(Para)magnetic solid lipid nanocarriers for early detection and treatment of solid tumours

Universidade do Minho Escola de Ciências Cátia Andreia Vieira Rocha October 2019 (Para)magnetic solid lipid nanocarriers for early detection and treatment of solid tumours Cátia Andreia Vieira Rocha (Para)magnetic solid lipid nanocarriers for early detection and treatment of solid tumours UMinho|2019 Universidade do Minho Escola de Ciências Cátia Andreia Vieira Rocha (Para)magnetic solid lipid nanocarriers for early detection and treatment of solid tumours Master thesis Master in Biophysics and Bionanosystems Work developed under the supervision of Doctor Juan Gallo Páramo And Professor Doctor Paulo José Gomes Coutinho October 2019 Cátia Andreia Vieira Rocha (Para)magnetic solid lipid nanocarriers for early detectio n and treatment of solid Master Thesis Master in Biophysics and Bionanosystems Work developed under the supervision of Doctor Juan Gallo Páramo and Professor Doctor Paulo José Gomes Coutinho October 2019 ii DIREITOS DE AUTOR E CONDIÇÕES DE UTILIZAÇÃO DO TRABALHO POR TERCEIROS Este é um trabalho académico que pode ser utilizado por terceiros desde que respeitadas as regras e boas práticas internacionalmente aceites, no que concerne aos direitos de autor e direitos conexos. Assim, o presente trabalho pode ser utilizado nos termos previstos na licença abaixo indicada. Caso o utilizador necessite de permissão para poder fazer um uso do trabalho em condições não previstas no licenciamento indicado, deverá contactar o autor, através do RepositóriUM da Universidade do Minho. Licença concedida aos utilizadores deste trabalho Atribuição-NãoComercial-SemDerivações CC BY-NC-ND https://creativecommons.org/licenses/by-nc-nd/4.0/ iii ACKNOWLEDGEMENTS During the development of this thesis all the challenges and objectives would not have been overcome without the unconditional support of everyone involved. I’d like to express my utmost appreciation to my main supervisor, Dr. Juan Gallo Páramo, for its patient guidance, support, availability and constructive suggestions throughout the whole process of developing this dissertation. I’d like to express my gratitude to Dr. Manuel Bañobre for his warm welcome into the AmTheNa group, his confidence, patience and willingness to teach me more in every step of the way. I’d like to thank Dr. Paulo Coutinho for his teachings and his guidance along the journey that led me to finish this thesis. I express my profound gratitude to my parents, because without them none of my academic accomplishments would be possible. For all their support, patience, confidence and continuous motivation throughout the six years of studies that led me here. I’d like to thank my brother Fábio for always helping me to relax and take my mind off stress. I’d like to express my gratitude to my boyfriend, Tadeu, for always being there for me, proportionating fun moments, listening, calming me and helping me throughout every step of this journey. I’d also like to thank my high school friends and my cousin for always being present and supporting me. I have to thank my master colleagues and most recent friends for their help, animated talks and friendship. I also show my appreciation to my dear friends from my bachelor, for comforting me even at long distances. I’d like to express my gratitude to Dr. Milene, Dr. Lorena, Dr, Stefania, Dr. Marta, Nuria and all the AmTheNa members for the sharing of ideas, good conversations and always being available to help me whenever I had difficulties. I’d also like to acknowledge everyone that was involved throughout this whole process, including the institution that accepted me, namely the group of Nanomedicine of INL, that provided all the possible resources needed to conclude this dissertation. iv STATEMENT OF INTEGRITY I hereby declare having conducted this academic work with integrity. I confirm that I have not used plagiarism or any form of undue use of information or falsification of results along the process leading to its elaboration. I further declare that I have fully acknowledged the Code of Ethical Conduct of the University of Minho. v Título: Nanopartículas (para)magnéticas para deteção precoce e tratamento de tumores sólidos. RESUMO Estratégias promissoras de entrega fármacos são cada vez mais relevantes para o tratamento de uma ampla variedade de doenças, como tumores sólidos. A entrega de fármacos é vista como uma solução inovadora para superar desafios atuais na medicina, como encontrar o equilíbrio perfeito entre ganho terapêutico e efeitos adversos induzidos pela terapia. Nanocompósitos híbridos (orgânicos/inorgânicos) têm vindo a ser explorados, de modo a combinar sinergicamente diagnóstico e terapia em formulações clinicamente eficazes. Neste trabalho foi desenvolvido um veículo de entrega de fármacos externamente controlado, com atividade anti-tumoral aprimorada. Nanotransportadores lipídicos sólidos biocompatíveis combinados com nanopartículas magnéticas, responsivas ou não responsivas, e doxorrubicina foram sintetizados e caracterizados físico-química e funcionalmente. A eficiência de encapsulação da doxorrubicina foi de quase 100 %. As nanopartículas obtidas são esféricas e seu tamanho varia entre 150 e 200 nm, sendo ideal para administração. O nanocompósito apresentou um comportamento misto, superparamagnético e paramagnético, ideal para obter contraste duplo no diagnóstico por ressonância magnética. A inclusão de magnetita no veículo de entrega faz com que este seja adequado para tratamentos termorregulados, devido ao seu grande potencial de aquecimento quando exposto a um campo magnético alternado. Portanto, a libertação de doxorrubicina foi maior quando a formulação foi exposta a hipertermia magnética. Os efeitos das nanopartículas lipídicas em células de cancro da mama triplo negativo (Hs578T) foram avaliados in vitro . Ensaios de toxicidade confirmaram a biocompatibilidade do veículo e das nanopartículas magnéticas encapsuladas, para concentrações clinicamente relevantes de iões de ferro e manganês. Células tratadas com nanotransportadores lipídicos sólidos e expostas a hipertermia (quimioterapia/hipertermia), mostraram um maior efeito anti-tumoral, confirmando o potencial das partículas comparado a tratamentos tradicionais. Os sistemas de entrega desenvolvidos representam avanços no teranóstico personalizado. Os resultados obtidos são promissores para o uso destes nanocompósitos no diagnóstico precoce, monitorização de doenças e tratamento sinergético de tumores sólidos, atingindo os objetivos de toxicidade sistémica reduzida, agente de contraste duplo, eficácia anti-tumoral aprimorada e efeitos adversos diminuídos na terapia do cancro. Palavras-chave: Hipertermia magnética, Nanopartículas magnéticas, Nanotransportadores lipídicos sólidos, Ressonância magnética, Teranóstico. vi Title: (Para)magnetic nanoparticles for early detection and treatment of solid tumours. ABSTRACT Nowadays, promising nano drug delivery strategies are becoming increasingly relevant to treat a wide range of diseases, such as solid malignant tumours. Drug delivery is sought as an innovative solution to overcome medical difficulties, as finding the perfect equilibrium between therapeutic gain and therapy induced adverse effects. Hybrid organic/inorganic nanocomposites are being explored to synergistically combine diagnostic and therapeutic capabilities into clinically effective formulations. Herein, an externally controlled theranostic drug delivery vehicle with enhanced antitumoral activity is developed. Biocompatible solid lipid nanocarriers combined with responsive or non-responsive magnetic nanoparticles and doxorubicin were synthetised and physico-chemically and functionally characterised. The nanoparticles are spherical, and their size ranged from 150 to 200 nm, ideal for administration. Doxorubicin presented an encapsulation efficiency of almost 100 %. The nanocomposite showed mixed superparamagnetic and paramagnetic behaviour, ideal to achieve a dual contrast in magnetic resonance imaging diagnosis. Moreover, the inclusion of magnetite into the delivery vehicle makes it suitable for thermo-regulated treatments, due to its great heating potential when exposed to an alternating magnetic field. Thus, doxorubicin release was more pronounced when exposed to hyperthermia treatment. An in vitro evaluation of the lipid nanoparticles effects was performed in triple negative breast cancer cells (Hs578T). Toxicity assays confirmed the biocompatibility of the vehicle and the encapsulated magnetic nanoparticles at clinically relevant concentrations of iron and manganese ions. By combining solid lipid nanocarriers loaded with doxorubicin and magnetic field exposure (chemotherapy/hyperthermia) in vitro , a better anti-tumoral effect was achieved demonstrating the potential of these particles to enhance the effect of traditional treatments. The designed delivery systems represent a step forward towards the final goal of personalised theranostics. The results attained are promising for the use of the presented nanocomposites in early diagnosis, disease monitoring and synergetic treatment of solid tumours. This may realize long sought after goals of reduced systemic toxicity, dual contrast agent efficiency, enhanced antitumoral efficacy and diminished adverse effects in cancer therapy. Keywords: Magnetic hyperthermia, Magnetic nanoparticles, Magnetic resonance imaging, Solid lipid nanocarriers, Theranostic. vii MOTIVATION The global cancer burden is estimated to have risen to 18.1 million new cases and 9,6 million deaths in 2018. In this same year, more than 58000 people were diagnosed with cancer in Portugal, being that half of those cases have led to death [1]. The vast majority of these cancers appear as solid tumours. Current treatment protocols (surgery, chemo-, radioand immunotherapy) have proved their utility in the clinic, but they are still far from being the solution to this problem. In this project an innovative tool against solid tumours through a combination of early diagnosis and treatment (theranosis) will be developed. The International Iberian Nanotechnology Laboratory (INL) has recently developed a multifunctional nanocomposite including magnetic nanoparticles dispersed in a lipid matrix [2]. Both the magnetic and lipid components of this system combine their properties in a synergic manner to provide a powerful weapon in the fight against cancer. The lipid part of the system can be used to encapsulate chemotherapeutic agents. This encapsulation has a double effect on the outcome of the treatment, on one hand increases the local concentration of the drug in the cells, and on the other protects healthy tissues from the deleterious effect of the drug, reducing then side effects. The magnetic part brings also interesting capabilities to the system. Magnetic nanoparticles (MNPs) have been used in the clinic since the late 1990’s as magnetic resonance imaging (MRI) contrast agents (CAs) and thus can be used for the non-invasive detection of tumours [3]. MNPs can as well be used to generate heat under alternating magnetic fields in a process called magnetic hyperthermia (MHT). This heat can be used for the direct ablation of tumours through hyperthermia, and/or as a stimulus for the controlled delivery of drugs [4]. Paramagnetic nanostructures have also been developed at INL and can be used as MRI CAs and at the same time as O2 generators to remediate tumour hypoxia and improve current treatments [5], [6]. Overall, this study will consist in the creation of a new nanocomposite, followed by its physicochemical and functional characterisation, focusing on its performance as a MRI contrast agent, magnetic hyperthermia effector and controlled drug delivery system. This can be an important tool in nanomedicine to help in the fight against cancer. viii TABLE OF CONTENTS ACKNOWLEDGEMENTS ............................................................................................................ III RESUMO ................................................................................................................................... V ABSTRACT ............................................................................................................................... VI MOTIVATION ........................................................................................................................... VII LIST OF FIGURES ..................................................................................................................... XI LIST OF TABLES ...................................................................................................................... XV LIST OF ABBREVIATIONS AND ACRONYMS ............................................................................. XVI CHAPTER 1 - INTRODUCTION................................................................................................... 1 1.1. THE ISSUE OF CANCER ................................................................................................... 1 1.1.1. Etiology and incidence ....................................................................................... 1 1.1.2. Tumour biology .................................................................................................. 2 1.1.3. Classic and emerging therapies ......................................................................... 3 1.2. NANOMEDICINE IN CANCER.............................................................................................. 4 1.3. (PARA)MAGNETIC NANOPARTICLES .................................................................................... 6 1.3.1. Iron oxide nanoparticles ..................................................................................... 7 1.3.2. Manganese nanoparticles .................................................................................. 9 1.3.3. Synthesis Methods ........................................................................................... 11 1.4. BIOMEDICAL APPLICATIONS ........................................................................................... 13 1.4.1. Drug Delivery ................................................................................................... 13 1.4.2. Magnetic Resonance Imaging ........................................................................... 16 1.4.3. Magnetic Hyperthermia .................................................................................... 19 1.5. OBJECTIVES ............................................................................................................... 21 CHAPTER 2 - CHARACTERISATION TECHNIQUES ................................................................... 23 2.1. DYNAMIC LIGHT SCATTERING .......................................................................................... 23 2.2. SPECTROSCOPY .......................................................................................................... 25 2.2.1. Ultraviolet-Visible absorption spectroscopy ........................................................ 25 2.2.2. Fluorescence spectroscopy .............................................................................. 27 2.2.3. Fourier Transform Infra-red spectroscopy ......................................................... 27 2.2.4. Inductively coupled plasma-atomic emission spectroscopy ................................ 29 2.3. MICROSCOPY ............................................................................................................. 30 2.3.1. Optical microscopy .......................................................................................... 31 2.3.2. Confocal microscopy ........................................................................................ 31 2.3.3. Transmission electron microscopy ................................................................... 32 2.3.3.1. STEM-EDX mode ...................................................................................... 34 2.4. THERMOGRAVIMETRIC ANALYSIS ...................................................................................... 35 2.5. MAGNETIC MEASUREMENTS ........................................................................................... 36 2.5.1. Vibrating sample magnetometer ....................................................................... 36 2.5.2. Superconducting quantum interference device ................................................. 37 2.6. X-RAY DIFFRACTION ...................................................................................................... 38 xv LIST OF TABLES Table 1 - Physico-chemical characteristics of doxorubicin [84], [85]. ........................................ 13 Table 2 - List of all equipment, materials, software and reagents used throughout the course of this project. ................................................................................................................................... 45 Table 3 - Percentage of dox and MNPs present in the SLNs. .................................................... 49 Table 4 - Parameters used to measure the size and zeta potential of the NPs. ......................... 50 Table 5 - Element concentrations used for the MRI studies for each formulation. ...................... 54 Table 6 - MRI specifications for T1 and T2 maps acquisition. ..................................................... 55 Table 7 - Concentration of ions for each SLN formulation......................................................... 62 Table 8 - DLS and ELS analysis results for each SLN formulation. ............................................ 70 Table 9 - Magnetic properties of each particle.......................................................................... 73 Table 10 - r1,2 and r2/r1 ratio for the different SLN formulations. ................................................ 82 Table 11 - r1,2 and r2/r1 ratio for the responsive SLN formulations in water and hydrogen peroxide. .............................................................................................................................................. 84 Table 12 - Concentrations of magnetic NPs and dox, used to treat the cells for MHT experiments. .............................................................................................................................................. 