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 the transferrin oligomer and its potential application in drug delivery
(USC Thesis Other)
Characterization of the transferrin oligomer and its potential application in drug delivery
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
CHARACTERIZATION OF THE TRANSFERRIN OLIGOMER AND ITS
POTENTIAL APPLICATION IN DRUG DELIVERY
By
Hsuan-Yao Wang
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PHARMACEUTICAL SCIENCES)
May 2019
Copyright 2019 Hsuan-Yao Wang
I
DEDICATION
To my grandmother, Wu-Kun Wang, for her unconditional love
II
ACKNOWLEDGEMENT
During my graduate study at School of Pharmacy, University of Southern
California, I received a lot of supports from many people in the department. I’m grateful
that I can meet these people in my lifetime, especially my advisor Dr. Wei-Chiang Shen.
He is dedicated to science, and is always passionate about the new findings or
technologies that could possibly be applied to treat diseases. He is very patient in
teaching students, which allows me to learn a lot of scientific knowledges in my
graduate study. He is also kind to people, which makes every student feel comfortable
in the lab. For me, he is not only my advisor, but also my friend and family member. He
is a raw model for all the young scientists, including me, and here I want to express my
gratitude for all the years of mentorship from him during my graduate study.
I would like to acknowledge my committee members, Dr. Curtis Okamoto, Dr.
Yong (Tiger) Zhang, Dr. Jennica Zaro, and Dr. Sarah Hamm-Alvarez, for their enormous
support, and the important scientific suggestions in qualifying exam or annual student
seminar meetings. Their advice and critique inspire me to see things from various
angles, and to pursue the knowledges in other scientific fields. I really appreciate their
support and mentor during my graduate study. I would also like to thank two core
facilities, USC NanoBiophysical Center and USC School of Pharmacy Translational
Research Lab, for their great technical supports, so that I can complete several
important studies in the dissertation. I would also like to express my gratitude to Dr.
Michael Hsiao and his lab members at Academia Sinica, Taipei, Taiwan, for their
generosity to form the collaboration and to provide the animal model for us to complete
my dissertation. I deeply appreciate it.
III
Furthermore, I really appreciate all the supports from Dr. Shen’s lab. I want to
thank an alumnus from Dr. Shen’s lab, Dr. Jun Wu at City of Hope, for his support in a
preliminary study for my dissertation project. I would also like to thank Daisy Shen for
her caring, so that I can feel the warmth in my graduate study. I also want to thank Hsin-
Fang Lee and Dr. Likun Fei for training me the techniques when I entered the lab. I am
grateful to have great peers in the lab when I joined the school, Dr. Zoe Folchman-
Wagner, Dr. Li Zhou, and Dr. Yuqian Liu. We together had a great time, and achieved
serval accomplishments in the graduate school. I will cherish our friendship forever. I
also want to thank Dr. Chunmeng Sun and Dr. Tzyy-Harn (Mindy) Yeh for their
supports, not only in my projects, but also in my life. In addition, I want to thank Yang Su
and Lina He for all their efforts, allowing me to have more time to focus on the research
projects.
Moreover, I want to thank my friends: Dr. Willy Tsai, Dr. Ingrid Lua, Justin Tang,
Dr. Wan Ting Chen, Jeff Kuo, Tina Hsu, Stephanie Tsai, and Pangyu Chen, my family
members at Orange County: Jerry Chen, Frenda Yang, and Max Chen, for all their
supports throughout my life in the US. I also want to thank my parents and my brother in
Taiwan for their support and understanding in my lifetime. Lastly, I am really grateful to
meet my wife, Xianhui (Sharon) Chen, in the program. It is her to make me believe that
there is always a silver lining behind the cloud, and I am the luckiest man to have her by
my side. I couldn’t reach this far without all the helps from all the people either inside or
outside of the program, and I will always remember the journey I have been through
during the days at USC
IV
TABLE OF CONTENTS
DEDICATION································ ································ ··········· I
ACKNOWLEDGEMENT ································ ····························· II
LIST OF FIGURES AND TABLES ································ ··············· IX
ABBREVIATIONS ································ ································ ···· XII
ABSTRACT ································ ································ ············· 1
CHAPTER 1. INTRODUCTION
1.1 The application of serum proteins in drug delivery and development ······· 2
1.2 Transferrin, a serum glycoprotein ································ ························· 3
1.2.1 Structure and function ································ ································ ········ 3
1.2.2 Transferrin and transferrin receptor-mediated endocytosis ························ 4
1.2.3 Transferrin receptor-mediated transcytosis ································ ············ 5
1.3 The application of Tf/TfR system in drug delivery and development ········· 7
1.3.1 Tumor targeting ································ ································ ················ 7
1.3.2 Half-life extension ································ ································ ············· 8
1.3.3 Delivery of protein drugs across gastrointestinal epithelium and blood
brain barrier ································ ································ ····················· 8
1.3.4 Protein prodrug activation ································ ································ ··· 10
1.4 Challenges of targeting Tf/TfR system in drug development ···················· 10
1.4.1 ·· Potential off-target effect ································ ································ ···· 10
1.4.2 Binding competition from the endogenous Tf ································ ·········· 11
1.4.3 Low transcytosis rate across the BBB ································ ··················· 11
1.4.4 Rapid recycling after endocytosis ································ ························· 12
V
CHAPTER 2: SYNTHESIS AND PHYSICOCHEMICAL PROPERTY
CHARACTERIZATION OF TF OLIGOMER
2.1 Background ································ ································ ························ 14
2.2 Materials and methods ································ ································ ········ 15
2.2.1 Preparation of NonR-AggTf and R-AggTf ································ ·············· 15
2.2.1.1 Iron-saturation of ApoTf ································ ··················· 15
2.2.1.2 SPDP conjugation on Tf ································ ··················· 15
2.2.1.3 Synthesis of NonR-AggTf ································ ················· 15
2.2.1.4 Synthesis of R-AggTf ································ ······················ 16
2.2.1.5 Purification of NonR-AggTf and R-AggTf ····························· 16
2.2.2 Dynamic light scattering analysis ································ ························· 17
2.2.3 Stability analysis ································ ································ ··············· 17
2.2.4 SEC-MALS analysis ································ ································ ·········· 17
2.2.5 Biosensor analysis ································ ································ ············ 18
2.3 Results ································ ································ ······························ 20
2.3.1 SDS-PAGE and Coomassie blue staining of NonR-AggTf and R-AggTf ······· 20
2.3.2 DLS analysis of MonoTf, NonR-AggTf, and R-AggTf ································ 20
2.3.3 SEC-MALS analysis of NonR-AggTf and the R-AggTf ······························ 20
2.3.4 Stability test ································ ································ ····················· 21
2.3.5 Binding affinity analysis of NonR-AggTf and R-AggTf to TfR by SPR ·········· 21
2.4 Figures ································ ································ ······························ 22
2.5 Discussion ································ ································ ························· 27
2.6 Summary ································ ································ ··························· 29
VI
CHAPTER 3: EVALUATION OF THE POTENTIAL APPLICATION OF TF
OLIGOMER IN TUMOR TARGETING
3.1 Background ································ ································ ························ 30
3.2 Materials and methods ································ ································ ········ 32
3.2.1 Cell culture ································ ································ ······················ 32
3.2.2 Protein labeling ································ ································ ················ 32
3.2.3 Pulse-chase assay ································ ································ ············ 33
3.2.4 TfR competitive binding assay ································ ····························· 34
3.2.5 Cellular uptake assay ································ ································ ········ 35
3.2.6 Animal ································ ································ ···························· 35
3.2.7 Pharmacokinetics study ································ ································ ····· 36
3.2.8 Biodistribution study ································ ································ ·········· 37
3.2.9 Tumor targeting and retention study ································ ····················· 37
3.2.10 Tumor targeting studies of NIR-NonR-AggTf with excess Tf blockage ········· 38
3.3 Results
3.3.1 Cellular retention of NonR-AggTf and R-AggTf in MDA-MB-231 cells ·········· 39
3.3.2 Binding profile of MonoTf, NonR-AggTf, and R-AggTf in MDA-MB-231 cells · 40
3.3.3 Cellular uptake of NonR-AggTf and R-AggTf in MDA-MB-231 cells ············· 40
3.3.4 Pharmacokinetics parameters of NonR-AggTf and R-AggTf ······················ 41
3.3.5 Biodistribution comparison between MonoTf and NonR-AggTf ··················· 42
3.3.6 Tumor targeting and retention effect of MonoTf and NonR-AggTf ··············· 43
3.3.7 Tumor targeting studies of NIR-NonR-AggTf with excess Tf blockage ········· 44
3.4 Figures and Tables ································ ································ ·············· 46
3.5 Discussion ································ ································ ························· 56
3.6 Summary ································ ································ ··························· 61
VII
CHAPTER 4: EVALUATION OF THE POTENTIAL APPLICATION OF
THE TF OLIGOMER IN TRANS-EPITHELIAL PROTEIN DRUG
DELIVERY
4.1 Background ································ ································ ························ 63
4.2 Material and methods ································ ································ ·········· 67
4.2.1 Cell culture ································ ································ ······················ 67
4.2.2 Oligonucleotide primers ································ ································ ····· 67
4.2.3 Molecular cloning ································ ································ ·············· 68
4.2.4 Recombinant protein production and preparation ································ ···· 70
4.2.5 Western blotting ································ ································ ··············· 71
4.2.6 Caco-2 cell-based pulse-chase and transcytosis assay ···························· 72
4.2.7 Nb2 cell-based proliferation assay ································ ························ 72
4.2.8 Pharmacokinetics of MonoTf and NonR-AggTf with oral administration ······· 73
4.3 Results ································ ································ ······························ 75
4.3.1 Transcytosis capability of NonR-AggTf and R-AggTf ································ 75
4.3.2 Comparison of the serum concentration of MonoTf and NonR-AggTf after
oral administration ································ ································ ············ 75
4.3.3 Cloning of hGH-C2-Tf, hGH-CH4C-Tf, hGH-hinge-Tf, and hGH-Lzip-Tf
expressing plasmids ································ ································ ·········· 76
4.3.4 Protein production and Western blotting analysis of recombinant hGH-Tf
fusion protein with various linkers ································ ························· 78
4.3.5 Biological activity evaluation of the hGH-Lzip-Tf fusion protein on Nb2 cells · 78
4.3.6 Comparison of the cellular retention and the transcytosis capability between
hGH-H4-Tf and hGH-Lzip-Tf in polarized Caco-2 cells ······························ 79
4.4 Figures and Table ································ ································ ··············· 81
4.5 Discussion ································ ································ ························· 93
4.6 Summary ································ ································ ··························· 97
VIII
CHAPTER 5: SUMMARY
5.1 Conclusion ································ ································ ························· 98
5.2 Summary graphics ································ ································ ·············· 100
5.3 Future Perspectives ································ ································ ············ 103
5.3.1 Optimization of the chemically-synthesized Tf oligomer ···························· 103
5.3.2 Exploration of the further application of leucine zipper dimerization linker ····· 103
5.3.3 Production of the recombinant Tf oligomer using other polymerization motifs 104
REFERENCES ································ ································ ········· 106
IX
LIST OF FIGURES AND TABLES
Figures
Figure 1.1: TfR-mediated endocytosis and recycling ································ ············ 5
Figure 1.2: The proposed model for TfR-mediated transcytosis in polarized GI
epithelium································ ································ ································ ···· 6
Figure 2.2.1: Schematic illustration of NonR-AggTf synthesis ································ 19
Figure 2.2.2: Schematic illustration of R-AggTf synthesis ································ ······ 19
Figure 2.4.1: SDS-PAGE and Coomassie blue analysis of NonR-AggTf and R-AggTf 22
Figure 2.4.2: DLS analysis of MonoTf, NonR-AggTf, and R-AggTf ·························· 23
Figure 2.4.3: SEC-MALS analysis of NonR-AggTf and R-AggTf ····························· 24
Figure 2.4.4: Stability test of NonR-AggTf in PBS at 37 ℃ ································ ····· 25
Figure 2.4.5: Binding affinity analysis of NonR-AggTf and R-AggTf to TfR by SPR ····· 26
Figure 3.4.1: Cellular Retention of NonR-AggTf and R-AggTf in MDA-MB-231 Cells ·· 46
Figure 3.4.2: Binding profile of MonoTf, NonR-AggTf, and R-AggTf to the TfR in MDA-
MB-231 cells ································ ································ ································ 47
Figure 3.4.3: Cellular uptake of MonoTf, NonR-AggTf and R-AggTf in MDA-MB-231
cells ································ ································ ································ ··········· 48
Figure 3.4.4: Pharmacokinetics study of MonoTf, NonR-AggTf, and NonR-AggTf ······ 49
Figure 3.4.5: Biodistribution study of MonoTf and NonR-AggTf ······························ 51
Figure 3.4.6: Tumor targeting and retention study of MonoTf and NonR-AggTf in MDA-
MB-231 xenograft mice model ································ ································ ········· 54
Figure 3.4.7: Biodistribution and tumor targeting study of NonR-AggTf in MDA-MB-231
xenograft mice model with excess Tf blockage ································ ··················· 55
X
Figure 4.4.1 Cellular retention of NonR-AggTf and R-AggTf in polarized Caco-2 cells 81
Figure 4.4.2: Transcytosis assay of MonoTf, NonR-AggTf and R-AggTf in polarized
Caco-2 cells ································ ································ ································ · 82
Figure 4.4.3: Serum concentration of MonoTf and NonR-AggTf in CF-1 wild type mice
after oral administration ································ ································ ·················· 83
Figure 4.4.4: Molecular cloning strategy for hGH-C2-Tf, hGH-CH4C-Tf, hGH hinge-Tf,
and hGH-Lzip-Tf expressing plasmids ································ ······························· 84
Figure 4.4.5: PCR confirmation of the potential clones for the hGH-C2-Tf expressing
plasmid after ligation and transformation ································ ··························· 85
Figure 4.4.6: PCR confirmation of the potential clones for the hGH-hinge-Tf expressing
plasmid after ligation and transformation ································ ··························· 85
Figure 4.4.7: PCR confirmation of the potential clones for the hGH-CH4C-Tf expressing
plasmid after ligation and transformation ································ ··························· 86
Figure 4.4.8: PCR confirmation of the potential clones for the hGH-Lzip-Tf expressing
plasmid after ligation and transformation ································ ··························· 86
Figure 4.4.9: The sequence data of hGH-Lzip-Tf expressing plasmid ······················ 87
Figure 4.4.10: Design of the linkers for hGH-Tf fusion proteins ······························· 88
Figure 4.4.11: Western blotting of the hGH-Tf fusion proteins with various linker ······· 89
Figure 4.4.12: Confirmation of the bioactivity of hGH moiety on hGH-Lzip-Tf fusion
protein ································ ································ ································ ········ 90
Figure 4.4.13: Cellular retention of hGH-H4-Tf and hGH-Lzip-Tf in polarized
Caco-2 cells ································ ································ ································ · 91
Figure 4.4.14: Transcytosis assay of hGH-H4-Tf and hGH-Lzip-Tf in polarized Caco-2
cells ································ ································ ································ ··········· 92
XI
Figure 5.2.1: The proposed mechanism of prolonged retention for chemically-
synthesized Tf oligomer in tumor cells ································ ····························· 100
Figure 5.2.2: Enhance protein degradation for chemically-synthesized Tf oligomer
during transcytosis process in polarized GI epithelium ································ ········ 101
Figure 5.2.3: Transcytosis of dimeric Tf-based fusion protein in polarized
Caco-2 cells ································ ································ ······························· 102
Tables
Table 3.4.1: PK parameters of MonoTf, NonR-AggTf, and R-AggTf by two-
compartmental and non-compartmental analysis ································ ················· 50
Table 3.4.2: Dunnett’s multiple comparison test for the biodistribution of
125
I-MonoTf · 52
Table 3.4.3: Dunnett’s multiple comparison test for the biodistribution of
125
I-NonR-
AggTf ································ ································ ································ ········· 53
Table 4.4.1: Description and the amino acid sequence for recombinant protein linkers
in various hGH-Tf fusion proteins ································ ································ ····· 88
XII
ABBREVIATIONS
2CA Two-compartmental analysis
8D3 A rat monoclonal antibody recognizing mouse TfR
AE Apical endosome
ApoTf Iron-free transferrin
BBB Blood brain barrier
BE Basolateral endosome
BioD Biodistribution
BM(PEG)3 1,11-bis(maleimido)triethylene glycol
BSA Bovine serum albumin
CE Common endosome
DLS Dynamic light scattering
DMEM Dulbecco's modified eagle medium
DMSO Dimethyl sulfoxide
DMT1 Divalent metal transporter 1
DNA Deoxyribonucleic acid
DTT Dithiothreitol
EC50 Half maximal effective concentration
EDC/NHS (1-Ethyl-3-(3-dimethylaminopropyl)-carbodiimide)/N-
hydroxysuccinimide crosslinking
EDTA Ethylenediaminetetraacetic acid
ELISA Enzyme-linked immunosorbent assay
FBS Fetal bovine serum
FcRn Neonatal Fc receptor
FDA U.S. Food and Drug Administration
G-CSF Granulocyte-colony stimulating factor
GI epithelium Gastrointestinal epithelium
h Hour
HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
hGH Human growth hormone
hGH-Tf Human growth hormone-transferrin fusion protein
hGH-C2-Tf Human growth hormone-transferrin fusion protein with the cyclo
linker, but with no thrombin cutting site
hGH-CH4C-Tf Human growth hormone-transferrin fusion protein with the helical
linker, and two cysteine residues at both ends of the linker
hGH-H4-Tf Human growth hormone-transferrin fusion protein with helical linker
hGH-hinge-Tf Human growth hormone-transferrin fusion protein with the IgG Fc
hinge motif as the linker
hGH-Lzip-Tf Human growth hormone-transferrin fusion protein with the leucine
zipper motif as the linker
HoloTf Iron-loaded transferrin
HRP Horseradish peroxidase
I.V. Intravenous
IC50 Half maximal inhibitory concentration
XIII
IgA Immunoglobulin A
IgD Immunoglobulin D
IgE Immunoglobulin E
IgG Immunoglobulin G
IgM Immunoglobulin M
Kd Dissociation constant
KD Equilibrium constant
KDa Kilo dalton
Lzip Leucine zipper linker
min Minute
MonoTf Transferrin monomer
MPS Mononuclear phagocytic system
MWCO Molecular weight cut-off
NaCl Sodium chloride
NaOH Sodium hydroxide
NCA Non-compartmental analysis
NIR Near infrared
NonR-AggTf Non-reducible form of transferrin oligomer
OX26 A mouse monoclonal antibody recognizing rat TfR
PBS Phosphate buffered saline
PCR Polymerase chain reaction
PEG Polyethylene glycol
PEI Polyethylimine
PK Pharmacokinetics
ProINS-Tf Proinsulin-transferrin fusion protein
R-AggTf Reducible form of transferrin oligomer
RPMI Roswell Park Memorial Institute medium
SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis
SEC-MALS Size exclusive chromatography with multi-angle light scattering
siRNA Small interfering ribonucleic acid
SPDP N-succinimidyl 3-(2-pyridyldithio) propionate
SPR Surface plasma resonance
STZ Streptozotocin
TCA Trichloroacetic acid
Tf Transferrin
TfR Transferrin receptor
Tf-CRM107 Tf-diphthria toxin conjugate
TMB 3,3',5,5'-Tetramethylbenzidine
TNBC Triple negative breast cancer
1
ABSTRACT
Transferrin/transferrin receptor (Tf/TfR) system has been exploited to develop
several novel strategies in drug delivery in the past decades, including tumor targeting
and trans-epithelial protein drug delivery. Although it is promising to apply this system to
develop new therapeutics in the pre-clinical settings, there is still no FDA-approved
therapy that is developed based on the Tf/TfR system. Several challenges remain
unsolved to utilize Tf/TfR system in drug development, and one of them is the rapid
recycling of Tf after endocytosis. This phenomenon may limit the therapeutic efficacy if
the drug payload requires intracellular action to achieve the therapeutic effect. Using Tf
oligomer as the carrier has been proposed to solve this issue since it was reported to
exhibit a prolonged cellular retention after endocytosis. However, in order to evaluate
the potential application of Tf oligomer in drug delivery, detailed investigation on
biological properties in vitro and the pharmacokinetics or biodistribution profiles in vivo
are required. In this dissertation, Tf oligomer was chemically synthesized, and the
biological properties were characterized in TfR
high
MDA-MB-231 cancer cells and in
polarized Caco-2 cells. The enhance binding affinity and prolonged cellular retention
observed for Tf oligomer in the MDA-MB-231 cells account for the higher tumor
retention in the xenograft mouse model. However, the prolonged cellular retention
observed in the polarized Caco-2 cells may lead to extensive protein degradation for Tf
oligomer during the transcytosis process. The findings in this dissertation indicate that,
compared with Tf monomer, the chemically-synthesized Tf oligomer could be a good
drug carrier for tumor targeting, but not a suitable carrier to deliver protein therapeutics
across the gastrointestinal epithelium.
