Content uploaded by Oluwadamilola Kolawole
Author content
All content in this area was uploaded by Oluwadamilola Kolawole on Oct 09, 2015
Content may be subject to copyright.
Self assembly of star shaped amphiphiles –
Opportunities for drug delivery
Aluko Oluwadamilola Miriam
A thesis submitted in partial fulfillment of the requirements for the degree of
Master of Science in Drug Delivery, University of London
September, 2010
Department of Pharmaceutics
The School of Pharmacy
University of London
29-39 Brunswick square
London WC1N 1AX
United Kingdom
ACKNOWLEDEMENTS
My sincere gratitude goes to everyone who made it possible for me to complete this thesis.
Firstly, I am very grateful to my supervisor Professor Ijeoma Uchegbu for her guidance and
constant motivation throughout my year of intensive research. Thank you for your support
and encouragement. Indeed, you have been my source of inspiration and determination for
excellence. I adore your wealth of wisdom and experience in the scientific research. You have
inculcated me with the spirit of perseverance and accomplishment, in addition to developing
my analytical skills and exposure.
My heartfelt thanks to the pleasant people that helped with various analyses, such as Mire
Zloh, Derek Marley, Dave McCarthy, Kersti Karu and other wonderful people that I cannot
afford to express all their names. Also, to my dear senior colleagues - Kar wai, Adeline,
Katerina, Mariarosa, Jayanat (tutor), Natrah, Funmi, you are highly cherished! You have
made life in the laboratory very excited and intriguing for me. Pleasant memories about you
linger on, in my heart.
To my beloved family, thank you so much for believing in me, you have invested your life
savings to give me this opportunity. Mum and Dad, you are the most sacrificing parents I
have encountered in the universe! Kayode and Nike, I will forever be indebted to you and
your children, thank God for blessing me with a marvelous brother and sister-in-law. Olamide
and Bukola thank you for your love and understanding throughout this period of sacrifice, you
are cherished! To Tayo and Bunmi (nephew), I love you dearly. To my fiancé, thank you for
your love, moral support and words of encouragement, you are irreplaceable!
Above all, I return the glory to Almighty God, in whom I live and have my being. You are the
vine and I am the branch. My inspiration comes from you, without whom I cannot achieve
anything meaningful in life. You are my refuge, fortress and source of excellence!
i
ABSTRACT
In order to study the influence of branching on the self-assembly of PEG amphiphiles,
polyethylene glycol (PEG)-based star-shaped amphiphiles have been synthesised and studied
as self-assembling systems. Palmitoyl (C16) groups were grafted to 8-armed PEG with differing
degrees of palmitoylation (P8PEG1 & P8PEG4). A novel amphiphilic linear analogue
(HDPEG) was also synthesised with hexadecyl (C16) pendant groups for comparison. These
amphiphiles were characterised using 1H Nuclear Magnetic Resonance, Fourier Transform-
Infrared and MALDI-TOF mass spectroscopy. The Pyrene probe was employed to evaluate self
assembly properties while Photon Correlation Spectroscopy was used to measure particle size
distribution. Molecular architecture and hydrophobic substitution had a profound effect on their
self-assembly behavior; as P8PEG4 with branched architecture and the greatest degree of
hydrophobic substitution had the lowest polydispersity index. Also, the critical micellar
concentration (CMC) for P8PEG4, P8PEG1 and HDPEG were 3, 8 and 15 μM respectively,
inferring greater micelle stability with branched architecture and increased hydrophobic
substitution. Particle size and morphology were confirmed by Transmission electron
microscopy as P8PEG4 and HDPEG formed mixtures of micelles and nanoparticles while a
novel core-shell nano- and micro self assembly was observed for P8PEG1. Preliminary drug
encapsulation studies on the amphiphiles loaded low amounts of Griseofulvin (0.04-0.09mg/ml
with 5mg/ml of polymers). These resulting stable aggregates of PEG-based amphiphiles may
be of benefit for drug delivery applications. However, future studies should focus on influence
of polymer architecture on drug encapsulation in order to improve their encapsulation
efficiency.
iI
CONTENTS PAGE
Acknowledgements ……………………………………………………………………... i
Abstract …………………………………………………………………………………. ii
Contents ……………………………………………………………………………….. iii-iv
List of Figures …………………………………………………………………………. v-vii
List of tables ……………………………………………………………………………. viii
Plagiarism statement …………………………………………………………………… ix
1 Introduction ……………………………………………………………………… 1-2
1.1 Nanosystems as drug delivery tools ……………………………………………… 2-3
1.1.1 Liposomes …………………………………………………………………………3-4
1.1.2 Polymeric nanoparticles …………………………………………………………… 4
1.1.3 Polymeric micelles ………………………………………………………………... 4-5
1.2 Self assembly properties of amphiphiles ……………………………...................... 6
1.2.1 Low molecular weight-amphiphiles ………………………………………………. 7-8
1.2.2 Super-amphiphiles ………………………………………………………………. 8-10
1.3 Molecular architecture of linear copolymers ………………………………….. 10-12
1.4 Molecular architecture of branched copolymers ……………………………….. 13-15
1.5 Aims and objectives of the work …………………………………………………. 16
2 Synthesis and Characterization of star-shaped and linear amphiphiles …….. 17
2.1 Materials ………………………………………………………………………….. 17
2.2 Methods …………………………………………………………………………… 17
2.2.1 Synthesis of star-shaped palmitoyl-8PEG (P8PEG1 & P8PEG4) …………….. 17-19
2.2.2 Synthesis of linear hexadecyl PEG (HDPEG) ……………………………………. 20
2.3 Structural Characterization ……………………………………………………….. 21
2.3.1 NMR-COSY spectroscopy ……………………………………………………….. 21
2.3.2 Determination of molecular weight using ESI- & MALDI- TOF MS ………….. 21-22
2.3.3 FT-IR spectroscopy ……………………………………………………………… 22
iii
2.4 Self assembly studies of PEG-based amphiphiles ……………………………… 22
2.4.1 Pyrene probe studies for CMC determination ………………………………… 22-23
2.4.2 PCS measurements ……………………………………………………………… 23
2.4.3 TEM with negative staining …………………………………………………….. 23
3 Drug encapsulation studies ……………………………………………………. 24
3.1 Materials ………………………………………………………………………… 24
3.2 Methods …………………………………………………………………………. 25
3.2.1 Preparation of drug-loaded polymer aggregates of P8PEG4 and HDPEG ………25
3.2.2 HPLC analysis of drug loaded onto P8PEG4 and HDPEG aggregates ………… 25
4 Results ………………………………………………………………………….. 26
4.1 Synthetic yield and physical properties ………………………………………… 26-27
4.2 Structural Characterization ……………………………………………………… 28
4.2.1 NMR and COSY data …………………………………………………………… 28-44
4.2.2 MALDI-TOF MS and ESI-MS data …………………………………………….. 45-50
4.2.3 FT-IR ……………………………………………………………………………. 51-54
4.3 Self assembly studies ……………………………………………………………. 54
4.3.1 Pyrene probe studies …………………………………………………………….. 54-57
4.3.2 PCS data …………………………………………………………………………. 57-61
4.3.3 TEM data ……………………………………………………………………….. 61-62
4.4 Griseofulvin encapsulation studies ………………………………………………. 63
4.4.1 Extent of drug loading / encapsulation efficiency for the amphiphiles ………...... 63
5 Discussion ……………………………………………………………………… 64-72
6 Conclusion ……………………………………………………………………..... 73
7 Future work …………………………………………………………………….. 73
References ………………………………………………………………………………. 74-82
iv
LIST OF FIGURES
Figure 1.1: Structure of the liposomes consisting of the phosphoilid bilayer ………………. 3
Figure 1.2: Schematic representation of a self-assembled block co-polymeric micelle …….. 5
Figure 1.3: The general scheme of micelle formation / demicellization from amphiphilic
molecules …………………………………………………………………………………….. 6
Figure 1.4: Schematic representation of the different aggregation morphologies found in low
molecular amphiphiles ……………………………………………………………………….. 7
Figure 1.5: Representation of the homopolymer and varieties of copolymers ………………. 8
Figure 1.6: Micelle-forming di-block and tri-block copolymers …………………………… 11
Figure 1.7: Mechanism of micelle formation from different types of amphiphilic copolymers
……………………………………………………………………………………………….. 12
Figure 1.8: Common varieties of chain architectures for highly branched molecules ……… 13
Figure 2.1: Reaction scheme for the synthesis of star-shaped palmitoyl-8PEG ……………. 19
Figure 2.2: Reaction scheme for the synthesis of linear hexadecylPEG ……………………. 20
Figure 4.1a: 1H NMR spectrum of 8PEG in CDCl3 ………………………………………… 29
Figure 4.1b: 1H-1H COSY spectrum of 8PEG in CDCl3 …………………………………… 30
Figure 4.2a: 1H NMR spectrum of PNS in CDCl3 …………………………………………. 31
Figure 4.2b: 1H-1H COSY spectrum of PNS in CDCl3 …………………………………….. 32
Figure 4.3a: 1H NMR spectrum of P8PEG1 in CDCl3 ……………………………………... 33
Figure 4.3b: 1H-1H COSY spectrum of P8PEG1 in CDCl3 ………………………………… 34
Figure 4.4a: 1H NMR spectrum of P8PEG4 in CDCl3 …………………………………… 35
Figure 4.4b: 1H-1H COSY spectrum of P8PEG4 in CDCl3 ……………………………… 36
v
Figure 4.5a: 1H NMR spectrum of SSAMPEG in CDCl3 ………………………………….38
Figure 4.5b: 1H-1H COSY spectrum of SSAMPEG in CDCl3 ……………………………. 39
Figure 4.6a: 1H NMR spectrum of HDA in CDCl3 ……………………………………….. 40
Figure 4.6b: 1H-1H COSY spectrum of HDA in CDCl3 …………………………………... 41
Figure 4.7a: 1H NMR spectrum of HDPEG in CDCl3 …………………………………….. 42
Figure 4.7b: 1H-1H COSY spectrum of HDPEG in CDCl3 ………………………………... 43
Figure 4.8a: MALDI-TOF Gaussian distribution curve for 8PEG ………………………… 45
Figure 4.8b: ESI-MS Gaussian distribution curve for PNS ………………………………... 46
Figure 4.8c: MALDI-TOF Gaussian distribution curve for P8PEG1 ………………………. 48
Figure 4.8d: MALDI-TOF Gaussian distribution curve for P8PEG4 ………………………. 48
Figure 4.8e: MALDI-TOF Gaussian distribution curve for SSAMPEG …………………… 49
Figure 4.8f: ESI-MS Gaussian distribution curve for HDA ………………………………... 50
Figure 4.8g: MALDI-TOF Gaussian distribution curve for HDPEG ………………………. 50
Figure 4.9a: FT-IR spectrum for 8PEG …………………………………………………...... 51
Figure 4.9b: FT-IR spectrum for PNS ……………………………………………………… 51
Figure 4.9c: FT-IR spectrum for P8PEG1 ………………………………………………..... 52
Figure 4.9d: FT-IR spectrum for P8PEG4 ………………………………………………..... 52
Figure 4.9e: FT-IR spectrum for SSAMPEG ………………………………………………. 53
Figure 4.9f: FT-IR spectrum for HDA ………………………………………………………53
Figure 4.9g: FT-IR spectrum for HDPEG ………………………………………………….. 54
vi
Figure 4.10: Effect of increasing concentrations of P8PEG4 on the emission spectra of Pyrene
……………………………………………………………………………………………… 55
Figure 4.11a: Determination of CMC of P8PEG1 from the I1/I3 ratio of pyrene …………. 56
Figure 4.11b: Determination of CMC of P8PEG4 from the I1/I3 ratio of pyrene …………. 56
Figure 4.11c: Determination of CMC of HDPEG from the I1/I3 ratio of pyrene …………. 57
Figure 4.12a (i): PCS size distribution curve of unfiltered P8PEG1 dispersion ………….. 58
Figure 4.12a (ii): PCS size distribution curve of filtered P8PEG1 dispersion ……………. 59
Figure 4.12b (i): PCS size distribution curve of unfiltered P8PEG4 dispersion …………. 59
Figure 4.12b (ii): PCS size distribution curve of filtered P8PEG4 dispersion ……………. 60
Figure 4.12c (i): PCS size distribution curve of unfiltered HDPEG ……………………… 60
Figure 4.12c (ii): PCS size distribution curve of filtered HDPEG ………………………... 61
Figure 4.13a: TEM image of filtered (0.45µm) P8PEG1 dispersion ……………………… 61
Figure 4.13b: TEM image of filtered (0.45µm) P8PEG4 dispersion ……………………… 62
Figure 4.13c (i): TEM image of filtered (0.45µm) HDPEG dispersion …………………… 62
Figure 4.13c (ii): TEM image of unfiltered HDPEG dispersion, with particle size correlation
with PCS data ………………………………………………………………………………. 62
Figure 5.1: Schematic representation of the molecular arrangement in the various self-
assemblies of PEG-based amphiphiles ................................................................................... 69
vii
LIST OF TABLES
Table 2.1: Molar feed ratio for the synthesis of palmitoyl-8PEG …………………………..18
Table 4.1a: Synthetic yield and degree of palmitoylation (using 1H NMR data) of PEG-based
amphiphiles …………………………………………………………………………………. 26
Table 4.1b: Physical properties of PEG-based amphiphiles ………………………………... 27
Table 4.2: CMC of PEG-based amphiphiles determined by Pyrene method ……………….. 45
Table 4.3a: PCS measurement of unfiltered polymer dispersion (2mg ml-1) ………………. 57
Table 4.3b: PCS measurement of filtered polymer dispersion (2mg ml-1) using 0.45 µm filter
………………………………………………………………………………………………. 58
Table 4.4: Drug Loading of PEG-based amphiphiles ………………………………………. 63
viii
PLAGIARISM STATEMENT
This thesis describes research conducted in the School of Pharmacy, University of London
under the supervision of Professor Ijeoma Uchegbu. I certify that the research described is
original and that I have written all the text herein and have clearly indicated by suitable
citation any part of this dissertation that has already appeared in publication.
