm
meter (World Precision Instruments, Sarasota, FL).

4.2.4. Proinsulin-specific Radioimmunoassay
Cells were treated with 100 pM of ProINS-Tf, in the presence or absence of 1000-
fold excess of Tf or BSA. Culture media were collected at different incubation
timepoints, and centrifuged to obtain the clear supernatants. To measure the remaining
ProINS-Tf in the cultured medium, the media were applied to a human ProINS-specific
radioimmunoassay (RIA) from Millipore according to the manufacturer’s instructions.
Quantification of ProINS-Tf was based on a standard curve generated by 10, 30, 100, and
300 pM of ProINS-Tf. This ProINS-specific RIA has less than 0.1% cross-reactivity with
human INS.

4.2.5. Insulin-specific Radioimmunoassay
Cells were treated with either ProINS-Tf or ProINS, and media were collected at
various timepoints after incubation. To determine the generated immunoreactive insulin-
transferrin (irINS-Tf) in the culture medium, the media were subjected to a human INS-
68
specific RIA (Millipore) based on the manufacturer’s instructions. Quantification of
irINS-Tf or irINS was based on a standard curve generated by 12, 30, 60, 120, 300, 600,
and 1200 pM of an INS standard (Millipore). This INS-specific RIA has less than 0.2%
cross-reactivity with human ProINS.

4.2.6. Lactate Dehydrogenase Release Assay
H4IIE cells were treated with ProINS-Tf alone, or in the presence of excess Tf or
BSA. Following 24 h of incubation, media were collected and centrifuged, and the
supernatants were applied to the lactate dehydrogenase (LDH) release assay. The LDH
levels in the media were measured using the commercial CytoTox-One Homogenous
Membrane Integrity Assay (Promega, Madison, WI) according to the manufacturer's
instructions.

4.2.7. Measurement of Alkaline Phosphatase Activity
Caco-2 cells were collected on different days during differentiation period by
trypsinization. Cell numbers were counted by Beckman Coulter Cell and Particle
Counters (Beckman Coulter, Brea, CA). Alkaline phosphatase (ALP) activity was
measured in an absorbance-based assay. After washing with PBS, cell pellets (250,000
cells per tube) were resuspended in 100 μl substrate buffer containing 10 nM
diethanolamine (Sigma), 0.5 mM MgCl
2
, 1 mg/ml para-nitrophenylphosphate (Sigma) at
pH 10.5. Cells were then incubated at room temperature for 20 min, followed by the
69
addition of 50 μl of 2M NaOH to stop the reaction. The solutions were loaded onto a 96-
well microplate and the absorbance was measured at 405 nm.

4.2.8. Statistical Analysis
All the experiments were performed with n=3 for each condition. Student's t-test
was used for statistical analysis, where differences with p values less than 0.05 were
considered statistically significant.

4.3. Results
4.3.1. Conversion of Proinsulin-Transferrin to Immunoreactive Insulin-Transferrin in
Hepatoma Cells
4.3.1.1. Decrease in Detection of Proinsulin-Transferrin Following Incubation with
Hepatoma Cells
When 100 pM of ProINS-Tf was incubated with H4IIE hepatoma cells, a gradual
decrease of ProINS-Tf was observed during the 24 h incubation. Detected by a ProINS-
specific RIA with less than 0.1% cross-reactivity to human INS, only 30% and 19% of
the fusion protein remained in the cell treatment medium after 12 and 24 h, respectively.
However, ProINS-Tf was relatively stable when incubated in cell-free medium.
Interestingly, the loss of ProINS-Tf during H4IIE incubation was significantly blocked by
adding excess Tf, leading to 91% remaining in the medium after 12 h and 61% after 24 h.
In contrast to Tf, the presence of excess BSA had no apparent effect on the changes in
ProINS-Tf concentration (Fig. 21).
70
A LDH release assay was performed in order to demonstrate that the cell
membrane remained undamaged during ProINS-Tf incubation. As shown in Fig. 22, no
significant difference in the levels of released LDH was observed in the treatment media
at 24 h timepoint compared to the non-treated control group.

Fig. 21. ProINS-Tf is progressively decreased during incubation with H4IIE
hepatoma cells. ProINS-Tf, in the presence or absence of 1000-fold excess of Tf or BSA
("ProINS-Tf+1000Tf" or "ProINS-Tf+1000BSA"), were incubated at 37°C with either
H4IIE cells or blank wells ("Non-Cell"). Culture media were collected at various
timepoints (0.5, 1, 3, 6, 9, 12, and 24 h) and applied to ProINS-specific RIA to measure
the concentration of ProINS-Tf. ProINS-Tf was quantified based on a standard curve of
ProINS-Tf according to the manufacturer’s instructions. Results were expressed as
average from three treatment groups with error bars indicating the standard deviation
(n=3). Data were analyzed using the Student’s t-test and marked with asterisks (*) to
indicate statistically significant differences compared to ProINS-Tf (*p<0.05, **p<0.01).

71

Fig. 22. Cell membrane integrity remained intact following ProINS-Tf treatment
in the presence or absence of excess Tf or BSA. Cells were treated with ProINS-Tf with
or without 1000-fold excess of Tf or BSA for 24 h, and the released LDH levels were
measured from culture medium. Data were normalized to non-treated groups, and
expressed as average with error bars indicating the standard deviation (n=3).

4.3.1.2. Increase in Detection of Immunoreactive Insulin-Transferrin Following
Incubation with Hepatoma Cells
In order to determine whether the gradual decline of ProINS-Tf was attributed to
a gradual conversion of ProINS-Tf to irINS-Tf during incubation with H4IIE hepatoma
cells, an INS-specific RIA with less than 0.2% cross-reactivity to human ProINS was
applied to detect the irINS-Tf in the cell treatment medium. As shown in Fig. 23, both 1
and 10 nM of ProINS-Tf showed considerable conversion to irINS-Tf during the 24 h
incubation with H4IIE cells. The conversion efficiency for 10 nM of ProINS-Tf was 8.3
± 0.77 % at 24 h, whereas the efficiency for 1 nM of ProINS-Tf was 21.6 ± 0.13 % at 24
h of incubation. It appeared that a lower concentration of ProINS-Tf showed a higher
conversion capacity. Therefore, the conversion efficiency might be higher than 21.6% for
72
100 pM of ProINS-Tf, which may explain the 80% of loss in ProINS-Tf at 24 h observed
in Fig. 21. However, no apparent conversion to irINS was observed for ProINS alone
under both 1 and 10 nM. In addition, equivalent counterparts from mock-transfection
(pcDNA3.1 vector only) also showed very little irINS-containing components in the
treatment medium. Therefore, these results indicated that there was no interference from
the protein impurities.

Fig. 23. ProINS-Tf was converted to irINS-Tf in H4IIE hepatoma cells. H4IIE
cells were incubated with either 1 nM (A) or 10 nM (B) of ProINS or ProINS-Tf at 37°C.
Counterpart solutions from mock transfection (transfected with pcDNA3.1 vector only),
in the equivalent amount to 10 nM of ProINS-Tf, was also incubated with H4IIE cells (B,
open square, dotted line). Cell treatment media were collected at the indicated timepoints
and subjected to human INS-specific RIA. To quantify the concentration of irINS or
irINS-Tf, a standard curve of INS was used according to the manufacturer’s instructions.
Results were expressed as average from three treatment groups with error bars indicating
the standard deviation (n=3).

As shown in Fig. 24A, 10 nM of ProINS-Tf had minimal irINS-containing
components at the initial time of incubation, implying that ProINS-Tf did not undergo
significant conversion to irINS-Tf during the production period in HEK293 cells. After 2
h of incubation, irINS-Tf was gradually detected and increased until up to 24 h of
73
incubation. In accordance with the TfR-mediated loss of ProINS-Tf detected by ProINS-
specific RIA, the generation of irINS-Tf was also remarkably suppressed by co-
incubating with excess Tf, but not affected by the addition of excess BSA. This suggested
that the conversion of ProINS-Tf to irINS-Tf was a TfR-mediated process. The cell-
treatment media after 24 h of incubation with 10 nM ProINS-Tf was also analyzed by
anti-Tf Western blot. Lane 3 and 4 in Fig. 24B showed that no bands corresponding to
free Tf were detected in the cell-treatment media at two different loading volumes,
indicating that the conversion products were irINS-Tf instead of irINS alone.

Fig. 24. Conversion of ProINS-Tf to irINS-Tf was a TfR-mediated process. (A)
H4IIE cells were incubated with 10 nM of ProINS in the presence or absence of 1000-
fold excess of Tf or BSA. Cell treatment media were collected at the indicated timepoints
and applied to human INS-specific RIA. Results were expressed as average from three
treatment group with error bars indicating the standard deviation (n=3). (B) ProINS-Tf
(10 nM) was incubated with H4IIE cells for 24 h. The collected cell treatment medium
was subjected to 8% non-reducing SDS-PAGE, following by anti-Tf immunoblotting.
Lane 1: human Tf (20 ng). Lane 2: ProINS-Tf fusion protein (13.4 ng, 15 μL). Lane 3:
H4IIE treated media (15 μL). Lane 4: H4IIE-treated media (30 μL).

74
4.3.1.3. Mechanism Studies of Proinsulin-Transferrin Conversion
To further understand the TfR-mediated conversion of ProINS-Tf to irINS-Tf in
hepatoma cells, this conversion process was evaluated by drug intervention that can alter
the intracellular protein trafficking and processing. The drug concentrations were tested
to make sure that the cell viability was not significantly compromised due to cytotoxicity.
As shown in Fig. 25 A and B, compared to the non-treated ProINS-Tf, 86.4% and
83.0% of the irINS-Tf conversion was inhibited when ProINS-Tf was co-incubated for 8
h with the lysosomotropic agents ammonium chloride (20 mM) and chloroquine (50 μM),
respectively. Similarly, the inhibition was 71.0% and 72.5% in the presence of
ammonium chloride and chloroquine at 12 h of incubation, respectively. In addition,
adding 80 μM of the microtubule-disrupting reagent nocozadole exhibited 50.9% and
37.2% of irINS-Tf conversion at 8 and 12 h, respectively. However, co-incubating
ProINS-Tf with brefeldin A (9.6 μg/mL) did not show significant change in irINS-Tf
conversion. Furthermore, Fig. 25C and D demonstrated that the conversion was
temperature-dependent, as incubating the cells at 16°C almost completely abolished the
conversion at both 8 and 12 h of incubation. These results suggested that the conversion
of ProINS-Tf was a pH-sensitive, microtubule- and temperature-dependent process.
75

Fig. 25. Conversion of ProINS-Tf was a pH-sensitive, microtubule- and
temperature-dependent process. ProINS-Tf (10 nM) was incubated in the absence or
presence of 20 mM ammonium chloride, 50 μM chloroquine, 80 μM nocodazole, or 9.6
μg/mL brefeldin A at 37°C for 8 h (A) and 12 h (B). ProINS-Tf (10 nM) was incubated at
37°C or 16°C for 8 h (C) and 12 h (D). The converted irINS-Tf in each treatment group
was measured by INS-specific RIA. All the data were expressed as the average
percentage of irINS-Tf conversion relative to ProINS-Tf without treatment. Data were
analyzed using Student’s t-test and marked with asterisks (*) to indicate statistically
significant differences (*p<0.05, **p<0.01).

76
4.3.2. Activation of Proinsulin-Transferrin in Hepatoma Cells
The previous section described a TfR-mediated conversion of ProINS-Tf to
irINS-Tf in H4IIE hepatoma cells. Correspondingly, ProINS-Tf fusion protein exhibited a
TfR-mediated enhanced activity in inhibiting glucose production in H4IIE cells (24-fold
higher than ProINS), despite its low intrinsic activity in promoting glucose uptake in
differentiated adipocytes (Chapter 3). These results suggested an activation of ProINS-Tf
due to its conversion to irINS-Tf during the incubation with H4IIE cells. Therefore,
several short-term activities of the H4IIE-pretreated ProINS-Tf were examined.
The first experiment was to examine whether the receptor binding abilities
increased after H4IIE incubation. Compared with ProINS-Tf alone, H4IIE-pretreated
ProINS-Tf exhibited an increased ability to compete with the radiolabeled tracer [
125
I-
Tyr14]-INS in IR binding. The increase was significant but relatively small, since the
pretreatment with H4IIE cells was only 1 h at 37°C. In contrast to ProINS-Tf, no
significant changes in the binding ability of ProINS were noticed (Fig. 26A). In addition,
this increased competitive binding was also temperature-dependent, since it only
occurred when the pretreatment was carried out at 37°C but not 4°C (Fig. 26B).

77

Fig. 26. H4IIE-pretreated ProINS-Tf exhibited an increased binding to IR. (A)
Following DMEM/0.1%BSA incubation for 30 min to remove serum, H4IIE cells were
first incubated with either buffer or 10 nM of ProINS/ProINS-Tf at 37°C for 1 h (shown
as "H4IIE-treated ProINS" or "H4IIE-pretreated ProINS-Tf"). Cells were then
equilibrated at 4°C for 15 min. Buffer-treated cells were incubated with [
125
I-Tyr14] INS
plus 10 nM of ProINS or ProINS-Tf (shown as "ProINS" or "ProINS-Tf"), while the two
pretreated group were incubated with [
125
I-Tyr14] INS only. Binding assays were the
same as described in Section 3.2.3. (B) The pretreatment was the same as in (A), except
that the pretreatment was carried out in either 37°C or 4°C (shown as "H4IIE-treated
37°C" or "H4IIE-pretreated 4°C". Results were expressed as cpm normalized by the
protein quantification. Data were analyzed using Student’s t-test and marked with
asterisks (*) to indicate statistically significant differences (*p<0.05, **p<0.01).

To further study the activation of ProINS-Tf in H4IIE cells over the 24 h
incubation, H4IIE cells were pretreated with either ProINS or ProINS-Tf for 24 h, and
then the cell treatment media were directly added to freshly starved H4IIE cells again for
a 5 min short-term incubation. Following the 5 min treatment, phospho-Akt levels were
measured and compared. Fig. 27 demonstrated that, compared with the blank-treated
counterparts, H4IIE-treated ProINS-Tf elicited a remarkably increased Akt
phosphorylation. On the contrary, neither H4IIE-pretreated ProINS nor H4IIE-pretreated
DMEM exhibited a significant increase, relative to the respective blank-pretreated
controls.
78

Fig. 27. H4IIE-pretreated ProINS-Tf elicited a significant increase in the 5 min
stimulation of Akt phosphorylation. ProINS, ProINS-Tf, or DMEM was first
preincubated with H4IIE cells for 24 h at 37°C. The collected cell treatment media were
immediately applied to fresh acquiescent H4IIE cells for 5 min incubation, and the Akt
phosphorylation was determined as described in Section 3.2.8. Results were quantified by
densitometry and presented as the relative intensity of phospho-Akt normalized by beta-
actin. Data were expressed as the average of relative intensity, with error bars indicating
the standard deviation (n=3). Statistical differences was marked with * indicated p<0.05,
and ** indicated p<0.01.

79
Lastly, ProINS-Tf that underwent 24 h preincubation with H4IIE cells was also
tested for the ability to promote glucose uptake in differentiated adipocytes. The H4IIE-
pretreated ProINS-Tf showed a 2-fold greater promotion of glucose uptake than the
blank-treated fusion protein. Interestingly, the glucose uptake levels stimulated by the
H4IIE-treated ProINS-Tf approximated to that stimulated by 1 nM of INS. Conversely,
no significant increase was observed in DMEM groups (Fig. 28).