2
CHAPTER 1: INTRODUCTION
1.1 The application of serum proteins in drug delivery and development
Protein molecules have been widely utilized as drug carriers to improve the
therapeutic efficacy of drug payloads. Serum proteins, including albumin or
immunoglobulins, have emerged as attractive drug carriers in developing new
therapeutics due to their long-lasting, nonimmunogenic, biocompatible, and
biodegradable characteristics [1]. Albumin, the most abundant serum protein in the
blood circulation with 66.5 KDa in molecule weight, has been utilized to associate, fuse,
or conjugate with various drug payloads to improve the therapeutic efficacy. For
example, myristic acid modified insulin analog, which was developed by Novo Nordisk
and marketed as Levemir®, can be highly associated with albumin to improve the PK
profile. Compared with the half-life of native human insulin, which is around 4 to 6
minutes, Levemir® has the extended half-life around 5 to 7 hours [2]. Therefore,
Levemir was approved by FDA in 2005 as a basal insulin analogue to treat type 1 or
type 2 diabetes. Another example of albumin-based drug is Abraxane® , an albumin
paclitaxel nanoparticle for treating solid tumors. Although the detailed mechanism is still
under debate, Abraxane can be accumulated in the tumor after injection, and the
paclitaxel can be release locally to achieve tumor killing effect [3-6]. Currently, Abraxane
has been approved for treating several types of cancer, including metastatic breast
cancer, metastatic non-small cell lung cancer, and metastatic adenocarcinoma of
pancreas in combination with gemcitabine.
3
Immunoglobulins, the second abundant serum protein, can be classified into five
main isotypes: IgM, IgG, IgA, IgD and IgE. Among them, IgG is the base structure for
most of antibody therapeutics currently on the market. The Fab fragment of the IgG can
recognize the target antigen, and the binding between Fab and antigen can elicit the
pharmacological effect to treat diseases [7]. The Fc fragment of the antibody can bind to
neonatal Fc receptor (FcRn), and the binding between Fc and FcRn is responsible for
the maintenance of IgG structure in the blood circulation. The function of Fc fragment
leads to new Fc-fused protein therapeutics developed to improve the pharmacokinetics
properties and therapeutic efficacy, and several of them have been approved by FDA
for clinic use, such as Eylea for age related macular degeneration, and Enbrel for
rheumatoid arthritis [8]. Nowadays, the monoclonal antibody or Fc-based therapeutics
have become two major modalities to treat various human diseases, suggesting the
success of using immunoglobulin as the platform in drug discovery and development [8,
9].
1.2 Transferrin, a serum glycoprotein
1.2.1 Structure and function
As one of the major serum proteins, transferrin (Tf) also attracted much attention
as a potential drug carrier for novel therapeutic development in past decades.
Transferrins (Tfs), known as iron-transporting proteins, are glycoproteins with 670-700
amino acid residues and around 80 KDa in molecular weight. Tfs are composed of two
evolutionary conserved lobes, the 336 amino acids N-lobe and 343 amino acid C-lobe,
and each lobe has a hydrophilic ferric ion binding site. Several members of Tf have
been found, and the serum transferrin (sTf) is the best well-known one. Serum Tf can
4
be found in the blood and other biological fluid, such as the bile and cerebrospinal fluid.
The serum Tf molecule has very high binding affinity to the Fe
3+
, and the Kd is estimated
around 10
-20
M under neutral pH. However, under acidic environment, the structure of Tf
is altered, leading to the decrease in binding affinity to Fe
3+
, resulting in Fe
3+
release.
The major biological function of Tf is to transport the Fe
3+
to the cells in demand.
Iron is critical for all living organisms, and it can serve as co-factor to support the
function of several essential enzymes, such as ribonucleotide reductase, a key enzyme
for the DNA replication process [10]. In addition, iron is a key co-factor for the heme
group in the red blood cells, and the heme group is responsible for carrying oxygen.
However, free iron ion could be toxic to the body, which may promote the free radical
formation and increase the oxidative stress to damage the tissues [11]. Therefore, with
the high binding affinity to the iron ion, the other function of Tf is to maintain the iron in a
redox-inactive state to protect the body.
1.2.2 Transferrin and Transferrin Receptor-mediated endocytosis
Transferrin receptor (TfR) is a type II transmembrane protein which is composed
of two identical monomers linked by disulfide bonds. The major biological functions of
TfR are iron uptake and cell growth regulation via Tf/TfR-mediated endocytosis. TfR has
high binding affinity to iron-loaded transferrin (holoTf), but 10 to 100-fold lower affinity to
iron-free transferrin (apoTf) under physiological pH [12, 13]. When the holoTf binds to
the TfR on the cell surface, the binding stimulates the initiation of clathrin-mediated
endocytosis. The AP2 adaptor protein and clathrin triskelion are gathered around the
holoTf/TfR complex, and clathrin-coated vesicle is formed through the invagination
process. The vesicle containing holoTf/TfR complex is further directed to early
5
endosome, where the environment is slightly acidic. Due the change in pH, the structure
shift of holoTf allows the release of ferric ion from the Tf molecule, and the free iron ion
is transported outside of the endosome by divalent metal transporter DMT1. Meanwhile,
the apoTf still remains the bound form to the TfR in the endosome after releasing ferric
ion, and the apoTf/TfR complex is subsequently transported back to the cell surface.
Once the vesicle fused with the cell membrane, the apoTf is released from the complex
back to the circulation system. The whole process is called Tf/TfR-mediated
endocytosis and recycling, which is illustrated in Figure.1.1.
Figure 1.1: TfR-mediated endocytosis and recycling [14]
1.2.3 Transferrin Receptor-mediated transcytosis
In polarized epithelium cells, not all the internalized holoTf will be recycled back
to the blood stream or the physiological fluid after binding to the TfR. More evidence
indicated that part of the Tf can be transcytosed across the polarized cells, including the
intestinal and alveolar epithelium. The possible mechanism for TfR-mediated
6
transcytosis has been proposed and illustrated in Figure.1.2. In polarized epithelium,
majority of the TfRs reside in the basolateral side of the cell surface, but some TfRs
appear at the apical cell membrane due to missorting events. After binding to the TfR at
the apical side, holoTf is internalized through TfR-mediated endocytosis, and then
transported to the apical endosome (AE). Inside the AE, holoTf releases iron ion, and
the apoTf become accumulated in the common endosome (CE) for longer period. On
the other side, divalent metal transporter (DMT) enters the common endosome through
endocytosis process, which also transports the ferric ion into the common endosome.
Once the ApoTf binds to the ferric ion to form holoTf, it will start the exocytosis process.
Subsequently, the holoTf is transported to the basolateral recycling route, and then
released into the blood circulation.
Figure 1.2: The proposed model for TfR-mediated transcytosis in polarized GI
epithelium. Adapted from “Transferrin Receptor-Mediated Transcytosis in Intestinal
7
Epithelial Cells for Gastrointestinal Absorption of Protein Drugs”, Chapter 3 in “Targeted
Delivery of Small and Macromolecular Drugs”, ISBN: 9781420087727, 2010.
1.3 The application of Tf/TfR system in drug delivery and development
1.3.1 Tumor targeting
One of the most well-known applications of utilizing Tf/TfR system in drug
development is tumor targeting. Tumor cells require higher amount of ferric ions as a
co-factor to support their rapid proliferation [10], and majority of malignant cells are
found TfR overexpressed on the surface to transport the ferric ions [15]. Therefore, TfR
becomes an attractive receptor to design novel targeting strategies to deliver
therapeutic agents to the tumor mass. Tf, or the TfR antibody have been directly fused
or conjugated to anti-tuomr agents, such as toxins or therapeutic agents, to achieve
tumor targeting, further leading to improved therapeutic efficacy. One of the examples is
Tf-CRM107, a Tf-diphthria toxin conjugate, which was developed to treat malignant
glioma [16]. In the preclinical studies, Tf-CRM107 is 1000 to 100000-fold more toxic to
the tumor cells compared with the CRM107 in the cell-based studies, and 10 to 100
more potent in the animal studies. In the phase II clinical trials, 35% of the evaluable
patients receiving Tf-CRM107 infusion showed complete or partial response rate,
suggesting that Tf-CRM107 could be an alternative therapy for glioma patients. In
addition to the direct fusion or conjugation to anti-tumor agents, Tf or TfR antibodies are
often used to modify the surface of various types of nanocarriers to achieve active
tumor targeting. Different modalities of therapeutic agents, including plasmid DNA,
siRNA, small molecules, have been encapsulated in the Tf-modified nanocarriers,
resulting in a better anti-tumor efficacy compared with non-decorated nanocarrier or free
8
drugs in various types of tumor bearing mouse models [17]. These results all indicate
that the TfR is a valuable surface receptor to design targeted therapies against cancer.
1.3.2 Half-life extension
The other mentionable application of using the Tf moiety in drug development is
to increase the half-life of therapeutic proteins or peptides. Fusing to the Tf moiety
significantly increase the molecule weight for small protein or peptide drugs, which
avoid the rapid clearance from the renal filtration [18]. In addition, similar to Fc-based or
albumin-based therapeutic fusion proteins, the Tf/TfR recycling pathway prevents the
Tf-based fusion protein from being directed to lysosomal compartment, resulting in
extended half-life in the circulation [19]. Although fusing Tf moiety may sometimes
attenuate the bioactivity of the protein drugs, the prolonged half-life of the fusion can
compensate for the lower bioactivity to reach comparable pharmacological effect in vivo
[20].
1.3.3 Delivery of protein drugs across gastrointestinal epithelium and blood brain
barrier
The large molecular size and the sensitivity to protease degradation limit the use
of most current protein drugs from oral administration. The invasive treatments and
injection site reactions often lead to reduced patient compliance to protein therapeutics.
Therefore, techniques to deliver protein drugs across GI epithelium is heavily desired.
The phenomenon of TfR-mediated transcytosis inspired scientists to develop novel
strategies to deliver protein therapeutics across the GI epithelium without disrupting the
tight junction. Previously, insulin was conjugated with Tf moiety, and this conjugate
9
achieved hypoglycemic effect after orally administered into STZ-induced diabetic rats
[21]. Recombinant fusion protein design was also applied to fuse Tf with granulocyte-
colony stimulating factor (G-CSF) or human growth hormone (hGH), and both fusion
proteins demonstrated the therapeutic efficacy after orally administered to BDF1 mice or
hypophysectomized rats, respectively [20, 22]. All these reports indicate that TfR is the
endogenous receptor that can be exploited to design novel platform to delivery protein
drugs across GI epithelium.
TfR is also expressed on the surface of brain endothelial cells, which is proposed
to be responsible for transporting ferric ion into the brain tissue. Although the
mechanism of TfR-mediated transcytosis in brain endothelium is not well characterized,
extensive studies have been performed to explore whether TfR-targeting moieties can
be used to delivery therapeutic proteins across BBB. Two TfR antibodies have been
raised to investigate thoroughly: OX26, a mouse monoclonal antibody recognizing rat
TfR [23], and 8D3, a rat monoclonal antibody recognizing mouse TfR [24]. Intravenous
injection of OX26 antibody into rats lead to the detection of antibody in brain vasculature
and the brain parenchyma [25], Different large molecule modalities, including
radiopeptides, enzymes, antisense oligos, have been modified with 8D3 antibody to
increase the penetration across the BBB and the amount accumulated in brain tissues.
These positive results all suggest that, by targeting the TfR on the brain endothelium, it
is possible to deliver protein therapeutics across BBB to treat neuronal diseases in the
central nervous system. However, more studies and researches are still required to
optimize these antibodies that can be really transcytosed across BBB rather than being
degraded in the brain endothelium.
10
1.3.4 Protein prodrug activation
Previously, a fusion protein has been constructed by fusing proinsulin and
transferrin moieties together in our laboratory. The native state of this proinsulin-
transferrin (ProINS-Tf) fusion protein possess no activity in inducing insulin receptor
(IR)-mediated signaling; however, it can be converted into an active form to elicit IR-
mediated signaling after incubated with H4IIE rat hepatoma cells [26]. Addition of
excess Tf can block the conversion, suggesting that the activation process of ProINS-Tf
requires the internalization event through TfR-mediated endocytosis, which the fusion
protein may encounter certain type of protein protease or convertase during the
recycling process [26]. The TfR-mediated protein prodrug conversion can be also
observed in vivo, and the activated ProINS-Tf can achieve hypoglycemic effect in both
STZ-induced and non-obese diabetic mouse models [27, 28]. Utilizing TfR-mediated
conversion is new ca oncept to design protein prodrug for therapeutic use, and the
ProINS-Tf could be a novel insulin analogue to maintain basal glucose in diabetic
patients.
1.4 Challenges of targeting Tf/TfR system in drug development
1.4.1 Potential off-target effect
Although several foreseeable applications of targeting Tf/TfR to design novel
therapeutics, some hurdles remain unsolved before moving into clinical stages. In anti-
cancer drug development, it is critical to design a new therapy that not only target the
tumor mass, but also avoid the off-target effect to the normal tissues. TfR is
overexpressed in majority of malignant cell types; however, TfR is also expressed in
11
several normal tissues, including bone marrow, erythrocytes, kidney tubular epithelium,
and liver Kupffer cells and hepatocytes [15]. A study was performed to investigate the
safety aspect of anti-TfR antibodies administration, and single dose of high affinity anti-
TfR antibody results in reticulocyte depletion in mice [29]. Although the anti-TfR is
designed for penetrating BBB, safety studies should be performed to address whether
the Tf-modified or TfR antibody-modified drug carriers would cause any impact on
normal tissues.
1.4.2 Binding competition from the endogenous Tf
Another issue of utilizing the Tf/TfR system in drug delivery is the competition
from the endogenous Tf to the binding site of TfR. Tf is abundant in the blood
circulation, with estimated concentration around 25 M in the plasma [13, 30].
Therefore, endogenous Tf would compete with those therapeutic agents or drug carriers
modified with Tf molecule to the binding site TfR, which could affect the delivery
efficiency to the target tissue. To avoid this limitation, TfR-targeted antibodies or TfR-
targeted peptides could be used to bind TfR with at other binding sites, rather than the
same binding site which the endogenous Tf would bind to.
1.4.3 Low transcytosis rate across the BBB
Targeting TfR is proposed to be one of the valuable strategies for delivering
protein therapeutics across the BBB; however, only 0.44% of the initial dose of OX26
antibody were detected after 24 hours post-injection [25]. Majority of the injected
antibody was associated with brain capillary, rather than entering brain areas [31]. In
order to optimize the transcytosis rate of the TfR antibody, Yu et al found that lower
12
affinity anti-TfR antibody has better transcytosis rate across the BBB compare with the
higher affinity one [32]. A following study further demonstrated that higher affinity TfR
antibody has an increased internalization incidence, but also higher increased
degradation rate compared to the lower affinity TfR antibody or the control isotype IgG
[33]. Therefore, strategies to find the most optimized parameters for TfR antibody to
penetrate the BBB is still required [34].
1.4.4 Rapid recycling after endocytosis
Several studies have been conducted to estimate the kinetics of the TfR-
mediated endocytosis and recycling, which the average cellular retention time for Tf-TfR
complexes range from 4 min to 22 min [35]. This short cellular retention time and rapid
recycling process limit the therapeutic efficacy if the drug payloads are required to be
released inside the cell to achieve their therapeutic function [36]. In addition, rapid
recycling may also decrease the chance and the amount of Tf-TfR complex being
transcytosed across the cell, which could restrict the application of Tf/TfR system in
delivering drug payload polarized epithelium.
Exploiting Tf/TfR systems to design novel drug carriers or therapeutics might be
a promising strategy to overcome the challenges in cancer treatments, diabetic
treatments, or in delivering protein therapeutics across physiological barriers. However,
the obstacles mentioned in the previous section are required to be solved. Several
possible solutions have been proposed, including using multivalent Tf to increase the
cellular retention time after binding to the TfR [37, 38]. Although some studies have
been done to discuss the potential of using multivalent Tf, or Tf oligomer, to replace Tf
monomer for optimizing the system, more detailed comparison in biological properties
13
between Tf monomer and Tf oligomer are needed. Series of experiments, especially
comprehensive in vivo studies, are required to evaluate the full potential of utilizing Tf
oligomer in drug delivery and development.
14
CHAPTER 2: SYNTHESIS AND PHYSICOCHEMICAL PROPERTY
CHARACTERIZATION OF TF OLIGOMER
2.1 Background
A multivalent Tf molecule has been introduced, and its capability in crosslinking
TfR to prolong its retention in the cells was described [37]. Therefore, using multivalent
Tf as an alternative ligand to target TfR to avoid the rapid recycling process was
proposed to improve the efficacy of Tf-based therapeutics. However, the method to
prepare multivalent Tf for further investigation is not yet determined. Several methods
for preparing protein oligomer or protein particles have been introduced, including
coacervate, desolvation, emulsification, self-assembly, and nanoprecipitation [39]. In
this dissertation, we adopted the method introduced previously from our laboratory [38],
trying to synthesize Tf oligomer by chemical conjugation. It was reported that chemical
conjugation may lead to a denser and tighter particle formation, which made the particle
easier to penetrate the tissues or tumor mass [39]. In addition, in the synthesis steps,
we used different crosslinker to generate reducible or non-reducible bond between Tf
moieties, so that we could elucidate the potential influence of the reducing environment
to the biological properties of Tf oligomer in the cell or in the body.
15
2.2 Materials and Methods
2.2.1 Preparation of NonR-AggTf and R-AggTf
2.2.1.1 Iron-saturation of ApoTf
10 mg of human Apo-Tf (Sigma-Aldrich, MO) was first dissolved in PBS (without
K
+
), and saturated with ferric ammonium citrate solution at 37 °C for 2 h. After
incubation, the mixture was poured into the presoaked Spectra/Por dialysis membrane
(MWCO 12-14 KDa, Spectrum Laboratory, CA) and dialyzed in 2 L of PBS twice at 4°C
overnight. After dialysis, this iron-saturated Tf monomer solution is referred as MonoTf
solution in this article.
2.2.1.2 SPDP conjugation on Tf
For the synthesis of both NonR-AggTf and R-AggTf, SPDP was first conjugated
to MonoTf. 10 mg of MonoTf in 1 mL PBS was mixed with 375 mg of N-succinimidyl 3-
(2-pyridyldithio) propionate (SPDP) (Thermo Fisher Scientific, IL), which was dissolved
in DMSO (Sigma-Aldrich, MO), to generate sulfhydryl-containing Tf. After the
conjugation reaction at 4 °C for 4 h, the solution was poured into a presoaked
Spectra/Por dialysis membrane (MWCO 12-14 kDa) and dialyzed twice in 2 L of PBS at
4 °C overnight. The level of SPDP modification was monitored by pyridine-2-thione
assay as described on the manufacturing manual, and the ratio between conjugated
SPDP to MonoTf was kept around 5:1.
2.2.1.3 Synthesis of NonR-AggTf
SPDP-conjugated Tf moieties were crosslinked with 1,11-
bis(maleimido)triethylene glycol, BM(PEG)3 (Thermo Fisher Scientific, IL), to generate
16
non-reducible Tf oligomer. In detail, the molar ratio between conjugated SPDP and
BM(PEG)3 linker was 2:1.1 in the reaction. The desired amount of BM(PEG)3 was
dissolved in DMSO and divided into five aliquots. In order to generate sulfhydryl-
activated Tf, SPDP-conjugated Tf solution was reduced by 25mM DTT at room
temperature for 30 min. Size-exclusion chromatography (Sephadex-G50, 1.5 cm x 15
cm, PBS mobile phase, pH 7.4) (Amersham) was used to isolate sulfhydryl-activated Tf
and remove DTT from the solution. For the linking process, sulfhydryl-activated Tf
solution was added with each BM(PEG)3 aliquot while stirring on ice at intervals of 3
min, 3 min, 3 min, and 10 min between each mixing. After adding the last aliquot, the
mixture was stirred at 4 °C overnight, followed by dialysis (MWCO 12-14 KDa) in PBS to
remove non-reacted BM(PEG)3. The reaction scheme for preparing NonR-AggTf is
shown in Figure 2.2.1.
2.2.1.4 Synthesis of R-AggTf
SPDP-conjugated Tf moieties were crosslinked with tetra(ethylene glycol) dithiol
(Sigma-Aldrich, MO) to generate reducible Tf oligomer. In detail, the molar ratio
between conjugated SPDP and tetra(ethylene glycol) dithiol was 2:1.1 in the reaction.
The desired amount of tetra(ethylene glycol) was dissolved in DMSO, and divided into
five aliquots. Subsequently, the SPDP-conjugated Tf solution was mixed with each
aliquot at 3 min per interval to perform the linking reaction on ice. After final addition of
the last aliquot, the mixture was agitated at 4 °C overnight, then followed by dialysis
(MWCO 12-14 KDa) in PBS to remove non-reacted tetra(ethylene glycol). The reaction
scheme for preparing NonR-AggTf is shown in Figure 2.2.2.
