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
/
Cationic cell penetrating peptides: characterization of transport properties in epithelial cells and their utilization as delivery systems for protein and peptide drugs
(USC Thesis Other)
Cationic cell penetrating peptides: characterization of transport properties in epithelial cells and their utilization as delivery systems for protein and peptide drugs
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
CATIONIC CELL PENETRATING PEPTIDES:
CHARACTERIZATION OF TRANSPORT PROPERTIES IN EPITHELIAL
CELLS AND THEIR UTILIZATION AS DELIVERY SYSTEMS FOR PROTEIN
AND PEPTIDE DRUGS
by
Leena Patel
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PHARMACEUTICAL SCIENCES)
August 2007
Copyright 2007 Leena Patel
ii
DEDICATION
To my loving parents,
Mr. Natverbhai Patel and Mrs. Pramilaben Patel
iii
ACKNOWLEDGEMENTS
I would like to heartily thank my dissertation advisor Dr. Wei-Chiang Shen
for mentoring me through the PhD program. Dr. Shen was kind enough to accept me
in his lab when I needed a new home. He has constantly supported me and
encouraged me to pursue my research interests. I will always cherish the numerous
and much needed pep talks he has given me to motivate me during the tough times in
my project.
I am deeply thankful to Dr. Vincent H. L. Lee; I owe it to him for giving me a
chance to make a bold change in my career from Pharmacy Administration to basic
science. It is also in the Leelab, that I cultivated crucial skills of a basic researcher
and also acquired the team spirit motto, which I will keep forever.
My committee members; Dr. Kwang-Jin Kim has kindly provided week after
week, the primary cells used in this dissertation and also helped in manuscript
discussions. Dr. Curtis Okamoto, Dr. Austin Mircheff and Dr. Austin Yang – I
sincerely appreciate your helpful comments and discussions.
Many thanks are due to Dr. Jeff. Wang, who selflessly assisted me in the
synthesis work, which is a major part of this thesis and also Daisy Shen for always
helping me with in vivo experiments and creating a fun and friendly environment in
the lab.
All my labmates and colleagues have been very helpful, especially Dr. Rana
Bahhady, Dr. Liyun Yuan, Dr. Ching-Jou Lim and Dr. Yun Bai. Thank you for the
iv
technical help, heated science discussions over coffee and most of all lasting
friendship.
My parents- Thank you very much for always insisting on an education. I
may take it for granted some times but I truly appreciate your sacrifices and the
hurdles that you’ve gone through just to see me, Gaurang and Sonal excel in school.
Finally, nothing would be possible without the unconditional support and
love of my husband, Dr. Ashutosh Kulkarni. Life is sweet with you dear!
v
TABLE OF CONTENTS
Dedication ii
Acknowledgements iii
List of Tables ix
List of Figures x
List of Schemes xii
Abbreviations xiii
Abstract xiv
PREFACE
Project goals xvi
Hypothesis and specific aims xvii
Overall outcome of the project xix
CHAPTER 1: Characterization of transport properties of cationic CPPs in
epithelial cells
I. INTRODUCTION 1
1. Need for protein and peptide drug delivery systems 1
2. Approaches to increase systemic bioavailability of protein and
peptides
2
3. Cell penetrating peptides as vectors 3
4. In vitro models for transport studies 4
5. Transport pathways in epithelial cells 7
II. EXPERIMENTAL 9
1. Synthesis and radiolabeling of peptides 9
2. Isolation and cell culture of RAECI and RAECII 10
3. Uptake and transport studies 11
4. Bidirectional transport and apparent permeability 11
5. Measurement of TEER 12
6. Effect of temperature on transport 13
7. Transport of fluid phase marker FD70 in presence of peptides 13
8. Pulse-chase assay 13
9. MTT assay
10. Statistical analysis
14
14
III. RESULTS 15
1. Transport of YGR9, YGR15 and YGK9 15
2. Comparison of transport in RAECI vs RAECII 16
3. Transport of L-isoform vs D-isoform 17
4. Concentration dependent transport and uptake of Tat-D and yr7 18
5. Bidirectional transport of Tat-D and yr7 19
vi
6. Apparent permeability of mannitol 20
7. Transient and reversible decrease in TEER over time 21
8. Temperature dependence in transport of Tat-D and yr7 23
9. Effect on transport of FD70
10. Time course of relative efflux after pulse-chase assay
11. Cytotoxicity of Tat-D and yr7
24
25
26
IV. DISCUSSION 26
V. SUMMARY 33
CHAPTER 2: Cationic CPP mediated peptide delivery: Desmopressin-
oligoarginine for oral delivery
I. INTRODUCTION 34
1. Desmopressin (DDAVP) as a peptide cargo 34
2. Disulfide bridging as a technique of conjugation 36
3. Delivery of peptide cargo using CPP 38
II. EXPERIMENTAL 39
1. Bioconjugation of DDAVP with Cr7 39
2. TLC analysis 42
3. MALDI-TOF mass spectroscopy 42
4. Radiolabeling of DDAVP and DDAVP-Cr7 42
5. In vitro transport assay 42
6. Size-exclusion analysis of transported conjugate 43
7. In vivo animal model-Brattleboro rats 43
8. Biological activity of DDAVP and DDVAP-Cr7 44
9. Oral delivery of DDAVP and DDAVP-Cr7 44
10. Effect of formulation 44
III. RESULTS 45
1. Bioconjugation of DDAVP with Cr7 45
2. Characterization of bioconjugate 47
a. Determination of Mwt of conjugate 47
b. Regeneration of desmopressin upon reduction 47
3. Transport of conjugate in MDCK and RAECII 49
4. Retention of biological activity 51
5. Oral activity and in vitro – in vivo correlation 52
6. Effect of formulation 53
7. Analysis of in vitro components 55
vii
IV. DISCUSSION 57
V. SUMMARY 61
CHAPTER 3: Cationic CPP mediated protein delivery: Insulin-oligoarginine
for pulmonary delivery
I. INTRODUCTION 62
1. Background information on insulin 63
2. Non-invasive modes of insulin delivery 63
a. Oral delivery 63
b. Pulmonary delivery 64
3. Chemical modifications and conjugation of insulin 65
II. EXPERIMENTAL 68
1. Synthesis, purification and radiolabeling of insulin bioconjugates 68
2. Characterization of bioconjugates 73
a. SDS-PAGE analysis of peak 73
b. MALDI-TOF mass spectroscopy 73
c. Anion-exchange chromatography 73
d. Radioiodination of insulin and conjugates 74
3. Alveolar epithelial and Caco-2 culture 74
4. Transport of insulin bioconjugates
5. TCA precipitate assay
75
75
6. Transport of In-cr9 vs In + cr9 76
a. Comparison of physical mixture vs covalent linkage 76
b. Uptake and transport of insulin: Pep1 mixture 76
7. Effect of In-cr9 on TEER of RAECII 76
8. Transport and uptake of In-cr9 in RAECI and Caco-2 cells 77
9. Effect of temperature on transport of In-cr9 77
10. Effect of biochemical modulators on transport of In-cr9 77
11. Effect of heparin and protamine on uptake and transport of In-cr9 78
12. Insulin receptor binding assay 78
a. HepG2 cell culture 78
b. Concentration dependent insulin binding 79
c. Competition assay 79
13. In vivo activity of In-cr9 79
a. Induction of diabetes in rats 79
b. Preparation of dosing solution 80
c. Spray instillation of In-cr9 into lungs of diabetic rats 81
viii
III. RESULTS 81
1. Biosynthesis of In-cr9, In-ctat and In-ck9 81
2. Characterization of bioconjugates 94
a. Modification of net charge 94
b. Determination of Mwt by SDS-PAGE 95
c. Determination of Mwt by MALDI-TOF mass spectroscopy 96
3. Higher transport of In-cr9 compared to In-tat and In-ck9 98
4. Linear increase in transport of In-cr9 over time 99
5. Importance of covalent conjugation 100
a. Physical mixture vs covalent linkage 100
b. Higher uptake but low transport with Pep1 101
6. Effect of In-cr9 on TEER of RAECII 103
7. Temperature dependent decrease in transport of In-cr9 104
8. Comparison of transport in other epithelial cell models 105
9. Effect of biochemical modulators 106
a. No effect of endocytosis inhibitors 106
b. No effect of macropinocytosis inhibitors 106
c. Minor decrease in transport with metabolic inhibitors 106
10. Importance of charge on uptake and transport of In-cr9 107
11. Lack of competition for receptor binding by In-cr9 109
12. Higher and prolonged effect of In-cr9 vs insulin by spray
instillation
111
IV. DISCUSSION 113
V. SUMMARY 124
CHAPTER 4: Future Directions
1. Large-scale synthesis of stable Insulin-CPP bioconjugates 125
2. Broader in vivo studies defining PK and PD parameters 125
3. Long term cytotoxicity effects of CPP in lungs 126
4. Application to other therapeutic macromolecules 126
REFERENCES 127
ix
LIST OF TABLES
1.1 In vitro lung epithelial cell culture models 6
1.2 List of CPPs used for transport studies 9
1.3 Apparent permeabilities of mannitol 21
1.4 MTT assay for various concentrations of yr7 and Tat-D peptide 26
3.1 List of reaction conditions and retention times 93
3.2 Effect of biochemical modulators on transport of In-cr9 107
x
LIST OF FIGURES
1.1 Major transport pathways in epithelial cells 8
1.2 Total transport of YGR9, YGK9 and YGR15 across RAECII 15
1.3 Comparison of total transport in RAECI vs RAECII 16
1.4 Transport of Tat, YGR9 and YGr9 across RAECII 17
1.5 Concentration dependent transport of yr7 and Tat-D peptides 18
1.6 Concentration dependent uptake 19
1.7 Bi-directional transport as a function of time 20
1.8 Effect of peptides on TEER 22
1.9 Effect of temperature on transport of peptides 23
1.10 Effect of yr7 or Tat-D on transport of FD70 24
1.11 Time course of basolateral peptide release after pulse-chase assay 25
2.1 Structure of DDAVP and DDAVP-Cr7 35
2.2 Reaction scheme of bioconjugation of desmopressin to Cr7 41
2.3 Elution profiles of TP-Cr7 and DDAVP-Cr7 46
2.4 Schematic of TLC analysis and MS chromatogram 48
2.5 Transport across RAECII and MDCK cells 50
2.6 In vivo biological activity of DDAVP-Cr7 51
2.7 Oral activity of DDAVP-Cr7 compared to DDAVP 52
2.8 Effect of various formulations on activity of DDAVP-Cr7 54
2.9 Elution profiles of transported DDAVP and DDAVP-Cr7 56
3.1 Schematic illustration of insulin structure 67
3.2 Schematic representation of insulin-CPP conjugation 70
3.3a Reversed phase HPLC chromatograph of insulin 84
3.3b Reversed phase HPLC chromatograph of tricitraconyl insulin 85
3.3c Reversed phase HPLC chromatograph of dicitraconyl insulin (InP4) 86
3.3d Reversed phase HPLC chromatograph of purified InP4 87
3.3e Reversed phase HPLC chromatograph of InP4-SPDP 88
xi
3.3f Reversed phase HPLC chromatograph of purified InP4-SPDP 89
3.3g Reversed phase HPLC chromatograph of In-cr9 90
3.3h Reversed phase HPLC chromatograph of In-ctat 91
3.3i Reversed phase HPLC chromatograph of In-ck9 92
3.4a Anion-exchange chromatography of In-cr9 94
3.4b SDS-PAGE analysis of In-cr9 95
3.4c MALDI-TOF mass spectra of In-cr9, In-ctat and In-ck9 97
3.5 Transport comparison of insulin bioconjugates in RAECII 98
3.6 Time-dependent transport of In-cr9 in RAECII 99
3.7 Comparison of covalent linkage with cr9 vs physical mixture 100
3.8 Uptake and transport of insulin in presence of Pep-1 102
3.9 Effect of In-cr9 on TEER of RAECII 103
3.10 Effect of temperature on transport of In-cr9 104
3.11 Transport of In-cr9 across RAECI and Caco-2 cells 105
3.12 Uptake and transport of In-cr9 in presence of heparin and protamine 108
3.13 Concentration dependent binding of insulin to receptors in HepG2
cells
109
3.14 Receptor binding competition assay for insulin and In-cr9 110
3.15 In vivo effect of In-cr9 after spray instillation in diabetic rats 112
3.16 Schematic representation of prolonged in vivo effect of In-cr9 122
xii
LIST OF SCHEMES
Scheme 1 Specific aims designed to test the hypothesis xix
Scheme 2 Summary of findings from chapter 1 33
Scheme 3 Summary of findings from chapter 2 61
Scheme 4 Summary of findings from chapter 3 124
xiii
ABBREVIATIONS
CPPs Cell penetrating peptides
Tat Transcriptional transactivator
PhRMA The Pharmaceutical Research and Manufacturers of America
MAP Model amphipathic peptides
MDCK Madin Darby canine kidney cells
RAECI Rat alveolar epithelial type I-like cells
RAECII Rat alveolar epithelial type II-like cells
TEER Transepithelial electrical resistance
PBS Phosphate buffered saline
BSA Bovine serum albumin
MTT 3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyl tetrazolium bromide
DMF Dimethylformamide
DTDP 2, 2’-dithiodipyridine
SPDP N-succinimidyl-3-(2-pyridyldithio)propionate
DDAVP Desmopressin
TFA Trifluoroacetic acid
MeCN Acetonitrile
DMMA Dimethylmaleic anhydride
Cit A Citraconic anhydride
InP4 Dicitraconyl insulin
Papp Apparent permeability
xiv
ABSTRACT
The recent interest in cationic cell penetrating peptide (CPPs) stems from
their potential utility as non-invasive drug delivery platforms. These peptides are
short sequences consisting mostly of arginine or lysine residues including the Tat
peptide, Antp peptide and different oligomers of arginine and lysine. By conjugating
CPPs to a wide range of impermeable cargo, their intracellular delivery has shown to
be increased in different types of cells by a less defined mechanism of transduction.
Despite their versatility, few studies have focused on their application for
transcellular delivery. Considering that the original protein from which the cationic
CPPs are derived, HIV-1 Tat protein, is able to exit the infected cell and internalize
into neighbouring cells, it is conceivable that CPPs can enter into differentiated
epithelial cells from the apical domain and exit on the basolateral domain. If this is
true, they can also ferry impermeable cargo across the epithelial cells, thus aid in
overcoming one of the notorious barriers for oral and pulmonary delivery of proteins
and peptides. This hypothesis was tested by investigating the transport of CPPs in
alveolar epithelial cells and also by conjugating CPPs to a peptide and a protein drug
intended for oral and pulmonary delivery respectively. The results presented in this
thesis show that CPPs can be transported across alveolar epithelial cells without
severely compromising cellular integrity. Additionally, oligoarginine conjugated to
protein drug, e.g., insulin, can facilitate the transport of insulin in vitro across
alveolar epithelial cells by a transcellular pathway and can enhance insulin transport
in vivo when delivered into the lungs of diabetic rats. However, conjugating
xv
oligoarginine to a peptide drug, desmopressin, improved in vitro transport marginally
but did not have any beneficial effect when delivered orally. From this work, the
potential for CPPs as delivery systems for transcellular transport of proteins,
especially through the pulmonary route, is realized.
xvi
PREFACE
I. PROJECT GOALS
The serendipitous discovery that the HIV-coded Tat protein can rapidly and
efficiently be taken up by intact cells in culture medium has led to the emergence of
a new and exciting class of delivery vectors referred to as cell penetrating peptides
(CPPs). CPPs are short oligopeptides of less than 30 amino acids, either cationic or
amphipathic in nature, with the ability to penetrate across cell plasma membrane in
an unusual way. Several studies have shown that covalent conjugation of CPPs such
as oligoarginine or Tat peptide to macromolecules such as proteins, peptides,
antisense agents and even nanoparticles, facilitate their transduction across cell
membrane as well as tissue barriers by an unknown mechanism. The main objective
of this project was to determine if CPPs could exhibit the same cell penetrating
capabilities across differentiated epithelial cells, which act as a major barrier for oral
and pulmonary delivery of macromolecules. Differentiated epithelial cells possess
two unique cell membrane domains, apical and basolateral, with junctional proteins
between the two domains and adjacent cells that all contribute to form a continuous
relatively impermeable barrier for macromolecules. The ability to break through this
barrier is significant for systemic delivery of therapeutics by the non-invasive
alternative routes of drug delivery. To establish a proof-of concept, impermeable
peptide and protein drugs were used as model macromolecules that were tethered to
CPPs using disulfide linkage to be transported across epithelial cells. The long-term
goal of this project is to determine the suitability of CPPs as delivery systems for
xvii
macromolecular drugs, specifically peptides and proteins for oral and/or pulmonary
delivery.
II. HYPOTHESIS AND SPECIFIC AIMS
Several studies have shown that intracellular delivery of impermeable
macromolecules can be achieved by covalent conjugation to CPP. Furthermore, it
has been shown that an even tougher cellular barrier such as blood brain barrier
(BBB) or skin barrier can be penetrated by CPPs. Conjugation of doxorubicin to
SynB1, an amphipathic CPP, resulted in a 30-fold increase in the transport across the
BBB whereas oligoarginine covalently conjugated to cyclosporin A has shown
efficient permeation through the skin tissue. Also, in vivo studies utilizing a Tat--
gal fusion show that Tat peptide can penetrate across most of the tissues including
liver, kidney lungs, heart, spleen and also into the brain. From the various
applications of CPP and the possibility of crossing barriers such as BBB and
epidermal and dermal layers, we hypothesized that cationic CPPs, tat peptide and its
analogues (e.g., oligoarginine), conjugated to protein or peptide drugs can enhance
the absorption of macromolecule drugs across cell barriers of polarized epithelial
cells without compromising cellular integrity and thus improve the bioavailability of
the conjugated drug. To test this hypothesis, we proposed three specific aims listed
below. The background, experimental design, results and discussions for each aim
are given in details in the ensuing chapters.
xviii
Specific aim # 1:
To characterize the transepithelial transport of unconjugated cationic CPPs across
alveolar epithelial cells.
In this aim, we determined the transport properties such as efficiency,
directionality, time and concentration dependency and cytotoxicty of various cationic
CPPs including, oligoarginine, oligolysine and tat peptide in an in vitro epithelial cell
model that represents the type II alveolar cells of the lungs. The purpose was to gain
an insight into the transport behaviors of the peptide alone and whether they are
suitable as delivery systems.
Specific aim # 2:
To enhance the oral delivery of desmopressin, a peptide drug by conjugation with
oligoarginine.
The focus of this aim was to apply the characteristic feature of enhanced
delivery of oligoarginine to a poorly permeable peptide drug. The disulfide approach
of covalent conjugation was utilized to link the desmopressin to oligoarginine. Both
in vitro transport and in vivo pharmacodynamic activity of the conjugate were
investigated
Specific aim # 3:
To enhance the pulmonary delivery of insulin, a protein drug by conjugation with
oligoarginine.
This aim was targeted at using CPP to enhance the transport of a protein
drug, insulin across the abundant alveolar epithelial cells. Disulfide linkage was also
xix
utilized in this case, however a bifunctional cross linker was used to introduce the
thiol linkage to insulin. Both in vitro and in vivo aspects of enhanced delivery were
investigated and the transport mechanism across epithelial cell was also studied. A
schematic of the specific aims is presented below.
III. OVERALL OUTCOME OF THE PROJECT
The highlight of this project is the successful enhancement in transport seen
for insulin conjugated to oligoarginine. We present both in vitro and in vivo evidence
for the concept of utilizing cationic CPPs as delivery vectors to cross the alveolar
epithelial cell barrier. However, the complexity and hardships for oral delivery of
macromolecules are evident with our example of desmopressin. While a modest
Scheme 1: Specific aims designed to test the hypothesis of this project
Cationic CPPs
Transepithelial Transport
Oral Delivery
•Peptide cargo
•Desmopressin
•Heptaarginine
Pulmonary Delivery
•Protein cargo
•Insulin
•Nonaarginine
2 3 1
Cationic CPPs
Transepithelial Transport
Oral Delivery
•Peptide cargo
•Desmopressin
•Heptaarginine
Pulmonary Delivery
•Protein cargo
•Insulin
•Nonaarginine
2 3 1
xx
increase in in vitro transport of conjugated desmopressin was observed, this effect
could not be translated in vivo due to interplay of several factors including over
dominance by proteolytic activities. Although our goal was to target both oral and
pulmonary epithelial barriers, our results favor utilizing CPPs for traversing the
pulmonary epithelium. The work presented here is a foundation for further
investigating the utility of CPPs as delivery systems for other macromolecular drugs
targeted to the lungs for systemic availability and optimizing the effect seen with
insulin.
1
CHAPTER 1
Transport of Cationic CPPs in Alveolar Epithelial Cells
I. INTRODUCTION
1. Need for protein and peptide drug delivery systems
Advances in cell and molecular biology has led to the identification of
several proteins and peptides that can be utilized as therapeutic agents for various
illnesses. The ability to produce these biotherapeutics in large scale through
recombinant DNA technology has further fueled a deep interest in developing them.
A 2006 survey by PhRMA reports that 418 biotechnology products that include
recombinant proteins, vaccines, monoclonal antibodies and hormones, are at various
stages of clinical development with 125 biotherapeutic products already approved
and available to patients (Tauzin 2006).
Despite the rapid progress in identifying and developing biotherapeutics, the
biggest challenge remains in achieving effective, convenient and targeted delivery of
these agents in the human body. This challenge is posed due to inherent physical
and chemical characteristics of protein and peptides. Their hydrophilicity, large
molecular size, aggregation and adsorptive nature, enzymatic, chemical and physical
instability, and bioincompatibilities are some of the reasons that hinder their
development. In addition, they exhibit undesirable pharmacokinetic characteristics
such as poor permeation across biological membrane barriers, short plasma half-lives
and also immunogenic tendencies (Kumar et al. 2006). These challenges restrict the
2
delivery of biotherapeutics to parenteral route of administration, which is
inconvenient and unacceptable to patients and often leads to poor patient
compliance.
To address these problems, delivery systems that improve the attributes in
part or whole of the therapeutic agent are warranted. In addition, from a patient
perspective, alternate routes of administration that are less invasive are desired. Oral,
buccal, nasal, pulmonary, transdermal, and ocular routes are among the routes that
have been explored for protein and peptide delivery. However, none of these routes
achieve the effectiveness in delivery seen with parenteral route because of two
distinct classes of barriers. Enzymatic inactivation and permeability constrains limit
their systemic availability, thus necessitating delivery strategies.
2. Approaches to increase systemic bioavailability of proteins and peptides
To circumvent enzymatic barriers, strategies utilized include use of enzyme
inhibitors or enzyme protection techniques such as encapsulation, and chemical or
physical modification of proteins and peptides. Enzyme inhibitors such as aprotinin,
amastatin, bestatin and puromicin have been extensively studied but show limited
practical or commercial acceptability because of their severe side effects including
systemic toxicity and disturbance in digestion of nutrient proteins (Lee 1988).
Chemical modification in the form of prodrug approach is successfully applied to
peptide drugs such as desmopressin, an analogue of vasopressin whereby a single
substitution of L-arginine with D-arginine and deamination of the N terminal leads to
3
protection against enzyme activity (Kahns et al. 1993). Similarly, enkephalins have
been derivatized with aldehydes and ketones to form Met-enkephalins and Leu-
enkephalins that are proteolytically stable than the parent compound (Bundgaard and
Rasmussen 1991).
On the other hand, permeability barriers are approached by (i) use of
molecular engineering, i.e covalent coupling of macromolecules to carriers that ferry
the cargo across mucosal and epithelial barrier, (ii) formulation engineering i.e non-
covalent modification of macromolecules with delivery systems such as
nanoparticles, liposomes, hydrogels, emulsions and mucoadhesives, or (iii) the use of
penetration enhancers that alter the structural integrity of mucosal and epithelial
membrane to facilitate transport (Shen 2003). All these strategies have their pros and
cons and are therefore sometimes used synergistically for a better overall outcome.
3. Cell penetrating peptides as delivery vectors
CPPS are short peptides classified as either cationic or amphipathic
depending on their sequence. The characteristic feature of cationic class of CPPs is
the presence of arginine moiety in the sequence, examples include: tat peptide,
derived from tat protein, Antp peptide derived from the Drosophila antennapedia
protein and synthetic oligoarginine and oligolysine of various lengths (Futaki 2002).
