tailored hollow mesoporous silica nanoplatforms with biological...
TRANSCRIPT
I
Tailored hollow mesoporous silica nanoplatforms with biological labels for colon targeted drug
delivery systems
by
Xiaodong She
(Masters of Engineering)
Submitted in fulfilment of the requirements for the degree of
Doctor of Philosophy
Deakin University
November, 2014
IV
Acknowledgement
Every PhD project involves a new journey that starts with an idea and travels its own
distinct path until a whole thesis is created. I love my PhD journey, with its unexpected
turns and twists, surprises, joy and excitement. But more than anything else, when I look
back at the journey I have taken, I would like to gratefully acknowledge a number of
people who played a crucial role in helping me with my project.
Herein I would like to acknowledge my principle supervisor Prof Lingxue Kong, who
made it possible for you to read this thesis. Thanks to him for his trust and consistent
support, for giving his precious time, and for seeing the potential of this project when it
was just an idea.
I would also like to make this opportunity to express my sincere thanks to Prof Wei Duan,
Dr. Sarah Shigdar in the School of Medicine, Dr. Leonora Velleman, Dr. Fenghua She
Dr. Weimin Gao and Dr Ludovic Dumee for their wonderful assistances, encouragement
and invaluable help. Special thanks to Lijue Chen, Chengpeng Li and Tao Wang whom I
had the immense honour and privilege to work with. Without their help, this doctoral
work would not have been accomplished. Marion Wright, Dr John Denman in Ian Wark
Research Institute, Dr Cara Doherty in CSIRO Materials Science & Engineering, and Dr
Haijin Zhu in Centre of Excellence for Electromaterials Science have provided a lot of
help regarding my experimental work, and hereby acknowledged.
Personally, I would like to thank my husband Canzhong HE for his encouragement and
company during my PHD journey, where he offered support and understanding in all
wethers. I would also like to thank my parents, sisters and other family members for their
consistent support.
V
Abstract
Nanoparticle based targeted drug delivery systems hold great promise for cancer therapy
because of their targeting functions, sustained drug release profiles, reduced side-effects
and ability to overcome multidrug resistance (MDR). Nevertheless, inadequate specificity
and stability, insufficient intratumor drug concentration, and high toxicity related to
carriers are still not well solved and thus restrict clinical translation of such systems. To
overcome these shortcomings, hollow mesoporous silica nanoparticles (HMSNs) were
chosen as carriers for engineering a novel drug delivery system with specifically targeting,
high drug loading, sustained release behaviour, and negligible toxicity.
HMSNs are one of the most promising carriers for effective drug delivery due to their
large surface area, high volume for drug loading and excellent biocompatibility. However
the non-ionic surfactant templated HMSNs often have a broad size distribution, a
defective mesoporous framework or a missed hollow structure because of the difficulties
involved in controlling the formation and organization of micelles for the growth of silica
framework. In this study, a novel “Eudragit assisted” strategy has been developed to
fabricate HMSNs by utilising the Eudrgit nanoparticles as cores and to assist the self-
assembly of micelle organisation. Highly dispersed mesoporous silica spheres with intact
hollow interiors and through pores on the shell were achieved after the removal of
Eudragit core and surfactant micelles. The HMSNs have shown a high surface area (670
m2/g), small diameter (120 nm) and uniform pore diameter (2.5 nm). Eudragit S100 was
proven to assist in the spontaneous organization of the Triton X-100 (TX100) micelles
into structurally ordered nanoscale architecture via hydrogen bonding. The HMSNs have
demonstrated non-cytotoxicity to colorectal cancer cells SW480. The high quality of
HMSNs and their excellent biocompatibility demonstrate that the Eudragit assisted
HMSNs can be potential intracellular vehicles for targeted drug delivery.
Besides biocompatibility, targeting functions are essential for drug delivery systems as
they have the potential to precisely target and kill cancerous cells while leaving normal
cells unharmed. Given that human epidermal growth-factor receptor (EGFR) is over
expressed by tumour cells and their natural ligand, epidermal growth factor (EGF) has
VI
emerged as an attractive targeting molecule labelled onto nanoparticles for cancer therapy.
To generate targeting function of the synthesized HMSNs, EGF was grafted to HMSNs.
HMSNs with and without EGF bio-ligands were covalently linked to a fluorescent dye
(FITC), making them detectable by fluorescence microscopy and flow cytometry. In vitro
studies with colorectal cancer cells SW 480 overexpressing EGFR demonstrated that
HMSNs with EGF bioconjugates (HMSN-EGF) achieved nearly 25 times higher cellular
internalization than the unfunctionalised control at a concentration of 10 μg/mL and an
incubation time of 30 mins. In addition, this superior cellular uptake of HMSN-EGF was
inhibited when EGFR in cancer cells was knocked down by free EGF. Similarly, the
targeting function of HMSN-EGF was very limited when they were applied to cells
without EGFR expression (SW 620), revealing the very high specificity of HMSN-EGF
to EGFR positive cells and the contribution of EGF-EGFR interaction.
On the other hand, the properties of grafted targeting ligands, in terms of effective
immobilization, stability and binding efficiency, significantly affect the subsequent
cellular interactions of HMSNs because the performance of biomaterials in a biological
environment is largely influenced by their surface properties. Although the utilization of
EGF for targeting delivery is well established, a precise control of the density of EGF
attachments on nanoparticles is limited. It is found that a small change in the quantity of
EGF labels on HMSNs may make a significant difference in the subsequent cell activities
of HMSN-EGF. To address this issue, hollow mesoporous silica nanoparticles
functionalized with amine group were conjugated with EGF using carbodiimide
chemistry. The time of flight secondary ion mass spectrometry (ToF-SIMS), a very
surface specific technique (penetration depth < 1.5 nm), was employed to study the
binding efficiency of the EGF to the nanoparticles. Principal component analysis (PCA)
was employed to track the relative surface concentrations of EGF on the surface of
HMSNs. Effective control of the quality and density of the resulting grafted EGF on
nanoparticles was achieved which would provide opportunities for engineering targeted
drug delivery systems with tailored targeting efficacy in the future.
Furthermore, the capacity to encapsulate drugs and perform a sustained release is one of
the most important requirements for a drug delivery system. To evaluate the drug loading
and releasing effectiveness of the HMSNs, the model drug, 5-fluorouracil (5-FU), was
VII
loaded in the resultant HMSNs and achieved a sustained release profile. However, for
drug delivery systems, an adequate intratumor concentration of targeted drug is essential
for effective cancer therapy, which requires a high drug loading capacity of drug carriers.
Moreover, it was reported that, in addition to sustained release, specific drug release rates
are always pursued for personal treatment as different stages of disease evolution need
different drug release rates. In this regard, smart delivery systems with a higher drug
encapsulation and more effective control of drug release rate were of great importance.
To meet these requirements, ways to enhance the loading capacity of HMSNs specifically
to 5-FU were investigated and the results showed that 5-FU loading capacity of HMSNs
can be tuned by introducing chemical groups onto HMSNs, varying pH values for loading
process and generating electrostatic attractions between 5-FU and HMSNs. The highest
5-FU encapsulation ability of HMSNs was achieved when these nanocariers conjugated
with amine groups (-NH2). The highest loading capacity of 288.9 mg(5-FU)/g(HMSNs) was
observed at pH 8.0 which is 2 times of that reported in a similar study. On the other hand,
the release rate of 5-FU from HMSNs was controlled by changing the structure parameter
of HMSNs in terms of pore size, shell thickness and particle diameter. Experimental
results revealed that a larger pore size resulted in a quicker release rate while a thicker
shell decreased the amount of drug released from HMSNs. Moreover, particle diameter
is not as important as pore size and shell thickness in determining the release profile of
HMSNs.
In addition to excellent loading and release, overcoming multidrug resistance (MDR) is
a new requirement set for modern drug delivery systems. MDR referring to the ability of
cancer cells to adapt and survive from chemotherapy is one of the most complex and
challenging problems in cancer treatments. Although nanopariticle based drug delivery
systems are often used to prevent MDR in cancer treatment by sidestepping drug
resistance mechanisms, the development of drug resistant genes by cancer cells ultimately
leads to unsatisfied outcomes. To overcome the 5-FU resistance, HMSNs were modified
to deliver siRNA to silence the gene expression involved in MDR. It is found that rather
than surface charge, pore size is a key parameter of HMSNs for effective siRNA
encapsulation. In particular, when the pore size of HMSNs was expanded form 2.5 nm to
4.3 nm, an obvious increase in siRNA loading was found. Almost all siRNA can be
encapsulated into HMSNs when HMSNs have a pore size larger than 4.3 nm and the
VIII
particle to siRNA ratio is 100.
In summary, high quality HMSNs were synthesized via a newly developed strategy
assisted by Eudragit. These HMSNs were proven to be non-cytotoxic to colorectal cancer
cells SW480. The HMSNs have a high drug and siRNA loading capacity, sustained and
controlled release, and efficient and selective cell internalization when bioconjugated
with Epidermal Growth Factor (EGF). The promising in vitro cell tests have revealed the
great potential of the HMSNs to be used for colorectal cancer therapy.
IX
CContents
Chapter 1 Introduction ................................................................................................... 1
1.1. Research aims and objects ................................................................................... 1
1.2. Outline of the thesis ............................................................................................. 3
Chapter 2 Literature review ........................................................................................ 5
2.1. Background ................................................................................................................. 5
2.2. Controlled and slowed drug delivery systems .......................................................... 7
2.2.1. Controlled and slowed drug delivery ..................................................................... 7
2.2.2. Nanoparticle based drug delivery systems ............................................................. 9
2.2.3. Active targeting drug delivery ............................................................................. 10
2.2.4. Platforms for nanoparticles based drug delivery systems .................................... 12
2.3. Preparation and modification of HMSNs ............................................................... 19
2.3.1. Architectural features of MSNs ........................................................................... 19
2.3.2. Fabrication of HMSNs ......................................................................................... 20
2.3.3. Modification of MSNs ......................................................................................... 24
2.3.4. Qualitative characterization of HMSNs ............................................................... 28
2.3.5. MSNs’ pharmacokinetics ..................................................................................... 33
2.4. Drug loading and release .......................................................................................... 34
2.5. Drug resistant phenomenon ..................................................................................... 37
Chapter 3 Materials and methodology ...................................................................... 41
3.1 Introduction ............................................................................................................... 41
3.2 Materials .................................................................................................................... 41
3.3 Preparation of versatile HMSNs ............................................................................. 42
3.3.1 Formation of Eudragit S-100 nanoparticles ......................................................... 42
3.3.2 Fabrication of HMSNs ......................................................................................... 42
3.3.3 Control of particle size ......................................................................................... 43
3.3.4 Control of shell thickness .................................................................................... 43
3.3.5 Pore expansion of HMSNs .................................................................................. 44
3.4 Modifications of HMSNs .......................................................................................... 44
3.4.1 Amine-, methyl- and cyano-functionalization ..................................................... 44
3.4.2 Carboxyl-modification (HMSN-COOH) ............................................................. 45
3.4.3 FITC-modification (FITC-HMSNs) .................................................................... 45
3.4.4 EGF labelling (HMSN-EGF) ............................................................................... 45
3.5 5-FU loading and release study ................................................................................ 46
3.5.1 Measurement of 5-FU oil : water partition coefficients ....................................... 46
X
3.5.2 5-FU loading ........................................................................................................ 46
3.5.3 5-FU release ......................................................................................................... 47
3.6 Cell experiments ........................................................................................................ 48
3.6.1 Cell culture ........................................................................................................... 48
3.6.2 Cellular uptake of HMSNs and EGF-HMSNs in colorectal cancer cells ............ 48
3.7 Characterizations ...................................................................................................... 49
3.7.1 Electron microscopy ............................................................................................ 49
3.7.2 Nitrogen absorption and desorption experiments ................................................ 49
3.7.3 Small angle X-Ray diffraction (SAXRD) ............................................................ 49
3.7.4 Thermo-gravimetric analysis (TGA) ................................................................... 49
3.7.5 Fourier transform infrared spectrometer (FTIR) .................................................. 50
3.7.6 Time of flight secondary ion mass spectrometry (ToF-SIMS) ............................ 50
3.7.7 Nuclear magnetic resonance (NMR) ................................................................... 51
3.7.8 Dynamic light scattering (DLS) ........................................................................... 51
3.7.9 Confocal microscopy ........................................................................................... 51
Chapter 4 Synthesis of tailored HMSNs for nanotheranostics ................................. 52
4.1 Introduction ............................................................................................................... 52
4.2 Synthesis of HMSNs with designed properties for drug delivery systems ........... 54
4.2.1 Preparation and characterization of HMSNs ....................................................... 54
4.2.2 Innovations of Eudragit-assisted HMSNs fabrication ......................................... 60
4.2.3 Mechanism of Eudragit-assisted HMSNs fabrication .......................................... 63
4.2.4 5-FU encapsulation and the in vitro release ......................................................... 69
4.3 Pore size tuning for controlled release .................................................................... 71
4.3.1 BET analysis of pore size .................................................................................... 73
4.3.2 PALS analysis of pore size ................................................................................. 74
4.3.3 Controlled and sustained release of HMSNs with different pore size ................ 76
4.4 Cytotoxicity and non-specific cellular uptake of HMSNs ..................................... 80
4.5 Conclusions ................................................................................................................ 83
Chapter 5 Modification of HMSNs for targeting delivery ...................................... 84
5.1 Introduction ............................................................................................................... 84
5.2 A facile, one step strategy for EGF labelling .......................................................... 85
5.2.1 Synthesis of amine functionalized HMSNs (HMSNs-NH2) ............................... 86
5.2.2 Preparation of EGF grafted HMSNs and characterization of EGF labelling ........... ............................................................................................................................. 92
5.2.3 The effect of EGF concentration on surface chemistries of the HMSNs ............. 95
5.2.4 Analysis on immonium ions distinguishing samples with different surface chemistry .............................................................................................................. 99
XI
5.3 Comparison of newly developed grafting method with other methods.............. 101
5.3.1 Superior EGF grafting efficiency by the new grafting method ......................... 101
5.3.2 Increase EGF labels on HMSNs for enhanced targeting efficiency .................. 103
5.4 Effectively target colorectal cancer cells by EGF grafted HMSNs ..................... 106
5.4.1 Effective targeting effect of HMSN-EGF to EGFR positive CRC cells ............ 106
5.4.2 Reduced targeting effect of HMSN-EGF by pretreating SW480 cells with free EGF .................................................................................................................... 109
5.4.3 Non-targeting effect of HMSN-EGF to EGFR negative cell line SW620 ......... 110
5.5 Conclusions .............................................................................................................. 111
Chapter 6 Functionalization of HMSNs for favourite 5-FU loading and siRNA encapsulation .......................................................................................... 113
6.1 Introduction ............................................................................................................. 113
6.2 Enhanced 5-FU loading capacity of HMSNs by precise functionalization ........ 114
6.2.1 Study of 5-FU partition coefficient .................................................................... 114
6.2.2 Preparation and characterization of functionalized HMSNs .............................. 117
6.2.3 Regulate 5-FU loading capacity of HMSNs by modifications .......................... 123
6.3 Control of 5-FU loading and release ..................................................................... 124
6.3.1 The effect of pH on drug loading and release .................................................... 124
6.3.2 The effect of particle size and shell thickness on drug loading and release ...... 126
6.3.3 The effect of EGF on HMSNs’ drug loading and release behaviours ............... 129
6.4 SiRNA encapsulation based on HMSNs ................................................................ 131
6.4.1 Increased surface charge of HMSNs for SiRNA loading .................................. 131
6.4.2 Pore size expansion of HMSNs for efficient SiRNA encapsulation .................. 135
6.5 Conclusions .............................................................................................................. 142
Chapter 7 Conclusions and future works ................................................................. 144
7.1 Overview of current work ...................................................................................... 144
7.2 Limitations and future work .................................................................................. 148
7.2.1 Targeting efficiency of EGF-HMSNs ................................................................ 148
7.2.2 siRNA release and knockdown efficiency ......................................................... 149
7.2.3 Preparation multi-anticancer drug loaded HMSNs ............................................ 149
7.2.4 Animal experiences............................................................................................ 151
Bibliography ................................................................................................................. 152
Publication List ............................................................................................................. 175
XII
List of Table Table 2. 1 Methods employed to introduce functional groups ....................................... 25
Table 4. 1 Proton diffusion coefficients (D) of Eudragit and Eudragit/TX100 mixture . 67
Table 4. 2 Physicochemical Properties of HMSNs ......................................................... 74
Table 4. 3 PALS data for HMSNs .................................................................................. 75
Table 4. 4 Loading capacity and other physical parameters of HMSNs ......................... 77
Table 4. 5 Values of diffusional exponent and corresponding drug release mechanism 79
Table 5. 1 Positive fragment ions (immonium ions) used in PCA ................................. 97
Table 6. 1 Partition coefficient of 5-FU and relevant parameters ................................. 115
Table 6. 2 Structure parameters of functionalized and non-functionalized HMSNs .... 120
Table 6. 3 Loading capacity of functionalized and non-functionalized HMSNs .......... 123
Table 6. 4 Parameters of HMSNs and HMSN-NH2 ...................................................... 132
Table 6. 5 siRNA encapsulation efficiency................................................................... 135
Table 6. 6 Structure parameter of HMSNs-S, HMSN-M and HMSNs-L ..................... 138
XIII
List of Figure
Figure 2. 1 Tine line showing advancements made in CRC treatment ............................. 5
Figure 2. 2 History of adjuvant therapy of colon cancer................................................... 6
Figure 2. 3 Drug’s pharmacokinetic fates of different administration modes .................. 7
Figure 2. 4 Schematic structure of EGFR ....................................................................... 10
Figure 2. 5 EGFR signalling pathway ............................................................................. 11
Figure 2. 6 Multifunctional carbon nanotubes in cancer therapy................................... 14
Figure 2. 7 Preparation of modified SWCNTs............................................................... 15
Figure 2. 8 Mesoporous silica nanoparticles as a platform for drug delivery ................. 16
Figure 2. 9 Transmission electron microscopy of mesoporous silica nanoparticles ....... 16
Figure 2. 10 Schematic representation of the drug delivery system by using mesoporous
silica nanoparticles .......................................................................................................... 18
Figure 2. 11 Different particle morphologies of MSN. ................................................... 19
Figure 2. 12 The synthesis process of HMSNs ............................................................... 21
Figure 2. 13 Schematic illustration of the formation process of HMSs .......................... 22
Figure 2. 14 Scheme of the synthetic procedure of HMSs ............................................. 23
Figure 2. 15 Formation (left) and microscopic structure (right) of hollow/rattle-type
MSNs .............................................................................................................................. 23
Figure 2. 16 Introduction of functional groups in different regions of MSN. ................ 24
Figure 2. 17 Functionalized hollow mesoporous silica nanoparticles ............................ 26
Figure 2. 18 TEM morphology of HMSNs ..................................................................... 28
Figure 2. 19 XRD patterns of mesoporous silica nanoparticles synthesized at different
pH .................................................................................................................................... 29
XIV
Figure 2. 20 Six categories of gas adsorption isotherms for porous materials with
different pore size. ........................................................................................................... 30
Figure 2. 21 Adsorption isotherms of nitrogen in mesoporous silica nanoparticles of
different pore sizes.. ........................................................................................................ 31
Figure 2. 22 Pore geometry ............................................................................................. 32
Figure 2. 23 Effect of pore geometry on shapes of hysteresis loop ................................ 32
Figure 2. 24 TEM images of MSNs (black dots) endocytosed by human cervical cancer
cells ................................................................................................................................. 34
Figure 2. 25 HMSNs based drug delivery system for simultaneous joint lubrication and
treatment .......................................................................................................................... 35
Figure 2. 26 HMSNs based drug delivery system for high-intensity focused ultrasound
(HIFU)-mediated intravenous drug delivery. .................................................................. 36
Figure 2. 27 Rational design of a multifunctional micellar nanomedicine for targeted co-
delivery of siRNA and DOX to overcome multidrug resistance .................................... 38
Figure 2. 28 System for delivery of doxorubicin (DOX)/siRNA combination based on
quantum dots (QDs) modified with β-cyclodextrin ........................................................ 40
Figure 4. 1 Fabrication possess of HMSNs ..................................................................... 54
Figure 4. 2 a, b & c) SEM images of HMSNs and d, e & f) TEM image of HMSNs
displaying hollow interior ............................................................................................... 56
Figure 4. 3 DLS curve of HMSNs .................................................................................. 57
Figure 4. 4 Small angle XRD pattern of HMSNs ........................................................... 58
Figure 4. 5 Nitrogen adsorption-desorption isotherm (A) and the BJH pore size
distribution of HMSNs (B) ............................................................................................. 59
XV
Figure 4. 6 TEM images (a) and SEM images (b & c) of HMSNs-S and TEM images
(d) of HMSNs-C and SEM images (e & f) ..................................................................... 61
Figure 4. 7 Zeta potential of HMSNs-S and HMSNs-N ................................................. 62
Figure 4. 8 Stability study of HMSNs-C and HMSNs-S ................................................ 62
Figure 4. 9 Formation schematics and of TX100/Eudragit S100 composite micelles (a)
and proposed reactions between TX100 and Eudragit S100 .......................................... 64
Figure 4. 10 DLS of Eudragit S100 nanoparticles, TX100 and their mixture ................ 65
Figure 4. 11 FTIR spectra of Eudragit S100 nanoparticles, TX100 and the mixture of
Eudragit and TX100 ........................................................................................................ 66
Figure 4. 12 1H NMR spectra of Eudragit S100 nanoparticles (upper) and
Eudragit/TX100 mixture (bottom); Inserts are TX100 molecular structure. .................. 67
Figure 4. 13 SEM images (a) and TEM image (b) of non-Eudragit assisted MSNs....... 68
Figure 4. 14 (a):5-FU loading profile and (b): 5-FU release profile of HMSNs ............ 70
Figure 4. 15 Three stages of the slow release kinetic model .......................................... 71
Figure 4. 16 Representative schematic illustration of controlling drug release rate by
tuning the pore size of HMSNs ....................................................................................... 72
Figure 4. 17 Adsorption-desorption isotherms (A) and corresponding pore size
distribution (B) of HMSNs ............................................................................................. 73
Figure 4. 18 The relationship between porosity and τn ................................................... 75
Figure 4. 19 In vitro release profiles of HMSNs with different pore size in PBS with a
pH of 5.0 ......................................................................................................................... 77
Figure 4. 20 In vitro release kinetics of HMSNs with different pore size in PBS with a
pH of 5.0 ......................................................................................................................... 79
Figure 4. 21 Cytotoxicity of HMSNs on SW 480 cells at different concentrations....... 80
XVI
Figure 4. 22 Cell uptake of FITC-HMSNs with varying time and concentration by
SW480 cells. ................................................................................................................... 82
Figure 5. 1. a, b & c) SEM images of HMSNs and d, e & f) TEM image of HMSNs
displaying hollow interior ............................................................................................... 88
Figure 5. 2 A) Nitrogen adsorption-desorption isotherm; B) Small angle XRD pattern of
HMSNs; C) BJH pore size distribution. .......................................................................... 89
Figure 5. 3 FTIR spectra of HMSNs-NH2 and HMSNs ................................................. 90
Figure 5. 4 TGA thermograms of HMSNs & HMSNs-NH2 (a) and 29Si single pulse
excitation (SPE) spectrum of HMSNs-NH2 .................................................................... 91
Figure 5. 5 Schematic diagram showing the process of EGF grafting onto HMSNs ..... 92
Figure 5. 6 Positive survey mass spectra for: (A) HMSNs surface; (B) HMSNs-NH2
surfaces; (C) EGF; (D) HMSNs-NH2-EGF..................................................................... 94
Figure 5. 7 Positive survey mass spectra of HMSN-NH2-EGF surface using different
EGF concentration .......................................................................................................... 95
Figure 5. 8 Plot of ratios of the intensities of the 28 m/z (from HMSNs) to 59 m/z, 70
m/z and 58 m/z (from EGF) ............................................................................................ 96
Figure 5. 9 Scores on PC1 and PC2 (a) and corresponding loading plots on PC1 (b) and
PC2 (c) of immonium ions from static positive spectra of HMSNs-NH2-EGF prepared
with EGF concentration .................................................................................................. 98
Figure 5. 10 Comparison of characteristic immonium ions intensities between samples
....................................................................................................................................... 100
Figure 5. 11 Routes of EGF lebelling based on carbodiimide chemistry ..................... 102
Figure 5. 12 SEM images of HMSN-COOH-EGF (a) and HMSN-NH2-EGF (b)........ 103
Figure 5. 13 Positive survey mass spectra .................................................................. 105
XVII
Figure 5. 14 Cellular uptake of EGF grafted and plain HMSNs in SW480 cells ......... 106
Figure 5. 15 Confocal microscopic images of SW480 cells after incubating with FITC
labelled HMSN-EGF (a & b) or HMSNs (c & d) with a concentration of 5mg/ml and an
incubation time of 2 hours ............................................................................................ 108
Figure 5. 16. The cellular uptake rate of HMSN-EGF and HMSNs in SW 480 cells with
or without 5 μM of free EGF pretreatment ................................................................... 109
Figure 5. 17. Confocal microscope images of EGF-HMSNs and HMSNs in SW 480
cells. .............................................................................................................................. 110
Figure 5. 18. The mean fluorescence intensity in the SW 620 cells treated with different
concentrations of HMSNs or EGF-HMSNs for 2 hours. .............................................. 111
Figure 6. 1 percentages of 5-FU dissolved in each phase with varying octanol to water
ratio ............................................................................................................................... 116
Figure 6. 2 TEM images of HMSNs (a), HMSN-NH2 (b), HMSN-COOH (c), HMSN-
CN (d) and HMSN-CH3 (e)........................................................................................... 117
Figure 6. 3 N2 adsorption-desorption isotherms (a) and the corresponding pore size
distributions (b) of plain and functionalized HMSNs ................................................... 119
Figure 6. 4 FTIR spectra of plain and functionalized HMSNs and plain HMSNs ....... 120
Figure 6. 5 TGA curves of functionalized HMSNs and plain HMSNs ........................ 122
Figure 6. 6 5-FU loading capacity of HMSNs and HMSN-NH2 at different pH values
(a) and UV spectra of remaining 5-FU after 24 hours loading by HMSN-NH2 at
different pH values (b) .................................................................................................. 125
Figure 6. 7 Two possible 5-FU anions (AN1 and AN3) and dianion ........................... 126
Figure 6. 8 SEM images of HMSNs with different particle size: 100 nm (a & d); 200 nm
(b & e) and 300 nm (c & f) ........................................................................................... 127
XVIII
Figure 6. 9 Loading capacities (a) and release profiles (b) at pH 5.5 of HMSNs with
varying particle size ...................................................................................................... 127
Figure 6. 10 SEM images of HMSNs with different shell thickness: 10 nm (a & d); 15
nm (b & e) and 30 nm (c & f) ....................................................................................... 128
Figure 6. 11 Loading capacities (a) and release profiles (b) at pH 5.5 of HMSNs with
varying shell thickness .................................................................................................. 129
Figure 6. 12 SEM images (a & b) and TEM image (c) of HMSN-EGF ....................... 130
Figure 6. 13 Loading (a, at pH 8.0) and release (b, at pH 5.5) profiles of HMSNs with
and without EGF attachments ....................................................................................... 130
Figure 6. 14 SEM image (a) and zeta potential (b) of HMSN-NH2 .............................. 132
Figure 6. 15 siRNA encapsulation efficiency of HMSN-NH2. Various weight ratio of
particles/siRNA from 5 to 200 were applied. ............................................................... 133
Figure 6. 16 Zeta potential of HMSNs, HMSN-NH2-1, HMSN-NH2-2 ....................... 134
Figure 6. 17 siRNA encapsulation efficiency of HMSN-NH2-2 (a) and HMSN-NH2-3
(b). Various weight ratio of particles/siRNA from 5 to 200 were applied. ................... 134
Figure 6. 18 SEM and TEM images of HMSNs-S (a & b), HMSNs-M (c & d) and
HMSNs-L (e & f) .......................................................................................................... 137
Figure 6. 19 Zeta potential (a) and TGA (b) of aminated HMSNs, HMSNs-S, ........... 139
Figure 6. 20 siRNA encapsulation efficiency of HMSNs-S, HMSNs-M and HMSNs-L.
Various weight ratio of particles/siRNA from 5 to 200 were applied. ......................... 141
Figure 6. 21 Proposed siRNA encapsulation of HMSNs with different pore sizes ...... 142
1
Chapter 1 Introduction
1.1. Research aims and objects
Colorectal cancer (CRC) remains a significant health burden in the world. For its
treatment, the anticancer drug of fluorinated pyrimidine 5-fluorouracil (5-FU) as the
backbone stands alone in CRC therapy [1]. Nearly every chemotherapy regimen for CRC
includes 5-FU. However, owing to its inadequate intratumor concentrations and severe
side effects, CRC patients pay great prices when receive the treatments. Obviously, the
therapeutic utility of 5-FU on CRC treatment has been hindered by lacking a safe delivery
system. The challenge is to extend its circulating half-life and to control its distribution
in tissues.
One way to address this challenge is to engineer a system where a constant dose of 5-FU
is delivered directly to colorectal cancer cells over an extended period [2, 3]. After
reviewing the relationship between drug’s pharmacokinetic fates and administration
modes, an optimally efficacious approach toward achieving the goal of continuous,
nontoxic bioactivity would involve controlled and sustained drug delivery systems based
on nanocarriers labelled with cell targeted molecules.
When compared to polymer and carbon nanotubes which have been widely investigated
as carriers in drug delivery systems, hollow mesoporous silica nanoparticles (HMSNs)
have the potential to serve as a versatile drug nanocarrier for smart drug delivery. For
example, HMSNs can level off the drug concentration at the targeted area as the drug
molecules released from the ordered pores by a tunable diffusion. This will reduce drug
dosage and side effects. Zero premature release before reaching the targeted sites can be
attained when using functionalized HMSNs as carriers. Furthermore, HMSNs have large
pore volume and surface area and are an ideal platform for cancer treatment because of
their higher drug loading [4].
2
However, owning to the lack of structure and morphology controlling, applications of
HMSNs for drug delivery are still limited. Thus, technology to fabricate, functionalize
and modify HMSNs with expected shapes, sizes and defined properties is vitally essential.
Additionally, as pointed out by Argyo [5] in a very recent review, cell type specificity is
a challenge that still needs to be overcome by using these types of materials as
nanoplatforms for drug delivery systems. This is because extensive studies on
nanoparticle based targeted drug delivery systems were relied on the enhanced-
permeation end-retention effect (EPR) [6, 7] of cancer cells. However, this effect is
limited just to vascularized tumours and is usually not sufficient for a complete
eradication of the cancer [8]. Base on the comprehensive literature review, we
hypothesized that active targeting of cancer cells by epidermal growth factor (EGF) on
the surface of the nanoparticles [9-11] could be an efficient strategy to provide an
enhanced accumulation of the nanoparticles in targeted cell population (CRC cells) and
thus represent a powerful tool to improve the efficacy of the cancer treatment.
Therefore the main objectives of this project are:
1. Engineering new ways to prepare HMSNs with designed properties and tailored
structure for targeted drug delivery and investigating the biosafety of these
synthesized HMSNs.
2. Detail studies on chemical modifications of HMSNs to achieve high 5-FU loading
and sustained release and the fabrication of HMSNs with controllable drug release
rate. On the other hand, the 5-FU behaviours at different pHs were different, which
is another factor that needs to be investigated to obtain a high 5-FU encapsulation
rate by HMSNs.
3. Grafting of targeting ligands on HMSNs to attain a high specificity of HMSNs for
further application. The relationship between grafted biological attachments and
properties as well as biological behaviours need to be fully studied.
4. In vitro studies with colorectal cancer cells to assess the biological activities of
3
HMSNs in biological environment are require to demonstrate HMSNs as an effective
vehicle for successful intracellular delivery of 5-FU for colorectal cancer therapy.
1.2. Outline of the thesis
This thesis consists of 7 chapters as follows:
Chapter 2 provides a comprehensive review of the challenges existing in colorectal cancer
treatments, the background information and recent studies in the areas of nanoparticle
based drug delivery systems, nanoplatforms used for sustained release, and biomarkers
identified for targeted drug delivery. Throughout this chapter, gaps in 5-FU delivery
specifically for colorectal cancer therapy are identified. In addition, feasible methods
proposed to fill these gaps are introduced and discussed.
Chapter 3 outlines all the materials and experimental procedures used in this thesis.
Techniques for the fabrication and modification of HMSNs as well as the strategies
applied for testing the drug loading and release properties, for introducing biological
targeting labels and for in vitro experiments are illustrated. Methods for the
characterization of HMSNs such as SEM, TEM, XRD, FTIR and TGA etc. are described
in details.
Chapter 4 focuses on the synthesis of HMSNs and the fabrication mechanism.
Characterizations of structure, morphology and biocompatibility are also studied in this
chapter.
Chapter 5 aims at the study of the targeting function of current drug delivery systems.
Extremely high specificity of HMSNs for targeting delivery is generated by introducing
a bio-ligand, EGF, onto HMSNs. Especially, the characterization of EGF attachments and
EGF concentration on the surface of nanoplatforms were fully investigated in this chapter.
4
Chapter 6 explores ways to effectively increase 5-FU and siRNA loading capacity of
HMSNs. In addition, strategies regarding better control of drug release kinetics from
HMSNs are investigated.
Chapter 7 summarizes the major findings and contributions of this project and identifies
some limitations in the current study. Future research directions in this area are also
proposed in this chapter.
5
Chapter 2 Literature review
2.1. Background
Colorectal cancer (CRC) manifesting as cancerous growths in the colon and rectum, is
the fourth most commonly diagnosed malignancy in the world. Over the past 25 years,
significant advances have been made in the treatment of CRC (Figure 2.1), leading to a
steady improvement of the treatment of this illness [12]. However, it still remains a
significant health burden in the world with over 1,000,000 cases and a disease-specific
mortality of nearly 33% even in developed countries [1]. Obviously, the treatment of CRC
is still full of challenges and many of them are still unsolved, such as toxic effects of
chemotherapy [13] and tumours developing resistance to drugs [14].
Figure 2. 1 Tine line showing advancements made in CRC treatment [12]
Among drugs for CRC treatment, the fluorinated pyrimidine 5-fluorouracil (5-FU) as the
backbone stands alone in CRC therapy (Figure 2.2). Fluorouracil inhibits thymidylate
synthase which is the rate-limiting enzyme for pyrimidine nucleotide synthesis [15].
Nearly all chemotherapy regimens shown to be effective for CRC treatments incorporate
5-FU [1, 16], highlighting its crucial and first-line role in treating colorectal cancer.
6
Figure 2. 2 History of adjuvant therapy of colon cancer [17]
Many 5-FU intravenous administration schedules are in clinical use, but the efficacy is
limited, because intravenous 5-FU injection is hard to reach optimal dose effectiveness at
targeted sites due to 5-FU’s very short half-life and quick metabolism in plasma
circulation [18]. Moreover, intravenous administration of 5-FU has been shown to cause
severe side effects [19] because it induces high drug concentration above the maximum
tolerable drug concentration in the plasma [20, 21]. Obviously, the therapeutic effect of
5-FU on cancer treatment has been hindered by the lack of a safe delivery system.
In summary, 5-FU plays an important role in CRC chemotherapy. However, the more
extensive use of 5-FU is limited by its inadequate intratumor concentrations and severe
side effects which may result from its short half-life and non-specific systemic organ
distribution. Therefore, new strategies for effective and safe 5-FU delivery is urgently
needed for colorectal cancer therapy [22].
