Share...
13 Apr 2014 03:09
http://dspace.wul.waseda.ac.jp/dspace/bitstream/2065/34622/3/Honbun-5042.pdf のHTMLバージョンです。
G o o g l eではファイルを自動的にHTMLに変換して保存しています。
Efficient Intracellular Delivery of Biomacromolecules by Liposomes Containing Amino-acid Based Lipids
Page 1
Efficient Intracellular Delivery of Biomacromolecules
by Liposomes Containing Amino-acid Based Lipids
アミノ酸型脂質を用いたリポソームによる
生体高分子の高効率な細胞内運搬
February 2009
早稲田大学大学院 理工学研究科
応用化学専攻 高分子化学研究
Yosuke OBATA
小幡 洋輔

Page 2
Prof. Dr. Shinji Takeoka
Prof. Dr. Hiroyuki Nishide
Prof. Dr. Kiyotaka Sakai
Assoc. Prof. Dr. Arianna Menciassi
Promoter :
Referees :

Page 3
i
Contents
Preface
Chapter 1: Molecular Self-Assembly and Physicochemical Properties of
Liposomes for Drug Delivery System
1. Introduction
1
2. Molecular Self-assembly Controlled by Intermolecular Interactions
2-1. Definition of Molecular Self-Assembly
2
2-2. Self-Assembly in Biological System
3
2-3. Intermolecular Interactions Dominating Self-Assembly
4
2-4. Thermodynamics of Molecular Assembly
7
3. Physicochemical Properties of Lipid Bilayer Vesicles
3-1. Chemical Structure of Amphiphiles
10
3-2. Phase Transition Behavior of Lipid Bilayer Membrane
12
3-3. Membrane Permeability
15
4. Drug Delivery Focused on Carriers Using Molecular Assembly
4-1. Drug Delivery System
16
4-2. Drug Carriers for Improving Drug Delivery
17
4-3. Gene Therapy
20
References
22
Chapter 2: Syntheses of Amino-acid Based Lipids Having a Cationic Head
Group and Evaluation of Their Properties for Gene Delivery
1. Introduction
25
2. Cationic Liposome Mediated Transfection
2-1. Cationic Liposomes for Cellular Transfection
26
2-2. Conventional Cationic Lipids for Transfection
27
3. Cationic Assemblies Containing Synthetic Amino-acid Based Lipids
3-1. Syntheses of Cationic Amino-acid Based Lipids
28
3-2. Morphological Study of Cationic Assemblies
29
3-3. Physicochemical Properties of Assemblies
31

Page 4
ii
3-4. Synthesis of Cationic Lipids Having a Hydrocarbon Spacer between
Head Group and Hydrophobic Moieties
32
4. Experimental Section
4-1. Syntheses of Amino-acid Based Cationic Lipids
33
4-2. General Methods
37
4-3. Synthesis of Cationic Lipids Having a Spacer between Cationic
Head Group and Hydrophobic Moieties
38
References
42
Chapter 3: Correlation between Structure of Amino-acid Base Lipid and
Transfection Efficiency in Vitro for Efficient Plasmid DNA
Delivery
1. Introduction
43
2. Barriers to Intracellular Trafficking in Gene Delivery
44
3. Evaluation of Cationic Assemblies Composed of Amino-acid Based Lipids
for Plasmid DNA Delivery in COS-7 cells
3-1. Lipoplex Formation
46
3-2. Influence of a Lipid-to-pDNA ratio on Dispersion State
48
3-3. Transfection Efficiency of Amino-acid Based Lipids
48
3-4. Fusogenic Potential of Cationic Assemblies
50
3-5. Influence of Serum Proteins on Gene Expression Efficiency
52
3-6. Cytotoxicity of Amino-acid Based Lipids
53
4. Neuronal Transfection with Amino-acid Based Lipids
4-1. Neuronal Transfection
54
4-2. Transfection Efficiency on Neuronal Cell Lines
55
5. Experiment Section
58
References
62
Chapter 4: Construction of Carriers for Introducing Biomacromolecules into
Cells by Cationic Amino-acid Based Lipids
1. Introduction
65
2. Indrodiction of Functional Biomacromolecules into Living Cells
2-1. Signification for Introduction of Biomacromolecules into Cells
66

Page 5
iii
2-2. A Human Homologous Recombination Protein; DMC1
66
2-3. Dmc1 Introduction into Living Cells
67
3. Construction of DMC1-Encapsulating Cationic Liposomes
3-1. Preparation of DMC1-Encapsulating Cationic Liposomes
68
3-2. Protease Digestion Assay
69
3-3. Intracellular DMC1 Transition by Cationic Liposomes in COS-1 cells 70
3-4. Western Blotting Analysis
73
3-5. Conclusion of DMC1-Encapsulating Liposomes
75
4. New Methodology of Gene Delivery Using Carbon Nanotubes
4-1. Biomedical Application of Carbon Nanotube with Cationic Amino-acid
Based Lipids
75
4-2. Lipid Wrapped MWNT
76
4-3. Complexation of L-MWNT with DNA
78
4-4. Gene Expression Efficiency of L-MWNT/pDNA complexes
80
4-5. Gene Expression of L-MWNT/pDNA Complexes under Magnetic Fields 81
4-6. Cytotoxicity of L-MWNT
83
4-7. Conclusion of Water Soluble L-MWNTs for Plasmid DNA Delivery
84
5. Experimental section
5-1. General Methods on DMC1-encapsulating Liposomes
85
5-2. General Methods on Lipidic Wrapped Carbon Nanotubes
88
Reference
91
Chapter 5: Charge Convertible Amphiphiles for Constructing pH-Sensitive
Liposomes
1. Introduction
95
2. Intracellular Drug Delivery Using pH-sensitive Liposomes
2-1. Intracellular Transition of Liposomes
96
2-2. Conventional pH-Sensitive Liposomes
97
3. Synthesis of Zwitterionic Lipids as Charge Convertible Component
3-1. Charge Convertible Liposomes for Improving Intracellular Drug Delivery 99
3-2. Zwitterionic Lipids in Response to Endosomal pH
100
4. Experimental Section
102
References
105

Page 6
iv
Chapter 6: Evaluation of Charge Convertible Liposomes Containing Synthetic
Zwitterionic Lipids for Efficient Systemic Drug Delivery
1. Introduction
107
2. Preparation of Doxorubicin-encapsulating Liposomes
2-1. Encapsulation of Small Molecular Weight Drugs
108
2-2. Syntheses of Charge Convertible Lipids
109
2-3. Zeta Potential Conversion of Charge Convertible Liposomes
110
2-4. Evaluation of Fusogenic Potential of Charge Convertible Liposomes
112
2-5. Characterization of DOX-encapsulating Liposomes
114
3. Evaluation of DOX-encapsulating Liposomes Containing Zwitterionic Lipids
3-1. Cellular Uptake Efficiency of Liposomes
116
3-2. Evaluation of DOX-Encapsulating Liposomes by Confocal Laser
Microscopy
117
3-3. In Vitro Pharmaceutical Activity of Liposomes
119
3-4. Blood Persistence and Biodistribution of Charge Convertible Liposomes 122
3-5. Conclusion
123
4. Experimental Section
6-1. Synthesis of Zwitterionic Lipids
124
6-2. General Methods
124
References
129
Chapter 7: Development of Injectable gene Carrier: Charge Convertible
Liposomes Encapsulated Plasmid DNA
1. Introduction
131
2. pDNA-Encapsulating Liposomes Containing Zwitterionic Lipids
2-1. Preparation of PLL-pDNA Complexes
132
2-2. Preparation of pDNA-encapsulating Liposomes
133
2-3. Zeta Potential of Liposomes at Various pHs
135
2-4. PLL/pDNA Release from Liposomes at Various pHs
137
3. Evaluation of pDNA-encapsulating Liposomes in vitro
3-1. Cellular Uptake Efficiency
138
3-2. Gene Expression Efficiency of pDNA-encapsulating Liposomes
139
3-3. Intracellular Gene Delivery of pDNA by Charge Convertible Liposomes 140
4. Experimental section
141

Page 7
v
References
145
Chapter 8: Future Prospects
1. Introduction
147
2. Development of Charge Convertible Liposomes in Response to Extracellular pH of
Solid Tumor
2-1. Tumor pH Targeting Nanotechnology
148
2-2. Construction of Gene or Drug Carrier in Response to Tumor pH
149
References
152
List of Achievements
Acknowledgement

Page 8
vi

Page 9
Preface
Recently, numerous biotechnologies have been overcome and clarified genome
information of conventional disease due to development of gene diagnosis. The therapies
to diseases are then shifted from drug therapy using small molecule drugs to gene therapy
using specific genome sequence. Beside the conventional disease, gene therapy expects
providing new treatment to inherited disorders, which had no treatment so far. Furthermore,
gene therapy offers personalized medication with concerning gender, species, life style,
character, and so on. Therefore, gene therapy is more and more attractive approach today.
The use of genes as therapeutic agents has attracted attention as a novel approach to
the treatment of both acquired and inherited diseases. In the case of inherited disorders, the
introduction of a normal copy of the affected gene can be effective as shown in the well
known case of gene therapy for severe combined immune deficiency due to adenosine
deaminase (ADA) deficiency, in which the normal gene for ADA is used to treat the
affected patient. For the treatment of acquired disorders, such as cancer and infectious
diseases, effective potential strategies involve not only the introduction of a therapeutic
genes for a cytokine or an antigen, but also the silencing of the expression of an abnormal
gene, whose expression is enhanced in the diseased tissue.
Gene delivery strategies for treatment fall into two categories: Viral and non-viral
vector. More than 70% of current gene therapies are performed using the viral vector.
However, these treatments are plagued by serious problems such as toxicity,
immunogenicity. Therefore, future practical applications of genes to medicine will keenly
require the development of safe non-viral gene delivery system.
For these backgrounds, the author construct functional liposomes composed of
amino-acid based lipids for efficient gene and drug delivery. First interest is enhancement
of transfection efficiency in vitro. The correlation between lipid structure and transfection
efficiency has clarified to enhance gene delivery on immobilized cell line by using cationic
liposomes, resulting in constructing liposome-based non-viral carriers with high
transferring capacity. The late interest is improvement of intracellular drug and gene
delivery using charge convertible liposomes. The charge convertible liposomes offer
systemic drug and gene delivery as well as improvement of intracellular delivery.
The author convinces that functional liposomes composed of the amino-acid based
lipids will necessary contribute to development of drug and gene carrier for constructing
practical drug and gene delivery system.
Yosuke Obata

Page 10

Page 11
Chapter 1
- 1 -
Chapter 1
Molecular Self-Assembly and Physicochemical Properties of
Liposomes for Drug Delivery System
1. Introduction
The biological world is rich with ordered assemblies of molecules. Indeed, the
assembly and function of supramolecular structures resulted from self-assembly or
self-organization is central to modern biology. The forces holding together these
assemblies are diverse: van der Waals, electrostatic, hydrophobic interactions, and
hydrogen bonds, all contribute to specific recognition between members of the assembly.
In this chapter, structures and techniques that can be used to fabricate supermolecules
created from huge numbers of component molecules are introduced. Furthermore,
programmed self-assemblies such as liposomes, micelles, and polymer-based particles are
also explained for construction of efficient drug delivery.

Page 12
Chapter 1
- 2 -
2. Molecular Self-Assembly Controlled by Intermolecular Interactions
2-1. Definition of Molecular Self-Assembly
Nanostructures are assemblies of bonded atoms
that have dimensions in the range of 1 to 102
nanometers1. Structures in this range of sizes can be
considered as exceptionally large, unexceptional, or
exceptionally small, depending on one's viewpoint,
synthetic and analytical technologies, and interests (Fig.
1-1)2. To solid-state physicists, materials scientists, and
electrical engineers, nanostructures are small. The
techniques, such as microlithography and deposition
from the vapor, that are used in these fields to fabricate
microstructures and devices require increasingly
substantial effort as they are extended to the range below 102 nm. To biologists,
nanostructures are familiar objects. A range of biological structures-from proteins through
viruses to cellular organelles-have dimensions of 1 to 102 nm. To chemists, nanostructures
are large. Considered as molecules, nanostructures require the assembly of groups of atoms
numbering from 103 to 109 and having molecular weights of 104 to 1010 Da. Synthetic
techniques that generate well-defined structures at the lower ends of these ranges are only
now being developed and the upper ends remain largely unexplored. Developing
techniques for synthesizing and characterizing ultra large molecules and molecular
assemblies-nanostructures-is one of the grand challenges now facing chemistry.
Nanostructures provide major unsolved problems in complexity and require new strategies
and technologies for their synthesis and characterization. The solutions to these problems
would be both interesting in themselves and essential elements in extending chemistry
Fig. 1-1 Comparison of the
relative sizes of structures
generated in biology.

Page 13
Chapter 1
- 3 -
toward problems in materials science and biology.
2-2. Self-Assembly in Biological System
The highly sophisticated functions seen in biological systems originate from their
well-designed molecular arrangements, which are formed through self-assembly and
self-organization (Fig.1-2). The characteristics of biological supermolecules mean that
they provide good targets to aim for and good design rules to use when designing artificial
functional systems. In this section, several examples of supermolecules in biology are
introducded. The well constructed example of a biological supermolecule is a cellular
membrane. Cellular membrane consists mainly of a fluidic lipid bilayer containing
proteins (Fig. 1-2). The lipids are self-assembled into the bilayer structure and the proteins
float within the lipid bilayer. The whole structure is formed through self-assembly
processes. The membrane protein is stably buried in the lipid bilayer due to the
amphiphilic nature of the membrane protein. The surfaces of some parts of the protein
have mainly hydrophobic amino-acid residues, and hydrophilic residues are located on the
other surfaces. The former parts are accommodated in the hydrophobic lipid bilayer and
the latter protein regions are exposed to the surface of the water. The major driving force
for lipid bilayer formation is hydrophobic interaction. This interaction is much less specific
and less directional than the hydrogen bonding and metal coordination interactions that are
Fig. 1-2 Structure of cellular membrane as a supramolecules

Page 14
Chapter 1
- 4 -
used in precisely programmed supramolecular assemblies. As shown in cellular membrane,
the intermolecular interactions are existed to construct molecular assembly.
2-3. Intermolecular Interactions Dominating Self-Assembly
In previous section, molecular assemblies in biological field are introduced. Then,
interactions dominating molecular self-assembly are the interaction that an electric dipole
to occur by heterogeneous distribution of the electronic density participates. Therefore, the
interaction between the molecules is basically equilibrium reaction. Furthermore, forces of
intermolecular interaction are considerably weaker than that of covalent bond, resulting in
repeating dissociation and binding. In this section, the force holding self-assembly are
introduced, and compared the intermolecular interaction.
Noncovalent interactions can generally divide into four forces: (i) ionic interactions,
(ii) hydrogen bonding, (iii) van der Waals force, and (iv) hydrophobic interaction (Fig. 1-3).
Among the intermolecular interaction, the strong interaction is the ionic interaction, which
is comparable with covalent interactions in vacuo (Table 1). However, the covalent
interaction in water shows range from 90 kcal/mol for single bond in relation to the
non-covalent interaction for providing 3 kcal/mol ion-ion interaction.

Page 15
Chapter 1
- 5 -
Ionic bonding
The electric dipoles of water molecules arrange in
the ion interval with decreasing electrostatic affinity
between the polarity bases.
Covalent bonding
(ca. 0.1 nm)
Hydrogen bonding
(ca. 0.2 nm)
Distance between hydrogen atom and
oxygen atom of hydrogen bonding is longer
than that of covalent bonding.
Hydrogen bonding in biological system
Hydrogen bonding
repuls
ion
attrac
tion
En
erg
y
van der Waals contact distance
distance between
centers of atoms
At very short distances any two atoms show a weak
bonding interaction due to their fluctuating electrical
charges. If the two atoms are too close together,
however, they repel each other very strongly.
van der Waals forces
Hydrophobic interaction
Hydrophobic group
Hydrophilic group
Hydrophobic interaction arise from the exclusion of
non-polar groups or molecules from aqueous
solution. This situation is more energetically
favorable because water molecules interact with
themselves or with other polar groups or molecules
preferentially.
Fig. 1-3 Intermolecular interaction holding molecular self-assembly.

Page 16
Chapter 1
- 6 -
Ion-ion interactions are non-directional in nature, meaning that the interaction can
occur in any orientation. Ion-dipole and dipole-dipole interactions, however, have
orientation-dependent aspects requiring two entities to be aligned such that the interactions
are in the optimal direction. Due to the relative rigidity of directional interactions, only
mutually complementary species are able to from aggregates, whereas non-directional
interactions can stabilize a wide range of molecular pairings. The strength of these
directional interactions depends upon the species involved. Ion-dipole interactions are
stronger than dipole-dipole interactions (50-200 and 5-50 kJ/mol, respectively) as ions
have a higher charge density than dipoles. Despite being the weakest directional interaction,
dipole-dipole interactions are useful for bringing species into alignment, as the interaction
requires a specific orientation of both entities. Electrostatic interactions play an important
role in understanding the factors that influence high binding affinities, particularly in
biological systems in which there is a large number of recognition processes that involve
charge-charge interactions; indeed these are often the first interactions between a substrate
and an enzyme.
The hydrogen bonding is also important non-covalent interaction due to its strength
and high degree of directionality. It represents a special kind of dipole-dipole interaction
between a proton donor and a proton acceptor. There are a number of naturally occurring
‘building block’ that are a rich source of hydrogen bond donors and acceptors (e.g. amino
acids, carbohydrates).
Van der Waals interactions are dispersion effects that comprise two components,
namely the London interaction and the exchange and repulsion interaction. This
interaction arise from fluctuations of the electron distribution between species that are in
close proximity to one another. As the electron cloud moves about a molecule’s momentary
location, an instantaneous dipole is between two adjacent species will align the molecules

Page 17
Chapter 1
- 7 -
such that a partial positive charge from one species will be attached to a partial negative
charge form another molecules.
Hydrophobic effects arise from the exclusion of non-polar groups or molecules
from aqueous solution. This situation is more energetically favorable because water
molecules interact with themselves or with other polar groups or molecules preferentially.
This phenomenon can be observed between dichloromethane and water which are
immiscible. The organic solvent is forced away as the intersolvent interactions between the
water molecules themselves are more favourable an important role in some supramolecular
chemistry, for example, the biding of organic molecules by cyclophanes and cyclodextrins
in water. Hydrophobic effects can be split into two energetic components, namely an
enthalpic hydrophobic effect and an entropic hydrophobic effect. The hydrophobic effect is
also very important in biological systems in the creation and maintenance of the
macromolecular structure and supramolecular assemblies of the living cell.
2.4. Thermodynamics of Molecular Assembly
All interactions in living systems occur in aqueous solution due to 50-60% of the
human body containing water. Under this condition, hydrophobic interaction plays
important roles in stabilization of the three dimensional conformation of protein, an
bonds
Covalent bond
Ionic bond
Hydrogen bond
Van der Waals attraction
(per atom)
Length [nm]
0.15
0.25
0.30
0.35
Strength [kcal/mol]
90
80
4
0.1
90
3
1
0.1
in vacuo
in water
Hydrophobic interaction
0.02
Table 1-1. Comparison between energy in covalent bond
and non-covalent bond

Page 18
Chapter 1
- 8 -
organization of the cells and tissues, and a formation of the various complex. A similar
phenomenon can be seen an amphiphile, which has both headgroup and hydrophobic
moieties in a molecule spontaneously assembles in water by hydrophobic interaction.
These phenomenon can be understood as the transfer to the thermodynamically stable state.
There are several literatures on the thermodynamics of self-assemble. Equilibrium
thermodynamics requires that in a system of molecules that form aggregated structures in
solution, the chemical potential of all identical molecules in different aggregates be the
same. This might be expressed as following equation:
μ = μ1
0 + kTlogX1 = μ2
0 + 1/2 kT logX2 = μ3
0 + 1/3 kT logX3 = ····· (1-1)
or
μ = μN = μN
0 + (kT/N)log(XN/N) = constant, N=1,2,3 ····· (1-2)
where μN is the chemical potential of a molecule in an aggregate (aggregation number N,
the same for all N), μN
0 is the standard part of the chemical potential (the mean interaction
free energy) per molecule in aggregates of aggregation number N, and XN is the
concentration (more strictly the activity) of molecules incorporated in aggregates of
number N (N=1, μ1
0 and X1 represented to isolated molecules, monomers in solution,
respectively). Equation (1-2) may also be derived using the familiar law of mass action.
Fig. 1-4 Association of N monomers into an aggregation (e.g. micelle). The
mean life time of an amphiphile molecular in a micelle is very short, 10-5
10-3 s-1.

Page 19
Chapter 1
- 9 -
Thus, reffering to Fig. 1-4,
rate of association = k1X1
N,
rate of dissociation = kN(XN/N),
where
k1/kN = exp[-N(μN
0 – μ1
0)/kT]
(1-3)
is the ratio of the two reaction rates (equilibrium constant). These combine to give equation
(1-2), which can also be expressed in the more useful forms:
XN = N {X1exp[(μ1
0 – μN
0)/kT]}N
(1-4)
and
XN = N {(XM/M)exp[MM
0 – μN
0)/kT]}N/M
(1-5)
where M is any arbitrary reference state of aggregates (or monomers) with aggregation
number M (or1). Equation (1-4) together with the conservation relating total solute
concentration C as
C = X1 + X2 + X3 + ····· = ΣXN (N = 1, 2, 3, ·····) (1-6)
Completely defines the system. Depending on how the free energies μ1
0, μN
0 are defined
the dimentionless concentrations C and XN can be expressed as volume fraction or mole
fraction. In particular, note that C and XN can never exceed unity. Equation (1-4) assumes
ideal mixing. Little more can be said about aggregates dispersion without specifying the
form and magnitude of mN0 as a function of N.
μN
0 = μ0
+ αkT/N
P
(1-7)

Page 20
Chapter 1
- 10 -
3. Physicochemical Properties of Lipid Bilayer Vesicles
3-1. Chemical Structure of Amphiphiles
A amphiphile consists of a hydrophilic head and a hydrophobic tail. The head has a
high affinity to water, while contact of the tail with water is energetically unfavorable and
this part is preferably solvated by nonpolar solvents. Molecules that have affinities for both
hydrophilic and hydrophobic media are called amphiphiles. Lipids are typical amphiphiles,
and many kinds of artificial amphiphiles are reported to form membrane-like structures in
aqueous solution based on their structure (Fig. 1-5). The former is based on the assumption
that, for any amphiphile molecule, the effects of specific interactions such as hydration,
hydrogen bonding, charge interactions, and van der Waals forces, etc., and relevant steric
contributions from the headgroup and hydrocarbon chains, can be indirectly accounted for
in terms of an optimal polar headgroup cross-sectional area (a0), a critical chain length (lc),
and hydrocarbon chain volume (v). Further, it was proposed than these geometric
properties of the lipid molecule define a so-called critical packing parameter (v/a0lc), from
which one can make generalized predictions about the ‘‘preferred shappes’’ of lipid
molecules and evaluate whether a particular aggregate structure is compatible with those
geometric properties 3,4. On the basis of suck considerations, it was determined that lipids
with critical packing parameter values of less than 0.5 are effectively cone-shaped and
should form micellar aggregates; those with packing parameter values between 0.5 and 1
are roughly cylindrical in shape and should form lamellar assemblies; and those with
packing parameter values greater than 1 are inverted cone-shaped and should form inverted
(typeII) phases. Despite the simplicity of this concept, it has proven to be useful for the
rationalization of many aspects of the nonlamellar phase behavior of many membrane
lipids. However, because of its inherently imprecise simplifying assumptions, this model
rends to reliably support only broad generalizations about the nonolamellar forming

Page 21
Chapter 1
- 11 -
tendencies of lipid molecles. In part this can be attributed to the fact that the model does
not adequately consider the headgroup repulsion and chain repulsion contributions to the
interaction free energy 5. Also, at temperatures above Tm, a lipid molecule is very flexible
and does not have a defined shape per se. Indeed, under suck conditions its ‘‘shape’’ is
better defined as that of the average volume it occupies while minimizing the packing free
energy and, to a large extent, this ‘‘shape’’ is determined by the phase state of the lipid
aggregate and not the other way around 6.
Fig. 1-5 Aggregate structures compatible with the varied packing properties of lipids and
other amphiphiles.
< 1/3
1/3 – 1/2
1/2 – 1
~ 1
> 1
Single –chained lipids
(detergents) with large
Head-group areas:
NaDS in low salt
Some lysophospholipids
Single-chained lipids with
Small head-group areas:
NaDS in high salt
Non ionic lipids
lysolecithin
Double-chained lipids with
large head-group areas, fluid chains:
Lecithin, sphingomyelin,
Phosphatidylserine in water
Phosphatidylglycerol
Phosphatidic acid
Some single-chained lipids with
very small head-group
Double-chained lipids with
small head-group areas,
Anionic lipids in high salt,
saturated frozen chains:
Phosphatidylethanolamine,
Phosphatidylserine + Ca2+
Double-chained lipids with
small head-group areas,
Non ionic lipids,
unsaturated chains, high T:
Unsat phosphatidylethanolamine,
Cardiolipin + Ca2+
Phosphatidic acid + Ca2+
cholesterol
Lipid
Critical packing
parameter
v/a0lc
Critical packing
shape
Structure formed

Page 22
Chapter 1
- 12 -
3-2. Phase Transition Behavior of Lipid Bilayer Membrane
Hydrated lipids may exist in one or more mesomorphic forms. Analysis of the
phase transitions between these forms is necessary because the state/fluidity of the bilayers
is important determinant of in vitro and in vivo liposomal stability and drug release profiles.
The phospholipids most commonly employed in the production of liposomes are the
phosphatidylcholines. A typical differential scanning calorimetry (DSC) trace for a
diacylphosphatidylcholine is shown in Fig. 1-6. Pure, synthetic, long chained
phospholipids can undergo a number of transitions at defined temperatures. Following
prolonged incubation at low temperatures, fully hydrated, long chain phosphatidylcholines
such as DPPC, are in the ordered, condensed crystalline subgel (Lc) state, in which the
hydrocarbon chains are in the fully extended, all trans conformation, and the polar head
groups are relatively immobile at the water interface 7-9.
On heating, Lc state phospholipids undergo a subtrasition to the Lβ state, in which there is
Fig. 1-6 A typical DSC trace of a diacylphosphaticylcholine.

Page 23
Chapter 1
- 13 -
increased head group mobility and water penetration into the interfacial region of the
bilayer. The subtrasition usually occur approximately 30oC below the temperature of the
main gel to liquid-crystalline phase transition10, although DPPC has a subtransition at 21oC,
approximately 20oC below the temperature of main phase transition11. Subtransitions can
be subdivided into:
(a) Type I ‘‘solid-solid’’ transitions between sugel and gel phase. These are exhibited by
saturated
phophatidylcholines
with
C16
to
C18
chains9,12
and
dipalmitoylphophatidylglycerol13, and are characterized by a small change in rotameric
disordering of hydrocarbon chains.
(b) Type II transitions occur in saturated phophatidylethanolamines and involve more
rotameric disordering, and more melting of the subgel phase into the liquid-crystalline
phase14.
Headgroup interactions are also important: the inclusion of small amounts of the
opposite steroisomer or cholesterol eliminate the subtransition12. DPPC exhibits a
sub-subtransition with a Tm of approximately 6.8oC15, without a prolonged period of
incubation at low temperature. On heating, Lβ gel state lipids undergo the pretransition to
the ‘‘rippled’’ gel state. The pretransition usually occurs between 5-10oC below the main
transition, with a smaller enthalpy, and may be due to rotation of the polar head groups or
co-operative movement of the hydrocarbon chains, prior to structural changes in the
lamellar lattice, with the bilayer reorganizing from o one-dimensional lamellar to a
two-dimensional monoclinic lattice consisting of no lamellar distorted by periodic
‘‘ripples’’. Heating Pβ’ state lipids results in co-porative ‘‘melting’’ of the hydrocarbon
chains (the main gel to liquid-crystalline phase transition) to give the Lc state. The
orientation of the carbon-carbon single bonds changes from trans to a situation where
gauche configurations are present. The intermolecular distance between molecules is

Page 24
Chapter 1
- 14 -
approximately
2
nm,
consequently rotation in one
molecular, impacts on adjacent
molecules, such that this
transition is a co-operative event.
The temperature of the main
phase transition (Tc) is largely
determined by the polar head group together with the length and degree of unsaturation of
the hydrocarbon chains. For phospholipids having the same head group and degree of
hydration, increasing saturation in the hydrocarbon chains increase the Tc, with
trans-saturated chains having a higher Tc
than those which are cis-unsaturated.
Phospholipids with longer hydrocarbon chains have a higher Tc and associated enthalpy
than those with shorter chains (Table 1-2)8,16. The effect of the head group on the main
transition depends on the ionic strength and composition of the aqueous phase, together
with pH for charged lipids (Table 1-3)8,11.
Lipids
DLPC (C12)
DMPC (C14)
DPPC (C16)
DSPC (C18)
Transition temperature (Tc)
[oC]
0
23.0
41.0
58.0
ΔH [kJ/mol]
12.1
28.0
36.4
44.8
DAPC (C20)
61.8
49.8
DBPC (C22)
75.0
62.3
Head group
Choline
Ethanol amine
Glycerol
Phosphatidic acid
41.4
64
57
41
ΔH [kJ/mol]
34.7
35.6
37.2
31.4
Phosphatidylserine
67
21.8
Sphigomyelin
58
12.1
Transition temperature (Tc)
[oC]
pH
7.4
7.4
1.1
7.0
6.5
9.1
2.0
7.0
12.0
67
54
43.1
33.9
37.7
23.9
7.4
13.0
41.3
32
28.5
33.5
Table 1-2. Gel to liquid-crystalline phase transition
temperature for 1,2-diacylphosphatidylcholine bilayers
Table 1-3. Effect of head group on the main transition temperature of
dipalmytoylphospholipids.

Page 25
Chapter 1
- 15 -
Cholesterol is often included in liposomal membrane formulations to modify
fluidity 17, allowing control of the rate of release of entrapped hydrophilic materials and
enhancing in vitro and in vivo stability. Inclusion of cholesterol in C14 to C20
phosphatidylcholine bilayers, at between 2-6 mol%, eliminates the phospholipids
pretransition 18,19. Cholesterol also produces a decrease in the Tc as the main transition
endotherm for DMPC and DPPC liposomes containing between 1-25 mol% cholesterol can
be deconvoluted into two or possibly three peaks, indicative of phase separation into
cholesterol-rich and cholesterol-poor regions within the bilayer. A distinct transition is not
detectable by DCS when phospholipid bilayers contain 50 mol% cholesterol19,20.
3-3. Membrane Permeability
The rate at which a molecule diffuses across
a lipid bilayer membrane varies depending on the
size of the molecules and its relative solubility in
hydrophobic region of bilayer. In general, the
smaller the molecule and the more soluble in the
bilayer membrane, the rapidly it will diffuse across
a bilayer membrane. Small non-polar molecules
easily dissolve and rapidly diffuse across them. Uncharged polar molecules also diffuse
rapidly across a bilayer membrane if they are small enough. For example, CO2 (Mw: 44),
ethanol (Mw: 46), and urea (Mw: 60) across rapidly, glycerol (Mw: 92) less rapidly, and
glucose (Mw: 180) hardly across as shown in Fig. 1-7. The permeability of protons (1.44 x
10-4 cm·s-1) is greater than those of cations (10-13 – 10-14 cm·s-1), or anions (10-11-10-12
cm·s-1) 21. Though water molecules are relatively insoluble in bilayer membrane, water
Fig. 1-7
Relative membrane
permeability in a lipid bilayer.

Page 26
Chapter 1
- 16 -
permeability across single-compartment vesicles, is in the 10-3 - 10-6 cm·s-1 range 22. This
has been rationalized in terms of transport via hydrogen bond exchange with water
molecules in the bilayer. In addition, it is thought that the dipolar structure of the water
molecules allows it to cross the regions of the bilayer containing the lipid head group
unusually rapidly. On the other hand, lipid bilayer membranes are highly impermeable to
all charged molecules (e.g. ions, amino acids). The charge and high degree of hydration of
such molecules prevents them from entering the hydrocarbon cores of the bilayer.
4. Drug Delivery Focused on Carriers Using Molecular Assembly
4-1. Drug Delivery System
Drug delivery is one of the important considerations in drug development and
therapeutics. New technologies are applied for improving pharmacokinetic profiles to
reduce side effects and to improve patient compliance. The deciphering of the human
genome has opened up multiple avenues for identifying new targets and developing novel
therapies 23. With the availability of a plethora of genomic data and high-throughput
screens, there is a compelling need to translate this “new knowledge” into successful
therapies. Intelligent design of drug delivery systems (DDS) to meet this need requires a
thorough understanding of the factors influencing the disease, drug, destination and
delivery. Drug delivery scientists have been successful in achieving spatial and temporal
control of drug from DDS, but the path for transport of drug molecules from the delivery
system to the site of action remains largely uncontrolled. Targeting and releasing the drug
as close as possible to the site of action can fulfill this objective. The efficient functioning
and regulation of the biological system are strictly controlled by structural hierarchy
varying from centimeter to nanometer scales 24. Many of the bioassemblies and
endogenous molecules (such as enzymes, hormones and nucleic acids) exist and function

Page 27
Chapter 1
- 17 -
at the nanoscale level. Majority of the therapeutic targets exist intracellularly or at the
surface of the cells. Hence the goal of the drug delivery system is to deliver drugs to the
intended target with minimal interference with normal cellular functioning and with the
least systemic side effects.
4-2. Drug Carriers for Improving Drug Delivery
The use of nanocarriers offers an interesting opportunity for drug and gene
delivery where the delivery system becomes an active participant, rather than passive
vehicle, in the optimization of therapy. Several families of molecular assemblies are
employed as nano drug carriers for either passive or active targeting. Liposomes,
polymeric nanoparticles, block copolymer micelles, and so on are colloidal molecular
assemblies for drug delivery system (Fig. 1-8) 25.
[Liposomes]
A liposome is a spherical vesicle of colloidal dimensions domposed of one or more
lipid or phospholipid bilayers. The size of liposomes varies from 20 nm to 100 μm,
although each lipid bilayer is about 4nm thick. Various amphipathic molecules have been
Fig. 1-8 Various drug carrier for efficient drug delivery system

Page 28
Chapter 1
- 18 -
used to fabricate liposomes, especially lecithins from natural or synthetic sources. To
increase machanical stability and decrease the leakage of the contens, charged
phospholipids like phosphatidylserine or phosphatidylglycerol and cholesterol can be
added. Depending on its physicochemical characteristics, a drug can become trapped in the
aqueous space, intercalated or dissolved within the lipid bilayer, or form ionic or
hydrophobic complexes with nucleic acids and other macromolecules without physical
entrapment. For medical use, a lioposome suspension must be precisely defined with
respect to drug and lipid concentrations, size distribution, percentage of entrapped drug,
pH, osmolarity, conductivity, and presence of degradation products. Specific methods exist
to manufacture liopsomes of a particular size, morphology and surface characteristics,
parameters that together will determine the biologic fate of the liposomes.
Due to similarities in both structure and chemical composition to biomembranes,
liposomes have been extensively investigated as carriers for enhancing the incorporation
of an assortment of drug molecules into target cells. Through the use of liposomes one can
achieve a variety of therapeutic goals, including enhanced drug uptake and reduced
toxicity. Numerous studies have been conducted on the use of liosomes to deliver
chemotherapeutics to tumor cells; however, in spite of som impressive results in aminal
models, acceptance of liposomes as drug delivery system for human use has only been
realized in a very few instances. In addition to a relateively high cost, additional
shortcomings associated with this form of drug delivery are short residency time in the
circulation due to rapid removal by phagocytic cells of the reticuloendothelial system
(RES), physical and chemical instability on storage, and the lack of generally uniform and
applicable techniques for their production on a commercial scale.