On the other hand, amphipathic peptides contain mostly lysine residue and have even
distribution of hydrophobic and hydrophilic amino acids resulting is an helical
4
structure, examples include, transportan, Pep-1 and the model amphipathic peptide
(MAP) (Fernandez-Carneado et al. 2004).
Cationic CPPs have gained considerable attention as non-invasive delivery
system due to their unique ability to translocate across the plasma membrane and
also assist attached payload in crossing the plasma membrane. Several studies show
intracellular delivery of macromolecules including protein and particulate cargo
using CPPs as delivery vectors (Dietz and Bahr 2004). However, their utility in
transcellular delivery is not fully delineated. The exact mechanism by which CPPs
are internalized into the cell is a matter of huge controversy and to-date there is no
consensus on it. Initially it was proposed that they enter cells through energy
independent, direct translocation mechanisms but recently endocytosis has been
implicated as the major or a partial route for cell entry by some, depending on the
type of CPP, cargo and cell system used (Futaki 2002; Suzuki et al. 2002; Zaro and
Shen 2003; Nakase et al. 2004; Trehin et al. 2004; Kaplan et al. 2005; Zaro and Shen
2005). Understanding the pathway of internalization will be very important in
utilizing and optimizing CPPs as delivery systems.
4. In vitro intestinal and lung epithelial models for transport studies
FDA’s regulation guidelines based on Biopharmaceutical Classification
System (BCS) require that a test molecule be classified based on their solubility and
permeability (Yu et al. 2002). The permeability aspect of the test molecule is
determined by measurement of apparent permeability in in vitro epithelial cell
5
models. From these epithelial cell models, predictions on the in vivo absorption and
transport mechanism can be made. For predicting intestinal absorption, the gold
standard is the Caco-2 cell model. Caco-2 is a heterogeneous cell line derived from a
human colorectal adenocarcinoma that exhibits many features of absorptive intestinal
cells. They are able to differentiate into polarized enterocytes, form well-developed
junctional complexes and express high levels of brush border hydrolases, and a
variety of transporters including efflux pump, P-glycoprotein (Hidalgo et al. 1989;
Augustijns et al. 1993). For these reasons, they serve as an ideal in vitro permeability
model (Artursson et al. 2001). Equally useful is the MDCK cell model, an
immortalized cell line derived from Madin-Darby Canine kidney (MDCK), which
forms tight monolayers and requires lesser time to become confluent compared to
Caco-2 cells (Cho et al. 1989).
To model pulmonary absorption of molecules, both cell lines and primary
cultures from different species have been tested. Table 1.1 below lists the various
models that have been characterized and applied for transport studies. Although a
gold standard for lung absorption model does not exist, primary cultures of alveolar
cells derived from rat lungs have been used for decades to study the transport of
proteins and peptides (Morimoto et al. 1993, 1994; Matsukawa et al. 1996;
Matsukawa et al. 1997; Matsukawa et al. 2000; Kim et al. 2003). There are several
properties that make it a suitable model for predicting in vivo lung absorption.
Firstly, alveolar epithelium represents more than 97% of the total lung epithelial
surface area and is therefore the major barrier to systemic delivery of
6
macromolecules. Additionally, in vitro these cells undergo phenotypic
transdifferentiation from type II-like cells to type I-like cells over time, analogous to
the in vivo situation, and also exhibit high transepithelial electrical resistance
(TEER > 2000 .cm
2
) characteristic of in vivo conditions (Cheek et al. 1989).
Recently, primary alveolar cultures of human origin have been developed, which
display similar characteristics to their rat counterparts, however, the difficulty in
obtaining human alveolar cells, precludes their extensive use (Elbert et al. 1999).
We utilized the primary rat alveolar epithelial cell model consisting of either
type II-like cells, (RAECII) or type I-like cells (RAECI) for studying the transport
characteristics of various cationic CPPs.
Table 1.1: In vitro lung epithelial cell culture models used for
biopharmaceutical research. Adapted from (Sakagami 2006)
Origin Epithelial cells
Human Animal
Primary cultured
cells
Tracheobronchial
Bronchial
Alveolar
Tracheal
Alveolar
Continuous
cultured cells
Cancer derived Calu-1, Calu-3,
Calu-6
H441, HBE1
A427, A549
Transformed
from normal
lungs
9HTE16o-,
16HBE14o-,
1HAEo-,
BEAS-2B,
CF/T43, AK-D
SOPC1
7
5. Transport pathways in epithelial cells.
Transport in epithelial cells can occur through one or a combination of the
following pathways depending on the physicochemical characteristics of the
molecule. Passive diffusion can occur either transcellularly or paracellulary, whereas
active transport can occur through active carrier i.e transporter or receptor-ligand
complex, or through efflux mechanisms (Artursson et al. 2001). Fig 1.1 depicts the
various pathways of drug transport in epithelial cells.
For peptides or peptidomimetics, the transport pathway in epithelial cells is
shown to depend on the molecular size and the charge state of the peptide (Dodoo et
al. 2000). Substrate dipeptides and tripeptides are generally transported by active
transport through transporters such as PepT-1 or PepT-2 (Thwaites et al. 1994)
whereas peptides with a length of 4-6 amino acids are suggested to be transported
paracellularly through the aqueous pores formed between tight junctional complexes
(Dodoo et al. 2000). CPPs generally have a sequence that is more than 6 residues in
length with shorter peptides (< 7 residues) showing no cell penetrating properties
(Mitchell et al. 2000). There are few studies that have focused on the transcellular
transport of CPPs using differentiated epithelial cells with contradictory results. In
one study, tat peptide and hCT, a CPP derived from human calcitonin are shown to
have uptake but no transcellular transport in MDCK cells, whereas another study
shows no permeation at all for tat peptide in MDCK cells (Violini et al. 2002; Trehin
et al. 2004). Similar contradictory results are also seen for Caco-2 cells. Both
8
paracellular and transcellular pathways are suggested for transport of different CPPs
(Lindgren et al. 2004; Zhang et al. 2004; Koch et al. 2005).
The goal of this chapter was to investigate the transport characteristics
of selected cationic CPPs in primary cultures of rat alveolar epithelial cells. Keeping
in mind, that these peptides are delivery systems that can be used to deliver cargo
into and across epithelial cells, it is crucial to map their transport pathway and also
determine their effect on the integrity and viability of epithelial cells. We chose 3
different types of cationic CPPs, oligoarginine, oligolysine and tat peptide as model
peptides. Based on results obtained from this chapter and supporting evidence from
the literature, we focused on oligoarginine as a candidate delivery system for
increasing delivery of peptide and protein drugs.
Efflux
Transport
(Active)
Paracellular
Transport
(Passive)
Transcellular
Transport
(Passive)
Influx
Transport
(Active)
Carrier mediated
endocytosis
(Active)
TJ
APICAL
BASOLATERAL
Facilitative
Transport
(Passive)
Efflux
Transport
(Active)
Paracellular
Transport
(Passive)
Transcellular
Transport
(Passive)
Influx
Transport
(Active)
Carrier mediated
endocytosis
(Active)
TJ
APICAL
BASOLATERAL
Facilitative
Transport
(Passive)
Fig. 1.1 Major transport pathways for compounds in epithelial cells
9
II. EXPERIMENTAL
1. Syntheses and radiolabeling of peptides
All the peptides used in transport studies are listed in table 1.2. In following
the norms in labeling, L-isoform of aminoacids are represented in upper case and D-
iosoform are represented in lower case letter. The peptides were synthesized by
solid-phase synthesis and purchased from Genemed (South San Francisco, CA).
Tyrosine residue was included in each peptide sequence to enable radiolabelling with
125
Iodine using the chloramine-T method (Sonoda and Schlamowitz 1970). Briefly,
1-2 mg of each peptide was dissolved in phosphate buffered saline (PBS), pH 7.0
and mixed with 40 µl Na
125
I (ICN, Irvine, CA) and 50 µl of oxidizing agent
chloramine-T solution in PBS (8mg/ml) and allowed to react for 10min. To the
reaction mixture, 50 µl of reducing agent sodium metabisulfite (4.8mg/ml) was
added and incubated for additional 10 min. The iodination was stopped by addition
of excess potassium iodide (10mg/ml).
No. Name Sequence M
r
§
pI
¥
CD
1 YGR9 YGRRRRRRRRR 1643.93 12.54 7.63
2 YGR15 YGRRRRRRRRRRRRRRR 2581.06 12.81 4.96
3 YGK9 YGKKKKKKKKK 1391.81 10.60 7.62
4 Tat-L YGRKKRRQRRR 1559.84 12.31 7.89
5 yr
7
yrrrrrrr 1274.50 12.40 9.73
6 Tat-D ygrkkrrqrrr 1559.84 12.31 7.89
§
M
r
Average molecular mass
¥
pI Isoelectric point
CD Charge density
Table 1.2. List of CPPs used for transport studies with their sequence, average
molecular mass M
r
, isoelectric point pI, and charge density CD. M
r
, pI, and CD
were theoretically calculated using tools from www.expasy.org.
10
Radiolabelled peptides were purified by size exclusion chromatography using
sephadex G-15 gel matrix and stored at -20ºC until further use.
2. Isolation and culture of type I (RAECI) and type II (RAECII) cells
Lungs of specific pathogen-free male Sprague-Dawley rats were perfused via
the pulmonary artery and lavaged with Ca
2+
/Mg
2+
-free Ringer’s solution. The lungs
were then instilled with 2.0-2.5 U/ml of porcine pancreatic elastase for 20 min at
37°C. Minced lung tissues of ~1 mm
3
blocks were sequentially filtered through 100
µm, 40 µm and 10 µm Nitex membranes (Tetko, Elmsford, NY), followed by
panning on rat IgG-coated bacteriologic plates for 1 hour at 37°C. The partially
purified type II pneumocytes were plated onto tissue culture-treated 12-mm
polycarbonate filters (Transwell, 0.4µm pore size; Corning-Costar, Cambridge, MA)
at a density of 10
6
cells/cm
2
in a culture medium composed of a 1:1 mixture of MEM
and Ham’s F-12 (Sigma) supplemented with 10% newborn bovine serum, 1.25
mg/ml bovine serum albumin (BSA), 100 U/ml penicillin and 100 ng/ml
streptomycin. Monolayers were fed with fresh culture medium on day 3. KGF
(10ng/ml) was added on day 3 to the culture medium to retain type II cell-like
phenotype, whereas culture medium without KGF was used throughout the culture
period to allow differentiation to type I cell-like phenotype. All monolayers were
utilized on days 5-7 in culture. Culture media were changed on day 5 for monolayers
that were utilized on day 6 or day 7.
11
3. Uptake and transport studies
Confluent monolayers of type II or type I cells were first pre-equilibrated
with serum free media for 1 hour at 37 ºC prior to dosing of peptides. Subsequently
the serum free medium was replaced with dosing solution of peptides. In the initial
studies using L-isoform peptides, 2 different concentrations (2 µg/ml and 10 µg/ml)
were tested. The protocol for D-isoform peptides was later modified to dose at
micromolar concentrations. Except for the time-dependent studies, monolayers were
incubated with the peptides for 1 hour after which medium in the receiver
compartment was removed and counted for radioactivity using a gamma counter. To
determine the total cellular uptake, that is membrane bound + intracellular
accumulation, the Transwell inserts were rinsed three times with ice-cold PBS, cut
out and counted for accumulated radioactivity using a gamma counter.
4. Bi-directional transport and apparent permeability
Confluent monolayers of type II cells with an initial TEER value of > 2.5 k
.cm
2
were dosed with 10 µM tat or yr7 in both apical and basolateral
compartments. For apically dosed monolayers, 750 µl of basolateral medium was
sampled for radioactivity every hour for four hours, and replaced with the same
amount of fresh media. Similarly, basolateral dosed monolayers were subjected to
apical sampling of 250 µl at same time intervals, with replenishment of 250 µl fresh
media. For mannitol permeability studies,
3
H-mannitol (0.5 µCi/ml) in the presence
or absence of 10 µM unlabelled Tat-D or yr7 was dosed in either apical or
12
basolateral compartment and sampled as mentioned above. Cumulative transport of
peptides or mannitol as function of time was plotted and the Apparent permeability
of Tat-D and yr7 individually or mannitol in the presence or absence of peptides, was
determined from bi-directional transport studies using the equation:
P
app
= (dc/dt x V)/ (A x C
o
).
Where:
Dc/dt is the unidirectional flux across the monolayers (nM/s)
V is the volume of receptor chamber; apical (0.5 ml), basal (1.5 ml)
A is the surface area of the monolayer; 1.13 cm
2
C
o
is the initial drug concentration in nM
5. Measurement of transepithelial electrical resistance (TEER)
To determine the effect of Tat-D or yr
7
on the TEER of monolayers, cells
were pre-equilibriated with serum free media for 1 hour at 37 ºC prior to dosing of
peptides either in the apical or basolateral compartment. Only monolayers with
TEER of > 2.5 k.cm
2
after pre-equilibration, were used for the study. TEER was
measured using a voltage-ohm meter (EVOM, World Precision Instruments,
Sarasota, FL) at intervals of 60 min up to 240 minutes. Results are presented as % of
initial control values.
13
6. Effect of temperature on transport
To determine the effect of temperature on the transport of Tat-D and yr
7
peptides, all monolayers were first preequilibriated with serum free media for 30 min
at 37 ºC and then another 30min at 37 ºC, 16 ºC or 4 ºC. After dosing 10 µM of
peptide solution in the apical compartment, the monolayers were incubated at their
respective temperatures for 1 hour. Radioactivity in the basolateral medium was
determined by gamma counter.
7. Transport of fluid phase marker FD70 in presence of peptides
Confluent monolayers were pre-equilibriated with phenol red-free media
containing 1mg/ml FITCDextran70 (FD70) or 10 µM of Tat-D or yr7 peptides. An
aliquot from the basolateral chamber was sampled every hour for 4 hours and
replaced with equal amount of fresh media. Fluorescence absorbance was measured
at excitation wavelength 490nm and emission wavelength 518 nm using TECAN
GENios Plus microplate reader.
8. Pulse-Chase assay
Monolayers grown on larger 6-well Transwells were utilized for pulse-chase
assay. Pre-equilibrated monolayers were pulsed with 10 µM of Tat-D or yr7 peptides
in the apical compartment for 1 hour, after which the dosing solution and the
basolateral medium was aspirated. Both compartments were replenished with fresh
medium. At predetermined time intervals, an aliquot from both compartments was
14
taken and counted for radioactivity. Equal volume of media was replaced
immediately in each compartment.
9. MTT assay
This assay was performed according to the method described in (Ishimoto et
al. 2006) with a few modifications. Briefly, freshly isolated alveolar type II cells
were grown in 24- well tissue culture cluster plates at a density of (0.5 x 10
6
cells/well). On day 3 of culture, cells were washed thrice with PBS and incubated
with various concentrations, 0, 1, 10, 25, 50, 100 and 200 µM of either tat or yr7
peptide in phenol red-free RPMI 1640 media for 4 hrs. As a positive control cells
were also incubated with 1% SDS. An MTT (3-(4,5-dimethylthiazole-2-yl)-2,5-
diphenyl tetrazolium bromide) assay was then performed as follows. Cells were
incubated with 1mg/ml of MTT in serum free RPMI 1640 media at 37 ºC for 2 hrs.
The cells were then washed with cold PBS and DMSO was added to dissolve the
formazan crystals. Absorbance was measured at 570 nm on a TECAN GENios Plus
microplate reader. Results are shown as a comparison of treated cells to non-treated
cells.
10. Statistical Analysis
Data is presented as mean ± stdev. For each experiment, samples (n) ranged from 3-4
for each group. Where appropriate, differences between groups were determined by
student’s t-test and a p value of <0.05 was considered statistically significant.
15
III. RESULTS
1. Transport of YGR9, YGR15 and YGK9
To compare the transport efficiencies of various cationic peptides across
alveolar epithelial cells, YGR9, YGR15 and YGK9 at 2 different concentrations, 2
µg/ml and 10 µg/ml were tested (Fig. 1.2). For both concentrations, YGR9 appeared
to be transported significantly higher than YGR15 and YGK9. There was no
statistical significant difference in the transport of YGR15 and YGK9. These values
represent total peptide transported as determined by total radioactivity count. To
differentiate intact versus degraded peptide, basolateral media was analyzed by size-
exclusion chromatography, however considerable amounts of transported peptides
were digested into fragments that were smaller than the exclusion limits of sephadex
0
10
20
30
40
50
60
YGR9 YGR15 YGK9
ng Peptide/Monolayer .
(2ug/ml)
(10ug/ml)
Fig. 1.2. Total transport of YGR9, YGR15 or YGK9 across RAECII.
RAECII monolayers were incubated with 2 µg/ml or 10 µg/ml
oligopeptides for 2 hours. Total amount of peptide transported was
determined by radioactivity. Data are presented as mean ± stdev. n=4.
16
G- 15 (data not shown). Sensitive methods of analysis to separate and quantify
degradation were not pursued.
2. Comparison of transport in RAECI vs RAECII
Efficiency of total transport of the oligopeptides tested above was compared
between RAECI and RAECII. For all three peptides 10 µg/ml concentrations were
used. RAECI also exhibited higher transport of YGR9 compared to YGR15 and
YGK9. No apparent difference in transport of YGR15 and YGK9 between RAECI
and RAECII was observed. However transport of YGR9 appeared to be significantly
higher in RAECII compared to RAECI.
0
10
20
30
40
50
60
YGR9 YGR15 YGK9
ng Peptide/Monolayer.
Type I cells
Type II cells
Fig. 1.3: Comparison of total transport in RAECI vs RAECII. Monolayers
were incubated with 10 µg/ml oligopeptides for 2 hours. Total amount of
peptide transported was determined by radioactivity. Data are presented as
mean ± stdev. n=4
17
3. Transport of L-isoform vs D-isoform
Transport of proteolyticaly stable YGr9 was compared to the transport of
YGR9 and TAT in RAECII. Total transport of YGr9 was 40% lower compared to
YGR9 and no difference when compared to TAT peptide. However upon analysis of
intact versus digested peptides using size exclusion chromatography, it was
determined that digestion of YGr9 still occurred to a great extent possibly due to
cleavage between tyrosine (Y) and glycine (G) residues.
0
10
20
30
40
50
60
70
80
90
TAT YGR9 YGr9
ng peptide/monolayer
10 ug/ml
Fig. 1.4: Transport of TAT, YGR9 and YGr9 across RAECII. RAECII
monolayers were incubated with 10 µg/ml of YGR9 and YGr9 for 2 hours.
Total amount of peptide transported was determined by radioactivity. Data
are presented as mean ± stdev. n=4
18
4. Concentration dependent transport and uptake of Tat-D and yr7
Since the presence of any L isoform of residues in the peptide sequence led to
extensive degradation of the peptides leading to a mixture of digested products, we
decided to use peptides that consisted exclusively of D amino acids. Peptide
sequence yr7 (yrrrrrrr) was used along with tat peptide (ygrkkrrqrrr) to further study
the transport characteristics in AECII. Unidirectional transport and uptake at a
concentration range 1 – 20 µM of yr7 and tat peptide was tested. In this
concentration range, the transport of both peptides was linear and non-saturable
indicative of a passive mechanism of transcellular or paracellular transport (Fig. 1.5)
Transport of yr7 was marginally higher than tat peptide at 10 and 20 µM. The
cellular uptake of both peptides was also concentration dependent, correlated
positively to transport across monolayers and did not saturate at the concentrations
tested (Fig. 1.6).
0
5
10
15
20
25
30
35
40
05 10 15 20 25
Concentration, [µM]
nmol/Monolayer.
Tat
yr7
Fig. 1.5: Concentration dependent transport in RAECII. Monolayers were dosed
with 1-20 µM of either yr7 or Tat-D peptide in the apical compartment and
incubated at 37 ºC for 1 hr. The amount of radiolabelled peptide accumulated in
the basolateral compartment was determined. Data are presented as mean ±
stdev. n=4
19
5. Bi-directional transport of yr7 and Tat-D
Bi-directional transport, i.e. apical to basolateral (A-B) and basolateral to
apical (B-A) in AECII was performed using 10 µM of yr7 or Tat-D peptides. A-B
and B-A transport of both peptides was symmetrical and time dependent (Fig. 1.7).
Apparent permeability for yr7 was calculated to be 1.58 x 10
-7
cm/s for A-B
direction and 2.26 x 10
-7
cm/s for B-A direction. The values for Tat-D were 1.00 x
10
-7
cm/s for A-B and 1.91 x 10
-7
cm/s for B-A direction. The B-A: A-B ratio is 1.4
and 1.9 for yr7 and Tat-D respectively. From the apparent permeability values and
the B-A: A-B permeability ratios the transport of yr7 is symmetrical whereas Tat-D
is asymmetrical with a higher B-A permeability than A-B.
0
20
40
60
80
100
15 10 20
Concentration, [µM]
nmol/Monolayer
Tat
yr7
Fig. 1.6: Concentration dependent uptake in RAECII. Monolayers in 12
were dosed with 1-20 µM of either yr7 or Tat-D peptide in the apical
compartment and incubated at 37 ºC for 1 hr. The amount of radiolabelled
peptide accumulated the cell monolayers was determined after washing the
monolayer inserts thrice with ice-cold PBS. Data are presented as mean ±
stdev. n=4
20
6. Apparent permeability of mannitol in presence and absence of peptides
To determine if Tat-D or yr7 influenced paracellular permeability, we tested
A-B and B-A transport of mannitol in the presence and absence of these peptides.
Table 1.3 shows the apparent permeability of mannitol as calculated from the
equation described in experimental section. There was no difference in the Papp of
mannitol, either in the A-B or B-A direction in the presence of 10 µM yr7. In the
presence of Tat-D, A-B Papp of mannitol did not change, however B-A Papp
increased moderately from 1.4 (± 0.15) x 10
-6
to 5.26 (± 0.17) x 10
-6
cm/s.
A-B
0
5
10
15
20
25
30
35
0 100 200 300
T ime [min]
nmol/cm
2
B-A
0
5
10
15
20
25
30
35
0 100 200 300
T ime [min]
nmol/cm
2
A-B
-5
5
15
25
35
0 100 200 300
T ime [Min]
nmol/cm
2
B-A
0
5
10
15
20
25
30
35
0 100 200 300
T ime [min]
nmol/cm
2
I
II
III IV
Fig. 1.7: Bidirectional transport of 10 µM Tat-D and yr7 peptides as a function
of time. (I) A-B transport of yr7, (II) B-A transport of yr7 (III) A-B transport
of Tat-D, (IV) B-A transport of Tat-D. Data represents mean ± stdev, n= 4
21
Substrate Apparent Permeability (cm/s)
A-B B-A
Mannitol 1.91(±0.35)x10
-6
1.40 (± 0.15) x 10
-6
Mannitol + yr7 (10µM) 1.67 (± 0.01) x 10
-6
1.58 (± 0.16) x 10
-6
Mannitol + Tat-D (10µM) 2.21 (± 0.35) x 10
-6
5.26 (± 0.17) x 10
-6
7. Transient and reversible decrease in TEER over time
Effect on the tight junctional barriers was studied by determining the change
in TEER over time when confluent monolayers were incubated with yr7 and Tat-D
both apically and basolaterally (Fig. 1.8). It was observed that for yr7, TEER
dropped to a maximum of 82 ± 11% of control for apical dosing and 75 ± 3% of
control for basolateral dosing, after 1 hr of incubation and gradually recovered to
control values in 4 hrs. A similar pattern was observed when monolayers were
incubated with mannitol or both mannitol and yr7. On the other hand, there was no
change in TEER when tat was incubated apically but a gradual decrease was
observed when incubated in the basolateral compartment. Monolayers incubated
with mannitol plus Tat-D also showed a drop in resistance up to 60 ± 4 % of control.
Taken together, these data indicate that yr7 and tat peptide do not decrease the
monolayer integrity at 10 µM concentrations and any change in TEER is transient
and reversible. It is important to note that monolayers that showed initial TEER
values between 2.5 -4.0 k.cm
2
were used for this study.
Table 1.3. Apparent permeability of mannitol in A-B and B-A obtained in RAECII
in the presence and absence of Tat-D and yr7 peptides.
