Development of hydroxyapatite-based nanoparticles for medical imaging diagnosis

Faculdade de Engenharia da Universidade do Porto Development of Hydroxyapatite-based Nanoparticles for Medical Imaging Diagnosis Ana Azevedo Moço Dissertação realizada no âmbito do Mestrado Integrado em Bioengenharia para candidatura à obtenção do grau de Mestre em Engenharia Biomédica Orientador:____________________________ Co-Orientador:_________________________ Porto, Junho 2013 ii © Ana Azevedo Moço, 2013 iii This thesis was supervised by: Supervisor Dr. Marta Laranjeira INEB – Instituto de Engenharia Biomédica, Universidade do Porto FEUP – Faculdade de Engenharia da Universidade do Porto Co-supervisor Professor Dr. Fernando Jorge Monteiro INEB – Instituto de Engenharia Biomédica, Universidade do Porto FEUP – Faculdade de Engenharia da Universidade do Porto The research described in this thesis was conducted at: INEB – Instituto de Engenharia Biomédica, Divisão de Biomateriais, Universidade do Porto, Portugal; FEUP – Faculdade de Engenharia da Universidade do Porto, Portugal. iv v Resumo O cancro é ainda uma das principais causas de morte e é facto que um diagnóstico precoce aumenta significativamente as possibilidades de sobrevivência do paciente, otimizando a eficácia do tratamento. Este objectivo pode ser atingido através de um equipamento adequado de detecção e um agente de contraste. A Ressonância Magnética (RM) é uma técnica clinica de imagem com alta resolução espacial e contraste dos tecidos moles que permite uma visualização não-invasiva e tomográfica de estruturas anatómicas. É hoje aceite que agentes de contraste magnéticos de dimensão nanométrica são capazes de maximizar a sensibilidade da RM encurtando os tempos de relaxividade dos protões da água presentes nas áreas dos tecidos onde o agente se acumula. Estes agentes caracterizam-se por apresentarem vantagens distintas em relação aos agentes de contraste convencionais, nomeadamente pela sua elevada área de superfície, pela possibilidade de entrega específica a uma localização tumoral-alvo e pela possibilidade de projetar o seu tempo de circulação sanguíneo ou outras características relevantes. A citotoxicidade é ainda uma questão importante que limita a aplicação clínica de muitos nano-agentes de contraste propostos até agora. A hidroxiapatite é um composto biocerâmico que não só incorpora iões facilmente, mas também confere biocompatibilidade ao sistema, o que faz a sua conjugação com iões magnéticos uma abordagem apelativa na imagem localizada de cancro. Tentativas consideráveis têm sido realizadas no âmbito deste tema, mas existe ainda uma necessidade para estandardizar procedimentos e comparar sistemas nano capazes de realçar o contraste tumoral e potencialmente poderiam permitir uma detecção específica de cancro. No presente trabalho, nanoparticulas de hidroxiapatite dopadas com iões magnéticos foram produzidas com êxito por um método otimizado e reproduzível de precipitação química. Apurou-se que todas as partículas de nanohidroxiapatite dopadas de Gd, Fe(II), Fe(III) e Co possuíam a forma de bastonete e um tamanho médio entre 30 e 60nm. A estrutura principal de hidroxiapatite não foi alterada pela percentagem de dopagem embora esta se tenha provado como diretamente proporcional aos níveis de momento magnético da amostra. Hidroxiapatite dopada com 10% Gd apresentou os níveis mais altos de vi momento magnético entre todos os iões dopantes e também os valores mais próximos dos de um agente de contraste já comercializado. A viabilidade e a morfologia de células humanas microvasculares endoteliais da derme não sugerem uma alteração significativa após incubação com os materiais sintetizados. Com efeito, os resultados mostram que as nanoparticulas de hidroxiapatite dopadas com 10% Gd são as mais promissoras dentro do grupo testado e, portanto, podem tornar-se um valioso biomaterial para agente de contraste de RM. vii Abstract Cancer is still one of the main causes of death and it is known that an early diagnostic significantly increases chances of survival of the patient, optimizing the efficacy of the treatment. This can be achieved with an appropriate detection device and contrast agent. Magnetic Resonance Imaging (MRI) is a clinical imaging technique that allows for noninvasive tomographic visualization of anatomic structures with high spatial resolution and soft tissue contrast. It is well accepted that nanosized magnetic contrast agents can maximize MRI sensitivity shortening the relaxation times of bulk water protons in tissue where the agent accumulates. They are characterized by distinct advantages over conventional contrast agents namely their high surface area, possibility of delivery to a specific cancer site by active targeting and the possibility of engineering their blood circulation half-life, among others. Cytotoxicity is still a major issue that limits the clinical application of many nano-contrast agents proposed so far. Bioceramic hydroxyapatite not only easily incorporates ions but also confers biocompatibility to the system which makes its conjugation with magnetically responsive ions an appealing approach in targeted cancer imaging. Considerable attempts have been made within this subject but there is a need to standardize procedures and compare nanosized systems capable of enhancing tumor contrast that would hereafter allow targeted cancer detection. In the present work, magnetic ion doped hydroxyapatite nanoparticles were successfully produced by an optimized and reproducible wet chemical precipitation method. It was found that all Gd, Fe(II), Fe(III) and Co doped nanohydroxyapatite particles were rod-like shaped and of average size between 30 and 60nm. Main hydroxyapatite structure was not altered by the percentage of doping although this percentage was proven to be directly proportional to the magnetic moment levels of the sample. 10% Gd doped HAP presented the highest magnetic moment levels between all doping ions and also the closest values to the ones of a commercialized contrast agent. Viability and morphology of human dermal microvascular endothelial cells did not seem to be significantly altered after incubation with the synthesized materials. viii In effect, the results show that 10% Gd doped nanohydroxyapatite is the most promising from the group of nanoparticles tested and therefore can be a valuable biomaterial contrast agent for MR imaging. ix Expect problems and eat them for breakfast. Alfred A. Montapert xvi xvii List of tables Table 1.1 Summary of distinct features of benign and malignant tumors (adapted from [5]) .............. 