7
2.2. Controlled and slowed drug delivery systems
2.2.1. Controlled and slowed drug delivery
One way to overcome the limitations associated with 5-FU for CRC chemotherapy is to
engineer a systems where a constant dose of 5-FU is delivered directly to colorectal
cancer cells over an extended period. With the development of pharmaceutics, a
medicinal agent’s therapeutic effect regarding its pharmacokinetic fate in vitro and in vivo
can be optimized by the utility of precise drug administrations [23]. Figure 2.3 depicts the
relationship between drug’s serum concentration and administration modes. Obviously,
an optimally efficacious approach toward achieving the goal of continuous, nontoxic
bioactivity would involve controlled and sustained administrations which provide a
considerably extended therapeutic delivery of the drugs.
Figure 2. 3 Drug’s pharmacokinetic fates of different administration modes [23]
8
Engineering sequential release dosage forms is beneficial for optimal therapy in terms of
safety, efficacy and patient compliance. Ideally, a controlled release dosage form will
maintain a therapeutic concentration of the drug in the blood throughout the dosing
interval [24, 25]. With this aim, techniques for the preparation of sustained release
formulations were investigated including preparation of prodrug [26, 27], incorporation
of drugs into carriers and coating with special materials [28, 29]. In addition, liquisolid
technique, referring to the control of the dissolution rate of drugs, is one of the best and
most successful methods because of its simplicity and low cost [30]. Particularly, a
“liquisolid system” involved in blending drug solutions with selected coating materials
or carriers before drying to form compactible powder mixtures [31]. It has been used to
produce sustained release formulations of water-soluble drugs such as propranolol
hydrochloride [24] and the release of drug from these formulations had been proven to
follow zero-order release kinetics [30-32].
At present, 5-FU sequential release pellets have been prepared [22, 33, 34] and evidenced
to exhibit inhibition for tumour growth and proliferation to some extent [35]. This
augmentative therapeutic efficiency is due to the extension of 5-FU plasma half-life by
the sequential release [34]. However, when compared to free drug, these pellets showed
limited advantages in minimising side effects of 5-FU because only limited control over
the release kinetics is possible [36], where an initially rapid release is presented and
followed by an inability to maintain a sufficient drug concentration for an extended period
of time [37]. This limitation is due, in part, to the poor water-soluble nature of 5-FU and
its low molecular weight (130.08). An alternative approach has been reported to control
the release of 5-FU by attaching the drug covalently to the network of selected materials
or carriers [38] via degradable linkages. Nevertheless, the dependence on temperature,
pressure or pH [39] and susceptibility to degradation by esterases [40] often present a
challenge, which finally leads to deviations from predicted release kinetics.
While sustain release formulations offer promising opportunities to enhance anti-tumour
efficacy of 5-FU with low systemic toxicity, the unpredictable tumour environment and
susceptibility of drugs to collapse on the way to tumour site make it difficult to precisely
control the drug release kinetics. Therefore, an alternative or complementary strategy is
9
required to be developed to overcome these shortcomings, where a biomaterial platform
is applied to enable sustainable, predictable, and tunable 5-FU release.
2.2.2. Nanoparticle based drug delivery systems
Recently, nanomedicine has been demonstrated to significantly enhance the efficacy of
cancer therapy because nanoparticles maintain a prolonged effective drug concentration
which renders cancer cells expose sustainedly to the drug [20]. The feasibility of
nanoparticle-based anticancer therapy has been proven in clinical use [6]. Owning to their
small size and the characteristic features of tumour biology, like the leaky vasculature
and the impair lymphatic system, nanoparticles can reach certain solid tumours and
release biologically active cargos by diffusion into the tumour tissues. This method is
called passive targeting approach [6]. Previously, passive targeting nanoparticles had
been used as platforms to successfully control and sustain drug release for therapy. For
example, it was shown that selective anticancer effect had been enhanced by passive
targeting self-assembled nanoparticles [41]. Similarly, silica nanoparticles as drug
carriers have effectively demonstrated heart-targeting profile by passive targeting
approach [42].
These inspired studies could help to engineer a 5-FU delivery system to the colon basing
on nanoparticles with small enough size to improve the transportation efficiency of 5-FU
to cancer cells. Although there have been some preliminary reports [43], the application
of this passive targeting approach in colon targeted drug delivery systems is still hindered.
Because of the random nature of this approach, it is hard to control the drug release
kinetics and this in turn decreases the efficacy of localizing nanomedicines at cancer cells
in vivo. As we know, our body is so perfectly structured that some physiological drug
barriers in our body are existing to protect organs from external toxins, like the
gastrointestinal barrier [44]. Moreover, cancer cells are smart enough that their survival
is achieved by presenting barriers to fight with anti-cancer drugs, such as drug resistant
genes and transport proteins for expelling drugs out of cancer cells. Unfortunately, as
colon is the distal segment of the gastrointestinal tract, when delivering 5-FU to the colon,
nanocarriers will encounter barriers motioned above on their route to the targeted site.
10
Therefore, it would be desirable to combine the rational design of nanocarriers with novel
nanoscale targeting molecules.
2.2.3. Active targeting drug delivery
To address this challenge, rather than passive delivery, it is essential to find an actively
targeting region on the surface of colorectal cancer cells with higher specificity and
binding affinity. Therefore, understanding the pathology of colorectal cancer becomes
imperative.
In recent years, overexpression of growth factors and receptors has been implicated in
tumorigenesis and tumour progression of CRC [45, 46]. Dysfunctions in signalling
pathways have also been demonstrated to lead to proliferation, survival and migration of
cancers [47, 48]. The overexpression of epidermal growth factor receptor (EGFR) has
been documented in 30 – 90% of cases of advanced CRC, which has shown to involve in
numerous cellular responses including proliferation and metastatic spread [49]. It thus
becomes vital to understand the EGFR signalling pathway in CRC during neoplastic
transformation as it may enable us to programme the nanocarriers so they can actively
bind to colorectal cancer cells.
Figure 2. 4 Schematic structure of EGFR
EGFR is an endogenous cell surface glycoprotein with a molecular weight of 170 KDa
[11]. It comprises three domains that include a 622 amino acid extracellular domain, a 23
11
amino acid hydrophobic transmembrane domain, and an intracellular tyrosine Kinase
domain with several phosphorylation sites (Figure 2.4) [49].
The EGFR signalling pathway consists of 4 sub-pathways: PI3K-mTOR pathway, JAK-
STAT pathway, PLC signalling pathway and RAS-ERK signalling pathway. On ligand
binding (signal activation), EGFR homo-dimerizes activate the intracellular kinase
domain. Then, importins (adopter proteins, like PI3K, JAK and PLC) bind with
transcription factors (like STAT and ERK), leading to the nucleus accumulation of EGFR.
As a result, tumour proliferation, metastasis and survival happen (Figure 2.5) [1].
Figure 2. 5 EGFR signalling pathway
(JAK, Janus kinase; PLC, phospholipase C; STAT, signal transducer and activator of transcription.)
With its association with tumour survival and proliferation, EGFR has the potential to
serve as an actively targeting region of colon targeted drug delivery system. Epidermal
growth factor (EGF), hepatin-binding EGF (HB-EGF), betacellulin, transforming
growth-factor-a (TGF-a), epiregulin and amphiregulin are six known endogenous ligands
of EGFR. Although any of these ligands can be used as a targeting agent for EGFR, EGF
with a good compromise between the binding affinity and the size of the molecule (EGF,
6.1 kDa) is one of the most commonly detected factors in humans [8]. It is hypothesized
12
that introducing EGF as a ligand to EGFR on the surface of nanocarriers will exhibit
specific delivery of 5-FU into colorectal cancer cells, and thus enhance the resultant anti-
cancer activity. To the best of our knowledge, at present there is not EGFR targeting 5-
FU delivery system being exploited specifically for colorectal cancer.
2.2.4. Platforms for nanoparticles based drug delivery systems
Before designing a colon targeted drug delivery system, several questions should be
addressed. How to achieve a tight association between the targeting molecules and the
surface of nanocarrier is one of key questions. Also, as for colon targeted drug delivery
system, biodegradability and toxicological harmlessness is another point that should be
considered beforehand because the materials to be used for drug delivery may accumulate
in the human body on prolonged use. As we mentioned above, nanoparticles exhibit slow-
release profile, but different functions can be observed even for the same drug delivery
system when using different nanocarriers. In this regard, it would therefore be desirable
to study materials used for drug delivery and find out the one that can achieve the expected
functions greatly.
In order to achieve an ideal colon targeted drug delivery system, a variety of materials
have been exploited to meet different therapeutic needs, many with great potential for
topical delivery. Reports on materials with controlled release properties for topical
delivery include: cellobiose-derived monomer [50], carbon nanotubes [51], metals [52,
53], chitosan [54, 55], pectins [54, 56], guar gum [57, 58], inulin, and dextran. These
materials can be divided into steroids, e.g. polymer, carbon nanotubes and inorganic
nanoparticles.
2.2.4.1.Polymer
Polymers are a kind of widely used materials for colon targeted drug delivery system due
to their inherent properties. For example, excellent stability and safety, less toxicity and
gel forming make the utilisation of polymers for colon specific delivery wide and
intensive. Moreover they are more cost-effective, as they are abundant in various
13
structures and exhibit diversified properties.
A large number of studies have been done in exploiting polymers as potential carriers for
colon specific delivery including chitosan [59], pectin [60, 61], sugar gum [58] and
dendrimer [62]. The effects of formulation parameters on drug release pattern are studied
by Das, Chaudhury et al. [54]. Elkhodairy, Barakat et al. [63] used low molecular weight
chitosan to prepare a new guar-based colon delivery formulation. Hiorth, Skoien et al.
[64] revealed that a high degree of swelling of chitosan-containing pellets may boost the
drug release. pH-responsive fluorinated dendrimer-based particulates were prepared by
Criscione, Le et al. [65] which exhibit controlled release of agents. Moreover, these dense
particulates are self-assembled and possess a high density of fluorine spins that can be
easily detected in vivo.
However, the issue associated with the utilisation of polymers is their high water
solubility. Polymers are easily assimilated by microorganisms or degraded by enzymes
on its way to the targeting site such as the human colon, resulting in breaking down of the
molecules back bone [66]. As a consequence, the molecular weight reduces as the
polymer undergoes a further degradation leading to the loss of mechanical strength,
unable to hold the targeting molecules (such as EGF) and drug (like 5-FU) any longer.
This is how the burst release and premature release happen.
2.2.4.2.Carbon nanotube
Carbon nanotubes (CNTs), regarded as one of the most promising nanomaterials, possess
the ability to deliver drug into living systems [67]. Owning to their unique
physicochemical properties like their inherent stability, novel structure, high aspect ratio
as well as high electrical and thermal conductivity, CNTs can not only detect the
cancerous cell, but also transport biologically active cargo into these living cells [68].
During the last several years, CNTs have been widely investigated in the field of cancer
treatment modality [69], containing gene [70] and photodynamic therapy [71], lymphatic
targeted chemotherapy [72], drug delivery [73] and thermal therapy [74] (Figure 2.6).
Research has found that CNTs present great potential to meet future drug delivery
14
challenges [75]. For instance, when the titanium alloys were implanted, the self-organized
TiO2 nanotubes are found to be a promising drug delivery system [76]. Again,
molecularly imprinted nanotubes are found to possess the potential of controlled drug
release [77].
Figure 2. 6 Multifunctional carbon nanotubes in cancer therapy [51]
Novel SWNT-based tumor-targeted drug delivery systems which are mainly made up of
functionalized SWNTs, tumor-targeting ligands, and anticancer drugs have gained
considerable interests [51]. By linking precise peptides or ligands on their surface,
functionalized CNTs can recognize cancer-specific receptors on the cell surface and cross
the mammalian cell membrane by endocytosis or other mechanisms to recognize cancer-
specific receptors on the cell surfaces. Finally, therapeutic drugs are released more safely
and effectively than untargeted ones for the purpose of targeted killing of cancer cells
[78]. A pH-triggered targeted drug delivery system based on modified single wall carbon
nanotubes is prepared by Zhang, Meng et al. (Figure 2.7). The loading efficiency and
release rate of the anticancer drug doxorubicin (DOX) can be controlled [79].
15
Figure 2. 7 Preparation of modified SWCNTs [79]
(SWCNTs: single wall carbon nanotubes; ALG: sodium alginate; CHI: chitosan; DOX: doxorubicin)
Inspired by these studies, it seems that CNTs could serve as a nanoplatform with specific
colon targeting function. However, the application of CNTs in biomedicine has been
concerned over their side effects and toxicity. Studies performed has found that the CNTs
toxicity not rely on the presence of metal impurities and their length [80] but the CNTs
themselves like physical form, concentration and functionalization degrees. In addition,
multi-walled carbon nanotubes could result in oxidative stress and cell death as they may
cause incomplete phagocytosis [81]. What is more, in the research performed by Ji, Liu
et al., correlation between toxicity of CNTs and the route of administration is suggested
[51]. Therefore, CNTs is not the most suitable nanoplatform due to its high toxicity.
2.2.4.3.Mesoporous silica nanoparticles
The synthesis and application of mesoporous silica nanoparticles (MSN) has attracted
great attention in the last two decades since the first ordered mesoporous silica molecular
sieves (the type of MCM-41) was synthesized by Kresge, Leonowicz et al. in 1992 [82].
The tunability of the structure, morphology, modification, functionalization, surface
properties, particle size and pore diameter are drawing increasing attention for the purpose
of their applications in the sectors of pharmaceutical [83], optics [84], environmental
protection [85] and catalysis [86]. Particularly, owning to their unique properties, MSNs
possess great potential to be applied in controlled release (Figure 2.8). In addition, in
order to develop the application of MSNs in biomedicine, research on the
16
biocompatibility [87], cellular uptake [88] and cytotoxicity [89] of MSNs has been
performed [90].
Figure 2. 8 Mesoporous silica nanoparticles as a platform for drug delivery
A=gatekeepers B=drugs C=stimuli-responsive linkers K=functional groups G=complexation with plasmid DNA [88]
Figure 2. 9 Transmission electron microscopy of mesoporous silica nanoparticles
The honeycomb-like structure of the nanoparticles is visualized with the parallel stripes (black arrow) and
the hexagonally packed light dots (red arrow) [88]
17
MSNs are synthesized inorganic nanoparticles with a honeycomb-like porous structure
(Figure 2.9), which is originated from the idea of extending the application of zeolites
[89]. When compared to most general organic materials like polysaccharides, liposomes
and micelles, MSNs possess many unique properties such as high surface area and pore
volume, ordered pores and sizes, tunable morphology [90]. Firstly, MSNs are widely used
in molecular sieving, ion exchange, adsorption [91] and separation. Soon, the well-
ordered sizes and pores extend their application to new fields. Specifically, MSNs were
used in sewage circulation purification by functionalization with mercaptopropyl groups.
Moreover MSNs are also used as templates for conductive polymers, or for controlled
polymerization processes [92]. There is also potential for MSNs to be applied for
macromolecule and micromolecule immobilization.
In particular, as there are mesopores (channels) on MSNs which are able to encapsulate
pharmaceutical cargoes, they are exploited to be used as carriers in order to attain the
controlled release in recent years. More importantly, more and more research proves that
MSNs with the diameter range of 100nm to 200nm can be taken by cells through the
endocytosis [93]. In addition, MSNs have significant advantages to be used in targeting
drug delivery systems [90]. For example, MSNs possess the capacity to level off the drug
concentration at the targeted area as the drug molecules are released from the ordered
pores or channels of MSNs by a tuneable diffusion. This gives rise to a reduction in
dosage and the preventions of side effects. Zero premature release before reaching the
targeted sites can be attained when using functioned MSNs as carriers in targeting drug
delivery systems. Furthermore, the large pore volume and large surface areas lead to a
higher drug loading [4]. Choi, Jaworski et al. functionalized MSNs for the purpose of
controlling drug release, and developed a drug delivery system using MSNs as carriers
based on the host-guest concept (Figure 2.10) [83] . Mamaeva and co-workers [94]
proposed MSNs as a platform for cancer treatments. In their study, MSNs based drug
delivery system can block the Notch signalling in cancer by controlling the delivery of γ-
secretase inhibitors to cancer cells.
18
Figure 2. 10 Schematic representation of the drug delivery system by using mesoporous
silica nanoparticles: (a) preparation of mesoporous silica nanoparticle, (b) immobilization
of 1 as gate molecule, (c) encapsulation of curcumin (2), (d) complex formation of 18-
crown-6 moiety with Cs+ ion and (e) release of curcumin (2) upon addition of K+ ion.
[83]
Most importantly, recent studies pointed out that MSNs are biocompatible, and particles
are degraded and excreted mainly through urine and faeces without being retained within
the organs [94]. Lu and co-workers [87] find that 95% degraded products of MSNs could
be safely eliminated out of body by renal excretion.
To summarize, great advancements are made in the synthesis of versatile drug
nanocarriers to date, including polymers, carbon nanotubes and MSNs [66]. However,
due to high toxicity of the nanotubes and premature leakages of the polymeric carriers
[61], the development of colon targeted drug delivery systems that can reduce the toxic
effect while augment therapeutic efficacy of 5-FU remains a significant challenge. From
the literature review, it can be found that MSN have been highlighted as compelling
carriers for anticancer drugs owning to their unique properties, such as high surface area,
ordered porous structure, and easily functionalized surface. More importantly, its
biocompatibility and physical and chemical stability make MSNs the potential drug
delivery vehicles that target colon cancer cells.
19
2.3. Preparation and modification of HMSNs
2.3.1. Architectural features of MSNs
During the MSN synthesis process, the silicate polymerization directed by templates can
be controlled in order to achieve different designed architectural features [95]. The
morphology of MSNs can be controlled in different ways, including varying templates,
controlling the synthesis pH and choosing additives to control the magnitude of the
interactions between the growing silica polymer and the assembled templates. Moreover,
the control of the rates of the silica source hydrolysis and condensation is important to
attain certain architectural features of MSNs. Sphere, rod, shuttle and lamel are the main
structures of MSNs (Figure 2.11).
Figure 2. 11 Different particle morphologies of MSN: (a) rod, (b) sphere, (c) lamel, (d)
shuttles, (e) core-shell and (d) hollow particles [4, 96].
Among the different MSNs, it is important to choose the right one that can achieve the
expected functions. For drug delivery, it was reported that the nanospheres exhibit a
higher efficiency of cell internalization of when compared with nanorods. This is based
on the positive relationship between the particle surface areas and the speed of cell
membrane enclosing a particle [88, 97]. Moreover, with the development of
nanomedicine, demands for nanoplatforms with specific performance such as large drug
20
loading gradually increase. To meet such new requirements, more and more new methods
are under investigation for developing new drug carriers. Recently, hollow MSNs
(HMSNs) have been studied as a new-generation drug platform for delivery systems
owning to their extraordinarily high loading capacity. Li and co-workers [98] exploited
HMSNs as drug carriers to increase the capacity of storing aspirin.
Based on these studies, it is proposed that hollow mesoporous silica nanospheres is an
ideal nanocarrier for colon targeting 5-FU delivery system as it exhibits spherical
morphology to enhance cell internalization and hollow cave in the middle to increase drug
loading. However, up to date, fabricating hollow MSNs with good stability and
dispersibility remains a main challenge owning to the complex effects of physicochemical
parameters on synthesis reactions. Thereby, developing new methods to fabricate hollow
MSNs is important for the further development of efficient drug delivery systems.
2.3.2. Fabrication of HMSNs
The formation of meso-structure materials is usually attained by the synthesis of liquid-
crystal-like arrays at low-temperature. The liquid-crystal-like arrays consist of precursors
and templating agents, typically an amphiphilic organic surfactant [99]. For hollow MSN,
the synthesis is usually achieved by using a dual template method. More specifically, a
hard or soft template (cave template) which is used for the generation of the interior cave
is dispersed in a neutral or charged surfactant (pore template). Then pore template
micelles are formed around the surface of cave templating agents. After that, silica matrix
coating on the core template is generated by adding silica source agents. Finally, both
core and cave templates are removed by thermal calcination or solvent extraction to obtain
hollow MSNs with hollow interiors and mesoporous shells [100] (Figure 2.12). Now,
soft-templating, hard-templating and select-etching approaches are investigated for
successfully fabricating hollow MSNs.
21
Figure 2. 12 The synthesis process of HMSNs
2.3.2.1.Soft-templating method
Soft-templating method refers to using surfactants as cave templates to build a hollow
interior cave after the removal of surfactants. For example, Feng and co-workers [101]
synthesized hollow MSNs through the aggregation of micelles. In their work, (-)-N-
dodecyl-N-methylephedrinium bromide (DMEB) was employed as a cave and pore
template and carboxyethylsilanetriol sodium salt (CSS) was used to stabilize the DMEB
micelles. Importantly, the particles size and mesoporous shell thickness could be changed
by altering the pH values during the synthesis possess (Figure 2.13). It has been suggested
that DMEB, CSS and pH values determine the morphology of hollow MSNs. Other
surfactants using for hollow MSNs fabrication through soft-templating route include
tetrapropylammonium hydroxide [98], hexadecane [102], Pluronic F127 [103] and
polystyrene latex [104].
22
Figure 2. 13 Schematic illustration of the formation process of HMSs [101]
Soft-templating method has been proved to be a simple and effective synthesis approach
of hollow MSNs. However, hollow MSNs obtain through this method tend to collapse in
the surfactant removing process. Moreover, it is hard to completely remove the cave
template through solvent extraction and this in turn will lead to severe side effects when
hollow MSNs are used to deliver drugs to human [100].
2.3.2.2.Hard-templating approach
Another strategy for the synthesising of HMSNs relies on employing some solid particles
like silica and hematite as hard templates to build a hollow interior cave after template
removal process (Figure 2.14). This fabrication route is made up of four steps: (a) the
formation of solid cave templates; (b) core template dispersion, sometimes
functionalization is needed to ensure core surfactants coat the cave templates favourably;
(c) silica mesoporous shell generation; and (d) the removal of solid cave. The diameter of
the spheres and the caves can be easily tuned by changing the solid core size. Zhang and
co-workers [105] prepared hollow MSNs with tunable shell porosity by the
transformation of solid silica particles to hollow structure in NaBH4 solution. Other hard
templates being used to create hollow structures are nanospheres of carbon, metals and
metal oxides [106]. In general, the challenge of this approach is to fabricate hollow MSNs
with small diameters (under 100nm).
23
Figure 2. 14 Scheme of the synthetic procedure of HMSs [107]
2.3.2.3.Select-etching technique
Figure 2. 15 Formation (left) and microscopic structure (right) of hollow/rattle-type
MSNs [108]
Recently, the fabrication of hollow MSNs through select-etching method was developed.
Hollow interiors are generated by making use of etching agents to etch the inner cores.
This strategy is based on the structural differences or compositional differences between
the core and shell structures [108]. Chen and co-workers [96, 108] prepared hollow/rattle-
type MSNs by using sodium carbonate solution to etch solid silica nanoparticles, resulting
in the building of a hollow cave in the middle (Figure 2.15). Gao and co-workers [109]
synthesized hollow MSNs with three different pore sizes by a structural difference based
selective etching methods. In their work, hollow MSNs exhibited the function of
controlling drug release rate which is achieved by changing the pore sizes. Similarly,
24
Zhang and co-workers [110] described a surface protected approach to prepare hollow
MSNs. The interior of solid silica spheres is selectively etched by NaOH when their near
surface layer is protected by poly (vinyl pyrrolidone).
2.3.3. Modification of MSNs
Figure 2. 16 Introduction of functional groups in different regions of MSN (a) at the
external surface, (b) at the pore entrances, or (c) within the walls [4].
After the preparation of MSNs, it is essential that these as synthesized nanocarriers
possess designed properties for specific drug delivery systems, such as high cellular
uptake, effective cell internalization, higher binding affinity with targeting molecules. In
this regard, modification of MSNs is always needed to make hollow MSNs smart carriers
for targeting.
As there are many hydroxyl groups existing on the surfaces, MSNs can be modified very
easily in different regions including pore walls, the interior/exterior surfaces and the pore
entrance (Figure 2.16). The introduction of functional groups results in more versatile
MSNs. For example, creating hydrophobic and hydrophilic surfaces, positively and
negatively charged surfaces, improving biocompatibility and targeting functions,
regulating drug release rates can be attained by the functionalization of MSNs. He and
co-worker [111] changed the surface charge of SBA-15-structured mesoporous silica
nanoparticles from -31.1 mV to 29.6 mV by grafting rhodamine B onto the surfaces. The
resulting MSNs showed an outstanding sustained-release property when used for drug
delivery. Again, Lu and co-workers [112] improved the dispersion of MSNs by modifying
25
the surfaces with trihydroxysilylpropyl methylphosphonate.
Methods used for the surface modification of MSNs include post synthesis grafting,
molecular imprinting, and organosiloxane/siloxane co-condensation. Some functional
groups have been widely introduced onto the surface of MSNs through one or more of
these methods (Table 2.1).
Table 2. 1 Methods employed to introduce functional groups onto
the MSNs surface [89]
Functional group Method
Ureidoalkyl Co-condensation
Mercaptoalkyl Co-condensation, grafting
Cyanoalkyl Co-condensation, grafting
Aminoalkyl Co-condensation, grafting
Allyl Co-condensation, grafting
Isicyanatoalkyl Grafting
Epoxyalkyl Grafting
Phosphonatoalkyl Grafting
Metal-aminoalkyl Molecular imprinting
2.3.3.1.Co-condensation
Inorganic–organic hybrid networks of MSNs are usually attained using co-condensation
of a tetraalkoxysilane and one or more organo-substituted trialkoxysilanes. This method
has the advantages of the homogeneous coverage of functional groups. However, the
selective functionalization of the external or internal surfaces of MSN is hard to achieve
through this method. The degree of organic functionalization can be controlled through
studying the electrostatic matching effects between substituent groups and the surfactant
[113, 114]. It is found that the anions which are less hydrated than Br- can displace it
from the cetyltrimethylammonium bromide (CTAB) micelle and bind tightly to the
26
cationic CTAB. The anionic lyotropic series reported by Larsen for interactions with
CTAB is: citrate < CO32- < SO4
2- < CH3CO2- < F- < OH- < HCO2
- < Cl- < NO3- < Br- <
CH3C6H4SO3- [113] .
2.3.3.2.Grafting
Grafting is a method commonly used for the post-synthesis modification of MSNs.
Introduction of some functional groups on the external or the internal surfaces of MSNs
is usually achieved by silylation on free silanol groups ( Si–OH) and germinal silanol
groups (-Si(OH)2). When compared to the co-condensed method, grafted materials
exhibit better hydrothermal stability. However, the problem that this method encountered
is the inhomogeneous surface coverage of functional groups. More specifically,
functional groups grafted onto the surfaces of MSN are mainly located on the external
surface and in close proximity to the pore entrances [84].
2.3.3.3.HMSNs modified with chemical groups for certain applications
Figure 2. 17 Functionalized hollow mesoporous silica nanoparticles [115]
27
HMSNs with a large cavity inside each original MSN, have recently been highlighted to
have great potential for biomedical applications due to their greatly enhanced drug
loading ability [115]. Additionally, with the availability of well-established techniques
for surface functionalization as illustrated above, HMSNs possessing biocompatibility
can be selectively modified for various applications, such as specific targeting and
imaging (Figure 2.17).
Recently, HMSNs have been modified and successively employed as highly attractive
multifunctional nanoplatforms for increasing surface hydrophobicity [116], enhancing
cellular uptake [117], cell targeting [106], bioimaging [115], and controlled drug release
[109]. For example, by integrating different types of chemical groups on the internal or
external surface, HMSNs can be used as a stationary phase in HPLC to separate both
biomolecules and small aromatic molecules, such as proteins [118]. By changing the type
(and length) of the employed alkyl chains, the accessible pore sizes of HMSNs can be
tailored for controlling the release kinetics of loaded cargos [119, 120]. Gao [109] and Jia
et al. [121] progressively increased the pore dimensions of HMSNs by extending the
etching time of HMSNs with the Na2CO3 and NaBH4 solutions, respectively and showed
sustained and controlled release of these modified HMSNs.
Nevertheless, although modifications have made HMSNs more versatile for applications,
they also have disadvantages that negatively affect HMSNs when they are exploited as
drug carriers. Firstly, if not well controlled, modifications would cause structure collapse
[102] of HMSNs. Further, the introduction of chemical groups on the surface of HMSNs
may bring additional toxicity which makes HMSNs unsuitable as carriers for drug
delivery systems. Moreover, in many cases, it is difficult to couple several functional
groups in sufficient concentration, since the number of attachment sites on the particle
surface is limited. In addition, each functionalization step might negatively affect the
suspension stability of the particulate system, depending on the physicochemical
properties of the added function. Therefore, all these issues should be considered when a
drug delivery system is engineered based on modified HMSNs and the development of
effective methods to deal with these problems are of great importance.
28
2.3.4. Qualitative characterization of HMSNs
2.3.4.1. Morphology
HMSNs as a kind of mesoporuous materials possess a porous shell with well-defined
cylindrical/hexagonal mesopores and a spherical hollow cavity. The characterization of
HMSNs is similar to other mesoporous materials. In general, a variety of techniques
including transmission electron microscopy (TEM), scanning electron microscopy (SEM),
electron diffraction (ED), X-ray diffraction (XRD), small-angle X-ray/neutron scattering
(SAXS/SANS), gas adsorption measurements and nuclear magnetic resonance (NMR)
were applied to study the morphology and microstructure of HMSNs [122]. Particularly,
the hollow structure can be clearly seen under TEM, where a contrast in colour between
the cavity and the shell was observed due the reduced density in the cavity (Figure 2.18).
Moreover, the existence of mesoporous shell of HMSNs is often confirmed by the
presence of low angel peaks (2θ < 5˚) in HMSNs’ XRD or SAXS patterns (Figure 2.19)
[123-125].
Figure 2. 18 TEM morphology of HMSNs [104, 126]
Conventionally, sorption isotherms represent the most widely used method to provide
detailed information for determining porous structure, pore geometry and size as well as
29
pore distribution of HMSNs [127, 128]. According to the IUPAC classification, there are
six categories of gas adsorption isotherms, Types I-V and IVc (Figure 2.20) [129], which
is a source to study structural information of porous materials due to their different
characteristic shapes.
Figure 2. 19 XRD patterns of mesoporous silica nanoparticles synthesized at different
pH [125]
On the basis of the diameter, pores are classified as micropores (< 2 nm), mesopores (2 -
50 nm), and macropores (> 50 nm). In detail, Type I isotherm exhibiting prominent
adsorption at low relative pressures and then levelling off, is usually related to adsorption
in micropores. However, it may also be considered to be indicative of mesoporous
materials with pore sizes close to the micropore range. Type II and III isotherms are often
observed for macroporous materials while Type IV in general, and IVc in particular, is
typical for many materials with accessible mesopores.
Adsorption on mesoporous solids is a multilayer adsorption process followed by capillary
condensation in mesopores, which give rise to Type IV isotherms [129]. It should be
noted that some porous materials may exhibit a combination of the six types of isotherms
as a result of the presence of several different types of pores in the structure.
30
Figure 2. 20 Six categories of gas adsorption isotherms for porous materials with different
pore size [129].
31
Based on the above illustration, nitrogen adsorption-desorption isotherm of HMSNs
should follow the type IV as illustrated in Figure 2.21, with a high pore volume and a
large surface area.
Figure 2. 21 Adsorption isotherms of nitrogen in mesoporous silica nanoparticles of
different pore sizes. Closed symbols, adsorption; open symbols, desorption [122].
2.3.4.1. Pore geometry and pore size
Pore geometry is an essential part of investigating HMSNs as pore types and shapes play
important roles in the applications of HMSNs. According to the types of pores that define
the networks (Figure 2.22), MSN are divided into several types, such as MCM-41, SBA-
15 and FDU-1 [130].
32
Figure 2. 22 Pore geometry [129]
Nitrogen Absorption and Desorption (BET/BJH) is commonly used to characterize the
pore geometry and pore size of MSNs. Isotherm linear plots (Figure 2.23, type H1-H4)
obtained from BET/BJH adsorption and desorption equipment provide information of
pore geometry [128, 129, 131]. Owning to different pore structure, the way of gas
evaporation differs from one another, which results in the change in shape of the
hysteresis loops on nitrogen adsorption-desorption isotherms (Figure 2.23). Therefore the
shape of a gas adsorption and desorption isotherm is one of the sources that provide
important qualitative structural information. By studying shapes of the corresponding
isotherms, the information about pore geometry can be obtained.
Figure 2. 23 Effect of pore geometry on shapes of hysteresis loop
33
For an ideal 5-FU delivery system, it is essential to make sure that the drug is easily loaded
and released in a designed profile. The pore connection would therefore be a vital element
that should be taken into consideration, as it plays a key role in assessing the feasibility
of hollow MSNs nanocarriers for 5-FU delivery. The accessibility of mesopores as well
as the pore size can be acquired through this method. However, this characterization
method is limited to pores with a size of 2-50 nm.
Positron annihilation lifetime spectroscopy (PALS) is another technique used to identify
the porosity of mesoporous materials. PALS is a powerful characterization giving useful
information on hierarchical porosity, especially on micropores and molecular scale free
volume elements that is information hardly available from other characterisation methods,
such as BET and XRD.
2.3.5. MSNs’ pharmacokinetics
The pharmacokinetics of MSNs is fundamentally significant to understand, interpret and
assess the effectiveness of particles as it is closely connected to the toxicity and
biocompatibility of MSNs as drug carriers. Thus, the study of MSNs’ pharmacokinetics
is vitally essential before the utilization of MSNs for delivery systems. The current
investigations on MSNs’ pharmacokinetics include blood circulation [132], retention,
biodegradation, excretion and bio-distribution.
MSNs have found to be biocompatible and well tolerated when injected in mice with a
dosage of 50 mg kg-1. This dosage is proved to be an adequate concentration to provide
effective cancer therapy. On the other hand, higher dosage of MSNs (1.5 g kg-1) would
cause death to mice. The biocompatibility of MSNs can be improved by coating
biocompatible polymers.
34
Figure 2. 24 TEM images of MSNs (black dots) endocytosed by human cervical cancer
cells [88]
The pathways of HMSNs for entry into cells are endocytosis but the surface
functionalization and change in structural properties of HMSNs can lead to different
pathways (Figure 2.24) [88]. Lu [87] found that MSNs were rapidly and completely
cleared from animal bodies through urine and feces. The biodegradation of HMSNs
follows a three-stage degradation, involving a fast bulk degradation stage in the first hour,
a decelerated degradation stage hindered by calcium/magnesium silicate layer and finally,
a slow diffusion stage on day-scale [133, 134]. The bio-distribution of HMSNs is mainly
in the liver and spleen [87, 135]. Moreover, the influence of surface potential on the
pharmacokinetics of HMSNs is significant. Studies conducted by Mou and co-workers
[136] reveal that HMSNs with positive charge on the surface undergo faster transportation
and excretion.
2.4. Drug loading and release
As for drug delivery system, high drug loading and controlled or sustained drug release
are highly desired. In this case, hollow MSNs possessing both nano-scaled particle size
and large pore and cave diameter are one of the promising candidates being pursued
(Figure 2.25). Size, surface potential and hydrophilicity are factors to be controlled to
make MSNs as ideal drug carriers with extraordinarily high load capacity as well as
controlled release profile [137].