Fig. 28. H4IIE-pretreated ProINS-Tf exhibited a significant increase in promoting
glucose uptake in adipocytes. ProINS-Tf or DMEM was preincubated with H4IIE cells
for 24 h, and the collected cell treatment media were applied to a 30 min glucose uptake
assay as described in Section 3.2.6. Non-treated group (shown as "NT") was treated with
buffer only. Data were expressed as an average of percentage of glucose uptake relative
to the non-treated group (n=3). Statistical differences was marked with * indicated
p<0.05, and ** indicated p<0.01.

80
4.3.3. Conversion of Proinsulin-Transferrin to Immunoreactive Insulin-Transferrin in
Caco-2 Cells
4.3.3.1. Conversion in Caco-2 Cells Seeded in Regular Plates
The conversion of ProINS-Tf fusion protein was further evaluated in Caco-2 cells
in addition to H4IIE hepatoma cells. Caco-2 cell differentiation during continuous culture
was examined by measuring the ALP activity on the cell surface. The ALP activity
significantly enhanced after 6 days of culture and reached the highest levels on day 12.
Thus, the ALP biomarker result suggested that the differentiated Caco-2 cell model in
this study exhibited enterocyte-like phenotype (Fig. 29A). Fig. 29B demonstrated that the
irINS-Tf conversion occurred only in the differentiated enterocyte-like Caco-2 cells but
not the regular Caco-2 cells. Furthermore, similar to H4IIE cells, the conversion of
ProINS-Tf was also a TfR-mediated process in differentiated Caco-2 cells, since it was
significantly blocked by the co-incubation with excess Tf but not BSA (Fig. 30).

81

Fig. 29. ProINS-Tf was converted to irINS-Tf in differentiated Caco-2 cells. (A)
ALP activity during Caco-2 differentiation upon confluence. Data were expressed as the
OD
405
values normalized by cell number (n=3). (B) For conversion measurement, 10 nM
of ProINS-Tf was incubated with either differentiated or regular Caco-2 cells for up to 24
h. The cell treatment media were collected at the indicated timepoints and the irINS-Tf
was quantified (n=3).

Fig. 30. Conversion of ProINS-Tf in differentiated Caco-2 cells was a TfR-
mediated process. ProINS-Tf (10 nM), in the presence or absence of excess Tf or BSA,
was incubated with differentiated Caco-2 cells. The converted irINS-Tf was quantified
from the collected treatment media (n=3).

82
4.3.3.2. Conversion in Caco-2 Cells Seeded in Trans-well Plates
In addition, the conversion of ProINS-Tf to irINS-Tf was also preliminarily
assessed when Caco-2 cells were seeded in Trans-well plates. Upon confluence, cell
monolayer was continuously grown in the plate for 2 or 3 weeks to form steady tight
junctions, and TEER was monitored every other day. The resistance increased from 80 Ω
cm
2
on day 2 to more than 400 Ω cm
2
on day 14, and remained stable after 14 days.
ProINS-Tf fusion protein was dosed on the apical side for transcytosis to the basolateral
side, or vice versa. Fig. 31 demonstrated that, in both directions, there was a slight
increase in the transcytosed irINS-Tf from 2 to 6 h followed by a significant increase
from 6 to 8 h. However, the amounts of irINS-Tf conversion detected during Caco-2
transcytosis in both directions were considerably lower than in regularly-seeded Caco-2
cells and hepatoma cells. The resistance remained unchanged at the end of the 8 h
incubation period, indicating that there was not significant leak of ProINS-Tf across the
monolayer.

83

Fig. 31. ProINS-Tf was slightly converted to irINS-Tf during transcytosis in
differentiated Caco-2 cells. Caco-2 monolayer was grown in Trans-well units. ProINS-Tf
(10 nM) was dosed to either the apical side (upper chamber) or basolateral side (lower
chamber) of the cells, and medium containing 1% BSA was added to the other side for
detection of transcytosed irINS-Tf. Media were collected at the indicated timepoints and
applied for irINS-Tf measurement. Data were presented as the average of three individual
wells with error bars indicating the standard deviation.

4.4. Discussion
During the 24 h incubation in H4IIE hepatoma cells, a progressive decline of
ProINS-Tf correlated with an increase in detection of irINS -Tf (Fig. 21 and 24). Thus, it
is believed that ProINS-Tf was converted to irINS-Tf by H4IIE cells. It has been shown
that ProINS-Tf had little effect in short-term INS-related activities including IR binding,
5 min stimulation of Akt phosphorylation, and 30 min promotion of glucose uptake
(Chapter 3). However, compared to the blank-treated counterparts, the H4IIE-pretreated
ProINS-Tf demonstrated enhanced activities in all of these three measurements (Fig. 26-
28). These findings indicated that the converted irINS-Tf was an active form that
accounted for the enhanced potency observed in both inhibition of glucose production
84
and stimulation of Akt phosphorylation (Chapter 3). Only long-term activities can benefit
from this conversion, because the conversion takes some time to produce sufficient
amount of irINS-Tf (Fig. 23). Conversely, short-term activities failed to allow enough
conversion, thus they predominantly depend on the intrinsic IR binding ability of
ProINS-Tf (Chapter 3).
The conversion was found to be inhibited in the presence of excess Tf, but not
excess BSA which blocked non-specific binding on cell membrane (Fig. 24). In addition,
the generation of irINS-containing components was only detected in treatment media
incubated with ProINS-Tf but not ProINS (Fig. 23). Therefore, the conversion of ProINS-
Tf to irINS-Tf was a Tf-TfR mediated process. As discussed earlier in Chapter 3, the
binding affinity of Tf to TfR is about 100-fold higher than that of ProINS to IR. Thus the
strong Tf-TfR binding may direct the endocytosis of ProINS-Tf to the Tf-TfR pathway
rather than the INS-IR pathway. The Tf-TfR pathway eventually allows the endocytosed
proteins to be released out of the cells, since it undergoes recycling instead of lysosomal
degradation [46, 102].
It is believed that the conversion is an intracellular processing event by certain
convertases in the TfR-associated intracellular compartments, instead of a non-specific
degradation by proteases inside the cell or on the cell membrane. This hypothetical
mechanism for ProINS-Tf conversion (Scheme 2) is supported by the following
observations. First, ProINS alone, which can only be internalized into the IR pathway,
was not converted to irINS after incubation (Fig. 23). Second, blocking the TfR on the
cell membrane by excess of Tf apparently inhibited both the loss of ProINS-Tf and the
85
generation of irINS-Tf (Fig. 21 and 24). Third, no increase of LDH leakage was detected
following 24 h incubation with ProINS-Tf (Fig. 22), which suggested that the cell
membrane integrity remained intact and no intracellular proteases were released to the
media. Fourth, a drastic inhibition of the irINS-Tf conversion occurred in the presence of
several reagents (or condition) that interfered with the Tf-TfR endocytosis and recycling
pathway, including ammonium chloride, chloroquine, nocodazole, and incubation at
16°C (Fig. 25).
Based on the well-accepted fast recycling and slow recycling routes of Tf-TfR
pathway, it is mostly likely that the conversion of ProINS-Tf to irINS-Tf occurred during
the slow recycling route (Scheme 2). This is supported by the following findings. First,
the conversion was remarkably suppressed by lysosomotropic agents (Fig. 25A and B),
which raised the vesicular pH and thus inhibited vesicular compartment fusion events that
primarily occur in the slow recycling pathway [105]. Second, the complete eradication of
conversion at 16°C (Fig. 25C and D) strongly demonstrated the involvement of
compartments in the late stage of endocytosis [105]. Third, the conversion was
suppressed by nocodazole, and the association of microtubule indicated that it might
occur far away from the cell membrane. Third, a lag time of 2 h was found in the
detection of irINS-Tf following incubation (Fig. 24A), and this was longer than expected
for the fast recycling which takes only 10 min per cycle [45].
It has been demonstrated by several groups that the recycling vesicles in the TfR
slow recycling pathway connected with the vesicles from protein secretory pathway at the
TGN in various liver cells [50, 95, 103]. In addition, various endoproteases including the
86
subtilisin-like prohormone convertase family are predominantly localized in the TGN,
and these convertases are accountable for the proprotein processing routed in the
secretory pathways [106-108]. Therefore, the TfR-mediated trafficking of ProINS-Tf
may have allowed the access of endocytosed ProINS-Tf with those endoproteases from
the secretory compartments and thus the conversion to irINS-Tf. However, further
investigation is needed to identify the enzyme(s) that are responsible for ProINS-Tf
conversion.
Lastly, the occurrence of ProINS-Tf conversion in Caco-2 cells are very exciting.
Differentiated Caco-2 cells exhibit an enterocyte-like phenotype, and this cell line is a
well-accepted epithelial model for evaluation of drug transport across GI epithelium
[109]. The selective irINS-Tf conversion in differentiated but not regular Caco-2 cells
presents a great possibility of ProINS-Tf activation during enterocyte transcytosis, which
indicates a potential advantage as an oral delivery agent. Furthermore, this result also
suggests that ProINS-Tf conversion is a particular process and does not just occur in any
type of cells, despite the ubiquitous TfR pathway. Currently, it has only been
demonstrated in hepatoma cells and differentiated Caco-2 cells.

87

Scheme 2. A hypothetical model for the intracellular activation of ProINS-Tf to
irINS-Tf in hepatoma cells. Following TfR binding, ProINS-Tf/TfR complex is
endocytosed to the early endosome, and undergoes TfR-mediated recycling pathway.
During the fast recycling, ProINS-Tf is rapidly recycled and released as an intact protein.
During the slow recycling, ProINS-Tf is delivered to the slow recycling compartments
which merge with vesicles from protein secretory pathway at the TGN, thereby allowing
ProINS-Tf to access to secretory proteases that can convert ProINS-Tf to irINS-Tf. The
converted irINS-Tf is subsequently recycled and released into the culture media. Reduced
conversion efficiency was observed under treatment (a) with lysosomotropic agents,
which inhibit vesicular fusion of the slow recycling pathway and inactivate pH-sensitive
proteases, and (b) with the microtubule-disrupting agent nocodazole or incubation at 16
°C, which both inhibit the slow recycling pathway [104].

88
4.5. Summary
In this chapter, following a 24 h incubation with H4IIE hepatoma cells, a
progressive decline of detection in ProINS-Tf along with a gradual increase of irINS-Tf
has been observed in cell treatment medium. Meanwhile, an activation of ProINS-Tf has
been demonstrated after incubation with H4IIE hepatoma cells, as evidenced by the
enhanced activities observed for H4IIE-pretreated ProINS-Tf in several short-term
assays. This indicates that the converted irINS-Tf is an active form which is responsible
for the enhanced potency of ProINS-Tf in long-term activities, such as inhibition of
glucose production and the prolonged stimulation of Akt phosphorylation.
Both the loss of ProINS-Tf and generation of irINS-Tf can be blocked in the
presence of excess Tf, which indicate that the conversion is mediated through Tf-TfR
pathway in hepatoma cells. Mechanism studies have shown that the conversion is an
intracellular processing event by certain convertases in the Tf-TfR slow recycling
compartments, since this process is pH-sensitive, microtubule- and temperature-
dependent. In addition to H4IIE hepatoma cells, this TfR-mediated conversion of
ProINS-Tf has also been found in differentiated but not regular Caco-2 cells. Therefore,
through the approach of fusion protein, this unique Tf-TfR mediated endocytosis and
recycling pathway provides a novel intracellular mechanism for proprotein activation.

89
CHAPTER 5: EVALUATION OF IN VIVO PHARMACOKINETICS AND
PHARMACODYNAMICS OF PROINSULIN-TRANSFERRIN FUSION
PROTEIN

5.1. Background
Due to the short plasma half-life of regular INS in vivo, prolonging its time-action
profile has been a major goal in the development of long-acting INS as a replacement for
endogenous basal INS [13]. During the past decade, several long-acting INS analogues
with either prolonged absorption or half-life have been developed, including insulin
glargine, insulin detemir, and the recently reported insulin degludec [110-112].
In normal physiology, endogenous INS from the pancreas is first delivered to the
liver where about 50% of the INS is cleared, and the remaining INS subsequently
circulates to the peripheral tissues including muscle and fat. As a result, the liver is
physiologically exposed to higher levels of INS than the periphery, which leads to a
greater effect on HPG than PGD [80]. However, all of the current INS and INS analogues
are administered through subcutaneous injection, which results in a relatively
underinsulinized liver and hyperinsulinized periphery [113, 114]. The resulting greater
exposure in the periphery accounts for the metabolic abnormalities (e.g., hypoglycemia,
dyslipidemia, and weight gain) associated with subcutaneous injected INS analogues
[115].
An INS therapeutic with greater effect on HGP than PGD is believed to mimic the
endogenous portal INS and offer an advantage over the current treatment regiments.
Therefore, in addition to the long-acting effects, INS analogues with specificity for the
liver have begun to draw awareness in recent years. For example, a thyroxyl-INS
90
complex has shown a greater access to the liver than other tissues, by taking advantage of
the open sinusoids of the liver [116]. In addition, an INS analogue with higher affinity for
a liver-dominant IR isoform has demonstrated a greater effect in the liver [117].
However, the PK performance of these analogues is either short- or intermediate-acting.
The ideal replacement for basal INS should be a long-acting INS analogue with
greater effect in the liver. ProINS-Tf has demonstrated a strong activity in hepatoma cells
due to its conversion and activation, but low activity in adipocytes (Chapter 3 and 4).
Considering these results combined with the potential half-life extension by Tf fusion, it
is anticipated that ProINS-Tf may elicit an extended in vivo efficacy and possibly a
greater effect in the liver. Therefore, in this chapter, ProINS-Tf has been evaluated for its
hypoglycemic activity, PK properties, and its efficacy in the liver following subcutaneous
and intravenous injection in diabetic mice.

5.2. Materials and Methods
5.2.1. Animals
Male CF-1 mice (6-7 weeks old, 25 g) were purchased from Charles Rivers
(Wilmington, MA) and male C57BL/6J mice (6-7 weeks old, 20-25 g) were purchased
from Jackson Laboratory (Bar Harbor, ME). All the mice were housed in the animal
facility of University of Southern California (USC) on a 12 h day and 12 h night cycle.
The animals rooms were maintained with room temperature at 22 ± 3°C and relative
humidity at 50 ± 20%. Mice had access to water and regular rodent diet (Labdiet,
Richmond, IN) ad libitum. Under fasting condition, mice were restrained from food and
91
provided with only water. All the animal studies were conducted according to the “Guide
for the Care and Use of Laboratory Animals” (NIH Publication No. 85-23, revised 1985).
The protocol of animal experiments was approved by the Institutional Animal Care and
Use Committee at USC.

5.2.2. Streptozotocin-induced Diabetes Mice Model
After delivery to the animal facility, mice were allowed to rest for 2-3 days.
Streptozotocin (STZ, Sigma) solution was freshly prepared in 100 mM sodium citrate
buffer (pH 4.5). Mice were first fasted for 4 to 6 h, followed by a single intraperitoneal
injection of 150 mg/kg STZ to induce diabetes. Subsequently, mice were provided with
food and water for about 5 days to develop diabetes. The BG levels were monitored via
tail vein using OneTouch glucose meter (Life Scan, Milpitas, CA) with a detection range
between 20 mg/dL and 600 mg/dL. Mice with BG levels more than 400 mg/dL were used
for experiments.