22 Table 1.2 Non-exhaustive list of MRI approved contrast agents used in clinical practice (Adapted from [16], [28] and [29]) ...................................................................................... 27 Table 1.3 A summary of hydroxyapatite-based magnetic nanoparticles non-commercialized developed for in vivo clinical imaging. ...................................................................... 29 xviii xix Abbreviations List of abbreviations BMA Bismethylammide BMDC Bone Marrow Derived Cells CT Computed Tomography DLS Dynamic Light Scattering DO3A [4,7-bis-carboxymethyl-10-(2-aminoethyl)-1,4,7,10-tetraazacyclododec-1-yl]-acetic acid DOTA 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acids DPDP Dipyridoxyl Diphosphate DTPA Diethylenetriaminepentaacetic Acid EPR Enhanced Penetration and Retention effect FBS Foetal Bovine Serum FDA U.S. Food and Administration FTIR Fourrier Transform Infrared Gd-BOPTA Gadobentate-dimeglumine HAp Hydroxyapatite hDMEC Human dermal microvascular endothelial cell hFOB Human fetal osteoblast cell line hUVEC Human umbilical vein endothelial cell HP Hydroxypropyl KB Human nasopharyngeal carcinoma L929 Normal mouse lung fibroblast cell line MDSC Myeloid Cell-Derived Suppressor Cells MGC-803 Human gastric carcinoma cells MMP Matrix Metalloproteinases MRI Magnetic Resonance Imaging MSC Mesenchymal Stem Cells nanoHAP Nano-hydroxyapatite NIRF Near-infrared Fluorescence NPs Nanoparticles OECD Organization for Economic Co-Operation and Development PAI Photoacoustic Imaging xx PET Positron Emission Tomography R3-IGF-1 Recombinant long R insulin-like growth factor-1 RES Reticuloendothelial System RF Radio-frequency rhEGF Human recombinant epidermal growth factor rhFGF-β Human recombinant fibroblast growth factor-β RT Room temperature Saos-2 Osteoblast-like cells SPECT Single-Photon Emission Computed Tomography SPIO Superparamagnetic Iron Oxides SQUID Superconducting quantum interference device SWNTs Single Walled Nantotubes TEM TIE2-expressing monocytes TR Repetition Time UE Electrophoretic mobility USI Ultrasound Imaging VEGF Vascular Endothelial Growth Factor XRD X-ray diffraction xxi Chapter 1 Introduction 1.1 - Cancer Disease 1.1.1 - Physiological characteristics of benign and malignant solid tumors Not only cancer is a major public health problem, with estimated 5.2 million new cases of cancer diagnosed in OECD countries[1] and an average of 83 million years of “healthy life” lost due to death and disability in 2008, but this also constituted a total economic impact of $895 billion worldwide in the same year[2]. So, even though research efforts to improve current cancer therapies over the past three decades have led to an improvement in patient survival, there is still a need for new and better strategies. But, in order to develop new therapies for cancer treatment, it is crucial to get a better insight of tumors. In common medical usage, a neoplasm, a mass of tissue that continuously replicates which is not influenced by the regulatory processes that control normal cell growth, is often referred to as a tumor. In general, all tumors have two basic components: the parenchyma, which is constituted by transformed or neoplastic cells and determines the neoplasm behavior, and the nonneoplastic stroma, which, in turn, consists of connective tissue, blood vessels and hostderived inflammatory cells, therefore providing support for the growth of parenchymal cells. However, neoplasms or tumors can be categorized into benign or malign, regarding their potential clinical behavior and performance according to some parameters: differentiation and anaplasia (lack of differentiation), rate of growth, local invasion, and metastasis. A benign tumor will remain localized, with no chance of spreading to other sites, can be surgically removed and it is considered to be relatively innocent. Nevertheless, benign tumors can evolve to a malignant state requiring regular follow-up. 22 Malignant tumors are referred to as cancers and most of them have the ability to invade and destroy adjacent structures and spread to distant sites (metastasize) to possibly cause death. In a more detailed perspective, as illustrated in Table 1, malignant tumors have distinct parenchymal cells, from well differentiated to completely undifferentiated states. It is well known that the more differentiated a cell is, the more likely it performs its normal functional activities. That said, anaplasia is a common feature in malignant tumors and it involves dedifferentiation, loss of the structural and functional differentiation of specialized cells or failure of differentiation in stem cells. Anaplastic nuclei are variable and peculiar in size and shape and mitoses are often numerous and distinctly atypical. Therefore, generally, the more rapidly growing and the more anaplastic a tumor, the less likely it is to have specialized functional activity. Malignant tumors also do not present an often feature in the benign ones: an enclosing fibrous capsule probably derived from the host stroma that isolates them from the hosting environment. Thus, most cancers grow progressively penetrating and destroying surrounding tissues without developing well-defined capsules. Metastasizing and local invasion constitute the most reliable features that distinguish malignant from benign tumors. However, not all cancers have the same ability to metastasize. Examples like basal cell carcinomas or primary brain tumors are highly invasive in their primary sites of origin but rarely metastasize[3, 4]. Table 1.1 Summary of distinct features of benign and malignant tumors (adapted from [5]) Because of their rapidly growing feature, most cancers tend to spread locally and to distant organism sites, process called metastasizing. Angiogenesis, known as the formation of new blood vessels from existing ones, is essential to the transformation of a small, often dormant cluster of cancer cells to a solid tumor and also to the metastasizing process, hypothesis stated for the first time in 1971 by Judah Folkman[6]. For solid tumors of 1-2mm3, simple diffusion is the way oxygen and nutrients reach the center of the tumor, but for solid tumors bigger than 2mm3, a state of cellular hypoxia begins, which initiates signaling events that upregulate the production of pro-angiogenic factors, such as vascular endothelial growth BENIGN TUMOR MALIGNANT TUMOR Differentiation and anaplasia Composed of well-differentiated cells that closely resemble their normal counterparts; mitoses are limited in number and of normal configuration. Characterized by a wide range of parenchymal cell differentiation; anaplasia is common; mitoses are often numerous and distinctly atypical. Rate of growth Most benign tumors increase in size slowly over the span of months to years. Most cancers progressively enlarge over time, some slowly, others rapidly. Local invasion Remains localized at its site of origin. Most cancers infiltrate, invade, or metastasize to distant sites. Metastasis Nonmetastatic Most cancers develop secondary implants (metastases) discontinuous with the primary tumor, in remote tissues. 