…………………… ………………..
Signature Date
ix
1. Introduction.
Over the past two decades, efforts have been made towards developing efficient pharmaceutical Drug
Delivery Systems (DDSs) to limit drug degradation and inactivation upon administration, avoid undesirable
side effects, and deliver sufficient drug concentrations timely at pathological sites (Gref et al., 1994).
Moreover, drug carriers meant for parenteral administration should possess small particle size; high drug
loading potential; be easy, and reasonably cheap to prepare; be biodegradable and biocompatible; show
prolonged circulation, and selective retention within pathological sites (in the body) in sufficient quantities
(Gref et al., 1994).
However, conventional chemotherapy in the form of simple formulations has failed to achieve optimal
therapeutic index and selective intracellular targeting (Feng, 2004). This is associated with the poor
physicochemical properties of the cytotoxic agents (such as poor aqueous solubility) as well as lack of
specificity of such formulations (Feng, 2004), where the former is not available at target sites in
therapeutic concentrations and the latter inflicts toxic effects on healthy tissues in addition to cancerous
ones. Thus improved drug delivery systems [DDSs] were developed to enhance efficacy (bioavailability) of
the incorporated drug as well as limit possible toxic effects (target-specific).
The significantly exploited varieties of DDSs for the encapsulation of many therapeutic and diagnostic
agents are the liposomes (Torchilin, 2005), polymeric nanoparticles (Feng et al., 2004), gold nanoparticles
(Bergen et al., 2006), magnetic nanoparticles (Shubayev et al., 2009), and micelles (Gaucher et al., 2005).
However, other forms of drug carriers such as carbon nanotubes, solid lipid nanoparticles, microcapsules,
nanoassemblies, and dendrimers are also being investigated for improved bioavailability and limited
toxicity (Breunig et al., 2008). These carriers may be slowly degrading, targeting (conjugation with specific
antibodies to elicit desired site specific activity), or stimuli-responsive (pH- or temperature-sensitive); and
are chosen for drug delivery based on desired features such as biocompatibility, biodegradability, efficacy
and targeting potential (Kreuter, 1994).
Recently, there is a growing interest towards drug carriers (particularly of polymeric nature) that self-
assemble into micelles due to incomparable advantages conferred by micelle formation relative to other
drug carriers. Micelles enhance the apparent solubility and bioavailability of hydrophobic drugs; they are
also targeted passively (preferential accumulation of drugs within leaky tumor vasculature based on
“enhanced permeability and retention” effect) or actively based on ligand conjugation (Torchilin, 2001).
Molecular self assembly into micellar structures is displayed by molecules and supramolecular entities
where they aggregate (linked by noncovalent bonds) under equilibrium conditions into desired structures
by virtue of their shape and this strategy is exploited in nanofabrication to design and develop
thermodynamically stable, complex and functional structures with promising encapsulation properties
(Whitesides et al., 1991).
The major drug delivery tools, the micelle-forming systems (including the linear and branched polymers)
as well as their pharmaceutical applications are detailed below in order to appreciate the evolution of the
amphiphilic polymer-based DDSs as promising nanocarriers.
1.1. Nanosystems as Drug Delivery tools
The nanosystems are desirable as DDSs because of their particle size in the nano-range, which enhance
therapeutic efficacy. For example, malignant tumors associated with other organs excluding brain may be
targeted passively, based on the enhanced permeability and retention effect (EPR), which takes advantage
of leaky tumor cells, where nanosystems within the range of 10 to 100 nm are retained within tumors,
thereby delivering cytotoxic agents in therapeutic concentrations (Jain et al., 2001). However, rapid
clearance by the reticuloendothelial system (RES) and macrophages are likely problems of passive delivery
of bioactive molecules to such organs except the carriers possess hydrophilic (stealth) surface, which
improves their retention within tumors (Jain et al., 2001). Thus, they should be appropriately formulated
to prevent toxicity associated with non-specific delivery. On the other hand, therapeutic failure associated
with RES clearance may be avoided by surface modification of nanocarriers. The extensively exploited
types are discussed briefly.
1.1.1. Liposomes
Liposomes (Figure 1.1) are phospholipid vesicles within the size ranges of 50 to 1000 nm, suitable for the
encapsulation of hydrophilic drugs (into its inner aqueous segment), though hydrophobic drugs may also
loaded onto its outer phospholipid bilayer (Lasic, 1993). Over the past two decades, they have been
proven as promising carriers, especially due to their efficacy, safety and targeting potential, where
targeting moieties, including antibodies are conjugated to liposomal surfaces (Torchilin, 2005). However,
they are limited by short plasma circulation time, which limits contact time with pathological sites / target
antigens (Lasic, 1993).
Figure 1.1: Structure of the liposome consisting of the phospholipid bilayer (Torchilin, 2001).
Thus long-circulating PEGylated liposomes (polyethylene glycol coating) were developed and explored
over the last decade, due to its prolonged plasmatic circulation and enhanced retention in different
diseased sites (e.g., solid tumors or metastatic areas) via EPR effect (Jain, 2001). The EPR effect is based
on the fact that pathological vasculature are leaky in nature and thus readily penetrated by small and
large molecules targeted passively especially to tumor sites based on the cutoff size (200-600 nm) of the
compromised tumor vasculature (Maeda et al., 2001).
1.1.2. Polymeric nanoparticles
Water soluble drug
embedded in the
polar region of the
phospholipid bilayer
Nanoparticles comprise of spherically structured particles (10-800 nm) used as colloidal carriers for the
delivery of both hydrophilic and hydrophobic therapeutic substances (Allemann et al., 1993). They have
potential ability to encapsulate/entrap drugs, prevent their degradation (stabilize); avoid systemic toxicity
/ side effects and ensure targeted / prolonged drug delivery (Allemann et al., 1993). Drug delivery using
nanoparticles (particularly with polymeric components) has been progressively explored over the last two
decades because they have been able to address some of the limitations of conventional formulations
such as short shelf life and immediate release profile (Kreuter, 2001). Moreover, they are easily prepared,
highly stable and non-immunogenic (Kreuter, 2001).
Novel large molecule therapeutics (proteins, genes, and antisense drugs) for central nervous system
disorders have also been delivered in a controlled fashion during in vivo studies (Huang et al., 2007). Their
particle size and surface properties can also be altered to achieve passive and active targeting respectively
(Kreuter, 2001). For example, intravenous injection of transferrin conjugated and PEGylated albumin
nanoparticles loaded with zidovudine to mice resulted in improved brain concentration (by about 20%)
and reduced distribution to reticuloendothelial system (RES), relative to other unmodified
nanoparticulate systems (Mishra et al., 2006). This has shown the therapeutic implications of the
physicochemical properties of DDSs.
1.1.3. Polymeric micelles (PMs)
The therapeutic application of hydrophobic agents, including anticancer drugs have been limited by poor
aqueous solubility, absorption and bioavailability (Lipinski et al., 2001) as well as embolic reaction and
local toxicity associated with drug aggregation during intravenous injection of such drugs (Lasic & Martins,
1995). Conversely, their hydrophobicity is critical for intracellular delivery, thus PMs were developed for
the formulation of poorly water soluble drugs such as cisplatin (Yokoyama et al., 1996).
Figure 1.2: Schematic representation of a self-assembled block co-polymeric micelle (Mishra et al.,
2010).
PMs (Figure 1.2) are colloidal dispersion (5-100 nm) consisting of a core and shell structure generated
from amphiphilic block copolymer (possessing two segments); where the inner hydrophobic core
encapsulates the drug and outer hydrophilic shell / corona shields the drug from the aqueous surrounding
as well as recognition by the RES (Lasic & Martins, 1995). Conversely, the core can consist of water soluble
polymer, and made hydrophobic by the chemical conjugation of hydrophobic drugs to such hydrophilic
core (Yokoyama et al., 1996). Alternatively, two oppositely charged polyions are complexed, forming
polyion complex (PIC) micelles with a hydrophobic core. For example, PIC spherical micelles with narrow
particle size distribution and potential for antisense DNA delivery were generated from cationic block
copolymers (poly [ethylene glycol]-poly [є-benzyloxycarbonyl-L-lysine]) conjugated with anionic
oligonucleotide (Kataoka et al., 1996).
This group of nanocarriers (PMs) cannot be fully appreciated without the knowledge of their self assembly
properties which generates the micelles. Nevertheless, the science of micellization originates from low
molecular weight surfactants, which also consist of the hydrophilic and hydrophobic segments. The two
categories of micelle-forming systems are detailed below.
1.2. Self-assembly properties of amphiphiles
Amphiphiles are molecules that consist of both the hydrophilic and hydrophobic segments. On the other
hand, micelles (Figure 1.3) are spherical colloidal dispersions within the size range of 5 – 100 nm
possessing two clearly distinct regions with different affinities towards a given solvent; and are
instantaneously formed by surfactants or amphiphilic materials under equilibrium conditions (Torchilin,
2001). The self assembly of these amphiphilic substances is concentration and temperature-dependent,
that is, they occur separately at low concentrations in the liquid medium and at high concentrations, they
begin to aggregate at a critical concentration known as the critical micellar concentration (CMC) while the
number of individual molecules forming a micelle is termed “aggregation number” (Rosen, 1999). The
most reliable technique for detecting the presence of the micellar aggregates is the fluorescence
spectroscopy using fluorescent probes such as pyrene, where an increase in fluorescence intensity is
associated with solubilisation within the non-polar micellar core at increased concentrations, is exploited
(Kalyanasundaram & Thomas, 1977; Dominguez et al., 1997) to investigate self-assembly properties. The
self-assembly features of surfactants and amphiphilic polymers are detailed below.
Figure 1.3: The general scheme of micelle formation / demicellization from amphiphilic molecules
(Torchilin, 2001).
1.2.1. Low molecular weight-amphiphiles
Surfactants with low molecular weight constitute this class of amphiphiles. They have been extensively
explored in the pharmaceutical industry as wetting and stabilizing agents (Torchilin, 2001); in addition to
micellization
demicellization
improving the aqueous solubility of hydrophobic drugs (Rangel-Yagui et al., 2005). The polar and non-
polar moieties of these amphiphiles facilitate their build up as aggregates in solutions or arrangement at
interfaces (Rangel-Yagui et al., 2005).
Figure 1.4: Schematic representation of the different aggregation morphologies found in low molecular
amphiphiles: A, spherical micelles; B, rod-like micelles; C, disk-shaped micelles; D, inverted micelles; E,
normal cylindrical hexagonal packing; F, lamella; G, inverted cylindrical hexagonal packing; H,
double bilayer formation in a spherical vesicle (Benito, 2006).
Low molecular weight amphiphilic molecules generate a variety of aggregates (micelles, vesicles and liquid
crystal phases) in aqueous solution (Figure 1.4) under the influence of hydrophobic interactions,
dependent on parameters such as concentration, structure, and temperature of the medium (Lasic, 1993).
This phenomenon is termed “self-assembly” and has enhanced the solubilisation and bioavailability of
therapeutic agents with poor aqueous solubility (Uchegbu et al., 2001).
Although low molecular weight surfactants are known to form small and narrowly distributed micelles,
their dilution in biological fluid (blood or gastrointestinal fluid) as well as premature release and
precipitation of incorporated hydrophobic drugs have limited their application ( Francis et al., 2004).