22
(A-B)
0
20
40
60
80
100
120
140
0 50 100 150 200 250 300
Time (Min)
% Control
yr7
Mannit ol
yr7 + M
(B-A)
0
20
40
60
80
100
120
140
0 50 100 150 200 250 300
Time (M in)
% Control
yr7
Mannit ol
yr7 + M
(A-B)
0
20
40
60
80
100
120
140
0 60 120 180 240
T ime (Min)
% Control
T at
Mannit ol
T at + M
(B-A)
0
20
40
60
80
100
120
0 60 120 180 240
T ime (Min)
% Control
Tat
M annitol
Tat + M
I II
III
IV
Figure 1.8: TEER of confluent monolayers at different time points after addition of 10μM peptide, mannitol or
mannitol and peptide in either the apical or basolateral compartment. (I) Apical dosing; (II) Basolateral dosing
of yr7, mannitol or Mannitol + yr7 (III) Apical dosing; (IV) Basolateral dosing of tat, mannitol or Mannitol +
Tat-D. Data are presented as mean ± stdev. n= 4.
23
8. Effect of temperature on transport of yr7 and Tat-D
Effect of temperature on transport of yr7 and Tat-D was studied by
comparing the transport at 37 ºC and 4º C. It is known that lowering temperature
does not affect the paracellular route of transport in epithelial cells, but lowering the
temperature below 18 ºC affects transcellular transport by endocytosis. Therefore to
further distinguish the role of transcellular transport from paracellular transport, the
effect of temperature was determined. Transport of both peptides was seen to be
temperature-dependent with 74 % and 65% decrease in transport at 4 ºC compared to
37 ºC for tat and yr7 respectively.
0
2
4
6
8
10
12
14
Tat-D yr7
nmoles/Monolayer
37 oC
4 oC
Fig. 1.9: Effect of temperature on transport of yr7 and Tat-D peptide.
Cells were incubated at 37 ºC or 4 ºC for 1 hr in the presence of 10 µM
peptide. Radioactivity in the basolateral compartment was determined.
Data represent mean ± stdev. n=3-4
24
9. Effect of Tat-D or yr7 on transport of FD70
To determine if Tat-D or yr7 influenced the transport of fluid phase marker
FITC-dextran (FD70; MW 70 kDa), we studied the transport of FD70 in the
presence of 10 µM Tat-D or yr7 peptides. Fig. X shows that there was no increase or
decrease in the transport of FD70 in the presence of Tat-D or yr7. Additionally, the
transport of Tat-D or yr7was not affected by the presence of FD70 (data not shown).
0
2
4
6
8
10
12
14
16
18
20
0 50 100 150 200 250 300
Time (Min)
ng FD70/monolayer
FD70
FD70 + yr7
FD70 + Tat-D
Fig 1.10. Effect of yr7 or Tat-D on the transport of FD70. Confluent RAECII
monolayers were incubated with FD70 (1mg/ml) and Tat-D or yr7 (10 µM).
Transport of FD70 was determined at various time points by measuring
fluorescence absorbance at Ex 490nm and Em 518 nm. Data are presented as
mean ± stdev and n = 3.
25
10. Pulse chase assay of yr7 and Tat-D
To investigate whether the tat peptides are taken up by the cells and then
rapidly released across the basolateral membrane, the monolayers were pulsed with
10 µM yr7 or tat apically, incubated for 1 hr and washed with ice-cold PBS to
remove non-absorbed radioactivity. Fresh media was supplemented in both apical
and basolateral chambers and the amount of radioactivity in basolateral chamber
determined. Yr7 release across the basolateral membrane was initially rapid for 6
hours and reached a plateau at 24 hours. Up to 48 hrs of chase, the values of
accumulation did not exceed 6 nM and 15 nM per monolayer for yr7 and tat
respectively.
0
2
4
6
8
10
12
14
16
18
20
0 10 20 30 40 50 60
Time (hrs)
nM/monolayer
Tat
yr7
Fig 1.11. Time course of basolateral release of Tat-D or yr7 after apical pre-
incubation for of 10 µM peptide for 1hr at 37 ºC. Radioactivity in the
basolateral compartment was determined at various time points. Data are
presented as mean ± stdev where n=3
26
11. Cytotoxicity of yr7 and Tat-D
To determine the effect of yr7 and tat peptide on the viability of alveolar
epithelial type II cells, MTT assay was performed. As shown in table 1.4, no toxicity
was observed upto a concentration of 100µM, however at 200µM, a 22 ± 2% and 10
± 6% drop in cell viability was observed for yr7 and tat respectively. As a positive
control, 1% SDS was also tested and showed a 90% drop in cell viability.
IV. DISCUSSION
Cationic CPPs were previously thought to penetrate through cell membrane
rapidly and efficiently through energy independent mechanisms (Vives et al. 1997;
Suzuki et al. 2002). This notion has been refuted lately, with more evidence
suggesting the existence of multiple mechanisms for cell entry (Zaro and Shen
2003). While several studies have focused on their intracellular transport pathway
MTT Assay
Control 100 ± 4 %
1% SDS 10 ± 1 %
1µM 10µM 25µM 50µM 100µM 200µM
yr
7
102 ± 6 % 104 ± 9 % 97 ± 20 % 83 ± 13 % 79 ± 8 % 78 ± 2 %
Tat-D 78 ± 17 % 97 ± 20 % 97 ± 12 % 87 ± 16 % 92 ± 14 % 90 ± 6 %
Table 1.4: MTT assay for various concentrations of yr7 and tat peptide in
AECII. Data are presented as % of control, whereby control indicates cells
incubated with phenol red-free RPMI medium. Values are mean ± stdev. n = 4
27
and applications for intracellular delivery, little is known about their transcellular
transport behavior and their suitability for transcellular delivery. In this chapter, we
studied the transport characteristics of selected cationic CPPs across alveolar
epithelial cells and their effect on cell viability.
We compared the transcellular transport efficiency of three different peptides,
oligoarginine, R9 and R15 and oligolysine, K9 in RAECII. To allow radiolabelling
of these peptides; we introduced tyrosine (Y) and glycine (G) moieties. Our results
indicate that YGR9 is transported more efficiently than YGK9. Although K9 and R9
have similar charge density and the molecular weight of YGK9 is in fact lower than
YGR9, the transport of YGR9 is significantly higher than YGK9, suggesting that the
increase in transport is due to structural features of arginine and not simply because
of charge. This finding is in accordance with several studies that show higher uptake
of oligoarginine by virtue of its guanidine group compared to oligolysine (Wender et
al. 2000; Zaro and Shen 2003). The length of oligoarginine and hence number of
guanidine group has been shown to correlate with the efficiency of uptake. A
comparison of R5, R7 and R9 showed that R9 had optimum efficiency in cell uptake
(Mitchell et al. 2000; Wender et al. 2000). To determine if further increasing the
length of oligoarginine would result in higher transport, we increased the length from
R9 to R15, however in our studies increasing the length of oligoarginine (YGR15)
did not result in increased transport.
Transport studies were performed in alveolar epithelial cells, to determine the
feasibility of using these peptides for systemic delivery by pulmonary route. The
28
alveolar epithelium forms the major transport barrier and it consists of type I and
type II cells. Type I cells occupy 95% of surface area and therefore contribute
significantly to the barrier. To determine if the transport of peptides was cell type
dependent, we compared the transport of YGR9, YGR15 and YGK9 in cell models
that represent type I cells (RAECI) to that of type II cells (RAECII). A similar
pattern was observed in RAECI to that of RAECII for all the three peptides, with
highest transport seen for YGR9. However the transport of YGR9 in RAECII was
moderately but statistically significantly higher than RAECI. This difference in
transport could possibly be explained by differential expression of proteoglycans,
suggested binding sites for CPPs, in RAECII compared to RAECI (Vaccaro and
Brody 1981). Alternatively, different cellular pathways could be involved in
transporting these peptides in the two cell types. Further studies are required to
implicate either of these possibilities. Nevertheless, the difference in efficiency of
transport is not sufficiently low in RAECI to suggest cell-dependent transport and
both cell types can potentially contribute in the transport of these peptides in lungs.
Analysis of the transported peptides indicated high metabolic activity with
substantial degradation of peptides. Therefore, we investigated the transport of the
metabolically stable D-isoform of R9 (r9). Transport of YGr9 was significantly
lower than YGR9, but degradation was still observed presumably because of
protease cleavage between L-isoform residues of tyrosine (Y) and glycine (G).
Alveolar epithelial cells have been shown to exhibit high aminopeptidase activity
(Forbes et al. 1999), furthermore it was demonstrated that incubation of CPPs
29
namely tat peptide, hCT (human calcitonin) derived CPP and penetratin with both
Calu-3 and MDCK epithelial cells results in metabolites formed due to N-terminal
endopeptidases with further degradation of metabolites occurring due to
aminopeptidases and/or carboxypeptidases (Trehin et al. 2004). Efficient metabolism
of delivery system is a prerequisite for cargo release when conjugated, however there
is potential for premature release prior to cell penetration, therefore it is necessary to
select or engineer CPPs that are metabolically stable, yet can release the attached
cargo. Two approaches to achieve stability during transport are to use protease
inhibitors along with the peptides or replace all L-isoform residues with D-isoforms.
We chose to use an exclusively D-isoform peptide for maximum proteolytic stability.
The peptide sequence was modified, by removing glycine and also a shorter length
r7 was maintained since heptaarginine is shown to be the minimum length required
to possess cell-penetrating properties. The resultant peptide yr7 was used along with
Tat-D (an exclusively D-isoform of tat peptide) for further studies in RAECII.
Both the uptake and transport of yr7 and Tat-D was seen to be dose
dependent at concentrations from 1-20 µM and did not saturate at these
concentrations suggesting no involvement of a saturable component such as a
transporter or receptor. Transport of yr7 was marginally higher than that of Tat-D at
10 µM and 20 µM, although there was no difference between the two in uptake at all
the concentration tested. The difference in transport can possibly be attributed to the
number of arginine residues. While yr7 has a total of 7 arginines, Tat-D has only 6
arginines. Interestingly, it was shown that heptaarginine (R6), had similar uptake to
30
tat peptide which also consists of 6 arginine residues (Wender et al. 2000), thus the
marginal increase could be due to the extra arginine in yr7.
Since alveolar epithelial cells are polarized, with different architecture and
membrane components on the apical versus basolateral surface, we sought to
determine if there was a difference in the directionality of transport in peptides. We
investigated the A-B and B-A transport of both peptides at 10 µM. The transport for
yr7 was time dependent and symmetrical with a B-A: A-B apparent permeability
ratio of 1.4. Additionally the permeability of mannitol, a paracellular marker did not
change in either direction in the presence of yr7 and its effect on TEER was not
significant, characterized by a transient but reversible decrease. These results suggest
that the transport of yr7 occurs through passive diffusion either in a transcellular or
paracellular route and yr7 does not affect the paracellular route of transport.
Tat-D also showed time dependent transport but with higher B-A
permeability compared to A-B as indicated by a 1.9 ratio of B-A: A-B apparent
permeability. This increase could be attributed to the activity of P-glycoprotein since
P-gp is localized apically in (Campbell et al. 2003), its substrates are cationic and
peptides as large as 12 amino acids such as yeast -matting factor can be transported
by P-gp (Raymond et al. 1992). However, a more likely explanation for the increase
is the modulating effect of Tat-D on the basolaterally accessible tight junctions. The
higher B-A permeability of Tat-D correlates well with an increase in permeability of
mannitol in the B-A direction and also an ~ 20% irreversible drop in TEER both in
the presence and absence of mannitol. A similar observation of higher B-A
31
permeability was made by two different groups when Tat-D peptide was incubated in
the basolateral compartment of MDCK monolayers or Caco-2 monolayers (Violini et
al. 2002; Trehin et al. 2004). Why this effect occurs only from the basolateral
domain and not from the apical domain is not clear.
A key finding in our studies is the effect of temperature on the transport of
yr7 and Tat-D peptides. From the bi-directional studies, we could not conclude if the
transport is paracellular or transcellular. Decreasing the temperature from 37 ºC to 4
ºC resulted in 74% and 65% decrease in transport for Tat-D and yr7 respectively.
This is a clear indication that plasma membrane is involved in the transport of the
peptides because at 4 ºC, the membrane fluidity is severely restricted, slowing down
transport through the membrane. If passive diffusion by only paracellular route was
involved, the decrease in transport is expected not to be more than 40% of 37 ºC
control (Matsukawa et al. 1996). Therefore, a transcellular component in transport is
suggested.
We further show the involvement of transcellular pathway in our pulse-chase
studies. It is seen that after apical pulse with the peptides, there is linear increase in
basolateral release up to 6hrs, which saturates at further time points. Since a
moderately higher A-B transport of yr7 is seen compared to Tat-D, it would be
expected that the basolateral release of yr7 would be moderately higher or similar to
that of Tat-D instead of lower release. This discrepancy can possibly be explained by
higher binding affinity of yr7 to intracellular or membrane components compared to
32
Tat-D that hinder the release of yr7 to basolateral compartment. However, we do not
have evidence to prove this suggestion.
Recent studies showed that oligoarginine and Tat peptide are internalized by
pinocytotic processes and are also able to stimulate macropinocytosis in Namalwa or
Hela cells (Nakase et al. 2004; Kaplan et al. 2005). This stimulation would lead to an
increased uptake of solutes through fluid-phase endocytosis. In addition, an increase
in paracellular transport of FITC Dextran 4.4 kDa, a fluid phase marker in nasal
epithelial cells was seen in the presence of high MW poly-L-arginine (Ohtake et al.
2002). We investigated whether low MW oligoarginine (yr7) or Tat-D could increase
the transcellular transport of FITC dextran (FD70) either through increased fluid-
phase endoyctosis or through paracellular route. Our results show that there was no
increase in FD70 transport in alveolar epithelial cells due to yr7 or Tat-D peptide.
Finally, we studied the short-term cytotoxicity of yr7 or Tat-D on RAECII up
to a concentration of 200 µM using MTT assay. For Tat-D peptide, no significant
cytotoxicity was observed up to 200 µM. At 200 µM, the cell viability reduced by
only 10 ± 6%. For yr7, statistically significant reduction in cell viability was seen at
100µM with 21 ± 8% decrease. Cytotoxicity studies of CPP are limited, however the
effect of Tat peptide on Hela, MDCK, Calu-3 and TR146 cells show that there was
no toxic effect up to 100 µM but toxicity was seen at a concentration as high as 1
mM (Vives et al. 1997; Trehin et al. 2004). In conclusion, our studies indicate that
oligoarginine and Tat peptide (D-isoforms) may have potential as delivery systems to
penetrate across alveolar epithelial cells.
33
V. SUMMARY
YR9, YK9 or YR15
L vs D-isoform
RAECII vs RAECI
Yr7 vs Tat-D
Mannitol permeability
FD70 permeability
Cytotoxicity
Bi-directional transport
TEER
YR9 most efficient
D-isoform more stable but less efficient
Higher transport in RAECII
Yr7 has marginally higher transport
Symmetrical or near symmetrical transport
Transient and reversible effect on TEER
No major effect on mannitol permeability
No effect on FD70 permeability
Yr7 > 100 μM. Tat > 200 μM.
Cationic CPP
Temperature Decrease in temp = decrease transport
YR9, YK9 or YR15
L vs D-isoform
RAECII vs RAECI
Yr7 vs Tat-D
Mannitol permeability
FD70 permeability
Cytotoxicity
Bi-directional transport
TEER
YR9 most efficient
D-isoform more stable but less efficient
Higher transport in RAECII
Yr7 has marginally higher transport
Symmetrical or near symmetrical transport
Transient and reversible effect on TEER
No major effect on mannitol permeability
No effect on FD70 permeability
Yr7 > 100 μM. Tat > 200 μM.
Cationic CPP
Temperature Decrease in temp = decrease transport
Scheme 2: Summary of findings in chapter 1
34
CHAPTER 2
Cationic CPP Mediated Peptide Delivery: Desmopressin-Oligoarginine
for Oral Delivery
I. INTRODUCTION
1. Desmopressin as a choice for peptide cargo
Desmopressin is a synthetic analogue of the natural antidiuretic
hormone vasopressin. Unlike vasopressin, desmopressin has no vasopressor activity
but has only antidiuretic activity (Vavra et al. 1968). The selective antidiuretic
activity is due to its ability to bind to only V-2 receptors and not to V-1 receptors. V-
2 receptors are G-protein coupled receptors present in the collecting ducts of the
kidney and are responsible for promotion of water reabsorption via stimulation of
cyclic AMP production (Kaufmann and Vischer 2003). Clinically, desmopressin is
used for the treatment of diabetes insipidus and nocturnal enuresis. It is also used as
a haemostatic drug in patients with mild haemophilia A, von Willebrand’s disease or
platelet dysfunction due to its ability to cause release of endogenous factor VIII, von
Willebrand factor and tissue plasminogen activator into the blood circulation
(Mannucci et al. 1977). It is usually administered orally or intranasally, however the
bioavailability of the peptide is only 0.1% for the oral route and 3.4% for the nasal
route in humans (Fjellestad-Paulsen et al. 1993). In rats, the oral bioavailability has
been shown to be even lower than human and the least among the different species
tested. The major reason attributed to the low bioavailability was poor permeability
properties of the rat intestinal mucosa to the hydrophilic drug (Lundin et al. 1994).
35
Various prodrug strategies have been applied to desmopressin to improve its
bioavailabilty (Kahns et al. 1993). For example, our lab has shown that reversible
lipidization of desmopressin leads to a 250-fold increase in its antidiuretic potency
when administered subcutaneously (Wang et al. 1993; Wang et al. 1999).
Desmopressin was chosen as a model of peptide drug to be modified using
CPPs for the following reasons. Firstly, it is a low molecular weight peptide with an
inherent disulfide bond, which can be used to ligate CPPs. Fig. 2.1 below, shows the
structure of desmopressin and the CPP conjugate. It also has a tyrosine residue that
can be radiolabelled with
125
I to trace the molecule. Secondly, as mentioned above, it
has low permeability and poor bioavailability when administered orally and
intranasally, therefore any enhancement in the permeability and increase in
bioavailability due to conjugation with CPP will be clearly evident. Furthermore it is
relatively cheap and a safe drug with a wide therapeutic window.
Also a well-established animal model for the human disease of diabetes
insipidus exists which can be used to study the in vivo effect of desmopressin and
s
s
OC -CH
2
-CH
2
Tyr
Phe
Gly
Asn Cys
Pro D-Arg Gly -NH
2
s
s
OC -CH
2
-CH
2
Tyr
Phe
Gly
Asn Cys
Pro D-Arg Gly -NH
2 Asn Cys Pro
D -Arg Gly -NH
2
Tyr Phe Gly CH
2
-CH
2
CO
S
Cys
D -(Arg )7
S
S
Cys
D -(Arg )7
S
Asn Cys Pro
D -Arg Gly -NH
2
Tyr Phe Gly CH
2
-CH
2
CO
S
Cys
D -(Arg )7
S
S
Cys
D -(Arg )7
S
Fig. 2.1: Structure of desmopressin and desmopressin conjugated to two Cr7 by
disulfide bond
36
desmopressin-CPP conjugates. Brattleboro rats exhibit hereditary diabetes insipidus
due to a congenital defect that prevents them from producing vasopressin and
concentrating urine (Trinh-Trang-Tan et al. 1982).
2. Reversible disulfide bridging as a strategy for drug delivery
A disulfide bond (-s-s-) is a covalent bond that forms when oxidation of two
sulfhydryl groups present in cysteine or other –SH containing molecules occurs. It is
deemed as an attractive strategy in drug delivery because the covalent linkage is
reversible, allowing for the release of biologically active molecule and also because
it is relatively stable in plasma. The reversibility of the disulfide bond is dictated by
the nature of the microenvironment in which the macromolecule exists. In the
extracellular space, oxidizing conditions prevail and therefore the covalent link is
maintained, at different subcellular levels reducing conditions cause cleavage of the
bond thus releasing the attached component.
Bacterial and plant toxins such as cholera, diptheria and ricin take advantage
of the reversible breakage of disulfide bond during the process of translocation
across the plasma membrane into the cytosol of host cells (Mandel et al. 1991; Ryser
et al. 1991; Watson and Spooner 2006). Similarly, to deliver membrane impermeable
macromolecules, a delivery enhancing system that assists in translocation across
plasma membrane is attached through disulfide bridging. Other creative applications
of disulfide bonding in addition to enhanced drug delivery include: targeted delivery,
improving pharmacokinetics and increasing stability (Saito et al. 2003). Thiol-based
37
strategy has been employed for years in cell systems as well as pre-clinical models
for a variety of macromolecule including peptides, proteins, oligonucleotides and
plasmid gene (Saito et al. 2003). However, it was only a few years ago that FDA
approved Mylotarg, the first drug containing synthetic disulfide chemistry between
an antibody and an anticancer agent (Niculescu-Duvaz 2000).
An important aspect to consider when using thiol conjugation technique is the
method used to introduce thiol groups in the two molecules to be conjugated. In case
of proteins and peptides that have cysteine residues, inherent thiol groups are
sufficient and may not require external addition of thiols. In the absence of cysteine
in the peptide or protein, or an inaccessible cysteine, two methods can be resorted to.
First, cysteine can be incorporated into a peptide during chemical synthesis of the
full-length peptide, or, for proteins, it can be introduced by site-directed amino acid
substitution. Secondly, the N-terminal residue of the peptide or protein or amino
group on lysine residue can be modified with commercially available
heterobifunctional cross-linkers that introduce free thiol groups to the protein or
peptide. These include N-succinimidyl-3-(2-pyridyldithio)propionate (SPDP), N-
succinimidyl S-acethylthioacetate (SATA) and 2-iminothiolane (Ghetie and Vitetta
2001). In this dissertation, we introduced cysteine to the delivery system (CPP)
during chemical synthesis, whereas for the cargo we used two separate approaches.
For the peptide cargo, desmopressin, we utilized the two inherent cysteines to
conjugate 2 molecules of Cr7. In the next chapter insulin is used as a protein cargo
38
and the heterobifunctional cross linker, SPDP is used to introduce a thiol group in the
lysine residue present in B-chain.
Another important factor to consider is the locations where reduction of the
disulfide bond is likely to occur. The transport pathway used by the disulfide
conjugate will influence where and to what extent reduction occurs. However, the
transport pathway used by CPP or CPP conjugates is not well understood. At the
cellular level, several redox enzymes and redox agents have been characterized that
could serve as potential sites for reduction. These include, surface associated redox
enzymes such as protein disulfide enzyme (PDI) (Mandel et al. 1991), NADH-
oxidase(NOX) (Morre et al. 1998) and thioredoxin (Sahaf et al. 1997) or cytosolic
redox potential maintained by GSH: GSSG ratio (Meister and Anderson 1983). In
systemic circulation, the enviroment is prevalently oxidizing, however there are low
concentrations of cysteine (~ 8 µM) and glutathione (~2 µM) that can gradually
reduce and release the disulfide linked molecules (Jones et al. 1998; Jones et al.
2000).
3. Delivery of peptide cargo using CPP.
In the last two decades, CPPs have been widely used to deliver a range of
molecules that vary greatly in size and nature both in vitro and in vivo. Several non-
permeable peptides have been conjugated to CPP for intracellular delivery and show
evidence of internalization and biological activity (Mae and Langel 2006). For
example, P16 peptide conjugated to different CPP is shown to inhibit
hypophosphorylation of the retinoblastoma protein causing an early cell cycle arrest
39
(Zaro and Shen 2005). Rothbard et al reported a striking study that partly led us to
propose testing a peptide cargo for delivery across epithelial cells. They used
cyclosporine A, a peptide-based hydrophilic drug to conjugate to oligoarginine and
tested the feasibility of transdermal delivery (Rothbard et al. 2000). Their results
showed efficient and effective delivery of cyclosporine A across cells into mouse
and human skin. We therefore decided to study the effect of conjugating peptide
drug, desmporessin to oligoarginine and determine if it was feasible to cross
epithelial barriers using this strategy.
II. EXPERIMENTAL
1. Bioconjugation of DDAVP with Cr7
To conjugate heptaarginine to DDAVP by the disulfide linkage, a cysteine
residue was added to the N-terminal of heptaarginine during solid phase synthesis of
the carrier, designated as Cr7. Cysteine moiety in Cr7 was activated using DTDP.