23 factor. In turn, VEGF regulates endothelial cell proliferation and enhances permeability of the microvessels, giving rise to neovasculatures different from those of normal tissues with an irregular shape, of high density and heterogeneity and poorly oxygenated, resulting in regions of hypoxia. The lack of oxygen stimulates the tumor cells to have an anaerobic metabolism that causes an extracellular acidosis in response to excessive production of lactic acid and CO2. It has been reported that extracellular normal tissue pH is around 7.4, whereas extracellular tumor pH is generally between 6.0 and 7.0 and this characteristic has been a source of interest especially in targeted cancer therapies investigation[7]. The initiation of the tumor angiogenesis is crucial to the exponential tumor growth and is called angiogenic switch. It can occur at different stages in the tumor-progression pathway and it is defined by several distinct phases such as vessel dilation, angiogenic sprouting, new vessel formation and maturation and recruitment of perivascular cells. In addition, cancer cells tend to compress lymphatic vessels forcing them to collapse, therefore creating lack of functional lymphatic system mostly within the tumor central area. This characteristic along with a leaky tumor vasculature induces an accumulation of vascular contents inside the tumor and an interstitial hypertension which hampers the transport of macromolecules from the capillaries to the interstitial fluid[8]. Also, the tumor microenvironment plays a major role in the modulation of the metastatic behavior of most cancers. It is now recognized that primary and metastatic tumors induce the mobilization of bone marrow-derived cells that help the tumor to progress, survive and grow at a secondary site. As portrayed in Figure 1.1, a plethora of cell types interact with each Figure 1.1 The primary tumor microenvironment (adapted from [9]) 24 other, responsible for the secretion of, for example, VEGF and matrix metalloproteinases, enabling angiogenesis malignant progression. Endothelial cells in the blood and lymphatic vessels, pericytes, fibroblasts, macrophages, neutrophils, mast cells, myeloid cell-derived suppressor cells, mesenchymal stem cells and TIE2-expressing monocytes are among the main cell types that interact with cancer cells[9]. 1.1.2 - Current tumor imaging perspective Early detection of cancer disease allows the maximum likelihood of successful treatment and recovery, since early localization of malignant solid tumors may allow a directed therapy to its site, optimizing the efficacy of the treatment. This can be achieved with an appropriate detection device. The type of information required or the characteristics of the biological target condition the performance of the in-vivo imaging technique used and the choice of the contrast agent or probe. For instance, MRI is better suited for clinical use than fluorescence imaging, since it is more powerful and safer on humans[10]. Imaging modalities play an important role in the clinical management of cancer, including screening, diagnosis, treatment planning and therapy monitoring. Table 2 summarizes the most commonly used bioimaging modalities in tumor diagnosis, paying special attention to current NP-based contrast agents for each technique (MRI, computed tomography, positron emission tomography, single-photon emission computed tomography, ultrasound imaging, photoacoustic imaging and near-infrared fluorescence). These most commonly used imaging modalities in clinical detection of cancer are all nonspecific, therefore they cannot distinguish between malignant or benign tumors and none leads to 100% accurate detection. Besides, most of the presented technologies and their contrast agents involve exposure to radiation or heavy metals, which present potential health risks to patients and limit the ways they can be applied. The goal of cancer imaging should be to detect and/or image the smallest possible number of tumor cells, since there are currently no reliable methods for in vivo imaging of tumors smaller than 1cm3, constituted with approximately 109 cells[11]. Also, there are certain types of cancer that are still currently hard to detect and image such as ovarian cancer because of its extra thin earliest proliferative lesion or pediatric cancer, since it is difficult to preform clinical trials in young ages and because they constitute a small market size[11]. 1.1.2.1 - Magnetic resonance imaging MRI is a non-invasive FDA approved for clinical use imaging technique (since 1985) that does not expose the patient to ionizing radiation and offers a high spatial resolution of approximately 50μm. It is also accepted that magnetic fields are not contraindicated for humans and MR scanners are safe with a static magnetic field of 8 Tesla, except for patients whose body contains magnetizable material such as implanted metallic medical devices, pacemakers, artificial limbs or tattoos[12, 13]. MRI is based on the detection of nuclear magnetic resonance signals that are emitted by hydrogen protons in body tissues, membrane lipids, proteins, etc., under the effect of a magnetic field and it is used for the diagnosis and monitoring of human diseases, since it enables whole-organisms anatomical imaging. MRI can quickly produce pictures of wide range 25 of bodily tissues, though the clarity of these pictures is sometimes insufficient for diagnoses, which can be improved through the appliance of a contrast agent that enhances the visibility of the tissues it is in contact with. Technically, the application of a transverse, or perpendicular to the main magnetic field, radiofrequency-field induces the transversal magnetization of the hydrogen protons. When these try to realign again with the longitudinal axis of the magnetic field, returning to a state of equilibrium (relaxation process), an electric field is created, generating the resonance signal (Mxy). In addition, three tissue parameters influence MRI image