2.2.1.5 Purification of NonR-AggTf and R-AggTf
17
After the crosslinking reaction, size-exclusion chromatography was performed by
flowing through Sephacryl S200HR column (2cm x 25 cm, PBS mobile phase, pH 7.4)
(GE Healthcare Life Sciences, United Kingdom) and 60 fractions were collected with 0.5
mL in each fraction. SDS-PAGE and Coomassie blue staining were performed to
analyze the protein sample in each fraction, and fractions with higher purity of R-AggTf
or NonR-AggTf were pooled together. The pooled samples were further concentrated by
Microsep
TM
Advance Centrifugal Device (10 MWCO, Pall Corporation, NY), and the final
products were sterilized, analyzed by SDS-PAGE, and stored at 4 °C.
2.2.2 Dynamic light scattering analysis
The protein samples were prepared at the final concentration of 1 mg/mL, and 50
uL of the samples were loaded into clear flat bottom/black 96-well plate in quintuple.
The hydrodynamic radius was determined by conducting dynamic light scattering (DLS)
analysis (Dynapro Plate Reader, Wyatt, Santa Barbara, CA).
2.2.3 Stability analysis
The protein samples were prepared at a final concentration of 1 mg/mL and
incubated at 37 °C up to 24 h. At 2 h, 6 h,12 h, and 24 h timepoints, the samples were
collected and analyzed by DLS.
2.2.4 SEC-MALS analysis
The average molecular weight for R-AggTf or NonR-AggTf was measured
through the combination of size exclusive chromatography with multi-angle light
scattering (SEC-MALS). 100 g of protein samples in 50 L were loaded to the Shodex
KW-804 column (Showa Denko K.K, Tokyo, Japan), and the PBS was used as the
18
mobile phase to separate the sample. The signal for protein sample was captured by
both UV and Wyatt MALS detector, the readout and the molecular weight were further
analyzed by ASTRA software provided by Wyatt.
2.2.5 Biosensor analysis
The biosensor-based affinity analysis, or surface plasma resonance (SPR)
assay, was performed as described previously with some modification [40]. In brief, the
sensor chips were prepared using amine coupling through EDC/NHS to attach ~3000
resonance units of recombinant human TfR protein (Abcam, Cat # ab182678,
Cambridge, UK) to the surface of Series S CM5 sensor chip (GE Healthcare, Chicago,
IL) [41]. MonoTf, R-AggTf, or NonR-AggTf at various protein concentration ranging from
0.16 g/mL to 20.48 g/mL were injected to the TfR-coupled flow-cells using the flow
rate of 50 L/min in the 4-min association phase. Subsequently, HEPES based saline
buffered (10mM HEPES, 3 mM EDTA, 150 mM NaCl, pH=7.4) was injected to the flow-
cells with the flow rate of 50 L/min in the 10-min dissociation phase. Afterward, the
flow-cells were regenerated by injecting a buffer containing 0.5 M MgCl2, 50 mM
HEPES (pH=7.4), 150 mM NaCl, and 0.005% Tween-20. Equilibrium dissociation
constants were calculated based on the association and dissociation rate constant.
19
Figure 2.2.1: Schematic Illustration of NonR-AggTf synthesis
Figure 2.2.2: Schematic Illustration of R-AggTf synthesis
20
2.3 Result
2.3.1 SDS-PAGE and Coomassie blue staining of NonR-AggTf and R-AggTf
After series of reaction to prepare NonR-AggTf or R-AggTf, the final product was
analyzed by SDS-PAGE followed by Coomassie Blue staining to confirm the purity. For
both products, the purity can reach around 90-95% based on the analysis from BioRad
ImageLab (Figure 2.4.1). Under the reducing environment, the NonR-AggTf still
maintained its oligomeric form (Figure 2.4.1 Lane 4), but the R-AggTf was dissociated
into monomeric form (Figure 2.4.1 Lane 8).
2.3.2 DLS analysis of MonoTf, NonR-AggTf, and R-AggTf
DLS was performed to examine the hydrodynamic radius of MonoTf, NonR-
AggTf, and R-AggTf (Figure 2.4.2). The MonoTf has the size around 8 nm in diameter
(Figure 2.4.2 black line). For NonR-AggTf final product, the size in average is around
48 nm in diameter, with 23.8% in polydispersity (Figure 2.4.2 red line). For R-AggTf
final product, the size in average is around 30 nm in diameter, with 23% in the
polydispersity but 2 peaks observed in the data (Figure 2.4.2 blue line).
2.3.3 SEC-MALS analysis of NonR-AggTf and the R-AggTf
Since the size for NonR-AggTf and R-AggTf is relatively larger than most of the
protein molecule, the SEC-MALS was performed to estimate the average molecular
weight for NonR-AggTf and R-AggTf. The continuous signal rather than a single peak
was acquired by the SEC-MALS analysis data, suggesting both NonR-AggTf and R-
AggTf are not homogeneous product. The estimated average molecular weight for
NonR-AggTf and R-AggTf are 1200 KDa and 1300 KDa, respectively (Figure 2.4.3).
21
2.3.4 Stability test
General stability test was performed to study whether NonR-AggTf is stable
under physiological temperature. NonR-AggTf was incubated under 37 °C up to 24 h,
and no significant difference in the hydrodynamic radius and percent of polydispersity
was observed at designated timepoints (Figure 2.4.4).
2.3.5 Binding affinity analysis of NonR-AggTf and R-AggTf to TfR by SPR
Labeling-free SPR assay was conducted to evaluate whether the binding
capability of the Tf units on NonR-AggTf or R-AggTf to the TfR were affected after
series of conjugation process. The SPR sensograms for MonoTf, NonR-AggTf, and R-
AggTf are shown in Figure 2.4.5. Compared with the MonoTf, which has the equilibrium
around 1.86*10
-8
M, the equilibrium constant (KD) for NonR-AggTf and R-AggTf is
1.97*10
-8
M and 2.72*10
-8
M. The results from the SPR suggested that, on the NonR-
AggTf and R-AggTf molecules, the binding capability of the Tf units to TfR were slightly
but significantly changed after the conjugation process.
22
2.4 Figures
Figure 2.4.1: SDS-PAGE and Coomassie blue analysis of NonR-AggTf and R-
AggTf. NonR-AggTf (Lane 3-4) and R-AggTf (Lane 7-8) were analyzed by SDS-PAGE
and followed by Coomassie blue staining to confirm the purity of the end products.
(Lane 1 and Lane 5: PageRuler
TM
Prestained Protein Ladder #SM0671/2, Lane 2 and 6:
MonoTf. Lane, Lane 3: NonR-AggTf in non-reducing sampling buffer, Lane 4: NonR-
AggTf in reducing sampling buffer, Lane 7: R-AggTf in non-reducing sampling buffer,
Lane 8: R-AggTf in reducing sampling buffer)
23
Figure 2.4.2: DLS analysis of MonoTf, NonR-AggTf, and R-AggTf. The
hydrodynamic radius of MonoTf, NonR-AggTf, and R-AggTf were determined by DLS.
NonR-AggTf has the average particle size around 48 nm in diameter, with 23.8% in
polydispersity (red line). R-AggTf has average particle size around 30 nm in diameter,
with 23% in polydispersity (blue line).
24
Figure 2.4.3: SEC-MALS Analysis of NonR-AggTf and R-AggTf. The sample of
NonR-AggTf and R-AggTf were subjected to SEC-MALS analysis to generate the
chromatograms. Based on the chromatograms and Astra software analysis, the
estimated average molecular weights for NonR-AggTf (blue line) and R-AggTf (red line)
are 1217 KDa and 1317 KDa, respectively.
25
Figure 2.4.4: Stability test of NonR-AggTf in PBS at 37 ℃. (A) NonR-AggTf was
incubated in PBS at 37 ℃ for 2 h, 6 h, 12 h and 24 h, and DLS was performed to
determine the diameter of NonR-AggTf at designated timepoints. (B) The average
diameter (black solid line) and polydispersity index (blue solid line) of NonR-AggTf were
determined by DLS after incubating in PBS at 37 ℃ from 2 h to 24 h. Date were
presented as mean values with error bars indicating the standard deviation (n=5)
26
Figure 2.4.5: Binding affinity analysis of NonR-AggTf and R-AggTf to TfR by SPR.
Various concentration of MonoTf, NonR-AggTf, R-AggTf were injected to bind the TfR
immobilized covalently on the chip, and the sensorgrams of MonoTf (A), NonR-AggTf
(B) or R-AggTf (C) were generated. The fitting curves were obtained based on the
sensorgrams plotted on the upper right to determine the KD
27
2.5 Discussion
Based on the reaction process illustrated in Figure 2.2.1 and Figure 2.2.2, the
NonR-AggTf and R-AggTf was synthesized by chemical conjugation, and the solutions
of NonR-AggTf and R-AggTf were further purified, concentrated to obtain the final
product. SDS-PAGE followed by Coomassie blue staining demonstrated that the Tf
oligomer formed by adding BM(PEG)3 as the crosslinker is non-reducible, indicating the
thioester bond was formed as we expected in the reaction scheme. Meanwhile, the Tf
oligomer formed by adding tetra(ethylene glycol)dithiol in the crosslinking reaction is
reducible when incubated with DTT, suggesting the disulfide bond was formed between
the Tf moieties. The plan for designing two forms of Tf oligomer is to study whether the
reducing environments in the cells (e.g. cytosol) [42] or in the body (e.g. protein disulfide
isomerase on hepatocyte membrane surface) [43, 44] would have any impact on the
biological properties of these two Tf oligomers, allowing us to elucidate the biological
process for Tf oligomers after binding to TfR on cell surface or being injected into the
body.
After both NonR-AggTf and R-AggTf were successfully synthesized and purified,
basic physical properties were characterized, including their particle size, molecular
weight, and stability. Compared with size of other types of protein
oligomer/nanoparticles prepared by coacervate or disolvation, which are around 100-
200 nm in diameter [39], our Tf oligomers have much smaller size around 50 nm in
diameter. The covalent bonding formed through our proposed reaction process may
result in tighter, denser, and smaller particle size as we observed in the DLS analysis.
Although the results from the SEC-MALS indicate that the average molecular weight of
28
R-AggTf is slightly higher than the NonR-AggTf, the DLS results shows the opposite
result that the NonR-AggTf have a larger particle size compared with the R-AggTf. This
difference may due to the longer linker we used to prepare the NonR-AggTf. Further,
the NonR-AggTf was used as the model to perform the preliminary stability study to test
whether the size or the polydispersity would be changed under physiological
temperature. As shown in Fig.2.4.4, no significant change was observed for the NonR-
AggTf up to 24 h, which is expected since the force to assemble Tf moieties is the
covalent bonding. Therefore, we think the R-AggTf should be also stable when
incubating under the physiological temperature in the PBS.
The surface modification with SPDP was involved in the synthesis of NonR-
AggTf and R-AggTf; therefore, it is critical to ensure the binding capability to the TfR
was not significantly affected after series of chemical conjugation steps. SPR, a
labeling-free assay, was performed to assess the binding affinity of the NonR-AggTf and
R-AggTf to the TfR. Since the product for NonR-AggTf and R-AggTf is heterogenous,
protein samples with the same amount in mass was injected to measure the binding
affinity of the Tf units on the NonR-AggTf and R-AggTf to the TfR. After fitting the
sensogram results, the KD for both MonoTf and NonR-AggTf are similar, suggesting the
binding affinity of Tf units on the NonR-AggTf to the TfR were not altered significantly.
Whereas, the KD for R-AggTf to the TfR was slightly increased, indicating the binding
affinity of Tf units to the TfR was slightly decreased after the R-AggTf was synthesized.
The SPR results demonstrated that, in general, both NonR-AggTf and R-AggTf still
maintain good binding capability to the TfR even after series of chemical reactions
during the synthesis process. Previously, a self-assemble Tf nanoparticle was
29
introduced by adding the DTT to break the intramolecule disulfide linkage in the Tf, then
resembled by incubating under oxidative environment [45]. The self-assembled Tf
particle still maintained the binding capability to the TfR. Our Tf oligomers and the self-
assemble Tf particle can still bind to the TfR, suggesting the association between Tf and
TfR is hard to be fully disrupted after chemical treatments on the Tf moieties in the
preparation process.
2.6 Summary
In this chapter, the NonR-AggTf and R-AggTf are synthesized by chemical
conjugation. Both NonR-AggTf and R-AggTf have small particle size around 50 nm and
30 nm in diameter, respectively. The results from SPR analysis suggest that both Tf
oligomers maintain their capability in binding to the TfR, which provides us good models
to further characterize the several biological characteristics of oligomeric Tf, such as
cellular trafficking behaviors after binding to the TfR on cell surface, or the
pharmacokinetics and biodistribution properties after injecting into animal body.
30
CHAPTER 3: EVALUATION OF THE POTENTIAL APPLICATION OF TF
OLIGOMER IN TUMOR TARGETING
3.1 Introduction
One of the major applications of using Tf/TfR system in drug development is to
deliver the drug payload to the tumor tissue. However, if a drug payload needs
intracellular action to achieve anti-tumor effect, the rapid of Tf/TfR-mediated recycling
would become the limiting factor [46]. Previously, a Tf oligomer-methotraxate conjugate
was synthesized in our lab, and the Tf oligomer-methotrexate conjugate demonstrated
better anti-proliferation effect compared with the Tf monomer-methotraxate conjugate in
MCF-7 breast cancer cells [38]. However, the biological process of Tf oligomer after
binding to the TfR is still largely unknown. In addition, the pharmacokinetics properties,
biodistribution profile, and the tumor targeting capability if Tf oligomer have not been
investigated. In order to evaluate whether the Tf oligomer could be an ideal carrier to
deliver therapeutic agents to tumor mass, more studies, especially animal experiments,
are required to translate the findings from cell-based models to in vivo models.
MDA-MB-231 was chosen as the cancer cell model to study the biological
characteristic of Tf oligomers in cell-based assays and the tumor targeting and retention
effect in the xenograft animal models in this chapter. MDA-MB-231 is a triple negative
breast cancer cell line, which has minimal expression in estrogen receptor,
progesterone receptor, and human epidermal growth factor receptor 2 [47]. Triple
negative breast cancer (TNBC) accounts for 10-15 % of diagnostic breast cancer cases
worldwide, and TNBC cases usually have high pathological grade [48]. The
31
chemotherapy is the current mainstay treatment for TNBC; however, systematic
chemotherapy often leads to several side effects, including nausea and vomiting [49,
50]. Therefore, new targeted therapies are desired for treating the patients with TNBC to
improve their quality of lives. As one of the TNBC cell lines, MDA-MB-231 cells were
often used as the model to evaluate the therapeutic efficacy of novel treatment for
TNBC in pre-clinical stage [51]. In addition, the MDA-MB-231 cell line is found to
overexpress TfR on the cell surface [52], which is a good cell model for us to study the
process after the Tf oligomers binding to the TfR on the cell surface.
In this chapter, several biological characteristics of Tf oligomer, including the
cellular retention, binding affinity, and cellular uptake amount, were studied in the MDA-
MB-231 cells. Furthermore, the comparison of the pharmacokinetics and biodistribution
profiles between Tf oligomer and monomer were conducted in wildtype mice. Lastly, the
tumor targeting and retention effects of the Tf oligomer were investigated in the MDA-
MB-231 xenograft mouse model, trying to evaluate the application of Tf oligomer in
tumor targeting.
32
3.2 Methods
3.2.1 Cell culture
Human mammary gland adenocarcinoma MDA-MB-231 cell line was purchased
from ATCC (Manassas, MA). MDA-MB-231 cells were grown in DMEM media
supplemented with 10% FBS, 2 mM L-glutamine, 50 U/mL penicillin, and 50 g/mL
streptomycin. MDA-MB-231 cells were incubated in humidified atmosphere at 37°C
containing 5% CO2, and subcultured prior to full confluence using trypsin/EDTA.
3.2.2 Protein labeling
For protein radio-iodination, MonoTf, NonR-AggTf, and R-AggTf were labeled
with
125
I by the Chloramine T method as previously described [53]. First, solutions
containing 750 g of MonoTf, NonR-AggTf, or R-AggTf were mixed with
125
I-Na
(PerkinElmer, MA) in 1.5 mL tubes. Subsequently, 50 L of Chloramine T (Sigma-
Aldrich, MO) solution (8 mg/mL) was added into the mixture to start the labeling
reaction. After a 10 min incubation on ice, 50 L of sodium metabisulfite solution (4.8
mg/mL) was added into the mixture, and the mixture was incubated on ice for 5 min to
stop the reaction. Subsequently, 100 L of potassium iodide solution (10 mg/mL) was
added to the mixture, and the mixture was incubated on ice for 5 min. The mixture was
next loaded onto a Sephadex-G50 size-exclusion column (0.8 cm x 20 cm, PBS mobile
phase, pH 7.4) to purify
125
I-Tf. Ten fractions were collected from the chromatography
with 1 mL for each fraction, and the gamma radiation from 5 L of each fraction was
measured by a Cobra II gamma counter (PerkinElmer, MA).
33
For protein labeling with biotin, 1 mL of MonoTf, NonR-AggTf, or R-AggTf
samples (2 mg/mL) were reacted with 25 L of 10 mM sulfo-NHS-LC-Biotin (Thermo
Fisher Scientific Inc), and the unreacted sulfo-NHS-LC-biotin molecules were separated
by dialysis. HABA/Avidin reagent (Sigma-Aldrich, MO) was used to determine the level
of biotin conjugated on the target protein. To prepare near infrared (NIR) labeled
protein, MonoTf or NonR-AggTf was reacted with IRDye 800CW NHS ester (LICOR,
Lincoln, NE) according to the manufacturer’s protocol. To prepare NIR-MonoTf, the
reactions were carried out at room temperature for 2 h with the molar ratio
(dye/transferrin unit) of 2:1 to achieve ~1:1 modification ratio. To prepare NIR-NonR-
AggTf, the reactions were carried out at room temperature for 2 h with the molar ratio
(dye/transferrin unit) of 4:1 to achieve ~1:1 modification ratio. The NIR-labeled proteins
were purified by Sephadex G-50 (MWCO: 12~14 KDa) size-exclusive chromatography.
The concentration of radio-iodine-, biotin-, or NIR-labeled proteins were determined by
MicroBCA
TM
protein assay kit (Thermo Fisher Scientific, Waltham, MA)
3.2.3 Pulse-chase assay
Full confluence of MDA-MB-231 cells in 6-well plates were prepared for the
pulse-chase assay. The cells were first washed with PBS to remove residual of Tf in the
culture media, and then incubated with serum-free dosing medium containing 3.2 g of
125
I-MonoTf,
125
I-NonR-AggTf, or
125
I-R-AggTf at 37 °C for 30 min to uptake the radio-
iodinated Tf (pulse phase). After the pulse phase, the dosing medium was removed,
and the cells were washed with ice-cold PBS, then incubated with serum-free medium
containing 320 ug MonoTf for different time length (15 min, 30 min, 1 h, 2 h, 3 h, or 24
h) to collect the recycled radio-iodinated Tf (chase phase). After the chase phase, the
34
cells were lysed with 1N NaOH to prepare the cell lysate. Radioactivity in the chase
medium and the cell lysate were measured by the Cobra II gamma counter, which
represent the amount of recycled radio-iodinated Tf and cell-associated radio-iodinated
Tf, respectively. The sum of the radio-iodinated Tf in recycled medium and cell lysate is
considered as 100%, and the percentage of recycled and cell-associated radio-
iodinated Tf can be obtained.
In order to investigate the percentage of intact and degraded radio-iodinated Tf
after recycling back to medium, trichloroacetic acid (TCA) protein precipitation was
performed. 1 mL of the chase medium with recycled radio-iodinated Tf was mixed with
250 mL of 6.25 M TCA solution, and the mixture was placed on ice for 10 min, and then
centrifuged with 13000 rpm at 4°C for 10 min to precipitate intact protein. The
radioactivity of the degraded protein in the supernatant portion and the intact protein in
the pellet was measured. The radioactivity values from supernatant represented the
amount of degraded radio-iodinated Tf recycled back to reservoir medium, while the
radioactivity value from the protein pellet represented the intact radio-iodinated-Tf in the
medium. The percentage of degraded and intact radio-iodinated Tf can be calculated.
3.2.4 TfR competitive binding assay
Full confluence of MDA-MB-231 cells in 6-well plates were incubated with
125
I-
MonoTf,
125
I-NonR-AggTf, or
125
I-R-AggTf (1.6 g/mL), with the addition of unlabeled
MonoTf (concentration ranged from 0.016 g/mL to 800 g/mL) in the serum-free
DMEM supplemented with 0.1% BSA at 4°C for 2 h. Subsequently, dosing medium was
removed, the cells were washed with ice-cold PBS and solubilized in 1N NaOH. The
35
radioactivity of
125
I-proteins in cell lysate samples were detected and measured by using
Packard gamma counter, and the concentration for MonoTf to inhibit 50% binding (IC50)
of
125
I-MonoTf,
125
I-NonR-AggTf, or
125
I-R-AggTf to the cell was determined by
Graphpad.