88 xvi LIST OF ABBREVIATIONS AND ACRONYMS 2D – Two dimensional; 3D – Tri dimensional; ADF – Annular dark-field; AMF - Alternating magnetic field; ATR - Attenuated total reflectance; BBB – Blood Brain Barrier; BF – Bright field; BOLD – Blood oxygen level dependency; BSE - Back-scattered electrons; CA – Contrast agent; CCDCharge couple device; DAPI - 4',6'-diamino-2-fenil-indol; DDS – Drug delivery system; DHA – Docosahexaenoic acid; DiO - 3,3'-Dioctadecyloxacarbocyanine Perchlorate; DLS – Dynamic light scattering; DMEM - Dulbecco’s modified Eagle’s medium; DMEM – Dulbecco’s modified eagle’s medium; DNA - Deoxyribonucleic acid; dox – Doxorubicin; DSC - Differential scanning calorimetry; EDX - Energy dispersive X-ray spectroscopy; EE – Encapsulation efficiency; ELS - Electrophoretic Light Scattering; EMA - European Medicines Agency; EPR – Enhanced permeability and retention; FBS - Fetal bovine serum; FC – Field cooling; xvii FDA – Food and Drug Administration; FIR - Far infrared; fMRI – Functional magnetic resonance imaging; FOVField of view; FT-IR – Fourier Transform Infrared spectroscopy; Glc – glucose; Hc – Coercivity; HPLC - High performance liquid chromatography; Hs578T - Triple negative breast carcinoma cell line; ICPAES – Inductively coupled plasma - atomic emission spectroscopy; INL - International Iberian Nanotechnology Laboratory; IR – Infrared or Inversion recovery; LDL - Low-density lipoprotein; LED – light-emitting diode; logD - Distribution coefficient; logP - Partition coefficient; M – Magnetization; MCT - Mercury-Cadmium-Telluride; MDR – Multi-drug resistance; MEMS - multi-echo multi-slice; MHT – Magnetic hyperthermia; MIR - middle infrared; MNP - Magnetic nanoparticle; MPRAGE - Magnetization prepared rapid gradient echo; Mr – Remanent magnetization; MRI – Magnetic resonance imaging; MS– Saturation magnetization; mSLNs - magnetic solid lipid nanocarriers; NA – Numerical aperture; NC – Nanocarrier; xviii NLC – Nanostructured Lipid Carrier; NMR - Nuclear magnetic resonance; NP - Nanoparticle; OA – Oleic acid; PBS - Phospate-buffered saline; PD – Proton density; PEG - Poly(ethylene) glycol; PI - Polydispersity Index; QD – Quantum Dot; R1 - Longitudinal relaxation rate; r1 - Longitudinal relaxivity; R2 - Transverse relaxation rate; r2 - Transverse relaxivity; RF - Radiofrequency; ROS – Reactive Oxygen Species; rpm – rotations per minute; SAR – Specific absorption rate; SE – Spin echo; SEM – Scanning electron microscopy or Standard error of the mean; SLN – Solid Lipid Nanocarrier; SPION - Superparamagnetic iron oxide nanoparticle; SQUID - Superconducting quantum interference device; STEM - Scanning transmission electron microscopy; T1 - Longitudinal relaxation time; T2 - Transverse relaxation time; TB – Blocking temperature; TE – Echo time; TEM –Transmission electron microscopy; TGA – Thermogravimetric analysis; TI – Inversion time; xix TR – Repetition time; UV-Vis – Ultraviolet visible; VEGF - Vascular endothelial growth factor; VSM - Vibrating-sample magnetometer or vibrating sample measurement; WHO - World Health Organization; XRD – X-ray diffraction; ZFC – Zero field cooling; ζ - potential - Zeta potential 1 CHAPTER 1 - INTRODUCTION 1.1. The issue of cancer 1.1.1. Etiology and incidence Cancer is a disease that is part of our current reality and has been a challenge to researchers and medical teams to find new strategies to overcome it. This is a problem with no prospects of having a longterm resolution, being therefore a disease of the present and the future. This disease is characterised by a rapid and uncontrolled division of abnormal cells, which can grow beyond their usual boundaries and spread to other parts of the body (metastasis) [7]. Cancer is caused by genetic changes, and in some cases they may be hereditary. There are several other etiologic factors that may lead to cancer, such as age, smoking, alcohol, diet, exposure to chemical agents and radiation, hormones and infections by viruses and bacteria [8]. Hence, the prevention of this disease can be possible to a certain extent through changes in diet and lifestyle, regular screening, chemoprevention and risk-reducing surgery [9]. Cancer is the second leading cause of death worldwide and its incidence has been increasing annually, along with the average age of the affected population. In 2018, there were an estimated 18.1 million new cases of cancer in the world [1]. Figure 1 shows the estimated age-standardized incidence rates worldwide, for all cancers in both sexes in 2018 [10]. In 2018, the World Health Organization (WHO) reported 9,6 million deaths due to cancer. In this same year, more than 58000 people were diagnosed with cancer in Portugal, being that half of those Figure 1Global age-standardized incidence of all cancers in both sexes in 2018 [10]. 2 cases have led to death [1]. The vast majority of these cancers appear as solid tumours such as lung, breast, prostate and colorectal tumours [11]. With the introduction of screening and the increased flow of information on cancer, it was possible to observe a gradual decrease in early and standardized mortality rates in Portugal. However, we continue to see an increase in the gross mortality rate, especially in males [12]. 1.1.2. Tumour biology Cancer is a process with several steps, by which cells go through metabolic and structural changes. These alterations have a great impact on the mechanisms that control cell proliferation, as the same time as they increase the resistance of cancer cells to the host’s immune system. The changes occur at the DNA (deoxyribonucleic acid) level, which contains the genetic information responsible for the synthesis of proteins. DNA molecules are modified at the level of their nucleotidic sequence (mutations) and at their chromatin structural organization. Modifications in the nucleotidic sequence lead to a defective gene reading by transcriptional mechanisms, resulting in the synthesis of proteins in disproportionate amounts or with a wrong amino acid structure. Alterations in the chromatin conformation result in difficulties in the process of DNA packaging in the nucleus, leading to malfunctions in DNA transcription, translation, replication and repair mechanisms [13]. The process of carcinogenesis can be defined in at least 3 steps: initiation, promotion and progression (figure 2) [14]. In the initiation phase the mutagenic agent, endogenous or exogenous, will interact with the DNA and induce physical, chemical or biological changes in the genes. These changes result in a deregulation of cell signalling mechanisms and in their survival and differentiation. At the end of this phase the cancer is not yet clinically detectable and may be reversible. However, lesions caused in proto-oncogenes and tumour suppressor genes will contribute to the development and evolution of the disease. During the promotion, mutations in tumour suppressor genes, which control replication through self-destruction programs such as RB1 (retinoblastoma gene) and TP53 (produces p53 protein, responsible for cell cycle blockade, DNA repair and apoptosis), will lead to its inactivation and consequently to a blockade of apoptosis and uncontrolled proliferation of pre-neoplastic cells. Progression is the last stage of neoplastic transformation, characterised by genetic and phenotypic changes and accelerated tumour growth [15]. The presence of a solid tumour induces an angiogenic response by the organism’s blood vessels. Angiogenesis is, in part, responsible for the transition from a minimally invasive and poorly vascularized tumour to a highly aggressive and vascularized one. At this stage the clinical prognosis is not favourable, representing in most cases host death [16]. 3 1.1.3. Classic and emerging therapies Nowadays, we are faced with several challenges in medicine, with cancer being one of the greatest and most problematic. Most classic therapies are proving to be insufficient, existing a growing need to develop new forms of treatment that are less invasive, more specific, and at the same time capable of eradicating the disease. Cancer therapies can be used for curative or palliative purposes, since there are several types of cancer for which there is no available cure. The treatments can be systemic or localised, with the most common being chemotherapy, radiotherapy, surgery and immunotherapy [17]. These can be used alone or combined, however they are quite invasive and can be cytotoxic, which may lead to the appearance of side effects. Chemotherapeutic drugs inhibit cell division, eventually leading to cellular death, especially in cancer cells. Nonetheless, a problem with chemotherapy is that it also damages healthy host cells, being particularly cytotoxic to the immune system, thus downregulating its response to the tumour [18]. Due to many chemotherapeutic agents being delivered to the tumour through the circulatory system, poorly vascularized tumours will be more inaccessible to the drug. The blood-brain barrier also represents an obstacle for chemotherapy, leading to drug delivery issues [19]. The principle of radiotherapy is to destroy cancer cells using ionizing radiation, while sparing the surrounding normal tissues. The efficiency of the treatment relies on the irreversible damage caused to the DNA directly, or indirectly through the formation of reactive oxygen species (ROS), which leads to cell death. However, despite all the technologic advances made in this field, the radiation can still affect Figure 2 - Carcinogenesis phases (initiation, promotion and progression) and events involved in each phase [14]. 4 normal surrounding tissues, leading to adverse side effects [4]. Therefore, the importance of controlling the tumour at a cellular level arises. In spite of all the advances made in classic cancer therapies, their efficacy seems to have reached a plateau for most solid tumours. However, new emerging therapies have a high probability of building onto the classical ones and reducing their flaws and limitations. These forefront ideas include examples such as genic therapy, photodynamic therapy, hyperthermia and targeted drug delivery. In this work we focus on the last two, particularly in magnetic hyperthermia (MHT) [4], [20]–[22]. On one hand, MHT is already being tested as a complementary treatment to radiotherapy, showing great results in clinical trials [23]–[26]. On the other hand, controlled drug delivery can increase significantly the accuracy and precision of the administration of chemotherapeutic drugs [27]. Moreover, both these techniques can be enhanced through nanotechnology, the first using MNPs (which can also be useful in diagnosis), the latter using organic and inorganic nanoparticles (NPs). 1.2. Nanomedicine in cancer The term nanotechnology was introduced in 1959 by the physicist Richard Feynman, in a talk entitled “There’s Plenty of Room at the Bottom” and has since been a highly explored subject and in constant growth [28]. Nanotechnology is the manipulation of matter at an atomic and molecular scale, which allows the exploration of nanomaterials with diameters between 1 and 100 nm [29]. Materials at the nanoscale often exhibit completely different behaviours from the same materials in bulk form, in this size scale the quantum effects rule the behaviour and properties of the particles. While on bulk materials the physical properties are constant independently of their size, in the nanoworld the size of the NPs can dictate their chemical and physical properties like fluorescence, melting point, electrical conductivity, magnetic properties and chemical reactivity [30]. Therefore, the properties of the materials change as their size approaches the nanoscale and as the percentage of atoms on the surface of the material becomes significant. Cancer nanotechnology, in particular, is evolving as a new interdisciplinary research to resolve cancer-related medical and pharmaceutical complications. Since the first proposal (1970s) of a targeted drug delivery system (DDS), the treatment and diagnosis of numerous diseases using nanotechnology has improved tremendously [31], [32]. The application of nanotechnology in medicine is an emerging field known as nanomedicine, and it can range from a biological application of nanomaterials to nanoelectronic biosensors (figure 3). Thus, nanomedicine can be defined as the use of nanostructured materials in medicine that according to their structure have unique medical effects [33]. Nowadays, a 5 significant number of companies is already pursuing nanomedicine activities, with the leading one being controlled drug delivery. Nanoscale DDSs for the treatment of cancers like breast, leukemia and Kaposi sarcoma, are already on the market and the number continues to grow [33], [34]. Despite recent advances in this field, the development of delivery systems with specific targeting abilities that can diminish the side effects caused by other therapies is still one of the challenges to nanomedicine. Hence, it’s very important to continue to investigate different nanomaterials and their capabilities, with the focus on achieving effective DDSs. Nanomaterials can be applied in many areas, but have a particular importance in nanomedicine, and they can be organic or inorganic. Lipid NPs, dendrimers, hydrogels and synthetic polymers are examples of organic nanomaterials. Examples of inorganic nanomaterials comprise mesoporous silica NPs, quantum dots (QDs), metallic, magnetic, paramagnetic and superparamagnetic iron oxide NPs (SPIONs), among others [27], [35]. Figure 4 illustrates a few examples of organic and inorganic NPs [35]. Certain nanomaterials allow efficient encapsulation and/or transport of drugs, proteins and nucleic acids, and can be conjugated with different ligands, such as peptides, antibodies or aptamers, increasing the binding affinity of the NPs to their targets [36]. Nanoscale DDS using organic and inorganic NPs are emerging technologies in nanomedicine for cancer theranosis (therapy plus diagnosis). A better understanding of tumour biology and the increased availability of versatile materials (hydrogels, polymers, lipids), have led to the development of Figure 4 - Illustration of organic (top) and inorganic (bottom) NPs [35]. Figure 3 - The role of nanomedicine in cancer treatment. Magnetic Hyperthermia (Para)magnetic nanoparticles TREATMENT OF CANCER Molecular diagnosis and imaging Early detection and follow-up Controlled drug delivery Gene therapy Nanoparticles Biomarkers NANOMEDICINE Biosensors Nanocarriers Nanomaterials 12 alkaline and acidic medium [48], [80]. A drawback in coprecipitation synthesis is the fact that the control of particle size distribution is limited, producing NPs with a wide size distribution and therefore with a non-ideal magnetic behaviour for many applications. To achieve monodisperse iron oxide MNPs with the coprecipitation method, two key stages are involved: a short burst of nucleation, that happens when the concentration of species reaches critical supersaturation, and then, a subsequent slow growth of the nuclei occurs by diffusion of the solutes to the surface of the crystal. To obtain monodisperse MNPs it’s crucial that these two stages are separated [48], [81]. Recently, the use of organic additives as stabilization agents, showed great results as a strategy to achieve more stable and monodisperse MNPs [82]. Thus, the selection of a proper surfactant is also an important issue for the stabilization of such particles [48]. Since coprecipitation processes require high-pH solutions, the by-products of this method may require further purification [79]. The hydrothermal method is another simple and widely used technique, that involves the production of NPs, usually iron oxides, by crystallization from an aqueous solution of ions at a high temperature under high vapor pressure [79]. The main processes performed with this method are hydrolysis and oxidation (where ferrous salts are used), or the neutralization of mixed metal hydroxides [80]. The reaction conditions, such as solvent, temperature and time have significant effects on the product. Thus, this technique allows a better control of the particles size, shape, composition and stoichiometry, producing MNPs with good monodispersity [79]. The size is mainly controlled through the rates of nucleation and grain growth, which are dependent on the reaction temperature. At higher temperatures, nucleation can be faster than grain growth, which results in a decrease in particle size. However, if the reaction time is extended, it will favour the grain growth and the particles obtained will be bigger. Surface capping agents, such as sodium oleate or PEG, can also be added during the reaction to stabilize the NPs and reduce the risk of agglomeration [48], [83]. The sonochemical method has been extensively used to synthetize MNPs with unusual magnetic properties. It involves the use of sonochemical techniques, such as a high-energy ultra-sonication, that creates cavitations, that is, the formation and growth of nuclei, and the implosive collapse of bubbles in liquid. The implosive collapse of the bubbles can provide localised heat with a temperature of about 5000 K [79]. This way we can obtain highly monodispersed oxide NPs from an aqueous solution of ions, with reduced agglomeration. However, sonochemical methods are not fit for large scale synthesis [52]. The methods mentioned above can complement each other, to easily produce MNPs with the desired physico-chemical properties. 13 1.4. Biomedical applications 1.4.1. Drug Delivery Drug delivery is a process through which a pharmaceutical compound is incorporated into a vehicle and administered into the body to safely achieve its desired therapeutic effect, with diminished side effects. As mentioned before, the encapsulation of drugs in DDSs improves the bioavailability, pharmacokinetics (concentration versus time) and pharmacodynamics (effect versus concentration) of therapeutics, among numerous other advantages [33]. The requirements for drug delivery are known as ADMET profile, standing for absorption, distribution, metabolism, excretion and toxicity. This profile is determined by drug characteristics like lipophilicity, solubility, partition (logP) and distribution (logD) coefficients, molecular weight and median lethal dose, and it influences the performance and pharmacological activity of the encapsulated compound. Doxorubicin (dox) was the drug chosen for this work due to its fluorescence properties, that allow the tracking of the drug in cells, and lipophilic character, that makes the encapsulation into the SLNs more efficient. Doxorubicin is a cytotoxic anthracycline antibiotic approved by the FDA, being already used to treat a wide variety of cancers. Some of the physicochemical characteristics of this drug are shown in table 1 [84], [85]. In relation to its pharmacokinetics, dox is administrated intravenously since it is not absorbed by the gastrointestinal tract. It has a rapid and wide distribution (5-10 min) with a complete half-life of 12-18 h and a binding to plasma proteins from 50 to 85 % [86], [87]. Dox is mainly metabolised in the liver. It enters the cells via passive diffusion and has a rapid uptake by cells, but it does not cross the blood-brain barrier (BBB) [87]. After its metabolism 50% of the drug is already eliminated by the liver but its primarily excretion paths are biliary and faecal excretion [86]. Dox toxicity in rats was defined with a median lethal dose of 21800 µg/kg [84]. Doxorubicin has multiples mechanisms of action that lead to cell death. Some mechanisms are based on its interaction with DNA-associated enzymes. Dox intercalates into DNA base pairs and inhibits Chemical structure Chemical formula C27H29NO11 Molecular weight 543.5 g/mol Melting point 229-231 ºC Water solubility 1.18 mg/mL (2% soluble) LogP 1.27 LogD 0.02 PKa (dissociation constant) Strongest acidic: 9.46 Strongest basic: 8.46 Table 1 - Physico-chemical characteristics of doxorubicin [84], [85]. 14 macromolecular biosynthesis, producing a wide range of cytotoxic effects against human tumour cell lines [88], [89]. It can also induct DNA double-strand breaks and chromosomal aberrations by its interaction with topoisomerase II [84]. Upon enzymatic reduction, dox is able to originate ROS (OH.) which may cause extensive cell damage and cell death [87]. Despite its advantages a great problem with this drug is that it also induces apoptosis and necrosis in healthy tissues causing great toxicity in major organs like the heart, brain, liver and kidneys [89]. This can cause acute side effects like cardiac arrhythmias, nausea, vomiting, flares among others, and late side effects like myelosuppression, alopecia and chronic cardiotoxicity that can lead to heart failure [87]. For these reasons, over the years a great number of researchers have been studying to devise a DDS capable of enhancing the efficacy and selectivity of dox, being that in clinic is frequently administered in a liposomal solution, Doxil®. The ideal DDS needs to ensure that the complex arrives and acts at the target location, and this targeting can be an active or passive process. In an active targeting, the nanocarrier (NC) surface is functionalized with ligands that bind to specific receptors in the target tissue, while in a passive targeting, the NC diffuses and accumulates at highly vascularized sites, such as tumours, where the overexpression of vascular endothelial growth factor (VEGF) increases vascular permeability [38]. The selective accumulation of the nanosystem in tumours, occurs via the enhanced permeability and retention (EPR) effect and active cellular uptake [90], [91]. Overall, these technologies are gathering great research attention, since they offer numerous advantages over conventional chemotherapy. They can be used solo (monotherapy) or synergistically combined with existing therapies, like magnetic hyperthermia, allowing a controlled release of the drug and enhancing its antitumoral efficiency [21]. Among the various nanoscale DDS, lipid nanocarriers (NCs) have emerged as central delivery systems, as they mimic in many ways the natural lipid environment found on biomembranes, presenting high biocompatibility and low toxicities when applied in vivo [92]. Moreover, in very limited cases, this kind of NCs show the capability to cross the BBB, which is relevant for the treatment of cerebral tumours. This is thought to happen because of their ability to mimic low-density lipoproteins (LDL) particles, being transported to the brain by a LDL receptor [93]. Examples of lipid NPs used as DDS are micelles, liposomes, ethosomes, nanoemulsions, nanostructured lipid carriers (NLCs) and solid lipid nanocarriers (SLNs) (figure 7). The last two particles are quite similar, differing only in the lipid matrix, which in SLNs contains only solid lipids, while in NLCs contains solid and liquid lipids [94]. 