Development of Janus Nanocomposites As A

Development of Janus Nanocomposites as a

Multifunctional Nanocarrier for Cancer Therapy

A dissertation Submitted to the Division of Research and Advanced Studies of the University of Cincinnati in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY in the School of Energy, Environmental, Biological & Medical Engineering of the College of Engineering and Applied Science 2013 by

Feng Wang

B.S., Nanjing University of Technology, China 2005

Committee Chair: Dr. Donglu Shi

ABSTRACT

With the advancement of nanotechnology, cancer therapy requires that the carrier at nanoscale integrates cell targeting, imaging, drug storage and controlled drug release simultaneously. Extensive efforts have been devoted to isotropic sphere nanoparticle based carrier systems for their uniform surface properties. However, multifunctionality also leads to a challenge issue: different moieties may interfere with each other in bioconjugation process due to the similar conjugation chemistry applied to the same surface. As a result, Janus nanoparticles which are anisotropic in shape, composition or surface chemistry have attracted increasing attention. Asymmetric composition could achieve multifunctionality simultaneously. More importantly, surfaces could be selectively loaded with targeting ligands, imaging probes or drugs, which made the Janus nanoparticles “truly multifunctional entities”.

A variety of fabrication methods have been studied to synthesize Janus nanoparticles for applications such as surfactants, magnetic-fluorescent display or imaging, and catalysts, etc. In contrast, exploration in the biomedical application field is rather limited. Based on our previous work on iron oxides@polystyrene matrix multifunctional nanoparticles and yolk-shell nanocomposites, we designed the polystyrene/Fe3O4@SiO2 superparamagnetic Janus nanocomposites (SJNCs). The SJNCs (~300 nm) are composed of a polystyrene (PS) core and a silica half shell embedded with iron oxide nanoparticles. We demonstrated the innovative dual

functionalities on independent surfaces were obtained simultaneously during one-pot facile synthesis, which is much more convenient than the previous report on generating the similar structure through selectively coating. PS surfaces were decorated with carboxyl groups and silanol groups on silica surfaces provided enormous opportunities for further functionalization.

To achieve cell targeting and controlled drug release, we conjugated folic acid (FA) to the PS and doxorubicin (DOX), an anti-cancer drug to the silica via a pH-sensitive hydrazone bond

(FA-SJNCs-DOX). Drug release behaviors in different buffer solutions (pH 5.0, 6.0 and 7.4) displayed clear pH dependence. To evaluate the in vitro cell targeting and drug release, cell cytotoxicity of FA-SJNCs-DOX against human breast cancer line MDA-MB-231 was tested.

Significant difference of the concentrations killing 50 % of the cells (IC50) was observed from targeted and non-targeted group. It is hypothesized that targeted nanocomposites (FA-SJNCs-

DOX) could be internalized by cancer cells via folate receptor mediated endocytosis and release the drug at a faster rate in endocytic compartments (pH 4.5~6.5), compared to under physiological condition (pH 7.4) from the non-targeted group (SJNCs-DOX), thus providing highly localized controlled drug delivery. More interestingly, the incorporated iron oxides could provide potential application such as MRI, magnetic targeting and hyperthermia, thus making the

Janus nanocomposites a truly versatile platform for wide biomedical application.

ACKNOWLEDGEMENT

I would like to express my sincere gratitude to my academic advisor, Dr. Donglu Shi, for his encouragement, support, patience, and valuable advice in academic research and scientific writing during my study in University of Cincinnati. I would acknowledge my committee members, Drs. Giovanni M. Pauletti, Vesselin Shanov, and Vikram K. Kuppa on for their suggestions on dissertation. This research has been supported by National Natural Science

Foundation of China (No.51003077; No.51173135), Shanghai Nano-program (No.

11nm0506100), and the Fundamental Research Funds for the Central Universities. I’m grateful to Drs. Yilong Wang and Giovanni M. Pauletti for their guidance in this highly interdisciplinary study. I have gained a lot of experience on material synthesis and tissue culture and most importantly, confidence to face challenges in unknown fields. I acknowledge Bing Han, Andrew

D. Gilpin, Xiaoping Chen and Dr. Bo Hu for helping with characterization and providing valuable suggestions on structure analysis. My gratitude also goes to lab-mates, Ronak Patel, Drs.

Sheng Tong, Hoonsung Cho and Chris Huth for their support in every way.

I would like to give my sincere thanks to my parents, my uncle and aunt, my cousin and my dear grandma who passed away last month, for their unconditional love, support and patience.

I also want to thank my friends in University of Cincinnati and in The Institute for

Biomedical Engineering and Nano Science, Tongji University, China, Xuecheng Dong,

Dingchuan Xue, Xiang Gao, Linqian Feng, Fei Yu, Hua Li, Peng He, Vibhor Chaswal, Guojun

iii

Zhang, Yin Chen, Mary Lim, Zhouyang Liu, Mingyu Zhong, Xiaojun Cai, Huiyun Wen and

Xuequan Li for their help and encouragement.

Contents ABSTRACT ...... i

ACKNOWLEDGEMENT ...... iii

Chapter 1. Introduction ...... 1

1.1. Targeting ...... 3

1.2. Drug Delivery ...... 5

1.2.1 Micelles ...... 5

1.2.2. Dendrimers ...... 7

1.2.3. Quantum Dots ...... 8

1.2.4. Inorganic Nanoparticles ...... 10

1.3. Magnetic Resonance Imaging ...... 15

1.4. Development of Janus Particles ...... 16

1.4.1. Introduction to Janus Particles ...... 16

1.4.2. Inorganic Janus Particles ...... 18

1.4.3. Polymeric Janus Particles ...... 20

1.4.3. Polymeric/Inorganic Janus Composites ...... 22

1.4.4. Applications of Janus Particles ...... 25

1.5. Research objectives and strategy: Multifunctional Janus Nanocomposites for Biomedical Application ...... 26

1.6. References ...... 27

Chapter 2. Synthesis and Application of Multifunctional Nanocarrier System for Cancer Therapy ...... 36

2.1. Background ...... 36

2.2. Experimental ...... 38

2.2.1. QD Conjugation to MNSs ...... 38

2.4. Conclusion ...... 43

2.5. References ...... 44

Chapter 3. Design and Development of Superparamagnetic Janus Nanoparticles as a Multi- functional Tool for Cancer Therapy ...... 45

3.1. Introduction ...... 45

3.2. Design and Development of Polystyrene/Fe3O4@Silica nanocomposites ...... 46

3.2.1. Experimental ...... 46

3.2.2. Results and Discussion ...... 48

3.3. Design and Development of Dual Functionalized Polystyrene/Fe3O4@Silica nanocomposites ...... 51

3.3.1. Experimental ...... 52

3.5. References ...... 69

Chapter 4. Drug Loading and Release Study on Silica Nanoparticles ...... 71

4.1. Introduction ...... 71

4.2. Experimental ...... 73

4.2.1. Mateials and Chemicals ...... 73

4.2.2. Methods...... 74

4.2.3. Characterization ...... 77

4.3. Results and Discussion ...... 79

4.3.1. Silica Nanoparticle Synthesis and Doxorubicin Conjugation ...... 79

4.3.2. Drug loading and release study ...... 81

4.3.3. Cell Cytotoxicity Analysis ...... 85

4.4. Conclusion ...... 90

4.5. References ...... 90

Chapter 5. Folic Acid Conjugation, Drug Loading and in-vitro Release and Cytotoxicity Study of Superparamagnetic Janus Nanocomposites ...... 92

5.1. Introduction ...... 92

5.2.2. Methods ...... 96

5.4. Conclusion ...... 115

5.5. References ...... 116

Chapter 6. Summary and Outlook ...... 119

Chapter 1. Introduction

With the rapid development of nanotechnology, there has been an increasing need that nanomateirals can address the key clinical issues in drug delivery, hyperthermia, imaging and cell targeting, preferably in a combined fashion, thus helping doctors with cancer diagnosis and treatment simultaneously. There has been extensive study on establishing nanocarrier systems achieving isolated functionalities. However, driven by the clinical application, more research is focusing on developing an integrated nanocarrier system to achieve drug delivery, imaging, cell targeting and hyperthermia simultaneously, which leads to the terminology of

“multifunctionality”.

Such nanocarrier systems must have the following features to achieve multifunctionality.

1. The surface of the nanocarrier system bears functional groups. This could allow further manipulation for attachment of biological moieties such as drugs and targeting ligands. The surface properties might be the most important to the nanocarriers because in most cases, bio- application is obtained through surface functionalization. Besides, surface properties determine the colloidal stability of the nanocarrier systems.

2. The nanocarriers need to be ready for imaging. One way is to label the nanocarriers with fluorescent markers or the nanocarriers themselves are fluorescent. The other way is through magnetic resonance imaging (MRI), which requires that nanocarriers have magnetic components.

3. The nanocarriers have to be non-toxic, preferably biodegradable to be considered for clinical application.

4. Some nanocarriers are designed to be hollow structured for drug storage and appeared quite promising regarding the examples of mesoporous silica nanoparticles and polymer micelles.

An idealized multifunctional nano-carrier designed for diagnostic and therapeutic purpose is illustrated in Figure 1.

Figure 1.1. Idealized schematic illustration of a nano-carrier for cancer diagnosis and treatment.1

In this study, a novel dual functional Janus polymer/inorganic magnetic nanocomposite was designed and synthesized. The two distinct surfaces were individually fuctionalized to achieve cell targeting, controlled drug delivery and potentially MRI simultaneously.

1.1. Targeting

One fact that people developing the drug delivery carriers have to face is that the delivery systems may be, as well, harmful to healthy tissues. The side effects from cancer therapy are often disturbing, sometimes even fatal. Nanocarrier systems with cell targeting ability offer site specificity and deliver cargos directly to tumor sites with much less damage to healthy tissues, thus improving therapeutic efficacy.2,3 Therefore, cell targeting is essential for both cancer diagnosis and treatment and becomes a key component of the multifunctional nanocarrier systems.4-6

Active targeting is usually achieved by conjugation of nanocarrier surfaces with targeting moieties. By far, the most widely used ligands are antibodies and antibody fragments. For example, Anti-EGFR antibodies have been conjugated to liposomes7,8 and nanoparticles9-11 for diagnostic and therapeutic purposes. Anti-prostate-specific membrane antigen (anti-PSMA) is another good example. Thomas and colleagues12 conjugated dendrimers with anti-PSMA antibodies and investigated the targeting effect on LNCap cells. It was shown that targeted dendrimers achieved specific binding followed by internalization compared to non-targeted group and competition control. Gao et al13 functionalized PEGylated quantum dots with anti-

PSMA antibodies (Figure 1.2) and performed in vivo imaging using a whole-body macro- illumination system. Only the targeted group showed a strong signal from the tumor sites.

Figure. 1.2. Schematic drawing illustrating the structure of a multifunctional QD probe. The

QDs are stabilized by the capping ligand; a copolymer layer enables encapsulation; the targeting ligands (antibodies, peptides, or small molecules) help achieve specific bonding and (PEG) improves hydrophilicity.13

In some cases, antibodies or antibody fragments are conjugated via carbodiimide- mediated reaction between amino groups of antibodies and carboxyl groups of nanocarriers.14

However, antibodies or antibody fragments could be conjugated to the carriers in any ways, i.e. through any amino groups on the backbone. This random conjugating could possibly block the binding sites of antibodies and reduce the targeting effectiveness.14 There have been reports that antibody activity decreased after binding to PEG chains.15-17 As a result, the maleimide chemistry becomes more favored by scientists. Thiol functional groups could be engineered away from the binding sites, thus maintaining the binding activity.

Other targeting ligands such as peptides18-22 and folic acid23-27 have also been used.

Peptides can be considered as the short version of antibodies, possessing similar receptor-

specificity but much easier to handle in chemistry. Folate receptors (FR), which specifically bind folic acid, facilitate fast internalization and cell surface recycling.28 This could in turn ensure accumulation of more nanocarriers at the tumor cells.29

1.2. Drug Delivery

Another key function of the nanocarrier system is to store and release drugs in a controlled fashion for localized delivery to the lesions. According to the structure, nanocarrier system can be divided into nanocapsules, in which drugs are confined and protected by polymer membranes, and nanoparticles, the matrix to which drugs can be physically or chemically attached.30

In conventional chemotherapy, the treatment affects both cancer cells and healthy cells throughout the body. However, because of the small size of nanocarrier systems, efficient uptake by a variety of cell types can be obtained and drugs are accumulated at a relatively high level at the desired sites.31-33 The use of nanocarrier systems offers potential higher efficacy and lower toxicity. Besides, some hydrophobic drugs can also be encapsulated for delivery. The nanoscaled drug-carrier complexes are expected to improve both the pharmacological and therapeutic properties of the drugs incorporated.34

1.2.1 Micelles

Block copolymers generally possess one hydrophobic end and one hydrophilic end. They

tend to self assemble in aqueous systems to form a core-shell structure with a hydrophobic core and a hydrophilic shell. The outer surfaces of micelles can be PEGylated to form a brush-like corona for improved water solubility and circulation half-life while the core provides accommodation for hydrophobic drugs. In addition, molecular weights as well as the chemical compositions of blocks can be readily manipulated to control the physical properties of micelles such as morphology and size35. In the case of cross-linking induced micellization, cross-linking agents are introduced into the block copolymer solutions, and self assembling will be initiated once cross-linking is finished.35 With these unique advantages, micelles have been widely investigated36-39 for drug delivery purpose and clinically approved models have been developed.40 Recently, the interest has shifted to active targeting by grafting targeting ligands and controlled release instead of physically diffusion. For example, Bae41 and Yang42 (Figure 1.3) both constructed the micelles with doxorubicin attached to the hydrophobic inner core via a pH- sensitive hydrazone bond and folic acid conjugated to hydrophilic outer surface. Once the micelles are internalized by tumor cells, the decrease of pH in endocytic compartments cleaves the hydrazone bond and releases the drug.

Figure 1.3. Schematic structure of the micelle delivery system.42

1.2.2. Dendrimers

Dendrimers are composed of a core molecule as the root from which branched structures generate in a repeating, symmetric fashion.43 Dendrimers are generally highly monodispersed with small sizes (~10 nm). Cargo molecules can be stored in the interior or attached to the stretching arms (Figure 1.4). The core can be small molecules, nanoparticles, other dendrimers or polymers. The surface groups are easily taliored to biocompatible polymers, targeting ligands, dyes or biomarkers.44

Figure 1.4. Schematic 2D presentation of a dendrimer G3 containing three generations.44

Initially cargos such as DNA,45 dyes or hydrophobic drugs46-48 were physically encapsulated in the vacancy among the dendritic arms. Later people began to covalently conjugate guest molecules to the periphery of dendrimers via biodegradable bonds so that the release can be controlled.49 For example, Duncan and co-workers50 conjugated cisplatin to

PAMAM dendrimers and found that the solubility improved and the systemic toxicity decreased.

Quintana et al.51 managed to conjugate methotrexate to the PAMAM dendrimer periphery via an ester bond which could undergo hydrolysis at biological conditions.

1.2.3. Quantum Dots

Quantum dots (QDs) are nano-sized spherical crystals with a core-shell structure ranged from 2-10 nm. The core is usually composed of semiconductor metal (CdS, ZnS, PbS, CdSe) and coated with a semiconductor shell to enhance optical property as well as stabilize the core.52 To make QDs water soluble, normally an amphiphilic polymer layer is coated on the surface of the

shell. By tailoring the ending groups of the polymer chains, various kinds of chemical reactive sites are generated for further modification for bio-application purpose.53

QDs have extraordinary optical properties for imaging: the emission is size related and can be tuned easily and; the fluorescence is brighter and more resistant to photobleaching.54,55

There has been investigations on utilizing QDs as the carrier for drug delivery.56,57 However, there are increasing concerns about QD-based nanocarrier systems because some researchers reported QDs may be toxic. But more information on the toxicology of QDs is needed before further development can be made.58

Figure 1.5. Schematic drawing illustrating popular methods of conjugating

biological molecules to QDs. The QDs are functionalized with mercaptoacetic

acid.59

1.2.4. Inorganic Nanoparticles

Compared with the polymer based nanoparticle drug delivery system, inorganic nanoparticles seem to be less favored by researchers. But they have unique advantages: they are easy to fabricate and often render multi functions such as contrasts and exothermic reactors.34

Iron Oxide Nanoparticles

At first, iron oxide nanoparticles with the size of 50-100 nm were prepared as contrast agents for magnetic resonance imaging (MRI) in 1980s with their applications for drug storage attracted much attention roughly at the same time.60,61 Since then, more researchers made efforts modifying the surface of iron oxide nanoparticles for more efficient drug delivery. In Lubbe and coworker’s 62-64 reports, the iron oxide nanoparticles (~200 nm) were coated with starch polymers for steric stabilization. Then mitoxantron, an anticancer drug was absorbed to the polymer layer via electrostatic interaction. Mitoxantrone was released within 60 min from the conjugates. When tested in vivo on tumor bearing rabbits under magnetic field, therapeutic effect was observed with visualization of iron oxide nanoparticle accumulation both histologically and by MRI.

Jain et al.65 coated iron oxide nanoparticles with oleic acid (OA) and then another layer of PEG-polypropylene oxide (PPO)-PEG block co-polymer. The anticancer drug doxorubicin

(DOX) was encapsulated in the OA layer while the PEG segments made the complexes hydrophilic. The drug was released by diffusion.

Wuang et al.66 first coated iron oxide nanoparticles (~10 nm) with Poly(methacrylic acid)

(P(MAA)) via a surface initiated atom transfer radical polymerization (ATRP), and then grafted

hydrazide ending groups to P(MMA). DOX was then covalently bonded to the iron oxide nanoparticles via a pH-sensitive hydrazone linkage (Figure 1.6). The in vitro release profiles proved that the conjugates released more drug in acidic environment, which was beneficial to increase drug accumulation in tumor sites at low treatment dosage.

Other methods of surface modification include turning hydroxyl groups into amino groups, carboxyl groups, aldehyde groups or thiol groups by using silane-type cross-linkers. For example, Shkilnyy et al.67 coated iron oxide nanoparticles with (3-

Glycidoxypropyl)trimethoxysilane, and then further conjugated DOX and amino group functionalized PEG. The covalent bond formed between amine and epoxide functional groups was stable in solution but subjected to degradation in the cancer cells. Cell cytotoxicity profiles indicated DOX release with significant time delay.

Due to the potential application of combined magnetically-induced hyperthermia that sensitizes tumor cells to cytotoxic drugs68 and MRI for diagnosis, iron oxide nanoparticles provide attractive possibilities as drug carriers.

Figure 1.6. Schematic illustration showing DOX conjugation to P(MMA) functionalized iron

oxide nanoparticles via hydrazone bonding.66

Mesoporous Silica Nanoparticles

Mesoporous silica nanoparticles (MSNs) are composed of hundreds of empty channels for foreign cargo storage. MSNs possess extremely high surface area for functionlization, large pore volume for encapsulation and tunable pore size with relatively narrow distribution. These advantages make MSAs a competitive candidate for drug delivery carrier.69

Many have made efforts exploring the relationship between the morphology of MSNs and the drug storage performances. It was found that mesoporous silica nanoparticles with channel-like pores packed in a hexagonal fashion loaded more drugs with a relatively longer release time period. On the other hand, pore size did not play an important role in determining drug loading efficiency and release behaviors.70

Meanwhile, chemically controlled drug release was also investigated. Lai and coworkers71 built a series of MSN delivery systems with gatekeepers that can respond to stimuli.

To evaluate the system, they encapsulated vancomycin in the channels and covalently bonded

CdS nanoparticles as the gatekeepers via a chemically cleavable disulfide linkage (Figure 1.7).

Drugs can be released upon the cleavage of disulfide bonding. It was found that in the absence of reducing agents, the system showed “zero premature release” in phosphate buffer saline solution in 12 h. In contrast, with disulfide reducing agents present, 85 % of the drug entrapped was released in the initial 24 h.

Figure 1.7. Schematic representation of the MSN-based release system with disulfide linked

CdS as gatekeepers.71

Despite the successful demonstration of drug delivery by MSNs, scholars are still investigating if MSN based system can deliver other types of biogenic agents into the cells and how to assess the long term bio-compatibility of those systems.69

Gold Nanoparticles

Researchers found gold nanoparticles an interesting candidate for drug delivery carrier because:

1. Gold nanoparticles are not difficult to make. Normally gold salts are reduced to produce gold nanoparticles.72

2. They are low toxic.73

3. Gold nanoparticle surfaces are easily modified with thiol-group containing polymers for stabilization. More importantly, further conjugation of bio-molecules such as DNA, drugs, targeting ligands and imaging probes can be carried out by manipulating the polymer layer.34,74

Gu and coworkers75 functionalized gold nanoparticles with vacomycin derivatives bearing thiol groups and achieved good antibiotic activity. Aryal et al.76 successfully coated gold nanoparticles with thiolated methoxy poly(ethylene glycol) and methyl thioglycolate (MTG).