5.2.3. Proteins
Recombinant human INS from E. coli was dissolved in 100 mM HCl (pH 3.0),
and further diluted in PBS to a stock of 15 μg/mL. Recombinant human ProINS was
dissolved in PBS to a stock of 100 μg/mL. ProINS-Tf and ProINS-Tf-TD (1:1.51) were
prepared as previously described in Chapter 2 and 3, respectively.

92
5.2.4. Measurement of In Vivo Pharmacokinetics
Following protein injection, 20 μL of blood was collected from the saphenous
vein at various timepoints. Blood was mixed with heparin (5U/mL blood) to prevent
coagulation, and centrifuged at 1,000 x g for 30 min to obtain clear plasma. Plasma
samples were frozen at -80°C for long-term storage.
Thawed plasma samples were diluted and subjected to human ProINS-specific
RIA to quantify the plasma protein levels. Quantification of plasma ProINS
concentrations was based on a standard curve generated by 2, 5, 10, 20, 50, 100, and 200
pM of ProINS, and quantification of ProINS-Tf concentrations was based on a standard
curve generated by 10, 30, 100, and 300 pM of ProINS-Tf. This ProINS-specific RIA has
less than 0.1% cross-reactivity with human INS. The PK parameters were calculated
according to equations from [118].

5.2.5. Measurement of In Vivo Hypoglycemic Efficacy
For experiments under fasting condition, protein injection was carried out on day
4 following STZ induction. Mice were first fasted for 2 h and given a single intravenous
or subcutaneous injection of proteins or buffer (vehicle group), and remained fasted until
12 h post-injection. On the other hand, for experiments carried out at free-feeding
condition, proteins or buffer were injected on day 3 to avoid extremely high BG levels
(and also to avoid the BG levels exceeding the detection range of the glucose meter), and
all the mice were given food freely during the recording period. BG levels were recorded
at the indicated timepoints post-injection.
93
5.2.6. Measurement of Liver-associated Enzyme Expression
Liver was extracted from the mice, perfused with PBS to completely remove the
blood, and immediately frozen in liquid nitrogen. Total RNA was extracted from frozen
liver tissues using TRIzol reagent from Invitrogen. The extracted RNA was treated with
TURBO DNase (Invitrogen) to deactivate the DNA contaminants in the RNA samples.
After cleanup, DNase-treated RNA was quantitatively and qualitatively tested by
spectrophotometer and agarose gel, respectively. Subsequently, RNA was reverse-
transcribed into cDNA using M-MLV reverse transcriptase (Promega). The total cDNA
was applied to TaqMan real-time PCR to quantify the target cDNAs using specific oligo
primer set (forward and reverse) and the fluorescent-labeled probe (Applied Biosciences).
Nuclease-free water was used during the entire experiment to avoid nucleotide
degradation. The target gene expression was normalized to the expression of
glyceraldehydate-3-phosphate dehydrogenase (GAPDH). The assay numbers for the
probes and primer sets of glucose-6-phosphatase (G6Pase), phosphoenolpyruvate
carboxykinase (PEPCK), and GAPDH were Mm00839363_m, Mm00440636_m1, and
Mm99999915_g1, respectively. Relative expression levels in treatment group were
compared to the control group and analyzed using comparative C
T
method [119].

5.2.7. Statistical Analysis
All the animal experiments were performed with n=3-5 mice for each treatment
group. Student's t-test was used for statistical analysis, where differences with p values
less than 0.05 were considered statistically significant.
94
5.3. Results
5.3.1. Pharmacokinetics
The PK profiles of ProINS-Tf fusion protein was first evaluated through
intravenous injection to normal CF-1 mice, and it was compared with ProINS that was
injected at the same molar ratio dose. The plasma concentrations of these two proteins
were measured by ProINS-specific RIA. Fig. 32 demonstrated that ProINS-Tf (0.5
mg/kg) exhibited a 6.8-fold longer elimination half-life and 50-fold longer distribution
half-life compared to ProINS (0.053 mg/kg, equimolar dose as ProINS-Tf) following
intravenous injection in CF-1 mice. Therefore, both the distribution and elimination rate
slowed down due to the increased size of protein.

Fig. 32. ProINS-Tf exhibited a longer half-life than ProINS through intravenous
injection to CF-1 mice. Either 0.053 mg/kg of ProINS or 0.5 mg/kg of ProINS-Tf fusion
protein (both equivalent to 5.62 nmole/kg of protein) was administered intravenously to
CF-1 mice, and blood was collected at the indicated timepoints. The distribution half-life
(t
1/2 α
) and elimination half-life (t
1/2 β
) of each protein were calculated and shown in the
figure. Data were expressed as average from four mice per group with error bars
indicating the standard deviation.
95

Furthermore, since the hypoglycemic activity was studied in STZ-induced
diabetic mice through subcutaneous injection, the PK profiles of ProINS-Tf and ProINS
were also compared in diabetic mice. Fig. 33A demonstrated that 0.21 mg/kg of ProINS
can be detected until 2 h post-injection, whereas 2 mg/kg of ProINS-Tf (equimolar dose
as 0.21 mg/kg of ProINS) was detectable up to 48 h post-injection. Similar to the
intravenously injected ProINS-Tf in CF-1 mice, ProINS-Tf also exhibited a longer
elimination half-life of 7.29 h associated with a quicker elimination rate of 0.095 h
-1
,
compared to a half-life of 0.5 h with an elimination rate of 1.42 h
-1
for ProINS,
respectively. In addition, ProINS-Tf (t
max
of 5.5 h) showed a longer absorption than
ProINS (t
max
of 0.3 h). The protracted absorption also resulted in a plateau period from 4
to 8 h post-injection for the PK of ProINS-Tf. Based on the curves of the two proteins,
ProINS-Tf exhibited greater exposure to the mice with a significantly higher area under
the curve (AUC) compared to ProINS (Table 4). Moreover, Fig. 33B showed that there
was no difference in the PK of ProINS-Tf between fasting and free-feeding conditions.

96

Fig. 33. ProINS-Tf exhibited prolonged PK profiles through subcutaneous
injection to STZ-induced diabetic mice. (A) Either 0.21 mg/kg of ProINS or 2 mg/kg
ProINS-Tf fusion protein (equivalent to 22.5 nmole/kg of protein) was administered
subcutaneously to STZ-induced diabetic mice, and blood was collected at the indicated
timepoints. (B) The PK profiles of 2 mg/kg subcutaneously administered ProINS-Tf in
fasted or free-fed mice. Data were expressed as average with error bars indicating the
standard deviation (n=4).

97

Table 4. PK analyses of subcutaneously administered ProINS and ProINS-Tf in STZ-
induced diabetic mice.

ProINS
(0.21 mg/kg, 22.5 nmole/kg)
ProINS-Tf
(2 mg/kg, 22.5 nmole/kg)
C
max
(ng·mL
-1
) 432.52 ± 84.19 5032.67 ± 161.91
t
max
(h) 0.31 ± 0.11 5.50 ± 1.91
t
1/2elim
(h) 0.50 ± 0.09 7.29 ± 0.15
K
el
(h
-1
) 1.42 ± 0.23 0.095 ± 0.002
AUC
0_inf
(h·ng·mL
-1
) 426.59 ± 56.51 77950.70 ± 2947.35

Equimolar dose of ProINS or ProINS-Tf fusion protein (equivalent to 22.5 nmole/kg of
protein) was administered subcutaneously to STZ-induced diabetic mice. All the data
were presented as average ± standard deviation from four mice per treatment group. C
max
,
maximal plasma concentration; t
max
, time to maximal plasma concentration; t
1/2elim
,
elimination half-life; K
el
, elimination rate constant; AUC
0-inf
, area under the plasma
concentration curve extrapolated to infinity.

5.3.2 Pharmacodynamics
5.3.2.1. Hypoglycemic Efficacy of Subcutaneously Administered Proinsulin-Transferrin
in Fasted Diabetic Mice
In order to demonstrate the in vivo efficacy of ProINS-Tf fusion protein in
glycemia control, its hypoglycemic effect was investigated in STZ-induced diabetic mice
that exhibit severe hyperglycemia (Fig. 34A). When mice were fasted for 14 h (2 h pre-
injection and 12 h post-injection), the BG of the vehicle-treated mice gradually
decreased, and the lowest levels were ~60% of the initial levels at 6 h post-injection.
However, it gradually returned to ~80% at 12 h post-injection, which was due to the
glucose produced from the liver under prolonged fasting [113]. Subcutaneous injection of
both ProINS and INS showed a rapid decrease in BG within the first two hours, but the
response diminished quickly with BG returning to the levels similar to the vehicle group.
98
The maximum response caused by ProINS and INS was 62% and 78% decrease in BG,
respectively. The longer half-life of ProINS compensated for its lower binding affinity to
the IR. Therefore, the efficacy of ProINS did not appear much lower than INS. When
mice were fasted, subcutaneous injection of 2 mg/kg ProINS-Tf elicited a delayed onset
but a sustained effect. The maximum effect started from 4 h and lasted until at least 12 h,
and there was no pronounced peak during this period. A 2 mg/kg dose of ProINS-Tf
reduced 73-81% of the BG, thus leading to a sustained FBG approximately to or even
below 100 mg/dL. Interestingly, the activity of ProINS-Tf at 12 h was comparable to that
of INS at 1 and 2 h. In addition, the hypoglycemic response of ProINS-Tf was in a dose-
dependent manner (Fig. 34B). Compared to 2 mg/kg of ProINS-Tf, a lower dose of 0.67
mg/kg showed a relatively less potent efficacy with only 60-70% decrease in BG, and the
effect significantly declined after 8 h post-injection.
99

Fig. 34. ProINS-Tf showed extended hypoglycemic efficacy following
subcutaneous injection in fasted STZ-diabetic mice. (A) Equimolar dose of INS (0.13
mg/kg), ProINS (0.21 mg/kg), or ProINS-Tf fusion protein (2 mg/kg) was administered
to diabetic mice. (B) Two doses of ProINS-Tf (0.67 mg/kg and 2 mg/kg) were injected to
diabetic mice. All the mice were fasted 2 h prior to injection and remained fasted during
the experimental period. The BG was recorded at the indicated timepoints. Results were
expressed as percentage relative to the initial BG levels. Data were shown as average
with error bars indicating the standard deviation (n=4-5).

100
5.3.2.2. Hypoglycemic Efficacy of Intravenously Administered Proinsulin-Transferrin in
Fasted Diabetic Mice
Since the immediate in vivo hypoglycemic response of ProINS-Tf was masked by
the protracted absorption phase (until 4 h post-injection) caused by subcutaneous
administration, the efficacy by intravenous injected ProINS-Tf in diabetic mice was
further evaluated. As shown in Fig. 35A, 0.026 mg/kg of INS demonstrated a very rapid
hypoglycemic effect showing 66% decrease in BG (relative to the initial levels) at both
0.5 and 1 h post-injection, and the BG started to return after 1 h. On the contrary, very
little effect was demonstrated by 0.4 mg/kg of ProINS-Tf within the first two hours
following injection (0%, 3%, and 19% decrease at 0.5, 1, and 2 h, respectively). Distinct
from ProINS-Tf, an equimolar dose of ProINS-Tf-TD fusion protein (ProINS-Tf treated
with trypsin in the w/w ratio of 1:1.51) exhibited a more rapid and potent effect than
ProINS-Tf, with 23%, 39%, and 55% decrease in BG at 0.5, 1, and 2 h post-injection,
respectively.
When the effect was continuously monitored at later timepoints (Fig. 35B),
ProINS-Tf demonstrated a significantly improved hypoglycemic effect with 61% BG
decrease at 4 h. This effect appeared to be in a sustained manner, as the BG decrease by
ProINS-Tf was 65% at 12 h. On the other hand, ProINS-Tf-TD showed a 76% decrease
at 4 h, but the effect seemed to slightly decline as demonstrated by 57% decrease in BG
at 12 h. In contrast to ProINS-Tf and ProINS-Tf-TD, the BG in INS-treated group all
recovered to the levels similar to vehicle group at 4 and 12 h post-injection. Both
Coomassie blue staining and anti-Tf Western blot (Fig. 36) confirmed that the
101
administered ProINS-Tf-TD fusion protein had a molecular weight lower than ProINS-
Tf, and so this ProINS-Tf-TD was an INS-like-Tf fusion protein.

Fig. 35. ProINS-Tf exhibited a delayed but prolonged hypoglycemic response
following intravenous injection to diabetic mice. Equimolar dose of INS (0.026 mg/kg),
ProINS-Tf (0.4 mg/kg), or ProINS-Tf-TD (0.387 mg/kg) was administered. All the mice
were fasted 2 h before injection and remained fasted during the experiment. (A) Short-
term (within 2 h) BG changes after protein injection. (B) Comparison of BG levels at 1,
4, and 12 h following protein injection. Data were shown as average with error bars
indicating the standard deviation (n=3-4). Statistical significance determined by Student's
t-test was shown by the asterisk with * indicating p<0.05 and ** indicating p<0.01
compared to ProINS-Tf group (shown in A).

102

Fig. 36. ProINS-Tf-TD showed a decreased molecular weight compared to
ProINS-Tf. (A) Coomassie blue staining of ProINS-Tf (lane 1) and ProINS-Tf-TD
(1:1.51) (lane 2). (B) Anti-Tf Western blot of Tf (lane 1), ProINS-Tf (land 2), and
ProINS-Tf-TD (1:1.51) (lane 3 and 4).

5.3.2.3. Hypoglycemic Efficacy of Subcutaneously Administered Proinsulin-Transferrin
in Free-Fed Diabetic Mice
When diabetic mice were fed freely, vehicle-treated mice showed stable BG
levels during the entire detection period, and it was higher than the BG of fasted mice due
to food intake. Subcutaneous injection of ProINS-Tf showed only 20% decrease in the
BG of free-fed mice from 2 h to at least 12 h post-injection (Fig. 37), which was
dramatically different from results in the fasted mice demonstrated earlier.

103

Fig. 37. ProINS-Tf demonstrated a weak hypoglycemic response following
subcutaneous injection to free-fed diabetic mice. ProINS-Tf (2 mg/kg) was administered,
and the BG levels were monitored. All the mice were freely fed during the experiment.
Data were shown as average with error bars indicating the standard deviation (n=4).

5.3.3 Inhibition of Hepatic Glucose Production by Proinsulin-Transferrin
The inhibition of HGP was assessed by the suppression of gene expression related
to glycogenolysis and gluconeogenesis in the liver [120]. G6Pase catalyzes the terminal
step in both gluconeogenic and glycogenolytic pathways [121], and PEPCK synthesizes
p-enolpyruvate from oxalacetate, which is the rate-controlling step of gluconeogenesis
[122]. In diabetic conditions, both mRNA levels of these two enzymes are elevated, and
they can be reduced by administration of INS [120].Therefore, the mRNA levels of these
two genes in the liver were measured after administration of ProINS-Tf. When mice were
fasted for prolonged hours, 2 mg/kg of ProINS-Tf significantly inhibited 89% of G6Pase
104
expression and 22% of PEPCK expression at 12 h post-injection. However, the inhibitory
effects on G6Pase by equimolar ProINS and INS were 59% and 16%, respectively, and
minimal effects on PEPCK was shown by ProINS and INS (Fig. 38). Therefore, the HGP
in the liver was effectively suppressed by ProINS-Tf when mice were under prolonged
fasting.