contrast, which are sensitivity to proton density, longitudinal relaxation times T1 (return of longitudinal magnetization to align with the external magnetic field) and transverse relaxation times T2 (vanishing of transverse magnetization), which are illustrated in Figure 1.2 (Adapted from [14]). However, here remains a possibility of poor contrast between healthy and diseased tissue due to small variation in their relaxation times. A method to improve image contrast and diagnostic value of MRI is to introduce contrast enhancer agents that increase relaxation rates of water protons in tissue where the agent accumulates. By manipulating TR and time delay after excitation before the signal is measured, the relative contributions of T1 and T2 can be manipulated to produce T1and T2weighted images. More specifically, the agents that increase T1 signal produce a bright contrast and those that reduce the T2 signal create a dark contrast. Therefore, agents that reduce T1 or T2 relaxation times are said to have individual relaxivities values [15], [16], [10]. 1.2 - Nanoparticles in tumor detection 1.2.1 - Biodistribution of the nanoparticles Nanoparticles constitute a very powerful tool in cancer therapy and diagnosis due to some key characteristics. Primarily, their size range, 10-100nm, allows NPs use in several Figure 1.2 Illustration of longitudinal relaxation time T1 and transverse relaxation time T2. RF excitations are applied with a TR (top) with each excitation producing a magnetic resonance signal (Mxy) that exponentially decays with a time constant T2 (bottom right). Between excitations, the magnetization also exponentially returns to its equilibrium state, pointing along the direction of the main magnetic field, with a time constant T1 (bottom left). 32 MRI is a non-invasive and commonly used imaging modality that can maximize its diagnostic value with the combined use of magnetic contrast agents. In addition, MRI contrast agents with nanometric dimensions constitute a very powerful tool in cancer therapy and diagnosis possessing distinct advantages over conventional ones due to some key characteristics such as their small size and high surface area, the possibility of delivery to a specific cancer site by active targeting and the possibility of engineering their blood circulation half-life, among other features. Although some nano-contrast agents have been proposed so far, cytotoxicity is still a major issue that endangers their clinical use. Hydroxyapatite is a bioceramic material that enables the inclusion of ions in its crystal lattice and also confers biocompatibility to the system which makes its conjugation with magnetically responsive ions an appealing approach in targeted cancer imaging. As previously described, some efforts to develop hydroxyapatite-based MRI contrast agents have been made. Nevertheless, there is a need to standardize the procedures of fabrication of the magnetic NPs. Also, it is not included in the literature a study that compares different doping agents produced by the same method and tested on the same parameters. This project main goal is to produce, characterize and compare hydroxyapatite nanoparticles doped with particular magnetic species (Gd, Fe(II), Fe(III) and Co) and ultimately infer about their potential as MRI contrast agents. This main goal can be divided into smaller tasks that consist on:  Choose the right method a optimizing the available conditions in order to produce doped nanoparticles with sizes between 10 and 100nm;  Characterize the produced NPs in terms of their size, magnetic properties and also characterize the process of ion doping and its influence on the HAP structure;  Evaluate any toxic effects of the doped NPs in a common cell line in a certain time frame. 33 Chapter 2 Materials and Methods 2.1 - Synthesis of the Hydroxyapatite-based nanoparticles The hydroxyapatite-based nanoparticles were prepared by the wet chemical precipitation method. In a typical synthesis procedure of pure hydroxyapatite nanoparticles, 40mL of 0.5M Ca(OH)2 (98% extra pure, Acros Organics) was prepared and heated to 100ºC to which 40mL of 0.3M H3PO4 (85 wt% solution in water, Acros Organics) was added at a rate of 500µL/min. At this point, the magnetic dopants Gadolinium chloride hexahydrate Cl3Gd.6H2O (99%, Sigma Aldrich), Cobalt nitrate hexahydrate Co(NO3)2.6H2O (98+%, Acros Organics), Iron nitrate nonahydrate Fe(NO3)3 9H2O (98+%, Acros Organics) and Iron chloride tetrahydrate FeCl2.4H2O (99+%, Acros Organics) were added simultaneously with the phosphate precursor H3PO4 dropwise to the Ca(OH)2 solution at the rate of 500µL/min to produce doped hydroxyapatite nanoparticles. The amount of dopant ions added depended on the required percentage of doping (2.5%, 5% and 10%) with respect to the atomic percentage of Ca2+. The applied nomenclature to the nanoparticles is formatted as XY_HAP, where X is the doped magnetic ion and Y represents the percentage of doping. All solutions were prepared with deionized water and the (Ca+doping ion)/P molar ratio was maintained at 1.67 in all procedures. The reaction was carried out at 100ºC in a water bath with constant stirring of 650rpm, as portrayed in figure 2.1. Throughout the reaction, pH was maintained at ~7.4 using ammonium hydroxide (25% solution in water, Acros Organics). After the reaction, the mixture was kept in these conditions for 2 hours and then incubated at room temperature overnight. Prior to the drying step at 60ºC, samples were centrifuged (5 min, ~2070g) and washed 3 times with hot deionized water. Finally, all nanoparticles were powdered using mortar and pestle and stored in a desiccator for further use. This protocol was adapted from the one described by Ashokan et al [17]. 