MTG was further modified with hydrazide groups to covalent bond DOX. In vitro drug release profiles indicated that more drug was released because the hydrazone bond was subjected to hydrolysis under mild acidic environment (Figure 1.8).

Figure 1.8. Scheme for conjugation of DOX to gold nanoparticles.76

Very interestingly, gold nanoparticles are also able to serve as photothermal agents with a tunable absorption maximum related with the particle size.77 El-Sayed et al.9 conjugated gold nanoparticles with an anti-epidermal growth receptor to target human cancer cells. Cell death was observed when the cancer cells incubated with gold nanoparticles were irradiated with a laser of 514 nm.

1.3. Magnetic Resonance Imaging

Magnetic Resonance Imaging (MRI) is an imaging technology used widely for medical purposes. Due to the presence of large amount of water in biological tissues, there are large quantities of protons with small magnetic moment. The protons are aligned in the presence of a large magnetic field but are excited by a radio frequency (RF) pulse. Protons will return to their equilibrium status after the RF pulse is removed. This relaxation process will generate frequency

78,79 signals that can be used to construct tissue images. The relaxation time can be divided into T1 and T2, in which T1 represents longitudinal (or spin-lattice) and T2 represents the transverse (or

80 spin-spin) relaxation times. Both T1 and T2 relaxation time can be reduced to enhance MRI.

Superparamagnetic iron oxides (SPIOs) have been studied for many years as contrast agents with

81 greater reduction on T2 relaxation time. SPIOs coated with different materials (dextran, starch,

PEG) in different sizes have been investigated.82,83 Currently SPIOs lend themselves as a key component of the nanocarrier systems so that they could incorporate the capabilities of MRI and

magnetic induced hyperthermia to achieve combined diagnostic and therapeutic functions.

1.4. Development of Janus Particles

1.4.1. Introduction to Janus Particles

As the understanding of nanoscaled materials goes deeper, many efforts were devoted to constructing nanocarrier systems that can achieve multifunctionalities. Because such systems can potentially revolutionize the approaches for diagnosis and treatment of cancer, it is reasonable to predict that nanoparticle based systems will continue to improve and eventually reach the clinic.84

However, most of the approaches used currently have been concentrated on surface functionalization of nanoparticles with different drugs,85 biological molecules, including DNA,86

RNA,87 peptide,88 antibodies,89 and imaging probes such as quantum dots.90 One of the major disadvantages in this approach is characteristically the single surface structure of the nanoparticles. A nanoparticle normally possesses a symmetrical geometry with limited surface functionalities available for multiple components. Furthermore, multi-functional conjugates on a single carrier could interact with each other, leading to adverse effects. The design and assembly of a single surface symmetrical carrier is also subject to difficulties in the structural and chemical arrangements of the functional components. For example, in Shkinlyy’s model,67 DOX and PEG were both cross-linked to silane and then conjugated to the surface of iron oxides (Figure 1.9).

Because the two species were attached to the iron oxide surfaces via the same silane-glass coupling chemistry, it is reasonable to speculate there was competition between DOX and PEG

and it was difficult to control the distribution of them. It was also possible that PEG chains would have steric repulsion for DOX.

Figure 1.9. Schematic drawing illustrating the modification of iron oxide surface with DOX and

PEG.67

It is, therefore, critical to develop multi-surface nanostructures for assembly of a variety of components on a clinically viable delivery system that can best utilize the intrinsic properties of nanomaterials. So far, scientists have well understood the fundamentals of isotropic nanoparticles. However, inspired by the new possibilities of desirable structures and behavior offered by anisotropic nanoparticles, researchers have begun to explore ways to fabricate asymmetric nanoparticles.91 Attempts have been made to generate anisotropy in shape as well as functionality on microparticles,92-94 but it has become extremely challenging when it comes to the nanoscale. Janus particles, named after the Roman god who has two distinct faces, possess multiple surface structures that are anisotropic in shape, composition, and surface chemistry.95

Their structural and functional asymmetry makes them ideal candidate for assembly of multiple components on a single-particle system.96 More importantly, the multiple surfaces with different functional groups can be used to selectively conjugate with targeting ligands, imaging probes, or drugs.97-99 These characteristics of the Janus particles make them “truly multifunctional entities”.91 In the following sections, approaches for fabricating Janus particles of both micro and nano size are briefly reviewed. As mentioned above, it still remains challenging to extend the existing methods for producing micro-sized Janus particles to fabricate Janus nanoparticles.

1.4.2. Inorganic Janus Particles

Generally inorganic Janus particles can be categorized into two types:95

1. The particle consists of two or more inorganic phase, each component expressing different optical, electrical or magnetic properties.

2. The particle is composed of one material phase but possesses distinct chemical properties on both sides.

There has been a large amount of literature reporting the design and fabrication of inorganic Janus particles. One flexible method widely employed is masking, which involves partially coverage of the matrix materials and subsequent growth of other inorganic phases on the exposed matrix via chemical functionalization,100 electrostatic attraction101 or metal evaporation.102 For example, Bae et al. first crystallized silica nanoparticles on a silicon wafer of several layer thick in a close packed pattern. Then the surfaces of silica nanoparticles were coated with octadecyltrichlorosilane self-assembled monolayers (OTS-SAMs), only left the areas

that neighboring nanoparticles touching uncoated. After the top layers of silica nanoparticles were removed by tape, TiO2 nanoparticles were deposited onto the uncoated areas of the silica nanoparticle surfaces.103 In another example of masking, McConnell et al.104 produced multiregional Janus nanoparticles by applying a horizontal substrate to protect the bottom of the matrix materials. Amino group functionalized silica nanoparticles were sank into horizontal styrene-acrylic acid random copolymer films swollen by solvent. Then gold nanoparticles were deposited onto exposed surfaces of silica nanparticles by electrostatic interaction (Figure 1.10).

Figure 1.10. Schematic illustration of the formation of multiregional Janus particles.104

Ligands can also be used to link the two distinct parts of a Janus nanoparticle. Zhang et al.105 employed hexadecyltrimethylammonium bromide (CTAB) as the linker as well as the template in aqueous solution to fabricate Janus nanoparticles with Fe3O4 heads and mesoporous silica tails. The length of silica tails could be finely tuned by adjusting the amount of tetraethyl orthosilicate added.

To produce Janus nanoparticles with one single material phase but two different surface

properties, another method called Pickering emulsion was developed by Granick and coworkers.95,106,107 In this model, nanoparticles are adsorbed onto the interface of a oil-in-water emulsion system. Then the oil phase is frozen and chemical modification is carried out on the particle surfaces exposed to water. In the last stage the oil phase is removed to release nanoparticles with asymmetric surface functional groups. For example, Granick et al. dispersed fused silica particles in molten wax and mixed with water to form an emulsion system. After wax was cooled down to freeze, the exposed surface of silica particles were functionalized with 3-

(aminopropyl)triethoxysilane to generate amino groups. The wax was later dissolved by chloroform and silica particles could be conjugated further with octadecyltrichlorosilane on the other side to make the silica particles amphiphilic.107 This technique could also be considered in the category of masking approach.

1.4.3. Polymeric Janus Particles

The most widely used method for producing polymeric Janus particles is phase separation, involving the interactions between polymers and monomers, among polymer chains and between the particles and aqueous phase.108,109 One of the common approaches features using polymers as the seeds for the further growth of another polymer via emulsion polymerization. Another method is based on copolymer blocks ABC. The middle block B is cross-linked with the two ends A and C to generate a two-module Janus particle. Although phase separation is an relatively easy process, it is difficult to control the morphology of the Janus particles and the production rate is often very low.95

Very similarly to inorganic Janus particle synthesis, the masking approach also finds its application in polymeric Janus particle fabrication. The mechanism is to functionalize partial surfaces of the pre-synthesized polymer particles, either by masking the other part in a substrate or carrying out the modification at interfaces. So only the exposed surfaces would be subjected to subsequent growth of other polymers.110,111 For example, Bradley and Rowe attached positively charged poly(2-vinylpyridine) microgel particles to negatively charged poly(2-vinylpyridine-co- styrene) latex particles and grafted poly (N-isopropylacrylamide) to the exposed surfaces to produce asymmetric polymeric composites.112

Microfluid can also be employed to prepare Janus particles. The dispersed phase containing monomers is injected perpendicular into an immiscible continuous phase to produce monomer droplets in a stream for polymerization.113 Lone et al. took water as the dispersed phase and hexane as the continuous phase. The microfluidic device produced water droplets containing light-responsive polymers and cross-linkers. Under UV irradiation, Janus structures were formed in the droplets (Figure 1.11).113

Figure 1.11. (a) Route to fabricate Janus particles using microfluid. (b) Optical microscopic image of Janus particles.113

Compared to phase separation and masking, which have limitation in production quantities, microfluid can be easily scaled up for larger amounts.95 However, the sizes from microfluid approach are a lot bigger due to the droplet sizes formed during injecting.

1.4.3. Polymeric/Inorganic Janus Composites

Combining advantages from both organic and inorganic composition, the polymeric/inorganic composites attracted extensive interests as they have potential application as catalysts, functional coatings and biosensors.91 Polystyrene/silica was investigated most among the model structures that have been developed, possibly because both have relatively simple and controllable fabrication process.

Figure 1.12. (a) Process to produce mushroom Janus nanocomposites. (b) Dark field STEM of

PS/Fe3O4@SiO2. (c) SEM and dark field STEM overlapping image of PS/Fe3O4@SiO2. (d) TEM

114 image of PS/hollow SiO2 after Fe3O4 is etched.

Immobilization (selective coating) remains useful in fabrication polymer-inorganic Janus nanocomposites. Feyen and coworkers managed to fabricate a mushroom ternary Janus strucute of Polystyrene/Iron [email protected] First oxides were immobilized onto the polystyrene nanoparticles through emulsion to serve as growth sites for silica. Then silica nanospheres were formed upon iron oxides via sol-gel reaction. The iron oxides could be etched away by hydrochloride to generate hollow silica nanoparticles (Figure 1.12). Gong et al. grew ZnO nanowires on part of the PS surfaces.115 Carboxyl group functionalized PS particles were sank in

2+ a polymer film and incubated in the ZnCl2 aqueous solution. As a result Zn ions deposited onto the exposed PS surfaces due to electrostatic attraction. Then ZnO nanowires could grow through an electrochemical process.

Figure 1.13. Formation of asymmetric PS/SiO2 composites by radiation initiated miniemulsion polymerization.116

Phase separation represents another important approach in generating polymer-inorganic composites. Because polystyrene can be synthesized easily through emulsion polymerization, phase separation often involves miniemulsion process. Ge et al.116 made silica nanoparticles aggregate by addition of very small amount of water. Methacryloxypropyltrimethoxysilane (MPS) was attached to the exposed surfaces of silica as cross-linkers. Subsequent growth of PS occurred at the sites of MPS via miniemulsion and the morphology of Janus structures could be tuned by adjusting the amount of monomers (Figure 1.13). Wang et al.117 used similar method to fabricate

PS/Silica asymmetric nanocomposites. The surfaces of PS were functionalized by carboxyl groups by using the initiator 4,4′-Azobis(4-cyanovaleric acid) and silica surfaces were modified

with amino groups by coupling of the 3-aminopropyl-triethoxysilane, thus making the nanocomposites dual-functional. They further extended this approach to synthesize ternary Janus structures.118 Oleic acid functionalized iron oxides were mixed with styrene (St) monomers and tetraethoxysilane (TEOS) to form a miniemulsion system under sonication. Emulsion polymerization began upon the addition of 2,2-azobis- (isobutyronitrile) and phase separation occurred due to the decrease of affinity between PS chains and hydrophobic components.

Eventually ternary PS/Fe3O4@SiO2 Janus nanocomposites were produced. Interestingly, iron oxides were embedded in silica, which was different from Teo’s report,119 where iron oxides were found inside PS spheres.

1.4.4. Applications of Janus Particles

Due to the distinct surface properties, Janus particles can be modified to be amphiphilic, thus making them extremely suitable as surfactants to emulsify immiscible liquids.120 For example, Kim and coworkers stabilized non-spherical emulsion drops with amphiphilic particles.121 On the other hand, Janus particles that have magnetic components possess potential application in manipulation of devices and biomedical treatment such as drug delivery and imaging.95 Kim et al. managed to control the movement direction of Janus particles in a microfluidic device by rotating the magnetic field.122 With quantum dots embedded in the

123 polymer nodule of the Fe3O4/polymer Janus particles, Yin et al. demonstrated interesting imaging capability of the Janus composites. Strictly aligned in the external magnetic field, when

the Fe3O4 doped side faced upwards, the particle displayed red-brown in daylight and black under UV. When the QD embedded part faces upwards, the particle appeared white in daylight

99 and blue under UV. Xu et al. fabricated Au/Fe3O4 nanoparticles and conjugated Herceptin and

Platin via stepwise functionalization to achieve targeted drug delivery. The conjugates released the drugs faster at mild acidic environment. The Janus nanoparticles were also found capable of

124 serving as catalyst. She et al. tested the catalytic properties of TiO2 coated Au nanoparticles in the reduction of 4-nitrophenol to 4-aminophenol and found that the Janus nanoparticles catalyzed the reaction as fast as the bare gold nanoparticles in the first cycle but could be reused for up to five cycles without much loss of activity.

1.5. Research objectives and strategy: Multifunctional Janus

Nanocomposites for Biomedical Application

Nanotechnology has provided new path into the fundamental biomedical studies.

Generally the common nanomaterials (silica, gold, iron oxides) used for biomedical applications were utilized of their intrinsic properties, such as magnetism, fluorescence and surface functional groups. However, the complicity of biological systems is in need of intelligent nanosystems that can integrate multiple functionalities. This requires that the nanocarrier incorporates several components simultaneously. Considering the complexity and potential adverse interactions among different components that the conventional single spherical nanoparticles would suffer

from, we decided to develop Janus nanocomposites with two distinct surfaces suitable for further functionalization. Iron oxide nanoparticles were to be embedded for the potential application of magnetic induced hyperthermia and MRI. Based on our previous experience on developing ternary Janus nanostructures and rendering dual functionalities, we decided to employ the phase separation approach combined with miniemulsion. Anticancer drugs and targeting moieties are individually conjugated to the two distinct surfaces for targeted, controlled drug delivery. In this dissertation, the work of producing superparamagnetic PS/Fe3O4@SiO2 Janus nanocomposites is presented. The surfaces of PS are conjugated with DOX and silica surfaces are attached with folic acid for targeting. The multifunctional nanocarrier is evaluated in vitro for its potential application in cancer therapy.

1.6. References

1. D. L. Shi, "Integrated Multifunctional Nanosystems for Medical Diagnosis and Treatment," Adv Funct Mater, 19[21] 3356-73 (2009). 2. L. Brannon-Peppas and J. O. Blanchette, "Nanoparticle and targeted systems for cancer therapy," Adv Drug Deliver Rev, 64 206-12 (2012). 3. L. Brannon-Peppas and J. O. Blanchette, "Nanoparticle and targeted systems for cancer therapy," Adv Drug Deliver Rev, 56[11] 1649-59 (2004). 4. O. C. Farokhzad, J. Cheng, B. A. Teply, I. Sherifi, S. Jon, P. W. Kantoff, J. P. Richie, andR. Langer, "Targeted nanoparticle-aptamer bioconjugates for cancer chemotherapy in vivo," Proc. Natl. Acad. Sci. U. S. A., 103[16] 6315-20 (2006). 5. F. Marcucci and F. i. Lefoulon, "Active targeting with particulate drug carriers in tumor therapy: fundamentals and recent progress," Drug Discov. Today, 9[5] 219-28 (2004). 6. O. C. Farokhzad, S. Jon, A. Khademhosseini, T.-N. T. Tran, D. A. LaVan, andR. Langer, "Nanoparticle- Aptamer Bioconjugates: A New Approach for Targeting Prostate Cancer Cells," Cancer Res., 64[21] 7668-72 (2004). 7. T. A. Elbayoumi, S. Pabba, A. Roby, andV. P. Torchilin, "Antinucleosome antibody-modified liposomes and lipid-core micelles for tumor-targeted delivery of therapeutic and diagnostic agents," J

Liposome Res, 17[1] 1-14 (2007). 8. T. A. Elbayoumi and V. P. Torchilin, "Tumor-specific antibody-mediated targeted delivery of Doxil (R) reduces the manifestation of auricular erythema side effect in mice," Int J Pharmaceut, 357[1-2] 272-79 (2008). 9. I. H. El-Sayed, X. H. Huang, andM. A. El-Sayed, "Selective laser photo-thermal therapy of epithelial carcinoma using anti-EGFR antibody conjugated gold nanoparticles," Cancer Lett, 239[1] 129-35 (2006). 10. X. M. Qian, X. H. Peng, D. O. Ansari, Q. Yin-Goen, G. Z. Chen, D. M. Shin, L. Yang, A. N. Young, M. D. Wang, andS. M. Nie, "In vivo tumor targeting and spectroscopic detection with surface- enhanced Raman nanoparticle tags," Nat Biotechnol, 26[1] 83-90 (2008). 11. S. Acharya, F. Dilnawaz, andS. K. Sahoo, "Targeted epidermal growth factor receptor nanoparticle bioconjugates for breast cancer therapy," Biomaterials, 30[29] 5737-50 (2009). 12. T. P. Thomas, A. K. Patri, A. Myc, M. T. Myaing, J. Y. Ye, T. B. Norris, andJ. R. Baker, Jr., "In vitro targeting of synthesized antibody-conjugated dendrimer nanoparticles," Biomacromolecules, 5[6] 2269-74 (2004). 13. X. H. Gao, Y. Y. Cui, R. M. Levenson, L. W. K. Chung, andS. M. Nie, "In vivo cancer targeting and imaging with semiconductor quantum dots," Nat Biotechnol, 22[8] 969-76 (2004). 14. A. P. Chapman, "PEGylated antibodies and antibody fragments for improved therapy: a review," Adv Drug Deliver Rev, 54[4] 531-45 (2002). 15. L. S. Lee, C. Conover, C. Shi, M. Whitlow, andD. Filpula, "Prolonged circulating lives of single-chain Fv proteins conjugated with polyethylene glycol: a comparison of conjugation chemistries and compounds," Bioconjug Chem, 10[6] 973-81 (1999). 16. T. Suzuki, N. Kanbara, T. Tomono, N. Hayashi, andI. Shinohara, "Physicochemical and Biological Properties of Poly(Ethylene Glycol)-Coupled Immunoglobulin-G," Biochim Biophys Acta, 788[2] 248-55 (1984). 17. K. Kitamura, T. Takahashi, T. Yamaguchi, A. Noguchi, A. Noguchi, K. Takashina, H. Tsurumi, M. Inagake, T. Toyokuni, andS. Hakomori, "Chemical-Engineering of the Monoclonal Antibody-A7 by Polyethylene-Glycol for Targeting Cancer-Chemotherapy," Cancer Res, 51[16] 4310-15 (1991). 18. N. Chanda, V. Kattumuri, R. Shukla, A. Zambre, K. Katti, A. Upendran, R. R. Kulkarni, P. Kan, G. M. Fent, S. W. Casteel, C. J. Smith, E. Boote, J. D. Robertson, C. Cutler, J. R. Lever, K. V. Katti, andR. Kannan, "Bombesin functionalized gold nanoparticles show in vitro and in vivo cancer receptor specificity," P Natl Acad Sci USA, 107[19] 8760-65 (2010). 19. V. Biju, D. Muraleedharan, K. Nakayama, Y. Shinohara, T. Itoh, Y. Baba, andM. Ishikawa, "Quantum dot-insect neuropeptide conjugates for fluorescence imaging, transfection, and nucleus targeting of living cells," Langmuir, 23[20] 10254-61 (2007). 20. G. von Maltzahn, Y. Ren, J. H. Park, D. H. Min, V. R. Kotamraju, J. Jayakumar, V. Fogal, M. J. Sailor, E. Ruoslahti, andS. N. Bhatia, "In vivo tumor cell targeting with "Click" nanoparticles," Bioconjugate Chem, 19[8] 1570-78 (2008). 21. X. A. Chen, X. H. Wang, Y. S. Wang, L. Yang, J. Hu, W. J. Xiao, A. Fu, L. L. Cai, X. Li, X. Ye, Y. L. Liu, W. S. Wu, X. M. Shao, Y. Q. Mao, Y. Q. Wei, andL. J. Chen, "Improved tumor-targeting drug delivery and therapeutic efficacy by cationic liposome modified with truncated bFGF peptide," J Control Release, 145[1] 17-25 (2010).