1.2.2. Super-amphiphiles
Polymers have been shown to form efficient DDSs for drugs (Feng et al., 2004), genes (Park et al., 2006)
and diagnostic agents (Shubayev et al., 2009). Spacer molecules such as amino acids or peptides are
usually inserted chemically between the polymer and the active agents to control the rate and site of the
release of the bioactive molecules; and such drug-polymer conjugates, which may contain acid-labile
linkage, are taken up into the cell by endocytosis (Gillies et al., 2004). However, carrier systems of
homopolymer origin have limited drug-loading potential as the drug payload enveloped by the polymer
depends on its number of constituent reactive moieties (Breunig et al., 2008). Moreover, they do not form
micelles because they consist of the same type of monomers polymerized. Thus they either have similar
affinity or repulsion towards the aqueous medium (depending on their hydrophilicity or hydrophobicity),
preventing the formation of the core and shell structure; which also limits their encapsulation efficiency.
Figure 1.5: Representation of the homopolymer and varieties of copolymers (Benito, 2006).
Hence, amphiphilic polymers were explored especially due to their self assembly feature, which generates
a core/shell structure. The hydrophobic domains of the amphiphiles associate in aqueous environment,
where they avoid contact with water molecules while the hydrophilic corona interacts favourably with
water molecules. This results in the formation of self-assemblies in the aqueous medium due to
hydrophobic interactions while the hydrophilic segment sustains the aqueous solubility of the polymer
(Rosler et al., 2001). Amphiphilic polymers are sub-divided as block copolymers and graft polymers. Block
copolymers (Figure 1.5) are formed by the copolymerisation of hydrophilic and hydrophobic monomers,
and are the most common molecular architecture of amphiphilic polymers (Krasia & Patrickios, 2002).
They may be made up of two (di-block) or three (tri-block) building blocks and examples are provided in
Figure 1.6. Graft copolymers are branched forms of block-copolymers where hydrophobic groups are
attached onto the hydrophilic polymer backbones (Burke et al., 2001).
They are referred to as super-amphiphiles because they are macromolecular analogues of low molecular
weight surfactants. Moreover, aqueous medium is not a prerequisite for their self-assembly, in contrast
to low molecular weight amphiphiles (Discher et al., 2000). A number of studies have been presented
where super-amphiphiles aggregated in other solvent systems apart from water (Din & Liu, 1998;
Breulmann et al., 2000). This is because they have the ability to form reverse micelles in non-aqueous
media, where the hydrophilic domain constitutes the core with hydrophobic shell. However, drug delivery
scientists are less interested in this type of micelles because the biological systems is aqueous in nature,
thus the behaviour of our delivery system in aqueous environment is of great importance. All the
parameters that influence the assembly behaviour of low molecular weight amphiphiles also play a role
in the self-aggregation process of amphiphilic block-copolymers; and analogous superstructures were also
observed in solution.
Their potential to form these supramolecular structures in aqueous solutions has generated progressive
interest particularly in the biochemical and biomedical fields in recent years (Gaucher et al., 2005;
Nakayama et al., 2006). The self-assembly behavior of these amphiphilic block co-polymers is dependent
on the relative length of the blocks (Hajduk et al., 1998; Burke et al., 2001, Kabanov & Alakhov, 2002;
Discher & Eisenberg, 2002) as well as their molecular architecture (Cheng et al., 2006; Chooi et al., 2010).
Both block copolymer and grafted polymers with amphiphilic nature aggregate in water to form polymeric
micelles (Wang et al., 2004), nanoparticles (Wang et al., 2001; Cheng et al., 2006) or bilayer vesicles (Wang
et al., 2001; Discher & Eisenberg, 2002). Apart from their small and narrowly distributed particle size,
polymeric nanoparticles and micelles generated from the self assembly of amphiphilic polymers have
been exploited due to their safety and stability in biological fluid (Torchilin, 2001; Francis et al., 2004),
relative to low molecular weight surfactants. Grafted polymer amphiphile such as quaternary ammonium
glycol chitosan (GCPQ) also possess greater solubilisation potential for hydrophobic drugs than low
molecular weight surfactants (Uchegbu et al., 2001). These nanostructures have shown promising clinical
applications, for example, block copolymer micelles, composed of poly (ethylene oxide) (hydrophilic
block) and poly (L-amino acid) (hydrophobic block) exhibit self assembly properties and accumulate
passively at solid tumors in vivo for prolonged periods, delivering doxorubicin (Kwon & Kataoka, 1995).
1.3. Molecular architecture of linear copolymers
Polymers are made of repeating units of low molecular weight moieties called monomers. Copolymers
consist of two different (hydrophilic and hydrophobic blocks) repeating units differing in their solubility
while similar repeating units constitute homopolymers (Odian, 2004).
Figure 1.6: Micelle-forming di-block and tri-block co-polymers (Torchilin, 2001).
Copolymers have subdivision based on the sequence in which the different repeating units appear
within the polymer chain. Alternating copolymers are such that their repeating units alternate within the
sequence while non-specific sequencing copolymers are called random or statistical polymer. Long
segments of a monomer, when represented in a block pattern, are called block-copolymer (Odian,
2004).
Figure 1.7: Mechanism of micelle formation from different types of amphiphilic co-polymers (Torchilin,
2001).
Polymers such as poly (ethylene oxide) (PEO) / poly (ethylene glycol) (PEG), poly (acrylic acid), and poly
vinyl pyridine constitute the hydrophilic segment of block copolymers while poly (L-amino acids), poly (L-
lactic acid), poly (ε-caprolactone) (PCL) and poly (propylene oxide) (PPO) make up their hydrophobic cores
(Kwon & Kataoka, 1995; Kim et al., 1998). Some examples of hydrophilic backbones employed in
biomedical research for synthesis of graft copolymers include glycol chitosan (Qu et al., 2006), poly
(propylenimine) (Chooi et al., 2010), and hydroxypropyl cellulose (Francis et al., 2003). Palmitoyl, cetyl,
and cholesteryl groups have been employed as hydrophobic segments (Qu et al., 2006; Wang et al., 2004;
Liu et al., 2004). The mechanism of micellization / self assembly by micelle-forming di-block, tri-block co-
polymers, and graft copolymers are presented in Figure 1.7.
1.4. Molecular architecture of branched copolymers
Although linear polymers are employed conventionally in the area of drug and gene delivery, recent
polymer-related researches have been directed towards polymers having varieties of functional groups,
including those with highly branched architecture. This resulted in the development of complex polymeric
architectures such as gradient polymers, polymer brushes, graft polymers, dendrimers and star polymers
(Peleshanko & Tsukruk 2008).They have also been classified as comb-like, ramified, star-shaped and
dendritic polymers by Benito, 2006. They all possess a central reactive part (backbones, single groups, or
nanoparticles) attached to terminal groups which are reactive, polar, or hydrophobic in nature
(Peleshanko & Tsukruk 2008). Their morphology and possible chain configuration depends on the
chemical composition of backbones, cores, branches, and terminal groups (Peleshanko & Tsukruk 2008).
Figure 1.8: Common varieties of chain architectures for highly branched molecules: (a) graft, (b) brush,
(c) dendrimer, (d) hyperbranched, (e) arborescent, and (f) star architectures (Peleshanko & Tsukruk,
2008).
These highly branched polymers (Figure 1.8) have architectural hindrances as well as a high concentration
of terminal functional groups, which confers different physicochemical properties compared to their
linear analogs (Roovers, 1985). This confers lower solution and melt viscosities than linear polymer (with
same molar mass), which is advantageous in the manufacturing and pharmaceutical research fields
(Peleshanko & Tsukruk, 2008).
Dendrimers are special class of branched polymers, in which the ramifications occur in each monomer,
resulting in branched branches (Newkome et al., 2001). They are the most popular class of the highly
branched molecules, which has been extensively studied for two decades (Peleshanko & Tsukruk 2008).
A lot of exciting results generated with respect to their chemical architectures, synthetic strategies,
encapsulating features, aggregation behavior, assembly in solutions and surfaces as well as potential
application as advanced drug delivery devices, nanocomposite materials, and catalytic systems have been
documented (Newkome et al., 2001; Tully & Frechet, 2001). Its unclear mechanism of interaction with a
complex biological environment as well as its overall high cost of synthesis has limited its application
(Peleshanko & Tsukruk 2008).
Thus other branched polymers (lower degree of branching and less regular architecture) such as star
shaped copolymers of 3-armed and 4-armed PEG grafted to PCL (Lu et al., 2006; Cheng et al., 2007; Sheng
& Ya, 2008) have been employed for their promising features such as self assembly into nanocarriers,
which enhances the encapsulation and delivery of hydrophobic agents. Moreover, prolonged duration of
stepwise reaction and purification critical for conventional dendrimers is avoided by their generation via
one-pot synthesis (Peleshanko & Tsukruk 2008).
Mikto (asymmetric/heteroarm) star polymers are star-shaped polymers with varying types of polymer
arms (different chemical entity and/or molecular weight) emanating from the core. They are quite
cumbersome to synthesize, as it involves several protection/deprotection strategies and combination of
polymerization reactions (Khanna et al., 2010). However, their self-assembly properties into micellar
aggregates occurring at lower concentration than that of linear polymers (greater micelle stability)
generated interest into their drug delivery and biological applications in the future (Khanna et al., 2010).
Poly (propylenimine) dendrimer amphiphiles derivatized with PEG and grafted with palmitoyl groups
have also been investigated for their self assembly potential and they formed nanoassemblies (Chooi et
al., 2010). The degree of palmitoylation as well as their highly branched nature influenced their self
assembly as well as encapsulation efficiency for cyclosporine. Those modified with three palmitoyl groups
were sterically hindered from self assembly into high capacity hydrophobic domains. However, further
increase in degree of hydrophobicity with more palmitoyl groups improved encapsulation (Chooi et al.,
2010). This showed the influence of palmitoylation on the modulation of self assembly features of highly
branched polymers.
Although, PEG-based 3- and 4 armed star shaped copolymers (with same type of multiple arms)
copolymerised with poly (ε-caprolactone) have been well characterized in terms of thermal and self
assembly features (Lu et al., 2006; Cheng et al., 2007; Sheng &Ya, 2008 ) and their potential clinical
applications such as sustained release features investigated by Dong et al., 2010, the self assembly
properties of star shaped PEG based copolymers (with greater number of arms grafted with hydrophobic
moiety such as palmitoyl group) has not been reported. Moreover, PEGylated dendrimer amphiphiles
grafted with palmitoyl groups have been reported to be non-hemolytic (Chooi et al., 2010), confirming
the biocompatibility of PEG-based polymeric systems. Thus exploration of 8-armed star-shaped PEG-
based graft copolymer is desired.
1.5. Aims and objectives of the work
The aim of this project was to investigate the self assembly features of the novel star shaped amphiphilic
graft copolymer, in comparison to its linear analogue. My working hypothesis is that palmitoylation of
these polymers will facilitate their self assembly into nanocarriers (with hydrophobic domains) having
potential for hydrophobic drug encapsulation. Moreover, their molecular architecture would also
influence their self assembly features.
The objectives of this project were:
To synthesize two star-shaped copolymers (P8PEG1 & P8PEG4) with different level of hydrophobic
substitution as well as a novel linear analogue with same molecular weight.
To chemically characterise the polymers in terms of their hydrophobic substitution using NMR,
FTIR and MALDI-TOF.
To evaluate the self assembly of the novel amphiphilic star shaped (P8PEG 1 and P8PEG4) and
linear copolymers (HDPEG).
To characterise the resulting polymer aggregates with respect to their size and morphology.
2. Synthesis and Characterization of star-shaped and linear amphiphiles
2.1. Materials
Material
Supplier
8-arm Polyethylene glycol (8PEG; MW 2000Da)
Jenkem Technology, U.S.A.
O-[N-succinimidyl) succinyl-aminoethyl]- O’-methylpolyethylene
glycol
Sigma-Aldrich Co., U.K.
(SSAMPEG; MW 2000Da)
Palmitic acid N-hydroxysuccinimide ester (PNS)
Sigma-Aldrich Co., U.K.
Hexadecyl amine (HDA)
Sigma-Aldrich Co., U.K.
Sodium bicarbonate (NaHCO3)
Sigma-Aldrich Co., U.K.
Triethyl amine (TEA)
Sigma-Aldrich Co., U.K.
Diethyl ether
V.W.R. International, U.K
Chloroform (CDCl3)
Fischer Scientific, U.K.
Deuterated Solvents
Sigma-Aldrich Co., U.K.
Pyrene
Fluka, U.K
2.2. Methods
2.2.1. Synthesis of star-shaped palmitoyl-8PEG (P8PEG1 & P8PEG4)
Two types of palmitoyl-8PEG [8-ARMPEG-[(NH2)8-x (Palmitoyl x] (P8PEG1 & P8PEG4) was prepared by
reacting PNS and 8ARM-PEG-(NH2)8 according to the molar feed ratio shown in Table 2.1.
Table 2.1: Molar feed ratio for the synthesis of Palmitoyl-8PEG (P8PEG)
Briefly, 8ARM-PEG-(NH2)8 and sodium bicarbonate (Table 2.1) were dissolved in a mixture of ethanol and
water. PNS solubilised in ethanol was added gradually into PEG solution, while stirring continuously over
1 h. The polymer aqueous mixture was stirred for 18 h, in a light-tight environment at room temperature
(25oC) (Chooi et al., 2010). Subsequently, ethanol was evaporated under reduced pressure at 45oC and
the resultant residue washed with diethyl ether (3 X 100 mL) to remove unreacted PNS.
The product was thoroughly dialysed (molecular weight cut off = 1 KDa) against water (5 L) with 6 changes
over 24 h. The resultant products were then lyophilized. The synthetic yield for P8PEG1 (off-white) and
P8PEG4 (cotton-like) were 66.5mg (12.01 %w/w) and 60.5mg (10.98 %w/w) respectively.
Polymer
Molar feed
ratio
(PNS:8PEG-
(NH2)8)
8-PEG-
(NH2)8
(mg)
PNS
(mg)
Sodium
bicarbonate
(mg)
Volume of
ethanol
solubilising PNS
(mL)
Volume of water to
ethanol solubilising
sodium bicarbonate
(mL)
P8PEG1
10:1
400
707.0
336.0
200
100ml: 100ml
P8PEG4
30:1
400
2121
1680.0
400
100ml: 50ml
Figure 2.1: Reaction scheme for the synthesis of star shaped Palmitoyl-8PEG.
2.2.2. Synthesis of linear hexadecyl PEG (HDPEG)
NaHCO3/
C2H5OH/H2O
8-arm PEG
Palmitic acid N-
hydroxysuccinimide ester
P8PEG1 or P8PEG4
The linear amphiphilic polymer (Figure 2.2) was prepared by dissolving HDA (120.7mg), SSAMPEG (200mg)
and triethyl amine (0.75ml) in chloroform (50ml) and stirred for 18hrs; chloroform evaporated and
resultant product purified and freeze-dried as described in section 2.2.1. Molar feed ratio of SSAMPEG to
HDA was 1:5. The synthetic yield for HDPEG (cotton-like) was 113.8mg (20 % w/w).
HDA
HDPEG
Figure 2.2: Reaction scheme for the synthesis of linear hexadecyl-PEG
TEA / CHCl3
HDA: SSAMPEG – 5:1
SSAMPEG
2.3. Structural Characterization
2.3.1. Nuclear Magnetic Resonance (NMR) - COSY Spectroscopy
1H NMR and 1H-1H COSY (Bruker AMX 400MHz spectrometer, Bruker Instruments, U.K.) experiments were
carried out on all polymer solutions. P8PEG1, P8PEG4 and HDPEG (3mg) were dissolved in Deuterated
chloroform (CDCl3) (0.6ml). The degree of Palmitoylation for P8PEG1 and P8PEG4 were calculated by
comparing the ratio of palmitoyl methyl protons (δ = 0.90 ppm) to ethylene glycol protons (δ = 3.0-3.7
ppm). The binding of hexadecyl groups to polyethylene glycol moiety of SSAMPEG in HDPEG (having one
possible point of substitution) is revealed by the 1H NMR spectrum [hexadecyl methyl protons (δ = 0.90
ppm) and ethylene glycol protons (δ = 3.0-3.7 ppm)] and confirmed from COSY spectrum.
2.3.2. Determination of molecular weight using ESI- and MALDI-TOF MS
Electrospray ionization (positive) mass spectrometry (ESI-MS) is appropriate for the determination of
molecular weight of compounds less than 1000 Da while the latter is appropriate for larger materials
(greater than 1000 Da). This procedure was carried out by using Finnigan Mat TSQ7000 (triple quadrupole)
mass spectrometer to confirm the mass of low molecular weight-materials used, that is PNS and HDA.
Samples (1mg) were dissolved in Acetonitrile or methanol (1:1 ratio, 1ml) and infused into a TSQ 7000
mass spectrometer operated in the EI mode at the rate of 1mLh-1, with a needle voltage of 4.46kV and
capillary temperature of 250oC. The instrument generates expressed mass distribution expressed in
Daltons.
Matrix-assisted laser desorption/ionization-time of flight mass spectroscopy (MALDI-TOF) is a procedure
used for the fast and accurate determination of molecular weight of polymers and employed to confirm
formation of P8PEG1, P8PEG4 and HDPEG. It involved preparation of samples (10 mg/ml) using
appropriate solvent (chloroform) and mixed with 50:50 acetonitrile: water in 0.1 % Trifluoroacetic acid. 1
µL of the mixture was spotted onto the MALDI target plate and allowed to dry before analysis. This
procedure is qualitative, thus weight of sample was not critical. The samples were run on the Voyager DE-
PRO in linear positive mode with an accelerating voltage of 20000 V and an extraction delay time 170
nsec. 200 shots per spectrum were obtained for each sample. The range was dependent upon the sample
itself shown in their respective spectra in the results section.
2.3.3. Fourier transform-Infrared (FT-IR) spectroscopy
The infrared absorption spectrum for the samples was obtained using the Perkin Elmer spectrum 100 FT-
IR spectrometer (Perkin Elmer, U.K.) equipped with a sampler accessory to perform Horizontal Attenuated
Total Reflectance (HATR) and a germanium crystal (4000 to 650 cm-1) and EZ Omnic IR software
(maximum resolution is 1 cm-1) averaging 512 scans at 2 cm-1 resolution. The spectra were viewed with
EZ Omnic Version 5.0 computer IR analysis program (Perkin Elmer, U.K.). A background spectrum was
recorded on a clean germanium spectral window before a sample spectrum was recorded. This was meant
to identify key functional groups present in the starting materials and finished product.
2.4. Self assembly studies of PEG-based amphiphiles
The aqueous solution of the polymer samples (2mg ml-1) were prepared by probe sonication (Soniprep
150 Instruments, Sanyo, U.K.) of the polymer samples in distilled water or specified diluents (pyrene and
methyl orange solution) for 10 minutes at 50 % of its maximum output prior to self assembly studies.
2.4.1. Pyrene probe
The preparation of dilute aqueous solution of Pyrene (2µM) involves dissolution of Pyrene in ethanol
(0.4mg mL-1); 100 µL of the solution is drawn into a 100 mL volumetric flask and the ethanol present was
evaporated off, under a stream of nitrogen. The residue was then made up to 100mL solution with distilled
water. The aqueous pyrene solution was then used as the solvent to prepare various concentrations
(10mg mL-1 to 1x10-4mg mL-1) of polymer dispersions by probe sonication as previously described. The
successive lower concentrations were produced by serial dilution of the stock solution. The fluorescence
emission spectra of the pyrene-polymer samples were recorded (350-450nm) at an excitation wavelength
of 335nm using the fluorescence spectrophotometer (Perkin Elmer LS50-B). The aqueous pyrene solution
also served as the control. The I1/I3 ratio was calculated from the ratio of the intensity of the first
(372.5nm) to the third (392.5nm) vibronic peaks in the Pyrene emission spectra (Uchegbu et al., 2001;
Kalyanasundaram et al., 1977).
2.4.2. Photon Correlation Spectroscopy (PCS) measurements
The PEG-based micellar aggregates were prepared at a concentration of 2mg ml-1 by sonication for 10
minutes. The hydrodynamic diameter of the self assembled systems was measured using photon
correlation spectroscopy (Malvern Zetasizer 3000HSA, Malvern Instruments Ltd., U.K.) at room
temperature (25oC) at a wavelength of 633 nm and the data analysed using the Contin method of data
analysis. Measurements were carried out in triplicates.
2.4.3. Transmission Electron Microscopy (TEM) with negative staining
A drop of the polymer solution filtered through 0.45 µm filter (2mg ml-1) was put on carbon coated grid
and excess sample filtered off with No. 1 Whatman Filter paper and negatively stained with 1% aqueous
uranyl acetate. The images were focused using the FEI CM120 BioTwin Transmission Electron Microscope
(Philips, Eindhoven, Netherlands) to visualize morphology and self assemblies of polymers in the aqueous
environment. Digital images were captured using a digital camera. This procedure was carried out in
triplicate.
3. Drug encapsulation studies
Amphiphilic polymers have been employed for the protection as well as controlled release of the
encapsulated hydrophobic agents such as insulin as demonstrated by Thompson et al., (2010). Thus the
synthesized and characterised amphiphilic star polymer (P8PEG4) served as the model for star shaped
amphiphiles and was evaluated for its drug encapsulation efficiency using griseofulvin as a model drug.
Results obtained for P8PEG4 was compared with that of the linear analogue (HDPEG).
The amount of encapsulated drug was determined using high performance liquid chromatography (HPLC).
This instrument is widely employed for the qualitative and quantitative analysis of pharmaceuticals and it
involves the elution of a sample based on its preferential affinity for the mobile or stationary phase. Both
phases are chosen with a good knowledge of the physicochemical properties of the sample. Stationary
phase consists of reversed phase octadecylsilane (retention based on hydrophobic vander Waals
interactions) packed in columns (Braithwaite & Smith, 1996). The detector employed was UV
spectrophotometer.
3.1. Materials
Material
Supplier
PEG-based amphiphiles (P8PEG4 & HDPEG)
Synthesized in section 2
Griseofulvin (99% HPLC grade)
Sigma-Aldrich Co., U.K.
Methanol
Sigma-Aldrich Co., U.K.
3.2. Methods
3.2.1. Preparation of drug loaded polymer aggregates of P8PEG4 and HDPEG
P8PEG4 was selected for the preparation of drug loaded polymer aggregates, representing the star
architecture because it displayed greater self-assembly features. Polymer dispersions (5mg ml-1) in
distilled water were prepared by probe sonication of P8PEG4 and HDPEG and transferred to glass vials
containing pre-weighed griseofulvin powder (5mg). The drug was loaded into the polymer aggregates by
probe sonication for 10 min at 50% maximum outputs and the amount of encapsulated drug analysed by
HPLC.
3.2.2 HPLC analysis of drug loaded onto P8PEG4 and HDPEG micellar aggregates
The drug-polymer dispersions (P8PEG4 and HDPEG) were filtered (0.45µm) to remove unloaded drug. The
filtrate for P8PEG4 was dissolved in methanol: water mobile phase (3:1) and analysed by the HPLC system
adapted to Walters 515 HPLC pump, Walters 717 plus autosampler and Walters 486 Tunable absorbance
detector. The reverse phase column (ONYX™), 10cm x 4.6 mm C18 maintained at 60oC. The flow rate for
the mobile phase was 1.5 ml min-1 and the volume of injection was 20 µl. The griseofulvin peak was
detected at 292nm with a retention time of 1.42 min. The data was analysed using waters Empower
computer software. A calibration curve for griseofulvin (r2 = 0.9993) was obtained from various
concentrations of standard griseofulvin solutions ranging from 1µg ml-1 to 100µg ml-1. The same method
was employed for filtrate-containing griseofulvin encapsulated within HDPEG, except that the mobile
phase was methanol: water: acetic acid – 45:54:1 and r2 = 0.9999 from the resulting calibration curve. The
retention time was obtained at 6.18 min. The equations of the straight line for interpreting encapsulated
drug for P8PEG4 and HDPEG were y = 57.653x and y = 63.876x respectively as different mobile phases
were used.
4. Results
4.1. Synthetic yield and physical properties
Three polyethylene glycol based (two star shaped and one linear) amphiphilic polymers were successfully
synthesized and their reaction yields and physical properties recorded in Table 4.1a & b. The yield was
determined by dividing the weight of the freeze dried product by the weight of the starting materials used
in the synthesis schemes and multiplying with 100 to obtain a percentage as shown below.
The mass and percentage yield of the product was ~ 66.50 mg and 12.01 % w/w respectively.
Percentage yield (% w/w) = Weight of reactants × 100 %
Weight of products
66.50mg × 100 % = 12.01 % w/w
554.00mg
Table 4.1a: Synthetic yield and degree of palmitoylation (using 1 H NMR data) of PEG based amphiphiles,
(n depicts the number of batches synthesized).
Polymer
Yield
(mg)
Yield
(% w/w)
Palmitoylation (mole %)
arms substituted
with palmitoyl
groups
P8PEG1
(n = 3)
66.5 ± 1.01
12.01 ±
0.786
10.615 ± 2.555
1 arm
P8PEG4
(n = 3)
60.5 ± 8.93
10.98 ±
7.02
52.15 ± 5.44
4 arms
HDPEG
(n = 3)
113.8 ± 3.46
20 ± 2.72
Not applicable
Not applicable
The Physical properties of the star-shaped and linear amphiphiles are presented in Table 4.1b.
Table 4.1b: Physical properties of PEG based amphiphiles.
Polymer
Lyophilized form
Aqueous solution
Filtered form
P8PEG1
Off-white
Opaque
translucent
P8PEG4
Cotton-like
Opaque
translucent
HDPEG
Cotton-like
Translucent (foamy)
Clear
4.2. Structural Characterisation
4.2.1. NMR and COSY data
The synthesis of the star-shaped amphiphiles were confirmed by the NMR spectra for 8arm PEG, PNS
(starting materials), P8PEG and P8PEG4 (products) as shown in Figures 4.1a-b, 4.2a-b, 4.3a-b, and 4.4a-b
respectively.
The proton assignments for 8PEG amine are listed:
δ2.9-3.0 ppm = CH2 (methylene group attached to OR group from the core)
δ3.3-3.7 ppm = CH2 (methylene group from the ethylene glycol moiety)
δ4.3ppm = CH2 (β-methylene group to the terminal amine group)
δ4.8 ppm = CH2 (α-methylene group to the terminal amine group)
The proton assignments for PNS are listed:
δ0.9 ppm = CH3 (terminal methyl group)
δ1.3-1.5 ppm = (CH2)12 (methylene)
δ2.6 ppm = CH2-C=O (α-palmitoyl deshielded by carbonyl)
δ2.8 ppm = CH2 (methylene group in the succinimidyl ring)
δ1.7 ppm = CH2-C=O (β-palmitoyl deshielded by carbonyl)
The proton for P8PEG1 and P8PEG4 have similar chemical shift but they differ in terms of intensity, which
is due to greater degree of palmitoylation in the latter. Their proton assignments are listed:
δ0.9 ppm = CH3 (palmitoyl methyl protons)
δ1.3 ppm = (CH2)13 (palmitoyl methylene protons)
δ1.6 ppm = CH2-CH2-C=O (β-palmitoyl methylene protons deshielded by carbonyl group)
δ2.2 ppm = CH2-C=O (α-methylene protons deshielded by carbonyl group)
δ3.0 ppm = -CH2- (terminal methylene protons adjacent to PEG core)
δ3.4-3.7 ppm = CH2 (PEG methylene protons attached to palmitoyl protons)
b
d
a
c
Figure 4.1a: 1H NMR spectrum of 8PEG in CDCl3
d
c
b
c
a
a
b
d
Figure 4.1b: 1H – 1H COSY of 8PEG in CDCl3
a
b
Hc coupled with Hd
Ha coupled to Hb
d
c
Figure 4.2a: 1H NMR spectrum of PNS in CDCl3
a
b
c
a
b
c
d
d
e
e
Figure 4.2b: 1H – 1H COSY of PNS in CDCl3
b
c
d
a
a
b
c
d
Ha coupled with Hb
Hb coupled with He
Hc coupled with He
e
e
Figure 4.3a: 1H NMR spectrum of P8PEG1 in CDCl3
a
b
c
d
a
b
c
d
e
e
f
f
Figure 4.3b: 1H – 1H COSY of P8PEG1 in CDCl3
a
a
b
b
c
c
d
d
e
e
Ha coupled to Hb
Hb coupled to Hf
Hc coupled to Hf
f
f
CDCl3
CDCl3
Figure 4.4a: 1H NMR spectrum of P8PEG4 in CDCl3
a
b
c
d
e
CDCl3
a
b
c
d
e
f
f
a
b
c
d
a
e
c
d
CDCl3
Ha coupled with Hb
Hb coupled with Hf
Hc coupled with Hf
CDCl3
e
b
f
f
Figure 4.4b: 1H – 1H COSY of P8PEG4 in CDCl3
Also, synthesis of the linear amphiphile was confirmed from the NMR spectra of SSAMPEG, HDA (starting
material) and HDPEG (product) as shown in Figures 4.5a-b, 4.6a-b and 4.7a-b respectively.
The proton assignments for SSAMPEG are listed:
δ1.8 = -CH3-O (terminal methoxyl group)
δ2.6-3.0 = -CH2- CH2- (attached to carbonyl group)
δ3.4-3.5 = -CH2 (in the succinimidyl ring)
δ3.6-3.8 = -CH2- (ethylene glycol protons)
The proton assignments for HDA are listed:
δ0.9 = CH3 (terminal methyl group)
δ1.3-1.5 = (CH2)14 (methylene)
δ2.7 = CH2-NH2 (methylene proton attached to the amine group)
The proton assignments for HDPEG are listed:
δ0.9 = CH3 (palmitoyl)
δ1.3 = (CH2)13 (palmitoyl)
δ1.5 = CH2-NH (β-methylene group from the hexadecyl moiety)
δ1.7 = CH2-NH (α-methylene group from the hexadecyl moiety)
δ2.5 = CH2-NH (methylene group attached to the PEG moiety by the amide linker)
δ3.4-3.7 = CH2 (ethylene glycol moiety attached to hexadecyl protons)
Figure 4.5a: 1H NMR spectrum of SSAMPEG in CDCl3
a
c
b2
b
d
d
c
b1
a
Figure 4.5b: 1H – 1H COSY of SSAMPEG in CDCl3
c
b1
d
a
c
d
Hb1 coupled with Hb2
b2
b2
a
b1
Figure 4.6a: 1H NMR spectrum of HDA in CDCl3
a
b
c
d
a
b
c
d
Figure 4.6b: 1H – 1H COSY of HDA in CDCl3
a
b
c
d
Ha coupled with Hb
coupled
a
b
c
d
Hb coupled with Hc
coupled
Hc coupled with Hd
Figure 4.7a: 1H NMR spectrum of HDPEG in CDCl3
a
a
b
b
c
c
d
d
e
e
f
f
Figure 4.7b: 1H – 1H COSY of HDPEG in CDCl3
The degree of hydrophobic substitution in P8PEG was calculated based on the chemical shift in the 1H
NMR spectra (Uchegbu et al., 2001) and slightly modified to suit the molecular architecture. The
a
b
f
d
e
a
b
c
d
e
Ha coupled to Hb
Hc coupled to Hd
c
f
Hb coupled to Hf
calculations were carried out using the ratio of palmitoyl methyl protons (δ = 0.9 ppm) to the polyethylene
glycol (core) protons (δ = 3.4-3.8 ppm) as expressed in the example below (Uchegbu et al., 2001).
Palmitoylation (mole %)
= Integration of palmitoyl methyl protons at 0.90 ppm/24 x 100% ……eq. 1
Integration of polyethylene glycol protons at 3.4-3.8 ppm / 16
= 4.85/24 x 100%
24.55/16
= 13.17 % ~ 1 arm palmitoyl substitution
If 100 % mole Palmitoylation = 8 arm of the 8PEG substitution with palmitoyl groups,
Then 13.17 % mole Palmitoylation depicts 1 arm substitution.
Also “24” represents the total methyl protons within the 8 arms and “16” represents the total ethylene
glycol (OCH2CH2) protons within one arm with four repeating units.
4.2.2. Mass Spectroscopy (MALDI-TOF MS & ESI-MS)
The nominal molecular weights of the three amphiphiles were determined by considering the molecular
weight of the abundant specie in the MALDI-TOF spectra (Figure 4.8a-d). This was used to confirm the
extent of Palmitoylation of the star shaped amphiphiles (P8PEG1 and P8PEG4) as well as the formation of
the linear amphiphile (HDPEG).
Table 4.2: Molecular weight distribution of PEG based amphiphiles
Polymer
Average MW (Da)
Palmitoylation
(mole %)
P8PEG1 (n = 3)
2244 ± 228.64
10.46 ± 2.555
P8PEG4 (n = 3)
3000 ± 234.00
56.15 ± 10.21
HDPEG (n = 3)
2183 ± 2.72
Linear amphiphile formed
Figure
4.8a: MALDI-TOF Gaussian distribution curve for 8 armed PEG (starting material for the star-shaped
amphiphile) showing average molecular weight of 2004 Da.
Ideal MW: 2000Da
Figure 4.8b: ESI-MS Gaussian distribution curve for PNS (starting material for the star-shaped amphiphile)
showing average molecular weight of 358 Da.
Ideal MW = 353.5Da
The nominal molecular weight for the star shaped amphiphiles was estimated based on an increase of
239 Daltons per arm of the polymer that is substituted with palmitoyl groups.
Thus, molecular weight increase of 1912 Da depicts 100% palmitoylation (8 arms substitution)
Then MW increase of 1000 Da (with respect to starting material with MW of 2,000) shows ~52%
Palmitoylation (4 arm substitution with palmitoyl groups), that is
% palmitoylation (With respect to MALDI-TOF data)
= max. palmitoylation achievable x MW increase ………………….eq. 2a
max. MW increase achievable
= 100 x 1000 % ~ 52% substitutions.
OR
Number of arms palmitoylated
= MW of resulting amphiphile – MW 8PEG …………………………eq. 2b
239
= 3234 - 2004 ~ 4 arms of the 8PEG substituted with palmitoyl groups.
239
Figure 4.8c: MALDI-TOF Gaussian distribution curve for P8PEG1 (star-shaped amphiphile) showing
nominal molecular weight of 2114 Da.
Figure 4.8d: MALDI Gaussian distribution curve for P8PEG4 (star-shaped amphiphile) showing nominal
molecular weight of 3234 Da.
Ideal MW = 2239 Da
Ideal MW = 3984 Da
The ESI-MS and MALDI-TOF molecular weight distribution curve of the linear amphiphilic polymer as well
as their starting materials are presented below (Figure 4.8e -g).
Figure 4.8e: MALDI Gaussian distribution curve for SSAMPEG (starting material for the linear polymer)
showing nominal molecular weight of 1871 Da.
SSAMPEG
Ideal MW: 2000
Figure 4.8f: MALDI Gaussian distribution curve for HDA (starting material for the linear polymer) showing
nominal molecular weight of 243.3gmol-1.
Figure 4.8g: MALDI Gaussian distribution curve for HDPEG (linear polymer) showing nominal molecular
weight of 2183.8 Da.
Ideal MW: 241.46gmol-1
4.2.3. Fourier Transform infrared spectroscopy (FT-IR)
The FT-IR spectrum for the star shaped polymers (P8PEG1 & P8PEG4) showed characteristic peaks
representing functional groups present in both starting materials (8PEG & PNS) as shown in Figure 4.9a-
d.
Figure 4.9a: FT-IR spectrum for 8PEG showing main functional groups
Figure 4.9b: FT-IR spectrum for PNS showing main functional groups.
-NH2
-CH3, -CH2-
-CH3, -CH2-
1750cm-1
C=O
-CH3, -CH2-
-CH2CH2
-NH
Figure 4.9c: FT-IR spectrum of P8PEG1 showing main functional groups resulting from 8PEG and PNS
Figure 4.9d: FT-IR spectrum of P8PEG4 showing main functional groups resulting from 8PEG and PNS
The FT-IR spectrum for the linear polymer (HDPEG) showed characteristic peaks representing functional
groups present in both starting materials (SSAMPEG & HDA) as shown in Figure 4.9e-g.
-CH3, -CH2-
-C=O
-NH
1640 cm-1
-CH3, -CH2-
-NH
C=O
Figure 4.9e: FT-IR spectrum for SSAMPEG showing main functional groups.
Figure 4.9f: FT-IR spectrum for HDA showing main functional groups.
-CH3, -CH2-
N-O
C=O
-CH3, -CH2-
CH2CH2
-CH3, -CH2-
-NH2
-CH2CH2
-CH2CH2
-NH2
Figure 4.9g: FT-IR spectrum for HDPEG showing main functional groups generated from SSAMPEG and
HDA.
4.3 Self assembly studies
4.3.1 Pyrene probe studies
Pyrene was employed as the hydrophobic probe for the self assembly studies. The Fluorescence intensity
of the Pyrene monomer is dictated by polarity of the surrounding environment and these variations in the
vibronic band intensity ratio (I1/I3) have been used to determine the CMC values of the linear and star
shaped amphiphiles (Kalyanasundaram & Thomas, 1977; Uchegbu et al., 2001). Pyrene has poor aqueous
solubility and its fluorescence intensity is minimal in aqueous medium due to self quenching
(Kalyanasundaram & Thomas, 1977). The I1/I3 ratio generated from the emission spectra of pyrene
changes as it partitions into the non polar regions of amphiphilic aggregates as shown in Figure 4.10.
-NH 3319cm-1 -
links both
moieties
-CH2CH2
from HDA
and PEG
-CH2CH2 from
HDA
-C=O
Figure 4.10: Effect of increasing concentrations of P8PEG4 (0.0001 – 10mg ml-1) on the emission spectra
of pyrene.
The critical micellar concentration (CMC) for the star shaped amphiphiles (P8PEG1 & P8PEG4) and the
linear amphiphile (HDPEG) evaluated from the pyrene probe analysis is presented in Table 4.2.
Table 4.2: Critical micellar concentration of polyethylene glycol amphiphiles determined using the Pyrene
probe method (n = 3).
Polymer
pyrene
Mean ± S.D
mg ml-1
μM
P8PEG1
0.02
8 ± 0.002
P8PEG4
0.009
3 ± 0.001
HDPEG
0.077
15 ± 0.003
The CMC was obtained from the point of intersection of the two tangential lines on the graph of I1/I3 ratio
against polymer concentration as shown in Figure 4.11a-c.
Figure 4.11a: Determination of CMC of P8PEG1 (0.0001 – 10mg ml-1) from the I1/I3 ratio of Pyrene
Figure 4.11b: Determination of CMC of P8PEG4 (0.0001 – 10mg ml-1) from the I1/I3 ratio of pyrene.
CMC
CMC
Figure 4.11c: Determination of CMC of HDPEG (0.0001 – 10mg ml-1) from the I1/I3 ratio of pyrene.
4.3.2 Photon Correlation Spectroscopy using Malvern Zetasizer
P8PEG1 and P8PEG4 appeared as opaque micellar liquids while HDPEG appeared as clear isotropic micellar
liquids when prepared as 2mg mL-1 aqueous solutions. The particle size and polydispersity of polymer
dispersions after sonication for ten minutes are presented in Table 4.3a-b.
Table 4.3a: PCS measurement of unfiltered polymer dispersions (2mg mL-1)
Polymer dispersion
Mean size (nm)
Polydispersity
P8PEG1
500.4 ± 91.4
0.829 ± 0.276
P8PEG4
417.3 ± 47.7
0.526 ± 0.109
HDPEG
571.1 ± 123
0.983 ± 0.057
Table 4.3b: PCS measurement of filtered polymer dispersions (2mg mL-1) using 0.45 µm filter
CMC
Polymer dispersion
Mean size (nm)
Polydispersity
P8PEG1
182.6 ± 7.2
0.727 ± 0.276
P8PEG4
77.7 ± 5.1
0.225 ± 0.109
HDPEG
178.4 ± 6.5
0.783 ± 0.057
Various aggregate populations with similar particle size range are represented by the different peaks
shown in the PCS size distribution graphs (Figure 4.12a-c):
Figure 4.12a (i): PCS size distribution curve of unfiltered P8PEG1 dispersion
Figure 4.12a (ii): PCS size distribution curve of filtered P8PEG1 dispersion
Figure 4.12b (i): PCS size distribution curve of unfiltered P8PEG4 dispersion
Figure 4.12b (ii): PCS size distribution curve of filtered P8PEG4 dispersion
Figure 4.12c (i): PCS size distribution curve of unfiltered HDPEG dispersion
Figure 4.12c (ii): PCS size distribution curve of filtered HDPEG dispersion
4.3.3 Transmission electron microscopy
The morphology of the polymer aggregates for the PEG-based amphiphiles as illustrated by TEM (Figure
4.12a-c) is presented below. The dark and light background observed in the images are associated with
the density of the sample as electrons are deflected from the more dense regions while electron pass
through the less dense portions of the sample and images appear on the fluorescent screen (Hyatt, 1986).
Figure 4.13a: TEM image of filtered (0.45µm) P8PEG1 (2mg mL-1) dispersion.
Size bar 1 micron
Figure 4.13b: TEM image of filtered (0.45µm) P8PEG4 (2mg mL-1) dispersion.
Figure 4.13c: TEM images of HDPEG (2mg mL-1) dispersions i) filtered
ii) Unfiltered, whose particle size correlate with PCS size distribution.
4.4 Griseofulvin encapsulation studies
Size bar 500 nm
Size bar 100nm [i]
Size bar 500 nm [ii]
4.4.1 Extent of drug loading / encapsulation efficiency for the amphiphiles
Results obtained from loading Griseofulvin onto PEG based amphiphiles at concentration of 5mg/ml is
presented in Table 4.4.
Table 4.4: Result of drug loading of PEG-based amphiphiles
Polymer
Drug loading per 5mg polymer
Encapsulation efficiency (%)
P8PEG4
0.04 ± 0.001mg
0.8
HDPEG
0.09 ± 0.002mg
1.79
5. Discussion
Poly (ethylene glycol) is a synthetic polymer approved by FDA for numerous biomedical and
pharmaceutical applications due to its hydrophilicity, solubility in water and organic solvents,
biocompatibility and safety (Zalipsky & Harris, 1997). It is also the hydrophilic polymer of choice for the
synthesis of various linear amphiphilic copolymers (Kwon & Kataoka, 1995). Recent advances in polymer
chemistry that affords the design and characterization of highly branched amphiphilic copolymers have
also employed it as the hydrophilic block (Lu et al., 2006; Cheng et al., 2007; Dong et al., 2010). Moreover,
PEG has been reported in several articles for prolonging the plasma circulating time of therapeutic agents
especially when conjugated with nanosystems such as doxil, a PEGylated liposomal formulation containing
doxorubicin approved for the treatment of ovarian carcinoma (Lasic, 1993).
PEG with a star architecture consisting of eight arms grafted to palmitoyl groups has been investigated in
this study, as a follow up to the findings from study carried out on palmitoylated dendrimer amphiphiles
with a flexible layer of PEG, which formed self-assembly with non-hemolytic property (Chooi et al., 2010).
The suitability of palmitoyl groups (facilitating self assembly) as the hydrophobic moiety for the
preparation of linear and branched graft copolymers has been established (Qu et al., 2006; Chooi et al.,
2010). Besides that, star shaped block copolymers (with fewer arms) have also self-assembled into
nanostructures (Cheng et al., 2006 & 2007).
We thus decided to investigate novel star shaped amphiphilic graft copolymers (with increased number
of arms). Two types of amphiphiles with different degrees of palmitoylation, P8PEG1 (palmitoylPEG with
one arm substituted with palmitoyl group) and P8PEG4 (palmitoylPEG with four arms substituted with
palmitoyl groups) were synthesized to determine the influence of hydrophobicity on their self-assembly
properties. The molar feed ratio and other reagents shown in Table 2.1 was used to control the
palmitoylation levels as their 8-armed PEG was grafted to one and four palmitoyl groups respectively. The
reaction stoichiometry of the starting materials was such that the proportion of PNS to 8-armed PEG was
tripled in order to improve the hydrophobic modification from one palmitoyl to four palmitoyl group-
substitution. This is because abundant (30) moles of the hydrophobic PNS reactant interacted with a mole
of the hydrophilic backbone (8-PEG), which favoured greater hydrophobic substitution in P8PEG4, relative
to P8PEG1. However, 10 moles of PNS was adequate to effect palmitoylation of one arm of 8-armed PEG
to synthesize P8PEG1.
It has been established that linear copolymers have different gel-sol transition properties from their star
shaped analogues (Lu et al., 2006). A novel amphiphilic linear graft copolymer, hexadecyl PEG (HDPEG)
was also synthesized in order to compare linear with branched PEG amphiphiles. The hydrophilic
backbone (PEG) for the linear and star shaped graft copolymer had same molecular weight (MW) as MW
is known to influence aggregation (Qu et al., 2006). This is meant to exclude the influence of molecular
weight on self-assembly features of our amphiphiles. The chain length for their hydrophobic pendant
group was also similar (palmitoyl and hexadecyl group [C16]) as hydrophobic chain length also affects
extent of aggregation (McCormick & Chang, 1994).
The amphiphiles were successfully synthesized and percentage yields were 10-20%, with HDPEG having
the greatest yield (20%), which could be due to its linear architecture; synthesis of the branched material
may be sterically hindered. The synthetic yields for the star shaped amphiphiles (P8PEG1 & P8PEG4) were
similar (12.01% vs. 10.98%), confirming the role of molecular architecture on the yield of amphiphiles.
The synthesis of P8PEG1, P8PEG4 and HDPEG (Figure 5.1) was revealed from 1H NMR & 1H-1H COSY (Fig.
4.1-4.7), and FTIR analysis (Fig. 4.9a-f). The average degree of palmitoylation of P8PEG1 and P8PEG4
(Table 4.1) evaluated using their 1H NMR data involved comparison between palmitoyl methyl protons (δ
~ 0.9 ppm) and PEG methylene protons (-O-CH2-, δ ~ 3.4-3.7 ppm), applying eq. 1, pg 44. It revealed that
P8PEG1 and P8PEG4 had one and four palmitoyl groups (grafted to the 8-armed PEG backbone)
respectively. The degree of hydrophobic modification of the PEG core was not applicable to HDPEG
because of its linear architecture but the grafting of the hexadecyl pendant group (δ ~ 0.9 ppm) to the
PEG backbone (-O-CH2-, δ ~ 3.4-3.7 ppm) in HDPEG was confirmed by the MALDI-TOF data, nominal MW
of 2,183 Da depicting MW increase of ~ 300 Da as the succinimidyl moiety is replaced by the hexadecyl
pendant group (Fig. 4.8e – g). Both moieties are linked by amide bond and are not shown from the 1H
NMR data (Fig. 4.4-4.6).
The most abundant specie in the MALDI-TOF spectrum was taken as the nominal MW. The MALDI-TOF
mass spectrometry data is the most definitive data set; and it was used to confirm the level of
palmitoylation. It showed nominal MW of 2,114 and 3234 Da for P8PEG1 and P8PEG4 respectively. (Fig.
4.8 c & d). The correlation of their molecular weight with their degree of palmitoylation was established
using eq. 2a & b; pg 47, which confirmed that the PEG backbone of P8PEG1 and P8PEG4 had one and four
palmitoyl groups respectively, grafted onto them (Fig. 4.8c & d). The degree of palmitoylation calculated
using the nominal MW of the star-like amphiphiles (associated with the most abundant specie in the
spectra) correlated with that obtained via 1H NMR data (Table 4.2 vs. Table 4.1).
Characteristic functional groups associated with the linear and star shaped amphiphiles were also
observed in their FTIR spectra to confirm identity of the starting materials and synthesis of the novel
amphiphiles (Fig. 4.9a-f). The FT-IR spectra represented in Fig. 4.9c & d (P8PEG1 & P8PEG4) showed
overlap of functional groups present in their starting materials (Fig. 4.9a & b). This confirms the formation
of the star-shaped amphiphiles evident by the –CH2CH2 stretch appearing at 2849-2917 cm-1 (distinctive
of the modified 8PEG) as well as –NH group (1640 cm-1), which links both segments. Further, FT-IR spectra
of HDPEG (Fig. 4.9g) showed peaks at 2882-2917 cm-1 associated with –CH2CH2 from the ethylene glycol
and hexadecyl moiety. The –NH group from HDPEG appeared at 3334 cm-1 due to the greater intensity
of such group in HDPEG, relative to P8PEG1 & P8PEG4 (Fig. 4.9g vs 4.9c & d). The appearance of FT-IR
spectrum at 2917cm-1 for P8PEG and HDPEG is due to the similarity in their chemical composition (PEG
backbone and C16-pendant group).
The physical appearance of the star shaped amphiphiles (P8PEG1 and P8PEG4) was similar in aqueous
solution as they were turbid in appearance while that of HDPEG was translucent in nature. This may be
associated with their molecular architecture. The filtered polymer dispersions of P8PEG1 and P8PEG4 was
translucent while that of HDPEG was clear, still justifying the influence of molecular architecture on
physical properties of amphiphiles in aqueous medium.
Amphiphiles (molecule possessing hydrophilic and hydrophobic segment) strives to be in a state of
minimal free energy level in an aqueous medium by shielding their hydrophobic domains and allowing the
hydrophilic portion to interact favourably (Florence & Attwood, 1998). They occur as monomers at low
concentrations due to minimal interaction between the hydrophobic groups. However, the critical
aggregation/micellar concentration (CAC/CMC) is reached when amphiphiles aggregate (associate) in the
bulk solution (generating hydrophobic core) due to increasing concentrations of amphiphilic molecules as
they cannot shield additional hydrophobic groups (Florence & Attwood, 1998).
The hydrophobic probe for micellar aggregates used in this study (pyrene) is sensitive to changes in
polarity of the surrounding medium and it preferentially partitioned into the hydrophobic cores of micellar
aggregates at higher amphiphile concentration and is retained within the aqueous medium (polar
environment) at lower amphiphile concentration (Kalyanasundaram & Thomas, 1977) due to decrease in
the magnitude of aggregates. Thus the ratio of the first to third emission peak (I1/I3) in their fluorescence
spectra (Figure 4.10) decreases with respect to increased concentration and vice versa; and this parameter
was used to establish the self-assembly properties of amphiphiles.