Page 29
Chapter 1
- 19 -
[Micelles]
Polymeric micelles are a visible form of targeted delivery system for
water-insoluble and amphiphilic drugs. Like surfactant-based micells, polymeric micelles
are core/shell structure; however, unlike their traditional counterparts, they are physically
more stable and capable of solubilizing substantial amounts of hydrophobic compounds
within their inner core. Due to their hydrophilic s hell and small size they may prolong the
circulation of drugs in body fluids and increase drug accumulation in tissues. The shell is
responsible for micelle stabilization and interaction with plasma proteins and cell
membranes. The hydrophobic core, usually made from a biodegradable polymer, serves as
a reservoir for an insoluble drug and shields the drug from the aqueous surroundings.
Polymers used for fabricating polymeric micelles have included pluronics, polyethylene
glycol (PEG)-lipid conjugates and pH-sensitive poly(N-isopropylacrylamide)-based
micelles or polyion complex micelles. Polymeric micelles have also use for encapsulating
and delivering photosensitizing drugs and dyes. However, the biodistribution
characteristics of the polymeric micelle formulating thus far investigated have not been
entirely satisfactory. For example, greater selectivity and effectiveness could be gained by
introducing targeting ligands and site-controlled releasing capabilities, respectively, and
excessive nanoparticulate accumulation and long-term toxic effects could be reduced by
utilizing biodegradable polymers with good clearance properties.
[Polymer-based nanoparticles]
In recent years, polymeric nanoparticles have received recognition as a promising
type of colloidal drug carrier. They have bee widely used for controlled drug delivery by
the intravenous, ocular and oral routes of administration. An additional positive feature of
this form of drug delivery is that they have shown potential as carrier for anticancer agents.

Page 30
Chapter 1
- 20 -
It has been verified that the tissue distribution characteristics, specificity and
pharmacokinetic properties of anticancer drugs can be better controlled upon their
incorporation into nanoparticles. Furthermore, their form of drug delivery may contribute
to reducing the side effects and toxicity normally associated with anticancer drugs, while
increasing therapeutic efficacy, and can lead to drug accumulation in the solid tumor since
they has been found to escape from the vasculature through the leaky endothelial tissue
that surrounds a solid tumor.
Nanoparticles can overcome the multidrug-resistance phenotype mediated by
P-glycoprotein and bring about an increase in drug content inside neoplastic cells. This
finding is of special significance for patients with cancer undergoing treatment with
paclitaxel and who, after some time, develop resistance to the drug. Other importance
advantages derived from the use of nanoparticles include the ease of their preparation, the
availability of well-defined biodegradable polymers for their manufacturing [e.g.
poly(D,L-lactide-co-glycolide) (PLGA)], and their high stability in biological fluids and
during storage. Moreover, nanoparticles can permeate cells and tissues to become
internalized and can efficiently deliver a particular drug to its target tissue without
clogging the capillaries.
4-3. Gene Therapy
Today, nucleic acids are readily manipulated by specific enzymes, produced in
large quantities and progress in automation has enables the sequencing of entire genomes,
including the human genome. Such progress coupled with an increase in the understanding
of the molecular mechanisms and genetic implications underlying manyu diseases has
triggered the hope that nucleic acids could treat disease at the source. However, many
hurdles remain that must be overcome in order for gene therapy to be of practical use in the

Page 31
Chapter 1
- 21 -
clinic. Gene delivery relies upon the encapsulation of a gene of interest, which is then
ideally delivered to target cells. After uptake up by endocytosis, the DNA molecules must
be released into the cell so that transcription and translation may occur to produce the
protein of interest. To achieve successful gene delivery, significant barriers must be
overcome at each step of this process in order to optimize gene activity while minimizing
the potential for inhibitory inflammatory responses. Particularly interest has been paid in
recent years to the development of efficient non-viral vectors. Viral vectors such as such as
adenovirus vectors, adeno-associated virus vectors or retrovirus vectors may provide
superior gene delivery to target cells compared to their non-viral counterparts, but viral
vectors also come with the significantly increased risk of triggering a specific immuno
response, which under wxtreme circumstances could result in death. Non-viral vectors may
trigger an inflammatory response but are not likely to elicit specific recognition, making
these types of vectors less hazardous in terms of antigen-specific immuno responses.
Although non-viral vectors are more appealing in this respect, there are several other
factors that must be considered in vector design, including specific cell targeting,
optimized uptake, and efficient intracellular release of the vector, in addition to minimizing
the immunoresponse. The development of novel non-viral vectors intended to optimize one
or more of these aspects of gene delivery is keenly awaited.

Page 32
Chapter 1
- 22 -
Reference
1. Moffat, A.S. MOSAIC 21 (1990) 30.
2. Whitesides, G.M., Mathias, J.P., Seto, C.T. Science 254 (1991) 1312–1319.
3. Israelachvilli, J.N., Mitchell, D.J. Ninham, B.W. Biochim. Biophys. Acta 470 (1997)
9185–201.
4 Israelachvilli, J.N., Marcelja, S., Horn, R.G. Q.Rev. Biophys. 470 (1980) 121–200.
5. Israelachvilli, J.N. ‘‘Intermolecular and surface Forces,’’ 2nd ed. Academic Press,
London.
6. Gruner, S.M. ‘‘Structure of Biological Membrane’’ P.L. Yeagle, ed. P211–250. CRC
Press, Boca Raton, FL.
7. Fuldner, H.H. Biochemistry 20 (1981) 5707–5710.
8.Cameron, D.G., Mantsch, H.H. Biophys. J. 38 (1982) 175–184.
9. Lewis, B.A., Dasgupta, S.K., Griffin, R.G. Biochemistry 23 (1984) 1988–1993.
10. Chen, S.C., Sturtevant, J.M., Gaffney, B.J. Proc. Natl, Sci. Sci. U.S.A. 77 (1980)
5060–5063.
11. Biltonen, R.L., Lichtenberg, D. Chem. Phys. Lipids 64 (1993) 129–142.
12. Finegold, L., Singer, M.A. Chem. Phys. Lipids 35 (1984) 291–297.
13. Blauroch, A.E. Biochemisty 25 (1986) 299–305.
14. Wilkinson, D.A., Nagle, J.F. Biochemisty 23 (1984) 1538–1541.
15. Slater, J.L, Huang, C. Biophys. J. 52 (1987) 667–670.
16. van Dijck, P.W.M., Kaper, A.J., Oonk, H.A., de Gier, J. Biochim. Biophys. Acta 1977,
470, 58–69.
17. Kirby, C., Clarke, J., Gregoriadis, G. Biochem J.186 (1980) 591–598.
18. Malcomson, R.J., Higinbotham, J., Beswick, P.H., Privat, P.O., Saunier, L. J. Membr.
Sci. 123 (1997) 243–246.

Page 33
Chapter 1
- 23 -
19. Mcmullen, T.P.W., Lewis, R.N.A.H., McElhaney, R.N. Biochemistry 32 (1993)
522–526.
20. Mabrey, S., Matteo, P.L., Stuetevant, J.M. Biophys. J. 17 (1977) 82a
21. Gutknech, J., Walter, A. Biochim. Biophys. Acta 644 (1981) 153–156.
22. Lawaczek, R. J. Membr. Biol. 51 (1979) 3–4.
23. Drews, I.I. Drug Discov. Today 5 (2000) 2–4
24. Esfand, R., Tomalia D.A. Drug Discov. Today 6 (2001) 427–436.
25. Ganta, S., Devalapally, H., Shahiwala, A., Amiji, M. J. Cont. Release 126 (2008)
187–204.

Page 34
Chapter 1
- 24 -

Page 35
Chapter 2
- 25 -
Chapter 2
Syntheses of Amino-acid Based Lipids Having a Cationic Head
Group and Evaluation of Their Properties for Gene Delivery
1. Introduction
Cationic liposomes are currently the most heavily focused and they are widely used for
the purpose of drug delivery today. The premise behind cationic liposomes is that most cell
surfaces (plasma membrane) contain negatively charged residues (e.g. sialic acid, membrane
protein)1, and these negatively biological cellular membrane could be targeted through
electrostatic interaction by a cationic liposomes mediated drug delivery system. The major
application of cationic liposomes is nonviral vectors for gene therapy.
In this chapter, the author initially synthesizes a series of amino-acid base lipids to
investigate gene transferring capacity based on cationic lipid structure. Indeed, for estimation of
gene transferring capacity, the characteristics of cationic liposomes based on lipid structure,
which influence gene expression efficiency, are introduced.

Page 36
Chapter 2
- 26 -
2. Cationic Liposome Mediated Transfection
2-1. Cationic Liposomes for Cellular Transfection
Cationic lipids represent the second group of synthetic vectors commonly used in gene
delivery. Since first being used for gene therapy in 1987 by Felgner et al.2, numerous cationic
lipids (also called cytofectin or lipofection reagents) have been synthesized and used for delivery
in cell culture, animals, and patients enrolled in phases I and II of clinical trials. Cationic lipids
are technically simple and quick to formulate, readily available commercially, and may be
tailored for specific applications. The cationic liposome formulation is mixed with DNA in a
dilute solution. The DNA will form a lipoplex with the liposomes through charge interaction as
DNA is a polyanion (Fig. 2-1). The liposomes are kept in slight excess such that the resulting
DNA-liposome complex retains a slightly net positive charge. The complex formation is due to
random charge interactions and is difficult if not impossible to control. The result is that
complexes will form with varying sizes and net charge, that relates to gene expression efficiency.
Quite often, the complexes are too large to be taken up by the cell. Previous studies using
lipoplexes suggested that the most effective complexes size are 300 nm or smaller.
+
+
+
+
+
+
Cationic liposome
DNA
DNA-liposome complex
Fig. 2-1. Schematic illustration of lipoplexes composed of positively charged
liposomes and negatively charged DNA.

Page 37
Chapter 2
- 27 -
2-2. Conventional Cationic Lipids for Transfection
Concerning cationic lipid structure, they are composed of a cationic head group attached
by a linker to lipid hydrophobic moieties. The positively charged head group is necessary for the
binding of nucleic acid phosphate groups for complexasation as described above for lipoplexes.
All cationic lipids are therefore positively charged amphiphiles systems. They can be classified
into various subgroups according to their basic structural characteristics (Chart 2-1).
(1) Monovalent aliphatic lipids characterized by a single amine function in their head group, e.g.
N[1-(2,3-dioleyloxy) propyl]-N,N,N-trimethylammonium chloride (DOTMA), 1,2-dioleyl-
3-trimethylammonium-propane
(DOTAP),
N-(2-hydroxyethyl)-
N,N-dimethyl-2,3-bis(tetradecyloxy-1-propanaminiumbromide) (DMRIE).
(2) Multivalent aliphatic lipids whose polar head groups contain several amine functions such as
the spermine group, e.g. dioctadecylamidoglycylspermine (DOGS).
(3) Cationic cholesterol derivatives, e.g. 3β-[N-(N0,N0-dimethylaminoethane)-carba-
moyl]cholesterol (DC-Chol), bis-guanidium-tren-cholesterol (BGTC).
Chart 2-1. Conventional cationic lipids having various cationic head group and hydrophobic
moieties for transfection..

Page 38
Chapter 2
- 28 -
3. Cationic Assemblies Containing Synthetic Amino-acid Based Lipids
3-1. Syntheses of Cationic Amino-acid Based Lipids
The author focuses on the cationic lipid-based plasmid DNA (pDNA) carriers without
using phospholipids. Then, the author synthesized a series of cationic lipids having a glutamate
backbone and a cationic amino acid such as lysine, histidine, or arginine as the head group as
shown in Scheme 2-1.
p-Tos
95oC, Benzene
+
R-OH
Glutamic acid
R=C14, C16, C18
n=13, 15, 17
Boc-Lys(Boc)-OSu
CH2Cl2 / TEA
TFA
Boc-His(1-Boc)-OSu
CH2Cl2 / TEA
TFA
TFA
Boc-Arg (Boc)2-OH
CH2Cl2 / TEA/BOP
O
O
O
N
H
C
O
(CH2)n
CH3
O (CH2)n
CH3
H2N
NH
NH
NH2
O
O
O
N
H
C
O
(CH2)n CH3
O (CH2)n CH3
H2N
NH
N
O
O
O
N
H
C
O
(CH2)n CH3
O (CH2)n CH3
H2N
H2N
O
O
O
H2N
(CH2)n
CH3
O (CH2)n
CH3
Glu2Cn
1a : n=13
1b : n=15
1c : n=17
2a : n=13
2b : n=15
2c : n=17
3a : n=13
3b : n=15
3c : n=17
O
O
O
N
H
C
O
(CH2)n CH3
O (CH2)n CH3
HN
NH
N
NH
Boc
Boc
Boc
O
O
O
N
H
C
O
(CH2)n CH3
O (CH2)n CH3
N
NN
Boc
Boc
O
O
O
N
H
C
O
(CH2)n CH3
O (CH2)n CH3
N
N
Boc
Boc
Scheme 2-1. Synthesis of amino-acid based cationic lipids bearing lysine, histidine or arginine
at head group and varying length of the hydrocarbon chain.

Page 39
Chapter 2
- 29 -
The author synthesized a series of amino-acid based cationic lipids that were expected to
display gene transferring capacity by utilizing the head group amino-acid (lysine, histidine or
arginine). Two long-chain alcohol molecules of varying chain length were esterified to the
carboxyl groups of glutamic acid (a-c). The yield of a-c after recrystallization was high (i.e.,
>70%). Protected amino-acid reagents, such as Boc-Lys(Boc)-OSu, Boc-His(1-Boc)-OSu or
Boc-Arg(Boc)2-OH, were reacted with the amino group of a-c via an amide linkage.
Deprotection with trifluoroacetic acid led to the formation of the amino-acid based cationic lipids.
Thus, the author succeeded in devising an efficient synthetic route to generate cationic lipids
using a minimal number of steps, followed by recrystallization to give a cost effective
purification method. The further information of synthesis will describe in experimental section.
2-2. Morphological Study of Cationic Assemblies
The cationic assemblies formed from the synthetic amino-acid based cationic lipids were
prepared by an extrusion method (Table 2-1) and were observed using TEM (Fig. 2-2). In
addition, the author confirmed that the structure of the cationic assemblies was influenced by the
head group of the lipids. Assemblies prepared from the lysine- (1a-1b) or arginine-type lipid
(3a-3c) formed unilamellar vesicles with a diameter of approximately 100 nm by DLS
measurements, but more than 100 nm in size for the cationic assembly of 1c. By contrast,
morphology of the histidine-type lipids (2b, 2c) was tube-like. The histidine-type lipid 2a
aggregated when dispersed in distilled water, and could not permeate through a membrane filter
of pore size 3 µm. No further characterization of 2a was performed in this study. Moreover,
morphologies of the unilamellar vesicles formed from lysine-, arginine-, or histidine-type lipids
were less influenced by variations in the length of alkyl chains.

Page 40
Chapter 2
- 30 -
The morphology of cationic assemblies was analyzed. The lysine- or arginine-type lipids
formed unilamellar vesicles, whereas the histidine-type lipid formed a tube-like structure. These
Lipid
composition
1c
1b
1a
3c
3b
3a
Particle size
(nm)
89 + 37
98 + 31
116 + 54
90 + 32
93 + 34
92 + 40
2c
2b
93 + 34
97 + 37
Table 2-1. The size distribution of cationic assemblies
constructed with the cationic lipids after extrusion.
Fig. 2-2. Transmission electron microscopic images of cationic assemblies containing the
synthetic lipids by varying head group and length of alkyl chain.
Lys
His
Arg
14
16
18
100 nm
100 nm
100 nm
100 nm
100 nm
100 nm
1a
1b
1c
2b
3a
3b
3c
100 nm
2c
100 nm
N.D.
100 nm
100 nm
100 nm
100 nm
100 nm
100 nm
1a
1b
1c
2b
3a
3b
3c
100 nm
2c
100 nm
N.D.
Head group
Length of alkyl chain

Page 41
Chapter 2
- 31 -
results indicate that the overall architecture of the assemblies is largely determined by the nature
of the hydrophilic head group. It was reported that assemblies formed from amino-acid based
lipids with an asymmetric carbon can adopt various structures such as a tube, fiber or ribbon 3-4.
Here, the author confirmed that the synthetic amino-acid based lipids with two asymmetric
carbon atoms would be expected to take various assembling structures. After extrusion, the
histidine-type lipid initially formed a vesicular structure that was observed by TEM (data not
shown). However, the ratio of tube-like assemblies gradually increased over a period of a week
during storage at 4oC. If the transition temperature is exceeded during the extrusion procedure, a
rearrangement of lipid molecules into unstable bilayer vesicles might occur. Subsequent
incubation at low temperatures will then cause the vesicles to gradually aggregate and fuse into a
more stable state i.e., in this case a tube-like structure 5. Therefore, the author concluded that 2a,
2b and 2c couldn’t form stable vesicle dispersions due to an unfavorable balance of hydrophilic
and hydrophobic characteristics, and the particular structure of the histidine moiety.
3-3. Physicochemical Properties of Assemblies
The author investigated transition temperature (Tc) because the membrane fluidity
influenced on gene transferring capacity. The transition temperature of the lysine-type lipids 1a,
1b and 1c was 26.3, 43.3 and 53.2oC, respectively (Table 2-2). For the arginine-type lipids 3a,
3b and 3c, the transition temperatures were 24.7, 42.5 and 52.7oC, respectively. Similar analyses
of 2b and 2c were conducted to obtain the transition temperatures of 34.4 and 42.3oC,
respectively. The results indicated that the transition temperature of the cationic assemblies
increased with increasing alkyl chain length. Thus, the hydrophobic moiety appears to have a
direct effect on fusogenic potential. There was no significant difference in transition temperature
between the lysine-type and arginine-type series of lipids having the same alkyl chain length.

Page 42
Chapter 2
- 32 -
3-4. Synthesis of Cationic Lipids Having a Hydrocarbon Spacer between Head Group and
Hydrophobic Moieties
For further explosion of appropriate cationic lipid structure, the author focused on spacer
between cationic moiety and hydrophobic moieties of the lipids oriented onto the liposomal
membrane because the hydrophobic component is preferably fused to cellular membrane. Few
reports suggested that the hydrophilic or hydrophobic spacers of the cationic lipids influenced on
the gene expression, though they evaluated them in the lipoplex system 6–8. Then, the author
synthesized cationic lipids having a hydrocarbon type spacer with varying the length as shown in
Scheme 2-2.
Table 2-2. A transition temperature of cationic assemblies by DSC analysis
Scheme 2-2. Syntheses of cationic lipids having a hydrocarbon type spacer with varying the
length between cationic head group and hydrophobic moieties.
O
O
O
O
H2N
( CH2 )15
( CH2 )15
CH3
CH3
O
O
O
O
H
N
( CH2 )15
( CH2 )15
CH3
CH3
C
O
NH-(CH2)n
O
O
O
O
H
N
( CH2 )15
( CH2 )15
CH3
CH3
C
O
(CH2)n
NH
NH
C
O H
N
NH2-(CH2)n-COOH
Boc-NH-(CH2)n-COOH
Boc
O
O
O
O
H
N
( CH2 )15
( CH2 )15
CH3
CH3
C
O
NH2-(CH2)n
O
O
O
O
H
N
( CH2 )15
( CH2 )15
CH3
CH3
C
O
(CH2)n
H2N
H2N
C
O H
N
Boc
Boc
4a: n = 3
4c: n = 7
4b: n = 5
E
aReaction conditions: (A) (Boc)2O/TEA/methanol, 60oC; (B) Boc-NH-(CH2)n-COOH/BOP/TEA/DCM; (C) TFA, 4oC; (D) Boc-Lys(Boc)-OSu/TEA/DCM;
(E) TFA, 4oC.
A
B
C
D
4d: n = 11
Compounds
Transition temp. (oC)
1c
1b
1a
3c
3b
3a
26.3
43.3
53.2
24.7
42.5
52.7

Page 43
Chapter 2
- 33 -
4. Experimental Section
4-1. Syntheses of Amino-acid Based Cationic Lipids
Synthesis of the hydrophobic moiety of cationic lipids.
a
(1,5-ditetradecyl-L-glutamate),
b
(1,5-dihexadecyl-L-glutamate), and
c
(1,5-dioctadecyl-L-glutamate) were synthesized as shown in Scheme. L-Glutamic acid (1 g, 6.8
mmol) and p-Tos (1.65 g, 8.16 mmol) were dissolved in benzene (200 mL) and refluxed for 1 hr
at 90oC. Tetradecylalcohol, hexadecylalcohol or octadecylalcohol (15 mmol) was added to the
solution, followed by stirring for 12 hr under reflux. The reaction mixture was evaporated and
then dissolved in chloroform (100 ml). The chloroform solution was treated with a sodium
carbonate solution (100 ml x 2) and washed with distilled water (100 ml x 1). After the
chloroform solution was evaporated, a, b and c were recrystalized from methanol (100 mL) at
4oC to obtain a white powder with a yield of 76%, 82% and 83 %, respectively.
1,5-Ditetradecyl-L-glutamate (a): Rf 0.75 (CHCl3:MeOH 4:1). 1H-NMR (CDCl3:MeOD 10:1, 500 MHz,
δ ppm): 0.88 (t, 6H, CH2CH3); 1.28 (m, 44H, CH2(myristoyl)), 1.58-1.66 (m,4H, COOCH2CH2); 2.01-2.10 (m,
2H, NH2CHCH2), 2.45 (t, 2H, CH2CO); 3.80 (t, 1H, NH2CH), 4.06-4.14 (t, 4H, COOCH2), 7.18, 7.73 (d,
2H, NH2). MS(ESI): (M+H)+ calcd. for C33H65NO4, 539.87; found, 540.5.
1,5-Dihexadecyl-L-glutamate (b): Rf 0.73 (CHCl3:MeOH 4:1). 1H-NMR (CDCl3:MeOD 10:1, 500 MHz,
δ ppm): 0.88 (t, 6H, CH2CH3); 1.26 (m, 52H, CH2(palmitoyl)), 1.55-1.64 (m,4H, COOCH2CH2); 2.21 (m, 2H,
NH2CHCH2), 2.51 (t, 2H, CH2CO), 4.04 (t, 1H, NH2CH), 4.10-4.14 (t, 4H, COOCH2), 7.18, 7.72 (d, 2H,
Fig. 2-3.
1H-NMR spectrum of 1,5-ditetradecyl-L-glutamate (a), 1,5-dihexadecyl-L-glutamate (b),
and 1,5-distearyl-L-glutamate (c).

Page 44
Chapter 2
- 34 -
NH2). MS(ESI): (M+H)+ calcd. for C37H73NO4, 595.58; found, 595.5.
1,5-Dihexadecyl-L-glutamate (c): Rf 0.75 (CHCl3:MeOH 4:1). 1H-NMR (CDCl3:MeOD 10:1, 500 MHz,
δ ppm): 0.88 (t, 6H, CH2CH3); 1.26 (m, 60H, CH2(stearyl)), 1.52-1.67 (m,4H, COOCH2CH2), 1.90-2.12 (m,
2H, NH2CHCH2), 2.47 (t, 2H, CH2CO), 3.59 (t, 1H, NH2CH), 4.06-4.15 (t, 4H, COOCH2), 7.18, 7.73 (d,
2H, NH2). MS(ESI): (M+H)+ calcd for C33H65NO4, 652.09; found, 652.5.
Syntheses of amino-acid based cationic lipids
Amino acid-based cationic lipids having a cationic head-group were synthesized as shown in
Scheme. a, b or c (1.68 mmol) and triethylamine (2 mmol) were dissolved in dichloromethane
and stirred for 1 hr at room temperature. Either Boc-Lys(Boc)-OSu, Boc-His(1-Boc)-OSu, or
Boc-Arg(Boc)2-OH was added to the solution and stirred for 6 hr at room temperature. The
reaction mixture was evaporated and then dissolved in chloroform (100 ml). The chloroform
solution was treated with a sodium carbonate solution (100 ml x 2) and washed with distilled
water (100 ml x 1). After the chloroform solution was evaporated, the amino group-protected
intermediates were recrystalized from methanol (50 mL) at 4oC. After deprotection with
trifluoroacetic acid (20 mL) for 2 hr at 4oC, 1a (1,5-ditetradecyl N-lysyl-L-glutamate), 1b
(1,5-dihexadecyl N-lysyl-L-glutamate) or 1c (1,5-dioctadecyl N-lysyl-L-glutamate) were
obtained as a white powder (yields: 61%, 90%, or 86%, respectively) after freeze-drying with
benzene. Using a similar method to that described above, 2a (1,5-ditetradecyl
N-histidyl-L-glutamate), 2b (1,5-dihexadecyl N-histidyl-L-glutamate) and 2c (1,5-dioctadecyl
N-histidyl-L-glutamate) were obtained with a yield of 68%, 81% and 80%, respectively. For the
synthesis of arginine-type lipids, BOP reagent was used for the amide linkage. Compounds 3a
(1,5-ditetradecyl N-arginyl-L-glutamate), 3b (1,5-dihexadecyl N-arginyl-L-glutamate) and 3c
(1,5-dioctadecyl N-arginyl-L-glutamate) were obtained with a yield of 40%, 55% and 70%,
respectively.

Page 45
Chapter 2
- 35 -
1,5-Ditetradecyl N-lysyl-L-glutamate (1a): Rf 0.14 (CHCl3:MeOH 4:1). 1H-NMR (CDCl3:MeOD 10:1,
500 MHz, δ ppm): 0.88 (t, 6H, CH2CH3); 1.26 (m, 44H, CH2(myristoyl)), 1.53 (m, 2H, NH2CHCH2CH2), 1.62
(m, 4H, COOCH2), 1.72 (m, 2H, NH2CH2CH2), 1.80-1.95 (m, 2H, NH2CHCH2), 1.96-2.25 (m, 2H,
NHCHCH2), 2.43 (m, 2H, CH2COO), 2.96 (m, 2H, NH2CH2), 3.97 (m, 1H, NH2CHCO), 4.00-4.20 (m,
4H, COOCH2CH2), 4.51 (q, 1H, NHCH). MS(ESI): (M+H)+ calcd. for C39H77N3O5, 668.05; found, 669.7.
1,5-Dihexadecyl N-lysyl-L-glutamate (1b): Rf 0.19 (CHCl3:MeOH 4:1). 1H-NMR (CDCl3:MeOD 10:1,
500 MHz, δ ppm): 0.88 (t, 6H, CH2CH3); 1.26 (m, 52H, CH2(palmitoyl)), 1.56 (m, 2H, NH2CHCH2CH2), 1.62
(m, 4H, COOCH2), 1.69 (m, 2H, NH2CH2CH2), 1.87 (m, 2H, NH2CHCH2), 1.94-2.25 (m, 2H,
NHCHCH2), 2.43 (m, 2H, CH2COO), 2.95 (m, 2H, NH2CH2), 3.97 (m, 1H, NH2CHCO), 4.02-4.16 (m,
4H, COOCH2CH2), 4.51 (q, 1H, NHCH). MS(ESI): (M+H)+ calcd. for C43H85N3O5, 724.15; found, 724.7.
1,5-Dioctadecyl N-lysyl-L-glutamate (1c): Rf 0.06 (CHCl3:MeOH 4:1). 1H-NMR (CDCl3:MeOD 10:1,
500 MHz, δ ppm): 0.88 (t, 6H, CH2CH3); 1.26 (m, 60H, CH2(stearyl)), 1.52 (m, 2H, NH2CHCH2CH2), 1.64
(m, 4H, COOCH2CH2), 1.72 (m, 2H, NH2CH2CH2), 1.88 (m, 2H, NH2CHCH2), 1.95-2.27 (m, 2H,
NHCHCH2), 2.44 (m, 2H, CH2COO), 2.94 (m, 2H, NH2CH2), 3.94 (m, 1H, NH2CHCO), 4.05-4.12 (m,
4H, COOCH2CH2), 4.52 (q, 1H, NHCH). MS(ESI): (M+H)+ calcd. for C43H85N3O5, 780.26; found, 780.7.
Fig. 2-4.
1H-NMR spectrum of 1,5-ditetradecyl N-lysyl-L-glutamate (1a), 1,5-dihexadecyl
N-lysyl-L-glutamate (1b), and 1,5-distearyl N-lysyl-L-glutamate (1c).
Fig. 2-5.
1H-NMR spectrum of 1,5-ditetradecyl N-histidyl-L-glutamate (2a), 1,5-dihexadecyl N-
histidyl-L-glutamate (2b), and 1,5-distearyl N-histidyl-L-glutamate (2c).

Page 46
Chapter 2
- 36 -
1,5-Ditetradecyl N-histidyl-L-glutamate (2a): Rf 0.14 (CHCl3:MeOH 4:1). 1H-NMR (CDCl3:MeOD
10:1, 500 MHz, δ ppm): 0.88 (t, 6H, CH2CH3); 1.26 (m, 44H, CH2(myristoyl)), 1.64 (m, 4H, COOCH2CH2),
1.96-2.25 (m, 2H, NHCHCH2), 2.40 (t, 2H, NHCHCH2CH2), 2.94-3.19 (q, 2H, NH2CHCH2), 3.93 (q, 1H,
NHCH), 4.05-4.16 (m, 4H, COOCH2), 4.52 (q, 1H, NH2CH), 7.00 (s, 1H, CH2C=CH), 7.82 (s, 1H,
N=CH). MS(ESI): (M+H)+ calcd. for C39H72N4O5, 677.01; found, 677.6.
1,5-Dihexadecyl N-histidyl-L-glutamate (2b): Rf 0.15 (CHCl3:MeOH 4:1). 1H-NMR (CDCl3:MeOD 10:1, 500
MHz, δ ppm): 0.88 (t, 6H, CH2CH3); 1.26 (m, 52H, CH2(palmitoyl)), 1.64 (m, 4H, COOCH2CH2), 1.96-2.29 (m, 2H,
NHCHCH2), 2.46 (t, 2H, NHCHCH2CH2), 3.20-3.49 (q, 2H, NH2CHCH2), 4.05-4.14 (m, 4H, COOCH2), 4.33 (q,
1H, NHCH), 4.52 (q, 1H, NH2CH), 7.82 (s, 1H, N=CH). MS(ESI): (M+H)+ calcd. for C43H80N4O5, 732.12 ; found,
733.7.
1,5-Dioctadecyl N-histidyl-L-glutamate (2c): Rf 0.13 (CHCl3:MeOH 4:1). 1H-NMR (CDCl3:MeOD
10:1, 500 MHz, δ ppm): 0.88 (t, 6H, CH2CH3); 1.26 (m, 60H, CH2(stearyl)), 1.52-1.63 (m, 4H,
COOCH2CH2), 1.94-2.22 (m, 2H, NHCHCH2), 2.37 (t, 2H, NHCHCH2CH2), 2.85-3.18 (q, 2H,
NH2CHCH2), 4.05-4.13 (m, 4H, COOCH2), 4.52 (q, 1H, NH2CH), 6.88 (d, 1H, CH2C=CH), 7.50 (d, 1H,
N=CH). MS(ESI): (M+H)+ calcd. for C47H88N4O5, 789.23; found, 789.7.
1,5-Ditetradecyl N-arginyl-L-glutamate (3a): Rf 0.24 (CHCl3:MeOH 4:1). 1H-NMR (CDCl3:MeOD
10:1, 500 MHz, δ ppm): 0.88 (t, 6H, CH2CH3); 1.27 (m, 44H, CH2(myristoyl)), 1.61 (m, 4H, COOCH2CH2),
1.72 (m, 2H, NH2CHCH2CH2), 1.78-1.92 (m, 2H, NH2CHCH2), 1.96-2.25 (m, 2H, NHCHCH2), 2.42 (t,
2H, NHCHCH2CH2), 3.18 (m, 2H, CH2NHC=N), 4.06-4.14 (m, 4H, COOCH2), 4.51 (q, 1H, CONHCH).
MS(ESI): (M+H)+ calcd. for C39H77N5O5, 696.06; found, 696.7.
1,5-Dihexadecyl N-arginyl-L-glutamate (3b): Rf 0.07 (CHCl3:MeOH 10:1). 1H-NMR (CDCl3:MeOD
4:1, 500 MHz, δ ppm): 0.88 (t, 6H, CH2CH3); 1.26 (m, 52H, CH2(palmitoyl)), 1.61 (m, 4H, COOCH2CH2),
1.72 (m, 2H, NH2CHCH2CH2), 1.86-2.02 (m, 2H, NHCHCH2), 1.94-2.26 (m, 2H, NHCHCH2), 2.43 (t,
2H, NHCHCH2CH2), 3.18 (m, 2H, CH2NHC=N), 3.96 (t, 1H, NH2CHCH2), 4.06-4.15 (m, 4H, COOCH2),
4.51 (q, 1H, NHCHCH2). MS(ESI): (M+H)+ calcd. for C43H85N4O5, 752.17; found, 752.8.
1,5-Dioctadecyl N-arginyl-L-glutamate (3c): Rf 0.05 (CHCl3:MeOH 101). 1H-NMR (CDCl3:MeOD 4:1,
Fig. 2-6.
1H-NMR spectrum of 1,5-ditetradecyl N-arginyl-L-glutamate (3a), 1,5-dihexadecyl N-
arginyl-L-glutamate (3b), and 1,5-distearyl N-arginyl -L-glutamate (3c).

