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

Prof. Dr. Shinji Takeoka

Prof. Dr. Hiroyuki Nishide

Prof. Dr. Kiyotaka Sakai

Assoc. Prof. Dr. Arianna Menciassi

Promoter :

Referees :

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

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

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

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

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

vi

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

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.

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.

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

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.

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.

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

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

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.

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[M(μM

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)

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

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

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.

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

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.

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.

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

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

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.

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.

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

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.

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.

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.

Chapter 1

- 24 -

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.

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.

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..

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.

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.

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

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.

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

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).

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.

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).

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).

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

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,

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.

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.

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.

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.

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.

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.

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.

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).

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.

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.

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 3a–3c 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;

1a–1c, and (B) arginine type lipids; 3a–3c.

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

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).

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;

1a–1c, and (B) arginine type lipids; 3a–3c in presence of fetal bovine serum.

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.

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

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.

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

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

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

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

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

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.

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.

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) 124–132.

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.

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.

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.

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.

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

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.

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

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.

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.

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

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.

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.

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,

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.

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.

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

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.

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.

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.

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.

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.

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.

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

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,

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

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.

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

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).

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.

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.

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)

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.

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.

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.

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,

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.

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

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

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

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

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

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

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.

Chapter 5

- 106 -

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.

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.

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.

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.

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.

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.

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.

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

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.

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.

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.

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)

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

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.

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 (%

)

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.

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

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,

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

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

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,

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.

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.

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.

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.

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)

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)

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

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

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.

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.

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.

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.

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.

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

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

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

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).

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.

Chapter 7

- 146 -

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.

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.

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.

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

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

.

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.

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.

Chapter 8

- 154 -

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. Acta–Biomembr. in press (2009).

(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.

(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)

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