3.2.5 Cellular uptake assay
Full confluence of MDA-MB-231 cells in 6-well plate were incubated with dosing
medium containing
125
I-MonoTf,
125
I-NonR-AggTf or
125
I-R-AggTf (3.2 g/mL) in the
serum-free DMEM supplemented with 0.1% BSA at 37°C for 30 min to 3 h.
Subsequently, the cell monolayers were washed with ice-cold PBS and then incubated
with Trypsin-EDTA at 37°C for 3 min to detach cells, as well as to remove the surface-
bound
125
I-protein. The cells were collected and centrifuged to separate
125
I-protein in
the pellet (Intracellular) from the supernatant, and the cell pellet was dissolved in 1N
NaOH. The radioactivity in the cell pellet portion was detected, and the
125
I-protein
amount was normalized with the total protein.
The cellular uptake assays with the non-labeled MonoTf competition were
conducted to mimic the physiological situation in the circulation, as well as to investigate
if the internalization process was TfR-mediated. In the dosing medium,
125
I-MonoTf,
125
I-
NonR-AggTf, or
125
I-R-AggTf Tf (3.2 g/mL), with 20-fold or 100-fold unlabeled MonoTf
in the serum-free DMEM supplemented with 0.1% BSA were added to the full
confluence of MDA-MB-231 or HeLa cells, and the cellular uptake assays were
performed as described in the previous paragraph.
3.2.6 Animal
36
For pharmacokinetics and biodistribution studies, male CF-1 mice (6-7 weeks
old, 25-30 g) were purchased from Charles River Laboratories (Wilmington, MA), and
the animal experiments were conducted at University of Southern California in
accordance with approved protocol by University of Southern California Institutional
Animal Care and Use Committee. For tumor targeting studies, NOD scid gamma
breeding pairs were purchased from The Jackson Laboratory (Bar Harbor, ME) and
bred in house at Academia Sinica, Taipei, Taiwan. To establish xenograft mouse model
of human breast cancer, MDA-MB-231 cells (10
6
cells/100 L PBS) were inoculated
subcutaneously at the backside of the female NOD scid gamma mice (6-8 weeks old,
~25 g), and the animal experiments were performed at the Genomic Research Center in
collaboration with Dr. Michael Hsiao at Genomics Research Center. The protocol has
been approved by Academia Sinica Institutional Animal Care and Use Committee. All
mice were housed at 12 h light/ 12 h dark cycles and had access to regular rodent diet
and water ad libitum.
3.2.7 Pharmacokinetics study
For pharmacokinetics studies, male CF-1 mice were administered with 0.5 mg/kg
of biotin-MonoTf, biotin-NonR-AggTf, or biotin-R-AggTf through lateral tail-vein injection.
At designated time points, 30 L of the blood sample was drawn from the saphenous
vein and mix with heparin solution as previously described [54]. The blood plasma
sample was isolated, diluted in PBS with 5% milk and 0.1% Tween-20, and 100 L of
the diluted samples were loaded into the StreptaWell (Roche Applied Science,
Penzberg, Germany). Subsequently, each well was incubated with anti-Tf antibody
(Abcam, Cambridge, United Kingdom, 1:1000 dilution) for 1 h, and followed by HRP-
37
conjugated rabbit anti-goat antibody (Sigma-Aldrich, 1:2000 dilution) incubation for 1 h.
TMB substrate (Kirkegaard & Perry Laboratories, Inc., MD) was added to develop color.
Subsequently, 0.25 M sulfuric acid solution was added to stop the reaction, and the
signal was detected at Ab450 by EnVision 2103 plate reader (Perkin Elmer, Waltham,
MA). The PK profile were analyzed by an Excel add-in software PKSolver using two-
compartmental model or non-compartmental model [55].
3.2.8 Biodistribution study
For Biodistribution studies, male CF-1 mice were administered with 2 mg/kg of
125
I-MonoTf or
125
I-NonR-AggTf through lateral tail-vein injection. At designated time
points, mice were euthanized, and blood, heart, lung, liver, spleen, kidney, muscle, and
brain were collected and weighted. For biodistribution analysis, tissue-associated
radioactivity was measured by the Cobra II gamma counter, and the radio-iodinated
protein amount in the tissue was estimated, then further normalized to the tissue weight.
For tissue retention analysis, the radio-iodinated protein amount in the tissues were
normalized to the tissue weight (T), the radio-iodinated protein amount in the plasma
were normalized with the plasma volume (P), and the T/P ratio can be obtained as
previously described [56].
3.2.9 Tumor targeting and retention study
NIR-MonoTf or NIR-NonR-AggTf Tf at the dose of 10 mg/kg was administered
into lateral tail vein when the tumor volume reached around 250 mm
3
(0.5*L*W
2
), and
the tumor targeting effect was assessed by detecting the near infrared signal using
Bruker In Vivo Xtreme (Billerica, MA). The mice fur was removed in order to achieve
38
better image quality without blocking the signal. At the designated time points post-
injection, an infrared filter set, excitation at 760 nm and emission at 830 nm, was used
to detect and acquire the fluorescence signal from IR800-labeled proteins, and all the
illumination settings (I.e. voltage lamp, exposure time, f/stop, field of views) were the
same across all the images. For ex vivo study, the tumor mass and blood plasma were
collected, weighed, and the fluorescence signal was acquired. The photon intensity
emitted from the tumor mass and plasma was analyzed by the imaging software
provided from Bruker, and then normalized to weigh and volume, respectively. To
analyze the retention effect of NIR-MonoTf and NIR-NonR-AggTf in the tumor mass, the
normalized signal from the tumor was divided by the normalized signal from the plasma
to obtain the T/P ratio as describe previously.
3.2.10 Tumor targeting studies of NIR-NonR-AggTf with excess Tf Blockage
NIR-NonR-AggTf at a dose of 5 mg/kg was co-administered with or without holo-
Tf at a dose of 250 mg/kg into MDA-MB-231 tumor bearing mice through lateral tail vein
injection, and the in vivo fluorescence images were acquired at 6 h post-injection.
Tumor mass and major organs, including lung, heart, liver, spleen, kidney, were
collected, and the NIR signals were acquired. The photon intensities emitted from the
tumor mass and different organs were analyzed, normalized to the weight.
39
3.3 Results
3.3.1 Cellular retention of NonR-AggTf and R-AggTf in MDA-MB-231 cells
The pulse-chase assays were performed in the MDA-MB-231 cells to confirm
whether the prolonged cellular retention of Tf oligomers could be observed in this TfR-
overexpressed breast cancer cell line. In MDA-MB-231 cells, around 70 % of the
125
I-
MonoTf was recycled back to the chase medium within 1 h of the chase phase (Figure
3.4.1 A, black solid line). Compared with the
125
I-MonoTf, significant longer cellular
retention in MDA-MB-231 cells was observed for
125
I-NonR-AggTf, with only 40% of the
125
I-NonR-AggTf was recycled to the chase medium at 1 h timepoint (Figure 3.4.1 A,
red dot line). Even at the 24 h of chase phase, still around 20% of the
125
I-NonR-AggTf
was associated with the cell, which was doubled compared with the percentage of cell-
associated
125
I-MonoTf. However, unlike the findings for NonR-AggTf, no prolonged
cellular retention in MDA-MB-231 cells was observed for
125
I-R-AggTf at any timepoints
of the chase phase (Figure 3.4.1 A, blue dashed line).
TCA precipitation assays were conducted to evaluate the percentage of
degraded radio-iodinated protein in the chase medium. No significant difference was
observed between the degraded percentage of
125
I-MonoTf and
125
I-NonR-AggTf before
3 h of the chase phase (Figure 3.4.1 B), but higher percent of degraded
125
I-NonR-
AggTf was observed at 24 h timepoint (Figure 3.4.1 B, red triangle). For
125
I-R-AggTf,
less percentage of the degraded
125
I-R-AggTf was found in the chase medium, and
similar percentage of degraded
125
I-R-AggTf was observed at 24 h of the chase phase
(Figure 3.4.1 B, blue square).
40
3.3.2 Binding profile of MonoTf, NonR-AggTf, and R-AggTf in MDA-MB-231 cells
Previously, the TfR crosslinking by multivalent Tf has been reported to cause the
prolonged retention and altered intracellular trafficking for the internalized ligand [37].
Therefore, competitive binding assays were performed to study whether Tf oligomers
can bind to multiple TfRs, exhibiting better binding affinity on the cell surface. In the
MDA-MB-231 cells, the
125
I-NonR-AggTf demonstrated a better binding profile
compared with the
125
I-MonoTf (Figure 3.4.2). Whereas, the
125
I-R-AggTf showed the
binding curve similar to the
125
I-MonoTf’s result (Figure 3.4.2, blue dashed line). After
fitting to the non-linear curve, the calculated IC50 for
125
I-MonoTf,
125
I-NonR-AggTf, and
125
I-R-AggTf were 2.9 g/mL, 25.4 g/mL, and 3.5 g/mL, respectively. These results
suggested that the concentration required for non-labeled MonoTf to replace 50% of the
125
I-NonR-AggTf binding to the TfR on the MDA-MB-231 cells was 25.4 g/mL, which
was around 8-fold higher than the result for
125
I-MonoTf.
3.3.3 Cellular uptake of NonR-AggTf and R-AggTf in MDA-MB-231 cells
The cell uptake assays were performed to characterize the internalized amounts
of
125
I-MonoTf,
125
I-NonR-AggTf, and
125
I-R-AggTf after binding to the MDA-MB-231
cells. The internalized amounts of
125
I-MonoTf increased rapidly from 0 to 30 min of the
incubation, and reached a plateau after 60 min (Figure 3.4.3 A, black solid line with
solid circle). Whereas slower but steady increase in the internalized
125
I-NonR-AggTf
amounts were observed throughout the study (Figure 3.4.3 B, red solid line with solid
triangle). As to
125
I-R-AggTf, the internalized amounts reached a plateau after 2 h of
incubation, but the amounts were much lower than that of
125
I-MonoTf or
125
I-NonR-
41
AggTf (Figure 3.4.3 B, blue solid line with solid circle). The internalization processes
of all radio-iodinated Tfs were significantly blocked by adding 100-fold of non-labeled
MonoTf (Figure 3.4.3 A, B, C, dotted lines), indicating the processes were through
TfR-mediated endocytosis.
In the body, there is significant amount of the endogenous serum Tf (~25 M) in
the blood circulation. In order to better mimic the environment in the body that the
injected Tf oligomer may need to compete the TfR binding site against the endogenous
serum Tf, cellular uptake assays with 20-fold non-labeled MonoTf in the dosing solution
were performed. In MDA-MB-231 cells, the internalized amounts of
125
I-MonoTf and
125
I-
R-AggTf were dramatically decreased with 20-fold non-labeled MonoTf competition
(Figure 3.4.3 A and C, solid line with open circle and square, respectively);
however, the phenomenon was not found in the
125
I-NonR-AggTf (Figure 3.4.3 B, solid
line with open triangle). These results indicated that the NonR-AggTf possessed a
better capability to compete the binding site against the endogenous MonoTf, and then
get internalized into the cells.
3.3.4 Pharmacokinetics parameters of NonR-AggTf and R-AggTf
After series of cell-based assays performed in MDA-MB-231 cells, the
comparison of the pharmacokinetics profiles between MonoTf and Tf oligomers was first
conducted to investigate the plasma concentration change of NonR-AggTf and R-AggTf
in the body overtime. After a single bolus intravenous dosing, the pharmacokinetics
curves for MonoTf, NonR-AggTf, and R-AggTf all demonstrated two phases of
exponential decline in plasma concentration, with the distribution phase between 0-2 h
post-injection, followed by the elimination phase after 2 h post-injection (Figure 3.4.4).
42
However, compared with the MonoTf, the plasma concentration of NonR-AggTf and R-
AggTf decreasec rapidly during the distribution phase, suggesting both NonR-AggTf
and R-AggTf entered the tissues rapidly after the injection. The PK results were fitted
with 2-compartmental model analysis by PKSolver to get the detailed PK parameters for
MonoTf, NonR-AggTf, and R-AggTf. The volume of the distribution of the peripheral
compartment is significantly higher for NonR-AggTf and R-AggTf. In addition, the half-
lives for NonR-AggTf and R-AggTf were shorter than the MonoTf. All the parameters
obtained from fitting the 2-compartmental model (2CA) and non-compartmental model
(NCA) are shown in Table 3.4.1
3.3.5 Biodistribution comparison between MonoTf and NonR-AggTf
The pharmacokinetics results revealed that a large amount of NonR-AggTf was
distributed to tissues after entering the blood circulation. Therefore, biodistribution
studies were performed to identify the major tissues that NonR-AggTf was resided after
injection. At 30 min post-injection, majority of the MonoTf was distributed to the liver,
spleen, and kidney (Figure 3.4.5 A, black bars). Majority of the NonR-AggTf was
distributed to the liver and the spleen, and the protein amounts were nearly 2-fold higher
compared with the MonoTf (Figure 3.4.5 A and B, black bars). Even at 2 h post
injection, the amounts of NonR-AggTf were still significantly higher than the amounts of
MonoTf in liver and in spleen (Figure 3.4.5 A and B, light gray bars). These results
supported the findings from the PK study that the volume of distribution for NonR-AggTf
was much higher than the MonoTf. However, even with a higher volume of distribution
observed in the PK study, we observed the trend that the amount of NonR-AggTf in the
heart tissue was lower than the MonoTf.
43
In order to evaluate the targeting and retention effect of NonR-AggTf in different
tissues, the radio-iodinated protein amount in the tissues were further normalized to
radio-iodinated protein amount in the plasma to obtain the Tissue/Plasma (T/P) ratios.
The T/P ratios of MonoTf ranged from 0.05 to 0.1 for all the tissues (except brain)
overtime, suggesting that MonoTf had more relative even distribution and retention in all
tissues (Figure 3.4.5 C). The statistic results by Dunnett’s multiple comparison test for
the MonoTf group also supported this finding (Table 3.4.2). Compared with MonoTf,
significant liver and spleen distribution was observed for NonR-AggTf at the 30 min post
injection (Figure 3.4.5 D, black bars). However, the retention effect was more obvious
in liver than in spleen for the NonR-AggTf, and the T/P ratio of liver did not decrease
rapidly overtime. The results also demonstrated that the kidney T/P ratio gradually
increased overtime, suggesting NonR-AggTf had a prolonged retention in the kidney
(Figure 3.4.5 D). The statistic results by Dunnett’s multiple comparison test for the
NonR-AggTf group showed the tissues and the timepoints that NonR-AggTf was
significantly distributed to in this biodistribution study (Table 3.4.3).
3.3.6 Tumor targeting and retention effect of MonoTf and NonR-AggTf
One of the applications of Tf-based carriers is to deliver the drug payload to the
tumor mass. In previous cell-based studies, prolonged cellular retention in the MDA-MB-
231 was found after the NonR-AggTf bound to the TfR on the cell surface. In addition,
with the enhanced binding profile, NonR-AggTf had a better capability to compete the
TfR binding site against MonoTf, leading to a higher internalization amount observed in
the competitive uptake assay in the MDA-MB-231 cells. In order to translate the results
from cell-based assay to in vivo models, near-infrared fluorescent dye labeled MonoTf
44
(NIR-MonoTf) and NonR-AggTf (NIR-NonR-AggTf) were administered into MDA-MB-
231 xenograft mouse model with the dose of 10 mg/kg by lateral tail-vein injection,
attempting to evaluate their tumor targeting and retention effects. In the real-time in vivo
imaging, no significant difference in the photon intensity at the tumor site was observed
between the mice injected with NIR-MonoTf or NIR-NonR-AggTf overtime (Figure 3.4.6
A). The quantified results from the ex vivo studies also suggested that the amounts of
NIR-MonoTf and NIR-NonR-AggTf distributed to the tumor mass were not significantly
different (Figure 3.4.6 B).
To evaluate the tumor retention effect, the NIR signals in the tumor mass were
normalized to the NIR signals in the plasma to obtain the Tumor/Plasma (T/P) ratios. As
shown in Figure 3.4.6 C, although no difference in the T/P ratio for tumor retention was
found at 1 h post injection, higher T/P ratio for NonR-AggTf was observed at 6 h and 24
h post injection, and there was a statistically-significant difference between NonR-AggTf
and MonoTf at 6 h post injection. In addition, the increase in T/P ratio for NonR-AggTf
appeared faster than the MonoTf overtime, indicating NonR-AggTf had a better
retention effect in the tumor mass.
3.3.7 Tumor targeting studies of NIR-NonR-AggTf with excess Tf blockage
To elucidate whether the targeting and retention effect of NonR-AggTf in tumor
mass was attributed to the active targeting effect from Tf-TfR interaction, TfR blockage
study as described previously was performed by co-administrating excess non-labeled
MonoTf and NIR-NonR-AggTf into MDA-MB-231 tumor bearing mice [19]. With excess
MonoTf competition, the signal of NIR-NonR-AggTf accumulated in the tumor mass was
partially abolished compared to the non-competition group at 6 h post-injection as
45
shown in Figure 3.4.7 A. Ex vivo experiments were followed to collect tumor, heart,
lung, liver, spleen, and kidney at 6 h post-injection, and the signal from NIR-NonR-
AggTf were quantified to compare the differences between two groups. As shown in
Figure 3.4.7 B and C, significant reduction in NIR signal was detected in the tumor and
kidney with excess Tf competition; whereas, the signal in liver and spleen did not exhibit
significant change even with excess fold of Tf competition. These results indicated that
the interaction between Tf units to TfR were essential for NonR-AggTf to maintain full
targeting capability to the tumor mass.
46
3.4 Figures and Tables
Figure 3.4.1 Cellular retention of NonR-AggTf and R-AggTf in MDA-MB-231 cells.
(A) In the pulse-chase assay, the sum of radioactivity value from the harvest medium
and cell lysate was considered as 100%. The percentage of the radioactivity detected in
the cell lysate was considered as “Cell Associated Transferrin”, and the results from
different timepoints was plotted. (*** here indicated the p value <0.01 between
125
I-MonoTf and
125
I-
NonR-AggTf groups. Some error bars may not be displayed because the values are lower than the scale
set of the graph) (B) Trichloroacetic acid (TCA) protein precipitation was added into the
harvest medium from pulse-chase assay to separate the degraded protein, and the
percentage of the degraded protein in the medium at different timepoints was plotted.
Data were presented as the mean values with the error bars indicating the standard
deviation (n=3). (*** here indicated the p value <0.001. Some error bars may not be displayed
because the values are lower than the scale set of the graph)
A B
47
Figure 3.4.2 Binding profile of MonoTf, NonR-AggTf, and R-AggTf to the TfR in
MDA-MB-231 cells.
125
I-MonoTf,
125
I-NonR-AggTf, or
125
I-NonR-AggTf with various
concentration of non-labeled MonoTf competition in the dosing medium were applied to
the binding assays in MDA-MB-231 cells. The results were analyzed by Prisim using
non-linear regression fitting to generate the fitting curves, and the IC50 was calculated.
Data were presented as the mean values with error bars indicating standard deviation
(n=3) (Some error bars may not be displayed because the values are lower than the scale set of the
graph).
A
48
Figure 3.4.3: Cellular uptake of MonoTf, NonR-AggTf and R-AggTf in MDA-MB-231
cells. Dosing medium containing (A)
125I
-MonoTf, (B)
125
I-NonR-AggTf, (C) or
125
I-R-
AggTf with various fold of non-labeled MonoTf competition were added to full confluent
MDA-MB-231 cells to perform the cellular uptake assay. The intracellular amounts of
radio-iodinated proteins at different timepoints were measured, and then normalized to
the cellular protein amounts. Data were presented as mean value with error bars
indicating the standard deviation. (n=3) (** here indicated the p value <0.01 between non-
competition and 100-fold non-labeled MonoTf competition groups. Some error bars may not be displayed
because the values are lower than the scale set of the graph)
A
B C
49
Figure 3.4.4: Pharmacokinetics study of MonoTf, NonR-AggTf, and NonR-AggTf.
CF male mice were administered with biotin-MonoTf (MonoTf), biotin-NonR-AggTf
(NonR-AggTf) or biotin-R-AggTf (R-AggTf), via intravenous tail-vein injection with the
dose of 0.5 mg/kg. Blood plasma was collected at 5 min, 1 h ,2 h, 4h, 8 h, and 16 h
post-injection, and the concentration of different forms of biotinylated-Tf was determined
by ELISA. Date were presented as the mean values with error bars indicating standard
deviation (n=4). (* here indicated the p value <0.01 between all groups. Some error bars may not be
displayed because the values are lower than the scale set of the graph)
50
Table 3.4.1: PK parameters of MonoTf, NonR-AggTf, and R-AggTf by two-
compartmental and non-compartmental analysis. (A) Illustration of two-compartment
model for PK analysis and associated parameters. (B) The data from Figure 4.4.1 was
analyzed by PKSolver using two-compartmental (2CA) or non-compartmental (NCA)
analysis to obtain the detailed PK parameters.
A
B
51
Figure 3.4.5: Biodistribution study of MonoTf and NonR-AggTf.