Page 47
Chapter 2
- 37 -
500 MHz, δ ppm): 0.88 (t, 6H, CH2CH3); 1.26 (m, 60H, CH2(stearyl)), 1.60 (m, 4H, COOCH2CH2), 1.52,
1.72 (m, 2H, NH2CHCH2CH2), 1.88-2.02 (m, 2H, NHCHCH2), 1.96-2.25 (t, 2H, NHCHCH2), 3.19 (m,
2H, CH2NHC=N), 4.00 (t, 1H, NHCHCH2), 4.05-4.16 (m, 4H, COOCH2), 4.51 (q, 1H, NH2CH).
MS(ESI): (M+H)+ calcd. for C39H77N5O5, 808.27; found, 808.8.
4-2. General Methods
Measurement of size distribution of cationic assemblies
The lyophilized lipid powder (10 mg) was hydrated in deionized water (1 mL) for 2 hr
and extruded with a LIPEX™ EXTRUDER (Northern Lipids Inc.; Vancouver, Canada) with
membrane filters (cellulose acetate filters with pore sizes of 3.00, 0.80, 0.65, 0.45, 0.22, 0.10 µm,
Millipore Inc., Bedford, MA) at 60oC. The cationic assemblies were prepared from the lysine-
(1a-1c), histidine- (2a-2c) or arginine- (3a-3c) type lipids. The lipid concentration of the cationic
assemblies was determined by estimating the lipid weigh after the freeze drying the cationic
assembling dispersion.
TEM observation of cationic assemblies
The cationic assemblies were observed by transmission electron microscopy (TEM). A
drop of the sample dispersion was placed on a 100 mesh copper grid, and then the excess
dispersion was removed with a filter paper. A drop of 2% phosphotungstic acid solution (pH 7.4)
was added to the grid and then dried for 12 hr in a desiccator. The morphology of the cationic
assemblies was observed by TEM (JM-1011, JEOL).
Calorimetric analysis of the cationic assemblies
A gel-to-lipid crystalline phase transition temperature (Tc) of the cationic assemblies
from the synthetic amino-acid based cationic lipids was investigated using a differential scanning
calorimeter (DSC). The dispersion containing 40 µmol-lipids of cationic assemblies was added
to a silver pan and sealed. A reference pan was mounted with 30 µL of distilled water. The

Page 48
Chapter 2
- 38 -
sample pan and the reference pan were placed in a DSC cell compartment (Differential Scanning
Colorimeter Q2000; TA Instruments, Newcastle, DE). The measurement was usually started
from 0oC, and the temperature was raised at a rate of 2oC/min up to 80oC. The transition
temperature was estimated from the DSC curve.
4-3. Synthesis of Cationic Lipids Having a Spacer between Cationic Head Group and
Hydrophobic moieties
The author constructed hydrocarbon chain spacers with different carbon lengths from an
ammonioalkanoic acid as shown in Scheme 2-2. 4-Aminobutyric acid (5 g, 48.5 mmol),
6-aminohexanoic acid (5 g, 38.2 mmol), 8-aminooctanoic acid (5 g, 31.4 mmol), or
12-aminolauric acid (5 g, 23.3 mmol) was dissolved in 100 mL methanol, and then triethylamine,
and 1.1-hold of di-t-butyl dicarbonate ((Boc)2O) to the ammonioalkanoic acid group were added
to the solution. After stirring for 12 hr at 60oC, the solvent was evaporated. Ethyl acetate (100ml)
was added to the remaining sticky liquid. The ethyl acetate solution was treated with a 0.2-N
hydrochloride solution (100 ml x 2) and washed with distilled water (100 ml x 2). After the
solution
was
evaporated,
N-butoxycarbonyl-aminobutyric
acid,
N-butoxycarbonyl-aminohexanoic acid, N-butoxycarbonyl-aminooctanoic acid and
N-butoxycarbonyl-aminolauric acid (Compound II in scheme 1) was recrystallized from hexane
(100 mL) at 4oC to obtain a white powder with yields of 71%, 72%, 35%, and 84%, respectively.
For the construction of 4a (1,5-dihexadecyl-N-lysyl-N-trityl-L-glutamate), 4b
(1,5-dihexadecyl-N-lysyl-N-pentyl-L-glutamate),
4c
(1,5-dihexadecyl-N-lysyl-N
-heptyl-L-glutamate), and 4d (1,5-dihexadecyl-N-lysyl- N-undecyl-L-glutamate), the
amino-group protected N-butoxycarbonyl-aminobutyric acid (0.751 g, 3.7 mmol),
N-butoxycarbonyl-aminohexanoic acid (0.855 g, 3.7 mmol), N-butoxycarbonyl-aminooctanoic
acid (0.958 g, 3.7 mmol) or N-butoxycarbonyl-aminolauric acid (1.17 g, 3.7 mmol) was
dissolved
in
dichrolomethane
(100
mL),
respectively.
Then,

Page 49
Chapter 2
- 39 -
benzotriazol-1-yloxytris(dimethylamino)phosphonium hexafluorophosphate (BOP, 4.07 mmol)
was added to the solution to activate the carboxyl group. After stirring with 1,5-dihexadecy
glutamate for 12 hr at room temperature, the solution was treated with a sodium carbonate
solution (100 ml x 2) and washed with distilled water (100 ml x 2). After the solution was
evaporated, 1,5-dihexadecyl-N-butoxycarbonyl-N-trityl-L-glutamate, 1,5-dihexadecyl-N-
butoxycarbonyl-N-pentyl-L-glutamate, 1,5-dihexadecyl-N-butoxycarbonyl- N-heptyl-L-glutamate,
or 1,5-dihexadecyl-N-butoxycarbonyl-N-undecyl-L-glutamate (Compound III in scheme 1) was
recrystallized from methanol (100 mL) at 4oC to obtain a white powder, respectively. The
powder was dissolved in chloroform (10 mL), and then trifluoroacetate (10 mL) was added to the
solution to remove the Boc group. After incubation for 2 hr at 4oC, the solution was treated with
a sodium carbonate solution (100 ml x 2) and washed with distilled water (100 ml x 2). After the
chloroform
solution
was
evaporated,
1,5-dihexadecyl-N-trityl-L-glutamate,
1,5-dihexadecyl-N-pentyl-L-glutamate,
1,5-dihexadecyl-N-heptyl-L-glutamate,
or
1,5-dihexadecyl-N-undecyl-L-glutamate (Compound IV in scheme 1) was obtained as a white
powder with yields of 58%, 48%, 56%, or 54% respectively.
1,5-dihexadecyl-N-trityl-L-glutamate (1 g, 1.47 mmol), 1,5-dihexadecyl-
N-pentyl-L-glutamate (1 g, 1.42 mol), 1,5-dihexadecyl-N-heptyl-L-glutamate (1 g, 1.37 mmol),
or 1,5-dihexadecyl-N-undecyl-L-glutamate (1g, 1.29 mmol) was added to dichloromethane
(100 mL), and then trimethylamine (1.51 mmol) and Boc-Lys(Boc)-OSu (1.5 mmol) were added
to each solution. After stirring for 12 hr, the solution was treated with a sodium carbonate
solution (100 ml x 2) and washed with distilled water (100 ml x 2). When the chloroform
solution was evaporated, 4a, 4b, 4c, or 4d was obtained after deprotection of Boc group with
TFA with yields of 57%, 35%, 45%, or 39%, respectively.

Page 50
Chapter 2
- 40 -
4a: Rf: 0.1 (CHCl3/CH3OH/H2O (65/25/4)), 1H-NMR (CDCl3, 500 MHz, δ ppm): 0.88 (t, 6H, CH2CH3);
1.20-1.34 (br, 54H, CH2), 1.42-1.52 (m, 2H, NHCH2CH2CH2CO), 1.54-1.71 (m, 6H, OCOCH2CH2,
NH2CH2CH2) 1.76-1.86 (m, 2H, NH2CH(CO)CH2), 1.98-2.13 (m, 2H, CONHCHCH2), 2.28 (t, 2H,
NHCOCH2), 2.35-2.47 (m, 2H, CH2COO), 2.92 (t, 2H, NH2CH2), 3.25-3.30 (m, 2H,
CONHCH2),3.43-3.47 (m, 1H, NH2CH), 4.05-4.12 (m, 4H, COOCH2), 4.50-4.57 (m, 1H, CONHCH),
7.36(d, 1H, CHNH), 7.82(t, 1H, CH2NH). MS(ESI): (M+H)+ calcd. for C47H92N4O6, 809.7; found, 809.9.
4b: Rf: 0.1 (CHCl3/CH3OH/H2O (65/25/4)), 1H-NMR (CDCl3, 500 MHz, δ ppm): 0.88 (t, 6H, CH2CH3);
1.20-1.35 (br, 58H, CH2), 1.40-1.51 (br, 4H, CONHCH2CH2, NH2CH2CH2) 1.52-1.68 (m, 6H,
OCOCH2CH2, NH2CH(CO)CH2), 1.81-1.99 (m, 2H, CONHCHCH2), 2.13-2.25 (br, 2H, NHCOCH2),
4a
4b
4c
4d
Fig. 2-7.
1H-NMR spectrum of cationic lipids (4a-4d) having a hydrocarbon type spacer between
head group and hydrophobic moieties of the lipid.

Page 51
Chapter 2
- 41 -
2.33-2.41 (m, 2H, CH2COO), 2.90-2.98 (br, 2H, NH2CH2), 3.16-3.25 (br, 2H, CONHCH2),3.75-3.90 (m,
1H, NH2CH), 4.02-4.13 (m, 4H, COOCH2), 4.53-4.56 (m, 1H, CONHCH), 7.09(d, 1H, CHNH), 8.19(t,
1H, CH2NH). MS(ESI): (M+H)+ calcd. for C49H96N4O6, 837.7; found, 837.8.
4c: Rf: 0.1 (CHCl3/CH3OH/H2O (65/25/4)), 1H-NMR (CDCl3, 500 MHz, δ ppm): 0.88 (t, 6H, CH2CH3);
1.14-1.34 (br, 60H, CH2), 1.40-1.52 (br, 2H, NHCOCH2CH2) 1.54-1.70 (m, 8H, OCOCH2CH2,
NH2CH2CH2, CONHCH2CH2), 1.75-1.82(m, 2H, NH2CH(CO)CH2), 1.92-2.24 (m, 2H, CONHCHCH2),
2.21 (t, 2H, NHCOCH2), 2.30-2.46 (m, 2H, CH2COO), 2.85 (t, 2H, NH2CH2), 3.16-3.24 (m, 2H,
CONHCH2), 3.34-3.76 (m, 1H, NH2CH), 4.02-4.15 (m, 4H, COOCH2), 4.57-4.61 (m, 1H, CONHCH),
6.47(d, 1H, CHNH), 7.52(t, 1H, CH2NH). MS(ESI): (M+H)+ calcd. for C51H100N4O6, 865.8; found, 866.1.
4d: Rf: 0.1 (CHCl3/CH3OH/H2O (65/25/4)), 1H-NMR (CDCl3, 500 MHz, δ ppm): 0.88 (t, 6H, CH2CH3);
1.14-1.34 (br, 68H, CH2), 1.40-1.52 (br, 2H, NHCOCH2CH2) 1.52-1.71 (m, 8H, OCOCH2CH2,
NH2CH2CH2, CONHCH2CH2), 1.78-1.86(m, 2H, NH2CH(CO)CH2), 1.94-2.24 (m, 2H, CONHCHCH2),
2.21 (t, 2H, NHCOCH2), 2.30-2.46 (m, 2H, CH2COO), 2.80 (t, 2H, NH2CH2), 3.10-3.24 (m, 2H,
CONHCH2), 3.33-3.40 (m, 1H, NH2CH), 4.02-4.14 (m, 4H, COOCH2), 4.58-4.62 (m, 1H, CONHCH),
7.60(d, 1H, CHNH), 8.21(t, 1H, CH2NH). MS(ESI): (M+H)+ calcd. for C55H108N4O6, 1026.8; found,
1026.9.

Page 52
Chapter 2
- 42 -
References
1. Vigneron, J.-P., Oudrhiri, N., Fauquet, M., Vergely, L., Bradly, J.-C., Basseville, M., Lehn, P.,
Lehn, J.M. Proc. Natl. Acad. Sci. U.S.A 93 (1996) 9682–9686
2. Felgner P.L, Gadek T.R., Holm M., Roman, R., Chan, H.W., Wenz M, Northrop, J.P., Ringold,
G.M., Danielsen, M. Proc. Natl. Acad. Sci. U.S.A 84 (1987) 7413–7417.
3. Shimizu, T. Polym. J. 35 (2003) 1–22.
4. Yamada, N., Ariga, K., Naito, M., Matsubara, K., Koyama, E. J. Am. Chem. Soc. 120 (1998)
12192–12199.
5. Lauf, U., Fahr, A., Westesen, K., Ulrich, A. S. Chem. Phys. Chem. 5 (2004) 1246–1249.
6. Palmer, L.R., Chen, T., Angela, M.I. Lam, D.B. Fenske, K.F. Wong, I. MacLachlan, P.R.
Cullis, P.R. Biochim. Biophys. Acta 1611 (2003) 204–216
7. Bajaj, A., Kondaiah, P., Bhattacharya, S. Biomacromol. 9 (2008) 991–999.
8. Bhattacharya, S., De, S. Chemistry - A European J. 5 (1999) 2335–2347.

Page 53
Chapter 3
- 43 -
Chapter 3
Correlation between Structure of Amino-acid Base Lipid and
Transfection Efficiency in Vitro for Efficient Plasmid DNA Delivery
1. Introduction
In previous chapter, the author synthesized a series of amino-acid based lipids with
varying head group as hydrophilic moiety and length of alkyl chains as hydrophobic moiety. The
previous studies indicated that lipid structure was most important factor to obtain high gene
expression.
In this chapter, gene expression efficiencies of cationic assemblies containing amino-acid
based lipids are introduced to clarify correlation between lipid structure and gene expression
efficiency in vitro. Then, assemblies containing amino-acid base lipids are investigated in terms
of complexation profiles with plasmid DNA (pDNA) and the dispersion state of the resulting
lipoplexes by altering the cationic head group and length of the alkyl chains of the cationic lipids.
Lipoplexes containing pDNA-encoding luciferase are prepared with a series of amino-acid based
lipids and various lipid-to-pDNA ratios. The appropriate structure is then soaked.

Page 54
Chapter 3
- 44 -
2. Barriers to Intracellular Trafficking in Gene Delivery
While cationic liposomes have the advantage of an excellent safety profile, the major
drawback to these vectors is that they are not as efficient in gene expression as the viral vector
(retrovirus, adenovirus, adeno-associated virus, herpes simplex, etc…) delivery system. In order
to improve these inefficiency, several investigators have examined the cellular trafficking of
DNA-liposome complexes. Lipoplex must surmount many barriers to successfully deliver DNA
to the cell nucleus. A diagram of the most likely mechanism of lipofection is shown in Fig. 3-1.
The first stage is binding of a lipoplex to the cell due to nonspecific electrostatic interactions
between cationic membranes and anionic proteoglycan residues on the cell surface 1. Cell
association is enhanced for positively charged lipoplexes and for large aggregates, the latter
probably due to a faster rate of settling on the cultured cells 2. The most efficient DNA/lipid
mixing ratio for lipofection is a moderate excess of cationic charge 3.
The primary pathway for lipoplex entry into cells is endocytosis 4,5. Fusion of lipoplexes
with the cell membrane also occurs, but there is no correlation between the degree of lipid
Lysosomes
Endosomes
Cells
Endocytosis
Nucleus
Lysosomes
Endosomes
Cells
Endocytosis
Nucleus
1 Formation of a homogeneous stable complex
2 Uptake of the complex by the cell (endocytosis)
3 Release/escape from the endosome
4 Uncoating of the lipid from the DNA
5 Transcription/expression
Gene expression
Fig. 3-1. Intracellular gene delivery by lipoplexes and potential barriers to provide efficient
gene transfer.

Page 55
Chapter 3
- 45 -
mixing and the lipofection efficiency 6. Possibly a fusion event releases DNA outside the cell, or
inside the cell, but too distant from the cell nucleus.
A recent report suggested that small particles (200 nm or below) enter cells mainly by
clathrin-madiated endocytosis and larger particles (around 500 nm) by caveolae-mediated
endocytosis, while vary large particles (1 μm or above) are not internalized at all 7. Small particle
are delivered to lysosomes for digestion with a few hours, but larger particles have an extended
residence time within endosomal compartments, increasing their probability of escape into the
cytosol 7. These results may explain the common empirical finding that lipoplex size has a
significant influence on transgene expression. There is an optimal lipoplex size for greatest
lipofection efficiency 8–10.
Release of lipoplexes from endosomal
compartments is widely considered a
rate-determining step for lipofection. Efficient
lipoplex formations are able to escape from
endosomes, while inefficient formulations
remain trapped 11,12. The endosome membrane
can be disrupted by lipid exchange or fusion
with lipoplexes as shown in Fig. 3-2. Inverse
hexagonal phase lipoplexes are reported to fuse
rapidly with anionic membrane, disrupting the
membrane and releasing DNA 13 and hexagonal
lipoplex structure has been correlated with
efficient lipofection 14,15. Lipid missing between
cationic lipids and anionic lipids found in the
endosomal membrane can also induce local
formation of a hexagonal phase, due to ion pair charge neutralization 16,17. This destabilization of
Fig. 3-2. Hypothesis of endosomal escape
of lipoplexes gene delivery systems.

Page 56
Chapter 3
- 46 -
bilayers is hindered by the presence of cylindrically shaped helper lipid, but is facilitated by
inverse conical shaped helper lipid. Formation of a non bilayer phase in lipoplex-membrane
mixtures appears to be critical for efficient endosome escape 17.
3. Evaluation of Cationic Assemblies Composed of Amino-acid Based Lipids for Plasmid
DNA Delivery in COS-7 cells
3-1. Lipoplex Formation
The author analyzed lipoplexes formation between the cationic assemblies by varying the
lipid-to-pDNA ratio using agarose gel retardation assays (Fig. 3-3). The abilities of all the
cationic assemblies were enough to form lipoplexes. Excess pDNA was barely detectable using a
lipid-to-pDNA ratio of 3 in the cationic assemblies of 1a, 1b and 3a-3c. For 1c, 2b and 2c, the
lipid-to-pDNA ratio had to be increased to 20, 10 and 10, respectively. Having prepared the
lipoplexes, spontaneous pDNA release into the solution was scarcely detectable even after
several hours.
1a
1b
1c
2b
2c
3a
3b
3c
1
3
5 10 20 50
DNA
1
3
5 10 20 50
DNA
1
3
5 10 20 50
DNA
1 3
5 10 20 50
DNA
1 3
5 10 20 50
DNA
1 3
5 10 20 50
DNA
1 3
5 10 20 50
DNA
1 3
5 10 20 50
DNA
Fig. 3-3. Gel retardation of lipoplexes constructed with cationic assemblies and pDNA at
various lipid-to-pDNA ratios (w/w).

Page 57
Chapter 3
- 47 -
All the cationic assemblies were confirmed to form lipoplexes with pDNA from a gel
retardation assay. When the lipid-to-pDNA ratio was >3, no excess pDNA was detected for 1a,
1b, 3a and 3b (the estimated N/P ratio of 2.7). Because the unilamellar vesicles comprise both an
outer and inner layer, the more accurate N/P ratio is about 1.4 if there were no change of inner-
and outer-lipid composition after lipoplex formation. Therefore, a proportion of the amino
groups of the outer layer of the liposome do not have access to the phosphate groups of the
pDNA. When 1c was utilized as a component of the cationic assembly, more lipids were required
to form lipoplexes. Indeed, the size of cationic assembly of 1c was large in comparison to that of
1a or 1b. If 1c would take a unilamellar structure, the portion of the amino groups of the vesicle,
which can access to pDNA, should be equal to those of 1a and 1b. However, more 1c lipid was
required to form a lipoplex with the pDNA. Therefore, the author regarded the cationic
assemblies of 1c as multilamellar structures. Therefore, the author defined the cationic
assemblies as multilamellar structures. Cationic assemblies from the histidine-type lipids formed
a stable complex with the pDNA at a lipid-to-pDNA ratio of 10 (N/P=8). More histidine-type
lipid based cationic assemblies were required to form lipoplex than the lysine- or arginine-type
lipids. Based on pKa of the different amino groups, electrostatic interaction of the tertiary amino
group of histidine will be much weaker than that of the primary amino group of lysine or
arginine. Thus, the author might anticipate that the competence of the histidine-type lipids to
form lipoplexes will be relatively low. Indeed, this conclusion is further supported by previous
studies using cationic amino-acid derivatives derived from tryptophan-type lipids, which
exhibited low levels of gene expression in transfection experiments 18. Moreover, tube-like
assemblies might be unable to form a lipoplex with pDNA. No further analysis of the histidine
type-lipids was performed in this study. Therefore, the analysis shows that assemblies with a
lysine or arginine head-group are the most appropriate to form lipoplexes.

Page 58
Chapter 3
- 48 -
3-2. Influence of a Lipid-to-pDNA Ratio on Dispersion State
In particular, the author investigated the lipoplexes composed of 1a and 3a with respect
to size distribution and zeta-potential (Fig. 3). The apparent size of the lipoplex composed of 1a
and pDNA continuously increased with increasing lipid-to-pDNA ratios from 1 to 5. However,
any further increase in the lipid-to-pDNA ratio resulted in a reduction in the size of lipoplex. The
90 nm unilamellar vesicles of 1a increased in size to 390 nm lipoplexes at a lipid-to-pDNA of 10.
The zeta-potential of the lipoplex composed of 1a dramatically increased as the lipid-to-pDNA
ratio was raised from 1 to 3. However, additional increases in the lipid-to-pDNA ratio from 3 to
50 resulted only in a gradual rise in zeta-potential, reaching a maximum of around +40 mV. The
size of the lipoplexes composed of 3a also increased as the lipid-to-pDNA ratio was raised to 5,
although the zeta-potential was constant (+25 mV) at ratios greater than 5.
3-3. Transfection Efficiency of Amino-acid Based Lipids
The author evaluated the gene transferring efficiency of lipoplexes composed of the
synthetic lipids using COS-7 cells (Fig. 3-5). Lipoplexes composed of each cationic assembly
and luciferase-encoding pDNA were prepared by varying the lipid-to-pDNA ratio with fixing the
Partic
le
S
iz
e
[nm]
0
400
600
800
200
-60
-20
20
-40
40
60
0
1000
size
zeta potential
Zeta potential [m
V]
Z
e
ta
p
o
tentia
l [m
V]
0
400
600
800
200
-60
-20
20
-40
40
60
0
1000
size
Zeta potential
P
a
rticle
S
ize [n
m]
A
B
1
3
5
10
20
2a / pDNA (w/w)
50
1
3
5
10
20
4a / pDNA (w/w)
50
Fig. 3-4. The size distribution and zeta-potential of the lipoplex constructed with (A) 1a or (B)
3a lipid. The lipoplexes were dispersed in distilled water at 37oC.

Page 59
Chapter 3
- 49 -
pDNA concentration of 2 μg/mL. The lipoplexes were then transfected into COS-7 cells. For the
lysine-type lipids, gene expression efficiency exhibited a maximum value at a 1a-to-pDNA ratio
of 10, which was 2.2-fold higher compared with that for LipofectamineTM2000 reagent in the
absence of FBS (Fig. 3-5(A)). The gene expression level after transfection with 1a, 1b and 1c
tended to increase with shorter alkyl chains. The gene transferring efficiency with 1c was the
lowest of the three lysine-type lipids. In addition, the author investigated the gene transferring
efficiencies of cationic assemblies prepared from the other cationic lipids having a histidine or an arginine as
the cationic head-group. Cationic assemblies formed from histidine-type lipids 2b or 2c did not give detectable
levels (ca. 200 RLU at all lipid-to-pDNA ratios) of gene expression. The gene expression efficiencies for the
arginine-type lipids were as high as those obtained for the lysine-type lipids. The arginine-type 3a3c also
showed a maximum gene expression efficiency at a lipid-to-pDNA ratio of 10 (Fig. 3-5(B)). Furthermore, the
gene expression efficiency for 3a-3c increased as the length of the alkyl chain was reduced.
Comparison of gene transfer efficiency with respect to the head group of the cationic
lipids is as follows: lysine ≥ arginine > histidine. Thus, both lysine and arginine form appropriate
head-groups to bind pDNA and display a high affinity for the cell membrane. Although the
1
102
103
Lu
cife
ra
s
e
ac
tiv
ity
[R
LU
/ μ
g-Pro
tein
]
1a
1b
1c
LA2000
pDNA
1
3
5
10
20
cationic lipid / pDNA (w/w)
10
50
104
A
1
102
103
Lu
cife
ra
s
e
ac
tiv
ity
[R
LU
/ μ
g-Pro
tein
]
3a
3b
3c
pDNA
10
104
B
LA2000
1
3
5
10
20
cationic lipid / pDNA (w/w)
50
Fig. 3-5. Gene expression efficiency of cationic assemblies containing (A) lysine type lipid;
1a1c, and (B) arginine type lipids; 3a3c.

Page 60
Chapter 3
- 50 -
lipoplex formation using histidine-type lipids was confirmed by gel retardation, the level of gene
expression was quite low. This low level of gene expression might be due to poor cellular uptake
of the tube-like assemblies, which were several micrometers in length and 10 nm in diameter.
Furthermore, after endocytosis the electrostatic interaction between the tertiary amino group of
the histidine and the phosphate group of pDNA is presumably very weak in the acidic
environment of the endosome. Thus, delivery of pDNA into the nucleus will be very inefficient.
The gene expression efficiencies of synthetic cationic lipids were also dependent on the
length of alkyl chain. Order of gene expression efficiency in terms of the number of carbon
atoms in the alkyl chain was as follows: 14 > 16 > 18. The gene expression efficiency of a
lipoplex is generally known to be dependent on the Tc of the cationic lipid19,20. Two independent
studies involving a series of cationic lipids of different alkyl chain length showed the expression
efficiency to be: C14 > Coleyl > C18 > C16 > C12 by Felgner et al.
21 or C14 > Coleyl > C16 > C18 by
Floch et al.
22,23. However, because lipoplexes utilize helper lipids, such as DOPE or cholesterol,
the fluidity of the liposomal membrane is not solely dependent on the fluidity of the pure
cationic lipids. Lipid membranes of high fluidity tend to promote fusion with a biological
membrane, leading to high gene expression efficiency19,20,24. Thus, the author also analyzed the
fusogenic potential of the cationic assemblies using a model membrane that mimics a
biomembrane. The fusogenic potentials of the lipoplexes from 1c (Tc; 53.2oC) or 3c (Tc; 52.7oC)
with low fluidity were quite low, resulting in quite low gene expression efficiency in comparison
with that of 1a (Tc; 26.3oC) or 3a (Tc; 24.7oC) of high fluidity. In this study, cationic assemblies
with high fluidity led to high gene expression.
3-4. Fusogenic Potential of Cationic Assemblies
The fusogenic potential of the cationic assemblies to biomembrane mimicking liposomes
was investigated (Fig. 3-6). The author confirmed that the difference in the fusogenic potential of
the prepared cationic assemblies was based on fluidity. The transition temperatures of the

Page 61
Chapter 3
- 51 -
lysine-type lipids 1a, 1b and 1c were 26.3, 43.3 and 53.2oC, respectively as described in Chapter
2. For the arginine-type lipids 3a, 3b and 3c, the transition temperatures were 24.7, 42.5 and
52.7oC, respectively. Similar analyses of 2b and 2c were conducted to obtain the transition
temperatures of 34.4 and 42.3oC, respectively. These results indicate that the transition
temperature of the cationic assemblies increased with increasing alkyl chain length. Thus, the
hydrophobic moiety appears to have a direct effect on fusogenic potential. There was no
significant difference in transition temperature between the lysine-type and arginine-type series
of lipids having the same alkyl chain length. Among the lysine-type lipids, 1a exhibited the
highest fusogenic potential (Fig. 3-6(A)). The fusogenic ratio of 1a gradually increased and
reached about 13% after incubation for 30 min at 37oC. 1b showed a slightly lower fusogenic
potential than 1a. However, the fusogenic potential of 1c was significantly diminished compared
to 1a and 1b. A similar trend was also observed for the arginine-type lipids (Fig. 3-6(B)). The
fusogenic potential of 3a and 3b were almost the same, whereas that of 3c was significantly
lower. The fusogenic potentials of the lysine-type lipids were slightly higher than those of the
arginine-type lipids. These data would highlight the influence of the hydrophobic moiety on
fusogenic potential.
Lip
id mixing ratio %
0
10
15
5
A
B
20
5
10
15
20
25
30
0
Lip
id
mixing
ra
tio
%
0
10
15
5
20
5
10
15
20
25
30
0
Time (min)
Time (min)
Fig. 3-6. (A) Fusogenic potential between the cationic assemblies from 1a (circle), 1b (square)
or 1c (triangle) and biomembrane mimicking liposome at 37oC. (B) Fusogenic potential
between the cationic assemblies from 3a (circle), 3b (square) and 3c (triangle).

Page 62
Chapter 3
- 52 -
3-5. Influence of Serum Proteins on Gene Expression Efficiency
In addition, gene expression efficiency with Lipofectamine2000 was considerably
reduced in the presence of FBS (Fig. 3-6(A)). By contrast, gene expression levels determined
using lipoplexes formed from 1a-1c displayed no significant difference in the presence/absence
of FBS. In addition, the gene expression efficiency with the arginine-type lipids was not
significantly affected by the presence of FBS (Fig. 3-6(B)). Consequently, in the presence of
FBS the expression level of 1a at a lipid-to-pDNA ratio of 10 was 12-fold higher than that
observed using lipofectamineTM2000.
A low level of gene expression efficiency in the presence of FBS has been a major barrier
to the medical application of gene therapy 25. In general, the phospholipid molecules in the
liposomal membrane are thought to be removed by high density lipoproteins (HDL) in serum 26.
This would result in poor fusogenic ability to the biological membrane in the presence of FBS,
leading to low gene expression efficiency, particularly when DOPE was utilized as a liposomal
component. In this study, the reduction in gene expression with DOPE-containing transgenic
1
102
103
Lu
cife
ra
s
e
ac
tiv
ity
[R
LU
/ μ
g-Pro
tein
]
1a
1b
1c
10
104
A
LA2000
pDNA
1
3
5
10
20
cationic lipid / pDNA (w/w)
50
1
102
103
Lu
cife
ra
s
e
ac
tiv
ity
[R
LU
/ μ
g-Pro
tein
]
3a
3b
3c
10
104
B
LA2000
pDNA
1
3
5
10
20
cationic lipid / pDNA (w/w)
50
Fig. 3-5. Gene expression efficiency of cationic assemblies containing (A) lysine type lipid;
1a1c, and (B) arginine type lipids; 3a3c in presence of fetal bovine serum.

Page 63
Chapter 3
- 53 -
reagents was also confirmed in the presence of FBS. However, cationic assemblies formed from
the synthetic lipids under optimal conditions displayed no reduction in gene expression
efficiency in the presence of FBS. This observation suggests that the amino-acid based cationic
lipids are not influenced by FBS. Thus, the author succeeded in constructing cationic carriers
based on non-DOPE.
3-6. Cytotoxicity of Amino-acid Based Lipids
The author investigated the cytotoxicities of the lipoplexes constructed with the various
synthetic cationic lipids (Fig. 3-6). All cationic lipids were revealed to have low toxicity.
Lipoplex composed of lipofectamine2000 reagent at a lipid-to-pDNA ratio of 3 was toxic to
around 100% of COS-7 cells at a lipid concentration of 100 μg/ml. By contrast, lipoplexes
constructed of the cationic assemblies of 1a, 1b, 3a or 3c exhibited significantly lower toxicities.
For example, more than half of the cells seeded on a culture dish survived even in the presence
of 1 mg/ml lipid concentration of lipoplex (i.e., ~50 times higher lipid concentration compared to
the experiment using LipofectamineTM2000).
20
40
60
80
100
Ce
ll Su
rviv
al %
0
120
1
0.1
100
10
1000
1a
3a
LA2000
1b
3b
Lipid [μg/ml]
Fig. 3-6. Cytotoxicity of COS-7 cells transfected with lipoplexes constructed using synthetic
cationic lipids and pDNA compared with lipofectamineTM2000 reagent. Lipoplexes were added
to COS-7 cells and incubated for 24 hr. Cytotoxicity was estimated using the WST assay.

Page 64
Chapter 3
- 54 -
The cytotoxicity of the lipoplexes was evaluated using COS-7 cells. Compared with
conventional LipofectamineTM2000, cytotoxicity of the lipoplexes constructed with the synthetic
amino-acid based cationic lipids was very low. Toxicity of the conventional reagent is probably
related to the presence of multivalent cationic compounds such as DOGS or DOSPA.
Specifically, after dissociation of the pDNA the multivalent cationic compounds aggregate with
various intracellular organelles, leading to cell death. By contrast, amino-acid based cationic
lipids are easily dissociated and metabolized. Furthermore, the ester bond in the linker region of
the amino-acid based cationic lipids is readily hydrolyzed in the cell, thereby lowering their
cytotoxicity 27.
4. Neuronal Transfection with Amino-acid Based Lipids
4-1. Neuronal Transfection
One of the current challenges for those studying the use of cationic liposomes for gene
delivery is the preparation of lipoplexes that appropriate for the transfection of neuronal cells and
primary cultures. Difficulties that are particular to the transfection of neuronal cells are caused
by the post-mitotic status of these cells and their high sensitivity to the cytotoxicity of the gene
carrier. Previous transfection studies on neuronal cells showed a low transfection efficiency
when
cationic
liposomes
composed
of
cationic
lipids
such
as
N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium
methyl-sulfate
(DOTAP),
N-[1-(2,3-dioleyl)propyl]-N,N,N-trimethylammonium
chloride
(DOTMA),
and
2,3-dioleoyloxy-N-[2(sperminecarboxamide)ethyl]-N,N-dimethyl-1-propanaminium
trifluoroacetate (DOSPA) were used28-31, even if a high transfection efficiency was obtained with
immortalized cell lines 27,32,33. With respect to cationic lipids, we have demonstrated the
importance of amino-acid-based cationic lipids for gene and protein delivery 34,35. Cationic lipids
that contain lysine or arginine in the headgroup have already been shown to give a higher
transfection efficiency than that of LipofectamineTM2000, which contains DOSPA, and their

Page 65
Chapter 3
- 55 -
cytotoxicity is also remarkably low. These data suggest that such lipids could be very effective
for the transfection of neuronal cells.
In this section, the author particularly analyzed the transfection efficiency of the cationic
amino-acid-based lipids in a neuronal cell line, with the ultimate aim of identifying effective gene carriers for
the central nervous system. The author used lipoplexes that were formed from cationic liposomes composed of
1,5-dihexadecyl N-arginyl-L-glutamate (3b) and pDNA. The lipoplexes were then investigated in terms of
lipoplex formation, transfection efficiency, intracellular trafficking, and cytotoxicity in SH-SY5Y cells in order
to allow the optimization of lipoplexes that contain amino-acid-based lipids for efficient neuronal transfection.
4-2. Transfection Efficiency on Neuronal Cell Line
The transfection efficiency of the lipoplexes at different lipid-to-DNA ratios was
investigated using SH-SY5Y cells. In the case of naked pDNA, no GFP expression was detected.
Among the ratios that were investigated (5, 10, 15, and 20), a lipid-to-DNA ratio of 15 gave the
highest transfection efficiency, with 16% of the cells expressing GFP (Fig. 3-7). A lipid-to-DNA
ratio of 15 corresponds to an N/P ratio (ratio of nitrogen atoms in the liposomes to phosphorus
atoms in the pDNA) of 12. Moreover, the number of cells that expressed GFP after transfection
with the 3b lipoplexes at this ratio was approximately four times greater than the number
obtained using LipofectamineTM2000 (4%).
5
10
15
20
Lipid-to-DNA ratio (wt/wt)
LA®2000
pDNA
only
Fig. 3-7. Gene expression efficiency of 3b liposomes in human neuronal SH-SY5Y cells.
The lipoplexes varied a lipid-to-pDNA ratio and added to the cells.