Fig. 38. Liver G6Pase and PEPCK mRNA expression was inhibited by ProINS-Tf
at 12 h post-injection to fasted diabetic mice. Results were expressed as the mRNA levels
of the treatment groups relative to the vehicle. Data were shown as average with error
bars indicating the standard deviation (n=4). Statistical significance was determined by
Student's t-test and marked with an asterisk (* p<0.05 and ** p<0.01) compared to the
vehicle.

105
5.4. Discussion
One of the most straightforward strategies to develop long-acting INS is to
prolong its plasma half-life. The best examples include INS detemir which is designed to
bind to albumin in the circulation and thus achieve a prolonged half-life [110], and the
newly developed ultra-long-acting INS degludec where modified INS molecule is
conjugated to the palmitic acid [111]. Human Tf has been used to prolong protein half-
life, owing to its large size, good stability, and most importantly, its receptor-mediated
recycling mechanism. The reported GLP-1 and hGH, when fused with Tf as fusion
proteins, have both exhibited increased half-life [69, 123]. In our case, ProINS-Tf had a
6.8- and 15-fold higher elimination half-life than ProINS alone in normal CF-1 mice and
STZ-induced diabetic mice, respectively (Fig. 32 and 33). The distinct differences in PK
parameters are probably attributed to either different mice model (normal mice vs.
diabetic mice) or the nonlinear PK of proteins (the doses for CF-1 mice were 4 times
lower than for diabetic mice) [124, 125]. In addition to the increased half-life, ProINS-Tf
also showed a greater AUC value and thus offered a greater exposure to the mice than
ProINS (Table 4). Corresponding to the improved PK profiles, the reduction in
hyperglycemia by ProINS-Tf was also extended to at least 12 h post-injection (only
recorded until 12 h to avoid long period of fasting). The FBG after ProINS-Tf injection
were maintained around the normoglycemia range without pronounced peak from 4 to 12
h (Fig. 34). Therefore, the extended and particularly peakless hypoglycemic efficacy of
ProINS-Tf implied its potential application as a long-acting INS analogue.
106
Devoid of the interference from subcutaneous absorption, intravenous
administration of ProINS-Tf showed very little reduction of BG within the first two hours
post-injection, whereas a remarkable reduction of BG was observed in INS-treated mice
during the initial hours (Fig. 35A). However, a considerable decline of BG was observed
at 4 h and it appeared to sustain to 12 h, whereas the BG of INS group returned to control
levels (Fig. 35B). Therefore, the delayed onset of an enhanced efficacy in the mice may
also result from an in vivo activation, which was similar to what was observed in vitro
where an activation of ProINS-Tf to irINS-Tf led to a delayed but increased potency
(Chapter 4). An activated form ProINS-Tf-TD has been demonstrated to behave like an
INS-like-Tf in vitro (Fig. 17 and 18), and its in vivo hypoglycemic response within two
hours post-injection was significantly stronger, albeit not as potent as INS, than ProINS-
Tf. Additionally, ProINS-Tf showed similar potency as ProINS-Tf-TD at later timepoints
(4 and 12 h, Fig. 35B), which further indicated that ProINS-Tf might be gradually
activated in vivo. Furthermore, several findings have implied that this in vivo activation
may occur in the liver. First, ProINS conversion and activation to irINS-Tf has been
demonstrated in hepatoma cells (Chapter 4). Second, HGP-related gene expression was
significantly suppressed by ProINS-Tf in vivo (Fig. 38). Lastly, the liver may be
beneficial for the TfR-mediated activation, since hepatocytes express high levels of TfR
and they are the most efficient non-endocrine cells with abundant ER and Golgi.
In addition to the prolonged half-life and extended hypoglycemic efficacy of
ProINS-Tf, there appears to be a correlation between the end-point effect (hypoglycemic
effect) and the inhibition of HGP in the liver. This was evidenced by the correspondence
107
of hypoglycemic efficacy with the suppression of liver G6Pase and PEPCK mRNA
expression in mice under prolonged fasting (Fig. 38), and the dramatic distinction
between fasted and free-fed mice (80% vs. 20% decrease in BG, Fig. 34 and 37). The
hypoglycemic effect by INS results from a combination of promoting PGD and inhibiting
HGP [84]. Promoting PGD in peripheral tissues relies on INS-stimulated GLUT4
transportation that is independent of BG levels [126], whereas the inhibiting HGP is
responsive to the BG due to the glucose-sensitive property of the liver [113]. The HPG
from the liver is the exclusive source for the BG during prolonged fasting (longer than 6
h). However, the HGP only constitutes part of the total BG at free-feeding conditions,
and it reduces because of the feedback from the dietary glucose [113]. Therefore, the
higher BG reduction by ProINS-Tf observed in prolonged fasting than free-feeding
condition can be attributed to its preferential effect in inhibiting HGP in the liver than
promoting PGD in the periphery. Furthermore, a dramatic difference of efficacy between
the two conditions would not be expected, if ProINS-Tf had a favored effect in promoting
PGD. Additionally, its relatively weak potency in the peripheral tissues was also
supported by its in vitro low activity measured in adipocytes (Fig. 16). This in vivo liver-
preferential effect of ProINS-Tf may possibly be relevant to the reported marginally
hepatospecific effect of ProINS [36, 127]. Another interpretation could be the in vivo
activation process discussed earlier. If an activation happened selectively in the liver, it
would lead to an enhanced activity locally in the liver.
Numerous studies have agreed that the liver plays an essential role in controlling
the FBG by regulating HGP [113, 120]. Diabetes due to either INS deficiency or INS
108
resistance shows excessive levels of HGP that correlate with a significantly elevated
FBG. Therefore, having a preferential or ideally specific effect in the liver will provide a
desirable control of hyperglycemia.

5.5. Summary
In this chapter, the PK/PD of ProINS-Tf fusion protein have been investigated in
a STZ-induced diabetic mice model. Compared with ProINS alone, ProINS-Tf has a
significantly prolonged elimination half-life and enhanced in vivo exposure.
Corresponding to these prolonged PK profiles, ProINS-Tf elicits extended PD effects
with the FBG maintaining within normoglycemia ranges in a peakless pattern between 4
and at least 12 h post-injection. The delayed occurrence of hypoglycemic efficacy
following intravenous injection indicates the possibility of an in vivo activation of
ProINS-Tf. Furthermore, the correlation of hypoglycemic efficacy with the inhibition of
HGP-related gene expression suggests that ProINS-Tf has a preferential effect in the
liver. Additionally, the drastically distinct efficacies between fasting and free-feeding
conditions also support this liver-preferential effect of ProINS-Tf. Taken together, the
extended and liver-preferential action in reducing hyperglycemia have suggested the
potential of ProINS-Tf as a promising candidate for a long-acting INS analogue.

109
CHAPTER 6: SUMMARY AND FUTURE PERSPECTIVES

6.1. Summary
6.1.1. Overall Goal of this Dissertation
The main goal of this dissertation is to develop a novel long-acting INS analogue
that can potentially be applied as a basal INS supplement for diabetes treatment. Both
type 1 and type 2 diabetes feature elevated FBG levels due to excessive glucose
production from the liver. The functionality of long-acting INS analogues is not only to
reduce the administration frequency of INS, but more importantly, to act on the liver to
control the FBG levels.
The native human INS is a potent hormone that can induce an acute glucose-
lowering effect. It is a small peptide, and its rapid clearance results in a short plasma half-
life [26]. The precursor peptide ProINS, albeit having a longer half-life than INS, exerts
low IR binding affinity and low in vitro biological potency [36]. In addition, INS and
ProINS regulate glucose homoeostasis by promoting PGD in the periphery and inhibiting
HGP in the liver. INS has no preference between these two actions, and ProINS has a
marginal in vivo hepatospecificity [36]. Therefore, INS and ProINS have undesirable
PK/PD properties when they were exogenously administered as therapeutics, which leads
to a great demand of novel long-acting INS analogues with improved long-lasting
functions.
Human Tf-based fusion protein strategy has shown successful improvement in the
PK of small peptides or proteins, such as G-CSF (2005), hGH (2010), GLP-1 (2010), and
110
exendin-4 (2010) [67-69]. In this dissertation, we have designed and produced a fifth Tf
recombinant fusion protein, ProINS-Tf. ProINS was selected because INS is a two-chain
peptide that is not a suitable candidate for fusion. ProINS-Tf is expected to exert
improved performance compared with INS or ProINS, by taking advantage of Tf
properties such as large size, high stability, and the unique receptor-mediated recycling
pathway.

6.1.2. Proinsulin-Transferrin is Produced using Recombinant DNA Technology
First of all, we employed a mammalian expression system combined with the his-
tag method to produce and purify ProINS-Tf fusion protein. ProINS-Tf was secreted by
HEK 293 cells to the culture media. The TFF-based concentration coupled with his-tag
purification enabled a thousand fold concentration and as high as 80% in purity (Fig. 8).
ProINS-Tf was shown as a major band of 89 kDa on both Coomassie blue staining and
Western blot against ProINS and Tf (Fig. 9).
Two in vitro bioassays corresponding to the effects of INS on PGD and HPG
were established to examine the biological potency of ProINS-Tf. ProINS-Tf
demonstrated a very low activity in the short-term (30 min incubation) promotion of
glucose uptake in adipocytes, which was similar to ProINS but about two orders of
magnitude lower than INS (Fig. 16). The low activity of ProINS-Tf in adipocytes
correlated with its low IR binding affinity (Fig. 12). In addition, ex vivo trypsin-digestion
was able to convert ProINS-Tf to an INS-like-Tf, and the latter exhibited increased
activity in promoting glucose uptake compared to non-digested ProINS-Tf (Fig. 17 and
111
18). These results indicated that ProINS-Tf intrinsically acted in resemblance of ProINS
and the activation to an INS-like molecule was essential for promoting glucose uptake.
On the other hand, ProINS-Tf elicited a remarkably strong activity in the long-
term (24 h incubation) inhibition of glucose production in hepatoma cells, which was 24-
and 6-fold stronger than ProINS and INS, respectively. Interestingly, this strong activity
was Tf-TfR mediated, since it was significantly blocked in the presence of excess Tf (Fig.
19). Distinct from both INS and ProINS, ProINS-Tf induced little stimulation of Akt
phosphorylation at 5 min incubation, however, phospho-Akt levels were significantly
enhanced at both 4 and 12 h of incubation (Fig. 20). This delayed but enhanced response
suggested an activation of ProINS-Tf by hepatoma cells.

6.1.3. Proinsulin-Transferrin is Converted to an Active Form of Insulin through
Transferrin Receptor Mediated Endocytosis and Recycling Pathway
In order to verify the activation hypothesis, the kinetic changes of ProINS-Tf in
the cell treatment medium were examined following 24 h incubation with the hepatoma
cells. A progressive decline in ProINS-Tf corresponded with a gradual generation of an
irINS-Tf in the medium, thereby confirming that ProINS-Tf was converted by the
hepatoma cells. It was observed that both responses could be blocked by excess of Tf but
not BSA, indicating that the conversion was attributed to Tf-TfR pathway (Fig. 21 and
24). Further intracellular trafficking studies indicated that the conversion occurred in the
slow recycling route of the Tf pathway during which the endocytosed ProINS-Tf might
gain access to the intracellular proteases (Fig. 25). The TfR-mediated recycling and the
112
final release of the endocytosed fusion protein outside the cells enabled the intracellularly
converted irINS-Tf to be detected in the cell treatment medium. A detailed mechanism
has been demonstrated in Scheme 2 of Chapter 4. Furthermore, the hepatoma cell
pretreated ProINS-Tf also demonstrated enhanced short-term activities, whereas non-
treated ProINS-Tf did not (Fig. 26-28). This strongly verified that the irINS-Tf was an
active form and the conversion is responsible for the strong activity observed in glucose
production assay. Based on these findings, the ex vivo trypsin digestion is no longer
necessary for ProINS-Tf activation.
Among many prodrug-activating methods, the intracellular activation strategy
employing the proteases in the protein secretory pathway has shown promise in the
conversion of ProINS and other proproteins [108, 128-130]. However, due to the limited
endocytosis and the inaccessibility to certain compartments inside the cells, most reported
approaches primarily rely on either gene delivery or cellular expression methods. Results
from our study suggest that, under the control of the Tf-TfR endocytosis and recycling
pathway, the Tf fusion protein approach can employ the intracellular machinery to
facilitate proprotein endocytosis, conversion and release, while circumventing the
complex problems with other methods.

6.1.4. Proinsulin-Transferrin Elicits Prolonged Pharmacokinetics and Extended
Pharmacodynamics with Liver-preferential Effects In Vivo
The in vivo PK/PD evaluation in STZ-induced diabetic mice provided very
promising findings for ProINS-Tf fusion protein. First, compared with INS and ProINS,
113
ProINS-Tf fusion protein elicited a 15-fold longer elimination half-life and a 180-fold
greater exposure (Fig. 33). The prolonged PK correlated with the extended PD that
demonstrated a peakless hypoglycemic efficacy from 4 to 12 h post-injection in fasted
mice (Fig. 34). Second, intravenously injected ProINS-Tf demonstrated little
hypoglycemic efficacy in fasted mice during the first two hours, whereas ProINS-Tf-TD
showed an immediate glucose-lowing response following intravenous injection. Results
of these two proteins coincided with the in vitro promotion of glucose uptake data. On the
other hand, a progressively increased efficacy of ProINS-Tf was observed at 4 and 12 h
post-injection, and this agreed with the kinetic pattern of Akt phosphorylation (Fig. 35).
These results suggested an activation of ProINS-Tf may also occur in vivo. In addition,
the hypoglycemic efficacy of ProINS-Tf correlated with the inhibition of HGP in the
liver, based on (1) the correspondence of glucose-lowing effect with the suppression in
HGP-related gene expression, and (2) the dramatically distinct response between fasted
and free-fed mice (Fig. 37 and 38). Therefore, we believe that ProINS-Tf had a
preferential effect in inhibiting HGP to promoting PGD.

6.1.5. Conclusion
In summary, this dissertation has presented the following three contributions to
the area of protein drug development. First, ProINS-Tf is a new member to Tf-based
fusion protein family, and it is the first proprotein-Tf fusion protein. Second, we have
proposed a novel intracellular proprotein activation mechanism utilizing the Tf-TfR
mediated endocytosis and recycling pathway. Thus, in addition to prolonging half-life
114
and facilitating oral delivery of protein drugs, results from this work open new doors for
application of Tf-based bifunctional proteins in prodrug activation [104]. Lastly, the
long-lasting PK/PD and the proposed liver-preferential effect suggest great promise of
ProINS-Tf fusion protein as a potential long-acting INS analogue for diabetes treatment.