12.4 mg of Cr7 dissolved in 200 µl DMF was reacted with 2-fold excess of DTDP
for 1.5 hrs at room temperature. The reaction mixture was diluted with 600 µl of
0.05M NH
4
HCO
3
and purified using size exclusion chromatography using Sephadex
G15 gel matrix (bed volume; 10ml) and 0.05M NH
4
HCO
3
as eluting solvent.
Fractions 5-9 were pooled and lyophilized. The percentage modification of Cr7 was
estimated by DTDP modification assay. Briefly, an aliquot of Cr7 dissolved in PBS
was reacted with 5 µl of 0.1M DTT. The change in absorbance at 343 nm before and
40
after 5min of DTT addition was monitored. The equation below was used to
calculate the amount of Cr7 activated.
Mol (DTDP) ABS /8080
---------------- = -----------------------------
Mol (Cr7) [Cr7 (mg/ml)]/ Mwt
Where ABS is the change in absorbance at 343 nm, 8080 is the extinction
co-efficient for pyridine-2-thione at 343 nm, Mwt is the molecular weight of Cr7 and
Cr7 (mg/ml) is the concentration of the aliquot reacted with DTT.
DDAVP (2mg/ml) was dissolved in PBS pH 7.4 and reacted with equimolar
DTT at 37 ºC for 30 min to reduce disulfide bond. Activated Cr7 was added
dropwise at a ratio of 4:1 (activated Cr7: DDAVP) and allowed to react at room
temperature for 1 hr. Generation of pyridine-2-thione was monitored by UV
spectroscopy at 343nm. After the reaction was complete, the mixture was purified
using size exclusion using sephadex G15 gel matrix (bed volume; 20ml) and 0.05M
NH
4
HCO
3
as eluting solvent. % Yield was estimated by absorbance at 280 nm. The
reaction scheme is shown in Fig 2.2
41
N
S
N
S
2,2’-
dithiodipyridine
Cys-hepta-d-Arg
N
O
S
N
Arg7
N
S
Modification of CysHepta-d-Arg:
+
=
N
O
S
N
Arg7 H +
S
N
H
TP-Cys-heptaArg
Thiopyridine
N
O
S
N
Arg7
N
S
TP-Cys-heptaArg
+
Reduced Desmopressin
=
Drug
SH
Drug-Cys-heptaArg
Disulfide linkage of Drug-CR7:
SH
Figure 2.2: Reaction scheme of bioconjugation of DDAVP to Cr7
42
2. TLC analysis
DDAVP, DDAVP-Cr7, reduced DDAVP-Cr7 or DTT were spotted on a TLC plate
and dried. The plate was developed using the organic layer of a mixture of n-
butanol:water: acetic acid (4:5:1) and dried before exposing the TLC plate to Iodine
to detect the peptides. Retention factor (RF) values were calculated based on the
migration of each spot relative to solvent front.
3. MALDI-TOF mass spectroscopy
The definite molecular mass identification of DDAVP-Cr7 was confirmed
using the Krotos Kompact MALDI-TOF Mass Spectrometer at the USC proteomics
core. Approximately 10 pmol of the conjugate was spotted on a matrix consisting of
10 mg/ml -cyano-4-hydroxycinnamic acid in 1% TFA /70% acetonitrile and
subjected to MALDI-TOF using an Axima-CFR in the linear mode.
4. Radiolabelling of DDAVP and DDAVP-Cr7
DDAVP and DDAVP-Cr7 were radiolabelled with
125
I using the chloramine-
T method described in chapter 1. Free
125
I was separated from radiolabelled products
using Sephadex G10 and 100% DMF as mobile phase
5. In vitro transport assay
MDCK type I cells were cultured according to the methods described in
(Taub et al. 1994). Confluent monolayers of MDCK cells grown on 12-well
43
Transwells were equilibriated with serum-free medium for one hour and dosed with
10 µg/ml of
125
I-DDAVP or equivalent DDAVP-Cr7. The basolateral compartment
was sampled after 2 hours and radioactive count determined. Similar transport
experiment was performed in RAECII model described in chapter 1.
6. Size-exclusion analysis of transported conjugate
To determine the percentage of intact conjugate from the total amount
transported, basolateral media collected from the transport experiment in RAECII
and MDCK cells was subjected to size exclusion chromatography using Sephadex
G10 (bed volume: 10ml) with 100% DMF as the eluting solvent.
7. In vivo animal model-Brattleboro rats
The Brattleboro rat lacks the vasopressin hormone and therefore is a good
model to test the activity of antidiuretics. It displays symptoms of diabetes insipidus
as characterized by excessive thirst and urination (Trinh-Trang-Tan et al. 1982). A
group of 4 male Brattleboro rats; 6-7 weeks old were obtained from Harlan S/D
(Indianapolis, IN) and housed in 4 separate metabolic cages. They were allowed to
acclimatize for 48 hours prior to using them for studies. Body weight, water intake
and urine output was measured daily. Animal experiments were approved by the
IACUC at USC.
44
8. Biological activity of DDAVP-Cr7 and DDAVP
To determine if the conjugate could exert antidiuretic activity, 5µg/kg of
DDAVP or DDAVP-Cr7 (5 µg/kg DDAVP equivalent) was administered by sc
injection. Water intake and urine output was measured every 24 hours post injection
for 5 days.
9. Oral delivery of DDAVP-Cr7 and DDAVP
After a 3-days wash out period, the same rats were used to test the oral
activity of drug and conjugate. The rats were fasted overnight before administering
25 µg/kg of DDAVP or DDAVP equivalent conjugate orally using a gavage needle.
Both DDAVP and conjugate were dissolved in PBS pH 7.4. Volume of urine output
was recorded at time intervals, 0, 1, 2, 3, 4, 5, 6, 12, 24, 48 and 72 hrs. The rats were
allowed to feed after 6 hours of oral dosing.
10. Formulation effect
In an attempt to improve the activity of the conjugate, various formulations
were tested. Different solutions consisting of 10 µg/ml DDAVP-Cr7 in one of the
following were formulated. Bowman-Birk inhibitor (6mg), Na
2
CO
3
(150mg/ml),
Liposyn (10%), heparin (~100µg) or poly-L-glutamic acid (~ 100µg; MW 7700).
Combinations of heparin (~100 µg) + Na
2
CO
3
(150mg/ml) or PLGA (~ 100 µg;
MW 7700) + Na
2
CO
3
(150mg/ml) in presence of 10 µg/ml DDAVP-Cr7 were also
tested.
45
III. RESULTS
1. Bioconjugation of DDAVP with Cr7
Reduced DDAVP has two free thiol groups to which Cr7 can be linked. Therefore
the expected ratio of DDAVP to Cr7 is 1:2. To introduce a better leaving group to
cysteine moiety, Cr7 was activated with DTDP to form TP-Cr7, and purified as
described in the experimental section. Fig. 2.3 A, below shows the size-exclusion
profile of TP-Cr7. Peak 1 represents TP-Cr7 whereas peak 2 represents pyridine-2-
thione that is formed upon reduction by DTT and elutes after the void volume. %
yield calculated from equation 2.1 ranged from 65-75 %.
Activated Cr7 was reacted with reduced DDAVP at a molar ratio of 4:1. To
purify DDAVP-Cr7 from the excess reactants and by-products, the reaction mixture
was passed through sephadex G25. (Fig 1b.) The percentage yield of the conjugate as
determined by absorbance as 280 nm was calculated to be 80%.
46
TP-Cr7 (G15; 10ml; 0.05M NH4HCO3)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
12 3456 789 10 11 12 13 14 15 16 17 18 19 20
Fraction (ml)
Absorbance
abs280
abs343
DDAVP-2CR7 (G15; 20ml; 0.05M NH4HCO3)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
1 2 34 56 789 10 11 12 13 14 15 16 17 18 19 20 21 22
Fraction (1ml)
Absorbance
Abs280
A.
B.
Fig 2.3. A. Elution profile of TP-Cr7 in Sephadex G15 gel matrix. B.
Elution profile of DDAVP-Cr7 in Sephadex G15 gel matrix. Absorbance
was measured by UV spectroscopy using wavelength 280 nm and 343
nm for TP-Cr7 and wavelength 280 nm for DDAVP-Cr7.
47
2. Characterization of DDAVP-Cr7
a. Determination of molecular weight
Mass spectrometric analysis (figure 2b) showed a peak with the expected size
of the conjugate (3498 Da) indicating successful conjugation of DDAVP to CR7 at a
ratio of 1:2. Unexpectedly, a peak corresponding to the dimer (2426 Da) of
heptaarginine was also detected. This suggests that conditions utilized for synthesis
were not sufficiently optimal for conjugate formation with an equal likelihood of
dimer formation and also, purification by sephadex G-15 column was insufficient.
Nevertheless, it was not the goal of this project to achieve a pure conjugate. The
presence of dimer in the mixture was not expected to interfere with the activity of the
conjugate or exert any biological activity of its own.
b. Regeneration of DDAVP upon reduction of conjugate
Using thin layer chromatography, it was determined that DDAVP-Cr7 can be
reduced in vitro to regenerate DDAVP. As seen in the schematic Fig. 2.4A, lane 3, a
spot corresponding to RF of DDAVP (0.1) was seen when DDAVP-Cr7 was reacted
with DTT. However the conjugate did not migrate from the loading line due to the
presence of Cr7.
48
1 2 3 4
(A)
(B)
RF
3498.55
2429.53
1245.09
0.9
0.1
0
1 2 3 4 1 2 3 4
(A)
(B)
RF
3498.55
2429.53
1245.09
0.9
0.1
0
Fig. 2.4: Schematic of TLC analysis, (A) lane 1 represents desmopressin, lane 2: desmopressin conjugate;
lane 3: DTT reduced conjugate; lane 4: DTT; RF: Retention factor. (B) MS spectra of Conjugate, 3498.55
corresponds to mass of conjugate, 2429.53 corresponds to dimer of carrier, 1245.09 corresponds to
fragment of carrier or reduced DDAVP
49
3. Transport of conjugate in MDCK cells and RAECII
Total amount of radiolabelled conjugate transported after 2 hours of
incubation at 37º C was higher in both the RAECII (Fig. 2.5A) and MDCK cells
(Fig. 2.5B) as compared to the native drug. In RAECII, the total conjugate
transported was 2.7 fold higher at a concentration of 10µg/ml and 3 fold higher at 20
µg/ml. In MDCK cells, the total amount of conjugate transported was 2.2 times
higher than the transport of DDAVP at a concentration of 10µg/ml.
50
0.0 50.0 100.0 150.0 200.0 250.0
DDAVP
DDAVP-Cr7
ng /well
20ug/ml
10ug/ml
*
*
0.0 50.0 100.0 150.0 200.0 250.0
DDAVP
DDAVP-Cr7
ng /well
20ug/ml
10ug/ml
0.0 50.0 100.0 150.0 200.0 250.0
DDAVP
DDAVP-Cr7
ng /well
20ug/ml
10ug/ml
*
*
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
1 4 0
1 6 0
1 8 0
D D A V P D D A V P -C r7
ng /well
*
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
1 4 0
1 6 0
1 8 0
D D A V P D D A V P -C r7
ng /well
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
1 4 0
1 6 0
1 8 0
D D A V P D D A V P -C r7
ng /well
*
A.
B.
Fig. 2.5. A. Transport of DDAVP and DDAVP-Cr7 across RAECII. B. Transport
across MDCK cells. Confluent monolayers of RAECII or MDCK were equilibriated
with serum free medium for one hour and dosed with 10 µg/ml or 20 µg/ml of
125
I-
DDAVP or equivalent conjugate in RAECII and 10 µg/ml in MDCK cells. The
basolateral compartment was sampled after 2 hours and the radioactive count
determined. Data represents mean ± stdev; n=4; *p < 0.001
51
4. Retention of biological activity of bioconjugate
To determine if DDAVP can be regenerated from conjugate in vivo,
Brattleboro rats were injected sc with DDAVP-Cr7. The volume of urine output was
used as a measure to determine the activity of DDAVP. As shown in Fig. 2.6,
DDAVP-Cr7 was able to elicit similar response in activity as DDAVP suggesting
that after s.c injection, the disulfide bonds between DDAVP and Cr7 are reduced
systemically to release Cr7 and regenerate DDAVP in a structurally active form.
This corroborates with the findings of in vitro regeneration of DDAVP from
DDAVP-Cr7 when reduced with DTT.
0
50
100
150
200
250
300
350
400
01 23 4 5 6
Days
ml
DDAVP (SC: 5ug/Kg)
DDAVP-Cr7 (SC: 5ug/Kg)
Fig 2.6. Subcutaneous activities of DDAVP-Cr7 and DDAVP. Fasted
Brattleboro rats were injected SC with DDAVP or DDAVP-Cr7 at a
concentration of 5 µg/kg. Volume of urine output was measured over a
period of 4 days. Baseline volume (day 0) is the urine output 24 hours
prior to dosing. Values represent mean ± stdev; n=4
52
5. Oral activity and in vitro-in vivo correlation
A five-fold higher dose of DDAVP was administered orally compared to s.c
to determine the oral effect of conjugate or native DDAVP. Urine output was
measured at 12, 24, 48 and 72 hours. As shown in Fig 2.7, the volume of urine at
time points measured was marginally lower for conjugate compared to native drug.
However the difference between the activity of DDAVP-Cr7 and DDAVP was not
statistically significant
0
50
100
150
200
250
300
0123 4567
Days
ml
DDAVP-Cr7 (Oral:25ug/kg)
DDAVP (Oral:25ug/kg)
Fig 2.7. Oral activities of DDAVP-Cr7 and DDAVP. Fasted
Brattleboro rats were orally administered with DDAVP or DDAVP-
Cr7 at a concentration of 25 µg/kg. Volume of urine output was
measured over a period of 5 days. Baseline volume (day 0) is the urine
output 24 hours prior to dosing. Values represent mean ± stdev; n=4
53
6. Effect of formulation
We expected to observe a better effect of DDAVP-Cr7 compared to DDAVP,
owing to the cell penetrating peptide’s ability to enhance penetration across the
epithelial barrier. However, since no statistically significant difference was observed
between the two, it was speculated that the conjugate does undergoes degradation in
the gut prior to its availability to the epithelial cells. Therefore we decided to test the
effect of various formulations that would protect the conjugate from metabolism.
Na
2
CO
3
was used to neutralize the acidic pH of stomach, heparin and poly L-
glutamic acid (PLGA) were used as complexation agents that would physically
complex with the positively charged conjugate, thus shielding the conjugate from
degradation enzymes. Bowman-Birk inhibitor was used as a non-specific protease
inhibitor and Liposyn was used as a solubilizer. Although the results shown in Fig.
2.8A. are preliminary based on n=1, it was evident that a single formulation factor
was not sufficient to mitigate the degradation activity in the gut. To determine if a
combination of factors would protect the conjugate, we formulated DDAVP-Cr with
both pH neutralizer and complexation agent (Heparin + Na
2
CO
3
or PLGA + Na
2
CO
3
)
Fig 2.8B shows that even a combined effect of complexation agent and acid
neutralizer did not significantly alter the activity of DDAVP-Cr7
54
0
50
100
150
200
250
300
350
0 10 20 30 40 50 60 70 80
Time (hrs)
Urine output (ml).
Na2CO3 (150mg/ml)
Liposyn (10%)
BBI (6mg)
PLGA (~100ug)
Heparin (~100ug)
DDAVP-Cr7
0
50
100
150
200
250
300
350
400
450
500
0 10 20 30 40 50 60
Time (hrs)
Unine output (ml)
DP-Cr7 + heparin + Na2co3
DP-Cr7 + PLGA+ Na2Co3
DP-Cr7
Fig 2.8. A. Effect of various formulations on the oral activity of DDAVP-Cr7. 10
µg/kg of DDAVP-Cr7 in formulation with either of the following: sodium
carbonate, heparin, PLGA, Liposyn, or BBI was administered orally to Brattleboro
rats. Values represent absolute urine output. n=1 B. Effect of combining 2
excipients on the activity of DDAVP-Cr7. Values represents mean ± difference of
mean. n=2
55
7. Analysis of in vitro transport components
Using size-exclusion chromatography, we analyzed the basolateral fluid of
the transport experiment in MDCK cells and RAECII to determine the intact versus
degradation products of DDAVP or DDAVP-Cr7. Fig 2.9 A below shows the profile
of both DDAVP-Cr7 and DDAVP eluent of transport in MDCK cells when passed
through a 20ml column of sephadex G-15. Qualititative analysis of the peak shows
that an equal amount of intact drug is transported in both cases however the
conjugate undergoes extensive degradation as evidenced by 2 major peaks beyond
the void volume. Fig 2.9B, shows the profile of eluent from transport in RAECII. In
these cell model also, an equal amount of intact drug was recovered however the
profiles of the degradation peaks were different from MDCK cells suggesting
involvement of different proteases or different pathways of transport.
56
Fig. 2.9. A. Elution profile of DDAVP and DDAVP-Cr7 transported
across RAECII cells. B. MDCK cells. Media collected from basolateral
compartment of transport study was passed through a size exclusion
column consisting of sephadex G10 gel matrix (bed volume 20ml) using
100% DMF as eluting solvent
[B] MDCK Cells
-200
0
200
400
600
800
1000
1200
16 11 16 21 26 31 36 41 46
Fraction (1ml)
cpm.
DDAVP-Cr7 (basal)
DDAVP (basal)
[A] RAECII
-100
0
100
200
300
400
500
600
15 9 13 17 21 25 29 33 37 41 45 49
Fraction (1ml)
cpm.
DDAVP-Cr7 (basal)
DDAVP (basal)
57
IV. DISCUSSION
In this chapter, we demonstrate the application of oligoarginine as a delivery
vector for a peptide drug desmopressin (DDAVP). We show that DDAVP can be
modified with 2 moieties of Cr7 using the two inherent thiols in its peptide sequence.
The rationale for using disulfide chemistry for conjugation was to ensure that
DDAVP is released from the vector to exert its biological activity. It has been
previously shown that the macrocyclic ring formed due to disulfide linkage is
necessary for the activity of desmopressin and related analogues (Langs et al. 1986).
Our results show that we can regenerate DDAVP from DDAVP-Cr7 in vitro using
reducing conditions as seen by TLC analysis. Upon subcutaneous injection of
DDAVP-Cr7 in Brattleboro rats, the activity of DDAVP is retained, as seen by a
decrease in urine output similar to that of DDAVP, suggesting that Cr7 is cleaved
from the conjugate to regenerate DDAVP with its macrocyclic ring structure. Similar
result was observed when DDAVP was conjugated to different lengths of fatty acid
using disulfide chemistry (Wang et al. 1993; Wang et al. 1999).
In vitro transport comparison in two different epithelial cell models, RAECII
and MDCK, indicated that the total transport of DDAVP-Cr7 was more than 2-fold
higher at the concentrations tested. Qualitative analysis of the basolateral media after
transport experiment suggested a high degree of metabolism for both DDAVP and
DAVP-Cr7 with different elution profiles of the metabolites. This can possibly be
explained by the increased susceptibility of DDAVP-Cr7 to proteases due to opening
of the ring structure. In alveolar epithelial cells, amino peptidases that localize in the
58
apical membrane are shown to metabolise vasopressin with cleavage occurring
between Tyr and Phe thus generating Cys-Tyr dipeptides (Yamahara et al. 1994).
Thus, it is likely that the Tyr-Phe site is more accessible to aminopeptidase in
DDAVP-Cr7 than DDAVP partly, contributing to the high amounts of metabolites.
On the other hand, other cytosolic or basolateral peptidases may also be involved in
degrading the conjugate. In the case of vasopressin, camostat mesylate, an
aminopeptidase inhibitor, was shown to decrease its cellular processing but not
completely abolish it implicating the involvement of other proteases (Yamahara et al.
1994). Since our aim was to determine the beneficial effect of Cr7 on permeability,
we did not further explore the in vitro degradation pathways of DDAVP-Cr7.
In parallel animal studies, we investigated whether the increase in transport in
vitro correlated with higher pharmacological activity in vivo when administered
orally. Our results indicate that there was a marginal improvement in activity for
DDAVP-Cr7 compared to DDAVP but the difference was not statistically
significant. It was tempting to speculate that the marginal increase in activity seen
could be due to the absorption enhancing effect of Cr7 and this effect could be
magnified if other factors that influence stability and integrity of the conjugate in the
gut could be controlled.
The gut is a complex system with several protective mechanisms that act in
concert to prevent systemic absorption of foreign substances. In an attempt to breach
these protective mechanisms and increase the dose available for absorption, we
tested a panel of formulations of our conjugate. To neutralize the acidic environment
59
of the stomach and hence prevent hydrolysis of the peptide conjugate, we
administered the conjugate in presence of sodium carbonate. -chymotrypsin has
been shown to catalyze the cleavage of orally administered DDAVP (Fredholt et al.
1999). To prevent the activity of this enzyme, we used Bowman-Birk protease
inhibitor derived from soybean which inhibits both chymotrypsin and trypsin activity
(Birk 1985). We also modified our conjugate by mixing it with negatively charged
heparin or PLGA. Mixing of conjugate to heparin or PLGA is expected to induce a
tight complex formation based on ionic interaction with the positively charged
oligoargininie that would protect it from enzymatic attack due to steric hindrance but
not interfere with its absorption. This complex formation has been shown to occur
with heparin and poly-lysine in solution (Gelman and Blackwell 1973). Also, due to
the additional hydrophilicity imparted to DDAVP by conjugating it to Cr7, its
increased solubility is likely to increase its susceptibility to lumenal degradation
enzymes and also increase repulsion from the lipophilic epithelial cell membrane.
For this reason, DDAVP was formulated in Liposyn a fatty acid based emulsion.
However, none of the above strategies seemed to increase the activity of DDAVP-
Cr7.
We therefore concluded that the modification of DDAVP with Cr7 does not
provide any practical advantage over DDAVP. Although higher permeability of the
conjugate was seen in vitro, the simultaneous enzymatic barrier, poses an equal
challenge in increasing its overall bioavailability. Nevertheless, we are cautious in
generalizing it to other cell penetrating peptides and other peptide drugs. It is likely
60
that other CPPs or peptide drugs may behave different from our model. Furthermore
optimization in terms of CPP selection or drug: CPP ratio may be required to see an
effect in vivo. Conjugation using disulfide linkage provides the advantage of
regenerating the drug, but it necessitates the attachment of 2 moieties for each
disulfide bond disrupted to stabilize the molecule. In the case of DDAVP-Cr7, two
molecules of Cr7 per DDAVP might not be the optimum ratio for stability or cell
membrane penetration. The distance between two thiol groups is only 4 amino acids
in DDAVP. This spacing may not be sufficient for optimal stability, when a highly
charged oligoarginine is introduced due to the repulsion from positively charged
arginines on each peptide chain. Also, as discussed above, opening the ring structure
may make it more susceptible to enzymatic degradation resulting in competing
pathways of enzymatic digestion or permeation across plasma membrane for the
conjugate.
In summary, we show that DDAVP can be derivatized with two moieties of
Cr7 by disulfide bridging. DDAVP can be regenerated in vitro as well as in vivo
from DDAVP-Cr7 in a biologically active form. Total transport of DDAVP-Cr7 is
higher than DDAVP in both MDCK and RAECII, however there is extensive
metabolism during the transport process. Oral activity of DDAVP-Cr7 is only
marginally better than DDAVP in Brattleboro rats. This activity could not be
increased by formulation manipulation. Therefore, oligoarginine may not be a
suitable delivery system for DDAVP or peptide drugs intended for oral delivery.
61
V. SUMMARY
Biosynthesis
Biological activity
Formulation effect
In vitro transport
Oral activity
1:2 ratio of DDAVP-Cr7
Retained, similar to DDAVP
3-fold higher in MDCK cells
No improvement over DDAVP
No effect on bioactivity
DDAVP
&
Oligoarginine
Biosynthesis
Biological activity
Formulation effect
In vitro transport
Oral activity
1:2 ratio of DDAVP-Cr7
Retained, similar to DDAVP
3-fold higher in MDCK cells
No improvement over DDAVP
No effect on bioactivity
DDAVP
&
Oligoarginine
Scheme 3: Summary of findings in chapter 2
62
CHAPTER 3
Cationic CPP Mediated Protein Delivery: Insulin-Oligoarginine for
Pulmonary Delivery
I. INTRODUCTION
1. Background information on insulin
Insulin is a molecule with a rich history, which was first discovered by
Banting and Best in the 1920s (Banting and Best 1990). It is the first protein to be
fully sequenced, isolated in a pure crystalline form, chemically synthesized and also
the first protein obtained by recombinant DNA technology. Its structure is conserved
across species with minor differences in bovine (three amino acids) and porcine
insulin (one amino acid). For this reason, initial source of insulin for clinical use
were derived from cow, pig, horse or fish pancreas, which were effective in
treatment but had problems of purity and immunogenicity. In 1982 Eli lily
introduced human insulin (Humulin) obtained through genetic engineering, with
lesser purity and immunogenicity issues (Williams et al. 1982).