22. B. Kang, M. A. Mackey, andM. A. El-Sayed, "Nuclear Targeting of Gold Nanoparticles in Cancer Cells Induces DNA Damage, Causing Cytokinesis Arrest and Apoptosis," J Am Chem Soc, 132[5] 1517-+ (2010). 23. S. Mohapatra, S. K. Mallick, T. K. Maiti, S. K. Ghosh, andP. Pramanik, "Synthesis of highly stable folic acid conjugated magnetite nanoparticles for targeting cancer cells," Nanotechnology, 18[38] (2007). 24. Y. Teow and S. Valiyaveettil, "Active targeting of cancer cells using folic acid-conjugated platinum nanoparticles," Nanoscale, 2[12] 2607-13 (2010). 25. C. Sun, R. Sze, andM. Q. Zhang, "Folic acid-PEG conjugated superparamagnetic nanoparticles for targeted cellular uptake and detection by MRI," J Biomed Mater Res A, 78A[3] 550-57 (2006). 26. Z. W. Zhang, J. Jia, Y. Q. Lai, Y. Y. Ma, J. Weng, andL. P. Sun, "Conjugating folic acid to gold nanoparticles through glutathione for targeting and detecting cancer cells," Bioorgan Med Chem, 18[15] 5528-34 (2010). 27. F. Porta, G. E. Lamers, J. Morrhayim, A. Chatzopoulou, M. Schaaf, H. den Dulk, C. Backendorf, J. I. Zink, andA. Kros, "Folic Acid-modified mesoporous silica nanoparticles for cellular and nuclear targeted drug delivery," Adv Healthc Mater, 2[2] 281-6 (2013). 28. C. M. Paulos, J. A. Reddy, C. P. Leamon, M. J. Turk, andP. S. Low, "Ligand binding and kinetics of folate receptor recycling in vivo: Impact on receptor-mediated drug delivery," Mol Pharmacol, 66[6] 1406-14 (2004). 29. M. Wang and M. Thanou, "Targeting nanoparticles to cancer," Pharmacol Res, 62[2] 90-99 (2010). 30. R. Singh and J. W. Lillard, "Nanoparticle-based targeted drug delivery," Exp Mol Pathol, 86[3] 215-23 (2009). 31. J. Panyam and V. Labhasetwar, "Biodegradable nanoparticles for drug and gene delivery to cells and tissue," Adv Drug Deliver Rev, 64 61-71 (2012). 32. M. P. Desai, V. Labhasetwar, E. Walter, R. J. Levy, andG. L. Amidon, "The mechanism of uptake of biodegradable microparticles in Caco-2 cells is size dependent," Pharm Res-Dordr, 14[11] 1568- 73 (1997). 33. Y. Matsumura and H. Maeda, "A New Concept for Macromolecular Therapeutics in Cancer- Chemotherapy - Mechanism of Tumoritropic Accumulation of Proteins and the Antitumor Agent Smancs," Cancer Res, 46[12] 6387-92 (1986). 34. T. Murakami and K. Tsuchida, "Recent advances in inorganic nanoparticle-based drug delivery systems," Mini-Rev Med Chem, 8[2] 175-83 (2008). 35. N. Nishiyama and K. Kataoka, "Current state, achievements, and future prospects of polymeric micelles as nanocarriers for drug and gene delivery," Pharmacol. Ther., 112[3] 630-48 (2006). 36. K. Emoto, Y. Nagasaki, andK. Kataoka, "Coating of surfaces with stabilized reactive micelles from poly(ethylene glycol)-poly(DL-lactic acid) block copolymer," Langmuir, 15[16] 5212-18 (1999). 37. C. Cheng, H. Wei, B. X. Shi, H. Cheng, C. Li, Z. W. Gu, S. X. Cheng, X. Z. Zhang, andR. X. Zhuo, "Biotinylated thermoresponsive micelle self-assembled from double-hydrophilic block copolymer for drug delivery and tumor target," Biomaterials, 29[4] 497-505 (2008). 38. M. L. Forrest, J. A. Yanez, C. M. Remsberg, Y. Ohgami, G. S. Kwon, andN. M. Davies, "Paclitaxel prodrugs with sustained release and high solubility in poly(ethylene glycol)-b-poly(epsilon- caprolactone) micelle nanocarriers: Pharmacokinetic disposition, tolerability, and cytotoxicity,"

Pharm Res-Dordr, 25[1] 194-206 (2008). 39. H. S. Yoo and T. G. Park, "Folate receptor targeted biodegradable polymeric doxorubicin micelles," J Control Release, 96[2] 273-83 (2004). 40. D. Sutton, N. Nasongkla, E. Blanco, andJ. M. Gao, "Functionalized micellar systems for cancer targeted drug delivery," Pharm Res-Dordr, 24[6] 1029-46 (2007). 41. Y. Bae, W. D. Jang, N. Nishiyama, S. Fukushima, andK. Kataoka, "Multifunctional polymeric micelles with folate-mediated cancer cell targeting and pH-triggered drug releasing properties for active intracellular drug delivery," Mol Biosyst, 1[3] 242-50 (2005). 42. X. Q. Yang, J. J. Grailer, S. Pilla, D. A. Steeber, andS. Q. Gong, "Tumor-Targeting, pH-Responsive, and Stable Unimolecular Micelles as Drug Nanocarriers for Targeted Cancer Therapy," Bioconjugate Chem, 21[3] 496-504 (2010). 43. C. Dufes, I. F. Uchegbu, andA. G. Schatzlein, "Dendrimers in gene delivery," Adv Drug Deliver Rev, 57[15] 2177-202 (2005). 44. S. Svenson, "Dendrimers as versatile platform in drug delivery applications," European Journal of Pharmaceutics and Biopharmaceutics, 71[3] 445-62 (2009). 45. J. Haensler and F. C. Szoka, "Polyamidoamine Cascade Polymers Mediate Efficient Transfection of Cells in Culture," Bioconjugate Chem, 4[5] 372-79 (1993). 46. J. F. G. A. Jansen, E. M. M. Debrabandervandenberg, andE. W. Meijer, "Encapsulation of Guest Molecules into a Dendritic Box," Science, 266[5188] 1226-29 (1994). 47. J. F. G. A. Jansen, E. W. Meijer, andE. M. M. Debrabandervandenberg, "The Dendritic Box - Shape- Selective Liberation of Encapsulated Guests," J Am Chem Soc, 117[15] 4417-18 (1995). 48. C. J. Hawker, K. L. Wooley, andJ. M. J. Frechet, "Unimolecular Micelles and Globular Amphiphiles - Dendritic Macromolecules as Novel Recyclable Solubilization Agents," J Chem Soc Perk T 1[12] 1287-97 (1993). 49. E. R. Gillies and J. M. J. Frechet, "Dendrimers and dendritic polymers in drug delivery," Drug Discov Today, 10[1] 35-43 (2005). 50. R. Duncan and N. Malik, "Dendrimers: Biocompatibility and potential for delivery of anticancer agents," Crs Bui Nat 105-06 (1996). 51. A. Quintana, E. Raczka, L. Piehler, I. Lee, A. Myc, I. Majoros, A. K. Patri, T. Thomas, J. Mule, andJ. R. Baker, "Design and function of a dendrimer-based therapeutic nanodevice targeted to tumor cells through the folate receptor," Pharm Res-Dordr, 19[9] 1310-16 (2002). 52. D. J. Norris and M. G. Bawendi, "Measurement and assignment of the size-dependent optical spectrum in CdSe quantum dots," Phys Rev B, 53[24] 16338-46 (1996). 53. X. Wu, H. Liu, J. Liu, K. N. Haley, J. A. Treadway, J. P. Larson, N. Ge, F. Peale, andM. P. Bruchez, "Immunofluorescent labeling of cancer marker Her2 and other cellular targets with semiconductor quantum dots," Nat Biotech, 21[1] 41-46 (2003). 54. R. E. Bailey and S. Nie, "Alloyed Semiconductor Quantum Dots: Tuning the Optical Properties without Changing the Particle Size," J. Am. Chem. Soc., 125[23] 7100-06 (2003). 55. W. C. W. Chan and S. Nie, "Quantum Dot Bioconjugates for Ultrasensitive Nonisotopic Detection," Science, 281[5385] 2016-18 (1998). 56. M. Adeli, F. Hakimpoor, M. Parsamanesh, M. Kalantari, Z. Sobhani, andF. Attyabi, "Quantum dot- pseudopolyrotaxane supramolecules as anticancer drug delivery systems," Polymer, 52[11] 2401-

13 (2011). 57. T. Q. Vu, R. Maddipati, T. A. Blute, B. J. Nehilla, L. Nusblat, andT. A. Desai, "Ligand-conjugated quantum dots for targeted drug delivery to nerve cells," Eng Med Biol Soc Ann 152-53 (2005). 58. W. H. De Jong and P. J. A. Borm, "Drug delivery and nanoparticles: Applications and hazards," Int J Nanomed, 3[2] 133-49 (2008). 59. A. M. Smith, G. Ruan, M. N. Rhyner, andS. Nie, "Engineering Luminescent Quantum Dots for In Vivo Molecular and Cellular Imaging," Ann. Biomed. Eng., 34[1] 3 - 14 (2006). 60. D. D. Stark, R. Weissleder, G. Elizondo, P. F. Hahn, S. Saini, L. E. Todd, J. Wittenberg, andJ. T. Ferrucci, "Superparamagnetic Iron-Oxide - Clinical-Application as a Contrast Agent for Mr Imaging of the Liver," Radiology, 168[2] 297-301 (1988). 61. K. J. Widder, R. M. Morris, G. Poore, D. P. Howard, andA. E. Senyei, "Tumor Remission in Yoshida Sarcoma-Bearing Rats by Selective Targeting of Magnetic Albumin Microspheres Containing Doxorubicin," P Natl Acad Sci-Biol, 78[1] 579-81 (1981). 62. C. Alexiou, W. Arnold, R. J. Klein, F. G. Parak, P. Hulin, C. Bergemann, W. Erhardt, S. Wagenpfeil, andA. S. Lubbe, "Locoregional cancer treatment with magnetic drug targeting," Cancer Res, 60[23] 6641-48 (2000). 63. A. S. Lubbe, C. Bergemann, W. Huhnt, T. Fricke, H. Riess, J. W. Brock, andD. Huhn, "Preclinical experiences with magnetic drug targeting: Tolerance and efficacy," Cancer Res, 56[20] 4694-701 (1996). 64. A. S. Lubbe, C. Bergemann, H. Riess, F. Schriever, P. Reichardt, K. Possinger, M. Matthias, B. Dorken, F. Herrmann, R. Gurtler, P. Hohenberger, N. Haas, R. Sohr, B. Sander, A. J. Lemke, D. Ohlendorf, W. Huhnt, andD. Huhn, "Clinical experiences with magnetic drag targeting: A phase I study with 4'- epidoxorubicin in 14 patients with advanced solid tumors," Cancer Res, 56[20] 4686-93 (1996). 65. T. K. Jain, M. A. Morales, S. K. Sahoo, D. L. Leslie-Pelecky, andV. Labhasetwar, "Iron oxide nanoparticles for sustained delivery of anticancer agents," Mol Pharmaceut, 2[3] 194-205 (2005). 66. S. C. Wuang, K. G. Neoh, E. T. Kang, D. E. Leckband, andD. W. Pack, "Acid-Sensitive Magnetic Nanoparticles as Potential Drug Depots," Aiche J, 57[6] 1638-45 (2011). 67. A. Shkilnyy, E. Munnier, K. Herve, M. Souce, R. Benoit, S. Cohen-Jonathan, P. Limelette, M. L. Saboungi, P. Dubois, andI. Chourpa, "Synthesis and Evaluation of Novel Biocompatible Super- paramagnetic Iron Oxide Nanoparticles as Magnetic Anticancer Drug Carrier and Fluorescence Active Label," J Phys Chem C, 114[13] 5850-58 (2010). 68. I. Hilger and W. A. Kaiser, "Iron oxide-based nanostructures for MRI and magnetic hyperthermia," Nanomedicine-Uk, 7[9] 1443-59 (2012). 69. I. I. Slowing, J. L. Vivero-Escoto, C. W. Wu, andV. S. Y. Lin, "Mesoporous silica nanoparticles as controlled release drug delivery and gene transfection carriers," Adv Drug Deliver Rev, 60[11] 1278-88 (2008). 70. B. G. Trewyn, C. M. Whitman, andV. S. Y. Lin, "Morphological control of room-temperature ionic liquid templated mesoporous silica nanoparticles for controlled release of antibacterial agents," Nano Lett, 4[11] 2139-43 (2004). 71. C. Y. Lai, B. G. Trewyn, D. M. Jeftinija, K. Jeftinija, S. Xu, S. Jeftinija, andV. S. Y. Lin, "A mesoporous silica nanosphere-based carrier system with chemically removable CdS nanoparticle caps for stimuli- responsive controlled release of neurotransmitters and drug molecules," J Am Chem Soc, 125[15]

4451-59 (2003). 72. M. C. Daniel and D. Astruc, "Gold nanoparticles: Assembly, supramolecular chemistry, quantum-size- related properties, and applications toward biology, catalysis, and nanotechnology," Chem Rev, 104[1] 293-346 (2004). 73. E. E. Connor, J. Mwamuka, A. Gole, C. J. Murphy, andM. D. Wyatt, "Gold nanoparticles are taken up by human cells but do not cause acute cytotoxicity," Small, 1[3] 325-27 (2005). 74. G. F. Paciotti, L. Myer, D. Weinreich, D. Goia, N. Pavel, R. E. McLaughlin, andL. Tamarkin, "Colloidal gold: a novel nanoparticle vector for tumor directed drug delivery," Drug Deliv, 11[3] 169-83 (2004). 75. H. W. Gu, P. L. Ho, E. Tong, L. Wang, andB. Xu, "Presenting vancomycin on nanoparticles to enhance antimicrobial activities," Nano Lett, 3[9] 1261-63 (2003). 76. S. Aryal, J. J. Grailer, S. Pilla, D. A. Steeber, andS. Q. Gong, "Doxorubicin conjugated gold nanoparticles as water-soluble and pH-responsive anticancer drug nanocarriers," J Mater Chem, 19[42] 7879- 84 (2009). 77. C. M. Pitsillides, E. K. Joe, X. B. Wei, R. R. Anderson, andC. P. Lin, "Selective cell targeting with light- absorbing microparticles and nanoparticles," Biophysical Journal, 84[6] 4023-32 (2003). 78. C. C. Berry, "Progress in functionalization of magnetic nanoparticles for applications in biomedicine," J Phys D Appl Phys, 42[22] (2009). 79. Q. A. Pankhurst, J. Connolly, S. K. Jones, andJ. Dobson, "Applications of magnetic nanoparticles in biomedicine," J Phys D Appl Phys, 36[13] R167-R81 (2003). 80. R. Weissleder, G. Elizondo, J. Wittenberg, C. A. Rabito, H. H. Bengele, andL. Josephson, "Ultrasmall Superparamagnetic Iron-Oxide - Characterization of a New Class of Contrast Agents for Mr Imaging," Radiology, 175[2] 489-93 (1990). 81. J. S. Weinstein, C. G. Varallyay, E. Dosa, S. Gahramanov, B. Hamilton, W. D. Rooney, L. L. Muldoon, andE. A. Neuwelt, "Superparamagnetic iron oxide nanoparticles: diagnostic magnetic resonance imaging and potential therapeutic applications in neurooncology and central nervous system inflammatory pathologies, a review," J Cerebr Blood F Met, 30[1] 15-35 (2010). 82. R. Lawaczeck, H. Bauer, T. Frenzel, M. Hasegawa, Y. Ito, K. Kito, N. Miwa, H. Tsutsui, H. Volger, andH. J. Weinmann, "Magnetic iron oxide particles coated with carboxydextran for parenteral administration and liver contrasting - Pre-clinical profile of SH U555A," Acta Radiol, 38[4] 584-97 (1997). 83. L. Babes, B. Denizot, G. Tanguy, J. J. Le Jeune, andP. Jallet, "Synthesis of iron oxide nanoparticles used as MRI contrast agents: A parametric study," J Colloid Interf Sci, 212[2] 474-82 (1999). 84. M. E. Gindy and R. K. Prud'homme, "Multifunctional nanoparticles for imaging, delivery and targeting in cancer therapy," Expert Opin Drug Del, 6[8] 865-78 (2009). 85. C. H. Lee, S. H. Cheng, I. P. Huang, J. S. Souris, C. S. Yang, C. Y. Mou, andL. W. Lo, "Intracellular pH- Responsive Mesoporous Silica Nanoparticles for the Controlled Release of Anticancer Chemotherapeutics," Angew Chem Int Edit, 49[44] 8214-19 (2010). 86. F. M. Kievit, O. Veiseh, N. Bhattarai, C. Fang, J. W. Gunn, D. Lee, R. G. Ellenbogen, J. M. Olson, andM. Q. Zhang, "PEI-PEG-Chitosan-Copolymer-Coated Iron Oxide Nanoparticles for Safe Gene Delivery: Synthesis, Complexation, and Transfection," Adv Funct Mater, 19[14] 2244-51 (2009). 87. G. Liu, J. Xie, F. Zhang, Z. Y. Wang, K. Luo, L. Zhu, Q. M. Quan, G. Niu, S. Lee, H. Ai, andX. Y. Chen, "N-

Alkyl-PEI-Functionalized Iron Oxide Nanoclusters for Efficient siRNA Delivery," Small, 7[19] 2742- 49 (2011). 88. I. Ojea-Jimenez, L. Garcia-Fernandez, J. Lorenzo, andV. F. Puntes, "Facile Preparation of Cationic Gold Nanoparticle-Bioconjugates for Cell Penetration and Nuclear Targeting," ACS Nano, 6[9] 7692- 702 (2012). 89. H. S. Cho, Z. Dong, G. M. Pauletti, J. Zhang, H. Xu, H. Gu, L. Wang, R. C. Ewing, C. Huth, F. Wang, andD. Shi, "Fluorescent, superparamagnetic nanospheres for drug storage, targeting, and imaging: a multifunctional nanocarrier system for cancer diagnosis and treatment," ACS Nano, 4[9] 5398- 404 (2010). 90. A. D. Quach, G. Crivat, M. A. Tarr, andZ. Rosenzweig, "Gold Nanoparticle-Quantum Dot-Polystyrene Microspheres as Fluorescence Resonance Energy Transfer Probes for Bioassays," J Am Chem Soc, 133[7] 2028-30 (2011). 91. M. Lattuada and T. A. Hatton, "Synthesis, properties and applications of Janus nanoparticles," Nano Today, 6[3] 286-308 (2011). 92. C. A. Serra and Z. Q. Chang, "Microfluidic-assisted synthesis of polymer particles," Chem Eng Technol, 31[8] 1099-115 (2008). 93. D. Dendukuri, D. C. Pregibon, J. Collins, T. A. Hatton, andP. S. Doyle, "Continuous-flow lithography for high-throughput microparticle synthesis," Nature Materials, 5[5] 365-69 (2006). 94. D. Dendukuri and P. S. Doyle, "The Synthesis and Assembly of Polymeric Microparticles Using Microfluidics," Adv Mater, 21[41] 4071-86 (2009). 95. J. Hu, S. X. Zhou, Y. Y. Sun, X. S. Fang, andL. M. Wu, "Fabrication, properties and applications of Janus particles," Chem Soc Rev, 41[11] 4356-78 (2012). 96. S. H. Hu and X. H. Gao, "Nanocomposites with Spatially Separated Functionalities for Combined Imaging and Magnetolytic Therapy," J Am Chem Soc, 132[21] 7234-7237 (2010). 97. K. H. Roh, D. C. Martin, andJ. Lahann, "Biphasic Janus particles with nanoscale anisotropy," Nat Mater, 4[10] 759-63 (2005). 98. K. H. Roh, M. Yoshida, andJ. Lahann, "Compartmentalized, multiphasic nanocolloids with potential applications in drug delivery and biomedical imaging," Materialwiss Werkst, 38[12] 1008-11 (2007). 99. C. J. Xu, B. D. Wang, andS. H. Sun, "Dumbbell-like Au-Fe3O4 Nanoparticles for Target-Specific Platin Delivery," J Am Chem Soc, 131[12] 4216-4217 (2009). 100. A. Perro, S. Reculusa, F. Pereira, M. H. Delville, C. Mingotaud, E. Duguet, E. Bourgeat-Lami, andS. Ravaine, "Towards large amounts of Janus nanoparticles through a protection-deprotection route," Chem Commun[44] 5542-43 (2005). 101. L. Nagle, D. Ryan, S. Cobbe, andD. Fitzmaurice, "Templated nanoparticle assembly on the surface of a patterned nanosphere," Nano Lett, 3[1] 51-53 (2003). 102. M. A. Correa-Duarte, V. Salgueirino-Maceira, B. Rodriguez-Gonzalez, L. M. Liz-Marzan, A. Kosiorek, W. Kandulski, andM. Giersig, "Asymmetric functional colloids through selective hemisphere modification," Adv Mater, 17[16] 2014-2018 (2005). 103. C. Bae, J. Moon, H. Shin, J. Kim, andM. M. Sung, "Fabrication of monodisperse asymmetric colloidal clusters by using contact area lithography (CAL)," J Am Chem Soc, 129[46] 14232-39 (2007). 104. M. D. McConnell, M. J. Kraeutler, S. Yang, andR. J. Composto, "Patchy and Multiregion Janus