115
6.2. Future Perspectives
The first direction of future work regards to the TfR-mediated activation of
ProINS-Tf fusion protein presented in this dissertation. This is the first proprotein
activation mechanism utilizing receptor-mediated endocytosis. To further understand this
activation and thus improve the conversion efficiency, more comprehensive studies are
necessary.
First of all, it will be interesting to examine whether the conversion also occurs in
cells other than hepatoma cells, especially cells derived from non-liver tissues involved in
glucose homeostasis (e.g., adipocytes, L6 rat skeletal myoblast cells, and C2C12 mouse
myoblast cells) [131]. Results from these cells will reveal the hepatospecificity for
ProINS-Tf conversion in vitro. Conversion in primary hepatocytes can also be tested to
compare with that in currently-used hepatoma cells.
Hepatocytes (hepatoma cells) are well-known for the endocrine secretion of
proteins, and additionally, ProINS-Tf conversion has been preliminarily demonstrated in
enterocyte-like Caco-2 cells. These suggest that secretory-like cells may be beneficial for
ProINS-Tf conversion, as they may either express specific proteases or offer a more
favorable environment for the access of endocytosed ProINS-Tf to certain proteases.
Additional work to study this correlation may help identify the protease(s) responsible for
the conversion.
Another issue that deserves detailed investigation is the characterization of irINS-
Tf detected by INS-specific RIA. The antibody used in this RIA method recognizes INS,
as well as the Des 64,65 intermediate (Fig. 2) with 76% cross-reactivity. Thus, the RIA
116
results do not ensure whether it is intact insulin-transferrin (INS-Tf) or one of the
immunoreactive intermediates. Identification of the conversion products can be achieved
using liquid chromatography combined with mass spectroscopy, which is a common
analytical method for large-molecule characterization [132].
The second direction of future work regards to the in vivo studies. The in vivo
results presented by this dissertation work clearly demonstrate prolonged PK and
extended PD properties of ProINS-Tf fusion protein in comparison to that of INS or
ProINS. The liver-preferential effect of ProINS-Tf is a very exciting finding, which
shows great promise in application as a basal INS replacement. Therefore, future work
should focus on elucidating the preference of ProINS-Tf in inhibiting HGP to promoting
PGD. One method is to employ the well-characterized hyperinsulinemic-euglycemic
clamp studies, which can directly compare the amount of HGP to PGD following protein
injection in vivo [133]. Another approach is to explore whether there is a stronger
ProINS-Tf-mediated IR signaling activation in the liver than periphery. In addition,
ProINS-Tf can also be assessed for the efficacy toward an excessive FBG in type 2
diabetic animal models, which may potentially expand the application of this fusion
protein. Furthermore, this liver-preferential effect is currently hypothesized to result from
an in vivo activation in the liver. Therefore, measuring the in vivo ProINS-Tf conversion
and verifying the activation will be two next subjects of interests after confirming the
liver-preferential effect.

117
Taken together, this dissertation work has characterized various aspects of
ProINS-Tf fusion protein, and illustrated the benefits of Tf-based fusion strategy. In the
future, with more understanding of the conversion process, this TfR-mediated activation
mechanism will be able to offer an important opportunity for proprotein activation. With
additional future studies regarding the in vivo liver-preference, ProINS-Tf will show
more promising prospects as a basal INS replacement.

118
REFERENCES

(1) Rader, R., BIOPHARMA: Biopharmaceutical products in the U.S. and European
markets. 2012, Rockville, MD: Biotechnology Information Institute.

(2) Leader, B., Baca, Q. J. and Golan, D. E. Protein therapeutics: a summary and
pharmacological classification. Nature Rev Drug Discovery. 2008, 7, 21-39.

(3) Tomlinson, I. M. Next-generation protein drugs. Nat Biotechnol. 2004, 22, 521-2.

(4) Frokjaer, S. and Otzen, D. E. Protein drug stability: a formulation challenge. Nat
Rev Drug Discov. 2005, 4, 298-306.

(5) Mahmood, I. and Green, M. D. Pharmacokinetic and pharmacodynamic
considerations in the development of therapeutic proteins. Clin Pharmacokinet.
2005, 44, 331-47.

(6) Brown, L. R. Commercial challenges of protein drug delivery. Expert Opin Drug
Deliv. 2005, 2, 29-42.

(7) Walsh, C. T., Posttranslational modification of proteins: expanding Nature's
inventory. 2005, Colorado: Roberts & Company.

(8) Graslund, S., Nordlund, P., Weigelt, J., Hallberg, B. M., Bray, J., et al. Protein
production and purification. Nat Methods. 2008, 5, 135-46.

(9) Schellekens, H. Bioequivalence and the immunogenicity of biopharmaceuticals.
Nat Rev Drug Discov. 2002, 1, 457-62.

(10) Barbosa, M. D. Immunogenicity of biotherapeutics in the context of developing
biosimilars and biobetters. Drug Discov Today. 2012, 16, 345-53.

(11) Marshall, S. A., Lazar, G. A., Chirino, A. J. and Desjarlais, J. R. Rational design
and engineering of therapeutic proteins. Drug Discov Today. 2003, 8, 212-21.

(12) Roberts, M. J., Bentley, M. D. and Harris, J. M. Chemistry for peptide and protein
PEGylation. Adv Drug Deliv Rev. 2002, 54, 459-76.

(13) Hirsch, I. B. Insulin analogues. N Engl J Med. 2005, 352, 174-83.

(14) Arakawa, T., Prestrelski, S. J., Narhi, L. O., Boone, T. C. and Kenney, W. C.
Cysteine 17 of recombinant human granulocyte-colony stimulating factor is
partially solvent-exposed. J Protein Chem. 1993, 12, 525-31.

119
(15) Culajay, J. F., Blaber, S. I., Khurana, A. and Blaber, M. Thermodynamic
characterization of mutants of human fibroblast growth factor 1 with an increased
physiological half-life. Biochemistry. 2000, 39, 7153-8.

(16) Sarkar, C. A., Lowenhaupt, K., Horan, T., Boone, T. C., Tidor, B., et al. Rational
cytokine design for increased lifetime and enhanced potency using pH-activated
"histidine switching". Nat Biotechnol. 2002, 20, 908-13.

(17) Overbay, D. K. and Manley, H. J. Darbepoetin-alpha: a review of the literature.
Pharmacotherapy. 2002, 22, 889-97.

(18) Espinal, J. Understanding insulin action: principles and molecular mechanisms.
Chichester: Ellis Horwood. 1989.

(19) Chien, Y. W. Human insulin: basic sciences to therapeutic uses. Drug Dev. Ind.
Pharm. 1996, 22, 752-789.

(20) De Meyts, P. and Whittaker, J. Structural biology of insulin and IGF1 receptors:
implications for drug design. Nat Rev Drug Discov. 2002, 1, 769-83.

(21) De Meyts, P., Van Obberghen, E. and Roth, J. Mapping of the residues
responsible for the negative cooperativity of the receptor-binding region of
insulin. Nature. 1978, 273, 504-9.

(22) De Meyts, P., Roth, J., Neville, D. M., Jr., Gavin, J. R. and Lesniak, M. A. Insulin
interactions with its receptors: experimental evidence for negative cooperativity.
Biochem Biophys Res Commun. 1973, 55, 154-161.

(23) White, M. F. The insulin signaling system and the IRS proteins. Diabetologia.
1997, 40 Suppl 2, S2-17.

(24) Clore, J. N. and Thurby-Hay, L. Which basal insulin should be used in patients
with type 2 diabetes mellitus? Curr Diab Rep. 2007, 7, 348-52.

(25) Galloway, J. A. New directions in drug development: mixtures, analogues, and
modeling. Diabetes Care. 1993, 16 Suppl 3, 16-23.

(26) Leahy, J. L. and Cefalu, W. T., eds. Intensive insulin therapy in type 1 diabetes
mellitus. Insulin Therapy. 2002, Marcel Dekker: NY. pp 87-112.

(27) Leahy, J. L. Basal-prandial insulin therapy: Scientific concept review and
application. Am J Med Sci. 2006, 332, 24-31.
120
(28) Owens, D. R. and Bolli, G. B. Beyond the era of NPH insulin--long-acting insulin
analogs: chemistry, comparative pharmacology, and clinical application. Diabetes
Technol Ther. 2008, 10, 333-49.

(29) McAulay, V. and Frier, B. M. Insulin analogues and other developments in
insulin therapy for diabetes. Expert Opin Pharmacother. 2003, 4, 1141-56.

(30) Day, C., Archer, H. and Bailey, C. J. Recent advances in insulin therapy. Br J
Cardio. 2003, 10.

(31) Chance, R. E. and Frank, B. H. Research, development, production, and safety of
biosynthelic human insulin. Diabetes Care. 1993, 16, 133-142.

(32) Given, B. D., Cohen, R. M., Shoelson, S. E., Frank, B. H., Rubenstein, A. H., et
al. Biochemical and clinical implications of proinsulin conversion intermediates. J
Clin Invest. 1985, 76, 1398-405.

(33) Schatz, H., Ammermann, S., Laube, H. and Federlin, K. Bioactivity and
pharmacokinetics of human proinsulin in comparison to human insulin after
intravenous and subcutaneous injection. Horm Metab Res. 1988, 20, 445-9.

(34) Peavy, D. E., Abram, J. D., Frank, B. H. and Duckworth, W. C. In vitro activity of
biosynthetic human proinsulin: receptor binding and biological potency of
proinsulin and insulin in isolated rat adipocytes. Diabetes. 1984, 33, 1062-67.

(35) Howey, D. C., Frank, B. H., Galloway, J. A., Root, M. A. and Spradlin, C. T.
Human proinsulin (rDNA) as an intermediate-acting insulin agonist: a comparison
with NPH (N) and regular (R) insulins (Abstract). Diabetes. 1984, 33 Suppl 1
23A.

(36) Galloway, J. A., Hooper, S. A., Spradlin, C. T., Howey, D. C., Frank, B. H., et al.
Biosynthetic human proinsulin. Review of chemistry, in vitro and in vivo receptor
binding, animal and human pharmacology studies, and clinical trial experience.
Diabetes Care. 1992, 15, 666-92.

(37) Glauber, H. S., Henry, R. R., Wallace, P., Frank, B. H., Galloway, J. A., et al. The
effects of biosynthetic human proinsulin on carbohydrate metabolism in non-
insulin-dependent diabetes mellitus. N Engl J Med. 1987, 316, 443-9.

(38) Huebers, H., Huebers, E., Rummel, W. and Crichton, R. R. Isolation and
characterization of iron-binding proteins from rat intestinal mucosa. Eur J
Biochem. 1976, 66, 447-55.

121
(39) Parkkinen, J., von Bonsdorff, L., Ebeling, F. and Sahlstedt, L. Function and
therapeutic development of apotransferrin. Vox Sang. 2002, 83 Suppl 1, 321-6.

(40) Lok, C. N. and Loh, T. T. Regulation of transferrin function and expression:
review and update. Biol Signals Recept. 1998, 7, 157-78.

(41) Morgan, E. H., Smith, G. D. and Peters, T. J. Uptake and subcellular processing
of 59Fe-125I-labelled transferrin by rat liver. Biochem J. 1986, 237, 163-73.

(42) Gomme, P. T., McCann, K. B. and Bertolini, J. Transferrin: structure, function
and potential therapeutic actions. Drug Discov Today. 2005, 10, 267-73.

(43) Morgan, E. H. and Baker, E. Iron uptake and metabolism by hepatocytes. Fed
Proc. 1986, 45, 2810-6.

(44) Mukherjee, S., Ghosh, R. N. and Maxfield, F. R. Endocytosis. Physiol Rev. 1997,
77, 759-803.

(45) Sonnichsen, B., De Renzis, S., Nielsen, E., Rietdorf, J. and Zerial, M. Distinct
membrane domains on endosomes in the recycling pathway visualized by
multicolor imaging of Rab4, Rab5, and Rab11. J Cell Biol. 2000, 149, 901-14.

(46) Widera, A., Norouziyan, F. and Shen, W. C. Mechanisms of TfR-mediated
transcytosis and sorting in epithelial cells and applications toward drug delivery.
Adv Drug Deliv Rev. 2003, 55, 1439-66.

(47) Mayor, S., Presley, J. F. and Maxfield, F. R. Sorting of membrane components
from endosomes and subsequent recycling to the cell surface occurs by a bulk
flow process. J Cell Biol. 1993, 121, 1257-69.

(48) Daro, E., van der Sluijs, P., Galli, T. and Mellman, I. Rab4 and cellubrevin define
different early endosome populations on the pathway of transferrin receptor
recycling. Proc Natl Acad Sci U S A. 1996, 93, 9559-64.

(49) Stoorvogel, W., Geuze, H. J., Griffith, J. M. and Strous, G. J. The pathways of
endocytosed transferrin and secretory protein are connected in the trans-Golgi
reticulum. J Cell Biol. 1988, 106, 1821-9.

(50) Fishman, J. B. and Fine, R. E. A trans Golgi-derived exocytic coated vesicle can
contain both newly synthesized cholinesterase and internalized transferrin. Cell.
1987, 48, 157-64.

(51) Brandsma, M. E., Jevnikar, A. M. and Ma, S. Recombinant human transferrin:
beyond iron binding and transport. Biotech Adv. 2011, 29, 230-8.
122
(52) Lim, C. J. and Shen, W. C. Transferrin-oligomers as potential carriers in
anticancer drug delivery. Pharm Res. 2004, 21, 1985-92.

(53) Koppu, S., Oh, Y. J., Edrada-Ebel, R., Blatchford, D. R., Tetley, L., et al. Tumor
regression after systemic administration of a novel tumor-targeted gene delivery
system carrying a therapeutic plasmid DNA. J Control Release. 2010, 143, 215-
21.

(54) Woodnutt, G., Violand, B. and North, M. Advances in protein therapeutics. Curr
Opin Drug Discov Devel. 2008, 11, 754-61.

(55) Schmidt, S. R. Fusion-proteins as biopharmaceuticals--applications and
challenges. Curr Opin Drug Discov Devel. 2009, 12, 284-95.

(56) Roopenian, D. C. and Akilesh, S. FcRn: the neonatal Fc receptor comes of age.
Nat Rev Immunol. 2007, 7, 715-25.

(57) Lencer, W. I. and Blumberg, R. S. A passionate kiss, then run: exocytosis and
recycling of IgG by FcRn. Trends Cell Biol. 2005, 15, 5-9.

(58) Schroeder, H. W. and Cavacini, L. Structure and function of immunoglobulins. J
Allergy Clin Immunol. 2010, 125, S41-52.

(59) Gershoni, J. M. Molecular decoys: antidotes, therapeutics and
immunomodulators. Curr Opin Biotechnol. 2008, 19, 644-51.

(60) Huang, C. Receptor-Fc fusion therapeutics, traps, and MIMETIBODY
technology. Curr Opin Biotechnol. 2009, 20, 692-9.

(61) Goldenberg, M. M. Etanercept, a novel drug for the treatment of patients with
severe, active rheumatoid arthritis. Clin Ther. 1999, 21, 75-87.

(62) Elsadek, B. and Kratz, F. Impact of albumin on drug delivery - new applications
on the horizon. J Control Release. 2012, 157, 4-28.

(63) Subramanian, G. M., Fiscella, M., Lamous-Smith, A., Zeuzem, S. and
McHutchison, J. G. Albinterferon alpha-2b: a genetic fusion protein for the
treatment of chronic hepatitis C. Nat Biotechnol. 2007, 25, 1411-9.

(64) Tomkin, G. H. Albiglutide, an albumin-based fusion of glucagon-like peptide 1
for the potential treatment of type 2 diabetes. Curr Opin Mol Ther. 2009, 11, 579-
88.

123
(65) Halpern, W., Riccobene, T. A., Agostini, H., Baker, K., Stolow, D., et al.
Albugranin, a recombinant human granulocyte colony stimulating factor (G-CSF)
genetically fused to recombinant human albumin induces prolonged myelopoietic
effects in mice and monkeys. Pharm Res. 2002, 19, 1720-9.