Structurally, human insulin consists of 51 amino acids, with 21 residues in
chain A and 30 residues in chain B with a molecular mass of 5808 kDa. It has two
inter-chain disulfide bonds that link chain A and chain B and one intrachain disulfide
bond linking cysteines at position 6 and 11 of chain A. At concentrations above
0.1mM and in the presence of Zinc, three insulin dimers assemble to form a hexamer
of insulin monomers (Blundell et al. 1972).
63
Insulin is used primarily to treat insulin-dependent diabetes mellitus (type I
diabetes) and in some patients with insulin resistant diabetes mellitus (type II
diabetes). It is a vital endocrine hormone that regulates blood glucose levels by
binding to cell surface receptors and triggering cellular events that lead to glucose
absorption from the blood. The major clinical problems identified with insulin
therapy are the mode of administration and the dosage regimen. Insulin is restricted
to parenteral route (subcutaneous) of administration due to its physicochemical
properties whereas the dosage regimen is problematic because of the non-
physiological time-action profile of slow onset and long duration seen with regular
insulin. Both clinical problems have extensively been investigated. Alternate routes
of delivery including oral, pulmonary, transdermal, buccal, ocular, and nasal have
been proposed (Lassmann-Vague and Raccah 2006). Whereas, modifications in the
dosage regimen that lead to better glycemic control are approached by altering
insulin molecule to generate insulin analogs such as Lispro (molecular switching of
B28Pro and B29Lys) and Aspart (changing B28Asp to B28Pro), or conjugating
insulin to specific moieties to generate insulin derivatives, such as pegylated insulin
or palmitoyl insulin (Bhatnagar et al. 2006).
2. Non-invasive modes of insulin delivery
a. Oral delivery
From a patient perspective, oral delivery is the most convenient and desired
route of administration especially when repeated administration is necessary. For
64
insulin specifically, oral route is rationally the best alternative to endogenous insulin
secretion because of the immediate access to the ultimate target, liver. Due to its
direct delivery into portal circulation, oral insulin regulates hepatic glucose output
directly and within minutes because of the high insulin sensitivity of the liver,
potentially resulting in replication of the early-phase secretion seen in normal
glucose homeostasis, and also an added benefit of reducing the risk of hypoglycemia
by virtue of not targeting peripheral glucose in skeletal muscle (Arbit 2004).
As much as insulin oral delivery is desired, it is still considered the ‘holy
grail’ with numerous attempts aimed at increasing its permeability across the
intestinal epithelium and reducing its susceptibility to enzymatic degradation with
modest success. Although a marketed product for oral insulin is not available, the
current state of research spans from proof-of concept studies to late stage clinical
trial.
b. Pulmonary delivery
Another non-invasive administration route that is popular for insulin delivery
is the pulmonary route as evidenced by the recent approval of Exubera, an inhaled
version of recombinant human insulin, in USA and Europe (Lenzer 2006). The
physiological attributes of the lung make it a suitable target for systemic delivery of
therapeutic molecules. The large absorptive surface area (~ 100m
2
), extensive
vascularization, and low enzymatic activity contribute to increased bioavailabilty of
insulin delivered through the pulmonary route. However, relative to systemic
delivery, the bioavailability is poor, partly due to the absorption barrier posed by the
65
epithelium lining the lungs. Particularly, the lower respiratory tract that provides the
large surface area for absorption consists of a tight alveolar epithelium that prevents
permeation of macromolecules. Competing pathways such as phagocytic uptake and
proteolytic degradation further limit the amount of intact insulin available for
absorption. Strategies employed to overcome these barriers include the use of
penetration enhancers (Kumar and Misra 2003) or incorporation of the
macromolecules into carrier systems such as nanospheres and liposomes (Liu et al.
1993; Garcia-Contreras et al. 2003).
3. Chemical modifications and conjugation of insulin
Insulin has been conjugated to numerous moieties for the purposes of
improving physicochemical properties, enhancing permeability across cell
membrane, increasing enzymatic stability, prolonging systemic availability and also
for bioanalytical purposes.
Modification of insulin with mPEG did not alter its structural properties or
potency, but increased insulin resistance to aggregation, eliminated antigenicity and
also prolonged the systemic circulation of insulin (Uchio et al. 1999). Similarly,
modification with non-esterified fatty acids, leads to prolonged activity due to their
binding affinity for albumin in vivo (Kurtzhals et al. 1996). Insulin has also been
modified with human serum albumin for a long lasting derivative and also a fusion
protein of insulin and albumin has been engineered (Duttaroy et al. 2005;
Thibaudeau et al. 2005). To increase permeability across epithelia, insulin has been
66
conjugated to transferrinn, deoxycholic acid, and tat peptide (Shah and Shen 1996;
Lee et al. 2005; Liang and Yang 2005). For bioanalytical, radioimmunoassays and
receptor binding studies, insulin has been modified with fluorophores such as FITC,
radioactive elements such as iodine and chromophores (Bredehorst et al. 1991; Hentz
et al. 1997). In addition, insulin derivatives of polysialic acid (Zhang et al. 2006),
silk sericin (Jain et al. 2003) saporin (Ippoliti et al. 1996), biotin (Fabry and
Brandenburg 1992), and ricin (Roth et al. 1981) have been synthesized and
characterized
Site-specific conjugations or modifications in insulin are generally obtained
through the three primary amines (A1, B1 and B29), tyrosines or cysteine reactive
groups (Fig. 3.1). Due to different pKas of the three amino groups and different
reactivity rates B1 > A1 > > B29, it is possible to make modification at a specific site
on insulin. Extensive structure activity studies have identified the receptor-binding
site of insulin and therefore to retain biological activity, conjugating at position B1
and B29 are preferred over position A1. A common strategy to gain site-specific
conjugation at either amino position is to reversibly block the non-target amino
groups by reacting with reagents that reversibly react with the amino group.
67
Several reagents are available to block protein amino groups that produce
derivatives, which vary widely in the conditions of their stability and their properties.
Maleic anhydride was commonly used because it could form an amide bond with one
of the carboxyl group whereas the protonated form of the second carboxyl group
could catalyze the hydrolysis of the amide bond with a half-life of 11 hours at pH 3.5
and 37 ºC (Butler et al. 1967). Such a long half-life for regenerating amino group is
not convenient and therefore derivatives of maleic anhydride, 2,3 dimethyl maleic
B29 B29
Fig. 3.1. Schematic illustration of insulin 3-D structure. Shaded region
indicates receptor-binding site of insulin. Red arrows indicate the three
amino groups present on insulin. Green arrows indicate disulfide bonds. A14,
A19, B16 and B26 represent tyrosine residues.
68
anhydride (DMMA) or 2-methyl maleic anhydride (citraconic anhydride, Cit A)
were introduced (Dixon and Perham 1968). With two methyl groups present,
DMMA has the disadvantage of rapid hydrolysis and therefore does not protect the
amino groups long enough for further modification, whereas Cit A has one methyl
group with an intermediate rate of hydrolysis but a higher probability of forming two
products depending on which carbonyl group of the reagent is attacked.
In this chapter, we conjugated insulin with various cationic CPPs by site-
specific conjugation at Lys B29 using a heterobifunctional cross-linker, SPDP and
Cit A as amino protecting reagent. The purpose of the conjugation was to determine
if CPP could enhance the transepithelial transport of insulin across RAECII. The
transport efficiency of oligoarginine, oligolysine and tat peptides, represented by cr9,
ck9 and ctat respectively, was determined. Oligoarginine, showed the highest
transport efficiency and was chosen for further mechanistic studies and also tested in
vivo in diabetic rats.
II. EXPERIMENTAL
1. Synthesis and purification of insulin-CPP conjugates
The modification method of insulin for specific conjugation at position B29
was adapted from (Naithani and Gattner 1982) with several changes in the
procedure. Recombinant human insulin (20mg/ml) was reacted with 15 μl citraconic
anhydride (100mg/ml) at pH 8 for 1.5 hrs at which point most of the insulin was
converted to A1, B1, B29-tricitraconylinsulin (step 1). The pH of reaction mixture
69
was adjusted to pH 5 and incubated at room temperature for 4.5 hours to partially
deblock insulin at position B29. The resulting A1,B1-dicitraconyl insulin (InP4) was
purified from the mixture of several species of modified insulin by reversed phase
HPLC using a C4 column with a mobile phase consisting of solvent A-5% MeCN,
20mM NH
4
HCO
3
and solvent B-95% MeCN, 20mM NH
4
HCO
3
eluted at a gradient
of 17.5-25% solvent B in 15min at a flow rate of 1ml/min (step 2). This intermediate
(InP4) was lyophilized and stored at –20 ºC or used immediately . To conjugate the
cysteine moiety of CPP to amino group on B29 of InP4, a heterobifunctional linker
SPDP, that is reactive towards amine and thiol groups was utilized. InP4 was
dissolved in phosphate buffer pH 8, was first reacted with aliquots of 5 µl SPDP
(50mg/ml DMF) every 20minutes for 2 hrs. This reaction was monitored by
intermittent HPLC analysis of reaction mixture. When more than 90% of Inp4 was
modified with SPDP, InP4-SPDP was isolated from the unreacted InP4 and excess
SPDP by reversed phase HPLC using a C18 column and mobile phase consisting
of solvent A-5% MeCN, 0.1% TFA; solvent B-95% MeCN, 0.1% TFA eluted at a
linear gradient of 15-50% solvent B in 15min at a flow rate of 1ml/min (step 3).
InP4-SPDP intermediate collected from HPLC was concentrated, adjusted to pH 8
using 1N NaOH and subsequently reacted with equimolar amounts of cr9, ck
9
or Tat-
D peptide for 1hr to obtain In-CPP conjugate (step 4).
70
Fig. 3.2. Schematic representation of insulin-CPP conjugation
STEP 1
Lys
B29
Ala
B
30
Asn
A21
NH
2
NH
2
NH
2
A-chain
B chain
Lys
B29
Ala
B
30
Asn
A21
NH
2
NH
2
NH
2
A-chain
B chain
Citraconic Anhydride
+
Insulin
Lys
B29
Ala
B3
0
Asn
A21
NH
NH
NH
A-chain
B chain
CitA
CitA
Lys
B29
Ala
B3
0
Asn
A21
NH
NH
NH
A-chain
B chain
CitA
CitA
O
O
O
C H
3
O
O
O
C H
3
Intermediate 1
pH 8
2h, RT
CitA
Lys
B29
Ala
B3
0
Asn
A21
NH
NH
NH
A-chain
B chain
CitA
CitA
Lys
B29
Ala
B3
0
Asn
A21
NH
NH
NH
A-chain
B chain
CitA
CitA
Intermediate 1
CitA
Lys
B29
Ala
B3
0
Asn
A21
NH
NH
NH
2
A-chain
B chain
CitA
CitA
Lys
B29
Ala
B3
0
Asn
A21
NH
NH
NH
2
A-chain
B chain
CitA
CitA
Intermediate 2
HPLC Purification
C4, 15-30% B 15 min
A: 10% MeCN,20mM NH
4
HCO
3
B: 95% MeCN,20mM NH
4
HCO
3
4.5h, RT
pH 5
STEP 2
71
Fig. 3.2. Schematic representation of insulin-CPP conjugation (continued)
STEP 3
Lys
B29
Ala
B
30
Asn
A21
NH
NH
NH
2
A-chain
B chain
CitA
CitA
Lys
B29
Ala
B
30
Asn
A21
NH
NH
NH
2
A-chain
B chain
CitA
CitA
Intermediate 2
N
O
O
O S
O
S
N
Lys
B29
Ala
B30
Asn
A21
NH
2
NH
2
N
H
A-chain
B chain
S
O
S
N
Lys
B29
Ala
B30
Asn
A21
NH
2
NH
2
N
H
A-chain
B chain
S
O
S
N
pH 8, N
2
1.5 Hr, RT
Intermediate 3
SPDP
HPLC Purification
C18, 10-100% B 15 min
A: 10% MeCN,0.1% TFA
B: 95% MeCN,0.1% TFA
+
72
Fig. 3.2. Schematic representation of insulin-CPP conjugation (continued)
STEP 4
Lys
B29
Ala
B30
Asn
A2
1
NH
2
NH
2
N
H
A-chain
B chain
S
O
S
N
Lys
B29
Ala
B30
Asn
A2
1
NH
2
NH
2
N
H
A-chain
B chain
S
O
S
N
NH
2
O
S H
[X]
n
Lys
B29
Ala
B
30
Asn
A21
NH
2
NH
2
N
H
A-chain
B chain
S
O
NH
2
O
S
[X]
n
+
Intermediate 3
C-CPP
HPLC Purification
C18, 17.5-25% B 15 min
A: 10% MeCN,0.1% TFA
B: 95% MeCN,0.1% TFA
pH 9
0.5Hr RT
Bioconjugate (In-CPP)
+
73
2. Characterization and radiolabeling of conjugates
a. SDS-PAGE analysis of In-CPP conjugate
SDS-PAGE was performed according to the method described by Laemmli
(Laemmli 1970). 6.5% stacking gel and 15 % separating gel were used under non-
denaturing conditions. Protein marker, insulin, and insulin conjugates were mixed
gently with 5x loading buffer devoid of any reducing agent and loaded into the wells
of the gel. Electrophoresis was run for approximately 2 hrs, starting at a constant
current of 15 mA for stacking gel and increased to 20 mA when samples entered the
separating gel. The polyacrylamide gel was stained with Coomassie blue for
detection.
b. Anion-exchange chromatography
Sephacel-DEAE matrix (2ml) was packed in a syringe plugged with fiber
glass and washed with 20mM Tris-HCl buffer. After loading insulin or insulin-cr9
conjugate to the matrix, 5ml of three different concentrations (0.01, 0.15 and 0.5M)
of NaCl in 20mM Tris-HCl buffer (pH 7.4) were used sequentially to elute insulin
and insulin-cr9. 1ml fractions were collected and the absorbance at 280nm
wavelength were analyzed by UV spectrometer.
c. MALDI-TOF mass spectroscopy
Molecular weight of insulin-CPP conjugates was obtained using Krotos
Kompact MALDI-TOF Mass Spectrometer (USC Proteomics Core). Approximately
10 pmol of the conjugate was spotted on a matrix consisting of 10mg/ml-cyano-4-
74
hydroxycinnamic acid in 1% TFA /70% acetonitrile and subjected to MALDI-TOF
using an Axima-CFR in the linear mode using an accelerating voltage of 25 kV.
d. Radioiodination of insulin and insulin-CPP conjugates
Insulin and insulin-CPP conjugates were radiolabelled with
125
I using
chloramine-T method (described in chapter 2) and purified using size-exclusion
chromatography using G-25 gel matrix. Since low quantities of conjugate in the µg
range were radiolabelled, elution was performed using PBS supplemented with 0.1%
BSA to reduce non-specific adsorption of conjugates to the gel matrix. During the
chloramine-T reaction, precipitation on the walls of the test tube was observed. The
quantity of protein lost due to precipitation and aggregation could not be determined
due to the µg level of quantity. Therefore specific activity for the conjugates could
not be determined and all experiments were performed by dosing equal cpm for
bioconjugates and insulin.
3. Alveolar epithelial and Caco-2 Cell culture
Primary rat alveolar type I and type II cells were cultured as described in
chapter 1. Confluent human colon carcinoma cells (Caco-2 cells) were detached
from 25 cm2 stock culture flasks using 0.5ml trypsin/EDTA incubated at 37 ºC for
15 min. The detached cells were resuspended in 5ml of DMEM supplemented with
10% fetal bovine serum (FBS), 1% L-glutamine and 1% essential amino acids. Cell
density was determined using a coulter counter. Suspended Caco-2 cells were seeded
in 12-well transwell plates at a density of 500,000 cells/well. 1.5 ml of supplemented
75
DMEM was added to the basolateral chambers of the transwells. Cells were allowed
to attach to inserts for 3 days prior to changing medium. Cells were allowed to grow
for 14 days postseeding by which time monolayers had developed TEER of
approximately 400.cm
2
. Medium was replaced every 2-3 days and one day prior to
performing transport experiment.
4. Transport of insulin bioconjugates
RAECII monolayers were pre-equilibriated with serum free medium on day 5
of culture for one hour. Monolayers were treated with ~ 0.6 * 10
5
cpm/ monolayer
of either insulin, In-cr9, In-ctat or In-ck9. After 2 hours of incubation at 37 ºC, media
in basolateral compartment was analyzed for radioactivity and subjected to
trichloroacetic acid (TCA) precipitate assay to determine the amount of intact
protein. Transport is indicated as % dose transported per cell monolayer. Statistical
analysis was performed by ANOVA with post hoc t-test.
For time dependent transport of In-cr9 or insulin, different monolayers were
used for each time point, with a total of n=3 for each time point. They were subjected
to the same treatment described above.
5. TCA precipitate assay
For all in vitro transport experiments, amount of intact protein transported in the
basolateral compartment was determined by the TCA precipitate assay. Briefly, after
determining the radioactivity in the basolateral media, samples were incubated with
76
15% TCA at 4 ºC for 15 minutes and then centrifuged for 15 min at 1500rpm to
precipitate intact protein. The supernatant was aspirated and the radioactivity in the
precipitate was considered as intact protein (native insulin or in conjugate form)
transported across the monolayers.
6. Transport of In-cr9 vs In + cr9
a. Comparison of physical mixture vs covalent linkage
Cells were incubated either with In-cr9 or insulin premixed with an equimolar
amount of cr9. Transport of intact protein was determined after 2 hours of incubation
b. Uptake and transport of insulin: Pep1 mixture
A mixture of radiolabeled insulin and Pep1 in the ratio of 1:4 was prepared and
dosed in RAECII. Uptake and transport were determined at 4 hrs and 24 hrs after
incubation at 37 ºC. For uptake studies, monolayers were washed thrice with ice-cold
PBS and the membrane cut out from the inserts. Radioactivity accumulated in the
membrane filter was determined. Amount of intact protein was determined by TCA
precipitate assay
7. Effect of In-cr9 on TEER of RAECII
The effect of In-cr9 or physical mixture of insulin and cr9 were investigated
on the integrity of the monolayers by monitoring TEER across RAECII monolayers.
Insulin was used as a control. An EVOM meter, connected to a pair of chopstick
electrodes was used to measure TEER values at 0, 60, 120, 180 and 240 min after
77
dosing. Monolayers were maintained at 37 ºC for the entire time period. TEER
values at 0 min was considered as 100% and the percentage change in TEER at the
remaining time points were calculated. TEER values ranged from 3.6 –2.5 k.cm
2
before and after the experiment.
8. Transport in RAECI and Caco-2 cells
RAECI grown in the absence of KGF as described in chapter 1 were used on
day 6 of seeding whereas Caco-2 cells were used on day 14 of seeding for transport
of In-cr9. Protocol described above for RAECII was used.
9. Effect of temperature on transport of In-cr9
RAECII monolayers were pre-equilibriated with serum free medium on day 5
of culture for one hour. Monolayers were then incubated at 37 ºC, 16 ºC or 4 ºC for
30 minutes, and were then dosed with 0.18 * 10
5
cpm/monolayer of In-cr9 that had
had been incubated at 37 ºC, 16 ºC or 4 ºC respectively, for 30 min prior to dosing.
The cells were incubated for an additional 2 hours at the respective temperature
before the basolateral media was collected and measured for radioactivity. TCA
precipitate assay was performed to determine intact protein.
10. Effect of biochemical modulators on transport of In-cr9
To study the effect of various modulators, monolayers were pre-incubated in
the apical compartment with one of the following. 10 mM or 50 mM of ammonium
78
chloride, 30 µM monensin, 33 µM nocodazole dissolved in DMF, 30 µM
chlorpromazine, 200 µM of EIPA, 0.1 µg/ml cytochalsin D, or 10 mM + 6 mM of
sodium azide and deoxyglucose for 1 hr after pre-equilibriating for 30 min with
serum free medium. After aspirating the media containing the biochemicals, dosing
solution consisting of radiolabeled In-cr9 combined with the same concentration of
the biochemical was incubated for an additional 2 hrs. Amount of intact protein
transported was determined as described above.
11. Effect of heparin and protamine on uptake and transport of In-cr9
Monolayers were pre-treated with 1 μM of heparin or protamine for 1 hr,
after which they were incubated with dosing media containing In-cr9 and heparin or
In-cr9 and protamine. After 2 hrs of incubation, monolayers were washed with ice-
cold PBS and membrane filters cut out from insert to determine the amount of
uptake. Amount of intact protein transported was determined as described above.
12. Receptor binding assay
a. HepG-2 cell culture
HepG2 human hepatoma cell line obtained from ATCC, (Rockville, MD) were
cultured in T-25 flasks with DMEM supplemented with 10 % fetal bovine serum,
2mM glutamine, 50 µg/ml streptomycin and 50 U/ml penicillin. Cells became
confluent in T-25 flasks after 4 days and were passed twice before using them for
binding studies.
79
b. Insulin concentration dependent binding
HepG2 cells were seeded on 24-well cluster plates at a density of 4 x 10
5
cells/well. Cells were used after 3 days of seeding when confluence was reached.
Monolayers were washed twice with PBS before equilibrating with DMEM
containing 0.1% BSA.
125
I-Tyr14 insulin dosing solution of various concentrations
was prepared in media consisting of 0.1% BSA. The cells were incubated with
dosing solution for 2 hours at 4 ºC after which the unbound insulin was removed by
washings with PBS. Monolayers were dissolved in 0.1% NaOH, counted for
radioactivity and the protein concentration determined by bicinchoninic acid assay
(BCA) (Pierce, Rockford IL). The data were plotted using GraphPad Prism and Kd
value determined by curve fitting.
c. Competition assay
The method for competition assay is similar to the concentration dependent
binding assay described above, except that one concentration of
125
I-Tyr14 insulin
(50 pM) and different concentrations (1 pM – 1 µM) of unlabeled insulin or In-cr9
were used to compete with the radiolabelled insulin.
13. In vivo activity of In-cr9
a. Induction of diabetes in rats
The animal studies described in this section was approved by the Institutional
Animal Care and Use Committee of University of Southern California, and the
80
procedures were conducted according to the guide for the Care and Use of
Laboratory Animals (National Research Council 1996). Sprague-Dawley male rats
aged 10 weeks, weighing 250-280 g were housed under standard laboratory
conditions of 65 ± 2% relative humidity, 23 ± 2 ºC temperature and 12 hr light dark
cycle. Rats were fed with standard rodent pellet diet and tap water adlibitum when
not under experiment. They were allowed to acclimatize for 3 days prior to starting
experiment. Rats were fasted overnight and baseline glucose determined before
intraperitoneal injection of freshly prepared streptozotocin (STZ, 60 mg/ml in pH 4.5
acetate buffer) to induce diabetes. Five days after STZ-treatment, fasted rats with a
plasma glucose level of > 400 mg/dl were selected for further studies.
b. Preparation of dosing solutions.
To obtain a pure In-cr9 conjugate required HPLC purification using organic
solvents. Exchanging the organic solvent with aqueous buffer led to extensive
protein loss through aggregation. Furthermore, the yield was not sufficient for in
vivo studies. Therefore, purification process mentioned in step 3 and 4 were not
performed when obtaining conjugate for animal experiments and a mixture of
conjugate was used. For control studies, 5mg of insulin was dissolved in 1ml PBS
(pH 7.4). The concentrations of insulin or In-cr9 were adjusted such that only100 µl
was used for spray instillation. The dose used was 0.375mg/kg, (~ 10.5 U/kg) of
insulin or insulin equivalent.
81
c. Spray instillation of In-cr9, insulin and control PBS
The diabetic rats were fasted overnight and divided into 3 groups with 4 rats per
group for control, insulin, or In-cr9. The animals were anesthetized by an
intraperitoneal injection of a mixture of xylazine (10mg/kg) and ketamine
(90mg/kg). 100 µl of either dosing solution was administered 5 minutes after
anesthesia using 1A-1B MicroSprayer
TM
(Penn Century Inc. Philadelphia, PA)
attached to a syringe. During administration, the animal was held in a lateral position
on a restrainer and the tongue pulled aside using blunt forceps to expose the larynx.