Particles with Tunable Optical Properties," Nano Lett, 10[2] 603-09 (2010). 105. L. Zhang, F. Zhang, W. F. Dong, J. F. Song, Q. S. Huo, andH. B. Sun, "Magnetic-mesoporous Janus nanoparticles," Chem Commun, 47[4] 1225-27 (2011). 106. J. Zhang, J. Jin, andH. Y. Zhao, "Surface-Initiated Free Radical Polymerization at the Liquid-Liquid Interface: A One-Step Approach for the Synthesis of Amphiphilic Janus Silica Particles," Langmuir, 25[11] 6431-37 (2009). 107. L. Hong, S. Jiang, andS. Granick, "Simple method to produce Janus colloidal particles in large quantity," Langmuir, 22[23] 9495-99 (2006). 108. K. M. Chen, Y. Zhu, Y. F. Zhang, L. Li, Y. Lu, andX. H. Guo, "Synthesis of Magnetic Spherical Polyelectrolyte Brushes," Macromolecules, 44[3] 632-39 (2011). 109. A. Misra and M. W. Urban, "Acorn-shape polymeric nano-colloids: synthesis and self-assembled films," Macromol Rapid Commun, 31[2] 119-27 (2010). 110. K. Nakahama, H. Kawaguchi, andK. Fujimoto, "A novel preparation of nonsymmetrical microspheres using the Langmuir-Blodgett technique," Langmuir, 16[21] 7882-86 (2000). 111. S. Y. Zhang, Z. Li, S. Samarajeewa, G. R. Sun, C. Yang, andK. L. Wooley, "Orthogonally Dual-Clickable Janus Nanoparticles via a Cyclic Templating Strategy," J Am Chem Soc, 133[29] 11046-49 (2011). 112. M. Bradley and J. Rowe, "Cluster formation of Janus polymer microgels," Soft Matter, 5[16] 3114-19 (2009). 113. S. Lone, S. H. Kim, S. W. Nam, S. Park, J. Joo, andI. W. Cheong, "Microfluidic synthesis of Janus particles by UV-directed phase separation," Chem Commun, 47[9] 2634-36 (2011). 114. M. Feyen, C. Weidenthaler, F. Schuth, andA. H. Lu, "Regioselectively Controlled Synthesis of Colloidal Mushroom Nanostructures and Their Hollow Derivatives," J Am Chem Soc, 132[19] 6791-99 (2010). 115. J. A. Gong, X. H. Zu, Y. H. Li, W. Mu, andY. L. Deng, "Janus particles with tunable coverage of zinc oxide nanowires," J Mater Chem, 21[7] 2067-69 (2011). 116. X. P. Ge, M. Z. Wang, Q. Yuan, H. Wang, andX. W. Ge, "The morphological control of anisotropic polystyrene/silica hybrid particles prepared by radiation miniemulsion polymerization," Chem Commun[19] 2765-67 (2009). 117. Y. Wang, H. Xu, W. Qiang, H. Gu, andD. Shi, "Asymmetric Composite Nanoparticles with Anisotropic Surface Functionalities," Journal of Nanomaterials, 2009 (2009). 118. Y. L. Wang, H. Xu, Y. S. Ma, F. F. Guo, F. Wang, andD. L. Shi, "Facile One-Pot Synthesis and Morphological Control of Asymmetric Superparamagnetic Composite Nanoparticles," Langmuir, 27[11] 7207-12 (2011). 119. B. M. Teo, S. K. Suh, T. A. Hatton, M. Ashokkumar, andF. Grieser, "Sonochemical Synthesis of Magnetic Janus Nanoparticles," Langmuir, 27[1] 30-33 (2011). 120. Q. Xu, X. Kang, R. A. Bogomolni, andS. Chen, "Controlled assembly of Janus nanoparticles," Langmuir, 26[18] 14923-8 (2010). 121. J. W. Kim, D. Lee, H. C. Shum, andD. A. Weitz, "Colloid surfactants for emulsion stabilization," Adv Mater, 20[17] 3239-43 (2008). 122. S. H. Kim, J. Y. Sim, J. M. Lim, andS. M. Yang, "Magnetoresponsive Microparticles with Nanoscopic Surface Structures for Remote-Controlled Locomotion," Angew Chem Int Edit, 49[22] 3786-90 (2010).

123. S. N. Yin, C. F. Wang, Z. Y. Yu, J. Wang, S. S. Liu, andS. Chen, "Versatile Bifunctional Magnetic- Fluorescent Responsive Janus Supraballs Towards the Flexible Bead Display," Adv Mater, 23[26] 2915-19 (2011). 124. Z. W. Seh, S. H. Liu, S. Y. Zhang, K. W. Shah, andM. Y. Han, "Synthesis and multiple reuse of eccentric Au@TiO2 nanostructures as catalysts," Chem Commun, 47[23] 6689-91 (2011).

Chapter 2. Synthesis and Application of

Multifunctional Nanocarrier System for Cancer

Therapy

2.1. Background

To address the critical issues in cancer diagnosis and treatment, our group first initiated the study on carbon nanotube-based multifunctional nanosystems for biomedical application in 2006, and expanded our interests to nanopheres later on. In 2010, we managed to develop a multifunctional nanocarrier system based on the iron oxide embedded polystyrene nanoparticles with surfaces functionalized with carboxyl groups. These magnetic nanospheres (MNSs) were obtained from our collaborator Pro. Gu from Shanghai Jiaotong University as a gift. The MNSs had an average diameter of around 150 nm and were composed of Fe3O4 nanoparticles (20 nm in diameter) embedded in polystyrene matrix. Quantum dots (QDs) were covalently conjugated to the surfaces of MNSs for fluorescent imaging. In addition, the surfaces of MNSs were manipulated for functionalizaiton of targeting moieties and anti-cancer drugs with a biodegradable polymer layer to incorporate cell targeting and drug storage capabilities.

polyethylene a oxide QDs

polystyrene ⊝ ⊝  ⊝ ⊝  + Fe3O4 amine-functionalized Quantum Dot (QD)

50 nm 50 nm

carboxylate-functionalized MNS QDs-conjugated MNSs

b PLGA + PTX + PLGA +

QDs QDs PLGA + Paclitaxel (PTX) PTX 5 nm

QDs-conjugated MNSs 5 n m PTX-PLGA-QD-MNSs

Figure 2.1. Schematic illustration of (a) QD conjugation onto MSNs and (b) PTX loading and

PLGA encapsulation.1

2.2. Experimental

2.2.1. QD Conjugation to MNSs

Carbolylate-functionalized MNSs were washed with deionized water (DI water) for three times and then activated with 50 mM of 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC) for 15 min before N-hydroxysuccinimide (NHS) was added to make a final concentration of 5mM. Then amino group functionalized QDs [Qdot 655ITK amino (PEG),

Invitrogen Corp.] were injected and the reaction was left overnight before QD-MNSs were collected and redispersed in DI water after washing with DI water using an external magnet.

2.2.2. PTX loading of QD-MNSs

2 mg of QD-MNSs, 100 ug of paclitaxel (PTX) and 100 ug of carboxyl functionalized poly (lactic-co-glycolic acid) (PLGA) were dispersed in 100 ul of acetonitrile to make an oil phase. The oil phase was injected slowly to 1ml of DI water under sonication. The emulsion was further sonicated for another 5 h to evaporate organic solvent. The resulting PLGA-PTX-QD-

MNSs were washed 5 times with PBS and stored at 4 °C. The procedure is illustrated in Figure

2.1 with transmission electron microscopy (TEM) images describing the nanoparticles at each stage.

2.2.3. Conjugation of anti-PSMA on the surface of PLGA-PTX-QD-MNSs

The PLGA-PTX-QD-MNSs were functionalized with amino groups using ethylene-diamine

in the presence of EDC and NHS coupling agents. Afterwards, the amino group functionalized

PLGA-PTX-QD-MNSs were incubated with an anti-PSMA solution in the presence of

EDC/NHS for 4 h at room temperature. The procedure is illustrated in Figure 2.2.

Figure 2.2. Schematic diagrams illustrating (a) PTX loading and carboxyl-ended PLGA functionalization on the surface of QD-MNSs; (b) amine-modification of PLGA-PTX- QD-

MNSs using ethylenediamine as the crosslinker; (c) conjugation of anti-PSMA to the PTX-

PLGA-QD-MNSs, and (d) the final structure of multifunctional nano-carrier system.1

2.2.4. Evaluation of the multifunctional nano-carrier system

The drug loaded, antibody functionalized MSNs were subjected to cell cytotoxicity assesment and cell targeting analysis on human PC-3MM2 prostate cancer cells.

2.3. Results and Discussion

As is shown in Figure 2.3, incubation of tumor cells with broad dosing range of unmodified MNS, QD-labeled, fluorescent MNSs, and PLGA-coated, fluorescent MNSs does not significantly affect cell viability as compared to vehicle control, which underlines the safety of drug-free nanocarriers up to 25 ng/mL. In contrast, PTX-PLGA-QD-MNSs have dose- dependently decreased cell viability after 96 h of incubation. The estimated concentration associated with 50% inhibition of the mitochondrial enzyme activity (IC50) was 125 ng/mLfor

PLGA-PTX- QD-MNS. Considering that the IC50 value of free paclitaxel for the same cancer cell line is 5 ng/mL,2 approximately 4 % (w/w) of the PTX is released from PTX-PLGA-QD-

MNS conjugates.

120

100

80 OriginalMNS QD-MNS 60 PLGA-QD-MNS PTX-PLGA-QD-MNS

40 Cell Viability (%) Viability Cell 20

0 0.001 0.01 0.1 1 10 Amount of MNSs (ug)

Figure 2.3. Dose-dependent effects of MNSs, QD-MNS, PLGA-QD-MNS, and PTX-

PLGA-QD-MNSs on viability of human PC3mm2 prostate cancer cells.1

Once the PTX-PLGA-QD-MNSs is incubated with the cancer cells, the paclitaxel loaded on the MNSs would start to be released from the MNSs as a result of the PLGA degradation into the cell medium. The release behavior of PTX from the MNS would be affected by various parameters, such as, temperature, the range of pH, the ingredients of the cell culture medium, and the degradation rate of PLGA. To investigate the effect of PLGA on drug storage and release behavior, the amount of PLGA was doubled or reduced by half from the original quantity. As a result, the cell viabilities decreased by increasing the amount of PLGA used for paclitaxel loading, indicating that larger volume of PLGA on the surface of MNSs may have enhanced paclitaxel loading efficiency on the MNSs. However, it was also noted that increased amount of

PLGA leads to more compact PLGA coating layer, which would take more time to degrade and release the PTX. The decrease of cell viability is a combined effect of higher loading capacity

and lower release speed of PTX.

a LNCaP cells PTX-PLGA-QD-MNSs

b LNCaP cells anti-PSMA-PTX-PLGA-QD-MNSs

c PC3mm2 cells PTX-PLGA-QD-MNSs

d PC3mm2 cells anti-PSMA-PTX-PLGA-QD-MNSs

Figure 2.4. in vitro targeting studies of anti-PSMA-conjugated PTX-PLGA-QD- MNSS binding activity in cultured PSMA-positive LNCaP prostate cancer cells, and PSMA-negative PC3mm2.1

Cell targeting studies have been carried out on the PSMA-positive LNCaP prostate cancer cells and PSMA-negative PC3mm2 prostate cancer cells. Anti-PSMA-PTX-PLGA-QD-

MNSs and non-targeted PTX-PLGA-QD-MNS, as a control, were incubated with the fixed

LNCaP and PC3mm2 prostate cancer cells, respectively. It is apparent that LNCaP cells are targeted successfully by the Anti-PSMA-PTX-PLGA-QD-MNSs (Figure 2.4b) while no fluorescent signals are detected in LNCaP cells incubated to non-targeted PTX-PLGA-QD-MNS

(Figure 2.4a). In contrast, PSMA-negative PC3mm2 human prostate cells express no fluorescent signals after incubation with and without anti-PSMA conjugated PLGA-PTX-QD-MNSs. (Figure

2.4c, d).

2.4. Conclusion

A multifunctional nanocarrier system is successfully established based on the iron oxides embedded polystyrene nanoparticles. QDs are decorated on the nanoparticle surfaces, followed by PTX loading and PLGA coating. Finally antibodies are conjugated on the surface of coated nanoparticles. The nanocarrier system is exposed to biomedical assessment and show satisfying results in cell killing and cell targeting analysis.

2.5. References

1. H. S. Cho, Z. Dong, G. M. Pauletti, J. Zhang, H. Xu, H. Gu, L. Wang, R. C. Ewing, C. Huth, F. Wang, andD. Shi, "Fluorescent, superparamagnetic nanospheres for drug storage, targeting, and imaging: a multifunctional nanocarrier system for cancer diagnosis and treatment," ACS Nano, 4[9] 5398- 404 (2010). 2. Y. Guo, D. L. Shi, H. S. Cho, Z. Y. Dong, A. Kulkarni, G. M. Pauletti, W. Wang, J. Lian, W. Liu, L. Ren, Q. Q. Zhang, G. K. Liu, C. Huth, L. M. Wang, andR. C. Ewing, "In vivo imaging and drug storage by quantum-dot-conjugated carbon nanotubes," Adv Funct Mater, 18[17] 2489-97 (2008).

Chapter 3. Design and Development of

Superparamagnetic Janus Nanoparticles as a Multi- functional Tool for Cancer Therapy

3.1. Introduction

In Chapter 2 we have shown the application of a multifunctional nanocarrier system for cancer therapy based on iron oxides embedded polystyrene nanoparticles. However, the singularity of surface functionality made incorporation of multi functionalities very complicated and it still remains unclear that whether different components will interfere with each other.

Therefore, we were determined to design and develop a Janus nanocomposite suitable for cancer diagnosis and treatment.

Silica was chosen as one of the components because of its low toxicity and the abundance of surface silanol groups for almost unlimited options for further functionalization. Polystyrene was another candidate due to the good bio-compatibility and potential different surface groups.

Finally Fe3O4 nanoparticles were incorporated because superparamagnetism would render the nanocomposites opportunities for application of MRI, magnetic targeting and hyperthermia, thus making the nanocomposites truly multifunctional.

3.2. Design and Development of Polystyrene/Fe3O4@Silica nanocomposites

The methods and mechanism of fabricating polymer/silica Janus nanocomposites have been studied extensively and are now well understood but it remains challenging to incorporate magnetic properties in the nanolcomposites.1-6 We tried to synthesize polystyrene/silica nanocomposites with Fe3O4 nanoparticles embedded in silica portion and obtained satisfying results.7 This could serve as a base for our further exploration for bi-functional nanocomposites.

3.2.1. Experimental

3.2.1.1. Materials and Chemicals

Ferric chloride hexahydrate (FeCl3 • 6H2O), ferrous chloride tetrahydrate (FeCl2 • 4H2O), oleic acid (OA), styrene (St), tetraethoxysilane (TEOS), sodium dodecyl sulfate (SDS), hexadecane

(HD), γ-Methacryloxypropyltrimethoxysilane (MPS), divinylbenzene (DVB) and 2’2-azobis-

(isobutyronitrile) (AIBN) were purchased from Sigma-Aldrich (USA). Ammonium hydroxide

(NH4OH, 29.79 wt%) was purchased from Fisher Scientific (USA). Styrene monomer was washed with 5 wt% sodium hydroxide solution and stored at -20 °C before use. All other chemical were used as received. Deionized (DI) water was used through experiment. Ultra high purity nitrogen gas was obtained from Wright Brother Gas.

3.2.1.2. Methods

Synthesis of Oleic Acid Functionalized Iron Oxide Nanoparticles (OAIOs)

A co-precipitation method was used to synthesize oleic acid surface functionalized iron oxide nanoparticles (~10 nm) (OAIOs). First, FeCl2 • 4H2O (7.72 g) and FeCl3 • 6H2O (24 g) were dissolve in 100 ml of DI water, which was heated to 80 °C with nitrogen bubbling for 0.5 h.

After 50 g of NH4OH was injected, 3.76 g of oleic acid was added and the dispersion was kept at

80 °C for 3h. When the dispersion was cooled to room temperature, products were washed with ethanol and DI water three times respectively and OAIOs were obtained. The OAIOs were dispersed in octane and dried under rotary evaporation for further use.

Fabrication of Polystyrene/Fe3O4@Silica nanocomposites

The continuous water phase was formed by dissolving 0.092 g of SDS in 39 g of DI water.

The dispersed oil phase was prepared by mixing 0.4 g of HD 2 g of TEOS, 0.1 g of MPS and 60 mg of OAIOs in 8 g of St monomers. The oil phase was injected dropwise into the water phase under powerful sonication, followed by 250 rpm mechanical stirring for 0.5 h. Then miniemulsion droplets were formed by sonicating the emulsions for 10 min at 500 W at a duty cycle of 50 % using a Scientz-IID sonifier (Ningbo, China) in an ice bath. The miniemulsion system was transferred to a three-headed flask. After the addition of 0.08 g of AIBN, the system was deoxygenized by nitrogen bubbling for 0.5 h with mechanical stirring at 250 rpm. Then the flask was moved into a 70 °C water bath with a condenser under nitrogen protection to start polymerization. After 60 min, 35 µl of NH4OH was added into the reaction. After another 5 h, the reaction system was taken out of the water bath and stirred at room temperature overnight.

Finally, nanocomposites were purified by washing with DI water using an external magnetic field and dried under vacuum at room temperature. Detailed illustration of the synthesis route is shown in Figure 3.1.

Figure 3.1. Schematic drawing illustrating the formation of Polystyrene/Fe3O4@Silica Janus nanocomposites via miniemulsion and sol-gel reaction.7

3.2.2. Results and Discussion

From the TEM image of the polystyrene/Fe3O4@silica nanocomposites (Figure 3.2a), it can be seen that the nanocomposites is composed of two spherical parts. The lighter portion indicates a diameter of 200~250 nm and EDX elemental analysis (Figure 3.2b) reveals that the composition is mainly carbon. The darker portion with black dots with the diameter of around

200 nm is composed of silicon and iron elements (carbon signal came from the coating on copper grid). Thus it is reasonable to conclude that the nanocomposites consist of two spherical parts, one with the composition of polystyrene polymer module and the other iron oxide embedded silica. The observance that Fe3O4 nanoparticles are incorporated in silica part is possibly related to the phase separation after polymerization started. The iron oxide nanoparticles had been coated with oleic acid, which was turned to oleate by ammonium hydroxide added in the sol-gel reaction. As a result, OAIOs turned from hydrophobic to hydrophilic, which made them more compatible with TEOS than polystyrene chains.

Figure 3.2. (a) TEM images of polystyrene/Fe3O4@silica Janus nanocomposites and (b) EDX elemental analysis of one Janus nanocomposite as displayed in the inset of part a.7

Figure 3.3. TEM images of polystyrene/silica/Fe3O4 nanocomposites when the phase separation was modified by (a) adding NH4OH 10 min after polymerization started at 70 °C and (b) DVB was included in the oil phase.7

To further investigate the phase separation in the reaction process, NH4OH was added earlier, 10 min after polymerization started. Compared to 60 min after polymerization started, the phase separation between PS chains and the other hydrophobic portion (St monomers, OAIOs and TEOS) was not completed at this stage. The addition of NH4OH started sol-gel process of

TEOS and silica began to form. OAIOs were trapped in the St monomers and as a result, the final products were polystyrene/silica/Fe3O4 nanocomposites (Figure 3.3a) instead of asymmetric nanocomposites with two modules. In another subsequent experiment,

divinylbenzene (DVB) was added to inhibit phase separation. Polystyrene chains would form internal cross-linking in the presence of DVB, thus reducing the mobility of polystyrene chains and phase separation was inhibited.8 The final products (Figure 3.3b) were also symmetric polystyrene/silica/Fe3O4 nanocomposites as we expected. These two experiments confirmed our theory that phase separation played an important role in forming the Janus structure of the nanocomposites.

3.3. Design and Development of Dual Functionalized

Polystyrene/Fe3O4@Silica nanocomposites

Although we successfully synthesized the polystyrene/Fe3O4@silica Janus nanocomposites, the polystyrene surfaces were not functionalized, which makes further application rather difficult because only silica surfaces could be modified. Modification of polystyrene surface with functional groups would involve multiple steps, which increases the complexity and makes the process less controllable. In this way, the Janus structure has no advantage over the symmetric structure. Based on the work of Wang et al.,9 we decided to introduce carboxyl groups onto the polystyrene by using carboxyl group ending initiators. It is hoped that carboxyl groups would be decorated onto the surface of polystyrene in the polymerization process thus synthesis as well as functionalization could be achieved in a one-pot reaction.

