Comprehensive Toxicity Assessment of PEGylated Magnetic Nanoparticles for in vivo applications

1 Comprehensive Toxicity Assessment of PEGylated Magnetic 1 Nanoparticles for in vivo applications. 2 Carlos Caroa,b,‡, David Egea-Benaventea,‡, Rocio Polvilloc, Jose Luis Royod, Manuel 3 Pernia Lealb,*, María Luisa García-Martína,*. 4 aBIONAND, Andalusian Centre for Nanomedicine and Biotechnology (Junta de 5 Andalucía-Universidad de Málaga), Málaga, Spain 6 bDepartamento de Química Orgánica y Farmacéutica, Universidad de Sevilla, 41012 7 Seville (Spain) 8 cAndalusian Centre for Developmental Biology, (Universidad Pablo de Olavide-Junta de 9 Andalucia-CSIC). 41013, Seville, Spain, 10 dArea of Biochemistry and Molecular Biology. School of Medicine. University of 11 Malaga. Malaga Spain. 12 ‡Carlos Caro and David Egea-Benavente contributed equally to this work. 13 * email: [email protected]; [email protected]. 14 Keywords: magnetic nanoparticles, contrast agents, MRI, toxicity, Zebra fish embryos. 15 Statistical summary. 16 Total number of words: 4921 17 figures: 7 18 19 20 21 22 2 Abstract 1 Magnetic nanoparticles (MNPs) represent one of the greatest promises for the 2 development of a new generation of diagnostic agents for magnetic resonance imaging, 3 with improved specificity and safety. Indeed, during the last decade the number of studies 4 published in this field has grown exponentially. However, the clinical translation 5 achieved so far has been very limited. This situation is likely related to the fact that most 6 studies are focused on the in vitro characterization of these new nanomaterials, and very 7 few provide an exhaustive in vivo characterization, where key aspects, such as 8 pharmacokinetics, bioavailability, and, most importantly, toxicity, are properly evaluated. 9 In this work, we propose a protocol for the comprehensive assessment of the toxicity of 10 MNPs, based on the use of Zebrafish embryos as an intermediate screening step between 11 cell culture assays and studies in rodents. MNPs with different cores, ferrite and 12 manganese ferrite oxide, and sizes between 3 and 20 nm, were evaluated. Cell viability 13 at a concentration of 50 μg/mL of PEGylated MNPs was above 90 % in all cases. 14 However, the exposure of Zebrafish embryos to manganese based MNPs at 15 concentrations above 100 g/mL showed a low survival rate (<50%). In contrast, no 16 mortality (survival rate ~100%) and normal hatching rate were obtained for the iron oxide 17 MNPs. Based on these results, together with the physicochemical and magnetic properties 18 (r2=153.6 mM-1·s-1), the PEGylated 20 nm cubic shape iron oxide MNPs were selected 19 and tested in mice, showing very good MRI contrast and, as expected, absence of toxicity. 20 1. Introduction 21 In the last decades, a large part of nanomedicine research has focused on the development 22 of engineered functional nanoparticles (NPs) for in vivo applications due to their 23 impressive magnetic, optical or thermal properties, together with improved 24 3 biocompatibility.[1-3] These characteristics make NPs excellent candidates for a new 1 generation of diagnostic and/or therapeutic agents, with applicability in many diseases, 2 such as cancer, cardiovascular or neurodegenerative diseases.[4, 5] Among these 3 engineered NPs are the magnetic NPs, which are being thorough studied as a potential 4 new generation of MRI contrast agents for clinical diagnosis.[6-8] However, despite the 5 considerable amount of work devoted to this field of research, a very limited clinical 6 translation has been achieved to date. Indeed, ferumoxytol is currently the only FDA- 7 approved magnetic NPs with potential application as MRI contrast agent in patients with 8 impaired renal function.[9, 10] The reasons underlying this situation are likely related to 9 the fact that many of the studies on MNPs lack crucial information about the in vivo 10 behavior of these new nanomaterials, such as their pharmacokinetics, bioavailability, and, 11 most importantly, their toxicity. Although toxicity is one of the key aspects in the 12 characterization of nanomaterials intended for in vivo use, most studies on engineered 13 NPs simply report in vitro cytotoxicity assays, typically MTT and LDH assays,[11] and 14 very few report in vivo toxicity assays.