125
I-MonoTf or
125
I-
NonR-AggTf (dose: 2 mg/Kg per mouse) were injected intravenously into the CF-1 mice
(n=4, around 30 g). The radioactivity and the weight of the organs were measured, and
the equivalent amount of
125
I-MonoTf (A) and
125
I-NonR-AggTf (B) were calculated and
normalized to the tissue weight. The radio-iodinated protein amount in the tissue was
divided by the radio-iodinated protein amount in the plasma to get the T/P ratio for
MonoTf (C) or NonR-AggTf (D) to evaluate the tissue targeting and retention effect.
Date were presented as the mean values with error bars indicating standard deviation
(n=4)
A B
C D
52
Table 3.4.2: Dunnett’s multiple comparison test for the biodistribution of
125
I-
MonoTf. At each time point, the data sets of all T/P ratio of
125
I-MonoTf in all tissues
was compared with the T/P ratio of
125
I-MonoTf in individual tissue using Dunnett’s
multiple comparison test. (** represents p<0.01)
53
Table 3.4.3: Dunnett’s multiple comparison test for the biodistribution of
125
I-
NonR-AggTf. At each time point, the data sets of all T/P ratio of
125
I-NonR-AggTf in all
tissues was compared with the T/P ratio of
125
I-NonR-AggTf in individual tissue using
Dunnett’s multiple comparison test. (* represents p<0.05, ** represents p<0.01, ***
represent p<0.001)
54
Figure 3.4.6: Tumor targeting and retention study of MonoTf and NonR-AggTf in
MDA-MB-231 xenograft mice model. (A) Tumor bearing NOD scid gamma mice (n=4
for each group) were intravenously injected with NIR-MonoTf or NIR-NonR-AggTf with
the dose of 10 mg/kg per mice to investigate the tumor targeting and retention effect.
Real-time images of the NIR signals from 1 h to 24 h post-injection were record in three
representative mice, and the black arrowheads represent the tumor inoculation site. (B)
Ex vivo studies were performed to collect the tumor mass at 1 h, 6 h, and 24 h post
injection, and the net photon number of NIR signals in the tumor mass were quantified
and normalized with the weight of the tumor mass. The results were presented with the
average ± SD in the bar chart. (C) The quantified signals in the tumor mass were
normalized with the quantified signals in blood plasma to evaluate the retention effect of
MonoTf and NonR-AggTf in the tumor mass at 1 h, 6 h, and 24 h post injection. The
ratios were plotted in the bar chart with the average ± SD. (* represent the p value < 0.05)
A
B C
55
Figure 3.4.7: Biodistribution and tumor targeting study of OligoTf in MDA-MB-231
xenograft mice model with excess Tf blockage. MDA-MB-231 tumor bearing NOD
scid gamma mice (n=4 for each group, ~25 g/mouse) were intravenously injected with
NIR-NonR-AggTf (5 mg/kg) or NIR-NonR-AggTf mixed with 6 mg MonoTf to evaluate
the tumor targeting effect. Images for NIR signals were taken at 6 h post injection, and
the representative image was shown in (A). The black arrows indicate the tumor
inoculation site. Ex vivo studies were performed, and the representative images of the
NIR signal in different tissues were shown in (B). The net photon of the tumors and
organs from the ex vivo studies were quantified and normalized with the weight, and the
normalized results were plotted into bar chart shown in (C). (* represent p<0.05, ** represent p<0.01)
B
A
NIR-NonR-AggTf
NIR-NonR-AggTf
+ MonoTf
NIR-NonR-AggTf
+ MonoTf
NIR-NonR-AggTf
C
56
3.5 Discussion
In this chapter, series of cell-based assays in MDA-MB-231 cell line were
performed to investigate the biological properties of NonR-AggTf and R-AggTf. Pulse-
chase assays were first conducted to study the cellular retention of NonR-AggTf and R-
AggTf. Significant phenomenon of prolonged cellular retention was observed for NonR-
AggT, but not for R-AggTf, in the MDA-MB-231 cells (Figure 3.4.1 A). In addition,
higher percentage of degraded NonR-AggTf was observed in the harvest medium in the
pulse-chase assay (Figure 3.4.1 B). The prolonged cellular retention and increased
protein degradation phenomenon fit the cellular trafficking patterns of several
crosslinked receptor reported previously, including neonatal Fc receptor (FcRn),
intercellular adhesion molecule-1 (ICAM-1), as well as TfR [57, 58]. Compared with the
single ligand-receptor complex, which is majorly recycled back to the membrane surface
after endocytosis, the crosslinked multiple ligand-receptor complex exhibits altered
intracellular trafficking, resulting in prolonged cellular retention and higher lysosomal
targeting rate for the ligands. The enhanced binding profile for the NonR-AggTf
observed in MDA-MB-231 cell-based competitive binding assay supported the concept
that NonR-AggTf may bind to multiple TfR on the cell surface, leading to TfR
crosslinking, and resulting in prolonged cellular retention of the NonR-AggTf (Figure
3.4.2). However, higher degree but not all the internalized NonR-AggTf molecules were
degraded in the pulse-chase assays, suggesting that the majority of NonR-AggTf
molecules entered the slow recycling route rather than entering the lysosomal
compartment (Figure 3.4.1). This finding may because the final product of the Tf
57
oligomers were varied in size (Figure 2.4.2); therefore, not every NonR-AggTf molecule
could induce TfR crosslinking event.
The prolonged cellular retention of NonR-AggTf may also lead to the
accumulation effect in NonR-AggTf internalized amount observed in the time-course
cellular uptake assays. Unlike MonoTf, which the amount reached a plateau after 1 h of
incubation (Figure 3.4.3 A), the amount of cell-associated NonR-AggTf kept increasing
throughout the study (Figure 3.4.3 B). In order to mimic the environment that the
exogenous Tf-based molecules need to compete the binding site with endogenous Tf
after administered into the body [13], 20-fold and 100-fold of non-labeled Tf monomer
were added to modify the time-course uptake assays. In MDA-MB-231 cells, 100-fold
non-labeled MonoTf competition resulted in the significant inhibition in cellular uptake
for all three types of Tf molecules, suggesting that the internalization events were TfR-
mediated processes (Figure 3.4.3). However, 20-fold non-labeled MonoTf competition
inhibited the internalized amount of
125
I-MonoTf and
125
I-R-AggTf, but not the amount of
125
I-NonR-AggTf in the MDA-MB-231 cells. This phenomenon could be attributed to the
higher binding affinity of NonR-AggTf to the MDA-MB-231 cells observed in the
competitive binding assay. The NonR-AggTf may bind to multiple TfRs on the cell
surface; therefore, small amount of the MonoTf couldn’t compete all the binding sites
between NonR-AggTf and TfRs. These results indicated that NonR-AggTf may resist to
the competition from endogenous serum Tf in the circulation after being administered
into the body, leading to a higher cellular uptake amount in the cells with high TfR
expression level.
58
In order to evaluate whether the Tf oligomers, especially the NonR-AggTf, have
advantages in tissue or tumor targeting, series of in vivo studies were performed. The
pharmacokinetics properties and the biodistribution profiles of MonoTf and Tf oligomer
were first investigated in the wild type mouse model. In the PK studies, all plots for the
plasma concentration of NonR-AggTf, R-AggTf, and MonoTf presented two-
compartmental pharmacokinetics curves, with the distribution phase for the first 2 h post
injection, then followed by the elimination phase (Figure 3.4.4). Subsequently,
biodistribution studies were performed to elucidate where the NonR-AggTf and MonoTf
will be distributed after being injected into the body. NonR-AggTf were majorly
distributed to the liver and the spleen, with significantly higher amounts than MonoTf
during the distribution phase (Figure 3.4.5 B). Based on these findings, the
physicochemical property of Tf oligomers seems to be the major attribute to determine
their PK and biodistribution profiles in the mice. Particles with size around 50-100 nm in
diameter are often trapped in the liver and spleen after injection because both organs
contain highly fenestrated vascular structure and enriched mononuclear phagocytic
system (MPS) [59, 60], which may account for the high volume of distribution and high
clearance rate for Tf oligomers in the peripheral compartment (Table 3.4.1). Since the
PK curve for NonR-AggTf showed a dramatic plasma concentration decrease in
distribution phase, it was understandable to observed that NonR-AggTf had the lowest
mean residence time (MRT) and area under the curve (AUC) value when analyzed by
non-compartmental model (Table 3.4.1).
In the biodistribution study, the radio-iodinated protein amounts in the tissue were
normalized to the radio-iodinated protein amounts in the plasma to obtain the T/P ratio.
59
In protein therapeutics development, tissue to plasma (or tissue to serum) ratio is a
number to assess whether an radio-iodinated therapeutic protein has a trend of tissue
targeting or tissue retention [56]. Also, it is a way to compare the tissue targeting effect
between two protein molecules if the binding targets are the same. The T/P ratio of
125
I-
MonoTf was relatively even in all tissues except the brain, indicating that there is no
specific organ or tissue distribution pattern for the MonoTf over time (Figure 3.4.5 C
and Table 4.4.1). Whereas, NonR-AggTf possessed significant targeting effect to the
liver and the spleen right after intravenous injection. After 2 h post-injection, the
targeting and retention effect of NonR-AggTf to the liver was still observed, but the
effect to the spleen was diminished. At 8 h post-injection, the T/P ration of NonR-AggTf
in the liver was still significantly higher. In addition to the liver tissue, significant higher
T/P ration was found in the kidney at 8 h post-injection. Two possible reasons may
explain the phenomenon. First, the increase in T/P ratio may due the prolonged
retention of NonR-AggTf in kidney tissue after binding to the TfR since TfR was reported
to express on the surface of the tubular epithelium [15]. The other explanation may due
to the increase degradation rate for NonR-AggTf after injected into the body, leading to
the increased residual NonR-AggTf amounts accumulated in the kidney over time.
The tumor targeting studies of NonR-AggTf and MonoTf was conducted in the
MDA-MB-231 xenograft mouse model. The amount of NonR-AggTf resided in the tumor
mass was comparable to that of MonoTf throughout the study (Figure 3.4.6 A and B).
Since majority of the NonR-AggTf would be cleared rapidly through the MPS after
injection, lower plasma concentration left available for further tumor targeting could
account for the compromised amount of NonR-AggTf in the tumor mass. Nevertheless,
60
the T/P ratios indicated that NonR-AggTf had a better retention effect in the tumor mass
compared with the MonoTf (Figure 3.4.6 C), which was similar to the findings from the
cell-based studies. However, strategies such as stealth coatings may be required to
avoid the fast clearance from MPS to increase the amount of multivalent protein
nanoparticle that could be distributed to the tumor mass.
Tumor targeting effect could be achieved by both passive or active targeting for
the surface-modified nanoparticles; however, the relative contribution of these two
targeting mechanisms is always under debate [61]. In our studies, the NonR-AggTf can
be also considered as a surface-modified nanoparticle with its size around 50 nm in
diameter, and with the capability to bind TfR. Therefore, in vivo TfR blockage was
performed in the MDA-MB-231 xenograft model to elucidate whether Tf-TfR binding was
involved in the tumor or liver targeting observed in the previous animal studies. It is
estimated that the total weight of Tf for a 25 g mouse is around 2.9 to 3.6 mg (Tf
concentration in the body is around 25 M, and the blood value for a 25 mg mouse
ranges from 1.46 – 1.8 mL). in other word, the NIR-NonR-AggTf with the injection dose
of 5 mg/kg already faced 23 to 28-fold of endogenous Tf competition. With the non-
labeled MonoTf co-administration, the Tf competition increased to nearly 75-fold. Co-
administration of 50-fold of non-labeled MonoTf significantly reduce the amount of NIR-
NonR-AggTf targeted to the tumor mass (Figure 3.4.7 B), and the signal was
decreased by nearly 35 % compared with the non-coadministration group (Figure 3.4.7
C). These results were similar to the data from the cell-based competitive binding
assays under 30-fold and 100-fold of non-labeled MonoTf competition (% of binding
dropped from 70% to 47%) (Figure 3.4.2), suggesting that the receptor-mediated active
61
targeting was essential for the NonR-AggTf to target the tumor mass. Although
transferrin receptor 2 (TfR2) is highly expressed in the liver, which can also bind to Tf
[62], the amount of NIR-NonR-AggTf in the liver did not change, indicating that the
passive targeting and the MPS recognition could be the major mechanism for the NonR-
AggTf to be distributed in the liver. The NIR signal in the kidney was significantly
decreased with non-labeled MonoTf co-administration (Figure 3.4.7 B and C). Similar
to the findings from the biodistribution study in the wildtype mice, two possible reasons
could explain this phenomenon. The first reason could be that the excess non-label
MonoTf could compete the binding between NIR-NonR-AggTf to the TfR on the surface
of the tubular epithelium in the kidney, which accounts for the decrease of the NIR
signal in kidney. The other possible reason was that the MonoTf competition may
attenuate the endocytosis of NIR-NonR-AggTf, leading to lower degradation rate of NIR-
NonR-AggTf in the cells, resulting to less NIR amount excreted through renal filtration.
3.6 Summary
In this chapter, the biological properties of Tf oligomer, especially the NonR-
AggTf, were characterized through both in vitro and in vivo studies. The NonR-AggTf
demonstrated an enhanced binding affinity to the TfR
high
MDA-MB-231 cells, which may
due to the binding between NonR-AggTf and multiple TfRs. The enhanced binding for
NonR-AggTf on the surface led to prolonged cellular retention and increased
degradation rate observed in the MDA-MB-231 cells. In addition, the enhanced binding
allowed the NonR-AggTf to better compete with Tf to the TfR binding site, leading to
higher internalized amount in the cellular uptake assay. The prolonged retention of
NonR-AggTf in the tumor mass was also observed in the MDA-MB-231 xenograft
62
model, and the active targeting property is crucial for the NonR-AggTf to achieve the
tumor targeting effect. These findings suggest that Tf oligomer could be a good carrier
to deliver drug payload to the TfR
high
tumor mass.
63
CHAPTER 4: EVALUATION OF THE POTENTIAL APPLICATION OF
THE TF OLIGOMER IN TRANS-EPITHELIAL PROTEIN DRUG
DELIVERY
4.1 Background
The other important application of using Tf as a drug carrier is to deliver protein
therapeutics across the gastrointestinal (GI) epithelium. Previously, oral administration
of insulin-Tf conjugate demonstrated a significant reduction of the blood glucose level in
STZ-induced diabetes rat model [21]. In addition, hGH and G-CSF was fused to Tf by
using DNA recombinant technique, and the fusion proteins can achieve weight gain or
myelopoietic effect after oral administration in animal model, respectively [20, 22].
These results suggest that, by either chemical conjugation or DNA recombinant, linking
Tf moiety to protein therapeutics is a promising strategy to deliver protein drugs through
oral administration. However, the bioavailability or Tf-based protein therapeutics after
oral dosing are still low, and two factors may limit the transcytosis efficiency for Tf-
based protein therapeutics: low TfR expression on the apical side of membrane and
rapid recycling to apical surface after endocytosis [63]. Therefore, there is a high
demand to find new ways to improve the transcytosis efficiency for Tf-based protein
drugs
In order to assess whether a drug candidate possess a better permeability
property or a transcytosis capability across the GI epithelium, Caco-2 cell is the classic
cell model to establish the polarized epithelium in the in vitro setting. Caco-2 cells have
been reported to develop the morphological characteristics of normal enterocytes when
64
grown on the dish with nitrocellulose filter [64]. The development of the tight junction in
the Caco-2 cells monolayer will increase the transepithelial electrical resistance (TEER),
and the polarized cell monolayer can be used to evaluate the permeability of therapeutic
agents [64]. The polarized Caco-2 monolayer has been also used to study the
transcytosis ability of insulin-Tf conjugate in our laboratory previously [65]; therefore,
this system is a suitable model for us to evaluate the transcytosis capability of Tf
oligomers
In previous chapters, the Tf oligomer was prepared by serial steps of chemical
reactions to link Tf moieties together. Although the chemical conjugation is the most
direct method to generate Tf oligomers, the end-products are always heterogenous. In
addition, the manufacturing process requires several steps of purification and dialysis,
which is time demanding, and the production yield is decreased due to complicated
manufacture process. Further, the chemical linker may be not biocompatible or
biodegradable, which may cause toxicity-related problems. In order to avoid these
issues, DNA recombinant technique could be a potential solution to produce the desired
protein molecules with homogenous end-product. Additionally, the recombinant DNA
techniques simplify the process for protein production and purification. Also. protein
molecules are composed of nature amino acids, which eliminates the concern of
potential toxicity problems after being administered to the body [66]. Furthermore,
protein produced from recombinant DNA technique leaves the surface reactive residues
available for further chemical conjugation, which can be linked to drug payloads,
reporters, or PEG for various purposes. However, there was no report of the methods to
65
produce recombinant Tf oligomer. Finding the proper strategies to oligomerize protein
moieties via recombinant DNA technique could be challenging.
Several proteins have been found to form oligomeric structure naturally, and the
oligomerization process is crucial for these proteins to involve in several molecular and
cellular actions, such as clathrin assembly or p53-mediated gene transcription [67, 68].
Different kind of protein domains responsible for oligomerization process, including
dimerization or tetramerization, have been found and characterized in detail to study the
oligomerization process in these protein structures. One of the important dimerization
domains is leucine zipper domain, which consists of amphipathic helix that dimerized
in parallel to form a coiled-coil structure [69]. The protein sequence of leucine zipper
motif is composed of heptad repeat of amino acids, which the positions of the heptad
repeats are labeled a-g [70]. The hydrophobic interaction formed between the nonpolar
acid residues at a position and leucine at d position pack in the canonical “knobs and
holes” structure, providing the stabilizing force for dimerization. Leucine zipper is a
critical motif for B-ZIP (basic-region leucine zipper) types of transcription factors in
eukaryotes. This motif allows the dimerization of B-ZIP proteins, such as c-Jun and c-
Fos, and the dimerized protein will further bind to the major groove of the DNA to start
the transcription process, which have critical roles in regulating cell growth [71].
The concept of forming “knobs and holes” through leucine zipper motif has been
also utilized in antibody engineering to form bispecific antibody. The first bispecific using
the leucine zipper sequence from c-fos and c-jun was introduced in 1992, and this anti-
CD3 Fab’-Fos and anti-IL-2R Fab’-Jun bispecific antibody was demonstrated to be
highly effective in recruiting cytotoxic T-cell to lyse with HuT102 cells in vitro [72]. This
66
innovative method suggested that incorporating leucine zipper motif in protein
engineering is an approachable strategy to design the desired dimerized protein for
novel biologics development.
In this chapter, the cellular retention of NonR-AggTf and R-AggTf was first
studied in polarized Caco-2 cells to investigate whether the similar effect observed in
MDA-MB-231 cells can be also seen in polarized Caco-2 cells. Furthermore, the
transcytosis assays were performed to compare the transcytosis capability between
MonoTf and two forms of Tf oligomers. Subsequently, NonR-AggTf and R-AggTf were
orally administered to wild type mice, attempting to compare the serum concentration of
MonoTf, NonR-AggTf, and R-AggTf after oral dosing.
In addition, the leucine zipper motif was applied as a recombinant linker in this
chapter, attempting to generate dimeric Tf-based recombinant fusion protein. After
protein production and characterization, the cellular retention and transcytosis capability
of monomeric Tf-based fusion protein and dimeric Tf-based fusion protein were
investigated in polarized Caco-2 cells, trying to propose a novel strategy to improve the
transcytosis efficiency for Tf-based recombinant fusion protein.
67
4.2 Methods
4.2.1 Cell culture
Human colorectal adenocarcinoma Caco-2 cell line and human embryonic kidney
cell HEK293 were purchased from ATCC (Manassas, MA). Caco-2 cells were grown in
DMEM media supplemented with 20% FBS, 2 mM L-glutamine, 50 U/mL penicillin, and
50 g/mL streptomycin. HEK293 cells were grown in DMEM media supplemented with
10% FBS, 2 mM L-glutamine, 50 U/mL penicillin, and 50 g/mL streptomycin. Rat
lymphoma Nb-2 cells were grown in RPMI-1640 media supplemented with 10% FBS, 2
mM L-glutamine, 50 U/mL penicillin, and 50 g/mL streptomycin. All cells were
incubated in humidified atmosphere at 37°C containing 5% CO2, and subcultured prior
to full confluence using trypsin/EDTA.