The microsprayer was then guided to the glottis region by illumination with a fiber
optic otoscope and gently pushed further into the trachea where the contents were
sprayed. The animal was held in a vertical position for 1 minute after administration
and then placed in a blanket and allowed to recover. A drop of blood from the tail
vein was withdrawn at predetermined time points to measure glucose using the ONE
TOUCH
®
ULTRA
TM
blood glucose monitoring system (Lifescan, Inc, Milpitas, CA).
The results were expressed as percentage of initial glucose levels.
III. RESULTS
1. Biosynthesis of In-cr9, In-ctat and In-ck9
Conjugation of cr9, tat or ck9 to insulin was achieved by using a
heterobifunctional cross-linker SPDP that is reactive to –NH2 and –SH groups.
Specific conjugation to lysine moiety in the B-chain of insulin was done by blocking
all three amino terminals on insulin with citraconyl anhydride followed by partial
82
deblocking of lysine amino group through pH manipulation. To obtain a
homogenous conjugate as far as possible, purification steps using reversed phase
HPLC for every intermediate were employed. Eluent peaks were collected at the
retention time, concentrated and used for subsequent reaction. Representative
chromatograms of each step are shown in Fig 3.3 a- i.
As seen in Fig 3.3a, free insulin eluted at a RT of 12.23, upon addition of
citraconyl insulin, 2 peaks were formed eluting at RT 7.30 and 8.08 (Fig 3.3b)
representing 2 different isomeric forms of tricitraconyl insulin. After lowering the
pH to 5.0 to deprotect amino group on lysine B29, two new peaks emerged at RT 8.5
and 9.3 (Fig 3.3c) with the formation of dictraconyl insulin (InP4). Additional small
peaks seen at later RT were a result of further hydrolysis of the protected amines,
resulting in monocitraconyl insulin. Both new peaks were pooled, lyophilized and
used for subsequent reactions. InP4 (Fig 3.3d) was reacted with SPDP crosslinker to
obtain InP4-SPDP (Fig 3.3e). Reaction with SPDP led to a great amount of
aggregation possibly forming dimers and trimers of cross-linked insulin. Therefore
the reaction mixture was purified to isolate InP4-SPDP from the aggregates, and the
by-products (Fig 3.3f). The acidic conditions due to 0.1% TFA were necessary to
hydrolyze the protected amine groups after SPDP linkage. After concentrating the
pure InP4-SPDP peak, the pH of the solution was raised to pH 8 and reacted with
equimolar concentrations of cr9, ctat or ck9 to generate In-cr9 (Fig 3.3g), In-ctat (Fig
3.3h) and In-ck9 (Fig 3.3i) respectively. For all three conjugates, two peaks,
83
designated as pk1 and pk2 were recovered and analyzed. Both peaks showed
characteristics of the conjugate and were therefore mixed and used for other studies.
Table 3.1 shows a summary of the conditions for each step and the retention
times for insulin, intermediates and insulin conjugates. All the conjugates
synthesized were radiolabelled with 125I as described in the experimental section
and purified by sephadex G25 gel matrix.
84
Fig. 3.3a. Reversed phase HPLC chromatogram of insulin dissolved in PBS
85
Fig. 3.3b. Reversed phase HPLC chromatogram of tricitraconyl-insulin
generated after reacting insulin with citraconic anhydride
86
Fig. 3.3c. Reversed phase HPLC chromatogram of tricitraconyl insulin
hydrolyzed to dicitraconyl insulin (InP4)
87
Fig. 3.3d. Reversed phase HPLC chromatogram of purified dicitraconyl insulin
(InP4) using a C18 column
88
Fig. 3.3e. Reversed phase HPLC chromatogram of InP4-SPDP obtained by
reacting InP4 with SPDP
89
Fig. 3.3f. Reversed phase HPLC chromatogram of purified InP4-SPDP
90
Fig. 3.3g. Reversed phase HPLC chromatogram of In-cr9
91
Fig. 3.3h. Reversed phase HPLC chromatogram of In-ctat
92
Fig. 3.3i. Reversed phase HPLC chromatogram of In-ck9
93
Table 3.1: List of reaction conditions and retention times for insulin bioconjugates and intermediates
Sample Name Description HPLC
Column
Mobile Phase
(Solvent A and Solvent B)
Gradient
15 min
(Solvent B)
Retention Times*
(min)
1 Insulin C4 A: 10%MeCN in 20mM NH4CO3;
B: 95%MeCN in 20mMNH4CO3
15-30% 12.23
2 In-CitA-Rxn1 pH 8 reaction (Rxn1) C4 A: 10%MeCN in 20mM NH4CO3;
B: 95%MeCN in 20mMNH4CO3
15-30% 7.30 and 8.08
3 In-CitA-Rxn2 pH 5 reaction (Rxn2) C4 A: 10%MeCN in 20mM NH4CO3;
B: 95%MeCN in 20mMNH4CO3
15-30% 8.50 and 9.34
4 InP4 Dicitraconyl-insulin isolated
fromRxn2 above
C18 A: 10%MeCN in 0.1%TFA;
B: 95%MeCN in 0.1% TFA
10-100% 13.03 and 13.20
5 InP4-SPDP InP4 reacted with SPDP C18 A: 10%MeCN in 0.1%TFA;
B: 95%MeCN in 0.1% TFA
10-100% 13.5 and 13.67
6 InP4-SPDP (pure) Separated fromunreacted
InP4 and excess SPDP
C18 A: 10%MeCN in 0.1%TFA;
B: 95%MeCN in 0.1% TFA
17.5- 25% 12.6 and 13.70
7 In-cr9 InP4 reacted with cr9 C18 A: 10%MeCN in 0.1%TFA;
B: 95%MeCN in 0.1% TFA
17.5- 25% 10.75 and 11.60
8 In-ctat InP4 reacted with ctat C18 A: 10%MeCN in 0.1%TFA;
B: 95%MeCN in 0.1% TFA
17.5- 25% 10.9 and 11.76
9 In-ck9 InP4 reacted with ck9 C18 A: 10%MeCN in 0.1%TFA;
B: 95%MeCN in 0.1% TFA
17.5- 25% 11.05 and 11.9
94
2. Characterization of insulin bioconjugates
a. Modification of net charge of insulin after conjugation.
At neutral pH insulin in negatively charged (-2.1) with an isolectric point of
5.3. Upon conjugation with positively charged oligoarginine, the conjugate will bear
a net positive charge (6.9) at neutral pH with a calculated isoelectric point of 9.2.
Using a cationic gel matrix, we determined if conjugation to cr9 increased the net
charge of the bioconjugate. As seen in Fig 3.4a, In-cr9 did not bind to sephacel
DEAE and was eluted in 20mM tris buffer (pH =7.0) in the presence of a minimum
salt concentration (0.01M NaCl). On the other hand, insulin, due to its net negative
charge bound to the gel matrix and eluted from the column when a higher
concentration gradient (0.5M NaCl) of salt was applied. UV spectroscopy at 280 nm
was used since cr9 alone does not have absorbance at this wavelength and would
therefore not interfere with the detection of In-cr9.
-0.05
0
0.05
0.1
0.15
0.2
0.25
01 2 3 45 6 7 8 9 10 11 12 13 14 15 16 17
Fraction (ml)
OD
280
Insulin-cr9
Insulin
0.01M Nacl 0.15M Nacl
0.5M Nacl
Fig. 3.4a. Ion exchange chromatography to determine the charge state of
In-cr9 compared to insulin.
95
b. Determination of molecular weight by SDS-PAGE
SDS-PAGE was employed to detect changes in molecular weight of insulin
upon conjugation to cr9. Since disulfide linkage was used for conjugation, the gel
was run under non-denaturing conditions in the absence of reducing agent DTT or-
merkaptoethanol. In Fig. 3.4b, lane 1 shows a broad range marker with the lowest
molecular weight of X. Lane 2 shows insulin whereas lanes 3-4 show peak 1 (pk1:
RT 10.75min) and lanes 5-6 show peak 2 (pk2: RT 11.60 min). A single band that is
in line with the lowest band of protein marker and slightly above that of insulin
indicative of an increase in molecular weight represents the conjugate with a single
chain of cr9 attached.
A similar increase in molecular weight was observed for both peaks (pk1 and
pk2) of In-ctat and In-ck9 conjugates (data not shown)
1
2 3
4
5 6
Fig 3.4b. SDS-PAGE analysis. Samples were loaded onto a
discontinous gel consisting of 6.5% stacking gel and 15% running gel.
The gel was stained with coommassie blue to detect protein bands.
Lane 1 represents broad range marker, lane 2 represents insulin and lane
3-4 represent pk1 of In-cr9 and lane 5-6 represent pk2 of In-cr9
96
c. Determination of molecular weight by MALDI-TOF mass spectroscopy
To further confirm that the conjugation ratio of insulin to cell penetrating
peptide was indeed 1:1, the conjugates were analyzed by MALDI-TOF. Fig 3.4c (i)
confirms that a purely monosubstituted conjugate was obtained for cr9 with an
expected average molecular weight of 7421 (observed 7466.29). Fig 3.4c (ii) shows
the chromatograph of In-ctat. The expected average molecular weight of In-ctat is
7338 but the observed peak is at 2399.74, which corresponds to In-ctat with a
mass:charge ratio of 1:3. Fig. 3.4c (iii) shows In-ck9 with an expected average
molecular weight of 7169 and a peak observed at average molecular weight 7165.01.
97
(i)
(ii)
(iii)
0
10
20
30
40
50
60
70
80
90
100
%Int.
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000
M ass/Charge
146685 mV Profiles 1-243: Apex
Proteomics Core 032806
CFR Test
Data: leenasampleBlinearmode20001.O12 28 M ar 2006 16:12 Cal: linear 22 Oct 2003 12:38
Kratos PC A xima CFRplus V 2.3.4: M ode linear, Power: 129, Blanked, P.Ext. @1533 (bin 101)
7466.29
5894.92
5759.89
6021.35
1520.51
7727.14 3740.41
Fig. 3.4c: MALDI-TOF Mass spectroscopy chromatographs of pk1 of
insulin-CPP bioconjugates. (i) In-cr9; (ii) In-ctat; (iii) In-ck9.
98
3. Higher transport of In-cr9 compared to In-tat and In-ck9
To compare the transport efficiency of the different conjugates
synthesized, pre-equilibriated RAECII were incubated at 37 ºC for 2 hours with
radiolabelled insulin, In-cr9, In-ctat or In-ck9. Fig 3.5 shows the intact amount as a
percentage of dose that was transported across each monolayer. The intact amount
was determined by TCA precipitate assay. Compared to insulin, all three conjugates
showed statistically significantly higher transport than free insulin. The magnitude of
increase however, varied, with a 27-fold increase in In-cr9, 19-fold increase in In-
ctat and 4-fold increase in In-ck9 relative to insulin.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Insulin In-cr9 In-Tat In-ck9
Conjugate
% of Dose/ monolayer
**
**
*
Fig 3.5: Transport comparison of insulin and insulin conjugated to D isoforms
of oligoarginine, tat peptide or oligolysine after 2-hour incubation in AECII
monolayers. Samples represent average transport ± stdev of intact protein as a
percentage of dosing solution. Intact amount of protein was measured by TCA
precipitation assay, n=3 * denotes p<0.05 ** p<0.005 versus insulin control
99
4. Linear increase in transport of In-cr9 over time
Transport of In-cr9 in RAECII over a period of 4 hours was measured to
determine if there was saturation in transport. As seen in fig 3.6, the transport of In-
cr9 had a lag phase of one hour followed by a linear increase in transport for the next
three hours and did not saturate for the time period tested.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
01 2 345
Time (Hrs)
% Transport/Monolayer
Insulin
Insulin-cr9
Fig 3.6: Time-dependent transport in RAECII. Monolayers were incubated for
pre-determined time intervals with insulin or In-cr9. Intact protein transported
was determined by TCA precipitate assay. Data represents mean ± stdev, n= 4
100
5. Importance of covalent conjugation
a. Physical mixture versus covalent linkage
To deliver cargo to intracellular compartments using cell-penetrating
peptides, it is necessary to covalently link the cargo and the peptide. To determine if
covalent linkage was necessary for transcellular transport, we compared the transport
of In-cr9 to a physical mixture of equimolar quantities of insulin and cr9. Fig 3.7
shows that a physical mixture of cr9 and insulin was not sufficient to increase the
transport of insulin and it was necessary to covalently link the cr9 peptide to insulin.
0
0.05
0.1
0.15
0.2
0.25
Insulin In-cr9 Insulin + cr9
% Dose/Monolayer
Fig 3.7: Comparison of covalent linkage to physical mixture. RAECII
monolayers were incubated with insulin, In-cr9 or a physical mixture of
insulin and cr9 at equimlar ratio for 2hrs at 37 ºC. Basolateral media
was analyzed for intact protein by TCA precipitate assay. Values
represet mean ±stdev n=3
101
b. Higher uptake but low transport with Pep-1.
Pep-1, also known as Chariot®, is the only known cell penetrating peptide
that can deliver macromolecules, including biologically active proteins and
oligonucleotides into cells without the need for covalent conjugation. By mixing
insulin with chariot in a 1:4 ratio, we studied the uptake and transport of insulin after
4 hour and 24 hours of incubation. At both time points, chariot significantly
increased the uptake of insulin (Fig 3.8A), however there was no statistically
significant difference in the transcellular transport of insulin at 4 hours or 24 hours
(Fig 3.8B).
102
A
B
Transport
0
2
4
6
8
10
12
14
16
4hrs 24hrs
Incubation time (Hrs)
ng Insulin/monolayer
Insulin
Insulin + Pep1
Uptake
0
10
20
30
40
50
4hrs 24hrs
Incubation time (Hrs)
ng Insulin/monolayer
Insulin
Insulin + Pep1
Fig. 3.8: (A) Uptake and (B) transport of insulin in presence of Pep1.
125
I-
Insulin was mixed with Pep1 in 1:4 ratios and incubated with RAECII at 37 ºC
for 4 or 24 hrs. Cell inserts were washed thrice in ice cold PBS and counted for
radioactivity to determine cell uptake. Amount of intact insulin transported was
measured by radioactivity count following TCA precipitate assay. Values are
respresented as mean ± stdev, where n = 3
103
6. Effect of In-Cr9 on TEER of RAECII
TEER is used as a measure of cell monolayer integrity and is negatively
correlated with paracellular permeability of molecules across the monolayer. Fig 3.9
shows the effect of In-cr9 on TEER of RAECII during a 4-hour incubation period.
All monolayers had initial TEER values of >3.0 k.cm
2
. As controls, monolayers
were also incubated with insulin or a physical mixture of insulin and cr9. In-cr9
caused a slow onset of decline in TEER that did not exceed 19% of control for the
four-hour incubation period.
.
0
20
40
60
80
100
120
01 23 45
Time (Hrs)
% Change in TEER
Insulin
In-cr9
Ins + cr9
Fig 3.9: Percentage change of transepithelial electrical resistance
(TEER) of primary cultured alveolar epithelial cells treated with
insulin, insulin-cr9 or insulin + cr9 for upto 4 hours. Samples
represent average change ± stdev, n= 4
104
7. Effect of temperature on transport of In-cr9
To determine the effect of reduced temperature on the transport of In-cr9,
cells were incubated with In-cr9 for 2 hour at 37 ºC, 16 ºC and 4 ºC. In principle,
endocytic process of vesicle formation occurs at 16 ºC, however vesicle fusion, (a
step necessary for cellular internalization) is arrested. At 4 ºC however, both vesicle
formation and vesicle fusion are inhibited. In addition, the membrane fluidity of
cells is reduced. To determine the involvement of endocytosis and the cell membrane
in transport of conjugate, effect of temperature was studied. The transport of intact
peptide decreased to 65% and 79% of control at 16 ºC and 4 ºC respectively (Fig
3.10).
0
20
40
60
80
100
Control 16 4
Temperature [
o
C]
% of Control
Fig 3.10: Transport of intact
125
I-Insulin across RAECII after incubation at
different temperatures. Monolayers were incubated for 2 hours at 37 ºC, 16
ºC or 4 ºC. Amount of intact protein transported was determined by TCA
precipitate assay. Values represent mean ± stdev where n = 4
105
8. Comparison of transport in other epithelial cell models
To find out if the increase in transport of In-cr9 was specific only to RAECII,
we studied the transport of the conjugate in two other epithelial cell models Caco-2
and RAECI (Fig 3.11). Caco-2 cells are routinely used as an in vitro model to test
permeability of molecules as a surrogate for oral delivery. Whereas RAECI is a cell
model that represents the type I cells present in pulmonary epithelium. An increase
in transport in both cell models was observed. Compared to insulin, there was a 2-
fold increase in transport for In-cr9 in Caco-2 cells and 5-fold increase in RAECI
cell model.
RAECI
0.00
0.05
0.10
0.15
0.20
0.25
0.30
Insulin Insulin-cr9
% dose/Monolayer
Caco-2 Cells
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Insulin Insulin-cr9
% dose/monolayer
Fig. 3.11: Transport of In-cr9 in RAECI and Caco-2 cell models.
Confluent monolayers were incubated with insulin or In-cr9 at 37 ºC
for 2 hrs. The amount of intact protein transported in each well was
determined by TCA precipitate assay. Values represent mean ±
stdev where n = 3-4
106
9. Effect of biochemical modulators
a. Inhibitors of endocytosis
Temperature dependence studies on the transport of In-cr9 indicated the
involvement of endocytosis. To further probe the nature of endocytosis and
subsequent transcytosis involved, we tested the transport of In-cr9 in presence of a
panel of endocytosis inhibitors (Table 3.2). There was no statistically siginificant
difference when cells were pre-treated with 10 mM or 50 mM NH
4
Cl, 30 µM
monensin and 30-µM chlorpromazine. There was a drastic increase, ~ 40% in
transport in the presence of 33 µM nocodazole. In the presence of protamine a
modest decrease in transport was observed compared to control
b. Inhibitors of macropinocytosis
Macropinocytosis is a specialized form of endocytosis that occurs in most cells.
To determine if macropinocytosis is involved in transport of In-cr9, cells were
pretreated with macropinocytosis inhibitors, EIPA and cytochalasin D. Compared to
control cells, no change in transport was observed.
c. Metabolic inhibitors
In the presence of 10mM sodium azide and 6mM deoxyglucose, 15% decrease in
transport was observed compared to control.
107
10. Effect of heparin and protamine on the uptake and transport of In-cr9
To determine if negatively charged or positively charged species could
compete with In-cr9 for binding/ uptake and transport, cells were pretreated with 1
μM heparin (negative charge) or 1 μM protamine (positive charge), and subsequently
dosed with In-cr9. Fig 3.12 shows the effect of heparin and protamine on the uptake
and transport of In-cr9. In the case of heparin, a 16 ± 4.6% decrease in uptake and
19.4 ± 4.5 % decrease in transport was observed. In the presence of protamine, the
decrease in uptake was 21.6 ± 11.3% whereas decrease in transport was 23.2 ± 8.6%.
Concentration % Intact
transport
Control - 100 ± 6
NH4Cl 10 mM 96 ± 3
NH4Cl 50 mM 113.8 ± 10
Monensin 30µM 96.7 ± 5
Nocodazole * 33µM 142 ± 12
Chlorpromazine 30µM 91 ± 15
Control - 100 ± 21
EIPA 200µM 128 ± 8
Cytochalasin D 0.1µg/ml 115 ± 21
Control - 100 ± 8
NaN3 + Deoxyglucose * 10 mM + 6 mM 85 ± 2
Protamine * 1 μM 78 ± 11
Table 3.2: Percentage transport of In-cr9 in RAECII in the presence of various
biochemical modulators of transport. Values represent mean ± stdev where n= 3-4
108
Cell Uptake
0
20
40
60
80
100
120
Control Heparin Protamine
% of Control.
(A)
Transport (Intact)
0
20
40
60
80
100
120
Control Heparin Protamine
% of Control
(B)
Fig. 3.12: (A) Uptake and (B) transport of In-cr9 in the presence of 1 μM
heparin or protamine. Confluent monolayers were pre-incubated with heparin
or protamine for 30 min followed by incubation with In-cr9 + heparin or In-cr9
+ protamine for an additional 2 hrs at 37 ºC. Cell inserts were washed thrice in
ice cold PBS and counted for radioactivity to determine cell uptake. Amount of
intact protein transported was measured by radioactivity count following TCA
precipitate assay. Values are represented as mean ± stdev, where n = 4
109
11. Lack of competition for receptor binding by In-cr9
To test if In-cr9 could compete with insulin to bind to insulin receptors in
vitro, we used HepG2 cells, which are shown to have a high concentration of insulin
receptors.
125
I-Tyr14-insulin which is insulin specifically labeled with
125
I at tyrosine
14 of B-chain was used as the radioligand. Preliminary saturation experiments were
performed with various concentrations of radiolabelled insulin (Fig 3.13). Using
non-linear regression curve fitting, Kd value of insulin was estimated to be 125 ± 29
pM and used as a reference for subsequent competition studies.
Fig 3.14A shows competition of receptor binding with increasing
concentrations of unlabelled insulin with an IC
50
of 1.2 (± 0.4) x 10
-9
M. Fig 3.14B
shows that there was no competition for receptor binding with increasing
concentrations of unlabelled In-cr9. upto 100 nM.
1 10 100 1000
0
5000
10000
CPM
[
125
I Insulin] (pM)
CPM (min) = 188
CPM (max) = 6886 +/- 1290
Kd = 125 +/- 39 pM
Fig 3.13: Concentration dependent binding of
125
I-(Tyr14)-insulin in HepG2 cells.
110
(A)
0.1 1 10 100 1000 10000 100000
0
50
100
% [I
125
] Insulin / Control
[Unlabeled Insulin] (pM)
0
20
40
60
80
100
120
140
1 10 100 1000 10000 100000 1000000
[In-cr9] (pM)
% Control
In-Cr9
(B)
Fig. 3.14: Receptor binding competition assay for (A) insulin (B) In-cr9 in
HepG2 cells.
111
12. Higher and prolonged effect of In-cr9 vs insulin after spray instillation
Streptozotocin induced diabetic rats were used as an in vivo model to test the
effect of In-cr9 when delivered to the lung using spray instillation technique. The %
drop in glucose level at various time intervals was used as the pharmacodynamic
effect of In-cr9 (fig 3.15). Rats were fasted overnight to stabilize baseline glucose
levels and initial glucose levels of all the experimental rats were above 400 mg/dL.
Initially, four rats each for the insulin and In-cr9 groups were used whereas 3 rats
were used for the control PBS group. However, in both insulin and In-cr9 groups,
one rat did not respond to treatment, therefore during data analysis, these rats were
considered as outliers. As seen in fig 3.15, instillation of In-cr9 led to an initial surge
in glucose level up to 30 min and gradually declined to 20% of initial values up to
300 min and then leveled off from 300 min to 460 min. At a similar dose, insulin
showed a 20% decrease in glucose levels at 120 min and continued to decrease up to
40% of initial values over the next 180 min. An equal volume of PBS did not show
any significant effect on glucose level for the time period tested.
112
112
0
20
40
60
80
100
120
140
0 100 200 300 400 500
Time (min)
% of Control.
Control (PBS)
Insulin (.375 mg/kg)
Conjugate (.375 mg/kg Eq-In)
Fig 3.15: Percentage change in glucose after spray instillation of insulin, In-cr9 or PBS in fasted STZ
induced diabetic rats. Values represent mean ± SE ; n=3
113
IV. DISCUSSION
In this chapter, the feasibility of using cationic CPP for the transcellular
delivery of insulin, as a model protein, across alveolar epithelial cells was
established.
First, insulin was bioconjugated with cr9, ctat or ck9 in a 1:1 ratio with site-
specific conjugation at Lys B29. Both DMMA and Cit A were tried as amino
protecting groups, but due to rapid hydrolysis of DMMA, it was difficult to control
subsequent SPDP reaction and therefore Cit A was used. As expected, 2 isomers of
tri- and dicitraconyl insulin resulted when Cit A was reacted with insulin at different
pH. Unexpectedly, two peaks also resulted in the final conjugates. The isomers are a
result of 2 possible forms of intermediates when Cit A reacts with insulin, however
in the acidic conditions of TFA, Cit A in the intermediate (InP4-SPDP) is expected
to hydrolyze and result only in one possible peak of each conjugate when reacted
with CPP. The presence of two peaks could then be either due to incomplete
hydrolysis of Cit A or additional conjugates of 1:2 or 1:3 ratio of insulin:CPP.