35
Figure 2. 25 HMSNs based drug delivery system for simultaneous joint lubrication and treatment [138]
Hollow MSNs have open pores for drug molecules to be encapsulated and have hollow
interiors for a high drug storage (Figure 2.26) which makes drug loading and release quite
different from polymeric carriers [139, 140]. The hollow cave and pore size would have
significant effect on drug loading and release capacity. It is reported that selective
adsorption of drug greatly depends on the pore size diameter [100]. Moreover, surface
potential and hydrophilicity also play important roles on drug loading capacity. In general,
hollow MSNs with both the opposite potential and similar hydrophilicity of drug
molecules are more likely to achieve higher drug adsorption. Therefore, such parameters
should be taken into account while designing nanocarriers for a given drug molecule. He,
Zhang et al. [111, 141] developed a drug delivery system by fabricating MSNs with
positive charge on their surface to load a negative charged drug for anti- hepatic fibrosis.
It is found that the positively charged MSNs present better drug loading capacity and
controlled release profiles than negatively charged ones. Gao and co-workers [109]
regulated the release rate of doxorubicin by tuning the pore diameter of hollow MSNs.
36
Figure 2. 26 HMSNs based drug delivery system for high-intensity focused ultrasound
(HIFU)-mediated intravenous drug delivery [142].(a) The schematic illustration of the
synthesis route of HMSN-LM and its applications in multi-drug co-delivery, temperature-
responsive release and HIFU treatment enhancement; TEM images of HMSNs (b) and
drug loaded HMSNs (c), and the insets are their corresponding schematic illustration.
Transmembrane transport of hollow MSNs is another element that affects drug release
kinetics in vivo. It is well established that the internalization of nanocarriers via
endocytosis are size and cell dependent. MSNs can be prepared with a wide range of
particle size (30-300nm) and tailored shapes (spheres, plates and rods). However, only
particular size and shape can exhibit good transmerbrane transport. Lu and co-workers
[143] found out MSNs with 50 nm in diameter exhibit the best cellular uptake efficiency.
This is also proved by Tang et al [100] who pointed out that the order of cellular uptake
efficacy of MSNs is 50 nm > 30 nm > 110 nm > 280 nm > 170 nm. Importantly, it is
suggested that the most effective size range of MSN in drug delivery is 50-150 nm from
this standing point of cellular uptake efficacy. In addition, for transmembrane transport,
the feature of the net negative surface charge of cell membrane attracts attention from
researchers [88, 144]. Nanocarriers are modified to possess positive charge on their
surfaces leading to high efficiency in the endocytic trafficking in mammalian cells [136,
37
145]. He et al [146] pointed out that the higher surface hydrophobicity and larger particle
size will prohibit the transmembrane delivery.
Inspired by these studies, as for 5-FU, in order to attain high adsorption, it is essential to
fabricate hollow MSNs with a particle size within the range of 50-150 nm. Also, hollow
MSN carriers possessing hydrophilic and positively charged groups on its mesopore walls
and surfaces would enhance their 5-FU loading capacity.
2.5. Drug resistant phenomenon
Systemic toxicity of anti-cancer drugs as well as the emergence of multidrug resistance
(MDR) by cancer cells have greatly limited the success of chemotherapy, making cancer
remains one of the major causes of mortality and morbidity in the world [147-150].
The systemic toxicity of 5-FU mainly resulted from its nonspecific distribution in a
biological system, leading to the death of normal cells [35, 151]. Nanocarriers such as
HMSNs have been developed that can offer targeting delivery of chemotherapy drugs to
cancer cells by taking the advantage of enhanced permeation and retention (EPR) e ect
[152, 153] in the tumour site. Further functionalization of the surface of HMSNs allows
for enhanced recognition and internalization of drug loaded HMSNs by target cell
population, thereby protecting normal tissues from exposing to the toxic effects of anti-
cancer drugs [154-157].
However, even though the toxicity of 5-FU would be overcome by a well-designed
HMSNs based targeted drug delivery system, the effective cancer treatment of 5-FU
cannot be maintained for a relative long period for a complete killing of the cancel cells.
Instead, the therapeutic effect of 5-FU decreased along the time, leading to treatment
failure and recurrence or relapse of tumours [158, 159]. This is mainly due to the cancer
cells developing resistance to the drug 5-FU, which is another obstacle to the successful
cancer treatment [160, 161].
38
Figure 2. 27 Rational design of a multifunctional micellar nanomedicine for targeted co-delivery of siRNA and DOX to overcome multidrug resistance [162]
(a) Chemical structure of functionalized copolymers with ligands at the end of the PEO block and
conjugated moieties on the PCL block. (b) Assembly of multifunctional micelles with DOX and siRNA in
the micellar core and RGD and/or TAT on the micellar shell. (c) Schematic diagram showing the proposed
model for the intracellular processing of targeted micelles in MDR cancer cells after receptor-mediated
endocytosis, leading to cytoplasmic siRNA delivery followed by P-gp down-regulation on the cellular and
nuclear membrane and endosomal DOX release, followed by DOX nuclear accumulation
Multidrug resistance (MDR) involves cancer cells overexpressing drug transporter
proteins, like P-glycoprotein (P-gp) to pump the drugs out of the cells [163, 164]. To
overcome this problem, combination delivery of chemotherapy drugs [165, 166] is
emerged as a strategy and is proved to be effective for some short-term treatments.
However, in many case, long-terms therapy is required for a full cure of cancers, where
co-delivery of chemotherapy drugs shows inability [167]. This is mainly due to the fact
that after a particular type of cancer has developed resistance to a given drug, the cancer
cells develop resistance to other drugs as well. Fortunately, the mechanism and genes
behind MDR are well explored and the inhabitation of P-gp functions was achieved by
specific small biological molecules [162]. Despite this effort, because of their nonspecific
action that caused intolerable toxicity, no P-gp inhibitor has been approved for clinic use.
Therefore, new strategies to inhibit the expression (rather than the function) of P-gp is of
great importance.
39
In this regard, the use of small interfering RNA (siRNA) for a potent post-transcriptional
gene silencing was developed [163]. However, the delivery of siRNA through systemic
administration is often obsessed with poor stability, low cellular uptake, and rapid
clearance of siRNA from the circulation [162]. Similarly, nanoplatforms can provide
efficient delivery of siRNA with high specificity and low off-target effect. A better
therapeutic effect in MDR tumours was observed by co-delivery of drugs and siRNA to
the same cancer cells [162, 167]. For example, as shown in Figure 2.27, siRNA was
successfully delivered by a multifunctional micellar system to overcome multidrug
resistance. Mao et al. [168] used functionalized quantum dots for simultaneous delivery
of siRNA and anticancer drug DOX, which showed effectively induced apoptosis and
suppressed tumour growth (Figure 2.28).
Inspired by the research illustrated above, we hypothesised that co-delivery of 5-FU and
siRNA using HMSNs as nanocarriers would simultaneously deal with the problems
regarding systemic toxicity and drug resistance. Nevertheless, co-delivery of drug and
siRNA is a challenging task due to their huge different properties [169]. Thus, when
engineering a co-delivery system, multiple factors need to be considered beforehand.
Firstly, mutual chemical compatibility should be ensured. The capacity to load and
release drug-siRNA combination then needs to be determined [170]. More importantly,
the differences in hydrophobicity, metabolic stability and molecular weight between
drugs and siRNA are very critical for nanocarriers [167, 171].
Based on this literature review, we propose to develop a co-delivery system through a
more feasible option. HMSNs will be loaded with 5-FU and siRNA separately. Then by
combining 5-FU loaded HMSNs and si-RNA loaded HMSNs with different ratio will be
delivered to colorectal cancer cells. This proposed strategy take the advantage of avoiding
co-encapsulation of 5-FU and si-RNA, which will provide a more-potent, independent
control and less-toxic means for cancer treatment.
40
Figure 2. 28 System for delivery of doxorubicin (DOX)/siRNA combination based on quantum dots (QDs) modified with β-cyclodextrin [168]
41
Chapter 3 Materials and methodology
3.1 Introduction
This chapter outlines the chemical reagents, HMSNs fabrication and modification
techniques, in vitro experimental methods as well as characterization technologies
applied in this thesis. Firstly, all the materials including chemicals and auxiliary reagents
are specified, followed by a full illustration of preparation and chemical functionalization
technologies for HMSNs. Additionally, strategies used for 5-FU loading and release study
are described. Furthermore, detailed in vitro experimental methods for assessing the
biological activity of HMSNs as drug carriers for targeted drug delivery system are
discussed. Finally, technologies utilised to characterize the morphology and properties of
samples in terms of particle size, pore size, surface area, modification efficiency, drug
loading capacity, release profile, biocompatibility and targeting ability are fully
introduced.
3.2 Materials
Tetraethyl orthosilicate (TEOS, Si(OCH2CH3)4), Triton X-100 (TX100), Poly(ethylene
glycol)20-block-poly(propylene glycol)70-block-poly(ethylene glycol)20 (P123),
Poly(ethylene glycol)106-block-poly(propylene glycol)70-block-poly(ethylene glycol)106
(F127), Tweens 20, ammonia solution (28-30%), Sulphuric acid, Sodium carbonate
(Na2CO3), Octadecyltrimethoxysilane (OTMS), N-Hydroxysuccinimide (NHS), 1-ethyl-
3-(3-dimethylaminopropyl) carbodiimide (EDC), (3-Aminopropyl)triethoxysilane
(APTES), 3-Cyanopropyltriethoxysilane (CPTES), Fluorescein isothiocyanate (FITC),
fluorinated pyrimidine 5-fluorouracil (5-FU), deuterium oxide, Epidermal growth factor
(EGF, human), 100× penicillin/streptomycin solution, Trypsin−EDTA Solution (0.25%
trypsin, 0.1% EDTA), paraformaldehyde (PFA), 4',6-Diamidino-2-Phenylindole (DAPI)
were all purchased from Sigma Aldrich (Sydney, Australia). Dulbecco's Modified Eagle
Medium (DMEM), Fetal Bovine Serum (FBS), Phosphate-Buffered Saline (PBS) and
42
GlutaMAX™ were from life technologies. Eudragit S100 which is made in Germany was
purchased from Shenzhen Youpuhui Pharmaceutical Co. Ltd (Shenzhen, China).
3.3 Preparation of versatile HMSNs
3.3.1 Formation of Eudragit S-100 nanoparticles
The synthesis of the Eudragit core particles were achieved by nanoprecipitation [172].
Specifically, 20 mL acetone solution of Eudragit S-100 (4 mg/mL) was added dropwise
to 100 mL deionised water under stirring. The suspension was left to proceed for 5 h
under agitation at room temperature before being used for HMSNs fabrication.
3.3.2 Fabrication of HMSNs
For the formation of the mesoporous silica shell, 120 mL Eudragit S-100 nanoparticle
solution (0.75 mg/mL) was ultrasonically dispersed for 3 min (100W), followed by
addition of 1.0 g TX100 under stirring. After TX100 was completely dissolved, 2.3 mL
TEOS was added and this solution was left to proceed for 24 h before being transferred
into a sealed container and hydrothermally treated overnight under 100 C. The resulting
nanoparticles (as-synthesized HMSNs) were collected by centrifugation (10000 rpm for
20 mins) and redispersed in Acetone for the removal of Eudragit core and TX100
templates. The particles were then washed 3 times with ethanol and deionized water
respectively before finally freeze-dried (deionized water was used as the solvent for all
freeze-drying procedures).
For stability study, paralleled samples were prepared using the same method illustrated
above to obtain as-synthesized HMSNs. Then these as-synthesized HMSNs were freeze-
dried before undergoing a calcination process, where 550 ˚C was applied for 6 h to
remove the Eudragit core and TX100 templates.
43
3.3.3 Control of particle size
The particle size of HMSNs was controlled by synchronously changing the amount of
TEOS and the concentration of Eudragit S-100 solution applied in HMSNs fabrication.
In detail, 20 mL acetone solution of Eudragit S-100 (4 mg/ml, 8 mg/ml and 16 mg/ml)
was added dropwise to 100 ml deionised water under stirring. The suspension was left to
proceed for 5 h under agitation at room temperature. Then the resulting Eudragit S-100
nanoparticles suspensions were ultrasonically dispersed for 3 min, followed by addition
of 1.0 g TX100 under stirring. After completely dissolving TX100, 2.3 mL, 3.5 mL and
8.0 mL TEOS were added, respectively and these solutions were left to proceed for 24 h
before transferred into sealed containers and hydrothermally treated overnight under 100
˚C. The resulting HMSNs with different size were collected by centrifugation and
redispersed in Acetone to remove Eudragit core and TX100 templates. Finally all the
particles were collected and freeze-dried.
3.3.4 Control of shell thickness
HMSNs with varying shell thickness (10 nm, 15 nm and 30 nm) were fabricated by using
different types of nanoparticles as cores. Typically, HMSNs with 10 nm shell thickness
was fabricated according to the method illustrated in 3.3.2 while those with 15 nm and 30
nm shell thickness were synthesised with the following procedures.
Six milliliter TEOS was added dropwise to a mixture containing 100 mL ethanol and 6
mL ammonia solution, which was hydrolysed and condensed for 0.5 and 1 h at 30 ˚C to
obtain silica nanoparticles with different diameters as cores for further HMSNs
fabrications. After that, 2 mL OTMS and 5 mL TEOS were firstly mixed and then added
to the above solutions, which was left to react for another 1 and 4 h at 30 ˚C under
magnetic stirring. Finally, the yield particles were collected by centrifugations and
redispersed in Na2CO3 solutions (0.6 M) for 1 h at 80 ˚C before calcinations and freeze-
drying.
44
3.3.5 Pore expansion of HMSNs
The control of the pore size of then HMSNs were achieved by varying the molecular
weight of non-ionic surfactants used for HMSNs fabrications. Larger pores were
developed by surfactants with longer molecular chain which forms bigger micelles that
finally evolved to pores of HMSNs. F127, P123 and Tweens 20 were used instead of
TX100 for HMSNs synthesis described in 3.3.2.
Another method for pore expansion is to apply different calcination time to HMSNs,
where small pores were collapsed and merged into larger pores. In detail, 6 mL TEOS
was added dropwise to a mixture containing 100 mL ethanol and 6 mL ammonia solution,
which was hydrolysed and condensed for 1 h at 30 ˚C. Then, 2 mL OTMS and 5 mL
TEOS were mixed first and added to the above solutions, which was left to react for
another 1 h at 30 ˚C under magnetic stirring. After that, the yield particles were collected
by centrifugations and redispersed in Na2CO3 solutions (0.6 M) for 1 h at 80 ˚C. The
resultant particles were divided into 3 groups which underwent calcinations for 3, 6 and
12 h at 550 ˚C before freeze-drying. HMSNs with pore size of 2.5, 4.3 and 20.1 nm were
acquired.
3.4 Modifications of HMSNs
3.4.1 Amine-, methyl- and cyano-functionalization
For amine-functionalization (HMSN-NH2), 50 mg HMSNs were ultrasonically dispersed
in 50 mL ethanol, 25 μL APTES were added and stirred for 24 h before centrifugations
and washed 3 times with ethanol then deionized water respectively. Finally these yielded
HMSN-NH2 were collected and freeze dried.
The experimental procedure for methyl- and cyano-functionalization (HMSN-CH3 and
HMSN-CN) is similar to amine-functionalization. The only difference is to use OTMS
45
and CPTES instead of APTES for chemical modification.
3.4.2 Carboxyl-modification (HMSN-COOH)
Fifty milligrams of HMSNs were ultrasonically dispersed in 50 mL ethanol, 25 μL
APTES were added and stirred for 24 h. The resulting nanoparticles were collected by
centrifugations and redispersed in sulphuric acid (50% v/v), heated at 150 ˚C for 4 h. The
final products were re-collected, washed and freeze dried.
3.4.3 FITC-modification (FITC-HMSNs)
Twenty milligrams of HMSNs were suspended in 20 mL ethanol, followed by addition
of 10 mL APTES, then this mixture was stirred for 24 h at room temperature. The
resulting amino functionalized HMSNs were collected by centrifugation and then
redispersed in 10 mL FITC ethanol solution (1 mg/mL) and stirred for another 24h. The
final FITC modified HMSNs (FITC-HMSNs) were collected for freeze-drying.
3.4.4 EGF labelling (HMSN-EGF)
Sixteen milligrams of NH2-HMSNs were ultrasonically dispersed in 1 mL deionised
water, to which 2.1 mg NHS and 17.3 mg EDC were added. After that, 0.4mL of EGF
solution (0.2 mg/mL in PBS solution) and 0.8 μL of triethyl amine were added. The
mixture was left to proceed for 1 h under stirring at room temperature. The EGF grafted
hollow mesoporous silica nanoparticles (EGF-HMSNs) were collected by centrifugation
and washed in deionised water. These EGF-HMSNs were redispersed in 8 mL FITC
ethanol solution (1 mg/mL) and stirred for 24h. The final FITC modified EGF-HMSNs
(EGF-HMSN-FITC) were collected by centrifugation and washed with ethanol then
deionized water. Finally, the product was freeze-dried.
46
3.5 5-FU loading and release study
3.5.1 Measurement of 5-FU oil : water partition coefficients
Measurement of 5-FU oil : water partition coefficients was conducted via shake flash
method according to OECD guideline 107 [173]. Firstly, n-octanol and Milli-Q water
were saturated, respectively prior to the experiments and herein their saturated solutions
were referred to as octanol and water. Then, 5-FU was dissolved in water to obtain a
concentration of 266.7 μg/mL. A combination of octanol and water with an octanol:water
ratio of 1:1, 1:2 and 2:1 were added to centrifuge tubes and a total volume of 40 mL was
maintained. Each ratio was carried out in triplicate. 1mL 5-FU was added to the centrifuge
tubes and the phases were mixed by vortex for 10 min before centrifuging at 1000 rpm
for a further 10 min. Both 5-FU concentrations in water and octanol phases were
determined via UV-Vis spectroscopy (Ocean Optics system, USA).
The partition coefficient Pow is calculated from the data of each run using the following
equation (Equ. 3.1):
3.1
3.5.2 5-FU loading
The loading of 5-FU was carried out by soaking HMSNs in a concentrated drug solution
at room temperature for 24 h while stirring to ensure the diffusion of the drug molecules
through the mesopores. Firstly, 5-FU was dissolved in deionized water to a concentration
of 3 mg/ml. 150 mg of HMSNs was ultrasonically dispersed in 50 mL of the 5-FU solution.
The mixture was stirred at room temperature. At timed intervals of 0, 2, 4, 6, 8, 12 and
24 h, 0.5 mL of the solution was extracted, centrifuged to collect the supernatant for UV-
Vis analysis at a wavelength of 266 nm while the 5-FU -loaded HMSNs precipitate was
placed back into the drug solution. At intervals of 0, 2, 4, 6, 8 and 12h, the HMSNs only
loaded a certain amount of 5-FU and they still have the capacity for further encapsulation
47
of the drug. Therefore, the 5-FU loaded HMSNs precipitate collected at intervals were
placed back to the mixture for further loading. After the loading, the 5-FU loaded HMSNs
were washed with deionized water before freezer-drying. The loading amount of 5-FU in
the HMSNs was calculated by subtracting the amount of 5-FU in the supernatant from
the amount in the original drug loading solution.
The drug loading capacity was calculated using the following equation:
3.5.3 5-FU release
To measure the drug release profile from 5-Fu loaded HMSNs, 30 mg of 5-FU-loaded
HMSNs were dispersed in 150 mL of phosphate buffer saline (PBS) solutions at pH 5.5
or pH 8.0 and the mixtures were gently stirred at 37 C. At timed intervals, 1 mL of the
solution was removed and centrifuged (13000 rpm for 5 mins) for UV-Vis analysis and 1
mL of fresh PBS was added to the drug release solution to keep the volume constant. For
drug release, it is important to make sure that a certain amount of the drug is released into
a constant volume. A changing volume can affect the release rate and release kinetics to
some extent. The method used in this work for measuring the drug release behavior of
HMSNs is a standard method which has been widely used [174-177].
The release percentage at each time point was analysed by UV-Vis at a wavelength of
266 nm and calculated using the following equation:
×100%
where Ct is the drug concentration at a time t, Ct-1, Ct-2 are the drug concentrations detected
at different time points previous to t (C0 =0). V1 is the total volume for drug release (150
mL), V2 is the volume that is taken for UV measurement (1 mL), W1 is the weight of drug
loaded HMSNs (0.03 g) and L is the drug loading capacity of HMSNs (194.5 mg(5-
FU)/g(HMSNs)). The 5-Fu release experiment was conducted in triplicate and the mean
48
results were reported.
3.6 Cell experiments
3.6.1 Cell culture
The colorectal cancer cell line, SW 480 and SW 620, with and without EGFR
overexpression, were cultured in DMEM containing 10% foetal bovine serum and 1%
penicillin/streptomycin supplemented with 1% GlutaMAX™. The culture was
maintained at 37 C in a humidified atmosphere containing 5% CO2.
3.6.2 Cellular uptake of HMSNs and EGF-HMSNs in colorectal cancer cells
SW480 and SW 620 were seeded into 24-well plates with 1×105 cells per well in 0.5 mL
of complete growth medium. After 24 hours of cell attachment at 37 in 5% CO2, the
cells were treated with 2, 5 or 10 μg/mL of EGF-HMSN-FITC or FITC-HMSNs for 30
min, 2 hours or 4 hours. The cells were washed with PBS three times and harvested by
trypsinization. Then the cells were resuspended in PBS containing 10% of FBS for flow
cytometric analysis. The green fluorescently-labelled live cells were counted as positive
cells which are associated with nanoparticles and the mean fluorescence intensity (MFI)
was calculated. The cells incubated with DMEM only were used as a negative control.
To visualize the uptake of HMSNs via confocal microscope, SW 480 and SW 620 cells
were seeded into 8-well chamber slides at 5 ×104 per well for 24 hours before incubation
with 10 μg/mL of EGF-HMSN-FITC or FITC-HMSNs for 30 min, 2 hours or 4 hours.
Cells were fixed with 4% paraformaldehyde and stained with DAPI for confocal
microscopy imaging. Confocal imaging of the cells was carried out on a laser scanning
confocal microscope.
49
3.7 Characterizations
3.7.1 Electron microscopy
The morphology of HMSNs was studied using scanning electron microscopy (SEM, Carl
Zeiss AG SEM Supra 55VP) and transmission electron microscopy (TEM, LaB6 JEOL
JEM-2100). Before imaging under SEM, all the samples were coated with carbon through
a BAL-TEC SCD 050 sputter coater (Leica Microsystems, Australia). TEM imaging were
conducted under 200 kV.
3.7.2 Nitrogen absorption and desorption experiments
Nitrogen absorption and desorption isotherms of HMSNs were conducted on a
Micromeritics Tristar 3000 (Particle & Surface Science, UK) equipment at 77K. Before
measurements, samples were degassed at 523 K for 1 h under nitrogen. The specific
surface areas of HMSNs were calculated using the Brunauer-Emmett-Teller (BET)
method. The Barrett-Joyner-Halenda (BJH) model was utilized to obtain the pore size
distributions from the desorption branch of isotherms.
3.7.3 Small angle X-Ray diffraction (SAXRD)
The mesoscopic ordering of the nanoparticles was confirmed by small angle X-ray
diffraction (SAXRD) using a Bruker D8 Advance (Bruker, Germany) X-ray
diffractometer with Cu-Kα radiation (λ=0.1542 nm). The instruments were operated at 40
kV and 40 mA current.
3.7.4 Thermo-gravimetric analysis (TGA)
TGA of HMSNs were conducted using a TG Simultaneous Thermal Analyser (STA 449C,
ZETZSCH, Germany). Samples with a weight of ~5 mg were placed into aluminium
oxide crucibles and heated under nitrogen condition at a heating rate of 10 ˚C/min from
room temperature to 600 ˚C.
50
3.7.5 Fourier transform infrared spectrometer (FTIR)
Fourier transform infrared (FTIR) spectra were measured with a Vertex 70 spectrometer
(Bruker, Germany). All the samples were compressed into KBr pellets and recorded at 64
scans from 4000cm-1 to 400cm-1 with a resolution of 4 cm-1.
3.7.6 Time of flight secondary ion mass spectrometry (ToF-SIMS)
ToF-SIMS experiments were performed using a Physical Electronics Inc. PHI TRIFT V
nanoTOF instrument (Physical Electronics Inc., Chanhassen, MN, USA) equipped with
a pulsed liquid metal 79+Au primary ion gun (LMIG), operating at 30 kV energy. Dual
charge neutralisation was provided by an electron flood gun (10 eV electrons) and 10 eV
Ar+ ions. Experiments were performed under a vacuum of 5x10-6Pa or better. “Bunched”
Au1 instrumental settings were used to optimise mass resolution for the collection of
spectra. Positive SIMS spectra were collected from areas of 100×100 micron, with an
acquisition time of 2 minutes.
In order to meet the static SIMS regime, the corresponding total primary ion dose of less
than 1 × 1012 ions cm-2 was typically achieved [178]. A mass resolution at nominal m/z
= 27 amu (C2H3+) was m/ m of > 5000. Ten positive ion mass spectra were collected
from areas that did not overlap for each sample. Sample spectra were processed and
interrogated using WincadenceN software (Physical Electronics Inc., Chanhassen, MN,
USA). Integrations were performed on all recognisable and unobscured immonium ions
in the 2-160 amu mass range. The peaks were normalised to the total intensity of all
selected peaks [179].
PCA was performed using Statistica software (StatSoft Pacific Pty Ltd, North Melbourne,
VIC, Australia).
51
3.7.7 Nuclear magnetic resonance (NMR)
The pulse-field gradient NMR (PFG-NMR) diffusion experiments were carried out on a
Bruker Avance III 500 MHz wide bore spectrometer (with proton Larmor frequency of
500.07 MHz) equipped with a 5 mm diff50 pulse-field gradient probe. Each sample was
packed in a 5 mm Schott E NMR tube to a height of 50 mm. The pulse-field gradient
stimulated echo (PFG-STE) pulse sequence [180] was used to obtain diffusion
coefficients and the 1H NMR signal was used for the determination of diffusion
coefficients of different molecules. The maximum gradient strength of the amplifier is
29.454 T/m.
3.7.8 Dynamic light scattering (DLS)
The particle size of Eudragit and micelle sizes of TX100 were investigated by dynamic
light scattering (Malvern ZetaSizer Nano). Each sample was tested in triplicate and the
mean result was reported.
Zeta potential of HMSNs were analysed on this instrument as well. Samples were
dispersed in ethanol with a concentration of 1 mg/ml. Measurements of each sample were
conducted in triplicate and the average value was considered.
3.7.9 Confocal microscopy
Confocal laser scanning microscope (CLSM) images of cells were took via a confocal
microscopy (Leica TCS SP5, Germany) using 40× objective, 488 nm/405 excitation.
52
Chapter 4 Synthesis of tailored HMSNs for nanotheranostics
4.1 Introduction
Hollow mesoporous silica nanoparticles (HMSNs) with large hollow interiors and
penetrating pores in the shell are exceptionally promising platforms in many current and
emerging fields such as pharmaceutics [1], optics [2], environmental protection [3] and
catalysis [4]. In particular, they were exploited as nanocarriers for drug delivery [5]
because their huge cavities provide large drug loading capacities [6] and the open
mesopores allows transportation of drugs [7]. In addition, owing to their hollow structure,
HMSNs also possess other characteristics such as a decreased particle density and
increased active area for drug loading. In this regard, HMSNs are superior to non-hollow
mesoporous silica nanoparticles (MSNs) and other corresponding mesoporous silica
nanoparticle composites [8]. However, owing to the complex effects of physicochemical
parameters on synthesis reactions, fabricating HMSNs with good stability and dispersion
remains a main challenge [5] which limits their further application in drug delivery.
Thereby, developing new, reliable and facile methodologies for synthesising discrete and
dispersible HMSNs is important and could contribute to the further development of drug
delivery systems.
One of the typical methods to produce mesoporous silica materials is self-assembly which
is a well-known synthesis strategy and occurs when molecules interact with one another
through a balance of attractive and repulsive interactions [9, 10]. For mesoporous silica
materials, self-assembly is achieved via a surfactant templating route which involves the
spontaneous organization of templates into structurally ordered micelles through van der
Waals and Coulomb interactions [11] or hydrogen bonding. As shown in previous studies,
non-ionic surfactants such as tri-block copolymers with certain hydrophilic/hydrophobic
ratios and neutral surfactants like Triton X-100 (TX100) have been demonstrated to be
extremely suitable for assembling ordered mesoporous silicates [12, 13]. However, due
to the difficulty in adjusting micelle structure and the poor understanding of the
53
mechanisms behind the self-assembly formation of mesoporous structure in sol–gel
chemistry, the controllable tailored design of non-ionic surfactant templated HMSNs is
still far from satisfactory. Therefore, there is immense value in the development of new
methodologies for regulating the structure of non-ionic surfactant micelles in order to
tailor HMSNs for certain applications.
Current efforts are ongoing to optimize the parameters and yield of mesoporous structures
by changing material ratios or fabrication parameters, such pH [14, 15] and synthesis
temperature [16]. From a technical point of view, an insight into the micelle organization
mechanism can contribute to the improvement of the silica structure including stability,
pore size and volume. In addition, customisation of the silica structure will allow
facilitation of the host-guest specific interaction for the intended purpose.
In this chapter Eudragit S100 was introduced as a core particle to bind and facilitate the
self-assembly of the non-ionic surfactant into stable composite micelles. The reactions
between surfactant molecules and Eudragit S100 was investigated by Fourier Transform
Infrared (FTIR), Dynamic Light Scattering (DLS) and Pulse Field Gradient NMR (PFG-
NMR) analyses. The addition of Eudragit S100 were inferred to result in the redistribution
of surfactant micelles by hydrogen-bonding. Uniform and highly ordered HMSNs were
achieved, demonstrating the efficiency of using Eudragit to assist in the morphological
formation of surfactant micelles and consequently, the silica shell structure.
The efficacy of Eudragit-assisted HMSNs as drug carriers was ascertained by examining
the loading and release behaviour of the first line anticancer drug, 5-FU. Additionally, in
vitro cytotoxicity and cellular uptake of these HMSNs by colorectal cancer cells SW480
were investigated.
54
4.2 Synthesis of HMSNs with designed properties for drug delivery
systems
4.2.1 Preparation and characterization of HMSNs
For HMSNs preparation, typically a hard or soft sphere is used to create the hollow cavity
in HMSNs while surfactants are employed to develop pores in the shell and this in turn,
involves a template-eliminated process [106, 107, 116, 181] . One of the typical methods
to remove the core template and the pore template is calcination, where the as-synthesized
HMSNs (with templates) subject to a very high temperature (>500 ˚C). However, during
high-temperature, irreversible particle aggregation occurs, leading to the formation of
nondispersible and conglutinated materials. Current efforts have been made in creating
the intrinsic space in HMSNs, but little attention has been paid to solve the problem of
particle dispersibility in solvents [182].
To obtain dispersible HMSNs, a new method was developed to synthesise hollow
mesoporous silica nanoparticles (HMSNs) by employing Eudragit S100 nanoparticles as
the dissolvable core templates. Both the Eudragit S100 core and the structure-directing
surfactant can be easily and synchronously removed through solvent extraction in acetone
rather than calcination.
Figure 4. 1 Fabrication possess of HMSNs
55
As a working model, the designed protocol as shown in Figure 4.1 depicts the fabrication
possess where Eudragit S 100 nanoparticles were employed as a core template and the
non-ionic surfactant Triton X-100 (TX100) was used as the pore template. Firstly, TX100
micelles were assembled on the surface of Eudragit S 100 nanoparticles. Subsequently,
TEOS was added as a silica source to form the mesoporous silica shell with embedded
TX100 micelles. TX100 micelles and Eudragit S 100 nanoparticles can be easily removed
by solvent extraction at the same time resulting in HMSNs with open mesopores and a
hollow cavity.
The microstructure of synthesized nanoparticles which are made up of a huge hollow
cavity and a mesopourous shell are proved by SEM and TEM images (Figure 4.2). The
spherical hollow interior can clearly be seen from the TEM images in Figure 4.2c-f, with
a shell thickness of 10 nm (Figure 4.2e). The synthesized particles were spherical in shape
with a mean diameter of 100nm. The rough surface of these hollow mesoporous silica
nanoparticles shown in Figure 4.2b further suggested their porous structures [183, 184].
56
Figure 4. 2 a, b & c) SEM images of HMSNs and d, e & f) TEM image of HMSNs displaying hollow interior
57
1 10 100 10000
5
10
15
20
25
Inte
nsity
(%)
Size (nm)
PDI: 0.048
Figure 4. 3 DLS curve of HMSNs
The particle distribution collected by dynamic light scattering (DLS) in Figure 4.3
definitely displays a narrow size distribution centred at 164 nm. The polydispersity index
(PDI) of these HMSNs is approximately 0.048, suggesting a high size uniformity in these
products [185]. It is found that the diameters showed here are slightly bigger than the size
was measured by TEM (~100nm). This is presumably due to the fact that the DLS
technique measured the particle size in solution where a hydrate layer exists on the
HMSNs, thereby producing a larger diameter than that in its dried form observed from
TEM or SEM.
The small angle X-ray diffraction (SAXRD) pattern of the synthesized HMSNs was
presented in Figure 4.4. Only one distinct broad Bragg diffraction peak was presented at
2θ=2-3 , which was the characteristic XRD pattern of a mesophase pore system [146,
186, 187]. The characteristic peak of mesoporous materials was centred at 2θ=2.1˚. This
result was in good agreement with other mesoporous silica nanoparticles synthesized by
nonionic surfactant templating [101, 118, 135, 188].
58
2 4 6 8
0
1100
2200
3300
4400
Inte
nsity
/(a.u
.)
2θ/degree
Figure 4. 4 Small angle XRD pattern of HMSNs
According to IUPAC classification, HMSNs exhibited a classical IV type N2 adsorption-
desorption isotherm with well-defined steps at relative pressures (P/P0) of 0.1-0.3 and 0.9-
1.0 corresponding to capillary condensation and desorption in open mesopores and
interstitial pores respectively [189]. The N2 adsorption-desorption isotherm for HMSNs
(Figure 4.5a) displayed two hysteresis loops at P/P0=0.3-0.5 and P/P0=0.9-0.96,
respectively, which further supports the XRD results that suggested HMSNs have a
mesoporous structure. The HMSNs were calculated to have a narrow pore size
distribution centred at 2.5 nm (Figure 4.5b). The specific surface area and pore volume
of HMSNs were 760m2/g and 0.45cm3/g, respectively.
59
0.0 0.2 0.4 0.6 0.8 1.0
650
700
750
800
850
P/P0=0.9
P/P0=0.96P/P0=0.5
P/P0=0.3
Vad
s/cm
3 g-1
P/P0
(A)
P/P0=0.1
5 10 15 20 25
0.0
0.5
1.0
1.5
2.0
dV/d
logD
(cm
3 /g)
pore size (nm)
(B)2.5nm
Figure 4. 5 Nitrogen adsorption-desorption isotherm (A) and the BJH pore size
distribution of HMSNs (B)
60
4.2.2 Innovations of Eudragit-assisted HMSNs fabrication
When compared to other conventional synthesis strategies of HMSNs, the use of Eudragit
as template is novel. Eudragit S100 is a co-polymer of methacrylic acid and methyl
methacrylate with the ratio of 1:2. It is stable under acidic pH (< 7) but becomes
dissolvable in base condition (pH >7) or in Acetone [190, 191]. Because of this property,
Eudragit S100 is often used as an enteric polymers for drug delivery [192-194]. By
applying Eudragit S 100 nanoparticles as a core template, the yield products HMSNs
don’t have to undergo the calcination process for the removal of the core template to
create a hollow cavity in HMSNs. Instead, Eudragit core can be easily removed by
soaking the as-synthesized HMSNs (HMSNs with Eudragit cores) in base solution or in
Acetone under room temperature. This solvent extraction process to clean the Eudragit
cores is facile and gentle, which can remarkably prevent the HMSNs from further damage
or aggregation caused by subjecting to a very high temperate. To our knowledge, this is
the first time to report the use of Eudragit S100 nanoparticles as a core template for
HMSNs synthesis. More importantly, Eudragit S100 is non-toxic, which is also superbly
useful for the fabrication of HMSNs as drug carriers.