[12] The use of mammalian animal models, 15 typically mice and rats, presents ethical issues due the large number of animals that should 16 be used for the toxicological screening of all the NPs that are being produced. To 17 overcome this limitation, several invertebrate animal models, such as Caenorhabditis 18 elegans,[13] Drosophila melanogaster,[14] have been used. However, from a 19 translational point of view, vertebrate animal models, specially the Zebrafish (Danio 20 rerio), are more appropriate because they exhibit a larger degree of similarity with 21 mammals in terms of early development and signaling repertoire.[15] Thus, the Zebrafish 22 model has emerged as an alternative for in vivo toxicological assays of new engineered 23 functional NPs,[16-19] being already well established as a model on environmental 24 sciences.[20, 21] Zebrafish has several advantages over mammalian models, such as the 25 4 possibility of performing high-throughput screening due to the high fecundity with rapid 1 embryos development, and the low cost of husbandry and housing. Therefore, the 2 Zebrafish model is an excellent option for in vivo toxicity screening of nanomaterials, 3 previous to in vivo experiments in mammalian models. Herein, we propose a protocol for 4 the comprehensive toxicity assessment, from in vitro to in vivo, of MRI contrast agents 5 based on magnetic NPs. MNPs with different cores, ferrite and manganese ferrite oxide, 6 and sizes between 3 and 20 nm, were studied. According to previous studies, the 7 incorporation of the paramagnetic manganese ion in the 3D structure of the NPs induces 8 an enhancement of the magnetic properties of the materials, which could lead to the 9 potential use of the MNPs as T1 or dual contrast agent, instead of the usual T2 contrast of 10 the iron oxide MNPs.[22] Following this hypothesis, we prepared both iron oxide and 11 ferrite manganese NPs and compared them in terms of magnetic properties and toxicity. 12 NPs were functionalized with a 3 kDa PEGylated ligand, which enhances their stability 13 in physiological medium without altering their properties. The toxicity of these 14 PEGylated NPs was first assessed in vitro on a mouse cell line. Then, Zebrafish embryos 15 were used for in vivo toxicity screening. Finally, the most promising magnetic NP was 16 selected and evaluated in mice, including in vivo pharmacokinetics and biodistribution 17 experiments by MRI, and histological analysis of tissue sections. 18 2. Results and Discussion 19 Synthesis of PEGylated magnetic NPs. 20 As commented above, in this work we prepared two different sets of magnetic NPs, one 21 of them based on iron oxide and the other one on iron manganese oxide. Each set of NPs 22 was divided on two sizes, small particles around 3 nm and large particles between 14 and 23 20 nm. The synthesis of the different NPs was carried out following the protocol described 24 5 by Hyeon and col.[23] with slight modifications (Figure 1). The subsequent 1 functionalization with a 3 kDa gallol-PEG-OH ligand yielded highly stable monodisperse 2 and water soluble NPs, hereinafter referred to as: Fe1 (3 nm ferrite NP), MnFe1 (3 nm 3 manganese ferrite NP), Fe2 (19 nm ferrite NP) and MnFe2 (14 nm manganese ferrite NP). 4 The presence of the gallol-PEG ligand at the nanoparticle surface was confirmed by FTIR 5 spectroscopy (Figure S1 and S2). The spectra of the functionalized MNPs showed the 6 main peaks of the corresponding gallol-PEG ligand. The hydrodynamic (HD) diameters 7 were measured by Dynamic Light Scattering (DLS) in two media, water and phosphate 8 buffered saline (PBS) to ensure that the stability of the particles remained unaltered at 9 physiological pH and ionic strength (Table S1). The small MNPs, Fe1 and MnFe1, 10 presented HD diameters around 21 nm in water, and slightly larger, between 24-30 nm, 11 in PBS. On the other hand, the large MNPs, Fe2 and