15 The current development in cancer nanotechnology focuses on cancer detection, diagnosis, and treatment in a targeted manner using a single delivery system. Therefore, the focus of this work will be SLNs simultaneously loaded with a chemotherapeutic drug and magnetic and/or paramagnetic NPs. Thus, this bionanosystem will be able to provide: non-invasive imaging by MRI, diagnosis and follow-up; controlled release with MHT induction; and a synergistic treatment combining chemotherapy (dox) and MHT. SLNs, also called lipospheres or solid lipid nanospheres, are solid lipids at human physiological temperature (37º), biodegradable, biocompatible and stable against aggregation or coalescence [46], [95]. In comparison with liposomes, SLNs present a higher stability, higher protection of the incorporated compound, higher entrapment efficiency for hydrophobic drugs, easy scale up potential and low production cost [96]–[98]. These NCs can be prepared by using solid lipids (at body and room temperature), including mono-, diand triglycerides, fatty acids or waxes, water and surfactants to stabilize the particle and form administrable emulsions [95], [99], [100]. SLNs can be produced by various methods such as: microemulsification, solvent displacement, ultrasonication, phase inversion, solvent emulsification evaporation, solvent emulsification diffusion, meltemulsification, cold homogenization and high-pressure homogenization (hot homogenization), with the latter being the most used due to its scale up potential [2], [95], [97], [99], [100]. Despite all the different mentioned methods, they can be summarized since all of these are based on lipid fusion. This process begins with the lipid phase being melted and posteriorly dispersed through high-speed agitation in an aqueous solution containing a surfactant, followed by a high-pressure homogenization, this way achieving a nanoemulsion. Lastly, emulsion droplets crystalize forming lipid NPs with a solid matrix. This matrix makes them ideal to encapsulate lipophilic drugs, such as doxorubicin, but the physico-chemical properties, polymorphic form and crystalline characteristics of the constituent lipids at biological Figure 7Types of lipid based nanocarriers. Adapted from [96]. 16 temperatures, can influence the drug loading into SLNs. This loading is done during the fusion step and before the high-shearing step in the lipid phase (for hydrophobic drugs) [97], [100]. The loading of dox in SLNs (dox-SLNs) to bypass MDR was addressed, showing that dox-SLNs efficiently enhance apoptotic cell death through higher accumulation of doxorubicin in dox-resistant breast cancer cells, overcoming chemoresistance [101]. Another study with dox-SLNs prepared via the solvent emulsification diffusion method, showed that the drug was released in higher rates at lower pH, as happens in tumour environment [102]. Mussi et al. studied dox-SLNs, evaluating at the same time the effect of docosahexaenoic acid (DHA), a fatty acid that increases the anticancer drugs activity and its encapsulation efficiency. This study showed greater dox release at lower pH and higher cytotoxicity in lung tumour cell line when DHA was incorporated, proving its efficiency as a nanoscale DDS against cancer [103]. Dox-SLNs were also studied for glioblastoma treatment, showing great drug cytotoxicity against glioma cell lines and increasing its permeation through a cell monolayer, assumed as a model of the BBB. These results validated SLNs as promising NC for dox delivery to the brain in glioblastoma treatment [104]. Overall, nanotechnology allied to drug delivery is very promising, and new formulations are now being intensively explored, thereby providing hope for new cancer treatment options in the near future. 1.4.2. Magnetic Resonance Imaging Nowadays, MRI is one of the most powerful imaging tools, not only in clinical diagnosis but also in biological research. MRI is a non-invasive technology to obtain real-time images of the internal anatomy and physiology of living organisms and it’s the most efficient available method to identify soft tissues. For that reason, it has been extensively used to study the brain, asses cardiac function and detect abnormalities in the organism [105]. In spite of all its capabilities, MRI still needs supplements to help distinguish abnormal soft tissues more accurately. For this effect, compounds known as contrast agents can be used. CAs are introduced into an organism to enhance imaging quality, being able to give anatomical and functional information. MRI is based on the nuclear magnetic resonance (NMR) principle in which the nuclear magnetic moment of water protons aligns with an external magnetic field and is later excited with a radiofrequency (RF) pulse, generating an electric signal of a specific frequency (Larmor frequency, 𝜔0) [106], [107]. This pulse disturbs the protons spins from the direction of the applied magnetic field, and on their way back to their original low-energy state, two relaxation times can be defined: the longitudinal relaxation time ( T1 , spin-lattice), which is based on energy transfer and reflects the length of time it takes to recover from a 17 decreased net magnetization in the direction of the applied magnetic field; and the transverse relaxation time ( T2 , spin-spin), which is based on spin dephasing and reflects the length of time it takes for a NMR signal to decay in the perpendicular direction to the applied magnetic field (figure 8) [108]. The spin dephasing can be caused by local inhomogeneities due to tissue-inherent factors or to external sources. When this process accounts for both of these factors, the images produced are called T2 * [108]. Depending on the nature of the CAs, they can preferentially modify the T1 or T2 relaxation times [109]. The contrast that can be seen in MR images is the result of local variations of T1 and T2 relaxation times of water molecules in adjacent regions. Therefore, as the human body is composed of 60% water, CAs able to reduce the water relaxation times and subsequently enhance MRI contrast are needed [106]. These CAs are often based in (para)magnetic NPs. When placed under an external magnetic field, each particle creates a magnetic field of its own that induces field inhomogeneities in surrounding areas, which will disturb the proton spins of surrounding water molecules [108]. These events will change the MR signal, measured as a shortening of T1 , when there is a close interaction between protons and CAs, and a shortening of T2 , when there is a large susceptibility difference between the particles and the surrounding medium that results in small magnetic field gradients [108]. These measures are able to intensify the signal differences between the healthy tissues and the diseased ones, with T1 CAs giving a bright contrast and the T2 CAs offering a dark contrast. The efficiency of a determined CA is given by its longitudinal and transversal relaxivity values, r1 e r2 respectively. These values are obtained by measuring the relaxation rates (inverse of relaxation times), R1 (1/ T1 ) or R2 (1/ T2 ), at different concentrations of CA. The higher the relaxivity of a given formulation, the higher its efficiency as an MRI CA. Thus, the smaller the relaxation time of CA, the greater its contrast enhancement efficiency [106]. Dual CAs are already being developed by several researchers [110], [111]. Figure 8 - Fundamentals of MRI based on nuclear magnetic resonance principle. B0, external magnetic field [108]. 18 These are able to enhance simultaneously T1 and T2 signals, which will allow the observation of different tissues with increased specificity, being able to give a more accurate diagnose without the need of injecting more than one kind of contrast. For the evaluation of these formulations their relaxivities are still considered, but so is the r2 / r1 ratio. This ratio is an important parameter defining CA performance: high r1 values and r2 / r1 ratios near 1 result in bright contrast in T1weighted imaging; high r2 values with high r2 / r1 ratios result in a dark contrast enhancement in T2weighted images; intermediate r2 / r1 ratios enable a dual T1 - T2 enhancement [112], [113]. Although ferric ions have 5 unpaired electrons that increase their r1 value, iron oxide NPs possess high MS and consequently a high r2 relaxivity and a large r2 / r1 ratio. For these reasons these NPs are not appropriate for T1 contrast enhancement [114]. SPIONs were already commercialized as T2 MRI CAs by AMAG Pharmaceuticals, Guerbet group and Bayer Schering Pharma [46], [52]. Due to their strong magnetization and superparamagnetic behaviour, these iron oxide NPs create local microscopic field inhomogeneities and activate the dephasing of protons. This way, the use of SPIONs results in a prolonged shortening of both T2 and T2* relaxation times in the region around them, creating hypointense dark domains in MR images (negative contrast) [105], [115]. Furthermore, the high susceptibility of SPIONs allows the efficient tracking of labelled cells and can induce perturbations of the magnetic field in normal tissues, creating a “blooming effect” that demolishes the background around lesions [109]. As a result, SPIONs can be used to detect infection and inflammation, with higher accuracy than gadolinium-based CAs (paramagnetic T1 CAs), since they can be phagocytised by macrophages leading to imaging hypointensity of macrophage infiltrated tissues [115], [116]. The larger magnetic moment of SPIONs compared to gadolinium CAs, explains the increased relaxivity of SPIONs compared with gadolinium at similar tissue concentrations [115]. Iron oxide NPs can also be used for functional MRI (fMRI) using the BOLD (Blood Oxygen Level Dependent) technique, being useful for the diagnosis of cardiovascular diseases and cancer [52]. Another important property of SPIONs is their potential for safe use in patients with chronic kidney disease, given its rapid removal through extravasations and renal clearance, which can be an advantage over the Gd-based CAs on the market [50], [115], [117], [118]. Despite all the advantages, these T2 CAs also have disadvantages that limit their clinical use. One example could be the fact that their negative contrast is sometimes confused with other hypointense areas (calcifications, bleeding and metal deposits), which can deceive the clinical diagnosis [109]. For this reason, dual CAs active in both T1 and T2 imaging acquisition modes are highly needed to unequivocally distinguish physiological interferences. 19 Currently, only T1 CAs are used in the clinic in Europe. These agents are based on a paramagnetic ion, Gd3+, that due to its short blood circulation times and toxicity has to be strongly chelated with organic molecules [118]. Unlike T2 CAs, T1 CAs present high r1 relaxivity and a small r2 / r1 ratio, producing a bright (hyperintense) positive contrast. Manganese-based T1 CAs have revealed a longer circulation half-time, that enhances the signal in tumours, a higher biocompatibility and they do not cause susceptibility artefacts such as T2 CAs [114]. MnO NPs are great candidates to substitute/complement T1 Gd CAs, used now in the clinic, since manganese, unlike Gd does not present significant toxic effects in the human body [114]. Studies have showed that nanosheets of responsive MnO2, undergo structural changes under pH and redox stimuli, to enhance their MRI performance over 50 fold [6], [119], [120]. MnO2 particles present an OFF-ON activatable character, since their paramagnetic character increases, upon reduction to Mn2+, enhancing their MR signal [5], [6]. Mn2+ ion exhibits very prominent contrasting effects, revealing detailed physiologic and biological information. This ion allows the visualization of subanatomic structures of the brain and its neurological activity, unlike Gd3+ and T2 CAs [109], [114]. 1.4.3. Magnetic Hyperthermia The use of hyperthermia in the treatment of tumours is an old practice that consists in rising the temperatures of a specific part of the body or the whole organism, above the physiological level (4145ºC). This treatment is considered as a supplementary treatment to chemotherapy, radiotherapy and surgery in cancer treatment [26], [121]. Conventional hyperthermia treatments, such as hot water bath, perfusion heating, high-frequency radiation or ultrasound, among others, were employed in the ablation of malignant tumours [122]. However, a problem with classic hyperthermia is the lack of homogeneity in the heat distribution profile, which can cause unwanted hot spots in healthy tissues and cold spots in the tumour (allowing its repopulation) [22]. Through the use of NPs, nanotechnology can improve hyperthermia, allowing for a more localised treatment (figure 9). Figure 9Nanoparticles concentrated inside the tumour can enhance the effects of conventional hyperthermia, while reducing possible side effects. NPTT: Nano-Photo-Thermal Therapy; NaRFA: Nano-Radio-Frequency Ablation; NMH: NanoMagnetic Hyperthermia; NUH: Nano-Ultra sound hyperthermia [22]. 20 Magnetic hyperthermia is a minimally invasive treatment, where MNPs produce heat through magnetic energy losses, that occur in the presence of an alternating magnetic field. This heat can be used for the ablation of tumours, since cancer cells are more sensitive to high temperatures (>41ºC) than normal cells [22], [80]. One of the most sought-after administration routes for MNPs is through intravenous injection. One option to accumulate the MNPs in the tumours after injection is the use of magnetic fields; MNPs are guided towards the target tissue due to the presence of a sufficiently strong magnetic field, that can make the NPs overcome the blood flow [4]. An adequate MNP functionalization can also improve the targeting of the NPs. On a recent study, Liao et al. studied the effect of iron oxide NPs functionalized with alginate and galactosamine as the cell-targeting ligand (Fe3O4@Alg-GA) in MHT in vitro . The results showed that this nanostructure is able to promote hyperthermia, significantly decreasing the cell viability in a human hepatocellular cell line (HepG2) [53]. Currently, iron oxide NPs are considered the favourite heating agent for MHT, because of their interesting size-dependent properties, biocompatibility, minimal toxicity and ease of excretion and functionalization. SPIONs are already commercialized for hyperthermia purposes by the European Institute of Science and Magforce (NanoTherm®) [52]. In multi-domain and single-domain particles, the main heating processes are hysteresis losses and relaxation losses (Néel or Brown relaxation), respectively [4], [22]. Thus, the heating of SPIONs is mainly due to relaxation losses during the reorientation of the magnetization under alternating field exposure. Néel relaxation results from random flips of the spins without particle rotation, being the only relaxation process in MHT with immobilized NPs [22]. Brown relaxation results from frictional forces if the particle can rotate in a medium of low viscosity [48]. The quantity of heat produced by the MNPs should be as high as possible, in order to diminish the dose, and it depends strongly on NPs properties such as: mean size, magnetic anisotropy (shape), MS and on the amplitude and frequency of the oscillating magnetic field [48]. An important characteristic of MNPs utilized in clinical hyperthermia is their specific absorption rate (SAR), defined as the rate at which electromagnetic energy is absorbed by unit mass of a biological material [80], expressed in W/g: 𝑆𝐴𝑅 = cdT 𝑚𝑁𝑃dt where c is the heat capacity, and 𝑑𝑇 is the temperature increment in a certain time (𝑑𝑡), and mNP is the mass of the NP in question per mL of water. The SAR, that results from loss processes, depends on the applied AMF frequency and on the magnetic field. The efficiency of MNPs in energy conversion is of great importance, and for this reason, the ideal particle would present a high SAR at low AMF, since it implies (1) 21 a lower time of residence of the NPs in the human body and lower dosages to be administered [4], [80], [123]. MHT is most of the times used in combination with other therapies such as radiotherapy and chemotherapy. The main aim of radiotherapy is to eliminate cancer, maintaining the therapeutic dose in the diseased cells, while reducing the radiation dose on adjacent healthy cells. Therefore, controlling and killing the tumours at a cellular level is of great importance. This can be achieved using both targeted MNP and an AMF, to sensitize tumour cells for posterior radiotherapy treatments. In a clinical trial, MaierHauff et al. investigated the efficacy of MHT combined with radiation therapy in the treatment of glioblastoma multiforme. The results showed that MHT was well tolerated by all patients and the median survival rate was greatly increased [124]. The use of MHT also reduced radiation dose, leading to a prolonged overall survival [125], [126]. Chemotherapy can also be enhanced with the application of concomitant MHT. MHT can increase the blood flow, perfusion and blood vessel pore size in the tumour, facilitating drug accumulation and penetration in solid tumour tissues. MHT can also improve drug cytotoxicity through an enhanced intracellular uptake of drugs, because of the increased cell membrane permeability. Inhibition of DNA repair and acceleration of cytotoxic reactions, can also contribute to increase drug toxicity in combined treatments [121]. Itoh et al. evaluated the thermal enhancement of cisplatin and adriamycin cytotoxicity in vitro, using human bladder cancer cell line (T24). The results showed that the combined chemotherapy and MHT treatment had a higher anti-tumour effect than either hyperthermia or chemotherapy alone. The survival rate was the same in combined and drug alone treatments, however the drug concentration was 10-fold higher in the latter. This indicates that the quantity of drugs used can be significantly reduced in the combined treatment, hence limiting chemotherapy side effects [127]. Overall, MNPs can be used in diverse biomedical applications, being extensively explored in the field of diagnosis and treatment of cancer. Recent research has been focused on the use of these NPs for early detection and treatment of solid tumours, creating in this way an innovative tool for the fight against cancer. 1.5. Objectives The main aim of this project is the development and initial validation of a novel theranostic nanocomposite system to help in the fight against cancer. The individual goals of the project are: • The design of a nanocomposite comprising a lipid matrix, a chemotherapeutic drug and a (para)magnetic core; 28 transitions are the strongest and most important in IR spectroscopy [144]. Generally, a molecule with N atoms has 3N degrees of freedom, that are the maximum number of potential transitions of that molecule. All molecules have three translational degrees of freedom, being able to move in three directions. But if the molecule is nonlinear it also has three rotational degrees of freedom and if it’s linear, it has two rotational degrees of freedom. Thus, nonlinear and linear molecules have 3N-6 and 3N-5 vibration modes, respectively [144]. When irradiated, molecules absorb frequencies that are characteristic of their structure, being possible to determine their vibrational frequencies and consequently the corresponding functional group. The most common vibration modes are: bending, that involves changes in the bond angles of the atoms and can be further classified as scissoring, rocking, wagging and twisting; stretching, that changes the length of the bonds between atoms; and torsional modes that involve the twisting of the backbone of the molecule (figure 14) [143], [144]. These motions can also be symmetrical or asymmetrical and in-plane or out-of-plane, and can only be IR active if the corresponding vibration changes the molecular dipole moment [145] . Opposed to other spectroscopies FT-IR has a high spectral resolution, good signal-to-noise ratios and an ability to measure a broad region of the spectrum, normally from 4000 to 400 cm-1, in little time [146]. The absorption bands of the spectra are characterised by its frequency, intensity and shape, which are unique features for each individual molecule. The band frequency indicates the presence of certain functional groups, the band shape gives information about the group functionality and the material purity, and the band intensity provides information on the amount and type of a determined functional group of the molecule, by comparison with other bands [144]. Figure 14 - Possible vibrational modes of molecules. Adapted from [145] . 29 Experimentally, first the background spectrum of the IR source is recorded. Next, the spectrum of the IR source with the sample in place is recorded. At this point, the radiation of different wavelengths goes through a Michelson interferometer, where the light is directed to a beam splitter that reflects half of it into a fixed mirror and other half into a movable mirror, creating two different optical paths. The beams from both mirrors are returned to the beam splitter and partially reflected to the detector. The Michelson interferometer is responsible for recombining these two beams, conducting them into the sample, and then into the detector that measures the optical path difference of the beams. Afterwards, the computer takes that raw absorption data (interferogram) and conducts a math process known as the Fourier Transform, to decompose the signal obtained into its constituent frequencies, this way generating a readable absorbance spectrum [147]. In figure 15 is shown a schematic representation of this process [148]. 2.2.4. Inductively coupled plasma-atomic emission spectroscopy Inductively coupled plasma-atomic emission spectroscopy (ICP-AES) is a widely used technique to identify and quantify the presence of metals in different samples. It is a commonly used technique in nanomedicine to quantify the mass percentage of a determined element in nanocomposites. It combines an induced coupled plasma, with a spectrophotometer to obtain a spectrum that is representative of ppm levels of trace elements [149]. These spectrophotometers are capable of running almost every metal in a large number of samples per run [150]. The source of an ICP-AES instrument is an argon-based plasma that is generated by a RF coil and when power is applied an alternated current oscillates and creates a magnetic field from the coil to the torch. This field helps to maintain the argon flow and breaks down the argon gas into ions and electrons, this process is called inductively coupled plasma [150], [151]. The plasma provides high thermal energies from 5500 to 6500 K, that can destroy to a great extent the molecular bonds of most elements, this way the analytes dry, dissociate and atomize [152]. To get the sample into the plasma it must be nebulized. The sample is pumped into a pneumatic nebulizer that forms an aerosol and then enters a spray chamber Figure 15Schematic representation of a FT-IR spectrometer [146]. 30 that selects very small droplets [150], [152]. These droplets are carried by the argon gas into the plasma and their atoms collide with energetically excited argon species, ionize and emit characteristic atomic and ionic spectra that are collected by a detector [149]. The emitted radiation is specific to the element and its intensity is proportional to the element concentration in the sample [150]. Before reaching the detector, the emitted light is focused onto the entrance slit of a spectrometer by using a convex lens arrangement (figure 16). The light is then separated by a diffraction grating into its component wavelengths in a spectrometer. The spectrometer can have different configurations to achieve different ends. In the case of the Echelle spectrometer represented in figure 16, the diffracted light from the grating focuses on a prism, suffering a secondary dispersion, this way avoiding spectral overlap [153]. With respect to the detection system the spectrometer can also be sequential or simultaneous. In the sequential spectrometers only the radiation from a selected wavelength is detected, being that this selection occurs by rotating the grating. In the simultaneous spectrometer the detector, normally a charge coupled device (CCD) detector is used and allows the simultaneous analysis of many wavelengths or elements [149], [153]. This equipment allows a large number of samples to be analysed with a very high sensitivity and a very broad dynamic range. 2.3. Microscopy Microscopy is a technical field widely used in research to study materials or specimens not visible to the naked eye, using microscopes. In this work optical and electron microscopy were used to study the NPs by themselves and applied in cells. Figure 16 - Schematic representation of an ICP-AES equipment with an Echelle spectrometer. Adapted from [151] and [153]. 31 All microscopes present a resolution limit that can be influenced by the source and the lenses conditions. Ernst Abbe discovered the optical equation (7), called Abbe diffraction limit, that indicates the requirements needed to obtain a clear image, without distortion caused by aberrations. This equation states that the resolution of a given instrument is proportional to the wavelength of the transmitted light, and inversely proportional to the numeric aperture (NA) [154]. 𝑑 = 𝜆 2𝑛 𝑠𝑖𝑛 𝜃 =𝜆 2𝑁𝐴 where 𝜆 is the wavelength of the light source, 𝑛 is the index of refraction of the medium between the object and the objective and 𝜃 is the acceptance angle of the objective lens. The 𝑛𝑠𝑖𝑛 𝜃 represents the numeric aperture, that measures the capacity of a lens to receive the transmitted light [154]. Since there’s a clear dependence of the resolution over the wavelength, optical microscopy was used to observe cells, which are bigger, and electron microscopy was used to observe NPs, which are smaller. 2.3.1. Optical microscopy In optical microscopy a beam of light is directed to the sample and transmitted through or reflected by it. The microscope allows its observer to analyse the effects of light as it interacts with the sample, obtaining a magnified image of the sample. For this work an inverted optical microscope with a camera was used. This microscope distinguishes itself from the regular upright microscopes, because its objectives are placed below the stage, while its light source and condenser are on top [155]. The light source, that can be a LED (light-emitting diode) illuminator or a halogen lamp, shines through the condenser lens that guides the light evenly across the field of view (FOV) of the microscope. The image is amplified by the objectives and eyepiece lens. These microscopes can have a digital camera connected to them, to make easier to take pictures of the sample. They normally support various methods like fluorescence microscopy, phase contrast method and bright field (BF), among others. 2.3.2. Confocal microscopy Confocal microscopy is a type of optical microscopy, that allows the acquisition of fluorescence micrographs with increased contrast and resolution, using a focused laser for excitation and a pinhole to block out-of-focus light [156]. Two dimensional (2D) images are acquired plane to plane, at different depths, which allows fluorescence detection in thick samples. Tri dimensional (3D) objects can be visualized by scanning several planes and stacking them, using a microscopy deconvolution software (zstack). This equipment allows the visualization of different sections of a sample using different (7) 32 fluorophores, through the illumination of the sample with a multiwavelength laser that uses various laser excitation lines [157]. Figure 17 shows a general and simplified scheme of a confocal microscope. A laser ③ is used to illuminate a source pinhole ④ that focus the light into a dichroic mirror or beam splitter ⑤. Then the light is reflected into an objective ⑥ that converges the light into a single point in the focal plane of the specimen. The emitted fluorescent light (purple line) from this focal point is focused at the detector pinhole ② and reaches the detector ①. The light emerging from objects outside that plane (red line) is largely excluded by the pinhole, not contributing to the final image [158], [159]. This microscopes can also include excitation and emission filters, allowing a more detailed fluorescence imaging [157]. This process contributes to an image with more resolution and less artefacts, ideal to the observation of cells and their components. 2.3.3. Transmission electron microscopy Transmission electron microscopy (TEM, an acronym that also stands for transmission electron microscope) is a microscopy technique in which an accelerated electron beam is transmitted through a sample. Through this technique it is possible to analyse the morphology, composition and the crystallographic structure of materials, with a resolution of approximately 0.2 nm [30]. This high resolution is obtained through the acceleration of the electrons. The wavelength of an electron is dependent upon accelerating voltage (generally from 80 to 300 kV), accordingly to the equation (8): 𝜆 = ℎ √2𝑚ⅇ𝑉 where ℎ is the Planck’s constant, 𝑚 the mass of the electron, ⅇ its electronic charge and 𝑉 the accelerating voltage applied. When the applied electron beam is accelerated with high values of voltage, TEM can obtain high resolution for extremely short wavelengths, and the previous equation can be approximated to the following (9): 𝜆~ 1,23 √𝑉 Figure 17 - General representation of a confocal microscope [157]. (8) (9) 33 Therefore, the desired electron wavelength can be achieved by varying the accelerating voltage of the electron beam. The higher the accelerating voltage, the smaller the wavelength of the electrons and the higher the possible achievable resolution [160]. Figure 18 shows the schematics representation of a TEM. The mechanism of imaging of TEM can be explained as follows: an electron gun generates an electron beam, that is controlled by many electromagnetic lenses and metal apertures, arranged sequentially along the microscope column [161]. The column should maintain a high vacuum, since the electrons cannot move in atmosphere. The focusing phenomenon occurs because the electrons behave as negatively charged particles, that are deflected by the magnetic and electric components of the microscope [160]. The condenser lens and aperture, exclude high angle electrons, modulating the electrons into a defined beam that focus in the sample [161]. The ideal thickness of the sample should be around 100 nm, so that the electrons can pass through [160]. The objective lens, right above the specimen stage, focuses the transmitted electrons into one point in the image plane obtaining the first image. With the objective aperture, a well-defined electron beam is attained and an electron diffraction pattern can be obtained [161]. Then, the projector lens magnifies the image and the diffraction pattern onto the phosphorescent screen, to translate the electron image information into a visible form [160]. Normally this equipment is fitted with electron detection system, mainly charged coupled devices. A CCD camera is positioned above the phosphorescent screen, to acquire digital images of the specimen. Additional lenses and detectors can be installed to perform different TEM modes, or simply to correct chromatic aberrations. Electrons are one type of ionizing radiation, that produces a wide range of secondary signals when interacting with the sample, which will be dependent on the thickness of the sample. The thicker the sample, the longer the electron path, and the more likely it is that an electron will experience multiple scattering events [161]. Primary and secondary electrons, transmitted and back-scattered electrons (BSE), low energy Auger electrons and X-rays are some examples of scattering events (figure 19) [160]. Figure 18 - Schematic representation of a TEM. Adapted from [159]. 34 BSE and transmitted electrons may experience elastic (no energy loss) or inelastic (energy loss) scattering, that can be useful in different TEM modes. Many of these signals are used in other techniques like scanning transmission electron microscopy (STEM) and energy dispersive X-ray spectroscopy (EDX), in the case of characteristic X-rays [162]. 2.3.3.1. STEM-EDX mode Scanning transmission electron microscopy is a mode that can be available in TEM. For this mode, the instrument needs to be fitted with STEM scan coils, which can scan a focused electron beam across the sample. The TEM would also need to be equipped with BF (bright field) and annular dark-field (ADF) detectors for imaging [163]. This technique can be combined with analytical techniques like EDX, to obtain a mapping of the elemental composition of the sample [160]. For the EDX an X-ray detector needs to be mounted in line-of-sight of the specimen. The characteristic X-rays generated by the interaction of electrons with the sample, are collected and counted according to their energy by the detector. This way, in a little time collection, an EDX spectrum can be obtained, allowing an elemental analysis of the sample [164]. Combining this technique with STEM, an X-ray spot spectrum can be attained from many locations across the sample. This enables a BF or ADF image acquisition with the elemental distributions marked in specific areas of the sample surface, throughout the image [161]. Electron microscopy techniques are contributing significantly in various areas of research, being particularly relevant in the study of the morphology and chemistry of NPs, so useful for drug delivery studies. Figure 19 - Types of signals produced by the interaction of a high-energy electron beam with the sample [160]. 35 2.4. Thermogravimetric analysis Thermogravimetric analysis (TGA, also thermogravimetric analyser) is an essential technique used for material characterisation regarding their composition. It allows to monitor the mass of a substance as a function of temperature over time [165]. The sample is positioned under a controlled isothermal or varying temperature in a given atmosphere [166]. The TGA relies on a high degree of precision in three different extents: weight, temperature and weight change with temperature over time. This permits the determination of degradation temperatures, the quantity loss of water or solvent, decarboxylation, the level of organic and inorganic components, the amount of residue remaining, among others [167], [168]. Figure 20 shows a schematic representation of thermogravimetric analyser with horizontal furnace, which helps minimize turbulence caused by thermal buoyancy and the purge gas [165]. This equipment consists of a sample pan that is supported by a precision balance ①. These components are contained inside a furnace, that can be vertical or horizontal, and achieve temperatures as high as 2700 K [166], [167]. A thermostated balance chamber ②, is used to minimize environmental influences in the balance cell and two adjustments ring weights ③, to help in the weight measurement. Temperature sensors are placed right above the sample pan ④, to measure its temperature, and in the furnace ⑤, to measure the atmospheric temperature inside the furnace. A furnace heater ⑥ is also present inside the furnace to either heat it or cool it [165]. A purge gas supply system consisting of a gas inlet ⑦ and a gas outlet ⑧, controls the sample environment. This gas may be inert or a reactive gas that flows over the sample through a reactive gas capillary ⑨ [168]. The descending TGA thermal curve indicates the weight loss occurred in function of temperature. For simultaneous detection of thermal events, the TGA can feature a DSC (differential scanning Figure 20 - Schematic representation of a TGA with a horizontal furnace. Adapted from [163]. 36 calorimetry) heat flow measurement [165]. This measurement can be attained in the presence of a DSC sensor that detects thermal events, like fusion and crystallization, being able to register accurately the exact transition temperatures. 2.5. Magnetic measurements 2.5.1. Vibrating sample magnetometer A vibrating-sample magnetometer (VSM) is a scientific equipment used to measure the magnetic properties of a material. In this technique a sample is vibrated sinusoidally at a small amplitude, perpendicularly to a homogeneous magnetic field [169]. The vibrating component causes a change in the magnetic flux, which generates an electrical signal in the coils based on Faraday’s Law of Induction (10). This law states that an electromotive force (𝜀 in volts) is induced in a conductor by a time varying magnetic flux [170]: 𝜀 = −𝑁 𝛥𝜙𝐵 𝛥𝑡 where 𝑁 is the number of turns of a coil, 𝛥𝜙𝐵 the change in magnetic flux and 𝛥𝑡 the change in time. Figure 21 shows a schematic representation of a VSM. First, the electromagnets create a uniform magnetic field, that magnetize the sample [170]. Then, the electrodynamic sample vibrator induces a sinusoidal vibration of the sample at constant frequency, with the help of a straw used as the sample holder. The oscillating magnetic field of the vibrating sample causes a change in the magnetic flux of the nearby pick up coils (detection coils). This change induces a voltage in the suitable located pick-up coils, that is proportional to the magnetization (M) of the sample. A second voltage can be induced in similar coils, called reference coils, by a reference sample that can be a permanent magnet. This will generate a reference signal that can make the measurements insensitive to changes of vibration amplitude and frequency, magnetic field inhomogeneities among other instabilities [169]. Both the signals coming from the different sets of coils are processed by a lock-in amplifier and then converted into a graph of magnetization versus magnetic field strength (H), normally referred to as hysteresis curve [171]. (10) Figure 21 - Schematic representation of a measurement process by VSM. Adapted from [168]. 37 2.5.2. Superconducting quantum interference device The superconducting quantum interference device (SQUID) is a very sensitive magnetometer able to detect changes in an applied magnetic flux. This device can measure the magnetic properties of a material with a field resolution of 10-17 T, based on superconducting loops containing Josephson junctions. To have superconductivity is necessary to refrigerate the system with liquid Helium, below a critical temperature of 4.2 K. A SQUID device consists of a superconductor ring interrupted by one or more Josephson junctions (figure 22A) [172]. The Josephson junctions create a resistive barrier between two superconducting regions, leading to electron tunnelling across the junction [173]. If a constant biasing current is maintained in the SQUID sensor, it develops a voltage across the junctions. The magnetic flux passing through the loop enhances the current in one junction and reduces it in the other, which causes the voltage across the junction to oscillate [173]. When an external magnetic flux is coupled into the loop, the voltage changes in a periodic manner that corresponds to an increase of one flux quantum. By measuring the change in voltage it is possible to determine the magnetic flux of the material coupled to the SQUID [174]. The measuring process in the SQUID is very similar to the one in VSM. The sample is put inside a straw and moved vertically along a set of three superconducting pick-up coils and subjected to a uniform magnetic field [175]. The coils are connected by a single piece of superconducting wire and are only sensitive to the magnetic fields of the sample, eliminating any contribution from external fields. The coils, the wire and the squid create a superconducting loop. Any change in the magnetic flux will produce a proportional change in the persistent current of the coils, consequently producing variations in the SQUID output voltage [175]. The output signal is then transformed into electronic signals, thus allowing the extraction of the magnetic properties of the material [176]. This process is illustrated in figure 22B. Figure 22 - A: dual Josephson junction of a SQUID device; B: schematic setup of a SQUID magnetometer and an inset showing the squid response in voltage versus the sample position. Adapted from [172] and [174]. A B B0 44 is fixed in the center of the coil as shown in figure 30, providing a good thermal isolation that diminishes the loss of heat generated by the sample and avoid interferencial heatings from the electronics. This configuration ensures experiment repeatibility and the validity of the field readings. The temperature sensors in this equipment are incorporated in the sample holder [192]. This particular configuration is the most reliable to calculate the SAR of magnetic colloids, but it is less versatile for other in vitro / in vivo experiments. Figure 30 - Schematic representation of a MHT applicator ideal for calorimetric studies. Adapted from [190]. 45 CHAPTER 3 - MATERIALS AND METHODS 3.1. Materials In this section, all the materials used for the development of this project, from equipment to chemicals, will be listed in table 2. Table 2 - List of all equipment, materials, software and reagents used throughout the course of this project. Equipment - Centrifuge miniSpin - eppendorf; - Centrifuge UNIVERSAL 320 – Hettich zentrifugen; - Compact Spectrofluorometer - FluoroMax-4 – Horiba Scientific; - Confocal microscope– LSM780 on inverted Axio Observer - Zeiss; - Digital Sonifier® - Branson Ultrasonics Corporation; - Dynamic Light Scattering – Zeta potential - SZ-100Z – Horiba Scientific; - Fluorescence plate reader (Biotek, model: SYNERGY H1); - Fourier Transform Infrared Spectroscopy – Vertex 80v – Bruker; - High Performance Liquid Chromatography – 1290 Infinity – Agilent Technologies; - Inductive Plasma Atomic Emission Spectrometer - ICPE 9000 - SHIMADZU; - Inverted microscope - Eclipse TS100 with a digital sight DSFi1 camera – Nikon; - Magnetic Hyperthermia applicator – Magnetherm version 1.5; - Magnetic Hyperthermia equipment – DM1 - nanoScale Biomagnetics; - MR solutions 3.0 Tesla benchtop magnetic resonance imaging system; - Precision balance – ABT 120-5DM – KERN; - Superconducting Quantum Interference Device – Vibrating Sample Measurement – MPMS - Quantum Design - Thermogravimetric Analyser – TGA/DSC 1/1100 SF STARe System - Mettler Toledo; - Transmission Electron Microscope – JEOL JEM 2100 200 kV (Cryo & Tomography); - Ultra-sound – Elmasonic P - Elma; - UV-Vis SpectroPhotoMeter - UV-2550 – SHIMADZU corporation; - Vibrating-sample magnetometer – model 3473-70 electromagnet - GMW magnet systems; - Vortex – VV3 - VWR; - X-Ray Diffraction System – X PERT PRO MRD – PANalytical Software - Image J (Fiji); - Microsoft Office 365; - Origin 9.0; - Paint 3D; - Zen 3.0 (Blue edition) Materials - Filters (30000 Da) – Millipore; - Pur-a-lyzer (60000 Da) – Sigma Aldrich; - Custom printed MRI holder; - Carbon coated copper grids, 400 mesh – Ted Pella,Inc. - CELLViewTM dish with 4 compartments and glass bottom – Greiner BioOne; - 24 well plates; - 96 well plates; Reagents/solutions - AlexaFluor 647 (Invitrogen) 46 - Ammonium hydroxide (Sigma-Aldrich); - anti-α-tubulin antibody (Sigma-Aldrich) - Aquabluer (MultiTarget Pharmaceuticals, LLC); - Carnauba Wax (Koster Keunen Holland BV); - Chloroform (analytical grade or better: Sigma-Aldrich); - DAPI (4',6'-diamino-2-fenil-indol) (Sigma-Aldrich) - DiO (3,3'-Dioctadecyloxacarbocyanine Perchlorate)(Iris Biotech); - Doxorrubicin (Adooq Bioscience); - Dulbecco’s modified Eagle’s medium (DMEM); - Fetal bovine serum (FBS); - Hexane (analytical grade or better: Sigma-Aldrich); - Iron(II) chloride tetrahydrate (Sigma-Aldrich); - Iron(III) chloride hexahydrate (Sigma-Aldrich); - Milli-Q water (ultra-pure); - Oleic Acid ≥99% (Sigma-Aldrich); - Penicillin-Streptomycin; - Phospate-buffered saline (PBS, ph - 7.4), (Sigma-Aldrich); - Potassium Permanganate (Sigma-Aldrich); - Shandon Immu-mount mounting media (Thermo Scientific); - Tween 80 (Sigma-Aldrich); 3.2. Methods In this section the methods used for the synthesis and characterisation of NPs are described. The procedures done for the drug studies and in vitro experiments are also described. 3.2.1. Synthesis of superparamagnetic iron oxide nanoparticles (SPIONs) Hydrophobic nanoparticles of magnetite (Fe3O4@OA) were prepared by coprecipitation method adapted from [193]. Briefly, 9.2 g of iron(II) chloride tetrahydrate and 15 g of iron(III) chloride hexahydrate were dissolved in 250 mL of mili-Q water and stirred for 10 min at 50 °C. After 10 min, 30 mL of ammonium hydroxide, NH4OH (12 mol/L), were added starting this way the coprecipitation that resulted in a dark precipitate of Fe3O4. Then 2.5 mL of oleic acid (OA) were added and the mixture was heated at 80 °C for 1 h. The excess of NH4OH and OA was removed by magnetic separation of Fe3O4@OA with the use of a magnet, followed by the decantation of the supernatant and the redispersion of the solid in fresh solvent. The washing procedure was repeated five times with milli-Q water and four times with hexane. Finally, the Fe3O4@OA NPs were dried, dispersed in chloroform and stored in the fridge [194]. Figure 31 illustrates this procedure. 47 3.2.2. Synthesis of manganese oxide nanoparticles Hydrophobic manganese oxide (MnO@OA) nanoparticles were prepared according to [195]. The procedure is illustrated in figure 32. For the preparation, 0.5 g of potassium permanganate (KMnO4) were dissolved in 250 mL of Milli-Q water and left stirring for about 30 mins. A total of 5 mL of OA were added and the sample was left stirring, at room temperature, for 24 h. The next day there was a brown 'solid' floating on a colourless solution, this solution was decanted and the solid washed twice with mili-Q water. After this, ethanol (EtOH) was added to remove any possible residue reactants. At this point the solid was resuspended in the solvent, transferred to falcon tubes and centrifuged for 5 min at 8500 rpm. The supernatant was once more discarded and the pellet resuspended in hexane, resulting in a dark-brown solution that was stored in the fridge until further use. Figure 31 - Illustration of the coprecipitation method, used for the synthesis of Fe3O4@OA NPs. Figure 32 - Illustration of the procedures used for the synthesis of MnO@OA NPs. Co-precipitation ↗T: 50 - 80ºC FeCl 2 .4H 2 O+ FeCl3.6H2O Stirring: 10 min - 50 ºC Heating: 1 h - 80 ºC Separation and washing NH4OH Nucleation OA Growth Colloidal Fe3O4@OA CHCL3 KMnO 4 + milli-Q OA Stirring: 30 min Stirring: 24 h milli-Q Washing EtOH H Centrifugation: 5 min - 8500 rpm Colloidal MnO@OA Hexane 48 3.2.3. Synthesis of manganese dioxide nanoparticles A sonochemical method was used for the preparation of manganese dioxide nanostructures (MnO2@Glc), based on the mild reduction of MnO4 by glucose (Glc) (figure 33) [5]. In a typical synthesis, 0.057 g of Glc were dissolved in 30 mL of mili-Q water. Then 0.25 g of KMnO4 were added and the purple solution was immediately sonicated for 20 min at 25% power, with a 10 mm tip in a digital sonifier. Then, for the purification of the samples, the solutions were centrifuged at 8500 rpm for 15 min, the supernatant was discarded and the pellet re-suspended in 40 mL of water. This process was repeated 2 more times and finally the pellet was resuspended in 30 mL of water. This solution was then centrifuged for 3 min at 3000 rpm to remove large aggregates. This time the supernatant was kept and stored until further use. 3.2.4. Synthesis of solid lipid nanocarriers Throughout the development of this work non-magnetic and magnetic SLNs (mSLNs) were synthetized. All the nanoparticles presented are made of a surfactant, Tween 80 that is approved by the FDA; a solid lipid matrix made of the biocompatible carnauba wax; and DiO, a green fluorescent lipophilic carbocyanine, useful to trace the NPs in cells. The mSLNs can contain Fe3O4, MnO and MnO2 combined in different ways. The final NPs comprise a (para)magnetic core and a chemotherapeutic drug, doxorubicin. A general procedure for the mSLNs will be explained, the details about each formulation can be found in table 2. A modified melt-emulsification method was used for the preparation of SLNs (figure 34). Initially, 200 mg of carnauba wax were mixed in a glass vial with a colloidal solution of MNPs, 0.25 mL of a chloroform solution of DiO (1 mg/mL), added only in final formulations, and dox. This mixture was then placed in the digital sonifier. At this point, 4.5 mL of Milli-Q water and 0.5 mL of a water solution of Tween 80 (50 mg/ml) were added to the vial and the sample was ultrasonicated with a 3 mm tip, for 2 min, at 25% Figure 33 - Illustration of the sonochemical method, used for the synthesis of MnO2@Glc NPs. Purification KMnO 4 Glc + milli-Q Sonication: 25 % power; 20 min. Centrifugation: 15 min; 8500 rpm 3 x Centrifugation: 3 min; 3000 rpm milli-Q Aqueous MnO2@Glc 49 power, at 20 second working intervals, while being simultaneously heated with a heat gun. Immediately after the sonication, the sample was immersed in ice to solidify the lipid NPs. Once cold, the formulation was centrifuged for 10 min, at 3000 RPM to remove big wax aggregates, the supernatant was stored in a glass vial and the pellet was discarded. Prior to the synthesis of our final SLNs, containing SPIONs, MnO and dox (SLN@Fe3O4-MnO-dox), several intermediary experiments were made along the way. First, it was necessary to test if the drug and the different MNPs could be efficiently encapsulated into the SLN’s alone (SLN@dox, SLN@Fe3O4, SLN@MnO, SLN@MnO2) and once they were, the following procedure was to encapsulate both magnetic and paramagnetic particles (SLN@Fe3O4-MnO, SLN@Fe3O4-MnO2). Table 3 shows the theoretical percentage of each compound that was used, in relation to the quantity of Carnauba wax. Table 3 - Percentage of dox and MNPs present in the SLNs. Formulation dox (%) Fe3O4 (%) MnO (%) MnO2 (%) SLN@dox 20 - - - SLN@ Fe3O4 - 30 - - SLN@ Fe3O4–dox 20 5 - - SLN@MnO2 - - - 10 SLN@ Fe3O4–MnO2-dox 10 5 - 20 SLN@MnO - - 22 - SLN@ Fe3O4–MnO - 10 20 - SLN@Fe3O4–MnO–dox 20 10 20 - Figure 34 - Illustration of the modified melt-emulsification method used for the synthesis of mSLNs. Wax + dox + DiO +MNPs Tween 80 + mili-Q Sonication: 25% power; 2 min (20 s ON/OFF); Centrifugation: 10 min; 3000 rpm 50 3.2.5. Physico-chemical characterisation techniques 3.2.5.1. DLS To measure the average size, PI and the zeta potential of the SLNs a Dynamic Light Scattering – Zeta potential, SZ-100Z equipment from Horiba Scientific with nanometric resolution (from 0.3 nm to 8 µm) was used. For all the measurements an electrode cell was used. For all the measurements 1 mL of sample was used, with a dilution of 1:100, since it was the ratio at which the lowest PI values could be obtained. For the DLS and ELS measurements for the SLNs containing both MnO and Fe3O4, only the refractive index of magnetite was used. For the SLNs containing either of those components it was used their respective refractive index. For the SLNs without MNPs a generic organic sample refractive index was used. All the samples were measured at 25ºC, 3 to 5 times, with a stabilization time of 90 seconds, being that the final values presented are the mean of all the measurements. For the size only the “mean average” of the peak was considered, and for the zeta only the ‘’mean zeta’’ value. The parameters used for the measurement of the size and zeta potential are listed in table 4. Table 4 - Parameters used to measure the size and zeta potential of the NPs. 3.2.5.2. UV-Vis absorption spectroscopy For the UV-Vis spectrum acquisition an UV-Vis SpectroPhotoMeter - UV-2550 from SHIMADZU corporation, with a spectral range from 190 to 1100 nm was used. The samples were highly diluted in water or hexane, so the references used were either water or hexane. These measurements were made only for manganese NPs and SLNs containing manganese, since this material presents a known peak in the UV-Vis spectrum. This way, by analysing the spectra obtained, it was possible to identify the presence Cell - Electrode cell: carbon, 6mm Detector angle - 173º Sample - Dilution - Refractive index - 1:100 - Magnetite: 1.900 – 1.700i - Manganese: 2.170 – 0.000i - Organic samples: 1.600 – 0.000i Dispersion medium - Refractive Index - Temperature - Viscosity Water - 1.333 - 25 ºC - 0.88 cP 51 of manganese in the samples analysed. The wavelength range used for these measurements was from 300 to 800 nm, the slit aperture was 5 nm and the cells used were quartz cells. 3.2.5.3. Fluorescence spectroscopy The acquisition of fluorescence spectra was made with a Compact Spectrofluorometer, FluoroMax4 from Horiba Scientific. This technique was used to analyse the samples of SLNs containing dox, to confirm the presence of the drug, since it is a fluorescent drug with an emission peak around 590 nm. The emission spectra of DiO, a fluorescent dye with an emission at around 506 nm, was also recorded. The spectra were acquired from 500 to 700 nm, using 5 nm entrance and exit slits, with λex of 480 nm for dox, and 489 nm for DiO. Both these wavelengths are very close to each other, making it difficult to read the spectra. For the spectra acquisition the samples were highly diluted, and a four opening disposable cell was used. 3.2.5.4. FT-IR For the spectra acquisition a FT-IR - Vertex 80v from Bruker was used (figure 35A). This system is equipped to perform in the middle infrared (MIR) and far infrared (FIR) ranges. For that effect a MIR source, made of silicon carbide is installed inside the equipment, and a FIR source, a water-cooled mercury lamp, is installed externally. The spectrophotometer is also equipped with a transmittance and an ATR (attenuated total reflectance) accessories, shown in figure 35B and 35C respectively. The operation mode used in this work was the MIR-ATR mode. The detector used was a very sensitive MCT (Mercury-Cadmium-Telluride) detector, cooled with liquid nitrogen. This technique was used to obtain information about the functional groups present in the SLNs and their constituents (Tween 80, Carnauba wax and MNPs) and compare their spectra. For sample preparation 500 µL of the MNPs and the SLNs were dried in the vacuum and put directly in the ATR. The Tween 80 and C. wax did not need any preparation. After a background acquisition, the sample is put in the diamond present in the ATR accessory, which has a higher refraction index than the sample, and is Figure 35 - A: Vertex 80v FT-IR system; B: accessory for the transmittance operation; C: accessory for ATR operation mode. A B C 52 then submitted to vacuum and the measurements are taken. It is to note that the FT-IR is a relative technique and the simple observation of its spectra is necessary, but not a sufficient condition to determine the presence of functional groups. 3.2.5.5. ICP-AES For the ICP-AES measurements standard solutions with predefined well-knonw concentrations of the elements to be analysed (Fe and Mn) needed to be prepared. This allows to stablish a calibration curve for each element, that would be used by the system to calculate the metals concentration. For the iron, the solutions prepared ranged from 0 to 10 ppm and for the manganese, the standards ranged from 0 to 0.5 ppm. For sample preparation 5 or 10 µL of MNPs or 100 µL of SLNs were pipetted into a 15 mL falcon together with 1 mL of hydrochloric acid and left to react overnight. Before the analysis 9 mL of mili-Q water were added to the falcon and placed in the carousel of the equipment to measure. To analyse the concentration of iron and manganese the emission wavelengths used were 235 nm and 257 nm, respectively. 3.2.5.6. TEM/ STEM-EDX For the development of this project a TEM, JEOL JEM-2100-HT (Cryo & Tomography), with an accelerating voltage between 80 kV and 200 kV and a resolution of 0.24 nm was used. This instrument is equipped with a high brightness LaB6 (lanthanum hexaboride) electron gun and a fast-readout OneView 4k x 4k CCD camera. The JEOL allows not only TEM, but also STEM with a resolution of >1.5 nm. It also has an EDX from Oxford Instruments, with a solid angle of 0.13 and a hard X-ray aperture for elemental analysis. This equipment was used to analyse the morphology of the synthetized MNPs and SLNs. Its major use was for the SLNs, to verify if the MNPs were efficiently encapsulated. When it was not possible to visualize the MNPs encapsulated, namely the case of MnO2, a STEM-EDX analysis was performed. This technique enabled the localization of MnO2 particles within the SLNs, through the elemental mapping of the sample. For sample preparation the samples were highly diluted into the respective solvents, and then 7 µL of these dilutions were loaded into carbon coated copper grids with 400 mesh (squares across the grid) from Ted Pella Inc., represented in figure 36, and dried in the vacuum [196]. In the microscope the sample was loaded with a single tilt holder, Figure 36 - Schematic representation of a carbon copper grid from Ted Pella [194]. 53 and the accelerating voltage was set at 200 kV, for both TEM and STEM-EDX analysis. STEM-EDX analysis were performed with BF contrast. The images were analysed and treated using Image J (Fiji) software, mainly for NPs measurements, brightness and contrast corrections. 3.2.5.7. TGA For this work a TGA/DSC 1, 1100 SF, STARe system from Mettler Toledo was used. For this analysis no sample preparation was needed, 50 µL of the liquid samples were loaded directly into the sample pan (figure 37) and its weight registered. The analysis was performed under a nitrogen atmosphere. In the case of the SLNs the method used consisted in a temperature increase from 25 to 900 ºC at 10ºC/min, with an approximately 30 min stop at 120 ºC to assure that all the water had evaporated. In the case of the SPIONs the method consisted in a temperature increase from 25 to 900 ºC at 5ºC/min, with a 10 min stop at 60 ºC to make sure that the CHCl3 had evaporated. 3.2.5.8. VSM To magnetically characterise MnO and MnO2 NPs a VSM 3473-70 electromagnet with a coil gap of 127 mm, from GMW was used. For sample preparation, 500 µL of sample were dried in the vacuum and then loaded into a gelatin capsule. The sample was weighted and fixed with cotton and placed in a glass straw that is the sample holder (figure 38A). Then the sample was carefully placed in the equipment, between the two electromagnets and the measurement was started (figure 38B). For these measurements a magnetic field of ± 2 T was applied. 3.2.5.9. SQUID To magnetically characterise SLNs a SQUID-VSM, MPMS magnetometer from Quantum Design was used. This equipment combines the SQUID technology with a vibrating sample measurement, also used in the VSM equipment. For sample preparation, 500 µL of sample were dried in the vacuum and then Figure 37 - TGA sample pan [163]. A B Figure 38 - A: sample holder and sample for VSM; B: sample placed in the VSM equipment. 60 61 CHAPTER 4 - RESULTS AND DISCUSSION 4.1. Synthesis of nanoparticles For the synthesis of MNPs (Fe3O4, MnO2 and MnO) the goal was to attain a solution with high concentration of particles and in great quantities. Since the NPs are stable in solution for long periods of time, having great quantities of these NPs will allow their utilisation for many ends, in this case to produce various SLNs. The surface of all MNPs were functionalized with either OA or Glc, to stabilize the final NPs and prevent their aggregation. This was a critical step to obtain stable monodisperse colloidal and aqueous suspensions over time. For the synthesis of spherical magnetite NPs functionalised with OA, a hydrothermal method was tested first. Even though the quantities obtained with this method were good, the concentration of magnetite obtained was small and not ideal for its encapsulation into SLNs. Therefore, a simpler, twostep coprecipitation method was tested. This last method offers an economic route to produce great quantities of monodisperse hydrophobic Fe3O4@OA NPs for biomedical applications. The OA is adsorbed on the surface of the NPs through chemical interaction between OA COO groups and Fe atoms. This way the hydrophobic tails of the OA face outwards forming hydrophobic Fe3O4@OA NPs (figure 43). These NPs were dispersed in a non-polar solvent, CHCl3, which is not only useful for their encapsulation into the SLNs, but also to prevent oxidation. The colloidal solution obtained had high concentrations of iron ([Fe]=67 mg/mL]). For the preparation of manganese oxide NPs, a simple method based on stirring was used. The main reaction is based on the reduction of the KMnO4 by the OA, being that the molar ratio of these two compounds crucial to determine the size and morphology of the nanostructures [195]. As the reaction is conducted in water, it is easier scale up, offering an economic route for the preparation of MnO. Therefore, with this method great quantities of a monodisperse colloidal solution of MnO@OA NPs were obtained, with a concentration of manganese of 44.6 mg/mL. Unlike magnetite, MnO NPs present an irregular shape that can vary by using different molar ratios of KMnO4/OA, in this case the particles present themselves in the form of nanosheets (figures 48A and 48B). For the synthesis of manganese dioxide NPs, various methods were tested along the way. The main problem with most of these methods was that the NPs presented very low concentrations with nonFigure 43 - Schematic representation of oleic acid-coated magnetite NPs. 62 uniform shapes. Also, despite the hydrophobic character of the prepared NPs, its encapsulation into the SLNs was not efficient. For these reasons hydrophilic MnO2@Glc NPs were produced using a sonochemical method that is an easy one step method to break KMnO4 chemical bounds. The reaction is based on a mild reduction of MnO4by Glc. With this method high quantities of “crushed paper” like NPs were obtained with a low manganese concentration of 0.412 mg/mL. Despite their hydrophilic character these were the only MnO2 NPs produced, that were efficiently loaded into the Carnauba matrix of the SLNs, yet in minimal concentrations. For the SLNs synthesis a modified melt emulsification method was used due to its simplicity and scalability [198]. With this method monodisperse aqueous solutions of SLNs containing MNPs and dox were obtained. The organic matrix of the NPs made of the biocompatible and FDA approved Carnauba wax, its ideal to incorporate the MNPs and the drug. The procedure allows to obtain SLNs with a regular size and zeta potential, however from batch to batch it is difficult to obtain constant concentrations of MNPs in the SLNs. As showed on chapter 3, different formulations had different percentages of magnetic content, but even when these theoretical percentages were the same in different batches, variable concentrations of MNPs were obtained. Table 7 shows the different values of concentration for each ion (Mn and Fe) in each formulation. Mn concentrations are always much lower than the theoretical values (as expected on the other hand at least for MnO2 due to their hydrophilic nature), which can limit the Mn NPs utility in biomedical applications. Table 7 - Concentration of ions for each SLN formulation. In the synthesis of SLNs containing dox the concentration of drug, fixed at 20% in relation to the Carnauba wax, was totally reproducible. Different batches with the same theoretical percentages of dox, showed constant encapsulation efficiency values of almost 100%. Nevertheless, the content of dox is spontaneously released over time until the quantity of dox inside and outside the lipid particles is the same, so the percentage obtained can be lower over time. For the final formulation (SLN@Fe3O4-MnOFormulation Fe (mg/mL) Mn (mg/mL) SLN@ Fe3O4 1.470 - SLN@ Fe3O4–dox 0.200 - SLN@MnO2 - 0.141 SLN@ Fe3O4–MnO2-dox 0.083 0.117 SLN@MnO - 0.289 SLN@ Fe3O4–MnO 0.347 0.129 SLN@Fe3O4–MnO–dox 0.222 0.042 63 dox) the EE was 99.6±0.2 %, which is an encouraging result. The concentration of DiO was fixed from the beginning to a level at which it is ideal for fluorescent optical imaging. 4.2. Physico-chemical characterisation 4.2.1. (Para)magnetic nanoparticles MNPs and lipid NPs were extensively characterised in the present study. Structural, chemical, magnetic and functional characterisation were performed through the techniques mentioned in chapter 3, to exploit the NPs capabilities and understand their behaviour, in order to validate the final bionanosystem for biomedical applications. Magnetite NPs stabilized with OA were first structurally characterised using TEM. The size obtained by TEM analysis corresponds only to the NP crystal core (without the organic coating). From figures 44A and 44B, acquired with 50k and 500 k times magnification respectively, it is possible to see that the magnetite NPs present a pseudo-spherical shape with a monodisperse distribution. The average size was determined by measuring 300 NPs from several TEM micrographs. The measurements revealed that the size of the crystals vary from 4 to 18 nm, and present a bell-shape size distribution that could be fitted to a Gaussian equation, characteristic of this kind of magnetic particles (figure 44C) [199]. The particles were measured, presenting an average diameter of 9.7±0.1 nm. The X-ray diffraction pattern of Fe3O4@OA NPs presents well-defined diffraction peaks, signifying that these NPs are highly crystalline. Figure 45 shows the XRD spectra obtained and the already documented magnetite pattern. The diffraction peaks with 2θ values of 30.1, 35.5, 37.1, 43.1, 53.5, 57 and 62.6 degrees are in agreement with the known Bragg reflections of magnetite (Crystallography Open Database), which are indexed as (2,2,0), (3,1,1), (2,2,2), (4,0,0), (4,2,2), (5,1,1) and (4,4,0) [200]. When in contact with oxygen magnetite can be easily oxidized and transformed into maghemite and hematite at high Figure 44 - A-B: TEM micrographs of magnetite NPs capped with OA; CSize distribution obtained by TEM of OA capped magnetite NPs, fitted to a Gaussian distribution. C A B 64 temperatures [201]. The absense of diffraction peaks characteristic of these coumpounds or any other secondary phase in the XRD spectrum, indicates that the synthetyzed product is purely magnetite. The thermogram obtained by thermogravimetric analysis for oleic acid-coated magnetite NPs is shown in figure 46. The TGA curve shows three distinguished steps of wheight loss. Over the temperature of 120 to 300 ºC, ①, the weight loss corresponds to the physic decomposition of the OA adsorbed on the surface of the NPs; from 300 to 500 º C, ②, to the chemical decomposition of this coating agent [202], [203]. The reason for the third weight loss its not clear, but one hypothesis can be that its due to the reducing gases produced by the OA degradation. Thus, the weight loss represented on ③, above 700 ºC may be of the magnetite itself due to its reduction by the reducing gases produced during the OA decomposition [203]. From this analysis it is possible to extrapolate that the Fe3O4@OA NPs contain aproximately 10% of OA. Figure 45 - X-ray diffraction pattern of oleate-coated magnetite. ① ② ③ Figure 46 - Thermogravimetric profile of the hydrophobic Fe3O4@OA in N2 atmosphere. 65 The magnetic properties oleate-capped magnetite NPs were measured with the SQUID-VSM equipment at 5 K and 300 K (room temperature). The field-dependent magnetization curves, represented on figure 47A, show the typical superparamagnetic behaviour of magnetite NPs. The hysteresis loop is more evident at 5 K, suggesting that at low temperature these particles are in a superparamagnetic magnetically blocked regime showing ferromagnetic-like behaviour (open hysteresis loop). However, at room temperature, they behave as a pure superparamagnetic, in which both remanent magnetization and coercivity are very close to 0 (Mr= 2.9 emu/g; Hc= 0.03 kOe). The Fe3O4@OA NPs present high saturation magnetization (Ms= 67 emu/g), significantly lower than the value of bulk magnetite (Ms= 92 emu/g). It is important to highlight that the Ms value was higher than that of Feridex® (Ms= 45 emu/g), which is a SPION formulation FDA approved as MRI contrast agent [204]. The reduced magnetization compared to the bulk material, may be attributed to the phenomenon of spin canting, defined as a nonuniform distribution of spins [205]. The distortion of the spin alignment is more pronounced at the surface of the particles, leading to a higher reduction of saturation magnetization in particles with a higher surface to volume ratio, like the oleate-capped magnetite NPs [205]. However it is important to note that the Ms value obtained for these particles increases, if only the magnetite core of the particles is taken into account for the magnetization normalization, instead of the total mass of the sample. Lastly, figure 47B shows the zero-field cooled and field cooled (ZFC-FC) magnetization curves under an applied magnetic field of 100 Oe. These curves reveal a blocking temperature (TB) of 150 K, which represents a magnetic transition from a magnetically blocked state to a superparamagnetic regime for the synthetised OA-capped SPIONs. Paramagnetic manganese oxide NPs (MnO@OA) were structurally characterised by TEM. The size obtained with this technique is attributed to the nanoparticle without the oleic acid coating as in the case of magnetite NPs explained before. As seen in figures 48A and 48B, acquired with 50 k and 150 k times Figure 47 - A: Field-dependent magnetization curves of OA-capped magnetite NPs at 5 and 300 K. The inset is a close-up of both curves; B: ZFC-FC magnetization curves under a magnetic field of 100 Oe. A B 66 magnification respectively, the MnO is presented as nanosheets that form spherical aggregates of themselves. The size of these NPs could not be perceived clearly, so the measurements where made along the major axis of the spherical form, formed by the nanosheets. The average size was determined by measuring 100 NPs in different micrographs. The size varied from 30 to 100 nm and the determined average size was 53.6±2.5 nm, still small enough to be considered for encapsulation. The absorbance of the colloidal suspension of MnO was acquired by UV-Vis spectroscopy. The spectrum shows a shoulder of absorption around 350-400 nm, which is typical of this kind of MNPs [206]. This absorbance may be due to the quantum confinement effects of MnO nanosheets, that are present at nanometric scale [207]. Figure 50 shows the X-ray diffraction pattern obtained for these particles and the already documented MnO pattern. The XRD spectrum of MnO@OA NPs did not present well-defined diffraction peaks, thus a background subtraction was made so the existing peaks became more distinguishable. The diffraction peaks with 2θ values of 35.2, 40.9 and 88.7 degrees are the only peaks in agreement with the known Bragg reflections of MnO (Crystallography Open Database), which are indexed as (1,1,1), (0,0,2) and C A B Figure 48 - A-B: TEM micrographs of MnO NPs capped with OA; CSize distribution obtained by TEM of OA capped MnO NPs, fitted to a Gaussian distribution. Figure 49 - UV-Vis spectrum of colloidal solution of MnO@OA nanosheets. 67 (0,0,4) [200]. The identified peaks are not sufficient to extrapolate from the XRD graph that the analysed sample is indeed manganese oxide. This could be due to the versatile nature of Mn and its many oxidation states, which in addition to with the nano size of the particles that broadens the peaks, makes it very difficult to unequivocally identify the phase of the particles. The magnetic behaviour of MnO@OA NPs was analysed by a vibrating sample magnetometer. This equipment is less sensible than the SQUID-VSM, but it is able to determine the magnetic character of a sample. The field-dependent magnetization curve of OA coated MnO NPs measured at room temperature is shown in figure 51. Through the observation of figure 51 it is possible to notice that no Ms or hysteresis are present and the curve is almost a straight line. This linear increase of the magnetization with the applied magnetic field indicates that the sample has a paramagnetic behaviour, in agreement with the reported magnetic behaviour for manganese oxide NPs, with a maximum magnetization of 0.15 emu/g at the highest field measured [208]. Figure 50 - X-ray diffraction pattern of oleate-coated manganese oxide nanosheets. Figure 51 - VSM graph for MnO@OA nanosheets. 68 Responsive manganese dioxide NPs functionalized with glucose where morphologically characterised using TEM. As seen in figures 52A and 52B, acquired with 50 and 120 k magnification respectively, the MnO2 is presented as nanosheets with an irregular shape resembling crushed paper sheets. The size of MnO2@Glc could not be perceived clearly, so the measurements where along the major axis of more external nanoparticles. The average size was determined by measuring 100 NPs from different TEM micrographs. The size varied from 15 to 30 nm and the determined average size was 21.8±0.3 nm, which is ideal for the encapsulation into the SLNs. The absorbance of the aqueous solution of MnO2 was obtained by UV-Vis spectroscopy. Like the MnO NPs, the spectrum shows a shoulder of absorption around 350-400 nm, which is characteristic of MnO2 nanosheets [5]. As before, quantum confinement effects are responsible for this behaviour. The XRD spectrum obtained for these particles and the documented MnO2 pattern are shown in figure 54. The spectrum of MnO2@Glc NPs did not present well-defined diffraction peaks, which as mentioned above may be due to the various oxidation states existing for Mn and to the effect of the nano size of the particles that broadens the peaks. The diffraction peaks with 2θ values of 37.2, 67.1 degrees C A B Figure 52 - A-B: TEM micrographs of MnO2 NPs capped with glucose; CSize distribution obtained by TEM of OA capped MnO NPs, fitted to a Gaussian distribution. Figure 53 - UV-Vis spectrum of colloidal solution of MnO2@Glc nanosheets. 69 are in agreement with the known Bragg reflections of MnO2 (Crystallography Open Database), indexed as (0,1,0) and (1,1,0) [200]. The information retrieved from this diffractogram is again not enough to extrapolate the crystalline structure of the sample. It is important to note that at the nanometric scale preferential crystallographic orientations can appear, that modify the peak intensity relations and hinder the phase identification by comparison with a given database. The magnetic behaviour of manganese dioxide NPs functionalized with glucose was analysed by a VSM. The magnetization curve of MnO2@Glc nanosheets, represented in figure 55, was measured at room temperature. The measured sample exhibits the typical paramagnetic behaviour (linear increase of the magnetization with the magnetic field) characteristic of manganese dioxide NPs, with a maximum magnetization of 0.8 emu/g at the highest magnetic field measured [208]. Figure 54 - X-ray diffraction pattern of glucose-coated manganese dioxide nanostructures. Figure 55 - VSM graph for MnO2@Glc nanosheets. 76 The magnetic properties of SLNs@MnO, were measured with the SQUID-VSM equipment at 5 and 300 K. The field-and temperature dependent magnetization curves are presented in figures 65A and 65B, respectivily. The magnetization curves show a linear M-H relationship, confirming the paramagnetic character of these NPs. The maximum magnetization is approximadely ten times greater at 5 K (0.5 emu/g) than at room temperature. According to the Curie law, PM = C/T, this is explained by the inversely proportional temperature dependence of the paramagnetic susceptibility (PM). The magnetization curve at 300 K also shows a very small diamagnetic contribution in the low field region that comes from the diagmagnetic behavior of the sample holder plus the cotton used to stick the powder to the gelatine capsule, which is only observed at low magnetic fields where the magnetization is very low for paramagnetic substances. As the magnetic field increases, the paramagentic behavior of the sample increases and the diamagnetic contribution is no longer appreciated. In the M vs H curve at 5K the paramagnetic susceptibility is higher and this diamagnetic contribution is not even observed (see inset of Figure 65). The ZFC-FC curves show a strong magnetization decay with temperature that obeys the Curie law characteristic of paramagnetic materials [216]. A peak is observed at low temperature whose origin is unclear. In order to elucidate the origin of this magnetic feature a more detailed magnetic characterization should be performed. Prior to the synthesis of the final NPs (SLNs@Fe3O4-MnO-dox) it was necessary to evaluate if the oleate-caped magnetite and manganese oxide NPs could be encapsulated together. Thus, SLNs@Fe3O4– MnO were observed by TEM and are shown in figures 66A and 66B, acquired at 50 k and 60 k magnification respectively. The particles are spherical, evenly disperse and their size is more similar to the one obtained by DLS (150-200 nm), suggesting a higher structural rigidity. The red arrows and the yellow arrows in both images point to MnO@OA and Fe3O4@OA NPs, respectively. Interestingly, the MNPs are efficiently and simultaneously encapsulated into the SLNs, which is a promising result. A B Figure 65 - Field-dependent magnetization curves of SLN@MnO at 300 K and at 5 K (inset); B: ZFC-FC magnetization curves at 100 Oe. 77 Finally, the SLNs@Fe3O4-MnO-dox were physico-chemically characterised. Figures 67A and 67B, show the TEM micrographs of these NPs, acquired with 20 k magnification. The SLNs present a spherical shape and some aggregation. This aggregation can be due to the presence of dox, but a stronger hypothesis is that it is caused by drying the samples in vacuum, as it was previously introduced. As in the dox free version of the SLNs, both MNPs were found to be efficiently encapsulated into the SLNs. In both images it is possible to identify the MnO@OA (red arrows) and the Fe3O4@OA (yellow arrows). The optical properties of this particles were studied by fluorescence spectroscopy. Figure 68 shows the fluorescence spectra for the SLNs@Fe3O4-MnO-dox, at an excitation wavelength of 480 nm for the dox and 489 nm for the fluorescent dye, DiO. Since both the excitation wavelengths are very close, their spectra are identical showing three different peaks. The first and less intense peak, at 506 nm corresponds to the fluorescence emission of DiO [217]. The next two peaks, at 557 and 590 nm, correspond to the fluorescence emission of dox, which matches the typical fluorescence spectrum of this A B A B Figure 66 - A,B: TEM micrographs of SLN@Fe3O4-MnO, the red arrows point to the MnO@OA NPs and the yellow arrows point to Fe3O4@OA NPs. Figure 67 - A,B: TEM micrographs of SLN@Fe3O4-MnO-dox, the red arrows point to the MnO@OA NPs and the yellow arrows point to Fe3O4@OA NPs. 78 drug [210], [211]. The observed optical behaviour confirms the presence of dox and DiO in the NPs, and the intensity of the peaks is in accordance with the quantities of each compound in the formulation. The TGA curve for the final SLNs presents two steps of weight loss and it is shown in figure 69. The first and biggest step of weight loss, ①, represents the degradation of the organic components of the SLNs. As stated before, this step of weight loss from 250 to 500 ºC corresponds to the degradation of the surfactant Tween 80, Carnauba wax, OA and doxorubicin [202], [212]–[215]. The second step of weight loss above 750 ºC, ②, may be due to the degradation of magnetite [203]. The inorganic content of SLN@Fe3O4-MnO-dox is approximately 47 %. The magnetic properties of SLNs containing manganese oxide, magnetite and dox were studied with the SQUID-VSM equipment at 5 and 300K. The field-dependent magnetization curves (figure 70A) and the magnetization curves under 100 Oe (figure 70B) confirm the mixed magnetic behaviour of these particles. As expected, the particles exhibit overimposed paramagnetic and superparamagnetic behaviours, due to the coexistence of paramagnetic MnO@OA and superparamagnetic Fe3O4@OA NPs. Figure 68 - Fluorescence spectra of SLNs@Fe3O4-MnO-dox at excitation wavelengths of 480 and 489 nm. ① ② Figure 69 - Thermogravimetric profile of SLNs@Fe3O4-MnO-dox in N2 atmosphere. 79 From figure 70A we can observe that the M-H magnetization curve at 5 K shows a ferromagnetic-like behaviour (open magnetic loop) (Mr= 0.08 emu/g; Hc= 0.16 kOe), ) that comes from the magnetite NPs, indicating that the nanoparticles are in a magnetically blocked state. At room temperature, these particles are within the superparamagnetic regime, in which the remanent magnetization and the coercivity are closer to 0 (Mr= 0.02 emu/g; Hc= 0.04 kOe). The paramagnetic contribution due to the presence of MnO NPs is appreciated at both 5 and 300K in the M-H curves, being dominant with increasing magnetic field. At 5K, the paramagnetic contribution becomes stronger and dominates the magnetic behaviour [218]. Thus, as in the case of SLNs@MnO the maximum magnetization is approximately ten times greater at 5 K (4 emu/g) than at room temperature (0.4 emu/g), as expected according to the Curie law [218]. The ZFC-FC curves also exhibit a mixed magnetic behaviour. At very low T, the magnetization decay with temperature is indicative of a dominant paramagnetic contribution coming from the paramagnetic MnO NPs, whereas as the temperature increases the typical superparamagnetic behaviour of Fe3O4 NPs emerge In fact, the ZFC magnetization curve reveals a blocking temperature at 97 K, above which the SLNs exhibit a superparamagnetic behaviour and below which they are in a blocked magnetic state, in agreeement with the ferromagnetic-like behaviour observed in the low field region of the field-dependent magnetization curve at 5K (left inset in figure 70). This combined magnetic behaviour can be exploited for biomedical applications, such as MRI diagnostic. The FT-IR spectra of Fe3O4@OA, MnO@OA, Tween 80 and Carnauba wax were acquired to ascertain the presence of common functional groups between these components and the SLNs. The spectrum of SLNs@Fe3O4-MnO was also acquired to evaluate if the presence of the dox in the SLNs@Fe3O4-MnO-dox could be observed (figure 71). The band at 560 cm-1 in the Fe3O4@OA spectrum corresponds to the symmetric Fe-O bonds stretching in the crystalline lattice of magnetite, and can also be found in the spectrum of the SLNs@Fe3O4-MnO-dox [203]. Around 1407 cm-1 a band appears that corresponds to the Figure 70 - Field-dependent magnetization curves of SLN@Fe3O4-MnO-dox at 300 K and at 5 K (left inset). The inset on the right shows a close-up of the two curves; B: ZFC-FC magnetization curves at 100 Oe. A B 80 CH3 umbrella mode of the OA, thus the presence of this group is also noticed in the MnO@OA spectrum [203]. In this last spectrum the peak around 425 cm-1 can be assigned to be the Mn-O stretching mode, which does not appear in any other spectra probably due to the small concentration of MnO in the SLNs [219]. Next in the MnO@OA spectrum two peaks at 1457cm-1 and 1710 cm-1 appear and are attributed to the asymmetric (-COO-) stretching mode and to the stretching vibration of (C=O) group, respectively. These functional groups are both present in the OA molecule, and the two peaks are also present in the wax and Tween 80 resulting in a more noticeable band in the SLNs [203]. However, the peak at 1457cm1 can also show a contribution from the bending vibration mode of (-CH3) groups that are present in the Carnauba wax and OA. The bands at around 2852 and 2922 cm-1 are attributed to the asymmetric and symmetric CH2 stretch, respectively [203]. These peaks are also present in the wax and Tween 80 with considerable intensity, resulting in their pronounced presence in both SLN formulations. In the Carnauba wax spectrum, the band around 721 cm-1 corresponds to the CH2 skeleton vibration, that can also be noticed in Tween 80, and consequently in the SLNs spectra [220]. The more distinct peak in the Tween 80 spectra is at 1100 cm-1 and corresponds to the stretching vibration of (C-O) groups, heavily present in Tween 80 chemical structure [220]. Consequently, these groups also appear very pronounced in both SLN formulations. In the Tween 80 and SLNs spectra, there is also a broad peak at around 3500 cm-1 that corresponds to the stretching vibration of (O-H) groups [220]. In the final nanoparticle formulation, containing both MNPs and dox, a new band appears at 1610 cm-1. This band corresponds to the bending vibration mode of (N-H) groups, present in dox [220]. In general, the results obtained by FT-IR are in accordance with what was expected, allowing the visualization of each component of the SLNs, in the final formulation. Figure 71 - FT-IR spectra of SLNs@Fe3O4-MnO, SLNs@Fe3O4-MnO-dox and their magnetic and organic components. 81 4.3. Functional Characterisation 4.3.1. Magnetic resonance imaging In order to evaluate the efficiency of the SLNs as contrast agents for MRI applications, both T1 and T2 maps for the most relevant particles will be analysed and their relaxivity values discussed. The longitudinal and transverse relaxation times are presented as a function of the concentration of Mn and Fe, respectively. The relaxivity values for all the particles were obtained from the slope of the graphs of relaxation rate ( R1/2 =1/ T1/2 s-1) versus ion concentration (mM) (annex 1 and 2). Figure 72 shows the T1 and T2 maps for the samples SLNs@Fe3O4-dox, SLNs@MnO and SLNs@Fe3O4MnO-dox. The relaxation times can be easily interpreted resorting to the calibration bar present in each map. First, the T2 map of the SLNs containing only 5% of iron (SLNs@Fe3O4-dox) was acquired to evaluate the CA efficiency of the encapsulated SPIONs at different concentrations. Figure 72A shows the results obtained for the transverse relaxation weighted map of these particles. The SLNs@Fe3O4-dox achieved a good T2 enhancement even at low concentrations and start to show some signal saturation around 60-75 µM under the conditions tested. The sample presents shorter relaxation times at higher concentrations, demonstrating its efficiency as a T2 CA. The r2 value in water for the SLNs@Fe3O4-dox was 183 mM-1 s-1, which is higher than the FDA approved superparamagnetic iron oxide CAs Feridex® ( r2 = 93 mM-1 s-1) and Resovist® ( r2 = 143 mM-1 s-1) [221]. Then, the T1 map of the SLNs containing manganese oxide (SLNs@MnO) was acquired to evaluate the contrast efficacy of the encapsulated MNPs at different concentrations. Figure 72B shows the 0,4 0,3 0,2 0,1 0 10 25 50 60 75 [Fe] µM T 2 weighted 3 2,25 1,50 0,75 0 88 175 263 438 [Mn] µM T 1 weighted 2,6 1,95 1,30 0,65 0 2,5 7,5 13 18 25 [Mn] µM T 1 weighted 2,3 1,72 1,15 0,57 0 13 40 66 93 133 [Fe] µM T 2 weighted A B C Figure 72 - A: MRI T2 map of SLNs@Fe3O4-dox; B: MRI T1 map of SLNs@MnO; C: MRI T1 and T2 maps of SLNs@Fe3O4-MnO-dox. 82 longitudinal relaxation weighted map measured for these particles, at increasingly higher concentrations. The T1weighted images show a great difference in contrast between the lower and higher concentrations, demonstrating the good performance of the particles as T1 contrast enhancers. The r1 value in water for the sample of SLNs@MnO was 1.25 mM-1 s-1, which is close to the manganese-based T1 CA Teslascan® ( r1 = 1.5 mM-1 s-1), approved by the FDA [221]. Finally, the T1 and T2 maps for the final particles (SLNs@Fe3O4-MnO-dox) were acquired and their relaxivities studied. Since the concentrations used for the SLNs containing only iron showed some saturation at higher concentration values, the range of iron concentrations used for this experiment was lower. Consequently, the range of concentration for Mn was also lower. These SLNs seem to attain a considerably good transverse and longitudinal enhancement even at a lower ranges of concentration, which is a great advantage since less quantity of solution would be needed for administration. The r1 value obtained for these particles was 13 mM-1 s-1 which is remarkably higher than Teslascan® and even Magnevist® ( r1 = 3.1 mM-1 s-1), a gadolinium-based T1 CA approved by the FDA [221]. The r2 value obtained was 318 mM-1 s-1 that is considerably higher than those of Feridex® and Resovist®, which is a good indicative of the efficacy of this formulation as CA for MRI. The r2 / r1 ratio is 24.5 mM-1 s-1, an intermediate value that indicates that the formulation in question can accomplish a dual T1 - T2 reduction. It is to note that the relaxivity values referred for commercial CA solutions are measured at 37 ºC, and as the relaxivity depends on the T (through the diffusion factor), actual values of the compounds presented in here could differ, although the differences are expected to be minimal given the small temperature difference between this temperature and that of the experiment (room temperature). Table 10 shows a summary of the relaxivities and r2/r1 ratios obtained for the different SLN formulations. Table 10 - r1,2 and r2/r1 ratio for the different SLN formulations. r1 (mM-1 s-1) r2 (mM-1 s-1) r2/r1 (mM-1 s-1) SLNs@Fe3O4-dox - 183 - SLNs@MnO 1.3 - - SLNs@Fe3O4-MnO-dox 13 318 24.5 Although the physico-chemical characterisation of responsive SLNs, the only formulations with MnO2, did not turn out exactly how it was expected, parallel MRI experiments were still performed to evaluate the responsive ability and CA efficiency of these formulations. The paramagnetic character of Mn ions is determined by the number of unpaired electrons, being that the higher the number of unpaired electrons, higher the paramagnetic character of the Mn ion [63]. Since in MnO2 structure, Mn4+ presents only 3 unpaired electrons, it has a weak paramagnetic character. Under certain biologically relevant stimuli, 83 MnO2 can be reduced to Mn2+ salts with 5 unpaired electrons, becoming highly paramagnetic. This reduction will enhance the T1 signal in MRI. Hydrogen peroxide (H2O2) was used to stimulate the reduction of MnO2 to Mn2+ as represented by the following reaction: Mn4+O2 + 2 H2O2 → Mn2+ + 2 H2O + 2 O2 Figure 73 presents the T1 and T2 maps for the samples SLNs@MnO2 and SLNs@Fe3O4-MnO2-dox in water and H2O2. The reaction with H2O2 should show major differences, particularly in T1 weighted images. However, to have deeper understanding of the process, differences in T2 signal were also studied. As starting point, the T1 map for the SLNs containing manganese dioxide (SLNs@MnO2) was acquired to evaluate the contrast efficacy and the O2 generation ability of the responsive SLNs at different concentrations. Figure 73A presents the longitudinal relaxation weighted map measured for these particles, at increasingly higher Mn concentrations in H2O and H2O2. At higher concentrations the T1weighted images present lower relaxation times, suggesting the NPs potential as CAs. Also, the contrast intensity is higher in H2O2, which indicates that the SLNs@MnO2 react with the H2O2 producing oxygen. The r1 value in water for the SLNs@MnO2 was 0.51 mM-1 s-1 in H2O versus 0.92 in H2O2, confirming the responsive character of the particles. Reported changes in r1 value upon reduction of MnO2 to Mn2+, are in the range of 50 to 100 fold. The changes observed in here are much more modest (2 fold) probably due to the encapsulation of MnO2 NPs inside the wax matrix that prevents its processing by H2O2. Nonetheless, the T1 and T2 maps for the SLNs@Fe3O4-MnO2-dox were acquired and are presented in figures 73B and 73C, respectively. The relaxation times are shorter at higher concentrations, for both T1 T 1 weighted 2,5 1,88 1,25 0,62 0 86 128 171 213 [Mn] µM H 2 O H 2 O 2 T 1 weighted H 2 O H 2 O 2 2 1,5 1,0 0,5 0 71 106 142 177 [Mn] µM 0,5 0,38 0,2 0,12 0 49 74 99 124 [Fe] µM H 2 O T 2 weighted H 2 O 2 A B C Figure 73 - A: MRI T1 map of SLNs@MnO2; B: MRI T1 map of SLNs@Fe3O4-MnO2-dox.; C: MRI T2 map of SLNs@Fe3O4-MnO2-dox. All images were acquired in water and in hydrogen peroxide. 84 and T2 -weighted images, suggesting that these particles have potential as dual contrast agents. The r1 value in water for the SLNs@Fe3O4-MnO2-dox was 2.2 mM-1 s-1 in H2O and 5.2 in H2O2, which is a great difference that confirms the responsive character of the NPs. Although no significant responsive behaviour is expected in T2 signal, the r2 relaxivities were also studied and showed a slight r2 difference between H2O (175 mM-1 s-1) and H2O2 (207 mM-1 s-1). The longitudinal relaxivity values obtained are both higher than the one of Teslascan®, and the transverse relaxivity values are also higher than the ones of Feridex® and Resovist®. The results obtained for these particles are great indicators of the dual CA potential of these SLNs and of their ability to produce oxygen in tumoral environment, however the coefficients of determination obtained for the relaxivity values were not great and more experiments need to be done to validate these results. The r2/r1 ratios have intermediate values that confirm the dual CA ability of this sample. Table 11 shows a summary of the relaxivities and r2/r1 ratios for the responsive formulations. Table 11 - r1,2 and r2/r1 ratio for the responsive SLN formulations in water and hydrogen peroxide. r1 (mM-1 s-1) r2 (mM-1 s-1) r2/r1 (mM-1 s-1) H2O H2O2 H2O H2O2 H2O H2O2 SLNs@MnO2 0.5 0.9 - - - - SLNs@Fe3O4-MnO2-dox 2.2 5.2 175 207 79.5 39.8 4.3.2. Magnetic hyperthermia efficiency The heating efficiency of SLNs@Fe3O4-MnO-dox NPs was evaluated using the highest field and frequency available (H=20 mT; f= 869 kHz). Figure 74 shows the heating curve for this sample, submitted to the AMF for 1h. Within 500 seconds of AMF induction, the temperature increased more than 3 ºC. The SAR value was calculated using the equation (1) present in chapter 1, where the heating capacity of the sample was assumed equal to that of water (4.18 J g-1 K-1). The SAR value obtained was of 277 W/g, more than 2 times higher than SAR reported for Feridex® (115 W/g), demonstrating the suitability of SPIONs as heat nanosources and the great heating potential of the final SLN formulation [222]. Figure 74 - Heating curve of the final formulation obtained after 1h of AMF induction. 85 4.3.3. Drug release studies The presence of MNPs in the SLNs can be profited to add a feature of control over the drug release process. The combination of passive drug release with magnetic hyperthermia will impact the release profile of doxorubicin. Therefore, the passive release of dox from the SLNs@Fe3O4-MnO-dox was monitored during 4 hours. Then, the amount of dox released for the same particles under continuous MHT (H=23 mT; f=174.5 kHz) was monitored in the same way. As shown in figure 75, the release of dox seems to be faster when it is induced by MHT, since the percentage of dox inductively released at the first time point is already higher than in the passive release. From 2 h it is possible to observe significant differences between the two modes of release (p<0.01), being that this difference is more significant at 3 h and 4h (p<0.001). At 4 h the percentage of drug released for the passive and induced release is about 8 and 14 %, respectively. These results suggest that the application of MHT can efficiently stimulate the release of the encapsulated drug. 4.3.4. In vitro studies 4.3.4.1. Cytotoxicity studies The cell viability indicator, AquaBluer™, was used to evaluate the cytotoxicity of different SLN formulations containing only the magnetic core, in a triple negative breast cancer cell line (Hs578T), which is one of the most aggressive cancers. Viable cells metabolize this reagent and turn it from its oxidized form (nonfluorescent, blue) to the reduced form (fluorescent, red). The fluorescence intensity of AquaBluer™ is proportional to number of viable cells in the sample, which allows to quantify the percentage of viable cells. The viability was assessed 48 h after the exposure to the NPs. The cell viability for different concentrations of each formulation (SLNs@Fe3O4, SLNs@MnO, SLNs@MnO2 and SLNs@Fe3O4-MnO) was compared with the non-treated (NT) cells. The treatment with Figure 75 - Comparison of passive versus MHT-induced release of doxorubicin during 4 h. Values represent mean ±SEM. *p<0.05; **p<0.01; ***p<0.001;****p<0.0001. 92 Cells treated with mSLNs without drug did not exhibit relevant toxicities, suggesting that the particles are safe for administration in clinically relevant dosages. Also, these lipidic NPs are efficiently internalized by cells, and the release and action of doxorubicin in the nucleus was confirmed. The encapsulation of dox intensified its effect in breast cancer cells. Further studies are required to unveil the mechanism behind this increased efficiency, but studies from the group with mSLNs point towards a faster delivery of doxorubicin to nuclei. The final mSLNs (SLNs@Fe3O4-MnO-dox) showed their capability as MHT effectors, since cell viability was significantly decreased when cells treated with these NPs were subjected to a AMF. Therefore, the main objective of this dissertation was fulfilled, since the presented bionanosystem presents a good performance as a dual MRI contrast agent, externally controlled drug delivery vehicle and MHT effector. The synergistic abilities of the nanocomposite represent a step forward for the treatment of solid malignant tumours, which may realize the dual goals of reduced systemic toxicity and enhanced anti-tumoral efficacy. 5.2. Future perspectives The results obtained raise new interesting questions about what could be optimised in the various steps of this work, and also what more could be done to validate the developed system as a new tool in the fight against cancer. It would be ideal to optimize the synthesis of the SLNs, making it more systematic in a way that the concentration of the encapsulated magnetic content could be accurately reproducible. This could be achieved by using a more stable melting mechanism, since this is a key step on the SLNs synthesis. Since experiments with encapsulated manganese dioxide NPs corroborated their responsive behaviour, it would be ideal to optimize the physico-chemical properties of the responsive SLNs. A new synthesis method of manganese dioxide NPs should be experimented, so it would be possible to obtain hydrophobic MnO2 NPs. This way the encapsulation of these particles together with magnetite would be more efficient, and the physico-chemical properties of the obtained SLNs would be more appropriate. It would also be interesting to evaluate the response of MnO2 NPs to pH changes, since the tumoral environment is an acidic one and MnO2 has been reported to be very sensitive to pH. Drug release experiments should be performed at least during 8 h to achieve a more complete drug release profile, and more MHT schemes should be tried to be able to induce significant drug release in 93 less time. Also, a fitting of the drug release curve with different mathematical models should be done to evaluate the type of release witnessed in each scenario. Doxorubicin was used as the proof of principle for these experiments, but it would also be interesting to try to encapsulate different chemotherapeutic drugs into the SLNs, even to encapsulate two or more drugs into the same SLNs. Furthermore, studying the MHT effects of the synthetized NPs in different cell lines is very important to validate the formulation. 94 95 CHAPTER 6 - BIBLIOGRAPHY [1] International Agency for research on Cancer - World Health Organization, “Cancer today,” Global Cancer Observatory , 2019. [Online]. Available: http://gco.iarc.fr/. [Accessed: 09-Jun-2019]. [2] A. Grillone et al. , “Active Targeting of Sorafenib: Preparation, Characterization, and In Vitro Testing of Drug-Loaded Magnetic Solid Lipid Nanoparticles,” Adv. Healthc. Mater. , vol. 4, no. 11, pp. 1681–1690, 2015. [3] J. Gallo, N. J. Long, and E. O. Aboagye, “Magnetic nanoparticles as contrast agents in the diagnosis and treatment of cancer.,” Chem. Soc. Rev. , vol. 42, no. 19, pp. 7816–7833, 2013. [4] M. Bañobre-López, A. Teijeiro, and J. Rivas, “Magnetic nanoparticle-based hyperthermia for cancer treatment,” Reports Pract. Oncol. Radiother. , vol. 18, no. 6, pp. 397–400, 2013. [5] M. Bañobre-Lopez, L. Garcia-Hevia, F. Cerqueira, F. Rivadulla, and J. Gallo, “Tunable performance of manganese oxide nanostructures as MRI contrast agents,” Chem. - A Eur. J. , vol. 24, no. 6, p. 1295, 2018. [6] J. Gallo, N. Vasimalai, M. T. Fernandez-Arguelles, and M. Bañobre-López, “Green synthesis of multimodal ‘OFF-ON’ activatable MRI/optical probes,” Dalt. Trans. , vol. 45, no. 44, pp. 17672– 17680, 2016. [7] National Cancer Institute, “What Is Cancer?,” 2015. [Online]. Available: https://www.cancer.gov/about-cancer/understanding/what-is-cancer. [Accessed: 10-May2018]. [8] National Cancer Institute, “Risk Factors for Cancer,” 2015. [Online]. Available: https://www.cancer.gov/about-cancer/causes-prevention/risk. [Accessed: 29-Jun-2018]. [9] PDQ Screening and Prevention Editorial Board, Cancer Prevention Overview (PDQ®): Patient Version . 2002. [10] International Agency for Research on Cancer - World Health Organization., “Estimated agestandardized incidence rates (World) in 2018, all cancers, both sexes, all ages,” Global Cancer Observatory , 2018. [Online]. Available: http://gco.iarc.fr/today/online-analysis-map. [Accessed: 31-Jul-2019]. [11] World Health Organization, “Cancer,” Feb-2018. [Online]. Available: http://www.who.int/en/news-room/fact-sheets/detail/cancer. [Accessed: 10-May-2018]. [12] N. Miranda et al. , “Portugal - Doenças Oncológicas em Números.” Direção Geral de Saúde, Lisboa, pp. 7–9, 2015. [13] P. Boyle and B. Levin, Mechanisms of carcinogenesis , vol. 159, no. 2. 2008. [14] R. Mehta, “The potential for the use of cell proliferation and oncogene expression as intermediate markers during liver carcinogenesis,” Cancer Lett. , vol. 93, no. 1, pp. 85–102, 1995. [15] C. Sonnenschein and A. Soto, “Carcinogenesis explained within the context of a theory of organisms,” Prog. Biophys. Mol. Biol. , vol. 122, no. 1, pp. 70–76, Oct. 2016. [16] S. Reuter, S. C. Gupta, M. M. Chaturvedi, and B. B. Aggarwal, “Oxidative stress, inflammation, and cancer: How are they linked?,” Free Radic. Biol. Med. , vol. 49, no. 11, pp. 1603–1616, 2010. 