To further demonstrate the above illustration, as-synthesized HMSNs (with Eudragit
cores) underwent solvent extraction in Acetone and calcination at 550 ˚C for 6 h
separately. Both treatments achieved the same goal, to remove the Eudragit cores, and the
yield products were named as HMSNs-S and HMSNs-C, respectively. The EM images
in Figure 4.6 definitely displayed the superiority of solvent extraction. It can be seen that
HMSNs-C were aggregated and some of them were damaged due to the high temperature.
In contrast, HMSNs-S exhibited an excellent dispersion revealing that the solvent
extraction method plays a crucial role in preparation of discrete HMSNs.
61
Figure 4. 6 TEM images (a) and SEM images (b & c) of HMSNs-S and TEM images (d) of HMSNs-C and SEM images (e & f)
62
-150 -100 -50 0 50 100 150
0
20000
40000
60000
80000
100000
120000-4.8 mV
Tota
l Cou
nts
Zeta potential (mV)
HMSNs-S HMSNs-C
-25.6 mV
Figure 4. 7 Zeta potential of HMSNs-S and HMSNs-C
In addition, the aqueous stability of HMSNs was thoroughly studied. The zeta potential
of HMSNs-S and HMSNs-C were -25.6 mV and -4.8 mV at pH 7.4, respectively (Figure
4.7). The higher surface charge of HMSNs-S was due to the abundant Si–OH groups on
the surface [195]. In comparison, when HMSNs-C were subject to calcination at 550 ˚C
for 6 h, some of the Si–OH groups condense with one another resulting in decreased Si–
OH groups on the surface and thus a decreased surface charge (-5.89 mV) was observed.
Figure 4. 8 Stability study of HMSNs-C and HMSNs-S
63
As shown in Figure 4.8, no obvious precipitation tendency of HMSNs-S was observed
over one week indicating its excellent stable dispersion in water. On the other hand,
HMSNs-C with a low surface charge exhibited decreased stability in aqueous solution,
which was proved by the presence of sediment at the bottom of the sample vial after 2
days of storage.
As such, it is demonstrated that the new Eudragit-assisted strategies for the fabrication of
HMSNs is superior to conventional methods used to fabricate HMSNs in preparation of
highly uniform, discrete and stable HMSNs, given that particle stability is one of the
critical considerations in the biomedical use of mesoporous silica nanoparticles.
Therefore, the aggregation state of HMSNs beginning at the synthesis process should be
taken into account before administration.
4.2.3 Mechanism of Eudragit-assisted HMSNs fabrication
Considering the successful results observed for Eudragit-assisted HMSNs fabrication, it
is important to further elucidate the fabrication mechanism. Generally, Eudragit S100
nanoparticles could be readily synthesized by nanoprecipitation [172] and TX 100
micelles are formed in water at a given concentration and temperature [196]. Hydrogen
bonds may be formed between Eudragit S100 nanoparticles and TX100 micelles due to
the existence of carboxyl groups in Eudragit molecules and hydroxyl groups in TX100
molecules.
If Eudragit S100 nanoparticles and TX 100 micelles are combined to form a core/shell
structure, hydrogen-bonding may be significantly complementary to the Coulombic
interaction between Eudragit S 100 nanoparticles and TX100 micelles. The interactions
between TX100 micelles and Eudragit S100 nanoparticles are proposed in Figure 4.9,
demonstrating the formation of TX100/Eudragit composite micelles. The hydroxyl
groups of surfactant micelles react with the carboxyl groups of Eudragit S100 particles to
form hydrogen bonds. These are critical properties for the formation of TX100 micelles
on the surface of Eudragit nanoparticles which finally form mesopores of HMSNs.
64
Figure 4. 9 Formation schematics and of TX100/Eudragit S100 composite micelles (a) and proposed reactions between TX100 and Eudragit S100 (equations 1-3)
To demonstrate this hypothesis, Eudragit S100 nanoparticles and TX 100 micelles were
prepared and their reactions were characterized by dynamic light scattering (DLS), FTIR
and PFG-NMR techniques. As shown in Figure 4.10, the formation of composite micelles
was tracked by DLS. Firstly, TX100 micelles with a diameter of approximately 8.7 nm
were formed in water. Subsequently, after adding Eudragit nanoparticles (diameter ~ 68
nm), the TX100/Eudragit composite micelles were formed with a diameter of ~ 92 nm
indicating the coupling between Eudragit nanoparticles and TX100 micelles.
From the FTIR results in Figure 4.11, the investigated composite micelles
(TX100/Eudragit S100) were found to display a broad and intense absorption (continuum)
in the 1500 – 800 cm-1 region, typical for short hydrogen bonds [197-201]. Broad
65
absorption arise from the bending and stretching vibrations of the O-H group involved in
the short OH….O hydrogen bonds. In addition, the red shift of the O-H stretch, which
varies from 3545 to 3500 cm-1, provides unambiguous information about the formation
of hydrogen bond [202, 203]. Thus, the extrinsic interaction between surfactant and
Eudragit led to the nanocoating of surfactant micelles around Eudragit nanoparticles
through self-assembly.
1 10 100 1000
0
5
10
15
20
25
Inte
nsity
(%)
size (nm)
Eudragit TX100 Eudragit/TX100
Figure 4. 10 DLS of Eudragit S100 nanoparticles, TX100 and their mixture
To provide more direct evidence of the interactions between Eudragit S100 nanoparticles
and TX100 micelles, 1H NMR spectra of Eudragit S100 nanoparticles and the mixture of
TX100/ Eudragit S100 were recorded in Figure 4.12 and the diffusion coefficients of
different species in both samples were listed in Table 4.1
66
0 500 1000 1500 2000 2500 3000 3500 4000 4500
Wavenumber(cm)
1730
1732 3500
Eudragit
TX100
Eudragit & TX100
3545
Figure 4. 11 FTIR spectra of Eudragit S100 nanoparticles, TX100 and the mixture of
Eudragit and TX100
As shown in Figure 4.12, the 1H NMR spectra of Eudragit S100 nanoparticles presented
two obvious peaks at 4.81 ppm (peak I) and 2.25ppm (peak II), which are assigned to the
residual water and acetone in the D2O solution, respectively. The three peaks labelled as
‘III’ are attributed to the Eudragit S100 molecules. The low intensity of the Eudragit peaks
is due to its poor solubility in D2O at room temperature. In Eudragit S100 nanoparticle
samples, the Eudragit S100 and the acetone molecules show similar diffusion coefficients,
which are largely different from the diffusion coefficient of H2O molecules. This result
suggests that the Eudragit molecules exist in the acetone phase, which is separated from
the bulk water phase. Remarkably, upon the addition of the TX100, the diffusion
coefficient of Eudragit S100 dropped by 75%, from 6.5×10-10 m2/s to 1.635×10-10 m2/s,
whereas the acetone molecules maintain a similar diffusion coefficient to that in the
system of pure Eudragit S100 nanoparticles. This is a strong indication of interactions
between the Eudragit and the TX100 molecules which have a much lower diffusion
coefficient (4.51 ×10-11 m2/s).
67
Figure 4. 12 1H NMR spectra of Eudragit S100 nanoparticles (upper) and Eudragit/TX100 mixture (bottom); Inserts are TX100 molecular structure.
Table 4. 1 Proton diffusion coefficients (D) of Eudragit and Eudragit/TX100 mixture
a The detailed assignment of each peak is shown in the inset of Figure 4.12. b The diffusion coefficient of TX100 molecules was obtained by fitting the sum of
the integrals of all the corresponding peaks to the diffusion equation.
Samples Peaks Assignment D (m2/s) SDb
Eudragit I H2O (in D2O) 1.22×10-9 1.22×10-3
II Acetone (in D2O) 7.25×10-10 4.65×10-3
III Eudragit 6.5×10-10 4.25×10-3
Eudragit/TX100 I' H2O (in D2O) 1.25×10-9 1.017×10-3
II' Acetone (in D2O) 7.74×10-10 5.32×10-3
III' Eudragit 1.635×10-10 7.4×10-3
1,2,3,4,5,6a TX100 4.51×10-11 b 6.1×10-3
68
The measured NMR diffusion coefficient of the Eudgragit in Eudragit/TX100 samples is
lower than that of the free Eudragit molecules (in the pure Eudragit sample), yet higher
than that of the TX100 molecules. This can be explained by the fast exchange process
between the free and associated Eudragit molecules. In the time scale of NMR
measurements (10 ms), the Eudragit molecules exhibit a very fast exchange between the
free and associated states. The measured diffusion coefficient (also termed as apparent
diffusion coefficient) is then a population-weighted value between the free and associated
values. Assuming that the associated Eudragit molecules are of the same diffusion
coefficient as the TX100 molecules (4.51 ×10-11 m2/s), it can be easily calculated that
about 19% of the Eudragit molecules are free and 81% of the Eudragit molecules are
associated with TX100 molecules in the equilibrium state.
Figure 4. 13 SEM images (a) and TEM image (b) of non-Eudragit assisted MSNs
To further confirm the effect of the coordination between Eudragit S-100 nanoparticles
and TX-100 micelles on the formation of the hollow mesoporous silica nanoparticles,
Eudragit S-100 was omitted during HMSNs fabrication while keeping other experimental
conditions the same. As seen from the SEM and TEM results (Figure 4.13), without
Eudragit S-100 nanoparticles, the mesoporous silica nanoparticles obtained did not
contain a hollow cavity. More importantly, the dispersibility, yield, size distribution and
overall quality of the structure of these nanopartilces were significantly inferior to the
HMSNs fabricated with the assistance from Eudragit. Thus, the role of Eudragit S-100 is
crucial in the formation of high quality mesoporous nanoparticles. More than being a
69
solid core particle to develop the hollow cavity, Eudragit S100 nanoparticles also assisted
in the cooperative assembly of TX100 micelles, which contributed to the excellent
morphology of the HMSNs.
4.2.4 5-FU encapsulation and the in vitro release
To assess the potential application of HMSNs in drug delivery, the drug loading and
release behaviour of HMSNs were investigated. 5-fluorouracil (5-FU) standing alone in
colorectal cancer therapy was chosen as a model drug for this study. The 5-FU loading
process predominantly relied on a physical adsorption mechanism in mesopores. The
results shown in Figure 4.14a indicated that HMSNs have a loading capacity of 194.5
mg(5-FU)/g(HMSNs) at 3 mg/mL of 5-FU loading solution. It has been found that a maximum
loading content can be obtained after 8 h.
Figure 4.14b presents the release of 5-FU from HMSNs in pH 5.5 and 8.0 PBS solutions.
The results demonstrated that the release pattern of 5-FU in both pHs is similar. There
were two stages during the release of 5-FU from HMSNs. The first stage of release was
initially rapid, and within this stage, nearly 23.38% and 26.58% of the 5-FU was released
into the aqueous media at pH 8.0 and 5.5, respectively. This may be due to the rapid
diffusion of the 5-FU molecules which were adsorbed on the surface and in pore entrances
of the HMSNs. When 5-FU loaded HMSNs were initially immersed into the PBS solution,
there is a concentration gradient of 5-FU between the PBS buffer and the HMSNs. Thus,
5-FU diffuses more quickly from the pores to the outside media driven by the diffusion
effect. After 1h, the second stage of release was relatively slow and the drug concentration
levelled off in 25 h which is an excellent release profile for cancer treatments.
70
0 5 10 15 20 25
0
10
20
Cum
ulat
ive
load
ed (%
)
Time (h)
(A)
0 1 2 3 4 15 20 25
0
10
20
30
40
50(B)
Time (hours)
Cum
ulat
ive
rele
ase
(%)
PH 8.0 PH 5.5
Figure 4. 14 (a):5-FU loading profile and (b): 5-FU release profile of HMSNs
71
On the other hand, when the pH was decreased from 8.0 to 5.5, the release rate was
accelerated and the total amount of 5-FU released increased. Near the isoelectric point of
5-FU (pKa = 8.0), the net charge of this drug molecule was decreased. Thus, electrostatic
attractions between the inorganic carrier and drug molecules were minimal [204].
However, at lower pH (pH =5.5) the elevated electrostatic repulsion would lead to a
higher transport of 5-FU through the mesoporous silica shell and thus a greater release
capacity [205]. The amount of 5-FU released from HMSNs in pH 8.0 and 5.5 PBS
solutions was ~37.98% and ~43.99% at 25h, respectively. These results illustrated that
these novel HMSNs exhibited great sustained release due to their mesoporous structure.
4.3 Pore size tuning for controlled release
An ideal delivery system is the one that can provide a considerably extended period of
effective drug delivery. The ideal slow release kinetic model [23] has three stages (Figure
4.15). The release rate in stage I is initially rapid and as a results, the drug concentration
in plasma can reach its therapeutic concentration rapidly. In stage II, the therapeutic
concentration is maintained for a period of time then drop off gradually to the minimum
effective concentration in stage III. The HMSNs based drug delivery system can exhibit
such a sustained release profile as demonstrated above.
Figure 4. 15 Three stages of the slow release kinetic model
I II III
therapeutic zone
72
However, with the further development of the cancer treatments, more requirements were
introduced to drug delivery systems. It was reported that, in addition to sustained release,
specific drug release rates are always pursued for personal treatment as different stages
of disease evolution need different drug release rates [206-208]. In this regard, smart
delivery systems with more effective control of drug release rate were of great importance.
Fortunately, the release rate of HMSNs based drug delivery system can be further
controlled and adjusted by changing the pore diameter (Figure 4.16). By varying the pore
size in the shell, the release rate of drug would be tuned to a specific rate for personal
treatments. It is anticipated that the release rate will increase along with the increase of
pore size because the space provided by small pores is limited for the diffusion of drug
molecules while large ones can supply bigger room for more drug molecules to go
through at the same time, which results in a quicker release speed.
Figure 4. 16 Representative schematic illustration of controlling drug release rate by
tuning the pore size of HMSNs
Three HMSNs with different pore size were synthesized using the same fabrication
system. The pore size of HMSNs was successfully tuned by using different surfactants
with different molecular weight (poly ethylene oxide block copolymers F127, P123 and
low molecular weight nonionic surfactant Tweens 20). The higher the molecular weight
73
of surfactant, the bigger the micelles would be form which were finally evolved to
mesopores of HMSNs.
The pore size of HMSNs can be investigated by a suite of characterisation techniques,
such as BET, XRD and PALS. These techniques provide different information whilst are
complementary to each other. BET and PALS were used to study the pore size distribution
of these three HMSNs (HMSN-1, HMSN-2 and HMSN-3). Combining these methods
would allow a complementary pore size analysis of materials.
4.3.1 BET analysis of pore size
0.0 0.2 0.4 0.6 0.8 1.0
Vad
s/cm
3 g-1
P/P0
HMSNs-2
HMSNs-3
HMSNs-1(a)
0 10 20 30 40 50 600
1
2
Por
e vo
lum
e/(c
m3 g
-1)
Pore size/nm
HMSNs-2 HMSNs-1 HMSNs-3
9.0nm
5.6nm(b)
Figure 4. 17 Adsorption-desorption isotherms (A) and corresponding pore size distribution (B) of HMSNs
Adsorption-desorption isotherms of HMSNs-1, HMSNs-2 and HMSNs-3 were recorded
in Figure 4.17a and the corresponding pore size distribution was shown in Figure 4.17b.
In addition, some physicochemical properties of HMSNs were listed in Table 4.2. The
adsorption-desorption isotherms (Figure 4.17a) of all HMSNs exhibit typical type IV
curves with well-defined capillary condensation steps. This suggested that all HMSNs
had mesoporous structure which was further confirmed by their high surface areas of 687,
589, 760 m2g-1and pore volumes of 0.73, 0.96, 0.45 cm3g-1 for HMSNs-1, HMSNs-2 and
HMSNs-3, respectively (Table 4.2). The pore size of HMSNs-1, HMSNs-2 and HMSNs-
3 was calculated and presented in Figure 4.17b. The appearance of sharp peaks in Figure
4.17b indicated that HMSNs-1 and HMSNs-2 had a majority of mesopores with the size
of 9.0nm and 5.6nm. However, for HMSNs-3, no peaks could be seen from Figure 4.17b.
74
The reason of this phenomenon may be that the average pore diameter of HMSNs-3 is
relatively small beyond the limit of this characterization. Compared with F127 and P123,
the molecular weight of Tweens 20 and TX100 is much smaller, which produces smaller
micelles during fabrication process and therefore results in smaller pore size of
nanoparticles. As pore size, surface area and pore volume of the final products highly
depend on the kind of surfactant used for fabrication, HMSNs synthesized with
surfactants having similar properties may have similar or even the same physicochemical
properties, including the surface area [209, 210].
Table 4. 2 Physicochemical Properties of HMSNs
Sample
name
surfactants BET
surface
area (m2/g)
Pore size
(nm)
Pore
volume
(cm3/g)
HMSNs-1 F127 589 9.0 0.96
HMSNs-2 P123 687 5.6 0.73
HMSNs-3 Tweens 20 760 - 0.45
4.3.2 PALS analysis of pore size
The PALS is another technique used to identify the porosity of mesoporous materials.
The results gained from PALS are presented in Table 4.3 which illustrates the multiple
scales of porosity readily detectable with positrons. It is known that for mesoporous silica
nanoparticles, τ3 is associated with micro-porosity in the amorphous silica network on
molecular scale (D=0.2-0.4nm) while τ4 provides information on bridging micropore
elements that connect mesopores (D≈1 nm). This information is not easily attainable from
BET results. τ5 is related to the mesopores (D=5-20nm) [211]. Figure 4.18 depicts the
relationship between porosity and τn.
75
Figure 4. 18 The relationship between porosity and τn [211]
Table 4. 3 PALS data for HMSNs
Sample τ3 (ns) D3
(nm)
I3 (%) τ4 (ns) D4
(nm)
I4 (%) τ5 (ns) D5
(nm)
I5 (%)
HMS-1
0.63
0.23
35.00
5.19
1.07
1.14
79.65
7.8
19.02
HMS-2
0.68
0.26
25.51
6.46
1.18
1.19
72.55
6.6
24.87
HMS-3
0.60
0.21
37.05
7.18
1.24
13.42
72.45
6.5
2.29
τn: o-Ps lifetimes; D: average free path of corresponding pore length; I: corresponding
intensity
76
As shown in Table 4.3, the pore size calculations were fitted to 3 other ortho-positronium
components that are associated with the pore size (τ3, τ4 and τ5). Each of the components
has an associated intensity value (I3, I4 and I5, relative number of pores) and a calculated
pore size (D3, D4, and D5) using the RTE spherical model. τ3 representing very small
micropores are very uniform between the samples, indicating that it is inherent to these
silica products. τ4 is also very small for pores with the diameter of around 1.20nm. It was
noticed that I4 of HMSNs-3 was much higher than that of other samples. This is a strong
indication of the fact that there is a large proportion of micropores existing in HMSNs-3
and this might be the reason for the absence of pore size distribution peak in Figure 4.17b.
Given that the pore size obtained from BET is calculated by BJH method whose
application is limited to mesopores (2 - 50nm) [212-214], micropores cannot be detected
via this method. The largest pore size D5 (calculated from Tau 5) is related to the pore
size from the BET data. When compared to BET results, the pore size D5 of HMSNs-1
and HMSNs-2 fluctuated to some extent. However only a range of pore sizes rather than
a uniform pore size can be seen from PALS [211]. In addition, it is noteworthy that I5 of
HMSNs-3 is only 2.29, so it is not a significant proportion of pores in this range for this
sample.
In conclusion, BET technique using BJH for pore size calculation is only effective to
evaluate the size of mesopores such as HMSNs-1 and HMSNs-2. When refers to material
with a large proportion of micropores like HMSNs-3, other methods (PALS et al.) should
be applied in the measurement of pore size. Further, the pore size of HMSNs can be
controlled by changing the surfactants used for synthesis. Under the reaction conditions
used here, Tweens 20 resulted in the formation of micropores (< 2nm), while P123 and
F127 tend to yield mesopores. The possible reason for this phenomenon would be
attributed to the surfactant behaviour during fabrication process.
4.3.3 Controlled and sustained release of HMSNs with different pore size
In order to investigate the release behaviour of HMSNs with different pore size, 5-FU had
been loaded into HMSNs-1, HMSNs-2 and HMSNs-3 to obtain 5-FU loaded HMSNs.
The loading capacity and some physical parameters are listed in Table 4.4. Three HMSNs
77
exhibit a similar loading capacity of around 10%. However, HMSNs are expected to
achieve a higher drug loading (20% ~ 50%) as they have the advantage of the high surface
area and high pore volume. The low drug encapsulation ability of HMSNs might be due
to the electrostatic repulsions between negatively charged silica nanoparticles and
negatively charged 5-FU molecules. Therefore the surface of the HMSNs needs to be
modified to improve 5-FU loading which will be specifically discussed in Chapter 6.
Table 4. 4 Loading capacity and other physical parameters of HMSNs
0 5 10 15 20 25
0
10
20
30
40
50
Cum
ulat
ive
rele
ase
(%)
Time (h)
HMSNs-3 HMSNs-2 HMSNs-1
Figure 4. 19 In vitro release profiles of HMSNs with different pore size in PBS with a pH of 5.0
The release of 5-FU from HMSNs with three pore sizes was examined in phosphate-
buffered solution (PBS) with a pH of 7.4. As evidenced in Figure 4.19, all HMSNs
displayed a sustained release and the release rate is related to the pore size of HMSNs. In
the first 2 hours, three HMSNs demonstrated a similar release rate because at the first
Samples Pore size (nm) Loading
capacity (%)
Zata potentials
(mV)
HMSNs-1 9.0 10.8 -5.32
HMSNs-2 5.6 9.3 -18.6
HMSNs-3 < 2 13.6 -27.1
78
stage of drug release (stage I presented in Figure 4.15), drug encapsulated on the surface
or at the entrances of pores are released. 5-FU loaded HMSNs have been washed at the
end of the drug loading procedure before freeze-drying. Therefore, some of the drugs
attached on the surface of HMSNs were washed away. However, the release of the drug
remained on the surface, in addition to those encapsulated in the pore entrance and near
the pore entrance leads to the rapid release behavior of HMSNs. At this release stage,
drug molecules were easily released from the HMSNs and the effect of pore size on drug
release is not significant. In addition, the concentration difference of the drug between
the environment (PBS solution) and inside the HMSNs is huge at this release stage, which
could account for the rapid release as well. As the size of the pores is much bigger than
that of the drug molecule (~0.5 nm), which provides enough room for the free
transportation of the drug, the drug can be encapsulated into the pores and the cavity of
HMSNs, not just attached on the surface. When the release proceeds further, drug stored
in the cavity of HMSNs passes through the pores to get released. As a result, pore size
becomes a more influential factor, determining the release speed of 5-FU. As expected,
HMSNs with larger pores present a faster release rate in addition to a higher cumulative
release rate. Within 24 hours, about 42.5%, 37.4% and 33.6% of 5-FU were released from
HMSNs-1, HMSNs-2 and HMSNs-3, respectively.
In the attempt to investigate the release kinetics of HMSNs, an empirical equation (Eq. 1)
was used to describe their release behaviour. Eq. 1 was introduced by Ritger and Peppas
in 1987 for description of solute release from non-swellable and swellable devices in
controlled release systems [215, 216].
(1)
The nature logarithm form of Eq. 1 is:
(2)
79
where Mt/M is the accumulative release percentage of 5-FU at time t; k is a constant and
n is the diffusional exponent indicating the transport mechanism of the drug from their
carriers. Table 4.5 lists the diffusional exponent values (n) and their corresponding drug
release mechanism based on literature [217, 218].
Table 4. 5 Values of diffusional exponent and corresponding drug release mechanism
Figure 4. 20 In vitro release kinetics of HMSNs with different pore size in PBS with a pH of 5.0
2.4
2.6
2.8
3
3.2
3.4
3.6
3.8
-1 0 1 2 3 4
ln (M
t/M∞)
lnt (hours)
HMSNs-1 HMSNs-2 HMSNs-3
Diffusional exponent, n
Drug release
mechanism Thin film Cylindrical
samples
Spherical samples
< 0.50
< 0.45
< 0.43
Quasi-Fickian
diffusion
0.50 0.45 0.43 Fickian diffusion
0.50< n < 1.00
0.45 < n < 1.00
0.43 < n < 1.00
Non-Fickian
diffusion
(Anomalous
transport)
1.00 1.00 1.00 Zero-order release
80
To obtain n values for three HMSNs, data extracted from drug release studies (Figure
4.19) were calculated according to Eq. 2. n of HMSNs (the slope) were graphically
determined by plotting versus ln t (Figure 4.20). The linearity of the regression
line was evaluated by the relevance coefficient (R2). As shown in Figure 4.20, the release
data of three HMSNs fit well to the Ritger-Peppas model which was evidenced by a high
R2 value (0.986, 0.981 and 0.943 for HMSNs-1, HMSNs-2 and HMSNs-3, respectively).
n is under 0.43 for all three HMSNs, demonstrating that three HMSNs follow the Quasi-
Fickian diffusion mechanism in which 5-FU molecules transport through the pores of
porous silica carriers without any relevant deformation and stresses during the whole
release process. Further, although the pore size regulates the speed of 5-FU molecules
passing through the pores, it shows no significant effect on their release mechanism. In
other words, the release rate of drug loaded HMSNs can be adjusted to a certain rate for
specific application while maintains their sustained release mechanism.
4.4 Cytotoxicity and non-specific cellular uptake of HMSNs
Figure 4. 21 Cytotoxicity of HMSNs on SW 480 cells at different concentrations
As nanocarriers for drug delivery systems, HMSNs are expected to be biocompatible with
100.00107.72 110.10
100.59
0
30
60
90
120
150
control 20 μg/ml 50 μg/ml 100 μg/ml
Cell
viab
ility
(%)
81
no toxic function on cancer cells. The biocompatibility of HMSNs was studied by an
MTT assay. Colorectal cancer cells SW 480 cells were cultured with HMSNs of varying
concentration for 24 hours. The cytotoxicity of HMSNs was expressed by the percentage
of cell viability compared to the control group which was not exposed to HMSNs (Figure
4.21). The viability of cells treated with HMSNs in the concentration range of 20 ~ 100
g/mL was similar to cells that were not exposed to HMSNs, demonstrating that HMSNs
are non-toxic and biocompatible with SW 480 cells. Therefore, HMSNs are a potential
nanocarrier for drug delivery systems to colorectal cancer cells.
To evaluate the natural ability of HMSNs to enter cancer cells, the cell uptake of FITC
modified HMSNs by SW480 cancer cells was investigated in vitro by flow cytometry and
fluorescence microscopy. After cell attachment in a 24 well culture plate at 37 C in 5%
CO2 for 24 hours, the SW480 cells (100000/well) were treated with FITC-HMSNs of
different concentration (2, 5, and 10 μg/mL) for various incubation times (30 min, 2 h
and 4 h).
As shown in Figure 4.22a, with increased HMSNs concentration and incubation time,
more particles were found to be internalised into SW480 cells, indicating that cell
internalisation of HMSNs are concentration and time dependent. The confocal
microscopy images and the corresponding 3D microscope projections showed in Figure
4.22b-i proved that the FITC-HMSNs can penetrate the plasma membrane of SW480 cells
and translocate into the cytoplasm. However, the uptake of HMSNs by SW480 cells is
relatively low, reaching a maximum of 50.8 % cells associated with HMSNs at 10 μg/mL
and 4 h incubation time (Figure 4.22a). Thus it is important to develop targeted drug
delivery systems that will improve the internalisation rate.
82
Figure 4. 22 Cell uptake of FITC-HMSNs with varying time and concentration by SW480 cells (A), confocal laser scanning microscope (CLSM) images of SW480 cells cultivated with FITC-HMSNs (10 μg/mL) for 4 hours (B-G) and corresponding 3D microscope projections (H-I). Blue: the nucleus (B & E); green: FITC (C & F); merging of image B and C (D); merging of image E and F (G); Scale bars are 40 μm for images B, C and D and 10 μm for images E-I.
1.7 2.21 3.393.05
9.29
28.5
7.99
21.3
50.8
0
10
20
30
40
50
60
2 5 10
% c
ells
asso
ciat
ed w
ith H
MSN
s
Concentration (μg/mL)
30 min
2 h
4h
B C D
E F G
A
83
4.5 Conclusions
Hollow mesoporous silica nanoparticles with large internal cavities were synthesized
through a novel and facile method by using Eudragit S100 nanoparticles as both a core
template and an assistant in the self-organization of surfactant micelles. The aggregation
problem has been effectively depressed through this “Eugragit assisted” strategy. The
HMSNs were found to exhibit a high drug loading and sustained release of chemotherapy
drug 5-FU. The HMSNs followed the Quasi-Fickian diffusion mechanism in which 5-FU
molecules transport through the pores of porous silica carriers without any relevant
deformation and stresses during the whole release process. Further, pore size is an
important parameter to regulate the release speed of 5-FU molecules from HMSNs. More
importantly, the synthesized HMSNs were found to exhibit nontoxic to colorectal cancer
cells and can enter colorectal cancer cells with a concentration and time dependent
manner. The successful synthesis of HMSNs in addition to their excellent
biocompatibility demonstrate the potential of HMSNs as effective drug carriers for
building versatile cancer-targeted drug delivery systems.
84
Chapter 5 Modification of HMSNs for targeting delivery
5.1 Introduction
Nanoparticles have been extensively studied as drug carriers for cancer treatments [6, 20,
219], because they can be passively accumulated in tumour tissues due to a phenomenon
called the enhanced-permeation end-retention effect (EPR) [6, 7]. However, this effect is
limited just to vascularized tumours and is usually not sufficient for a complete
eradication of the cancer [8]. Numerous research groups are therefore trying to develop
an efficient system for the active targeting of cancer cells by binding various ligands on
the surface of the nanoparticles [9-11]. Such systems could provide a specific
accumulation of the nanoparticles in targeted cells and thus represent a powerful tool to
improve the efficacy of the cancer treatment.
The over expression of certain receptors on the surface of tumour cells can be exploited
for targeting by drug carriers. EGFR, as one of receptors, has shown to involve in cellular
responses including proliferation and metastatic spread [49]. As a result, EGFR is one of
the most heavily investigated tumour targeting biomarkers for constructing nanoparticles
for targeted cancer therapeutic applications [7, 11, 220]. EGF, hepatin-binding EGF (HB-
EGF), betacellulin, transforming growth-factor-a (TGF-a), epiregulin and amphiregulin
are six known endogenous ligands of EGFR. Although any of these ligands can be used
as a targeting agent for EGFR, EGF with a good compromise between the binding affinity
and the size of the molecule (EGF, 6.1 kDa) is one of the most commonly detected factors
in humans [8]. It is hypothesized that EGF as a ligand to EGFR on the surface of
nanocarriers will exhibit specific delivery of anticancer drug, 5-FU, into colorectal cancer
cells, and thus enhance the resultant anti-cancer activity and reduce the side effects of
anticancer drugs.
85
Several surface modification methods have been implemented to covalently couple EGF
to the surface of nanoparticles [221] and carbodiimide chemistry has been found to be a
preferential method for the binding of EGF to silica nanoparticles [75, 221, 222]. For a
HMSNs based drug delivery system, the challenge lies in successfully maintaining an
intact structure while combining the targeting ligands and drugs into nanocarriers. In
many cases, HMSNs based targeted delivery systems end up with collapsed HMSNs
because too many functionalization steps were applied and each of them might involve
several times of centrifugations and re-dispersions which negatively affect the stability
of HMSNs. Therefore, facile method with less steps for EGF immobilization is of great
importance.
In order to protect the structure of HMSNs, a facile, single step strategy for EGF grafting
will be introduced, where EGF is covalently linked to the surface of amine functionalized
HMSNs by using aqueous 1-(3-(dimethylamino)propyl)-3- ethylcarbodiimide
hydrochloride (EDC) and N-Hydroxysuccinimide (NHS) as promoters. The covalent
attachment of the EGF to the particle surface was first confirmed by time of flight
secondary ion mass spectrometry (ToF-SIMs). Such functionalized particles are proved
by cell experiments to be suitable candidates as drug carriers for targeting delivery due
to the combination of the large loading capacity and the targeting function. Importantly,
the control of the quantity of EGF attachments on the surface of HMSNs was achieved
by changing the EGF concentration during the grafting. An increase in EGF on HMSNs
will increase the targeting efficiency to colorectal cancer cells.
5.2 A facile, one step strategy for EGF labelling
The covalent binding of proteins, such as EGF, to nanoparticles is highly challenging
since their active conformation must be preserved during the binding process, which
limits the choice of the reaction conditions. Several surface modification methods have
been implemented to covalently couple EGF to the surface of nanoparticles [221] and
carbodiimide chemistry is found to be the most effective method for the binding of EGF
to silica nanoparticles [75, 221, 222].
86
Carbodiimide chemistry is to generate crosslink between carboxylic groups and primary
amines using EDC and NHS and is a versatile and powerful tool for preparing
biomolecular probes, crosslinking peptides and proteins, and immobilizing
macromolecules for use in numerous cell biology detection and protein analysis. EGF as
a protein, contains both carboxylic acids and primary amines (C-and N-termini, and also
in amino acids presenting in side-chains of EGF). Thus, HMSNs bearing either amine or
carboxyl groups can be easily crosslinked to EGF by the utility of EDC and NHS.
However, most previous research only focused on generating carboxyl functionalized
silica nanoparticles for the binding of EGF [75, 221, 222] and the comparatively simple
route of preparing amine functionalized ones were not reported. In particular, HMSNs
with a hollow cavity are often obsessed with the problem of collapse when undergoing
too many times of centrifugation or modification steps. Compared to amine group
functionalization, carboxyl group functionalization of silica nanoparticles is relatively
complicated as it involves more than one step [223, 224]. For example, Kralj et al
achieved carboxyl-functionalized nanoparticles using a follow-up reaction of amine-
functionalized surfaces [225] and Yiu et al firstly modified silica nanoparticles with
cyano groups, then oxidated them to carboxyl groups [226]. As a result, carboxyl
functionalization of silica nanoparticles is more time consuming and might negatively
affect the physicochemical properties (such as suspension stability) of the products. To
overcome these shortcomings, amine modification was employed to produce HMSNs
with functional groups while maintaining an intact structure for EGF labelling.
5.2.1 Synthesis of amine functionalized HMSNs (HMSNs-NH2)
HMSNs-NH2 were synthesized by co-condensation method in which NH2 groups were
introduced during the fabrication process. This results in a more homogeneous coverage
of functional groups onto HMSNs. As shown in TEM and SEM images (Figure 5.1), all
morphologic features in plain HMSNs are remained in HMSNs-NH2, such as the hollow
structure (Figure 5.1c-f), the particle diameter (~150 nm), and a good dispersion, making
them very suitable for targeted drug delivery systems. The deliberately broken HMSNs-
NH2 shown in Figure 5.1c and f further proved their hollow structure [102]. It is
87
noteworthy that most HMSNs-NH2 have intact structure (Figure 5.1a and d). Similar to
HMSNs’ characterization, XRD and BET were performed to confirm the mesophase of
HMSNs-NH2. A broad Bragg diffraction peak within the range of 2θ=0-3˚ was found in
Figure 5.2b, illustrating that HMSNs-NH2 have a mesoporous structure [146, 186, 187].
BET results were collected and presented in Figure 5.2a, in addition to a classical IV type
isotherm, a hysteresis loop at relative pressures (P/P0) of 0.45-0.9 was observed which
further proved the mesoporous structure of HMSNs [189]. The pores of HMSNs-NH2
have a narrower size distribution centred at 3.0 nm (Figure 5.2c). The specific surface
area and pore volume of HMSNs-NH2 are 544m2/g and 0.57cm3/g, respectively.
It is noteworthy that, compared to Figure 4.4, the Bragg diffraction peak of HMSNs-NH2
(Figure 5.2b) shifted to a lower angle, which could be due to the grafting of –NH2
functional groups. According to the TEM, SEM and XRD results, the introduction of –
NH2 onto HMSNs didn’t have significant influence on the porous structure of HMSNs
which is used for the drug loading and controlled release. However, due to the coverage
of –NH2, slight decrease in the surface area has been identified via BET results.
88
Figure 5. 1. a, b & c) SEM images of HMSNs-NH2 and d, e & f) TEM image of HMSNs displaying hollow interior; c & f are broken HMSNs-NH2 showing their hollow structure
89
0.0 0.2 0.4 0.6 0.8 1.00
10
20
30
40
50
V ads/c
m3 g-1
P/P0
A
Figure 5. 2 A) Nitrogen adsorption-desorption isotherm; B) Small angle XRD pattern of HMSNs-NH2; C) BJH pore size distribution.
0 2 4 6 8 10
0
1000
2000
3000
4000
5000
Inte
nsity
/(a.u
.)
2θ/degree
B
0 5 10 15 200.0
0.5
1.0
1.5
2.0
dV/d
logD
Pore size (nm)
C
90
NH2 groups were qualified via FTIR and quantified by thermogravimetric analysis (TGA)
and NMR techniques. FTIR spectra of HMSNs-NH2 and the control sample (HMSNs)
are shown in Figure 5.3. Both samples have strong bands located at around 460 cm-1 (δO–
Si–O), 1100 (ʋasym. Si–O–Si), and 3300-3500 cm-1 (ʋSi–OH), confirming the presence of the
SiO2 inorganic phase. In particular, the bands with the highest intensity, centred at 1089
and 1073 of HMSNs and HMSNs-NH2, respectively are associated with the SiO2
stretching vibration [227]. The NH2 functionalization is clearly evidenced by the bands
detected at around 1500 cm-1, which is unambiguously attributed to the NH bending
vibration [227-230]. Results obtained by FTIR prove the effective NH2 functionalization
of mesoporous silica nanoparticles.
0 1000 2000 3000 4000
3311
1089
465
Wavenumber(cm-1)
HMSN HMSN-NH2
NH2
460
1073
3445
Figure 5. 3 FTIR spectra of HMSNs-NH2 and HMSNs
The TAG thermograms and 29Si single pulse excitation (SPE) spectrum were given in
Figure 5.4 for the evaluation of NH2 quantity. A weight loss of 14.6% was found in
HMSNs while HMSNs-NH2 presented an increased weight loss percentage of 21.5%
(Figure 5.4a). The higher weight loss of HMSNs-NH2 is due to the degradation of grafted
functional groups, inferring that around 5.9% (w/w) of HMSNs-NH2 was related to NH2
immobilization. The 29Si SPE of HMSNs-NH2 spectrum is shown in Figure 5.4b in which
Qm (m=2-4 indicating 2, 1 or 0 –OH groups are linked to Si) and T signals provide
91
information of silanol groups and grafted functional groups (-NH2), respectively [228,
231]. The peaks at -110, -100 and -90 ppm are associated to Q4 [(SiO)4Si], Q3
[(SiO)3SiOH] and Q2 [(SiO)2Si(OH)2], given that –OH groups of the HMSNs react with
amine silane (APTES) for –NH2 immobilization which results in a decrease in the number
of –OH groups on HMSNs. Thus, the presence of signals with decreased amount of –OH
groups (Q4 & Q3) in addition to T signal can reveal the successful grafting of –NH2 [231-
235]. The relative concentration of -NH2 in HMSNs (15%) was calculated by taking the
signal summation (T + Q2 + Q3 + Q4) as 100%.
100 200 300 400 500 600 7000
20
40
60
80
100
Wei
ght (
%)
Temperature (0C)
HMSNs HMSNs-NH2
weight loss:HMSNs 14.6%HMSNs-NH2 21.5%
(A)
100 50 0 -50 -100 -150 -200
T
Q4
Q3
Q2
46%
34%
5%
Chemical Shift (ppm)
15%
(B)
Figure 5. 4 TGA thermograms of HMSNs & HMSNs-NH2 (a) and 29Si single pulse excitation (SPE) spectrum of HMSNs-NH2
92
5.2.2 Preparation of EGF grafted HMSNs and characterization of EGF labelling
Figure 5. 5 Schematic diagram showing the process of EGF grafting onto HMSNs
A facile, single-step process was developed to graft EGF onto HMSNs (Figure 5.5).
After grafting, the characterization of EGF attachment is a crucial subsequent step.
However, since the amount of EGF on HMSNs is very low, the determination of EGF
immobilization emerges to be a big challenge as most characterization methods, such as
thermo-gravimetric analysis (TGA), energy-dispersive X-ray spectroscopy (EDX),
electron energy loss spectroscopy (ELLS) and even X-ray photoelectron spectroscopy
(XPS) have their limit. As EGF is used as a targeting ligand to target cancer cells [75],
cell and/or even animal experiments were carried out by researchers to evaluate the
effectiveness of EGF and thus validate their success and effectiveness of the grafting
[236-238]. For example, comparison between EGF and non-EGF grafted samples in
cellular uptake by cancer cells was largely used in the illustration of effective EGF
grafting. Even though these passive methods are useful, they cannot provide direct
information on stability and density of targeting ligands on the surface of nanoparticles
and thus fail to control the quantity of targeting molecules on nanoparticles for specific
application.
Time of flight secondary ion mass spectrometry (ToF-SIMS) as a highly specific
analytical tool in the study of proteins, has been shown to be a powerful technique in the
detection of extremely low amount of proteins which were undetectable by XPS [239].
Inspired by the high chemical selectivity and surface sensitivity of ToF-SIMS [239-242],
it was hypothesized that ToF-SIMS would allow for the observation of successful EGF
attachment to the HMSNs in addition to the quantity of the EGF immobilization.
93
In this regard, ToF-SIMS analysis was performed on HMSN surfaces before and after
EGF grafting. Amino acids presenting in the backbone of proteins such as EGF
are cracked into smaller immonium ions (amino acid mass fragments) during the
measurement of TOF-SIMS. Identification of characteristic amino acid mass fragments
by ToF-SIMS lead to its extremely high sensitivity (better than XPS) for the detection of
proteins [240].
Figure 5.6 displays the positive survey ToF-SIMS spectra (m/z 0 - 200) for the
unmodified HMSNs, amine group grafted HMSNs (HMSNs-NH2), EGF and HMSNs
labelled with EGF (HMSN-NH2-EGF). For the HMSN surfaces (Figure 5.6a) the intense
ion signals at nominal m/z 28, 29 and 45 amu can be attributed mainly to Si+, SiH+ and
SiHO+ fragments [239]. Some peaks raised by the amine groups on HMSNs-NH2 surface
are shown in Figure 5.6b. When compared to HMSNs, HMSNs-NH2 have new peaks at
15, 41, 43, 55, 57 and 69 amu which correspond to HN+, C3H5+, C2H5N+, C3H5N+,
C3H7N+ and C3H3NO+ fragments, confirming the successful grafting of NH2.
A control spectrum of the EGF is presented in Figure 5.6c. The main peaks at nominal
m/z 15, 27, 41, 43 and 55 happen to overlap with those of HMSNs-NH2. However in
addition to these peaks, intense peaks at nominal m/z 30, 70, 72, 73, 84, 86, 110 and 130
are observed which correspond to CH4N+, C4H8N+, C4H10N+, C2H7N3+, C5H10N+,
C5H12N+, C5H8N3+ and C9H8N+ and can be attributed to fragments from the EGF
backbone [242].
Figure 5.6d presents the ToF-SIMS spectra for EGF grafted HMSN surfaces. Ion signals
of the major fragments for both HMSNs and EGF were detected. For example, ion signals
of unmodified HMSNs reported above at nominal m/z 28, 29 and 45 amu can be seen. In
addition, the ion signals at nominal m/z 30, 70, 72, 73, 84 and 86 amu which were
assigned to fragments of EGF from the spectra of pure EGF (Figure 5.6c) are present
thereby suggesting successful EGF attachment. Interestingly, besides these characteristic
ion signals of HMSNs and EGF, new intense ion signals at nominal m/z 58 and 129 were
detected which are related to C3H8N+ and C5H13N4+ fragments. After grafting onto the
HMSNs surface, the conformation and orientation of EGF might change which can
subsequently lead to different distributions of the amino acid fragments detected in a
94
static ToF-SIMS experiment [241]. Therefore, these new peaks could be seen as
characteristic peaks of EGF grafting using the methods we illustrated above. These ToF-
SIMS results demonstrate the successful EGF labelling of HMSNs.
Figure 5. 6 Positive survey mass spectra for: (A) HMSNs surface; (B) HMSNs-NH2 surfaces; (C) EGF; (D) HMSNs-NH2-EGF.
95
5.2.3 The effect of EGF concentration on surface chemistries of the HMSNs
The stability and density of EGF on the surface of HMSNs may play a very crucial role
in targeting the cancer cells, though only one EGF molecule is needed to bind to the EGFR
on the cell to initiate internalisation. However, the more EGFs grafted onto the HMSNs,
the higher the probability of HMSNs-NH2-EGF targeting the cancer cells. It is
consequently important to quantify the EGF on the surface of HMSNs. In this regard,
EGF grafting was explored by varying the concentration of EGF during the attachment
procedure from 0.1 to 0.8 mg/ml and analysing the surfaces with ToF-SIMS (Figure 5.7).
Figure 5. 7 Positive survey mass spectra of HMSN-NH2-EGF surface using different EGF concentration: 0.1,0.2,0.4 and 0.8 mg/ml.
96
Figure 5. 8 Plot of ratios of the intensities of the 28 m/z (from HMSNs) to 59 m/z, 70 m/z and 58 m/z (from EGF)
It appears that the EGF concentration has a limited effect on the resulting surface as no
significant differences can be indentified between samples prepared with 0.1, 0.2, 0.4 and
0.8 mg/ml of EGF (Figure 5.7). Differences, however, emerge upon close inspection of
the ratios of 28/59, 28/70 and 28/58 mass fragments. 28 m/z is an ion signal specific for
HMSNs while 59 m/z, 70 m/z and 58 m/z are specific for EGF, thus the ratios of 28/58,
28/70 and 28/58 track the relative amount of EGF grafted on the surface of HMSNs.
Figure 5.8 displays that the ratio of 28/59 decreases with an increase in EGF concentration,
therefore suggesting that an increase in EGF concentration can lead to a greater amount
of EGF to be attached onto the surface of HMSNs. As ToF-SIMS is a very surface specific
technique (penetration depth < 1.5 nm) [241], an increase in EGF surface coverage would
result in less silica but more protein being detected, thus the ratio of 28/59 decreased.
Moreover, the same trend can be seen for the 28/70 and 28/58 ratios.
The differences between the samples shown in Figure 5.7 are developed from comparing
only a few peaks selected from ToF-SIMs spectra and the information contained in
unselected peaks is disregarded. Although this univariate analysis is useful, it is not
enough to properly evaluate the EGF attachment of these surfaces, because the remaining
peaks can provide important information for fully classifying the differences of surface
chemistries between samples. Thereby, PCA was employed to extract the most influential
factors from the complex mass spectra and to help in data interpretation. In the attempt to
0
2
4
6
8
10
12
14
16
0 0.2 0.4 0.6 0.8 1
28/59 28/70 28/58
97
classify the TOF-SIMS data by PCA, 35 peaks listed in Table 5.1 were used. The selection
of these peaks is based on the assignments taken from literature [243-246]. Only the
characteristic nitrogen containing peaks were studied which arise from amino acid
fragments ions so they are indicators of EGF immobilization.
Table 5. 1 Positive fragment ions (immonium ions) used in PCA
NO. Fragment m/z No Fragment m/z
1 CH4N+ 30 18 C4H10N+ 72
2 C2H4N+ 42 19 C2H7N3+ 73
3 CH3N2+ 43 20 C3H8NO+ 74
4 CH4N2+ 44 21 C4H6NO+ 84
5 C2H6N+ 44 22 C5H10N+ 84
6 C2H3S+ 47 23 C5H12N+ 86
7 C3H4N+ 54 24 C3H7N2O+ 87
8 C3H6N+ 56 25 C3H8NO2+ 88
9 C2H4NO+ 58 26 C4H10N3+ 100
10 C3H8N+ 58 27 C4H11N3+ 101
11 CH5N3+ 59 28 C4H8NO2+ 102
12 C2H5S+ 61 29 C7H7O+ 107
13 C4H6N+ 68 30 C5H8N3+ 110
14 C4H5O+ 69 31 C8H10N+ 120
15 C3H4NO+ 70 32 C5H11N4+ 127
16 C4H8N+ 70 33 C5H13N4+ 129
17 C3H3O2+ 71 34 C9H8N+ 130
35 C9H8O+ 132
98
Figure 5.9 displays the scores plot of positive fragment ions on PC1 and PC2 for the EGF
grafted HMSN samples and their corresponding loading plots show the influence of each
amino acid peak on PC1 and PC2. Each sample was characterised by 35 groups of positive
fragment ions derived from replicated positive ion mass spectra. About 91 % of the total
variances were captured by PC1 and PC2 revealing that these two-PC’s models retain the
most of the original information.
-0.08 -0.04 0.00 0.04 0.08-0.12
-0.08
-0.04
0.00
0.04
0.08
EGF-0.1 EGF-0.2 EGF-0.4 EGF-0.8
PC2
(34.
4%)
PC1 (56.7%)
(a)
-0.015
-0.010
-0.005
0.000
0.005
0.010
0.015
0.020
0.025
C4H8N+
CH5N3+
C3H8N+C2H4N
+
Load
ing
on P
C1
(56.
72%
) CH4N+ (b)
-0.025
-0.020
-0.015
-0.010
-0.005
0.000
0.005
0.010
0.015
0.020
0.025
C3H8N+
Load
ing
on P
C2
(34.
44%
)
CH4N+
(c)
Figure 5. 9 Scores on PC1 and PC2 (a) and corresponding loading plots on PC1 (b) and PC2 (c) of immonium ions from static positive spectra of HMSNs-NH2-EGF prepared
with EGF concentration of 0.1, 0.2, 0.4 and 0.8 mg/ml
99
As shown in Figure 5.9a, with an increase in EGF concentration, a decrease of PC1 scores
was observed. The EGF concentration of 0.1 and 0.2 mg/ml groups having the highest
PC1 scores overlaps with both PC’s, which reflects their similar surface chemistries in
terms of nitrogen species. Marked differences are observed when the EGF concentration
is increased further. When the concentration increases to 0.4 mg/ml, the samples have
negative scores on both PC1 and PC2 and that are separated from low concentration ones
(0.1 and 0.2 mg/ml). In contrast, the samples with 0.8 mg/ml EGF concentration have
negative scores on PC1 but positive scores on PC2 and that are well separated from other
three samples, signifying that they have quite different surfaces. This may be due to more
EGF proteins being attached onto the HMSNs surfaces. Figure 5.9b and c indicate a high
negative loading of C4H8N+ (at m/z 70 amu) on PC1 and C3H8N+ (at m/z 58 amu) on PC2,
which reflects these two fragments contribute the most to the differences between samples
with different amount of EGF attachments.
In summary, PCA on the ToF-SIMS data demonstrates EGF has been attached onto the
NH2 modified HMSNs surface and different concentration of EGF grafting leads to
different surface chemistries of the final products.
5.2.4 Analysis on immonium ions distinguishing samples with different surface chemistry
The loadings of immonium ions on PC1 and PC2 point at the peaks (CH4N+, C2H4N+,
C3H8N+, CH5N3+ and C4H8N+) that discriminate the groups (Figure 5.9 b and c). Therefore,
the relative intensities of these ions, which have the most impact on distinguishing
samples with different surface chemistry, are closely related to the amount of EGF that
has been grafted onto the nanoparticles. The detailed effect of EGF concentration on the
amount of EGF attachments can be observed by comparing the relative intensities of these
main immonium ions between different samples. In order to gain an acute insight into the
quantity of EGF on samples, Si+ and SiOH+ fragments are included in the analysis. Thus,
each sample was characterised by 7 groups of immonium ions derived from replicated
positive static secondary ion mass spectra.
100
Figure 5. 10 Comparison of characteristic immonium ions intensities between samples
Shown in Figure 5.10, unmodified HMSNs had significantly high Si+ and SiOH+
intensities but low immonium ions intensities (almost zero), consistent with the fact that
no EGF protein is present on the surface of this sample. NH2 functionalized HMSNs
present a similar trend to unmodified HMSNs, however these samples have relatively
high CH4N+ and C2H4N+ signals due to the presence of amine functional groups on the
surface. On the contrary, all EGF grafted samples had enhanced immonium ion signals
and decreased silica intensities. Further, with the increase in EGF concentration, the Si+
and SiOH+ intensities decreased inferring that EGF concentration had a positive effect on
facilitating the EGF grafting.
Significant differences can be seen by close inspecting the changes of immonium ions
signals between samples. Firstly, CH5N3+ and C4H8N+ fragments are the EGF indicators
since HMSNs-NH2 didn’t show such fragments. The intensities of these two immonium
ions increased along with an increase in EGF concentration revealing more EGF has been
grafted onto the surface of HMSNs, which is consistent with the gradual decrease of silica
signals. If the intensities of these two fragments of pure EGF were set as 100%, at the
highest EGF concentration (0.8 mg/ml), the detected intensities of these two fragments
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
CH4N C2H4N C3H8N CH5N3 C4H8N Si SiOH
Nor
mal
ised
cou
nts
+SIMS Species
EGF ProteinPlain HMSNsNH2 functionalised HMSNsNH2 +EGF 0.1NH2 +EGF 0.2NH2 +EGF 0.4NH2 +EGF 0.8
0
0.01
0.02
CH5N3 C4H8N
101
of HMSNs-NH2-EGF are 66.7 % and 78.6%, respectively revealing more than 50% of
EGFs was grafted on HMSNs. Secondly, the CH4N+ and C2H4N+ fragments presented in
EGF and the amine group grafted samples varied differently. Along with an increase in
EGF concentration, although a gradually growing trend was not observed with these two
ions signal intensities, four EGF grafted HMSNs showed improved signals compared to
pure EGF and HMSNs-NH2 samples. These enhanced intensities were attributed to the
synergy of joint signals of CH4N+ and C2H4N+ generated by EGF and free amine groups.
Lastly, C3H8N+ fragment was uniquely shown in EGF grafted samples.
ToF-SIMS combined with PCA were proved in this study to be able to evaluate the
binding efficiency between EGF and silica nanoparticles. This method could be extended
as a general approach to analyse protein biomarkers in targeted drug delivery systems.
5.3 Comparison of newly developed grafting method with other
methods
5.3.1 Superior EGF grafting efficiency by the new grafting method
As mentioned before, HMSNs with a cavity inside tend to form a defect structure as too
many functional steps or harsh experimental procedures (such as high temperature) are
involved in the process which will cause the collapse of HMSNs, a main issue that need
to be considered. Therefore, in order to protect HMSNs from damage, a more facile
process with fewer steps to graft EGF onto HMSNs was developed. To compare, EGF
grafted HMSNs were prepared via two routes and final products were named as HMSN-
COOH-EGF (produced from commonly use route) and HMSN-NH2-EGF (fabricated
from the newly developed route) and characterized through ToF-SIMS (Figure 5.11). As
shown from ToF-SIMS results, ion signals of major fragments of both silica and EGF
were detected in both HMSN-COOH-EGF and HMSN-NH2-EGF. The intense ion signals
at nominal m/z 28 can be attributed to Si+, while peaks at nominal m/z 58, 70, 71, and 72
amu correspond to EGF. Importantly, although EGF can be attached to HMSNs through
either of the routes, it is worth noticing that HMSN-NH2-EGF have demonstrated
enhanced EGF signals and decreased silica signals inferring more EGF were introduced
102
to covered the surface of HMSNs. Therefore, less silica signals were detected in this
sample. This result proved that new EGF immobilization method has a superior EGF
grafting efficiency when compared to the commonly used ones.
Figure 5. 11 Routes of EGF lebelling based on carbodiimide chemistry: (A) A commonly used route; (B) Newly developed route with reduced steps to protect HMSNs from damage and Positive survey mass spectra for: (C) HMSN-COOH-EGF surface; Insets: High mass resolution of static ToF-SIMs in the 70-72 amu region. (D) HMSNs-NH2-EGF surfaces;
Further, SEM images were taken to study the morphology of EGF grafted HMSNs
through different routes (Figure 5.12). After EGF grafting, HMSN-NH2-EGF remain the
intact structure while HMSN-COOH-EGF show broken particles. This is due to more
modification steps applied and thus more centrifugation rounds were involved in route 1
(Figure 5.11).
103
Figure 5. 12 SEM images of HMSN-COOH-EGF (a) and HMSN-NH2-EGF (b)
5.3.2 Increase EGF labels on HMSNs for enhanced targeting efficiency
The use of TOF-SIMs provides opportunities to determine relative EGF concentration on
HMSNs. Therefore, the amount of EGF grafted onto HMSNs was compared by changing
the EGF concentration during modification process. In 5.2.3 and 5.2.4, it was reported
that increasing EGF concentration can result in higher EGF attachments on nanoparticles
via route 2 illustrated in Figure 5.11. Again, 4 different EGF concentrations for EGF
grafting (0.1, 0.2, 0.4 and 0.8 mg/ml) were compared based on carboxyl group
functionalized HMSNs (route 1 in Figure 5.11). Unfortunately, no differences can be
found from these four spectra (Figure 5.13a). Moreover, as shown in Figure 5.13b,
although different EGF concentration were employed, all samples were overlapped along
both PC’s indicating their similar surface chemistries (a comparison to Figure 5.9). From
Figure 5.13a and b, 4 important ion signals were selected: 28 m/z (specific for HMSNs),
30 m/z, 44 m/z and 58 m/z (correspond to EGF). The ratio of silica to protein: 28/30,
28/44 and 28/58 of different EGF grafted samples was calculated and presented in Figure
5.13c (a comparison to Figure 5.8). When the EGF concentration was increased gradually
from 0.1-0.8 mg/ml, rather than a regular trend, the ratio of 28/58, 28/44 and 28/30
fluctuated. Thereby an effective control of EGF quantity on HMSNs is failed with the
commonly used route for EGF immobilization.
104
A new grafting method was developed specific to improve the EGF grafting efficiency of
HMSNs in 5.3.1. This new strategy for EGF labelling can tailor make HMSN-NH2-EGF
with a certain amount of EGF attachments on HMSNs (illustrated in 5.2.3). This will
contribute to more versatile applications of EGF grafting. A precise control of EGF on
HMSNs could lead to an effective regulation of nanoparticles entering the cells and
thereby controlling the drug concentration in targeted cells. It is known that the cancer
development is a multistep process where different drug concentration and exposure time
were required at different stages.
105
-0.15 -0.10 -0.05 0.00 0.05 0.10 0.15-0.04
-0.03
-0.02
-0.01
0.00
0.01
0.02
0.03
0.04
EGF-0.1 EGF-0.2 EGF-0.4 EGF-0.8
PC
2 (7
.9%
)
PC1 (82.6%)
(B)
Figure 5. 13 Positive survey mass spectra (a) and scores on PC1 and PC2 (b) of HMSN-COOH-EGF surface using different EGF concentration: 0.1,0.2,0.4 and 0.8 mg/ml; and plot of ratios of the intensities of the 28 m/z (from HMSNs) to 30 m/z, 44 m/z and 58 m/z (from EGF) (c)
0
2
4
6
8
10
0 0.2 0.4 0.6 0.8 1
Inte
nsity
EGF Concentration (mg/ml)
(C)
28/30
28/44
28/58
106
5.4 Effectively target colorectal cancer cells by EGF grafted HMSNs
5.4.1 Effective targeting effect of HMSN-EGF to EGFR positive CRC cells
Figure 5. 14 Cellular uptake of both EGF grafted and plain HMSNs in SW480 cells
Targeting is essential for drug delivery systems as it has the potential to precisely target
and kill cancerous cells while leaving normal cells unharmed. To configure HMSNs for
targeted drug delivery, EGF was grafted onto HMSNs (HMSN-EGF) to target the EGFR
positive colorectal cancer cells (SW 480 cells). HMSNs with and without EGF attachment
were covalently linked to a fluorescent dye (FITC), making them visible by fluorescence
microscopy and flow cytometry. SW 480 cells were treated with varying doses (1, 2, 5,
10 and 20 μg/ml) of HMSN-EGF for 2 hours in addition to non-EGF grafted HMSNs for
control.
With increased nanoparticle concentration, more HMSNs and EGF-HMSNs are found to
be associated with SW480 cells (Figure 5.14), indicating that the uptake of both HMSNs
and EGF-HMSNs by the cells are concentration dependent. It was found for nanoparticle
concentrations below 20 μg/ml, the cellular uptake of HMSN-EGF was higher than
unmodified HMSNs (Figure 5.14). In order for nanoparticles to internalise into > 90 %
of cells only 5 μg/mL of EGF-HMSNs was required while for unmodified HMSNs 20
0.00
20.00
40.00
60.00
80.00
100.00
120.00
0 5 10 15 20 25
% C
ells
ass
ocia
ted
with
HM
SNs
Concentration (μg/ml)
HMSNs
HMSN-EGF
107
μg/mL was necessary to achieve the same result. This improved internalisation was due
to the ability of the targeting molecules, EGF, to attach to their receptors, EGFR, which
were found to be overexpressed on cancer cells. Therefore EGF attachment can facilitate
the cellular uptake of nanoparticles and the EGF-EGFR reaction could account for this
observation.
On the other hand, at a concentration of 20 μg/mL both HMSNs and HMSN-EGF were
found to be taken up by ~ 100% of cells. However it is more preferable to deliver a smaller
quantity of HMSNs to minimise side effects to patients. Additionally, due to the targeting
function of EGF it will be possible to reduce systemic elimination of HMSNs from the
body and the problematic delivery of chemotherapeutic drugs to normal cells.
Figure 5.15 a-f displays confocal microscopic images of SW480 cell incubated with FITC
labelled HMSN-EGF and HMSNs (10 μg/mL) for 0.5, 2 and 4 hours. Green dots are
nanoparticles and cell nucleus were visualized in blue colour. As observed in Figure 5.15
a-f, with increased time, more HMSNs and EGF-HMSNs are taken up by SW480 cells,
demonstrating that the uptake of both HMSNs and EGF-HMSNs by the cells are time
dependent. Also, the uptake of HMSN-EGF into target cells was significantly higher than
that of HMSNs when the same incubation time was applied, consistent with the fact that
EGF attachment enhances the capacity of nanoparticle entering the targeted cells (EGFR
positive cells).
To further prove that the particles are actually penetrating into the cells, the cytoplasm of
the cells was stained and 3D projections of the cellular uptake of EGF-HMSNs were
presented in Figure 5.15 g-h. The cells were sliced with a slice thickness of 1 μm and all
the slices were combined and presented as a 3D projection. Green dots were seen within
the 3D projections indicating the EGF-HMSNs have been internalized into the cells.
108
Figure 5. 15 (a-c): the confocal laser scanning microscope (CLSM) images of SW480 cells cultivated with 10 μg/mL FITC-HMSNs; (d-f): CLSM images of SW480 cells cultivated with 10 μg/mL FITC-EGF-HMSNs; and (g-h): 3D microscope projections of SW 480 cells cultivated with 10 μg/mL FITC-EGF-HMSNs for 4h.Green dots are nanoparticles while cell nucleus were visualized in blue colour.
109
5.4.2 Reduced targeting effect of HMSN-EGF by pretreating SW480 cells with free EGF
Figure 5. 16. The cellular uptake rate (mean fluorescence intensity) of HMSN-EGF and HMSNs in SW 480 cells with or without 5 μM of free EGF pretreatment
As HMSN-EGF had a higher uptake in SW 480 cells, to validate that this enhanced uptake
rate is facilitated specifically by the EGF targeting bioconjugates, SW480 cells were
treated with 5 μM of free EGF before incubation with 5 μg/mL of HMSN-EGF and plain
HMSNs. As seen in Figure 5.16, when cells were pretreated with free EGF, the ability to
internalize HMSN-EGF decreased while the capacity to internalize plain HMSNs
remained the same. After pretreating with free EGF, the mean fluorescence intensity (MFI)
value of cells treated with HMSN-EGF decreased about 25%. However, no obvious
differences can be seen in the uptake of plain HMSNs between cells with and without
free EGF pretreatments. This demonstrated that part of the EGF receptors on the cell
surface can be occupied by free EGF which may inhibit the binding between HMSN-EGF
and EGFR. This result can be further confirmed by the confocal microscope images, as
seen in Figure 5.17, the uptake of the nanoparticles into target cells was visualized by
luminescence of green emitting FITC (green dots) and the cell nucleus was shown in blue
colour. After 2 hours of treatment with HMSNs, only a few of the cells were seen
associated with HMSNs while the uptake of HMSN-EGF was much higher. After
0.0
500.0
1000.0
1500.0
2000.0
2500.0
3000.0
3500.0
HMSNs HMSN-EGF
Cells without freeEGF pretreating
Cells with free EGFpretreating
25%
110
pretreating with free EGF, cells associated with HMSN-EGF decreased indicating the
uptake of HMSN-EGF was inhibited.
HMSNs HMSN-EGF HMSN-EGF + free EGF
Figure 5. 17. Confocal microscope images of EGF-HMSNs and HMSNs in SW 480 cells.
5.4.3 Non-targeting effect of HMSN-EGF to EGFR negative cell line SW620
The cellular uptake studies on the SW 480 cells have demonstrated that HMSN-EGF
exhibited a higher uptake rate than HMSNs and the cellular uptake can be inhibited by
pre-treating the cells with 5 μM of free EGF. To further confirm that the higher uptake
of HMSN-EGF was mediated specifically via EGFR, cellular uptake experiments were
conducted on SW 620 cells which do not express EGFR.
As shown in Figure 5.18, the SW 620 cells were treated with different concentrations of
HMSNs or HMSN-EGF for 2 hours and the MFI were measured by flow cytometer.
Similar to the cellular uptake in SW 480 cells, the cell uptake of both HMSN-EGF and
HMSNs were concentration dependant: the mean fluorescence intensity increased as the
concentration of nanoparticles increased. However, the difference between HMSN-EGF
and HMSNs is not obvious in this cell line as the MFI of cells treated with HMSNs and
HMSN-EGF on each concentration point was close to each other.
111
Figure 5. 18. The mean fluorescence intensity in the SW 620 cells treated with different concentrations of HMSNs or EGF-HMSNs for 2 hours.
5.5 Conclusions
To precisely kill colorectal cancer cells, HMSNs were successfully synthesized and
bioconjugated with specific receptor ligand EGF. The construction of a targeting
molecule (EGF) to the surface of HMSNs was achieved through a newly developed
strategy, where the carboxyl groups of EGF were conjugated to an amine terminated
mesoporous silica surface. This method to graft EGF on silica surface proved be facile
and effective and can protect HMSNs from damage during modification. The accurate
control of the number of EGF attachments on HMSNs was achieved via this method,
proving its superiority to commonly used ones.
The characterization of EGF attachments was carried out using ToF-SIMS. The ToF-
SIMS data revealed the successful grafting of EGF attachments. PCA was implemented
to further extract and clarify the information from the complex ToF-SIMs data and
illustrated that EGF concentration played a crucial role in the binding efficiency of the
EGF to the nanoparticles. In all cases, the combination of ToF-SIMS and PCA proved
0
200
400
600
800
1000
1200
1400
0 5 10 15 20 25
MFI
Particle concentration
HMSNs-FITC
EGF-HMSNs-FITC
112
the effectiveness of the method in classifying relative surface concentration of EGF
attachments.
In vitro experiments present direct evidences that HMSN-EGF can efficiently and
selectively enter the EGFR positive colorectal cancer cells (SW480). However, the
targeting function of the current drug delivery system had no effect on EGFR negative
cancer cell lines SW620. More importantly, cellular uptake results on SW 480 and SW
620 cells suggested that the EGF-EGFR endocytosis pathway accounted for the improved
internalisation of HMSN-EGF. Therefore, by utilizing the EGF-EGFR interactions, the
specific delivery of drugs in colorectal cancer cells can be achieved, which could
guarantee a more precise killing of cancer cells without damaging the normal cells.
113
Chapter 6 Functionalization of HMSNs for favourite 5-FU loading and siRNA
encapsulation
6.1 Introduction
Inadequate intracellular concentration of anti-cancer drug is always associated to the
failure of cancer therapy as drug concentration is a key parameter for successful treatment.
The excellent targeting function of current drug delivery system was demonstrated based
on HMSNs in last chapter, which can boost the number of nanoparticles to be internalized
by targeted cell population and thus increasing the amount of drug transported and
accumulated into cancer cells. However, even though with high surface area, high volume
and large cavity, HMSNs showed a relative low 5-FU loading capacity (~ 10%) which
may be due to the electrostatic repulsion between the drug and the carriers. In this case,
compared to those carriers with a higher drug loading capacity, more HMSNs was
required to be taken up by cancer cells to achieve an effective cure concentration for
cancer therapy. From a technical point of view, it is more preferable to deliver a smaller
quantity of HMSNs to minimise patient side effects [247]. Therefore, increasing the drug
loading capacity of HMSNs is essential for further optimizing the current drug delivery
system. The encapsulation of drugs in nanoparticles can be improved by functionalization
[248] or by optimizing the drug encapsulation conditions. Precisely controlling the
grafting of chemical groups on HMSNs would result in an enhancement in drug loading.
On the other hand, drug resistance is another issue associated with cancer therapy. Even
though a very high 5-FU loading and excellent targeting behaviour of HMSNs could be
achieved, the therapeutic effect of 5-FU is still limited due to the appearance of drug
resistance in cancer cells. To overcome this problem, siRNA encapsulation appeared to
be another potential application for HMSNs. Therefore, in this chapter, the studies focus
on the modifications of HMSNs for enhanced 5-FU loading and effective siRNA
114
encapsulation. Different functionalizations of HMSNs are introduced in this chapter to
make HMSNs more versatile for particular applications and ways of improving siRNA
encapsulation of HMSNs are explored.
6.2 Enhanced 5-FU loading capacity of HMSNs by precise functionalization
6.2.1 Study of 5-FU partition coefficient
The partition coefficient (Pow) is defined as the quotient of two concentrations of a test
substance (like 5-FU) in a two-phase system, where two largely immiscible solvents (n-
octanol and water) were mixed with a fixed volume ratio. The logarithm of Pow between
n-octanol and water (log Pow) has often been used as a key parameter to describe the
lipophilicity of a molecule. In the pharmaceutic field, logP is utilized to estimate the
biological activities of the drugs, such as solubility [249], biodegradation rate, bio-
accumulation, membrane permeability [250], and drug absorption and toxicity
predictions. Therefore, the investigation of the Pow of 5-FU is essential to predict its
transport and activity, thereby providing information for the subsequent modification of
HMSNs specific for high 5-FU loading.
The shake flask method is one of the common and standard experimental procedures
adopted for logPow study. This method is applied to determine the solubility and
hydrophobicity of compounds with logP values ranging from –2 to 4. The determination
is based on the principle that hydrophobic substances soluble in the lipid phase show a
positive logPow (log Pow > 0), while negative logPow values (log Pow < 0) typifies polar
compounds soluble in the water phase.
According to OECD guideline 107 [173], an effective partition coefficient study refers to
all obtained log Pow values falling within a range of -2 ~ 4 with a offset of ± 0.3 units. As
shown in Table 6.1, the log Pow values of 5-FU with all ratios (octanol : water) are within
the range mentioned above and are consistently negative, indicating that 5-FU is
hydrophilic. Specifically, when the volume of water decreased from 26.67 mL to 13.33
mL, the 5-FU concentration in water phase increased significantly from 7.747 μg/mL to
115
13.17 μg/mL. This is due to the fact that most 5-FU molecules prefer congregating in the
water phase, thus with a decrease in the water volume, a more concentrated 5-FU solution
is observed. In contrast, no obvious difference was found in 5-FU concentration in the
octanol phase. With varying octanol volume, from 26.67 mL to 13.33 mL, 5-FU
concentration remained at around 2.2 ~ 2.6 μg/mL, which can be explained by the
hydrophilic property of 5-FU.
Table 6. 1 Partition coefficient of 5-FU and relevant parameters
1:2 (O:W)
1:1 (O:W)
2:1 (O:W)
Water volume (mL) 26.67 20 13.33
Oil volume (mL) 13.33 20 26.67
C water (μg/mL)
7.747
9.925 13.17
C octanol (μg/mL) 2.279 2.380
2.629
Pow 0.2942 0.2398 0.1996
log Pow -0.5314 -0.6201 -0.6997
As shown in Figure 6.1, even though the volume of water and octanol changes, it appears
that in all ratios, around 75% of 5-FU are dissolved in the water phase while the other
~25% is found in the octanol phase. Therefore, a more hydrophilic surface of HMSNs
would increase the loading of 5-FU.
116
Figure 6. 1 percentages of 5-FU dissolved in each phase with varying octanol to water ratio
77.48% 74.44% 75.82%
22.52% 25.56% 24.18%
0%
20%
40%
60%
80%
100%
120%
1:2 (O:W) 1:1 (O:W) 2:1 (O:W)
Oil
Water
117
6.2.2 Preparation and characterization of functionalized HMSNs
Figure 6. 2 TEM images of HMSNs (a), HMSN-NH2 (b), HMSN-COOH (c), HMSN-CN (d) and HMSN-CH3 (e)
118
Surface-functionalized HMSNs were synthesized by post grafting method where
chemical groups were introduced to the surface of as-synthesized HMSNs using
appropriate silanes. CPTES, APTES and TEMS were used to prepare HMSN-CN,
HMSN-NH2 and HMSN-CH3 while HMSN-COOH was obtain by oxidizing HMSN-CN.
In all cases, plain HMSNs without any modification was used as a control. The
morphology of modified and plain HMSNs were characterized by TEM and BET. As
shown in Figure 6.2, all modified HMSNs (Figure 6.2 b-e) had a similar morphology to
that of plain HMSNs (Figure 6.2 a) revealing that the structure of these nanoparticles can
be well preserved during all functionalization processes. In particular, the surface-
functionalized HMSNs maintained a hollow structure and a very spherical shape with a
mean diameter of ~120 nm and a shell thickness of ~10 nm. In addition, all HMSNs
displayed a highly uniform size, shell thickness and more importantly, a mono-dispersion
which is one of the main requirements for drug delivery systems.
The presence of mesoporous structure in all HMSNs was confirmed by BET (Figure 6.3).
The nitrogen adsorption/desorption experiments demonstrated that both modified and
plain HMSNs showed characteristic isotherm of mesoporous materials, a type IV
isotherm, indicating the existence of mesopores. Well-defined steps occurred in the
relative pressure of 0.35-0.55 and 0.9-1.0 were related to capillary condensation and
desorption in open mesopores and interstitial pores, respectively [189]. As seen in Figure
6.3b, narrow pore size distributions are presented in HMSNs before and after surface
modifications. The average pore size of all HMSNs was calculated using the desorption
branch of the isotherm by the typical Barrett-Joyner-Halenda (BJH) method, according
to which HMSNs, HMSN-NH2, HMSN-CN, HMSN-COOH and HMSN-CH3 have a
pore size of 2.78 nm, 2.47 nm, 2.28 nm, 2.61 nm and 2.51 nm, respectively.
119
0.0 0.2 0.4 0.6 0.8 1.0
1.000.90
0.55
Vol
ume
abso
rbed
Relative pressure (P/P0)
HMSNs HMSN-CH3
HMSN-CN HMSN-COOH HMSN-NH2
0.35
(A)
0 2 4 6 8 10 12 14 16 18 20
HMSNs HMSN-CH3
HMSN-NH2
HMSN-COOH HMSN-CN
pore size (nm)
2.78nm (B)
2.51nm
2.47nm
2.61nm
2.28nm
Figure 6. 3 N2 adsorption-desorption isotherms (a) and the corresponding pore size distributions (b) of plain and functionalized HMSNs
120
Table 6. 2 Structure parameters of functionalized and non-functionalized HMSNs
0 500 1000 1500 2000 2500 3000 3500 4000 4500
NH2
CH3
COOH
Wavenumber(cm-1)
HMSNs HMSN-COOH HMSN-CN HMSN-CH3
HMSN-NH2
CN
δO–Si–O
νO–Si–O
νSi–OH
Figure 6. 4 FTIR spectra of plain and functionalized HMSNs and plain HMSNs
HMSNs HMSN-CH3
HMSN-CN
HMSN-NH2
HMSN-COOH
Plain HMSNs
Pore size (nm) 2.51 2.28 2.47 2.61 2.78
Pore volume (cm3/g)
0.43 0.50 0.43 0.47 0.53
Surface area (m2/g)
434 514 413 498 820
121
Structure parameters of the HMSNs with and without modifications were summarized in
Table 6.2. According to Table 6.2, not significant differences can be found in the pore
size and pore volume between all kinds of nanoparticles, proving that surface
functionalization had a limited effect on the two parameters of final products. The
maintaining of pore size and pore volume guarantees large spaces in functionalized
HMSNs for drug loading that ideally should be same as non-functionalized ones.
However, due to the coverage of functional groups, the surface area of functionalized
samples decreased when compared to plain HMSNs. On the other hand, due to being
different batches, the surface area of the plain HMSNs is slightly different from that
presented in Section 4.2.1 even though the same fabrication procedures were applied.
Compared with F127 and P123, the molecular weight of Tweens 20 and TX100 is much
smaller, which produces smaller micelles during the fabrication and therefore results in
smaller pore size of nanoparticles. As pore size, surface area and pore volume of the final
products highly depend on the kind of surfactant used for fabrication, HMSNs
synthesized with surfactants having similar properties may have similar or even the same
physicochemical properties, including the surface area. These results were consistent
with similar studies [228].
Different chemical groups of -NH2, -CN, -COOH and –CH3 were grafted onto the surface
of HMSNs separately and the successful grafting was evidenced by FTIR technique.
FTIR spectra of modified HMSNs and the control sample (HMSNs) were shown in Figure
6.4. All samples have strong bands located at around 460 (δO–Si–O), 1100 (ʋasym. Si–
O–Si), and 3300-3500 cm-1 (ʋSi–OH), confirming the presence of the SiO2 inorganic phase
[227]. The -NH2 functionalization is clearly evidenced by the bands detected at around
1500 cm-1 and 3000 cm-1, which are unambiguously attributed to the NH bending
vibration and C-N stretching vibration, respectively [227-230]. Similarly, -CN and –
COOH functionalizations were proved by the observation of relative characteristic peaks.
The peak at 2200cm-1 corresponding to cyano groups (CN) was observed in the spectrum
of HMSN-CN while HMSN-COOH was shown at a peak of 1730cm-1 relating to C=O
vibration of COOH. The bands appeared at around 3000 cm-1 in the spectrum of HMSN-
CH3 are assigned to C-H vibration of methyl groups (–CH3) confirming the existence of
–CH3 on the surface of HMSN-CH3. Results obtained by FTIR proved the effective
surface functionalization of HMSNs, thereby confirming the presentence of -NH2, -CN,
122
-COOH and –CH3 on the surface of HMSN-NH2, HMSN-CN, HMSN-COOH and
HMSN-CH3, respectively.
Thermogravimetric analysis (TGA) was carried out to further confirm the effective
surface modification of functionalized HMSNs. All samples were tested under nitrogen
condition at the temperature ranging from room temperature up to 650 C. The weight
loss of samples obtained from TGA arose from the decomposition of chemical groups
thereby a difference in the weight loss between samples can indicate the relative amount
of chemical groups grafted onto particles. As seen from Figure 6.5, the total weight loss
of plain HMSNs was ~8.0% due to the dehydroxylation of silanol groups (Si-OH). In
contrast, all functionalized HMSNs showed a total weight loss of ~20% which was
mainly from thermal cracking of grafted functional groups. Therefore, the success of the
grafting has been proven by TGA results which was in accordance with the FTIR results
illustrated above and the relative amount of functional groups of modified HMSNs was
~12% (w/w).
100 200 300 400 500 60060
80
100
Temperature (0C)
Wei
ght (
%)
HMSNs HMSN-CN HMSN-NH2
HMSN-CH3
HMSN-COOH
Figure 6. 5 TGA curves of functionalized HMSNs and plain HMSNs
123
6.2.3 Regulate 5-FU loading capacity of HMSNs by modifications
Table 6. 3 Loading capacity of functionalized and non-functionalized HMSNs
The 5-FU loading capacity of functionalized and non-functionalized HMSNs has been
monitored using a UV spectrophotometer. An initial drug concentration of 3 mg/ml was
applied and the concentration change after 24 hours of loading was tracked. The loading
capacity of all HMSNs were calculated and listed in Table 6.3. As expected, a difference
in the amount of 5-FU loading after surface functionalization was observed, which
resulted from the different hydrophobicity and hydrophilicity of the functional groups
[228]. The plain HMSNs presented a loading percentage of 18.34% which increase to
28.89% (the best) after amine functionalization. It was reported that by using amine
functionalized mesoporous silica nanoparticles (MSNs, non-hollow structure)
specifically for 5-FU encapsulation, a 5-FU loading of 124 mg 5-FU/g MSNs was
achieved [251]. Therefore, by designing and fabricating MSNs with different structure
such as the hollow MSNs, the 5-FU loading can be significantly improved. This
enhanced loading capacity could be attributed to the hollow structure of HMSNs and the
high efficiency of functionalization.
Furthermore, -CH3 functionalization decreases the loading capacity of HMSNs to some
extent (from 18.34% to 12.73%). 5-FU is a water-soluble drug therefore it is harder to
approach HMSNs functionalized with methyl groups which is slightly more hydrophobic.
On the other hand, -CN and -COOH functionalized HMSNs showed moderate loading
ability with a loading percentage of 22.54% and 20.73%, respectively. Although HMSN-
COOH is hydrophilic which is similar to 5-FU, its loading capacity is inferior to HMSN-
NH2. This can be explained by the different surface charge of HMSN-NH2 and HMSN-
COOH. Given that 5-FU is negatively charged, it is much easier for 5-FU to get inside
HMSNs HMSN-CH3
HMSN-CN
HMSN-NH2
HMSN-COOH
Plain HMSNs
Loading capacity (%) 12.73 22.54 28.89 20.73 18.34
Loading capacity (mg 5-FU/g HMSNs)
127.3 225.4 288.9 207.3 183.4
124
the positively charged HMSN-NH2 via electrostatic attractions. However, the existence
of negative charge of –COOH groups may repel 5-FU molecules and decrease the 5-FU
loading capacity of HMSN-COOH.
In summary, the loading capacity of HMSNs specific to 5-FU can be improved via
surface functionalization. The presence of –NH2 groups on the surface of nanoparticles
resulted in a highest 5-FU loading capacity. A combination of a similarity in
hydrophilicity and a presence of reverse charge of the drug gives rise to this highest
loading capacity of HMSN-NH2. This finding in addition to the higher EGF
immobilization by HMSN-NH2 illustrated in Chapter 5 will definitely render the current
drug delivery system more versatile for various applications.
6.3 Control of 5-FU loading and release
6.3.1 The effect of pH on drug loading and release
The above study illustrated that HMSN-NH2 can significantly increase the amount of 5-
FU encapsulated into nanoparticles due to the similar hydrophilicity to drug and the
existence of positive charge provided by –NH2 groups. Therefore the control of the
intensity of electrostatic force between nanoparticles and 5-FU may further regulate 5-
FU loading capacity of particles.
In this regard, the loading of 5-FU by HMSN-NH2 was carried out in 4 different pHs, 4.0,
5.5, 7.4 and 8.5. As shown in Figure 6.6, as pH increases, the loading of 5-FU capacity
of HMSN-NH2 increases and attains the maximum loading capacity of 27.52% at pH of
8.5. HMSN-NH2 is positively charged at a wider pH range because of the protonation of
the amine groups while under different pH values, the behaviour of 5-FU in terms of
surface charge is different [252, 253]. As shown in Figure 6.7, the deprotonation of 5-FU
can occur at N1 or N3 or both to form the monoanions AN1, AN3 and dianion under
alkaline condision. It is reported that at pH 7-10, AN1, AN3 and dianion of 5-FU coexist
as a mixture in aqueous medium [252, 253]. Therefore, at acidic pHs (4.0 and 5.5), 5-FU
becomes less deprotonated or non-deprotonated resulting in a very low surface charge of
125
5-FU. Therefore, lower drug loading capacity of HMSN-NH2 (18.56% and 19.03%) was
observed at these two pHs. Nevertheless, when pH increased to 7.4 and 8.5, the
deprotonation of 5-FU enhanced thereby enhancing electrostatic attractions between
HMSN-NH2 and 5-FU. Instead of being a neutral molecule in the medium, most 5-FU
molecules transferred to monoanions and dianion, which generated negative charges and
led to more 5-FU being encapsulated into the nanoparticles.
18.56
20.92
27.03 27.52
19.0320.45
19.4218.47
0
5
10
15
20
25
30
8.57.45.5
Load
ing
capa
city
(%)
pH values
HMSN-NH2
HMSNs
4.0
A
200 300 400 5000.0
0.5
1.0
Abs
orba
nce
(a.u
.)
Wavelength (nm)
pH 8.5 pH 5.5 pH 7.4 pH 4.0
B
Figure 6. 6 5-FU loading capacity of HMSNs and HMSN-NH2 at different pH values
(a) and UV spectra of remaining 5-FU after 24 hours loading by HMSN-NH2 at
different pH values (b)
126
Figure 6. 7 Two possible 5-FU anions (AN1 and AN3) and dianion
On the other hand, no obvious differences can be seen in the loading capacity of HMSNs
at various pHs. Due to the presence of negatively charged silanolate groups on the surface,
the loading capacity of HMSNs was inferior to HMSN-NH2 at all pHs. The encapsulation
of 5-FU by HMNSs might be mainly based on hydrogen bonding rather than electrostatic
force. In addition, under alkaline pHs where 5-FU possesses more negative charges, a
slightly decrease in drug loading capacity of HMSNs was found. This could be due to
the repulsive interaction between carriers and the drug.
6.3.2 The effect of particle size and shell thickness on drug loading and release
To study the effect of particle size on 5-FU loading and release behaviours, HMSNs with
different particle size (100, 200 and 300 nm) were synthesized and their morphology
were shown by SEM images (Figure 6.8). As seen from Figure 6.8, all HMSNs exhibit
very spherical shape with different diameters. In addition, the size of each kind of
AN3
1 2
3 34
5
6
AN1
Dianion
+H+
-H+
+H+
-H+ 1 3
3
1
127
HMSNs is uniform and is at the target size 100, 200 and 300 nm. A mono-dispersion is
found consistently for all particles.
Figure 6. 8 SEM images of HMSNs with different particle size: 100 nm (a & d); 200 nm (b & e) and 300 nm (c & f)
20.3621.45
19.23
100 nm 200 nm 300 nm0
10
20
5-Fu
load
ing
capa
city
(%)
Particle size (nm)
A
0 300 600 900 1200 1500
0
10
20
30
40
50
C
umul
ativ
e re
leas
e (%
)
Release time (mins)
100 nm 200 nm 300 nm
B
Figure 6. 9 Loading capacities (a) and release profiles (b) at pH 5.5 of HMSNs with varying particle size
Drug loading and release experiments were performed on HMSNs with varying particle
size and their loading capacities and release behaviours are shown in Figure 6.9. HMSNs
with particle size of 100, 200 and 300 nm obtained similar 5-FU loading capacities of
~20% and cumulative release percentages of ~40% (Figure 6.9), indicating particle size
128
is not an important factor of HMSNs in determining 5-FU loading and release ability. On
the other hand, three kinds of HMSNs displayed sustained release profiles with a
maximum release percentage approached at 5 hours and maintained for another 20 hours.
Similarly, HMSNs with a varying shell thickness were prepared for testing the effect of
shell thickness on drug loading and release performances. As presented in Figure 6.10,
the shell thickness of HMSNs can be changed from ~10 nm to ~30 nm while the particle
size remains the same as ~120 nm. HMSNs with different shell thickness showed
spherical appearance and hollow structure. Along with the increase in shell thickness, a
decrease in cavity size was found, which resulted from the fact that the size of the
nanoparticles used to develop cavities in HMSNs were different.
Figure 6. 10 SEM images of HMSNs with different shell thickness: 10 nm (a & d); 15
nm (b & e) and 30 nm (c & f)
The loading and release performance of HMSNs with different shell thickness were
summarized in Figure 6.11. Firstly, the drug loading capacity of HMSNs with a shell
thickness of 10, 15 and 30 nm were 18.42%, 16.93% and 12.16%, respectively. The
ability to encapsulate 5-FU decreases with an increase in shell thickness and HMSNs
with the maximum shell thickness (30 nm) shows an obvious decrease in 5-FU loading
capacity. This might be due to the reduced size of cavities in this kind of HMSNs and
therefore less space in HMSNs available for drug storage. Secondly, the release
129
behaviour of HMSNs with shell thickness of 10 and 15 nm generates a similar release
behaviour with a final released drug proportion of ~ 40%. However, even though a
sustained release profile was presented for HMSNs with 30 nm shell, only ~ 15% of
loaded drug can be release from the carriers. In addition, the release rate was slower than
the other two samples, which could be due to the thicker shell that provides a longer way
for drug molecules to transport.
18.4216.93
12.16
10 nm 15 nm 30 nm0
4
8
12
16
20
5-Fu
load
ing
capa
city
(%)
Shell thickness (nm)
A
-60 0 60 120 180 240 1200 1800
0
10
20
30
40
50
Cum
ulat
ive
rele
ase
(%)
Release time (mins)
10nm 15nm 30nm
B
Figure 6. 11 Loading capacities (a) and release profiles (b) at pH 5.5 of HMSNs with
varying shell thickness
6.3.3 The effect of EGF on HMSNs’ drug loading and release behaviours
EGF attachment had been proved to be an excellent targeting ligand in chapter 5 which
enables HMSNs to precisely target cancer cells. After targeting, an effective killing of
cancer cells should be guaranteed by releasing the anti-cancer drug pre-loaded in EGF-
HMSNs. Therefore, it is important to study the drug loading and release behaviours of
EGF-HMSNs. Figure 6.12 shows the morphology of EGF-HMSNs. After EGF grafting,
EGF-HMSNs maintain the spherical and hollow structure as plain HMSNs in addition to
a good dispersion. The loading and release profiles of EGF-HMSNs and HMSNs are
given in Figure 6.13, where a decreased loading capacity is found in EGF-HMSNs. EGF
appeared on the surface of nanoparticles, which may block some pores of HMSNs and
hinder the drug to get inside the nanocarriers. Therefore, the 5-FU loading capacity of
nanocarriers decreases from 20% to 12.5% after EGF conjugation. On the other hand,
130
the EGF attachments have limited effect on the final drug release percentage which was
proved by a similar cumulative release percentage found in both HMSNs and EGF-
HMSNs. Nevertheless, the release rate of EGF-HMSNs is slowed down by the EGFs. As
seen in Figure 6.13B, HMSNs can attain a maximum drug concentration in the first 5
hours of the release process while EGF-HMSNs exhibit a much slower release profile
with a highest release rate at 25 hours.
Figure 6. 12 SEM images (a & b) and TEM image (c) of HMSN-EGF
0 5 10 15 20 25
0
5
10
15
20
Load
ing
capa
city
(%)
Loading time (h)
EGF non-EGF
A
0 5 10 15 20 25
0
10
20
30
40
50
Release time (h)
Cul
mat
ive
rele
ase
(%)
non-EGF EGF
B
Figure 6. 13 Loading (a, at pH 8.0) and release (b, at pH 5.5) profiles of HMSNs with
and without EGF attachments
131
6.4 SiRNA encapsulation based on HMSNs
With targeting ligand EGF on the surface, HMSN-NH2-EGF can precisely target and kill
cancer cells by releasing the anti-cancer drug into target cells. However, the cancer cells
develop drug resistance after the drugs are continuously applied.
Multidrug resistance (MDR) referring to the ability of cancer cells to adapt and survive
from chemotherapy is one of the most complex and challenging problems in cancer
treatments [14, 109, 254, 255]. Although nanopariticle based drug delivery systems are
often used to prevent MDR in cancer treatment by sidestepping drug resistance
mechanisms, the development of drug resistant genes by cancer cells ultimately leads to
unsatisfied outcomes. A novel approach to address MDR is to take advantage of the
ability of nanocarriers to deliver nucleic acids (such as siRNA) to cancer cells, which
knock down genes associated with MDR.
To overcome the 5-FU resistance, HMSNs were modified for siRNA delivery to silence
the gene expression involved in MDR. As siRNA is negatively charged, HMSNs were
pre-functionalized to be positively charged and thus offer the electrostatic force for
siRNA loading.
6.4.1 Increased surface charge of HMSNs for SiRNA loading
HMSNs-NH2 possessing abundant amine groups on the surface thereby render HMSNs
positive surface charge for siRNA loading. An additional advantage of HMSNs-NH2 is
the enhanced loading capacity specifically to 5-FU. In this regard, HMSNs-NH2 were
applied to load siRNA and could serve as carriers for co-delivery of 5-FU and siRNA.
Amino-functionalized HMSNs with an average diameter of ~100 nm were obtained via
grafting 3-aminopropyltriethoxysilane (APTES) on HMSNs (Figure 6.14a). The particles
displayed a high surface area of 680 m2/g, a pore size of 2.5 nm and a pore volume of
1.12 cm3/g. The corresponding values after amino-functionalization were 380 m2/g, 2.3
132
nm and 0.78 cm3/g respectively (Table 6.4). Zeta potentials of particles before and after
functionalization are -27.1 mV and 0.936 mV respectively (Figure 6.14b). The increase
in zeta potential and decrease in surface area is due to the introduction of –NH2 groups.
-150 -100 -50 0 50 100 150
HMSNs HMSN-NH2
Zata potential (mV)In
tens
ity
Inte
nsity
-27.1 0.936
Figure 6. 14 SEM image (a) and zeta potential (b) of HMSN-NH2
Table 6. 4 Parameters of HMSNs and HMSN-NH2
Samples Pore size
(nm)
Pore volume
(cm3/g)
Surface area
(m2/g)
HMSNs
2.5
1.12
680
HMSN-NH2
2.3
0.78
380
Thereafter, various amount of HMSN-NH2 (0.5, 1, 2, 5, 10 and 20 μg) were incubated
with 0.1 μg of siRNA to obtain particles/siRNA weight ratio from 5 to 200. The
incubation was carried out in phosphate-buffered saline (PBS pH 7.4) for 2 h and
encapsulation efficiency of siRNA by HMSN-NH2 were characterized by agarose gel
electrophoresis. As shown in Figure 6.15, the loading of siRNA to HMSN-NH2 was
133
negligible (black bands at the bottom of each column indicate unloaded siRNA). With the
highest ratio of 200, a black band with faded colour (decreased intensity) can been seen
in the gel showing very low amount of siRNA was loaded. This result reveals that siRNA
can be adsorbed by HMSN-NH2 but only limited siRNA was encapsulated at a weight
ratio of 200.
Figure 6. 15 siRNA encapsulation efficiency of HMSN-NH2. Various weight ratio of
particles/siRNA from 5 to 200 were applied.
The poor siRNA loading capacity could be due to the low surface charge of HMSN-NH2.
Although HMSN-NH2 presented positive charges, the potential is only 0.936 mV which
might not be high enough to adsorb siRNA. As it is expected a high siRNA loading of
HMSN-NH2 will guarantee an effective delivery of siRNA for the inhibition of drug
resistant genes expression, HMSN-NH2-1, HMSN-NH2-2 and HMSN-NH2-3 with
different surface charge were prepared and used for siRNA loading. Figure 6.16 displayed
the potential value of plain HMSNs, HMSN-NH2-1, HMSN-NH2-2 and HMSN-NH2-3;
the increased charge of HMSN-NH2-2 and HMSN-NH2-3 were achieved by increasing
the amount of APTES used in functionalization stages.
134
-27.1
0.936
11.1
40.2
-30
-20
-10
0
10
20
30
40
50
HMSN-NH2-3HMSN-NH2-2HMSN-NH2-1
Zeta
-pot
entia
l (m
V)
HMSNs
Figure 6. 16 Zeta potential of HMSNs, HMSN-NH2-1, HMSN-NH2-2
and HMSN-NH2-3
The siRNA encapsulation efficiency was studied using the same method. We
hypothesized that an enhanced loading capacity would come with the increase in surface
charge of nanoparticles. Nevertheless, gel electrophoresis results revealed that loading of
siRNA by particles was still negligible (Figure 6.17 and Table 6.5). These results
suggested that the encapsulation of siRNA by HMSNs might not only associated with the
surface charge of particles. Instead, other parameters such as pore size might also
important for achieving an effective siRNA encapsulation.
Figure 6. 17 siRNA encapsulation efficiency of HMSN-NH2-2 (a) and HMSN-NH2-3
(b). Various weight ratio of particles/siRNA from 5 to 200 were applied.
135
Table 6. 5 siRNA encapsulation efficiency
samples Pore size (nm) Potential (mv) SiRNA Loading
HMSNs 2.3 -27 None
HMSNs-NH2-1 2.3 0.936 limited
HMSNs-NH2-2 2.3 11.1 limited
HMSNs-NH2-3 2.3 40.2 limited
6.4.2 Pore size expansion of HMSNs for efficient SiRNA encapsulation
On the basis of poor siRNA loading capacity illustrated above, it is suggested that besides
positive charged surface, larger pores might be necessary for a high siRNA loading by
HMSNs. As the length of siRNA is about 3nm, the relative small pore size (2.3 nm) of
current used particles would significantly reduce the adsorption of siRNA. When siRNA
was mixed with nanoparticles, they can only attach to the surface or stay near the surface
of nanoparticles rather than enter the pores. Similar results were found by other
researchers when non-hollow mesoporous silica nanoparticles were employed to load
siRNA [256]. Consequently, pore expansion of HMSNs is required before siRNA
loading.
Three HMSNs (HMSNs-S, HMSNs-M and HMSN-L) with the same particle size, shell
thickness but different pore size (small, middle and large) were prepared and their
structure parameters were presented in Table 6.6. As shown in Figure 6.18, all HMSNs
136
possess a large cavity inside and mesoporous shell outside, highlighting their distinctive
hollow structure. These HMSNs with very spherical shape exhibited a good dispersion.
Successful pore expansion of HMSNs were confirmed by scanning electron microscope
(SEM) (Figure 6.18 a, c & e) where middle and large pores of HMSNs-M and HMSNs-
L are clearly seen from the images. Similarly, in transmission electron microscopy (TEM)
images (Figure 6.18 b, d & f), small pores of HMSNs-S were hardly seen as white dots
indicating pores were missing in Figure 6.18 b. In contrast, white dots appeared in Figure
6.18 d, and become larger in Figure 6.18 f, revealing an increase in pore size of HMSNs.
137
Figure 6. 18 SEM and TEM images of HMSNs-S (a & b), HMSNs-M (c & d) and HMSNs-L (e & f)
138
The corresponding structure parameters of HMSNs-S, HMSNs-M and HMSN-L were
listed in Table 6.6. As expected, three HMSNs achieve the same particle size of ~120 nm
and shell thickness of ~10 nm, which is essential for investigating the effect of pore size
on siRNA loading capacity as keeping other parameters the same are necessary for
effective comparisons. The pore size of HMSNs-S, HMSNs-M and HMSNs-L were 2.5
nm, 4.3 nm and 20.1nm respectively. With an increase in pore size, pore volume
increased from 1.12 cm3/g to 1.45 cm3/g while the surface area decreased from 680 m2/g
to 513 m2/g [113].
Table 6. 6 Structure parameter of HMSNs-S, HMSN-M and HMSNs-L
Samples HMSNs-NH2-S HMSNs-NH2-M HMSNs-NH2-L
Pore size (nm) 2.5 4.3 20.1
Particle size (nm) 120 120 120
Shell thickness (nm)
10 10 10
Pore volume (cm3/g)
1.12 1.23 1.45
Surface area (m2/g)
680 604 513
To evaluate the capacity as siRNA carriers, these three HMSNs were aminated for the
presence of positive charges to adsorb negatively charged siRNA through electrostatic
interaction. The degree of amination of three kinds of HMSNs were evaluated by zeta-
potential and thermogravimetric analysis (TGA). The zeta potential results (Figure 6.19a)
demonstrate that three HMSNs possess a similar positive charge of around ~13 mV
revealing their close amount of amine groups. This amount of grafted amine groups was
further tracked via TGA. As shown in Figure 6.19b, the plain HMSNs have a total weight
loss of ~12.6 wt% under nitrogen condition, which is due to the dehydroxylation of
silanol groups. HMSNs-S, HMSNs-M and HMSNs-L show an increased weight loss
139
proving their successful modifications. Further, a similar total weight loss percentage (~
18%) of HMSNs-S, HMSNs-M and HMSNs-L again illustrates the fact that they have a
close number of functional groups.
-27.1
13.5 12.7 13.1
-30
-20
-10
0
10
20
HMSNs
HMSNs-LHMSNs-M
Zeta
-pot
entia
l (m
V)
HMSNs-S
100 200 300 400 500 600 70060
80
100
4500C
Temperature (0C)
Wei
ght (
%)
weight loss:HMSNs 14.6%HMSNs-S 19.8%HMSNs-M 19.3%HMSNs-L 19.4%
HMSNs HMSNs-S HMSNs-M HMSNs-L
2500C
Figure 6. 19 Zeta potential (a) and TGA (b) of aminated HMSNs, HMSNs-S,
HMSNs-M and HMSNs-L
140
Results from siRNA loading (Figure 6.20) demonstrate that HMSNs-M and HMSNs-L
have significantly higher siRNA loadings than HMSNs-S. When the pore size of HMSNs
was expanded form 2.5 nm to 4.3 nm, an obvious increase in siRNA loading was found.
Almost all siRNA could be encapsulated into HMSNs-M at the particle to siRNA ratio
of 100. This can be explained that the 4.3 nm pores of HMSNs-M are large enough for
the siRNA to pass through freely which enable HMSNs-M to encapsulate siRNA into the
pores as well as the hollow cavity (Figure 6.21). On the other hand, when the pore size
was further expanded to 20.1 nm, the siRNA loading of HMSNs-L remain the same as
that of HMSNs-M. This could be due to that both HMSNs-M and HMSNs-L had reached
their loading limit. In other words, the maximum loading capacities of both HMSNs-M
and HMSNs-L were similar and all their active space for siRNA loading had already been
occupied by siRNA at a weight ratio of 100. Therefore, for siRNA loading, pore size is
a significant parameter only when the pore size of the carriers is under 4.3 nm. When
HMSNs have a pore size larger than 4.3 nm, further expansion of the pore size will not
affect the siRNA loading anymore.
141
Figure 6. 20 siRNA encapsulation efficiency of HMSNs-S, HMSNs-M and HMSNs-L. Various weight ratio of particles/siRNA from 5 to 200
were applied.
142
Figure 6. 21 Proposed siRNA encapsulation of HMSNs with different pore sizes (on the
surface VS inside the pores and hollow cavity)
6.5 Conclusions
The presence of functional groups with different hydrophobicity and hydrophilicity on
HMSNs led to the variation of 5-FU loading. As 5-FU is more hydrophilic, HMSNs-NH2
showed the highest 5-FU loading capacity of 28.89%, while the presence of the more
hydrophobic group –CH3 on HMSNs have decreased the HMSNs’ loading capacity from
18.34% to 12.73%. In addition, due to the electrostatic attraction between nanoparticles
143
and 5-FU, the presence of positive charge on HMSN-NH2 contributed to its highest
loading capacity. The loading and release behaviours of HMSNs can also be well
controlled via adjusting the structure parameters of HMSNs: particle size, pore size and
shell thickness. For siRNA encapsulation, pore size, rather than the intensity of positive
charge existing on HMSNs, turned out to be a crucial parameter in determining the
siRNA encapsulation efficiency of HMSNs. Almost all siRNA can be encapsulated into
HMSNs with a pore size of 4.3 nm at the particle to siRNA ratio of 100.
144
Chapter 7 Conclusions and future works
7.1 Overview of current work
Nanoparticle based drug delivery systems have the potential to revolutionize cancer
therapy, because nanoparticles maintain a prolonged effective drug concentration which
allows cancer cells to be exposed sustainedly to the drug. Owning to their small size and
the characteristic features of tumour biology, like the leaky vasculature and the impair
lymphatic system, nanoparticles can reach certain solid tumours and release biologically
active cargos into the tumour cells. Several therapeutic nanocarriers including carbon
nanotubes, dendrimers, polymeric nanoparticles and inorganic nanoparticles, have been
reviewed in Chapter 2 and it was found that HMSNs possess the ability to maintain
optimum therapeutic efficacy with nontoxicity due to their large surface area, high
volume for drug loading and excellent biocompatibility. However, the non-ionic
surfactant templated HMSNs often have a broad size distribution or a defective
mesoporous structure because of the difficulties involved in controlling the formation
and organization of micelles for the growth of silica framework.
In Chapter 4, a novel “Eudragit assisted” strategy has been developed to fabricate hollow
mesoporous silica nanoparticles (HMSNs) by utilising the Eudrgit nanoparticles as cores
and to assist in the self-assembly of micelle organization. The significant achievements
of this newly developed synthesis route include:
1. Uniform, dispersible hollow mesoporous silica nanoparticles with large internal
cavities were achieved with this newly developed strategy. The method using
Eudragit S100 nanoparticles as both a core template and an assistant in the self-
organization of surfactant micelles proved an excellent method for HMSNs synthesis.
The mesoporous structure, a high surface area and a small pore size of HMSNs were
suggested by BET results, where IV type N2 adsorption- desorption isotherm, a
surface area of 760m2/g and a pore size of 2.5 nm were presented.