(66) Duttaroy, A., Kanakaraj, P., Osborn, B. L., Schneider, H., Pickeral, O. K., et al.
Development of a long-acting insulin analog using albumin fusion technology.
Diabetes. 2005, 54, 251-8.

(67) Amet, N., Wang, W. and Shen, W. C. Human growth hormone-transferrin fusion
protein for oral delivery in hypophysectomized rats. J Control Release. 2010, 141,
177-82.

(68) Bai, Y., Ann, D. K. and Shen, W. C. Recombinant granulocyte colony-stimulating
factor-transferrin fusion protein as an oral myelopoietic agent. Proc Natl Acad Sci
U S A. 2005, 102, 7292-6.

(69) Kim, B. J., Zhou, J., Martin, B., Carlson, O. D., Maudsley, S., et al. Transferrin
fusion technology: a novel approach to prolong biological half-life of
insulinotropic peptides. J Pharmacol Exp Ther. 2010, 334, 682-92.

(70) Wurm, F. M. Production of recombinant protein therapeutics in cultivated
mammalian cells. Nat Biotechnol. 2004, 22, 1393-8.

(71) Sethuraman, N. and Stadheim, T. A. Challenges in therapeutic glycoprotein
production. Curr Opin Biotechnol. 2006, 17, 341-6.

(72) Amet, N., Lee, H. F. and Shen, W. C. Insertion of the designed helical linker led
to increased expression of tf-based fusion proteins. Pharm Res. 2009, 26, 523-8.

(73) Winter, J., Lilie, H. and Rudolph, R. Renaturation of human proinsulin--a study
on refolding and conversion to insulin. Anal Biochem. 2002, 310, 148-55.

(74) Scheich, C. and Sievert, V. An automated method for high-throughput protein
purification applied to a comparison of His-tag and GST-tag affinity
chromatography. BMC Biotech. 2003, 3, 12-9.

(75) Mason, A. B., He, Q. Y., Halbrooks, P. J., Everse, S. J., Gumerov, D. R., et al.
Differential effect of a his tag at the N- and C-termini: functional studies with
recombinant human serum transferrin. Biochemistry. 2002, 41, 9448-54.

124
(76) MacGillivray, R. T., Moore, S. A., Chen, J., Anderson, B. F., Baker, H., et al.
Two high-resolution crystal structures of the recombinant N-lobe of human
transferrin reveal a structural change implicated in iron release. Biochemistry.
1998, 37, 7919-28.

(77) Baldi, L., Hacker, D. L., Adam, M. and Wurm, F. M. Recombinant protein
production by large-scale transient gene expression in mammalian cells: state of
the art and future perspectives. Biotechnol Lett. 2007, 29, 677-84.

(78) Huang, B. C., Davern, S. and Kennel, S. J. Mono and bivalent binding of a scFv
and covalent diabody to murine laminin-1 using radioiodinated proteins and SPR
measurements: Effects on tissue retention in vivo. J Immunol Methods. 2006,
313, 149-60.

(79) Du, T., Hu, K., Yang, J., Jin, J., Li, C., et al. Bifunctional CD4-DC-SIGN Fusion
Proteins Demonstrate Enhanced Avidity to gp120 and Inhibit HIV-1 Infection and
Dissemination. Antimicrob Agents Chemother. 2012, Jun 11.

(80) Duckworth, W. C., Bennett, R. G. and Hamel, F. G. Insulin degradation: progress
and potential. Endocr Rev. 1998, 19, 608-24.

(81) Probst, I., Hartmann, H., Jungermann, K. and Creutzfeldt, W. Insulin-like action
of proinsulin on rat liver carbohydrate metabolism in vitro. Diabetes. 1985, 34,
415-9.

(82) Freychet, P. The interactions of proinsulin with insulin receptors on the plasma
membrane of the liver. J Clin Invest. 1974, 54, 1020-31.

(83) Saltiel, A. R. and Pessin, J. E., eds. Hepatic regulation of fuel metabolism.
Mechanisms of Insulin Action, ed. C. Clark and C.B. Newgard. 2007, Spinger:
New York. pp 90-97.

(84) Sonksen, P. and Sonksen, J. Insulin: understanding its action in health and
disease. Br J Anaesth. 2000, 85, 69-79.

(85) Kahn, C. R. The molecular mechanism of insulin action. Annu Rev Med. 1985,
36, 429-51.

(86) White, M. F. The insulin signalling system and the IRS proteins. Diabetologia.
1997, 40, S2-17.

(87) Harmon, A. W., Paul, D. S. and Patel, Y. M. MEK inhibitors impair insulin-
stimulated glucose uptake in 3T3-L1 adipocytes. Am J Physiol Endocrinol Metab.
2004, 287, E758-66.
125

(88) Kitazawa, M., Ohizumi, Y., Oike, Y., Hishinuma, T. and Hashimoto, S.
Angiopoietin-related growth factor suppresses gluconeogenesis through the
Akt/forkhead box class O1-dependent pathway in hepatocytes. J Pharmacol Exp
Ther. 2007, 323, 787-93.

(89) Greene, M. W., Sakaue, H., Wang, L., Alessi, D. R. and Roth, R. A. Modulation
of insulin-stimulated degradation of human insulin receptor substrate-1 by Serine
312 phosphorylation. J Biol Chem. 2003, 278, 8199-211.

(90) Pullen, R. A., Lindsay, D. G., Wood, S. P., Tickle, I. J., Blundell, T. L., et al.
Receptor-binding region of insulin. Nature. 1976, 259, 369-73.

(91) Limbird, L. E., Cell surface receptors: A short course on theory and methods. 3rd
Ed. 2005, Springer. 219.

(92) Jonasson, P., Nilsson, J., Samuelsson, E., Moks, T., Stahl, S., et al. Single-step
trypsin cleavage of a fusion protein to obtain human insulin and its C peptide. Eur
J Biochem. 1996, 236, 656-61.

(93) Davis, R. J., Corvera, S. and Czech, M. P. Insulin stimulates cellular iron uptake
and causes the redistribution of intracellular transferrin receptors to the plasma
membrane. J Biol Chem. 1986, 261, 8708-11.

(94) Ronnett, G. V., Knutson, V. P. and Lane, M. D. Insulin-induced down-regulation
of insulin receptors in 3T3-L1 adipocytes. Altered rate of receptor inactivation. J
Biol Chem. 1982, 25, 4285-91.

(95) Ciechanover, A., Schwartz, A. L., Dautry-Varsat, A. and Lodish, H. F. Kinetics of
internalization and recycling of transferrin and the transferrin receptor in a human
hepatoma cell line. Effect of lysosomotropic agents. J Biol Chem. 1983, 258,
9681-9.

(96) Levy, J. R., Ullrich, A. and Olefsky, J. M. Endocytotic uptake, processing, and
retroendocytosis of human biosynthetic proinsulin by rat fibroblasts transfected
with the human insulin receptor gene. J Clin Invest. 1988, 81, 1370-7.

(97) Desbuquois, B., Chauvet, G., Kouach, M. and Authier, F. Cell itinerary and
metabolic fate of proinsulin in rat liver: in vivo and in vitro studies.
Endocrinology. 2003, 144, 5308-21.

(98) Bergenstal, R. M., Cohen, R. M., Lever, E., Polonsky, K., Jaspan, J., et al. The
metabolic effects of biosynthetic human proinsulin in individuals with type I
diabetes. J Clin Endocrinol Metab. 1984, 58, 973-9.
126

(99) Revers, R. R., Henry, R., Schmeiser, L., Kolterman, O., Cohen, R., et al. The
effects of biosynthetic human proinsulin on carbohydrate metabolism. Diabetes.
1984, 33, 762-70.

(100) Lavelle-Jones, M., Scott, M. H., Kolterman, O., Rubenstein, A. H., Olefsky, J. M.,
et al. Selective suppression of hepatic glucose output by human proinsulin in the
dog. Am J Physiol. 1987, 252, E230-6.

(101) Li, H. and Qian, Z. M. Transferrin/transferrin receptor-mediated drug delivery.
Med Res Rev. 2002, 22, 225-50.

(102) Qian, Z. M., Li, H., Sun, H. and Ho, K. Targeted drug delivery via the transferrin
receptor-mediated endocytosis pathway. Pharmacol Rev. 2002, 54, 561-87.

(103) Willingham, M. C., Hanover, J. A., Dickson, R. B. and Pastan, I. Morphologic
characterization of the pathway of transferrin endocytosis and recycling in human
KB cells. Proc Natl Acad Sci U S A. 1984, 81, 175-9.

(104) Wang, Y., Chen, Y. S., Zaro, J. L. and Shen, W. C. Receptor-mediated activation
of a proinsulin-transferrin fusion protein in hepatoma cells. J Control Release.
2011, 155, 386-92.

(105) Grant, B. D. and Donaldson, J. G. Pathways and mechanisms of endocytic
recycling. Nat Rev Mol Cell Biol. 2009, 10, 597-608.

(106) Smeekens, S. P., Montag, A. G., Thomas, G., Albiges-Rizo, C., Carroll, R., et al.
Proinsulin processing by the subtilisin-related proprotein convertases furin, PC2,
and PC3. Proc Natl Acad Sci U S A. 1992, 89, 8822-6.

(107) Seidah, N. G., Chretien, M. and Day, R. The family of subtilisin/kexin like pro-
protein and pro-hormone convertases: divergent or shared functions. Biochimie.
1994, 76, 197-209.

(108) Vollenweider, F., Irminger, J. C., Gross, D. J., Villa-Komaroff, L. and Halban, P.
A. Processing of proinsulin by transfected hepatoma (FAO) cells. J Biol Chem.
1992, 267, 14629-36.

(109) Shah, D. and Shen, W. C. Transcellular delivery of an insulin-transferrin
conjugate in enterocyte-like Caco-2 cells. J Pharm Sci. 1996, 85, 1306-11.

(110) Sorensen, A. R., Stidsen, C. E., Ribel, U., Nishimura, E., Sturis, J., et al. Insulin
detemir is a fully efficacious, low affinity agonist at the insulin receptor. Diabetes
Obes Metab. 2010, 12, 665-73.
127
(111) Birkeland, K. I., Home, P. D., Wendisch, U., Ratner, R. E., Johansen, T., et al.
Insulin degludec in type 1 diabetes: a randomized controlled trial of a new-
generation ultra-long-acting insulin compared with insulin glargine. Diabetes
Care. 2011, 34, 661-5.

(112) Gerich, J. E. Insulin glargine: long-acting basal insulin analog for improved
metabolic control. Curr Med Res Opin. 2004, 20, 31-7.

(113) Cherrington, A. D. Banting Lecture 1997. Control of glucose uptake and release
by the liver in vivo. Diabetes. 1999, 48, 1198-214.

(114) DeFronzo, R. A., Jacot, E., Jequier, E., Maeder, E., Wahren, J., et al. The effect of
insulin on the disposal of intravenous glucose. Results from indirect calorimetry
and hepatic and femoral venous catheterization. Diabetes. 1981, 30, 1000-7.

(115) Pickup, J. C. and Renard, E. Long-acting insulin analogs versus insulin pump
therapy for the treatment of type 1 and type 2 diabetes. Diabetes Care. 2008, 31,
S140-5.

(116) Shojaee-Moradie, F., Powrie, J. K., Sundermann, E., Spring, M. W., Schettler, A.,
et al. Novel hepatoselective insulin analog: studies with a covalently linked
thyroxyl-insulin complex in humans. Diabetes Care. 2000, 23, 1124-9.

(117) Vienberg, S. G., Bouman, S. D., Sorensen, H., Stidsen, C. E., Kjeldsen, T., et al.
Receptor-isoform-selective insulin analogues give tissue-preferential effects.
Biochem J. 2011, 440, 301-8.

(118) Hedaya, M. A., Basic Pharmacokinetics. 2011, CRC Press. 595.

(119) Schmittgen, T. D. and Livak, K. J. Analyzing real-time PCR data by the
comparative C(T) method. Nat Protoc. 2008, 3, 1101-8.

(120) Wu, C., Okar, D. A., Kang, J. and Lange, A. J. Reduction of hepatic glucose
production as a therapeutic target in the treatment of diabetes. Curr Drug Targets
Immune Endocr Metabol Disord. 2005, 5, 51-9.

(121) van Schaftingen, E. and Gerin, I. The glucose-6-phosphatase system. Biochem J.
2002, 362, 513-32.

(122) Yang, J., Kalhan, S. C. and Hanson, R. W. What is the metabolic role of
phosphoenolpyruvate carboxykinase? J Biol Chem. 2009, 284, 27025-9.

128
(123) Chen, X., Lee, H. F., Zaro, J. L. and Shen, W. C. Effects of receptor binding on
plasma half-life of bifunctional transferrin fusion proteins. Mol Pharm. 2011, 8,
457-65.

(124) Krippendorff, B. F., Kuester, K., Kloft, C. and Huisinga, C. Nonlinear
pharmacokinetics of therapeutic proteins resulting from receptor mediated
endocytosis. J Pharmacokinet Pharmacodyn. 2009, 36, 239-60.

(125) Meibohm, B., ed. Pharmacokinetics and Pharmacodynamics of Biotech Drugs:
Principles and Case Studies in Drug Development. 2006, Wiley-VCH Verlag
GmbH,. Weinheim. 426.

(126) Olson, A. L. and Pessin, J. E. Structure, function, and regulation of the
mammalian facilitative glucose transporter gene family. Annu Rev Nutr. 1996,
16, 235-56.

(127) Galloway, J. A., Root, M. A., Bergstrom, R., Spradlin, C. T., Howey, D. C., et al.
Clinical pharmacologic studies with human insulin (recombinant DNA). Diabetes
Care. 1982, 5 Suppl 2, 13-22.

(128) Arcelloni, C., Falqui, L., Martinenghi, S., Stabilini, A., Pontiroli, A. E., et al.
Processing and release of human proinsulin-cleavage products into culture media
by different engineered non-endocrine cells: a specific assessment by capillary
electrophoresis. J Endocrinol. 2000, 166, 437-45.

(129) Short, D. K., Okada, S., Yamauchi, K. and Pessin, J. E. Adenovirus-mediated
transfer of a modified human proinsulin gene reverses hyperglycemia in diabetic
mice. Am J Physiol. 1998, 275, E748-56.

(130) Abai, A. M., Hobart, P. M. and Barnhart, K. M. Insulin delivery with plasmid
DNA. Hum Gene Ther. 1999, 10, 2637-49.

(131) Jansson, P. A., Larsson, A., Smith, U. and Lonnroth, P. Lactate release from the
subcutaneous tissue in lean and obese men. J Clin Invest. 1994, 93, 240-6.

(132) Chen, G. and Pramanik, B. N. LC-MS for protein characterization: current
capabilities and future trends. Exp Rev Proteomics. 2008, 5, 435-44.

(133) Ayala, J. E., Bracy, D. P., McGuinness, O. P. and Wasserman, D. H.
Considerations in the Design of Hyperinsulinemic-Euglycemic Clamps in the
Conscious Mouse. Diabetes. 2006, 55, 390-7.

129
ALPHABETIZED REFERENCES

Abai, A. M., Hobart, P. M. and Barnhart, K. M. Insulin delivery with plasmid DNA. Hum
Gene Ther. 1999, 10, 2637-49.

Amet, N., Lee, H. F. and Shen, W. C. Insertion of the designed helical linker led to
increased expression of tf-based fusion proteins. Pharm Res. 2009, 26, 523-8.