96 [17] National Cancer Institute, “Treatment for Cancer,” 2015. [Online]. Available: https://www.cancer.gov/about-cancer/treatment. [Accessed: 10-May-2018]. [18] C. Rébé and F. Ghiringhelli, “Cytotoxic effects of chemotherapy on cancer and immune cells: how can it be modulated to generate novel therapeutic strategies?,” Futur. Oncol. , vol. 11, no. 19, pp. 2645–2654, Oct. 2015. [19] A. Urruticoechea, R. Alemany, J. Balart, A. Villanueva, F. Vinals, and G. Capella, “Recent Advances in Cancer Therapy: An Overview,” Curr. Pharm. Des. , vol. 16, no. 1, pp. 3–10, Jan. 2010. [20] S. S. Lucky, K. C. Soo, and Y. Zhang, “Nanoparticles in photodynamic therapy,” Chem. Rev. , vol. 115, no. 4, pp. 1990–2042, 2015. [21] J. W. Park, “Liposome-based drug delivery in breast cancer treatment,” Breast Cancer Res. , vol. 4, no. 3, p. 95, Jun. 2002. [22] J. Beik et al. , “Nanotechnology in hyperthermia cancer therapy: From fundamental principles to advanced applications,” J. Control. Release , vol. 235, pp. 205–221, 2016. [23] M. Johannsen, B. Thiesen, P. Wust, and A. Jordan, “Magnetic nanoparticle hyperthermia for prostate cancer,” Int. J. Hyperth. , vol. 26, no. 8, pp. 790–795, 2010. [24] M. Johannsen et al. , “Clinical hyperthermia of prostate cancer using magnetic nanoparticles: Presentation of a new interstitial technique,” Int. J. Hyperth. , vol. 21, no. 7, pp. 637–647, 2005. [25] K. Maier-Hauff et al. , “Efficacy and safety of intratumoral thermotherapy using magnetic iron-oxide nanoparticles combined with external beam radiotherapy on patients with recurrent glioblastoma multiforme,” J. Neurooncology , vol. 103, no. 2, pp. 317–324, 2011. [26] P. Kaur, M. D. Hurwitz, S. Krishnan, and A. Asea, “Combined hyperthermia and radiotherapy for the treatment of cancer.,” Cancers (Basel). , vol. 3, no. 4, pp. 3799–823, Sep. 2011. [27] S. Senapati, A. K. Mahanta, S. Kumar, and P. Maiti, “Controlled drug delivery vehicles for cancer treatment and their performance,” Signal Transduct. Target. Ther. , vol. 3, no. 1, p. 7, 2018. [28] E. Drexler, “‘There’s Plenty of Room at the Bottom’ (Richard Feynman, Pasadena, 29 December 1959),” Metamodern - The Trajectory of Technology , 2009. [Online]. Available: http://metamodern.com/2009/12/29/theres-plenty-of-room-at-the-bottom”-feynman-1959/. [Accessed: 29-May-2018]. [29] K. E. Drexler and M. Minsky, Engines Of Creation: The coming Era of Technology , 1st ed. New York: Anchor books: Doubleday, 1986. [30] E. Boysen and R. Booker, Nanotechnology for dummies , 2nd ed. Indianapolis: Wiley Publishing, Inc., 2005. [31] G. Gregoriadis, “Drug entrapment in liposomes,” FEBS Lett. , vol. 36, no. 3, pp. 292–296, 1973. [32] P. A. Kramer, “Albumin microspheres as vehicles for achieving specificity in drug delivery,” Journal of Pharmaceutical Sciences , vol. 63, no. 10. pp. 1646–1647, 1974. [33] V. Wagner, A. Dullaart, A. Bock, and A. Zweck, “The emerging nanomedicine landscape,” Nat. Biotechnol. , vol. 24, no. 10, pp. 1211–1217, 2006. [34] Grand View Research, “Controlled Release Drug Delivery Market Analysis Report By Technology (Implants, Transdermal, Microencapsulation, Targeted Delivery), By Release Mechanism, By 97 Application, And Segment Forecasts, 2018 - 2025,” 2018. [35] M. Rajabi and S. A. Mousa, “Lipid Nanoparticles and their Application in Nanomedicine,” Curr. Pharm. Biotechnol. , vol. 17, no. 8, pp. 662–672, May 2016. [36] J. R. Kanwar, R. R. Mohan, R. K. Kanwar, K. Roy, and R. Bawa, “Applications of aptamers in nanodelivery systems in cancer, eye and inflammatory diseases.,” Nanomedicine (Lond). , vol. 5, no. 9, pp. 1435–1445, 2010. [37] A. S. Hoffman, “The origins and evolution of ‘controlled’ drug delivery systems,” J. Control. Release , vol. 132, no. 3, pp. 153–163, 2008. [38] Y. Malam, M. Loizidou, and A. M. Seifalian, “Liposomes and nanoparticles: nanosized vehicles for drug delivery in cancer,” Trends Pharmacol. Sci. , vol. 30, no. 11, pp. 592–599, Nov. 2009. [39] K. ichi Ogawara, K. Un, K. ichi Tanaka, K. Higaki, and T. Kimura, “In vivo anti-tumor effect of PEG liposomal doxorubicin (DOX) in DOX-resistant tumor-bearing mice: Involvement of cytotoxic effect on vascular endothelial cells,” J. Control. Release , vol. 133, no. 1, pp. 4–10, 2009. [40] B. Issa, I. M. Obaidat, B. A. Albiss, and Y. Haik, “Magnetic nanoparticles: Surface effects and properties related to biomedicine applications,” Int. J. Mol. Sci. , vol. 14, pp. 21266–21305, 2013. [41] L. Kafrouni and O. Savadogo, “Recent progress on magnetic nanoparticles for magnetic hyperthermia,” Prog. Biomater. , vol. 5, no. 3–4, pp. 147–160, 2016. [42] L. R. McDowell, Minerals in animal and human nutrition. , 2nd ed. Elsevier Science, 2003. [43] N. Abbaspour, R. Hurrell, and R. Kelishadi, “Review on iron and its importance for human health,” J. Res. Med. Sci. , vol. 19, no. 2, pp. 164–174, 2014. [44] S. Puig, L. Ramos-Alonso, A. M. Romero, and M. T. Martínez-Pastor, “The elemental role of iron in DNA synthesis and repair,” Metallomics , vol. 9, no. 11, pp. 1483–1500, 2017. [45] WHO, “WHO/CDC expert consultation agrees on best indicators to assess iron deficiency, a major cause of anaemia,” Media center , 2004. [Online]. Available: https://www.who.int/mediacentre/news/notes/2004/anaemia/en/?fbclid=IwAR0u_RoM1Zuc7 6tkoP2sRd93IjKkW7r42nN3mppcoR9BJrcheP53pTqYQm0. [Accessed: 10-Aug-2019]. [46] R. Kammari, N. G. Das, and S. K. Das, “Nanoparticulate Systems for Therapeutic and Diagnostic Applications,” in Emerging Nanotechnologies for Diagnostics, Drug Delivery and Medical Devices , 1st ed., Cambridge: Elsevier, 2017, pp. 105–144. [47] A. K. Gupta and M. Gupta, “Synthesis and surface engineering of iron oxide nanoparticles for biomedical applications,” Biomaterials , vol. 26, no. 18, pp. 3995–4021, Jun. 2005. [48] A. H. Lu, E. L. Salabas, and F. Schüth, “Magnetic nanoparticles: Synthesis, protection, functionalization, and application,” Angewandte Chemie - International Edition , vol. 46, no. 8. pp. 1222–1244, 2007. [49] G. F. Goya and M. P. Morales, “Field Dependence of Blocking Temperature in Magnetite Nanoparticles,” J. Metastable Nanocrystalline Mater. , vol. 20–21, pp. 673–678, 2004. [50] A. Jordan et al. , “Presentation of a new magnetic field therapy system for the treatment of human solid tumors with magnetic fluid hyperthermia,” J. Magn. Magn. Mater. , vol. 225, no. 1–2, pp. 118–126, 2001. 98 [51] S. M. Moghimi, A. C. Hunter, and J. C. Murray, “Long-circulating and target-specific nanoparticles: theory to practice.,” Pharmacol. Rev. , vol. 53, no. 2, pp. 283–318, 2001. [52] M. Mahmoudi, S. Sant, B. Wang, S. Laurent, and T. Sen, “Superparamagnetic iron oxide nanoparticles (SPIONs): Development, surface modification and applications in chemotherapy,” Adv. Drug Deliv. Rev. , vol. 63, no. 1–2, pp. 24–46, Jan. 2011. [53] K. Wu et al. , “Functionalized magnetic iron oxide/alginate core-shell nanoparticles for targeting hyperthermia,” Int. J. Nanomedicine , vol. 10, p. 3315, May 2015. [54] C. Lundquist, C. Loo, I. Meraz, J. Cerda, X. Liu, and R. Serda, “Characterization of Free and Porous Silicon-Encapsulated Superparamagnetic Iron Oxide Nanoparticles as Platforms for the Development of Theranostic Vaccines,” Med. Sci. , vol. 2, no. 1, pp. 51–69, Feb. 2014. [55] R. Wang, Y. Hu, N. Zhao, and F. J. Xu, “Well-Defined Peapod-like Magnetic Nanoparticles and Their Controlled Modification for Effective Imaging Guided Gene Therapy,” ACS Appl. Mater. Interfaces , vol. 8, no. 18, pp. 11298–11308, 2016. [56] J. He, M. Huang, D. Wang, Z. Zhang, and G. Li, “Magnetic separation techniques in sample preparation for biological analysis: A review,” J Pharm Biomed Anal , vol. 101, pp. 84–101, 2014. [57] S. Chandra, S. Nigam, and D. Bahadur, “Combining unique properties of dendrimers and magnetic nanoparticles towards cancer theranostics,” J. Biomed. Nanotechnol. , vol. 10, no. 1, pp. 32–49, 2014. [58] B. S., W. Y., and H. X., “Nanotechnology for enrichment and detection of circulating tumor cells,” Nanomedicine , vol. 10, no. 12, pp. 1973–1990, 2015. [59] J. L. Greger, “Dietary Standards for Manganese: Overlap between Nutritional and Toxicological Studies,” J. Nutr. , vol. 128, pp. 368S-371S, 1998. [60] K. M. Erikson, T. Syversen, J. L. Aschner, and M. Aschner, “Interactions between excessive manganese exposures and dietary iron-deficiency in neurodegeneration,” Environ. Toxicol. Pharmacol. , vol. 19, no. 3, pp. 415–421, May 2005. [61] J. L. Aschner and M. Aschner, “Nutritional aspects of manganese homeostasis,” Mol. Aspects Med. , vol. 26, pp. 353–362, 2005. [62] J. S. Schneider, E. Decamp, A. J. Koser, S. Fritz, T. Syversen, and T. R. Guilarte, “Effects of Chronic Manganese Exposure on Cognitive and Motor Functioning in Non-Human Primates,” Brain Res. Rev. , vol. 1118, no. 1, pp. 222–231, 2006. [63] L. García-Hevia, M. Bañobre-Lopez, and J. Gallo, “Recent progress on manganese-based nanostructures as responsive MRI contrast agents,” Chem. - A Eur. J. , vol. 24, no. 2, pp. 2–13, 2018. [64] G. B. Gerber and A. Le, “Carcinogenicity , mutagenicity and teratogenicity of manganese compounds,” Oncol. Hematol. , vol. 42, pp. 25–34, 2002. [65] A. Takeda, “Manganese action in brain function,” Brain Res. Rev. , vol. 41, pp. 79–87, 2003. [66] J. Huang et al. , “HSA coated MnO nanoparticles with prominent MRI contrast for tumor imaging.,” Chem. Commun. (Camb). , vol. 46, no. 36, pp. 6684–6, Sep. 2010. [67] H. Bin Na et al. , “Development of a T1 contrast agent for magnetic resonance imaging using MnO 99 nanoparticles,” Angew. Chemie - Int. Ed. , no. 46, pp. 5397–5401, 2007. [68] J. Gallo, I. S. Alam, I. Lavdas, M. Wylezinska-Arridge, E. O. Aboagye, and N. J. Long, “RGD-targeted MnO nanoparticles as T 1 contrast agents for cancer imaging – the effect of PEG length in vivo,” J. Mater. Chem. B , vol. 2, no. 7, pp. 868–876, Jan. 2014. [69] C. Tu, E. A. Osborne, and A. Y. Louie, “Synthesis and characterization of a redoxand lightsensitive MRI contrast agent.,” Tetrahedron , vol. 65, no. 7, p. 1241, Feb. 2009. [70] M. M. Ali et al. , “Synthesis and relaxometric studies of a dendrimer-based pH-responsive MRI contrast agent.,” Chem. - A Eur. J. , vol. 14, no. 24, pp. 7250–8, Aug. 2008. [71] W. Zhu et al. , “Modulation of Hypoxia in Solid Tumor Microenvironment with MnO 2 Nanoparticles to Enhance Photodynamic Therapy,” Adv. Funct. Mater. , vol. 26, no. 30, pp. 5490–5498, Aug. 2016. [72] T. Lin, “Oxygen generating manganese dioxide nanoparticles for enhanced photodynamic therapy to bladder cancer by ameliorating hypoxia,” Eur. Urol. Suppl. , vol. 17, no. 2, p. e664, Mar. 2018. [73] A. Z. Abbasi et al. , “Hybrid manganese dioxide nanoparticles potentiate radiation therapy by modulating tumor hypoxia,” Cancer Res. , vol. 76, no. 22, pp. 6643–6656, 2016. [74] P. Prasad et al. , “Multifunctional albumin-MnO2nanoparticles modulate solid tumor microenvironment by attenuating hypoxia, acidosis, vascular endothelial growth factor and enhance radiation response,” ACS Nano , vol. 8, no. 4, pp. 3202–3212, 2014. [75] A. B. Chin and I. I. Yaacob, “Synthesis and characterization of magnetic iron oxide nanoparticles via w/o microemulsion and Massart’s procedure,” J. Mater. Process. Technol. , vol. 191, no. 1– 3, pp. 235–237, 2007. [76] E. H. Kim, H. S. Lee, B. K. Kwak, and B. K. Kim, “Synthesis of ferrofluid with magnetic nanoparticles by sonochemical method for MRI contrast agent,” in Journal of Magnetism and Magnetic Materials , 2005, vol. 289, pp. 328–330. [77] M. Kimata, D. Nakagawa, and M. Hasegawa, “Preparation of monodisperse magnetic particles by hydrolysis of iron alkoxide,” Powder Technol. , vol. 132, no. 2–3, pp. 112–118, 2003. [78] S. Basak, D. R. Chen, and P. Biswas, “Electrospray of ionic precursor solutions to synthesize iron oxide nanoparticles: Modified scaling law,” Chem. Eng. Sci. , vol. 62, no. 4, pp. 1263–1268, 2007. [79] W. Wu, Q. He, and C. Jiang, “Magnetic iron oxide nanoparticles: synthesis and surface functionalization strategies.,” Nanoscale Res. Lett. , vol. 3, no. 11, pp. 397–415, Oct. 2008. [80] S. Laurent et al. , “Magnetic Iron Oxide Nanoparticles: Synthesis, Stabilization, Vectorization, Physicochemical Characterizations, and Biological Applications,” Chem. Rev. , vol. 108, no. 6, pp. 2064–2110, Jun. 2008. [81] P. Tartaj, M. P. Morales, S. Veintemillas-Verdaguer, T. Gonzalez-Carreño, and C. J. Serna, “Synthesis, Properties and Biomedical Applications of Magnetic Nanoparticles,” in Handbook of magnetic materials , 191st ed., Amsterdam: Elsevier, 2007, pp. 235–237. [82] J. Lee, T. Isobe, and M. Senna, “Magnetic properties of ultrafine magnetite particles and their slurries prepared via in-situ precipitation,” Colloids Surfaces Physicochem. Eng. Asp. , vol. 109, pp. 121–127, Apr. 1996. 100 [83] Y. V. Kolenko et al. , “High-temperature magnetism as a probe for structural and compositional uniformity in ligand-capped magnetite nanoparticles,” J. Phys. Chem. C , vol. 118, no. 48, pp. 28322–28329, 2014. [84] DRUGBANK, “Doxorubicin,” 2019. [Online]. Available: https://www.drugbank.ca/drugs/DB00997. [Accessed: 04-Aug-2019]. [85] A. C. Alves et al. , “Influence of doxorubicin on model cell membrane properties: Insights from in vitro and in silico studies,” Sci. Rep. , vol. 6343, no. 7, pp. 1–11, 2017. [86] A. Kumar, B. Gautam, C. Dubey, and P. K. Tripathi, “A Review: Role of Doxorubicin in Treatment of Cancer,” Int. J. Pharm. Sci. Res. , vol. 5, no. 10, p. 4105, 2014. [87] P. A. J. Speth, Q. G. C. M. van Hoesel, and C. Haanen, “Clinical Pharmacokinetics of Doxorubicin,” Clin. Pharmacokinet. , vol. 15, no. 1, pp. 15–31, 1988. [88] F. Suzuki et al. , “Induction of tumor-specific cytotoxicity and apoptosis by doxorubicin,” Anticancer Res. , vol. 25, no. 2A, pp. 887–894, 2005. [89] O. Tacar, P. Sriamornsak, and C. R. Dass, “Doxorubicin: an update on anticancer molecular action, toxicity and novel drug delivery systems,” J. Pharm. Pharmacol. , vol. 65, no. 2, pp. 157– 170, 2013. [90] L. Qiu, N. Jing, and Y. Jin, “Preparation and in vitro evaluation of liposomal chloroquine diphosphate loaded by a transmembrane pH-gradient method,” Int. J. Pharm. , vol. 361, no. 1–2, pp. 56–63, 2008. [91] Y. M. Ning et al. , “Liposomal doxorubicin in combination with bortezomib for relapsed or refractory multiple myeloma,” Oncology , vol. 21, no. 12, pp. 1503–1508, 2007. [92] V. Sharma, P. Saxena, L. Singh, and P. Singh, “Self emulsifying drug delivery system: a novel approach.,” J. Pharm. Res. , vol. 5, no. 1, pp. 500–504, 2012. [93] M. I. Alam et al. , “Strategy for effective brain drug delivery,” Eur. J. Pharm. Sci. , vol. 40, no. 5, pp. 385–403, 2010. [94] T. Bezerra Soares et al. , “Lipid nanocarriers loaded with natural compounds: Potential new therapies for age related neurodegenerative diseases?,” Prog. Neurobiol. , 2018. [95] K. Mäder and W. Mehnert, “Solid lipid nanoparticles: production, characterization and applications.,” Adv. Drug Deliv. Rev. , vol. 47, no. 2–3, pp. 165–96, 2001. [96] L. Battaglia et al. , “Techniques for the Preparation of Solid Lipid Nano and Microparticles,” in Application of Nanotechnology in Drug Delivery , 1st ed., InTech, 2014, pp. 51–75. [97] M. Üner and G. Yener, “Importance of solid lipid nanoparticles (SLN) in various administration routes and future perspective,” Int. J. Nanomedicine , vol. 2, no. 3, pp. 289–300, 2007. [98] H. L. Wong, R. Bendayan, A. M. Rauth, Y. Li, and X. Y. Wu, “Chemotherapy with anticancer drugs encapsulated in solid lipid nanoparticles,” Adv. Drug Deliv. Rev. , vol. 59, no. 6, pp. 491–504, 2007. [99] S. . Wissing, O. Kayser, and R. . Müller, “Solid lipid nanoparticles for parenteral drug delivery,” Adv. Drug Deliv. Rev. , vol. 56, no. 9, pp. 1257–1272, 2004. [100] J. Pardeike, A. Hommoss, and R. H. Müller, “Lipid nanoparticles (SLN, NLC) in cosmetic and 101 pharmaceutical dermal products,” International Journal of Pharmaceutics , vol. 366, no. 1–2. pp. 170–184, 2009. [101] K. W. Kang et al. , “Doxorubicin-loaded solid lipid nanoparticles to overcome multidrug resistance in cancer therapy,” Nanomedicine Nanotechnology, Biol. Med. , vol. 6, no. 2, pp. 210–213, Apr. 2010. [102] R. K. Subedi, K. W. Kang, and H. K. Choi, “Preparation and characterization of solid lipid nanoparticles loaded with doxorubicin,” Eur J Pharm Sci , vol. 37, no. 3–4, pp. 508–513, 2009. [103] S. V. Mussi, R. C. Silva, M. C. de Oliveira, C. M. Lucci, R. B. de Azevedo, and L. A. M. Ferreira, “New approach to improve encapsulation and antitumor activity of doxorubicin loaded in solid lipid nanoparticles,” Eur. J. Pharm. Sci. , vol. 48, no. 1–2, pp. 282–290, Jan. 2013. [104] L. Battaglia et al. , “Solid Lipid Nanoparticles for Potential Doxorubicin Delivery in Glioblastoma Treatment: Preliminary In Vitro Studies,” J. Pharm. Sci. , vol. 103, no. 7, pp. 2157–2165, Jul. 2014. [105] B. M. Dale, M. A. Brown, and R. C. Semelka, MRI: Basic Principles and Applications , 5th ed. Chichester, UK: John Wiley & Sons, Ltd, 2015. [106] J. Rivas, Y. V. Kolen’ko, and M. Bañobre-López, “Magnetic Nanocolloids,” in Nanocolloids: A Meeting Point for Scientists and Technologists , M. Sánchez-Domínguez and C. Rodríguez-Abreu, Eds. Elsevier, 2016, pp. 75–120. [107] H. Yoshioka, P. M. Schlechtweg, and K. Kose, “Magnetic Resonance Imaging,” in Imaging of Arthritis and Metabolic Bone Disease , B. N. Weissman, Ed. Elsevier, 2009, pp. 34–48. [108] M. Bañobre-López, Y. Piñeiro, M. A. López-Quintela, and J. Rivas, “Magnetic Nanoparticles for Biomedical Applications,” in Handbook of Nanomaterials Properties , B. Bhushan, D. Luo, S. R. Schricker, W. Sigmund, and S. Zauscher, Eds. Berlin, Heidelberg: Springer, 2014, pp. 457–493. [109] H. Bin Na and T. Hyeon, “Nanostructured T1 MRI contrast agents,” J. Mater. Chem. , vol. 19, no. 35, pp. 6267–6273, 2009. [110] T. H. Shin et al. , “T 1 and T 2 dual-mode MRI contrast agent for enhancing accuracy by engineered nanomaterials,” ACS Nano , vol. 8, no. 4, pp. 3393–3401, Apr. 2014. [111] N. A. Keasberry, M. Bañobre-López, C. Wood, G. J. Stasiuk, J. Gallo, and N. J. Long, “Tuning the relaxation rates of dual-mode T 1 / T 2 nanoparticle contrast agents: a study into the ideal system,” Nanoscale , vol. 7, no. 38, pp. 16119–16128, Sep. 2015. [112] N. Lee and T. Hyeon, “Designed synthesis of uniformly sized iron oxide nanoparticles for efficient magnetic resonance imaging contrast agents,” Chemical Society Reviews , vol. 41, no. 7. pp. 2575–2589, 07-Apr-2012. [113] N. Guldris et al. , “Influence of the separation procedure on the properties of magnetic nanoparticles: Gaining in vitro stability and T1 -T2 magnetic resonance imaging performance,” J. Colloid Interface Sci. , vol. 472, pp. 229–236, Jun. 2016. [114] Y.-K. Peng, S. C. E. Tsang, and P.-T. Chou, “Chemical design of nanoprobes for T 1-weighted magnetic resonance imaging,” Mater. Today , vol. 19, no. 6, 2015. [115] A. Neuwelt, N. Sidhu, C. A. A. Hu, G. Mlady, S. C. Eberhardt, and L. O. Sillerud, “Iron-based superparamagnetic nanoparticle contrast agents for MRI of infection and inflammation,” Am. J. 108 [215] A. Lanz-Landázuri, A. Martínez De Ilarduya, M. García-Alvarez, and S. Muñoz-Guerra, “Poly(β,Lmalic acid)/Doxorubicin ionic complex: A pH-dependent delivery system,” React. Funct. Polym. , vol. 81, no. 1, pp. 45–53, 2014. [216] I. Uddin, P. Poddar, and A. Ahmad, “Extracellular Biosynthesis of Water Dispersible, Protein Capped Mn5O8 Nanoparticles Using the Fungus Fusarium oxysporum and Study of Their Magnetic Behavior,” J. Nanoeng. Nanomanufacturing , vol. 3, no. 2, pp. 91–97, Mar. 2013. [217] ThermoFisher, “Fluorescence SpectraViewer - PT,” DiO . [Online]. Available: https://www.thermofisher.com/pt/en/home/life-science/cell-analysis/labelingchemistry/fluorescence-spectraviewer.html?SID=srch-svtool&UID=275lip. [Accessed: 29-Sep2019]. [218] A. Puthirath Balan et al. , “Magnetic Properties and Photocatalytic Applications of 2D Sheets of Nonlayered Manganese Telluride by Liquid Exfoliation,” ACS Appl. Nano Mater. , vol. 1, no. 11, pp. 6427–6434, Nov. 2018. [219] M. Balamurugan, G. Venkatesan, S. Ramachandran, and S. Saravanan, “Synthesis and Characterization of Manganese Oxide Nanoparticles,” in Synthesis and Fabrication of Nanomaterials , V. Rajendran, P. Paramasivam, and K. E. Geckeler, Eds. Bloomsbury Publishing India, 2015, pp. 311–314. [220] E. Pretsch, J. Seibl, and W. Simon, “Tablas combinadas,” in Tablas para la determinación estructural por métodos espectroscópicos , 3rd ed., A. Fernandéz and R. Álvarez, Eds. Springer - Verlag Ibérica, 1998, pp. 16–50. [221] M. Rohrer, H. Bauer, J. Mintorovitch, M. Requardt, and H. J. Weinmann, “Comparison of magnetic properties of MRI contrast media solutions at different magnetic field strengths,” Invest. Radiol. , vol. 40, no. 11, pp. 715–724, Nov. 2005. [222] V. Nandwana et al. , “Theranostic magnetic nanostructures (MNS) for Cancer,” Cancer Treat. Res. , vol. 166, pp. 51–83, 2015. 109 ANNEXES A B C D A B C Annex 2 - Relaxation rates versus ion concentrations graphs for SLNs@MnO (A), SLNs@Fe3O4 (B) and SLNs@Fe3O4-MnO-dox (C) and (D). The slope in each graph represents the longitudinal and transverse relaxations times at a magnetic field of 3T. Annex 1 - Relaxation rates versus ion concentrations graphs for SLNs@MnO3 (A) and SLNs@Fe3O4-MnO3-dox (B) and (C), in water and hydrogen peroxide. The slopes in each graph represents the longitudinal and transverse relaxations times at a magnetic field of 3T. 110 A B C D Annex 3 - Micrographs representative of cell viability of Hs578T cells treated with SLNs@Fe3O4 (A), SLNs@Fe3O4-MnO (B), SLNs@MnO (C), SLNs@MnO2 (D) at a ion concentration of 1.5 µg/mL (100x magnification). 111 DAPI DiO Dox α-tubulin Merge NT SLNs@Fe 3 O 4 -MnOdox Annex 4 - Confocal micrographs of Hs570T cells non-treated (NT) and after exposure to SLNs@Fe3O4-MnO-dox. Scale bar corresponds to 20 µm.