Page 66
Chapter 3
- 56 -
Previous studies on the effect of the lipid-to-DNA ratio on the size and zeta potential of
the lipoplexes indicated that lowering the lipid-to-DNA ratio decreases the zeta potential, which
results in aggregation. As a consequence, at a low lipid-to-DNA ratio, cellular uptake is likely to
be suppressed both by the weak positive charge of the lipoplexes, which would result in low
electrostatic interaction of the lipoplexes with the negatively-charged cell membrane, and by the
aggregation of the lipoplexes, because of low phagocytic uptake due to the large aggregates.
Indeed, lipoplexes with lipid-to-DNA ratios from 1 to 10 showed a lower transfection efficiency
than lipoplexes with a lipid-to-DNA ratio of 15. On the other hand, increasing the lipid-to-DNA
ratio to more than 20 would result in the presence of free 3b liposomes. The free cationic
liposomes would be endocytosed preferably by the cells because their size (ca. 100 nm) is
smaller than that of the lipoplexes (ca. 400 nm). Therefore, the free liposomes should inhibit the
cellular uptake of the 3b lipoplexes, and result in a decrease in the transfection efficiency.
Wangerek et al. reported the transfection efficiency of SH-SY5Y cells with the lipid-based
transfection reagents Lipofectin® and LipofectamineTM2000 36. They detected less than 5%
GFP-positive cells, which was similar to the value that the author obtained in this study using
LipofectamineTM2000. Considering the higher transfection efficiency that was obtained with 3b
compared to LipofectamineTM2000, 3b liposomes could be expected to show a higher
transfection efficiency than commercially-available transfection reagents. The 3b liposomes
were also tested on glioblastoma multiform cells, and 5% of the transfected cells were found to
express GFP (data not shown). An interesting result is the fact that the transfection efficiency
varies between different cell lines, as shown by COS-7, SH-SY5Y, and glioblastoma multiform
cells. This difference may be caused by different cellular uptake efficiencies of the lipoplexes.
Because the charge of the cell membrane would be different in each cell line37,38, each cell
membrane type would show a different degree of electrostatic interaction with the lipoplexes,
and this could affect the cellular uptake efficiency. Another reason for the difference could be
the duplication time of each cell type, which would affect the entry of the pDNA into the nucleus

Page 67
Chapter 3
- 57 -
39,40. In summary, 3b liposomes could achieve higher transfection efficiencies for neuronal cells
than commercially-available transfection reagents.
In addition, the author investigated the effect of the concentration of lipoplexes with a
lipid-to-pDNA ratio of 15 on the transfection efficiency for neuronal cells. An increase in GFP
expression was obtained as the concentration of the lipoplexes in the medium was increased
from 2 to 20 μg/mL as a pDNA concentration. As the amount of the lipoplexes was increased,
GFP expression increased correspondingly until pDNA concentration of 20 μg/mL,
approximately 25% of the cells were GFP-positive (Fig. 3-7).
Fig. 3-7. (A) Microscopic images of neuronal SH-SY5Y cells at 48 hr after the addition of the
lipoplexes with a lipid-to-pDNA ratio of 15. The pDNA concentration was varied from 2 to 20
μg/mL and the amount of lipoplexes changed correspondingly. (B) Transfection efficiency of the
Arg-Glu2C16 lipoplexes with a lipid-to-pDNA ratio of 15 in SH-SY5Y cells.
20
16
8
4
2
Amount of DNA (μg/mL)
Fluorescent
image
Bright field
A
10
5
0
25
20
15
30
4
8
12
16
20
pDNA [μg/mL]
0
Prop
ortio
n
of G
F
P-pos
itive c
e
lls
(%
)
B

Page 68
Chapter 3
- 58 -
In this study, expression of the exogenous gene increased as the amount of lipoplexes
was increased, over a DNA concentration range of 2–20 μg/mL. This is presumably due to
increased uptake by endocytosis when the cells are exposed to a higher concentration of
lipoplexes. Owing to the low cytotoxicity of the Arg-Glu2C16 liposomes, it was possible to apply
a large amount of 3b lipoplexes to the cells in order to increase expression of the exogenous gene.
By contrast, conventional transgenic reagents such as LipofectamineTM2000 could not be applied
at such high concentrations due to their high cytotoxicity. Therefore, the author stress the
advantages of using 3b for neuronal transfection in terms of both transfection efficiency and
cytotoxicity. The association of liposomes with transferrin (Tf) has been used to improve
transfection efficiencies further 41,42. When neuronal cells were transfected with
Tf-DOTAP/cholesterol liposomes, four-fold higher expression of the exogenous gene was
observed than in cells that have been transfected with DOTAP/cholesterol liposomes. This
enhancement is due to increased cellular uptake of the Tf-lipoplexes. It should be possible to
enhance the transfection efficiency of neuronal cell lines further by modifying the Arg-Glu2C16
lipoplexes with Tf.
5. Experimental Section
Gel retardation assay of the lipoplexes
The lipoplexes prepared by varying the lipid-to-pDNA ratio were applied to a 1%
seaplaque GTG agarose gel (Takara Bio. Inc., Otsu, Japan) in 1% Tris-acetate-EDTA
(Invitrogen). After electrophoresis for 30 min, the agarose gel was rinsed in an ethidium bromide
(EtBr) solution for 10 min. The gel was then rinsed in distilled water for 15 min and visualized
on a UV transilluminator (GelDoc XR system; Bio-Rad, Hercules, CA).
Measurement of gene expression efficiency of the lipoplexes by COS-7 cells
COS-7 cells (transformed African green monkey kidney fibroblast cells) were utilized for

Page 69
Chapter 3
- 59 -
evaluating the gene expression efficiency of various lipoplexes. The COS-7 cells (1 x 104 cells)
were seeded on 96-well plates and incubated at 37oC under 5% CO2 for 12 hr. The COS-7 cells
were cultured with DMEM containing 10% FBS. The medium in the cell culture dishes was
changed to a fresh medium (100 μL) in the presence of lipoplexes containing 2 μg/mL of pDNA
as a final concentration. After incubation at 37oC for 24 hr, the cells were washed twice with an
ice-cold PBS solution and then lysed with a lysis buffer from the luciferase assay kit (Promega,
Madison, WI). The luciferase activity in a 10 μL aliquot of the cell lysate was measured with a
Microlumat Plus (EG&G Berthold, BadWilbad, Germany). The protein concentration of each
well lysate was determined by a standard protein assay (Bio-Rad Protein Assay; Bio-Rad). The
luciferase activity in each sample was normalized to the relative light unit (RLU) per microgram
of protein.
Fusion assay of the cationic assemblies with model membrane
The membrane fusion between the cationic assemblies and liposomes with
biomembrane-mimicking lipid composition were investigated to explore the gene transfer
mechanism of the prepared lipoplexes by using a fluorescence resonance energy transfer (FRET)
assay. The model liposomes constructed with DOPC/DOPE/DOPS/cholesterol (45:20:20:15,
wt%) were labeled with N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)-phosphatidylethanolamine
(NBD-PE) and rhodamine-phosphatidylethanolamine (Rho-PE), of which concentrations were 1
mol%. For the fusion assay, 500 μl of the labeled liposomes (500 μM as lipid concentration) was
added to 500 μL of the prepared lipoplexes (500 μM as lipid concentration). The mixed solution
was then incubated at 37oC for an appropriate period of time. Fluorescence intensity was
recorded using an excitation wavelength of 460 nm and an emission wavelength of 525 nm. The
time course of the lipid mixing percentage between the model membrane and the cationic
assemblies was determined using the following equation:
Lipid mixing % = ((Ft F0 ) / (FTX F0 )) x 100

Page 70
Chapter 3
- 60 -
where Ft is the fluorescence intensity at an appropriate time and F0 is the fluorescence intensity
just after dilution with distilled water and FTX is the fluorescence intensity after the addition of
Triton X-100 to solubilize the liposomes.
Cytotoxicity of the lipoplexes in COS-7 cells
The cytotoxicity of the lipoplexes constructed with the cationic assemblies and pDNA
was investigated with COS-7 cells. The COS-7 cells (1 x 104 cells) were seeded on a 96-well
plate and incubated at 37oC under 5% CO2 for 12 hr. The COS-7 cells were cultured with
DMEM containing 10% FBS. The medium in the cell culture dish was changed to a fresh
medium (100 μL) containing an appropriate concentration of the lipoplexes diluted with a
medium containing 10% FBS. After incubation at 37oC for 24 hr, the medium was changed to a
medium (110 μL) containing a tetrazolium salt (WST-1) and incubated for 30 min. Formazan
(absorbance at 540 nm) is then produced by succinate tetrazolium reductase in living cells. The
absorbance at 540 nm was monitored using a microplate reader (Perkin Elmer Japan Co. Ltd,
Tokyo, Japan).
Neuronal transfection
SH-SY5Y cells were utilized for evaluating the transfection efficiency of the lipoplexes.
The SH-SY5Y cells (2 x 104 cells) were seeded on 24-well plates and incubated under standard
conditions for 24 h. The medium in the cell culture dishes was exchanged with fresh medium
(500 μL) that contained lipoplexes at the different lipid-to-pDNA ratios. After incubation for 4 hr,
the medium was exchanged with fresh medium that contained 10% FBS. The cells were
incubated for a further 44 hr and then washed twice with phosphate-buffered saline (PBS). Cells
were observed using a Nikon TE2000U fluorescent microscope equipped with a Nikon DS-5MC
USB2 cooled CCD camera (Nikon, Tokyo, Japan). The total number of cells and the number of
GFP-positive cells in three fields were counted to estimate the transfection efficiency. All the

Page 71
Chapter 3
- 61 -
experiments that involved the lipoplexes were performed in triplicate. In order to determine the
optimal lipid-to-DNA ratio, the concentration of pDNA in the medium was fixed at 8 μg/mL.
This concentration was also used in the control experiments with LipofectamineTM2000.
Furthermore, in order to investigate the appropriate amount of lipoplexes for neuronal
transfection, the concentration of the lipoplexes with a lipid-to-pDNA ratio of 15 was varied
from 2 to 20 μg/mL and the transfection study was again performed.

Page 72
Chapter 3
- 62 -
References
1. Mislick, K.A., Baldeschwieler, J.D. Proc. Natl. Acad. Sci. U.S.A. 93 (1999) 12349–12354.
2. Girão da Cruz, M.T., Simões, S., Pires, P.P.T., Nir, S., Pedrosa de Lima, M.C. Biochim.
Biophys. Acta 1510 (2001) 136–141.
3. Sakurai, F., Inoue, R., Nisino, Y., Okuda, A., Mtsumoto, O., Taga, T., Yamashita, F., Takakura,
Y., Hashida, M. J. Cont. Release 66 (2000) 255–269.
4. Zabner, J., Fasbender, A.J., Moninger, T., Poellinger, K.A., Welsh, M.J. J. Biol. Chem. 270
(1995) 18997–19007.
5. Friend, D.S., Papahadjopoulos, D., Debs, R.J. Biochim. Biophys. Acta 1278 (1996) 41–50.
6. Stegmann, T., Legendre, J.-Y. Biochim. Biophys. Acta 1325 (1996) 71–79.
7. Rejman, J., Oberle, V., Zuhorn, I.S., Hoekstra, D. Biochem. J. 377 (2004)159–169.
8. Kawaura, U., Noguch, A., Furuno, T., Nakanishi, M. FEBS Lett. 421 (1998) 69–72.
9. Mui, B., Ahkong, Q.F., Chow, L., Hope, M.J. Biochim. Biophys. Acta1467 (2000) 281–292.
10. Lin, A.J., Slack, N.L., Ahmad, A., George, C.X., Samuel, C.E., Safinya, C.R. Biophys. J. 84
(2003) 3307–3316.
11. Zuidam, N.J., Hirsch-Lerner, D., Margulies, S., Barenholz, Y. Biochim. Biophys. Acta 1419
(1999) 207–220.
12. Ross, P.C., Hui, S.W. Gene Ther. 6 (1999) 651–659.
13. Lin, A.J., Slack, N.L., Ahmad, A., Koltover, C.X., George, C.X., Samuel, C.E., Safinya, C.R.
J. Dug Target 8 (2000) 13–27.
14. Koltover, I., Salditt, T., Rädler, J.O., Safinya, C.R. Science 281 (1998) 78–81.
15. Smisterova, J., Wagenaar, A., Stuart, M.C.A., Polushkin, E., Ten Brinke, G., Hulst, R.,
Engberts, J.B.F.N., Hoekstra, D. J. Biol. Chem. 276 (2001) 47615–47622.
16. Hafez, I.M., Maure, N., Cullis, P.R. Gene Ther. 8 (2001) 1188–1196.
17. Zuhorn, I.S., Bakowsky, U., Polushkin, E., Visser, W.H., Stuart, M.C.A., Engberts, J.B.F.N.,
Hoekstra, D. Mol. Ther. 11 (2005) 801–810.

Page 73
Chapter 3
- 63 -
18. Heyes, J. A., Duvaz, D. N., Cooper, R. G., Springer, C. J. J. Med. Chem. 45 (2002) 99–114.
19. Rädler, J. O., Koltover, I., Salditt, T., Safinya, C. R. Science 275 (1997) 810–814.
20. Zantl, R., Baicu, L., Artzner, F., Sprenger, I., Rapp, G., Rädler, J. O. J. Phys. Chem. B 103
(1999) 10300–10310.
21. Felgner, P. L., Gadek, T. R., Holm, M. Roman., R. Chan, H. W., Wenz, M., Northrop, J. P.,
Ringold, G. M., Danielsen, M. Proc. Natl. Acad. Sci. U.S.A. 84 (1987) 7413–7417.
22. Floch, V., Audrezet, M. P., Guillaume, C., Gobin, E., Bolch, G. L., Clement, J. C., Yaouanc,
J. J., Abbayes, H. D., Mercier, B., Leroy, J. P., Abgrall, J. F., Ferec, C. Biochim. Biophys. Acta.
1371 (1998) 53–70.
23. Floch, V., Loisel, S., Guenin, E., Herve, A. C., Clement, J. C., Yaouanc, J. J., Abbayes, H. D.,
Ferec, C. J. Med. Chem. 43 (2000) 4617–4628.
24. Nakanishi, M., Noguchi, A. Adv. Drug Deliv. Reviews 52 (2001)197–207.
25. Bailey, A. L., Cullis, P. R. Biochemistry 36 (1997)162–164.
26. Allen, T. M., Cleland, L. G. Biochim. Biophys. Acta. 597 (1980) 418–426.
27. Leventis, R., Silvirus, J. R. Biochim. Biophys. Acta 1023 (1990) 124132.
28. Wangerek LA, Dahl HM, Senden TJ, Carlin JB, Jans DA, Dunstan DE, J. Gene. Med. 3
(2001) 72–81.
29. da Cruz M.T., Simões, S., de Lima M.C.P. Exp. Neurology 187 (2004) 65–75.
30. Giraõ da Cruz, M.T., Cardoso, A.C.L, de Almeida, L.P., Simões, S., de Lima, M.C.P. Gene
Ther. 12 (2005) 1242–1252.
31. Ajmani, P.S., Hughes, J.A., Neurochim. Res. 24 (1999) 699–703.
32. Behr, J.P., Demeneix, B., Leffler, J.P., Perez-Mutul, J. Proc. Natl. Acad. Sci. U.S.A. 86
(1989) 6982–6986.
33. Behr, J.P. Bioconjugate Chem. 5 (1994) 382–389.
34. Obata, Y., Suzuki, D., Takeoka, S. Bioconjugate Chem. 19 (2008) 1055–1063.
35. Obata Y, Ikegaya N, Sakane I, Kurumizaka H, Takeda N, Takeoka S. Biochim. Biophys. Acta.

Page 74
Chapter 3
- 64 -
under revision (2008).
36. Lee, C.H., Ni, Y.H., Chen, C.C., Chou, C., Chang, F.H. Biochim. Biophys. Acta 1611 (2003)
55–62.
37. Williams, T.P., McGee, J.R. J. Biol. Chem. 257 (1982) 3491–3500.
38. Chakravarthy, B.R., Spence, M.W., Clarke, J.T.R., Cook, H.W. Biochim. Biophys. Acta 812
(1985) 223–233.
39. Zuhorn, I.S., Hoekstra, D. J. Mem. Biol. 189 (2002) 167–179.
40. Brunner, S., Sauer, T., Carotta, S., Cotton, M., Saltik, M., Wagner, E. Gene Ther. 7 (2000)
401–407.
41. Cheng, P.W. Human Gene Ther. 7 (1996) 275–282.
42. Simões, S., Slepushkin, V., Gaspar, R., de Lima, M.C.P, Duzgunes, N. Gene Ther. 5 (1998)
955–64.

Page 75
Chapter 4
- 65 -
Chapter 4
Construction of Carriers for Introducing Biomacromolecules into
Living Cells by Cationic Amino-acid Based Lipids
1. Introduction
The technologies for efficiently introducing functional biomacromolecules (e.g., genes,
proteins or their complexes) into living cells have attracted a great deal of interest. The
importance of these delivery systems has increased in terms of the versatile number of
applications, including gene therapy and protein supplemental therapy.
The aim of the present study is the preparation of DMC1-encapsulating cationic
liposomes containing novel amino acid-based lipids. The physicochemical properties, the
transfection ability of DMC1-encapsulating liposomes, and intracellular trafficking of the
introduced DMC1 are examined for effective protein delivery. This study is the first step toward
the development of a novel liposome-based DMC1 delivery system for genetically repairing
living cells.

Page 76
Chapter 4
- 66 -
2. Introdiction of Functional Biomacromolecules into Living Cells
2-1. Signification for Introduction of Biomacromolecules into Cells
Considerable progress has been made toward the development of effective transfection
reagents for the delivery of transcriptionally active DNA into cultured cells 1–4 and today,
plasmid transfection is a routine laboratory procedure used in most modern biomedical
laboratories. Often, the primary reason for performing DNA transfection is to express a desired
protein in the transfected cell to investigate its function. In this respect DNA transfection
technology can be considered an indirect protein delivery system. New methodologies to deliver
functional proteins into cells are presently being evaluated, but are still lacking in convenience
and effectiveness. The most actively studied approach uses a class of peptides that are 10–35
amino acids long and called “protein transduction domains” (PTD)1 5 or “membrane transport
signals”. The PTD derived from HIV-TAT 5,7,8, HSV-VP22 9, and antennapedia 10,11, or synthetic
PTD isolated from phage display libraries 12 are characterized by a high content of positively
charged arginine and lysine residues, which are potentially important for contact with the cell
membrane. The mechanism of action of PTD and membrane transport signals is not well
understood, however, their protein delivery efficiency varies depending on the protein delivered
12,13.
2-2. A Human Homologous Recombination Protein; DMC1
A human DMC1 protein (DMC1) is specifically expressed in meiosis and is known to be
a key component of the homologous recombination system 14–18. DMC1 was first discovered in
yeast 18 and subsequently found in mammals 19. Previous studies indicated that DMC1 knockout
mice displayed asynapsis and sterility due to defective meiotic recombination 20,21. Defects in the
homologous recombination cause accumulation of DNA double strand breaks and cell death
during mitotic cell division 22,23, indicating that homologous recombination also functions in the
mitotic double strand break repair.

Page 77
Chapter 4
- 67 -
2-3. Dmc1 introduction into living cells
The therapeutic concept is that of ‘gene repair’, which is an approach to ‘treat a diseased
gene itself’ at the molecular level. A higher organism is inherently provided with a number
systems to remedy a gene when it is miscopied in the replication process or damaged by some
form of stress. Making effective use of these systems should be a promising approach for
creating a therapeutic technology of gene repair. Therefore, DMC1-mediated homologous
recombination might be useful for treating damaged or improperly repaired DNA in living cells
providing DMC1 protein can be introduced into the mitotic cells without causing cytotoxicity.
Introducing template DNA for homologous pairing alone and its effect on gene repair have been
previously studied 24,25. The ultimate goal of our project is to deliver DMC1 together with an
oligonucleotide as template DNA. These homologous recombination systems are required for the
gene repair process. However, it would be undesirable for the introduced repair system to
continue operating once the damaged gene had been repaired. Direct insertion of DMC1 protein
is more appropriate than genetically mediating translation via an introduced gene because in the
latter method expression might persist for an extended period of time. Furthermore, the author
would anticipate the effect of the introduced protein to be immediately apparent.
Because genomic DNA is localized in the nucleus, DMC1 protein must be delivered to
this cellular compartment in order for it to function. Unlike gene delivery, viral vectors are not
appropriate for delivering proteins because their capsid shells are incapable of enveloping the
target protein. The introduction of functional proteins into living cells is currently performed by
microinjection and electroporation 26–28. However, these techniques have significant drawbacks
e.g., cytotoxicity, complex procedures, requirement for expensive equipment. Alternative
carriers have also been developed such as liposomes 29,30, polymer-based particles 31 and gels 32.
In particular, cationic liposomes can be used to form complexes between positively charged
liposomes and anionic proteins 29,33. It has also been suggested that cationic liposomes would
enhance cellular uptake by electrostatic interaction with the negatively charged cellular surface

Page 78
Chapter 4
- 68 -
34–36. Some of these liposome carriers are commercially available. However, most of these
preparations simply involve mixing with protein, which interact at the surface at variable ratios.
These intricate systems are often limited to in vivo experiments, because the cellular uptake of
the proteins is sometimes poor as a result of aggregation within the complex. Moreover, proteins
located on the surface of cationic liposomes are subject to degradation by proteases and are
thereby readily cleared by the biological defense systems. To protect functional proteins from
protease digestion, encapsulation into liposomes to generate an efficient protein delivery system
is examined. Furthermore, when cationic liposomes are introduced into living cells, membrane
fusion between the cationic liposomal components and the endosomal membrane of early
endosomes would be expected to occur, leading to the release of encapsulated proteins into the
cytoplasm. In the case of DMC1, the released protein would be spontaneously transferred to the
nucleus. The delivery of the functional DMC1 to nucleus is anticipated to provide an active
repair system.
3. Construction of DMC1-Encapsulating Cationic Liposomes
3-1. Preparation of DMC1-Encapsulating Cationic Liposomes
The author prepared cationic liposomes having an
average diameter of 244 + 95 nm for DMC1 introduction into
living cells. The encapsulation efficiency of DMC1 was
estimated to be ~12%. An efficiency of this magnitude might
be expected for liposomes with ca. 250 nm diameter, +29 mV
zeta potential. The diameter of the DMC1-encapsulating
liposomes was monitored to establish the dispersion stability
of the cationic liposomes by DLS measurements. Using this
data, the author clarified that no aggregation of liposomes
occurred after one month storage at 4oC. TEM observations
Fig. 4-1.
Transmission
electron microscopic images
of the DMC1-encapsulating
cationic liposomes stained
with
1%
sodium
phosphotungstic acid.

Page 79
Chapter 4
- 69 -
of the cationic liposomes verified they had a size range of 100-200 nm (Fig. 4-1). The zeta
potential of +24 mV indicates that the cationic moiety of Lys-Glu2C16 is located on the surface
of the liposome.
3-2. Protease Digestion Assay
The author confirmed DMC1 encapsulation into the cationic liposomes using proteinase
K. In Fig. 4-2, naked DMC1, DMC1-encapsulated cationic liposomes, and a mixture of naked
DMC1 and empty cationic liposomes are represented as I, II and III, respectively. Samples in the
right column of Fig. 4-2 were treated with proteinase K, whereas those in the left column were
untreated. In the absence of proteinase K two bands running with an apparent molecular mass of
31 kDa and 40 kDa were visible. The upper band corresponds to native DMC1 (theoretical mass
of 37 kDa), while the lower band is a degradation product generated in the purification process.
When the protease was added to DMC1, the naked DMC1 was completely digested (Fig. 4-2
right column, lane I). However, DMC1-encapsulated in liposomes retained the banding pattern
of the untreated samples except for an additional band corresponding to the protease (Fig. 4-2
right column, lane II, 30 kDa). These results clearly indicate that DMC1 encapsulated in the
liposomes was protected from proteolysis by the membrane bilayer. This conclusion is also
supported by the result of the cationic liposome/DMC1 complex, where the DMC1 adsorbed on
the outer surface of the liposome was subject to complete digestion in the presence of proteinase
K (Fig. 4-2 right column, lane III).
DMC1 encapsulation was evaluated using a protease digestion assay, in which
non-encapsulated DMC1 was degraded by the addition of exogenous proteinase K. It was
confirmed that DMC1 was effectively encapsulated into the cationic liposomes composed of
DOPC/cholesterol/Lys-Glu2C16/PEG-Glu2C18 and little DMC1 was electrostatically adhered to
the outer leaflet of the liposomes. The results suggested that protease couldn’t permeate through
the liposomal membrane, resulting in encapsulated DMC1 was effectively protected from

Page 80
Chapter 4
- 70 -
protease. By contrast, as indicated BioPORTER system, DMC1 on the surface of the liposomes
easily degradated from protease, implying the BioPORTER system was limited in vitro use.
3-3. Intracellular DMC1 Transition by Cationic Liposomes in COS-1 Cells
Localization of Cy3-DMC1 upon introduction into COS-1 cells via endocytosis was
investigated using confocal laser scanning microscopy (Fig. 4-3). After introduction of DMC1
into COS-1 cells, the endosomes or lysosomes were stained with LysoTracker®. The author
investigated the transfer of DCM1 into the cells by three different methodologies involving the
addition of (i) DMC1 alone (ii) DMC1-encapsulated in liposomes (iii) BioPORTER/DMC1
complex (only mix the empty cationic liposomes with DMC1). The cellular introduction of
DMC1 was undetectable after addition of DMC1 alone, indicating the protein could not undergo
endocytosis or permeate through the membrane. By contrast, DMC1-encapsulated liposomes
facilitated the efficient introduction of DMC1 into the cells. Furthermore, localization of DMC1
kDa
200
116
66
42
30
17
I
II
III
I
II
III
Protease (–)
Protease (+)
DMC1
DMC1
Protease K
Fig. 4-2. Resistance of DCM1 to the addition of exogenous proteinase K of (I) DMC1 alone, (II)
DMC1-encapsulating liposomes, and (III) complex between vesicles and DMC1. All samples
contained 1 μg DMC1 and were incubated with or without protease at 37oC for 10 min.

Page 81
Chapter 4
- 71 -
was different from that of the endosomes, suggesting that DMC1 is able to efficiently transfer
from the endosome into the cytoplasm. By contrast, although addition of the
BioPORTER/DMC1 complex assisted the transfer of DMC1 into the cell, almost all of the
internalized protein was located in the endosome (yellow dots in Fig. 4).
DMC1 alone could not permeate through the cellular membrane of COS-1 cells as shown
by our microscopic observations (Fig. 4-3). The author compared the utility of cationic
liposomes with a BioPORTER reagent, which is commercially available and has been
extensively analyzed 26,33,37. Comparison of the cellular uptake of DMC1 mediated either by our
cationic liposomes or by the BioPORTER system showed that the former method achieved the
highest level of DMC1 introduction into COS-1 cells. The effectiveness of the cationic
liposomes in delivering DMC1 is largely due to the narrow size distribution of the
DMC1-encapsulating liposomes. The relatively poor introduction of DMC1 into the target cells
a
b
c
a’
b’
c’
Phase contrast
Fluorescent image
Fig. 4-3. Confocal microscopic images of COS-1 cells after the addition of (a, a’) DMC1, (b, b’)
DMC1-encapsulating liposomes, or (c, c’) BioPORTER/DMC1 complexes, containing 500 ng Cy3-
DMC1. Endosomes in the cells were stained with LysoTracker® Green. The left fluorescence images
(a, b, c) show localization of the Cy3- DMC1 (Red) or endosomes (Green) and the right images (a’,
b’, c’) were incorporated by phase contrast.

Page 82
Chapter 4
- 72 -
using a complex of BioPORTER/DMC1 is probably due to aggregation within the preparation.
Indeed, the cellular uptake of BioPORTER/DMC1 complexes might be reduced as a result of
low cellular phagocytic capacity of the aggregated complexes as indicated by Lee et al.
31.
Therefore, the author compared the uptake efficiency of the DMC1-encapsulating liposomes
with that of the BioPORTER/DMC1 complex. In addition to introducing DCM1 into COS-1
cells, the DMC1-encapsulating cationic liposomes achieved effective DMC1 introduction into
the NIH3T3 cell line (data not shown). Therefore, delivery of DMC1 by our cationic liposomes
might be independent of the cell line. Importantly, upon addition of the DMC1-encapsulating
liposomes to COS-1 cells, a large proportion of the intracellular DMC1 was released from the
endosomes into the cytoplasm or nucleus. Thus, the author conclude cationic liposomes facilitate
DMC1 release from the endosomes by membrane fusion. By contrast, large amounts of the
DMC1 introduced by BioPORTER remained in the endosomes. The trigger for the endosomal
release of DMC1 is an electrostatic fusion process driven by the neutralization of the positively
charged liposomal membrane and negatively charged endosomal membrane. Therefore, the low
endosomal release of DMC1 introduced by BioPORTER might be explained from the
electrostatic repulsion of the anionic DMC1 of the complexes and the endosomal membrane. In
addition, the author also attempted the introduction of DMC1 using cationic lipid-based protein
carriers, Profect-P1 (Nacalai Tesque, Inc.; Kyoto, Japan) and SAINT-MIX™ (COSMO BIO
CO. Ltd.; Tokyo, Japan). However, in both cases the resultant level of intracellular DMC1 was
quite low (data not shown). Indeed, of all the experiments performed using lipid-based protein
carriers, our cationic liposomes showed the highest capacity for DMC1 introduction into living
cells.
3-4. Western Blotting Analysis
To clarify the localization of DMC1 in the cells after addition of the
DMC1-encapsulating liposomes and the BioPORTER/DMC1 complexes, further investigation

Page 83
Chapter 4
- 73 -
was performed by Western blotting analysis. Both cytoplasmic and nuclear fraction was collected
after the introduction of DMC1 into the COS-1 cells, confirmed by the presence of tubulin and
histone, respectively. Our results show that DMC1 is localized in both the cytoplasm and nucleus
after the addition of the DMC1-encapsulating liposomes in the absence of FBS (Fig. 4-4). By
contrast, DMC1 introduced using BioPORTER/DMC1 complex in the absence of FBS was
exclusively localized in the cytoplasm. Furthermore, the amount of intracellular DCM1 was
considerably lower than cells treated with DMC1-encapsulating liposomes. The author also
studied the effect of FBS on the introduction of DMC1 into COS-1 cells. DMC1 was also
localized in both the cytoplasm and nucleus after the addition of DMC1-encapsulating liposomes
in the presence of FBS. Furthermore, the amount of DMC1 in the cytoplasm and nucleus were
increased by comparison with the same experiment conducted in the absence of FBS.
Intriguingly, analysis of COS-1 cells after treatment with BioPORTER/DMC1 complex in the
presence of FBS failed to detect any internalized DMC1. Therefore, the dynamics of DMC1
introduction by the DMC1-encapsulating liposomes into COS-1 cells was difference from that of
BioPOTER/DMC1 complex. Western blot analysis showed that the band derived from tubulin as
a marker for the cytoplasmic fraction was also found in the nuclear fraction after treatment with
DMC1-encapsulating liposomes in the presence of FBS. Presumably this observation was due to
the adsorption of the empty cationic liposomes to the nuclear envelope with tubulin (Fig. 4-4.
See experiment involving the addition of the empty cationic liposomes). The serum in the
medium may facilitate aggregation with cellular proteins.

Page 84
Chapter 4
- 74 -
Nonetheless, nuclear localization of DMC1 could not be detected by microscopic
observations, further analysis in terms of DMC1 localization in COS-1 cells was carefully
carried out by Western blotting. Exogenous DMC1 introduced by cationic liposomes into the
cells was localized within the nucleus. The author compared the amount of DMC1 delivered by
DMC1-encapsulating liposomes with that by BioPORTER/DMC1 complexes. Estimations based
on the DMC1 concentration by Western blotting using densitometry measurements showed the
DMC1-encapsulating liposomes to be ~10-fold more efficient than the BioPORTER/DMC1
complex. Furthermore, unlike other cationic liposomes, the author were able to confirm our
cationic liposome preparation promoted transfer of DMC1 into the cytoplasm and nucleus. By
contrast, the BioPORTER system was unable to deliver DMC1 into the nucleus.
Histone
DMC1
Tubulin
DMC1
only
Vesicle
only
Vesicles
DMC1-encapsulating
liposomes
BioPORTER/DMC1
FBS
+
+
+
+
+
Cyt Nuc
Cyt Nuc
Cyt Nuc
Cyt Nuc
Cyt Nuc
Fig. 4-4. Western blotting analysis for DMC1 localization in COS-1 cells after the addition of DMC1,
DMC1-encapsulating liposomes, or the BioPORTER/DMC1 complex in the absence or presence of
FBS. Equal amounts of protein from the cell lysates were subjected to Tris/Tricine gel
electrophoresis. Both cytoplasmic and nuclear fraction was confirmed by the presence of tubulin and
histone, respectively.

Page 85
Chapter 4
- 75 -
3-5. Conclusion of DMC1-Encapsulating liposomes
In conclusion, the author have prepared stable DMC1-encapsualting liposomes by an
extrusion method for construction of a gene therapy system to promote genetic recombination.
The encapsulation of DMC1 into the cationic liposomes was confirmed by means of a protease
digestion assay. Moreover, the cationic liposomes mediated endosomal release of DMC1 as
demonstrated by the use of biomembrane mimicking anionic liposomes. As a result of the
facilitated fusogenic potential to the endosomal membrane, our cationic liposome displayed
superior intracellular DMC1 delivery by comparison with the established BioPORTER system.
By comparison with conventional protein delivery reagents, our cationic liposomes would offer
efficient DMC1 delivery to both the cytoplasm and nucleus of COS-1 cells with relatively low
levels of toxicity, even in presence of FBS. The author is currently investigating the
recombination ability of cells after introduction of DMC1 for genetic repair mediated gene
therapy.
4. New Methodology of pDNA Introduction into Cells Using Carbon Nanotubes
4-1. Biomedical Application of Carbon nanotubes with Cationic Amino-acid Based Lipids
The use of carbon-based nanostructures, such as carbon nanotubes, in biomedicine is
increasingly attracting attention 38. One key advantage of carbon nanotubes is their ability to
translocate through plasma membranes, allowing their use for the delivery of therapeutically
active molecules in a manner that resembles cell-penetrating peptides. Moreover, exploitation of
their unique electrical, optical, thermal, and spectroscopic properties in a biological context is
hoped to yield great advances in the detection, monitoring, and therapy of disease. Here the
author offers a preparation of lipid wrapped multi wall carbon nanotube for plasmid DNA
delivery.
Recently, solubilized CNTs were investigated as drug, protein, and gene carriers, with
their efficiencies comparable to liposomes, micells, polymer-based particles. In particular,

Page 86
Chapter 4
- 76 -
positive charged functional CNTs (f-CNTs) are investigated as gene carriers because their ability
to bind DNA 38,40,41
, as similar way to lipoplexes between cationic liposomes and DNA. Some
researchers have already applied soluble f-CNTs in pre clinical studies, and demonstrated the
efficiency of CNTs in cancer therapy 42.
4-2. Lipid Wrapped MWNT
The author prepared L-MWNT having an average diameter of 164 + 70 nm and a zeta
potential of about +42 mV, indicating their positive charge. After preparation of L-MWNT, their
diameter was monitored by a DLS measurement in order to know the dispersion stability. No
aggregation was found for more than three month at 4oC (Fig. 4-6). Transmission electron
microscopy (TEM) of L-MWNTs indicated a length of about 200–400 nm and a radius of about
5–10 nm (Fig. 4-6): the average length of major and minor axis of the L-MWNT from DLS were
Fig. 4-5. Several types of functionalized CNTs that can be adequately and individually
dispersed in biological environments. Background image: differential interference
contrast image of epithelial lung carcinoma (A549) cell culture.