3.3.1. Experimental

3.3.1.1. Materials and Chemicals

Ferric chloride hexahydrate (FeCl3 • 6H2O), ferrous chloride tetrahydrate (FeCl2 • 4H2O), oleic acid (OA), styrene (St), tetraethoxysilane (TEOS), sodium dodecyl sulfate (SDS), hexadecane (HD) and 4,4’-azobis(4-cyanovaleric acid) (ACVA) were purchased from Sigma-

Aldrich (USA). Ammonium hydroxide (NH4OH, 29.79 wt%), hydrofluoric acid (HF) and toluene was purchased from Fisher Scientific (USA). Styrene monomer was washed with 5 wt% sodium hydroxide solution and stored at -20 °C before use. All other chemical were used as received. Deionized (DI) water was used through experiment. Ultra high purity nitrogen gas was obtained from Wright Brother Gas.

3.3.1.2. Methods

Synthesis of Oleic Acid Functionalized Iron Oxide Nanoparticles

Oleic acid surface functionalized iron oxide nanoparticles (~10 nm) (OAIOs) were

10 fabricated using a method previously reported. In a typical reaction, FeCl2 • 4H2O (7.72 g) and

FeCl3 • 6H2O (24 g) were dissolve in 100 ml of DI water, kept at 80 °C with nitrogen bubbling for 0.5 h and followed by addition of NH4OH (50 g). Then oleic acid (3.76 g) was added and the dispersion was kept at 80 °C for another 3 h. After cooling down to room temperature, the precipitates OAIOs were washed with ethanol and DI water, dispersed in octane and dried by

rotary evaporation.

Synthesis of Polystyrene/Fe3O4@SiO2 Superparamagnetic Janus Nanocomposites (SJNCs)

The continuous water phase was formed by dissolving 0.092 g of SDS in 39 g of DI water.

The dispersed oil phase was prepared by dissolving 25 mg of OAIOs, 0.4 g of HD and 2 g of

TEOS in 8 g of St monomers. Then the oil phase was injected dropwise into the water phase under sonication, followed by mechanical stirring at 250 rpm for 0.5 h. The miniemulsion droplets were formed by sonicating the emulsions for 10 min at 500 W at a duty cycle of 50 % using a Scientz-IID sonifier (Ningbo, China) in an ice bath. Then the miniemulsion system was transferred to a three-headed flask. After 0.12 g of ACVA was neutralized by 1 ml of NaOH solution (0.856 M) and added into the flask, the system was deoxygenized by nitrogen bubbling for 0.5 h with 250 rpm mechanical stirring. Then the flask was moved into an 80 °C water bath with a condenser under nitrogen protection to start St monomers polymerization. After 60 min,

50 µl of NH4OH was added into the reaction. After another 5 h, the reaction system was taken out of the water bath and stirred at room temperature overnight. Finally, SJNCs were purified by washing with DI water using an external magnetic field and dried under vacuum at room temperature. Schematic drawing of the reaction process is illustrated in Figure 3.4.

3.3.1.3. Characterization

Transmission Electron Microscopy (TEM) and Energy Dispersive X-ray Analysis (EDX)

The obtained SJNCs were dispersed in DI water and dried onto carbon coated copper grid for TEM examination. TEM images were obtained on a Philips Tecnai 20 transmission electron

microscope with an accelerating voltage of 200 kV and a JEOL JEM2010 electron microscope equipped with EDX.

STEM and Elemental Mapping

STEM images and elemental mapping were obtained on a Zeiss Ultra55 equipped with

Oxford Aztec X-Max 50 EDS system operated at an accelerating voltage of 1 kV.

Dynamic Light Scattering (DLS)

Size distribution of SJNCs dispersed in DI water was measured using a Malvern

Zetasizer (Nano ZS).

Fourier Transformed Infrared Spectroscopy

The SJNCs were pelleted with potassium bromide powder and infrared spectra were obtained using a transmission method on a Nicolete 6700 FTIR from Thermo Scientific.

Thermogravimetric Analysis (TGA)

A thermogravimetric analyzer (NETZSCH TG209) was used to characterize the composition of JSNPs. The measurement was performed under nitrogen atmosphere from ambient to 600 °C with a heating rate of 10 °C/min.

Selective Etching

To conduct etching with hydrofluoric acid, the Janus nanocomposites were dispersed in 5%

HF solution in DI water and rotated overnight. Then the products were collected with a centrifuge at 12000 rpm after dialysis in DI water in a dialysis bag with the cutoff of 3000 Da.

When etched with toluene, the Janus nanocomposites were dispersed in toluene and rotated overnight. The products were magnetically washed 5 times with DI water.

Magnetic Property Measurment

The magnetic properties were measured at room temperature using a Vibrating Sample

Magnetometer (VSM 7407, Lake Shore, USA).

Magnetic Resonance Imaging (MRI)

MRI images were obtained on a MRI imaging machine from Agilent, USA (7.0 Telsa/160 mm) with a field of view of 45 mm x 45 mm.

Figure 3.4. Schematic illustration of the formation of polystyrene/Fe3O4@silica superparamagnetic Janus nanocomposites (SJNCs).11

3.3.2. Results and Discussion

As shown in Figure 3.4, the one-pot synthesis process is described as follows: (1) The miniemulsion system is created by miniemulsifying the pre-emulsified mixture of oil phase and water phase using a sonifier. The oil phase is formed by mixing the hydrophobic components including St monomers, HD, TEOS and OAIOs, while the water phase is formed by dissolving

surfactant SDS in DI water. (2) Once the miniemulsion droplets system is established, ACVA is added to initiate the polymerization of St monomers inside the droplets. This is different from conventional emulsion, which occurs in the monomer-swelled micelles.12 With the progressing of polymerization, St monomers in the miniemulsion droplets gradually transfer into polystyrene chain, which will decrease its affinity to the other hydrophobic portion.13 At this intermediate stage, the hydrophobic mixture of remaining St monomers, OAIOs and TEOS is located eccentrically on the PS particle surface. (3) Upon the phase separation of other hydrophobic components from PS particle, NH4OH is added to initiate the sol-gel reaction of TEOS to

14 generate silica. With addition of NH4OH, oleic acid on Fe3O4 surface is transformed to oleate, thus significantly lowering the hydrophobicity of OAIOs. As a result, OAIOs tend to be more compatible with TEOS and produce Fe3O4@Silica hybrid. As the initiator ACVA introduces carboxyl groups on the PS particle surface, which may react with the silanol groups on silica surface to form -Si-O-C- bond, the Fe3O4@Silica hybrid tend to grow along the PS particle surface and form a shell.11 The observation of half shell instead of complete coverage on PS particle is still under investigation.

Figure 3.5. TEM images of (A) oleic acid modified Fe3O4 nanoparticles (OAIOs) (scale bar denotes 50 nm) and (B) the dual functionalized polystyrene/Fe3O4@silica Janus nanoparticles, the scale bar is 200 nm (inset image is the magnified image of one nanocomposite particle, scale bar is 100 nm).11

Figure 3.6. (A) STEM image of the polystyrene/Fe3O4@silica Janus nanoparticles, scale bar denotes 200 nm and (B) elemental mapping of one nanocomposite particle as pointed out by the red arrow in (A).11

To understand the microstructure and chemical composition, TEM, STEM, EDX and Zeta

Sizer were employed for analysis. As can be seen in Figure 3.5, the Superparamagnetic Janus

Nanocomposites (SJNCs) are consisted of a polystyrene core covered by a half shell of silica embedded with Fe3O4 nanoparticles. From TEM image, it can be derived that the SJNCs have diameters of around 300 nm, with the PS core of about 200 nm, and the Fe3O4@Silica hybrid shell of around 100 nm thick. This is consistent with the DLS data (Figure 3.7), displaying an average size of 313 nm with a polydispersity index (PDI) of 0.090. This indicates very narrow distribution of Janus nanocomposite diameters, i.e. excellent monodispersity. As shown in the

EDS spectra (Figure 3.6B), the core shows mainly signal from C and O with minor signals of Si and Fe coming from the hybrid shell at the back of the core. On the other hand, there are stronger signals from Si and Fe from the shell, compared to the core portion. These results indicate that the core of SJNCs is composed of pure polymer, which is polystyrene in this case, while the shell is composed of iron oxide and silica. The observation that iron oxide nanoparticles are present in silica shell other than polystyrene core is consistent with the results in our previous paper.7

Similarly, it is proposed that it was the phase separation between inorganic and organic components during polymerization followed by sol-gel reaction of TEOS that leaded to the formation of SJNCs. As mentioned before, when polystyrene chains were formed during the miniemulsion polymerization, they tended to drive away the St monomers eccentrically, which also contained TEOS and OAIOs. The addition of NH4OH initiated sol-gel reaction of TEOS as well as the transformation of oleic acid to oleate. As a result, OAIOs had increased affinity to

TEOS and were incorporated in the silica rather than PS after condensation. It is possible under

favorable thermodynamic condition (80 °C ), the silanol groups (-Si-OH) on the silica surface reacted with the carboxyl groups (-COOH) introduced by ACVA on the PS surface and formed -

Si-O-C- bonds. Therefore, once silica was formed from condensation, it grew along the curvature of PS particle surface to make a half shell to generate bi-functionalized SJNCs exhibiting anisotropy in both composition and surface properties. The mechanism that PS particle is only half instead of completely covered by Fe3O4@SiO2 hybrid shell remains unclear and we will investigate into it further.

Figure 3.7. DLS data of Janus nanocomposites suspended in DI water.11

Figure 3.8. (A) FTIR spectra of Polystyrene/Fe3O4@Silica Janus nanoparticles and (B) magnified area (1450 cm-1 – 1800 cm-1) of spectra (A).11

The SJNCs was characterized with FTIR technique to study composition and chemical structure. As can be seen in Figure 3.8A, peaks from silica vibrations in the range of 800-1250 cm-1 are very strong, while the peaks at 2850 cm-1 and 2920 cm-1 represent -CH stretching. The peak at 1712 cm-1 (Figure 3.8B) can be assigned to carboxyl groups on the PS surface, with its weak intensity due to low quantities. Besides, the peaks at 1450 cm-1 and 1500 cm-1 are attributed to the C=C frame stretching vibration in aromatic ring, while the peaks at 700 cm-1 and

760 cm-1 are characteristic of single-replaced phenyl ring.15, 16 The peak from -Si-O-C- bond is at around 1100 cm-1 but cannot be observed, possibly due to the overlapping with the strong signal from silica vibrations in the same range.11

Figure. 3.6. TEM images of asymmetric PS/Fe3O4@SiO2 nanocomposites when NH4OH was added (A) 10 min and (B) 90 min after St monomer started polymerization at 80 °C. Both scale bars denote 200 nm.

In the synthesis process, there are two key factors for generating the Janus structure: 1) formation of miniemulsion, 2) the phase separation between PS partlicles and other hydrophobic portions. When the miniemulsion system was formed, all hydrophobic components including St monomers, HD, OAIOs and TEOS were incorporated in the miniemulsion droplets stabilized by

the surfactant SDS. The amount of SDS was below the critical micelle concentration (CMC) so that SDS did not form micelles alone. The polymerization of St monomers took place in each miniemulsion droplets as in the in situ reactor.12 This is the major difference between miniemulsion and conventional emulsion system. In conventional emulsion, polymerization happens in micelles, and monmers transfer from monomer droplets or from dissolved portion in the aqueous solvents as for the monomers with relatively high solubility in water. But in miniemulsion in this reaction, HD served as co-stabilizer and impeded mass transfer between droplets or from the solvents. Reaction would stop simultaneously when the reagents in the droplets were consumed up. The phase separation occurred between OAIOs and polymer.

Meanwhile, silica condensed from TEOS separated from PS particles, which is critical to the morphology of SJNCs. With the polystyrene chains increasing in numbers, the OAIOs and

TEOS experienced phase separation and were kept together with the remaining St monomers.

With the addition of NH4OH, TEOS began sol-gel reaction while oleic acid was converted into oleate ions, which transformed OAIOs into hydrophilic and thus were more compatible with hydrolyzed TEOS.7 Hydrophilic iron oxide nanoparticles might serve as nucleation sites for silica and facilitate the formation of Fe3O4@SiO2 half shell on PS particles. According to the theory mentioned above, the morphology can be manipulated by changing the time of NH4OH addition. It is shown in Figure 3.9A that if NH4OH was added much earlier (10 min after polymerization), the final products displayed as composite nanoparticles with all components

(polystyrene, silica, and Fe3O4) rather than asymmetric structure because there was not enough time for TEOS/OAIOs mixture to be driven away from polystyrene particles. As a result, all

components were forced to concentrate and form one composite particle. In contrast, if NH4OH was added later, i.e. 100 min after polymerization started, the final products still displayed Janus structure (Figure 3.9B). However, the oval-shaped Fe3O4/silica hybrid formed a heterodimer with

PS particle instead of covering the PS particle as a half shell. It is possible that OAIOs/TEOS mixture already experienced excessive phase separation from PS particle when NH4OH was added, so that the silica generated from condensation could not grow along the PS particle curvature. Instead, the oval-shaped iron oxide/silica hybrid only covered a small area of PS particle surface to produce a heterodimer.

Figure 3.10. (A) TEM image of SJNCs after etching with HF, (B) EDX elemental analysis spectra of the particles in (A), (C) TEM image of the SJNCs after toluene etching and (D) EDX

elemental spectra analysis of the shells in (C).

To further investigate the chemical composition of the SJNCs, we also conducted selective etching. HF could dissolve silica as well as iron oxide nanoparticles. As shown in

Figure 3.10A, only the cores were left under TEM observation. The EDX spectra in Figure

3.10B proved that the component of the cores were polystyrene because only carbon was observed. On the other hand, toluene can easily dissolve polystyrene, which lead to the image in

Figure 3.10C. Only the hybrid shells with black dots inside were left. The EDX spectra proved that the components were silica and iron oxide nanoparticles. These results proved our assumption of the chemical composition of the SJNCs.

Figure 3.11. TGA curve of the superparamagnetic Janus nanocomposites.11

To determine the amount of polymeric content in SJNPs, TGA was performed and it can be seen from Figure 3.11 that there was a weight of about 14.2 % in the range of 250 - 400 °C , which can be attributed to the polystyrene. The remaining mass of around 80 % could be considered as iron oxides and silica.

The magnetic properties of OAIOs and SJNPs were characterized by magnetization measurements at room temperature. As is shown in Figure 3.12, both OAIOs and SJNPs were superparamagnetic, exhibiting negligible magnetic remanence, which endows SJNPs the ability to be redispersed after external magnetic field is removed. The superparamagnetism provided opportunities for applications in MRI, hyperthermia and magnetic targeting.

Figure 3.12. Magnetization curves of (a) oleic acid modified iron oxide nanoparticles and (b) superparamagnetic Janus nanoparticles at room temperature.

Figure 3.13. T2 MR images of SJNCs with different concentrations. DI water was used as a control sample.

The potential application of SJNCs as MRI contrast agents was tested. As is shown in

Figure 3.13, as the concentration of SJNCs increased, the images darkened when subjected to

MRI due to the presence of superparamagnetic iron oxide nanoparticle components.

3.4. Conclusion

Dual functionalized superparamagnetic Janus Polystyrene/Fe3O4@Silica Nanoparticles have been synthesized via one-pot miniemulsion method based on previous work. The Janus nanoparticles showed anisotropy in composition, geometry and surface functionality. The reaction mechanism has been studied. The formation of miniemulsion system and phase separation was identified as the key factosr for the generating SJNCs. Compared to the similar structures produced by Feyen via stepwise selective coating,17 our approach is more convenient, which achieved desired structure in a one-pot synthesis. More importantly, the SJNCs were superparamagnetic and bi-functional simultaneously, making them truly multifunctional. The

SJNCs could serve as an excellent platform for bio-applications such as drug delivery, gene transfection and protein purification.

3.5. References

1. S. H. Hu and X. H. Gao, "Nanocomposites with Spatially Separated Functionalities for Combined Imaging and Magnetolytic Therapy," J Am Chem Soc, 132[21] 7234-37 (2010). 2. C. Zhang, B. Liu, C. Tang, J. Liu, X. Qu, J. Li, andZ. Yang, "Large scale synthesis of Janus submicron sized colloids by wet etching anisotropic ones," Chem Commun (Camb), 46[25] 4610-2 (2010). 3. J. Ge, Y. Hu, T. Zhang, andY. Yin, "Superparamagnetic composite colloids with anisotropic structures," J Am Chem Soc, 129[29] 8974-5 (2007). 4. W. Qiang, Y. Wang, P. He, H. Xu, H. Gu, andD. Shi, "Synthesis of asymmetric inorganic/polymer nanocomposite particles via localized substrate surface modification and miniemulsion polymerization," Langmuir, 24[3] 606-8 (2008). 5. X. Ge, M. Wang, Q. Yuan, andH. Wang, "The morphological control of anisotropic polystyrene/silica hybrid particles prepared by radiation miniemulsion polymerization," Chem Commun (Camb)[19] 2765-7 (2009). 6. B. Liu, C. Zhang, J. Liu, X. Qu, andZ. Yang, "Janus non-spherical colloids by asymmetric wet-etching," Chem Commun (Camb)[26] 3871-3 (2009).

7. Y. L. Wang, H. Xu, Y. S. Ma, F. F. Guo, F. Wang, and D. L. Shi, "Facile One-Pot Synthesis and Morphological Control of Asymmetric Superparamagnetic Composite Nanoparticles," Langmuir, 27[11] 7207-12 (2011). 8. T. Gong, D. Yang, J. H. Hu, W. L. Yang, C. C. Wang, andJ. Q. Lu, "Preparation of monodispersed hybrid nanospheres with high magnetite content from uniform Fe3O4 clusters," Colloid Surface A, 339[1-3] 232-39 (2009). 9. Y. Wang, H. Xu, W. Qiang, H. Gu, andD. Shi, "Asymmetric Composite Nanoparticles with Anisotropic Surface Functionalities," Journal of Nanomaterials, 2009 (2009). 10. Y. L. Wang, F. Wang, B. D. Chen, H. Xu, andD. L. Shi, "Facile one-pot synthesis of yolk-shell superparamagnetic nanocomposites via ternary phase separations," Chem Commun, 47[37] 10350-52 (2011). 1. S. H. Hu and X. H. Gao, "Nanocomposites with Spatially Separated Functionalities for Combined Imaging and Magnetolytic Therapy," J Am Chem Soc, 132[21] 7234-+ (2010). 2. C. Zhang, B. Liu, C. Tang, J. Liu, X. Qu, J. Li, andZ. Yang, "Large scale synthesis of Janus submicron sized colloids by wet etching anisotropic ones," Chem Commun (Camb), 46[25] 4610-2 (2010). 3. J. Ge, Y. Hu, T. Zhang, andY. Yin, "Superparamagnetic composite colloids with anisotropic structures," J Am Chem Soc, 129[29] 8974-5 (2007). 4. W. Qiang, Y. Wang, P. He, H. Xu, H. Gu, andD. Shi, "Synthesis of asymmetric inorganic/polymer nanocomposite particles via localized substrate surface modification and miniemulsion polymerization," Langmuir, 24[3] 606-8 (2008). 5. X. Ge, M. Wang, Q. Yuan, andH. Wang, "The morphological control of anisotropic polystyrene/silica hybrid particles prepared by radiation miniemulsion polymerization," Chem Commun (Camb)[19] 2765-7 (2009). 6. B. Liu, C. Zhang, J. Liu, X. Qu, andZ. Yang, "Janus non-spherical colloids by asymmetric wet-etching," Chem Commun (Camb)[26] 3871-3 (2009). 7. Y. L. Wang, H. Xu, Y. S. Ma, F. F. Guo, F. Wang, andD. L. Shi, "Facile One-Pot Synthesis and Morphological Control of Asymmetric Superparamagnetic Composite Nanoparticles," Langmuir, 27[11] 7207-12 (2011). 8. T. Gong, D. Yang, J. H. Hu, W. L. Yang, C. C. Wang, andJ. Q. Lu, "Preparation of monodispersed hybrid nanospheres with high magnetite content from uniform Fe3O4 clusters," Colloid Surface A, 339[1-3] 232-39 (2009). 9. Y. Wang, H. Xu, W. Qiang, H. Gu, andD. Shi, "Asymmetric Composite Nanoparticles with Anisotropic Surface Functionalities," Journal of Nanomaterials, 2009 (2009). 10. Y. L. Wang, F. Wang, B. D. Chen, H. Xu, andD. L. Shi, "Facile one-pot synthesis of yolk-shell superparamagnetic nanocomposites via ternary phase separations," Chem Commun, 47[37] 10350-52 (2011). 11. F. Wang, G. M. Pauletti, J. Wang, J. Zhang, R. C. Ewing, Y. Wang, andD. Shi, "Dual surface-functionalized Janus nanocomposites of polystyrene/Fe(3)O(4)@SiO(2) for simultaneous tumor cell targeting and stimulus- induced drug release," Adv Mater, 25[25] 3485-9 (2013). 12. K. Landfester, "Polyreactions in miniemulsions," Macromol Rapid Comm, 22[12] 896-936 (2001). 13. F. Montagne, O. Mondain-Monval, C. Pichot, andA. Elaissari, "Highly magnetic latexes from submicrometer oil in water ferrofluid emulsions," J Polym Sci Pol Chem, 44[8] 2642-56 (2006). 14. A. K. F. Dyab, M. Ozmen, M. Ersoz, andV. N. Paunov, "Fabrication of novel anisotropic magnetic microparticles," J Mater Chem, 19[21] 3475-81 (2009). 15. L. B. Feng, H. Li, M. J. Yang, andX. W. Wang, "Synthesis of SiO2/polystyrene hybrid particles via an esterification method," Colloid Polym Sci, 288[6] 673-80 (2010). 16. K. Zhang, W. Wu, H. Meng, K. Guo, andJ. F. Chen, "Pickering emulsion polymerization: Preparation of polystyrene/nano-SiO2 composite microspheres with core-shell structure," Powder Technol, 190[3] 393- 400 (2009). 17. M. Feyen, C. Weidenthaler, F. Schuth, andA. H. Lu, "Regioselectively Controlled Synthesis of Colloidal Mushroom

Nanostructures and Their Hollow Derivatives," J Am Chem Soc, 132[19] 6791-99 (2010).