34 2.2 - Physicochemical Characterization 2.2.1 - Size, morphology and zeta potential Transmission electron microscope (TEM) was employed for high resolution imaging of pure and doped nanoHAP. A Jeol JEM 1400 microscope was used to observe nanoparticle size and morphology. To prepare samples for observation, the powders were suspended in deionized water using a concentration of 0.05g/L and the dispersant sodium dodecyl sulfate (SDS, 20% solution in water, Sigma Aldrich) was added in a concentration of 10x10-3 mol/L. Solutions were then sonicated for 15 seconds (20x) (Sonics&Materials, VC50) and applied to a copper grid. Images were examined using ImageJ® software to quantify nanoparticles dimensions. The zeta potential and size of the synthesized nanoparticles were measured with a ZetaSizer Nano ZS. This device uses Dynamic Light Scattering (DLS) and calculates the particle size distribution from their diffusion under Brownian motion and applying the Stokes-Einstein relationship, D=kT/3πηDH, where D is the translational diffusion coefficient of a spherical particle in a fluid, k is the Boltzmann’s constant, T is the absolute temperature, η is the viscosity and DH is the hydrodynamic diameter. On the other hand, zeta potential is calculated through the particles electrophoretic mobility which is the movement of a charged particle relative to the liquid when an electric field is applied. A laser beam hits the moving particles and induces a change of frequency by the Doppler Effect. This difference is proportional to the electrophoretic mobility (UE) and zeta potential (z) is analyzed according to Henry’s equation, UE=2εzf(Ka)/3η, where ε is the dielectric constant, f(Ka) is Henry’s function and η is the viscosity[50]. For this characterization, the powders were suspended in deionized water and prepared the same way as for TEM observations. Figure 2.1 Representation of the nanoparticles fabrication conditions. a-Peristaltic pump; btemperature controller and heating plate with magnetic stirrer; c-pH meter. a. b. c. 35 2.2.2 - Chemical profile The infrared spectra that contain the species characteristic chemical bonds were recorded with a Fourrier Transform Infrared (FTIR) spectrometer Perkin-Elmer 2000 with a 4cm-1 spectral resolution and 100 scans per sample. In order to perform the analysis, ~2mg of each powdered sample was mixed with ~200mg of potassium bromide KBr to maintain low levels of humidity and discs were produced in a uniaxial press (Grasedy Specac). 2.2.3 - Crystal phase analysis In order to infer about the particles crystallinity before and after doping, X-ray diffraction (XRD) measurements were made. For this technique, samples were sintered to improve the obtained signal. The nanoHAP powders were heated up to 1000ºC at a rate of 4ºC/min and then heat-treated for 1h at this temperature, being the appropriate sintering temperature for nanosized HAP prepared by this method[41]. The x-ray diffraction patterns were obtained in a PANalytical X’Pert PRO TCU 1000 diffractometer, using monochromatic Cu radiation (λ = 1.541874Å). Data were acquired for 2Θ values between 8º and 100º with steps of 0,008º. Commercialized undoped nanoHAP was analyzed by other team member and was used as a control after heat-treatment. 2.2.4 - Magnetic properties The magnetic properties of the doped and undoped nanoparticles were studied by superconducting quantum interference device (SQUID, Quantum design Ref: MPMS-5S) in an applied magnetic field of•±50 kOe at 300K. The strength of the magnetic moment is presented as a mass magnetization (M) – magnetic moment per total mass of each sample. For these measurements, the previously prepared powders were weighed and inserted in a see-through capsule and, in turn, the capsule was inserted in a see-through straw that is introduced into the SQUID magnetometer for further analysis. 2.3 - Biological response 2.3.1 - Cell culture and maintenance Human Dermal Microvascular Endothelial Cells (hDMECs) were purchased from ScienCell Research Laboratories (Catalog number: 2000) and cultured in Endothelial Cell Basal medium (EBM®-2, Clonetics Lonza) supplemented with 2% foetal bovine serum (FBS), 0.04% hydrocortisone, 0.4% human recombinant fibroblast growth factor-β (rhFGF-β), 0.1% vascular endothelial growth factor (VEGF), 0.1% ascorbic acid, 0.1% Gentamicin Sulfate and Amphotericin-B, 0.1% heparin, 0.1% human recombinant epidermal growth factor (rhEGF) and 0.1% recombinant long R insulin-like growth factor-1 (R3-IGF-1) . The medium was renewed every 2-3 days until confluence was reached and cells were used until passage 7. All the procedures described with live cells were performed in sterile conditions on a laminar flow chamber and incubation was carried out in a 5% CO2 and 37 °C atmosphere. 36 2.3.2 - Cell metabolic activity evaluation In order to estimate the in vitro cell viability of the nanoHAP with the highest doping concentration (10%) of Gd, Fe(II), Fe(III) and Co, a resazurin (Sigma) assay was performed. This method is based on a cell viability indicator that uses natural reducing power of living cells to convert resazurin to the fluorescent product, resofurin. Fluorescence acts as a proportional response of cell metabolic activity[51]. Each well of the 96-well plate was seeded with 5000 hDME cells. After 24h of culture, the medium in the wells was refreshed with the medium containing autoclaved nanoparticles at a concentration of 500µg/mL. Finally, cells were incubated with new medium and resazurin assay was performed at the time points of 4h and 24h. Briefly, after previously removing the culture medium form the wells, resazurin was added in fresh medium at a final concentration of 10% (v/v) and incubated for 4h, at 37ºC. 100μL of medium was then extracted from each sample and fluorescence was measured at λex=540nm and λem=590nm in a microplate reader. 2.3.3 - Cellular morphology After 4 and 24 hours of culture, cell morphology was observed by staining the F-actin with Alexa Fluor ® 488 phalloidin (Invitrogen, Molecular Probes) and counterstaining the nuclei with DAPI (4',6-diamidino-2-phenylindole, Merck). Culture medium was removed from the wells and samples were washed 2x with PBS and fixated with 4% formaldehyde at room temperature (RT) for 15 minutes. After washing with PBS twice, cells were permeabilized for 30 minutes with 0.1% (v/v) Triton X-100 and washed again 3x with PBS. The samples were then incubated with the phalloidin diluted 1:100 in PBS for 30 minutes in the dark at RT and washed 3x with PBS. To stain the nuclei of the cells, samples were incubated with DAPI at a concentration of 0.1µg/mL in PBS for 10 min and washed with PBS 3x. Two drops of FluoromountTM were added and the samples were stored at 4ºC in the dark until observation under an inverted fluorescence microscope with proper light filters. 2.3.4 - Statistical analysis Data from the resazurin assay were analyzed using Scheffe test for multiple comparisons. Results were considered significant when p<0.05. Calculations and descriptive statistic were performed using SPSS-Software for Windows® (version 17.0) and Microsoft Office Excel 2010®. All data are presented as average values ± standard deviation. 