Amet, N., Wang, W. and Shen, W. C. Human growth hormone-transferrin fusion protein
for oral delivery in hypophysectomized rats. J Control Release. 2010, 141, 177-82.

Arakawa, T., Prestrelski, S. J., Narhi, L. O., Boone, T. C. and Kenney, W. C. Cysteine 17
of recombinant human granulocyte-colony stimulating factor is partially solvent-exposed.
J Protein Chem. 1993, 12, 525-31.

Arcelloni, C., Falqui, L., Martinenghi, S., Stabilini, A., Pontiroli, A. E., et al. Processing
and release of human proinsulin-cleavage products into culture media by different
engineered non-endocrine cells: a specific assessment by capillary electrophoresis. J
Endocrinol. 2000, 166, 437-45.

Ayala, J. E., Bracy, D. P., McGuinness, O. P. and Wasserman, D. H. Considerations in
the Design of Hyperinsulinemic-Euglycemic Clamps in the Conscious Mouse. Diabetes.
2006, 55, 390-7.

Bai, Y., Ann, D. K. and Shen, W. C. Recombinant granulocyte colony-stimulating factor-
transferrin fusion protein as an oral myelopoietic agent. Proc Natl Acad Sci U S A. 2005,
102, 7292-6.

Baldi, L., Hacker, D. L., Adam, M. and Wurm, F. M. Recombinant protein production by
large-scale transient gene expression in mammalian cells: state of the art and future
perspectives. Biotechnol Lett. 2007, 29, 677-84.

Barbosa, M. D. Immunogenicity of biotherapeutics in the context of developing
biosimilars and biobetters. Drug Discov Today. 2012, 16, 345-53.

Bergenstal, R. M., Cohen, R. M., Lever, E., Polonsky, K., Jaspan, J., et al. The metabolic
effects of biosynthetic human proinsulin in individuals with type I diabetes. J Clin
Endocrinol Metab. 1984, 58, 973-9.

Birkeland, K. I., Home, P. D., Wendisch, U., Ratner, R. E., Johansen, T., et al. Insulin
degludec in type 1 diabetes: a randomized controlled trial of a new-generation ultra-long-
acting insulin compared with insulin glargine. Diabetes Care. 2011, 34, 661-5.

130
Brandsma, M. E., Jevnikar, A. M. and Ma, S. Recombinant human transferrin: beyond
iron binding and transport. Biotech Adv. 2011, 29, 230-8.

Brown, L. R. Commercial challenges of protein drug delivery. Expert Opin Drug Deliv.
2005, 2, 29-42.

Chance, R. E. and Frank, B. H. Research, development, production, and safety of
biosynthelic human insulin. Diabetes Care. 1993, 16, 133-142.

Chen, G. and Pramanik, B. N. LC-MS for protein characterization: current capabilities
and future trends. Exp Rev Proteomics. 2008, 5, 435-44.

Chen, X., Lee, H. F., Zaro, J. L. and Shen, W. C. Effects of receptor binding on plasma
half-life of bifunctional transferrin fusion proteins. Mol Pharm. 2011, 8, 457-65.

Cherrington, A. D. Banting Lecture 1997. Control of glucose uptake and release by the
liver in vivo. Diabetes. 1999, 48, 1198-214.

Chien, Y. W. Human insulin: basic sciences to therapeutic uses. Drug Dev. Ind. Pharm.
1996, 22, 752-789.

Ciechanover, A., Schwartz, A. L., Dautry-Varsat, A. and Lodish, H. F. Kinetics of
internalization and recycling of transferrin and the transferrin receptor in a human
hepatoma cell line. Effect of lysosomotropic agents. J Biol Chem. 1983, 258, 9681-9.

Clore, J. N. and Thurby-Hay, L. Which basal insulin should be used in patients with type
2 diabetes mellitus? Curr Diab Rep. 2007, 7, 348-52.

Culajay, J. F., Blaber, S. I., Khurana, A. and Blaber, M. Thermodynamic characterization
of mutants of human fibroblast growth factor 1 with an increased physiological half-life.
Biochemistry. 2000, 39, 7153-8.

Daro, E., van der Sluijs, P., Galli, T. and Mellman, I. Rab4 and cellubrevin define
different early endosome populations on the pathway of transferrin receptor recycling.
Proc Natl Acad Sci U S A. 1996, 93, 9559-64.

Davis, R. J., Corvera, S. and Czech, M. P. Insulin stimulates cellular iron uptake and
causes the redistribution of intracellular transferrin receptors to the plasma membrane. J
Biol Chem. 1986, 261, 8708-11.

Day, C., Archer, H. and Bailey, C. J. Recent advances in insulin therapy. Br J Cardio.
2003, 10.

131
DeFronzo, R. A., Jacot, E., Jequier, E., Maeder, E., Wahren, J., et al. The effect of insulin
on the disposal of intravenous glucose. Results from indirect calorimetry and hepatic and
femoral venous catheterization. Diabetes. 1981, 30, 1000-7.

De Meyts, P., Roth, J., Neville, D. M., Jr., Gavin, J. R. and Lesniak, M. A. Insulin
interactions with its receptors: experimental evidence for negative cooperativity.
Biochem Biophys Res Commun. 1973, 55, 154-161.

De Meyts, P., Van Obberghen, E. and Roth, J. Mapping of the residues responsible for
the negative cooperativity of the receptor-binding region of insulin. Nature. 1978, 273,
504-9.

De Meyts, P. and Whittaker, J. Structural biology of insulin and IGF1 receptors:
implications for drug design. Nat Rev Drug Discov. 2002, 1, 769-83.

Desbuquois, B., Chauvet, G., Kouach, M. and Authier, F. Cell itinerary and metabolic
fate of proinsulin in rat liver: in vivo and in vitro studies. Endocrinology. 2003, 144,
5308-21.

Du, T., Hu, K., Yang, J., Jin, J., Li, C., et al. Bifunctional CD4-DC-SIGN Fusion Proteins
Demonstrate Enhanced Avidity to gp120 and Inhibit HIV-1 Infection and Dissemination.
Antimicrob Agents Chemother. 2012, Jun 11.

Duckworth, W. C., Bennett, R. G. and Hamel, F. G. Insulin degradation: progress and
potential. Endocr Rev. 1998, 19, 608-24.

Duttaroy, A., Kanakaraj, P., Osborn, B. L., Schneider, H., Pickeral, O. K., et al.
Development of a long-acting insulin analog using albumin fusion technology. Diabetes.
2005, 54, 251-8.

Elsadek, B. and Kratz, F. Impact of albumin on drug delivery - new applications on the
horizon. J Control Release. 2012, 157, 4-28.

Espinal, J. Understanding insulin action: principles and molecular mechanisms.
Chichester: Ellis Horwood. 1989.

Fishman, J. B. and Fine, R. E. A trans Golgi-derived exocytic coated vesicle can contain
both newly synthesized cholinesterase and internalized transferrin. Cell. 1987, 48, 157-
64.

Freychet, P. The interactions of proinsulin with insulin receptors on the plasma
membrane of the liver. J Clin Invest. 1974, 54, 1020-31.

132
Frokjaer, S. and Otzen, D. E. Protein drug stability: a formulation challenge. Nat Rev
Drug Discov. 2005, 4, 298-306.

Galloway, J. A. New directions in drug development: mixtures, analogues, and modeling.
Diabetes Care. 1993, 16 Suppl 3, 16-23.

Galloway, J. A., Hooper, S. A., Spradlin, C. T., Howey, D. C., Frank, B. H., et al.
Biosynthetic human proinsulin. Review of chemistry, in vitro and in vivo receptor
binding, animal and human pharmacology studies, and clinical trial experience. Diabetes
Care. 1992, 15, 666-92.

Galloway, J. A., Root, M. A., Bergstrom, R., Spradlin, C. T., Howey, D. C., et al. Clinical
pharmacologic studies with human insulin (recombinant DNA). Diabetes Care. 1982, 5
Suppl 2, 13-22.

Gerich, J. E. Insulin glargine: long-acting basal insulin analog for improved metabolic
control. Curr Med Res Opin. 2004, 20, 31-7.

Gershoni, J. M. Molecular decoys: antidotes, therapeutics and immunomodulators. Curr
Opin Biotechnol. 2008, 19, 644-51.

Glauber, H. S., Henry, R. R., Wallace, P., Frank, B. H., Galloway, J. A., et al. The effects
of biosynthetic human proinsulin on carbohydrate metabolism in non-insulin-dependent
diabetes mellitus. N Engl J Med. 1987, 316, 443-9.

Graslund, S., Nordlund, P., Weigelt, J., Hallberg, B. M., Bray, J., et al. Protein production
and purification. Nat Methods. 2008, 5, 135-46.

Given, B. D., Cohen, R. M., Shoelson, S. E., Frank, B. H., Rubenstein, A. H., et al.
Biochemical and clinical implications of proinsulin conversion intermediates. J Clin
Invest. 1985, 76, 1398-405.

Goldenberg, M. M. Etanercept, a novel drug for the treatment of patients with severe,
active rheumatoid arthritis. Clin Ther. 1999, 21, 75-87.

Gomme, P. T., McCann, K. B. and Bertolini, J. Transferrin: structure, function and
potential therapeutic actions. Drug Discov Today. 2005, 10, 267-73.

Grant, B. D. and Donaldson, J. G. Pathways and mechanisms of endocytic recycling. Nat
Rev Mol Cell Biol. 2009, 10, 597-608

Greene, M. W., Sakaue, H., Wang, L., Alessi, D. R. and Roth, R. A. Modulation of
insulin-stimulated degradation of human insulin receptor substrate-1 by Serine 312
phosphorylation. J Biol Chem. 2003, 278, 8199-211.
133

Kitazawa, M., Ohizumi, Y., Oike, Y., Hishinuma, T. and Hashimoto, S. Angiopoietin-
related growth factor suppresses gluconeogenesis through the Akt/forkhead box class O1-
dependent pathway in hepatocytes. J Pharmacol Exp Ther. 2007, 323, 787-93.

Krippendorff, B. F., Kuester, K., Kloft, C. and Huisinga, C. Nonlinear pharmacokinetics
of therapeutic proteins resulting from receptor mediated endocytosis. J Pharmacokinet
Pharmacodyn. 2009, 36, 239-60.

Halpern, W., Riccobene, T. A., Agostini, H., Baker, K., Stolow, D., et al. Albugranin, a
recombinant human granulocyte colony stimulating factor (G-CSF) genetically fused to
recombinant human albumin induces prolonged myelopoietic effects in mice and
monkeys. Pharm Res. 2002, 19, 1720-9.

Harmon, A. W., Paul, D. S. and Patel, Y. M. MEK inhibitors impair insulin-stimulated
glucose uptake in 3T3-L1 adipocytes. Am J Physiol Endocrinol Metab. 2004, 287, E758-
66.

Hedaya, M. A., Basic Pharmacokinetics. 2011, CRC Press. 595.

Hirsch, I. B. Insulin analogues. N Engl J Med. 2005, 352, 174-83.

Howey, D. C., Frank, B. H., Galloway, J. A., Root, M. A. and Spradlin, C. T. Human
proinsulin (rDNA) as an intermediate-acting insulin agonist: a comparison with NPH (N)
and regular (R) insulins (Abstract). Diabetes. 1984, 33 Suppl 1 23A.

Huang, C. Receptor-Fc fusion therapeutics, traps, and MIMETIBODY technology. Curr
Opin Biotechnol. 2009, 20, 692-9.

Huang, B. C., Davern, S. and Kennel, S. J. Mono and bivalent binding of a scFv and
covalent diabody to murine laminin-1 using radioiodinated proteins and SPR
measurements: Effects on tissue retention in vivo. J Immunol Methods. 2006, 313, 149-
60.

Huebers, H., Huebers, E., Rummel, W. and Crichton, R. R. Isolation and characterization
of iron-binding proteins from rat intestinal mucosa. Eur J Biochem. 1976, 66, 447-55.

Jansson, P. A., Larsson, A., Smith, U. and Lonnroth, P. Lactate release from the
subcutaneous tissue in lean and obese men. J Clin Invest. 1994, 93, 240-6.

Jonasson, P., Nilsson, J., Samuelsson, E., Moks, T., Stahl, S., et al. Single-step trypsin
cleavage of a fusion protein to obtain human insulin and its C peptide. Eur J Biochem.
1996, 236, 656-61.

134
Kahn, C. R. The molecular mechanism of insulin action. Annu Rev Med. 1985, 36, 429-
51.

Kim, B. J., Zhou, J., Martin, B., Carlson, O. D., Maudsley, S., et al. Transferrin fusion
technology: a novel approach to prolong biological half-life of insulinotropic peptides. J
Pharmacol Exp Ther. 2010, 334, 682-92.

Koppu, S., Oh, Y. J., Edrada-Ebel, R., Blatchford, D. R., Tetley, L., et al. Tumor
regression after systemic administration of a novel tumor-targeted gene delivery system
carrying a therapeutic plasmid DNA. J Control Release. 2010, 143, 215-21.

Lavelle-Jones, M., Scott, M. H., Kolterman, O., Rubenstein, A. H., Olefsky, J. M., et al.
Selective suppression of hepatic glucose output by human proinsulin in the dog. Am J
Physiol. 1987, 252, E230-6.

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

Leahy, J. L. Basal-prandial insulin therapy: Scientific concept review and application.
Am J Med Sci. 2006, 332, 24-31.

Leahy, J. L. and Cefalu, W. T., eds. Intensive insulin therapy in type 1 diabetes mellitus.
Insulin Therapy. 2002, Marcel Dekker: NY. pp 87-112.

Lencer, W. I. and Blumberg, R. S. A passionate kiss, then run: exocytosis and recycling
of IgG by FcRn. Trends Cell Biol. 2005, 15, 5-9.

Levy, J. R., Ullrich, A. and Olefsky, J. M. Endocytotic uptake, processing, and
retroendocytosis of human biosynthetic proinsulin by rat fibroblasts transfected with the
human insulin receptor gene. J Clin Invest. 1988, 81, 1370-7.

Li, H. and Qian, Z. M. Transferrin/transferrin receptor-mediated drug delivery. Med Res
Rev. 2002, 22, 225-50.

Lim, C. J. and Shen, W. C. Transferrin-oligomers as potential carriers in anticancer drug
delivery. Pharm Res. 2004, 21, 1985-92.

Limbird, L. E., Cell surface receptors: A short course on theory and methods. 3rd Edition
ed. 2005, Springer. 219.

Lok, C. N. and Loh, T. T. Regulation of transferrin function and expression: review and
update. Biol Signals Recept. 1998, 7, 157-78.

135
MacGillivray, R. T., Moore, S. A., Chen, J., Anderson, B. F., Baker, H., et al. Two high-
resolution crystal structures of the recombinant N-lobe of human transferrin reveal a
structural change implicated in iron release. Biochemistry. 1998, 37, 7919-28.

Mahmood, I. and Green, M. D. Pharmacokinetic and pharmacodynamic considerations in
the development of therapeutic proteins. Clin Pharmacokinet. 2005, 44, 331-47.

Marshall, S. A., Lazar, G. A., Chirino, A. J. and Desjarlais, J. R. Rational design and
engineering of therapeutic proteins. Drug Discov Today. 2003, 8, 212-21.

Mason, A. B., He, Q. Y., Halbrooks, P. J., Everse, S. J., Gumerov, D. R., et al.
Differential effect of a his tag at the N- and C-termini: functional studies with
recombinant human serum transferrin. Biochemistry. 2002, 41, 9448-54.