Page 87
Chapter 4
- 77 -
clarified to be reasonable. The L-MWNT was prepared using Arg-Glu2C16 and PEG5000-Glu2C18
with a simple procedure including a sonication step followed by centrifugation for purification.
In order to attain soluble CNTs, surfactants such as Triton X-100 and sodium
dodecylsulfate (SDS) are generally used. However, resulting soluble CNTs with the surfactants
lack stability and Cytocompatibility 48. The author used two types of lipids, Arg-Glu2C16 and
PEG-Glu2C18, to solubilize CNTs in water for safer amphiphiles 49-52. Arg-Glu2C16 provided the
positive charge to 0.3 % PEG-lipid included membrane. PEG-lipid was necessary to obtain
stable CNTs in water, because significant aggregation occurred within 1 week for the dispersion
of only Arg-Glu2C16 wrapped MWNTs. Therefore, the long stability of L-MWNTs was due to
the electrostatic repulsion from Arg-Glu2C16 and the steric hindrance of the PEG chains on the
surface of the L-MWNT. This phenomenon supports that the lipids are uniformly oriented onto
the surface of MWNTs. Regarding to the orientation of the mixed lipids onto CNTs, two
mechanisms have previously supposed concerning SWNTs using surfactants, but not lipids. The
TEM
(a)
(b)
(c)
+
+
+
+
+
+
+
+
+
+
= Arg-Glu2C16
= PEG-Glu2C18
Fig. 4-6. (a) Schema of the molecular assembly of lipidic wrapped multi-walled carbon nanotube
(L-MWNTs). (b) Dispersion of L-MWNTs in milliQ water for 3 month at room temperature. (c)
Transmission microscopic image of L-MWNTs.

Page 88
Chapter 4
- 78 -
first possibility sustains that SWNTs are encapsulated in a cylindrical surfactant micelle 53,54, the
second one provides a hemimicellar adsorption of surfactant molecules onto a SWNTs 43. Since
the lipids structure would display cyclindrical structure, the first orientation should be suggested.
4-3. Complexation of L-MWNT with DNA
The complexes between the positively charged L-MWNTs and pDNA, at different
L-MWNT/pDNA mixing ratios, were prepared and analyzed using an agarose gel retardation
assay (Fig. 4-7 (a)). The complexes of L-MWNT with pDNA were confirmed and excess pDNA
was barely detectable at a ratio of L-MWNT/pDNA of 50 and above. Moreover, spontaneous
pDNA release from the complexes into solution did not occur even after several hours.The
assembling states of the complexes at various L-MWNT/pDNA ratios were visualized by
transmission electron microscopy (Fig. 4-7. (b)). At L-MWNT/pDNA ratios of 5 and 50, some
complexes tended to aggregate. On the contrary, few aggregation of L-MWNT/pDNA complexes
was detected at a ratio of 25.
Complexes between positively charged L-MWNT and pDNA were prepared successfully
by electrostatic interaction. Concerning the condensation of pDNA with using lipid wrapped
CNTs by Shi et al., efficient intracellular siRNA delivery of phospholipid-coated SWNTs. In
this case, siRNA was covalently linked to PEG-lipids via a cleavable disulfide bonding 55. By
contrast, we adopted the simpler a non-covalent approach for condensation of pDNA to CNTs by
electrostatic interaction. By the electrophoresis assay of the complexes with different
L-MWNT/pDNA ratios, the author investigated the ability of the L-MWNTs to make complexe
with pDNA. Generally, the ratio of the nitrogen group of the cationic carriers to the phosphate
groups of pDNA (N/P ratio) influences the size, zeta potential, and structure of the complexes,
resulting in the N/P ratio-dependent gene expression as usually shown in the cationic liposomes
or polymer-based particles for gene delivery 56, 57. The mixing ratio of f-CNTs with pDNA also
influenced the gene expression efficiency in human lung carcinoma cells 39. The Arg-Glu2C16

Page 89
Chapter 4
- 79 -
concentration was separately calculated with a high performance liquid chromatography to
estimate the effective N/P ratio. L-MWNT/pDNA complexes with a mixing ratio of 25 (w/w)
resulted to have a N/P ratio of 16, at which the complexes showed the maximum gene expression
efficiency. Complexes between L-MWNT and pDNA prepared at a mixing ratio of 50 showed
visible aggregation (data not shown), suggesting that the ratio of the cationic charge from the
Arg-Glu2C16 and the anionic charge from pDNA are approximately 1 at a mixing ratio of about
50. On the contrary, complexes with a mixing ratio of 25 were stable in water, indicating that the
cationic moieties which didn’t contribute to the binding with DNA would contribute to
electrostatic repulsion that prevents the aggregation of the complexes.
pDNA
only
5
10
25
50
75
100
125
well
s.c.
o.c.
L-MWNT / pDNA (w/w)
5
25
50
L-MWNT / pDNA (w/w)
100 nm
200 nm
200 nm
(a)
(b)
Fig. 4-7. (a) Gel retardation assay of the L-MWNT/pDNA complexes at different mixing ratio (w/w).
(b) Transmission microscopic image of the L-MWNT/pDNA complexes at different mixing ratios.

Page 90
Chapter 4
- 80 -
4-4. Gene Expression Efficiency of L-MWNT/pDNA Complexes
The gene expression efficiency of the L-MWNT/pDNA complexes at different
L-MWNT/pDNA ratios on neuronal SH-SY5Y cell line was investigated. After the addition of
necked pDNA to the cells, no gene expression occurred. In the case of incubation of cells with
L-MWNT/pDNA complexes, GFP positive cells were detectable and the highest gene expression
was obtained at a L-MWNT/DNA ratio of 25 (being tested from 1 to 50; Fig. 4-8). The GFP
expression efficiency was quantitatively estimated by the ratio of GFP positive cells to the total
cell number. 6% cells expressed GFP after the addition of L-MWNT/pDNA complexes at the the
mixing ratio of 25. A comparable commercially available transgenic reagent,
LipofectamineTM2000, produced 4% of GFP positive cells, so demonstrating a comparable gene
expression efficiency: L-MWNTs, prepared with a simple non-covalent approach, were found to
be an efficient gene carrier n neuroblastoma cells.
0
Ra
tio
o
f G
F
P
pos
itiv
e c
e
ll (%
)
pDNA
only
5
10
25
50
1
L-MWNT / pDNA ratio (w/w)
LA2000
4
2
8
6
Fig. 4-8. Transfection efficiency of the L-MWNT/pDNA complexes with varying the mixing ratio
between L-MWNT and pDNA in SH-SY5Y cells.

Page 91
Chapter 4
- 81 -
The author achieved gene expression using L-MWNT in neuronal SH-SY5Y cells for
providing obtaining a maximum value of 5% GFP positive cells at a mixing ratio of 25. The
complexes with the mixing ratios lower than 10 reduced phagocytic ability because of the
aggregation of complexes themselves and higher than 50. Moreover, complexes with a mixing
ratio of 50 tended to provide toxicity (thus reducing gene expression efficiency). Comparing the
N/P ratio with other gene carriers such as poly-L-lysine and polyethyleneimine, the N/P ratio of
L-MWNT showing efficient gene expression is higher than that of the other polymer-based
transfection agents. In the case of cationic polymers, the polyplexes formed small structure 58,
whereas, the complexes of L-MWNT with pDNA slightly form a more condensed structure
because of the physical strength of the CNTs. The author demonstrated for the first time effective
gene expression on neuronal cell lines using L-MWNT with a non-covalent approach.
4-5. Gene Expression of the L-MWNT/pDNA Complexes under Magnetic Fields
The gene expression efficiency of the L-MWNT/pDNA complexes under a magnetic
field was evaluated. In this study the magnetic properties of L-MWNTs were focused to enhance
the gene expression efficiency. The magnetic field was applied under the SH-SY5Y cells
incubated with the L-MWNT/pDNA complexes with a mixing ratio of 25 for 4 hr. A dramatic
enhancement of gene expression under a magnetic filed was served (Fig. 4-9 (a)). The GFP
positive cells with a magnetic field were increased (13%) in comparison to the study without
application of the magnetic filed (6 %, Fig. 4-9 (b)). The author successfully obtained an
enhancement of gene expression of the magnetically L-MWNT/pDNA complex by a magnetic
field.

Page 92
Chapter 4
- 82 -
The enhancement of gene expression (twice respect to the control) of the L-MWNT/DNA
under a static magnetic field was successfully obtained. The enhancement of gene expression
under the magnetic field could be due to the increment of cellular uptake of the L-MWNT/pDNA
complexes because of a pulling effect by the permanent magnet. The useful magnetic properties
of CNTs can be exploited to the targeted drug therapy, as already described for other magnetic
particles 59. Cai. et al. for the first time reported magnetic-driven enhancement of gene
expression by using Ni doped carbon nanotubes 60. The magnetic spearing into cellular
membrane was effective to enhance gene expression. Since the spearing into cellular membrane
was due to the hydophobicity of the Ni-CNTs, the toxicity of these carriers could be relatively
high. The author demonstrated enhancement of the gene expression, driven by a permanent
magnetet, with L-MWNTs, that showed no toxicity even at high concentration.
Magnet (–)
Magnet (+)
6
3
0
15
12
9
G
F
P ex
pres
s
e
d c
e
ll (%
)
pDNA only
18
L-MWNT (25)
Magnetic field (–)
Magnetic field (+)
Fig. 4-9. (a) Microscopic images of neuronal SH-SY5Y cells at 48 hr after the addition of the
L-MWNT/pDNA complexes at a L-MWNT/pDNA ratio of 25. When the complexes were added to the
cells, a magnetic field was applied under the dishes for 4 hr for the investigation of the influence of the
L-MWNT magnetic properties on gene transfection efficiency. (b) Transfection efficiency of the
L-MWNT/pDNA ratio on SH-SY5Y cells in the presence of a static magnetic filed.

Page 93
Chapter 4
- 83 -
4-6. Cytotoxicity of L-MWNT
The cytotoxicity of the L-MWNT/pDNA complexes at a mixing ratio of 25 was
investigated in SH-SY5Y cells at a concentration between 0.1–5000 μg/mL (Fig. 4-10). At a
concentration of 5000 μg/mL, approximately 50% of the cells were survived, denoting a little
cytotoxicity of the complexes. However, LA2000 showed 40% of cell death even at a
concentration of 2 μg/mL. Therefore, the considerable lower cytotoxicity of the L-MWNT/DNA
complexes was confirmed.
The low cytotoxicity of the L-MWNT/pDNA complexes on SH-SY5Y cells was
demonstrated. So far, the toxicity and immunoresponse of CNTs have been a major issue to
construct CNT mediated drug delivery. Some reports indicate that CNTs induced dermatitis and
keratosis; furthermore, the SWNT exposure produced 40% cellular viability in the SWNT
concentration of 240 μg/mL on HaCaT cells, which was a quite low concentration in comparison
with those ones tested in the case of L-MWNT/DNA complexes (2 μg/mL). The cytotoxicity of
20
40
60
80
100
Ce
ll v
ia
b
ility
(%
)
0
120
1
0.1
100
10
1000
L-MWNT [μg/mL]
10000
Fig. 4-10. Cytotoxicity of the L-MWNT/pDNA complexes with a mixing ratio of 25 at different
concentrations.

Page 94
Chapter 4
- 84 -
MWNT could be due to the direct interaction with the cellular membrane due to the strong
hydrophobic properties of MWNT, resulting in degradation of cells. The wrapping lipids on the
MWNT dramatically reduced the cytotoxicity of MWNT by masking the hydrophobicity of
CNTs. Owing to low cytotoxicity of L-MWNT, the author found the high transfection efficiency
by varing the amount of the L-MWNT/pDNA complex (mixing ratio: 25) to neuronal cells. An
increment of the gene expression was found increasing the pDNA concentration in the medium
from 2 to 20 μg/mL. At 20 μg/mL of pDNA concentration in medium 12% of GFP positive cells
were obtained. Due to the low cytotoxiciy of the L-MWNT/pDNA complex, enhancement of the
gene expression increasing the L-MWNT/pDNA application to the cells was achieved.
4-7. Conclusion of Water Soluble L-MWNTs for Plasmid DNA Delivery
The ability of lipid wrapped MWNTs (obtained with the amino-acid based lipid Arg-Glu2C16) were
investigated for the first time in neuronal transfection. The lipid wrapping was effectively performed with the
cationic lipid Arg-Glu2C16 and PEG-lipid PEG-Glu2C18 to obtain stable and positively charged L-MWNTs.
L-MWNTs were binded to pDNA enconding GFP. When the complexes L-MWNTs/pDNA were applied to
SH-SY5Y cells, a maximum gene expression (6%) was revealed at a L-MWNT/DNA mixing ratio of 25. In
addition, the gene expression was further increased thanks to the application of a static magnetic field, thus
obtaining 12% GFP positive cells. The transfection efficiency was significanlty higher than that of
LipofectaimneTM2000 (a commercial available transgenic reagent). Intracellular DNA delivery was
investigated using chloroquine as endosomolytic reagent and via fluorescent microscopy using labeled pDNA,
demonstrating that L-MWNTs can effectively deliver DNA into nucleus. The cytotoxicity, investigated with
MTT assay, was found to be considerable lower in comparison with naked MWNTs or LipofectaineTM2000
reagent. Concluding the exploitation of L-MWNTs for the construction of CNT-based gene carriers was
demonstrated.

Page 95
Chapter 4
- 85 -
5. Experimental Section
5-1. General Methods on DMC1-encapsulating Liposomes
Preparation of cationic liposomes
DMC1-encapsulating cationic liposomes were prepared as follows; DOPC (58 mg),
cholesterol (29 mg), 1,5-dihexadecyl N-lysyl-L-glutamate (107 mg) and PEG5000-Glu2C18 (5.2
mg) were dissolved in t-butylalcohol and the resulting solution was freeze-dried. The mixed lipid
powder (75 mg) was hydrated in 20 mM HEPES buffer (pH 7.5, 1.5 mL) containing DMC1 or
Cy3-DMC1 (f.c. 18 μM) for 4 hr. The resulting solution was extruded with a LIPEX™
EXTRUDER (Northern Lipids Inc., Burnaby, BC) equipped with membrane filters (pore size 3.0,
0.80, 0.65, 0.45, 0.22 μm, cellulose acetate filters, Millipore Inc., Tokyo, Japan). The liposomes
were washed twice by ultracentrifugation (100,000 x g, 30 min). The DMC1 concentration
encapsulated in the liposomes was determined by quantification with a high-pressure liquid
chromatography system (Shimadzu Co., Kyoto, Japan) using a mobile phase of 36% acetonitrile
and 0.1% trifluoroacetic acid at 1 mL/min on a TSK-GEL G3000PWXL column (TOSOH, Tokyo,
Japan). The dispersion (20 μL) of the DMC1-encapsulating liposomes was solubilized with a 2%
Triton-X solution and then the resulting solutions were injected onto the HPLC system to
calculate the DMC1 concentration with a calibration curve at a concentration range from 50 to
500 μg/mL. Moreover, the author compared the DMC1-encapsulating liposomes with
BioPORTER as a well known reagent for protein introduction into cells, which was prepared
according to the manufacture’s instructions (Gene Therapy Systems, San Diego, CA). Briefly, a
BioPORTER dry film is resuspended with 250 μL of methanol and vortexed for 20 s. 10 μL of
the resulting solution was transferred into an Eppendorf tube and the solvent was evaporated at
room temperature for 2 hr. The lipid film was hydrated in 100 μL of HBS (10 mM HEPES, 150
mM NaCl, pH 7.0) containing the appropriate amounts of DMC1 or Cy3-DMC1. The solution
was pipetted up and down and incubated at room temperature for 5 min. Moreover, the solution
was vortexed for 5 s. Finally, 900 μL of the medium with or without 10% fetal bovine serum

Page 96
Chapter 4
- 86 -
(FBS) was added to the complexes for the following analyses.
DMC1 degradation against exogenous protease
The resistance of the encapsulated DMC1 against exogenous protease was assessed using
the following samples; naked DMC1 alone, DMC1-encapsulating liposomes, and the empty
cationic liposomes with DMC1. The samples (DMC1, 1.0 μg) were incubated with proteinase K
(66 μg/ml, 2.5 μL, Roche Diagnostics, Indianapolis, IN) in a total volume of 9.5 μL HEPES
buffer (pH 7.4) for 10 min at 37oC. Then, phenylmethanesulfonyl fluoride (PMSF, 100 mM, 5
μl) was added to the resulting solution to stop the reaction. In the case of the
DMC1-encapsulating liposomes, 2% sodium dodecyl sulfate (SDS) was added to the solution
and heated to 80oC for 10 min before analysis by SDS-PAGE. The samples were subjected to
electrophoresis on a 10% SDS-polyacrylamide gel (NuPAGE 10% Bis-Tris Gel; Invitrogen,
Carlsbad, CA). Bands on the gel were visualized by staining with Coomassie brilliant blue.
Intracellular DMC1 trafficking
COS-1 cells (1 x 105 cells) were seeded on 35-mm cell culture dishes and incubated in a
5% CO2 incubator for 24 hr at 37oC. The medium in the cell culture dish was exchanged with
fresh DMEM medium (1 mL) containing the Cy3-labeled DMC1 encapsulating liposomes
([DMC1] = 500 μg/mL as a final concentration) in the absence of FBS. After incubation for 4 hr,
the cells were washed twice with PBS, and then the medium was exchanged with 2 mL DMEM
medium containing Hoechst 33342 (f.c. 5 μg/ml) as a nucleus dying reagent. After incubation for
30 min, the cells were washed with PBS. Then, 2 mL of DMEM medium containing
LysoTracker® Green DND-26 (f.c. 150 nM; Invitrogen, Carlsbad, CA) as an endosome dying
reagent was added to the dishes. After incubation for 1 hr with the dye reagent, the cells were
washed twice with cold PBS and resuspended in 1 mL DMEM medium in the absence of FBS.
Then the cells were observed with a laser confocal microscope (DIGITAL ECLIPSE C1si-Ready,

Page 97
Chapter 4
- 87 -
Nikon Co., Ltd., Tokyo, Japan).
Western blotting analysis
For Western blot analysis, COS-1 cells (1 x 106 cells) were seeded on 100-mm cell
culture dishes and incubated in a 5% CO2 incubator for 24 hr at 37oC. The medium in the cell
culture dish was exchanged with fresh DMEM medium (1 mL) containing the
DMC1-encapsulating liposomes or BioPERTER/DMC1 complexes prepared according to the
attached instructions ([DMC1] = 500 μg/mL as a final concentration) in the absence or presence
of FBS. After incubation for 4 hr, the cells were washed twice with PBS and treated with a
trypsin-EDTA solution. The resulting solution was centrifuged (2000 x g, 5 min) to collect the
cells. The cells were then solubilized with a lysis buffer (15 mM Tris-HCl, pH 7.5, 60 mM KCl,
15 mM NaCl, 5 mM MgCl2, 1 mM CaCl2, 1 mM DTT, 250 mM sucrose, 0.6% NP40)
supplemented with a protease inhibitor (1 mM PMSF). After centrifugation (2000 x g, 5 min) of
the mixed solution the supernatant was collected as a cytoplasmic fraction. The precipitate was
treated with a lysis buffer (100 mM Tris-HCl, pH 6.8, 20% glycerol, 4% SDS) and heated for 10
min at 100oC to solubilize the nuclear envelope. The resulting solution was then subjected to
ultracentrifugation (25,000 x g, 30 min) to obtain the nuclear fraction. The protein concentration
of each fraction was determined using a protein assay kit (BioRad, Hercules, CA). Equal
quantities of the cellular proteins were mixed with an equal volume of sample buffer containing
SDS (100 mM Tris-HCl, pH 6.8, 0.2% bromophenol blue, 20% glycerol, 200 mM
mercaptoethanol, 4% SDS) and denatured by heat treatment at 100oC for 10 min. The denatured
proteins were separated by 10% SDS-PAGE and electroblotted onto a polyvinylidene difluoride
(PVDF) membrane (BioRad). The membrane was probed with monoclonal anti-tubulin beta
(mouse IgG isotype; Sigma-Aldrich, St Louis, MO), anti-acetyl histone H4 (Upstate Biotech, Inc,
Lake Placid, NY) and anti-DMC1 antibodies. Anti-DMC1 antibody was affinity purified from
anti-DMC1 polyclonal rabbit serum (Medical & Biological Laboratories, Nagoya, Japan). The

Page 98
Chapter 4
- 88 -
amount of DMC1 in the cytoplasmic and nuclear fractions could not be accurately determined by
comparing the intensity of the band on the gel. This was because we were unable to estimate the
amount of proteins present in both the cytoplasm and nucleus, despite applying equivalent
amounts of protein to each lane.
5-2. General Methods on Lipid Wrapped Carbon Nanotubes
Preparation of lipid wrapped MWNTs
The mixed lipids containing Arg-Glu2C16 and PEG-Glu2C18 were prepared after
freeze-drying from t-butyl alcohol, with a molar ratio of 10:0.03. Five mL of MilliQ water with 5
mg of CNTs and 20 mg of mixed lipids were sonicated for 30 min with a Branson Digital
Sonifier S-250 D by supplying a power of 100 W. After sonication, the resulting solutions were
centrifuged (1,000 g, 10 min, 4oC) twice to remove unsuspended MWNTs and impurities. The
resulting supernatant was ultracentrifuged (30,000 g, 30 min, 4oC) to separate lipidic wrapped
MWNTs (L-MWNTs) from unbounded lipids. At this condition, unbound lipids do not
precipitate and the resulting pellet was composed of L-MWNT complexes only. After
re-suspending the pellet in an equal volume of water, a dispersion of L-MWNTs was obtained.
The concentration of L-MWNTs was determined by estimating the L-MWNTs weigh of
L-MWNT after freeze-drying the dispersion.
Size distribution and zeta potential of the L-MWNTs
A dispersion of L-MWNTs (10 μL) containing 4 mg/mL of L-MWNTs was diluted with
2 mL of water, and the average particle diameter was measured with a dynamic light scattering
spectrophotometer (N4 PLUS Submicron Particle Size Analyzer, Beckman-Coulter, Fullerton,
FL). The zeta,potential of the L-MWNTs was measured with a Zetasizer (Zetasizer4, Malvern,
U.K.). The L-MWNTs suspension (0.1 mg/mL) in water weas loaded in a capillary cell mounted
on the apparatus and measured three times.

Page 99
Chapter 4
- 89 -
Observation of the L-MWNTs with a transmission electron microscopy
A drop of the sample dispersion was placed on a 100 mesh copper grid, and then the
excess dispersion was removed with a filter paper. The morphology of the L-MWNTs was
observed by TEM (JM-1011, JEOL). In the case of the L-MWNTs/pDNA complexes with
different mixing ratios, the complexes were placed on the copper grid, followed by the removal
of excess water, after incubation for 15 min to allow complex formation.
Preparation of L-MWNT/pDNA complexes
The L-MWNTs were mixed with pDNA that encoding GFP. The mixed solution was
gently agitated and incubated at room temperature for 15 min and then diluted with an
appropriate amount of DMEM for the analysis of the gene transfection efficiency.
LipofectamineTM2000 was prepared for the gene transfer study according to the manufacturer’s
guidelines as a comparable lipid based transgenic reagent.
Transfection of neuronal cells with L-MWNT/pDNA complexes
SH-SY5Y cells were used in order to evaluate the gene expression efficiency of the
L-MWNT/pDNA complexes. SH-SY5Y cells (2 x 104 cells) were seeded on 24-well plates and
incubated at 37oC under 5% CO2 for 24 hr with DMEM containing 10% FBS. The medium in
the cell culture dishes was exchanged with a fresh medium (500 μL) containing the
L-MWNT/pDNA complexes at the different lipid-to-pDNA ratios (16 μg/mL of pDNA as a final
concentration). After incubation for 4 hr, the medium was exchanged with a fresh medium
containing 10% FBS. The cells were further incubated for 44 hr, then washed twice with a PBS
solution and finally replaced in DMEM containing 10% FBS (500 μL). Cells were observedwith
a Nikon TE2000U fluorescent microscope equipped with a Nikon DS-5MC USB2 cooled CCD
camera. Total cell number and GFP positive cell number were counted in three different fields

Page 100
Chapter 4
- 90 -
for estimation if the gene expression efficiency; experiments were performed in triplicate. For
the measurement of the gene expression under the presence of a magnetic filed, the complexes
were incubated with the SH-SY5Y cells for 4 hr over a permanent magnet (Eclipse Magnetics
Ltd, N125, diameter 12.7 mm, height 9.5 mm, residual magnetic flux density 1.23 T). The
medium was then exchanged with a DMEM in absence of FBS, followed by 44 hr incubation.
Cytotoxicity of the L-MWNT/pDNA complexes
The cytotoxicity of the L-MWNT/pDNA complexes was investigated on SH-SY5Y cells:
1 x 104 cells were seeded in a 96-well plate and incubated at 37oC under 5% CO2 for 24 hr with a
DMEM containing 10% FBS. The medium in the cell culture dish was then exchanged with a
fresh medium (100 μL) containing a appropriate concentration of the complexes with a
L-MWNT-to-pDNA ratio of 25. After incubation at 37oC for 4 hr, the medium was exchanged
with 100 μL of DMEM 10% FBS, and the cells were further incubated for 20 hr. Cells were
finally incubated with a 0.5 mg/mL MTT (3-(4,5-dimetylthiazol-z-yl)-2,5-dipheyl tetrazolium
bromide) reagent for 2 hr. After cell treatment with 100 μL of DMSO, absorbance at 550 nm was
measured with a VERSAMax microplate reader (Molecular Devices, Sunnyvale, CA, USA).

Page 101
Chapter 4
- 91 -
References
1. Felgner, P. L., Gadek, T. R., Holm, M., Roman, R., Chan, H. W., Wenz, M., Northrop, J. P.,
Ringold, G. M., Danielsen, M. Proc. Natl. Acad. Sci. U. S. A. 84 (1987) 7413–7417.
2. Felgner, J. H., Kumar, R., Sridhar, C. N., Wheeler, C. J., Tsai, Y. J., Border, R., Ramsey, P.,
Martin, M., Felgner, P. L. J. Biol. Chem. 269 (1994) 2550–2561.
3. Gao, X., Huang, L. Gene Ther. 2 (1995) 710–722.
4. Boussif, O., Lezoualc’h, F., Zanta, M. A., Mergny, M. D., Scherman, D., Demeneix, B., Behr,
J. P. Proc. Natl. Acad. Sci. U. S. A. 92 (1995) 7297–7301.
5. Schwarze, S. R., Ho, A., Vocero-Akbani, A., Dowdy, S. F. Science 285 (1999) 1569–1572
6. Rojas, M., Donahue, J. P., Tan, Z., Lin, Y. Z. Nat. Biotechnol. 16 (1998) 370–375.
7. Fawell, S., Seery, J., Daikh, Y., Moore, C., Chen, L. L., Pepinsky, B., Barsoum, J. Proc. Natl.
Acad. Sci. U. S. A. 91 (1994) 664–668.
8. Vocero-Akbani, A. M., Heyden, N. V., Lissy, N. A., Ratner, L., Dowdy, S. F. Nat. Med. 5
(1999) 29–33.
9. Elliott, G., O’Hare, P. Cell 188 (1997) 223–233.
10. Perez, F., Joliot, A. H., Bloch-Gallego, E., Zahraoui, A., Triller, A., Prochiantz, A. J. Cell Sci.
102 (1992) 717–722
11. Derossi, D., Joliot, A. H., Chassaing, G., Prochiantz, A. J. Biol. Chem. 269 (1994)
10444–10450.
12. Mi, Z., Mai, J., Lu, X., Robbins, P. D. Mol. Ther. 2 (2000) 339–347.
13. Schwarze, S. R., Hruska, K. A., Dowdy, S. F. Trends Cell Biol. 10 (2000) 290–295.
14. Bishop, D.K., Park, D., Xu, L., Kleckner, N. Cell 69 (1992) 439–456.
15. Li, Z., Golub, E.I., Gupta, R., Radding, C.M. Proc. Natl. Acad. Sci. U.S.A. 94 (1997)
11221–11226.
16. Hong, E.L., Shinohara, A., Bishop, D.K. J. Biol. Chem. 276 (2001) 41906–41912.

Page 102
Chapter 4
- 92 -
17. Sehorn, M.G., Sigurdsson, S., Bussen, W., Unger, V.M., Sung, P. Nature 429 (2004)
433–437.
18. Bugreev, D.V., Golub, E.I., Stasiak, A.Z., Stasiak, A., Mazin, A.V. J. Biol. Chem. 280 (2005)
26886–26895.
19. Habu, T., Taki, T., West, A., Nishimune, Y., Morita, T. Nucleic Acids Res. 24 (1996) 470–477.
20. Pittman, D.L., Cobb, J., Schimenti, K.J., Wilson, L.A., Cooper, D.M., Brignull, E., Handel,
M.A., Schimenti, J.C. Mol. Cell 1 (1998) 697–705.
21. Yoshida, K., Kondoh, G., Matsuda, Y., Habu, T. Nishimura, Y., Morita, T. Mol. Cell 1 (1998)
707–718.
22. Sonoda, E., Sasaki, M.S., Buerstedde, J.-M., Bezzubova, O., Shinohara, A., Ogawa, H.
Takata, M., Yamaguch, Y., Takeda, S. EMBO J. 17 (1998) 598–608.
23. Sonoda, E., Takata, M., Yamashita, Y.M., Morrison, C. Takeda, S. Proc. Natl. Acad. Sci.
U.S.A. 98 (2001) 8388–8394.
24. Thorpe, P.H., Stevenson, B.J., Porteous, D.J. J. Gene Med. 4 (2002) 195–204.
25. Tsuchiya, H., Uchiyama, M., Hara, K., Nakatsu, Y., Tsuzuki, T., Inoue, H., Harashima, H.
Kamiya, H. Biochemistry 47 (2008) 8754–8759.
26. Marrero, M.B., Schieffer, B., Paxton, W.G.., Schieffer, E. Bernstein, K.E. J. Biol. Chem. 270
(1998) 15734–15738.
27. Merkle, D., Zheng, D., Ohrt, T., Crell, K., Schwille, P. Chembiochem 23 (2008) 1251–1259.
28. Abarzua, P. LoSardo, J.E., Gubler, M.L., Neri, A. Cancer Res. 55 (1995) 3490–3494.
29. Zelphatti, O., Wang, Y., Kitada, S., Reed, J.C., Felgner, P.L. Corbeil, J. J. Biol. Chem. 276
(2001) 35103–35110.
30. Debs, R.J., Freedman, L.P., Edmunds, S., Gaensler, K.L., Düzgünes, N., Yamamoto, K.R. J.
Biol. Chem. 265 (1990) 10189–10192.
31. Lee, A.L.Z., Wang, Y., Ye, W.-H., Yoon, H.S., Chan, S.Y., Yang, Y.-Y. Biomaterials 29
(2008) 1224–1232.

Page 103
Chapter 4
- 93 -
32. Ayame, H., Morimoto, N., Akiyoshi, K. Bioconjugate Chem. 19 (2008) 882–889.
33. Dalkara, D., Zuber, G., Behr, J.P. Mol. Ther. 9 (2004) 964–969.
34. Zhou, X., Huang, L. Biochim. Biophys. Acta 1189 (1994) 195–203.
35. Zuhorn, I.S., Hoekstra, D. J. Membr. Biol. 189 (2002)167–179.
36. Friend, D.S., Papahadjopoulos, D., Debs, R.J. Biochim. Biophys. Acta 1278 (1996) 41–50.
37. Page, B., Page, M., Noel, C. Int. J. of Oncology 3 (1993) 473-476.
38. Lacerda, L., Raffa, V., Prato, M., Bianco, A., Kostarelos, K. Nanotoday 2 (2007) 38–43.
39. Singh, R., Pantarotto, D., McCarthy, D., Chaloin, O., Hoebeke, J., Partidos, C.D., Briand, J-P.,
Prato, M., Bianco, A., Kostarelos, K. J. Am. Chem. Soc. 127 (2005) 4388–4369.
40. Erhardt, M., Singh, R., MacCarthy, D., Erhardt, M., Briand, J-P., Prato, M., Kostarelos, K.,
Bianco, A. Angew. Chem. Int. Ed. 43 (2004) 5242–5246.
41. Gao, L., Nie, L., Wang, T., Qin, Y., Guo, Z., Yang, D., Yan, X. ChemBioChem 7 (2006)
239–242.
42. Zhang, Z., Yang, X., Zhang, Y., Zeng, B., Wang, S., Zhu, T., Rpden, R.B.S., Chen, Y., Yang,
R. Cli. Cancer Res. 12 (2006) 4933–4939.
43. Islam, M.F., Rojas, E., Bergey, D.M., Johnson, A.T., Yodh, A.G. Nano Lett. 3 (2003)
269–273.
44. Obata, Y., Suzuki, D., Takeoka, S. Bioconjugate Chem. 19 (2008) 1055–1063.
45. Obata, Y., Saito, S., Takeda, N., Takeoka, S. Submitted for publication (2008).
46. Okamura, Y., Maekawa, I., Teramura, Y., Maruyama, H., Handa, M., Ikeda, Y., Takeoka, S.
Bioconjugate Chem. 16 (2005) 1589–1596.
47. Obata, Y., Ikegaya, N., Sakane, I., Kurumizaka, H., Takeda, N., Takeoka, S. Submitted for
publication (2008).
48. Farhood, H., Serbina, N., Huang, L. Biochim. Biophys. Acta 1235 (1995) 280–295.
49. Shvedova, A.A., Castranova, V., Kisin, E.R., Schwegler-Berry, D., Murray, A.R.,
Gandelsman, V.Z., Maynard, A., Baron, P. J. Toxicol. Environ. Health, Part A 66 (2003)

Page 104
Chapter 4
- 94 -
1909–1926.
50. Ciofani, G., Obata Y., Sato I., Okamura, Y., Raffa, V., Menciassi, A., Dario, P., Takeda, N.,
Takeoka, S. J Nanopart Res. 2008, DOI: 10.1007/s11051-008-9415-y.
51. Ciofani, G.., Raffa, V., Obata, Y., Menciassi, A., Dario, P., Takeoka, S. Curr. Nanoscience 4
(2008) 212–218.
52. Liang, X.F., Wang, H.J., Luo, H., Tian, H., Zhang, B.B., Hao, L.J., Teng, J.I., Chang, J.
Langmuir 24 (2008) 7147–7153.
53. Matarredona, O., Rhoads, H., Li, Z., Harwell, J.H., Balzano, L., Resasco, D.E. J. Phys. Chem.
B 107 (2003) 13357–13367.
54. Yurekli, K., Mithell, C., Krishnamoorti, R. J. Am. Chem. Soc. 126 (2004) 9902–9903.
55. Wong Shi Kam, N., Liu, Z., Dai, H. J. Am. Chem. Soc. 127 (2005) 12492–12493.
56. Lee, C.H., Ni, Y.H., Chen, C.C., Chou, C., Chang, F.H. Biochim. Biophys. Acta 1611 (2003)
55–62.
57. Medina-Kauwe, L.K., Xie, J., Hamm-Alvarez, S. Gene Ther. 12 (2005) 1734–1751.
58. Kogure, K., Moriguchi, R., Sasaki, K., Ueno, M., Futaki, S., Harashima, H. J. Cont. Release
98 (2004) 317–323.
59. Isojima, T., Lattuada, M., Vander Sande, J. B., Alan Hatton, T. ACS Nano DOI:
10.1021/nn800089z (2008).
60. Cai, D., Mataraza, J.M., Qin, Z-H., Huang, Z., Huang, J., Chiles, T.C., Carnahan, D., Kempa,
K., Ren, Z. Nature Methods 2 (2005) 449–454.