37 Chapter 3 Results and Discussion 3.1 - Physicochemical Characterization 3.1.1 - Size, morphology and zeta potential Both size and morphology of nanoparticles synthesized as MRI contrast agents influence their interaction not only with each other in a suspension, but also with the cells they are in contact with. Undoped HAP Gd10_HAP Gd5_HAP Gd2.5_HAP 38 Fe(II)10_HAP Fe(II)5_HAP Fe(II)2.5_HAP Fe(III)10_HAP Fe(III)5_HAP Fe(III)2.5_HAP Co10_HAP Co5_HAP Co2.5_HAP Figure 3.2 TEM images of the undoped and doped nanoHAP Figure 3.1 NanoHAP dimensions obtained with TEM images (Figure 3.1) observation through Image J® software 0,0 10,0 20,0 30,0 40,0 50,0 60,0 70,0 80,0 Size (nm) Spherical shape Rod-like shape (a) Rod-like shape (b) 39 The TEM image (Figure 3.1) of undoped hydroxyapatite indicates the formation of spherical nanoHAP of average size ~11 nm. On the other hand, all images of doped nanoparticles show a predominant more elongated shape that can be identified as a rod-like shape. In addition, these species present slightly higher dimensions, as also shown in Figure 3.2. The studied doped nanoparticles present longitudinal dimensions between ~34 and ~59nm and cross sectional dimensions between ~15 and ~29nm. These differences in size and morphology from the doped and undoped species might have been induced by differences in electric charge and ionic radius between the Ca2+ ions and the Gd3+/Fe2+/Fe3+/Co2+ doping ions. This may be influencing the crystal growth of nanoHAp, leading to altered size and shape of doped NPs compared to undoped NPs, as suggested by other authors in similar studies[47]. Considering the main application of the synthesized nanoparticles in tumor cells imaging and future applications in tumor cell specific targeting, ligand coated NPs can be efficiently directed toward and subsequently internalized by tumor cells. However, cellular uptake of 0 50 100 150 200 250 300 Size (nm) -60 -50 -40 -30 -20 -10 0 Zeta Potential (mV) Figure 3.3 ZetaSizer results displaying (A) the average particle sizes and (B) the average zeta potential of the doped and undoped nanoHAP A B A 40 NPs is size-dependent and some authors have found 40-60nm to be the optimal NP size range for endocytosis and mediation of other signaling processes [52, 53]. Consequently, the results obtained through TEM observation are in agreement with those described in the literature. In relation to the shape of the particles, although there is a lot of unconformity among authors, recent findings indicate that rod-like shaped nanoparticles are significantly better in cancer targeting, exhibiting higher specific uptake and lower nonspecific uptake in breast cancer cells compared with spherical nanoparticles [54]. This also constitutes a positive indicator that the doped nanoHAP produced in this work is suitable for targeted cancer imaging. Comparing the TEM results to the ones obtained with the ZetaSizer, size measurements differ. As portrayed in Figure 3.3A, the studied doped and undoped nanoparticles present sizes between a range of 155 and 244nm. Size between both methods differs due to the fact that ZetaSizer uses DLS technology whose approach recognizes each particle as a perfect sphere that does not represent the actual shape of the synthesized NPs. Also, the ZetaSizer measurements most likely represent the size of NPs aggregates that were not effectively dispersed after sonication. Concerning the measured zeta potential of the samples, it fluctuates between -45 and - 34mV with small deviations (Figure 3.3B). All preparations consisted on stable colloidal suspensions either at or above critical micelle concentration for the negatively charged surfactant SDS which explains the strongly negative zeta potential obtained in all suspensions[55]. 41 3.1.2 - Chemical profile 4000 3700 3400 3100 2800 2500 2200 1900 1600 1300 1000 700 400 Transmittance (%) Wavenumber (cm-1) Undoped HAP Gd10_HAP Gd5_HAP Gd2.5_HAP 4000 3700 3400 3100 2800 2500 2200 1900 1600 1300 1000 700 400 Transmittance (%) Wavenumber (cm-1) Undoped HAP Fe(II)10_HAP Fe(II)5_HAP Fe(II)2.5_HAP 4000 3700 3400 3100 2800 2500 2200 1900 1600 1300 1000 700 400 Transmittance (%) Wavenumber (cm-1) Undoped HAP Fe(III)10_HAP Fe(III)5_HAP Fe(III)2.5_HAP A B C OH - H2O HPO4 2CO3 2PO4 3- 48 Fe(II)10_HAP Fe(III)10_HAP Co10_HAP Undoped HAP TCPS A 49 Figure 3.8 Morphology observation of hDMECs cultured for 4 and 24h with doped and undoped nanoHAP. Double nuclei characteristic is highlighted in picture A (Fe(II)10_HAP after 4h of culture). FActin is shown in green and cells nuclei in blue. Magnification 10x. The results found seem to be in accordance to the ones extracted from the cell metabolic activity assay. Interestingly, it is possible to observe in each substrate what resembles cells going through the mitosis process with double nuclei in the same cytoplasmic area (highlighted by white arrows in Figure 3.8 and by white square in Figure 3.8A). Another interesting characteristic is the heterogenic morphology of hDMECs in both time points, existing clear differences in the size of the cells which also suggests that these cells are proliferating at different stadiums. Nonetheless, there does not seem to be a morphological change between time points in all conditions which might suggest that these NPs are biocompatible for this cell type in the studied time frame. However, the toxic effect of NPs in cells is not necessarily connected with the death of cell populations. They can also be associated with deviations in the fate of the cells, for instance alter their capacity to differentiate, as shown by Pisanic et al.[67]. For this reason, further tests should be addressed in order to fully comprehend the synthesized NPs effect in the cultured hDMECs. One important parameter is the oxidative stress caused within the cells that could lead to adverse biological effects[68], therefore the intracellular reactive oxygen species released in cells incubated with the NPs could be further investigated. 50 51 Chapter 4 Conclusions and future perspectives In this study, both hydroxyapatite and magnetic ion doped hydroxyapatite nanoparticles were successfully produced by a well-controlled wet chemical precipitation method. The obtained magnetically doped particles have a rod-like shape with sizes that vary approximately between 30 and 60nm that do not fluctuate with doping percentage. These characteristics are considered optimal for cell endocytosis and cancer targeting. It was shown that the doping concentration with respect to the atomic percentage of Ca2+ does not significantly alter the HAP structure although it has a predominant role in the magnetic properties of the nanoparticles. 