145
2. Eudragit S-100 is a widely used pharmaceutical polymer for preparing enteric
formulations [257, 258] and as carriers for pH dependent drug delivery systems [192-
194] due to its pH dependent solubility [190, 191]. The utilisation of Eudragit S100
nanoparticles as a core template for the fabrication of HMSNs affords the advantage
of non-toxicity [259], potential sites (carboxyl groups) for hydrogen bonding [260],
and most importantly easier core removal. Eudragit core can be easily removed by
soaking the as-synthesized HMSNs (HMSNs with Eudragit cores) in base solution
or in Acetone under room temperature. This solvent extraction process to clean the
Eudragit cores is facile and gentle, which remarkably preserved HMSNs form further
damage or aggregation caused by calcination, where HMSNs subjected to a very high
temperate. To our knowledge, this is the first time to report the use of Eudragit S100
nanoparticle as a core template for HMSNs synthesis.
3. The synthesized HMSNs were found to exhibit great sustained release of anti-cancer
drug 5-FU and nontoxic to colorectal cancer cells. The successful synthesis of
HMSNs in addition to their excellent biocompatibility demonstrate the potential of
HMSNs as effective drug carriers for building versatile cancer-targeted drug delivery
systems.
Many chemotherapeutic drugs for cancer treatments with low specificity often kill
healthy cells as well and thereby causing immense toxicity to the patient. Therefore,
targeting functions are essential for drug delivery systems as they have the potential to
precisely target and kill cancerous cells while leaving normal cells unharmed. To
configure HMSNs for targeted drug delivery, HMSNs were bioconjugated with specific
receptor ligand EGF in Chapter 5. The construction of a targeting molecule (EGF) to the
surface of HMSNs was achieved through a new method, where the carboxyl groups of
EGF conjugated to an amine terminated mesoporous silica surface. This method proved
to be facile and effective to graft EGF on silica surface and can protect HMSNs from
damage during modification. Three principal conclusions have been reached on the
preparation of EGF-HMSNs and their bioactivity in Chapter 5:
146
1. The current EGF grafting method can tune the quantity of EGF attachments on the
surface of HMSNs by changing the EGF concentration during the grafting stages,
which have not been previously reported in the open literature. A small change in
the properties of targeting ligand in terms of effective immobilization, stability,
concentration and binding efficiency can have a great influence on the subsequent
performance of nanoparticulate targeted drug delivery systems in the biological
environment. Although the utilisation of EGF for targeting delivery is well
established, a precise control of the density of EGF attachments on nanoparticles is
very limited. Therefore, the successful controlled grafting of EGF onto HMSNs by
the strategy illustrated in Chapter 5 proved the superiority of this method to
commonly used ones.
2. Characterization of EGF attachments was first carried out using ToF-SIMS. The
ToF-SIMS data revealed the successful grafting of EGF attachments. PCA was
implemented to further extract and clarify the information from the complex ToF-
SIMs data and illustrated that EGF concentration played a crucial role in the binding
efficiency of the EGF to the nanoparticles. In all cases, the combination of ToF-
SIMS and PCA proved to be very effective in classifying relative surface
concentration of EGF attachments.
3. In vitro experiments present direct evidences that HMSN-EGF can efficiently and
selectively enter the EGFR positive colorectal cancer cells (SW480) and they had no
effect on EGFR negative cancer cell line SW620. More importantly, cellular uptake
results on SW 480 cells and SW 620 cells suggested that the EGF-EGFR endocytosis
pathway accounted for the improved internalisation of HMSN-EGF. Therefore, by
utilizing the EGF-EGFR interactions, the specific delivery of drugs in colorectal
cancer cells can be achieved. This guarantees a more precisely killing of cancer cells
without killing the normal cells.
For drug delivery system, a high drug loading is always being pursued to guarantee an
adequate intracellular concentration of anti-cancer drug in tumour area. The capacity of
nanoparticles to include drugs can be improved by functionalization [248] or by
optimizing the drug encapsulation conditions. In Chapter 6, the focus was on the
147
modifications of HMSNs for enhanced 5-FU loading. Different functionalization of
HMSNs is introduced in this chapter to make HMSNs more versatile for particular
application. The significant achievements include:
1. The presence of different functional groups on HMSNs led to the variation of 5-FU
loading, which resulted from the different hydrophobicity and hydrophilicity of the
functional groups. As 5-FU is more hydrophilic, the presence of –NH2 groups on the
surface of nanoparticles resulted in the highest 5-FU loading capacity of 28.89%,
while the more hydrophobic group –CH3 decreased the HMSNs’ loading capacity
from 18.34% to 12.73%. In addition, the presence of positive charge on HMSN-NH2
contributed to its highest loading capacity. This is because of the electrostatic
attraction between nanoparticles and 5-FU.
2. The loading and release behaviours of HMSNs can also be controlled via adjusting
the structure parameters of HMSNs: particle size, pore size and shell thickness.
Studies in Chapter 6 revealed that HMSNs with particle size of 100, 200 and 300 nm
achieved similar 5-FU loading capacities of ~20% and cumulative release
percentages of ~40%, indicating particle size is not an important factor of HMSNs
in determining 5-FU loading and release ability. However, the capacity of HMSNs
to encapsulate 5-FU decreased with an increase in shell thickness and HMSNs with
the maximum shell thickness (30 nm) shown an obvious decrease in 5-FU loading
and release. Similarly, pore size is another significant factor for release, with an
increase in pore size, the release rate is accelerated (data was shown in Chapter 4).
3. To overcome the 5-FU resistance, HMSNs were modified for the delivery of siRNA
to silence the genes expression involved in multidrug resistance (MDR). Rather than
the intensity of positive charge existing on HMSNs, pore size turned out to be a
crucial parameter in determining the siRNA encapsulation efficiency of HMSNs.
Poor siRNA encapsulation was observed with HMSNs having a pore size of 2.3 nm,
which was due to the fact that siRNA was only adsorbed on the surface of HMSNs
and cannot pass through the pores to get into the cavity. When the pore size of
HMSNs was expanded to >4 nm, siRNA encapsulation was boosted. Almost all
siRNA can be encapsulated into HMSNs with a pore size of 4.3 nm at the particle to
148
siRNA ratio of 100. This can be explained that the 4.3 nm pores are large enough for
the siRNA to pass through freely which enables HMSNs to encapsulate siRNA into
the pores as well as the hollow cavity.
7.2 Limitations and future work
Although achievements have been made as outlined in the previous sections, there are
various limitations in the current project mainly due to the shortage of experimental and
characterization equipment. Animal experiments are highly desirable to this work even
though cell experiments were carried out for testing the biological behaviour of
investigated drug delivery system. However, the limitations do not reduce the
significance of the results which have been achieved as the cell experiments are as
important as animal ones in assessing the performances of current targeted drug delivery
system based on HMSNs.
7.2.1 Targeting efficiency of EGF-HMSNs
It is demonstrated in Chapter 5 that HMSNs bioconjugated with EGF can selectively and
efficiently target cancer cells with an overexpression of EGFR. Given that most cancer
cells overexpressed EGFR, these results in addition to the great biocompatibility of
HMSNs suggested the feasibility of future applications of HMSNs in targeted drug
delivery systems for cancer treatment. More importantly, the use of TOF-SIMs provides
opportunities to determine relative EGF concentration on HMSNs. Therefore, the amount
of EGF grafted onto HMSNs can be compared by changing EGF concentration during
modification process. Nevertheless, the subsequent targeting efficiency of EGF-HMSNs
with a varying amount of EGF attachments on the surface in cancer cells have not been
investigated in the current study. It is noteworthy that a precise control of EGF on HMSNs
could lead to an effective regulation of nanoparticles entering the cells and thereby
controlling the drug concentration in targeted cells. Therefore, a further investigation of
targeting efficiency of EGF-HMSNs with different EGF concentration on the surface is
of great importance to control drug concentration in cancer cells. It is known that the
cancer development is a multistep process where different drug concentration and drug
149
exposure time were required for different stages. Thus, looking at the effect of the number
of EGF ligands on the targeting behaviour of EGF-HMSNs would help engineer a
versatile drug delivery system with controlled targeting efficiency.
7.2.2 siRNA release and knockdown efficiency
In Chapter 6, siRNA delivery has been highlighted as an essential task for HMSNs to
overcome multidrug resistance. Although an effective siRNA loading by functionalized
HMSNs was clearly illustrated in this chapter, the release profile and gene knockdown
efficiency of siRNA loaded HMSNs were not studied, where further study associated with
siRNA delivery in vitro and in vivo is necessary to fully assess the potential of HMSNs
as siRNA carriers.
Drug resistance at the tumour level is complex, often involving multiple and dynamically
acquired multidrug resistance (MDR) mechanisms as a result of the expression of drug
e ux pumps, anti-apoptotic proteins, oncogenes, and regulators of drug metabolism.
Therefore, further study, including how to enhance the cellular uptake efficiency of
siRNA loaded HMSNs and effectively silence the targeted gene in vivo, needs to be
carried out in the near future. Also, how to control the release of siRNA from HMSNs
and protect siRNA from enzymatic degradation in vitro and in vivo should be addressed
to guarantee the most optimal siRNA knocking down.
7.2.3 Preparation multi-anticancer drug loaded HMSNs
As a drug carrier, high drug-loading capacity is being pursued. In Chapter 6, the structure
parameters of HMSNs can be controlled to load as many drug molecules as possible. In
addition, the surface potential and hydrophilicity of HMSNs can also be altered for
attracting drug molecules to enter the pores by precise surface modifications. However,
all the studies mentioned above is designed specifically for only one drug, 5-FU. As
pointed out by preclinical studies, a combination of chemotherapy drugs have led to
improved anti-cancer effects and the use of combination therapy for colon cancer is well
150
established worldwide. For instance, irinotecan is commonly used with 5-FU, and have
shown superior survival benefits to regimen of 5-FU only. In Australia, most frequently
used first line chemotherapy regimens involve the combination of 5-FU, oxaliplation and
leucovorin, becacizumab or irinotecan. Therefore, there exists a need to further explore
HMSNs as a useful platform for multidrug-combined therapy.
To enable the encapsulation of more than one drug in HMSNs more functionalizations
need to be introduced on the surface or in the pores of HMSNs. However, for a
nanoparticle based drug delivery system, the challenge lies in successfully combining
several functions into one system. For example, an ideal drug delivery system refers to
one that with targeting actions, sufficient therapeutic concentration of drug, nontoxicity
and adequate biological activity. In many cases, it is difficult to couple several functional
groups of sufficient concentration, since the number of attachment sites on the particle
surface is limited. Moreover, each functionalization step might negatively affect the
suspension stability of the particulate system, depending on the physicochemical
properties of the added function. Keep this in mind, further study would focus on the
structure optimisation of HMSNs rather than introducing functional groups. A novel kind
of core–shell dual-mesoporous structure with smaller mesopores in the shells and ordered
larger mesopores in the core has been constructed by Niu [261], which have exhibited the
capacity to simultaneously loaded two kinds of drugs with different molecular sizes and
different hydrophobicity. These core/shell hierarchical structures provide an inspiration
for future investigation into HMSNs. In Chapters 4 and 6, it has been demonstrated that
the pore size, shell thickness as well as particle size of HMSNs can be easily controlled,
which would open up opportunities to build up more versatile HMSNs with tailed
structure for multidrug encapsulation.
Another method proposed for the future co-delivery based on HMSNs is that HMSNs
would be divided into separate groups and each group will undergo respective specific
functionalization and drug loading. Finally, all the groups would mixed up with certain
ratio to achieve a multidrug loaded HMSNs mixture. This method takes the advantage of
high flexibility, specificity and more importantly, avoiding multi-functionalization on one
HMSNs and thereby increasing the efficiency while reducing toxicity.
151
7.2.4 Animal experiences
In this study, HMSNs have been demonstrated as a vehicle for successful intracellular
delivery of anticancer drug 5-FU. The current HMSNs based targeted drug delivery
system has shown sustained release behaviour, negligible toxicity and specifically
targeting and killing of colorectal cancer cells in vitro, which suggests that this system is
promising in the application for colorectal cancer therapy. However, in vitro study is not
enough for fully evaluate the biological behaviours of a drug delivery system as results
got form cell experiments could be quite different from outcomes obtained from
experiments performed on live mammals. Therefore, further validation of this targeted
drug delivery system in vivo is essential for its clinical translation.
Specifically, even though EGF-HMSNs have been selectively taken up by colorectal
cancer cells, it cannot guarantee a high selective accumulation of these nanoparticles in
solid tumours of live animal models, because the bio-distribution of nanoparticles in vivo
can be further influenced by biological environments [262]. In addition, there are findings
[263] that provide strong evidence that nanoparticles not only actively interact with
cancer cells, but also engage in and mediate in vivo behaviours. Thus, the interplay
between biological e ects and particles will undoubtedly be an important aspect that
needs to be further investigated so as to identify the full potential of the current delivery
system.
152
Bibliography
1. Walther, A., et al., Genetic prognostic and predictive markers in colorectal cancer.
Nature Reviews Cancer, 2009. 9(7): p. 489-499.
2. Kumar , K., Colon targeting drug delivery system: A review on recent approaches
International Journal of Pharmaceutics 2011. 2(1): p. 11.
3. Rajguru, V.V., et al., An overview on colonic drug delivery system. International
Journal of Pharmaceutical Sciences Review & Research, 2011. 6(2): p. 197-204.
4. Slowing, I.I., et al., Mesoporous silica nanoparticles: Structural design and
applications. Journal of Materials Chemistry, 2010. 20(37): p. 7924-7937.
5. Argyo, C., et al., Multifunctional mesoporous silica nanoparticles as a universal
platform for drug delivery. Chemistry of Materials, 2014. 26(1): p. 435-451.
6. Peer, D., et al., Nanocarriers as an emerging platform for cancer therapy. Nature
Nanotechnology, 2007. 2(12): p. 751-760.
7. Peng, X.H., et al., Targeted delivery of cisplatin to lung cancer using ScFvEGFR-
heparin- cisplatin nanoparticles. ACS Nano, 2011. 5(12): p. 9480-9493.
8. Byrne, J.D., T. Betancourt, and L. Brannon-Peppas, Active targeting schemes for
nanoparticle systems in cancer therapeutics. Advanced Drug Delivery Reviews,
2008. 60(15): p. 1615-1626.
9. Fay, F. and C.J. Scott, Antibody-targeted nanoparticles for cancer therapy.
Immunotherapy, 2011. 3(3): p. 381-394.
10. Zhang, Q., et al., Biocompatible, uniform, and redispersible mesoporous silica
nanoparticles for cancer-targeted drug delivery in vivo. Advanced Functional
Materials, 2014. 24(17): p. 2450-2461.
11. Sandoval, M.A., et al., EGFR-targeted stearoyl gemcitabine nanoparticles show
enhanced anti-tumor activity. Journal of Controlled Release, 2012. 157(2): p. 287-
296.
12. Hompes, R. and C. Cunningham, Colorectal cancer: Management. Medicine,
2011. 39(5): p. 254-258.
13. Eng, C., Toxic effects and their management: Daily clinical challenges in the
treatment of colorectal cancer. Nature Reviews Clinical Oncology, 2009. 6(4): p.
207-218.
153
14. KAISER, J., Combining Targeted Drugs To Stop Resistant Tumors. Science, 2011.
331: p. 1542.
15. Longley, D.B., D.P. Harkin, and P.G. Johnston, 5-Fluorouracil: mechanisms of
action and clinical strategies. Nature Reviews Cancer, 2003. 3(5): p. 330-338.
16. Guo, Y., et al., Capecitabine plus irinotecan versus 5-FU/leucovorin plus
irinotecan in the treatment of colorectal cancer: A meta-analysis. Clinical
Colorectal Cancer, 2014. 13(2): p. 110-118.
17. Kim, G.P. and A. Grothey, Current Challenges in the Adjuvant Therapy of Colon
Cancer. Challenges in Colorectal Cancer. 2008: Blackwell Publishing Ltd. 133-
152.
18. Enhanced Anti-Cancer Effect of 5-Fluorouracil Loaded into Thermo-Responsive
Conjugated Linoleic Acid-Incorporated Poloxamer Hydrogel on Metastatic
Colon Cancer Models. Journal of Nanoscience and Nanotechnology, 2011. 11: p.
1425-1428.
19. Dev, R.K., V. Bali, and K. Pathak, Novel microbially triggered colon specific
delivery system of 5-Fluorouracil: Statistical optimization, in vitro, in vivo,
cytotoxic and stability assessment. International Journal of Pharmaceutics, 2011.
411(1-2): p.142-151.
20. Feng, S.S., L. Zhao, and J. Tang, Nanomedicine for oral chemotherapy.
Nanomedicine, 2011. 6(3): p. 407-410.
21. Krishnaiah, Y.S.R., et al., In vitro drug release studies on guar gum-based colon
targeted oral drug delivery systems of 5-fluorouracil. European Journal of
Pharmaceutical Sciences, 2002. 16(3): p. 185-192.
22. Wu, J., et al., The mechanisms of 5-FU-PLA-O-CMC-NPS-mediated inhibition of
the proliferation of colorectal cancer cell line SW480. Tumor Biology, 2014.
35(6): p. 6095-9103
23. Neuse, E.W., Synthetic polymers as drug-delivery vehicles in medicine. Metal-
Based Drugs, 2008. 2008.
24. Javadzadeh, Y., L. Musaalrezaei, and A. Nokhodchi, Liquisolid technique as a
new approach to sustain propranolol hydrochloride release from tablet matrices.
International Journal of Pharmaceutics, 2008. 362(1-2): p. 102-108.
25. Asaduzzaman, M., et al., Development of sustain release matrix tablet of
ranolazine based on methocel K4M CR: In vitro drug release and kinetic
approach. Journal of Applied Pharmaceutical Science, 2011. 1(8): p. 131-136.
154
26. Taylor, H.E. and K.B. Sloan, 1-Alkylcarbonyloxymethyl prodrugs of 5-
fluorouracil (5-FU): Synthesis, physicochemical properties, and topical delivery
of 5-FU. Journal of Pharmaceutical Sciences, 1998. 87(1): p. 15-20.
27. Sloan, K.B., E.F. Sheretz, and R.G. McTiernan, The effect of 5-fluorouracil on
inhibition of epidermal DNA synthesis in vivo: A comparison of the effect of
formulations and a prodrug of 5-FU. Archives of Dermatological Research, 1990.
282(7): p. 484-486.
28. Koehler, K.C., et al., A Diels-Alder modulated approach to control and sustain
the release of dexamethasone and induce osteogenic differentiation of human
mesenchymal stem cells. Biomaterials, 2013. 34(16): p. 4150-4158.
29. Min, Y. and P.T. Hammond, Catechol-modified polyions in layer-by-layer
assembly to enhance stability and sustain release of biomolecules: A bioinspired
approach. Chemistry of Materials, 2011. 23(24): p. 5349-5357.
30. Pavani, E., S. Noman, and I.A. Syed, Liquisolid technique based sustained release
tablet of trimetazidine dihydrochloride. Drug Invention Today, 2013. 5(4): p. 302-
310.
31. Nokhodchi, A., et al., Liquisolid compacts: The effect of cosolvent and HPMC on
theophylline release. Colloids and Surfaces B: Biointerfaces, 2010. 79(1): p. 262-
269.
32. Ganesh, N.S., et al., Formulation and evaluation of sustained release lornoxicam
by liquisolid technique. International Journal of Pharmaceutical Sciences Review
and Research, 2011. 11(1): p. 53-57.
33. Elyagoby, A., N. Layas, and T.W. Wong, Colon-specific delivery of 5-
fluorouracil from zinc pectinate pellets through In Situ intracapsular
ethylcellulose-pectin plug formation. Journal of Pharmaceutical Sciences, 2013.
102(2): p. 604-616.
34. He, W., et al., Study on colon-specific pectin/ethylcellulose film-coated 5-
fluorouracil pellets in rats. International Journal of Pharmaceutics, 2008. 348(1-
2): p. 35-45.
35. Bose, A., A. Elyagoby, and T.W. Wong, Oral 5-fluorouracil colon-specific
delivery through in vivo pellet coating for colon cancer and aberrant crypt foci
treatment. International Journal of Pharmaceutics, 2014. 468(1-2): p. 178-186.
155
36. Wei, H., et al., In-vitro and in-vivo studies of pectin/ethylcellulose-film-coated
pellets of 5-fluorouracil for colonic targeting. Journal of Pharmacy and
Pharmacology, 2008. 60(1): p. 35-44.
37. Kaiser, N., et al., 5-Fluorouracil in vesicular phospholipid gels for anticancer
treatment: Entrapment and release properties. International Journal of
Pharmaceutics, 2003. 256(1-2): p. 123-131.
38. Cardillo, J.A., et al., An intravitreal biodegradable sustained release naproxen
and 5-fluorouracil system for the treatment of experimental post-traumatic
proliferative vitreoretinopathy. British Journal of Ophthalmology, 2004. 88(9): p.
1201-1205.
39. Kulthe, S.S., et al., Modulated release of 5-fluorouracil from pH-sensitive and
colon targeted pellets: An industrially feasible approach. Drug Development and
Industrial Pharmacy, 2013. 39(1): p. 138-145.
40. Fan, L.F., et al., Biphasic drug release: Permeability and swelling of
pectin/ethylcellulose films, and in vitro and in vivo correlation of film-coated
pellets in dogs. Chemical and Pharmaceutical Bulletin, 2008. 56(8): p. 1118-1125.
41. Won, Y.W., et al., Self-assembled nanoparticles with dual effects of passive tumor
targeting and cancer-selective anticancer effects. Advanced Functional Materials,
2012. 22(6): p. 1199-1208.
42. Galagudza, M.M., et al., Passive targeting of ischemic myocardium with the use
of silica nanoparticles. Nanotechnologies in Russia, 2010. 5(11): p. 844-850.
43. Perše, M., et al., Protective effect of fullerenol nano particles on colon cancer
development in dimethylhydrazine rat model. Digest Journal of Nanomaterials and
Biostructures, 2011. 6(4): p. 1543-1551.
44. Klopman, G. and H. Zhu, Recent methodologies for the estimation of n-
octanol/water partition coefficients and their use in the prediction of membrane
transport properties of drugs. Mini-Reviews in Medicinal Chemistry, 2005. 5(2):
p. 127-133.
45. Smyth, E.C., F. Sclafani, and D. Cunningham, Emerging molecular targets in
oncology: Clinical potential of MEeT/hepatocyte growth-factor inhibitors.
OncoTargets and Therapy, 2014. 7: p. 1001-1014.
46. Etienne-Grimaldi, M.C., et al., Molecular patterns in deficient mismatch repair
colorectal tumours: Results from a French prospective multicentric biological
and genetic study. British Journal of Cancer, 2014. 110(11): p. 2728-2737.
156
47. Roskoski, R., The ErbB/HER family of protein-tyrosine kinases and cancer.
Pharmacological Research, 2014. 79: p. 34-74.
48. Liu, J., et al., LMO1 is a novel oncogene in colorectal cancer and its
overexpression is a new predictive marker for anti-EGFR therapy. Tumor
Biology, 2014. 35(8): p. 8161-8167.
49. Saif, M.W., Colorectal cancer in review: The role of the EGFR pathway. Expert
Opinion on Investigational Drugs, 2010. 19(3): p. 357-369.
50. Sintov, A., et al., Enzymatic cleavage of disaccharide side groups in insoluble
synthetic polymers: A new method for specific delivery of drugs to the colon.
Biomaterials, 1993. 14(7): p. 483-490.
51. Ji, S.R., et al., Carbon nanotubes in cancer diagnosis and therapy. Biochimica et
Biophysica Acta - Reviews on Cancer, 2010. 1806(1): p. 29-35.
52. Gasser, G., I. Ott, and N. Metzler-Nolte, Organometallic anticancer compounds.
Journal of Medicinal Chemistry, 2011. 54(1): p. 3-25.
53. Baláž, P. and J. Sedlák, Arsenic in Cancer Treatment: Challenges for Application
of Realgar Nanoparticles (A Minireview). Toxins, 2010. 2(6): p. 1568-1581.
54. Das, S., A. Chaudhury, and K.Y. Ng, Preparation and evaluation of zinc-pectin-
chitosan composite particles for drug delivery to the colon: Role of chitosan in
modifying in vitro and in vivo drug release. International Journal of Pharmaceutics,
2011. 406(1-2): p. 11-20.
55. Saboktakin, M.R., et al., Synthesis and characterization of superparamagnetic
chitosan-dextran sulfate hydrogels as nano carriers for colon-specific drug
delivery. Carbohydrate Polymers, 2010. 81(2): p. 372-376.
56. Perera, G., J. Barthelmes, and A. Bernkop-Schnurch, Novel pectin-4-
aminothiophenole conjugate microparticles for colon-specific drug delivery.
Journal of Controlled Release, 2010. 145(3): p. 240-246.
57. Yellela, S., Guar gum-based colon-specific drug delivery systems for the
treatment of Inflammatory Bowel Diseases. Inflammatory Bowel Diseases, 2009.
15(12): p. S56-S56.
58. Shyale, S., et al., Pharmacokinetic evaluation and studies on the clinical efficacy
of guar gum-based oral drug delivery systems of albendazole and albendazole-
beta-cyclodextrin for colon-targeting in human volunteers. Drug Development
Research, 2006. 67(2): p. 154-165.
157
59. Wang, M.J., et al., Optimizing Preparation of NaCS-Chitosan Complex to Form
a Potential Material for the Colon-Specific Drug Delivery System. Journal of
Applied Polymer Science, 2010. 117(5): p. 3001-3012.
60. Bigucci, F., et al., Pectin-based microspheres for colon-specific delivery of
vancomycin. Journal of Pharmacy and Pharmacology, 2009. 61(1): p. 41-46.
61. Bigucci, F., et al., Chitosan/pectin polyelectrolyte complexes: Selection of suitable
preparative conditions for colon-specific delivery of vancomycin. European
Journal of Pharmaceutical Sciences, 2008. 35(5): p. 435-441.
62. Saboktakin, M.R., et al., Synthesis and Characterization of Chitosan Hydrogels
Containing 5-Aminosalicylic Acid Nanopendents for Colon: Specific Drug
Delivery. Journal of Pharmaceutical Sciences, 2010. 99(12): p. 4955-4961.
63. Elkhodairy, K.A., N.S. Barakat, and F.K. Alanazi, Solubilization and
Amorphization of Non Steroidal Anti-Inflammatory Drug with Low Molecular
Weight Chitosan for a New Guar-Based Colon Delivery Formulation. Letters in
Drug Design & Discovery, 2011. 8(3): p. 292-301.
64. Hiorth, M., T. Skoien, and S.A. Sande, Immersion coating of pellet cores
consisting of chitosan and calcium intended for colon drug delivery. European
Journal of Pharmaceutics and Biopharmaceutics, 2010. 75(2): p. 245-253.
65. Criscione, J.M., et al., Self-assembly of pH-responsive fluorinated dendrimer-
based particulates for drug delivery and noninvasive imaging. Biomaterials, 2009.
30(23-24): p. 3946-3955.
66. Sinha, V.R. and R. Kumria, Polysaccharides in colon-specific drug delivery.
International Journal of Pharmaceutics, 2001. 224(1-2): p. 19-38.
67. Yinghuai, Z., et al., Substituted carborane-appended water-soluble single-wall
carbon nanotubes: New approach to boron neutron capture therapy drug delivery.
Journal of the American Chemical Society, 2005. 127(27): p. 9875-9880.
68. Pastorin, G., et al., Double functionalisation of carbon nanotubes for multimodal
drug delivery. Chemical Communications, 2006. 21(11): p. 1182-1184.
69. Chen, J.Y., et al., Functionalized Single-Walled Carbon Nanotubes as Rationally
Designed Vehicles for Tumor-Targeted Drug Delivery. Journal of the American
Chemical Society, 2008. 130(49): p. 16778-16785.
70. Cox, B.J., N. Thamwattana, and J.M. Hill, Mechanics of spheroidal fullerenes and
carbon nanotubes for drug and gene delivery. Quarterly Journal of Mechanics and
Applied Mathematics, 2007. 60: p. 231-253.
158
71. Levi-Polyachenko, N,H., Merkel, E.J., Jones, B.T., Carroll, D.L and Setwart IV,
J.H., Nanotubes for rapid photothermal intracellular drug delivery. Molecular
Pharmaceutics, 2009. 6(4): p. 1062-1099.
72. Yang, D., et al., Hydrophilic multi-walled carbon nanotubes decorated with
magnetite nanoparticles as lymphatic targeted drug delivery vehicles. Chemical
Communications, 2009. 7(29): p. 4447-4449.
73. Ciofani, G., et al., Boron Nitride Nanotubes: A Novel Vector for Targeted
Magnetic Drug Delivery. Current Nanoscience, 2009. 5(1): p. 33-38.
74. Cai, K.Y., et al., Temperature-Responsive Controlled Drug Delivery System
Based on Titanium Nanotubes. Advanced Engineering Materials, 2010. 12(9): p.
B565-B570.
75. Bhirde, A.A., et al., Targeted killing of cancer cells in vivo and in vitro with EGF-
directed carbon nanotube-based drug delivery. ACS Nano, 2009. 3(2): p. 307-
316.
76. Liang, Y.Q., et al., Characterization of self-organized TiO2 nanotubes on Ti-4Zr-
22Nb-2Sn alloys and the application in drug delivery system. Journal of Materials
Science-Materials in Medicine, 2011. 22(3): p. 461-467.
77. Yin, J.F., et al., Molecularly imprinted nanotubes for enantioselective drug
delivery and controlled release. Chemical Communications, 2010. 46(41): p.
7688-7690.
78. Cho, S.J., et al., Silica coated titania nanotubes for drug delivery system.
Materials Letters, 2010. 64(15): p. 1664-1667.
79. Zhang, X., et al., Targeted delivery and controlled release of doxorubicin to
cancer cells using modified single wall carbon nanotubes. Biomaterials, 2009.
30(30): p. 6041-6047.
80. Simon-Deckers, A., et al., In vitro investigation of oxide nanoparticle and carbon
nanotube toxicity and intracellular accumulation in A549 human pneumocytes.
Toxicology, 2008. 253(1-3): p. 137-146.
81. Cheng, C., et al., Toxicity and imaging of multi-walled carbon nanotubes in
human macrophage cells. Biomaterials, 2009. 30(25): p. 4152-4160.
82. Kresge, C.T., et al., Ordered mesoporous molecular sieves synthesized by a
liquid-crystal template mechanism. Nature, 1992. 359(6397): p. 710-712.
159
83. Choi, Y.L., et al., Controlled release using mesoporous silica nanoparticles
functionalized with 18-crown-6 derivative. Journal of Materials Chemistry, 2011.
21(22): p. 7882-7885.
84. Radu, D.R., et al., Gatekeeping Layer Effect: A Poly(lactic acid)-coated
Mesoporous Silica Nanosphere-Based Fluorescence Probe for Detection of
Amino-Containing Neurotransmitters. Journal of the American Chemical Society,
2004. 126(6): p. 1640-1641.
85. Mercier, L. and T.J. Pinnavaia, Access in mesoporous materials: Advantages of a
uniform pore structure in the design of a heavy metal ion adsorbent for
environmental remediation. Advanced Materials, 1997. 9(6): p. 500-503.
86. Corma, A., From microporous to mesoporous molecular sieve materials and their
use in catalysis. Chemical Reviews, 1997. 97(6): p. 2373-2419.
87. Lu, J., et al., Biocompatibility, biodistribution, and drug-delivery efficiency of
mesoporous silica nanoparticles for cancer therapy in animals. Small, 2010.
6(16): p. 1794-1805.
88. Vivero-Escoto, J.L., et al., Mesoporous silica nanoparticles for intracellular
controlled drug delivery. Small, 2010. 6(18): p. 1952-1967.
89. Slowing, I.I., et al., Mesoporous silica nanoparticles as controlled release drug
delivery and gene transfection carriers. Advanced Drug Delivery Reviews, 2008.
60(11): p. 1278-1288.
90. He, Q. and J. Shi, Mesoporous silica nanoparticle based nano drug delivery
systems: Synthesis, controlled drug release and delivery, pharmacokinetics and
biocompatibility. Journal of Materials Chemistry, 2011. 21(16): p. 5845-5855.
91. Liu, X., et al., Adsorption of CO2, CH4 and N2 on ordered mesoporous silica
molecular sieve. Chemical Physics Letters, 2005. 415(4-6): p. 198-201.
92. Polarz, S. and A. Kuschel, Chemistry in confining reaction fields with special
emphasis on nanoporous materials. Chemistry - A European Journal, 2008.
14(32): p. 9816-9829.
93. Zhao, Y.L., et al., PH-operated nanopistons on the surfaces of mesoporous silica
nanoparticles. Journal of the American Chemical Society, 2010. 132(37): p.
13016-13025.
94. Mamaeva, V., et al., Mesoporous Silica Nanoparticles as Drug Delivery Systems
for Targeted Inhibition of Notch Signaling in Cancer. Molecular Therapy, 2011.
19(8): p. 1538-1546
160
95. Huh, S., et al., Organic Functionalization and Morphology Control of
Mesoporous Silicas via a Co-Condensation Synthesis Method. Chemistry of
Materials, 2003. 15(22): p. 4247-4256.
96. Chen, D., et al., Facile and scalable synthesis of tailored silica "nanorattle"
structures. Advanced Materials, 2009. 21(37): p. 3804-3807+3724.
97. Gao, H., W. Shi, and L.B. Freund, Mechanics of receptor-mediated endocytosis.
Proceedings of the National Academy of Sciences of the United States of America,
2005. 102(27): p. 9469-9474.
98. Li, Y., et al., Hollow spheres of mesoporous aluminosilicate with a three-
dimensional pore network and extraordinarily high hydrothermal stability. Nano
Letters, 2003. 3(5): p. 609-612.
99. Huo, Q., D.I. Margolese, and G.D. Stucky, Surfactant Control of Phases in the
Synthesis of Mesoporous Silica-Based Materials. Chemistry of Materials, 1996.
8(5): p. 1147-1160.
100. Tang, F., L. Li, and D. Chen, Mesoporous silica nanoparticles: Synthesis,
biocompatibility and drug delivery. Advanced Materials, 2012. 24(12): p. 1504-
1534.
101. Feng, Z., et al., A facile route to hollow nanospheres of mesoporous silica with
tunable size. Chemical Communications, 2008. 21(23): p. 2629-2631.
102. Kao, K.C., C.J. Tsou, and C.Y. Mou, Collapsed (kippah) hollow silica
nanoparticles. Chemical Communications, 2012. 48(28): p. 3454-3456.
103. Mandal, M. and M. Kruk, Family of single-micelle-templated organosilica hollow
nanospheres and nanotubes synthesized through adjustment of
organosilica/surfactant ratio. Chemistry of Materials, 2012. 24(1): p. 123-132.
104. Mei, X., et al., Hollow mesoporous silica nanoparticles conjugated with pH-
sensitive amphiphilic diblock polymer for controlled drug release. Microporous
and Mesoporous Materials, 2012. 152: p. 16-24.
105. Zhang, T., et al., Formation of hollow silica colloids through a spontaneous
dissolution-regrowth process. Angewandte Chemie - International Edition, 2008.
47(31): p. 5806-5811.
106. Lou, X.W., L.A. Archer, and Z. Yang, Hollow micro-/nanostructures: Synthesis
and applications. Advanced Materials, 2008. 20(21): p. 3987-4019.
161
107. Zhao, W., et al., Fabrication of uniform hollow mesoporous silica spheres and
ellipsoids of tunable size through a facile hard-templating route. Journal of
Materials Chemistry, 2009. 19(18): p. 2778-2783.
108. Chen, Y., et al., Hollow/rattle-type mesoporous nanostructures by a structural
difference-based selective etching strategy. ACS Nano, 2010. 4(1): p. 529-539.