4.2.2 Oligonucleotide primers
C2 linker_F: 5'-pTCGAGGCTGGTTGTAAAAATTTTTTCACTAGTTGTGGTTCTC-3'
C2 linker_R: 5'-pTCGAGAGAACCACAACTAGTGAAAAAATTTTTACAACCAGCC-3'
hinge linker_F: 5'-
GCAACTCGAGGACAAAACTCACACATGCCCACCGTGCCCACTCGAGTTGC-3'
hinge linker_R: 5'-
GCAACTCGAGTGGGCACGGTGGGCATGTGTGAGTTTTGTCCTCGAGTTGC-3'
CH4C linker_F: 5'-
GCAACTCGAGTGCGCTGAAGCTGCAGCCAAAGAAGCTGCAGCCAAAGAGGCCGC
AGCTAAAGAAGCCGCAGCTAAAGCTTGCCTCGAGTTGC-3'
68
CH4C linker_R: 5'-
GCAACTCGAGGCAAGCTTTAGCTGCGGCTTCTTTAGCTGCGGCCTCTTTGGCTGC
AGCTTCTTTGGCTGCAGCTTCAGCGCACTCGAGTTGC-3'
L Zipper_F 80bp: 5'-
GCAACTCGAGCGCGTCCAGCAGAAGCACCACCTGGAGAATGAGAAGACGCAGCT
CATTCAGCAGGTGGAGCAGCTTAAGC-3'
L Zipper_R 80 bp: 5'-
GCAACTCGAGGCACTTGACCTTGTAGGCGTCTCTCTCGCGGGCCAGCCGGGACA
CCTCCTGCTTAAGCTGCTCCACCTGC-3'
L Zipper PCR_F: 5'-GCAACTCGAGCGCGTCCAGC-3'
L Zipper PCR_R: 5'-GCAACTCGAGGCACTTGACC-3'
hGH_NF: 5'-CTAAAGGACCTAGAGGAAGGCATC-3'
Tf_CR: 5'-CTCACAGTTTTATCAGGGACCTCGAG-3'
4.2.3 Molecular cloning
All the oligonucleotides required for molecular cloning were synthesized by
Integrated DNA Technologies (Stokie, IL). The hGH-H4-Tf expressing plasmid in pcDNA
3.1 (+) backbone (Invitrogen, Carlsbad, CA) as previously described [19] was used to
construct the following fusion protein expressing plasmids: hGH-C2-Tf, hGH-CH4C-Tf,
hGH-hinge-Tf, and hGH-Lzip-Tf. The oligonucleotide for C2, CH4C, and hinge linker
were heated at 95 °C for 2 min to disrupt any potential secondary structures, and then
slowly cooled down to room temperature to allow the annealing process. The annealed
69
oligonucleotides were further digested by restriction enzyme XhoI (NEB, Ipswich, MA),
and then purified by QIAquick
®
PCR purification kit (Qiagen, Hilden, Germany). The
purified XhoI-digested oligonucleotides were served as “insert” in the ligation process.
The hGH-H4-Tf expressing plasmid were first digested by XhoI restriction enzyme to
remove the DNA for the H4 linker, and then incubated with calf-intestinal alkaline
phosphatase (CIP) (NEB, Ipswich, MA) to remove the terminal phosphate groups. The
XhoI-digested hGH-H4-Tf plasmid DNA was extracted by QIAquick gel extraction kit
(Qiagen, Hilden, Germany), and then served as the “vector” for the ligation process. The
cloning strategies for hGH-C2-Tf, hGH-CH4C-Tf, hGH-hinge-Tf expressing plasmid
construction are illustrated in Figure 4.4.4 A.
The leucine zipper interface from MAFB were chosen to design the leucine
zipper linker inserted between hGH and Tf moieties [73, 74]. For hGH-Lzip-Tf
expressing plasmid construction, the two-step PCR were performed to generate the full
length of oligonucleotide for leucine zipper linker. The full-length leucine zipper linker
DNA fragment was digested with XhoI restriction enzyme, and then ligated to the vector
as described previously. The cycles for two-step PCR were listed below, and the cloning
strategy for hGH-Lzip-Tf expressing plasmid construction is illustrated in Figure 4.4.4 B
70
PCR cycle:
The insert DNA samples and vector DNA were mixed, and the DNA ligase was
added to conduct DNA ligation at room temperature for 16 h. The ligation products were
transformed to DH5 , and the transformed DH5 bacterial cells were grown on LB agar
with 100 g/mL of ampicillin. Single colonies were picked, and PCR using hGH_NF and
Tf_CR as the primer sets was conducted to examine whether the DNA sequence for
various linkers were inserted between the hGH and Tf open reading frame. Using hGH-
Tf expressing plasmid as the template, the PCR DNA product with the size around 270
bp was served as negative control. On the other hand, the PCR product using hGH-H4-
Tf as the template was served as the positive control for the linker DNA sequence
insertion, which the product size is around 330 bp.
4.2.4 Recombinant protein production and preparation
For small scale protein production, HEK293 cells were seeded in the 6-well
plates. When reaching 70-80% confluence, the cells were transfected with protein
expressing plasmids by polyethylimine (PEI)-mediated transfection method, and the
1. 95°C 3 min
2. 95°C 30 sec
3. 67°C 1 min
4. 72°C 1 min
Repeat the steps from 2-4 for 4 more cycles
5. 72°C 2 min
6. 95°C 5 min
7. 95°C 30 sec
8. 61°C 30 sec
9. 72°C 1 min
Repeat the steps from 7-9 for 34 more cycles
10. 72°C 5 min
11. 4°C Stop
71
cells in each well were incubated in the 2 mL of CD293 medium (Invitrogen, Carlsbad,
CA) supplemented with 4mM L-Glutamine. After 4 days of protein production, the
CD293 medium was harvested, sterilized, and stored at 4 °C for further analysis.
For large scale protein production, HEK293 cells were seeded in 8 T-175 flasks.
PEI-mediated transfection was performed when the cell confluence reached around 70-
80%. The transfected cells in each flask were incubated with 63 mL of CD923 medium
supplemented with 4mM L-Glutamine to harvest the protein. At day 4 post-transfection,
the first batch of CD293 medium were harvest, and the second batch of CD293 medium
was added for another 3-day incubation. At day 7 post-transfection, the second batch of
the CD293 medium was harvested, and the protein samples were concentrated by lab
scale tangential flow filtration (Millipore, Burlington, MA) from 1 L down to 30 mL.
Subsequently, dialysis was performed to switch the buffer system to PBS, and the
protein samples were sterilized and stored at -20 °C for further analysis.
4.2.5 Western blotting
The fusion protein samples were mixed with reducing or non-reducing protein
sampling buffer, and the SDS-PAGE and protein transfer was conducted for Western
blotting. For antibody information, the goat anti-Tf polyclonal antibody (Sigma-Aldrich,
T2027) with the dilution ratio 1:5000 was used to detect the Tf moiety of the fusion
proteins, and the horseradish peroxidase (HRP)-conjugated rabbit anti-goat IgG
antibody (Sigma-Aldrich, A5420) with the dilution ratio 1:10000 was used to detect the
anti-Tf antibody and to develop chemiluminescence signal when reacting with ECL
reagents.
72
4.2.6 Caco-2 cell-based pulse-chase and transcytosis assay
To establish the polarized Caco-2 cells for the pulse-chase assay and
transcytosis assays, 0.5*10
6
Caco-2 cells were seeded in each transwell in the 6-well
plate setting, and the transepithelial electrical resistance value was measured
throughout the 21-day culture period to monitor the polarity establishment in Caco-2
cells. All the Caco-2 cell-based pulse-chase assays and transcytosis assays were
performed after the polarity is well-established in the Caco-2 cell monolayer.
To study the cellular retention of hGH-H4-Tf and hGH-Lzip-Tf in Caco-2 cells, the
apical chambers were added with
125
I-hGH-H4-Tf or
125
I-hGH-Lzip-Tf with the Tf moiety
amount equivalent to 4 g in the serum-free DMEM supplemented with 0.1% BSA. After
incubating at 37 °C for 30 min, the dosing medium was removed, and the cells were
washed with ice-cold PBS, then replenish with harvest medium (serum-free DMEM
supplemented with 0.1% BSA and 400 g of non-labeled MonoTf). At 1 h of the chase
phase, the medium in apical and basolateral reservoir were harvest, and the cells were
dissolved in 1N NaOH. The radioactivity of all three samples were measured, and the
percentage of radio-iodinated protein was in three compartments were calculated.
To evaluate the transcytosis capability of hGH-H4-Tf and hGH-Lzip-Tf in Caco-2
cells, the apical chambers were added with
125
I-hGH-H4-Tf or
125
I-hGH-Lzip-Tf with the
Tf moiety amount equivalent to 4 g in the serum-free DMEM supplemented with 0.1%
BSA, and the rest of the experimental procedures are the same as the transcytosis
assay described previously.
4.2.7 Nb2 cell-based proliferation assay
73
Nb2 cells were seeded (15,000/well) in 96-well plates in 200 L assay medium
(RPMI1640 medium supplemented with 2 mM L-glutamine, 10% horse serum, 50
units/mL penicillin, 50 g/mL streptomycin, and 50 mM 2-mercaptoethanol), and the
seeded cells were cultured for 24 h. The fusion protein samples and recombinant hGH
(Somatropin, LGC Standard USP1615708, United States Pharmacopeia Convention,
Rockville, MD) were diluted into the serum-free assay medium, and the equivalent hGH
doses ranging from 0.03 ng/mL to 100 ng/mL were added to the serum-starved Nb2
cells. After 4-day incubation, cells were incubated with 20 L of resazurin solution
(Biotium, Hayward, CA) for 16 h. The absorbance was measured at 570 nm and 595 nm
using a Genios microplate reader (Tecan, San Jose, CA) and normalized against the
vehicle control. The data were collected as triplicates and analyzed by sigmoidal curve
fitting in GraphPad Prism6.
4.2.8 Pharmacokinetics of MonoTf and NonR-AggTf with oral administration
For oral pharmacokinetics studies, male CF-1 mice were administered with 40
mg/kg of biotin-MonoTf or biotin-NonR-AggTf in PBS mixed with 30 mg/mL sodium
bicarbonate through gavage feeding. At 8 hours post feeding, whole blood sample were
drawn from cardiac puncture, and the blood serum sample were isolated. The serum
sample was diluted in PBS containing 5% milk and 0.1% Tween-20 with 1:1 ratio, and
100 L of the diluted samples were loaded into the StreptaWell (Roche Applied
Science, Penzberg, Germany). Subsequently, each well was incubated with anti-Tf
antibody (Abcam, Cambridge, United Kingdom, 1:1000 dilution) for 1 h, and followed by
HRP-conjugated rabbit anti-goat antibody (Sigma-Aldrich, 1:2000 dilution) incubation for
1 h. TMB substrate (Kirkegaard & Perry Laboratories, Inc., MD) was added to develop
74
color. Subsequently, 0.25 M sulfuric acid solution was added to stop the reaction, and
the signal was detected at Ab450 by EnVision 2103 plate reader (Perkin Elmer, Waltham,
MA).
75
4.3 Results
4.3.1 Transcytosis capability of NonR-AggTf and R-AggTf
Since the prolonged cellular retention was observed in MDA-MB-231 cells for
NonR-AggTf, series assays were performed to address whether the prolonged cellular
retention of NonR-AggTf or R-AggTf could be observed in polarized Caco-2 cells.
Pulse-chase assays in polarized Caco-2 cells were first conducted to study the cellular
retention pattern of NonR-AggTf and R-AggTf. At 6 h of the chase phase, although
majority of the endocytosed
125
I-NonR-AggTf and
125
I-R-AggTf were recycled back to
the medium at the apical side, higher percentage of the
125
I-NonR-AggTf and
125
I-R-
AggTf were found to be associated with the cells or be transported to the medium at the
basolateral side compared with the
125
I-MonoTf (Figure 4.4.1). In order to study whether
higher cellular retention of NonR-AggTf and R-AggTf may result in higher transcytosis
efficiency in polarized Caco-2 cells, the transcytosis assays were conducted to examine
the amount of
125
I-MonoTf,
125
I-NonR-AggTf and
125
I-R-AggTf transported across the
polarized Caco-2 epithelium over time. Compared with the
125
I-MonoTf, the total protein
amount of
125
I-NonR-AggTf and
125
I-R-AggTf released to the basolateral medium were
significantly higher over time (Figure 4.4.2 A). However, the TCA protein precipitation
assays revealed that the amount of intact
125
I-NonR-AggTf and
125
I-R-AggTf were much
less than the amount of intact
125
I-MonoTf (Figure 4.4.2 B), suggesting the protein
degradation rates were much higher for
125
I-NonR-AggTf and
125
I-R-AggTf during the
transcytosis process.
4.3.2 Comparison of the serum concentration of MonoTf and NonR-AggTf after
oral administration
76
In order to find out whether the NonR-AggTf with a prolonged cellular retention
effect may lead to higher serum concentration detected after oral administration, the
biotin-MonoTf and biotin-NonR-AggTf were administered through gavage feeding in the
CF-1 wildtype mouse. At 8 h post feeding, although the serum concentration of biotin-
MonoTf had a large variation, the serum concentration of biotin-NonR-AggTf was nearly
non-detectable (Figure 4.4.3). Combined with the findings from the in vitro transcytosis
assay, Tf oligomer may not be a suitable carrier to deliver protein therapeutics across
the GI epithelium.
4.3.3 Cloning of hGH-C2-Tf, hGH-CH4C-Tf, hGH-hinge-Tf, and hGH-Lzip-Tf
expressing plasmids
In previous sections and chapters, chemically-synthesized Tf oligomer was used
to perform several studies to evaluate its potential application in drug delivery. However,
protein oligomerization through chemical reactions has several challenges as
mentioned in the background part at the beginning of this chapter. Some of the issues,
such as heterogenous end-products and complicated production process, could be
resolved by using recombinant DNA technique to produce the protein of desired.
However, the methods of using recombinant DNA technique to produce Tf oligomer is
still unknown. To test whether Tf can be oligomerized or dimerized through fusing to the
natural protein oligomerization motifs, the leucine zipper motif protein sequences were
planned to insert into the Tf-based model fusion protein.
Several recombinant protein-based linkers containing cysteine residues were
designed to be inserted between hGH and Tf moieties, including the cyclo linker without
the thrombin cutting site [44], helical linker with cysteine residues at both ends, Fc hinge
77
minimal domain [75], and leucine zipper motif with cysteine residue at the c terminus.
The purpose of the introducing the cysteine residues was to stabilize the possible
dimeric structure through disulfide linkage. The cloning strategies of hGH-Tf fusion
protein with various linkers were depicted as shown in Figure 4.4.4
The ligated product was transformed into the DH5 bacteria, and the single
colonies were picked to perform PCR using hGH_NF and Tf_CR as the forward and
reverse primer to confirm the insertion of the oligonucleotides between hGH open
reading frame and Tf open reading frame. For hGH-C2-Tf expressing plasmid
construction, 3 possible clones were obtained (#2, #5, and #6) after the ligation and
transformation process (Figure 4.4.5). After sequencing, clone #5 was the correct one
containing hGH-C2-Tf expressing plasmid. For hGH-hinge-Tf expressing plasmid
construction, 8 possible clones (#2-#8, and #11) were obtained. Among them, clone #4
and clone #6 had the correct hGH-hinge-Tf expressing plasmid (Figure 4.4.6). For
hGH-CH4C-Tf expressing plasmid construction, 4 possible clones (#1, #2, #4 and #7)
were obtained. Among them, the clone #1 contained the correct hGH-CH4C-Tf
expressing plasmid (Figure 4.4.7). For hGH-Lzip-Tf expressing plasmid construction,
11 possible clones were obtained. (#1-#3, #5-#12). Among them, clone #9 and #10
contained the correct hGH-Lzip-Tf expressing plasmid (Figure 4.4.8 and Figure 4.4.9).
In total, four fusion protein expressing plasmid were constructed to investigate whether
different linker design with cysteine residues could lead to protein dimerization. The
expected amino acid sequence for various recombinant protein linkers are shown in
Table 4.4.1, and the proposed linker designs are illustrated in Figure 4.4.10.
78
4.3.4 Protein production and Western blotting analysis of recombinant hGH-Tf
fusion protein with various linkers
In order to confirm whether the hGH-Tf fusion protein with various linkers can
form a stable dimerized protein through disulfide linkage, Western blotting was
performed to analyze the recombinant fusion protein products produced from HEK293
cells. As shown in Figure 4.4.10 B, in reducing and non-reducing sampling buffer, only
one form, the monomeric form, of the hGH-H4-Tf fusion protein can be detected by
Western blotting. The dimerization phenomenon of the fusion protein was not found by
adding two cysteine residues in the cyclo linker, either with or without the thrombin
cutting site (Figure 4.4.10 A and B). The dimeric form of the recombinant fusion protein
under the non-reducing condition became detectable after the linker was changed to
CH4C linker or the Fc mini hinge domain linker. However, the major form of these two
recombinant hGH-Tf fusion proteins were still in monomeric form. Different from these
fusion proteins listed above, the major form of the recombinant hGH-Tf fusion protein
with Lzip linker under non-reducing environment was in dimeric form (Figure 4.4.10 B).
This dimeric structure of fusion protein could be dissociated to monomeric form by
adding the reducing agent DTT, suggesting the disulfide linkage was one of the critical
forces to maintain the dimeric structure.
4.3.5 Biological activity evaluation of the hGH-Lzip-Tf fusion protein on Nb2 cells
In order to confirm the biological activity of the hGH moiety of the hGH-Lzip-Tf
dimeric fusion protein, Western blotting and Nb2 based proliferation assay were
performed. Western blotting was first performed to confirm that the structure of hGH
was not significantly disrupted after the insertion of different linkers in the fusion proteins
79
(Figure 4.4.11 A). Subsequently, the Nb2 proliferation assay was performed to
compare the biological activity of the hGH moieties on hGH-H4-Tf and hGH-Lzip-Tf
fusion proteins. Proliferation of Nb2 cells was observed after adding the hGH-Lzip-Tf in
the dosing medium, suggesting the biological activity of hGH moiety on the hGH-Lzip-Tf
fusion protein was maintained. However, compared with the biological activity of native
hGH and hGH-H4-Tf fusion protein, the EC50 of hGH-Lzip-Tf was the highest (12.34
ng/mL vs 1.36 ng/mL for hGH and 2.22 ng/mL for hGH-H4-Tf) (Figure 4.4.11 B),
indicating that the activity of the hGH moiety on the dimeric fusion protein was partially
attenuated.
4.3.6 Comparison of the cellular retention and the transcytosis capability between
hGH-H4-Tf and hGH-Lzip-Tf in polarized Caco-2 cells
In previous chapter, the prolonged cellular retention of oligomeric Tf was
observed in polarized Caco-2 cells. Here, we tried to confirm whether the hGH-Lzip-Tf
fusion protein with a dimeric Tf moiety demonstrated prolonged cellular retention or not
when comparing with the hGH-H4-Tf, a monomeric Tf-based fusion protein. Pulse-chase
assay in polarized Caco-2 cells were performed to compare the cellular retention
pattern between hGH-Lzip-Tf and hGH-H4-Tf. At 1 h of the chase phase, no significant
difference was observed in the percentage of cellular associated
125
I-hGH-H4-Tf and
125
I-hGH-Lzip-Tf (Figure 4.4.12), suggesting the dimeric Tf moiety may not prolong the
cellular retention in the polarized epithelium.
The transcytosis capabilities of hGH-H4-Tf and hGH-Lzip-Tf were also evaluated
in the polarized Caco-2 cells. Compared with the
125
I-hGH-H4-Tf fusion protein, the total
protein amount in mass of
125
I-hGH-Lzip-Tf transcytosed across the polarized epithelium
80
was around 2-fold higher. TCA protein precipitation assays were performed to measure
the intact amount of protein in the total transcytosed protein in the basolateral reservoir.
No significant difference in intact protein percentage was found between hGH-H4-Tf and
hGH-Lzip-Tf (Figure 4.4.13 B); in other word, the intact protein amount of transcytosed
hGH-Lzip-Tf is 2-fold higher the hGH-H4-Tf (Figure 4.4.13 A).
81
4.4 Figures and Table
Figure 4.4.1 Cellular retention of NonR-AggTf and R-AggTf in polarized Caco-2
cells. The dosing medium containing
125
I-MonoTf,
125
I-NonR-AggTf, or
125
I-R-AggTf
were added at the apical reservoir at the start of the pulse phase, and the percentage of
the radio-iodinated proteins in different compartment were measured at 6 h of the chase
phase. The data were represented as the mean values with error bars indicating the
standard deviation (n=3) (** here indicated the p value <0.01. Some error bars may not be displayed
because the values are lower than the scale set of the graph).
82
Figure 4.4.2: Transcytosis assay of MonoTf, NonR-AggTf and R-AggTf in
polarized Caco-2 cells. (A) The dosing medium containing radio-iodinated Tfs were
added at the apical reservoir, and the medium at the basolateral reservoir was
harvested after 30, 1 h, 2 h, 4 h, 6 h of incubation to measure the radioactivity, and to
determine the transcytosed total protein amount. (B) The TCA protein precipitation
assays were performed to measure the amount of intact radio-iodinated proteins in the
basolateral reservoir at designated timepoints. The data were represented as the mean
values with error bars indicating the standard deviation (n=3). (*** here indicated the p value
<0.001. Some error bars may not be displayed because the values are lower than the scale set of the
graph).
A
B
83
Figure 4.4.3: Serum concentration of MonoTf and NonR-AggTf in CF-1 wild type
mice after oral administration. CF-1 wild type mice were orally administered with
biotin-MonoTf and biotin-NonR-AggTf at the dose of 40 mg/Kg through gavage feeding,
and the serum concentration of biotin-MonoTf and biotin-NonR-AggTf were measured at
8 h post feeding. The data were represented as the mean value with error bars
indicating the standard deviation (n=3).
84
Figure 4.4.4: Molecular cloning strategy for hGH-C2-Tf, hGH-CH4C-Tf,hGH hinge-
Tf, and hGH-Lzip-Tf expressing plasmids. The cloning process for generating hGH-
C2-Tf, hGH-CH4C-Tf, hGH hinge-Tf (A), or generating hGH-Lzip-Tf (B) expressing
constructs.
A
B
85
Figure 4.4.5: PCR confirmation of the potential clones for the hGH-C2-Tf
expressing plasmid after ligation and transformation. Nine single colonies were
picked, and 4 of them have linker insertion. Among these 4 clones, clone #5 contained
the correct sequence for the hGH-C2-Tf expressing plasmid.
Figure 4.4.6: PCR confirmation of the potential clones for the hGH-hinge-Tf
expressing plasmid after ligation and transformation. Eleven single colonies were
picked, and 9 of them have linker insertion. Among these 9 clones, clone #4 and #6
contained the correct sequence for the hGH-hinge-Tf expressing plasmid.
M - + 1 2 3 4 5 6 7 8 9 10 11
M - + 1 2 3 4 5 6 7 8 9
86
Figure 4.4.7: PCR confirmation of the potential clones for the hGH-CH4C-Tf
expressing plasmid after ligation and transformation. Seven single colonies were
picked, and 4 of them have linker insertion. Among these 4 clones, clone #4 contained
the correct sequence for the hGH-CH4C-Tf expressing plasmid.
Figure 4.4.8: PCR confirmation of the potential clones for the hGH-Lzip-Tf
expressing plasmid after ligation and transformation. Twelve single colonies were
picked, and eleven of them have linker insertion. Among these 11 clones, clone #49 and
#10 contained the correct sequence for the hGH-Lzip-Tf expressing plasmid
M + - 1 2 3 4 5 6 7
M + - 1 2 3 4 5 6 7 8 9 10 11 12
87
Figure 4.4.9: The sequence data of hGH-Lzip-Tf expressing plasmid. The plasmid
sequencing was performed from the 5’ of the fusion protein transcript. The DNA
sequence with bright green shading was the hGH moiety, and the sequence with light
blue shading was the Tf moiety. The DNA sequence with the yellow shading indicated
the XhoI cutting site. The leucine zipper motif was inserted between these two moieties,
which the DNA sequence is marked with the grey shading, and the amino acid
sequence for the leucine zipper linker is marked below the DNA sequence.
88
Table 4.4.1: Description and the amino acid sequence for recombinant protein
linkers in various hGH-Tf fusion proteins
Figure 4.4.10: Design of the linkers for hGH-Tf fusion proteins. The graphic
illustration of the hGH-Tf with various linkers introduced in the dissertation.
89
Figure 4.4.11: Western blotting of the hGH-Tf fusion proteins with various linker.
SDS-PAGE and Western blotting were performed to analyze the recombinant hGH-Tf
fusion protein products with C2 linker (A), or CH4C, hinge, and leucine zipper linkers
(B). Under non-reducing condition, majority of the fusion proteins are in monomeric
structure, except for the hGH-Lzip-Tf fusion protein, which was majorly in dimeric form.
After incubating with the reducing agent, the hGH-Lzip-Tf fusion protein was detected in
the monomeric form, suggesting the disulfide linkage was one of the crucial bonding to
maintain the dimeric structure for hGH-Lzip-Tf fusion protein.
A B
90
Figure 4.4.12: Confirmation of the bioactivity of hGH moiety on hGH-Lzip-Tf
fusion protein. (A) Western blotting was performed to study whether the structure of
hGH can be recognized by anti-hGH antibody after inserting various linkers in the hGH-
Tf fusion protein. The loaded samples for lane 1-6 were: MonoTf, hGH-H4-Tf, hGH-C2-
Tf, hGH-CH4C-Tf, hGH-hinge-Tf, and hGH-Lzip-Tf. (B) Nb-2 cell-based proliferation
assay was performed to evaluate the bioactivity of hGH moieties on hGH-H4-Tf and
hGH-Lzip-Tf fusion proteins. The EC50 for hGH, hGH-H4-Tf, and hGH-Lzip-Tf are 1.36
ng/mL, 2.22 ng/mL, and 12.34 ng/mL, respectively. The data were represented as the
mean value with error bars indicating the standard deviation (n=3) (Some error bars may not
be displayed because the values are lower than the scale set of the graph).
A B
91
Figure 4.4.13: Cellular retention of hGH-H4-Tf and hGH-Lzip-Tf in polarized Caco-2
cells. Pulse-chase assay was performed to evaluate the cellular retention of
125
I-hGH-
H4-Tf and
125
I-hGH-Lzip-Tf in the polarized Caco-2 cells. At 1 h of chase phase, the
radio-iodinated proteins in apical medium, basolateral medium, and the cells were
measured, and the percentage of cell-associated
125
I-hGH-H4-Tf or
125
I-hGH-Lzip-Tf was
quantified. The data were represented as the mean value with error bars indicating the
standard deviation (n=3)
92
Figure 4.4.14: Transcytosis assays of hGH-H4-Tf and hGH-Lzip-Tf in polarized
Caco-2 cells. (A) Transcytosis assays were performed to compare the transcytosis
capability of hGH-H4-Tf and hGH-Lzip-Tf in polarized Caco-2 cells. At the desinated
timepoints, the medium in the basolateral reservoir was harvested, and the intact
amount of radio-iodinated proteins were determined after TCA protein precipitation. (B)
TCA protein precitation was performed to calculate the percentage of intact radio-
iodinated protein in the medium of the basolateral reservoir at designated timepoints in
the transcytosis assays (Some error bars may not be displayed because the values are lower than
the scale set of the graph).
A B
93
4.5 Discussion
Prolonged cellular retention of NonR-AggTf and R-AggTf was also observed in
the polarized Caco-2 cells. The prolonged cellular retention indeed led to higher NonR-
AggTf and R-AggTf protein amounts transcytosed across the polarized Caco-2 cells;
however, majority of them were in degraded form (Figure 4.4.2). In the in vivo studies,
the amount of intact NonR-AggTf in the serum was also non-detectable after the oral
administration (Figure 4.4.3), which reflected the results observed in the in vitro
transcytosis assay. Although the detail mechanism of TfR-mediated transcytosis from
apical to basolateral membrane in polarized gastrointestinal epithelium is still unclear,
we propose a possible explanation for the increase degradation rate for Tf oligomers
during transcytosis based on some findings in brain endothelial cells. Previously, the
transcytosis capability of monovalent and bivalent TfR antibodies were compared in
brain endothelial-based transcytosis assays. Bivalent TfR antibody was demonstrated to
have better binding affinity to crosslink TfR. However, compared with the monovalent
TfR, higher level of the bivalent antibody was transported to the lysosomal
compartments rather than transcytosed across the endothelial cells [76]. In addition,
enhanced binding affinity to the TfR for TfR antibody increased the retention in the brain
endothelial cells, but decreased the transcytosis efficiency to the brain parenchyma [32].
Further researches on this high-affinity anti-TfR antibody revealed that the degradation
rate for TfR/anti-TfR antibody complex was increased during the transcytosis process in
the polarized brain endothelial model [33]. Therefore, either TfR receptor crosslinking or
increasing the binding affinity to the TfR led to lower transcytosis efficiency in the
polarized brain endothelial cells, even the internalized amount was increased, and
94
cellular retention was prolonged. Therefore, we proposed that, higher percentage of the
Tf oligomer was transported to the common endosome rather than the recycling
endosome after TfR-mediated endocytosis. In the common endosome, the Tf oligomer
encountered more TfR molecules from the basolateral membrane, and further resulted
in TfR crosslinking. Lastly, the Tf oligomer/TfR complex was transported to lysosomal
compartment for protein degradation, leading to lower intact protein amount
transcytosed across the polarized Caco-2 cells or GI epithelium.
In this chapter, we also attempted to develop a novel method to produce
recombinant Tf oligomer. Since there was no report regarding the strategy of preparing
recombinant Tf oligomer, leucine zipper dimerization motif was utilized to test whether
incorporating a natural dimerization domain in the Tf-based fusion protein may generate
Tf dimer-based or Tf oligomer-based fusion proteins. Several protein linkers with
cysteine residues, including the rigid helical linker, cyclo linker without thrombin cutting
site, Fc minimal hinge motif, and leucine zipper motif from MAFB transcription factor,
were designed and inserted between the hGH and Tf moieties (Table 4.4.1). The
Western blotting results demonstrated that only the fusion protein with leucine zipper
motif as the linker significantly increased the yield of the dimeric structure. Others were
majorly in the monomeric form (Figure 4.4.10). These results indicated that, in order to
form a dimeric or even oligomeric structure during the recombinant protein production
process, strong interactions between the oligomerization domains were essential to
bring fusion protein molecules in close proximity, then the disulfide linkage can be
formed to stabilize the structure. Without the protein oligomerization motif in the fusion
95
protein, only minimal intermolecular disulfide linkage or intramolecular disulfide bond
would be formed during the protein production (Figure 4.4.10) [44].
The Nb2 proliferation assay was performed to evaluate whether the function of
hGH moieties were totally diminished after the fusion protein dimerization. Although the
hGH-Lzip-Tf demonstrated its bioactivity to induce Nb2 cell proliferation, the EC50 is
much higher compared with the hGH or monomeric hGH-Tf fusion protein (Figure
4.4.11), suggesting the potency was decreased after dimerization. It has been reported
that hGH can form non-covalent dimer, as well as covalent dimer in the pituitary gland
[77, 78]. However, compared with the monomeric hGH, the hGH dimer only exhibited
10% of the growth promotion effect [79]. This phenomenon may account for the lower
bioactivity for hGH-Lzip-Tf observed in the Nb2 cell-based proliferation assay.
Nevertheless, the dimeric structure of hGH-Lzip-Tf fusion protein could be dissociated
into monomeric form since several reducing environments have been reported in the
body, such as high concentration of glutathione in the liver [80], or protein disulfide
isomerase on the surface of the hepatocyte [43].
Lastly, we performed the Caco-2 cell-based cellular retention and transcytosis
assays to compare the transcytosis capability of monomeric hGH-H4-Tf and dimeric
hGH-Lzip-Tf recombinant fusion protein. The results from pulse-chase assay at 1 h time
point did not reveal significant difference in cellular retention between monomeric hGH-
H4-Tf and dimeric hGH-Lzip-Tf fusion protein (Figure 4.4.12). In the transcytosis assay,
the percentage of degraded hGH-Lzip-Tf was similar to that of hGH-H4-Tf over time
(Figure 4.4.13 B), suggesting that the dimeric Tf may be able to avoid the excessive
protein degradation in lysosome, which was observed for the Tf oligomer. Under these
96
circumstances, the intact transcytosed hGH-Lzip-Tf protein amount was doubled
compared with the hGH-H4-Tf overtime (Figure 4.4.13 A). We proposed that, without
affecting the intracellular trafficking patterns of TfR in polarized Caco-2 cells, one TfR
can bind to one molecule of monomeric hGH-H4-Tf, or one molecule of dimeric hGH-
Lzip-Tf. Transporting one molecule of dimeric hGH-Lzip-Tf across the polarized cells
equals to two molecules of the hGH-Tf fusion protein; therefore, the amount in weight is
doubled in the transcytosis assay. We expect that the leucine zipper linker design
allows doubled amount of fusion protein transcytosed across the GI epithelium. The
hGH-Lzip-Tf can be dissociated from the dimeric structure back to the monomeric form
due to the reducing environment in the liver, leading to reversed bioactivity for the fusion
protein. With the doubled amount of the fusion protein and the reversed bioactivity of
hGH moiety, the therapeutic effect of the dimeric hGH-Lzip-Tf may be better than the
monomeric hGH-H4-Tf fusion protein after oral administration. Nonetheless, more
studies are required to investigate if the dimerized fusion protein will become the
monomeric form after entering the body.
97
4.6 Summary
Extensive protein degradation for chemically-synthesized Tf oligomer was
observed during the transcytosis process in polarized Caco-2 cells, suggesting that Tf
oligomer is not a proper carrier to deliver protein therapeutics across the GI epithelium.
A novel method was introduced to dimerized Tf-based fusion protein through leucine
zipper motif, and no increase protein degradation rate was observed for dimeric hGH-Tf
fusion protein during the transcytosis process, leading to double amount in mass of
hGH-Tf fusion protein transported across the polarized Caco-2 cells. These findings
suggest that fusion protein dimerization through leucine zipper may be a novel strategy
to increase the TfR-mediated transcytosis efficiency.
98
Chapter 5. Summary
5.1 Conclusion
In this dissertation, we generated the Tf oligomer through chemical conjugation
to investigate whether the Tf oligomer have any advantages to be utilized in drug
delivery and development. Series of studies were conducted to bridge the gap between
cell-based results and in vivo findings for the Tf oligomer. Tf oligomer synthesized by
chemical conjugation could be considered as a Tf-based protein nanoparticle, which
possessed higher binding capability to non-polarized TfR
high
breast cancer MDA-MB-
231 cell. Multiple ligand-receptor crosslinking led to prolonged cellular retention, gradual
increase in cellular uptake, better competition to the binding site with existence of Tf,
and higher intracellular degradation rate than the Tf monomer. Similar findings were
observed in MDA-MB-231 xenograft animal studies with increased retention of NonR-
AggTf in the tumor mass. Co-administration with excess fold of Tf led to the attenuated
targeting effect of NonR-AggTf to the tumor mass, but not liver or spleen, suggesting
that active targeting and the ability to compete with endogenous serum Tf were crucial
for NonR-AggTf to target the tumor mass in vivo. In polarized Caco-2 cells; however,
the prolonged cellular retention effect for Tf oligomer may result in higher protein
degradation rate during the transcytosis process, which may account for the non-
detectable level of NonR-AggTf in the serum after oral administration. These findings
indicate that the chemically-synthesized Tf oligomer could be a good carrier to deliver a
drug payload to target and to be retained in the tumor mass, but not be suitable to
deliver protein therapeutics across polarized GI epithelium.
99
A novel method was introduced to dimerize Tf-based recombinant fusion protein
by inserting a leucine zipper linker in this dissertation. Protein degradation rate were not
increased for the dimeric fusion protein during the transcytosis process, and the amount
of the dimeric hGH-Tf fusion protein in mass transported across the polarized Caco-2
cells is two-fold higher compared with the monomeric form of the fusion protein,
suggesting fusion protein dimerization through a recombinant protein linker could be a
novel strategy to improve the transcytosis efficiency for Tf-based fusion protein.
100
5.2 Summary graphic
Figure 5.2.1: The proposed mechanism of prolonged retention for chemically-
synthesized Tf oligomer in tumor cells. Compared with the Tf monomer, Tf oligomer
has a higher binding affinity to the tumor cells through multiple ligand-receptor
association. The enhanced binding allows Tf oligomer to better compete with the
binding site against endogenous Tf, resulting in TfR crosslinking on the surface. The Tf
oligomer/TfR complex exhibits a prolonged cellular retention and higher degradation
rate after endocytosis, and the prolonged retention can be also observed in the TfR
high
tumor mass. These findings suggest that the Tf oligomer could be a good carrier to
achieve tumor targeting and prolonged tumor retention effect
101
Figure 5.2.2: Enhance protein degradation for chemically-synthesized Tf oligomer
during transcytosis process in polarized GI epithelium. In polarized enterocytes, the
prolonged retention of Tf oligomer may enhance the degradation process during the
transcytosis, which decrease the TfR-mediated transcytosis efficiency. Therefore, Tf
oligomer is not a suitable carrier to transport protein therapeutic across the GI
epithelium.
102
Figure 5.2.3: Transcytosis of dimeric Tf-based fusion protein in polarized Caco-2
cells. The leucine zipper motif can serve as a recombinant protein linker to dimerize the
Tf-based fusion protein. No increased protein degradation rate is observed for the
dimerized form of hGH-Tf fusion protein during the transcytosis process, leading to
increased amount of hGH-Tf in mass transcytosed across the polarized epithelium.
Fusion protein dimerization though recombinant linker could be a novel strategy to
increase the transcytosis efficiency for Tf-based fusion protein.
103
5.3 Future Perspectives
5.3.1 Optimization of the chemically-synthesized Tf oligomer
The rapid clearance by MPS after I.V. injection is the major hurdle if the Tf
oligomer is going to be utilized as a drug carrier to target the tumor mass. In order to
avoid the recognition by MPS and increase the half-life and AUC for Tf oligomer,
“stealth” technologies could be applied, such as PEGylation [81]. It has been proven
that adding a stealth layer on protein nanocage can significantly increase the half-life,
as well as tumor targeting efficiency and amount [82]. Therefore, adding a stealth layer
could be the first step to optimize the Tf oligomer-based tumor targeting. However,
since the surface lysine residues of the Tf units have been used during the synthesis of
Tf oligomer, other reaction group should be chosen to perform PEGylation conjugation.
In addition, further validation in the binding properties to the TfR may be required for
pegylated Tf oligomer.
5.3.2 Exploration of the further application of leucine zipper dimerization linker
In this dissertation, hGH-Tf was used as the model fusion protein to evaluate if
the insertion of leucine zipper motif could be used to improve the transcytosis efficiency.
However, some preliminary studies in animal models are still required to evaluate
whether the dimerized form will be dissociated to monomeric form after entering the
body. If yes, further studies can be conducted to evaluate the practicality of using the
leucine zipper linker to improve the transcytosis efficiency, also the therapeutic efficacy
of Tf-based fusion protein after oral dosing. If not, the protein drug payload may need to
be changed since the bioactivity of dimerized hGH moiety is significantly diminished [77,
104
83]. Other protein therapeutics that have an enhanced bioactivity when dimerized can
be used to explore the potential application of leucine zipper linker in Tf-based fusion
protein design. Tumor necrosis factor-related apoptosis-inducing ligand, TRAIL, is a
member of tumor necrosis factor family protein, which has been reported to induce p53-
independent cancer apoptosis signaling [84]. Although there was a clinical study of
using recombinant of human native TRAIL (dulanermin) to treat cancer, the results
indicated that the instability of the non-covalent linkage and small molecular weight
could be the major reasons of not achieving desired anti-tumor efficacy [85]. Previously,
it has been reported that fusing leucine zipper motif to TRAIL significantly increase the
anti-tumor efficacy in vitro and in vivo [86]. Therefore, TRAIL could be a good candidate
payload to fuse with Tf moiety with the insertion of the leucine zipper linker, exploring
the potential application of dimeric Tf-based fusion protein. Compared with the TRAIL
ligand alone, there could be three possible advantages with the TRAIL-Lzip-Tf fusion
protein: 1. Our leucine zipper linker can form inter-molecular disulfide linkage to stabilize
the recombinant fusion protein in the blood circulation, and the dimerized TRAIL ligand
may elicit stronger anti-tumor efficacy in the animal model. 2. Fusing with Tf moiety
increases the size of the molecule, avoids the renal filtration, and extends the half-life.
3. The Tf moieties can serve as the ligand to increase the targeting effect to the tumor
mass. Even there are some off-target incidences, the TRAIL has been reported to have
no cytotoxic effect to the normal cells [87, 88].
5.3.3 Production of the recombinant Tf oligomer using other polymerization
motifs
105
In this dissertation, we only used one of the protein oligomerization motifs to
generate recombinant dimeric Tf-based fusion protein. In natural, there are several
oligomerization motifs that could be utilized to generate recombinant protein oligomer,
such as collagen-derived trimerization domain [89], tetramerization domain of p53 [68],
or the J-chain polymerization domain of IgM [90, 91]. It will be interested to find the
minimal domain that could be utilized in protein oligomerization from the perspective of
protein engineering. In addition, since the active amino acid residues on the protein
surface (e.g. surface lysine residues) are still available for further conjugation process,
the recombinant oligomer may carry more drug payloads compared with the chemically-
synthesized Tf oligomer, which may result in better therapeutic effect. Also, the potential
applications of recombinant Tf oligomer could be further exploited to design novel
therapeutics.
106
REFERENCES
1. Kratz, F. and B. Elsadek, Clinical impact of serum proteins on drug delivery. J Control Release,
2012. 161(2): p. 429-45.
2. Morales, J., Defining the role of insulin detemir in Basal insulin therapy. Drugs, 2007. 67(17): p.
2557-84.
3. Schnitzer, J.E., gp60 is an albumin-binding glycoprotein expressed by continuous endothelium
involved in albumin transcytosis. Am J Physiol, 1992. 262(1 Pt 2): p. H246-54.
4. Hawkins, M.J., P. Soon-Shiong, and N. Desai, Protein nanoparticles as drug carriers in clinical
medicine. Adv Drug Deliv Rev, 2008. 60(8): p. 876-85.
5. Hare, J.I., et al., Challenges and strategies in anti-cancer nanomedicine development: An industry
perspective. Adv Drug Deliv Rev, 2017. 108: p. 25-38.
6. Podhajcer, O.L., et al., The role of the matricellular protein SPARC in the dynamic interaction
between the tumor and the host. Cancer Metastasis Rev, 2008. 27(3): p. 523-37.