10% Gd doped HAP presented the highest magnetic moment levels compared to the Fe(II), Fe(III) and Co ones and the closest values to Dotarem®. Viability and morphology of human dermal microvascular endothelial cells did not seem to be significantly altered after incubation with the synthesized materials. The results achieved in the present project suggest that from the produced and tested magnetically responsive nanoparticles, Gd10_HAP seems to detain the highest potential for medical applications in the future as a versatile and customizable paramagnetic T1 MRI contrast agent. Future perspectives To increase the knowledge about the synthesized nanoparticles, XRD analysis should be further completed along with the Rietvelt method in order to fully understand the aspects in which the ion doping alters the nanohydroxyapatite core structure. In order to quantify the effective concentration of doping agent in each sample, an Inductively Coupled Plasma (ICP) could be performed. A curious complement could be the obtainment of MR images of the samples to compare the contrast provided by all the doped nanoparticles, also in comparison to Dotarem®. 52 Next step would also be to continue the in vitro work, repeating the cell viability studies possibly prolonging them and performing a cellular biodistribution assay to infer about the cellular uptake and stability of the nano-contrast. Haematocompatibility studies should also be annexed to this work, keeping in mind the in vivo application of these agents in the blood stream. An interesting and following phase would be to create a cell tumor model and optimize a suitable cancer target ligand to explore the targeted cancer imaging promising approach. References [1] OECD. Health at a Glance 2011. [2] Society AC. The global economic cost of cancer. 2010. [3] Robinson J, Dahiya M. Basal cell carcinoma with pulmonary and lymph node metastasis causing death. Archives of Dermatology. 2003;139:643-8. [4] Patchell RA. The management of brain metastases. Cancer Treatment Reviews. 2003;29:533-40. [5] Kumar V, Abbas AK, Fausto N, Mitchell R. Robbins basic pathology Elsevier Health Sciences; 2007. [6] Folkman J. Tumor angiogenesis: therapeutic implications. The New England journal of medicine. 1971;285:1182-6. [7] Deng Z, Zhen Z, Hu X, Wu S, Xu Z, Chu PK. Hollow chitosan-silica nanospheres as pHsensitive targeted delivery carriers in breast cancer therapy. Biomaterials. 2011;32:4976-86. [8] Blanco M, Teijón C, Olmo R, Teijón J. Targeted nanoparticles for cancer therapy, recent advances in novel drug carrier systems. InTech; 2012. [9] Joyce J, Pollard J. Microenvironmental regulation of metastasis. Nature Reviews Cancer. 2009;9:239-52. [10] Hahn M, Singh A, Sharma P, Brown S, Moudgil B. Nanoparticles as contrast agents for invivo bioimaging: current status and future perspectives. Analytical and bioanalytical chemistry. 2011;399:3-27. [11] Frangioni J. New technologies for human cancer imaging. Journal of Clinical Oncology. 2008;26:4012-21. [12] Atkinson I, Renteria L, Burd H, Pliskin N, Thulborn K. Safety of human MRI at static fields above the FDA 8T guideline: Sodium imaging at 9.4T does not affect vital signs or cognitive ability. Journal of Magnetic Resonance Imaging. 2007;26:1222-7. [13] Administration USFaD. Radiation-Emitting Products and Procedures: MRI (Magnetic Resonance Imaging). Updated in 2012. [14] Paus T, Collins DL, Evans AC, Leonard G, Pike B, Zijdenbos A. Maturation of white matter in the human brain: a review of magnetic resonance studies. Brain Research Bulletin. 2001;54:255-66. [15] Mornet S, Vasseur S, Grasset F, Duguet E. Magnetic nanoparticle design for medical diagnosis and therapy. Journal of Materials Chemistry. 2004;14:2161-75. [16] Yan G-P, Robinson L, Hogg P. Magnetic resonance imaging contrast agents: Overview and perspectives. Radiography. 2007;13, Supplement 1:e5-e19. [17] Ashokan A, Menon D, Nair S, Koyakutty M. A molecular receptor targeted, hydroxyapatite nanocrystal based multi-modal contrast agent. Biomaterials. 2010;31:2606-16. [18] Palazzo B, Iafisco M, Laforgia M, Margiotta N, Natile G, Bianchi CL, et al. Biomimetic hydroxyapatite–drug nanocrystals as potential bone substitutes with antitumor drug delivery properties. Advanced Functional Materials. 2007;17:2180-8. [19] Fukumori Y, Ichikawa H. Nanoparticles for cancer therapy and diagnosis. Advanced Powder Technology. 2006;17:1-28. [20] Greish K. Enhanced permeability and retention (EPR) effect for anticancer nanomedicine drug targeting. Methods in Molecular Biology. 2010;624:25-37. [21] Ohgushi M, Nagayama K, Wada A. Dextran-magnetite: A new relaxation reagent and its application to T2 measurements in gel systems. Journal of Magnetic Resonance 1978;29:599601. [22] Ferrucci J, Stark D. Iron oxide-enhanced MR imaging of the liver and spleen: review of the first 5 years. American Journal of Roentgenology. 1990;155:943-50. [23] Weissleder R, Elizondo G, Wittenberg J, Rabito CA, Bengele HH, Josephson L. Ultrasmall superparamagnetic iron oxide: characterization of a new class of contrast agents for MR imaging. Radiology. 1990;175:489-93. 54 [24] Lee H, Lee E, Kim D, Jang N, Jeong Y, Jon S. Antibiofouling polymer-coated superparamagnetic iron oxide nanoparticles as potential magnetic resonance contrast agents for in vivo cancer imaging. Journal of the American Chemical Society. 2006;128:7383-9. [25] HyoSook L, Huiping S, Yuqiang H, Kwak B. Synthesis of MRI contrast agent by coating superparamagnetic iron oxide with chitosan. IEEE Transactions on Magnetics. 2005;41:4102-4. [26] Kim D, Mikhaylova M, Wang F, Kehr J, Bjelke B, Zhang Y, et al. Starch-coated superparamagnetic nanoparticles as MR contrast agents. Chemistry of Materials. 2003;15:4343-51. [27] Bulte J, Cuyper M. Magnetoliposomes as contrast agents. In: Nejat D, editor. Methods in Enzymology: Academic Press; 2003. p. 175-98. [28] Schwert D, Davies J, Richardson N. Non-Gadolinium-based MRI contrast agents In: Krause W, editor.: Springer Berlin / Heidelberg; 2002. p. 165-99. [29] Josephson L. Magnetic Nanoparticles for MR Imaging. In: Ferrari M, Lee A, Lee LJ, editors. BioMEMS and Biomedical Nanotechnology: Springer US; 2006. p. 227-37. [30] Que EL, Gianolio E, Baker SL, Wong AP, Aime S, Chang CJ. Copper-responsive magnetic resonance imaging contrast agents. Journal of the American Chemical Society. 2009;131:852736. [31] Nidhin M, Nazeer SS, Jayasree RS, Kiran MS, Nair BU, Sreeram KJ. Flower shaped assembly of cobalt ferrite nanoparticles: application as T2 contrast agent in MRI. RSC Advances. 2013;3:6906-12. [32] Aime S, Castelli DD, Crich SG, Gianolio E, Terreno E. Pushing the sensitivity envelope of Lanthanide-based magnetic resonance imaging (MRI) contrast agents for molecular imaging applications. Accounts of Chemical Research. 