Page 105
Chapter 5
- 95 -
Chapter 5
Charge Convertible Amphiphiles for Constructing pH-Sensitive
Liposomes
1. Introduction
Nanotechnology has shown tremendous promise in target-specific delivery of drugs and
genes in the body. Although passive and active targeted-drug delivery has addressed a number of
important issues, additional properties that can be included in nanocarrier systems to enhance the
bioavailability of drugs at the disease site, and especially upon cellular internalization, are very
important. A nanocarrier system incorporated with stimuli-responsive property (e.g., pH,
temperature, or redox potential), for instance, would be amenable to address some of the
systemic and intracellular delivery barriers.
In this chapter, the author discusses the role of stimuli-responsive nano carrier systems
for drug and gene delivery. The advancement in material science has led to design of a variety of
materials, which are used for development of nanocarrier systems that can respond to biological
stimuli. Temperature, pH, and hypoxia are examples of “triggers” at the diseased site that could
be exploited with stimuliresponsive nanocarriers. With greater understanding of the difference
between normal and pathological tissues and cells and parallel developments in material design,
there is a highly promising role of stimuli-responsive nanocarriers for drug and gene delivery.

Page 106
Chapter 5
- 96 -
2. Intracellular Drug Delivery Using pH-Sensitive Liposomes
2-1. Intracellular Transition of Liposomes
In 1980, it was suggested that there might be a possible clinical application of liposome,
which was designed to release their contents in an environment such as tumor tissue with a
slightly decreased pH of around 6
1
. An advanced approach of pH-sensitive liposome is the
intracellular pH-triggered release of liposome compounds from the endocytotic pathway to the
cytosol. For this purpose one should understand the cellular mechanisms of uptake and
intracellular trafficking of colloids and particles such as liposomes
2-5
.
In general, the first step of the cellular uptake is independent of the mocrotubular system
and is followed by fusion of the interiorized PMV with plasma membrane vesicle wit the early
endosome. The early endosome is a compartment for rapid sorting of compounds such as
receptor proteins, which are then recycled to the plasma membrane, and other endocytosised
material such as liposomes, which are destined for the transport to the interior of the cell, i.e. to
the perinuclear region. Liposomes remain in the early endosome only for 2-3 min, at a maximum
Fig. 5-1. Pathway of endocytosised liposomes.

Page 107
Chapter 5
- 97 -
10 min at a slightly acidic pH of 6.3-6.8. Then, they are transported by carrier vesicles, actively
mediated by the microtubular system, to the late endosomes and to the lysosomes. The contents
of the early endosome are accumulated and concentrated in the late endosome. These vesicles
already contain hydrolases, which are delivered by fusion with vesicles from the golgi appartus.
Late endosomes are also called pre-lysosomes, because they are likely to initiate the degradative
process of their contents, which is finished in the lysosomes. Before the degradation happens,
liposomes, or at least their active compounds, should escape from the endosomal system to the
cytoplasm. The pH in the carrier vesicles decreases to a value of about 5.5 within a
transportation time of between 20 min and 3 hr, depending on the cell type. This drop in pH is
used in the concept of pH-sensitive liposomes to trigger the endosomal escape of their
compounds.
2-2. Conventional pH-Sensitive Liposomes
Numerous pH-sensitive liposomes with formulations of pH-sensitive lipids or mixtures of
lipids with macromolecules were developed to improve the delivery of active substances to the
site of action in the target cell. pH-Sensitive liposomes are formed from components which adopt
a lamellar phase at the physiological pH around 7.4. By decreasing the pH to a critical value, or a
range around 5.5, the liposomes become fusogenic and favorably fuse with the endosomal
membrane. This is the desired mechanism to release the liposomal contents into the cytoplasm.
The predominant reason for fusogenicity is the transition from a lamellar to hexagonal HII phase,
leading to a larger space of the lipophilic part of the lipid structure. Simultaneously-normally an
unwanted side-effect-membrane defects occur, which result in the release of entrapped
substances into the liposome-surrounding compartment, i.e. the endosome vesicle itself.
Therefore, the quantification of both the drug release and the ability to undergo membrane fusion
are important in vitro tests for developing pH-sensitive liposomes. Compounds which change
their molecular shape as a result of a decreased pH are needed for the desired structural changes,

Page 108
Chapter 5
- 98 -
as shown for different lipid mixture in Fig. 5-2.
One possibility is that mixtures of lipids which are shaped conically and inverted
conically at neutral pH, adopt an overall lamellar structure. Cones have a small polar headgroup
and take up a larger space at the hydrophobic tails. Polyunsaturated phosohatidylethanolamine
has pronounced properties of this type of membrane lipids. As a single compound, it forms a
Fig. 5-2. Transition from lamellar to non-lamellar phases in pH-sensitive lipid mixtures.

Page 109
Chapter 5
- 99 -
lamellar phase at a pH > 9
6
. In the pH range of interest (4.5-7.5) it is zwitterionic. In most cases
DOPE is used, which is conically shaped even at temperatures far below 0
o
C
7
. As the thermal
motion influences the dimension of the molecule at the hydrophobic tail, lipid mixtures with
DOPE are sensitive to temperature
8
.
As s second compound, an inverted conically shaped lipid, normally acidic and
containing a carboxyl group, is needed. Numerous suitable lipids have been studied in the
literature. In this section, only two species are discussed in great detail. One lipid of this type is
cholesterolhemisuccinate (CHEMS)
9
, which stabilizes the lamellar phase in DOPE/CHEMS
mixtures when its amount exceeds 20 mol%
10
. CHEMS itself exhibits pH-sensitive
polymorphism
11
. By decreasing the pH to a range around the pKa value of 5.8, the weak acid is
increasingly protonated, loses its charge, and reduces its hydration. As a result, the headgroup
vecomes smaller and the inverted cone is transformed to a cylinder- or cone-like structure
9,12
.
The overall lamellar structure of DOPE/CHEMS is therefore lost.
Recently developed systems contain binary mixtures of anionic and cationic lipids. At
neutral pH the mixture is overall anionic. Upon decreasing the pH, parts of the acidic lipids
become protonated and uncharged. Best fusogenicity of the formulation is achieved when the net
surface charge of the pH-sensitive liposome membrane is zero. The critical acidic pH can be
triggered by the lipid composition from 4–6.7
13
. Most of the pH-sensitive liposome formulation
are anionic at physiological pH 7.4. Repulsive forces of negative charges of liposomal and
cellular surfaces may reduce their interaction, which limits the cellular uptake of the liposomes.
3. Synthesis of Zwitterionic Lipids as Charge Convertible Component
3-1. Charge Convertible Liposomes for Improving Intracellular Drug Delivery
For introduction of drug and gene into living cells, cationic liposomes preferably utilized
due to high cellular uptake efficiency. The endosomal escape membrane fusion occurs between
the cationic liposomal components and the endosomal membrane of early endosomes (pH

Page 110
Chapter 5
- 100 -
6.0-6.5), leading that the liposomal contents are able to release into cytoplasm without
degradation. The electrostatic interaction is quite reasonable to release contents. Therefore,
cationic liposomes are mostly utilized to enhance the intracellular drug or gene delivery [9–13].
However, the cationic liposomes lack blood circulating ability after an intravenous injection
because they are tend to aggregate with serum proteins and to adhere electrostatically to vascular
endothelium cells, then a little of them are delivered to the target tissues. Then, author developed
charge convertible liposomes for efficient extra- and intracellular drug delivery.
3-2. Zwitterionic Lipids in Response to Endosomal pH
The author basically synthesized a series of zwitterionic lipids having glutamic acid(s) at
head group as shown in Chart 3-1 to soak appropriate structure for improving intracellular drug
delivery.
Chart 5-1. A series of zwitterionic lipids having glutamic acid at head group.
N
O
O
O
O
C
HO
N
H
C
O
HOOC
NH2
H2N
(CH2)n
(CH2)n
CH3
CH3
N
O
O
O
O
C
HO
H2N
NH
C
O
NH2
HOOC
(CH2)n
(CH2)n
CH3
CH3
N
O
O
O
O
C
HO
H
N
C
O
HOOC
NH2
NH
(CH2)n
(CH2)n
CH3
CH3
C
O
HOOC
NH2
N
O
O
O
O
C
HO
NH
C
O
HOOC
NH
C
O
(CH2)n
(CH2)n
HOOC
CH3
CH3
NH2
NH2
N
O
O
O
O
C
HO
H
N
C
O
HOOC
NH2
N
H
C
O
(CH2)n
(CH2)n
NH2
HOOC
CH3
CH3
N
O
O
O
O
C
HO
NH
C
O
HOOC
NH
C
O
(CH2)n
(CH2)n
HOOC
CH3
CH3
NH2
NH2
I
II
III
IV
V
VI

Page 111
Chapter 5
- 101 -
As comparable zwitterionic lipids, the author focused on head group of the lipids;
1,5-dihexadecyl N-lysyl-L-glutamate. The two amino groups of 1,5-dihexadecyl
N-lysyl-L-glutamate was linked to α- or γ-carboxylic group of glutamic acid via amino linkage.
When all the compounds were synthesized as described in Experimental section, the each
zwitterionic lipids were mixed with cholesterol for mole ratio of 1-to-1 and prepared liposomes
by extrusion. The resulting liposomes were analyzed in zeta potential and fusogenic potential to
anionic membrane at various pHs.
Compound I or II liposomes displayed positive zeta potential at pH7.4. Then, the zeta
potential was increased with decreasing pH. The fusogenic potential was almost independent on
pH. By contrast, when two glutamic acids was inserted to 1,5-dihexadecy N-lysyl-L-glutamate as
shown in compound III–VI, minus zeta potential at pH7.4 was provided. Indeed, charge
Fig. 5-3. Zeta potential and fusogenic potential.of liposomes composed of synthetic
zwitterionic lipids and cholesterol at various pHs.
pH
10
20
30
40
50
Lipid mix
ing ra
tio
%
3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
-20
0
40
60
80
Zeta pote
n
tial [m
V]
0
-40
100
20
pH
10
20
30
40
50
Lipid mix
ing ra
tio
%
3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
-20
0
40
60
80
Z
e
ta po
te
n
tial [m
V
]
0
-40
100
20
pH
10
20
30
40
50
Lipid
mix
ing ra
tio
%
3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
-20
0
40
60
80
Z
e
ta
po
te
ntia
l [m
V
]
0
-40
100
20
pH
10
20
30
40
50
Lipid
mix
ing ra
tio
%
3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
-20
0
40
60
80
Z
e
ta
poten
tial [m
V]
0
-40
100
20
pH
10
20
30
40
50
Lipid
mix
ing ra
tio
%
3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
-20
0
40
60
80
Zeta po
te
n
tia
l [m
V
]
0
-40
100
20
pH
10
20
30
40
50
Lipid mix
ing ra
tio
%
3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
-20
0
40
60
80
Z
e
ta
po
te
ntia
l [m
V
]
0
-40
100
20
I
II
III
IV
V
VI

Page 112
Chapter 5
- 102 -
conversion from negative to positive based on pH was obtained in compound III-VI. The pH
where zeta potential converts from negative to positive was 6.0, 6.5, 6.0, or 5.7 in case of
compound III, IV, V, or VI, respectively. Therefore, the author was able to notice control the pH.
Furthermore, these liposomes containing compound III-VI showed low fusion to anionic
membrane at pH7, where the liposomes displayed negative zeta potential. That was due to
electrostatic repulsion between anionic membrane and negatively charged liposomes containing
synthetic zwitterionic lipids. As a result, the author was able to confirm head group structure
dependent dynamics of zeta potential and fusogenic potential to anionic membrane.
4. Experimental Section
Syntheses of zwitterionic lipids
All compounds as shown in Chart 5-1 were synthesized as following scheme.
[Compound I]
NH2
O
O
O
O
(CH2)15
(CH2)15
CH3
CH3
Z-Lys(Boc)-OSu
N
O
O
O
O
C
H
O
N
H
C
N
(CH2)15
(CH2)15
CH3
CH3
H
C
O
O
O
O
H2, Pd-black
N
O
O
O
O
C
H
O
H2N
N
(CH2)15
(CH2)15
CH3
CH3
H
C
O
O
Boc-Glu(OtBu)-OSu
N
O
O
O
O
C
H
O
NH
N
(CH2)15
(CH2)15
CH3
CH3
H
C
O
O
C
O
NH
C
C
O
O
O
O
TFA
N
O
O
O
O
C
H
O
NH
H2N
(CH2)15
(CH2)15
CH3
CH3
C
O
NH2
HOOC
C37H73NO4
Exact Mass: 595.55
Mol. Wt.: 595.98
C56H99N3O9
Exact Mass: 957.74
Mol. Wt.: 958.40
C48H93N3O7
Exact Mass: 823.70
Mol. Wt.: 824.27
C62H116N4O12
Exact Mass: 1108.86
Mol. Wt.: 1109.60
C48H92N4O8
Exact Mass: 852.69
Mol. Wt.: 853.27

Page 113
Chapter 5
- 103 -
[Compound2]
[Compound3]
[Compound 4]
NH2
O
O
O
O
(CH2)15
(CH2)15
CH3
CH3
Boc-Lys(Z)-OSu
N
O
O
O
O
C
H
O
N
H
N
(CH2)15
(CH2)15
CH3
CH3
H
H2, Pd-black
Boc-Glu(OtBu)-OSu
C37H73NO4
Exact Mass: 595.55
Mol. Wt.: 595.98
C
O
O
C
O
O
N
O
O
O
O
C
H
O
N
H
H2N
(CH2)15
(CH2)15
CH3
CH3
C
O
O
N
O
O
O
O
C
H
O
N
H
NH
(CH2)15
(CH2)15
CH3
CH3
C
O
O
C
O
NH
C
C
O
O
O
O
N
O
O
O
O
C
H
O
H2N
NH
(CH2)15
(CH2)15
CH3
CH3
C
O
NH2
HOOC
TFA
C56H99N3O9
Exact Mass: 957.74
Mol. Wt.: 958.40
C48H93N3O7
Exact Mass: 823.70
Mol. Wt.: 824.27
C62H116N4O12
Exact Mass: 1108.86
Mol. Wt.: 1109.60
C48H92N4O8
Exact Mass: 852.69
Mol. Wt.: 853.27
NH2
O
O
O
O
(CH2)15
(CH2)15
CH3
CH3
Boc-Lys(Z)-OSu
N
O
O
O
O
C
H
O
N
H
N
(CH2)15
(CH2)15
CH3
CH3
H
Boc-Glu(OtBu)-OSu
C37H73NO4
Exact Mass: 595.55
Mol. Wt.: 595.98
C
O
O
C
O
O
H2, Pd-Black
C56H99N3O9
Exact Mass: 957.74
Mol. Wt.: 958.40
N
O
O
O
O
C
H
O
H2N
N
(CH2)15
(CH2)15
CH3
CH3
H
C
O
O
N
O
O
O
O
C
H
O
H
N
N
(CH2)15
(CH2)15
CH3
CH3
H
C
O
O
C
O
NH
C
C
O
O
O
O
TFA
N
O
O
O
O
C
H
O
H
N
H2N
(CH2)15
(CH2)15
CH3
CH3
C
O
NH
C
C
O
O
O
O
N
O
O
O
O
C
H
O
H
N
NH
(CH2)15
(CH2)15
CH3
CH3
C
O
NH
C
C
O
O
O
O
Boc-Glu-OtBu
C
O
NH
C
TFA
N
O
O
O
O
C
H
O
NH
NH
(CH2)15
(CH2)15
CH3
CH3
C
O
NH2
HOOC
C
O
H2N
COOH
O
O
C
O
O
C51H91N3O7
Exact Mass: 857.69
Mol. Wt.: 858.28
C65H114N4O12
Exact Mass: 1142.84
Mol. Wt.: 1143.62
C57H108N4O10
Exact Mass: 1008.81
Mol. Wt.: 1009.49
C71H131N5O15
Exact Mass: 1293.96
Mol. Wt.: 1294.82
C53H99N5O11
Exact Mass: 981.73
Mol. Wt.: 982.38

Page 114
Chapter 5
- 104 -
Indeed, the author will introduce synthesis of compound VI in chapter 6.
NH2
O
O
O
O
(CH2)15
(CH2)15
CH3
CH3
Boc-Lys(Z)-OSu
N
O
O
O
O
C
H
O
N
H
N
(CH2)15
(CH2)15
CH3
CH3
H
Boc-Glu-OtBu
C37H73NO4
Exact Mass: 595.55
Mol. Wt.: 595.98
C
O
O
C
O
O
H2, Pd-Black
C56H99N3O9
Exact Mass: 957.74
Mol. Wt.: 958.40
N
O
O
O
O
C
H
O
H2N
N
(CH2)15
(CH2)15
CH3
CH3
H
C
O
O
N
O
O
O
O
C
H
O
NH
N
(CH2)15
(CH2)15
CH3
CH3
H
C
O
O
TFA
N
O
O
O
O
C
H
O
NH
H2N
(CH2)15
(CH2)15
CH3
CH3
Boc-Glu(OtBu)-OSu
TFA
C51H91N3O7
Exact Mass: 857.69
Mol. Wt.: 858.28
C
O
NH
C
O
O
C
O
O
C
O
NH
C
O
O
C
O
O
N
O
O
O
O
C
H
O
NH
NH
(CH2)15
(CH2)15
CH3
CH3
C
O
NH
C
O
O
C
O
O
C
O
NH
C
C
O
O
O
O
N
O
O
O
O
C
H
O
NH
NH
(CH2)15
(CH2)15
CH3
CH3
C
O
H2N
COOH
C
O
NH2
HOOC
C71H131N5O15
Exact Mass: 1293.96
Mol. Wt.: 1294.82
C53H99N5O11
Exact Mass: 981.73
Mol. Wt.: 982.38
C57H108N4O10
Exact Mass: 1008.81
Mol. Wt.: 1009.49
C64H111N4O12•
Exact Mass: 1127.82
Mol. Wt.: 1128.59

Page 115
Chapter 5
- 105 -
References
1. Yatvin, M.B., Kreutz, W., Horowitz, B.A., Shinitzki, M. Science 210 (1980) 1253–1255.
2. Silverstein, S.C., Steinmann, R.M., Cohn, Z.A. Annu. Rev. Biochem. 46 (1977) 669–722.
3. Kelly, R.B. Cell 61 (1990) 5–8.
4. Mellmann, I. Annu. Rev. Cell Dev. Biol. 12 (1996) 575–625.
5. Düzgünes, N., Nir, S. Adv. Drug Deliv. Rev. 40 (1996) 3–18.
6. Papahadjopoulos, D. Biochim. Biophys. Acta 163 (1968) 240–254.
7. Cullis, P.R., de Kruiff, B. Biochim. Biophys. Acta 559 (1980) 359–369.
8. Torchilin, V.P., Omelyanenko, V.G.., Lukyanov, A.N. Anal. Biochem. 204 (1992) 107–109.
9. Ellens, H., Benz, J., Szoka, F.C. Biochemistry 23 (1984) 1532–1538.
10. Lai, M.-Z., Vail, W.J., Szoka, F.C. Biochemistry 24 (1985) 1654–1661.
11. Hafez, I., Cullis, P.R. Biochim. Biophys. Acta 1463 (2000) 107–114.
12. Van Bambeke, F., Kerkhofs, A., Schanck, A., Remacle, C., Sonveaux, E., Tulkens, P.M.,
Lipids 35 (2000) 213–216.
13. Hafez, I., Ansell, S., Cullis, P.R. Biophys. J. 79 (2000) 1438–1446.

Page 116
Chapter 5
- 106 -

Page 117
Chapter 6
- 107 -
Chapter 6
Evaluation of Charge Convertible Liposomes Containing Synthetic
Zwitterionic Lipids for Efficient Systemic Drug Delivery
1. Introduction
Conventional anticancer drugs have many adverse effects resulting from non-selective
toxicity and distribution of the drug to normal cells. Liposomes were one of the first
nanoparticulate drug delivery systems to show increased delivery of small molecule anticancer
drugs to solid tumors. Drug-encapsulating liposomes with diameters in the range of 100 nm can
accumulate in solid tumors via the enhanced permeability and retention (EPR) effect, which
occurs when nanoparticles extravasate from the circulation into tumors through gaps in the tumor
vasculature endothelium. After delivery of their contents, the liposomes are introduced cancer
tumor by endocytosis. Then, the active drug release into content expects to show high
pharmaceutical effects. For this background, the author prepares Doxorubicin-encapsulating
liposomes containing charge convertible lipids as drug carrier model. The application of charge
convertible liposomes containing zwitterionic lipids for drug delivery is described in this chapter.

Page 118
Chapter 6
- 108 -
2. Preparation of Doxorubicin-Encapsulating Liposomes
2-1. Encapsulation of Small Molecular Weight Drugs
The recognition that a variety of chemotherapeutic drugs could be accumulated within
LUVs exhibiting transmembrane pH gradients followed earlier studies on membrane potentials
and the uptake of weak bases used for measurement of ΔpH 1. The technique is based on the
membrane permeability of the neutral form of weakly basic drugs such as doxorubicin. When
doxorubicin (pKa=8.6) is incubated in the presence of LUVs exhibiting a ΔpH (interior acidic),
the neutral form of the drug will diffuse down its concentration gradient into the LUV interior,
where it will be subsequently protonated and trapped (the charged form is membrane
impermeable). As long as the internal buffer (300 mM citrate pH4) is able to maintain the ΔpH,
diffusion of neutral drug will continue until either all the drug has been taken up, or the buffering
capacity of the interior has been overwhelmed. This process is illustrated in Fig. 6-1 for the
uptake of doxorubicin in to LUVs, where it is seen that uptake is dependent on time, temperature,
and lipid composition 2. Under appropriate conditions high drug-to-lipid ratios can be achieved
with high trapping efficiencies and excellent drug retention. The method for encapsulation is
described below.
Fig. 6-1. Diagrammatic representations of drug uptake in response to transmembrane pH gradients.

Page 119
Chapter 6
- 109 -
The standard pH gradient methods, where in ΔpH is established by buffer exchange on a gel
exclusion column, is summarized in the left column. A second method for generating ΔpH
involves the initial formation of a transmembrane gradient of ammonium sulfate, which leads to
an acidified vesicle interior as neutral ammonia leaks from the vesicles. Transmembrane pH
gradients can also be established by ionofores in response to transmembrane ion gradients (e.g.
Mn2+, represented as solid circles in Fig. 6-1)
These techniques has been applied to a variety of drugs including doxorubicin 3-6,
epirubicin 3, ciprofloxacin 7,8, and vincristine 7. A variety of alkylammonium salts (e.g.
methylammonium salfate, porpyl ammonium salfate, amylammonium salfate) can be used in
place of ammonium sulfate. Some drugs, such as doxorubicin, precipitate and form a gel in the
vesicle interior 4,5, while others, such as ciprofloxacin , don’t 8. The physical state of
encapsulated drug will clearly affect retention and therefore may impact efficacy.
2-2. Syntheses of Charge Convertible Lipids
1,5-Dihexadecyl
N-glutamyl-L-glutamate
(L1)
and
1,5-dihexadecyl
N,N-diglutamyl-lysyl-L-glutamate (L2) were synthesized as shown in Scheme 6-1. The two
zwitterionic lipids was included to liposomal component to compare DPPC-containing
liposomes.

Page 120
Chapter 6
- 110 -
2-3. Zeta Potential Conversion of Charge Convertible Liposomes
The zeta,potentials of the L1- or L2-containing liposomes were measured at various pH
values (Fig. 6-2). The liposomes converted their zeta ,potential from negative to positive when
the pH value was decreased from 7.4 to 4.0. At pH4.0, the zeta potentials of the L1- or
L2-containing liposomes were +21 mV or +59 mV, respectively. Whereas, at pH7.4, the zeta
potentials of the L1 or L2-containing liposomes were -22 mV or -17 mV, respectively. In
addition, the pH values of the L1- or L2-containig liposomes where the zeta potential became
zero were around 4.6 or 5.6, respectively. On the other hand, the DPPC-containing liposomes
showed a constant zeta potential value in all pH regions (pH4.0–7.4).
H2N
O
O
O
O
(CH2)15
(CH2)15
CH3
CH3
H2N
OH
O
O
OH
N
O
O
O
O
(CH2)15
(CH2)15
CH3
C
HO
N
N
CH3
N
O
O
O
O
(CH2)15
(CH2)15
CH3
C
HO
N
CH3
H
C
O
NH2
N
H
C
O
HOOC
NH2
Reaction conditions: (A) Hexadecylalcohol, p-Tos, Benzene; (B) Boc-Glu(OtBu)-OSu, TEA, DCM; (C) TFA, r.t.; (D) Boc-Lys(Boc)-OSu, TEA, DCM;
(E)TFA, 4oC; (F) Boc-Glu(OtBu)-OSu, TEA, DCM; (G) TFA, r.t.
Boc
Boc
HOOC
N
O
O
O
O
(CH2)15
(CH2)15
CH3
C
HO
H2N
H2N
CH3
N
O
O
O
O
(CH2)15
(CH2)15
CH3
C
HO
N
CH3
H
C
O
N
N
H
C
O
OC
N
OC
Boc
Boc
But
But
N
O
O
O
O
(CH2)15
(CH2)15
CH3
C
HO
N
OC
CH3
Boc
But
N
O
O
O
O
(CH2)15
(CH2)15
CH3
C
HO
H2N
HOOC
CH3
A
B
C
D
E
F
G
L1
L2
Chart. 6-1. Syntheses of amino-acid based zwitterionic lipids having glutamic acid headgroup of the
lipids.

Page 121
Chapter 6
- 111 -
The author confirmed that both L1- and L2-containing liposomes converted their zeta
potential according to the surrounding pH. The structure of the synthetic zwitterionic lipid
should influence the dynamics of the zeta potential of the liposomes. At pH7.4, the zeta potential
of the L2-containing liposomes was higher than that of the L1-containing liposomes, indicating
the amino group of the glutamtic acid of L2 would be located at the outer surface than that of L1,
though the carboxyl groups of both L1- and L2-containing liposomes would be located at the
most outer surface. Among the zeta potential measurement, the amino-groups of L1 and L2
would be fully protonated, therefore, the zeta potential conversion depending on pH might be
due to the conversion of protonation degree of the carboxyl groups oriented at the liposomal
surface. In addition, the L2-containing liposomes showed the zeta potential zero at around pH5.6
compared to pH4.6 in the L1-containing liposomes. The pH where the zeta potential became
zero was dependent on the lipid structure, because the neighboring carboxylate anion would
restrict the dissociation of the carboxylic acid. This phenomenon could be generally realized by
4.0
6.0
0
-20
-40
60
40
20
pH
Z
e
ta
-potential (m
V
)
7.5
80
7.0
5.0
Fig. 6-1.
The zeta, potential of the liposomes composed of (open circles)
DPPC/cholesterol/PEG-Glu2C18, (closed circles) L1/cholesterol/PEG-Glu2C18, and (squares)
L2/cholesterol/PEG-Glu2C18. Liposomes were dispersed in 20 mM acetate buffer and measured at
37oC.

Page 122
Chapter 6
- 112 -
referring to the difference of pKa between acrylic acid (pKa 4.5) and polyacrylic acid (pKa 6.8).
Moreover, the charge conversion of the liposomes containing the synthetic zwitterionic lipids
was also recognized in the membrane modified with the PEG-lipids. It is indicated that the
charge conversion of the synthetic lipids triggered by pH change could affect the outer of PEG
layer. In this study, the pH dependent zeta potential of the liposomes could be demonstrated by
using the L1 and L2 type lipids, resulting in the providing pH-sensitive liposomes.
2-4. Evaluation of Fusogenic Potential Charge Convertible Liposomes
The fusogenic potential of the pH-sensitive liposomes was the most important property to
release contens from endosome to cytosplasm. As a model anionic membrane of endosome, the
negatively charged liposomes (DPPC/DPPG) were used. As shown in Fig. 6-2, the
DPPC-containing liposomes showed a little fusogenic potential to the negatively charged
membrane, whereas the pH-sensitive liposomes containing L1 or L2 showed a high fusogenic
potential at low pH regions, where the zeta potential of the liposomes was positive. The fusion of
the L1-containing liposomes occurred at pH values lower than pH5.5, and the maximum value of
41 % was observed at pH4.5 for 1 hr incubation. In the case of the L2-containing liposomes, the
pH was 6.0 and the maximum value was 36% at pH5.0.

Page 123
Chapter 6
- 113 -
The membrane fusion preferably occurs between the positively charged and negatively
charged membranes, and the model membrane was conventionally used for evaluation of the
endosomal release efficiency of the cationic liposome-mediated gene delivery 9. In our study, the
fusion between the negatively charged liposomes and the positively charged pH-sensitive
liposomes was expected by converting of their zeta potential by lowing pH. The fusion was
clearly increased when the pH value became less than pH5.0 or 6.0 for L1 or L2-containing
liposomes, respectively. Since, the zeta potential of the L1 or L2-containing liposomes became
positive in these pH regions, the positively charged liposomes should be adhered to the
negatively charged model membrane, followed by the membrane fusion after the flip-flop
movement of the lipid components. In addition, the fusogenic potential was maximized at pH4.5
for L1 and pH5.0 for L2, because the complexes between the negatively charged liposomes
(DPPC/DPPG, –20 mV) and positively charged L1 and L2-containing liposomes (6 mV at pH4.5
4.0
5.0
6.0
20
10
0
50
40
30
pH
Lip
id M
ixing
ra
tio (%
)
7.5
7.0
Fig. 6-2.
Fusogenic potential of the liposomes composed of (open circles)
DPPC/cholesterol/PEG-Glu2C18, (closed circles) L1/cholesterol/PEG-Glu2C18, and (squares)
L2/cholesterol/PEG-Glu2C18 to anionic liposomes (DPPC/DPPG) at various pH values.

Page 124
Chapter 6
- 114 -
for L1, 9 mV at pH5 for L2) easily formed by appropriate electrostatic interaction. Comprared
the fusogenic potentials of L1- or L2-containing liposomes with other cationic liposomes, the
similar values were reported so far 10. Therefore, the positively charged liposomes from the
synthetic cationic lipids would be able to show a high drug delivering capacity as the similar
way to the conventional cationic liposomes. Furthermore, L2 is more attractive than L1 because
the L2-containing liposomes could easily fuse the anionic membrane in the early endosomal pH
(5.5–6.0). On the other hand, the L1-containing liposomes might be sensitive to the late
endosomal pH (4.0–5.0). Therefore, a high fusogenic ability to the endosomal membrane of the
pH-sensitive liposomes containing L1 or L2 were imitatively estimated.
2-5. Characterization of DOX-Encapsulating Liposomes
The author prepared DOX-encapsulting liposome by using a transmembrane pH gradient
method. Although DOX is one of the most widely used anti-cancer agents, its clinical
application is still limited by its side effects such as myelosuppression, gastrointestinal toxicity,
and especially cardiotoxicity 11,12. As a model drug carrier, the author prepared the
DOX-encapsulating pH-sensitive liposomes containing L1 or L2 with an average diameter of
149 + 69 nm or 98 + 48 nm, respectively (Table 6-1). The DPPC-containing liposomes were also
prepared as pH-insensitive liposomes (120 + 47 nm). TEM observation of the
DOX-encapsulating liposomes supported the existence of approximately 100 nm liposomes. The
encapsulating efficiency of DOX was more than 90% for all liposome samples. The diameter of
the DOX-encapsulating liposomes was measured by a DLS measurement to monitor the
dispersion stability. All the liposome samples were clarified to show no aggregation of
liposomes for more than one month at 4oC. During preparation of the liposomes, the
lipid-to-DOX was 0.1. In addition, the DOX concentration of the liposomal dispersion was
adjusted to less than 0.34 mM because aggregation between DOX and liposomes occured when
DOX concentration became more than 0.5 mM (data not shown). Aggregation was also reported

Page 125
Chapter 6
- 115 -
in a previous paper, as the bundle formation of DOX fibers with liposomes 13.
After preparation of the DOX-encapsulating liposomes containing L1 or L2, the
retention profiles of DOX by the pH-sensitive liposomes were invesitigated in various pHs. The
liposome dispersion was incubated in acetate buffer wtih various pH for 1 hr and then the
amount of DOX released from the liposomes was measured after separation of the released
DOX from the DOX-encapsulating liposomes. Interestingly, any DOX release was not
confirmed for the DOX-encapsulating liposomes containing L1 or L2. It is indicated that the
membrane permeability of DOX was quite low even in acidic pH.
DPPC / cholesterol / PEG-Glu2C18
(5/5/0.03)
L1 / cholesterol / PEG-Glu2C18
(5/5/0.03)
L2 / cholesterol / PEG-Glu2C18
(5/5/0.03)
Membrane components
Size [nm]
120 + 47
149 + 69
98 + 48
TEM image
100 nm
100 nm
100 nm
Table 6-1. Characteristics of DOX-encapsulating liposomes containing DPPC, L1, or L2.

Page 126
Chapter 6
- 116 -
3. Evaluation of DOX-Encapsulating Liposomes Containing Zwitterionic Lipids
3-1. Cellular Uptake Efficiency of Liposomes
The efficiency of the cellular uptake of the DOX-encapsulating liposomes was
represented as a weight ratio of DOX to total proteins from HeLa cells (Fig. 6-3). In all liposome
samples such as DPPC-, L1-, or L2-containing liposomes, efficiency at 37oC was significantly
higher than that at 4oC, indicating the endocytosis-dependent cellular uptake. Then, the author
represented the substantial efficiency of the cellular uptake of DOX as the difference of the
efficiency between at 4oC and 37oC. Comparing the substantial efficiency among the liposome
samples, the almost equal uptake levels suggested that the liposomes composed of those
synthetic lipids didn’t affect the endocytosis-dependent cellular uptake cellular uptake at
physiological conditions (pH7.4, 37oC). At pH7.4, the investigated liposomes diplayed
comparable physical properties to the conventional liposomes for minus zeta potential.
5
0
15
Cellu
lar u
p
ta
ke
e
fficienc
y
[N
g-D
O
X
/m
g
-P
ro
te
in
]
10
20
DPPC
L1
L2
4oC
37oC
Fig. 6-3. Cellular uptake of the liposomes composed of DPPC/cholesterol/PEG-Glu2C18,
L1/cholesterol/PEG-Glu2C18, and L2/cholesterol/PEG-Glu2C18 on HeLa cells under (ash bar) 4oC or
(closed bar) 37oC for 2 hr.