Mayor, S., Presley, J. F. and Maxfield, F. R. Sorting of membrane components from
endosomes and subsequent recycling to the cell surface occurs by a bulk flow process. J
Cell Biol. 1993, 121, 1257-69.

McAulay, V. and Frier, B. M. Insulin analogues and other developments in insulin
therapy for diabetes. Expert Opin Pharmacother. 2003, 4, 1141-56.

Meibohm, B., ed. Pharmacokinetics and Pharmacodynamics of Biotech Drugs:
Principles and Case Studies in Drug Development. 2006, Wiley-VCH Verlag GmbH,.
Weinheim. 426.

Morgan, E. H. and Baker, E. Iron uptake and metabolism by hepatocytes. Fed Proc. 1986,
45, 2810-6.

Morgan, E. H., Smith, G. D. and Peters, T. J. Uptake and subcellular processing of 59Fe-
125I-labelled transferrin by rat liver. Biochem J. 1986, 237, 163-73.

Mukherjee, S., Ghosh, R. N. and Maxfield, F. R. Endocytosis. Physiol Rev. 1997, 77,
759-803.

Olson, A. L. and Pessin, J. E. Structure, function, and regulation of the mammalian
facilitative glucose transporter gene family. Annu Rev Nutr. 1996, 16, 235-56.

Overbay, D. K. and Manley, H. J. Darbepoetin-alpha: a review of the literature.
Pharmacotherapy. 2002, 22, 889-97.

Owens, D. R. and Bolli, G. B. Beyond the era of NPH insulin--long-acting insulin
analogs: chemistry, comparative pharmacology, and clinical application. Diabetes
Technol Ther. 2008, 10, 333-49.

136
Parkkinen, J., von Bonsdorff, L., Ebeling, F. and Sahlstedt, L. Function and therapeutic
development of apotransferrin. Vox Sang. 2002, 83 Suppl 1, 321-6.

Peavy, D. E., Abram, J. D., Frank, B. H. and Duckworth, W. C. In vitro activity of
biosynthetic human proinsulin: receptor binding and biological potency of proinsulin and
insulin in isolated rat adipocytes. Diabetes. 1984, 33, 1062-67.

Pickup, J. C. and Renard, E. Long-acting insulin analogs versus insulin pump therapy for
the treatment of type 1 and type 2 diabetes. Diabetes Care. 2008, 31, S140-5.

Probst, I., Hartmann, H., Jungermann, K. and Creutzfeldt, W. Insulin-like action of
proinsulin on rat liver carbohydrate metabolism in vitro. Diabetes. 1985, 34, 415-9.

Pullen, R. A., Lindsay, D. G., Wood, S. P., Tickle, I. J., Blundell, T. L., et al. Receptor-
binding region of insulin. Nature. 1976, 259, 369-73.

Qian, Z. M., Li, H., Sun, H. and Ho, K. Targeted drug delivery via the transferrin
receptor-mediated endocytosis pathway. Pharmacol Rev. 2002, 54, 561-87.

Rader, R., BIOPHARMA: Biopharmaceutical products in the U.S. and European
markets. 2012, Rockville, MD: Biotechnology Information Institute.

Revers, R. R., Henry, R., Schmeiser, L., Kolterman, O., Cohen, R., et al. The effects of
biosynthetic human proinsulin on carbohydrate metabolism. Diabetes. 1984, 33, 762-70.

Roberts, M. J., Bentley, M. D. and Harris, J. M. Chemistry for peptide and protein
PEGylation. Adv Drug Deliv Rev. 2002, 54, 459-76.

Ronnett, G. V., Knutson, V. P. and Lane, M. D. Insulin-induced down-regulation of
insulin receptors in 3T3-L1 adipocytes. Altered rate of receptor inactivation. J Biol
Chem. 1982, 25, 4285-91.

Roopenian, D. C. and Akilesh, S. FcRn: the neonatal Fc receptor comes of age. Nat Rev
Immunol. 2007, 7, 715-25.

Saltiel, A. R. and Pessin, J. E., eds. Hepatic regulation of fuel metabolism. Mechanisms
of Insulin Action, ed. C. Clark and C.B. Newgard. 2007, Spinger: New York. pp 90-97.

Sarkar, C. A., Lowenhaupt, K., Horan, T., Boone, T. C., Tidor, B., et al. Rational
cytokine design for increased lifetime and enhanced potency using pH-activated
"histidine switching". Nat Biotechnol. 2002, 20, 908-13.

137
Schatz, H., Ammermann, S., Laube, H. and Federlin, K. Bioactivity and
pharmacokinetics of human proinsulin in comparison to human insulin after intravenous
and subcutaneous injection. Horm Metab Res. 1988, 20, 445-9.

Scheich, C., and Sievert, V. An automated method for high-throughput protein
purification applied to a comparison of His-tag and GST-tag affinity chromatography.
BMC Biotech. 2003, 3, 12-9.

Schellekens, H. Bioequivalence and the immunogenicity of biopharmaceuticals. Nat Rev
Drug Discov. 2002, 1, 457-62.

Schmidt, S. R. Fusion-proteins as biopharmaceuticals--applications and challenges. Curr
Opin Drug Discov Devel. 2009, 12, 284-95.

Schmittgen, T. D. and Livak, K. J. Analyzing real-time PCR data by the comparative
C(T) method. Nat Protoc. 2008, 3, 1101-8.

Schroeder, H. W. and Cavacini, L. Structure and function of immunoglobulins. J Allergy
Clin Immunol. 2010, 125, S41-52.

Seidah, N. G., Chretien, M. and Day, R. The family of subtilisin/kexin like pro-protein
and pro-hormone convertases: divergent or shared functions. Biochimie. 1994, 76, 197-
209.

Sethuraman, N. and Stadheim, T. A. Challenges in therapeutic glycoprotein production.
Curr Opin Biotechnol. 2006, 17, 341-6.

Shah, D. and Shen, W. C. Transcellular delivery of an insulin-transferrin conjugate in
enterocyte-like Caco-2 cells. J Pharm Sci. 1996, 85, 1306-11.

Shojaee-Moradie, F., Powrie, J. K., Sundermann, E., Spring, M. W., Schuttler, A., et al.
Novel hepatoselective insulin analog: studies with a covalently linked thyroxyl-insulin
complex in humans. Diabetes Care. 2000, 23, 1124-9.

Short, D. K., Okada, S., Yamauchi, K. and Pessin, J. E. Adenovirus-mediated transfer of
a modified human proinsulin gene reverses hyperglycemia in diabetic mice. Am J
Physiol. 1998, 275, E748-56.

Smeekens, S. P., Montag, A. G., Thomas, G., Albiges-Rizo, C., Carroll, R., et al.
Proinsulin processing by the subtilisin-related proprotein convertases furin, PC2, and
PC3. Proc Natl Acad Sci U S A. 1992, 89, 8822-6.

Sonksen, P. and Sonksen, J. Insulin: understanding its action in health and disease. Br J
Anaesth. 2000, 85, 69-79.
138
Sonnichsen, B., De Renzis, S., Nielsen, E., Rietdorf, J. and Zerial, M. Distinct membrane
domains on endosomes in the recycling pathway visualized by multicolor imaging of
Rab4, Rab5, and Rab11. J Cell Biol. 2000, 149, 901-14.

Sorensen, A. R., Stidsen, C. E., Ribel, U., Nishimura, E., Sturis, J., et al. Insulin detemir
is a fully efficacious, low affinity agonist at the insulin receptor. Diabetes Obes Metab.
2010, 12, 665-73.

Stoorvogel, W., Geuze, H. J., Griffith, J. M. and Strous, G. J. The pathways of
endocytosed transferrin and secretory protein are connected in the trans-Golgi reticulum.
J Cell Biol. 1988, 106, 1821-9.

Subramanian, G. M., Fiscella, M., Lamous-Smith, A., Zeuzem, S. and McHutchison, J.
G. Albinterferon alpha-2b: a genetic fusion protein for the treatment of chronic hepatitis
C. Nat Biotechnol. 2007, 25, 1411-9.

Tomkin, G. H. Albiglutide, an albumin-based fusion of glucagon-like peptide 1 for the
potential treatment of type 2 diabetes. Curr Opin Mol Ther. 2009, 11, 579-88.

Tomlinson, I. M. Next-generation protein drugs. Nat Biotechnol. 2004, 22, 521-2.

van Schaftingen, E. and Gerin, I. The glucose-6-phosphatase system. Biochem J. 2002,
362, 513-32.

Vienberg, S. G., Bouman, S. D., Sorensen, H., Stidsen, C. E., Kjeldsen, T., et al.
Receptor-isoform-selective insulin analogues give tissue-preferential effects. Biochem J.
2011, 440, 301-8.

Vollenweider, F., Irminger, J. C., Gross, D. J., Villa-Komaroff, L. and Halban, P. A.
Processing of proinsulin by transfected hepatoma (FAO) cells. J Biol Chem. 1992, 267,
14629-36.

Walsh, C. T., Posttranslational modification of proteins: expanding Nature's inventory.
2005, Colorado: Roberts & Company.

Wang, Y., Chen, Y. S., Zaro, J. L. and Shen, W. C. Receptor-mediated activation of a
proinsulin-transferrin fusion protein in hepatoma cells. J Control Release. 2011, 155,
386-92.

White, M. F. The insulin signaling system and the IRS proteins. Diabetologia. 1997, 40
Suppl 2, S2-17.

139
Willingham, M. C., Hanover, J. A., Dickson, R. B. and Pastan, I. Morphologic
characterization of the pathway of transferrin endocytosis and recycling in human KB
cells. Proc Natl Acad Sci U S A. 1984, 81, 175-9.

Winter, J., Lilie, H. and Rudolph, R. Renaturation of human proinsulin--a study on
refolding and conversion to insulin. Anal Biochem. 2002, 310, 148-55.

Widera, A., Norouziyan, F. and Shen, W. C. Mechanisms of TfR-mediated transcytosis
and sorting in epithelial cells and applications toward drug delivery. Adv Drug Deliv
Rev. 2003, 55, 1439-66.

Woodnutt, G., Violand, B. and North, M. Advances in protein therapeutics. Curr Opin
Drug Discov Devel. 2008, 11, 754-61.

Wu, C., Okar, D. A., Kang, J. and Lange, A. J. Reduction of hepatic glucose production
as a therapeutic target in the treatment of diabetes. Curr Drug Targets Immune Endocr
Metabol Disord. 2005, 5, 51-9.

Wurm, F. M. Production of recombinant protein therapeutics in cultivated mammalian
cells. Nat Biotechnol. 2004, 22, 1393-8.

Yang, J., Kalhan, S. C. and Hanson, R. W. What is the metabolic role of
phosphoenolpyruvate carboxykinase? J Biol Chem. 2009, 284, 27025-9.

Abstract (if available)

Abstract Long-acting insulin (INS) analogues that exhibit prolonged time-action profiles and liver-specificity are currently in great demand for diabetes treatment. Native INS and its protein precursor proinsulin (ProINS) are both small peptides with short in vivo half-life and efficacy. Human transferrin (Tf) is a stable and large-sized plasma protein, and it has been demonstrated to prolong the half-life of small proteins. With the purposes of improving the therapeutic application of INS, a proinsulin-transferrin (ProINS-Tf) recombinant fusion protein has been designed and developed. This fusion protein is produced using recombinant fusion technology combined with his-tag purification method. ProINS-Tf exhibits a low activity in the 30 min promotion of glucose uptake in adipocytes, which corresponds with a low binding affinity to insulin receptor. Additionally, ProINS-Tf can be activated by ex vivo trypsin digestion. These results suggest that, similar to ProINS, the intrinsic potency of ProINS-Tf is low and an activation is required to achieve biological activity. On the other hand, ProINS-Tf elicits a Tf receptor (TfR) dependent enhanced activity in the 24 h inhibition of glucose production in hepatoma cells. Radioimmunoassays clearly demonstrate a TfR-mediated conversion and activation of ProINS-Tf to an immunoreactive insulin-transferrin fusion protein during the 24 h incubation with hepatoma cells. Therefore, we have proposed an intracellular ProINS-Tf activation mechanism that is mediated through TfR-mediated endocytosis and recycling pathway. Furthermore, compared to ProINS and INS, subcutaneously injected ProINS-Tf exerts an extended hypoglycemic efficacy with a prolonged half-life in fasted diabetic mice. The correlation of hypoglycemic efficacy with the suppression of liver-associated enzyme expression suggests a liver-preferential effect by ProINS-Tf. In addition, intravenously injection of trypsin-digested ProINS-Tf shows an immediate hypoglycemic response, whereas ProINS-Tf exhibits a delayed and long-lasting hypoglycemic efficacy. These observations imply that an activation of ProINS-Tf may also occur in vivo. Taken together, results from this dissertation have presented three contributions to the therapeutic protein discovery and development of. First, ProINS-Tf is a novel fusion protein and the first proprotein within the Tf-based fusion protein family. Second, a novel receptor-mediated intracellular proprotein activation mechanism is discovered. Lastly, ProINS-Tf shows great promise as a potential long-acting INS analogue for diabetes treatment.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Mechanism study of proinsulin-transferrin fusion protein as a long-acting and liver-targeting insulin prodrug

PDF

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

PDF

Reconstruction of a proinsulin-transferrin fusion protein

PDF

Preparation and characterization of Tf-G-CSF fusion protein

PDF

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

PDF

Human growth hormone-transferrin recombinant fusion protein for oral delivery

PDF

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

PDF

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

PDF

Insulin sensitivity in cognition, Alzheimer's disease and brain aging

PDF

Targets of protein nitration during aging

PDF

Pten regulates beta-cell regeneration intrinsically and independently of development

PDF

Flipping the switch on protein activity activity: elastin-like polypeptides assemble into cell switches and vesicles

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

Regulation of mitochondrial bioenergetics via PTEN (phosphatase and tensin homolog deleted on chromosome 10)/estrogen-related receptor alpha (ERRα) signaling

PDF

Site-specific PEGylation of recombinant immunotoxin αFAP-PE38 for half-life extension

PDF

Protein aggregation: current scenario and recent developments

PDF

Harnessing environmental and culture conditions to alter fungal ‘omics’

PDF

Steatosis induces Wnt/β-catenin pathway to stimulate proliferation of hepatic tumor initiating cells and promote liver cancer development

PDF

Energy metabolism and inflammation in brain aging: significance of age-dependent astrocyte metabolic-redox profile

PDF

Genetic engineering of fungi to enhance the production and elucidate the biosynthesis of bioactive secondary metabolites

Asset Metadata

Creator Wang, Yan (author)

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

School School of Pharmacy

Degree Doctor of Philosophy

Degree Program Molecular Pharmacology and Toxicology

Publication Date 08/23/2013

Defense Date 07/11/2012

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

Tag Diabetes,insulin,OAI-PMH Harvest,prodrug activation,proinsulin,recombinant fusion protein,transferrin

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Shen, Wei-Chiang (committee chair), Haworth, Ian S. (committee member), Stiles, Bangyan L. (committee member), Wang, Clay C.C. (committee member)

Creator Email helen.y.wang@gmail.com,wang23@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-91307

Unique identifier UC11289906

Identifier usctheses-c3-91307 (legacy record id)

Legacy Identifier etd-WangYan-1165.pdf

Dmrecord 91307

Document Type Dissertation

Rights Wang, Yan

Type texts

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

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA