Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Characterization of rice-derived proinsulin-transferrin fusion protein and its oral bioactivity as a basal insulin analogue
(USC Thesis Other)
Characterization of rice-derived proinsulin-transferrin fusion protein and its oral bioactivity as a basal insulin analogue
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
CHARACTERIZATION OF RICE-DERIVED PROINSULIN-TRANSFERRIN FUSION
PROTEIN AND ITS ORAL BIOACTIVITY AS A BASAL INSULIN ANALOGUE
By
Yu-Sheng Chen
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
(PHARMACEUTICAL SCIENCES)
December 2016
Copyright 2016 Yu-Sheng Chen
i
DEDICATION
To my family,
thank you for all your support and unconditional love.
ii
ACKNOWLEDGEMENTS
First, I would like to express my greatest appreciation to my dissertation advisor,
Dr. Wei-Chiang Shen, as an excellent mentor and role model to show me what
characteristics a good scientist should be equipped with. During my master and Ph.D.
training, his infectious enthusiasm and insightful wisdom in science, his patience and
kindness to people have a tremendous influence of my growth. In the meantime, I would
also like to give my deepest gratitude to Mrs. Daisy Shen for her great support and care in
my life. I feel very fortunate to be under Dr. Shen and Daisy's care throughout the past
seven years. As an international student, I always feel like I am home in Dr. Shen’s
laboratory. Therefore, when it is time to leave USC, I have exactly the same feeling as I
was taking the first plane from Taiwan to pursuit my graduate degree in the United
States.
I also want to thank my committee members, Dr. Curtis Okamoto, Dr. Austin
Mircheff, Dr. Bangyan Stiles and Dr. Julio Camarero. They provided me a lot of help and
advices for both my dissertation projects as well as my future career path. I would also
like to sincerely thank Dr. Jennica Zaro for her insightful suggestions and guidance for
my research projects. She is always so kind and patient when I need her help. Although
she is not working in USC now, I wish her all the best and success in her future
endeavors. At the same time, I also want to express my deep appreciation to Ventria
Bioscience for the support from their innovative transgenic rice expression system.
iii
Without their help, I would never be able to complete my dissertation on time.
In the past years, I have always appreciated the help and support from all the
labmates I have met in Dr. Shen’s lab. I really appreciate Dr. Yan Wang for teaching the
technical skills and knowledge when I just joined the laboratory. I also want to thank Dr.
Xiaoying Chen and Dr. Robert Mo for sharing their past experience with me. I sincerely
appreciate the enormous help I got from Hsin-Fang Lee and Dr. Likun Fei. Last, a very
special thank you goes to Dr. Juntang Shao. Thank her for always being there when I
need help. I am so glad that we could share these unforgettable moments in the past
years.
Finally, I want to sincerely thank my family. My parents, A-Ping Chen and Yun-
Shih Shih, constantly support me to pursuit my career goal as a pharmaceutical scientist. I
thank my brother and sister for their encouragement when I felt stressful. Without their
support, it would be impossible for me to achieve my Ph.D. degree.
iv
TABLE OF CONTENTS
DEDICATION i
ACKNOWLEDGEMENTS ii
TABLE OF CONTENTS iv
LIST OF TABLES viii
LIST OF FIGURES ix
LIST OF SCHEMES xi
ABBREVIATIONS xii
ABSTRACT xv
CHAPTER 1: INTRODUCTION 1
1.1. Oral Delivery of Protein Therapeutics 1
1.1.1. Protein Therapeutics 1
1.1.2. Advanced Strategies for Oral Delivery of Protein Therapeutics 3
1.2. Human Serum Transferrin 4
1.2.1. Background 4
1.2.2. Transferrin as Drug Carrier to Improve Drug Targeting and Delivery 5
1.3. Proinsulin-Transferrin (ProINS-Tf) Recombinant Fusion Protein 7
1.3.1. Diabetes and Insulin 7
1.3.2. ProINS-Tf Fusion Protein from Mammalian Cells 8
1.3.3. TfR-Mediated Conversion of ProINS-Tf Fusion Protein 9
1.3.4. Liver-Preferential Effect of ProINS-Tf Fusion Protein 10
1.3.5. Oral Delivery of ProINS-Tf 11
CHAPTER 2: PREPARATION OF PROINSULIN-TRANSFERRIN FUSION
PROTEINS FROM HEK293 CELLS AND TRANSGENIC RICE SEEDS 16
2.1. Background 16
2.2. Materials and Methods 19
2.2.1. ProINS-Tf Fusion Protein from HEK293 cells 19
v
2.2.1.1. Cell Culture 19
2.2.1.2. Transfection and Expression of Fusion Protein 19
2.2.1.3. Protein Concentration and His-Tag Purification 20
2.2.2. ProINS-Tf Fusion Protein from Transgenic Rice Seeds 21
2.2.2.1. Extraction and Purification of ProINS-Tf from Rice Seed Flour 21
2.3. Results 22
2.3.1. ProINS-Tf Fusion Protein from HEK293 cells 22
2.3.1.1. Increased Purity of HEK-ProINS-Tf 22
2.3.1.2. Total Production Yield of HEK293-derived Fusion Protein 22
2.3.2. ProINS-Tf Fusion Protein from Transgenic Rice Seeds 23
2.3.2.2. S-200HR Size Exclusion Column for Protein Purification 23
2.3.2.3. DEAE Ion Exchange Column for Protein Purification 23
2.3.2.4. Protein Yield of Rice-derived Fusion Protein 24
2.4. Discussion 25
CHAPTER 3: CHARACTERIZATION AND COMPARISON STUDIES OF RICE-
AND HEK293-DERIVED PROINSULIN-TRANSFERRIN FUSION PROTEINS 33
3.1. Background 33
3.2. Materials and Methods 34
3.2.1. Cell Culture 34
3.2.2. Molecular Characterization of Rice-Derived ProINS-Tf 34
3.2.3. Transferrin Receptor Binding Assay 36
3.2.4. Intracellular Conversion of ProINS-Tf in H4IIE Cells 37
3.2.5. Phosphorylation of AKT induced by ProINS-Tf Fusion Proteins 37
3.2.6. Glucose Production Inhibition Assay 38
3.3. Results 38
3.3.1. Molecular Characterization of Rice-Derived ProINS-Tf 38
3.3.2. Transferrin Receptor Binding Assay 39
3.3.3. Intracellular Conversion of ProINS-Tf in H4IIE Cells 39
3.3.4. Phosphorylation of AKT induced by ProINS-Tf Fusion Proteins 40
3.3.5. Glucose Production Inhibition Assay 40
3.4. Discussion 41
vi
CHAPTER 4: IN VIVO STUDIES AND ORAL ADMINISTRATION OF RICE-
DERIVED PROINSULIN-TRANSFERRIN FUSION PROTEIN 48
4.1. Background 48
4.2. Materials and Methods 48
4.2.1. Streptozotocin-Induced Type 1 Diabetic Mouse Model 48
4.2.2. Hypoglycemic Effect of Rice-ProINS-Tf (s.c.) on Type 1 Diabetic Mice 49
4.2.3. Extra-Long Term Hypoglycemic Effect of Rice-Derived ProINS-Tf 50
4.2.4. Oral Hypoglycemic Effect of Rice-Derived ProINS-Tf 50
4.3. Results 51
4.3.1. Hypoglycemic Effect of ProINS-Tf on Type 1 Diabetic Mice 51
4.3.2. Extra-Long Term Hypoglycemic Effect of Rice-Derived ProINS-Tf 52
4.3.3. Oral Hypoglycemic Effect of Rice-Derived ProINS-Tf 52
4.4. Discussion 53
CHAPTER 5: TRANSFERRIN-POLYLYSINE CONJUGATES AS A NOVEL ORAL
PROTEIN CARRIER WITH IMPROVED TRANSCYTOSIS ACROSS GI BARRIER 62
5.1. Background 62
5.2. Materials and Methods 64
5.2.1. Cell Culture 64
5.2.2. Disulfide-Bond Linked Tf-PLL Conjugates (Tf-ss-PLL) 65
5.2.2.1. Synthesis of Tf-ss-PLL 65
5.2.2.2. Caco-2 Cellular Binding Assay of Tf-ss-PLL 66
5.2.2.3. Transcytosis Assay of Tf-ss-PLL 66
5.2.3. Acid-Liable Tf-PLL Conjugates (Tf-NTA-PLL) 67
5.2.3.1. Synthesis of Tf-NTA-PLL 67
5.2.3.2. Caco-2 Binding Assay and pH Sensitivity Test of Tf-NTA-PLL 68
5.2.3.3. Transcytosis Assay of Tf-NTA-PLL 69
5.3. Results 69
5.3.1. Disulfide-Bond Linked Tf-PLL Conjugates (Tf-ss-PLL) 69
5.3.1.1. Synthesis of Tf-ss-PLL 69
5.3.1.2. Cellular Binding Assay of Tf-ss-PLL 70
5.3.1.3. Transcytosis Assay of Tf-ss-PLL 71
vii
5.3.2. Acid-Liable Tf-PLL Conjugates (Tf-NTA-PLL) 71
5.3.2.1. Cellular Binding Assay of Tf-NTA-PLL 71
5.3.2.2. Transcytosis Assay of Tf-NTA-PLL 72
5.4. Discussion 73
CHAPTER 6: SUMMARY AND FUTURE PERSPECTIVES 87
6.1. Rice-ProINS-Tf as an Oral Deliverable Basal Insulin Analog 87
6.1.1. The Rice-derived ProINS-Tf Fusion Protein 87
6.1.2. Transcytosis of Tf-NTA-PLL as an Enhanced Oral Protein Drug Carrier 89
6.1.3. Final Conclusion 90
6.2. Future Perspectives 90
REFERENCES 95
ALPHABETIZED REFERENCES 103
viii
LIST OF TABLES
Table 3-1. Competitive TfR binding assay in HeLa cells 44
Table 3-2. Glucose production inhibition in H4IIE cells 47
ix
LIST OF FIGURES
Figure 1-1. The expression plasmid and structure of ProINS-Tf fusion protein 15
Figure 2-1. His-Tag purification of HEK-ProINS-Tf 28
Figure 2-2. Concentration estimation of HEK-ProINS-Tf using anti-Tf
Western blot 29
Figure 2-3. S-200HR size exclusion chromatography of Rice-ProINS-Tf 30
Figure 2-4. DEAE ion exchange chromatography for Rice-ProINS-Tf
purification 31
Figure 2-5. Anti-(Pro)INS Western blot for estimating the conc. of
Rice-ProINS-Tf 32
Figure 3-1. Molecular characterization of Rice-ProINS-Tf 43
Figure 3-2. Competitive TfR binding assay in HeLa cells 44
Figure 3-3. TfR-mediated conversion of Rice-ProINS-Tf in H4IIE cells 45
Figure 3-4. Stimulation of AKT phosphorylation in H4IIE cells 46
Figure 3-5. Glucose production inhibition in H4IIE cells 47
Figure 4-1. Hypoglycemic efficacy under feeding condition (subcutaneous
injection) 56
Figure 4-2. Hypoglycemic efficacy under fasting condition (subcutaneous
injection) 57
Figure 4-3. Extra-long acting hypoglycemic effect of Rice-ProINS-Tf 58
Figure 4-4. Hypoglycemic effect of Rice-ProINS-Tf via oral administration 59
Figure 4-5. Liver targeting effect of orally administered Rice-ProINS-Tf 60
Figure 5-1. The TEER curve of Caco-2 cell on the 6-well Transwell plates 78
x
Figure 5-2. Characterization of Tf-ss-PLL 81
Figure 5-3. Caco-2 binding assay of Tf-ss-PLL 82
Figure 5-4. Caco-2 binding assay of Tf-NTA-PLL 83
Figure 5-5. pH sensitive dissociation between Tf-6xHis and Ni
2+
-NTA-PLL 84
Figure 5-6. Transcytosis assay of Tf-ss-PLL 85
Figure 5-7. Transcytosis assays of Tf-NTA-PLL 86
xi
LIST OF SCHEMES
Scheme 1-1. TfR-mediated endocytosis and recycling pathway 14
Scheme 2-1. Extraction and purification process of Rice-ProINS-Tf 27
Scheme 4-1. Comparison of pathways of Rice- and HEK-ProINS-Tf in liver
hepatocytes 61
Scheme 5-1. Transcytosis of Tf-PLL conjugates across Caco-2 epithelial cells 77
Scheme 5-2. Synthesis process of Tf-ss-PLL conjugate 79
Scheme 5-3. Synthesis process of Tf-NTA-PLL conjugates 80
xii
ABBREVIATIONS
AUC Area under the curve
Apo-Tf Tf without iron ions
BA Bioavailability
BCA Bicinchoninic acid
BSA Bovine serum albumin
CHO Chinese hamster ovary
CPDB Carboxypeptidase B
CPE Carboxypeptidase E
CPP Cell penetrating peptide
DEAE Diethylaminoethyl
DMEM Dulbecco’s modified Eagle medium
DMT-1 Divalent metal transporter-1
DTT Dithiothreitol
ER Endoplasmic reticulum
FBS Fetal bovine serum
FDA Food and Drug Administration
G6P Glucose-6-phosphate
GAPDH Glyceraldehyde-3-phosphate dehydrogenase
G-CSF Granulocyte-colony stimulating factor
GH Growth hormone
GI Gastrointestinal
xiii
GLP-1 Glucagon-like peptide-1
HBSS Hank's balanced salt solution
HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
His-Tag Hexahistidine tag
Holo-Tf Transferrin with diferric ions
hz Hydrazone
i.p Intraperitoneal
IDE Insulin degrading enzyme
IR Insulin receptor
irINS-Tf Immuno-reactive insulin-transferrin
MTT 3-(4,5-dimeth- ylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
MW Molecular weight
NTA Nitrilotriacetic acid
pAKT phosphorylated AKT
PD Pharmacodynamics
PEI Polyethylenimine
PLL Poly-L-lysine
PK Pharmacokinetics
ProINS-Tf Proinsulin-transferrin
RIA Radio-immunoassay
s.c. Subcutaneous
SD Standard deviation
SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis
xiv
STZ Streptozotocin
TEER Transepithelial electrical resistance
Tf Transferrin
TFF Tangential flow filtration
TfR Transferrin receptor
TGN Trans-Golgi network
xv
ABSTRACT
Since proinsulin-transferrin (ProINS-Tf) fusion protein was designed and
expressed from the mammalian HEK293 cells (HEK-ProINS-Tf), many interesting and
attractive features have been discovered. First, it has been found that ProINS-Tf could be
converted into an activated form when incubated with some specific cell lines or tissue
cultures through TfR-mediated endocytosis and recycling pathway. Second, in the
previous in vivo studies on streptozotocin-induced type-1 diabetic mice via subcutaneous
injections, ProINS-Tf has been proven as an extra-long acting insulin analogue with
liver-specific insulin action. However, there are still some other interesting features that
have not yet been fully explored and characterized. One of the ultimate goals of this
project is to develop ProINS-Tf into an oral deliverable insulin analogue. However, to
achieve this goal, two of the major challenges, including its low production yield from
HEK293 cells and the limited transcytosis rate of Tf carrier across the intestine, need to
be overcome.
First, to improve the protein productivity, an alternative protein expression
system, transgenic rice (Oryza sativa L.), was applied. The results indicated the
transgenic rice successfully expressed the intact and active rice-derived ProINS-Tf (Rice-
ProINS-Tf). Although the insulin-like bioactivity of Rice-ProINS-Tf was slightly lower
in vitro, its potency of in vivo blood glucose control was considerably stronger than that
produced from HEK293 cells (HEK-ProINS-Tf). Overall, it was demonstrated that rice-
ProINS-Tf exhibited a great similarity compared with the HEK-ProINS-Tf. Our oral
xvi
delivery studies in type 1 diabetic mice further demonstrated the oral hypoglycemic effect
of Rice-ProINS-Tf with oral bioactivity ≈ 1 %.
On the other hand, the transcytosis rate of Tf-carried fusion proteins across
intestinal epithelium is limited by the polarized distribution of TfR on the basolateral cell
membrane. Therefore, to increase the transcytosis rate, a novel transcytosis mechanism of
Tf across intestinal epithelium was proposed. In this subproject, Tf was conjugated with
positively charged poly-L-lysine (PLL) through a pH sensitive nitrilotriacetic acid (NTA)
linker. It was demonstrated PLL could increase the apical binding and internalization of
Tf. Subsequently, with the endosomal dissociation of the NTA linker, Tf can be
exocytosed using the basolaterally recycled TfR.
In this dissertation, the two major challenges of oral delivery of ProINS-Tf were
addressed and two separate approaches have been proposed and investigated to solve
these issues. In the future, we expect the findings from these two strategies could be
combined to develop ProINS-Tf as a low-cost, long-acting, and liver-targeting basal
insulin analogue with higher oral therapeutic effect for diabetic patients. Additionally, the
findings from this dissertation could be further applied to any other Tf-based fusion
proteins to increase their productivity as well as the oral bioavailability.
1
CHAPTER 1: INTRODUCTION
1.1. Oral Delivery of Protein Therapeutics
1.1.1. Protein Therapeutics
The first human protein therapeutic, Humulin®, was created at Genentech(1),
developed by Eli Lilly, and approved by the US Food and Drug Administration (FDA) in
1982. Since then, with the development of recombinant protein technology, protein
therapeutics have become one of the most attractive and quickly growing class of
therapeutic reagents. So far, protein therapeutics have already played significant and
diversified roles in the comprehensive field of medicine. Since the approval of
Humulin®, over 100 recombinant biologics-based new molecular entities (NMEs) have
been approved in the European Union and the United States(2). According to the distinct
functions and applications, these protein therapeutics were further categorized into four
different major groups, which include: I) protein therapeutics with enzymatic or
regulatory activity, II) protein therapeutics with special targeting activity, III) protein
vaccines, and IV) protein diagnostics (3).
Proteins and peptides, as the building blocks of life with dynamic and diverse
roles in the body, are now becoming a very promising group of therapeutic entities. These
biological macromolecules have displayed several advantages over the small-molecule
drugs. The most important advantages of protein therapeutics are their high specificity
and high potency due to the structural complexity that cannot easily be mimicked by
small-molecule drugs. As a result, with the increased specificity, protein therapeutics also
2
display lower adverse effects which are usually caused by the non-specific binding to
other undesired targets. However, these structural advantages also raise some issues, such
as protein stability, folding complexity, as well as the limited selections of administration
approach(4, 5).
Currently, most of the protein therapeutics are produced using recombinant
protein technology, because of the following reasons. First, compared with proteins
extracted and purified from their native sources, recombinant proteins usually could be
expressed from more inexpensive and efficient production platforms(6). Additionally, it
could reduce the risks of contamination or infection from these native sources. Second,
recombinant protein expression technology allows the expressed protein of interest with
exact sequence as human protein; therefore, could exhibit better specificities and reduce
the chances of immunogenicity(7). Third, the recombinant technology allows partial
modification and optimization of biologics products. Therefore, those modified proteins
could exhibit improved characteristics, such as stronger target binding and higher
specificity. Finally, the recombinant DNA technology could fuse DNA sequences of
different proteins, and therefore, allow the production of fusion proteins presenting
multiple functions from different protein domains. For example, several Fc fusion protein
therapeutics have been approved by FDA(8). In these fusion proteins, antibody Fc
domain was fused with many protein drugs majorly to extend their plasma half-lives and
increase the in vivo efficacies.
Today, many different kinds of protein expression system, such as bacteria, yeast,
3
insect cells and mammalian cells, have been developed and utilized for recombinant
biologics production. To select an optimal expression system, many factors, such as
protein quality, productivity, cost of production, and protein glycosylation, need to be
taken into consideration. Among all these expression systems, the mammalian cell
expression systems, such as Chinese hamster ovary (CHO) and mouse myeloma cell,
NS0, can provide better protein folding, proper post-translational modifications, and thus,
less chance of immunological side effects(9). Therefore, although with higher costs of
production, approximately 70 % of commercial recombinant protein therapeutics are still
produced using mammalian cell expression system(10).
Although the market of protein therapeutics has been dramatically increased
because of the development of recombinant protein technology, there are still many
limitations and challenges that needed to be overcome. For example, the dependence on
invasive administrations for systemic delivery has always been one of the major
limitations of protein therapeutics. Numerous efforts have been made to develop effective
non-invasive approaches for systemic drug administration of biologics. However, it
remains to be an unsolved problem for biopharmaceutical scientists, and the invasive
injections are still the major administration routes for protein therapeutics.
1.1.2. Advanced Strategies for Oral Delivery of Protein Therapeutics
Most common routes of non-invasive systemic drug administration include oral,
sublingual, inhalation and topical delivery. Among all these routes, the oral route is
always the most preferred approach for drug administration due to its convenience which
4
leads to highest patient compliance. However, for biological therapeutics, the unfavorable
physicochemical properties for drug absorption, including large molecular size and high
of hydrophilicity, as well as the acidic and enzymatic degradation in the gastrointestinal
(GI) tract make it a very difficult challenge to develop a protein drug delivery system
with an acceptable oral bioavailability(11).
To address these issues above and deliver protein therapeutics via oral route,
different strategies have been proposed and investigated. Some of them focus on
preventing enzymatic or acidic degradation in the GI tract, such as using protease
inhibitors or formulation vesicles(12). Others aim at enhancing the drug absorption across
the intestinal epithelium. For example, chitosan and its derivatives can transiently open
tight junctions between epithelial cells and subsequently increase the paracellular
transport of macromolecules(13). Vitamin B12 is another example which has been
applied as a drug carrier to improve the oral absorption of insulin(14). In some cases,
combination of different approaches, such as permeation enhancer-coated nanoparticles,
could bring more advantages of different strategies together to achieve better oral
bioactivities(15, 16).
1.2. Human Serum Transferrin
1.2.1. Background
Human serum transferrin (Tf) is an 80 kDa endogenous serum glycoprotein which
controls the transport of iron ions into the cells through Tf receptor (TfR)-mediated
endocytosis and recycling pathway; thus, is responsible for maintaining the balance of
5
free iron ions in the body(17). The x-ray crystal structure report revealed that the Tf
molecule is mainly composed by two similar lobes, the N-lobe and C-lobe, which are
linked by a short peptide(18). Each lobe of Tf can be further divided into two domains,
between which there is a ferric binding site in the cleft. The uptake and release of ferric
ion by Tf is pH-sensitive and can cause a simultaneous conformation change of the Tf
structure(19). In human body, TfR is expressed in most cells, such as red blood cells,
intestinal cells, monocytes and the blood-brain barrier.
The schematic mechanism of TfR-mediated endocytosis and recycling pathway is
shown as Scheme 1-1(20). Briefly, Tf without iron ions (apo-Tf) exhibits strong binding
affinity with free iron ions to form a diferric-Tf (holo-Tf). Subsequently, it binds to TfR
on the cell surface at the extracellular pH of 7.4, and is internalized into the cell via TfR-
mediated endocytosis mechanism. Due to the acidic environment of endosomes (pH ≈
5.5) and the triggered conformation change of Tf, iron ions are then released from the
complex and transported out of endosome. Finally, the apo-Tf-TfR complex is recycled
back to the cell surface; and the apo-Tf is dissociated from TfR at the neutral pH into the
blood circulation. The released apo-Tf is ready for another cycle of iron ions uptake.
1.2.2. Transferrin as Drug Carrier to Improve Drug Targeting and Delivery
With the advantages of the TfR-mediated endocytosis and recycling pathway, Tf
has been demonstrated as a multi-functional protein drug carrier that can improve
pharmacokinetic properties of different therapeutics(20). First of all, Tf has been utilized
as an ideal protein drug carrier for the extension of half-life for many biologics. Short
6
half-lives of proteins/peptides are usually caused by their fast renal clearance as well as
the enzymatic degradation in the body. Therefore, after recombinantly fused or
chemically conjugated with Tf, the increased molecular size and the TfR-mediated
recycling mechanism can dramatically decrease the systemic clearance of carried
biological therapeutics in the body. For example, growth hormone-Tf fusion protein (GH-
Tf, T
1/2
= 4.97 h) exhibited 20-fold longer half-life when compared to native GH protein
(T
1/2
< 0.25 h) in CF1 mice(21). Also, the recombinant fusion of glucagon-like peptide-
1(GLP-1) to transferrin exhibited a significant improvement of the half-life of GLP-1
from 1-2 min to 44 h(22).
In addition, due to the relatively higher expression level of TfR on malignant cells
and the efficient intracellular targeting capability of Tf resulting from TfR-mediated
internalization mechanism, Tf has become a very attractive cancer cell-targeting carrier.
So far, Tf has been conjugated with different cargos, including chemotherapeutics,
polymers, nanoparticles and gene therapy vectors, to increase their intracellular drug
concentration in cancer cells and reduce the non-specific toxicity in the body(23).
In addition, several studies have reported that TfR can be transcytosed across
physiological barriers in the body including blood brain barrier as well as the intestinal
epithelium (24). Therefore, the potential use of Tf to improve the drug transport across
these endothelial or epithelial cells have been intensively investigated. So far, Tf has been
also utilized as a drug carrier to increase the oral bioactivities of many protein
therapeutics, such as growth hormone (GH) and granulocyte-colony stimulating factor
7
(G-CSF)(25, 26).
1.3. Proinsulin-Transferrin (ProINS-Tf) Recombinant Fusion Protein
1.3.1. Diabetes and Insulin
Diabetes mellitus is a group of metabolic diseases characterized by hyperglycemia
resulting from the failures in secretion or action of insulin in the body; and thus causes a
series of chronic complications. Based on the report of National Diabetes Fact Sheet,
2014(27), approximately 21 million people in the United States have been diagnosed with
diabetes. Among all these patients, about 28.7 percent of them are receiving insulin
substitution therapy to maintain their blood glucose level within the target range. The
number of diabetes patients is continuously growing year by year. Therefore, insulin and
insulin analogues have become one of the most prescribed biotherapeutics in the world
nowadays.
Insulin, a hormone secreted from beta cells of the islets of Langerhans in the
pancreas, plays a dominant role in the regulation the metabolism of carbohydrate and fat
in the body. It is a post-translational modified molecule composed by two peptides, A-
chain and B-chain, which are bound together by two disulfide bonds. Insulin was first
extracted and purified from bovine pancreas as a drug for the treatment of diabetes in
1922(28). Since recombinant human insulin was first produced from E. coli by
Genentech in 1978, the development of recombinant DNA technology and protein
systems has provided not only alternative sources for human insulin manufacturing but
also the possibility for molecular design of new insulin analogues(29).
8
Currently, most of the commercialized insulin analogues are recombinantly
produced either from Escherichia coli or Saccharomyces cerevisiae(30). The two-
peptide-chain structure of insulin makes it more difficult for industrial production. To
acquire an active insulin molecule, two major approaches are applied. One of them
involves the separate production and purification of insulin A and B chains followed by
co-incubation under an optimum condition for the formation of disulfide bonds and
proper protein folding(31). The other method is carried out by expression of single-chain
insulin precursor, proinsulin, followed by proteolytic removal of the C-peptide. As a
result, insulin, unlike other single-chain peptide/protein therapeutics, cannot be simply
expressed and purified. These additional manufacturing processes could lead to many
issues, such as decreased production yield and increased chance of contamination.
Therefore, to meet to the growing market demand from the increasing number of diabetes
patients, biopharmaceutical scientists are still trying to find easier and less expensive
methods which allow the large-scale insulin production.
1.3.2. ProINS-Tf Fusion Protein from Mammalian Cells
As mentioned above, connecting insulin with Tf could bring multiple benefits,
such as prolonged half-life and increased oral bioavailability. However, insulin is not
suitable to recombinantly fused with Tf due to its two-chain structure connected by its
inter-peptide disulfide bonds. Therefore, to develop a long acting oral deliverable insulin
analogue by recombinant fusion with Tf, the single-chain protein precursor, ProINS, was
selected as a substitution of insulin. As a result, the expression plasmid containing DNA
sequence of proinsulin-transferrin (ProINS-Tf) was previously designed and constructed
9
in our laboratory, and the ProINS-Tf fusion protein successfully expressed from human
embryonic kidney HEK293 cells (Fig. 1-1) post polyethylenimine (PEI)-mediated
transient transfection(32). The hexahistidine peptide was fused to the C-terminal end of
Tf for protein purification purpose.
1.3.3. TfR-Mediated Conversion of ProINS-Tf Fusion Protein
It has been well known that ProINS exhibited only 1-10% potency of insulin
when tested in different in vitro assays including IR binding and other insulin-related
activities(33). Therefore, the original idea of this ProINS-Tf project was to treat ProINS-
Tf fusion protein with some specific enzymes to convert the ProINS moiety into an
activated form. Surprisingly, it was observed that the untreated ProINS-Tf exhibited a
delayed but strong bioactivity in AKT phosphorylation and glucose production inhibition
in hepatoma cells when compared with insulin and ProINS. However, it was also found
that the fusion protein showed a very low activity in promoting glucose uptake in the
differentiated NIH 3T3 adipocytes(32).
From the further studies, it was demonstrated that the ProINS moiety of ProINS-
Tf fusion protein could be intracellularly converted into an active form via TfR-mediated
endocytosis pathway. However, the TfR-mediated endocytosis mechanism is not the only
key component of conversion process of ProINS-Tf. In a conversion screening assay with
different cell lines, the result indicated that HeLa cells with highest TfR expression had
relatively low conversion rate(34). Oppositely, H4IIE cells and mouse primary
hepatocytes, although with lower TfR expression level, exhibited much higher conversion
10
rate compared to others. The results implied that the conversion and activation of
ProINS-Tf depend on not only the TfR mediated internalization mechanism but also the
proteolytic process of specific enzyme(s) that are present only in certain cell lines.
1.3.4. Liver-Preferential Effect of ProINS-Tf Fusion Protein
As the first organ encountered by insulin after the secretion, the liver is the
primary site of insulin action, uptake and degradation after its secretion from the
pancreatic ß cells(35). During fasting, insulin reduces the hepatic glucose production
through inhibiting either glycogenolysis or gluconeogenesis in the liver. When under
feeding condition, insulin in the liver can stimulate the glucose storage as glycogen.
According to the previous studies, 50 - 80 % of the endogenously released insulin in the
portal vein is extracted from the liver(36). The rest of insulin passed through the liver acts
on peripheral tissues, such as adipose and muscle, to stimulate glucose uptake from the
circulation. Therefore, under normal physiological conditions, the liver plays a critical
role on maintaining the blood glucose level within a narrow range (80 - 120 mg/dL).
However, for type 1 diabetic patients, the current insulin replacement therapies via
subcutaneous administration fail to mimic the liver-preferential effects of native insulin,
and a large portion of administered insulin therapies acts on the peripheral tissues. The
long term over-exposure of peripheral tissues to insulin caused by the non-physiological
distribution of administered insulin usually leads to some adverse effects, such as weight
gain, and eventually increases the risks of cardiovascular diseases for diabetes
patients(37).
11
Based on the previous reports from our laboratory, ProINS-Tf has been
demonstrated as an insulin analogue with high liver-preferential bioactivity. First of all,
the ex vivo conversion study in precision-cut mouse liver slices demonstrated that liver is
capable of converting ProINS-Tf into an active form(34). From the in vivo studies on
type-1 diabetic mice, it has been further demonstrated that ProINS-Tf could promote
higher hepatic glycogen accumulation compared with the saline and insulin groups(38).
Furthermore, it was also noticed that ProINS-Tf fusion protein induced strong insulin
receptor phosphorylation level in the liver but not in the skeletal muscles. All these
findings above suggested that liver is most likely the key organ which is responsible for
the conversion of ProINS-Tf in the body, probably due to the presence of
proteases/convertases, as well as the high TfR expression level in the hepatocytes.
1.3.5. Oral Delivery of ProINS-Tf
As diabetic patients rely on daily multiple injections of insulin therapeutics to
maintain their blood glucose within a normal range throughout their lifetime, insulin-
based therapeutics are always one of the first choices for the protein oral delivery studies.
Numerous efforts and various strategies have been made to overcome the physiological
barriers and to develop effective and safe insulin therapeutics that can be non-invasively
delivered(39). However, like other protein therapeutics, the peptide-based structure of
insulin makes it vulnerable to be degraded in the GI route after administered orally. So
far, there are no FDA-approved oral deliverable insulin product on the market, since the
bioavailabilities in the clinical trials are always lower than expected(40). As a result,
invasive administrations, such as insulin injections or pumps, are currently still the major
12
route of insulin delivery.
Our previous in vitro and in vivo studies have demonstrated ProINS-Tf fusion
protein as a highly potential candidate for long-acting basal insulin replacement therapy
because of its proven prolonged half-life and liver-preferential action(38). However, one
of the prospective applications of ProINS-Tf, oral delivery, has not yet been investigated
and developed. Fifteen years ago, insulin has been conjugated with Tf via a disulfide
bond linker and the oral hypoglycemic effect has been demonstrated in the type-1
diabetic rat model(41). The major problem of the insulin-Tf conjugates is the
heterogeneity of the product caused by the random chemical cross-linking reaction. Now,
the problem has been solved by the development of recombinant fusion protein
technology. However, to further develop the ProINS-Tf fusion protein as an oral
deliverable therapeutic protein, there are still some existing and potential issues that need
to be overcome.
First, one of the existing critical obstacles in developing oral ProINS-Tf is the
limited expression yield of ProINS-Tf from HEK293 cells which subsequently raises two
issues: the quantity of the fusion protein and the cost of production. So far, there are three
different Tf-based fusion proteins, including GH-Tf, G-CSF-Tf and ProINS-Tf, which
have been previously designed and synthesized in our laboratory. Although these three
fusion proteins were produced using the same expression technique and process, the
expression level of ProINS-Tf is only 5 - 10 % compared to the other two fusion proteins.
The relatively low expression level of ProINS-Tf is possibly caused by the more complex
13
structure of ProINS domain (three disulfide bonds within a short peptide) that increase
the chance of protein misfolding and the subsequent elimination of misfolded proteins by
the endoplasmic reticulum (ER)-associated protein degradation pathway(42).
Another potential issue of the oral delivery of ProINS-Tf was the limited
transcytosis rate of Tf across intestinal epithelium. Although many reports have already
demonstrated that Tf can be used as a carrier of protein oral delivery due to the TfR-
mediated transcytotic mechanism, it is also well known that TfR is predominantly
distributed on the basolateral surface of intestinal epithelium(43). Also, based on the
results of transcytosis and recycling assay on Caco-2 cells by Lim et al., apically applied
Tf recycled predominantly (~90 %) back to the domain of administration, suggesting
separate Tf recycling pathways from the apical and basolateral membranes(44). Although
in some previous reports, significant difference could be observed when comparing the
oral bioactivity of native form and Tf-carried protein drug after oral administration in
vivo, high dosage of the Tf-carried protein drug was usually needed to achieve the desired
therapeutic effect. The major portion of the drug was actually digested and degraded in
the gastrointestinal tract due to its low absorptive efficiency.
In conclusion, these two issues above, including the low expression level of
ProINS-Tf from HEK293 cells as well as the limited Tf transcytosis rate across the
intestinal epithelium, were the most critical problems that needed to be resolved. To
tackle these two issues and investigate the feasibility of oral delivery of ProINS-Tf, two
subprojects were proposed and conducted in the following chapters.
14
Scheme 1-1. TfR-mediated endocytosis and recycling pathway
Free iron ion can bind to apo-Tf to form holo-Tf which subsequently bind to TfR on the
cell surface (pH 7.4). The holo-Tf-TfR complex will be internalized into the cell via TfR-
mediated endocytosis pathway. Due to the conformation change of Tf in the acidic
endosomal environment (pH 5.5), the iron ions are released from Tf and transported out
of endosome into cytosol by divalent metal transporter-1 (DMT-1). The apo-Tf-TfR
complex will be recycled back to the cell surface. The apo-Tf and TfR are eventually
separated at pH 7.4 and will be ready for next cycle of iron uptake.
15
Figure 1-1. The expression plasmid and structure of ProINS-Tf fusion protein
The pcDNA3.1(+) vector that inserted with DNA sequence of preproinsulin-Tf-6xHis via
its EcoRV, XhoI and XbaI restriction enzyme cleavage sites. After the transient
expression from HEK293 cells, ProINS-Tf-6xHis fusion protein is produced and purified
using Ni-NTA column. *Tf structure acquired from PDB 3V83 (45)
Transient Expression
from HEK293 Cells
16
CHAPTER 2: PREPARATION OF PROINSULIN-TRANSFERRIN FUSION
PROTEINS FROM HEK293 CELLS AND TRANSGENIC RICE SEEDS
2.1. Background
ProINS-Tf fusion protein is a novel long-acting insulin analogue which has been
successfully expressed through transient expression of HEK293 mammalian cells.
According to the previous reports, ProINS-Tf was a prodrug-like insulin analogue in
which the proinsulin domain could be intracellularly activated into an active form via
TfR-mediated endocytosis and recycling pathway in the liver(32, 38). Therefore, the in
vitro pretreatment with proteolytic enzymes for proinsulin activation is no longer
necessary to convert ProINS-Tf fusion protein into an active form. Moreover, the
previous in vivo studies suggested ProINS-Tf was a potential candidate for basal insulin
replacement because of its liver-preferential effect and the extra-long-acting glycemic
control effect. However, one of the major short comings of this HEK293 cells-expressed
ProINS-Tf (HEK-ProINS-Tf) is its relatively low expression yield from the HEK293
transient expression system. Therefore, the cost of protein production became a hurdle for
this project.
To overcome the hurdle, many different strategies have been proposed and
investigated, but most of them failed to increase the expression yield. For example, the
inhibitors of insulin degrading enzyme (IDE), such as bacitracin and phenanthroline(46,
47), have been previously added to the protein production CD293 medium during the
expression process. However, no increase of production level of ProINS-Tf has been
17
noticed. In addition, mild hypothermia is another common approach to increase the
expression level of recombinant protein(48). The production yields of ProINS-Tf under
different temperatures from 32 – 37 °C have also been investigated, but there was no
significant difference between the tested groups. Additionally, the attempt has been made
to replace the signal peptide of ProINS-Tf with that of highly expressed GH-Tf fusion
protein; however, there was no difference in the protein expression level after the
substitution of signal peptide.
One of the possible causes of low expression level of ProINS-Tf is the complex
structure of ProINS. The short poly-peptide of ProINS containing three disulfide bonds
make it difficult to achieve the native folding state(49). It has been demonstrated that the
formation of all three disulfide bonds in the insulin/ProINS structure, including the two
inter-chain disulfide bonds: Cys(B
19
)-Cys(A
20
), Cys(B
7
)-Cys(A
7
) as well as the intra-A
chain disulfide bond: Cys(A
6
)-Cys(A
11
), are the essential attribute for the optimal insulin
bioactivity(50). The misfolding of ProINS caused by the disulfide bond mispairing could
cause the induction of ER stress response and reduction of protein secretion(51).
Therefore, methods that could enhance the intercellular protein folding were also tested.
For example, it has been previously demonstrated that the insertion of designed 𝛼-helical
linker could significantly increase expression of Tf-based fusion proteins, because it
could increase the distance between domains and provide larger conformational space for
proper protein folding(52). To confirm this theory, the same 𝛼-helical linker was inserted
between ProINS and Tf domains, and the results suggested the insertion of two copies of
this 𝛼-helical linker could also double the production yield of ProINS-Tf. However, this
18
2-fold increase was still not high enough for the required amount of fusion protein for the
oral delivery studies.
To further increase the productivity of ProINS-Tf and reduce the production cost,
we decided to look for an alternative protein expression platform. Among all different
kinds of protein expression system, transgenic plant expression system has always been
one of very attractive options for manufacturing recombinant proteins, because of its cost
effectiveness and high scale-up capacity(53). So far, some recombinant insulin analogues
have been successfully expressed from some plant protein expression systems. For
example, it was reported that an insulin precursor was expressed from Arabidopsis
thaliana seeds, followed by treated with trypsin to generate an active product, DesB30-
insulin(54). The in vitro and in vivo biological activity of this product was demonstrated.
A fusion protein consisting of cholera toxin B subunit and furin-cleavable proinsulin has
also been expressed from both tobacco and lettuce chloroplasts(55). Based on the in vivo
data, both intraperitoneal (i.p.) injection of the purified protein and oral delivery of
unprocessed fusion protein encapsulated in plant cells displayed hypoglycemic effects on
mice. Both of these two expression systems above showed a great production yield of
target protein.
In our case, transgenic rice (Oryza sativa L.) was selected as our first alternative
expression system, because of its previous success on production of human Tf with a
high production yield by Ventria Bioscience (1% seed dry weight)(56). Most importantly,
according to the previous reports, this low-cost rice-derived human Tf has shared great
19
structural and functional similarities with native Tf(57, 58). Therefore, the first objective
of this study was to confirm if ProINS-Tf could be successfully expressed from the
transgenic rice seeds with high production yield and high product quality. In this chapter,
the methods used to prepare and purify ProINS-Tf fusion proteins from two different
hosts will be shown and discussed.
2.2. Materials and Methods
2.2.1. ProINS-Tf Fusion Protein from HEK293 cells
2.2.1.1. Cell Culture
HEK293 (ATCC, Manassas, VA) cells were used for transient expression of
ProINS-Tf. For maintaining HEK293 cells, cells were cultured in cell culture flasks with
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) at 37°C and 5% CO
2
. For cell passaging, HEK293 cells were first washed
with PBS and then treated with 0.05% trypsin-EDTA at 37 °C for 5 min for cell
detachment from the surface. The detached cells were diluted with complete growth
medium, and cell densities were counted using Beckman Coulter Cell and Particle
Counter. Appropriate numbers of cell were then transferred to new flasks with fresh
complete cell growth medium.
2.2.1.2. Transfection and Expression of Fusion Protein
HEK293 cells maintained in T-175 flasks at ~80% confluence were transiently
transfected with pcDNA 3.1(+) plasmid containing the fusion gene of preproinsulin-Tf-
20
6xHis via PEI-mediated transfection. The cell culture medium was first replaced with
serum free DMEM. The DNA plasmid and PEI (w/w ratio of DNA to PEI = 1 to 2.5)
were mixed in 150 mM NaCl. After 15 min incubation at room temperature, the
DNA/PEI mixture was added to the flasks. After 6-h incubation at 37 °C, the transfection
medium was replaced by serum-free CD 293 medium supplemented with 4 mM L-
glutamine. The conditioned CD293 medium (Thermo Fisher Scientific Inc.) was
collected on Day 4 and Day 7 post-transfection.
2.2.1.3. Protein Concentration and His-Tag Purification
The secreted ProINS-Tf in the collected CD293 medium was concentrated using a
tangential flow filtration system (TFF, Millipore) and buffer-exchanged to the prepared
binding buffer (50 mM sodium phosphate, 300 mM NaCl, 10 mM imidazole, pH 7.0).
The concentrated sample was then loaded to the column pre-packed with nickel
nitrilotriacetic acid (Ni
2+
-NTA) agarose (Thermo Fisher Scientific Inc.). The column was
washed with wash buffer (50 mM sodium phosphate, 300 mM NaCl, 20 mM imidazole,
0.05% Tween-20, pH 7.0) to remove non-specifically bound proteins, and subsequently
eluted by elution buffer (50 mM sodium phosphate, 300 mM NaCl, 250 mM imidazole,
0.05% Tween-20, pH 7.0). The eluted protein was loaded into Spectra/Por dialysis
membrane (MWCO 12-14 kDa, Spectrum Laboratory, Rancho Dominguez, CA) and
dialyzed using against mannitol buffer (50 mM sodium phosphate, 50 mg/mL mannitol
and 0.01% Tween-20 at pH 7.0) at 4 °C overnight to removed excess imidazole. The
dialyzed protein was finally stored at -80 °C for future use.
21
2.2.2. ProINS-Tf Fusion Protein from Transgenic Rice Seeds
2.2.2.1. Extraction and Purification of ProINS-Tf from Rice Seed Flour
Through the collaboration between Dr. Wei-Chiang Shen’s group at University of
Southern California and Ventria Bioscience, the transgenic rice of ProINS-Tf was
previously developed using Ventria Bioscience's ExpressTec recombinant expression
system(59). The process to extract and purify Rice-ProINS-Tf from the transgenic rice
flour was briefly described in Scheme. 2-1. First, to extract the target protein from the
rice seeds, 200 g of transgenic rice flour was mixed with 2 L of 25 mM Tris-HCl buffer
pH 7.5 at room temperature for 30 min, followed by centrifuged at 10,000 g / 4°C for 2 h
to remove the insoluble starch. The supernatant was concentrated using TFF followed by
a centrifugal filter (MWCO 30 kDa, Millipore Amicon) at 5000 g until the volume
reached ~2 mL. The concentrated protein extract was first purified with Sephacryl S-200
HR size exclusion column (Bed height: 100 cm, GE Healthcare Life Sciences) to remove
the large and small size impurities. The fractions that contained ProINS-Tf fusion protein
were collected and loaded to the prepared Diethylaminoethyl (DEAE) Sepharose Fast
Flow chromatography (GE Healthcare Life Sciences). The column was first washed with
25 mM Tris-HCl buffer, pH 7.5 and the bound protein was eluted by serial elution with
25 mM Tris-HCl buffers containing 20 - 150 mM NaCl. The fractions containing purified
ProINS-Tf were concentrated and buffer-exchanged to mannitol buffer (50 mM sodium
phosphate, 50 mg/mL mannitol and 0.01% Tween-20 at pH 7.0) using centrifugal filter
(MWCO 30 kDa, Millipore Amicon). The final concentrated product was stored at - 80
°C for future experiments.
22
2.3. Results
2.3.1. ProINS-Tf Fusion Protein from HEK293 cells
2.3.1.1. Increased Purity of HEK-ProINS-Tf
HEK-ProINS-Tf fusion protein was purified from the TFF-concentrated medium
using His-Tag purification method (Fig. 2-1). Before purification, significant amounts of
protein impurities were present in TFF- concentrated samples, and the His-tagged HEK-
ProINS-Tf band could hardly be identified on Coomassie Blue stained gels. When
comparing the loaded protein and the flow-through, it showed that the targeted fusion
protein (arrow) disappeared from loaded protein sample and bound to Ni-NTA agarose
column. After the washing step with wash buffer to remove some non-specifically bound
protein impurities, the fusion protein was eluted from the column by elution buffer.
Based on the Coomassie Blue stained gel and purity calculation from BCA assay, the
protein purity was significantly improved from low than 5 % before to around 80 % after
the His-Tag purification process.
2.3.1.2. Total Production Yield of HEK293-derived Fusion Protein
The final concentration of ProINS-Tf was determined based on the result from
anti-human Tf Western blot (Fig. 2-2). The concentration of purified HEK-ProINS-Tf
was calculated by Image Lab software (Bio-Rad) using human Tf as standard curve. The
concentration of HEK-ProINS-Tf was 2953 nM and the final volume was about 0.8 mL.
Therefore, the final production yield of HEK-ProINS-Tf from 2 L conditioned medium
(16 T-175 flasks) was about 210 µg.
23
2.3.2. ProINS-Tf Fusion Protein from Transgenic Rice Seeds
2.3.2.2. S-200HR Size Exclusion Column for Protein Purification
The protein extracted from transgenic rice flour in the Tris-HCl buffer was
concentrated using TFF concentration system followed by a centrifugal filter until the
volume reached 2.5 mL. The concentrated protein sample were then loaded to a packed
Sephacryl S-200 HR column (100 mL, 60 cm height, equilibrated with Tris-HCl buffer,
pH 7.5), and proteins were separated using Tris-HCl buffer, pH 7.5 as the mobile phase
(2 mL/fraction, 50 fractions). The protein concentration in each fraction was determined
by BCA assay (Fig. 2-3a), and the purification results were further analyzed using 10 %
SDS-PAGE followed by Coomassie Blue Staining and anti-(Pro)INS Western blot (Fig.
2-3b). The results indicated that S-200HR could efficiently separate the large protein
aggregates and small peptides from the fusion protein product. However, a large portion
of Tf-like impurity (only detectable by anti-Tf antibody) was still remained with Rice-
ProINS-Tf. The 25
th
-28
th
fractions that contain highest amount of ProINS-Tf, determined
by anti-(Pro)INS Western blot were collected and pooled together for further purification
using DEAE ion exchange chromatography.
2.3.2.3. DEAE Ion Exchange Column for Protein Purification
The samples purified from S-200HR size exclusion chromatography were loaded
into the prepacked DEAE column (5 mL slurry in Tris-HCl buffer, pH 7.5). The column
was first washed twice with 10 mL of Tris-HCl buffer with 50 mM NaCl to remove most
of Tf impurity. The column was then eluted by stepwise increase of ion strength (75 mM
- 150 mM NaCl). The column was finally cleaned up using Tris-HCl buffer containing
24
500 mM NaCl. The result of Coomassie Blue stained gel (Fig. 2-4) indicated most of Tf
impurities could be removed by the wash buffer, although some Rice-ProINS-Tf was also
eluted in the 2
nd
wash. Most of Rice-ProINS-Tf could be eluted by the Tris-HCl buffer
containing 75 mM NaCl. The fractions that contain highest amount of Rice-ProINS-Tf,
determined by anti-(Pro)INS Western blot, were mixed together. The sample was then
concentrated and buffer-exchanged to mannitol buffer (50 mM sodium phosphate, 50
mg/mL mannitol and 0.01% Tween-20 at pH 7.0) using centrifugal concentration filter
(MWCO 30 kDa, Millipore Amicon).
2.3.2.4. Protein Yield of Rice-derived Fusion Protein
The concentration of final product of Rice-ProINS-Tf fusion protein was
determined by using anti-insulin Western blot as mentioned previously, and the HEK-
ProINS-Tf was used as the standard protein to create a standard curve (Fig 2.5). The
ProINS-Tf concentration of the final product was estimated to be 15.75 µM with a
volume of 2.2 mL. From the estimation, the final amount of Rice-ProINS-Tf extracted
and purified from 200 g of transgenic rice flour was around 2.7 mg in this batch. The
purity of protein was around 40 % based on the concentration of Rice-ProINS-Tf (1.24
mg/mL) estimated from anti-INS Western blot and total protein concentration (3.21
mg/mL) from BCA assay. Therefore, the final yield of Rice-ProINS-Tf from 200 g rice
flour was 2.73 mg. The purity could be further increased if necessary; however, the final
yield of protein would be compromised, depending on how high of the purity to be
achieved.
25
2.4. Discussion
In this chapter, the preparation and purification processes of both HEK-ProINS-Tf
and Rice-ProINS-Tf have been described in detail. Initially, both of them had their own
purity issues needed to be solved. For HEK-ProINS-Tf, the purity was able to be
significantly improved from 1 % to 80 % by applying His-Tag purification technique. For
Rice-ProINS-Tf, the purification process was more challenging. First, the fusion protein
needs to be purified from the rice flour-extracted protein without using His-tag
purification. The other critical difficulty of the purification of Rice-ProINS-Tf was the
extract contained a large portion of Tf-like molecule that is probably the digested product
during the intracellular synthesis process and exhibited very similar physicochemical
properties as Rice-ProINS-Tf. This Tf-like impurity must be removed from the fusion
protein because it would interfere the activation process of ProINS-Tf by blocking the
TfR binding and the receptor mediated recycling pathway.
To test whether the Tf-like impurity was a proteolytic product during the
extraction and purification process, protease inhibitor cocktail (Sigma, P9599) has been
added in the extraction buffer; however, there was no difference between the ratio of the
amount of Rice-ProINS-Tf to Tf in the protein extract. This result suggested the Tf
impurity has already existed in the endosperm cells, probably as a byproduct from the
biosynthesis process. Therefore, to remove the Tf-like impurities and achieve the optimal
quality of final product, different purification techniques had been tested. Based on our
findings, the Rice-ProINS-Tf could be efficiently purified using size exclusion
chromatography followed by cation exchange chromatography. The S-200HR size-
26
exclusion column was applied to remove most of the large- and small-molecule
impurities that have very strong binding affinity to the following ion exchange column.
This process could significantly minimize the cost from DEAE resin. Later, the DEAE
column was able to separate Tf impurity from ProINS-Tf product with little unavoidable
lost in the final yield.
27
Scheme 2-1. Extraction and purification process of Rice-ProINS-Tf
Transgenic rice flour of Rice-ProINS-Tf was mixed with Tris-HCl buffer at pH 7.5 at 25
°C for 30 min. After the centrifugation process to remove the insoluble starch, the
supernatant was concentrated using TFF and centrifugal concentration tube. The final
product of Rice-ProINS-Tf was finally purified from the protein concentrate using a S-
200 HR column followed by DEAD ion exchange chromatography.
28
Figure 2-1. His-Tag purification of HEK-ProINS-Tf
SDS-PAGE followed by Coomassie Blue staining was used to analyze the efficacy of
His-Tag purification process. Lane 1: human Tf, 2: TFF-concentrated sample, 3: flow-
through, 4: washing buffer, 5: 1
st
elution, 6: 2
nd
elution, 7: 3
rd
elution, 8: 4
th
elution, 9: 5
th
elution, 10, marker.
29
Figure 2-2. Concentration estimation of HEK-ProINS-Tf using anti-Tf Western blot
SDS-PAGE followed by anti-human Tf Western blot was used to estimate the final
concentration of purified product. Lane 1: marker, Lane 2 - Lane 5: 4 µL of human Tf
standards (125 nM, 250 nM, 500 nM, 1000 nM, respectively); elution, Lane 6 – Lane 8:
10-fold diluted purified HEK-ProINS-Tf (2 µL, 4 µL, 8 µL, respectively)
30
Figure 2-3. S-200HR size exclusion chromatography of Rice-ProINS-Tf
(a) After the S-200HR size exclusion chromatography purification, the concentration of
each collected fraction was determined using BCA assay. (b) 10% SDS-PAGE followed
by Coomassie Blue Staining and Anti-(Pro)INS WB. The asterisked fractions were
collected and pooled together for further DEAE purification step.
(a)
(b)
31
Figure 2-4. DEAE ion exchange chromatography for Rice-ProINS-Tf purification
The samples from DEAE chromatography were loaded to 10% SDS-PAGE followed by
Coomassie Blue staining. The arrow indicated the target Rice-ProINS-Tf fusion protein.
The asterisked lanes indicated the fractions that were collected for the future use.
32
Figure 2-5. Anti-(Pro)INS Western blot for estimating the conc. of Rice-ProINS-Tf
Lane 1-4: Rice-ProINS-Tf (50-fold diluted; 2, 4, 8, 16 µL, respectively); Lane 5: MW
Marker; Lane 6-9: HEK-ProINS-Tf (295.3 nM; 2, 4, 8, 16 µL, respectively). The final
concentration of Rice-ProINS-Tf estimated by Image Lab software was 15.75 µM.
1 2 3 4 5 6 7 8 9
33
CHAPTER 3: CHARACTERIZATION AND COMPARISON STUDIES OF RICE-
AND HEK293-DERIVED PROINSULIN-TRANSFERRIN FUSION PROTEINS
3.1. Background
In Chapter 2, the production and purification processes of HEK-ProINS-Tf and
Rice-ProINS-Tf have been both described. In this chapter, the physicochemical properties
and in vitro bioactivities of Rice-ProINS-Tf were characterized and compared with those
of HEK-ProINS-Tf.
Human Tf has been previously expressed from transgenic rice expression system
(Oryza sativa L.)(56) as a safe and economic alternative of human or animal plasma-
derived transferrin. The physicochemical, structural and functional properties of rice-
derived Tf have been extensively characterized and compared with either native human
Tf or mammalian cell-expressed human Tf. Based on the results of previous comparison
studies, there are only some minor differences between rice-derived Tf and
native/mammalian-expressed Tf(57, 58). Among all the observed differences, the most
obvious and important one is the status of N-glycosylation. However, it was shown that
the absence of N-glycosylation in rice-derived recombinant human Tf did not cause a
dramatic structural or functional change.
Based on the observation of rice-derived human Tf, it was expected that Rice-
ProINS-Tf should also exhibit similar intrinsic properties as those of HEK-ProINS-Tf,
except the status of N-glycosylation. The results from the comparison studies in this
34
chapter would be used to evaluate the overall product quality of Rice-ProINS-Tf and to
see if the transgenic rice expression system could be utilized as an alternative source of
recombinant ProINS-Tf fusion protein.
3.2. Materials and Methods
3.2.1. Cell Culture
The rat hepatoma H4IIE cell line used in the in vitro experiments was purchased
from ATCC. The H4IIE cells were cultured in DMEM supplemented with 10% FBS, 2
mM L-glutamine, and 50 U/mL penicillin - 50 µg/mL streptomycin. The HeLa cells were
cultured in RPMI supplemented with 10% FBS, 2 mM L-glutamine, and 50 U/mL
penicillin - 50 µg/mL streptomycin. During the cell passage process, the cells were
treated with 0.5% trypsin-EDTA at 37 °C for 5 min to detach the cells from the flask for
both of the cell lines.
3.2.2. Molecular Characterization of Rice-Derived ProINS-Tf
For Coomassie Brilliant Blue stained SDS-PAGE, the prepared proteins including
native human Tf (Sigma), rice-derived Tf (Invitria, Junction City, KS), purified HEK-
ProINS-Tf or purified Rice-ProINS-Tf was each mixed with non-reducing 6x sample
buffer, respectively, and boiled for 10 min. The samples were first resolved by 10 %
SDS-PAGE, and then stained with Coomassie Brilliant Blue staining buffer (0.1%
Coomassie Brilliant Blue R250 dye, 10% acetic acid, 40% methanol) for 45 min. The
stained gel was then washed with destaining buffer (7 % acetic acid, 40% methanol) for
at least 1 h until most of the background color had been removed. The gel was then
35
washed with ddH
2
O overnight, and the molecular weight of each sample was estimated
using Image Lab™ software (Bio-Rad).
For anti-human Tf Western blot, after the SDS-PAGE, the protein sample bands
were transferred from the gel to the 0.45 µm polyvinylidene fluoride membrane. The
membrane was blocked with blocking buffer (5% non-fat milk in Tris-buffered saline
with 0.05% Tween-20; 0.05% TBS-T) at room temperature for 1 h and incubated with
1:10,000 diluted anti-human Tf antibody (T2027, Sigma, St. Louis, MO) in the same
buffer. After overnight incubation at 4°C, the membrane was washed with TBS-T buffer
trice and subsequently incubated with horseradish peroxidase (HRP)-conjugated mouse
anti-goat secondary antibody (Bio-Rad, Hercules, CA) with the dilution of 1:10,000 at
room temperature for 1 h. After final washing with 0.05% TBS-T buffer, the protein
bands on the membrane were detected using enhanced chemiluminescence (ECL, GE
Healthcare) and analyzed using Quantity One 1-D Analysis software (Bio-Rad).
To detect of the ProINS moiety of Rice-ProINS-Tf, the membrane was first
blocked with 5 % bovine serum albumin (BSA) in TBST buffer, and then incubated with
anti-(Pro)INS antibody (ab8304, Abcam, Cambridge, MA) with the dilution of 1:3000 at
4°C overnight. After the same washing step, the membrane was then blotted with anti-
mouse HRP-conjugated secondary antibody (Bio-Rad) with the dilution of 1:5,000 at
room temperature for 1 h. The protein bands were detected using ECL-plus (GE
Healthcare) and analyzed using Quantity One 1-D Analysis software (Bio-Rad).
36
3.2.3. Transferrin Receptor Binding Assay
The purified protein samples, including HEK293-ProINS-Tf and Rice-ProINS-Tf,
were labeled with radioactive
125
I using Chloramine T method as previously
described(60). In brief, 1 mg of each protein was first mixed and reacted with Na
125
I
(PerkinElmer, Waltham, MA) and oxidizing agent Chloramine-T (Sigma) on ice for 5
min. Subsequently, sodium metabisulfite was added and gently mixed on ice for another
5 min. Excess potassium iodide was added to stop the radiolabeling reaction. Finally, the
125
I-labeled protein was purified by G-50 size exclusion chromatography using PBS as
mobile phase, and the radioactivity of
125
I-labeled protein in each fraction was counted
using a gamma counter (Packard, Downers Grove, IL). The fractions of first peak
containing
125
I-labeled protein sample was collected and stored at 4 °C for the future
experiments.
For this competitive TfR binding assay, HeLa cells were selected due to its high
TfR expression level which could lead more accurate measurement. HeLa cells were
seeded into 12-well plate and cultured until confluence. Before the TfR binding assay, the
cells were washed three times with PBS followed by incubation in RPMI with 0.1% BSA
at 37 °C for 30 min for stripping serum Tf. The cells were then treated with the dosing
solutions that contained 10 nM
125
I-labeled Tf and unlabeled Tf, HEK-ProINS-Tf or
Rice-ProINS-Tf at 0.1, 1, 10, 100 and 1000 nM in RPMI with 0.1 % BSA. After the
incuvation at 4 °C for 1 h, the cells were wash three times with ice-cold PBS, and
subsequently dissolved in 0.1 N NaOH. The radioactivity of cell lysate was measured
using a gamma counter (Packard, Downers Grove, IL), and then normalized with the total
37
cell protein amount that was measured using BCA assay.
3.2.4. Intracellular Conversion of ProINS-Tf in H4IIE Cells
A conversion study was performed to confirm the in vitro activation of Rice-
ProINS-Tf and to compare its conversion rates with that of HEK-ProINS-Tf. In this
study, H4IIE rat hepatoma cells were treated with dosing solutions containing ProINS-Tf
fusion protein from transgenic rice or HEK293 cells, in the presence or absence of 1000-
fold excess of apo-Tf (Sigma) at 37 °C. At the indicated time points, cell culture media
were collected and subjected to insulin-specific radioimmunoassay (RIA, Millipore) that
exhibited less than 0.2% cross-reactivity with human proinsulin. The insulin
concentrations of media were determined by insulin-specific RIA according to the
manufacturer’s instructions.
3.2.5. Phosphorylation of AKT induced by ProINS-Tf Fusion Proteins
To characterize the in vitro insulin-like bioactivity, AKT phosphorylation, was
applied as a measurement of insulin signaling pathway activation. H4IIE cells were
treated with DMEM with 0.1 % BSA (Sigma) for 18 h for serum starvation. The serum-
deprived cells were then treated with human insulin (Sigma), Rice-ProINS-Tf, HEK-
ProINS-Tf, or medium only in DMEM with 0.1% BSA. At the specific time points, cells
were lysed with cell extraction buffer (Invitrogen) supplemented with protease inhibitor
cocktail (Sigma) and phenylmethanesulfonyl fluoride (Sigma). The cell lysates were
subjected to Western blot analysis against anti-phospho-AKT (Ser 473) antibody (#4060,
Cell Signaling Technology, Danvers, MA) and anti-GAPDH antibody (#5174, Cell
38
Signaling Technology, Danvers, MA). The Western blots were developed using
Amersham ECL Plus kits (GE Health Care, Piscataway, NJ), and quantified using
Quantity One 1-D Analysis software (Bio-Rad, Hercules, CA).
3.2.6. Glucose Production Inhibition Assay
H4IIE cells in 24-well plates were treated with serial dilutions of human insulin,
Rice-ProINS-Tf, HEK-ProINS-Tf or rice-derived Tf (Invitria, Junction City, KS) in
serum-free DMEM with 0.1 % BSA at 37 °C for 24 h. After the treatment, the dosing
solution was then replaced with glucose production medium consisting of serum-,
glucose- and phenol red-free DMEM supplemented with 2 mM sodium pyruvate and 40
mM sodium DL-lactate. After the incubation at 37 °C for 3 h, the glucose production
medium was then harvested, and glucose concentration of each sample was measured by
using the Amplex Red Glucose/Glucose Oxidase Kit (Invitrogen). The cells were then
lysed with 0.1 N NaOH, and total cellular protein amount was determined using protein
bicinchoninic acid assay for data normalization.
3.3. Results
3.3.1. Molecular Characterization of Rice-Derived ProINS-Tf
Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE)
followed by Coomassie Blue staining as well as Western blotting were applied to
determine whether ProINS-Tf were successfully expressed from transgenic rice seeds.
The result of Coomassie Blue staining (Fig. 3-1a) showed that the estimated molecular
weight of human Tf (79.3 kDa) was close to the expected value (around 79 kDa). The
39
molecular weight of Rice-ProINS-Tf (79.3 kDa) are higher than the one of rice-derived
Tf (68.4 kDa) but lower than that of HEK-ProINS-Tf (88.1 kDa). The lower molecular
weight suggested the absence of glycosylation of Rice-ProINS-Tf, as previously shown in
the rice-derived human Tf(56). The results of Western blotting (Fig. 3-1b&c) showed that
Rice-ProINS-Tf molecule can be recognized by both anti-Tf and anti-(Pro)INS
antibodies. Both results have confirmed the structural features of Rice-ProINS-Tf and
demonstrated that ProINS-Tf fusion protein could be successfully expressed from Oryza
sativa L. transgenic rice expression system.
3.3.2. Transferrin Receptor Binding Assay
The results of TfR receptor binding assay (Fig. 3-2) showed that Tf (Kd = 12.44 ±
1.29 nM) exhibited 3-fold stronger binding affinity to TfR than HEK-ProINS-Tf (Kd =
34.13 ± 2.06 nM) and Rice-ProINS-Tf (Kd = 34.90 ± 3.56 nM). The TfR binding curves
of HEK-ProINS-Tf and Rice-ProINS-Tf are almost overlapped. The data suggested the
lack of glycosylation of Rice-ProINS-Tf did not influence the binding to TfR. This result
was consistent with the previous findings of transgenic rice-derived Tf.
3.3.3. Intracellular Conversion of ProINS-Tf in H4IIE Cells
In the conversion study of ProINS-Tf (Fig. 3-3) in H4IIE cells, the increased
concentration of RIA-detected insulin concentration demonstrated that Rice-ProINS-Tf
exhibited a similar conversion characteristic as HEK-ProINS-Tf. Also, the 12 h
conversion of Rice-ProINS-Tf was slightly higher than that of HEK-ProINS-Tf with the
incubation in H4IIE cells, but not statistically significant. In the presence of excess Tf,
40
the conversions for both Rice- and HEK- ProINS-Tf were significantly blocked. This
result suggested that both conversions were controlled by TfR-mediated endocytosis and
recycling pathway.
3.3.4. Phosphorylation of AKT induced by ProINS-Tf Fusion Proteins
To characterize the in vitro insulin-like bioactivity of different proteins,
phosphorylation of AKT stimulated by insulin, Rice-ProINS-Tf or HEK-ProINS-Tf was
investigated and compared. The result of AKT phosphorylation study (Fig 3-4) showed
that the H4IIE cells treated with 1 nM Insulin exhibited a rapid onset and strong
phosphorylated AKT (pAKT) signal at 5 min, 1 h and 4 h time points. The pAKT signal
of insulin treatment group disappeared at 24 h time point. However, for both Rice-
ProINS-Tf and HEK-ProINS-Tf groups, the pAKT signal started showing up at 1 h time
points, and sustained until 24 h of incubation. This lag time of AKT phosphorylation
suggested Rice-ProINS-Tf, as previously reported in HEK-ProINS-Tf, had low intrinsic
activity initially, and required conversion step to become an active form and display its
insulin-like activities. The result also confirmed the long acting effect of Rice-ProINS-Tf,
although it seemed slightly lower than that of HEK-ProINS-Tf at each time point.