109. Gao, Y., et al., Controlled intracellular release of doxorubicin in multidrug-
resistant cancer cells by tuning the shell-pore sizes of mesoporous silica
nanoparticles. ACS Nano, 2011. 5(12): p. 9788-9798.
110. Zhang, Q., et al., Permeable silica shell through surface-protected etching. Nano
Letters, 2008. 8(9): p. 2867-2871.
111. He, Q., et al., An anti-ROS/hepatic fibrosis drug delivery system based on
salvianolic acid B loaded mesoporous silica nanoparticles. Biomaterials, 2010.
31(30): p. 7785-7796.
112. Lu, J., et al., Mesoporous silica nanoparticles as a delivery system for
hydrophobic anticancer drugs. Small, 2007. 3(8): p. 1341-1346.
113. Radu, D.R., et al., Fine-tuning the degree of organic functionalization of
mesoporous silica nanosphere materials via an interfacially designed co-
condensation method. Chemical Communications, 2005. (10): p. 1264-1266.
114. Huh, S., et al., Cooperative Catalysis by General Acid and Base Bifunctionalized
Mesoporous Silica Nanospheres. Angewandte Chemie International Edition,
2005. 44(12): p. 1826-1830.
115. Shi, S., F. Chen, and W. Cai, Biomedical applications of functionalized hollow
mesoporous silica nanoparticles: Focusing on molecular imaging. Nanomedicine,
2013. 8(12): p. 2027-2039.
116. Mei, X., et al., Facile preparation of coating fluorescent hollow mesoporous silica
nanoparticles with pH-sensitive amphiphilic diblock copolymer for controlled
drug release and cell imaging. Soft Matter, 2012. 8(19): p. 5309-5316.
117. Lin, Y.-S., et al., Synthesis of hollow silica nanospheres with a microemulsion as
the template. Chemical Communications, 2009. (24): p. 3542-3544.
118. Yasmin, T. and K. Müller, Synthesis and characterization of surface modified
SBA-15 silica materials and their application in chromatography. Journal of
Chromatography A, 2011. 1218(37): p. 6464-6475.
162
119. Zhou, X., et al., Synthesis of hollow mesoporous silica nanoparticles with tunable
shell thickness and pore size using amphiphilic block copolymers as core
templates. Dalton Transactions, 2014. 43(31): p. 11834-11842.
120. Kao, K.-C. and C.-Y. Mou, Pore-expanded mesoporous silica nanoparticles with
alkanes/ethanol as pore expanding agent. Microporous and Mesoporous
Materials, 2013. 169(0): p. 7-15.
121. Jia, L., et al., Successfully tailoring the pore size of mesoporous silica
nanoparticles: Exploitation of delivery systems for poorly water-soluble drugs.
International Journal of Pharmaceutics, 2012. 439(1–2): p. 81-91.
122. Selvam, P., S.K. Bhatia, and C.G. Sonwane, Recent advances in processing and
characterization of periodic mesoporous MCM-41 silicate molecular sieves.
Industrial and Engineering Chemistry Research, 2001. 40(15): p. 3237-3261.
123. Zhang, H., et al., Opening and closing of nanocavities under cyclic loading in a
soft nanocomposite probed by real-time small-angle X-ray scattering.
Macromolecules, 2013. 46(3): p. 900-913.
124. Tse, N.M.K., et al., High-throughput preparation of hexagonally ordered
mesoporous silica and gadolinosilicate nanoparticles for use as MRI contrast
agents. ACS Combinatorial Science, 2012. 14(8): p. 443-450.
125. Wang, J., et al., Biphasic synthesis of colloidal mesoporous silica nanoparticles
using primary amine catalysts. Journal of Colloid and Interface Science, 2012.
385(1): p. 41-47.
126. Zhai, W., et al., Degradation of hollow mesoporous silica nanoparticles in human
umbilical vein endothelial cells. Journal of Biomedical Materials Research - Part
B Applied Biomaterials, 2012. 100(5): p. 1397-1403.
127. Morishige, K. and M. Tateishi, Accurate Relations between Pore Size and the
Pressure of Capillary Condensation and the Evaporation of Nitrogen in
Cylindrical Pores. Langmuir, 2006. 22(9): p. 4165-4169.
128. Ravikovitch, P.I. and A.V. Neimark, Experimental Confirmation of Different
Mechanisms of Evaporation from Ink-Bottle Type Pores: Equilibrium, Pore
Blocking, and Cavitation. Langmuir, 2002. 18(25): p. 9830-9837.
129. Kruk, M. and M. Jaroniec, Gas Adsorption Characterization of Ordered
Organic−Inorganic Nanocomposite Materials. Chemistry of Materials, 2001.
13(10): p. 3169-3183.
163
130. Shaobin, W., Ordered mesoporous materials for drug delivery. Microporous and
Mesoporous Materials, 2009. 117(1-2): p. 1-9.
131. Sarkisov, L. and P.A. Monson, Modeling of Adsorption and Desorption in Pores
of Simple Geometry Using Molecular Dynamics. Langmuir, 2001. 17(24): p.
7600-7604.
132. Ma, Z., et al., Impact of shape and pore size of mesoporous silica nanoparticles
on serum protein adsorption and RBCS hemolysis. ACS Applied Materials and
Interfaces, 2014. 6(4): p. 2431-2438.
133. Cauda, V., A. Schlossbauer, and T. Bein, Bio-degradation study of colloidal
mesoporous silica nanoparticles: Effect of surface functionalization with organo-
silanes and poly(ethylene glycol). Microporous and Mesoporous Materials, 2010.
132(1-2): p. 60-71.
134. He, Q., et al., The three-stage in vitro degradation behavior of mesoporous silica
in simulated body fluid. Microporous and Mesoporous Materials, 2010. 131(1-3):
p. 314-320.
135. He, Q., et al., In vivo biodistribution and urinary excretion of mesoporous silica
nanoparticles: Effects of particle size and PEGylation. Small, 2011. 7(2): p. 271-
280.
136. Souris, J.S., et al., Surface charge-mediated rapid hepatobiliary excretion of
mesoporous silica nanoparticles. Biomaterials, 2010. 31(21): p. 5564-5574.
137. Kwon, S., et al., Luminescent mesoporous nanoreservoirs for the effective loading
and intracellular delivery of therapeutic drugs. Acta Biomaterialia, 2014. 10(3):
p. 1431-1442.
138. Liu, G., et al., Charged polymer brushes-grafted hollow silica nanoparticles as a
novel promising material for simultaneous joint lubrication and treatment.
Journal of Physical Chemistry B, 2014. 118(18): p. 4920-4931.
139. Sun, Y., et al., Nanoassembles constructed from mesoporous silica nanoparticles
and surface-coated multilayer polyelectrolytes for controlled drug delivery.
Microporous and Mesoporous Materials, 2014. 185: p. 245-253.
140. Lu, F., et al., Iron oxide-loaded hollow mesoporous silica nanocapsules for
controlled drug release and hyperthermia. Chemical Communications, 2013.
49(97): p. 11436-11438.
164
141. He, Q., et al., Rhodamine B-co-condensed spherical SBA-15 nanoparticles: Facile
co-condensation synthesis and excellent fluorescence features. Journal of
Materials Chemistry, 2009. 19(21): p. 3395-3403.
142. Zhang, K., et al., A continuous tri-phase transition effect for HIFU-mediated
intravenous drug delivery. Biomaterials, 2014. 35(22): p. 5875-5885.
143. Lu, F., et al., Size effect on cell uptake in well-suspended, uniform mesoporous
silica nanoparticles. Small, 2009. 5(12): p. 1408-1413.
144. Slowing, I., B.G. Trewyn, and V.S.Y. Lin, Effect of surface functionalization of
MCM-41-type mesoporous silica nanoparticles on the endocytosis by human
cancer cells. Journal of the American Chemical Society, 2006. 128(46): p. 14792-
14793.
145. Chung, T.H., et al., The effect of surface charge on the uptake and biological
function of mesoporous silica nanoparticles in 3T3-L1 cells and human
mesenchymal stem cells. Biomaterials, 2007. 28(19): p. 2959-2966.
146. He, Q., et al., Intracellular localization and cytotoxicity of spherical mesoporous
silica nano-and microparticles. Small, 2009. 5(23): p. 2722-2729.
147. Munagala, S., et al., Synthesis and evaluation of Strychnos alkaloids as MDR
reversal agents for cancer cell eradication. Bioorganic and Medicinal Chemistry,
2014. 22(3): p. 1148-1155.
148. Radadiya, A., et al., Synthesis and 3D-QSAR study of 1,4-dihydropyridine
derivatives as MDR cancer reverters. European Journal of Medicinal Chemistry,
2014. 74: p. 375-387.
149. Wu, W.D., et al., Cordycepin down-regulates multiple drug resistant (MDR)/HIF-
1α through regulating AMPK/mTORC1 signaling in GBC-SD gallbladder cancer
cells. International Journal of Molecular Sciences, 2014. 15(7): p. 12778-12790.
150. Liu, Z., et al., Sinomenine sensitizes multidrug-resistant colon cancer cells (Caco-
2) to doxorubicin by downregulation of MDR-1 expression. PLoS ONE, 2014.
9(6).
151. Kim, Y.J., et al., Comparison of capecitabine and 5-fluorouracil in
chemoradiotherapy for locally advanced pancreatic cancer. Radiation Oncology,
2013. 8(1): p. 160
152. Mnif, N., et al., Effects of nanoparticle treatment and compatibilizers on the
properties of (PP/EPR)/nano-CaCO3 blends. International Journal of Material
Forming, 2008. 1(SUPPL. 1): p. 639-643.
165
153. Stylianopoulos, T., EPR-effect: Utilizing size-dependent nanoparticle delivery to
solid tumors. Therapeutic Delivery, 2013. 4(4): p. 421-423.
154. Kim, S.S., et al., A nanoparticle carrying the p53 gene targets tumors including
cancer stem cells, sensitizes glioblastoma to chemotherapy and improves survival.
ACS Nano, 2014. 8(6): p. 5494-5514.
155. Gao, Z., et al., Nanoparticle-based pseudo hapten for target-responsive cargo
release from a magnetic mesoporous silica nanocontainer. Chemical
Communications, 2014. 50(47): p. 6256-6258.
156. Lin, X., et al., Carbon nanoparticle-protected aptamers for highly sensitive and
selective detection of biomolecules based on nuclease-assisted target recycling
signal amplification. Chemical Communications, 2014. 50(57): p. 7646-7648.
157. Yang, C., et al., Target-induced strand release and thionine-decorated gold
nanoparticle amplification labels for sensitive electrochemical aptamer-based
sensing of small molecules. Sensors and Actuators, B: Chemical, 2014. 197: p.
149-154.
158. Chen, Z., et al., Drug resistance reversed by silencing LIM domain-containing
protein 1 expression in colorectal carcinoma. Oncology Letters, 2014. 8(2): p.
795-798.
159. Li, X., et al., Reversal of multidrug resistance of gastric cancer cells by
downregulation of CIAPIN1 with CIAPIN1 siRNA. Molecular Biology, 2008.
42(1): p. 91-97.
160. Shen, D.Y., et al., Inhibition of Wnt/β-catenin signaling downregulates P-
glycoprotein and reverses multi-drug resistance of cholangiocarcinoma. Cancer
Science, 2013. 104(10): p. 1303-1308.
161. Shi, X., et al., AQP5 silencing suppresses p38 MAPK signaling and improves drug
resistance in colon cancer cells. Tumor Biology, 2014. 35(7): p. 7035-7045
162. Xiong, X.B. and A. Lavasanifar, Traceable multifunctional micellar nanocarriers
for cancer-targeted co-delivery of MDR-1 siRNA and doxorubicin. ACS Nano,
2011. 5(6): p. 5202-5213.
163. Hua, J., D.G. Mutch, and T.J. Herzog, Stable suppression of MDR-1 gene using
siRNA expression vector to reverse drug resistance in a human uterine sarcoma
cell line. Gynecologic Oncology, 2005. 98(1): p. 31-38.
164. Logashenko, E.B., et al., Silencing of MDR 1 gene in cancer cells by siRNA.
Nucleosides, Nucleotides and Nucleic Acids, 2004. 23(6-7): p. 861-866.
166
165. Shanmugam, V., et al., Oligonucleotides-assembled au nanorod-assisted cancer
photothermal ablation and combination chemotherapy with targeted dual-drug
delivery of doxorubicin and cisplatin prodrug. ACS Applied Materials and
Interfaces, 2014. 6(6): p. 4382-4393.
166. Zhou, L., et al., Combination of chemotherapy and photodynamic therapy using
graphene oxide as drug delivery system. Journal of Photochemistry and
Photobiology B: Biology, 2014. 135: p. 7-16.
167. Li, J., et al., Recent advances in delivery of drug-nucleic acid combinations for
cancer treatment. Journal of Controlled Release, 2013. 172(2): p. 589-600.
168. Li, J.M., et al., Multifunctional QD-based co-delivery of siRNA and doxorubicin
to HeLa cells for reversal of multidrug resistance and real-time tracking.
Biomaterials, 2012. 33(9): p. 2780-2790.
169. Kanazawa, T., et al., Prolongation of life in rats with malignant glioma by
intranasal siRNA/drug codelivery to the brain with cell-penetrating peptide-
modified micelles. Molecular Pharmaceutics, 2014. 11(5): p. 1471-1478.
170. Kim, C., et al., Synergistic induction of apoptosis in brain cancer cells by targeted
codelivery of siRNA and anticancer drugs. Molecular Pharmaceutics, 2011. 8(5):
p. 1955-1961.
171. Deng, Z.J., et al., Layer-by-layer nanoparticles for systemic codelivery of an
anticancer drug and siRNA for potential triple-negative breast cancer treatment.
ACS Nano, 2013. 7(11): p. 9571-9584.
172. Vollrath, A., et al., Labeled nanoparticles based on pharmaceutical EUDRAGIT®
S 100 polymers. Macromolecular Rapid Communications, 2010. 31(23): p. 2053-
2058.
173. Guideline for the testing of chemicals - partition coefficient (n-octanol/water):
shake flask method. 1995, Paris: OECD Organisation for Economic Cooperation
and Development
174. Zhang, P., T. Wu, and J.L. Kong, In situ monitoring of intracellular controlled
drug release from mesoporous silica nanoparticles coated with pH-responsive
charge-reversal polymer. ACS Applied Materials and Interfaces, 2014. 6(20): p.
17446-17453.
175. Gui, R., Y. Wang, and J. Sun, Embedding fluorescent mesoporous silica
nanoparticles into biocompatible nanogels for tumor cell imaging and
167
thermo/pH-sensitive in vitro drug release. Colloids and Surfaces B: Biointerfaces,
2014. 116: p. 518-525.
176. Feng, W., et al., Polyelectrolyte multilayer functionalized mesoporous silica
nanoparticles for pH-responsive drug delivery: Layer thickness-dependent
release profiles and biocompatibility. Journal of Materials Chemistry B, 2013.
1(43): p. 5886-5898.
177. Zhao, C.X., L. Yu, and A.P.J. Middelberg, Magnetic mesoporous silica
nanoparticles end-capped with hydroxyapatite for pH-responsive drug release.
Journal of Materials Chemistry B, 2013. 1(37): p. 4828-4833.
178. Briggs, D., Surface Analysis of Polymers by XPS and Static SIMS. 1998:
Cambridge University Press.
179. Jasieniak, M., et al., Surface Analysis of Biomaterials, in Handbook of Surface
and Interface Analysis. 2009, CRC Press. p. 529-564.
180. Cotts, R.M., et al., Pulsed field gradient stimulated echo methods for improved
NMR diffusion measurements in heterogeneous systems. Journal of Magnetic
Resonance (1969), 1989. 83(2): p. 252-266.
181. Tan, B. and S.E. Rankin, Dual latex/surfactant templating of hollow spherical
silica particles with ordered mesoporous shells. Langmuir, 2005. 21(18): p. 8180-
8187.
182. Meng, H., et al., Development of pharmaceutically adapted mesoporous silica
nanoparticles platform. Journal of Physical Chemistry Letters, 2012. 3(3): p. 358-
359.
183. Patil, A., et al., Encapsulation of water insoluble drugs in mesoporous silica
nanoparticles using supercritical carbon dioxide. Journal of Nanomedicine and
Nanotechnology, 2011. 2(3): 1000111
184. Lin, Y.S., K.R. Hurley, and C.L. Haynes, Critical considerations in the
biomedical use of mesoporous silica nanoparticles. Journal of Physical Chemistry
Letters, 2012. 3(3): p. 364-374.
185. Chang, B., et al., General one-pot strategy to prepare multifunctional
nanocomposites with hydrophilic colloidal nanoparticles core/mesoporous silica
shell structure. Journal of Colloid and Interface Science, 2012. 377(1): p. 64-75.
186. Liu, J., et al., Yolk-shell hybrid materials with a periodic mesoporous organosilica
shell: Ideal nanoreactors for selective alcohol oxidation. Advanced Functional
Materials, 2012. 22(3): p. 591-599.
168
187. Yokoi, T., et al., Synthesis of mesoporous silica nanospheres promoted by basic
amino acids and their catalytic application. Chemistry of Materials, 2010. 22(13):
p. 3900-3908.
188. He, Q., et al., An anticancer drug delivery system based on surfactant-templated
mesoporous silica nanoparticles. Biomaterials, 2010. 31(12): p. 3335-3346.
189. Xia, X., et al. Resonant micro-cantilever chemical sensor with one-step synthesis
of -COOH functionalized mesoporous-silica nanoparticles for detection of trace-
level organophosphorus pesticide. 2012. Taipei.
190. De Arce Velasquez, A., et al., Novel Pullulan-Eudragit® S100 blend
microparticles for oral delivery of risedronate: Formulation, in vitro evaluation
and tableting of blend microparticles. Materials Science and Engineering C, 2014.
38(1): p. 212-217.
191. Mustafin, R.I., et al., Comparative pharmacokinetic assessment of diclofenac
sodium polycomplex drug delivery systems based on Eudragit copolymers.
Pharmaceutical Chemistry Journal, 2014. 48(1): p. 1-4.
192. Fontana, M.C., et al., Controlled release of raloxifene by nanoencapsulation:
Effect on in vitro antiproliferative activity of human breast cancer cells.
International Journal of Nanomedicine, 2014. 9(1): p. 2979-2991.
193. Hsu, F.Y., D.S. Yu, and C.C. Huang, Development of pH-sensitive
pectinate/alginate microspheres for colon drug delivery. Journal of Materials
Science: Materials in Medicine, 2013. 24(2): p. 317-323.
194. Coco, R., et al., Drug delivery to inflamed colon by nanoparticles: Comparison
of different strategies. International Journal of Pharmaceutics, 2013. 440(1): p. 3-
12.
195. Rosenholm, J.M. and M. Lindén, Towards establishing structure–activity
relationships for mesoporous silica in drug delivery applications. Journal of
Controlled Release, 2008. 128(2): p. 157-164.
196. Lu, Q., D. Chen, and X. Jiao, Fabrication of mesoporous silica microtubules
through the self-assembly behavior β-cyclodextrin and triton X-100 in aqueous
solution. Chemistry of Materials, 2005. 17(16): p. 4168-4173.
197. Singh, R.N., et al., Molecular structure, heteronuclear resonance assisted
hydrogen bond analysis, chemical reactivity and first hyperpolarizability of a
novel ethyl-4-{[(2,4-dinitrophenyl)-hydrazono]-ethyl}-3,5-dimethyl-1H-pyrrole-
169
2- carboxylate: A combined DFT and AIM approach. Spectrochimica Acta - Part
A: Molecular and Biomolecular Spectroscopy, 2012. 92: p. 295-304.
198. Szafran, M., et al., Molecular structures and hydrogen bonding in the 1 : 1 and
1 : 2 complexes of pyridine betaine with 2,6-dichloro-4-nitrophenol; an example
of strongly coupled hydrogen bonds, O-H···O=C-O-H···O. Journal of Molecular
Structure, 1997. 416(1-3): p. 145-160.
199. Szafran, M., et al., Conformational analysis of bis(trigonelline) hydrochloride,
perchlorate and their monohydrates by the B3LYP calculations, X-ray diffraction
and vibrational spectra. Journal of Molecular Structure, 2006. 784(1-3): p. 98-
108.
200. Barczyński, P., et al., Spectroscopic and structural investigation of 2,5-dicarboxy-
1- methylpyridinium inner salt. Spectrochimica Acta - Part A: Molecular and
Biomolecular Spectroscopy, 2014. 121: p. 586-595.
201. Barczyński, P., et al., Comparison of low-barrier hydrogen bonds in acid salts of
carboxylic acids and basic salts of betaines - FTIR study. Journal of Molecular
Structure, 1999. 484(1-3): p. 117-124.
202. He, Y., B. Zhu, and Y. Inoue, Hydrogen bonds in polymer blends. Progress in
Polymer Science (Oxford), 2004. 29(10): p. 1021-1051.
203. Gorgojo, P., et al., Exfoliated zeolite Nu-6(2) as filler for 6FDA-based
copolyimide mixed matrix membranes. Journal of Membrane Science, 2012. 411-
412: p. 146-152.
204. Wang, Y., et al. Formulation optimization for high drug loading colonic drug
delivery carrier. 2010. Yantai.
205. Moorthy, M.S., et al., Design of a Novel Mesoporous Organosilica Hybrid
Microcarrier: A pH Stimuli-responsive dual-drug-delivery vehicle for
intracellular delivery of anticancer Agents. Particle and Particle Systems
Characterization, 2013. 30(12): p. 1044-1055.
206. Tan, L., et al., Uniform double-shelled silica hollow spheres: Acid/base selective-
etching synthesis and their drug delivery application. RSC Advances, 2013. 3(16):
p. 5649-5655.
207. Chen, B., et al., Hollow mesoporous silicas as a drug solution delivery system for
insoluble drugs. Powder Technology, 2013. 240: p. 48-53.
170
208. Cao, X., et al., In vitro release and in vitro-in vivo correlation for silybin
meglumine incorporated into Hollow-type mesoporous silica nanoparticles.
International Journal of Nanomedicine, 2012. 7: p. 753-762.
209. Wan, Y., Y. Shi, and D. Zhao, Designed synthesis of mesoporous solids via
nonionic-surfactant-templating approach. Chemical Communications, 2007, 7(9):
p. 897-926.
210. Khimyak, Y.Z. and J. Klinowski, Formation of mesoporous silicates using Triton
XN surfactants in the presence of concentrated mineral acids. Journal of Materials
Chemistry, 2000. 10(8): p. 1847-1855.
211. Hill, M.R., et al., Internal and external surface characterisation of templating
processes for ordered mesoporous silicas and carbons. Journal of Materials
Chemistry, 2009. 19(15): p. 2215-2225.
212. Van Erp, T.S. and J.A. Martens, A standardization for BET fitting of adsorption
isotherms. Microporous and Mesoporous Materials, 2011. 145(1-3): p. 188-193.
213. Ladavos, A.K., et al., The BET equation, the inflection points of N 2 adsorption
isotherms and the estimation of specific surface area of porous solids.
Microporous and Mesoporous Materials, 2012. 151: p. 126-133.
214. Sing, K.S.W., et al., Assessment of Mesoporosity, in Adsorption by Powders and
Porous Solids: Principles, Methodology and Applications: Second Edition. 2013,
Elsevier Inc. p. 269-302.
215. Ritger, P.L. and N.A. Peppas, A simple equation for desciption of solute release I.
Fickian and non-Fickian release from non-swellable devices in the form of slabs,
spheres, cylinders or discs. Journal of Controlled Release, 1987. 5(1): p. 23-36.
216. Ritger, P.L. and N.A. Peppas, A simple equation for description of solute release
II. Fickian and anomalous release from swellable devices. Journal of Controlled
Release, 1987. 5(1): p. 37-42.
217. Salome, A.C., C.O. Godswill, and I.O. Ikechukwu, Kinetics and mechanisms of
drug release from swellable and non swellable matrices: A review. Research
Journal of Pharmaceutical, Biological and Chemical Sciences, 2013. 4(2): p. 97-
103.
218. Kosmidis, K., et al., Analysis of Case II drug transport with radial and axial
release from cylinders. International Journal of Pharmaceutics, 2003. 254(2): p.
183-188.
171
219. Bhattacharyya, S., et al., Inorganic nanoparticles in cancer therapy.
Pharmaceutical Research, 2011. 28(2): p. 237-259.
220. Seshacharyulu, P., et al., Targeting the EGFR signaling pathway in cancer
therapy. Expert Opinion on Therapeutic Targets, 2012. 16(1): p. 15-31.
221. Kralj, S., et al., Targeting EGFR-overexpressed A431 cells with EGF-labeled
silica-coated magnetic nanoparticles. Journal of Nanoparticle Research, 2013.
15(5): 1666.
222. Waddell, J.T., B. Liesenfeld, and C.D. Batich. Immobilization of EGF onto silica
nanoparticles. 2005. Boston, MA.
223. Xie, M., et al., Hybrid nanoparticles for drug delivery and bioimaging:
Mesoporous silica nanoparticles functionalized with carboxyl groups and a near-
infrared fluorescent dye. Journal of Colloid and Interface Science, 2013. 395: p.
306-314.
224. Sun, L., et al., Study on a carboxyl-activated carrier and its properties for papain
immobilization. Journal of Chemical Technology and Biotechnology. 2012. 87(8):
p. 1083-1088.
225. Kralj, S., M. Drofenik, and D. Makovec, Controlled surface functionalization of
silica-coated magnetic nanoparticles with terminal amino and carboxyl groups.
Journal of Nanoparticle Research, 2011. 13(7): p. 2829-2841.
226. Yiu, H.H.P., P.A. Wright, and N.P. Botting, Enzyme immobilisation using SBA-
15 mesoporous molecular sieves with functionalised surfaces. Journal of
Molecular Catalysis B: Enzymatic, 2001. 15(1–3): p. 81-92.
227. Blin, J.L., et al., Influence of alkyl peptidoamines on the structure of
functionalized mesoporous silica. Chemistry of Materials, 2004. 16(24): p. 5071-
5080.
228. Sevimli, F. and A. Yilmaz, Surface functionalization of SBA-15 particles for
amoxicillin delivery. Microporous and Mesoporous Materials, 2012. 158: p. 281-
291.
229. Shadjou, N. and M. Hasanzadeh, Amino functionalized mesoporous silica
decorated with iron oxide nanoparticles as a magnetically recoverable
nanoreactor for the synthesis of a new series of 2,4-diphenylpyrido[4,3-
d]pyrimidines. RSC Advances, 2014. 4(35): p. 18117-18126.
172
230. Alesker, M., et al., Hybrid silica nanoparticles traceable by fluorescence and FT-
IR spectroscopy: Preparation, characterization and preliminary biological
studies. Journal of Materials Chemistry, 2011. 21(29): p. 10883-10893.
231. Nedd, S., et al., Using a reactive force field to correlate mobilities obtained from
solid-state 13C NMR on mesoporous silica nanoparticle systems. Journal of
Physical Chemistry C, 2011. 115(33): p. 16333-16339.
232. Vasil'ev, S.G., et al., A Solid-State NMR Investigation of MQ Silicone Copolymers.
Applied Magnetic Resonance, 2013. 44(9): p. 1015-1025.
233. Moreau, J.J.E., et al., Lamellar bridged silsesquioxanes: Self-assembly through a
combination of hydrogen bonding and hydrophobic interactions. Chemistry - A
European Journal, 2005. 11(5): p. 1527-1537.
234. Hu, L., et al., Preparation and characterization of tungsten carbide confined in
the channels of SBA-15 mesoporous silica. Journal of Physical Chemistry B, 2007.
111(14): p. 3599-3608.
235. Huang, J., et al., Multifunctional mesoporous silica supported palladium
nanoparticles as efficient and reusable catalyst for water-medium Ullmann
reaction. New Journal of Chemistry, 2012. 36(6): p. 1378-1384.
236. Shevtsov, M.A., et al., Superparamagnetic iron oxide nanoparticles conjugated
with epidermal growth factor (SPION-EGF) for targeting brain tumors.
International Journal of Nanomedicine, 2014. 9(1): p. 273-287.
237. Matsumoto, R., et al., Targeting of EGF-displayed protein nanoparticles with
anticancer drugs. Journal of Biomedical Materials Research - Part B Applied
Biomaterials, 2014. 102(8): p. 1792-1798.
238. Chen, C., et al., EGF-functionalized single-walled carbon nanotubes for targeting
delivery of etoposide. Nanotechnology, 2012. 23(4). p. 045104.
239. Flavel, B.S., et al., Grafting of poly(ethylene glycol) on click chemistry modified
Si(100) surfaces. Langmuir, 2013. 29(26): p. 8355-8362.
240. Wagner, M.S., T.A. Horbett, and D.G. Castner, Characterization of the structure
of binary and ternary adsorbed protein films using electron spectroscopy for
chemical analysis, time-of-flight secondary ion mass spectrometry, and
radiolabeling. Langmuir, 2003. 19(5): p. 1708-1715.
241. Tidwell, C.D., et al., Static time-of-flight secondary ion mass spectrometry and x-
ray photoelectron spectroscopy characterization of adsorbed albumin and
fibronectin films. Surface and Interface Analysis, 2001. 31(8): p. 724-733.
173
242. Henry, M. and P. Bertrand, Surface composition of insulin and albumin adsorbed
on polymer substrates as revealed by multivariate analysis of ToF-SIMS data.
Surface and Interface Analysis, 2009. 41(2): p. 105-113.
243. Wagner, M.S. and D.G. Castner, Characterization of Adsorbed Protein Films by
Time-of-Flight Secondary Ion Mass Spectrometry with Principal Component
Analysis. Langmuir, 2001. 17(15): p. 4649-4660.
244. Henry, M., C. Dupont-Gillain, and P. Bertrand, Conformation Change of Albumin
Adsorbed on Polycarbonate Membranes as Revealed by ToF-SIMS. Langmuir,
2003. 19(15): p. 6271-6276.
245. Lhoest, J.B., et al., Fibronectin adsorption, conformation, and orientation on
polystyrene substrates studied by radiolabeling, XPS, and ToF SIMS. Journal of
Biomedical Materials Research, 1998. 41(1): p. 95-103.
246. Ratner, B.D., et al., Biomaterials Science - An Introduction to Materials in
Medicine (2nd Edition). Elsevier.
247. Sardan, M., et al., Noncovalent functionalization of mesoporous silica
nanoparticles with amphiphilic peptides. Journal of Materials Chemistry B, 2014.
2(15): p. 2168-2174.
248. Kamarudin, N.H.N., et al., Role of 3-aminopropyltriethoxysilane in the
preparation of mesoporous silica nanoparticles for ibuprofen delivery: Effect on
physicochemical properties. Microporous and Mesoporous Materials, 2013. 180:
p. 235-241.
249. Noble, A., Partition coefficients (n-octanol-water) for pesticides. Journal of
Chromatography, 1993. 642(1-2): p. 3-14.
250. Sato, S., et al., Enhancing effect of N-dodecyl-2-pyrrolidone on the percutaneous
absorption of 5-fluorouracil derivatives. Chemical and Pharmaceutical Bulletin,
1998. 46(5): p. 831-836.
251. Tomoiaga, A.M., et al., Investigations on nanoconfinement of low-molecular
antineoplastic agents into biocompatible magnetic matrices for drug targeting.
Colloids and Surfaces B: Biointerfaces, 2013. 111: p. 52-59.
252. Wierzchowski, K.L., E. Litonska, and D. Shugar, Infrared and ultraviolet studies
on the tautomeric equilibria in aqueous medium between monoanionic species of
uracil, thymine, 5-fluorouracil, and other 2,4-diketopyrimidines. Journal of the
American Chemical Society, 1965. 87(20): p. 4621-4629.
174
253. Markova, N., V. Enchev, and G. Ivanova, Tautomeric equilibria of 5-fluorouracil
anionic species in water. Journal of Physical Chemistry A, 2010. 114(50): p.
13154-13162.
254. Meng, H., et al., Codelivery of an optimal drug/siRNA combination using
mesoporous silica nanoparticles to overcome drug resistance in breast cancer in
vitro and in vivo. ACS Nano, 2013. 7(2): p. 994-1005.
255. Zhang, X., et al., Biofunctionalized polymer-lipid supported mesoporous silica
nanoparticles for release of chemotherapeutics in multidrug resistant cancer cells.
Biomaterials, 2014. 35(11): p. 3650-3665.
256. Na, H.K., et al., Efficient functional delivery of siRNA using mesoporous silica
nanoparticles with ultralarge pores. Small, 2012. 8(11): p. 1752-1761.
257. Sun, H., et al., Preparation and in vitro/in vivo characterization of enteric-coated
nanoparticles loaded with the antihypertensive peptide VLPVPR. International
Journal of Nanomedicine, 2014. 9(1): p. 1709-1716.
258. Xu, M., et al., Preparation and evaluation of colon adhesive pellets of 5-
aminosalicylic acid. International Journal of Pharmaceutics, 2014. 468(1-2): p.
165-171.
259. Dandekar, P., et al., Toxicological evaluation of pH-sensitive nanoparticles of
curcumin: Acute, sub-acute and genotoxicity studies. Food and Chemical
Toxicology, 2010. 48(8-9): p. 2073-2089.
260. Maghsoodi, M. and F. Sadeghpoor, Preparation and evaluation of solid
dispersions of piroxicam and Eudragit S100 by spherical crystallization technique.
Drug Development and Industrial Pharmacy, 2010. 36(8): p. 917-925.
261. Niu, D., et al., Synthesis of Core−Shell Structured Dual-Mesoporous Silica
Spheres with Tunable Pore Size and Controllable Shell Thickness. Journal of the
American Chemical Society, 2010. 132(43): p. 15144-15147.
262. Miller, L., et al., Synthesis, characterization, and biodistribution of multiple 89Zr-
labeled pore-expanded mesoporous silica nanoparticles for PET. Nanoscale,
2014. 6(9): p. 4928-4935.
263. Huang, X., et al., The shape effect of mesoporous silica nanoparticles on
biodistribution, clearance, and biocompatibility in vivo. ACS Nano, 2011. 5(7):
p. 5390-5399.
175
Publication List
1. X. D. She, C. Z. He, Z. Peng, L. X. Kong, Molecular-Level Dispersion of Graphene into Epoxidized Natural Rubber: Morphology, Interfacial Interaction and Mechanical Reinforcement, Polymer, 2014, 55 (26), 6803-6810
2. X. D. She, L.J. Chen, L. Velleman, C. P. Li, H. J. Zhu, C. Z. He, T. Wang, S. Shigdar, W. Duan, L. X. Kong, Fabrication of High Specificity Hollow Mesoporous Silica Nanoparticles Assisted by Eudragit for Targeted Drug Delivery, Journal of Colloid and Interface Science, 2015, 445, 151-160
3. C. P. Li, J. Vongsvivut, X. D. She, Y. Z. Li, F. H. She and L. X. Kong, New insight into non-isothermal crystallization of PVA/graphene composites, Physical Chemistry Chemical Physics, 2014, 16 (40), 22145-22158
4. X. D. She, L. Velleman, L. X. Kong, New technologies and tools: Self-Assembly synthesis of hollow mesoporous silica nanoparticles for colon targeting drug delivery systems, Clinical and experimental pharmacology and physiology, 2013, 40(supplement. s1), 52-52
5. C. P. Li, M. She, X. D. She, J. Dai, L. X. Kong, Functionalization of polyvinyl alcohol hydrogels with graphene oxide for potential dye removal, Journal of Applied Polymer Science, 2014, 131(3). DOI: 10.1002/APP.39872