To rule out the second possibility and also further characterize the conjugate,
SDS-PAGE and MALDI-TOF analysis were employed. Both methods indicated a
1:1 ratio of conjugation as seen by a slight upwards shift in bands in SDS-PAGE and
a corresponding peak of expected molecular weight for each conjugate in the mass
spectra. The difference between the average expected and obtained mwt for In-cr9 is
45 da whereas for In-ck9 is 4 da. This difference could likely be due to positive ions
on the two-carboxyl terminals of the two chains in insulin, presence of 2 Na
+
ions
114
(46 da) for In-cr9. For In-ck9, possibly the carboxyl terminal of ck9 is also ionized
with a total of 3 H
+
ions (3 da). Ion-exchange chromatography also revealed an
increase in overall positive charge for In-cr9 compared to insulin at neutral pH.
The transcellular transport of all three conjugates, In-cr9, In-ctat and In-ck9
was compared to that of insulin in RAECII. It is observed that the oligoarginine
conjugate had the highest efficiency in transport, which is consistent with the results
in chapter 1 where oligoarginine transport efficiency was higher than oligolysine or
Tat peptide. This is also consistent with other cellular uptake studies of conjugated
and unconjugated peptides that show oligoarginine as a more efficient CPP than
oligolysine and tat peptide (Wender et al. 2000; Zaro and Shen 2003). Thus, In-cr9
was chosen for further in vitro and in vivo studies.
A time-dependent study showed that the transport of In-cr9 did not saturate
up to 4 hours, also consistent with a non-saturable transport mechanism that was
seen in chapter 1 for yr7. Covalent linkage to CPP is almost always necessary to
facilitate delivery of proteins across plasma membrane. However, Morishita et al
recently showed that insulin absorption across the intestine could be increased by D-
isoforms of oligoarginine namely, r6, r8 and r10 by mixing these peptides in insulin
formulation (Morishita et al. 2007). We determined if a simple physical mixture of
insulin and cr9 was sufficient, to result in a similar increase in in vitro transport
across RAECII. We did not observe any increase in transport of insulin, possibly
because of the difference in model or due to difference in concentrations used. Our
concentration for transport studies was in µM range whereas they used oligoarginine
115
in the mM range. At such high concentrations, it is likely that paracellular transport
is increased, with likelihood of cytotoxicity concerns. To further investigate if a
physical mixture is sufficient for transport, Pep1, the only known CPP that increases
protein cell uptake by non-covalent means was examined (Morris et al. 2001). There
was an increase in cellular uptake however no increase in transcellular transport was
observed, further confirming the need for covalent conjugation for transcellular
transport.
In chapter 1, we reported that YGR9 could be transported across both RAECI
and RAECII to almost the same extent. This finding is crucial for in vivo
applications of CPP because in the in vivo setting, type I cells co-exist with the type
II cells to form a continuous and heterogeneous epithelial barrier. We investigated if
the conjugate form, In-cr9 could also be transported across RAECI. While an
increase in transport is seen in RAECI also, the magnitude of increase is much lower
than that of RAECII (5- versus 19 to 27-fold increase respectively). This implies
that the in vivo increase in activity we see is as a result of increased transport across
both type I and type II cells, but it also points towards different dynamics in
processing of the conjugate. As mentioned in chapter 1, some of the factors that
could contribute to the difference are, differential expression of glycosaminoglycans-
suggested binding sites, different membrane turnover rates or different intracellular
pathways of transport. Williams et al made a similar observation for cationic ferritin
when comparing type II cells and type I cells (Williams 1984). Type II cells
internalized and transported cationic ferritin much more efficiently than type I cells,
116
presumably due to different pathways of internalization. Surprisingly, in Caco-2
cells, the increase in transport of In-cr9 was only 2-fold higher. One possible reason
for lower transport in Caco-2 cells compared to alveolar cells is the relative thickness
of these epithelial cells. Both type I and type II alveolar cells are <5 μm thick
whereas Caco-2 cells are ~ >20 μm. Thus, transcellular movement across Caco-2
cells is likely to be hindered much more than in alveolar cells. Another difference
that may play a role is that Caco-2 cells are derived from an established cell line
whereas RAEC are primary cultures with possibly different expression of cell
surface components responsible in facilitating transport.
To further probe the pathways by which RAECII, enhance transport of In-
cr9; we studied the involvement of endocytosis. Transduction is considered as a
unique mechanism of cell entry for CPPs but endocytosis has also been implicated as
a concomitant pathway of entry (Zaro and Shen 2003, 2005). Additionally,
conjugation of CPPs to various payloads has resulted in a purely endocytic uptake of
the conjugate due to the nature of the cargo (Futaki 2002; Dietz and Bahr 2004). Our
first line of investigation involved the effect of temperature on transport. Endocytosis
and in particular vesicle fusion is arrested at 16 ºC, resulting in decreased uptake and
transport at this temperature. Our results show a 65% decrease in transport at 16 ºC,
which could be attributed to endocytic arrest or arguably a decrease in passive
paracellular transport suggested to be in the range of ~ 40% at reduced temperatures
(Matsukawa et al. 1996). We ruled out a decrease due to only paracellular transport
because we see a further reduction in transport at 4 ºC (79%), which implies that the
117
transport process involves more than just paracellular pathway, or rather the plasma
membrane is involved since membrane fluidity is highly restrictive at 4 ºC. It has
been shown previously by Zaro and Shen that the involvement of plasma membrane
could be in the form of vesicles and hence endocytosis or the unique transduction
mechanism for oligoarginine and its conjugates (Zaro and Shen 2003, 2005). Further
tests were carried out to investigate the nature of endocytosis involved.
In principle, endocytosis could result from specific or non-specific binding to
cell surface components followed by vesicle formation which follow one of the three
pathways, i.e recycled back to the membrane, sorted to lysosomes for degradation or
trafficked to the opposite membrane whereby the contents are released out of the
cell. For our purpose the third pathway, i.e transcytosis is of significance since
maximum delivery to the basal compartment is desired. Transcytosis can occur as a
result of both non-specific (adsorptive, fluid-phase) and specific (receptor-mediated)
endocytosis. Cationic CPPs have been suggested to bind non-specifically to
proteoglycans present on plasma membrane of cells such as CHO cells resulting in
adsorptive endocytosis (Sandgren et al. 2002; Fuchs and Raines 2004). To determine
the role of adsorptive endocytosis and consequently transcytosis in RAECII, we
performed transport studies in the presence of heparin or protamine. Heparin is
polyanionic in nature and has been shown to inhibit adsorptive-mediated transcytosis
of polylysine conjugated to tyramine in MDCK cells (Taub et al. 1994). On the other
hand, protamine has been used to inhibit adsorptive mediated endocytosis of a
cationic peptide in Caco-2 cells (Sai et al. 1998). Heparin would therefore neutralize
118
the positive charge of In-cr9 whereas protamine a polycation would compete with
the anionic binding sites on the cell surface. Both reagents showed a modest yet
statistically significant decrease in uptake and transport of In-cr9 suggesting that the
transcellular transport is partly but not wholly mediated by adsorptive transcytosis.
Additionally, depleting the cellular energy source by incubating with the metabolic
poison sodium azide and 2-deoxyglucose led to a ~15% decrease in transport.
Vesicular endocytosis and transcytosis involve a number of steps and cellular
machinery. Using a limited approach, we looked at the effect of known specific
inhibitors of endocytosis on In-cr9 transport. Ammonium chloride is a
lysosomotropic amine that alkalinizes the intracellular acidic vesicles whereas
monensin is an ionophore that interferes with vesicular Na
+
/ H
+
exchange (Seglen et
al. 1979; Deffebach et al. 1996). Both these reagents disrupt the transcytosis and
secretory pathways involving the acidic organelles, including endoplasmic reticulum,
Golgi apparatus and endosomes. However, neither ammonium chloride nor
monensin had an effect on the transport of intact In-cr9. Chlorpromazine interferes
with the formation of clathrin-coated pits and hence functions as an inhibitor of
clathrin-mediated endocytosis, but it did not affect the transcytosis of In-cr9. On the
other hand, nocodazole an inhibitor of microtubules polymerization led to a
significant increase in transport of In-cr9. Microtubules are important cytoplasmic
elements that facilitate the movement of vesicles in the cytoplasm, inhibition of
microtubule formation is thus supposed to limit vesicular transcytosis if any. Our
results contradict this phenomenon, instead of inhibition or no-effect we observe a
119
statistically significant increase in transport, however we can explain the increase in
transport by the drastic drop in TEER observed with nocodazole treatment. This
explanation is further supported by the findings of Birukova et al who showed that
nocodazole promotes barrier dysfunction of pulmonary endothelial cells monolayers
by cell retraction and paracellular gap formation (Birukova et al. 2004).
Macropinocytosis in another likely vesicular pathway of transport for In-cr9.
Recent studies have suggested that oligoarginine in not only internalized by
macropinocytosis but it also induces the formation of macropinosomes, a pre-
requisite for macropinocytosis (Nakase et al. 2004). Macropinocytosis occurs in
almost all cells and is used as mechanism of cell entry by several pathogens. In
A549, human alveolar type II cells, this pathway has been exploited by
mycobacterium tuberculosis (Garcia-Perez et al. 2003). To determine if the
macropinocytosis pathway played a role, we investigated the transport in presence of
inhibitors, cytochalasin D and ethylisopropylamiloride (EIPA). Cytochalasin D
inhibits f-actin polymerization that is a necessary step for filopodia formation
whereas EIPA inhibits macropinocytosis by inhibiting Na
+
/H
+
exchange (Nakase et
al. 2004). Both inhibitors did not show any decrease in transport of In-cr9 suggesting
that the role of macropinocytosis in its transcellular transport is limited.
Taken together, the data on transmembrane transcellular transport does not
point to any particular type of endocytosis/transcytosis as a predominant pathway of
transport of In-cr9 in RAECII. Adsorptive endocytosis seems to play a role but does
120
not sum up to the ~79% decrease in transport seen at 4 ºC leaving room for a
possible role of the unique transduction mechanism purported for cationic CPP.
The role of paracellular pathway cannot be ignored in this context in light of
the cationic nature of our macromolecule. It has been shown that cationic moieties,
such as poly-L-lysine, poly-L-arginine, protamine and cationic derivatives of
chitosan promote paracellular transport by modulating tight junctional complexes
(Peterson and Gruenhaupt 1990; Ohtake et al. 2002). However, our results on the
effect on TEER show that cr9 in itself does not decrease the TEER to any significant
level, neither does it increase the transport of free insulin. With the In-cr9 conjugate,
a gradual decrease in TEER is seen, but it does not decrease below ~80% of control
and hence would not likely contribute significantly to paracellular transport. The
decrease in TEER seen with In-cr9 but not when insulin is mixed with cr9 is
nonetheless interesting. We speculate that In-cr9 is interacting directly or indirectly
with cellular components that modulate tight junctions probably due to different
pathways followed.
The conjugation strategy of insulin we chose involved using Lys B29 because
of ease of modification of -amino group and also because Lys B29 is peripheral to
the receptor binding region of insulin. To determine if In-cr9 conjugate retains its
biological activity, we performed insulin receptor competition binding assay in
HepG2 cells. Our preliminary binding studies of radiolabeled insulin show a Kd
value of 125 pM for insulin in HepG2 cells. Competition binding studies with
unlabelled insulin showed an IC
50
value of 1.2 ± 0.4 nM that is consistent with
121
literature reports for insulin receptor binding in HepG2 cells (Mei et al. 1999; Lee et
al. 2005). However, In-cr9 did not seem to compete for receptor binding up to 1 µM.
The probable explanation for a lack of receptor binding is steric hindrance caused
either by the molecular weight or charge repulsion of cr9. It is known that shortening
of the B-chain of insulin by 5 amino acids, increases receptor binding affinity of
insulin due to conformational changes that further exposes the N-terminal residues of
the A-chain (Zakova et al. 2004). Attaching small lipophilic moieties such as
deoxycholic acid (374.6 da) at position B29 does not interfere with receptor
binding(Lee et al. 2005). However cr9 is neither lipophilic nor is it small in size,
therefore the molecular size, hydrophilicity and charge could contribute towards
conformational changes around the B-chain that decrease receptor binding affinity.
Receptor binding is necessary for insulin to exert its biological activity; the in
vitro data does not support receptor binding of In-cr9, however in vivo biological
activity in terms of glucose depression is seen after spray instillation in diabetic rats.
This clearly indicates that conformationally stable insulin is regenerated from In-cr9
in vivo and further reinforces the significance of the reversible disulfide linkage,
which is able to release a bioactive form of macromolecule.
Finally, the intention of conjugating a cationic CPP to insulin was to increase
its permeability across the alveolar epithelium, in deed the in vivo data show that In-
cr9 has better pharmacodynamic activity than insulin upon spray instillation.
Furthermore the activity of In-cr9 has a sustained effect that plateaus to ~20%
baseline glucose levels at 6-hours post injection. Several reasons can explain the
122
prolonged effect of In-cr9 as depicted in Fig 3.16. Firstly, time-dependent in vitro
transport studies showed that In-cr9 transport increases over time with no saturation
component. This continuous non-saturable absorption and supply of insulin could be
one contributing factor.
Reduction of the disulfide bond between insulin and cr9 could occur
intracellularly, in the epithelial cells with release of unconjugated insulin in the
bloodstream. On the other hand, the conjugate could be released into the bloodstream
Fig. 3.16: Schematic representation of proposed reasons for prolonged effect of
In-cr9 in vivo. The non-saturable transport seen in vitro could result in a continous
supply of insulin after (1) epithelial intracellular reduction of In-cr9 disulfide bond
or (2) intact In-cr9 is released into the bloodstream. Once in the blood stream, In-
cr9 could accumulate in the (3) lungs, (4) liver or (5) plasma that would act as a
depot. Through the circulation of In-cr9 in the liver, the disulfide bond could be
-S—S-
-SH
HS-
HS-
DISULFIDE
REDUCTION
DEPOT
Lung Liver Plasma
3 4 5
NON-SATURABLE
TRANSPORT
-S—S-
HS-
1 2
6
-S—S- -S—S-
-SH -SH
HS- HS-
HS- HS-
DISULFIDE
REDUCTION
DEPOT
Lung Liver Plasma
3 4 5
NON-SATURABLE
TRANSPORT
-S—S- -S—S-
HS- HS-
1 2
6
123
intact. In this case, complexation of In-cr9 with anionic components in the lung
endothelial basement membrane, liver or in plasma such as heparin could act as a
depot from which In-cr9 dissociates slowly. The hypothesis of a probable depot
effect is based on previous finding which showed that cationic polylysine tends to
accumulate in the lung and liver due to its positive charge (Ekrami et al. 1993). It is
also likely that the disulfide bond reduction in the liver would mediate as a rate-
limiting step in the release of free insulin.
In summary, we have established that cationic CPPs serve as a good delivery
vector for insulin to enhance its permeability across pulmonary epithelium. Higher
efficiency in transport is observed when insulin is conjugated to oligoarginine as
compared to tat or oligolysine. The relative transport of In-cr9 is higher in RAECII
compared to RAECI and Caco-2 cells. The transport pathway of In-cr9 across
RAECII is mainly transcellular with a modest but statistically significant
contribution by adsorptive transcytosis. It does not severely affect the TEER and
hence integrity of the monolayers and also, covalent linkage between insulin and cr9
is necessary to increase permeability but the covalent linkage has to be reversible,
since the conjugate does not bind to the receptor. Lastly, a positive in vitro- in vivo
correlation is seen with In-cr9 showing a better pharmacodynamic activity than
insulin in diabetic rats when administered by the pulmonary route. We conclude that
oligoarginine has great potential to serve as a delivery system for the transcellular
transport of other macromolecular biotherapeutics across the pulmonary epithelia.
124
V. SUMMARY
Biosynthesis
Transport efficiency
Cell dependency
In-cr9 vs In + cr9
Receptor binding
In vivo activity
Transport pathway
Cell integrity
I:1 ratio insulin :CPP
In-cr9 > In-ctat > In-ck9 > In
RAECII > RAECI >Caco-2 cells
Covalent linkage necessary
Transcellular
Minor effect on TEER
No in vitro receptor binding
In-cr9 >> Insulin
Insulin
&
Oligoarginine
Transcytosis? Partial adsorptive transcytosis
Biosynthesis
Transport efficiency
Cell dependency
In-cr9 vs In + cr9
Receptor binding
In vivo activity
Transport pathway
Cell integrity
I:1 ratio insulin :CPP
In-cr9 > In-ctat > In-ck9 > In
RAECII > RAECI >Caco-2 cells
Covalent linkage necessary
Transcellular
Minor effect on TEER
No in vitro receptor binding
In-cr9 >> Insulin
Insulin
&
Oligoarginine
Transcytosis? Partial adsorptive transcytosis
Scheme 4: Summary of finding in chapter 3
125
CHAPTER 4
Future Directions
1. Large-scale synthesis of stable insulin-CPP bioconjugates
The bioconjugation method described in chapter 3 for relatively
homogeneous insulin conjugates involved analytical scale synthesis. To further study
the beneficial effect of this conjugation especially in vivo, it is necessary to produce
higher amounts of pure conjugates. To produce larger quantities, the method can be
validated for semi-preparative or preparative scale synthesis of the conjugate;
alternatively, a molecular genetic approach can be utilized to create a fusion protein
of insulin and the cationic CPP. The problems faced in synthesizing in small scale
such, as protein aggregation, adsorption, and solubility are likely to be present when
preparing larger amounts of conjugates. However, with larger quantities it will be
possible to systematically investigate the stability and formulate the conjugate in an
optimal way.
2. Broader in vivo studies defining pharmacokinetic and pharmacodynamic
parameters.
The prolonged effect of In-cr9 seen after pulmonary delivery is very
interesting and needs further investigation. Studying the biodistribution of the
conjugate will help in determining if there is any preferential organ uptake. Previous
work done in our lab indicates that cationic modification of proteins shows a
preferential targeting to lungs and liver (Ekrami et al. 1993). A preferential retention
in the lung or liver would support our explanation of depot effect due to electrostatic
126
complexation with lung endothelial components or accumulation in the liver would
suggest a possible role for the slow disulfide cleavage in the reductive environment
of the liver. Profiling the circulating plasma insulin levels at various time points will
also help in explaining the sustained effect observed.
3. Long term cytotoxicity effects of CPPs in lungs
The current reluctance of embracing inhalation insulin in clinical settings is
due to the lack of comprehensive long-term cytotoxicity studies of pulmonary
insulin. Although insulin is a physiological protein, the repetitive use through
pulmonary route is bound to pose a problem for the delicate lung tissues. From our
studies it is clear that the CPPs are relatively non-toxic in vitro, for short term as
indicated by MTT assay and reversal of TEER properties. However, a major concern
is the long-term effects of cationic CPP exposure. Higher molecular weight Poly- L
lysine at high concentrations is shown to induce airway hyperresponsiveness in rats,
a characteristic feature of asthma. It will be important to determine if the smaller
cationic CPP would induce a similar response if used repeatedly.
4. Application to other therapeutic macromolecules
The increase in insulin transport seen due to its conjugation with
oligoarginine is encouraging and makes a sound case for investigating the feasibility
of applying this drug delivery strategy to other macromolecules intended for
systemic delivery through the lungs.
127
REFERENCES
Arbit E. (2004) The physiological rationale for oral insulin administration. Diabetes
Technol Ther 6, 510-517.
Artursson P., Palm K. and Luthman K. (2001) Caco-2 monolayers in experimental
and theoretical predictions of drug transport. Adv Drug Deliv Rev 46, 27-43.
Augustijns P. F., Bradshaw T. P., Gan L. S., Hendren R. W. and Thakker D. R.
(1993) Evidence for a polarized efflux system in CACO-2 cells capable of
modulating cyclosporin A transport. Biochem Biophys Res Commun 197, 360-365.
Banting F. G. and Best C. H. (1990) Pancreatic extracts. 1922. J Lab Clin Med 115,
254-272.
Bhatnagar S., Srivastava D., Jayadev M. S. and Dubey A. K. (2006) Molecular
variants and derivatives of insulin for improved glycemic control in diabetes. Prog
Biophys Mol Biol 91, 199-228.
Birk Y. (1985) The Bowman-Birk inhibitor. Trypsin- and chymotrypsin-inhibitor
from soybeans. Int J Pept Protein Res 25, 113-131.
Birukova A. A., Smurova K., Birukov K. G., Usatyuk P., Liu F., Kaibuchi K., Ricks-
Cord A., Natarajan V., Alieva I., Garcia J. G. and Verin A. D. (2004) Microtubule
disassembly induces cytoskeletal remodeling and lung vascular barrier dysfunction:
role of Rho-dependent mechanisms. J Cell Physiol 201, 55-70.
Blundell T. L., Cutfield J. F., Dodson E. J., Dodson G. G., Hodgkin D. C. and
Mercola D. A. (1972) The crystal structure of rhombohedral 2 zinc insulin. Cold
Spring Harb Symp Quant Biol 36, 233-241.
Bredehorst R., Wemhoff G. A., Kusterbeck A. W., Charles P. T., Thompson R. B.,
Ligler F. S. and Vogel C. W. (1991) Novel trifunctional carrier molecule for the
fluorescent labeling of haptens. Anal Biochem 193, 272-279.
Bundgaard H. and Rasmussen G. J. (1991) Prodrugs of peptides. 9. Bioreversible N-
alpha-hydroxyalkylation of the peptide bond to effect protection against
carboxypeptidase or other proteolytic enzymes. Pharm Res 8, 313-322.
Butler P. J., Harris J. I., Hartley B. S. and Leberman R. (1967) Reversible blocking
of peptide amino groups by maleic anhydride. Biochem J 103, 78P-79P.
128
Campbell L., Abulrob A. N., Kandalaft L. E., Plummer S., Hollins A. J., Gibbs A.
and Gumbleton M. (2003) Constitutive expression of p-glycoprotein in normal lung
alveolar epithelium and functionality in primary alveolar epithelial cultures. J
Pharmacol Exp Ther 304, 441-452.
Cheek J. M., Kim K. J. and Crandall E. D. (1989) Tight monolayers of rat alveolar
epithelial cells: bioelectric properties and active sodium transport. Am J Physiol 256,
C688-693.
Cho M. J., Thompson D. P., Cramer C. T., Vidmar T. J. and Scieszka J. F. (1989)
The Madin Darby canine kidney (MDCK) epithelial cell monolayer as a model
cellular transport barrier. Pharm Res 6, 71-77.
Deffebach M. E., Bryan C. J. and Hoy C. M. (1996) Protein movement across
cultured guinea pig trachea: specificity and effect of transcytosis inhibitors. Am J
Physiol 271, L744-752.
Dietz G. P. and Bahr M. (2004) Delivery of bioactive molecules into the cell: the
Trojan horse approach. Mol Cell Neurosci 27, 85-131.
Dixon H. B. and Perham R. N. (1968) Reversible blocking of amino groups with
citraconic anhydride. Biochem J 109, 312-314.
Dodoo A. N., Bansal S., Barlow D. J., Bennet F. C., Hider R. C., Lansley A. B.,
Lawrence M. J. and Marriott C. (2000) Systematic investigations of the influence of
molecular structure on the transport of peptides across cultured alveolar cell
monolayers. Pharm Res 17, 7-14.
Duttaroy A., Kanakaraj P., Osborn B. L., Schneider H., Pickeral O. K., Chen C.,
Zhang G., Kaithamana S., Singh M., Schulingkamp R., Crossan D., Bock J.,
Kaufman T. E., Reavey P., Carey-Barber M., Krishnan S. R., Garcia A., Murphy K.,
Siskind J. K., McLean M. A., Cheng S., Ruben S., Birse C. E. and Blondel O. (2005)
Development of a long-acting insulin analog using albumin fusion technology.
Diabetes 54, 251-258.
Ekrami H., Kennedy A. R., Witschi H. and Shen W. C. (1993) Cationized Bowman-
Birk protease inhibitor as a targeted cancer chemopreventive agent. J Drug Target 1,
41-49.