The CMC is identified as the critical concentration at which self-assembly of amphiphile monomers into
micellar aggregates persist and demicellization ceases.
The increased I1/I3 (in the lower concentration range) observed (Fig. 4.11 a-c) is due to the free monomers
in the bulk solution and the CMC is attained at the point of gradient change, after which aggregation
persists due to association of the monomers at higher amphiphile concentration with resultant decrease
in the I1/I3 ratio as pyrene partitions into the more hydrophobic environment (Kalyanasundaram &
Thomas, 1977).
Their molecular architecture as well as their degree of hydrophobic substitution influenced their CMC
values (Table 4.2) as the PEG molecular weights was kept constant. Their CMC values presented in the
order of increasing CMC values and decreasing micellar stability are: P8PEG4 < P8PEG1 < HDPEG.
However, the CMC values of both linear and star-like amphiphiles was lower than that of Pluronic block
copolymer F127 (Alexandridis et al., 1994) and Polysorbate 20 (Florence & Attwood, 1998) with CMC of
550 µM and 53 µM respectively, inferring that their micelles were more stable and resistant to dilution
in biological fluid than these triblock copolymers and low molecular weight surfactants.
P8PEG4 with four arms substituted with palmitoyl groups had greater degree of unfavourable interaction
of its hydrophobic alkyl chains and water molecules at a lower concentration (3 µM) than P8PEG1 (8 µM)
with decreased hydrophobicity. However, the CMC value of the star shaped polymers (P8PEG1 & P8PEG4)
was lower than that of the HDPEG (15 µM) with the linear molecular architecture. As aggregation is driven
by the entropy gain achieved when water molecules are freed from the hydrophobic cavity and is able to
hydrogen bond freely. We speculate that the less extended branched polymer which interacts less with
water self-assemble faster than the linear polymer which interacts more with water molecules.
Nevertheless, the CMC values observed for these polyethylene glycol based amphiphiles (3-15 µM)
suggest that they have solubilising potential and would be resistant to extreme dilution in vivo. This will
allows the incorporated hydrophobic agents to remain solubilised in the micellar aggregates on dilution
of the aggregates.
Figure 5.1: Schematic representation of the molecular arrangement in the various self-assemblies of PEG-
based amphiphiles.
The freshly prepared unfiltered P8PEG4 and HDPEG formed aggregates with particles size range (400 –
550 nm) and the polymers with lesser degree of hydrophobic substitution (P8PEG1 & HDPEG) had high
polydispersities whereas P8PEG4 had narrow particle size distribution as shown in Table 4.3a. The particle
size population for P8PEG1 (Figure 4.12a) consisting of various aggregate sizes (10-50nm, 100-500nm and
moderate volume beyond 500nm) was associated with limited hydrophobic interaction between fewer
palmitoyl groups, which limits their self assembly into aggregates with similar particle size range. A similar
situation was observed with the linear polymer (Fig. 4.12c) with same extent of hydrophobic modification
(C16 carbon chain length) with a very high polydispersity close to 1.00. On the other hand, there was
greater degree of intra and intermolecular association (hydrophobic) between more palmitoyl groups in
P8PEG4 (Figure 4.12b), which allow for formation of nano-assemblies with similar particle sizes.
From the PCS data (Table 4.3b), all the filtered (0.45µm) polymer dispersion had smaller mean particle
size of 77-182 nm (Fig. 4.12 a-c [ii]) than the unfiltered ones, retaining conventional polymeric micelles.
However, their mean particle size distribution was greater than that of linear and branched palmitoylated
amphiphiles studied, which include quaternary ammonium palmitoyl glycol (Qu et al., 2006) and
palmitoylated poly (propyleneimine) dendrimer amphiphiles (Chooi et al., 2010) with mean particle size
range of 30-75 nm and 10-20nm respectively.
TEM was used to understand the morphology of the polymers and also established its relationship with
the molecular architecture. Both P8PEG4 and HDPEG formed spherical nanosystems comprising of
mixtures of polymeric micelles and larger nanoparticles (Fig. 4.13b-c). This observation is not consistent
with findings by Thompson et al., (2008); where palmitoyl-grafted comb-shaped poly (allylamine)-based
amphiphile yielded distinctive dense nanoparticles while the polymeric micelles were observed with cetyl-
grafted polymers. Conversely, star shaped poly (ε-caprolactone)-based polymer studied by Chen et al.,
(2008) yielded only polymeric micelles. This observation may be due to the complex interplay between
the molecular architecture of the PEG core and level of hydrophobic substitution of the PEG backbone.
P8PEG1 yielded novel core-shell polymeric nano- and micro-self-assemblies (Fig. 4.13a), consisting
majorly of particles of about 500 nm (and aggregates of 1.5 µm). This is an interesting finding because
these self-assembles have not been reported in literature. This observation may be due to electrostatic
repulsion between the seven underivatised PEG arms (hydrophilic head groups), which limits inter and
intramolecular association, preventing the formation of conventional polymeric micelles.
Particle size distributions from TEM were similar to that obtained with PCS data (Fig. 4.13a-c vs Table 4.3a-
b). P8PEG4 had smaller aggregates of about 76nm as well as larger aggregates of 626nm. TEM image of
P8PEG4 showed a high axial ratio structures, which may be due to the hydrophobic interaction resulting
from the palmitoylation of 8-armed PEG (four palmitoyl group evenly distributed across the 8 arms of the
PEG). Moreover, their hydrogen bonding potential facilitates the close packing of their molecules. On the
other hand, HDPEG had aggregates of 20nm because the larger aggregates had been filtered. However,
TEM image of the unfiltered HDPEG showed 460nm and 1µm (Fig. 4.13c [i & ii]), which is representative
of their particle size distribution observed with PCS.
The choice of Griseofulvin for the study is due to its poor physicochemical properties (solubility of 9µg ml-
1 at 25oC; permeability: log P 2.18), which results in erratic absorption and justifies the indication of more
effective antifungal agents for dermatophytic infections (Salonen et al., 2005). It has been discovered that
the extent of drug solubilisation within colloidal aggregates may be improved by formation of larger
aggregates from individual monomers (Qu et al., 2006). Thus larger micellar aggregates (>100 nm) formed
by these PEG-based amphiphiles were thought to be suitable for the encapsulation of hydrophobic drugs
(griseofulvin). P8PEG4 (0.8 %) and HDPEG (1%) encapsulated low levels of the drug (Table 4.4) and the
maximal drug concentration obtained was 0.04-0.09 mg/ml with 5mg/ml polymers. Moreover, P8PEG4
was expected to have greater drug loading potential than linear PEG-PLA block copolymer which loaded
0.2mg/ml of Griseofulvin per 5mg/ml of the copolymer with EE of 4-6.5 % (Pierri & Avgoustakis, 2005).
This is a preliminary study as drug encapsulation will require optimization.
The consideration of the novel star-shaped and linear amphiphiles for oral and topical delivery is due to
the pressing need to discover more effective Griseofulvin formulations requiring low therapeutic dose.
Moreover, the oral and topical routes are more convenient than the parenteral route. Nanoparticulate
and micellar formulations have been used to improve oral bioavailability of Griseofulvin (Tobio et al.,
2000; Pierri & Avgoustakis, 2005). Thus these PEG-based amphiphiles may be potential drug carrier for
improved oral delivery of Griseofulvin.
However, their micellar stability suggests their potential for parenteral administration. Their hydrophobic
core may be sterically stabilized by the PEG (hydrophilic corona), which ensures colloidal stability,
improved plasmatic circulation and targeting potential. Thus anticancer and anti-inflammatory agents
may also be loaded and targeted passively or actively to diseased sites.
6. Conclusion
The PEG-based amphiphiles have shown self assembly properties with CMC values within the range of 3-
15 µM. Their degree of hydrophobic modification have also influenced their self-assembly potential, as
the amphiphile with the greatest degree of hydrophobicity (P8PEG4) had the least CMC value (3 µM),
showing greater micelle stability and longevity in biological fluid stability. However, all their CMC values
reveals that stable aggregates are formed which may be of benefit for drug delivery applications.
This is the first report where linear and branched PEGs have been compared and interesting finding
revealed. Molecular architecture had a profound influence on self assembly as the branched amphiphile
(P8PEG4) formed more stable aggregates. Furthermore, a greater degree of hydrophobic substitution
(P8PEG4) improved micellar stability and ensured low polydispersity of particle size distribution. A new
morphology was discovered with P8PEG1, which comprises of core-shell nano- and micro-self assemblies.
Thus the versatility of amphiphilic polymers has been demonstrated because they form a variety of
supramolecular systems (micelles, nanoparticles, and core-shell nano- & micro- core-shell self assemblies)
in aqueous medium and this feature can be exploited to deliver therapeutic and diagnostic agents with
differing physicochemical properties.
7. Future works
This will focus on the influence of amphiphile architecture on drug encapsulation. This is desired as a
result of their potential for the delivery of variety of hydrophobic agents. Increasing polymer
concentrations would also be explored to determine their influence on drug loading / encapsulation
efficiency (EE). The method of encapsulation as well as HPLC protocol for drug analysis will also require
optimization. Also, the chemical composition of the amphiphiles may be modified to improve their drug
loading capacity.
References
Allemann, E., Gurny, R., Doelker, E., 1993. Drug-loaded nanoparticles – preparation methods and drug targeting
issues. Eur. J. Pharm. Biopharm. 39 (5), 173-191.
Benito, S.M., 2006. Functionalized polymer nanocontainers for targeted drug delivery. Available online at:
http://www.edoc.unibas.ch/456/1/DissB_7623.pdf. Accessed 31 July, 2010.
Bergen, J.M., von Recum, H.A., Goodman, T.T., Massey, A.P., Pun, S.H., 2006. Gold nanoparticles as a versatile
platform for optimizing physicochemical parameters for targeted drug delivery. Macromol. Biosci. 6, 506-
516.
Braithwaite, A., Smith, F.J., 1996. Chromatographic methods. 5th edition, London: Blackie Academic &
Professional.
Breulmann, M., Forster, S., Antonietti, M., 2000. Mesoscopic surface patterns formed by block copolymer
micelles. Macromolecular Chem. & Phy. 201, 204-211.
Breunig, M., Bauer, S., Goepferich, A., 2008. Polymers and nanoparticles: Intelligent tools for intracellular
targeting? Eur. J. Pharm. & Biopharm. 68, 112-128.
Burke, S., Shen, H., Eisenberg, A., 2001. Multiple vesicular morphologies from block copolymers in solution.
Macromolecular Symposia 175, 273-283.
Chen, W-Q., Wei, H., Li, S-L., Feng, J., Nie, J., Zhang, X-Z., Zhuo, R-X., 2008. Fabrication of star-shaped, thermo-
sensitive poly (N-isopropyl acrylamide)-cholic acid- poly (ε-caprolactone) copolymers and their self
assembled micelles as drug carriers. Polymer 49, 3965-3972.
Cheng, W.P., Gray, A.I., Tetley, L., Hang, T.L.B., Schatzlein, A.G., Uchegbu, I.F., 2006. Polyelectrolyte
Nanoparticles with high drug loading enhance the oral uptake of hydrophobic compounds.
Biomacromolecules 7, 1509-1520.
Cheng, F.L., Sheng, R.G., Ya, Q.Z., Zong, H.L., Jian, R.G., 2007. Micellization and gelation of aqueous solutions of
star-shaped PEG-PCL block copolymers consisting of branched 4-arm poly(ethylene glycol) and
polycaprolactone blocks. Eur. Polym. J. 43, 1857-1865.
Chooi, K., Gray, A.I., Tetley, L., Fan, Y., Uchegbu, I.F., 2010. The molecular shape of Poly (propylenimine)
Dendrimer Amphiphiles has a profound effect on their self assembly. Langmuir 26(4), 2301-2316.
Din, J., Liu, G., 1998. Water soluble Hollow nanospheres as potential drug carriers. J. Phy. Chem. B 102(31), 6107-
6113.
Discher, B.M., Hammer, D.A., Bates, F.S., Discher, D.E., 2000. Polymer vesicles in various media. Current Opinion
in Colloid & Interface Sci. 5, 125-131.
Discher, D.E., Eisenberg, A., 2002. Polymer vesicles. Science 297, 967-973.
Dominguez, A., Fernandez, A., Gonzalez, N., Iglesias, E., Montenegro, L., 1997. Determination of critical micelle
concentration of some surfactants by three techniques. J. Chem. Ed. 74, 1227-1231.