Page 127
Chapter 6
- 117 -
Though, the liposomes containing L1 or L2 were found to be introduced into the HeLa
cells by endocytosis, the cellular uptake of the liposomes could be enhanced the
receptor-mediated endocytosis if the liposomes would be modified with ligands such as tumor
specific ligands 14,15. In this study, however, liposomes without any ligands were knowingly
utilized to evaluate the possibility of prepared liposomes, of which the cellular uptake was
endocytosis-dependent manner.
3-2. Evaluation of DOX-Encapsulating Liposomes by Confocal Laser Microscopy
The author traced the delivery of DOX after the DOX-encapsulating liposomes were
endocytosised by HaLa cells. After the application of the DOX–encapsulating liposomes to
HeLa cells, endosomes or lysosomes were stained with LysoTracker®. In case of the
DPPC-containing liposomes, many yellow dots were confirmed (Figure 4 (a)), representing the
co-localization of DOX and endosome because DOX itself shows red fluorescence and
LysoTracker® shows green fluorescence. It is indicated that DOX remained inside the endosomes.
Whereas, in cases of L1- or L2-containing liposomes, a lot of red stains were confirmed,
suggesting that DOX molecules should be released from the endosomes. Moreover,
accumulation of DOX at nucleus was dramatically increased by the pH-sensitive L1- or
L2-containing liposomes in comparison with the pH-insensitive DPPC-containing liposomes.
Then, the author quantively analyzed the fluorescent intensity of DOX at each nucleus. The
average fluorescent intensity from the cells after introducing the DOX-encapsulating L1- or
L2-containing liposomes was about 5 times higher than that of the DPPC-containing liposomes
(Figure 4(b)). Therefore, the author confirmed that an efficient DOX release from endosomes
and DOX accumulation at nucleus by the L1- or L2-containing liposomes. Little difference of
DOX accumulation at nucleus was confirmed between L1 and L2.

Page 128
Chapter 6
- 118 -
Considering that the improved DOX release from endosomes using the synthetic
zwitterionic lipids; L1 and L2, in spite of no difference in cellular uptake of the liposomes.
Around 2 hr after the addion of the liposomes, a large number of the liposomes endocytosised by
the cells would be transferred to lysosomes 16. Then, the liposomes would be collapsed by
0.2
0
0.6
Fluo
re
s
c
en
t In
ten
s
ity
fro
m
A
DR
P
e
r area
a
t n
u
c
le
u
s
[-]
0.4
1.0
0.8
1.2
DPPC
L1
L2
Fig. 6-4. (A) Confocal microscopic images of HeLa cells in the presence of (I)
DPPC/cholesterol/PEG-Glu2C18,
(II)
L1/cholesterol/PEG-Glu2C18,
and
(III)
L2/cholesterol/PEG-Glu2C18 after incubation at 37oC for 2 hr. (B) Fluorescent intensity of DOX at
nucleus per 30 cells after the addition of liposomes composed of (open bar)
DPPC/cholesterol/PEG-Glu2C18, (ash bar) L1/cholesterol/PEG-Glu2C18, and (closed bar)
L2/cholesterol/PEG-Glu2C18.
(A)
(B)

Page 129
Chapter 6
- 119 -
digestive enzymes of the lysosomes, resulting in release of the encapsulated molecules from
liposomes inside the lysosomes. The hydrophobic molecules such as DOX would readily
penetrate through the endosomal membrane to reach hydrophobic organelles such as nucleus. In
case of the DPPC-containing liposomes, the DOX distribution would be followed by such
mechanism. Whereas, the pH-sensitive L1- or L2-containing liposomes facilitated the DOX
release within 2 hr. After the endocytosis of the L1- or L2-containing liposomes, they would
be fused to the endosomal membrane when the endosomal pH was decreased to 4–6 where the
zeta potential of the L1- or L2-containing liposomes would be cationic. Though the
L2-containing liposomes showed the fusogenic potential at a higher pH region than that of
L1-containing liposomes, the drug accumulation at nucleus were almost the same values. It is
considered that the endosomal pH transition of HeLa cells would be so quick that the author
couldn’t follow the significant difference of the intracellular dynamics between L1 and L2 in
this study. The pH-sensitive liposomes composed of the cationic/anionic lipids were reported as
drug and gene carriers by enhancing the drug release in endosomes as a result of the pH-induced
liposomal aggregation 17,18. However, release of encapsulated contents in endosomes should
reduce the efficacy because they would be degradated by lysosomal enzymes. On the other hand,
our constructed pH-sensitive liposomes are able to facilitate the escape of liposomal contents
from liposomal degradation. Therefore, the pH-sensitive liposomes containing L1 or L2 are
strongly anticipated to provide the high efficacy of drugs or gene than the conventional
pH-sensitive liposomes.
3-3. In Vitro Pharmaceutical Activity of Liposomes
In order to study the effect of the DOX-encapsuting pH-sensitive liposomes on HeLa
cells, the author added the DOX-encapsulating liposomes or free DOX to the cells, and then
investigated the cellular growth (Fig. 6-5). The cell viability was about 60 % at 24 hr after the
addition of the DPPC-containing liposomes ([DOX]=10 μg/mL). Whereas, in cases of L1- or

Page 130
Chapter 6
- 120 -
L2-containing liposomes, the cell viabilities were 14%, or 8%, respectively, and the free DOX
led completely to cellular death in this concentration. In addition, the cytotoxicity of the
synthetic lipids was investigated on HeLa cells. At 100 μg/mL of lipid concentration, the DPPC-,
L1, or L2-containing liposomes showed around 60% of cell viability, indicating low cytotoxicity
of the synthetic lipids and the DPPC have widely utilized as a liposomal component in clinical
trials because of its low toxicity.
The author confirmed that in vitro the DOX efficacy by delivering the L1 and
L2-containing liposomes in comparison of the pH-insensitive DPPC-containing liposomes. So
far, the stealth liposomes composed of phosphocholine (PC) type-lipid have extensively been
utilized for drug carriers for ovarian cancer, AIDS-related kaposi sarcoma, and multiple
myeloma 19,20. Even though poor relase from the PC-type liposomes, pharmaceutical effects
were reported 21. Based on these knowledges, our constructed pH-sensitive liposomes would
show drastic enhancement of the drug efficacies. Though the similar DOX accumulation was
observed at nucleus between the L1- and L2-containing liposomes for 2 hr incubation, the cell
growth inhibition of the L2-containing liposomes was higher than that of the L1-containing
liposomes. It is suggested that the total DOX release for 48 hr from the L2-containing liposomes
would be higher than that of the L1-containing liposomes. The highly cancer cell growth
inhibition by small amount of the DOX-encapsulating liposomes should lead to reduce the
amount of DOX injection, resulting in the low side effects. With respect to pH-sensitive
liposomes for drug delivery, the liposomes composed of DOPE are generally utilized as working
on drug release to cytoplasm by fusion to endosomal membrane 22-24. However, the utility of
DOPE-containing liposomes is restricted in the presence of FBS because DOPE doesn’t work as
a fusogenic component as a result of aggregation with serum proteins 25,26. On the other hand,
the author achieved a high drug efficacy with L1- or L2-containing liposomes even in presence
of FBS, indicating that the L1 or L2 zwitterionic lipids would act as serum compatible
components for efficient intracellular drug delivery.

Page 131
Chapter 6
- 121 -
(a)
(b)
0
20
40
60
80
100
120
0.1
1
10
DOX (μg/mL)
C
e
ll viability (%
)
Fig. 6-5. (a) Cell viability of HeLa cells after the addition of DOX-encapsulating liposomes
composed
of
DPPC/cholesterol/PEG-Glu2C18,
L1/cholesterol/PEG-Glu2C18,
and
L2/cholesterol/PEG-Glu2C18. The amount of DOX was varied from 0.1 to 10 μg/mL for evaluating
drug efficacy with the liposomes (n=3). (b) The cytotoxicity of the empty liposomes composed of
DPPC/cholesterol/PEG-Glu2C18, L1/cholesterol/PEG-Glu2C18, and L2/cholesterol/PEG-Glu2C18 on
HeLa cells. The amount of the lipids was varied from 1 to 500 μg/mL (n=3).
0
20
40
60
80
100
120
1
10
100
Lipid (μg/mL)
Cell viability (%
)

Page 132
Chapter 6
- 122 -
3-4. Blood Persistence and Biodistribution of Charge Convertible Liposomes
Finally, the author investigated the persistence of the L2-containing liposomes in blood
circulation after intravenous injection. 0.3 mol% of PEG-Glu2C18 were inserted to total
membrane in this case, the PEG chain of molecular wight was ca. 5.0 kDa. The half-life of the
L2-containing liposomes was estimated to be 290 + 27 min, indicating that the low uptake of the
liposomes into reticular endothelial cells during blood circulation (Fig. 6-6(a)). Moreover, the
author investigated in vivo biodistribution of the L2-containing liposomes after the intravenous
injection. Then, the liposomes were mainly distributed in a spleen and a liver (Fig. 6-6(b)).
Recently, a LC-MS/MS analysis has been developed for estimating a circulating time and
biodistribution after the intravenous injection of biomaterials because of more rapid extraction
and directly calculation without labeling 27,28. Then, the circulation time of the L2-containing
liposomes was considerablely long compared with that shown in the conventional cationic
liposomes. Moreover, the cationic liposomes after intravenous injection were intensively
0.1
1
10
100
0
5
10
15
20
25
Time (hr)
% o
f L2
c
o
n
c
. afte
r inje
ctio
n
(a)
0
40
100
120
L2
c
o
n
c
. (μ
g/g)
80
60
20
Kidney
Lung
Liver
Spleen
(b)
Fig. 6-6. (a) Time course of the plasma concentration of the liposomes composed of
L2/cholesterol/PEG-Glu2C18 after administration via tail vein. (b) Biodistribution of the liposomes
composed of L2/cholesterol/PEG-Glu2C18 at 24 hr after the administration.

Page 133
Chapter 6
- 123 -
localized in a lung as a result of the poor circulation capability 29. Whereas, the distribution of
the L2-containing liposomes in lung was quite low, indicating that the L2-containing liposomes
behave as long circulating liposomes in blood in comparison with the conventional cationic
liposomes. Then, the biodistribution and half-time of the L2-containing liposomes is similar to
that of the conventional long circulating liposomes containing PC, because the author have
previously reported the half-time of the DPPC-containing liposomes (PEG 0.3 mol%) as 268 + 2
min with the same condition 30. Therefore, the half-time of the L2-containing liposomes was the
similar to that of the DPPC-containing liposomes, which was extent demonstrated their
posibility. Therefore, the high performance of the pH-sensitive liposomes composed of the
charge convertible zwitterionic lipids is expected from the improvement of the blood circulation
and intracellular drug delivery.
3-5. Conclusion
Liposomes composed of the designed amino-acid based zwitterionic lipids converted
their zeta potentials when exposed to acidic pH similar to the endosomal pH. Accompanied with
the zeta potential conversion, the fusogenic potential of the pH-sensitive liposomes to the
negaitively charged membrane increased. When the pH-sensitive liposomes were uptaken into
HeLa cells by endocytosis, the liposomes effectively fused with the endosomal membrane by the
zeta potential convertion of the liposomes from negative to positive. Moreover, this fusion
facilitated the DOX release to cytoplasm, followed by accumulation at nucleus. The
pH-sensitive liposomes showed the more efficient DOX toxicity to the HeLa cells compared
with the conventional DPPC liposomes. Moreover, the L2-containing liposomes provided the
higher DOX effeicacy than L1-containing liposomes. The L2-containing liposomes provided the
long blood circulation and appropriate biodistribution of long circulating liposomes. The above
findings indicate that the pH-sensitive liposomes composed of the zwitterinic lipids are useful as
drug or gene carriers for efficient in vivo applications. Currently, the author are exploring

Page 134
Chapter 6
- 124 -
DOX-encapsulating pH-sensitive liposomes composed of the synthetic lipids for cancer therapy.
4. Experimental Section
4-1. Synthesis of Zwitterionic Lipids
To obtain L1, 1,5-dihexadecyl-L-glutamate (1.0 g, 1.7 mmol) and triethylamine (260 μl,
2.0 mmol) were dissolved in dichloromethane (30 mL) and stirred for 1 hr at room temperature.
Boc-Glu(OtBu) was added to the solution and reacted by stirring for 6 hr at room temperature.
After deprotection with trifluoroacetic acid (20 mL), L1 was obtained as white powder (790 mg,
yield 65%). To obtain L2, the 1,5-dihexadecyl N-lysyl-L-glutamte (1.0 g, 1.4 mmol) and
triethylamine (388 μl, 3.0 mmol) were dissolved in dichloromethane (30 mL) and stirred for 1 hr
at room temperature. Boc-Glu(OtBu) (1.3 g, 3.0 mmol) was added to the solution and reacted by
stirring for 6 hr at room temperature. After deprotection with trifluoroacetic acid (20 mL) at
room temperature, L2 was obtained as white powder (745 mg, yield 55%).
L1: TLC (silica gel, chloroform/methanol = 4/1, v/v): Rf 0.24. 1H NMR (CDCl3, 500 MHz) δ:,0.88 (t, 6H),
1.26 (s, 52H), 1.60 (m, 4H), 1.70-2.20 (m, 4H), 2.32-2.42 (m, 4H), 3.40 (m, 1H), 4.07-4.12 (t, 4H), 4.51
(m, 1H). ESI-MS (M+H)+ calcd. 725, found 725.5.
L2: TLC (silica gel, chloroform/methanol = 4/1, v/v): Rf 0.05. 1H NMR (CDCl3, 500 MHz) δ:0.84(t, 6H),
1.23 (br, 52H), 3.88, 3.93 (t, 1H), 4.34(q, 1H), 4.44 (q, 1H). ESI-MS (M+H)+ calcd. 982, found 982.2.
4-2. General Methods
Preparation of liposomes containing synthetic lipid
DPPC-, L1-, or L2-containing liposomes (Table 6-1) were prepared as follows; DPPC,
L1, or L2 (434 μmol), cholesterol (434 μmol), and PEG5000-Glu2C18 (2.6 μmol) were dissolved
in t-butylalcohol at 60oC and then, lyophlilized. The mixed lipid powders were hydrated in 20
mM phosphate buffer (PB, pH7.5) for 2 hr and extruded with a LIPEX™ EXTRUDER
(Northern Lipids Inc.) at 60oC and using membrane filters (pore size 0.65, 0.30, 0.22, 0.10 μm,

Page 135
Chapter 6
- 125 -
cellulose acetate filters, Millipore Inc.). The liposome dispersions were ultracentrifugated
(100,000 x g, 30 min). After removal of the supernatant, the pellet was dispersed in 20 PB
(pH7.5) for characterization of the liposomes.
Measurement of zeta potential of liposomes
Zeta ,potentials of the resulting liposomes at various pHs were measured with a Zetasizer
(Zetasizer4, Malvern, U.K.). The liposome dispersions ([lipid] = 0.1 mg/mL) in 20 mM acetate
buffer (pH4.0, 4.5, 5.0 5.5, 6.0, 6.5, and 7.0) or 20 PB (pH7.4) were loaded in the capillary cell
mounted on the apparatus and measured three times at 37oC.
Fusogenic potential of liposomes
Membrane fusion between the pH-sensitive liposomes and liposomes with
biomembrane-mimicking lipid composition were investigated to demonstrate the fusogenic
potantial of the liposomes composed of the synthetic lipids by using a fluorescence resonance
energy transfer (FRET) assay. The negatively charged liposomes composed of DPPC/DPPG (5/1,
mol) were labeled with N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)-phosphatidylethanolamine
(NBD-PE) and rhodamine-phosphatidylethanolamine (Rho-PE), both of which concentrations
were 1 mol%. For the fusion assay, 50 μl of the labeled liposomes (1 mM as lipid concentration)
and 50 μL of the pH-sensitive liposomes (9 mM as lipid concentration) were added to 1mL
acetate buffer with various pH values. The mixed dispersion was then incubated at 37oC for 1 hr,
and then, the solution pH was adapted to pH7.5 after the 100 μL of the mixed dispesion was
diluted with 2 mL of 20 mM PB (pH7.5). The degree of the fusogenic potential was estimated
from the degree of the lipid dilution at the membrane of the negatively charged liposomes,
corresponding to the recovery of NBD flureorescence. Therefore, the fluorescence intensity was
recorded using an excitation wavelength of 460 nm and an emission wavelength of 525 nm. The
fusogenic potential of the liposomes according to pH was determined by using the following

Page 136
Chapter 6
- 126 -
equation (1):
Lipid mixing % = ((Ft F0 ) / (FTX F0 )) x 100 (1)
where Ft is the fluorescence intensity at 1 hr, F0 is the fluorescence intensity just after dilution
with distilled water, and FTX is the fluorescence intensity after the addition of 0.1% Triton X-100
to solubilize the liposomes.
Cellular uptake of theDOX-liposomes
HeLa cells (1 x 104 cells) were seeded per well in a 24-well plate cell culture dish and
incubated in a 5% CO2 at 37oC. After overnight incubation, the medium in the cell culture dish
was exchanged with a fresh DMEM containing 500 ng/ml DOX in the presence of 10% FBS.
After incubation at 4oC or 37oC for 2 hr, the cells were washed twice with a cold PBS solution,
and they were dissolved in 500 μL 0.1% SDS-containing 5 mM Tris buffer (pH7.5). The amount
of DOX in the cells was fluorometrically determined for the lysate with an excitation wavelength
of 460 nm and an emission wavelength at 535 nm with a fluorecence spectrometer (Shimadzu
Corp., Japan). The protein concentration of the cells was determined using a standard protein
assay (Bio-Rad Protein Assay, Bio-rad, Hercules, CA). Thus, the degree of cellular uptake of the
liposomes introduced into the HeLa cells was quantatatively evaluated and represented as the
DOX-ng per cellular protein.
Observation of intracellular DOX trafficking
HeLa cells (5 x 104 cells) were seeded in a 35-mm cell glass bottom dish and incubated in
a 5% CO2 incubator at 37oC. The medium in the cell culture dish was changed to fresh MEM
medium (1 mL) containing DOX-encapsulating liposomes ([DOX]=1 μg/mL). After incubation
for 2 hr, the cells were twice washed with PBS, and then the medium was exchanged with 1 mL
MEM medium containing LysoTracker® Green DND-26 (f.c. 15 μM) as an endosomal dying
reagent. After incubation for 30 s with LysoTracker® Green DND-26, then the cells were

Page 137
Chapter 6
- 127 -
observed with a laser confocal microscope (LSM5 Pascal, Carl Zeiss Co., Ltd. Germany). After
observation, the author measured the fluorescence intensisity of DOX at the nucleus by using
LSM Image Browser (Carl Zeiss Co., Ltd. Germany) for 30 cells.
Cell viability after the addition of the DOX-liposomes or the DOX-nonencapauslting empty
liposomes
HeLa cells (1 x 104 cells) were seeded per well in a 24-well plate cell culture dish and
incubated in a 5% CO2 at 37oC. After overnight incubation, the medium in the cell culture dish
was exchanged with a fresh DMEM containing DOX or the DOX-encapsulating liposome
dispesion with varing the DOX concentration diluted with DMEM. Cell cytotoxicity was
measured at 24 hr using a WST assay (TetraColorONE, SEIKAGAKU Co., Ltd. Japan). Cell
viability (%) was calculated from the ratio of the cells number after the addition of DOX to the
cell number without DOX. In addition, the empty liposomes composed of
DPPC/cholesterol/PEG-Glu2C18, L1/cholesterol/PEG-Glu2C18, or L2/cholesterol/PEG-Glu2C18
were added to the HeLa cells to investigate the cytotoxicity of the synthetic lipids. The each
liposome solution with varing the lipid concentration (0.1 – 500 μg/mL) was added to the HeLa
cells and invesitigated the cytotoxicity with the methods as described above.
In vivo evaluation of the pH-sensitive liposomes after intravenous injection
Liposomes composed of L2/cholesterol/PEG-Glu2C18 were administered via tail vein (40
mg/kg) to three 25 male wistar rats (250 + 10 g, Charles River, Japan). The blood sample (ca.
300 μL) was collected just before injection, and measured to monitor the blood circulation of the
pH-sensitive liposomes. The blood sample was centrifuged (2000 x g, 20 min) to obtain blood
plasma, and stored at –80oC until the lipid concentration was measured. At 24 hr after injection,
the animals were sacrifized to collect kidney, lung, liver, and spleen. The amounts of L2 in blood
plasma and organ samples were measured by a LC-MS/MS system (HPLC; LC-20A system,

Page 138
Chapter 6
- 128 -
Shimadzu Corp., Japan, MS: API2000, Applied Biosystems/MDS SCIEX). A CAPLCELLPAK
C18 MG 3 µm 2.0 × 50 mm column was used (Shiseido Co., Ltd. Japan) for chromatographic
separation. Samples were eluted by appling the linear gradient stating from a mixture of a 15%
solution A (0.1% acetic acid in water) and a 85% solution B (0.1% acetic acid in methanol) to a
100% solution B in 5 min. The gradient was followed by a 5 min plateau at the 100% solution B
before going back to the initial solvent mixture in 5 min. The eluated solution was transferred to
a mass–spectrometry system, where the atmospheric pressure ionization in the positive
electrospraymode was used. The peak areas of MS/MS transitions; 982→84.2, were summed for
L2 concentration. For quantification the peak area of L2 was calculated from the validated
calibration curve range of 1-300 μg/mL. The correlation coefficients (r
2) of the calibration curves
were better than or equal to 0.992.

Page 139
Chapter 6
- 129 -
References
1. Cullis, P.R. J. Liposome Res. 10 (2000) 501–512.
2. Myayer, L.D., Bally, M.B., Cullis, P.R. Biochim. Biophys. Acta 857 (1986) 123–126.
3. Hope, M.J., Wong, K.F. In liposomes in biomedical applications, P121, Harwood Academic
Publishers.
4. Lasic, D.D., Ceh, B., Stuart, M.C., Guo, L., Frederik, P.M., Barenholz, Y. Biochim. Biophys.
Acta 1239 (1995) 145–156.
5. Lasic, D.D., Frederik, P.M., Stuart, M.C., Barenholz, Y., McIntosh, T.J. FEBS Lett. 312
(1995) 255–258.
6. Haran, G., Cohen, R., Bar, L.K., Barenholz, Y. Biochim. Biophys. Acta 1151 (1993) 201–215.
7. Maure-Spurej, E., Wong, K.F., Maure, E., Maure, N., Fenske, D.B., Cullis, P.R. Biochim.
Biophys. Acta 1416 (1999) 1–10.
8. Maure, N., Wong, K.F., Hope, M.J., Cullis, P.R. Biochim. Biophys. Acta 1374 (1998) 9–20.
9. Rajesh, M., Sen, J., Srujan, M., Mukherjee, K., Sreedhar, B., Chaudhuri, A. J. Am. Chem. Soc.
129 (2007) 11408–11420.
10. Obata, Y., Suzuki, D., Takeoka. S. Bioconjugate Chem. 19 (2008) 1055–1063.
11. Voûte, P.A., Souhami, R.L., Nooij, M., Somers, R., Cortés-Funes, H., van der Eijken, J.W.,
Pringle, J., Hogendoorn, P.C., Kirkpatrick, A., Uscinska, B.M., van Glabbeke, M., Machin,
D., Weeden, S. Ann. Oncol. 10 (1999) 1211–1218.
12. Bruynzeel, A.M., Niessen, H.W., Bronzwaer, J.G.., van der Hoeven, J.J., Berkhof, J., Bast, A.,
van der Vijgh, W.J., van Groeningen, C.J. Br. J. Cancer 97 (2007) 1084–1089.
13. Li, X., Hirsh, D. J., Cabral-Lilly, D., Zirkel, A., Gruner, S.M., Janoff, A.S., Perkins, W.R.
Biochim. Biophys. Acta 1415 (1998) 23–40.
14. Park, J. W., Hong, K., Kirpotin, D. B., Colbern, G.., Shalaby, R., Baselga, J., Shao, Y.,
Nielsen, U. B., Marks, J. D., Moore, D., Papahadjopoulos, D., Benz, C. C. Clin. Cancer Res.
8 (2002) 1172–1181.

Page 140
Chapter 6
- 130 -
15. Kirpotin, D., Park, J. W., Hong, K., Zalipsky, S., Li, W. L., Carter, P., Benz, C. C.,
Papahadjopoulos, D.Biochemistry 36 (1996) 66–75.
16. Simões, S., Moreira, J. N., Fonseca, C., Düzgüneş, de Lima, M.C.P. Adv. Drug Deliv. Rev.
56 (2004) 947–965.
17. Shi, G., Guo, W., Stephenson, M.R., Lee, R.J. J. Cont. Release 80 (2002) 309–319.
18. Hafez, I. M., Ansell, S. Cullis, P. R. Biophys. J. 79 (2000) 1438–1446.
19. Green, A.E., Rose, P.G. Int. J. Nanomedicine 1 (2006) 229–239.
20. Udhrain, A., Skubitz, K.M., Northfelt, D.W. Int. J. Nanomedicine 2 (2006) 345–352.
21. Recchia, F., Filippis, S.D., Saggio, G.., Amiconi, G., Cesta, A., Carta, G., Rea, S. Anticancer
Drugs. 14 (2003) 633–638.
22. Ulrich, S.H., Schubert, R., Peschka-Süss, R. J. Cont. Release 110 (2006) 490–504.
23. Shin, J., Shum, P., Thompson, D. H. J. Cont. Release 91 (2003) 187–200.
24. Reddy, J. A., Low, P. S. J. Cont. Release 64 (2000) 27–37.
25. Bailey, A.L., Cullis, P.R. Biochemistry 36 (1997) 1628–1634.
26. Turek, J., Dubertret, C., Jaslin, G., Antonakis, K., Scherman, D., Pitard, B. J. Gene Med. 2
(2000) 32–40.
27. Chin, D.L., Lum, B.L., Sikic, B.I. J Chromatogr B Analyt Technol Biomed Life Sci. 5 (2002)
259–269.
28. Liu, Y., Yang, Y., Liu, X., Jiang, T. Talanta 74 (2008) 887–895.
29. Kim, H.S., Song, I. H., Kim, J. C., Kim, E. J., Jang, D. O., Park, Y. S. J. Cont. Release 115
(2006) 234–241.
30. Okamura, Y., Maekawa, I., Teramura, Y., Maruyama, H., Handa, M., Ikeda, Y., Takeoka, S.
Bioconjugate Chem. 16 (2005) 1589–1596.

Page 141
Chapter 7
- 131 -
Chapter 7
Development of Injectable Gene Carrier: Charge Convertible
Liposomes Encapsulated Plasmid DNA
1. Introduction
In chapter 6, the author introduced function of charge convertible liposomes for
anti-cancer drug delivery with improving intracellular delivery. On the other hand, one of the
major challenges for gene therapy is systemic delivery of a nucleic acid directly into an affected
tissue. Then, the charge convertible liposomes containing zwitterionic lipids anticipate
circulating in blood because the net charge of the charge convertible liposomes at physiological
pH is negative. Therefore, the author applies charge convertible liposomes into gene carriers.
This requires developing a liposome which is able to protect the encapsulated nucleic acid from
degradation during circulation, while delivering the gene of interest to the specific tissue and
specific subcellular compartment. In this chapter, fundamental properties of charge convertible
liposomes for systemic gene delivery are discussed.

Page 142
Chapter 7
- 132 -
2. pDNA-Encapsulating Liposomes Containing Zwitterionic Lipids
2-1. Preparation of PLL-pDNA Complexes
Plasmid DNA encoding luciferase was initially condensed with poly-L-lysine (Mw: 28
kDa) as a polycation by electrostatic interaction. Then, the author investigated the complexes in
term of size and zeta potential by varying a N/P ratio (Fig. 2-1). The apparatus diameter of the
complexes was about 50 nm among the N/P ratio was less than 0.8. By contrast, the size was
increased in N/P ratio of 0.8–1.0. The complexes with a N/P ratio of 0.9 showed highest size for
about 220 nm. At then, the size of the complexes was decreased when the PLL was further added
to the solution. The complexes with a N/P ratio of 1.2–1.5 showed constant size; 60 nm.
Considering the zeta potential of the PLL/pDNA complexes based on N/P ratio, the complexes
with a N/P ratio of 0.8 showed negative charge. When further PLL was added to DNA, the zeta
potential of the complexes converted from negative to positive and obtained the complexes with
+8 mV at the N/P ratio of 0.9. Indeed, the zeta potential of the complexes with a N/P ratio of
more than 1.0 showed constant zeta potential; +40 mV.
0
-20
-40
60
40
20
Ze
ta
p
o
tentia
l (m
V
)
50
100
150
200
250
0
A
p
p
a
re
nt
d
ia
m
eter (n
m
)
0.6
0.8
1.0
1.2
1.4
N/P ratio
(A)

Page 143
Chapter 7
- 133 -
The author particularly investigated the PLL/pDNA complexes with a N/P ratio of 1.2 by TEM.
Then, the condensed PLL/pDNA complexes with 70 nm size were visualized (Fig. 7-1 (B)). The
size was almost corresponded to size measurement by DLS.
Several ternary systems, using cationic condensing molecules, have been described in the
past few years including systems based PEI 1,2, PLL 3,4, spermidine 5. Particularly, PLL was used
as condensation of pDNA for encapsulation into anionic liposomes 6–8. The author could prepare
PLL/pDNA complexes with positive net charge and nano size. Then, the PLL/pDNA complexes
with a N/P ratio of 1.2 was selected to encapsulate them into anionic liposomes. For efficient
encapsulation of PLL/pDNA complexes into anionic liposomes, the author considered to need
PLL/pDNA complexes having cationic net charge and as small size as possible. Therefore, the
complexes with a N/P ratio of 1.2 was particularly selected for encapsulation.
2-2. Preparation of pDNA-Encapsulating Liposomes
The pDNA-encapsulating liposomes were prepared using lipid film. As comparable
liposomes, DPPC-liposome was prepared. Indeed, L1 and L2-liposomes were compared to
Fig. 7-1. (A) Apparatus size and zeta potential of PLL/pDNA complexes with varying N/P ratio. (B)
TEM image of PLL/pDNA complexes with a N/P ratio of 1.2
100 nm
(B)

Page 144
Chapter 7
- 134 -
conventional pH-sensitive liposomes composed of DOPE/CHEMS. When the
pDNA-encapsulating liposomes were prepared, all liposomes showed negative zeta potential,
implying the PLL/pDNA complexes or PLL shouldn’t be adhered at outer leaflet of the
liposomes (Table 7-1). The size of the pDNA-encapsulating liposomes was around 180 nm. The
encapsulating efficiency pDNA was 20% in the case of DPPC-liposomes. By contrast, L1- or
L2-liposomes provided around 40% encapsulating efficiency, that was about 2-fold higher than
that of DPPC-liposomes. Furthermore, the pDNA encapsulation efficiency in the case of
DOPE/CHEMS liposomes was about 55%. The author furthermore investigated morphology of
pDNA-encapsulating liposomes with a transmission electron microscopy. Then, all the
liposomes were visualized to form vesicular structures, showing that there was no great
difference of liposomal structure between liposomal components. Having prepared the liposomes,
DOPE/CHEMS liposome was gradually detected aggregates. Therefore, the author evaluated
their properties within 2 week after preparation.
The author was able to prepare pDNA-encapsulating liposomes containing the synthetic
zwitterionic lipids by extrusion. If DNA could be effectively encapsulated within the liposomes,
it would be useful for practical gene therapy because the liposomal membrane is able to inhibit
the access of DNase to the encapsulated DNA during circulation. In this study, PLL/pDNA
should be efficiently encapsulated into the liposomes by electrostatic interaction between anionic
liposomal membrane and cationic PLL/pDNA complexes under the sonication. Furthermore, if
the PLL/PDNA complexes were adhered at the outer leaflet of the liposomes, the PLL/pDNA
complexes would be dissociated from liposomes at the treatment for ultracentrifugation at pH12.
That was because almost amino group of PLL should be deprotonated based on pKb, resulting in
PLL couldn’t form complexes pDNA and liposomes electrostatically. Therefore, the author
could prepare pDNA-encapsulating liposome with negative net charge. The order of pDNA
encapsulating efficiency was corresponded to order of negative zeta potential of the liposomes.
The lowest zeta potential of DOPE/CHEMS liposomes (-41mV) among applied liposomes

Page 145
Chapter 7
- 135 -
provided highest encapsulating efficiency. On the contrary, the DPPC-liposomes having weak
negative net charge (-4 mV) lead to low pDNA encapsulation. Therefore, the author could
confirm that PLL/pDNA complexes should be adhered to inner leaflet of liposomes
electrostatically. The pDNA-encapsulating liposomes having around 180 nm and negative net
charge could effectively and simply prepare.
2-3. Zeta Potential of Liposomes at Various pHs
The zeta,potentials of the L1- or L2-containing liposomes and conventional pH-sensitive
liposomes (DOPE/CHEMS) were compared (Fig. 7-2).The conversion dynamics of the zeta
potential of the L1- or L2-liposomes were described in chapter 6. By contrast, DOPE/CHEMS
liposomes showed negative zeta potential at all pHs. At pH7.4, the zeta potential of
DOPE/CHEMS liposomes showed around -40mV. That was due to proton dissociations of
DPPC / cholesterol / PEG-Glu2C18
L1 / cholesterol / PEG-Glu2C18
L2 / cholesterol / PEG-Glu2C18
Membrane components
Size
[nm]
169 + 45
187 + 42
179 + 56
TEM image
pDNA-lipid
[μg/μmol]
2.5
4.6
3.2
DOPE / CHEMS / PEG-Glu2C18
198 + 84
8.6
Zeta potential
[mV]
-4.3 + 0.3
-27.5 + 2.0
-25.0 + 1.1
-41.3 + 1.8
Encapsulation
efficiency [%]
20
42
43
55
Table 7-1. Characteristics of comparable PLL/pDNA-encapsulating liposomes

Page 146
Chapter 7
- 136 -
CHEMS molecules. When pH was decreased, the zeta potential was increased to -8 mV at pH4.0.
However the positive zeta potential was not detect in of the DOPE/CHEMS liposomes.
The great difference of zeta potential dynamics between charge convertible liposomes
containing L1 or L2, and DOPE/CHEMS liposomes was confirmed. DOPE/CHEMS liposomes
showed negative zeta potential at various pH though the value was changed with pH condition.
That would be due to the conversion of protonation degree of carboxylic group of CHEMS
molecules. At acidic condition, parts of carboxylic group of CHEMS molecules oriented on outer
leaflet of liposomes should be protonated, resulting in increase of zeta potential. However, the
DOPE/CHEMS liposomes have an ability to convert their zeta potential from minus to positive
as shown in the charge convertible liposomes containing L1 and L2.
4.0
6.0
0
-20
-40
60
40
20
pH
Zeta po
tential (m
V)
7.5
80
7.0
5.0
Fig. 7-2.
The zeta, potential of the liposomes composed of (open circles)
DPPC/cholesterol/PEG-Glu2C18, (closed circles) L1/cholesterol/PEG-Glu2C18, (squares)
L2/cholesterol/PEG-Glu2C18, and (triangles) DOPE/CHEMS/PEG-GLu2C18. Liposomes were
dispersed in 20 mM acetate buffer and measured at 37oC.