Elbert K. J., Schafer U. F., Schafers H. J., Kim K. J., Lee V. H. and Lehr C. M.
(1999) Monolayers of human alveolar epithelial cells in primary culture for
pulmonary absorption and transport studies. Pharm Res 16, 601-608.
129
Fabry M. and Brandenburg D. (1992) Design and synthesis of a novel biotinylated
photoreactive insulin for receptor analysis. Biol Chem Hoppe Seyler 373, 143-150.
Fernandez-Carneado J., Kogan M. J., Pujals S. and Giralt E. (2004) Amphipathic
peptides and drug delivery. Biopolymers 76, 196-203.
Fjellestad-Paulsen A., Hoglund P., Lundin S. and Paulsen O. (1993)
Pharmacokinetics of 1-deamino-8-D-arginine vasopressin after various routes of
administration in healthy volunteers. Clin Endocrinol (Oxf) 38, 177-182.
Forbes B., Wilson C. G. and Gumbleton M. (1999) Temporal dependence of
ectopeptidase expression in alveolar epithelial cell culture: implications for study of
peptide absorption. Int J Pharm 180, 225-234.
Fredholt K., Ostergaard J., Savolainen J. and Friis G. J. (1999) alpha-Chymotrypsin-
catalyzed degradation of desmopressin (dDAVP): influence of pH, concentration and
various cyclodextrins. Int J Pharm 178, 223-229.
Fuchs S. M. and Raines R. T. (2004) Pathway for polyarginine entry into mammalian
cells. Biochemistry 43, 2438-2444.
Futaki S. (2002) Arginine-rich peptides: potential for intracellular delivery of
macromolecules and the mystery of the translocation mechanisms. Int J Pharm 245,
1-7.
Garcia-Contreras L., Morcol T., Bell S. J. and Hickey A. J. (2003) Evaluation of
novel particles as pulmonary delivery systems for insulin in rats. AAPS PharmSci 5,
E9.
Garcia-Perez B. E., Mondragon-Flores R. and Luna-Herrera J. (2003) Internalization
of Mycobacterium tuberculosis by macropinocytosis in non-phagocytic cells. Microb
Pathog 35, 49-55.
Gelman R. A. and Blackwell J. (1973) Heparin-polypeptide interactions in aqueous
solution. Arch Biochem Biophys 159, 427-433.
Ghetie V. and Vitetta E. S. (2001) Chemical construction of immunotoxins. Mol
Biotechnol 18, 251-268.
Hentz N. G., Richardson J. M., Sportsman J. R., Daijo J. and Sittampalam G. S.
(1997) Synthesis and characterization of insulin-fluorescein derivatives for
bioanalytical applications. Anal Chem 69, 4994-5000.
130
Hidalgo I. J., Raub T. J. and Borchardt R. T. (1989) Characterization of the human
colon carcinoma cell line (Caco-2) as a model system for intestinal epithelial
permeability. Gastroenterology 96, 736-749.
Ippoliti R., Lendaro E., Bellelli A., Fiani M. L., Benedetti P. A., Evangelista V. and
Brunori M. (1996) A saporin-insulin conjugate: synthesis and biochemical
characterization. Nat Toxins 4, 156-162.
Ishimoto N., Nemoto T., Nagayoshi K., Yamashita F. and Hashida M. (2006)
Improved anti-oxidant activity of superoxide dismutase by direct chemical
modification. J Control Release 111, 204-211.
Jain S., Hreczuk-Hirst D. H., McCormack B., Mital M., Epenetos A., Laing P. and
Gregoriadis G. (2003) Polysialylated insulin: synthesis, characterization and
biological activity in vivo. Biochim Biophys Acta 1622, 42-49.
Jones D. P., Carlson J. L., Mody V. C., Cai J., Lynn M. J. and Sternberg P. (2000)
Redox state of glutathione in human plasma. Free Radic Biol Med 28, 625-635.
Jones D. P., Carlson J. L., Samiec P. S., Sternberg P., Jr., Mody V. C., Jr., Reed R. L.
and Brown L. A. (1998) Glutathione measurement in human plasma. Evaluation of
sample collection, storage and derivatization conditions for analysis of dansyl
derivatives by HPLC. Clin Chim Acta 275, 175-184.
Kahns A. H., Buur A. and Bundgaard H. (1993) Prodrugs of peptides. 18. Synthesis
and evaluation of various esters of desmopressin (dDAVP). Pharm Res 10, 68-74.
Kaplan I. M., Wadia J. S. and Dowdy S. F. (2005) Cationic TAT peptide
transduction domain enters cells by macropinocytosis. J Control Release 102, 247-
253.
Kaufmann J. E. and Vischer U. M. (2003) Cellular mechanisms of the hemostatic
effects of desmopressin (DDAVP). J Thromb Haemost 1, 682-689.
Kim K. J., Matsukawa Y., Yamahara H., Kalra V. K., Lee V. H. and Crandall E. D.
(2003) Absorption of intact albumin across rat alveolar epithelial cell monolayers.
Am J Physiol Lung Cell Mol Physiol 284, L458-465.
Koch A. M., Reynolds F., Merkle H. P., Weissleder R. and Josephson L. (2005)
Transport of surface-modified nanoparticles through cell monolayers. Chembiochem
6, 337-345.
Kumar T. M. and Misra A. (2003) Influence of absorption promoters on pulmonary
insulin bioactivity. AAPS PharmSciTech 4, E15.
131
Kumar T. R., Soppimath K. and Nachaegari S. K. (2006) Novel delivery
technologies for protein and peptide therapeutics. Curr Pharm Biotechnol 7, 261-
276.
Kurtzhals P., Havelund S., Jonassen I., Kiehr B., Ribel U. and Markussen J. (1996)
Albumin binding and time action of acylated insulins in various species. J Pharm Sci
85, 304-308.
Laemmli U. K. (1970) Cleavage of structural proteins during the assembly of the
head of bacteriophage T4. Nature 227, 680-685.
Langs D. A., Smith G. D., Stezowski J. J. and Hughes R. E. (1986) Structure of
pressinoic acid: the cyclic moiety of vasopressin. Science 232, 1240-1242.
Lassmann-Vague V. and Raccah D. (2006) Alternatives routes of insulin delivery.
Diabetes Metab 32, 513-522.
Lee S., Kim K., Kumar T. S., Lee J., Kim S. K., Lee D. Y., Lee Y. K. and Byun Y.
(2005) Synthesis and biological properties of insulin-deoxycholic acid chemical
conjugates. Bioconjug Chem 16, 615-620.
Lee V. H. (1988) Enzymatic barriers to peptide and protein absorption. Crit Rev Ther
Drug Carrier Syst 5, 69-97.
Lenzer J. (2006) Inhaled insulin is approved in Europe and United States. Bmj 332,
321.
Liang J. F. and Yang V. C. (2005) Insulin-cell penetrating peptide hybrids with
improved intestinal absorption efficiency. Biochem Biophys Res Commun 335, 734-
738.
Lindgren M. E., Hallbrink M. M., Elmquist A. M. and Langel U. (2004) Passage of
cell-penetrating peptides across a human epithelial cell layer in vitro. Biochem J 377,
69-76.
Liu F. Y., Shao Z., Kildsig D. O. and Mitra A. K. (1993) Pulmonary delivery of free
and liposomal insulin. Pharm Res 10, 228-232.
Lundin S., Folkesson H. G., Pierzynovski S. G. and Bengtsson H. I. (1994)
Gastrointestinal absorption and plasma clearance rates of [D-Arg8]vasopressin
analogues in the rat. Peptides 15, 809-814.
132
Mae M. and Langel U. (2006) Cell-penetrating peptides as vectors for peptide,
protein and oligonucleotide delivery. Curr Opin Pharmacol 6, 509-514.
Mandel R., Ryser H. J., Niaki B., Ghani F. and Shen W. C. (1991) Isolation of
variants of Chinese hamster ovary cells with abnormally low levels of GSH:
decreased ability to cleave endocytosed disulfide bonds. J Cell Physiol 149, 60-65.
Mannucci P. M., Ruggeri Z. M., Pareti F. I. and Capitanio A. (1977) 1-Deamino-8-d-
arginine vasopressin: a new pharmacological approach to the management of
haemophilia and von Willebrands' diseases. Lancet 1, 869-872.
Matsukawa Y., Lee V. H., Crandall E. D. and Kim K. J. (1997) Size-dependent
dextran transport across rat alveolar epithelial cell monolayers. J Pharm Sci 86, 305-
309.
Matsukawa Y., Yamahara H., Lee V. H., Crandall E. D. and Kim K. J. (1996)
Horseradish peroxidase transport across rat alveolar epithelial cell monolayers.
Pharm Res 13, 1331-1335.
Matsukawa Y., Yamahara H., Yamashita F., Lee V. H., Crandall E. D. and Kim K. J.
(2000) Rates of protein transport across rat alveolar epithelial cell monolayers. J
Drug Target 7, 335-342.
Mei H., Yu C. and Chan K. K. (1999) NB1-C16-insulin: site-specific synthesis,
purification, and biological activity. Pharm Res 16, 1680-1686.
Meister A. and Anderson M. E. (1983) Glutathione. Annu Rev Biochem 52, 711-760.
Mitchell D. J., Kim D. T., Steinman L., Fathman C. G. and Rothbard J. B. (2000)
Polyarginine enters cells more efficiently than other polycationic homopolymers. J
Pept Res 56, 318-325.
Morimoto K., Yamahara H., Lee V. H. and Kim K. J. (1993) Dipeptide transport
across rat alveolar epithelial cell monolayers. Pharm Res 10, 1668-1674.
Morimoto K., Yamahara H., Lee V. H. and Kim K. J. (1994) Transport of
thyrotropin-releasing hormone across rat alveolar epithelial cell monolayers. Life Sci
54, 2083-2092.
Morishita M., Kamei N., Ehara J., Isowa K. and Takayama K. (2007) A novel
approach using functional peptides for efficient intestinal absorption of insulin. J
Control Release 118, 177-184.
133
Morre D. J., Chueh P. J., Lawler J. and Morre D. M. (1998) The sulfonylurea-
inhibited NADH oxidase activity of HeLa cell plasma membranes has properties of a
protein disulfide-thiol oxidoreductase with protein disulfide-thiol interchange
activity. J Bioenerg Biomembr 30, 477-487.
Morris M. C., Depollier J., Mery J., Heitz F. and Divita G. (2001) A peptide carrier
for the delivery of biologically active proteins into mammalian cells. Nat Biotechnol
19, 1173-1176.
Naithani V. K. and Gattner H. G. (1982) Preparation and properties of
citraconylinsulins. Hoppe Seylers Z Physiol Chem 363, 1443-1448.
Nakase I., Niwa M., Takeuchi T., Sonomura K., Kawabata N., Koike Y., Takehashi
M., Tanaka S., Ueda K., Simpson J. C., Jones A. T., Sugiura Y. and Futaki S. (2004)
Cellular uptake of arginine-rich peptides: roles for macropinocytosis and actin
rearrangement. Mol Ther 10, 1011-1022.
Niculescu-Duvaz I. (2000) Technology evaluation: gemtuzumab ozogamicin,
Celltech Group. Curr Opin Mol Ther 2, 691-696.
Ohtake K., Natsume H., Ueda H. and Morimoto Y. (2002) Analysis of transient and
reversible effects of poly-L-arginine on the in vivo nasal absorption of FITC-dextran
in rats. J Control Release 82, 263-275.
Peterson M. W. and Gruenhaupt D. (1990) Protamine increases the permeability of
cultured epithelial monolayers. J Appl Physiol 68, 220-227.
Raymond M., Gros P., Whiteway M. and Thomas D. Y. (1992) Functional
complementation of yeast ste6 by a mammalian multidrug resistance mdr gene.
Science 256, 232-234.
Roth R. A., Maddux B. A., Wong K. Y., Iwamoto Y. and Goldfine I. D. (1981)
Insulin-ricin B chain conjugate. A hybrid molecule with ricin-binding activity and
insulin biological activity. J Biol Chem 256, 5350-5354.
Rothbard J. B., Garlington S., Lin Q., Kirschberg T., Kreider E., McGrane P. L.,
Wender P. A. and Khavari P. A. (2000) Conjugation of arginine oligomers to
cyclosporin A facilitates topical delivery and inhibition of inflammation. Nat Med 6,
1253-1257.
Ryser H. J., Mandel R. and Ghani F. (1991) Cell surface sulfhydryls are required for
the cytotoxicity of diphtheria toxin but not of ricin in Chinese hamster ovary cells. J
Biol Chem 266, 18439-18442.
134
Sahaf B., Soderberg A., Spyrou G., Barral A. M., Pekkari K., Holmgren A. and
Rosen A. (1997) Thioredoxin expression and localization in human cell lines:
detection of full-length and truncated species. Exp Cell Res 236, 181-192.
Sai Y., Kajita M., Tamai I., Wakama J., Wakamiya T. and Tsuji A. (1998)
Adsorptive-mediated transcytosis of a synthetic basic peptide, 001-C8 in Caco-2
cells. Pharm Res 15, 1305-1309.
Saito G., Swanson J. A. and Lee K. D. (2003) Drug delivery strategy utilizing
conjugation via reversible disulfide linkages: role and site of cellular reducing
activities. Adv Drug Deliv Rev 55, 199-215.
Sakagami M. (2006) In vivo, in vitro and ex vivo models to assess pulmonary
absorption and disposition of inhaled therapeutics for systemic delivery. Adv Drug
Deliv Rev 58, 1030-1060.
Sandgren S., Cheng F. and Belting M. (2002) Nuclear targeting of macromolecular
polyanions by an HIV-Tat derived peptide. Role for cell-surface proteoglycans. J
Biol Chem 277, 38877-38883.
Seglen P. O., Grinde B. and Solheim A. E. (1979) Inhibition of the lysosomal
pathway of protein degradation in isolated rat hepatocytes by ammonia,
methylamine, chloroquine and leupeptin. Eur J Biochem 95, 215-225.
Shah D. and Shen W. C. (1996) Transcellular delivery of an insulin-transferrin
conjugate in enterocyte-like Caco-2 cells. J Pharm Sci 85, 1306-1311.
Shen W. C. (2003) Oral peptide and protein delivery: unfulfilled promises? Drug
Discov Today 8, 607-608.
Sonoda S. and Schlamowitz M. (1970) Studies of 125I trace labeling of
immunoglobulin G by chloramine-T. Immunochemistry 7, 885-898.
Suzuki T., Futaki S., Niwa M., Tanaka S., Ueda K. and Sugiura Y. (2002) Possible
existence of common internalization mechanisms among arginine-rich peptides. J
Biol Chem 277, 2437-2443.
Taub M. E., Wan J. and Shen W. C. (1994) Transepithelial transport of tyramine
across filter-grown MDCK cells via a poly(D-lysine) carrier. Pharm Res 11, 1250-
1256.
Tauzin B. (2006) Medicines in development, pp 1-52. PhRMA.
135
Thibaudeau K., Leger R., Huang X., Robitaille M., Quraishi O., Soucy C., Bousquet-
Gagnon N., van Wyk P., Paradis V., Castaigne J. P. and Bridon D. (2005) Synthesis
and evaluation of insulin-human serum albumin conjugates. Bioconjug Chem 16,
1000-1008.
Thwaites D. T., Hirst B. H. and Simmons N. L. (1994) Substrate specificity of the
di/tripeptide transporter in human intestinal epithelia (Caco-2): identification of
substrates that undergo H(+)-coupled absorption. Br J Pharmacol 113, 1050-1056.
Trehin R., Krauss U., Beck-Sickinger A. G., Merkle H. P. and Nielsen H. M. (2004)
Cellular uptake but low permeation of human calcitonin-derived cell penetrating
peptides and Tat(47-57) through well-differentiated epithelial models. Pharm Res 21,
1248-1256.
Trehin R., Nielsen H. M., Jahnke H. G., Krauss U., Beck-Sickinger A. G. and
Merkle H. P. (2004) Metabolic cleavage of cell-penetrating peptides in contact with
epithelial models: human calcitonin (hCT)-derived peptides, Tat(47-57) and
penetratin(43-58). Biochem J 382, 945-956.
Trinh-Trang-Tan M. M., Sokol H. W., Bankir L. and Valtin H. (1982) Homozygous
Brattleboro rats lack normal nephron heterogeneity as a consequence of their urine
concentrating defect. Ann N Y Acad Sci 394, 524-528.
Uchio T., Baudys M., Liu F., Song S. C. and Kim S. W. (1999) Site-specific insulin
conjugates with enhanced stability and extended action profile. Adv Drug Deliv Rev
35, 289-306.
Vaccaro C. A. and Brody J. S. (1981) Structural features of alveolar wall basement
membrane in the adult rat lung. J Cell Biol 91, 427-437.
Vavra I., Machova A., Holecek V., Cort J. H., Zaoral M. and Sorm F. (1968) Effect
of a synthetic analogue of vasopressin in animals and in patients with diabetes
insipidus. Lancet 1, 948-952.
Violini S., Sharma V., Prior J. L., Dyszlewski M. and Piwnica-Worms D. (2002)
Evidence for a plasma membrane-mediated permeability barrier to Tat basic domain
in well-differentiated epithelial cells: lack of correlation with heparan sulfate.
Biochemistry 41, 12652-12661.
Vives E., Brodin P. and Lebleu B. (1997) A truncated HIV-1 Tat protein basic
domain rapidly translocates through the plasma membrane and accumulates in the
cell nucleus. J Biol Chem 272, 16010-16017.
136
Wang J., Shen D. and Shen W. C. (1999) Preparation, purification, and
characterization of a reversibly lipidized desmopressin with potentiated anti-diuretic
activity. Pharm Res 16, 1674-1679.
Wang J., Zhao G., Shen W., Ochoa M., Moore D., Hubbard J. W. and Hintze T. H.
(1993) Effects of an orally active NO-releasing agent, CAS 936, and its active
metabolite, 3754, on cardiac and coronary dynamics in normal conscious dogs and
after pacing-induced heart failure. J Cardiovasc Pharmacol 22 Suppl 7, S51-58.
Watson P. and Spooner R. A. (2006) Toxin entry and trafficking in mammalian cells.
Adv Drug Deliv Rev 58, 1581-1596.
Wender P. A., Mitchell D. J., Pattabiraman K., Pelkey E. T., Steinman L. and
Rothbard J. B. (2000) The design, synthesis, and evaluation of molecules that enable
or enhance cellular uptake: peptoid molecular transporters. Proc Natl Acad Sci U S A
97, 13003-13008.
Williams D. C., Van Frank R. M., Muth W. L. and Burnett J. P. (1982) Cytoplasmic
inclusion bodies in Escherichia coli producing biosynthetic human insulin proteins.
Science 215, 687-689.
Williams M. C. (1984) Endocytosis in alveolar type II cells: effect of charge and size
of tracers. Proc Natl Acad Sci U S A 81, 6054-6058.
Yamahara H., Morimoto K., Lee V. H. and Kim K. J. (1994) Effects of protease
inhibitors on vasopressin transport across rat alveolar epithelial cell monolayers.
Pharm Res 11, 1617-1622.
Yu L. X., Amidon G. L., Polli J. E., Zhao H., Mehta M. U., Conner D. P., Shah V.
P., Lesko L. J., Chen M. L., Lee V. H. and Hussain A. S. (2002) Biopharmaceutics
classification system: the scientific basis for biowaiver extensions. Pharm Res 19,
921-925.
Zakova L., Barth T., Jiracek J., Barthova J. and Zorad S. (2004) Shortened insulin
analogues: marked changes in biological activity resulting from replacement of
TyrB26 and N-methylation of peptide bonds in the C-terminus of the B-chain.
Biochemistry 43, 2323-2331.
Zaro J. L. and Shen W. C. (2003) Quantitative comparison of membrane
transduction and endocytosis of oligopeptides. Biochem Biophys Res Commun 307,
241-247.
Zaro J. L. and Shen W. C. (2005) Evidence that membrane transduction of
oligoarginine does not require vesicle formation. Exp Cell Res 307, 164-173.
137
Zaro J. L. and Shen W. C. (2005) Cytosolic delivery of a p16-peptide oligoarginine
conjugate for inhibiting proliferation of MCF7 cells. J Control Release 108, 409-
417.
Zhang X., Wan L., Pooyan S., Su Y., Gardner C. R., Leibowitz M. J., Stein S. and
Sinko P. J. (2004) Quantitative assessment of the cell penetrating properties of RI-
Tat-9: evidence for a cell type-specific barrier at the plasma membrane of epithelial
cells. Mol Pharm 1, 145-155.
Zhang Y. Q., Ma Y., Xia Y. Y., Shen W. D., Mao J. P. and Xue R. Y. (2006) Silk
sericin-insulin bioconjugates: synthesis, characterization and biological activity. J
Control Release 115, 307-315.
Abstract (if available)
Abstract
The recent interest in cationic cell penetrating peptide (CPPs) stems from their potential utility as non-invasive drug delivery platforms. These peptides are short sequences consisting mostly of arginine or lysine residues including the Tat peptide, Antp peptide and different oligomers of arginine and lysine. By conjugating CPPs to a wide range of impermeable cargo, their intracellular delivery has shown to be increased in different types of cells by a less defined mechanism of transduction. Despite their versatility, few studies have focused on their application for transcellular delivery. Considering that the original protein from which the cationic CPPs are derived, HIV-1 Tat protein, is able to exit the infected cell and internalize into neighbouring cells, it is conceivable that CPPs can enter into differentiated epithelial cells from the apical domain and exit on the basolateral domain. If this is true, they can also ferry impermeable cargo across the epithelial cells, thus aid in overcoming one of the notorious barriers for oral and pulmonary delivery of proteins and peptides. This hypothesis was tested by investigating the transport of CPPs in alveolar epithelial cells and also by conjugating CPPs to a peptide and a protein drug intended for oral and pulmonary delivery respectively. The results presented in this thesis show that CPPs can be transported across alveolar epithelial cells without severely compromising cellular integrity. Additionally, oligoarginine conjugated to protein drug, e.g., insulin, can facilitate the transport of insulin in vitro across alveolar epithelial cells by a transcellular pathway and can enhance insulin transport in vivo when delivered into the lungs of diabetic rats. However, conjugating oligoarginine to a peptide drug, desmopressin, improved in vitro transport marginally but did not have any beneficial effect when delivered orally.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Cell penetrating peptide-based drug delivery system for targeting mildly acidic pH
PDF
Characterization and identification of endogenous factor(s) that enhance insulin transport across primary rat alveolar epithelial cell monolayers
PDF
Cell penetrating peptide-based polyplexes for sirRNA delivery
PDF
Nuclear localization of cell-penetrating peptides
PDF
Comparison of cellular uptake of arginine-rich cell-penetrating peptides
PDF
pH-sensitive cytotoxicity of a cell penetrating peptide fused with a histidine-glutamate co-oligopeptide
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
Characterization of the transferrin oligomer and its potential application in drug delivery
PDF
Reversible lipidization of peptide and protein drugs for prolonging the plasma half-lives and passive liver targeting
Asset Metadata
Creator
Patel, Leena
(author)
Core Title
Cationic cell penetrating peptides: characterization of transport properties in epithelial cells and their utilization as delivery systems for protein and peptide drugs
School
School of Pharmacy
Degree
Doctor of Philosophy
Degree Program
Pharmaceutical Sciences
Publication Date
07/27/2009
Defense Date
05/16/2007
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
cell penetrating peptides,OAI-PMH Harvest,oligoarginine,protein and peptide drug delivery,protein transduction domains
Language
English
Advisor
Shen, Wei-Chiang (
committee chair
), Kim, Kwang-Jin (
committee member
), Mircheff, Austin K. (
committee member
), Okamoto, Curtis Toshio (
committee member
)
Creator Email
leenapat@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-m686
Unique identifier
UC1224372
Identifier
etd-Patel-20070727 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-523375 (legacy record id),usctheses-m686 (legacy record id)
Legacy Identifier
etd-Patel-20070727.pdf
Dmrecord
523375
Document Type
Dissertation
Rights
Patel, Leena
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Repository Name
Libraries, University of Southern California
Repository Location
Los Angeles, California
Repository Email
cisadmin@lib.usc.edu
Tags
cell penetrating peptides
oligoarginine
protein and peptide drug delivery
protein transduction domains