Dong, P-W., Wang, X-H., Gu, Y-C., Wang, Yu-J., Wang, Yi-J., Gong, C-Y., Luo, F., Guo, G., Zhao, X., Wei, Y-Q., Qian,
Z-Y., 2010. Self-assembled biodegradable micelles based on star-shaped PCL-b-PEG copolymers for
chemotherapeutic drug delivery. Colloids and Surfaces A: Physicochem. Eng. Aspects 358, 128-134.
Feng, S.S., 2004. Nanoparticles of biodegradable polymers for new-concept chemotherapy. Expert Rev. Med.
Dev. 1, 115-125.
Florence, A.T., Attwood, D., 1998. Physicochemical Principles of Pharmacy.The Macmillan Press: Basingstoke,
U.K.
Francis, M.F., Piredda, M., Winnik, F.M., 2003. Solubilization of poorly soluble drugs in micelles of
hydrophobically modified hydroxypropylcellulose copolymers. J. Control. Rel. 93, 59-68.
Francis, M.F., Cristea, M., Winnik, F.M., 2004. Polymeric micelles for oral drug delivery: Why and how. Pure Appl.
Chem. 76, 1321-1335.
Gaucher, G., Dufresne, M-H., Sant, V.P., Kang, N., Maysinger, D., Leroux, J-C., 2005. Block copolymer micelles:
preparation, characterization and application in drug delivery. J. Control. Rel. 109,169-188.
Gillies, E.R., Goodwin, A.P., Frechet, J.M.J., 2004. Acetals as pH-sensitive linkages for drug delivery. Bioconjug.
Chem. 16, 122-130.
Gref, R., Minamitake, Y., Peracchia, M.T., Trubetskoy, V., Torchilin, V., Langer, R., 1994. Biodegradable long-
circulating polymeric nanospheres. Science 263, 1600-1603.
Hajduk, D.A., Kossuth, M.B., Hillmyer, M.A., Bates, F.S., 1998. Complex phase behaviour in aqueous solutions of
poly (ethylene oxide)-poly (ethyl ethylene) block copolymers. J. Phys. Chem. B 102, 4269.
Huang, R-Q., Qu, Y-H., Ke, W-L., Zhu, J-H., Pei, Y-Y., Jiang, C., 2007. Efficient gene delivery targeted to the brain
using a transferrin-conjugated polyethylene glycol-modified polyamidoamine dendrimer. The FASEB
Journ. – Research Commun. 21, 1117-1125.
Hyatt, M.A., 1986. Basic techniques for transmission electron microscopy. Orlando Academic press.
Jain, R.K., 2001. Delivery of molecular and cellular medicine to solid tumours. Adv. Drug Deliv. Rev. 46, 149-168.
Kabanov, A.V., Alakhov, V.Y., 2002. Pluronic® Block copolymers in drug delivery: from Micellar nanocontainers
to biological response modifiers. Crit. Rev. Therapeutic Drug Carrier Sys. 19, 1.
Kalyanasundaram, K., Thomas, J.K., 1977. Environmental effects on the vibronic band intensities in pyrene
monomer fluorescence and their application to studies of micellar systems. J. Am. Chem. Soc. 99, 2039-
2044.
Kataoka, K., Togawa, H., Harada, A., Yasugi, K., Matsumoto, T., Katayose, S., 1996. Spontaneous formation of
polyion complex micelles with narrow distribution from antisense oligonucleotide and cationic block
copolymer in physiological saline. Macromolecules 29, 8556-8557.
Khanna, K., Varshney, S., Kakkar, A., 2010. Miktoarm star polymers: advances in synthesis, self-assembly, and
application (review). Polymer chemistry (DOI: 10.1039/copy00082e). Accessed August 1, 2010.
Kim, S.Y., Shin, I.G., Lee, Y.M., Cho, C.G., Sung, Y.K., 1998. Methoxy poly (ethylene glycol) and ε-caprolactone
amphiphilic block copolymeric micelle containing indomethacin II. Micelle formation and drug release
behaviours. J. Contr. Rel. 51, 13-22.
Krasia, T.C., Patrickios, C.S., 2002. Synthesis and aqueous solution characterization of amphiphilic diblock
copolymers containing carbazole. Polymer 43,2917-2920.
Kreuter, J., 1994. Nanoparticles. In Colloidal drug delivery systems, J, K., Ed. Marcel Dekker: New York, 219-342.
Kreuter, J., 2001. Nanoparticulate systems for brain delivery of drugs Adv. Drug Deliv. Rev. 47, 65—81.
Kwon, G.S., Katoka, K., 1995. Block copolymer micelles as long-circulating drug vehicles. Adv. Drug Deliv. Rev.
16, 295-309.
Lasic, D.D., 1993. Liposomes: From Physics to Application. Elsevier, Amsterdam: 1-580.
Lasic, D.D., Martin, F.J., 1995. Stealth Liposomes. Boca Raton, FL: CRC Press.
Lipinski, C.A., Lombardo, F., Dominy, B.W., Feeney, P.J., 2001. Experimental and computational approaches to
estimate solubility and permeability in drug discovery and developement settings. Adv. Drug Deliv. Rev.
46, 3-26.
Liu, X.M., Pramoda, K.P., Yang, Y.Y., Chow, S.Y., He, C.B., 2004. Cholesteryl-grafted functional amphiphilic poly
(N-isopropylacrylamide-co-N-hydroxylmethylacrylamide): synthesis, temperature-sensitivity, self-
assembly and encapsulation of hydrophobic agent. Biomaterials 25, 2619-2628.
Lu, C.F., Guo, S-R., Zhang, Y., Yin, M., 2006. Synthesis and aggregation behaviour of four types of different shaped
PCL-PEG block copolymers. Polym. Int. 55,694-700.
Maeda, H., Sawa, T., Konno, T., 2001. Mechanism of tumor-targeted delivery of macromolecular drugs, including
the EPR effect in solid tumor and clinical overview of the prototype polymeric drug SMANCS. J. Control.
Rel. 74, 47-61.
McCormick, C.L., Chang, Y., 1994. Water-soluble copolymers 58. Associative interactions and photophysical
behaviors of amphiphilic behaviors of amphiphilic terpolymers prepared by modification of maleic-
anhydride ethyl vinyl ether copolymers. Macromolecules 27, 2151-2158.
Mishra, V., Mahor, S., Rawat, A., Gupta, P.N., Dubey, P., Khatri, K., Vyas, S.P., 2006. Targeted brain delivery of
AZT via transferrin anchored pegylated albumin nanoparticles. J. Drug Target. 14, 45-53.
Mishra, B., Patel, B.B., Tiwari, S., 2010. Colloidal nanocarriers: a review on formulation technology, types and
applications towards targeted drug delivery. Nanomed.: Nanotechnol., biol. & med. 6, 9-24.
Nakayama, M., Okano, T., Miyazaki, T., Kohori, F., Sakai, K., Yokoyama, M., 2006. Molecular design of biodegrable
polymeric micelles for temperature-responsive drug release. J. Control. Rel. 115,46-56.
Newkome, G.R., Moorefield, C.N., Vogtle, F., 2001. Dendrimers and dendrons: concepts, synthesis and
applications. New York: Wiley-VCH.
Odian, G.G., 2004. Principles of polymerization. 4th ed., 1-50.
Park, T.G., Jeong, J.H., Kim, S.W., 2006. Current status of polymeric gene delivery system. Adv. Drug Deliv. Rev.
58, 467-486.
Peleshanko, S., Tsukruk, V.V., 2008. The architectures and surface behavior of highly branched molecules. Prog.
Polym. Sci. 33, 523-580.
Pierri, E., Avgoustakis, K., 2005. Poly (lactide)-poly(ethylene glycol) micelles as a carrier for griseofulvin. J.
Biomed. Mater. Res. Part A 75A (3), 639-647.
Qu, X., Khutoryanskiy, V.V., Stewart, A., Rahman, S., Papahadjopoulos-Sternberg, B., Dufes, C., McCarthy, D.,
Wilson, C.G., Lyons, R., Carter, K.C., Schatzlein, A., Uchegbu, I.F., 2006. Carbohydrate-Based Micelle
Clusters which enhance hydrophobic drug bioavailability by up to 1 order of magnitude.
Biomacromolecules 7, 3452-3459.
Rangel-Yagui, C., Pessoa-Jr, A., Tavares, L.C., 2005. Micellar solubilization of drugs. J. Pharm. Pharmaceut. Sci. 8,
147-163.
Roovers, J., 1985. (In: Kroschwitz JI, editor). Encyclopedia of polymer science and engineering, New York: Wiley-
Interscience, 2,478-499.
Rosen, M.J., Mathias, J.H., Davenport, L., 1999. Abberant aggregation behavior in cationic germini surfactants
investigated by surface tension, interfacial tension and fluorescence methods. Langmuir 15, 7340-7346.
Rosler, A., Vandermeulen, G.W.M., Klok, A.A., 2001. Advanced drug delivery devices via self assembly of
amphiphilic block copolymers. Adv. Drug Deliv. Rev. 53,95-108.
Salonen, J., Laitinen, L., Kaukonen, A.M., Tuura, J., Bjorkqvist, M., Heikkila, T., Heikkila, H.V., Hirvonen, J., Lehto,
V.P., 2005. Mesoporous silicon microparticles for oral drug delivery: Loading and release of five model
drugs. J. Control. Rel. 108 (2-3), 362-374.
Sheng, R.G., Ya, Q.Z., 2008. Concentration controlled multilevel self assembly of 3-armed poly (ethylene glycol)-
b-poly (ε-caprolactone) block copolymers investigated by AFM. J. Polym. Sci. Polym. Phys. 46,1412-1418.
Shubayev, V.I., Pisanic II T.R., Jin, S., 2009. Magnetic nanoparticles for theragnostics. Adv. Drug Deliv. Rev. 61 (6),
467-477.
Thompson, C.J., Ding, C., Qu, Xiaozhong., Yang, Z., Uchegbu, I.F., Tetley, L., Cheng, W.P., 2008. The effect of
polymer architecture on the nano self-assemblies based on novel comb shaped amphiphilic poly
(allylamine). Colloid Polym. Sci. 286, 1511-1526.
Thompson, C.J., Tetley, L., Cheng, W.P., 2010. The influence of polymer architecture on the protective effect of
novel comb shaped amphiphilic poly (allylamine) against in vitro enzymatic degradation of insulin –
Towards oral insulin delivery. Int. J. Pharm. 383 (1-2), 216-227.
Tobio, M., Sanchez, A., Vila, A., Soriano, I., Evora, C., Vila-Jato, J.L., Alonso, M.J., 2000. The role of PEG on the
stability in digestive fluids and in vivo fate of PEG-PLA nanoparticles following oral administration. Colloids
Surfaces B: Biointerfaces 18, 315-323.
Torchilin, V.P., 2001. Structure and design of polymeric surfactant-based drug delivery systems. J. Control. Rel.
73, 137-172.
Torchilin, V.P., 2005. Recent advances with liposomes as pharmaceutical carriers. Nat. Rev. Drug Discov. 4, 145-
160.
Tully, D.C., Frechet, J.M.J., 2001. Dendrimers at surfaces and interfaces and interfaces: chemistry and
applications. Chem. Commun. 14,1229-1239.
Uchegbu, I.F., Sadiq, L., Arastoo, M., Gray, A.I., Wang, W., Waigh, R., Schatzlein, A.G., 2001. Quaternary
ammonium palmitoyl glycol chitosan – a new polysoap for drug delivery. Int. J. Pharm. 224,185-199.
Wang, W., McConaghy, A.M., Tetley, L., Uchegbu, I.F., 2001. Control of polymer molecular weight may be used
to control the size of palmitoyl glycol chitosan polymeric vesicles. Langmuir 17, 631-636.
Wang, W., Qu, X., Gray, A.I., Tetley, L., Uchegbu, I.F., 2004. Self assembly of cetyl linear polyethyleneimine to
give micelles, vesicles and dense nanoparticles. Macromolecules 37, 9114-9122.
Whitesides, G.M., Mathias, J.P., Seto, C.T., 1991. Molecular self-assembly and nanochemistry: a chemical
strategy for the synthesis of nanostructures. Science 254 (5036), 1312-1319.
Yokoyama, M., Okano, T., Sakurai, Y., Suwa, S., Kataoka, K., 1996. Introduction of cisplatin into polymeric
micelles. J. Control. Rel. 39, 351-356.
Zalipsky, S., Harris, J.M., 1997. Introduction to chemistry and biological applications of poly (ethylene glycol). In:
Harris JM, Zalipsky S, editors. Poly (ethylene glycol) chemistry and biological applications, ACS symposium.
Washington, DC: American Chemical Society, 680, 1–13.