3.3.5. Glucose Production Inhibition Assay
The inhibition effect of glucose production by 24-h incubation of Rice-ProINS-Tf
was investigated and compared with that of Insulin, HEK-ProINS-Tf and rice-derived Tf
(Fig. 3-5 and Table 3-2). Following the 24-h incubation, rice-derived Tf did not exert any
inhibitory effect on glucose production (IC
50
> 100 nM). The inhibitory effect of Rice-
41
ProINS-Tf (IC
50
= 379.8 ± 60.2 pM) was stronger compared to that of insulin (IC
50
=
2088 ± 888 pM) but was slightly weaker than that of HEK-ProINS-Tf (IC
50
= 101.2 ±
13.3 pM). This result, consistent with the result from AKT phosphorylation study,
suggested that Rice-ProINS-Tf showed a weaker in vitro insulin-like bioactivity in H4IIE
cells, when compared with HEK-ProINS-Tf.
3.4. Discussion
Based on the results in this chapter, Rice-ProINS-Tf exhibited a similar binding
affinity to TfR when compared with HEK-ProINS-Tf. In addition, Rice-ProINS-Tf, like
the previously reported HEK-ProINS-Tf, could be converted into an active form by TfR-
mediated endocytosis and recycling pathway when incubated with H4IIE cells. The
conversion rate of Rice-ProINS-Tf was similar to that of HEK-ProINS-Tf during a 12 h-
incubation at 37 °C. In the AKT phosphorylation assay, phospho-AKT could be only
detected at 1 h and 4 h, but not at the 5 min incubation time point in Rice-ProINS-Tf-
treated H4IIE cells. This lag time of AKT phosphorylation suggested Rice-ProINS-Tf,
like HEK-ProINS-Tf, exhibited very low intrinsic activity initially and required a
conversion step into an active form to display its effects. Rice-ProINS-Tf also showed a
slightly lower activity in the inhibition of glucose production when compared to HEK-
ProINS-Tf. The differences of insulin-like bioactivity between two fusion proteins could
be explained by the lack of glycosylation on Tf domain in Rice-ProINS-Tf. It further
influences the inter-domain interaction between ProINS and Tf that might block the
receptor binding site of activated ProINS.
42
The data from the in vitro comparison studies in this chapter suggested that Rice-
ProINS-Tf exhibited a similar TfR-binding affinity, a close conversion rate, but relatively
lower in vitro insulin bioactivity, when compared with HEK-ProINS-Tf. Although these
results were not exactly what we expected, it would be interesting to carry out further
studies to investigate the differences on the in vivo bioactivities.
43
Figure 3-1. Molecular characterization of Rice-ProINS-Tf
(a) 10 % SDS-PAGE followed by Coomassie Blue staining. The indicated molecular
weights were estimated by Image Lab Software. (b) Western blot probed with anti-
ProINS antibody. 0.1 µg, 0.2 µg, 0.5 µg and 1 µg of Rice-ProINS-Tf were loaded,
respectively. (c) Western blot probed with anti-Tf antibody. 0.1 µg, 0.2 µg, 0.5 µg and 1
µg of Rice-ProINS-Tf were loaded, respectively.
1. Marker
2. Rice Tf (68.4 kDa)
3. Rice-ProINS-Tf (79.3 kDa)
4. Human Serum Tf (79.3 kDa)
5. HEK-ProINS-Tf (88.1 kDa)
1 2 3 4 5
(a)
(b) (c)
44
Figure. 3-2 and Table 3-1. Competitive TfR binding assay in HeLa cells
HeLa cell on 12-well plates were treated with the dosing solutions containing 10 nM
125
I-labeled Tf with unlabeled Tf, HEK-ProINS-Tf or Rice-ProINS-Tf (0.1, 1, 10, 100,
1000 nM) in RPMI with 0.1 % BSA at 4 °C for 1 h. After the treatment, the cells were
lysed with 0.1 N NaOH. The radioactivity of cell lysate was measured using a gamma
counter, and then normalized with the total cell protein amount. The indicated IC
50
values
(mean ± SD) were calculated using GraphPad Prism (GraphPad Software Inc., San
Diego, CA).
Protein IC
50
(nM)
Human Tf 12.44 ± 1.29
Rice-ProINS-Tf 34.13 ± 2.06
HEK-ProINS-Tf 34.90 ± 3.56
45
Figure 3-3. TfR-mediated conversion of Rice-ProINS-Tf in H4IIE cells
H4IIE cells were treated with 10 nM of Rice-ProINS-Tf or HEK-ProINS-Tf in the
presence or absence of 1000-fold excess of apo-Tf at 37 °C. At the specific time-points,
media were collected and insulin concentrations were measured using INS-specific RIA.
The concentration of irINS-Tf was estimated using a human insulin standard curve from
12 pM to 1.2 nM according to the manufacturer’s instructions. The data represents the
increased insulin concentration from the initial time point (n=3).
46
Figure 3-4. Stimulation of AKT phosphorylation in H4IIE cells
Serum-deprived H4IIE cells were treated with 1 nM of INS, Rice-ProINS-Tf, HEK-
ProINS-Tf or DMEM w/ 0.1% BSA at 37 °C for 5 min, 1 h, 4 h, and 24 h. After lysed
with protein extraction buffer in the presence of protease and phosphatase inhibitors, 10
µg of total protein were loaded to 10% SDS-PAGE and probed with anti-phospho-AKT
antibody. The data were expressed as the intensity (mean ± SD, n=3) determined using
Quantity One 1-D Analysis software (Bio-Rad, Hercules, CA) with immunoblot against
anti-GAPDH antibody as a loading control.
47
Figure 3-5. and Table 3-2. Glucose production inhibition in H4IIE cells
H4IIE cells were treated with various concentrations of insulin, Rice-ProINS-Tf, HEK-
ProINS-Tf or rice-derived Tf at 37 °C for 24 h, and then incubated with glucose
production medium for additional 3 h. Glucose concentration in each sample was
measured using the Amplex Red Glucose/Glucose Oxidase Kit. The result was expressed
as percentage of total glucose output in the absence of any protein (n=3). All data were
normalized with total cell protein amount. The indicated IC
50
values (mean ± SD) were
calculated using GraphPad Prism (GraphPad Software Inc., San Diego, CA).
Protein IC
50
(pM)
Insulin 2088 ± 888
Rice-ProINS-Tf 379.8 ± 60.2
HEK-ProINS-Tf 101.2 ± 13.3
Rice Tf > 100 nM
48
CHAPTER 4: IN VIVO STUDIES AND ORAL ADMINISTRATION OF RICE-
DERIVED PROINSULIN-TRANSFERRIN FUSION PROTEIN
4.1. Background
In the previous two chapters, it was demonstrated that ProINS-Tf was
successfully extracted and purified from the transgenic rice expression system. Its
functional bioactivities, including TfR binding, TfR-mediated conversion in H4IIE cells
and its insulin-like activities were all confirmed and were compared with those of HEK-
ProINS-Tf. Although Rice-ProINS-Tf displayed a similar TfR binding on HeLa cells and
TfR-mediated conversion on H4IIE rat hepatoma cell, the insulin-like activities of Rice-
ProINS-Tf seems slightly lower than that of HEK-ProINS-Tf. In this chapter, the data
from the in vivo comparison studies as well as the oral hypoglycemic effect of Rice-
ProINS-Tf will be presented and discussed.
4.2. Materials and Methods
4.2.1. Streptozotocin-Induced Type 1 Diabetic Mouse Model
The animal experimental protocol was approved by the Institutional Animal Care
and Use Committee (IACUC) at the University of Southern California. All animal studies
were performed according to guidelines from National Institutes of Health. Male
C57BL/6J mice (6-week old) purchased from The Jackson Laboratory (Bar Harbor, ME)
were used for pharmacodynamics studies. The mice were housed on a 12-h day and 12-h
night cycle with room temperature maintained at 22 ± 3°C and relative humidity at 50 ±
20%, and fed with a standard laboratory rodent diet (Labdiet, Richmond, IN).
49
Type 1 diabetes of C57BL/6J mouse model was induced using high-dose
streptozotocin (STZ; sigma, 150 mg/kg) as previously described(38). After 6-h fasting,
the mice were intraperitoneally injected with freshly prepared STZ solution in 100 mM
sodium citrate buffer, pH 4.5. After waiting for 3-5 days for diabetes induction, mice with
blood glucose (BG) levels higher than 300 mg/dL were used for experiments. Blood
samples were collected from the tail and blood glucose levels were measured using
OneTouch Ultra glucose meter (LifeScan, Milpitas, CA; detection range: 20-600 mg/dL).
4.2.2. Hypoglycemic Effect of Rice-ProINS-Tf (s.c.) on Type 1 Diabetic Mice
For the in vivo studies for free-fed diabetic mice (Fig. 4-1), the experiments were
carried out on three days after STZ induction to avoid blood glucose levels beyond the
detection range. The mice were subcutaneously injected with prepared protein samples of
insulin, Rice-ProINS-Tf or HEK-ProINS-Tf (22.5 nmol/kg) or PBS. Mice were kept for
12 h post injection and had free access of water and food during the recording period.
For the studies under fasting condition (Fig. 4-2), the experiments were performed
on Day 5 after STZ induction. The mice were fasted from 2 h prior to injection, and then
subcutaneously injected with prepared protein samples of insulin, Rice-ProINS-Tf or
HEK-ProINS-Tf (22.5 nmol/kg) or PBS. After the injections, the mice were remained
fasted until 12 h post-administration with ad libitum access to water under fasting
condition. The blood glucose levels were measured from tail vein at the indicated time
points.
50
4.2.3. Extra-Long Term Hypoglycemic Effect of Rice-Derived ProINS-Tf
To test the in vivo long-term hypoglycemic efficacy of Rice-ProINS-Tf, the
experiment was also conducted in STZ-induced diabetic C57BL/6J mice. The mice were
fasted from 2 h prior to injection. After the injection of the Rice-ProINS-Tf (135
nmol/kg), insulin Glargine (135 nmol/kg, Lantus
®
, Sanofi) or saline, the mice were kept
under 8-h fasting/feeding cycle until 56-h post injection(61). The blood glucose
concentration was measured using ONE-TOUCH blood glucose meter (Lifescan, CA).
4.2.4. Oral Hypoglycemic Effect of Rice-Derived ProINS-Tf
For studying the hypoglycemic effect of Rice-ProINS-Tf via oral administration,
the experiments were conducted five days after the STZ induction. The diabetic mice
were fasted for 3 h for gastric emptying, and then the prepared proteins or buffer were
administrated using oral gavage needles or subcutaneous injection as described in Fig. 4-
4. Mice were kept fasted with free access of water until 12 h post-administration time
point. The blood glucose levels were measured using OneTouch Ultra glucose meter
from tail veins at the indicated time points.
To investigate the liver-targeting effect and further estimate oral bioactivity of
Rice-ProINS-Tf, a similar oral delivery study was conducted with some minor changes.
In this experiment, the mice were fasted for 3 h for gastric emptying, and then the
prepared proteins or buffer were administrated orally or subcutaneously as described in
Fig. 4-5. Two concentrations (11.25 or 22.5 nmol/kg) of Rice-ProINS-Tf were
subcutaneously injected for estimating the oral bioactivity. After the drug
51
administrations, the mice were kept under fasting condition for 8 h followed by free
feeding condition for additional 8h. The blood glucose levels were measured at the
indicated time points.
4.3. Results
4.3.1. Hypoglycemic Effect of ProINS-Tf on Type 1 Diabetic Mice
In order to test the in vivo biological activity of Rice-ProINS-Tf, the
hypoglycemic effects in type-1 diabetic mice via subcutaneous injections the were
examined. For the study on free-fed mice (Fig. 4-1), insulin exhibited a quick onset but
short duration of hypoglycemic effect. The blood glucose levels rapidly decreased within
the first hour and then quickly returned back to the initial blood glucose level after the 2 h
time point. For Rice-ProINS-Tf and HEK-ProINS-Tf groups, both of them showed a
slower onset but sustained effect to at least 12 h post-injection. However, the maintained
glucose levels from 4 h to 12 h of Rice-ProINS-Tf and HEK-ProINS-Tf were around 250
mg/dL (56 % of initial level) and 350 mg/dL (78 % of initial level), respectively. For the
study under fasting condition (Fig 4.2), the insulin groups also showed a quick onset of
blood glucose reduction, and the blood glucose levels start to elevated to around 200
mg/dL after 2 h post-injection. Both ProINS-Tf groups exhibited slow but long-term
control of blood glucose levels. The hypoglycemic effect of Rice-ProINS-Tf was stronger
than that of HEK-ProINS-Tf as well. Both calculated results of area under curve (AUC)
of blood glucose levels also indicated that, at 22.5 nmol/kg dose, Rice-ProINS-Tf
exhibited a significantly stronger effect on overall blood glucose control when compared
with insulin and HEK-ProINS-Tf.
52
4.3.2. Extra-Long Term Hypoglycemic Effect of Rice-Derived ProINS-Tf
The extra-long term hypoglycemic effect of Rice-ProINS-Tf was compared with
that of insulin Glargine, one of the most prescribed commercial long acting insulin
analogues. The insulin Glargine group showed a much quicker onset with relatively
shorter duration of blood glucose effect (< 16 h). On the other hand, the Rice-ProINS
exhibited a slightly slow onset, but its blood glucose lowering effect under fasting
conditions could sustain 40 h or longer post-injection in the type 1 diabetic mice. This
result suggested that Rice-ProINS-Tf, although exhibited a slightly stronger peripheral
effect than HEK-ProINS-Tf (Fig 4.1), still maintained a hepato-preferential effect in
diabetic mice (Fig 4.3).
4.3.3. Oral Hypoglycemic Effect of Rice-Derived ProINS-Tf
Oral hypoglycemic efficacy of Rice-ProINS-Tf was also investigated in type 1
diabetes mouse model (Fig. 4-4). Compared to the negative control group (oral PBS),
there is no significant decrease in blood glucose levels of STZ-induced diabetic mice
after oral administration of insulin (800 nmol/kg). However, the oral administration of
Rice-ProINS-Tf (800 nmol/kg) exhibited a slow but significant decrease in blood glucose
levels from 4 h at least 12 h time point in comparison with either oral PBS or oral insulin
(800 nmol/kg) group. The blood glucose curve of oral Rice-ProINS-Tf was almost
overlapped with that of subcutaneously injected Rice-ProINS-Tf group (22.5 nmol/kg).
In another experiment (Fig. 4-5a), Rice-ProINS-Tf (11.25 and 22.5 nmol/kg) were
subcutaneously injected to estimate the oral bioactivity of Rice-ProINS-Tf. The blood
53
glucose curve of oral Rice-ProINS-Tf was slightly higher than that of subcutaneously
injected Rice-ProINS-Tf groups at low dose (11.25 nmol/kg), but it still significantly
lower than the those of control groups (oral insulin and PBS) under fasting condition.
However, when mice were under feeding condition for 8 to 16 h time point, all the blood
glucose levels went up, and there was no significant difference between mice with
subcutaneous injection and oral administration.
4.4. Discussion
The presented results in Chapter 3 and Chapter 4 have demonstrated that Rice-
ProINS-Tf exhibited a similar in vitro and in vivo characteristics and bioactivity as HEK-
ProINS-Tf with some minor differences, such as lack of glycosylation of Rice-ProINS-
Tf. According to the data from AKT phosphorylation study and glucose production
inhibition assay, the in vitro insulin activity of Rice-ProINS-Tf was slightly lower than
that of HEK-ProINS-Tf. However, the in vivo potency of Rice-ProINS-Tf in blood
glucose control was considerably higher than that of HEK-ProINS-Tf. This interesting
discrepancy between in vitro and in vivo could be explained by the dual receptor targeting
of ProINS-Tf fusion protein. Recently, the relationships between the binding affinities of
two protein domains to two different domain receptors and the pharmacokinetic profiles
of bifunctional (drug and carrier) fusion proteins have been studied(62). Based on that
study, a fusion protein with stronger binding affinity of drug domain to drug receptor
displayed more receptor-mediated endocytosis, which would lead to more lysosomal
degradation and thus a shorter half-life. On the other hand, the stronger binding affinity
of carrier domain to carrier receptor could help the recycling the fusion protein back to
54
cell surface, avoid its lysosomal degradation, and therefore, prolong the plasma half-life.
According to the results from the competitive TfR binding assay, there is no
significant difference between the TfR binding affinities of Rice- and HEK-ProINS-Tf.
However, the lower AKT phosphorylation and lower inhibition of glucose production
suggested that the converted Rice-ProINS-Tf had lower binding affinity to insulin
receptor (IR). Therefore, Rice-ProINS-Tf with weaker binding affinity to IR should have
lower IR-mediated endocytosis and lysosomal degradation, and higher TfR-mediated
recycling. As a result, more intracellular Rice-ProINS-Tf could be recycled by TfR, and
repeatedly bind to IR on the cell surface. Also, due to the lower IR binding, relatively
more Rice-ProINS-Tf could be released from liver to peripheral tissues compared to
HEK-ProINS-Tf. This might be the reason why Rice-ProINS-Tf exhibited a higher
hypoglycemic effect when the mice were under free-fed condition11. The proposed
comparison model of the receptor binding and intracellular processing of Rice- and HEK-
ProINS-Tf is shown on Scheme 4-1.
The results of oral delivery study demonstrated that Rice-ProINS-Tf, like other
Tf-based fusion proteins, displayed its oral bioactivity on hypoglymic effect. Based on
the results from the oral delivery study (Fig. 4-5), the blood glucose lowing effect of
orally administrated Rice-ProINS-Tf at 800 nmol/kg was slightly lower than that of
subcutaneously injected Rice-ProINS-Tf at 11.25 nmol/kg. This result suggested that the
oral bioactivity of Rice-ProINS-Tf was approximately 1 %, when compared with the
subcutaneously injected groups.
55
When the mice were under feeding conditions, no significant oral hypoglycemic
effect of Rice-ProINS-Tf was observed (Fig. 4-5a). However, the subcutaneously injected
high dose insulin Glargine group (Fig. 4-5b), exhibited a strong postprandial
hypoglycemic effect which is usually caused by the high peripheral insulin effect. The
result of oral study suggested the oral Rice-ProINS-Tf, although could be activated by
trypsin or other enzymes in the GI tract, still exhibited the liver-preferential effect. The
result met our expectation that the orally administrated Rice-ProINS-Tf would be
absorbed into the portal vein and directly transported to the liver, as the same route the
native insulin after the secretion from the pancreas.
In conclusion, we have demonstrated the feasibility of using rice expression
system (Oryza sativa L.) for the production of an intact and effective ProINS-Tf fusion
protein. The in vitro and in vivo studies suggested that Rice-ProINS-Tf has some minor
differences from HEK-ProINS-Tf, such as apparent absence of glycosylation and slightly
lower IR binding affinity. Overall, Rice-ProINS-Tf still shares great similarities with
HEK-ProINS-Tf on their intrinsic characteristics and bioactivities. This result also
suggested that this transgenic rice expression system is most likely suitable for the
production of other Tf-based fusion proteins.
56
Figure 4-1. Hypoglycemic efficacy under feeding condition (subcutaneous injection)
STZ-induced type 1 diabetic mice (n≥3 per group) under free-fed condition were
subcutaneously injected with protein (22.5 nmol/kg) or PBS, and had free access of food
and water during the experiment. Blood glucose level was monitored using OneTouch
Ultra glucose meter at indicated time points as presented as mean ± SD. AUC of blood
glucose level (mean ± SD) from 0 h to 12 h time points in right panel were calculated
using GraphPad Prism (GraphPad Software Inc., San Diego, CA). *P < 0.05; **P < 0.01.
57
Figure 4-2. Hypoglycemic efficacy under fasting condition (subcutaneous injection)
STZ-induced type 1 diabetic mice (n ≥ 3) were fasted for 2 h before the subcutaneous
injection of protein (22.5 nmol/kg) or PBS. Mice were kept fasted with free access of
water until 12 h post injection. Blood glucose level was monitored using OneTouch Ultra
glucose meter at indicated time points as presented as mean ± SD. AUC of blood glucose
level (mean ± SD) from 0 h to 12 h time points in right panel were calculated using
GraphPad Prism (GraphPad Software Inc., San Diego, CA). *P < 0.05; **P < 0.01.
58
Figure 4-3. Extra-long acting hypoglycemic effect of Rice-ProINS-Tf
STZ-induced type 1 diabetic mice (n ≥ 3 per group) were subcutaneously injected with
Rice-ProINS-Tf (135 nmol/kg), Glargine (135 nmol/kg), or saline followed by kept under
8-h fasting/feeding cycle until 56 h post injection. Blood glucose level was monitored
using OneTouch Ultra glucose meter at indicated time points as presented as mean ± SD.
59
Figure 4-4. Hypoglycemic effect of Rice-ProINS-Tf via oral administration
STZ-induced type 1 diabetic mice (n ≥ 3 per group) were orally or subcutaneously (s.c.)
administered with the indicated proteins or buffer. After the injection, the mice were kept
fasted with free access of water for 12 h. The blood glucose level was monitored using
OneTouch Ultra glucose meter at indicated time points as presented as mean ± SD.
60
Figure 4-5. Liver targeting effect of orally administered Rice-ProINS-Tf
(a) STZ-induced type 1 diabetic mice (n ≥ 3 per group) were administered (orally or s.c.)
with the prepared protein or buffer. Mice were kept fasted with free access of water for 8
h, followed kept under free feeding condition for another 8 h. (b) The diabetic mice (n ≥
3 per group) were s.c. injected with HEK-ProINS-Tf (135 nmol/kg), insulin Glargine
(135 nmol/kg), or vehicle followed by kept under 8-h fasting plus 8-h feeding condition
(n ≥ 3 per group). Blood glucose level was monitored at the indicated time points using
OneTouch Ultra glucose meter as presented as mean ± SD.
(a)
(b)
61
Scheme 4-1. Comparison of pathways of Rice- and HEK-ProINS-Tf in liver
hepatocytes
After TfR-mediated intracellular conversion, due to the stronger binding affinity to IR,
most of the activated HEK-ProINS-Tf (blue arrows) would immediately bind to IR, get
endocytosed by IR and subsequently degraded intracellularly in the liver. In contrast, the
weaker IR binding affinity of the converted Rice-ProINS-Tf would cause less IR binding
and more protein recycling by TfR. Therefore, more activated Rice-ProINS-Tf could
repeatedly bind to IR resulting from more intracellular IR dissociation and TfR-mediated
recycling. Consequently, more converted Rice-ProINS-Tf could be released to peripheral
tissues and lead to relatively higher peripheral bioactivity than HEK-ProINS-Tf.
62
CHAPTER 5: TRANSFERRIN-POLYLYSINE CONJUGATES AS A NOVEL
ORAL PROTEIN CARRIER WITH IMPROVED TRANSCYTOSIS ACROSS GI
BARRIER
5.1. Background
Protein and peptide biologics have become an important category of drugs for the
treatment of human diseases(63). One of the limitations in developing proteins and
peptides into therapeutic drugs is the lack of a general method for their non-invasive
delivery(64). Among all non-invasive routes, the oral delivery of proteins and peptides
has long been recognized as an ultimate challenge in drug delivery(65). The advantages
in oral delivery of protein drugs have been well-recognized(66), and the feasibility of oral
delivery of proteins has been demonstrated in various technologies(67). However, there is
no FDA-approved drug delivery system currently available on the market that can
enhance the oral bioavailability of protein drug. One of the hurdles in developing oral
protein delivery system is that many of the delivery techniques are based on non-
physiological approaches, such as penetrating enhancers and surfactants, that may cause
irritation to the GI tract or some safety issues in chronic applications(68). Therefore, it is
desirable that the GI absorption of protein drugs can be relied on natural mechanisms in
GI physiology for the transcellular protein transport. Transcytosis is the major
physiological process for the release of proteins across either epithelial or endothelial cell
layers (69). Conceivably, transcytosis of a protein drug across intestinal epithelial cells
from mucosal to serosal surface via a physiological pathway should be an ideal route for
oral protein drug delivery.
63
Several studies in the past decades have focused on developing transcytosis-based
systems for oral protein drug delivery. In general, there are two different approaches in
those studies, i.e., l). receptor-mediated transcytosis; such as transferrin(24, 70), vitamin
B-12(71, 72), and cholera toxin B subunits(73), and 2). adsorptive transcytosis such as
protamine(74) and cell-penetration peptides(75). Even though limited success in oral
bioavailability of protein drug has been achieved in those reports, a highly efficient
transcytotic pathway is yet to be identified. Under the physiological conditions, the
release of proteins in epithelial cell layers, such as the secretion of IgA via pIgR-
mediated transcytosis(76), is mostly from serosal-to-mucosal surfaces. In addition, the
polarity of epithelial cells will restrict the receptor distribution between serosal and
mucosal membranes(44) and make the receptor-mediated transcytosis inefficient. On the
other hand, adsorptive transcytosis relies on non-specific binding of the protein molecule
on the surface of epithelial mucosal membrane. Even with a very high rate of surface
binding and endocytosis, most of the non-specific membrane-bound protein will be
degraded in lysosomes rather than transcytosed to the serosal membrane(77). Therefore,
the low efficiency in mucosal-to-serosal transcytosis in intestinal epithelial cells is the
major hurdle in the development of transcytosis-based oral protein delivery.
In this chapter, a novel approach was proposed to increase the transcytosis of a
protein molecule across the epithelial cells. The approach is based on the increase of
mucosal uptake of the protein molecules via adsorptive endocytosis and the serosal
release of the protein molecules via receptor-mediated recycling (Scheme 5-1). The
transport process involves an intracellular exchange of the protein drug from an
64
adsorptive carrier, e. g., cationic oligolysine, to a receptor-mediated carrier, e.g.,
transferrin. Such a “relay transcytosis” mechanism has been found in the transcytotic
process of several macromolecular ligands in GI epithelium. For examples, cobalamin
(vitamin B-l2) has been shown to be internalized as the intrinsic factor complex via
intrinsic factor receptor from the mucosal membranes of intestinal epithelial cells(78).
Once internalized, the complex will be degraded and cobalamin will be released from
lysosomes to cytoplasm as transcobalamin complex. The transcobalamin complex will be
released to the blood circulation from the serosal membrane. To demonstrate the
proposed mechanism, Tf and poly-L-lysine (PLL) conjugated by two different
intracellular cleavable linkers, disulfide or acid liable linker, were used as the model
drugs. The cellular binding and transcytosis of these two Tf-PLL conjugates model were
investigated.
5.2. Materials and Methods
5.2.1. Cell Culture
All the binding and transcytosis experiments were performed on Caco-2 cells,
obtained from the American Type Culture Collection (ATCC). Caco-2 cells were grown
in DMEM (Invitrogen) supplemented with 20% (v/v) fetal bovine serum (FBS), 2 mM L-
glutamine, 50 unit/ml penicillin, 50 µg/ml streptomycin and 0.1 mM non-essential amino
acids. HEK293 cells purchased from ATCC were utilized for the expression of
recombinant protein, and were grown in DMEM supplemented with 10% (v/v) FBS, 2
mM L-glutamine, 50 unit/ml penicillin and 50 µg/ml streptomycin.
65
For transcytosis assay, Caco-2 cells were seeded on polycarbonate membrane
filters (0.4 µm pore size) in 6-well Transwell plates (Corning, Corning, New York) as
previously described. The Caco-2 cells were grown for 21 days after confluence, with
culture medium replaced every 2 days as well as 24 h before each experiment. The
transepithelial electrical resistance (TEER) levels were measured using an epithelial
voltohmmeter (World Precision Instruments, West Haven, CT) to ensure complete
differentiation of Caco-2 cells into enterocyte-like cells. One example of TEER curve
during the differentiation was shown in Fig. 5-1. The Transwell membranes were
confirmed with TEER levels higher than 500 ohms•cm2 before used for transcytosis
assays.
5.2.2. Disulfide-Bond Linked Tf-PLL Conjugates (Tf-ss-PLL)
5.2.2.1. Synthesis of Tf-ss-PLL
As shown in Scheme. 5-2, human serum Tf (Sigma) was covalently linked to PLL
(K21, synthesized by Alamanda Polymers, Huntsville, AL) via a reducible disulfide
linkage with a bifunctional cross-linking reagent, N-succinimidyl-3-(2-pyridyldithio)
propionate (SPDP, Thermo Fisher Scientific). First, both Tf and PLL were reacted with
SPDP at 4 °C for 4 h, followed by size exclusion chromatography with PBS as the mobile
phase for purification. The modification ratio of Tf:SPDP and PLL:SPDP was determined
to be 1:3 and 1:1.5, respectively. Then, Tf-PDP was treated with dithiothreitol (DTT) to
expose sulfhydryl groups, and purified again using size exclusion columns (GE
Healthcare, Piscataway, NJ). Sulfhydryl-containing Tf was reacted with PLL-PDP at 4
°C for 24 h to generate disulfide bond-linked Tf-PLL conjugates (Tf-ss-PLL). 2,4,6-
66
Trinitrobenzene sulfonic acid (TNBSA, Sigma) assay using a standard curve generated
with Nα-Acetyl-L-lysine methyl ester hydrochloride was applied to confirm the increased
amino groups of Tf-ss-PLL by following the manufacturer’s instructions. SDS-PAGE
followed by Coomassie Blue staining were applied to further demonstrate if Tf and PLL
were successfully conjugated.
5.2.2.2. Caco-2 Cellular Binding Assay of Tf-ss-PLL
The binding assays on Caco-2 cells were performed to confirm the increased
binding of Tf to cell membrane due to the conjugation of PLL. Caco-2 cells were seeded
in 6-well plates; the experiments were performed after 3 days of seeding when cell
confluence was reached. Caco-2 cells in 6-well plates were first incubated in HBSS at 37
°C for 30 min to remove the serum endogenous Tf. After the incubation, cell monolayers
were washed twice with 1 mL of ice-cold HBSS, and treated with the prepared dosing
solutions in HBSS. After 2 h of incubation at 4 °C, the dosing solutions were removed,
and the cell monolayers were washed three times with ice-cold PBS. The cells were then
dissolved in 1 mL of 1 N NaOH, and radioactivity of the cell lysates was counted using a
gamma counter (Packard). Micro BCA Protein Assay Kit (Thermo Fisher Scientific Inc.)
was used to measure the concentration of cell lysates for normalizing the results.
5.2.2.3. Transcytosis Assay of Tf-ss-PLL
Prior to the transcytosis experiments, cell monolayers were washed with Hank's
balanced salt solution (HBSS) with 10 mM HEPES at pH 7.4, and then incubated with
the same solution to at 37 °C for 30 min to remove the serum Tf. Then, the dosing solution
67
(2 µg/mL
125
I-labeled Tf or Tf-ss-PLL in HBSS with 10 mM HEPES) was added to the
apical compartment of filter-growth Caco-2 cells. After 8-h incubation at 37 °C, the
buffers from the basolateral compartment containing transcytosed
125
I-labeled proteins
were collected; the radioactivity levels were counted as total transcytosed proteins that
included both intact and degraded proteins. The collected samples were subsequently
subjected to 15% trichloroacetic acid (TCA) precipitation to precipitate the intact
proteins(79). The radioactivity in the protein precipitates was measured as intact
transcytosed proteins. The TEER of cell monolayers was measured using an epithelial
voltohmmeter before and after the treatments to ensure the integrity of tight junctions.
The Caco-2 cell monolayers on membrane supports were washed three times with ice-
cold HBSS and cut off from the Transwell inserts. The radioactivity in the collected
membrane was also counted to estimate the cell-associated protein.
5.2.3. Acid-Liable Tf-PLL Conjugates (Tf-NTA-PLL)
5.2.3.1. Synthesis of Tf-NTA-PLL
pH-sensitive nitrilotriacetic (NTA) linker was applied for Tf-PLL conjugation,
because of its efficient cleavage of coordination bonds between polyhistidine-tag and
NTA-coordinated nickel ion as pH 6 or lower. Human Tf with a hexahistidine-tag on the
N-terminal end (Tf-6xHis) was recombinantly expressed and purified as previously
described(32). Briefly, HEK293 cells (ATCC, Manassas, VA) were transiently
transfected with pcDNA3.1(+) DNA plasmid containing the gene of Tf-6xHis via PEI-
mediated transfection. The conditioned CD293 medium (Thermo Fisher Scientific Inc.)
was collected twice on Day 4 and Day 7 post-transfection, and concentrated using
68
tangential flow filtration (TFF, Millipore). The column packed with Ni-NTA agarose
(Thermo Fisher Scientific Inc.) was used for protein purification.
NTA-modified PLL (NTA-PLL) was synthesized as shown in Scheme. 5-3. First,
Nα,Nα-Bis(carboxymethyl)-L-lysine (Sigma) was reacted with succinimidyl 4-(p-
maleimidophenyl) butyrate (SMPB, Thermo Fisher Scientific Inc.) at room temperature
for 4 h to generate NTA-MPB. PLL was reacted with SPDP at 4°C for 4 h, and
subsequently purified by G25 size exclusion column. The final ratio of PLL:SPDP was
1:2. The SPDP-modified PLL was reduced by the treatment with 25 mM DTT to generate
free sulfhydryl groups. Finally, the sulfhydryl-containing PLL was reacted with NTA-
MPB to form NTA-PLL. Tf-NTA-PLL complex was prepared by mixing Tf-6xHis with
NTA-PLL and NiSO
4
(Sigma).
5.2.3.2. Caco-2 Binding Assay and pH Sensitivity Test of Tf-NTA-PLL
Caco-2 cell binding assay was performed as described in 5.3.2.2. To confirm the
pH-sensitive dissociation between Tf-6xHis and Ni
2+
-NTA-PLL,
125
I-labeled Tf-6xHis
was first coincubated with NTA-PLL plus nickel ions (Tf-6xHis:NTA-PLL:Ni
2+
ratio =
1:10:40) in PBS (pH7.4) at 25 °C for 30 min, and then loaded into cation exchange spin
columns (Thermo Fisher Scientific Inc.). After spinning down the mixture and removing
the flow-through, the columns were washed twice with 200 µL of PBS, and subsequently
eluted with 200 µL of citric acid-Na
2
HPO
4
butters at various pH from 7.0 to 5.5. The
radioactivity of each eluate was counted using a gamma counter as the dissociated Tf-
6xHis.
69
5.2.3.3. Transcytosis Assay of Tf-NTA-PLL
Prior to the transcytosis experiments, cell monolayers were washed with Hank's
balanced salt solution (HBSS) with 10 mM HEPES at pH 7.4, and then incubated with
the same solution to at 37 °C for 30 min to remove the serum Tf. Then, the dosing solution
(2 µg/mL
125
I-labeled Tf-6xHis with or without Ni
2+
-NTA-PLL in HBSS with 10 mM
HEPES) was added to the apical or basolateral compartment of filter-growth Caco-2
cells. After 8-h incubation at 37 °C, the buffers from the opposite compartment
containing the transcytosed
125
I-labeled proteins were collected; the radioactivity levels
were counted as total transcytosed proteins that included both intact and degraded
proteins. The collected samples were subsequently subjected to 15% TCA precipitation to
precipitate intact proteins. The radioactivity in the protein precipitates was measured as
intact transcytosed proteins. The TEER of cell monolayers was measured using an
epithelial voltohmmeter before and after the treatments to ensure the integrity of tight
junctions. The Caco-2 cell monolayers on membrane supports were washed three times
with ice-cold HBSS and cut off from the Transwell inserts. The radioactivity of the
collected membrane was counted as cell-associated protein.
5.3. Results
5.3.1. Disulfide-Bond Linked Tf-PLL Conjugates (Tf-ss-PLL)
5.3.1.1. Synthesis of Tf-ss-PLL
In order to test whether Tf and PLL were conjugated with the disulfide bond
linker, the number of primary amine groups were measured using TNBSA assay. The
results in Fig. 5-2a indicated that Tf-ss-PLL group exhibited additional 60 amine groups
70
per molecule when compared with Tf-PDP group. The increased number of amine groups
suggested that Tf was successfully conjugated with PLL, and the conjugation ratio of
Tf:PLL was about 1:3 which was consistent with the ratio of 1:3 in Tf:PDP after the
SPDP modification. SDS-PAGE followed by Coomassie Blue staining was used to
further confirm the disulfide bond conjugation. From the 10 % SDS-PAGE in Fig. 5-2b, a
slight increase of molecular weight could be noticed on the major band of non-reduced
Tf-ss-PLL, when compared with non-reduced Tf. Some Tf oligomers, possibly due to the
multiple PDP modifications of PLL, were also noticed. However, after the treatment with
DTT, there is no difference between Tf and Tf-ss-PLL groups.
5.3.1.2. Cellular Binding Assay of Tf-ss-PLL
Caco-2 monolayers were treated with dosing solutions containing
125
I-labeled Tf
or Tf-ss-PLL at 4 °C to compare the cell membrane binding between these two proteins
(Fig. 5-3). Based on the result, it was confirmed that the cell membrane binding of Tf can
be significantly enhanced after covalently conjugated to PLL. After 2-h incubation, the
cell membrane-bound amount of Tf-ss-PLL (9192.17 ± 658.05 ng/mg cell protein) was
18.3 fold higher than that of Tf (500.84 ± 92.84 ng/mg cell protein). In addition, for
125
I-
Tf-ss-PLL treated groups with the increased concentration of cysteamine, the counted
radioactivity levels of cell lysates were gradually decreased. Because the radioactive
125
I
atoms were only labeled on the tyrosine residues of Tf, these results further confirmed
that the
125
I-Tf can be released from PLL when
125
I-Tf-ss-PLL was exposed within a
reducing environment.
71
5.3.1.3. Transcytosis Assay of Tf-ss-PLL
In this study, the effect of disulfide bond conjugated-PLL on the apical-to-
basolateral (A-to-B) transcytosis of Tf was investigated in 23-day cultured Caco-2 cells
in Transwell plates (Fig. 5-4). After the 8-h treatment, the cell monolayer apically treated
with
125
I-Tf-ss-PLL (73.32 ± 17.32 ng) exhibited 8.74-fold higher cell-associated protein,
compared with
125
I-Tf group (8.43 ± 1.23 ng). Also,
125
I-Tf-ss-PLL (37.32 ± 1.75 ng)
displayed a 2.93-fold higher amount of total A-to-B transcytosed protein than
125
I-Tf
(12.71 ± 0.80 ng). However, after the TCA precipitation, there is no significant difference
between the amounts of transcytosed intact protein of these two groups (p=0.053, n=3).
The amount of transcytosed intact protein on
125
I-Tf-ss-PLL group (1.81 ± 0.41 ng) was
even lower than that of
125
I-Tf group (2.47 ± 0.03 ng). After the treatment, no significant
decrease was noticed on the TEER of each membrane.
5.3.2. Acid-Liable Tf-PLL Conjugates (Tf-NTA-PLL)
5.3.2.1. Cellular Binding Assay of Tf-NTA-PLL
Caco-2 binding assay was performed to confirm the successful conjugation
between Tf-6xHis and Ni
2+
-NTA-PLL via the coordination bonds (Fig. 5-5). Based on
the results, the control group without NTA modification on PLL (187.71 ± 19.96 ng/mg
cell protein) or without nickel ions (199.03 ± 16.68 ng/mg cell protein) displayed no
significant increase on cellular binding of
125
I-Tf-6xHis when compared to
125
I-Tf-6xHis
alone (165.28 ± 25.60 ng/mg cell protein). On the other hand, the binding of
125
I-Tf-
6xHis on the cell membrane was significantly increased when it was coincubated with
NTA-PLL plus nickel ions (Tf-6xHis:NTA-PLL:Ni
2+
ratio = 1:10:40; 407.08 ± 6.95
72
ng/mg cell protein). In addition, the binding was further increased in the presence of
higher concentration of NTA-PLL and Ni
2+
(Tf-6xHis:NTA-PLL:Ni
2+
ratio = 1:40:160;
549.64 ± 32.91 ng/mg cell protein).
To demonstrate the pH-sensitive dissociation of between Tf-6xHis and Ni
2+
-
NTA-PLL (Fig. 5-6), same amount of Tf-NTA-PLL was first immobilized on the cation
ion exchange spin columns and eluted with buffers at different pH, respectively (Fig. 5-
6). The 5-fold increase of the radioactivity in the eluate at pH 5.5 as compared with that
at pH 6.0 or higher suggested that the pH sensitive dissociation occurred between pH 5.5
and 6.0.
5.3.2.2. Transcytosis Assay of Tf-NTA-PLL
Transcytosis assays were also conducted to investigate the transcytosis rate of Tf-
NTA-PLL (Fig. 5-7). After the 8-h treatment, the Caco-2 cell monolayer treated with 2
µg/mL
125
I-Tf-6xHis showed significantly higher amount of cell-associated protein when
coincubated with Ni
2+
-NTA-PLL (A-to-B: 5.68-fold higher; B-to-A: 5.45-fold higher).
However, unlike the transcytosis assay for disulfide bond conjugated Tf-PLL, the
increase of uptake of
125
I-Tf-6xHis did not cause the increase of degraded protein. Also,
the radioactivity from TCA precipitated pellets indicated that
125
I-Tf-6xHis with NTA-
PLL and nickel ions had a 2.68-fold higher rate in A-to-B transcytosis of intact protein
but no statistically significant increase has been noticed for B-to-A transcytosis. In this
experiment, no decease of TEER was detected after the 8-h treatment.
73
5.4. Discussion
In this chapter, we have conjugated polycationic PLL to Tf with two different
linkers to investigate whether the transcytosis rate of Tf could be enhanced. The proposed
mechanism, as shown on Scheme 5-1, was inspired by some previous studies. First, in
some natural cases, transcytosis of molecule across epithelial/endothelial barrier can be
achieved by incorporating two separated endocytic and exocytic mechanisms. For
example, in human’s antenatal period, maternal IgG is transported across the
syncytiotrophoblast by fluid phase endocytosis from maternal blood followed by neonatal
Fc receptor-mediated exocytosis to fetal circulation(80). Second, in epithelial cells,
apical-internalized endosomes could reach the compartment, called common endosomes
or subapical endosomes, that is also accessible for basolateral-internalized
endosomes(81). A more concrete evidence from Hughson’s previous results has proven
that, on polarized Caco-2 cells, apically endocytosed horseradish peroxidase was able to
be transported to the same intracellular compartment that contained basolaterally
endocytosed anti-TfR antibody(82). Therefore, these findings above support the
feasibility to utilize these common endosomes as transfer stations for Tf-based drug
transcytosis.
In this study, PLL was chosen to increase the mucosal uptake of Tf because of the
following reasons. First, PLL has abundant amine groups on the side chain of lysine
residues which allow easy modifications by commercial crosslinkers, such as SPDP and
SMPB. Second, unlike other arginine-rich cell penetrating peptides (CPPs), PLL has been
shown to be internalized predominantly by endocytosis mechanism rather than through
74
direct membrane transduction(83, 84). Also, it has shown that PLL exhibited a similar
endocytosis mechanism as horseradish peroxidase but with much higher endocytosis
rate(85). Therefore, when conjugated to PLL, Tf could be apically internalized into
epithelium via PLL-mediated absorptive endocytosis process, and subsequently
transported to the common endosomes where the basolaterally endocytosed and recycled
TfR presents.
In order to be transcytosed, the dissociation of Tf and PLL is another critical step
which allows internalized Tf to be associated with the TfR-mediated recycling pathway.
In this report, two of the commonly used cleavable linkers were incorporated in this
study(86), and these two different linkers brought totally different results on the
transcytosis studies. The cleavage of disulfide bond linker depends on the reductive
environment during the endocytic pathway(87). Although Tf-PLL conjugated with
disulfide linkers displays a much stronger cellular binding and uptake, the transcytosis of
intact Tf was unexpectedly decreased (Fig. 5-4). The increased degraded protein but
decreased intact protein of Tf-ss-PLL on the basolateral side implied that most of
internalized Tf was most likely degraded into small fragments by the lysosomal enzymes
and then released back to the cell culture medium. Therefore, it is highly possible that the
disulfide bond linker could not be cleaved efficiently in the endosomes or the cleavage
via reducing reaction might happen in the undesirable intracellular compartments, such as
lysosomes(88).
On the other hand, for acid-liable linker, the cleavage is dependent on the
75
decreased pH within endosomal vesicles. Currently, the most commonly used acid liable
linker for drug-carrier conjugation is hydrazone (hz) linker(89). In this project, hydrazone
linker was also used to conjugate Tf and PLL (Tf-hz-PLL). However, the results of
transcytosis assay on Tf-hz-PLL was similar to that on Tf-ss-PLL. The low transcytosis
of Tf-hz-PLL might be due to the same reason: inefficient cleavage under endosomal pH
due to the slow hydrolysis, i.e. 50% release after 6-h incubation at 37 °C/pH 5(90).
Therefore, to overcome the issue, another acid liable linker, NTA linker, was selected.
His-tagged technology is one of the major approaches for protein purification in
biotechnology(91). The potential application of using the coordination interaction
between polyhistidine-tag and Ni
2+
-NTA group as a pH sensitive linker for drug delivery
has also been exploited recently(92). Unlike the hydrazone linker which is cleaved via
slow hydrolysis process, the dissociation between His-tag and Ni-NTA group relies on
the protonation of histidine (pKa 6.0)(93). Thus, this NTA linkage could be dissociated
rapidly in the common endosomes at mildly acidic pH (pH 5.8)(94). Such a pH-
dependent dissociation can efficiently release Tf from PLL to avoided the lysosomal
degradation.
The coincubation with Ni-NTA-PLL could significantly increase the binding and
uptake of Tf-6xHis from the both sides of epithelial cell surface. However, only A-to-B
transcytosis of intact protein was statistically increased (Fig. 5-7a). The low B-to-A
transcytosis was most likely caused by the polarized distribution of TfR on the cell
membrane of filter-grown Caco-2 cells(43). Therefore, the results suggested that the
76
strategy could be only applied to improve the transcytosis of cargos from apical to
basolateral side. Also, the results indirectly implied that the TfR-mediated recycling is a
limiting step in the proposed relay transcytosis process.
77
Scheme 5-1. Transcytosis of Tf-PLL conjugates across Caco-2 epithelial cells
Tf-PLL conjugates with an endosomal cleavable linker can be bound to the apical
membrane of epithelial cells and internalized into cells via a non-specific adsorptive
endocytosis mechanism provided by the positive charges in PLL. The endocytosed Tf-
PLL conjugate can be distributed to common endosome that is also accessible to
basolaterally endocytosed and recycled Tf and TfR. In the common endosomes, Tf will
be separated from PLL by the cleavage of linker, bind the unoccupied TfR, and
subsequently be released to basolateral surface of epithelium by TfR-mediated recycling
pathway. On the other hand, membrane-bound PLL in the endosomes will be eventually
transported to and degraded in the lysosomes.
78
Figure 5-1. The TEER of Caco-2 cell monolayers on the 6-well Transwell plates
Caco-2 cells were seeded on the inner chamber in 6-well Transwell plates as previously
described. The Caco-2 cells were grown for 21 days after confluence, with fresh culture
medium replaced every two days. The TEER levels were measured using an epithelial
voltohmmeter.
79
Scheme 5-2. Synthesis process of Tf-ss-PLL conjugates
Tf and PLL were reacted with SPDP to form Tf-PDP and PLL-PDP, respectively. Tf-
PDP was treated with DTT to generate the free sulfhydryl group, and subsequently,
reacted with PLL-PDP to form Tf-ss-PLL conjugates.
80
Scheme 5-3. Synthesis process of Tf-NTA-PLL conjugates
Nα,Nα-Bis(carboxymethyl)-L-lysine was first reacted with SMPB to produce NTA-MPB.
PLL was reacted with SPDP to produce PLL-PDP and then treated with DTT to generate
free sulfhydryl groups. Sulfhydryl-containing PLL was subsequently reacted with NTA-
MPB to form NTA-PLL. Coordination bonds between Tf-6xHis and NTA-PLL can be
formed by directly mixing NTA-PLL with Tf-6xHis and NiSO4
81
(a) (b)
Figure 5-2. Characterization of Tf-ss-PLL
(a) Number of primary amino group per molecule was determined by TNBSA assay. 100
µg/mL Tf-PDP or Tf-ss-PLL was reacted with TNBSA followed by 2-h incubation at 37
°C. 10% SDS and 1 N HCl were added into the samples. Number of amines contained in
the sample was estimated by measuring the absorbance at 335 nm and using a standard
curve of Nα-Acetyl-L-lysine methyl ester hydrochloride. Single asterisks (*) indicates
significance with P < 0.05, as determined by the Student’s t-test (n=3) (b) 10 % SDS-
PAGE followed by Coomassie Blue staining. 6 µg of indicated proteins with or without
reducing agent DTT were loaded.
1. Tf-ss-PLL (non-reducing)
2. Tf (non-reducing)
3. Tf (reducing)
4. Tf-ss-PLL (reducing)
M. Marker
82
Figure 5-3. Caco-2 binding assay of Tf-ss-PLL
Caco-2 cells were treated with HBSS containing 10 µg/mL
125
I-Tf or
125
I-Tf-ss-PLL at 4
°C for 2 h. Serial dilutions of cysteamine from 1 µM to 1 mM were added to the cell
culture medium. The data represent the radioactivity of cell lysates normalized by total
cell protein amount determined using MicroBCA assay. Double asterisks (**) indicates
significance with P < 0.01 as determined by the Student’s t-test (n=3)
83
Figure 5-4. Transcytosis assay of Tf-ss-PLL
Caco-2 cell monolayers grown on 6-well Transwell inserts were apically treated with 2
µg/mL
125
I-Tf or
125
I-Tf-ss-PLL. After 8-h incubation at 37 °C, samples were collected
from the basolateral compartment, and subjected 15% TCA precipitation. The
radioactivity levels of cell monolayer on the membrane support, collected basolateral
solution and TCA-precipitated pellet were counted by gamma counter as the amount of
cell-associated protein, degraded transcytosed protein and intact transcytosed protein,
respectively. Triple asterisks (***) indicated P < 0.01 and 0.001, respectively, as
determined by the Student’s t-test analysis (n=3).
84
Figure 5-5. Caco-2 binding assay of Tf-NTA-PLL
(a) Caco-2 cells in 6-well plated were incubated with HBSS containing 0.8 µg/mL of
125
I-
Tf-6xHis alone, or in the absence or presence of PLL, NTA-PLL or NiSO4 at 4 °C for 2
h. The data represent the membrane-bound
125
I-Tf-6xHis normalized by total cell protein
amount determined using MicroBCA assay. double asterisks (**) and triple asterisks
(***) indicated P < 0.01 and 0.001, respectively, as determined by the Student’s t-test
(n=3)
85
Figure 5-6. pH sensitive dissociation between Tf-6xHis and Ni
2+
-NTA-PLL
Cation exchange spin column were loaded with same amount of
125
I-Tf-NTA-PLL
followed by elution with 200 µL of buffers at pH 7.0 to 5.5. The radioactivity of eluates
was counter using the gamma counter. The data represent the amount of
125
I-Tf-6xHis in
each eluate. The double asterisks (**) indicated P < 0.01, as determined by the Student’s
t-test (n=3)
86
(a) A-to-B Transcytosis (b) B-to-A Transcytosis
Figure 5-7. Transcytosis assays of Tf-NTA-PLL
Caco-2 cell monolayers grown on 6-well Transwell were apically (a) or basolaterally (b)
treated with 2 µg/mL
125
I-Tf-6xHis in the presence or absence of NTA-PLL and NiSO
4
(molecular ratio of Tf-6xHis:NTA-PLL:Ni
2+
= 1:50:200). After 8-h incubation at 37 °C,
the samples were collected from the medium in the opposite compartment, and subjected
15% TCA precipitation. The radioactivities of cell monolayer on the membrane supports
and TCA-precipitated pellets were counted by a gamma counter to determine the amount
of cell-associated protein and intact transcytosed protein, respectively. No decrease of
TEER was noticed after the 8-h treatments. Single asterisks (*) and triple asterisks (***)
indicated P < 0.05 and 0.001, respectively, as determined by the Student’s t-test analysis
(n=3).
125
I-Tf-6xHis
125
I-Tf-6xHis
+ Ni
2+
-NTA-PLL
87
CHAPTER 6: SUMMARY AND FUTURE PERSPECTIVES
6.1. Rice-ProINS-Tf as an Oral Deliverable Basal Insulin Analog
The results from previous publications indicated that ProINS-Tf fusion protein is
an insulin analogue with several advantages over commercial products currently on the
market. The major features of ProINS-Tf that have been demonstrated are the liver
preferential activity, the extra-long acting effect on blood glucose control and the lower
risk of severe hypoglycemia (34, 38). However, one of the potential applications of
ProINS-Tf that had not been investigated was the feasibility for oral administration and
absorption. In this dissertation, two major critical issues of developing the oral delivery
of ProINS-Tf were addressed and investigated separately.
6.1.1. The Rice-derived ProINS-Tf Fusion Protein
The first encountered issue for the oral delivery of ProINS-Tf fusion protein was
its low production yield from the transient expression system of mammalian HEK293
cells. Many different strategies have been proposed and conducted to solve the issue.
Eventually, the transgenic rice expression system was selected as an alternative system
for ProINS-Tf production due to its more productive and economic features when
compared with the transient expression system of HEK293 cells.
In Chapter 2, it was demonstrated that ProINS-Tf could be successfully produced
from the transgenic rice protein expression system, and the preparation processes of both
HEK-ProINS-Tf and Rice-ProINS-Tf were both described in detail. Although the
88
transgenic rice of Rice-ProINS-Tf could take months to grow from sowing to harvesting
stage, the final production yield was much higher than that of our lab-scale production
from HEK293 mammalian cells. Usually, the monthly production yield of HEK-ProINS-
Tf from our laboratory has been only 1-2 mg. On the other hand, 10-30 mg of purified
Rice-ProINS-Tf could be monthly acquired from rice flour using our lab-scale extraction
and purification equipment. Therefore, in terms of final production performance, the
results suggested that Rice-ProINS-Tf still exhibits strong advantages of low cost, high
yield and high scale-up capacity compared with mammalian cell culture production
system. The rice expression system has provided us with an efficient platform to prepare
the fusion protein for the following in vitro and in vivo experiments.
Based on the results described in Chapter 3 and 4, it was demonstrated that Rice-
ProINS-Tf exhibited very similar characteristics and functional activities as HEK-
ProINS-Tf. Although from the in vitro studies, the insulin-like activity of Rice-ProINS-Tf
was slightly lower, its in vivo effect in blood glucose control in the type-1 diabetic mice
was considerably higher than that of HEK-ProINS-Tf in the following in vivo studies.
The possible explanations for this in vitro and in vivo discrepancy include a higher
bioactivity of Rice-ProINS-Tf in peripheral tissues than that of HEK-ProINS-Tf or a
prolonged half-life of Rice-ProINS-Tf compared to HEK-ProINS-Tf in the body.
Finally, for oral delivery study, the orally administered Rice-ProINS-Tf (800
nmol/kg) exhibited a significant decrease in the blood glucose levels from 4 h to at least
12 h time point in the type 1 diabetic mice, when compared with either oral PBS or oral
89
insulin control group. The oral bioavailability of Rice-ProINS-Tf was approximately 1 %
in comparison with the subcutaneously injected Rice-ProINS-Tf groups. The blood
glucose lowering effect was completely eliminated under the feeding condition,
suggesting that orally administered Rice-ProINS-Tf also exhibited a liver-targeting
bioactivity. Therefore, the results demonstrated that Rice-ProINS-Tf can be developed
into an oral deliverable basal insulin analog.
6.1.2. Transcytosis of Tf-NTA-PLL as an Enhanced Oral Protein Drug Carrier
Tf, with its receptor-mediated transcytosis mechanism across intestinal
epithelium, has long been proven as a potential oral delivery carrier for different protein
therapeutics. However, one of the major rate-limiting issues of Tf in oral delivery is the
polarized expression of TfR on the basolateral surface instead of the mucosal surface of
the intestinal epithelium. To overcome this issue in attempt to develop a more efficient
oral delivery carrier, a novel transcytosis mechanism was proposed in the dissertation.
The proposed transcytosis mechanism is composed of three major steps: apical
binding and endocytosis by a positively charged peptide, intra-endosomal cleavage of the
linker to release the cargo, and basolateral exocytosis by TfR-mediated recycling
pathway. The concept was successfully demonstrated when Tf was conjugated to PLL
using a pH-sensitive NTA linker. The results of Caco-2 transcytosis assay suggested that
transcytosis rate of histidine-tagged Tf could be unidirectionally increased from apical to
basolateral compartment when co-incubated with Ni
2+
-NTA-PLL. The rate of basolateral
to apical transcytosis was not increased possibly due to the lack to TfR-mediated
90
recycling mechanism on the apical surface.
In conclusion, the transcytosis mechanisms of Tf and PLL were restricted by the
low apical expression level of TfR and the low lysosomal escape of PLL, respectively. In
Chapter 4, Tf and PLL was conjugated together with a cleavable linker to overcome each
of the restrictions and to take the advantages of their complementary properties. The
results from this in vitro model have successfully demonstrated that the intestinal
epithelial transcytosis of Tf could be significantly increased, when conjugated to PLL
with a pH-sensitive NTA linker.
6.1.3. Final Conclusion
To summarize this dissertation, the presented data suggest the transgenic rice
system (Oryza sativa L.) with the advantages of low cost and high productivity is an
alternative expression system for ProINS-Tf production. The in vivo data further
confirmed that Rice-ProINS-Tf exhibits an oral blood glucose control effect with oral
bioactivity ≈ 1.0 %. In the future, it is expected that the concept of proposed transcytosis
mechanism could be applied to Rice-ProINS-Tf to further enhance its oral bioactivity.
Hopefully, this rice-derived ProINS-Tf fusion protein could eventually be developed into
a low-cost, long-acting, and liver-preferential insulin therapeutic with high oral
bioavailability.
6.2. Future Perspectives
In this dissertation, the transgenic rice (Oryza sativa L.) has been demonstrated as
91
an alternative expression system of ProINS-Tf to solve the low yield and high cost issues
of HEK293 expression platform. However, this expression system is not perfect. In the
purification process of Rice-ProINS-Tf, it has been noticed that the Tf-like impurities in
the rice-extracted protein could be one of the potential issues of Rice-ProINS-Tf final
product. As ProINS-Tf needs to be converted into an active form by TfR-mediated
intracellular pathway, with the presence of excess Tf, the in vitro and in vivo bioactivities
of Rice-ProINS-Tf might be more or less compromised. Although most of Tf-like
impurities could be removed through the presented purification processes, to completely
remove the impurity would likely cause more than 50 % loss of the final product due to
the very similar physicochemical properties of ProINS-Tf and Tf-like impurities.
Therefore, to thoroughly solve this issue, we could select another rice clone with higher
Rice-ProINS-Tf without or with lower Tf impurities in the rice-extracted protein. In
addition, the transgenic rice is not the only system which has been used for Tf expression.
Most recently, it has been demonstrated that both Tf and Tf fusion protein containing
GLP-1 or its derivatives could be expressed from transgenic tobacco plants(95).
Conceivably, if the high Tf-like impurities issue cannot be resolved in other transgenic
clones of rice expression system, the transgenic tobacco plants should be considered as
another alternative expression system for ProINS-Tf production.
In the oral delivery studies of Rice-ProINS-Tf, the data from orally and
subcutaneously injected groups suggested the oral bioactivity of Rice-ProINS-Tf was
around 1 %. Since the fusion protein was only administered with PBS, therefore, it is
highly possible that the oral bioactivity could be further improved by applying some
92
formulations such as absorption enhancers or enzymatic inhibitors(67). For example,
complexation hydrogels, p(MAA-g-EG) microparticles, have been applied to insulin-Tf
conjugates by Kavimandan et al(96). The results suggested that the hydrogels could
successfully improve the absorption of insulin-Tf conjugates, most likely through the
enhancement of paracellular transport induced by the dilation of tight junction. The
second reason might be the increased stability of the conjugate against proteolytic
degradation in the GI tract.
It was also reported that insulin-Tf oligomer was more effective than monomeric
insulin-Tf conjugate in controlling the blood glucose levels when orally administered to
STZ-induced diabetic rats, possibly due to the alteration of TfR trafficking and increased
stability of the oligomer in the GI tract(44). Based on the similar concept, adding Zn
2+
ion into the formulation might be also beneficial for the oral absorption of ProINS-Tf. It
has been previous reported that during the biosynthesis mechanism of native insulin,
ProINS is first translated and secreted into Zn
2+
- and Ca
2+
-rich secretory vesicles and
assembled into a Zn
2+
- and Ca
2+
-containing hexameric form, (Zn
2+
)
2
(Ca
2+
)(ProINS)
6
(97).
The hexameric form of ProINS will be subsequently converted to the insulin hexamer, by
removing of the C-peptides through the action of the proteolytic enzymes. Therefore,
adding Zn
2+
ions into the ProINS-Tf formulation should be also able to induce the
formation of ProINS-Tf hexamer. The orally administered ProINS-Tf hexamer could still
be absorbed through TfR-mediated transcytosis, and the ProINS moieties in the hexamer
core could be protected by the surrounding enzyme-resistant Tfs from the degradation in
the GI tract. In addition, theoretically, this hexameric structure could be utilized to
93
achieve a further extended effect when compared with ProINS-Tf monomer.
The successful expression of ProINS-Tf from transgenic rice further ensured the
possibility of using the same platform to produce other Tf-fusion proteins, such as GH-Tf
and G-CSF-Tf. Many reports have suggested that protein drugs encapsulated in plant
cells could be protected from acidic and enzymatic degradation in the stomach(98). Once
it entered intestinal gut, the intracellular protein therapeutic would be subsequently
released into the lumen due to the digestion of the plant cell wall by the intestinal
microbes(99). Due to some potential issues, such as the abundant starch in the rice flour,
it might not be an ideal approach for oral administration of the unprocessed and
unpurified rice flour of ProINS-Tf into the body. However, for some other Tf-carried
protein therapeutics, e.g. GH-Tf fusion protein, the extraction and purification process
may not be necessary. The unprocessed rice flour of GH-Tf could provide not only an
ideal protective system to the fusion protein but the necessary nutrients for the GH-
deficient patients.
For the subproject of Tf-PLL conjugates, the data we obtained supported the
proposed hypothesis in which PLL could be utilized to increase the apical binding and
uptake of Tf into the epithelial model and eventually promote the transcytosis across the
cellular barrier. However, this concept could not be further investigated in the in vivo
models because of the toxicity and carcinogenicity caused by the nickel ion(100). The
quick dissociation between Tf and PLL in the intracellular vesicle is an essential step to
allow Tf to bind to the TfR in the endosomes and achieve better transcytosis rate.
94
Therefore, other appropriate linkers which can be cleaved efficiently and rapidly under
the endosomal pH should be further investigated.
To design and develop a desirable pH sensitive linker between Tf and PLL with
no toxic effect for this mechanism, a concept from a recent publication regarding a pH-
sensitive histidine–glutamic acid (HE) copolymer could be utilized (101-103). To form
the linkage, Tf recombinantly fused with the HE polypeptide can be directly mixed with
PLL. At physiological pH 7.4, as the imidazole side chain of histidine (pKa ≈ 6.0) in the
HE oligopeptide sequence is neutral, Tf and PLL could be bound together via the
electrostatic interaction between the anionic γ-carboxylate side chain of glutamic acid
(pKa ≈ 4.3) and cationic ε-amino side chain of lysine (pKa ≈ 10.5). However, once being
exposed to the mildly acidic pH in the endosomes, some of the histidine residues of the
HE polypeptide would be protonated and thus neutralize the negatively charged glutamic
acid side chain. Therefore, the charge interaction between HE polypeptide and PLL
would be disrupted. This pH-sensitive linkage between HE polypeptide and PLL should
be able to bring a very similar effect as the NTA linker, since both of the pH sensitive
dissociation mechanisms rely on the protonation of histidine groups. Most importantly,
no toxic nickel ions will be needed anymore. In the future, it is expected that this
proposed mechanism could be utilized to improve the epithelial transcytosis and oral
absorption of not only in ProINS-Tf but also other Tf-based fusion proteins that have
been previously constructed, such as GH-, G-CSF- and GLP-1-Tf fusion proteins.
95
REFERENCES
1. D. V. Goeddel et al., Expression in Escherichia coli of chemically synthesized
genes for human insulin. Proc. Natl. Acad. Sci. U.S.A. 76, 106–110 (1979).
2. M. S. Kinch, An overview of FDA-approved biologics medicines. Drug
Discovery Today. 20, 393–398 (2015).
3. B. Leader, Q. J. Baca, D. E. Golan, Protein therapeutics: a summary and
pharmacological classification. Nat Rev Drug Discov. 7, 21–39 (2008).
4. S. Frokjaer, D. E. Otzen, Protein drug stability: a formulation challenge. Nat Rev
Drug Discov. 4, 298–306 (2005).
5. E. H. Moeller, L. Jorgensen, Alternative routes of administration for systemic
delivery of protein pharmaceuticals. Drug Discovery Today: Technologies. 5,
e89–e94 (2008).
6. A. L. Demain, P. Vaishnav, Production of recombinant proteins by microbes and
higher organisms. Biotechnology Advances. 27, 297–306 (2009).
7. M. Baker, H. M. Reynolds, B. Lumicisi, C. J. Bryson, Immunogenicity of protein
therapeutics: The key causes, consequences and challenges. Self/Nonself. 1, 314–
322 (2014).
8. D. M. Czajkowsky, J. Hu, Z. Shao, R. J. Pleass, Fc-fusion proteins: new
developments and future perspectives. EMBO Mol Med. 4, 1015–1028 (2012).
9. K. H. Khan, Gene expression in Mammalian cells and its applications. Adv
Pharm Bull. 3, 257–263 (2013).
10. F. M. Wurm, Production of recombinant protein therapeutics in cultivated
mammalian cells. Nat Biotechnol. 22, 1393–1398 (2004).
11. B. F. Choonara et al., A review of advanced oral drug delivery technologies
facilitating the protection and absorption of protein and peptide molecules.
Biotechnology Advances. 32, 1269–1282 (2014).
12. K. Park, I. C. Kwon, K. Park, Oral protein delivery: Current status and future
prospect. Reactive and Functional Polymers. 71, 280–287 (2011).
13. K. Sonaje et al., Opening of Epithelial Tight Junctions and Enhancement of
Paracellular Permeation by Chitosan: Microscopic, Ultrastructural, and
Computed-Tomographic Observations. Mol. Pharmaceutics, 120413125623001–
9 (2012).
14. A. K. Petrus, A. R. Vortherms, T. J. Fairchild, R. P. Doyle, Vitamin B12 as a
Carrier for the Oral Delivery of Insulin. ChemMedChem. 2, 1717–1721 (2007).
96
15. J. W. Loh, J. Schneider, M. Carter, M. Saunders, L.-Y. Lim, Spinning Disc
Processing Technology: Potential for Large-Scale Manufacture of Chitosan
Nanoparticles. J Pharm Sci. 99, 4326–4336 (2010).
16. A. Verma et al., Vitamin B12 functionalized layer by layer calcium phosphate
nanoparticles: A mucoadhesive and pH responsive carrier for improved oral
delivery of insulin. Acta Biomaterialia. 31, 288–300 (2016).
17. K. Thorstensen, I. Romslo, The role of transferrin in the mechanism of cellular
iron uptake. Biochem. J. 271, 1–9 (1990).
18. K. Mizutani, M. Toyoda, B. Mikami, X-ray structures of transferrins and related
proteins. BBA - General Subjects. 1820, 203–211 (2012).
19. H. M. Baker, B. F. Anderson, E. N. Baker, Dealing with iron: common structural
principles in proteins that transport iron and heme. Proc. Natl. Acad. Sci. U.S.A.
100, 3579–3583 (2003).
20. Z. M. Qian, H. Li, H. Sun, K. Ho, Targeted drug delivery via the transferrin
receptor-mediated endocytosis pathway. Pharmacol. Rev. 54, 561–587 (2002).
21. X. Chen, J. L. Zaro, W.-C. Shen, Pharmacokinetics of recombinant bifunctional
fusion proteins. Expert Opinion on Drug Metabolism & Toxicology. 8, 581–595
(2012).
22. B. J. Kim et al., Transferrin Fusion Technology: A Novel Approach to
Prolonging Biological Half-Life of Insulinotropic Peptides. Journal of
Pharmacology and Experimental Therapeutics. 334, 682–692 (2010).
23. T. R. Daniels et al., The transferrin receptor and the targeted delivery of
therapeutic agents against cancer. Biochimica et Biophysica Acta (BBA) -
General Subjects. 1820, 291–317 (2012).
24. D. Shah, W.-C. Shen, Transcellular Delivery of an Insulin-Transferrin Conjugate
in Enterocyte-like Caco-2 Cells. J Pharm Sci. 85, 1306–1311 (1996).
25. Y. Bai, D. K. Ann, W.-C. Shen, Recombinant granulocyte colony-stimulating
factor-transferrin fusion protein as an oral myelopoietic agent. Proc. Natl. Acad.
Sci. U.S.A. 102, 7292–7296 (2005).
26. N. Amet, W. Wang, W.-C. Shen, Human growth hormone–transferrin fusion
protein for oral delivery in hypophysectomized rats. Journal of Controlled
Release. 141, 177–182 (2010).
27. Centers for Disease Control and Prevention, National Diabetes Statistics Report,
2014, 1–12 (2016).
28. G. Wilcox, Insulin and insulin resistance. Clin Biochem Rev. 26, 19–39 (2005).
97
29. I. S. Johnson, The trials and tribulations of producing the first genetically
engineered drug. Nat Rev Drug Discov. 2, 747–751 (2003).
30. J. Nielsen, Production of biopharmaceutical proteins by yeast. Bioengineered. 4,
207–211 (2014).
31. N. A. Baeshen et al., Cell factories for insulin production. Microb. Cell Fact. 13,
141 (2014).
32. Y. Wang, Y.-S. Chen, J. L. Zaro, W.-C. Shen, Receptor-mediated activation of a
proinsulin-transferrin fusion protein in hepatoma cells. Journal of Controlled
Release. 155, 386–392 (2011).
33. D. E. Peavy, J. D. Abram, B. H. Frank, W. C. Duckworth, In vitro activity of
biosynthetic human proinsulin. Receptor binding and biologic potency of
proinsulin and insulin in isolated rat adipocytes. Diabetes. 33, 1062–1067 (1984).
34. J. Shao, J. L. Zaro, W.-C. Shen, Proinsulin–Transferrin Fusion Protein Exhibits a
Prolonged and Selective Effect on the Control of Hepatic Glucose Production in
an Experimental Model of Type 1 Diabetes. Mol. Pharmaceutics. 13, 2641–2646
(2016).
35. A. D. Cherrington, Control of glucose uptake and release by the liver in vivo.
Diabetes. 48, 1198–1214 (1999).
36. W. C. Duckworth, R. G. Bennett, F. G. Hamel, Insulin Degradation: Progress
and Potential. Endocrine Reviews. 19, 608–624 (1998).
37. D. Russell-Jones, R. Khan, Insulin-associated weight gain in diabetes – causes,
effects and coping strategies. Diabetes Obes Metab. 9, 799–812 (2007).
38. Y. Wang, J. Shao, J. L. Zaro, W.-C. Shen, Proinsulin-transferrin fusion protein as
a novel long-acting insulin analog for the inhibition of hepatic glucose
production. Diabetes. 63, 1779–1788 (2014).
39. S. Kalra, B. Kalra, N. Agrawal, Oral insulin. Diabetol Metab Syndr. 2, 66 (2010).
40. E. Zijlstra, L. Heinemann, L. Plum-Mörschel, Oral insulin reloaded: a structured
approach. J Diabetes Sci Technol. 8, 458–465 (2014).
41. C. Q. Xia, J. Wang, W. C. Shen, Hypoglycemic effect of insulin-transferrin
conjugate in streptozotocin-induced diabetic rats. Journal of Pharmacology and
Experimental Therapeutics. 295, 594–600 (2000).
42. H. Hoelen et al., Proteasomal Degradation of Proinsulin Requires Derlin-2,
HRD1 and p97. PLoS ONE. 10, e0128206–16 (2015).
43. D. Shah, W.-C. Shen, The Establishment of Polarity and Enhanced Transcytosis
98
of Transferrin Receptors in Enterocyte-like Caco-2 Cells. Journal of Drug
Targeting. 2, 93–99 (1994).
44. C.-J. Lim, W.-C. Shen, Comparison of monomeric and oligomeric transferrin as
potential carrier in oral delivery of protein drugs. Journal of Controlled Release.
106, 273–286 (2005).
45. N. Noinaj et al., Structural basis for iron piracy by pathogenic Neisseria. Nature.
482, 53–58 (2012).
46. E. Yodoya, K. Rikyu, A. Yamamoto, S. Muranishi, Effects of various protease
inhibitors on the intestinal absorption of insulin. Regional difference between the
small and the large intestine. Drug Delivery System. 8, 115–119 (1993).
47. S. Harada, R. M. Smith, L. Jarett, 1,10-Phenanthroline increases nuclear
accumulation of insulin in response to inhibiting insulin degradation but has a
biphasic effect on insulin's ability to increase mRNA levels. DNA Cell Biol. 13,
487–493 (1994).
48. C.-Y. Lin et al., Enhancing Protein Expression in HEK-293 Cells by Lowering
Culture Temperature. PLoS ONE. 10, e0123562–19 (2015).
49. L. Haataja et al., Disulfide Mispairing During Proinsulin Folding in the
Endoplasmic Reticulum. Diabetes. 65, 1050–1060 (2016).
50. S.-G. Chang, K.-D. Choi, S.-H. Jang, H.-C. Shin, Role of disulfide bonds in the
structure and activity of human insulin. Mol. Cells. 16, 323–330 (2003).
51. S. Oyadomari et al., Targeted disruption of the Chop gene delays endoplasmic
reticulum stress-mediated diabetes. J. Clin. Invest. 109, 525–532 (2002).
52. N. Amet, H.-F. Lee, W.-C. Shen, Insertion of the Designed Helical Linker Led to
Increased Expression of Tf-Based Fusion Proteins. Pharm. Res. 26, 523–528
(2008).
53. J. K.-C. Ma, P. M. W. Drake, P. Christou, The production of recombinant
pharmaceutical proteins in plants. Nat. Rev. Genet. 4, 794–805 (2003).
54. C. L. Nykiforuk et al., Transgenic expression and recovery of biologically active
recombinant human insulin from Arabidopsis thaliana seeds. Plant Biotechnol. J.
4, 77–85 (2006).
55. D. Boyhan, H. Daniell, Low-cost production of proinsulin in tobacco and lettuce
chloroplasts for injectable or oral delivery of functional insulin and C-peptide.
Plant Biotechnol. J. 9, 585–598 (2010).
56. D. Zhang et al., Expression, purification, and characterization of recombinant
human transferrin from rice (Oryza sativa L.). Protein Expression and
99
Purification. 74, 69–79 (2010).
57. D. Zhang et al., Characterization of transferrin receptor-mediated endocytosis
and cellular iron delivery of recombinant human serum transferrin from rice
(Oryza sativa L.). BMC Biotechnology. 12, 1–1 (2012).
58. A. N. Steere et al., Biochemical and structural characterization of recombinant
human serum transferrin from rice (Oryza sativa L.). Journal of Inorganic
Biochemistry. 116, 37–44 (2012).
59. R. Alfano, B. A. Youngblood, D. Zhang, N. Huang, C. C. MacDonald, Human
leukemia inhibitory factor produced by the ExpressTec method from rice ( Oryza
sativaL.) is active in human neural stem cells and mouse induced pluripotent
stem cells. Bioengineered. 5, 180–185 (2014).
60. S. Sonoda, M. Schlamowitz, Studies of 125I trace labeling of immunoglobulin G
by chloramine-T. Immunochemistry. 7, 885–898 (1970).
61. P. D. Home, S. G. Ashwell, An overview of insulin glargine. Diabetes Metab Res
Rev. 18, S57–S63 (2002).
62. X. Chen, H.-F. Lee, J. L. Zaro, W.-C. Shen, Effects of Receptor Binding on
Plasma Half-Life of Bifunctional Transferrin Fusion Proteins. Mol.
Pharmaceutics. 8, 457–465 (2011).
63. P. J. Carter, Introduction to current and future protein therapeutics: A protein
engineering perspective. Exp. Cell Res. 317, 1261–1269 (2011).
64. I. Mahmood, M. D. Green, Pharmacokinetic and pharmacodynamic
considerations in the development of therapeutic proteins. Clin Pharmacokinet.
44, 331–347 (2005).
65. M. Goldberg, I. Gomez-Orellana, Challenges for the oral delivery of
macromolecules. Nat Rev Drug Discov. 2, 289–295 (2003).
66. P. Fonte, F. Araujo, S. Reis, B. Sarmento, Oral Insulin Delivery: How Far are
We? J Diabetes Sci Technol. 7, 520–531 (2013).
67. A. Muheem et al., A review on the strategies for oral delivery of proteins and
peptides and their clinical perspectives. SAUDI PHARMACEUTICAL
JOURNAL. 24, 413–428 (2016).
68. U. B. Kompella, V. H. Lee, Delivery systems for penetration enhancement of
peptide and protein drugs: design considerations. Advanced Drug Delivery
Reviews. 46, 211–245 (2001).
69. P. L. Tuma, A. L. Hubbard, Transcytosis: crossing cellular barriers.
Physiological Reviews. 83, 871–932 (2003).
100
70. B. Melanie, Transferrin' the load. Nat Rev Drug Discov. 4, 537–537 (2005).
71. K. S. Ramanujam, S. Seetharam, M. Ramasamy, B. Seetharam, Expression of
cobalamin transport proteins and cobalamin transcytosis by colon
adenocarcinoma cells. Am. J. Physiol. 260, G416–22 (1991).
72. B. Seetharam, R. R. Yammani, Cobalamin transport proteins and their cell-
surface receptors. ERM. 5, 1–18 (2003).
73. W. I. Lencer, S. Moe, P. A. Rufo, J. L. Madara, Transcytosis of cholera toxin
subunits across model human intestinal epithelia. Proc. Natl. Acad. Sci. U.S.A.
92, 10094–10098 (1995).
74. M. W. Peterson, D. Gruenhaupt, Protamine increases the permeability of cultured
epithelial monolayers. J. Appl. Physiol. 68, 220–227 (1990).
75. M. Kristensen, H. M. Nielsen, Cell-Penetrating Peptides as Carriers for Oral
Delivery of Biopharmaceuticals. Basic Clin Pharmacol Toxicol. 118, 99–106
(2015).
76. F.-E. Johansen, C. S. Kaetzel, Regulation of the polymeric immunoglobulin
receptor and IgA transport: new advances in environmental factors that stimulate
pIgR expression and its role in mucosal immunity. Mucosal Immunology. 4, 598–
602 (2011).
77. R. Tréhin et al., Metabolic cleavage of cell-penetrating peptides in contact with
epithelial models: human calcitonin (hCT)-derived peptides, Tat(47-57) and
penetratin(43-58). Biochem. J. 382, 945–956 (2004).
78. R. Green, Ins and outs of cellular cobalamin transport. Blood. 115, 1476–1477
(2010).
79. A. J. Link, J. LaBaer, Trichloroacetic Acid (TCA) Precipitation of Proteins. Cold
Spring Harbor Protocols. 2011, pdb.prot5651–pdb.prot5651 (2011).
80. D. C. Roopenian, S. Akilesh, FcRn: the neonatal Fc receptor comes of age. Nat
Rev Immunol. 7, 715–725 (2007).
81. K. E. Mostov, M. Verges, Y. Altschuler, Membrane traffic in polarized epithelial
cells. Current Opinion in Cell Biology. 12, 483–490 (2000).
82. E. J. Hughson, Endocytic pathways in polarized Caco-2 cells: identification of an
endosomal compartment accessible from both apical and basolateral surfaces. J.
Cell Biol. 110, 337–348 (1990).
83. H. D. Herce et al., Arginine-Rich Peptides Destabilize the Plasma Membrane,
Consistent with a Pore Formation Translocation Mechanism of Cell-Penetrating
Peptides. Biophysj. 97, 1917–1925 (2009).
101
84. J. L. Zaro, W.-C. Shen, Quantitative comparison of membrane transduction and
endocytosis of oligopeptides. Biochemical and Biophysical Research
Communications. 307, 241–247 (2003).
85. H. J. Ryser, I. Drummond, W. C. Shen, The cellular uptake of horseradish
peroxidase and its poly(lysine) conjugate by cultured fibroblasts is qualitatively
similar despite a 900-fold difference in rate. J. Cell. Physiol. 113, 167–178
(1982).
86. M. Ritchie, L. Tchistiakova, N. Scott, Implications of receptor-mediated
endocytosis and intracellular trafficking dynamics in the development of
antibody drug conjugates. mAbs. 5, 13–21 (2014).
87. J. Yang, H. Chen, I. R. Vlahov, J.-X. Cheng, P. S. Low, Evaluation of disulfide
reduction during receptor-mediated endocytosis by using FRET imaging. Proc.
Natl. Acad. Sci. U.S.A. 103, 13872–13877 (2006).
88. C. D. Austin et al., Oxidizing potential of endosomes and lysosomes limits
intracellular cleavage of disulfide-based antibody-drug conjugates. Proc. Natl.
Acad. Sci. U.S.A. 102, 17987–17992 (2005).
89. P. Trail, Antibody Drug Conjugates as Cancer Therapeutics. Antibodies. 2, 113–
129 (2013).
90. M. Huan et al., In Vitro and In Vivo Antitumor Activity of a Novel pH-Activated
Polymeric Drug Delivery System for Doxorubicin. PLoS ONE. 7, e44116–11
(2012).
91. S. Knecht, D. Ricklin, A. N. Eberle, B. Ernst, Oligohis-tags: mechanisms of
binding to Ni2+-NTA surfaces. J. Mol. Recognit. 22, 270–279 (2009).
92. R. K. June, K. Gogoi, A. Eguchi, X.-S. Cui, S. F. Dowdy, Synthesis of a pH-
Sensitive Nitrilotriacetic Linker to Peptide Transduction Domains To Enable
Intracellular Delivery of Histidine Imidazole Ring-Containing Macromolecules.
J. Am. Chem. Soc. 132, 10680–10682 (2010).
93. J. A. Bornhorst, J. J. Falke, in Applications of Chimeric Genes and Hybrid
Proteins Part A: Gene Expression and Protein Purification (Elsevier, 2000), vol.
326 of Methods in Enzymology, pp. 245–254.
94. E. Wang et al., Apical and basolateral endocytic pathways of MDCK cells meet
in acidic common endosomes distinct from a nearly-neutral apical recycling
endosome. Traffic. 1, 480–493 (2000).
95. M. E. Brandsma et al., Plant-derived recombinant human serum transferrin
demonstrates multiple functions. Plant Biotechnol. J. 8, 489–505 (2010).
96. N. J. Kavimandan, E. Losi, N. A. Peppas, Novel delivery system based on
102
complexation hydrogels as delivery vehicles for insulin-transferrin conjugates.
Biomaterials. 27, 3846–3854 (2006).
97. M. F. Dunn, Zinc–Ligand Interactions Modulate Assembly and Stability of the
Insulin Hexamer – A Review. Biometals. 18, 295–303 (2005).
98. K.-C. Kwon, H. Daniell, Low-cost oral delivery of protein drugs bioencapsulated
in plant cells. Plant Biotechnol. J. 13, 1017–1022 (2015).
99. K.-C. Kwon, H. Daniell, Oral Delivery of Protein Drugs Bioencapsulated in
Plant Cells. Mol Ther. 24, 1342–1350 (2016).
100. V. Bencko, Nickel: a review of its occupational and environmental toxicology. J
Hyg Epidemiol Microbiol Immunol. 27, 237–247 (1983).
101. L. Fei, L.-P. Yap, P. S. Conti, W.-C. Shen, J. L. Zaro, Tumor targeting of a cell
penetrating peptide by fusing with a pH-sensitive histidine-glutamate co-
oligopeptide. Biomaterials. 35, 4082–4087 (2014).
102. J. L. Zaro, L. Fei, W.-C. Shen, Recombinant peptide constructs for targeted cell
penetrating peptide-mediated delivery. Journal of Controlled Release. 158, 357–
361 (2012).
103. C. Sun, W.-C. Shen, J. Tu, J. L. Zaro, Interaction between Cell-Penetrating
Peptides and Acid-Sensitive Anionic Oligopeptides as a Model for the Design of
Targeted Drug Carriers. Mol. Pharmaceutics. 11, 1583–1590 (2014).
103
ALPHABETIZED REFERENCES
Amet, N., Lee, H.-F., and Shen, W.-C. (2008). Insertion of the Designed Helical Linker
Led to Increased Expression of Tf-Based Fusion Proteins. Pharm. Res. 26, 523–528.
Amet, N., Wang, W., and Shen, W.-C. (2010). Human growth hormone–transferrin
fusion protein for oral delivery in hypophysectomized rats. Journal of Controlled Release
141, 177–182.
Austin, C.D., Wen, X., Gazzard, L., Nelson, C., Scheller, R.H., and Scales, S.J. (2005).
Oxidizing potential of endosomes and lysosomes limits intracellular cleavage of
disulfide-based antibody-drug conjugates. Proc. Natl. Acad. Sci. U.S.a. 102, 17987–
17992.
Baeshen, N.A., Baeshen, M.N., Sheikh, A., Bora, R.S., Ahmed, M.M.M., Ramadan,
H.A.I., Saini, K.S., and Redwan, E.M. (2014). Cell factories for insulin production.
Microb. Cell Fact. 13, 141.
Bai, Y., Ann, D.K., and Shen, W.-C. (2005). Recombinant granulocyte colony-
stimulating factor-transferrin fusion protein as an oral myelopoietic agent. Proc. Natl.
Acad. Sci. U.S.a. 102, 7292–7296.
Baker, H.M., Anderson, B.F., and Baker, E.N. (2003). Dealing with iron: common
structural principles in proteins that transport iron and heme. Proc. Natl. Acad. Sci. U.S.a.
100, 3579–3583.
Baker, M., Reynolds, H.M., Lumicisi, B., and Bryson, C.J. (2014). Immunogenicity of
protein therapeutics: The key causes, consequences and challenges. Self/Nonself 1, 314–
322.
Bencko, V. (1983). Nickel: a review of its occupational and environmental toxicology. J
Hyg Epidemiol Microbiol Immunol 27, 237–247.
Bornhorst, J.A., and Falke, J.J. (2000). Purification of proteins using polyhistidine
affinity tags. In Applications of Chimeric Genes and Hybrid Proteins Part a: Gene
Expression and Protein Purification, (Elsevier), pp. 245–254.
Boyhan, D., and Daniell, H. (2010). Low-cost production of proinsulin in tobacco and
lettuce chloroplasts for injectable or oral delivery of functional insulin and C-peptide.
Plant Biotechnol. J. 9, 585–598.
Brandsma, M.E., Diao, H., Wang, X., Kohalmi, S.E., Jevnikar, A.M., and Ma, S. (2010).
Plant-derived recombinant human serum transferrin demonstrates multiple functions.
Plant Biotechnol. J. 8, 489–505.
104
Carter, P.J. (2011). Introduction to current and future protein therapeutics: A protein
engineering perspective. Exp. Cell Res. 317, 1261–1269.
Centers for Disease Control and Prevention (2016). National Diabetes Statistics Report,
2014. 1–12.
Chang, S.-G., Choi, K.-D., Jang, S.-H., and Shin, H.-C. (2003). Role of disulfide bonds in
the structure and activity of human insulin. Mol. Cells 16, 323–330.
Chaturvedi, K., Ganguly, K., Nadagouda, M.N., and Aminabhavi, T.M. (2013).
Polymeric hydrogels for oral insulin delivery. Journal of Controlled Release 165, 129–
138.
Chen, X., Lee, H.-F., Zaro, J.L., and Shen, W.-C. (2011). Effects of Receptor Binding on
Plasma Half-Life of Bifunctional Transferrin Fusion Proteins. Mol. Pharmaceutics 8,
457–465.
Chen, X., Zaro, J.L., and Shen, W.-C. (2012). Pharmacokinetics of recombinant
bifunctional fusion proteins. Expert Opinion on Drug Metabolism & Toxicology 8, 581–
595.
Choonara, B.F., Choonara, Y.E., Kumar, P., Bijukumar, D., Toit, du, L.C., and Pillay, V.
(2014). A review of advanced oral drug delivery technologies facilitating the protection
and absorption of protein and peptide molecules. Biotechnology Advances 32, 1269–
1282.
Craik, D.J., Fairlie, D.P., Liras, S., and Price, D. (2012). The Future of Peptide-based
Drugs. Chemical Biology & Drug Design 81, 136–147.
Czajkowsky, D.M., Hu, J., Shao, Z., and Pleass, R.J. (2012). Fc-fusion proteins: new
developments and future perspectives. EMBO Mol Med 4, 1015–1028.
Daniels, T.R., Bernabeu, E., Rodríguez, J.A., Patel, S., Kozman, M., Chiappetta, D.A.,
Holler, E., Ljubimova, J.Y., Helguera, G., and Penichet, M.L. (2012). The transferrin
receptor and the targeted delivery of therapeutic agents against cancer. Biochimica Et
Biophysica Acta (BBA) - General Subjects 1820, 291–317.
Demain, A.L., and Vaishnav, P. (2009). Production of recombinant proteins by microbes
and higher organisms. Biotechnology Advances 27, 297–306.
Duckworth, W.C., Bennett, R.G., and Hamel, F.G. (1998). Insulin Degradation: Progress
and Potential. Endocrine Reviews 19, 608–624.
Dunn, M.F. (2005). Zinc–Ligand Interactions Modulate Assembly and Stability of the
Insulin Hexamer – A Review. Biometals 18, 295–303.
105
Fei, L., Yap, L.-P., Conti, P.S., Shen, W.-C., and Zaro, J.L. (2014). Tumor targeting of a
cell penetrating peptide by fusing with a pH-sensitive histidine-glutamate co-
oligopeptide. Biomaterials 35, 4082–4087.
Fonte, P., Araujo, F., Reis, S., and Sarmento, B. (2013). Oral Insulin Delivery: How Far
are We? J Diabetes Sci Technol 7, 520–531.
Frokjaer, S., and Otzen, D.E. (2005). Protein drug stability: a formulation challenge. Nat
Rev Drug Discov 4, 298–306.
Goeddel, D.V., Kleid, D.G., Bolivar, F., Heyneker, H.L., Yansura, D.G., Crea, R.,
Hirose, T., Kraszewski, A., Itakura, K., and Riggs, A.D. (1979). Expression in
Escherichia coli of chemically synthesized genes for human insulin. Proc. Natl. Acad.
Sci. U.S.a. 76, 106–110.
Goldberg, M., and Gomez-Orellana, I. (2003). Challenges for the oral delivery of
macromolecules. Nat Rev Drug Discov 2, 289–295.
Green, R. (2010). Ins and outs of cellular cobalamin transport. Blood 115, 1476–1477.
Haataja, L., Manickam, N., Soliman, A., Tsai, B., Liu, M., and Arvan, P. (2016).
Disulfide Mispairing During Proinsulin Folding in the Endoplasmic Reticulum. Diabetes
65, 1050–1060.
Herce, H.D., Garcia, A.E., Litt, J., Kane, R.S., Martin, P., Enrique, N., Rebolledo, A., and
Milesi, V. (2009). Arginine-Rich Peptides Destabilize the Plasma Membrane, Consistent
with a Pore Formation Translocation Mechanism of Cell-Penetrating Peptides. Biophysj
97, 1917–1925.
Hoelen, H., Zaldumbide, A., van Leeuwen, W.F., Torfs, E.C.W., Engelse, M.A., Hassan,
C., Lebbink, R.J., de Koning, E.J., Resssing, M.E., de Ru, A.H., et al. (2015).
Proteasomal Degradation of Proinsulin Requires Derlin-2, HRD1 and p97. PLoS ONE
10, e0128206–e0128216.
Home, P.D., and Ashwell, S.G. (2002). An overview of insulin glargine. Diabetes Metab
Res Rev 18, S57–S63.
Huan, M., Zhang, B., Teng, Z., Cui, H., Wang, J., Liu, X., Xia, H., Zhou, S., and Mei, Q.
(2012). In Vitro and In Vivo Antitumor Activity of a Novel pH-Activated Polymeric Drug
Delivery System for Doxorubicin. PLoS ONE 7, e44116–11.
Hughson, E.J. (1990). Endocytic pathways in polarized Caco-2 cells: identification of an
endosomal compartment accessible from both apical and basolateral surfaces. J. Cell
Biol. 110, 337–348.
Johansen, F.-E., and Kaetzel, C.S. (2011). Regulation of the polymeric immunoglobulin
106
receptor and IgA transport: new advances in environmental factors that stimulate pIgR
expression and its role in mucosal immunity. Mucosal Immunology 4, 598–602.
Johnson, I.S. (2003). The trials and tribulations of producing the first genetically
engineered drug. Nat Rev Drug Discov 2, 747–751.
June, R.K., Gogoi, K., Eguchi, A., Cui, X.-S., and Dowdy, S.F. (2010). Synthesis of a
pH-Sensitive Nitrilotriacetic Linker to Peptide Transduction Domains To Enable
Intracellular Delivery of Histidine Imidazole Ring-Containing Macromolecules. J. Am.
Chem. Soc. 132, 10680–10682.
Kalra, S., Kalra, B., and Agrawal, N. (2010). Oral insulin. Diabetol Metab Syndr 2, 66.
Kavimandan, N.J., Losi, E., and Peppas, N.A. (2006). Novel delivery system based on
complexation hydrogels as delivery vehicles for insulin-transferrin conjugates.
Biomaterials 27, 3846–3854.
Khan, K.H. (2013). Gene expression in Mammalian cells and its applications. Adv Pharm
Bull 3, 257–263.
Kim, B.J., Zhou, J., Martin, B., Carlson, O.D., Maudsley, S., Greig, N.H., Mattson, M.P.,
Ladenheim, E.E., Wustner, J., Turner, A., et al. (2010). Transferrin Fusion Technology:
A Novel Approach to Prolonging Biological Half-Life of Insulinotropic Peptides. Journal
of Pharmacology and Experimental Therapeutics 334, 682–692.
Kinch, M.S. (2015). An overview of FDA-approved biologics medicines. Drug
Discovery Today 20, 393–398.
Knecht, S., Ricklin, D., Eberle, A.N., and Ernst, B. (2009). Oligohis-tags: mechanisms of
binding to Ni2+-NTA surfaces. J. Mol. Recognit. 22, 270–279.
Kompella, U.B., and Lee, V.H. (2001). Delivery systems for penetration enhancement of
peptide and protein drugs: design considerations. Advanced Drug Delivery Reviews 46,
211–245.
Kristensen, M., and Nielsen, H.M. (2015). Cell-Penetrating Peptides as Carriers for Oral
Delivery of Biopharmaceuticals. Basic Clin Pharmacol Toxicol 118, 99–106.
Kwon, K.-C., and Daniell, H. (2015). Low-cost oral delivery of protein drugs
bioencapsulated in plant cells. Plant Biotechnol. J. 13, 1017–1022.
Kwon, K.-C., and Daniell, H. (2016). Oral Delivery of Protein Drugs Bioencapsulated in
Plant Cells. Mol Ther 24, 1342–1350.
Leader, B., Baca, Q.J., and Golan, D.E. (2008). Protein therapeutics: a summary and
pharmacological classification. Nat Rev Drug Discov 7, 21–39.
107
Lencer, W.I., Moe, S., Rufo, P.A., and Madara, J.L. (1995). Transcytosis of cholera toxin
subunits across model human intestinal epithelia. Proc. Natl. Acad. Sci. U.S.a. 92,
10094–10098.
Lim, C.-J., and Shen, W.-C. (2005). Comparison of monomeric and oligomeric
transferrin as potential carrier in oral delivery of protein drugs. Journal of Controlled
Release 106, 273–286.
Link, A.J., and LaBaer, J. (2011). Trichloroacetic Acid (TCA) Precipitation of Proteins.
Cold Spring Harbor Protocols 2011, pdb.prot5651–pdb.prot5651.
Loh, J.W., Schneider, J., Carter, M., Saunders, M., and Lim, L.-Y. (2010). Spinning Disc
Processing Technology: Potential for Large-Scale Manufacture of Chitosan
Nanoparticles. J Pharm Sci 99, 4326–4336.
Ma, J.K.-C., Drake, P.M.W., and Christou, P. (2003). The production of recombinant
pharmaceutical proteins in plants. Nat. Rev. Genet. 4, 794–805.
Mahmood, I., and Green, M.D. (2005). Pharmacokinetic and pharmacodynamic
considerations in the development of therapeutic proteins. Clin Pharmacokinet 44, 331–
347.
Melanie, B. (2005). Transferrin' the load. Nat Rev Drug Discov 4, 537–537.
Mizutani, K., Toyoda, M., and Mikami, B. (2012). X-ray structures of transferrins and
related proteins. BBA - General Subjects 1820, 203–211.
Moeller, E.H., and Jorgensen, L. (2008). Alternative routes of administration for systemic
delivery of protein pharmaceuticals. Drug Discovery Today: Technologies 5, e89–e94.
Mostov, K.E., Verges, M., and Altschuler, Y. (2000). Membrane traffic in polarized
epithelial cells. Current Opinion in Cell Biology 12, 483–490.
Muheem, A., Shakeel, F., Jahangir, M.A., Anwar, M., Mallick, N., Jain, G.K., Warsi,
M.H., and Ahmad, F.J. (2016). A review on the strategies for oral delivery of proteins
and peptides and their clinical perspectives. Saudi Pharmaceutical Journal 24, 413–428.
Nielsen, J. (2014). Production of biopharmaceutical proteins by yeast. Bioengineered 4,
207–211.
Noinaj, N., Easley, N.C., Oke, M., Mizuno, N., Gumbart, J., Boura, E., Steere, A.N., Zak,
O., Aisen, P., Tajkhorshid, E., et al. (2012). Structural basis for iron piracy by pathogenic
Neisseria. Nature 482, 53–58.
Nykiforuk, C.L., Boothe, J.G., Murray, E.W., Keon, R.G., Goren, H.J., Markley, N.A.,
and Moloney, M.M. (2006). Transgenic expression and recovery of biologically active
108
recombinant human insulin from Arabidopsis thaliana seeds. Plant Biotechnol. J. 4, 77–
85.
Oyadomari, S., Koizumi, A., Takeda, K., Gotoh, T., Akira, S., Araki, E., and Mori, M.
(2002). Targeted disruption of the Chop gene delays endoplasmic reticulum stress-
mediated diabetes. J. Clin. Invest. 109, 525–532.
Park, K., Kwon, I.C., and Park, K. (2011). Oral protein delivery: Current status and future
prospect. Reactive and Functional Polymers 71, 280–287.
Peavy, D.E., Abram, J.D., Frank, B.H., and Duckworth, W.C. (1984). In vitro activity of
biosynthetic human proinsulin. Receptor binding and biologic potency of proinsulin and
insulin in isolated rat adipocytes. Diabetes 33, 1062–1067.
Peterson, M.W., and Gruenhaupt, D. (1990). Protamine increases the permeability of
cultured epithelial monolayers. J. Appl. Physiol. 68, 220–227.
Petrus, A.K., Vortherms, A.R., Fairchild, T.J., and Doyle, R.P. (2007). Vitamin B12 as a
Carrier for the Oral Delivery of Insulin. ChemMedChem 2, 1717–1721.
Qian, Z.M., Li, H., Sun, H., and Ho, K. (2002). Targeted drug delivery via the transferrin
receptor-mediated endocytosis pathway. Pharmacol. Rev. 54, 561–587.
Ramanujam, K.S., Seetharam, S., Ramasamy, M., and Seetharam, B. (1991). Expression
of cobalamin transport proteins and cobalamin transcytosis by colon adenocarcinoma
cells. Am. J. Physiol. 260, G416–G422.
Ritchie, M., Tchistiakova, L., and Scott, N. (2014). Implications of receptor-mediated
endocytosis and intracellular trafficking dynamics in the development of antibody drug
conjugates. mAbs 5, 13–21.
Roopenian, D.C., and Akilesh, S. (2007). FcRn: the neonatal Fc receptor comes of age.
Nat Rev Immunol 7, 715–725.
Russell-Jones, D., and Khan, R. (2007). Insulin-associated weight gain in diabetes –
causes, effects and coping strategies. Diabetes Obes Metab 9, 799–812.
Ryser, H.J., Drummond, I., and Shen, W.C. (1982). The cellular uptake of horseradish
peroxidase and its poly(lysine) conjugate by cultured fibroblasts is qualitatively similar
despite a 900-fold difference in rate. J. Cell. Physiol. 113, 167–178.
Seetharam, B., and Yammani, R.R. (2003). Cobalamin transport proteins and their cell-
surface receptors. Erm 5, 1–18.
Shah, D., and Shen, W.-C. (1994). The Establishment of Polarity and Enhanced
Transcytosis of Transferrin Receptors in Enterocyte-like Caco-2 Cells. Journal of Drug
109
Targeting 2, 93–99.
Shah, D., and Shen, W.-C. (1996). Transcellular Delivery of an Insulin-Transferrin
Conjugate in Enterocyte-like Caco-2 Cells. J Pharm Sci 85, 1306–1311.
Shao, J., Zaro, J.L., and Shen, W.-C. (2016). Proinsulin–Transferrin Fusion Protein
Exhibits a Prolonged and Selective Effect on the Control of Hepatic Glucose Production
in an Experimental Model of Type 1 Diabetes. Mol. Pharmaceutics 13, 2641–2646.
Sonaje, K., Chuang, E.-Y., Lin, K.-J., Yen, T.-C., Su, F.-Y., Tseng, M.T., and Sung, H.-
W. (2012). Opening of Epithelial Tight Junctions and Enhancement of Paracellular
Permeation by Chitosan: Microscopic, Ultrastructural, and Computed-Tomographic
Observations. Mol. Pharmaceutics 120413125623001–120413125623009.
Sonoda, S., and Schlamowitz, M. (1970). Studies of 125I trace labeling of
immunoglobulin G by chloramine-T. Immunochemistry 7, 885–898.
Steere, A.N., Bobst, C.E., Zhang, D., Pettit, S.C., Kaltashov, I.A., Huang, N., and Mason,
A.B. (2012). Biochemical and structural characterization of recombinant human serum
transferrin from rice (Oryza sativa L.). Journal of Inorganic Biochemistry 116, 37–44.
Sun, C., Shen, W.-C., Tu, J., and Zaro, J.L. (2014). Interaction between Cell-Penetrating
Peptides and Acid-Sensitive Anionic Oligopeptides as a Model for the Design of
Targeted Drug Carriers. Mol. Pharmaceutics 11, 1583–1590.
Trail, P. (2013). Antibody Drug Conjugates as Cancer Therapeutics. Antibodies 2, 113–
129.
Tréhin, R., Nielsen, H.M., Jahnke, H.-G., Krauss, U., Beck-Sickinger, A.G., and Merkle,
H.P. (2004). Metabolic cleavage of cell-penetrating peptides in contact with epithelial
models: human calcitonin (hCT)-derived peptides, Tat(47-57) and penetratin(43-58).
Biochem. J. 382, 945–956.
Tuma, P.L., and Hubbard, A.L. (2003). Transcytosis: crossing cellular barriers.
Physiological Reviews 83, 871–932.
Verma, A., Sharma, S., Gupta, P.K., Singh, A., Teja, B.V., Dwivedi, P., Gupta, G.K.,
Trivedi, R., and Mishra, P.R. (2016). Vitamin B12 functionalized layer by layer calcium
phosphate nanoparticles: A mucoadhesive and pH responsive carrier for improved oral
delivery of insulin. Acta Biomaterialia 31, 288–300.
Wang, E., Brown, P.S., Aroeti, B., Chapin, S.J., Mostov, K.E., and Dunn, K.W. (2000).
Apical and basolateral endocytic pathways of MDCK cells meet in acidic common
endosomes distinct from a nearly-neutral apical recycling endosome. Traffic 1, 480–493.
Wang, Y., Chen, Y.-S., Zaro, J.L., and Shen, W.-C. (2011). Receptor-mediated activation
110
of a proinsulin-transferrin fusion protein in hepatoma cells. Journal of Controlled Release
155, 386–392.
Wang, Y., Shao, J., Zaro, J.L., and Shen, W.-C. (2014). Proinsulin-transferrin fusion
protein as a novel long-acting insulin analog for the inhibition of hepatic glucose
production. Diabetes 63, 1779–1788.
Wilcox, G. (2005). Insulin and insulin resistance. Clin Biochem Rev 26, 19–39.
Wurm, F.M. (2004). Production of recombinant protein therapeutics in cultivated
mammalian cells. Nat Biotechnol. 22, 1393–1398.
Xia, C.Q., Wang, J., and Shen, W.C. (2000). Hypoglycemic effect of insulin-transferrin
conjugate in streptozotocin-induced diabetic rats. Journal of Pharmacology and
Experimental Therapeutics 295, 594–600.
Yang, J., Chen, H., Vlahov, I.R., Cheng, J.-X., and Low, P.S. (2006). Evaluation of
disulfide reduction during receptor-mediated endocytosis by using FRET imaging. Proc.
Natl. Acad. Sci. U.S.a. 103, 13872–13877.
Zaro, J.L., and Shen, W.-C. (2003). Quantitative comparison of membrane transduction
and endocytosis of oligopeptides. Biochemical and Biophysical Research
Communications 307, 241–247.
Zaro, J.L., Fei, L., and Shen, W.-C. (2012). Recombinant peptide constructs for targeted
cell penetrating peptide-mediated delivery. Journal of Controlled Release 158, 357–361.
Zhang, D., Lee, H.-F., Pettit, S.C., Zaro, J.L., Huang, N., and Shen, W.-C. (2012).
Characterization of transferrin receptor-mediated endocytosis and cellular iron delivery
of recombinant human serum transferrin from rice (Oryza sativa L.). BMC
Biotechnology 12, 1–1.
Zhang, D., Nandi, S., Bryan, P., Pettit, S., Nguyen, D., Santos, M.A., and Huang, N.
(2010). Expression, purification, and characterization of recombinant human transferrin
from rice (Oryza sativa L.). Protein Expression and Purification 74, 69–79.
Zijlstra, E., Heinemann, L., and Plum-Mörschel, L. (2014). Oral insulin reloaded: a
structured approach. J Diabetes Sci Technol 8, 458–465.
Abstract (if available)
Abstract
Since proinsulin-transferrin (ProINS-Tf) fusion protein was designed and expressed from the mammalian HEK293 cells (HEK-ProINS-Tf), many interesting and attractive features have been discovered. First, it has been found that ProINS-Tf could be converted into an activated form when incubated with some specific cell lines or tissue cultures through TfR-mediated endocytosis and recycling pathway. Second, in the previous in vivo studies on streptozotocin-induced type-1 diabetic mice via subcutaneous injections, ProINS-Tf has been proven as an extra-long acting insulin analogue with liver-specific insulin action. However, there are still some other interesting features that have not yet been fully explored and characterized. One of the ultimate goals of this project is to develop ProINS-Tf into an oral deliverable insulin analogue. However, to achieve this goal, two of the major challenges, including its low production yield from HEK293 cells and the limited transcytosis rate of Tf carrier across the intestine, need to be overcome. ❧ First, to improve the protein productivity, an alternative protein expression system, transgenic rice (Oryza sativa L.), was applied. The results indicated the transgenic rice successfully expressed the intact and active rice-derived ProINS-Tf (Rice-ProINS-Tf). Although the insulin-like bioactivity of Rice-ProINS-Tf was slightly lower in vitro, its potency of in vivo blood glucose control was considerably stronger than that produced from HEK293 cells (HEK-ProINS-Tf). Overall, it was demonstrated that rice-ProINS-Tf exhibited a great similarity compared with the HEK-ProINS-Tf. Our oral delivery studies in type 1 diabetic mice further demonstrated the oral hypoglycemic effect of Rice-ProINS-Tf with oral bioactivity ≈ 1%. ❧ On the other hand, the transcytosis rate of Tf-carried fusion proteins across intestinal epithelium is limited by the polarized distribution of TfR on the basolateral cell membrane. Therefore, to increase the transcytosis rate, a novel transcytosis mechanism of Tf across intestinal epithelium was proposed. In this subproject, Tf was conjugated with positively charged poly-L-lysine (PLL) through a pH sensitive nitrilotriacetic acid (NTA) linker. It was demonstrated PLL could increase the apical binding and internalization of Tf. Subsequently, with the endosomal dissociation of the NTA linker, Tf can be exocytosed using the basolaterally recycled TfR. ❧ In this dissertation, the two major challenges of oral delivery of ProINS-Tf were addressed and two separate approaches have been proposed and investigated to solve these issues. In the future, we expect the findings from these two strategies could be combined to develop ProINS-Tf as a low-cost, long-acting, and liver-targeting basal insulin analogue with higher oral therapeutic effect for diabetic patients. Additionally, the findings from this dissertation could be further applied to any other Tf-based fusion proteins to increase their productivity as well as the oral bioavailability.
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
Proinsulin-transferrin recombinant fusion protein: mechanism of activation and potential application in diabetes treatment
PDF
Reconstruction of a proinsulin-transferrin fusion protein
PDF
Human growth hormone-transferrin recombinant fusion protein for oral delivery
PDF
Proinsulin-transferrin fusion protein as a liver targeting insulin therapeutic in Type 1 diabetes treatment
PDF
Characterization of the transferrin oligomer and its potential application in drug delivery
PDF
Preparation and characterization of Tf-G-CSF fusion protein
PDF
The intracellular trafficking of transferrin in Caco-2 cells and its implication to transferrin-mediated oral drug delivery
PDF
Developing recombinant single chain Fc-dimer fusion proteins for improved protein drug delivery
PDF
Cationic cell penetrating peptides: characterization of transport properties in epithelial cells and their utilization as delivery systems for protein and peptide drugs
PDF
Characterization and identification of endogenous factor(s) that enhance insulin transport across primary rat alveolar epithelial cell monolayers
PDF
Characterization of actin based motility in mammalian cells through LIM and SH3 domain protein 1 (LASP1) and elastin like polypeptide (ELP) fusion protein
Asset Metadata
Creator
Chen, Yu-Sheng
(author)
Core Title
Characterization of rice-derived proinsulin-transferrin fusion protein and its oral bioactivity as a basal insulin analogue
School
School of Pharmacy
Degree
Doctor of Philosophy
Degree Program
Pharmaceutical Sciences
Publication Date
11/14/2018
Defense Date
10/24/2016
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
OAI-PMH Harvest,oral insulin,oral protein delivery,proinsulin-transferrin fusion protein,transcytosis,transgenic rice
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Shen, Wei-Chiang (
committee chair
), Mircheff, Austin (
committee member
), Okamoto, Curtis (
committee member
)
Creator Email
cys840227@gmail.com,yushengc@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-672149
Unique identifier
UC11336949
Identifier
etd-ChenYuShen-4923.pdf (filename),usctheses-c16-672149 (legacy record id)
Legacy Identifier
etd-ChenYuShen-4923-0.pdf
Dmrecord
672149
Document Type
Dissertation
Rights
Chen, Yu-Sheng
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
Tags
oral insulin
oral protein delivery
proinsulin-transferrin fusion protein
transcytosis
transgenic rice