7. Schroeder, H.W. and L. Cavacini, Structure and function of immunoglobulins. J Allergy Clin
Immunol, 2010. 125(2 Suppl 2): p. S41-52.
8. Czajkowsky, D.M., et al., Fc-fusion proteins: new developments and future perspectives. EMBO
Mol Med, 2012. 4(10): p. 1015-28.
9. Weiner, G.J., Building better monoclonal antibody-based therapeutics. Nat Rev Cancer, 2015.
15(6): p. 361-70.
10. Gomme, P.T., K.B. McCann, and J. Bertolini, Transferrin: structure, function and potential
therapeutic actions. Drug Discov Today, 2005. 10(4): p. 267-73.
11. van Campenhout, A., et al., Transferrin modifications and lipid peroxidation: implications in
diabetes mellitus. Free Radic Res, 2003. 37(10): p. 1069-77.
12. Aisen, P., A. Leibman, and J. Zweier, Stoichiometric and site characteristics of the binding of iron
to human transferrin. J Biol Chem, 1978. 253(6): p. 1930-7.
13. Qian, Z.M., et al., Targeted drug delivery via the transferrin receptor-mediated endocytosis
pathway. Pharmacol Rev, 2002. 54(4): p. 561-87.
14. Li, H., H. Sun, and Z.M. Qian, The role of the transferrin-transferrin-receptor system in drug
delivery and targeting. Trends Pharmacol Sci, 2002. 23(5): p. 206-9.
15. Gatter, K.C., et al., Transferrin receptors in human tissues: their distribution and possible clinical
relevance. J Clin Pathol, 1983. 36(5): p. 539-45.
16. Weaver, M. and D.W. Laske, Transferrin receptor ligand-targeted toxin conjugate (Tf-CRM107)
for therapy of malignant gliomas. J Neurooncol, 2003. 65(1): p. 3-13.
17. Daniels, T.R., et al., The transferrin receptor and the targeted delivery of therapeutic agents
against cancer. Biochim Biophys Acta, 2012. 1820(3): p. 291-317.
18. Strohl, W.R., Fusion Proteins for Half-Life Extension of Biologics as a Strategy to Make Biobetters.
BioDrugs, 2015. 29(4): p. 215-39.
19. Chen, X., et al., Effects of receptor binding on plasma half-life of bifunctional transferrin fusion
proteins. Mol Pharm, 2011. 8(2): p. 457-65.
20. Bai, Y., D.K. Ann, and W.C. Shen, Recombinant granulocyte colony-stimulating factor-transferrin
fusion protein as an oral myelopoietic agent. Proc Natl Acad Sci U S A, 2005. 102(20): p. 7292-6.
21. Xia, C.Q., J. Wang, and W.C. Shen, Hypoglycemic effect of insulin-transferrin conjugate in
streptozotocin-induced diabetic rats. J Pharmacol Exp Ther, 2000. 295(2): p. 594-600.
22. Amet, N., W. Wang, and W.C. Shen, Human growth hormone-transferrin fusion protein for oral
delivery in hypophysectomized rats. J Control Release, 2010. 141(2): p. 177-82.
107
23. Jefferies, W.A., et al., Transferrin receptor on endothelium of brain capillaries. Nature, 1984.
312(5990): p. 162-3.
24. Kissel, K., et al., Immunohistochemical localization of the murine transferrin receptor (TfR) on
blood-tissue barriers using a novel anti-TfR monoclonal antibody. Histochem Cell Biol, 1998.
110(1): p. 63-72.
25. Friden, P.M., et al., Anti-transferrin receptor antibody and antibody-drug conjugates cross the
blood-brain barrier. Proc Natl Acad Sci U S A, 1991. 88(11): p. 4771-5.
26. Wang, Y., et al., Receptor-mediated activation of a proinsulin-transferrin fusion protein in
hepatoma cells. J Control Release, 2011. 155(3): p. 386-92.
27. Wang, Y., et al., Proinsulin-transferrin fusion protein as a novel long-acting insulin analog for the
inhibition of hepatic glucose production. Diabetes, 2014. 63(5): p. 1779-88.
28. Liu, Y., et al., Biodistribution, activation, and retention of proinsulin-transferrin fusion protein in
the liver: Mechanism of liver-targeting as an insulin prodrug. J Control Release, 2018. 275: p.
186-191.
29. Couch, J.A., et al., Addressing safety liabilities of TfR bispecific antibodies that cross the blood-
brain barrier. Sci Transl Med, 2013. 5(183): p. 183ra57, 1-12.
30. Sharma, G., et al., The Role of Cell-Penetrating Peptide and Transferrin on Enhanced Delivery of
Drug to Brain. Int J Mol Sci, 2016. 17(6).
31. Moos, T. and E.H. Morgan, Restricted transport of anti-transferrin receptor antibody (OX26)
through the blood-brain barrier in the rat. J Neurochem, 2001. 79(1): p. 119-29.
32. Yu, Y.J., et al., Boosting brain uptake of a therapeutic antibody by reducing its affinity for a
transcytosis target. Sci Transl Med, 2011. 3(84): p. 84ra44.
33. Bien-Ly, N., et al., Transferrin receptor (TfR) trafficking determines brain uptake of TfR antibody
affinity variants. J Exp Med, 2014. 211(2): p. 233-44.
34. Paterson, J. and C.I. Webster, Exploiting transferrin receptor for delivering drugs across the
blood-brain barrier. Drug Discov Today Technol, 2016. 20: p. 49-52.
35. Mayle, K.M., A.M. Le, and D.T. Kamei, The intracellular trafficking pathway of transferrin.
Biochim Biophys Acta, 2012. 1820(3): p. 264-81.
36. Chiu, R.Y., et al., Improving the systemic drug delivery efficacy of nanoparticles using a
transferrin variant for targeting. J Control Release, 2014. 180: p. 33-41.
37. Marsh, E.W., et al., Oligomerized transferrin receptors are selectively retained by a lumenal
sorting signal in a long-lived endocytic recycling compartment. J Cell Biol, 1995. 129(6): p. 1509-
22.
38. Lim, C.J. and W.C. Shen, Transferrin-oligomers as potential carriers in anticancer drug delivery.
Pharm Res, 2004. 21(11): p. 1985-92.
39. Lohcharoenkal, W., et al., Protein nanoparticles as drug delivery carriers for cancer therapy.
Biomed Res Int, 2014. 2014: p. 180549.
40. West, A.P., et al., Mutational analysis of the transferrin receptor reveals overlapping HFE and
transferrin binding sites. J Mol Biol, 2001. 313(2): p. 385-97.
41. Fischer, M.J.E., Amine Coupling Through EDC/NHS: A Practical Approach, in Surface Plasmon
Resonance. Methods in Molecular Biology. 2010, Humana Press. p. 55-73.
42. Schafer, F.Q. and G.R. Buettner, Redox environment of the cell as viewed through the redox state
of the glutathione disulfide/glutathione couple. Free Radic Biol Med, 2001. 30(11): p. 1191-212.
43. Noiva, R., Protein disulfide isomerase: the multifunctional redox chaperone of the endoplasmic
reticulum. Semin Cell Dev Biol, 1999. 10(5): p. 481-93.
44. Chen, X., et al., Design of an in vivo cleavable disulfide linker in recombinant fusion proteins.
Biotechniques, 2010. 49(1): p. 513-8.
108
45. Wang, K., et al., Self-assembled IR780-loaded transferrin nanoparticles as an imaging, targeting
and PDT/PTT agent for cancer therapy. Sci Rep, 2016. 6: p. 27421.
46. Yoon, D.J., et al., Intratumoral therapy of glioblastoma multiforme using genetically engineered
transferrin for drug delivery. Cancer Res, 2010. 70(11): p. 4520-7.
47. Holliday, D.L. and V. Speirs, Choosing the right cell line for breast cancer research. Breast Cancer
Res, 2011. 13(4): p. 215.
48. Foulkes, W.D., I.E. Smith, and J.S. Reis-Filho, Triple-negative breast cancer. N Engl J Med, 2010.
363(20): p. 1938-48.
49. Lebert, J.M., et al., Advances in the systemic treatment of triple-negative breast cancer. Curr
Oncol, 2018. 25(Suppl 1): p. S142-S150.
50. Nurgali, K., R.T. Jagoe, and R. Abalo, Editorial: Adverse Effects of Cancer Chemotherapy: Anything
New to Improve Tolerance and Reduce Sequelae? Front Pharmacol, 2018. 9: p. 245.
51. Chavez, K.J., S.V. Garimella, and S. Lipkowitz, Triple negative breast cancer cell lines: one tool in
the search for better treatment of triple negative breast cancer. Breast Dis, 2010. 32(1-2): p. 35-
48.
52. Barth, B.M., et al., Bioconjugation of calcium phosphosilicate composite nanoparticles for
selective targeting of human breast and pancreatic cancers in vivo. ACS Nano, 2010. 4(3): p.
1279-87.
53. Sonoda, S. and M. Schlamowitz, Studies of 125I trace labeling of immunoglobulin G by
chloramine-T. Immunochemistry, 1970. 7(11): p. 885-98.
54. Hem, A., A.J. Smith, and P. Solberg, Saphenous vein puncture for blood sampling of the mouse,
rat, hamster, gerbil, guinea pig, ferret and mink. Lab Anim, 1998. 32(4): p. 364-8.
55. Zhang, Y., et al., PKSolver: An add-in program for pharmacokinetic and pharmacodynamic data
analysis in Microsoft Excel. Comput Methods Programs Biomed, 2010. 99(3): p. 306-14.
56. Vugmeyster, Y., et al., Biodistribution of [125I]-labeled therapeutic proteins: application in
protein drug development beyond oncology. J Pharm Sci, 2010. 99(2): p. 1028-45.
57. Ghaffarian, R. and S. Muro, Distinct subcellular trafficking resulting from monomeric vs
multimeric targeting to endothelial ICAM-1: implications for drug delivery. Mol Pharm, 2014.
11(12): p. 4350-62.
58. Weflen, A.W., et al., Multivalent immune complexes divert FcRn to lysosomes by exclusion from
recycling sorting tubules. Mol Biol Cell, 2013. 24(15): p. 2398-405.
59. Gustafson, H.H., et al., Nanoparticle Uptake: The Phagocyte Problem. Nano Today, 2015. 10(4):
p. 487-510.
60. Blanco, E., H. Shen, and M. Ferrari, Principles of nanoparticle design for overcoming biological
barriers to drug delivery. Nat Biotechnol, 2015. 33(9): p. 941-51.
61. Li, R., et al., Be Active or Not: the Relative Contribution of Active and Passive Tumor Targeting of
Nanomaterials. Nanotheranostics, 2017. 1(4): p. 346-357.
62. Fleming, R.E., et al., Transferrin receptor 2: continued expression in mouse liver in the face of iron
overload and in hereditary hemochromatosis. Proc Natl Acad Sci U S A, 2000. 97(5): p. 2214-9.
63. Norouziyan, F., W.C. Shen, and S.F. Hamm-Alvarez, Tyrphostin A8 stimulates a novel trafficking
pathway of apically endocytosed transferrin through Rab11-enriched compartments in Caco-2
cells. Am J Physiol Cell Physiol, 2008. 294(1): p. C7-21.
64. Hidalgo, I.J., T.J. Raub, and R.T. Borchardt, Characterization of the human colon carcinoma cell
line (Caco-2) as a model system for intestinal epithelial permeability. Gastroenterology, 1989.
96(3): p. 736-49.
65. Xia, C.Q. and W.C. Shen, Tyrphostin-8 enhances transferrin receptor-mediated transcytosis in
Caco-2- cells and inreases hypoglycemic effect of orally administered insulin-transferrin
conjugate in diabetic rats. Pharm Res, 2001. 18(2): p. 191-5.
109
66. Khan, S., et al., Role of Recombinant DNA Technology to Improve Life. Int J Genomics, 2016.
2016: p. 2405954.
67. Liu, S.H., et al., Regulation of clathrin assembly and trimerization defined using recombinant
triskelion hubs. Cell, 1995. 83(2): p. 257-67.
68. Chène, P., The role of tetramerization in p53 function. Oncogene, 2001. 20(21): p. 2611-7.
69. Jones, S., An overview of the basic helix-loop-helix proteins. Genome Biol, 2004. 5(6): p. 226.
70. Fujii, Y., et al., Structural basis for the diversity of DNA recognition by bZIP transcription factors.
Nat Struct Biol, 2000. 7(10): p. 889-93.
71. Chiu, R., et al., The c-Fos protein interacts with c-Jun/AP-1 to stimulate transcription of AP-1
responsive genes. Cell, 1988. 54(4): p. 541-52.
72. Kostelny, S.A., M.S. Cole, and J.Y. Tso, Formation of a bispecific antibody by the use of leucine
zippers. J Immunol, 1992. 148(5): p. 1547-53.
73. Amoutzias, G.D., et al., Choose your partners: dimerization in eukaryotic transcription factors.
Trends Biochem Sci, 2008. 33(5): p. 220-9.
74. Kataoka, K., et al., MafB, a new Maf family transcription activator that can associate with Maf
and Fos but not with Jun. Mol Cell Biol, 1994. 14(11): p. 7581-91.
75. Caaveiro, J.M., M. Kiyoshi, and K. Tsumoto, Structural analysis of Fc/FcγR complexes: a blueprint
for antibody design. Immunol Rev, 2015. 268(1): p. 201-21.
76. Niewoehner, J., et al., Increased brain penetration and potency of a therapeutic antibody using a
monovalent molecular shuttle. Neuron, 2014. 81(1): p. 49-60.
77. Becker, G.W., et al., Chemical, physical, and biological characterization of a dimeric form of
biosynthetic human growth hormone. Biotechnol Appl Biochem, 1987. 9(6): p. 478-87.
78. Nagatomi, Y., et al., Reversible dimerization of 20 kilodalton human growth hormone (hGH).
Growth Horm IGF Res, 2000. 10(4): p. 207-14.
79. Lewis, U.J., et al., An interchain disulfide dimer of human growth hormone. J Biol Chem, 1977.
252(11): p. 3697-702.
80. DeLeve, L.D. and N. Kaplowitz, Glutathione metabolism and its role in hepatotoxicity. Pharmacol
Ther, 1991. 52(3): p. 287-305.
81. Suk, J.S., et al., PEGylation as a strategy for improving nanoparticle-based drug and gene
delivery. Adv Drug Deliv Rev, 2016. 99(Pt A): p. 28-51.
82. Lee, N.K., et al., Ferritin nanocage with intrinsically disordered proteins and affibody: A platform
for tumor targeting with extended pharmacokinetics. J Control Release, 2017. 267: p. 172-180.
83. Bustamante, J.J., et al., Human growth hormone: 45-kDa isoform with extraordinarily stable
interchain disulfide links has attenuated receptor-binding and cell-proliferative activities. Growth
Horm IGF Res, 2010. 20(4): p. 298-304.
84. Wang, S. and W.S. El-Deiry, TRAIL and apoptosis induction by TNF-family death receptors.
Oncogene, 2003. 22(53): p. 8628-33.
85. Kelley, S.K., et al., Preclinical studies to predict the disposition of Apo2L/tumor necrosis factor-
related apoptosis-inducing ligand in humans: characterization of in vivo efficacy,
pharmacokinetics, and safety. J Pharmacol Exp Ther, 2001. 299(1): p. 31-8.
86. Walczak, H., et al., Tumoricidal activity of tumor necrosis factor-related apoptosis-inducing
ligand in vivo. Nat Med, 1999. 5(2): p. 157-63.
87. Pitti, R.M., et al., Induction of apoptosis by Apo-2 ligand, a new member of the tumor necrosis
factor cytokine family. J Biol Chem, 1996. 271(22): p. 12687-90.
88. Griffith, T.S., et al., Intracellular regulation of TRAIL-induced apoptosis in human melanoma cells.
J Immunol, 1998. 161(6): p. 2833-40.
89. Cuesta, A.M., et al., Improved stability of multivalent antibodies containing the human collagen
XV trimerization domain. MAbs, 2012. 4(2): p. 226-32.
110
90. Yoo, E.M., et al., Structural requirements for polymeric immunoglobulin assembly and
association with J chain. J Biol Chem, 1999. 274(47): p. 33771-7.
91. Hiramoto, E., et al., The IgM pentamer is an asymmetric pentagon with an open groove that
binds the AIM protein. Sci Adv, 2018. 4(10): p. eaau1199.
Abstract (if available)
Abstract
Transferrin/transferrin receptor (Tf/TfR) system has been exploited to develop several novel strategies in drug delivery in the past decades, including tumor targeting and trans-epithelial protein drug delivery. Although it is promising to apply this system to develop new therapeutics in the pre-clinical settings, there is still no FDA-approved therapy that is developed based on the Tf/TfR system. Several challenges remain unsolved to utilize Tf/TfR system in drug development, and one of them is the rapid recycling of Tf after endocytosis. This phenomenon may limit the therapeutic efficacy if the drug payload requires intracellular action to achieve the therapeutic effect. Using Tf oligomer as the carrier has been proposed to solve this issue since it was reported to exhibit a prolonged cellular retention after endocytosis. However, in order to evaluate the potential application of Tf oligomer in drug delivery, detailed investigation on biological properties in vitro and the pharmacokinetics or biodistribution profiles in vivo are required. In this dissertation, Tf oligomer was chemically synthesized, and the biological properties were characterized in TfRʰⁱᵍʰ MDA-MB-231 cancer cells and in polarized Caco-2 cells. The enhance binding affinity and prolonged cellular retention observed for Tf oligomer in the MDA-MB-231 cells account for the higher tumor retention in the xenograft mouse model. However, the prolonged cellular retention observed in the polarized Caco-2 cells may lead to extensive protein degradation for Tf oligomer during the transcytosis process. The findings in this dissertation indicate that, compared with Tf monomer, the chemically-synthesized Tf oligomer could be a good drug carrier for tumor targeting, but not a suitable carrier to deliver protein therapeutics across the gastrointestinal epithelium.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Comparison of the pharmacokinetic profiles between monomeric and oligomeric transferrin
PDF
Characterization of rice-derived proinsulin-transferrin fusion protein and its oral bioactivity as a basal insulin analogue
PDF
The intracellular trafficking of transferrin in Caco-2 cells and its implication to transferrin-mediated oral drug delivery
PDF
Proinsulin-transferrin recombinant fusion protein: mechanism of activation and potential application in diabetes treatment
PDF
Optimization of nanomedicine based drug delivery systems for the treatment of solid tumors
PDF
Human growth hormone-transferrin recombinant fusion protein for oral delivery
PDF
Mechanism study of proinsulin-transferrin fusion protein as a long-acting and liver-targeting insulin prodrug
PDF
Cationic cell penetrating peptides: characterization of transport properties in epithelial cells and their utilization as delivery systems for protein and peptide drugs
PDF
Developing recombinant single chain Fc-dimer fusion proteins for improved protein drug delivery
PDF
Preparation and characterization of Tf-G-CSF fusion protein
PDF
Cellular uptake mechanism of elastin-like polypeptide fusion proteins
PDF
Cell penetrating peptide-based drug delivery system for targeting mildly acidic pH
PDF
Multivalent smart elastin-like polypeptide therapeutics with drug delivery and biosensing applications.
PDF
Reversible lipidization of peptide and protein drugs for prolonging the plasma half-lives and passive liver targeting
PDF
Cell penetrating peptide-based polyplexes for sirRNA delivery
PDF
Sustained and targeted delivery of rapamycin using FKBP-elastin like polypeptide fusion proteins
PDF
Proinsulin-transferrin fusion protein as a liver targeting insulin therapeutic in Type 1 diabetes treatment
PDF
Factors that may impact drug disposition and metabolism
PDF
Characterization and identification of endogenous factor(s) that enhance insulin transport across primary rat alveolar epithelial cell monolayers
PDF
Analysis of endocytic and trafficking pathways of potential candidates for drug delivery in HeLa and lacrimal gland acinar cells
Asset Metadata
Creator
Wang, Hsuan-Yao
(author)
Core Title
Characterization of the transferrin oligomer and its potential application in drug delivery
School
School of Pharmacy
Degree
Doctor of Philosophy
Degree Program
Pharmaceutical Sciences
Publication Date
04/25/2021
Defense Date
03/08/2019
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
drug delivery,multivalent binding,OAI-PMH Harvest,oligomer,protein nanoparticle,transcytosis,transferrin,transferrin receptor,tumor targeting
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Shen, Wei-Chiang (
committee chair
), Okamoto, Curtis (
committee member
), Zhang, Yong (
committee member
)
Creator Email
hsuanyaw@usc.edu,sean.hsuanyao.wang@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c89-150770
Unique identifier
UC11662618
Identifier
etd-WangHsuanY-7299.pdf (filename),usctheses-c89-150770 (legacy record id)
Legacy Identifier
etd-WangHsuanY-7299.pdf
Dmrecord
150770
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Wang, Hsuan-Yao
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
drug delivery
multivalent binding
oligomer
protein nanoparticle
transcytosis
transferrin
transferrin receptor
tumor targeting