2009;42:822-31. [33] Piecuch J. Extraskeletal of a porous hydroxyapatite ceramic. Journal of Dental Research. 1982;61:1458-60. [34] Grote J. Reconstruction of the ossicular chain with hydroxyapatite implants. The Annals of otology, rhinology & laryngology Supplement. 1986;123:10-2. [35] Holmes RE. Bone regeneration within a coralline hydroxyapatite implant. Plastic and reconstructive surgery. 1979;63:626-33. [36] Habibovic P, Barrère F, Van Blitterswijk CA, de Groot K, Layrolle P. Biomimetic hydroxyapatite coating on metal implants. Journal of the American Ceramic Society. 2002;85:517-22. [37] Cook SD, Thomas KA, Delton JE, Volkman TK, Whitecloud TS, Key JF. Hydroxylapatite coating of porous implants improves bone ingrowth and interface attachment strength. Journal of Biomedical Materials Research. 1992;26:989-1001. [38] Xie G, Sun J, Zhong G, Liu C, Wei J. Hydroxyapatite nanoparticles as a controlled-release carrier of BMP-2: absorption and release kinetics in vitro. Journal of Materials Science: Materials in Medicine. 2010;21:1875-80. [39] Zhu SH, Huang BY, Zhou KC, Huang SP, Liu F, Li YM, et al. Hydroxyapatite nanoparticles as a novel gene carrier. Journal of Nanoparticle Research. 2004;6:307-11. [40] Morrissey R, Rodríguez-Lorenzo LM, Gross KA. Influence of ferrous iron incorporation on the structure of hydroxyapatite. Journal of Materials Science: Materials in Medicine. 2005;16:387-92. [41] Ferraz MP, Monteiro FJ, Manuel CM. Hydroxyapatite nanoparticles: A review of preparation methodologies. Journal of applied biomaterials & biomechanics 2004;2:74-80. [42] Cao L-y, Zhang C-b, Huang J-f. Synthesis of hydroxyapatite nanoparticles in ultrasonic precipitation. Ceramics International. 2005;31:1041-4. [43] Kumar R, Prakash KH, Cheang P, Khor KA. Temperature Driven Morphological Changes of Chemically Precipitated Hydroxyapatite Nanoparticles. Langmuir. 2004;20:5196-200. [44] Wu H-C, Wang T-W, Sun J-S, Wang W-H, Lin F-H. A novel biomagnetic nanoparticle based on hydroxyapatite. Nanotechnology. 2007;18:165601. [45] Hou CH, Hou SM, Hsueh YS, Lin J, Wu HC, Lin FH. The in vivo performance of biomagnetic hydroxyapatite nanoparticles in cancer hyperthermia therapy. Biomaterials. 2009;30:3956-60. [46] Petchsang N, Pon-On W, Hodak JH, Tang IM. Magnetic properties of Co-ferrite-doped hydroxyapatite nanoparticles having a core/shell structure. Journal of Magnetism and Magnetic Materials. 2009;321:1990-5. 55 [47] Chen F, Huang P, Zhu YJ, Wu J, Zhang CL, Cui DX. The photoluminescence, drug delivery and imaging properties of multifunctional Eu3+/Gd3+ dual-doped hydroxyapatite nanorods. Biomaterials. 2011;32:9031-9. [48] Panseri S, Cunha C, D'Alessandro T, Sandri M, Giavaresi G, Marcacci M, et al. Intrinsically superparamagnetic Fe-hydroxyapatite nanoparticles positively influence osteoblast-like cell behaviour. Journal of Nanobiotechnology. 2012;10:32. [49] Khemtong C, Kessinger CW, Ren J, Bey EA, Yang S-G, Guthi JS, et al. In vivo offresonance saturation magnetic resonance imaging of αvβ3-targeted superparamagnetic nanoparticles. Cancer Research. 2009;69:1651-8. [50] Limited MI. Zetasizer Series: Dynamic Light Scattering and Zeta Potential. 2012. [51] Ansar Ahmed S, Gogal Jr RM, Walsh JE. A new rapid and simple non-radioactive assay to monitor and determine the proliferation of lymphocytes: an alternative to [3H]thymidine incorporation assay. Journal of Immunological Methods. 1994;170:211-24. [52] Zhang S, Li J, Lykotrafitis G, Bao G, Suresh S. Size-dependent endocytosis of nanoparticles. Advanced Materials. 2009;21:419-24. [53] Jiang W, KimBetty YS, Rutka JT, ChanWarren CW. Nanoparticle-mediated cellular response is size-dependent. Nature Nanotechnology. 2008;3:145-50. [54] Barua S, Yoo J-W, Kolhar P, Wakankar A, Gokarn YR, Mitragotri S. Particle shape enhances specificity of antibody-displaying nanoparticles. Proceedings of the National Academy of Sciences. 2013. [55] Welzel T, Meyer-Zaika W, Epple M. Continuous preparation of functionalised calcium phosphate nanoparticles with adjustable crystallinity. Chemical Communications. 2004:12045. [56] Mondejar SP, Kovtun A, Epple M. Lanthanide-doped calcium phosphate nanoparticles with high internal crystallinity and with a shell of DNA as fluorescent probes in cell experiments. Journal of Materials Chemistry. 2007;17:4153-9. [57] Ribeiro N, Sousa SR, Monteiro FJ. Influence of crystallite size of nanophased hydroxyapatite on fibronectin and osteonectin adsorption and on MC3T3-E1 osteoblast adhesion and morphology. Journal of Colloid and Interface Science. 2010;351:398-406. [58] Pramanik N, Mishra D, Banerjee I, Maiti TK, Bhargava P, Pramanik P. Chemical synthesis, characterization, and biocompatibility study of hydroxyapatite/chitosan phosphate nanocomposite for bone tissue engineering applications. International journal of biomaterials. 2009;2009:512417. [59] Gibson IR, Best SM, Bonfield W. Chemical characterization of silicon-substituted hydroxyapatite. Journal of Biomedical Materials Research. 1999;44:422-8. [60] Gibson IR, Best SM, Bonfield W. Chemical characterization of silicon-substituted hydroxyapatite. Journal of Biomedical Materials Research. 1999;44:422-8. [61] Nkeng P, Koenig JF, Gautier JL, Chartier P, Poillerat G. Enhancement of surface areas of Co3O4 and NiCo2O4 electrocatalysts prepared by spray pyrolysis. Journal of Electroanalytical Chemistry. 1996;402:81-9. [62] Larkin P. Infrared and Raman Spectroscopy; Principles and Spectral Interpretation: Elsevier Science; 2011. [63] Morrissey R, Rodriguez-Lorenzo LM, Gross KA. Influence of ferrous iron incorporation on the structure of hydroxyapatite. Journal of Materials Science: Materials in Medicine. 2005;16:387-92. [64] Huang J, Zhong X, Wang L, Yang L, Mao H. Improving the magnetic resonance imaging contrast and detection methods with engineered magnetic nanoparticles. Theranostics. 2012;2:86-102. [65] DailyMed. DOTAREM (gadoterate meglumine) injection 2013. [66] Maeda H. The enhanced permeability and retention (EPR) effect in tumor vasculature: the key role of tumor-selective macromolecular drug targeting. Advances in Enzyme Regulation. 2001;41:189-207. [67] Pisanic Ii TR, Blackwell JD, Shubayev VI, Fiñones RR, Jin S. Nanotoxicity of iron oxide nanoparticle internalization in growing neurons. Biomaterials. 2007;28:2572-81. [68] Eom H-J, Choi J. Oxidative stress of CeO2 nanoparticles via p38-Nrf-2 signaling pathway in human bronchial epithelial cell, Beas-2B. Toxicology Letters. 2009;187:77-83.