Page 147
Chapter 7
- 137 -
2-4. PLL/pDNA Release from Liposomes at Various pHs
The author measured pDNA release profile from PLL/pDNA-encapsulating liposomes in
response to acidic pH. In case of DPPC-containing liposomes and L2-liposomes, there was no
release of pDNA at all pHs. By contrast, drastic pDNA release at acidic condition was confirmed
in the DOPE/CHEMS liposomes. Indeed, L1-liposomes showed pDNA release only at pH5.5 for
providing 23% release (Fig. 7-3).
pDNA release form the PLL/pDNA-encapsulating liposomes was only confirmed in
DOPE/CHEMS liposomes and L1-containing liposomes. Generally, DOPE-containing
liposomes could be achieved release of contents at acidic condition due to hexagonal transition
0
20
40
60
80
100
6.0
7.0
pH
Le
akag
e
ra
tio / %
4.0
5.0
Fig. 7-3.
pDNA release from the
liposomes composed of (open circles)
DPPC/cholesterol/PEG-Glu2C18, (closed circles) L1/cholesterol/PEG-Glu2C18, (squares)
L2/cholesterol/PEG-Glu2C18, and (triangles) DOPE/CHEMS/PEG-GLu2C18.

Page 148
Chapter 7
- 138 -
of liposomal structure 9,10. In this case, at acidic condition, DOPE/CHEMS membrane would
convert their vesicular structure from vesicle to other structure due to protonation of CHEMS
molecules. Therefore, the author furthermore analyzed apparatus size of each liposomes at all
pHs. Then, DOPE/CHEMS liposomes detected aggregates at pH 4.0 and 4.5, where pDNA
release occurred. Therefore, pDNA release from DOPE/CHEMS would be due to transition of
membrane structure at pH4.0 and 4.5. In the case of L1-liposomes, pDNA release from the
liposomes was confirmed at around 5.5. the pDNA release would be due to change of membrane
permeability of L1-containign liposomes.
3. Evaluation of PLL/pDNA-Encapsulating Liposomes in Vitro
3-1. Cellular Uptake Efficiency
The cellular uptake efficiency of the liposomes was compared in HeLa cells (Fig. 7-4).
At pH7.4, the investigated liposomes showed comparable cellular uptake efficiency. The
DPPC-liposomes showed slightly higher cellular uptake efficiency than that of L1-,
L2-liposomes, and DOPE/CHEMS liposomes.
0
0.2
0.4
0.6
0.8
1
DPPC
DOPE/CHEMS
L2
Cell a
s
so
c
ia
ted
lipid [n
m
o
l
/ mg
p
ro
tein]
L1
Fig. 7-4. Cellular uptake of the liposomes composed of DPPC/cholesterol/PEG-Glu2C18,
L1/cholesterol/PEG-Glu2C18, L2/cholesterol/PEG-Glu2C18, and DOPE/CHEMS/PEG-Glu2C18 on
HeLa cells at 37oC for 2 hr.

Page 149
Chapter 7
- 139 -
That would be due to higher zeta potential of DPPC liposomes than that of other
liposomes. The electrostatic repulsion between liposomes and cellular membrane would be low
in the case of DPPC liposomes, resulting in higher cellular uptake efficiency was obtained.
3-2. Gene Expression Efficiency of pDNA-Encapsulating Liposomes
The gene expression efficiencies of pDNA-encapsulating liposomes was investigated
(Fig. 7-5). Then, L2-liposomes showed highest gene expression efficiencies among prepared
PLL/pDNA-encapsulating liposomes. The order of efficiencies was L2 < DOPE/CHEMS < L1
< DPPC. When PLL/pDNA complexes with a N/P ratio of 1.2, cyototoxicity was confimed
(data not shown). The DPPC-containing liposomes showed 3-fold higher gene expression
efficiency than that with addition of pDNA alone. The gene expression efficiency of
L1-liposomes was 8-fold higher than that of L2-liposomes. Indeed, the gene expression
efficiency of L2-liposomes was 2.2 fold-higher than that of DOPE/CHEMS liposomes.
2
0
6
L
u
cife
ra
s
e
ac
tivity [ x 1
0
4
RLU
/m
g
-Pro
te
in
]
4
10
8
12
Cells
PLL/pDNA
L1
pDNA
DPPC
L2 DOPE/CHEMS
Fig. 7-5. Transfection efficinecy of the liposomes composed of DPPC/cholesterol/PEG-Glu2C18,
L1/cholesterol/PEG-Glu2C18, L2/cholesterol/PEG-Glu2C18, and DOPE/CHEMS/PEG-Glu2C18 on
MDA-mb231 cells.

Page 150
Chapter 7
- 140 -
The author successfully achieved high gene expression with charge convertible
liposomes. After addition of pDNA, gene expression couldn’t lead, indicating pDNA shouldn’t
permeate through plasma membrane of cells. In the case of pDNA/pDNA complexes, some
cytotoxicity was confirmed though the author could obtain some gene expression. For the gene
therapy, the cytotoxicity should reduce as possible as. Considering higher gene expression
efficiency of L2-liposomes than L1-liposomes, quick pDNA release from endosomes might lead
to gene expression because cellular uptake efficiency was almost equivalent. As described in
fusogenic potential of the liposomes, L2-liposomes were faster able to fuse to endosomal
membrane than L1-liposomes. The difference of pDNA release profile into cytoplasm would
occur to provide discrimination of gene expression efficiency between L1 and L2. The author
could control intracellular gene delivery using synthetic zwitterionic lipids. Furthermore, the
obtaining L2-mediated gene expression efficiency was higher than that of DOPE/CHEM
liposomes. The DOPE/CHEMS liposomes are currently used for gene and drug delivery from a
lot of researchers 8,9,11. Therefore, the author could display high performance of L2-liposomes
for gene delivery..
3-3. Intracellular Gene Delivery of pDNA by Charge Convertible Liposomes
The intracellular pDNA transition by the charge convertible liposomes containing L2 was
particularly investigated using a confocal micro scope. After adding Cy3 labeled
pDNA-encapsulating liposomes, endosomes were stained with LysoTracker green. When DPPC
liposomes were added to HeLa cells, the localization of Cy3-pDNA and endosomes was
corresponded, indicating pDNA should be in endosomes. By contrast, in the case of L2
liposomes, the localization of pDNA and endosomes were different, indicating pDNA was able
to escape from endosomes. Indeed, the efficiency of endosomal escape of pDNA with
DOPE/CHEMS liposomes was poor in comparison with L2-liposomes.

Page 151
Chapter 7
- 141 -
The author was able to confirm pDNA release form endosomes after addition of
PLL/pDNA-encapsulating liposomes containing L2. As expected above experiment, the trigger
of pDNA release into cytoplasm would be due to fusion between endosomal membrane and
charge convertible liposomes containing L2. Indeed, DOPE/CHEMS liposomes showed pDNA
release at pH4.0, 4.5 displayed poor ability to release pDNA into cytoplasm in cells. That was
because pDNA would be release into endosome from DOPE/CHEMS liposomes, resulting in
pDNA couldn’t release into cytoplasm.
4. Experimental Section
Size and zeta potential of PLL/pDNA complexes
To prepare pDNA/PLL complex, stocked pDNA and PLL solusion was diluted with 20
HEPES buffer (20 mM, pH 7.5) to obtain a final DNA concentration of 80 μg/mL and varying
amounts of PLL solution each, and the mixture was vortexed immediately for a few seconds.
Then the mean particle diameter was measured by a dynamic light scattering spectrophotometer
(N4 PLUS SubmicronParticle Size Analyzer; Beckman-Coulter Inc., Fullerton,FL). The
Fig. 7-6.
Intracellular pDNA transition by the liposomes composed of (Left)
DPPC/cholesterol/PEG-Glu2C18,
(center)
L2/cholesterol/PEG/Glu2C18,
and
(right)
DOPE/CHEMS/cholesterol/PEG-Glu2C18. pDNA and endosomes were stained with Cy3 and
Lysotracker, respectively.
L2
DOPE/CHEMS
DPPC

Page 152
Chapter 7
- 142 -
zeta-potential of the complex was measured with a Zetasizer (Zetasizer4; Malvern, UK). The
suspension of complex in HEPES (40 μg/mL as pDNA concentration, 1 mL) was loaded in the
capillary cell and measured at 37 °C.
Preparation of PLL/pDNA-encapsulating liposomes
Each mixed lipid was dissolved in chloroform (1mL) and dried to form a thin lipid film
around the wall of round-bottomed flask using a rotary evaporator. Residual chloroform was
removed under vacuum for overnight, and then pDNA/PLL complexes with a N/P ratio of 1.2
was added to the lipid film and sonicated with a bath sonicator (110 W, US-1, SND Co. Ltd.,
Nagano, Japan) for 1 min. The suspension was further stirred for 4 hr and extruded with a LIPEX
EXTRUDER (Northern Lipids Inc.; Vancouver, Canada) with membrane filters (polycarbonate
filters with pore sizes of 0.20, 0.10 µm, Millipore Inc., Bedford, MA). In the case of a 0.1 μm
pore sized filter, extrusion was performed three times. The liposome was diluted with 200 mM
NaOH/KCl buffer (pH 12) in order to remove PLL/pDNA complexes on the surface of the
liposomes, and then ultracentrifugated (160,000 x g, 30 min). The supernatant was removed and
resuspended with 20 mM HEPES buffer. The concentration of pDNA was determined by
spectrofluorometry using a DNA-binding dye; Hoechst 33258 (DOJINDO LABORATORIES,
Kumamoto, Japan). pDNA-encapsulating liposome (7 μL) was solubilized with equal volume of
0.1% Trition X-100 and 12 μL of the mixture was added to 48 μL of TNE (1 mM Tris-HCl, 10
mM NaCl, 0.1 mM EDTA) buffer containing 1mg/mL polyacrylic acid, which dissociate PLL
from PLL/pDNA complexes. Then 50,μL of the mixture was added to 250 ng/mL Hoechst 33258
(50,μL) and the amount of pDNA was determined by micro plate reader (Ex : 360 nm, Em : 460
nm).
Measurement of size distribution and zeta potential of the liposome
The various liposome dispersions (10 μg/mL as lipid concentration) were diluted with

Page 153
Chapter 7
- 143 -
HEPES buffer (20 mM, pH 7.4, 1mL). Then the mean particle diameter was measured by a
dynamic light scattering spectrophotometer as described above. The zeta potential of the
liposomes dispersed in HEPES buffer (20 mM, pH 7.4, 1mL) was measured with a Zetasizer.
The liposome dispersed in water (100 µg/mL as lipid concentration, 1 mL) was loaded in the
capillary cell and measured three times at 37 °C.
Measurement of pDNA release from liposomes at various pHs
For pDNA/PLL complex release assay, 50 μL of prepared pDNA-encapsulating liposome
(1.5 μM as lipid concentration) was added to 950 μL of 20 mM acetate buffer, and incubated at
37oC for 1 hr. Then, the resulting solution was ultracentrifugated (160,000 x g, 30 min) and the
supernatant was removed. Next, the precipitation was resuspended with 60 μL of TNE buffer.
Then, the suspended solutions were solubilized with 60 μL of 0.1%Triton X-100. 100 μL of the
mixture was diluted with 100 μL of TNE buffer, and Hoechst 33258 was added to it, and then the
amount of pDNA was determined by micro plate reader as described above. Next, pDNA/PLL
complex release (%) was determined by following formula:
Release (%) = 100 – (IpH I0)/(I7. 4 I0) x 100
Where IpH and I7.4 represents fluorescence of solubilized liposome after incubated at each pH and
pH 7.4 respectively for 1h. IT represents fluorescence after the addition of 10%Triton-X 100.
Gene expression efficiency of pDNA-encapsulating liposomes
For the transfection assay, HeLa cells (5 x 103 cells) were seeded per 96 wells plate and
incubated for 24 hr (37oC, 5% CO2). pDNA-Encapsulating liposome was added to the cells and
further incubated for 48 hr (37oC, 5% CO2). The cells were washed with PBS twice, and then
lysed with a lysis buffer from the luciferase assay kit (Promega, Madison, WI). The luciferase
activity in a 20 µL aliquot of the cell lysate was measured with a microplate reader. The protein
concentration of each well lysate was determined as described above. The luciferase activity in

Page 154
Chapter 7
- 144 -
each sample was normalized to obtain the relative light unit (RLU) per miligram of protein.
Intracellular pDNA trafficking
To observe the intracellular distribution of pDNA encapsulated into liposome, pDNA was
labeled with Cy3 by using Label IT Cy3 Labeling Kit (Mirus biocorp. Madison, WI).
Cy3-pDNA/PLL-encapsulating liposomes were then prepared as described above. HeLa cells (1
x 105 cells) were seeded on 35-mm glass bottom dish, and incubated for 24 hr (37oC, 5% CO2).
Cy3-pDNA/PLL-encapsulating liposomes (0.2 μg) were then added to the cells and incubated
for 6 hr (37oC, 5% CO2). The cells were washed with PBS twice, and then exchanged with a
DMEM containing FBS (10%) and LysoTracker Green DND (150 nM) used for staining
intracellular acidic compartments such as endosome or lysosome. The localization of Cy3-pDNA
and endosome was observed by confocal microscopy FV1000 (OLYMPUS, Tokyo Japan).

Page 155
Chapter 7
- 145 -
References
1. Murray, K.D., Etheridge, C.J., Shah, S.I., Matthews, D.A., Russell, W., Gurling, H.M., Miller,
A.D. Gene Ther. 8 (2001) 453–60.
2. Tagawa, T., Manvell, M., Brown, N., Keller, M., Perouzel, E., Murray, K.D., Harbottle, R.P.,
Tecle, M., Booy, F., Brahimi-Horn, M.C., Coutelle, C., Lemoine, N.R., Alton, E.W.F.W.,
Miller, A.D. Gene Ther. 9 (2002) 564–576.
3. Lee, M., Rentz, J., Han, S.O., Bull, D.A., Kim, S.W. Gene Ther. 10 (2003) 585–593.
4. Lee, C.H., Ni, Y.H., Chen, C.C., Chou, C., Chang, F.H. Biochim Biophys Acta 1611(2003)
55–62.
5. Gao, X., Huang, L. Biochemistry 35 (1996) 1027–1036.
6. Liu, F., Huang, L. J. Cont. Release 78 (2002) 259–266.
7. Lee R.J., Huang, L. J. Biol. Chem. 271 (1996) 8481–8487.
8. Kogure, K., Moriguchi, R., Sasaki, K., Ueno, M., Futaki, S., Harashima. H. J. Cont. Release
98 (2004) 317–323
9. Ellens, H., Benz, J., Szoka, F.C. Biochemistry 23 (1984) 1532–1538.
10. Wheeler, J. J., Palmer, L., Ossanlou, M., MacLachlan, I., Graham, R. W., Zhang, Y. P., Hope,
M. J., Scherrer, P., Cullis, P. R. Gene Ther. 6 (1999) 271–281.
11. Collins, D.S., Findlay, K., Harding, C.V. Immunol. 148 (1992) 3336–3341.

Page 156
Chapter 7
- 146 -

Page 157
Chapter 8
- 147 -
Chapter 8
Future Prospect
1. Introduction
The author has been constructed endosomal pH-sensitive liposomes for improving
intracellular drug and gene delivery by using zwitterionic amino-acid based lipids. Indeed,
responsible pH of charge convertible liposomes was found to be controllable by varying head
group structure. On the other hand, pH decrease also occurs in solid tumor, infection,
inflammation. Therefore, as future prospects, the author suggests that charge convertible
liposomes in response to extracellular pH of solid tumor expect to enhance cellular uptake
efficiency of the liposomes into the cancer cells. In this chapter, the author describes
characteristics of pH profile around solid tumor and discusses about required gene carriers in
response to extracellular pH.

Page 158
Chapter 8
- 148 -
2. Development of Charge Convertible Liposomes in Response to Extracellular pH of Solid
Tumor
2-1. Tumor pH Targeting Nanotechnology
Tumor-targeting approaches have been developed for improved efficacy and reduced
toxicity by altering biodistribution of cancer drugs and by using specific cell surface interactions.
Solid tumors are often characterized by overexpression of specific antigens or receptors on cell
surfaces
1–6
. Antigens and receptors help in transmitting signals from the surrounding
environment that are essential for the growth of tumor cells. Targeting antigens or receptors has
been extensively investigated as an important delivery mode by using macromolecular or
nano-sized carriers to tumor cells. Nanocarriers attached with surface ligands or antibodies
exploit these receptor-mediated uptake pathways that are recognized and internalized by the
tumor cells
7–11
. However, these approaches have achieved limited success in clinic, most likely
because of significant heterogeneity in both solid tumor cell types and cell surface markers
1–3
.
Additionally, both the presence of antigens and the expression of receptors on surface of these
tumor cells are transient and dynamic
1–6
. The heterogeneity of cancer cells may explain the
reasons for the unexpected results of targeting strategy
12
.
Fig. 8-1. pHe map.of the H2-resonance of IEPA in a coronal slice through an
MDAmb-435 tumor.

Page 159
Chapter 8
- 149 -
The extracellular pH (pHe) of normal tissues and blood pH are kept constant at pH 7.4 and their
intracellular pH (pHi) at 7.2. However, in most tumors the pH gradient is reversed (pHi > pHe).
Particularly, tumor pHe is lower than normal tissues
13–16
. Although there is a distribution in in
vivo, pHe measurements made by using needle type microelectrodes on human patients having
various solid tumors (adenocarcinoma, squamous cell carcinoma, soft tissue sarcoma, and
malignant melanoma) and in readily accessible areas (limbs, neck, or chest wall), shows the
mean pH value to be 7.0 with a range between 5.7 and 7.8
15
. This variation is dependent upon
tumor histology, tumor volume, and location inside a tumor. Recent measurements of pHe by
noninvasive technology such as
19
F,
31
P, or
1
H probes by magnetic resonance spectroscopy in
human tumor xenografts and in animals further proved consistently low pHe
16,17
. Reported pHe
data on human and animal solid tumors either by invasive or noninvasive methods showed that
more than 80% of all measured values are below pH 7.2
16,17
. The primary reason for this
imbalance in cancer pH is the high rate of glycolysis in cancer cells, both in aerobic and
anaerobic conditions
18–20
. It is also proposed that the acidic milieu benefits the cancer cells by
generating an invasive environment that tears down the extracellular matrix and destroys the
surrounding normal tissue cells
21
.
2-2. Construction of Gene or Drug Carrier in Response to Tumor pH
As described above, the tumor pH is locally decreased compared with normal tissues.
The author considers that the cellular uptake into tumor cells will be able to facilitate using
charge convertible liposomes. Having changed their net charge from negative to positive only
around tumor, the liposomes are expected both long circulating in blood and enhancement of
cellular uptake (Fig. 8-2). The design is corroborated PEGylated liposomes as anionic liposomes
and cationic liposomes.

Page 160
Chapter 8
- 150 -
For this aim, the author has to construct charge convertible liposomes in response to
pH6.4–7.2. Concerning other drug carriers in response to tumor pH, Lee et al. reported
intelligent micelles using PEG shielding method
22–24
.
Fig. 8-2. Design of charge convertible liposomes containing zwitterionic lipids. The liposomes
circulate in blood stable, and facilitate cellular uptake by varying their net charge from negative
to positive.
Fig. 8-3. Schematic diagram depicting the central concept of pH-induced biotin repositioning on
the micelle.
Normal
Lymphatic (active)
Circulation
Accumulation around tumor
Cellular uptake
Tumor
Blood
Lymphatic (poor)
Target cells
pH6.2-7.0
pH7.4

Page 161
Chapter 8
- 151 -
While above pH 7.0, biotin that is anchored on the micelle core via a pH-sensitive molecular
chain actuator (polyHis) is shielded by PEG shell of the micelle; biotin is exposed on the micelle
surface (6.5<pH<7.0) and can interact with cells, which facilitates biotin receptor-mediated
endocytosis. When the pH is further lowered (pH<6.5), the micelle destabilizes, resulting in
enhanced drug release and disrupting cell membranes such as endosomal membrane
22
.

Page 162
Chapter 8
- 152 -
References
1. Scholler, N., Fu, N., Yang, Y., Ye, Z., Goodman, G.E., Hellstrom, K.E., Hellstom, I. Proc.
Natl. Acad. Sci. U. S. A. 96 (1999) 11531–11536.
2. Chaidarun, S.S., Eggo, M.C., Sheppard, M.C., Stewart, P.M. Endocrinology 135 (1994)
2012–2021.
3. Parker, N., Turk, M.J., Westrick, E., Lewis, J.D., Low, P.S., Leamon, C.P., Anal. Biochem.
338 (2005) 284–293.
4. Muss, H.B., Thor, A.D., Berry, D.A., Kute, T., Liu, E.T., Koerner, F., Cirrincione, C.T.,
Budman, D.R., Wood, W.C., Barcos, M., Henderson, I.C. Engl. J. Med. 330 (1994)
1260–1266.
5. Daniels, R.A., Turley, H., Kimberley, F.C., Liu, X.S., Mongkolsapaya, J., Ch'En, P., Xu, X.N.
Jin, B.Q., Pezzella, F., Screaton, G.R. Cell Res. 15 (2005) 430–438.
6. Muller, C., Schubiger, P.A., Schibli, R. Eur. J. Nucl. Med. Mol. Imaging 33 (2006)
1162–1170.
7. Cirstoiu-Hapca, A., Bossy-Nobs, L., Buchegger, F., Gurny, R., Delie, F. Int. J. Pharm. 331
(2007) 190–196.
8. Rapoport, N., Gao, Z., Kennedy, A. J. Natl. Cancer Inst. 99 (2007) 1095–1106.
9. Ponce, A.M., Vujaskovic, Z., Yuan, F., Needham, D., Dewhirst, M.W. Int. J. Hypertherm. 22
(2006) 205–213.
10. Ko, J., Park, K., Kim, Y.S., Kim, M.S., Han, J.K., Kim, K., Park, R.W., Kim, I.S., Song,
H.K., Lee, D.S., Kwon, I.C. J. Control. Release 123 (2007) 109–115.
11. Bae, Y., Fukushima, S., Harada, A., Kataoka, K. Angew. Chem. Int. Ed. Engl. 42 (2003)
4640–4643.
12. Chaudhry, A., Carrasquillo, J.A., Avis, I.L., Shuke, N., Reynolds, J.C., Bartholomew, R.,
Larson, S.M., Cuttitta, F., Johnson, B.E., Mulshine, J.L. Clin. Cancer Res. 5 (1999)
3385–3393.

Page 163
Chapter 8
- 153 -
13. Engin, K., Leeper, D.B., Cater, J.R., Thistlethwaite, A.J., Tupchong, L., McFarlane, J.D. Int.
J. Hypertherm. 11 (1995) 211–216.
14. Volk, T., Jahde, E., Fortmeyer, H.P., Glusenkamp, K.H., Rajewsky, M.F. Br. J. Cancer 68
(1993) 492–500.
15. van Sluis, R., Bhujwalla, Z.M., Raghunand, N., Ballesteros, P., Alvarez, J., Cerdan, S.,
Galons, J.P., Gillies, R.J. Magn. Reson. Med. 41 (1999) 743–750.
16. Ojugo, A.S., McSheehy, P.M., McIntyre, D.J., McCoy, C., Stubbs, M., Leach, M.O., Judson,
I.R., Griffiths, J.R. NMR Biomed. 12 (1999) 495–504.
17. Leeper, D.B., Engin, K., Thistlethwaite, A.J., Hitchon, H.D., Dover, J.D., Li, D.J., Tupchong,
L. Int. J. Radiat. Oncol. Biol. Phys. 28 (1994) 935–943.
18. Tannockand, I.F., Rotin, D. Cancer Res. 49 (1989) 4373–4384.
19. Hobbs, S.K., Monsky, W.L., Yuan, F., Roberts, W.G., Griffith, L., Torchilin, V.P., Jain, R.K.
Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 4607–4612.
20. Stubbs, M., Mcsheehy, R.M.J., Griffiths, J.R., Bashford, L. Opinion 6 (2000) 15–19.
21. Yamagata, M., Hasuda, K., Stamato, T., Tannock, I.F.Br. J. Cancer 77 (1998) 1726–1731.
22. Lee, E.S., Gao, Z, Bae, Y.H. J. Cont. Release 132 (2008) 164–170.
23. Lee, E.S., Na, K., Bae, Y.H. Nano Lett. 5 (2005) 325–329.
24. Sethuraman, V.A., Bae, Y.H. J. Control. Release 118 (2007) 216–224.

Page 164
Chapter 8
- 154 -

Page 165
List of Achievements
(Research Paper)
1. Y. Obata, D. Suzuki, S. Takeoka, ‘‘Evaluation of cationic assemblies constructed with
amino-acid based lipids for plasmid DNA delivery’’, Bioconjugate Chem. 19 (2008)
1055–1063.
2. G. Ciofani, V. Raffa, Y. Obata, A. Menciassi, P. Dario, S. Takeoka, ‘‘Magnetic driven alginate
nanoparticles for targeted drug delivery’’, Current Nanoscience 4 (2008) 212–218.
3. H. Sakai, Y. Seisi, Y. Obata, S. Takeoka, H, Horinouchi, E. Tsuchida, K. Kobayashi, ‘‘Fluid
resuscitation with artificial oxygen carriers in hemorrhaged rats: Profiles of
hemoglobin-vesicle degradation and hematopsieses for 14 days’’, Shock, 31 (2009) 192–200.
4. G. Ciofani, V. Raffa, J. Yu, Y. Chen, Y. Obata, S. Takeoka, A. Menciassi, A. Cuschieri, ‘‘Boron
nitride nanotubes: A novel vector for targeted magnetic drug delivery’’, Current Nanoscience
5 (2009) 33–38.
5. G. Ciofani, V. Raffa, Y. Obata, I. Sato, A. Menciassi, P. Dario, N. Takeda, S. Takeoka,
‘‘Realization, characterization and functionalization of lipidic wrapped carbon nanotubes’’, J.
Nanopart. Res. 11 (2009) 477–484.
6. Y. Obata, S. Saito, N. Takeda, S. Takeoka, ‘‘Plasmid DNA-encapsulating liposomes: Effect of
a spacer between the cationic moiety and the hydrophobic moieties of the lipids on gene
expression efficiency’’, Biochim. Biophys. ActaBiomembr. in press (2009).

Page 166
(Patents)
1. PCT Patent: WO2006/098415 ‘‘Drug Carrier’’ S. Takeoka, Y. Okamura, H. Kanazawa, S.
Hisamoto, K. Kubota, Y. Obata
2. PCT Patent: WO2006/118327 ‘‘Cationic Amino Acid Type Lipid’’ S. Takeoka, Y. Obata
3. 国内出願 2007–210953 「pH 応答性集合体」 出願人: 武岡 真司, 小幡 洋輔
4. PCT Patent: WO2008/062911 ‘‘Reagent for Introduction of Protein or Gene’’ S. Takeoka, N.
Takeda, H. Kurumizaka, I. Sakane, N. Ikegaya, Y. Obata, S. Saito.
5. PCT Patent: WO2008/143339 ‘‘Amphipathic Molecule, Molecular Aggregate Comprising the
Amphipathic Molecule, and Use of the Molecular Aggregate’’ S. Takeoka, Y. Obata, S.
Tajima
(International Symposium)
Y. Obata, S. Takeoka. ‘‘Evaluation of release behavior of macromolecules from pH-responsive
liposomes’’, 3rd IUPAC-sponsored International Symposium on Macro- and Supramolecular
Architectures and Materials, Tokyo (Japan), May 2006, Tokyo
D. Suzuki, Y. Obata, S. Takeoka. ‘‘Synthesis of aminolipids having disulfide bond and controlled
release of the lipid’’, 3rd IUPAC-sponsored International Symposium on Macro- and
Supramolecular Architectures and Materials, Tokyo (Japan), May 2006.
Y. Obata, S. Takeoka. ‘‘Evaluation of release behavior of macromolecules from pH-sensitive
liposomes’’, YoungChem2006, Pułtusk (Poland), Oct. 2006.
Y. Obata, S. Takeoka. ‘‘Structural study of amino-acid based cationic lipids for development of
plasmid DNA carriers’’, 235th American Chemical Society National Meeting, New Orleans
(U.S.A), April 2008.
Y. Obata, S. Takeoka, ‘‘Release behavior from the pH-sensitive liposomes composed of the
zwitterionic lipids for anti-cancer drugs delivery’’, 11th Liposome Research days Conference,
Yokohama (Japan), July 2008.
S. Saito, Y. Obata, S. Takeoka ‘‘Syntheses of amino-acid based cationic lipids bearing a spacer
for efficient pDNA delivery’’, 11th Liposome Research days Conference, Yokohama (Japan),
July 2008.

Page 167
(国内学会発表)
1. 小幡 洋輔, 武岡 真司.「両イオン性アミノ酸型脂質から成る pH 応答性リポソームの
構築」第 57 回高分子討論会 (2008 年 9 月, 大阪)
2. 小幡 洋輔, 武岡 真司「アミノ酸型脂質から構成される pH 応答性リポソームの機能
評価」第 24 回日本 DDS 学会 (2008 年 6 月, 東京)
3. 小幡 洋輔, 田島 祥二, 武岡 真司. 「両イオン性アミノ酸型脂質から成るリポソーム
の特性と薬物運搬体としての評価」第 57 回高分子年次大会 (2008 年 5 月, 横浜)
4. 小幡 洋輔, 田島 祥二, 武岡 真司. 「アミノ酸型脂質から成る pH 応答性 遺伝子運搬
体の構築と機能評価」第 29 回日本バイオマテリアル学会 (2007 年 11 月, 大阪)
5. 小幡 洋輔, 田島 祥二, 武岡 真司. 「アミノ酸型脂質から構成される pH 応答性リポ
ソームの特性」第 23 回日本 DDS 学会 (2007 年 6 月, 熊本)
6. 小幡 洋輔, 田島 祥二, 武岡 真司. 「カチオン性アミノ酸型脂質の遺伝子運搬能評
価」遺伝子デリバリー研究会 第 7 回シンポジウム (2007 年 5 月, 東京)
7. 小幡 洋輔, 田島 祥二, 武岡 真司. 「アミノ酸型脂質から成る遺伝子運搬体の構築」
日本化学会第 87 春季年会 (2007 年 3 月, 大阪)
8. 小幡 洋輔, 田島 祥二, 武岡 真司. 「アミノ酸型脂質から成るカチオン性集合体の遺
伝子運搬能評価」第 28 回日本バイオマテリアル学会 (2006 年 11 月, 東京)
9. 小幡 洋輔, 武岡 真司. 「アミノ酸型脂質から構成される pH 応答性リポソームの特
性」第 22 回日本 DDS 学会 (2006 年 6 月, 東京)
10. 小幡 洋輔, 武岡 真司. 「アミノ酸型脂質を膜成分とした pH 応答性リポソームの調
製と高分子量内包物放出特性」第 55 回高分子学会年次大会 (2006 年 5 月, 名古屋)
11. 小幡 洋輔, 鈴木 大祐, 武岡 真司. 「アミノ酸型脂質から構成される pH 応答性リポ
ソームの特性」日本化学会第 86 春季年会 (2006 年 3 月, 千葉)
12. 小幡 洋輔, 武岡 真司. 「アミノ酸型脂質による pH 崩壊性リポソームの構築」第 54
回高分子討論会 (2005 年 9 月, 山形)
13. 小幡 洋輔, 武岡 真司. 「カチオン性アミノ酸を極性頭部とした両親媒性分子の合成
と分子集合特性の評価」第 54 回高分子学会年次大会 (2005 年 5 月, 横浜)
14. 小幡 洋輔, 武岡 真司. 「アミノ酸型脂質の合成と小胞体物性」日本膜学会第 27 年
会 (2005 年 5 月, 東京)
15. 小幡 洋輔, 武岡 真司, 西出 宏之, 酒井 宏水, 土田 英俊. 「ヘモグロビン小胞体(人
工赤血球)投与後のサイトカイン放出挙動」第 53 回高分子学会年次大会 (2004 年 5 月,
神戸)

Page 168
Acknowledgement
The presented thesis is the collection of the studies which have been carried out under the
guidance of Professor Dr. Shinji Takeoka, Department of Applied Chemistry in Waseda
University, during 2003–2009. The author expresses the greatest acknowledgement to Professor
Dr. Takeoka for his valuable suggestions, discussions and continuous encouragement throughout
this study. The author also expresses his sincere gratitude to Professor Dr. Hiroyuki Nishide for
his valuable advice as well as kindest supports on the work. The author thanks Professor Dr.
Kiyotaka Sakai, as an expert of biomedical engineering, for his effort as a member of the judging
committee for the doctoral theses. The author thanks Associate Professor Dr. Arianna Menciassi,
as an expert of biomedical robotics, Polo Sant' Anna Valdera, Scuola Superiore Sant'Anna for
her effort as a member of the judgment committee for the doctoral theses.
The author acknowledges to Dr. Hiromi Sakai for his valuable suggestions and discussions for
constructing systemic drug delivery. The author also acknowledges to Dr. Keitaro Sou for his
valuable advises and discussions for liposomal drug delivery.
The author expresses the greatest acknowledgement to Dr. Yosuke Okamura as a specialist in
bioengineering and biotechnology, particularly in liposomal research. The discussions about
liposomal research with Dr. Okamura were considerably valuable and their suggestions
admittedly preceded this research into practical drug delivery.
The author acknowledges to Dr. Yuji Teramura, Dr. Ichiro Takemura, Dr. Shinsuke Ishihara for
fruitful discussion about polymer chemistry and chemistry-based molecular medicine. The
author also expresses the acknowledgement to Mr. Daisuke Niwa, Mr. Kohei, Kubota, and Mr.
Ippei Maekawa, Miss Izumi Sato for continuous encouragement throughout this study.
The author is particularly very much to active colleagues; Mr. Daisuke Suzuki, Miss. Namiko
Ikegaya, Mr. Shunsuke Saito, Mr. Shoji Tajima, and Mr. Satoru Nakagawa for continuous
support throughout this study. The author furthermore thanks to colleagues of Takeoka
laboratory.
Finally, the author expresses his gratitude heartily to his parents Mr. Kazuo Obata, Mrs. Setsuko
Obata for continuous encouragement and financial support. The author also thanks to Miss.
Yukiko Obata, and Mrs. Ayako Nishioka for continuous encouragement.
Feb. 2009