Chapter 4. Drug Loading and Release Study on Silica

Nanoparticles

4.1. Introduction

Doxorubicin (DOX) (Figure 4.1), with the trade name Adriamycin, is a cancer chemotheraputic drug. It was discovered in the 1950s and has been widely studied by investigators throughout the world since then. DOX is often used in its hydrochloride salt form, interacting with DNA by intercalation and inhibition of macromolecular biosynthesis, thus preventing DNA chains from replication. 1-4

Figure 4.1. Chemical structure of doxorubicin.

However, DOX has many adverse effects, among which is cardiotoxicity could cause death. Therefore, DOX has earned itself the name of “Red Death”.5 We selected DOX as the model drug to evaluate the bio-application because the ketone group in doxorubicin can form a hydrazone bond (C=N) with a hydrazide (-NHNH2). As mentioned in Chapter 1, hydrazone bond is relatively stable at physiological condition (pH 7.2 ~ 7.4) but could break at endocytic contion (pH 4.5 ~ 6.5). This offers excellent opportunities for carrier systems because highly controlled release of doxorubicin could significantly reduce the side effects. Hydrazone bond based carriers have been investigated by many researchers.6-11 Hereby we proposed that the

Fe3O4@SiO2 hybrid half shell in the superparamagnetic Janus nanoparticle be loaded with DOX via a hydrazone bond. To avoid potential interference from the polystyrene core, we decided to try drug loading and release and cell cytoxicity study on silica nanoparticles first as a simplified model. Then the reaction could be applied on iron oxide@silica half shells because both silica surfaces share the same properties.

4.2. Experimental

4.2.1. Mateials and Chemicals

Tetraethoxysilane (TEOS), molecular sieves (3Å), 3-(Triethoxysilyl)propyl isocyanat, sodium dodecyl sulfate (SDS), and adipic acid dihydrazide, Hank's Balanced Salts, sodium bicarbonate, D-(+)-glucose, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) and 3-

(4,5-Dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) were purchased from

Sigma Aldrich and were used as received. Ammonium hydroxide (NH4OH, 29.79 wt%), hydrochloric acid (HCl, 37 %), phosphate buffered saline (PBS) and the DMEM media with high glucose (Hyclone) were purchased from Fisher Scientific (USA). HEPES buffer solution (10mM) and acetate buffer solution (10mM) were prepared in our laboratory. Doxorubicin hydrochloride

(DOX) was purchased from Meiji Seika Kaisha, Ltd. (Tokyo, Japan). The human breast cancer cell line MDA-MB-231 was obtained from American Type Culture Collection (ATCC) and cultured in DMEM with high glucose supplemented with 10% fetal bovine serum, 1% MEM-

Non Essential Amino Acids , 1% Penicillin Streptomycin and 1% l-glutamine. HBSS buffer was made by dissolving Hank's Balanced Salts (0.98 %), sodium bicarbonate (0.037 %), D-(+)- glucose (0.35 %) and HEPES (0.286 %) in autoclaved water and titrating pH to 7.35 with NaOH.

MTT stock solution was made by dissolving MTT powder in PBS buffer solution (5 mg mL-1) and stored in Dark. Lysis buffer was made by dissolving SDS (20 %) in DMF/water mixture

(50:50). Assay material CellTiter-Glo™ Luminescent Cell Viability Assay Kit was purchased

from Promega (Madison, WI). Deionized (DI) water was used through experiment. Molecular sieves were heated at 150 °C for 2 h, 350 °C for 3 h, cooled under vacuum and added into MeOH to remove absorbed water before use.

4.2.2. Methods

Fabrication of Silica Nanoparticles

In a typical reaction, 134.6 ml of EtOH, 7.0 ml of DI water and 2.81 ml of ammonium hydroxide were mixed well in a three-headed flask, followed by the addition of 6.74 ml of TEOS.

The system was stirred at 300 rpm at room temperature for 22 hours. The silica nanoparticles were obtained after washing the products with ethanol and DI water three times respectively using centrifugation at 10000 rpm. The silica nanoparticles were dried at room temperature under vacuum for further surface modification.

Synthesis of Silane-hydrazide cross-linker

52.26 mg (0.3 mmol) of adipic acid dihydrazide was dissolved in 4ml of DMSO. 74.2 µl

(0.3 mmol) of 3-(Triethoxysilyl)propyl isocyanat was dissolved in 500 µl of DMSO and added in to the adipic acid dihydrazide solution dropwise (one drop per 10 secs) under vigorous stirring.

After reaction overnight, DMSO was removed under reduced pressure at 80 °C and the oily residue was used without purification.

Surface Modification of Silica Nanoparticles

A standard silane coupling method was employed to modify the silica nanoparticles. 40

mg of silica nanoparticles were dispersed in 16 ml of EtOH/H2O (95:5) mixture. The as- synthesized silane-hydrazide was dissolved in 2 ml of EtOH and 70 µl of the solution was added into the silica dispersion under stirring. The pH value was tuned to 5.0 by HCl and the dispersion was stirred for another 4 hours at room temperature. Then the pH was tuned back to neutral by adding NaOH and stirred overnight. The surface modified silica nanoparticles were washed by centrifuging at 10000 rpm with EtOH and DI water for four times respectively and dried at room temperature under vacuum for drug loading.

DOX Conjugation to Surface Modified Silica Nanoparticles

First, 4 mg of DOX was dissolved in 15 ml of molecular sieve treated MeOH. The obtained silica-hydrazide nanoparticles (35 mg) were added to make dispersion. Then, 4 ul of trifluoroacetic acid (TFA) was added to change the pH to around 5.0. The mixture was kept stirring in dark at room temperature for 36 h. Finally, the dark red powders were washed with

MeOH till no absorbance of DOX at 480 nm was detected in the supernatant on a UV-Vis spectrophotometer. The silica-DOX conjugates were dried at room temperature under vacuum for drug release study. The drug loading capacity was determined using the subtraction method.

Unconjugated drug was calculated by measuring the DOX absorbance at 480 nm of the supernatant collected from all washes using a pre-established calibration curve of DOX solution in MeOH. The experimental flow is depicted in Figure 4.2.

Evaluation of pH sensitivity

To investigate the release behavior of DOX from silica-DOX conjugates at different pH values, three buffer media were prepared: acetate buffer (0.1 M, pH=5.0), acetate buffer (0.1 M,

pH=6.0) and 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) buffer (0.01 M, pH=7.4). 5 mg of silica-DOX conjugates were dispersed in three 1.5 ml microcentrifuge tubes, which contained 1 ml of the above buffer media respectively. Then the tubes were rotated at an incubator set at 37 °C . At certain time intervals, the dispersions were subjected to centrifugation

(10000 rpm) and the supernatant was taken for UV-Vis analysis to determine the amount of DOX released according to the pre-established calibration curve. Fresh buffer mediums of the same volume were added and the tubes were rotated continuously.

In-vitro Cell Cytotoxicity Assay

The cytotoxicity of SiO2-DOX conjugates against MDA-MB-231 cell was studied using both the MTT assay and the luminescent ATP-based assay kit (CellTiter GLO, Promega,

Madison, WI) according to the manufacturer's instructions. First, MDA-MB-231 cells were seeded (2500 cells per well) in a 96-well plate for 24 h. Then, the media was replaced with

HBSS buffer solution containing free DOX, SiO2-DOX and SiO2-NHNH2. Thereafter, cells were incubated for 4 h, washed with warm HBSS buffer twice and further incubated with serum- containing media for another 48 h. Then the plate was proceeded for assays for cytotoxicity data.

Triton X-100 was used as the toxic control.

As for the MTT assay, 25 µl of MTT stock solution was added into each well and incubated for 3 h. After that, 100 µl of lysis buffer was added into each well and mixed well using the pipette gun. After incubation overnight, the absorbance at 570 nm was recorded on a

Spectra Max 250 (Molecular Devices) with a cutoff at 690 nm.

The luminescence assay was carried out according to the manufacture’s instruction. The

96-well plate was equilibrated at room temperature before 100 µl of reconstituted assay solution

(CellTiter GLO) was added into each well. Then the plate was shaken for 5 min for mixing well and equilibrated for another 10 min. Finally luminescence was recorded on a plate reader.

All results were expressed as mean ± standard deviation. Statistical analysis was performed using GraphPad Prism 5. Comparisons between groups were performed using

Students t-test at a significance level of p < 0.05 (GraphPad Prism 5).

Figure 4.2. Schematic illustration of the synthesis strategy of doxorubicin loaded silica nanoparticles.

4.2.3. Characterization

Scanning Electron Microscopy (SEM)

The obtained silica nanoparticles were dispersed in DI water and dried onto SEM sample holder without gold sputtering for examination. Images were obtained on a field emission scanning electron microscope (FESEM, Leo 1530) with an accelerating voltage of 20 kV.

Transmission Electron Microscopy (TEM)

Transmission electron microscopy (TEM) images were obtained using a Tecnai F30 transmission electron microscope operated at 300 kV.

Dynamic Light Scattering (DLS)

Size distribution of SJNCs dispersed in DI water was measured using a Malvern

Zetasizer (Nano ZS).

Fourier Transformed Infrared Spectroscopy (FTIR)

The as-synthesized and hydrazide modified silica nanoparticles were pelleted with potassium bromide powder and infrared spectroscopy was obtained using a transmission method on a Nicolete 6700 FTIR from Thermo Scientific.

UV-Vis Spectroscopy

UV-Vis spectra of hydrazide modified and DOX loaded silica nanoparticles were recorded on a Hitachi U-3000 spectrophotometer using a cuvette with the path length of 1 cm and nominal volume of 350 µl .

MTT Assay

The absorbance of MTT assay was recorded on Spectra Max 250 from Molecular

Devices.

Luminescence Assay

Luminescence was obtained on a POLARstar OPTIMA microplate reader (BMG

Labtechnologies, Chicago, USA).

4.3. Results and Discussion

4.3.1. Silica Nanoparticle Synthesis and Doxorubicin Conjugation

As can be seen in Figure 4.3, the silica nanoparticles synthesized from sol-gel reaction are highly spherical in shape and about 120 nm in diameter. The hydrodynamic diameter obtained from DLS show a value of 175.3 nm with a poly-dispersity index (PDI) of 0.125. The

DLS data is displayed in Figure 4.4.

Figure 4.3. (A) SEM and (B) TEM images of silica nanoparticles. Both scale bars are 200 nm.

Figure 4.4. DLS Data of silica nanoparticles.

As is shown in Figure 4.2, -N=C=O groups are electrophiles and are reactive toward the amino groups on adipic acid dihydrazide, which are nucleophiles. The reactant 3-

(Triethoxysilyl)propyl isocyanat was added dropwise to ensure that the amino groups from adipic acid dihydrazide were excessive so that the opportunity that one adipic acid dihydrazide molecule is connected to two 3-(Triethoxysilyl)propyl isocyanat molecules can be minimized.

DMSO was chosen as the solvent because isocyanate groups are also reactive to water.

Figure 4.5 shows the FTIR spectra of silica nanoparticles before and after modification with silane-hydrazide. It is obvious that compared to unmodified silica nanoparticles, the hydrazide modified silica nanoparticles exhibit an apparent peak at 1550 cm-1, which arises from the ending amino groups of hydrazide. The peak at 1670 cm-1 can be attributed to C=O groups in the silane-hydrazide cross-linker. These peaks indicate that the silica nanoparticles had already been successfully modified with hydrazide and were ready for drug loading.

Figure 4.5. FTIR spectra of (a) unmodified silica nanoparticles and (b) hydrazide modified silica nanoparticles.

4.3.2. Drug loading and release study

To confirm that DOX is not physically attached to the silica surface, unmodified silica nanoparticles were subjected to the same procedure for DOX conjugation and visualized comparison showed a clear difference. It can be seen in Figure. 4.6B that after extensive MeOH washing, unmodified silica nanoparticles remained white, while hydrazide modified silica nanoparticles turned red, which is a clear sign of DOX loading. This indicates that physically attached DOX was already washed away in both cases. Thus it is reasonable to conclude that negligible DOX was physically attached to the hydrazide modified silica surface.

Figure 4.6. Photographs of unmodified silica nanoparticles and silica-hydrazide nanoparticles after DOX conjugation (A) after first centrifugation washing and (B) after 10 times of centrifugation washing with MeOH.

Figure 4.7 shows the UV-Vis spectra of silica-hydrazide nanoparticles and silica-DOX conjugates. It is obvious that after DOX conjugation, the spectra of silica-DOX nanoparticles exhibit a broad shoulder with the peak at around 517 nm compared to the hydrazide modified silica nanoparticles. This peak deviates from the characteristic peak of pure DOX at around 480 nm, which is possibly due to the covalent bonding between DOX and silica. Similar results were reported by Q. Hu12 when DOX was conjugated through an adamantyl group. These results imply a conjugation of DOX to the silica surface via covalent bonding, which is hydrazone in this case. The DOX loading capacity was determined to be 2.94 ± 0.26 % (w/w) using the subtraction method. Unconjugated DOX was calculated by measuring the DOX absorbance of the supernatant collected from all washes using a pre-established calibration curve of DOX

solution in MeOH (Figure 4.8A). Then the loaded DOX was determined by subtract the unconjugated amount from the starting amount of DOX.

Figure 4.7. UV-Vis spectra of (a) silica-hydrazide nanoparticles and (b) DOX conjugated silica nanoparticles.

Figure 4.8. Calibration curve of doxorubicin in (A) MeOH and (B) DI water.

Hydrazone bond is acidic cleavable and can undergo hydrolysis to release DOX. The amount of DOX released was calculated by measuring absorbance at 480 nm using the pre- established calibration curve of DOX in DI water (Figure 4.9b). Figure 4.9 demonstrates that initial DOX release rate is 10-fold faster at pH 5.0 than at pH 7.4, which mimics normal physiological condition. At pH 7.4, the accumulative DOX release after 120 h was 9.1 % (w/w).

On the contrary, at pH 5.0 and pH 6.0, which are similar to the environment in endosomes, lysosomes, and tumor sites, DOX releases were at much faster rates and the accumulative releases reached 73.5 % (w/w) and 53.0 % (w/w) after 120 h, respectively. This observation confirms the acid-sensitive cleavage of hydrazone bond to release DOX. As pH decreased, the concentration of H+ increased in the buffer solutions and the hydrolysis of hydrazone bond became faster thus more DOX was released from the conjugates.

Figure 4.9. DOX release profiles from DOX conjugated silica nanoparticles at (a) pH 5.0 and (b) pH 6.0 and (c) pH 7.4.

4.3.3. Cell Cytotoxicity Analysis

After silica nanoparticles were loaded with DOX, the zeta potential increased from -27.6 mv to +35.1 mv (Table 4.1). The charged surfaces have a high chance aggregating toward the protein in the media because the protein in the media are large surface charged molecules and can easily capture the conjugates via the electrostatic attraction. Thus we decided to avoid this issue by not protein free buffer solutions. Hank’s buffered salt solution (HBSS) can maintain the level of pH and osmotic balance as well as provide cells with water and other essential inorganic ions. The SiO2-DOX could be dispersed in HBSS and incubated with the cells for a certain period of time, during which absence of protein would not cause significant adverse effects on the cells. Considering that 2 h of incubation is too short for DOX to be released at a pH of 7.4 and 8 h of incubation is too long that cells may die, we tried two time frames: 4 h and 6 h.

Table 4.1. Zeta potential of unmodified SiO2, hydrazide modified SiO2, and SiO2-DOX conjugates.

SiO2 SiO2-NHNH2 SiO2-DOX

Zeta Potential (mv) -27.6 -2.19 +35.1

Figure 4.10. Cell viability comparison of MDA-MB-231 cells after cells were incubated with

HBSS buffer solutions and media for 4 h and 6 h, respectively. ***p < 0.001; ns (not significant), p > 0.05.

As can be seen in Figure 4.10, after cells were incubated with HBSS and media for 6 h, the cell viabilities showed significant difference. On the other hand, after incubation with HBSS and media for 4 h, cell viabilities were not significantly different. Based on this result, we decided to choose HBSS as the incubation buffer for drug release for a time period of 4 h.

Figure 4.11. Cell cytotoxicity data of free DOX against MDA-MB-231 cells showing different values between MTT assay (hollowsquares) and luminescence assay (solid spheres).

Figure 4.12. UV-Vis spectra of doxorubicin hydrochloride in DI water.

To measure the cell cytotoxicity, we first tried the widely-used MTT assay. In this assay,

3-(4,5-Dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) solution is incubated with cells and is reduced to purple formazan in living cells.13 Then lysis buffer (SDS solution in

DMF/water in our case) is added to dissolve the formazan and the solution turns purple. Finally the absorbance of the purple solution is quantified by measuring on a plate reader at 570 nm. The cell viability can be calculated from the absorbance ration of target well over untreated well

(100 % viability).

However, after MTT assay was applied to test the cytotoxicity of free DOX against

MDA-MB-231 cells, it was found that the absorbance values first decreased with increasing amount of DOX added but started to rise with increasing dosage of DOX at the high concentration end (Figure 4.11). This is contrary to the assumption that fewer cells would survive with increasing dosage of DOX, which was subsequently confirmed by optical microscopy. By checking the absorbance spectra of DOX (Figure 4.12), it was noticed that DOX still has some absorbance at 570 nm, which I had thought would be negligible in the MTT assay.

Then we decided to turn to other assays which have no interference with absorbance of DOX.

The CellTiter-Glo™ Luminescent Cell Viability Assay Kit is based on the reaction of the reagent with ATP in presence turning into luminescence which can be quantified. This would indirectly quantify the living cells. The cell cytotoxicity profile appeared more reasonable and was consistent with optical microscopic observation after we changed to the luminescent kit.

Based on the sigmoidal fitting of the data points, it was calculated that the IC50 value of free

DOX against MDA-MB-231 cells was 3.3 ± 0.3 µg ml-1.

Figure 4.13. Cell cytotoxicity curves of free DOX, SiO2-DOX conjugates (SiO2-DOX) and hydrazide modified SiO2 (SiO2-NHNH2) against MDA-MB-231 cells. The connecting lines are constructed from non-linear dose-response fit.

Cell cytotoxicity analysis was also conducted on SiO2-DOX conjugates and hydrazide modified SiO2. As can be seen in Figure 4.13, cell viability decreased with increasing dosage of

SiO2-DOX, which is obviously inhibitor-response characteristic. On the contrary, the viabilities of the cells treated with hydrazide modified SiO2 remained above 90 % for all dosages. This combination proved that it was the DOX released from the SiO2-DOX conjugates that caused the cell death. It was calculated that the IC50 value of SiO2-DOX conjugates against MDA-MB-231 cells 679.8 ± 41.0 µg ml-1. According the IC50 value of free DOX, about 3.3 µg of DOX was released and diffused into the nucleus of MDA-MB-231 cells, which gave a percentage of 16.3 % release from the total DOX incorporation. Considering that only about 1 % of DOX could be released in 4 h at pH 7.4, which was the pH condition of HBSS buffer solution, we hypothesized

that some SiO2-DOX conjugates were internalized by cells and released DOX at a much faster speed in endocytic compartments (endosomes & lysosomes) because the pH was lower (pH 4.5 ~

6.5). In this way, more DOX was released and caused higher cell cytotoxicity.

4.4. Conclusion

Silica nanoparticles have been fabricated and utilized as the base to study the theory and feasibility of drug delivery via hydrazone bonding. A new silane-hydrazide cross-linker has been synthesized and used to modify the surface of silica nanoparticles. Doxorubicin has been conjugated to the hydrazide modified silica nanoparticles and drug release behaviors of silica/DOX conjugates showed strong pH dependence. It was confirmed that SiO2-DOX conjugates showed cell cytotoxicity against MDA-MB-231. It was further hypothesized that some SiO2-DOX conjugates were internalized by cells and released DOX faster in a mild acidic endocytic environment.

4.5. References

1. W. J. Pigram, W. Fuller, andL. D. Hamilton, "Stereochemistry of Intercalation - Interaction of Daunomycin with DNA," Nature-New Biol, 235[53] 17-& (1972). 2. R. L. Momparler, M. Karon, S. E. Siegel, andF. Avila, "Effect of Adriamycin on DNA, Rna, and Protein- Synthesis in Cell-Free Systems and Intact-Cells," Cancer Res, 36[8] 2891-95 (1976). 3. F. A. Fornari, J. K. Randolph, J. C. Yalowich, M. K. Ritke, andD. A. Gewirtz, "Interference by Doxorubicin with DNA Unwinding in Mcf-7 Breast-Tumor Cells," Mol Pharmacol, 45[4] 649-56 (1994).

4. J. W. Lown, "Anthracycline and Anthraquinone Anticancer Agents - Current Status and Recent Developments," Pharmacol Therapeut, 60[2] 185-214 (1993). 5. J. E. Groopman, "How Doctors Think." Houghton Mifflin, (2007). 6. C. H. Lee, S. H. Cheng, I. P. Huang, J. S. Souris, C. S. Yang, C. Y. Mou, andL. W. Lo, "Intracellular pH- Responsive Mesoporous Silica Nanoparticles for the Controlled Release of Anticancer Chemotherapeutics," Angew Chem Int Edit, 49[44] 8214-19 (2010). 7. J. E. Lee, D. J. Lee, N. Lee, B. H. Kim, S. H. Choi, andT. Hyeon, "Multifunctional mesoporous silica nanocomposite nanoparticles for pH controlled drug release and dual modal imaging," J Mater Chem, 21[42] 16869-72 (2011). 8. X. Q. Yang, J. J. Grailer, S. Pilla, D. A. Steeber, andS. Q. Gong, "Tumor-Targeting, pH-Responsive, and Stable Unimolecular Micelles as Drug Nanocarriers for Targeted Cancer Therapy," Bioconjugate Chem, 21[3] 496-504 (2010). 9. D. X. Lu, X. T. Wen, J. Liang, Z. W. Gu, X. D. Zhang, andY. J. Fan, "A pH-Sensitive Nano Drug Delivery System Derived From Pullulan/Doxorubicin Conjugate," J Biomed Mater Res B, 89B[1] 177-83 (2009). 10. S. C. Wuang, K. G. Neoh, E. T. Kang, D. E. Leckband, andD. W. Pack, "Acid-Sensitive Magnetic Nanoparticles as Potential Drug Depots," Aiche J, 57[6] 1638-45 (2011). 11. Y. L. Chang, X. L. Meng, Y. L. Zhao, K. Li, B. Zhao, M. Zhu, Y. P. Li, X. S. Chen, andJ. Y. Wang, "Novel water-soluble and pH-responsive anticancer drug nanocarriers: Doxorubicin-PAMAM dendrimer conjugates attached to superparamagnetic iron oxide nanoparticles (IONPs)," J Colloid Interf Sci, 363[1] 403-09 (2011). 12. Q. D. Hu, H. Fan, Y. Ping, W. Q. Liang, G. P. Tang, andJ. Li, "Cationic supramolecular nanoparticles for co-delivery of gene and anticancer drug," Chem Commun, 47[19] 5572-74 (2011). 13. T. Mosmann, "Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays," J Immunol Methods, 65[1-2] 55-63 (1983).

Chapter 5. Folic Acid Conjugation, Drug Loading and in-vitro Release and Cytotoxicity Study of

Superparamagnetic Janus Nanocomposites

5.1. Introduction

Folic acid (Figure 5.1) is a vitamin essential for cell proliferation. The cellular uptake of folic acid is mediated by the protein known as the folate receptor.1 Folate receptors can facilitate cellu internalization after binding to folic acid via receptor-mediated endocytosis.1, 2, 3, 4, 5 It has been known that folate receptors have two isoforms: type-α and type-β, both of which have high affinity towards folic acid.6 The type-α folate receptors are only present in certain epithelial tissues and the type-β folate receptors are present in almost all normal tissues, both at very low levels. However, both folate receptors are significantly overexpressed in maliganant tissues. 1, 6, 7,

8, 9 For this reason, folic acid has been considered a targeting ligand and conjugated to proteins, liposomes and anti-cancer drugs for selective tumor targeting.10, 11, 12, 13, 14, 15, 16 Folic acid has two carboxyl groups, α- and γ-, in which the γ-carboxyl group possesses a much higher level of reactivity toward amino group in a coupling reaction mediated by carbodiimide.10 More

importantly, after folic acid is conjugated to a carrier or a foreign molecule via the γ-carboxyl group, the α-carboxyl group still retains high reactivity toward the folate receptor.17, 18

Figure 5.1. Structure of a folic acid molecule.

Compared to tumor-specific antibodies, folic acid targeting has some advantages:1, 9, 19

1) Low immunogenicity. Antibodies may receive host immune responses which produces a

barrier for clinical use;

2) Low molecular weight. The small molecule can rapidly diffuse through a tumor mass;

3) Relatively high chemical and physical endurance. Folic acid can retain chemical activity

and targeting ability after conjugation reaction and freezing-thawing cycles;

4) Simple conjugation chemistry. Carbodiimide-mediated coupling of carboxyl toward

amino groups have been well defined with relatively high yield;

5) Highly specific to tumors with high affinity

Therefore, we decided to use folic acid as the targeting ligand and doxorubicin as the model drug. Because the polystyrene surfaces were also decorated with carboxyl groups, poly(ethylene glycol) bis-amine with (NH2-PEG-NH2) was taken as a linker. First, folic acid was

linked to one amino ending group of NH2-PEG-NH2 and then the PEG-FA was functionalized to the polystyrene surfaces via the other amino group. As for the drug loading, a silane bearing hydrazide ending group was synthesized from 3-(triethoxysilyl)propyl isocyanat and adipic acid dihydrazide via carbodiimide-mediated coupling first. Then the Fe3O4/silica shell surfaces were modified with this silane-hydrazide crosslinker. Finally doxorubicin was conjugated to the hybrid shell surfaces via hydrazone bonds. It was hypothesized that the targeted superparamagnetic

Janus nanocomposites (SJNCs) bearing anti-cancer drug could be internalized by tumor cells via folate receptor-mediated endocytosis and release doxorubicin faster in endocytic compartments to achieve cytotoxicity. In this way, less amount of drug was released during physiological conditions due to the relatively high stability of hydrazone bonds and side effects of doxorubicin could be reduced. The schematic drawing of surface manipulation of SJNCs and cell internalization is depicted in Figure 5.2.

Figure 5.2. Schematic illustration of folic acid conjugation and drug loading onto SJNCs and subsequent cell internalization through endocytosis.20

5.2. Experimental

5.2.1. Materials

Triethylamie (TEA), folic acid (FA), 3-(triethoxysilyl)propyl isocyanat, adipic acid dihydrazide, trifluoroacetic acid (TFA), Hank's Balanced Salts, sodium bicarbonate, D-(+)- glucose and 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) were purchased from

Sigma-Aldrich (USA). Diethyether, tetrahydrofuran (THF), phosphate buffered saline (PBS) and the DMEM media with high glucose (Hyclone) were purchased from Fisher Scientific (USA). 1-

Ethyl-3-(3-dimethylaminopropyl) carbodiimide HCl (EDC) and N-Hydroxysuccinimide (NHS) were purchased from Thermo Fisher Scientific (USA). Poly(ethylene glycol) bis-amine with molecular weights of 1000 Da (NH2-PEG-NH2) was obtained from YareBio (China).

Doxorubicin hydrochloride (DOX) was purchased from Meiji Seika Kaisha, Ltd. (Tokyo, Japan).

Acetate buffer solutions (10 mM) were prepared in our laboratory. The nucleus staining material

Prolong Gold Antifade Regent with DAPI was purchased from Invitrogen. The human breast cancer cell line MDA-MB-231 was obtained from American Type Culture Collection (ATCC) and cultured in DMEM with high glucose supplemented with 10% fetal bovine serum, 1%

MEM-Non Essential Amino Acids , 1% Penicillin Streptomycin and 1% l-glutamine. HBSS buffer was made by dissolving Hank's Balanced Salts (0.98 %), sodium bicarbonate (0.037 %),

D-(+)-glucose (0.35 %) and HEPES (0.286 %) in autoclaved water and titrating pH to 7.35 with

NaOH. Assay material CellTiter-Glo™ Luminescent Cell Viability Assay Kit was purchased from Promega (Madison, WI). All chemical were used as received. Deionized (DI) water was

used through experiment.

5.2.2. Methods

Synthesis of PEG-FA and further conjugation onto Polystyrene Surface of SJNCs

Folic acid was conjugated to the NH2-PEG-NH2 through a carbodiimide-mediated coupling reaction. In a typical reaction, 50 mg of NH2-PEG-NH2 was dissolved in 8 ml of DMSO.

Then 22mg of folic acid, 47.8 mg of EDC and 15mg of NHS were dissolved in 10ml of DMSO followed by addition of 20 ul of TEA and the mixture was magnetically stirred for 1 h. The resulting yellow solution was added dropwise (one drop per 2 secs) into the NH2-PEG-NH2 solution under vigorous stirring and the reaction mixture was stirred overnight before DMSO was removed by distillation under reduced pressure at 80 °C. The yellow oily residue was dissolved in THF and PEG-FA was obtained by precipitation after the addition of diethyether.

Conjugation of PEG-FA onto Polystyrene Surface of SJNCs

For conjugation of PEG-FA onto the PS surface of SJNCs, 40 mg of SJNCs were dispersed in 10 ml of DI water, followed by dissolution of 80 mg of EDC. The suspension was stirred for 15 min. Then, 12 mg of NHS and 15 mg of PEG-FA were added. The reaction system was stirred overnight and subjected to magnetic wash until no absorbance can be detected at

270nm in the supernatant by UV-Vis spectroscopy. The resulting FA-PS/Fe3O4@SiO2 nanocomposites (FA-SJNCs) were dried under vacuum at room temperature.

Synthesis of Silane-hydrazide Crosslinker

In a typical reaction, 52.26 mg of adipic acid dihydrazide was dissolved in 4 ml of

DMSO. Then 74.2 ul of 3-(Triethoxysilyl)propyl isocyanat was dissolved in 500 ul of DMSO and added in to the adipic acid dihydrazide solution dropwise (one drop per 10 secs) under vigorous stirring. After reaction overnight, DMSO was removed under reduced pressure at 80 °C and the waxy residue was dissolved in 2 ml of EtOH.

Surface Modification of Silica Shell of SJNCs with Silane-hydrazide

A standard silane coupling method was employed to modify the silica surface. 40 mg of

FA-NPs were dispersed in 16 ml of EtOH/H2O (95:5) mixture. The 60 ul of the above silane- hydrazide solution in EtOH was added into the FA-SJNCs dispersion under stirring. The pH was tuned to 5 by HCl and the dispersion was stirred for 4 h at room temperature. Then the pH was tuned back to neutral by adding NaOH and stirred overnight. The surface modified FA-SJNCs-

NHNH2 nanocomposites were washed by magnetic separation with EtOH and DI water for four times respectively and dried at room temperature under vacuum for drug loading.

Doxorubicin loading onto FA-SJNCs-NHNH2 Nanocomposites

DOX was conjugated to the surface of silica half shell through the formation of hydrazone bonds. Briefly, 35 mg of FA-SJNCs-NHNH2 nanocomposites were dispersed in 15 ml of DOX solution in anhydrous methonal (0.33 mg ml-1) and 3 ul of TFA was added to tune the pH to 5. The mixture was stirred in darkness for 40 h at room temperature and intense magnetic wash with anhydrous methanol was carried out till no absorbance at 480 nm could be detected in the supernatant on a UV-Vis spectrophotometer. The FA-SJNCs-DOX conjugates were dried at room temperature under vacuum. The details of the reactions are shown in Figure 5.3.

Evaluation of pH sensitivity

The pH-responsive DOX release profile from the FA-SJNCs-DOX conjugates was studied using UV-Vis spectrophotometer. First, 10mg of FA-SJNCs-DOX conjugates were dispersed in 5 ml of buffer solutions of different pH conditions (PBS buffer, pH 7.4 and acetate buffer (pH 5.0 and 6.0)) in glass vials and placed on an electric rotator in an incubator set at

37 °C . At selected time intervals, 1 ml of supernatant was removed from the vial after magnetic wash and replaced by fresh buffer solution with the same volume. The absorbance of released

DOX was measured by a UV-Vis spectrophotometer at 480 nm and the amount was calculated according to a pre-established DOX calibration curve.

In-vitro Cell Cytotoxicity Assay

The cytotoxicity of SJNCs-DOX conjugates against MDA-MB-231 cell was studied using the luminescent ATP-based assay kit (CellTiter GLO, Promega, Madison, WI) according to the manufacturer's instructions. First, MDA-MB-231 cells were seeded (2500 cells per well) in

96-well plates for 24 h. Then, the media was replaced with HBSS buffer solution containing free

DOX, SJNCs, FA-SJNCs-DOX and SJNCs-DOX. In the FA competition experiment, free folic acid as the inhibitor was added to the FA-SJNCs-DOX containing HBSS buffer solution to make a 1 mM concentration. Thereafter, cells were incubated for 4 h, washed with warm HBSS buffer twice and incubated with fresh media for another 48 h. Finally, luminescence was recorded on a plate reader after applying the assay kit. Triton X-100 was used as the toxic control. All results were expressed as mean ± standard deviation. Statistical analysis was performed using GraphPad

Prism 5. Comparisons between groups were performed using t-tests with a value of p < 0.05 considered statically significant.

Fluorescent Microscopy

MDA-MB-231 cells were seeded in an 8-well chamber, each with the dimension of 1 cm x 0.75 cm, at a density of 3500 cells/well. Then 200 ul of DOX solution, FA-SJNCs-DOX and

SJNCs-DOX dispersion were incubated with the cells at the concentrations of 5.8 ug ml-1, 278 ug ml-1 and 278 ug ml-1 respectively. After 2 h, the wells were aspirated and washed with HBSS buffer twice, and the cells were fixed with 200 ul of 4 % formaldehyde for 30 min. After that, the formaldehyde was aspirated, the walls of the chamber were removed and the slide was rinsed with PBS (pH 7.4) several times. Finally, the cell nuclei were stained with Prolong Gold Antifade

Regent with DAPI and mounted for fluorescent microscope observation.

Figure 5.3 Schematic drawing of (A) folic acid functionalization of polystyrene core surface and

20 (B) DOX loading onto the Fe3O4@silica hybrid shell surface.

5.2.3. Characterization

Mass Spectroscopy

The samples were analyzed using Bruker BIFLEX III MALDI instrument from Bruker

Daltonics (Billerica, MA, USA). A saturated solution of 2-(4-hydroxyphenylazo)-benzoic acid

(HABA) in methanol was used as Matrix.

Fourier Transformed Infrared Spectroscopy (FTIR)

UV-Vis Spectroscopy

UV-Vis spectra of hydrazide modified and DOX loaded silica nanoparticles were recorded on a Hitachi U-3000 spectrophotometer using a cuvette with the path length of 1 cm and nominal volume of 350 ul.

Luminescence Assay

Luminescence was obtained on a POLARstar OPTIMA microplate reader (BMG

Labtechnologies, Chicago, USA).

5.3 Results and Discussion

5.3.1 PEG-FA Synthesis and Conjugation

As shown in Figure 5.3A, poly(ethylene glycol) bis-amine was going to react with folic

100

acid to form PEG-FA and then be functionalized to the polystyrene surfaces. However, poly(ethylene glycol) bis-amine has two amino ending groups, both of which have the same reactivity towards carboxyl groups. So it is highly possible that one molecule of poly(ethylene glycol) bis-amine crosslinks with two molecules of folic acid, which will make FA-PEG-FA. In that case, the product would lose the other amino ending groups, which was supposed to be reacting with the carboxyl groups on polystyrene core surfaces. For this reason, the carboxyl groups (mainly γ-) of folic acid was activated with EDC/NHS first and then added dropwise into the PEG solution to make sure that the activated folic acid molecules were surrounded by far excessively more PEG molecules. In this way, there was less chance that one PEG molecule linked to two folic acid molecules. Considering that the molar ration of FA/PEG was 1:1 and unreacted folic acid would not dissolve in THF, it was hypothesized that the final product would contain only PEG-FA rather than a mixture of PEG-FA and FA-PEG-FA. The final product was subjected to mass spectroscopy examination and the result (Figure 5.4) showed that the poly(ethylene glycol) bis-amine with an average molecular weight of 1000 Da had a relatively narrow molecular weight distribution (Figure 5.4a). After crosslinking with folic acid, the molecular weight increased by around 440 Da, which was close to the molecular of folic acid.

Besides, there were no obvious peaks near 1800, which eliminated the presence of FA-PEG-FA.

So the product consisted of one molecule of folic acid connecting one molecule of PEG.

101

Figure 5.4. MS spectra of (a) NH2-PEG-NH2 and (b) PEG-FA

The product was also subjected to FTIR analysis. As shown in Figure 5.5, the PEG-FA displays a series of peaks between 1250 cm-1 to 1750 cm-1, which is characteristic of folic acid.

These peaks are complicated to be clearly indentified and there’s no need for doing so. The spectra also shows a clear peak at 1106 cm-1, which is attributed to ether groups (-C-O-C). This is an obvious sign of presence of PEG in the product, as the same peak can be located in pure

PEG.

102

Figure 5.5. FTIR spectra of (a) folic acid, (b) NH2-PEG-NH2 and (c) PEG-FA.

After PEG-FA was functionalized to the polystyrene surfaces via carbodiimide-mediated coupling reaction, the FA-SJNCs were examined by UV-Vis spectroscopy. As can be seen in

Figure 5.6, after PEG-FA conjugation, the curve of FA-SJNCs shows a broad should centered at

363 nm, which is characteristic of folic acid.13, 21 The low intensity was possibly due to the small amount of PEG-FA functionalized on the surfaces. Therefore, the UV-Vis spectra confirmed the successful conjugation of PEG-FA on polystyrene surfaces.

103

Figure 5.6. UV-Vis spectra curves of (a) SJNCs and (b) FA-SJNCs.

5.3.2. Doxorubicin Loading and Release at Different pH Values

The same silane-hydrazide described in Chapter 4 was synthesized and conjugated to the

Fe3O4@Silica half shell surfaces. As can be seen in Figure 5.7, compared to FA-SJNCs, the

-1 hydrazide modified FA-SJNCs (FA-SJNCs-NHNH2) exhibit a small peak around at 1550 cm , which is attributed to the ending amino groups of hydrazide. The relatively high and broad peak at 1670 cm-1 could be assigned to C=O bonds in the amide groups in silane-hydrazide. These peaks provide proof that the ybrid shell surfaces have already been successfully modified with hydrazide for drug loading. However, after DOX loading, the Janus nanocomposites did not

104

show much difference from the hydrazide modified nanocomposites, because doxorubicin did not have significantly different peaks that can be identified by FTIR.

Figure 5.7. FTIR spectra of (a) SJNCs, (b) FA-SJNCs, (c) FA-SJNCs-NHNH2 and (d) FA-

SJNCs-DOX.20

The same set of nanocomposites were also subjected to UV-Vis examination and results are shown in Figure 5.8. It is obvious that after DOX loading, the spectra of FA-SJNCs-DOX generated a broad shoulder with the peak at around 517 nm compared to the hydrazide modified

Janus nanocomposites. The deviation of this peak from the characteristic peak of pure DOX at

105

around 480 nm is perhaps due to the covalent bonding formed between DOX and silica surfaces.

Similar results were found by Q. Hu.22 These results clearly confirmed the successful conjugation of DOX to the hybrid shell surface via covalent bonding. In this case hydrazone bonding was formed. After using subtraction method, the DOX loading capacity was determined to be 2.08 ± 0.2 wt%. Similarly, unconjugated DOX was calculated by measuring the DOX absorbance of the supernatant collected from all washes using a pre-established calibration curve of DOX solution in MeOH (Figure 4.9a). Then the loaded DOX was determined by subtract the unconjugated amount from the starting amount of DOX.

Figure 5.8. UV-Vis spectra of (a) SJNCs, (b) FA-SJNCs, (c) FA-SJNCs-NHNH2 and (d) FA-

SJNCs-DOX.20

106

The purpose of conjugating DOX using hydrazone bonding was to utilize the pH responsive hydrazone bonding. Therefore, the drug release behaviors were monitored at different pH values. The amount of DOX released was calculated by measuring absorbance at 480 nm using the pre-established calibration curve of DOX in DI water (Figure 4.9b). It can be seen in

Figure 5.9 that at pH 7.4, which simulated the physiological condition, accumulative DOX release after 195 h was about 25.1 %. In contrast, at pH 5.0 and pH 6.0, which were similar to the environment in endocytic compartments and tumor sites, DOX releases were much faster and the accumulative releases reached 82.6 % and 47.1 % after 195 h, respectively. This observation was consistent with the acid-sensitive nature of hydrazone bonding. As environment became more acidic, the concentration of H+ increased in the buffer solutions and the hydrolysis of hydrazone bond accelerated, thus releasing more DOX from the conjugates.

107

Figure 5.9. DOX release profiles from DOX conjugated SJNCs at (a) pH 5.0 and (b) pH 6.0 and

5.3.3 Cell Cytotoxicity Analysis

Although DOX release profiles gave satisfying results regarding the pH sensitivity of hydrazone bonding, the targeting effects from the folic acid on polystyrene surfaces were yet to be tested. Targeting effect and pH-triggered DOX release were then combined and evaluated against the human breast cell line MDA-MB-231 overexpressing folate receptors. Based on the work presented in Chapter 4, the incubation time of cells with drug bearing nanocomposites was decided to be 4 h. To avoid the potential aggregation issue from the electrostatic attraction

108

between the nanocomposites and the protein in the cell culture media, HBSS buffer solution was used to disperse nanocomposites and nanocomposites/DOX conjugates. The cells were incubated with nanocomposites/DOX conjugates for 4 h and replaced with fresh media. Cell viabilities were examined after another 48 h.

Two control experiments were designed to test the targeting effect of folic acid:

1) The drug bearing and folic acid conjugated nanocomposites (FA-SJNCs-DOX)

were subjected to a free folic acid competition experiment. Folic acid was

dissolved in the HBSS buffer to make a concentration of 1 mM. Then FA-

SJNCs-DOX were dispersed in this buffer and proceeded to incubation with

cells.

2) The drug bearing nanocomposites without folic acid (SJNCs-DOX) were

dispersed in HBSS buffer and subjected to subsequent incubation with cells.

109

Figure 5.10. Cytotoxicity assay curves of (A) targeted FA-SJNCs-DOX versus targeted FA-

SJNCs-DOX in free folic acid competition and (B) targeted FA-SJNCs-DOX versus non-targeted

SJNCs-DOX against MDA-MB-231 cell line. Cytotoxicity assay curves for DOX and nanocomposites bearing folic acid but none drug (FA-SJNCs-NHNH2) are displayed as reference in both graphs.20

The results of cytotoxicity assay on different experimental groups are shown in Figure

5.10. As can be seen in Figure 5.10A and Figure 5.10B, the curve of targeted FA-SJNCs-DOX lies to the right of the curve for DOX because the FA-SJNCs-DOX were much heavier than free

DOX. All curves showed similar trends that as the amount of materials incubated with cells increased, the cytotoxicity increased and the viability of cells decreased, which is inhibitor-

110

response characteristic. However, compared with targeted FA-SJNCs-DOX incubated with cells in HBSS buffer without folic acid, the cytotoxicity curve for the targeted FA-SJNCs-DOX in free folic acid competition experiment showed less cytotoxicity when treated with the same dosage.

The same trend was observed when non-targeted SJNCs-DOX were incubated with cancer cells.

The viability of the non-targeted group was much higher than the targeted group. To further analyze the cytotoxicity statistically, the IC50 values of three groups were calculated and displayed in Table 5.1. As shown in Figure 5.11, the IC 50 value of targeted FA-SJNCs-DOX conjugates is calculated at 255.3 ± 55.1 ug ml-1. This value is significantly lower than that of the non-targeted group (IC50 = 781.2 ± 163.0 ug ml-1, p < 0.001) and the free FA competition group

(IC50 = 1030.2 ± 416.1 ug ml-1, p < 0.05). These results clearly indicate the cell targeting effects of the dual-functionalized SJNCs.

DOX FA-SJNCs- FA-SJNCs-DOX- SJNCs-DOX

DOX FA Competition

IC 50 (ug ml-1) 3.3 ± 0.3 255.3 ± 55.1 781.2 ± 163.0 1030.2 ± 416.1

Table 5.1. IC 50 values of DOX, targeted FA-SJNCs-DOX, FA-SJNCs-DOX in free folic acid competition and non-targeted SJNCs-DOX against MDA-MB-231 cells.20

Compared to the non-targeted and the free FA competition groups, much more targeted

FA-SJNCs-DOX conjugates were internalized by tumor cells through the folate-receptor- mediated endocytosis. Once the conjugates were in the endocytic compartments, FA-SJNCs-

111

DOX conjugates were able to release DOX at a much faster rate, thus significantly enhance the efficacy of DOX and cause cell death. The highly selective targeting of FA-SJNCs-DOX conjugates could also reduce cardiac side effects of DOX in cancer therapy. The IC50 value for

DOX on MDA-MB-231 cell line is 3.3 ± 0.3 ug ml-1. 1.28 wt. % of DOX was found to be released from the FA-SJNCs-DOX conjugates in the targeted group.

Figure 5.11. Comparsion of IC 50 values from different experimental groups. *p < 0.05, ***p <

0.001.20

To ensure that the presence of free FA alone did not significantly affect the cell viability in the competition group, another control was conducted. MDA-MB-231 cells were incubated

112

with DOX in the presence of 1mM free FA. As can be seen in Figure 5.12, the two cytotoxicity assay curves for DOX and DOX + 1 mM FA are superimposed. The IC50 value for DOX + 1 mM FA is 4.4 ± 1.1 µg ml-1, which is not significantly different from that of DOX (p = 0.084).

This indicates that free FA alone had little effect on the viabilities of MDA-MB-231 cells.

Figure 5.12. Cytotoxicity profiles of DOX and DOX in the presence of 1 mM free FA against

MDA-MB-231 cell line. The connecting lines are constructed from non-linear dose-response fit.

To further understand the internalization behaviors of the conjugates into the cancer cells, free DOX, FA-SJNCs-DOX and SJNCs-DOX were incubated with MDA-MB-231 cells and then observed under fluorescent microscope.

113

Figure 5.13. Fluorescent microscopy images of MDA-MB-231 cells incubates with DOX, FA-

SJNCs-DOX and SJNCs-DOX for 2 h under (A) 40 x and (B) 400 x magnification. (a) DOX fluorescence (red); (b) the nuclei stained with DPAI (blue); and (c) merged image of a and b.

114

As can be seen in Figure 5.13, the fluorescence from free DOX is located only in the nuclei sites, which means that most of the DOX present diffused into the nuclei. However, in the targeted FA-SJNCs-DOX group, fluorescence comes from perinuclear regions. I suppose there is also fluorescence from nuclei, but the low intensity compared with that from the FA-SJNCs-

DOX aggregates made it less obvious. In the non-targeted SJNCs-DOX group, little fluorescence from aggregates was observed. The fluorescence pattern was very similar to cells treated with free DOX. The most majority of fluorescence came from nuclei but the intensity was much lower.

So it is reasonable to hypothesize that targeted FA-SJNCs-DOX were able to be internalized by the cancer cells and aggregate inside the cells, which generated bright fluorescent areas in perinuclear regions. On the other hand, very few of the non-targeted SJNCs-DOX were internalized by the cells. Only the DOX released in the media outside the cells were able to diffuse into the nuclei. The amount of DOX released at neutral pH was much less compared to free DOX, which made the fluorescence intensity a lot lower. This hypothesis is in agreement with the results from cytotoxicity assays.

5.4. Conclusion

The supermagnetic Janus nanocomposites of polystyrene/Fe3O4@silica were individually functionalized with FA and DOX. The release behaviors of the conjugates exhibited obvious pH- dependence. The cytotoxicity profiles of the conjugates against MDA-MB-231 breast cancer

115

cells confirmed the targeted and stimulus-induced drug delivery. This novel Janus nanostructure with dual functional groups presents high possibilities in clinical cancer diagnosis and treatment.

5.5. References

1. R. J. Lee and P. S. Low, "Folate as a targeting device for proteins utilizing folate receptor-mediated endocytosis," Methods Mol Med, 25 69-76 (2000). 2. B. Stella, S. Arpicco, M. T. Peracchia, D. Desmaele, J. Hoebeke, M. Renoir, J. D'Angelo, L. Cattel, andP. Couvreur, "Design of folic acid-conjugated nanoparticles for drug targeting," J Pharm Sci-Us, 89[11] 1452-64 (2000). 3. C. P. Leamon and P. S. Low, "Delivery of Macromolecules into Living Cells - a Method That Exploits Folate Receptor Endocytosis," P Natl Acad Sci USA, 88[13] 5572-76 (1991). 4. A. C. Antony, "The Biological Chemistry of Folate Receptors," Blood, 79[11] 2807-20 (1992). 5. K. G. Rothberg, Y. S. Ying, J. F. Kolhouse, B. A. Kamen, andR. G. W. Anderson, "The Glycophospholipid- Linked Folate Receptor Internalizes Folate without Entering the Clathrin-Coated Pit Endocytic Pathway," J Cell Biol, 110[3] 637-49 (1990). 6. J. F. Ross, P. K. Chaudhuri, andM. Ratnam, "Differential Regulation of Folate Receptor Isoforms in Normal and Malignant-Tissues in-Vivo and in Established Cell-Lines - Physiological and Clinical Implications," Cancer, 73[9] 2432-43 (1994). 7. M. R. Buist, C. F. M. Molthoff, P. Kenemans, andC. J. L. M. Meijer, "Distribution of Ov-Tl-3 and Mov18 in Normal and Malignant Ovarian Tissue," J Clin Pathol, 48[7] 631-36 (1995). 8. P. Garinchesa, I. Campbell, P. E. Saigo, J. L. Lewis, L. J. Old, andW. J. Rettig, "Trophoblast and Ovarian- Cancer Antigen-Lk26 - Sensitivity and Specificity in Immunopathology and Molecular- Identification as a Folate-Binding Protein," Am J Pathol, 142[2] 557-67 (1993). 9. M. D. Salazar and M. Ratnam, "The folate receptor: What does it promise in tissue-targeted therapeutics?," Cancer Metast Rev, 26[1] 141-52 (2007). 10. K. Kono, M. J. Liu, andJ. M. J. Frechet, "Design of dendritic macromolecules containing folate or methotrexate residues," Bioconjugate Chem, 10[6] 1115-21 (1999). 11. X. N. Fei, Y. Liu, andC. Li, "Folate Conjugated Chitosan Grafted Thiazole Orange Derivative with High Targeting for Early Breast Cancer Cells Diagnosis," J Fluoresc, 22[6] 1555-61 (2012). 12. A. J. Shen, D. L. Li, X. J. Cai, C. Y. Dong, H. Q. Dong, H. Y. Wen, G. H. Dai, P. J. Wang, andY. Y. Li, "Multifunctional nanocomposite based on graphene oxide for in vitro hepatocarcinoma diagnosis and treatment," J Biomed Mater Res A, 100A[9] 2499-506 (2012). 13. S. Mahajan, V. Koul, V. Choudhary, G. Shishodia, andA. C. Bharti, "Preparation and in vitro evaluation of folate-receptor-targeted SPION-polymer micelle hybrids for MRI contrast enhancement in cancer imaging," Nanotechnology, 24[1] 015603 (2013). 14. J. Hou, Q. Zhang, X. Li, Y. Tang, M. R. Cao, F. Bai, Q. Shi, C. H. Yang, D. L. Kong, andG. Bai, "Synthesis of

116

novel folate conjugated fluorescent nanoparticles for tumor imaging," J Biomed Mater Res A, 99A[4] 684-89 (2011). 15. M. Barz, F. Canal, K. Koynov, R. Zentel, andM. J. Vicent, "Synthesis and In Vitro Evaluation of Defined HPMA Folate Conjugates: Influence of Aggregation on Folate Receptor (FR) Mediated Cellular Uptake," Biomacromolecules, 11[9] 2274-82 (2010). 16. J. M. Saul, A. Annapragada, J. V. Natarajan, andR. V. Bellamkonda, "Controlled targeting of liposomal doxorubicin via the folate receptor in vitro," J Control Release, 92[1-2] 49-67 (2003). 17. S. Wang, R. J. Lee, C. J. Mathias, M. A. Green, andP. S. Low, "Synthesis, purification, and tumor cell uptake of 67Ga-deferoxamine--folate, a potential radiopharmaceutical for tumor imaging," Bioconjug Chem, 7[1] 56-62 (1996). 18. N. C. Fan, F. Y. Cheng, J. A. A. Ho, andC. S. Yeh, "Photocontrolled Targeted Drug Delivery: Photocaged Biologically Active Folic Acid as a Light-Responsive Tumor-Targeting Molecule," Angew Chem Int Edit, 51[35] 8806-10 (2012). 19. P. S. Low, W. A. Henne, andD. D. Doorneweerd, "Discovery and development of folic-acid-based receptor targeting for Imaging and therapy of cancer and inflammatory diseases," Accounts Chem Res, 41[1] 120-29 (2008). 1. R. J. Lee and P. S. Low, "Folate as a targeting device for proteins utilizing folate receptor-mediated endocytosis," Methods Mol Med, 25 69-76 (2000). 2. B. Stella, S. Arpicco, M. T. Peracchia, D. Desmaele, J. Hoebeke, M. Renoir, J. D'Angelo, L. Cattel, andP. Couvreur, "Design of folic acid-conjugated nanoparticles for drug targeting," J Pharm Sci-Us, 89[11] 1452-64 (2000). 3. C. P. Leamon and P. S. Low, "Delivery of Macromolecules into Living Cells - a Method That Exploits Folate Receptor Endocytosis," P Natl Acad Sci USA, 88[13] 5572-76 (1991). 4. A. C. Antony, "The Biological Chemistry of Folate Receptors," Blood, 79[11] 2807-20 (1992). 5. K. G. Rothberg, Y. S. Ying, J. F. Kolhouse, B. A. Kamen, andR. G. W. Anderson, "The Glycophospholipid-Linked Folate Receptor Internalizes Folate without Entering the Clathrin-Coated Pit Endocytic Pathway," J Cell Biol, 110[3] 637-49 (1990). 6. J. F. Ross, P. K. Chaudhuri, andM. Ratnam, "Differential Regulation of Folate Receptor Isoforms in Normal and Malignant-Tissues in-Vivo and in Established Cell-Lines - Physiological and Clinical Implications," Cancer, 73[9] 2432-43 (1994). 7. M. R. Buist, C. F. M. Molthoff, P. Kenemans, andC. J. L. M. Meijer, "Distribution of Ov-Tl-3 and Mov18 in Normal and Malignant Ovarian Tissue," J Clin Pathol, 48[7] 631-36 (1995). 8. P. Garinchesa, I. Campbell, P. E. Saigo, J. L. Lewis, L. J. Old, andW. J. Rettig, "Trophoblast and Ovarian-Cancer Antigen-Lk26 - Sensitivity and Specificity in Immunopathology and Molecular-Identification as a Folate- Binding Protein," Am J Pathol, 142[2] 557-67 (1993). 9. M. D. Salazar and M. Ratnam, "The folate receptor: What does it promise in tissue-targeted therapeutics?," Cancer Metast Rev, 26[1] 141-52 (2007). 10. K. Kono, M. J. Liu, andJ. M. J. Frechet, "Design of dendritic macromolecules containing folate or methotrexate residues," Bioconjugate Chem, 10[6] 1115-21 (1999). 11. X. N. Fei, Y. Liu, andC. Li, "Folate Conjugated Chitosan Grafted Thiazole Orange Derivative with High Targeting for Early Breast Cancer Cells Diagnosis," J Fluoresc, 22[6] 1555-61 (2012). 12. A. J. Shen, D. L. Li, X. J. Cai, C. Y. Dong, H. Q. Dong, H. Y. Wen, G. H. Dai, P. J. Wang, andY. Y. Li, "Multifunctional nanocomposite based on graphene oxide for in vitro hepatocarcinoma diagnosis and treatment," J Biomed Mater Res A, 100A[9] 2499-506 (2012). 13. S. Mahajan, V. Koul, V. Choudhary, G. Shishodia, andA. C. Bharti, "Preparation and in vitro evaluation of folate- receptor-targeted SPION-polymer micelle hybrids for MRI contrast enhancement in cancer imaging,"

117

Nanotechnology, 24[1] 015603 (2013). 14. J. Hou, Q. Zhang, X. Li, Y. Tang, M. R. Cao, F. Bai, Q. Shi, C. H. Yang, D. L. Kong, andG. Bai, "Synthesis of novel folate conjugated fluorescent nanoparticles for tumor imaging," J Biomed Mater Res A, 99A[4] 684-89 (2011). 15. M. Barz, F. Canal, K. Koynov, R. Zentel, andM. J. Vicent, "Synthesis and In Vitro Evaluation of Defined HPMA Folate Conjugates: Influence of Aggregation on Folate Receptor (FR) Mediated Cellular Uptake," Biomacromolecules, 11[9] 2274-82 (2010). 16. J. M. Saul, A. Annapragada, J. V. Natarajan, andR. V. Bellamkonda, "Controlled targeting of liposomal doxorubicin via the folate receptor in vitro," J Control Release, 92[1-2] 49-67 (2003). 17. S. Wang, R. J. Lee, C. J. Mathias, M. A. Green, andP. S. Low, "Synthesis, purification, and tumor cell uptake of 67Ga-deferoxamine--folate, a potential radiopharmaceutical for tumor imaging," Bioconjug Chem, 7[1] 56- 62 (1996). 18. N. C. Fan, F. Y. Cheng, J. A. A. Ho, andC. S. Yeh, "Photocontrolled Targeted Drug Delivery: Photocaged Biologically Active Folic Acid as a Light-Responsive Tumor-Targeting Molecule," Angew Chem Int Edit, 51[35] 8806-10 (2012). 19. P. S. Low, W. A. Henne, andD. D. Doorneweerd, "Discovery and development of folic-acid-based receptor targeting for Imaging and therapy of cancer and inflammatory diseases," Accounts Chem Res, 41[1] 120-29 (2008). 20. F. Wang, G. M. Pauletti, J. Wang, J. Zhang, R. C. Ewing, Y. Wang, andD. Shi, "Dual surface-functionalized Janus nanocomposites of polystyrene/Fe(3)O(4)@SiO(2) for simultaneous tumor cell targeting and stimulus- induced drug release," Adv Mater, 25[25] 3485-9 (2013). 21. D. Cheng, G. B. Hong, W. W. Wang, R. X. Yuan, H. Ai, J. Shen, B. L. Liang, J. M. Gao, andX. T. Shuai, "Nonclustered magnetite nanoparticle encapsulated biodegradable polymeric micelles with enhanced properties for in vivo tumor imaging," J Mater Chem, 21[13] 4796-804 (2011). 22. Q. D. Hu, H. Fan, Y. Ping, W. Q. Liang, G. P. Tang, andJ. Li, "Cationic supramolecular nanoparticles for co-delivery of gene and anticancer drug," Chem Commun, 47[19] 5572-74 (2011).

118

Chapter 6. Summary and Outlook

In Summary, a unique Janus assembly comprised of polystyrene/Fe3O4@silica was developed. The mechanism of the reaction was studied and phase separation was found to play an important role in forming the Janus structure. Our experimental data demonstrate that this versatile system possesses dual surfaces for effective conjugation of functionally different chemical moieties. Surface immobilization of FA targeting ligands on the carboxyl- functionalized PS matrix effectively enhances tumor-selective targeting and internalization.

Attachment of DOX via a pH-sensitive hydrazone linker to the silica surface enables controlled, stimulus-induced release of the cytotoxic drug after internalization under acidic conditions in endosomal compartments. Silica-embedded Fe3O4 nanoparticles render this Janus nanostructure superparamagnetic suitable for multimodal imaging and hyperthermia-induced sensitization of tumor cells. Consequently, this novel nanoassembly with dual surface functionalities offers an innovative drug delivery platform for multidimensional cancer treatment.

However, to further understand about the Janus structure and explore its possibilities for bio-applications, more researches are needed in the following areas:

1. Size is still considered a limitation factor for effective cellular uptake and it is our hope to reduce the size of SJNCs to below 150 nm by modifying the ratio of St monomers over. As the number of oil phase droplets in the miniemulsion system is decided by the concentration of SDS below its critical micelle concentration, reducing the amount of St monomers will result in less

St in each droplets. Theoretically the PS modules would be smaller. We have tried with less

119

TEOS and observed thinner shells. Another way is to increase the amount of SDS used in the reaction to create more droplets.

2. Investigate the force that binds the silica shell to the St core and stability of the linkage.

Currently the reason that silica attaches to the St core and only forms a half shell is not clear.

3. Replace DOX with other fluorescent probes to monitor the internalization of SJNCs and targeting effect. DOX is released and diffuse in the cells, so it’s ambiguous to locate the nanocomposites in the cells upon the fluorescence from DOX.

4. Investigate the effects of geometry on drug loading and cellular internalization. The novelty of this composite comes from its Janus core/half-shell structure. No study has shown how this structure would affect the drug loading and cellular uptake behaviors. We would consider computational simulation as a powerful tool for this study.

5. Silica surfaces provide abundant of opportunities for further functionalization. Other biological molecules or conjugating mechanism could be considered. For example, silanes with thiol ending group could be used to modify the silica shell, which could link to molecules also bearing thiol groups and form a disulfide bond.

120