MnFe2, presented HD diameters 12 around 56 nm in water, and 63 and 82 nm in PBS, respectively. Therefore, the presence 13 of salt produced small differences in HD diameters, which can be explained by a 14 modification of the hydration/solvation state of the NPs. Moreover, the stability of the 15 MNPs under physiological conditions was studied by measuring the HD diameters in PBS 16 during 24 h, resulting in HD diameters values similar to the initial ones (Figure S3). In 17 addition, the zeta potential of the PEGylated MNPs was slightly negative, which, together 18 with the above results, indicate that these MNPs are highly stable and do not form 19 aggregates in physiological media. To evaluate the potential of these MNPs as MRI 20 contrast agents, T1 and T2 relaxivities (r1 and r2) were measured at low (1.44 T) and high 21 (9.4 T) magnetic fields, showing similar trends in both cases. The smaller PEGylated Fe1 22 and MnFe1 exhibited the lower r2 values, 37.4 and 19.0 mM-1·s-1 at 9.4 T and 12 and 6.0 23 mM-1·s-1 at 1.44 T for the Fe1 and MnFe1, respectively. These values are in good 24 agreement with previously reported results for similar magnetic nanoparticles, which 25 6 described the correlation between size and magnetic properties, such as r2 and mass 1 magnetization.[24, 25] On the other hand, as expected, the larger MNPs showed higher 2 r2 values compared to the smaller ones, 154 and 51 mM-1·s-1 at 9.4 T and 80 and 37 mM- 3 1·s-1 at 1.44 T for the Fe2 and MnFe2, respectively (Figures S4-S6). The considerable 4 higher r2 values of the Fe2 are not only due to the larger size, but also to the fact that they 5 exhibit a cubic shape, whereas MnFe2 are spherical. The cubic shape affects the magnetic 6 properties inducing a more ferromagnetic behaviour and therefore higher magnetic 7 susceptibility.[26] Regarding T1 relaxivity, small and large MNPs behaved differently 8 (Figures S7-S8). At low magnetic field, small MNPs behaved as dual T1/T2 contrast 9 agents with r2/r1 ratios of 4.8 and 3.8 for Fe1 and MnFe1, respectively. In contrast, large 10 MNPs behaved exclusively as T2 contrast agents. At high magnetic field, all MNPs 11 showed r1 values close to 0, restricting their use to T2 contrast agents (Table S2 and S3). 12 The cytotoxicity of these engineered MNPs was evaluated in vitro using the mouse 13 microglial cell line N13. Cytotoxicity was tested in the concentration range of MNPs 14 between 0.1 to 100 μg/mL (Figure S10). Cultured cells showed no significant cytotoxicity 15 after 24 h of exposure to any of the MNPs, with cell viability values above 90% at a 16 concentration of 50 μg/mL of PEGylated MNPs. These results are in good agreement with 17 previously reported data in which not only iron oxide NPs, but also manganese ferrite 18 NPs did not show any substantial toxicity in different cell lines, such as SMMC-7721[27] 19 or PC-3[24]. Then, before testing the biocompatibility and bioavailability of these 20 potential MRI contrast agents in mice, an in vivo toxicity screening was performed on 21 Zebrafish. This approach consisted on the evaluation of the hatching and survival rates of 22 Zebrafish embryos exposed to different concentrations of engineered MNPs (0.01, 0.1, 1, 23 10, 100 µg/mL) at different times post fertilization (Figure 2a). Regarding the hatching 24 process, control embryos showed around 50% of hatching at 48 h post fertilization (hpf) 25 7 (data not shown), in agreement with reported hatching values for normal Zebrafish 1 embryos.[28] Therefore, 48 hpf was selected as the time to assess whether premature 2 hatching occurred in the groups exposed to the MNPs. Higher concentrations of MNPs 3 (10 µg/ml and 100 µg/ml) showed an increased hatching rate in comparison with control 4 non-exposed embryos. As previously reported, this early hatching was probably due to 5 the adsorption of particles on the chorion, which block the pores with the consequent 6 restriction of oxygen and nutrients (Figure 2a).[29] No mortality or malformations were 7 observed in the embryos exposed to different doses of particles at 48 hpf. Another key 8 parameter in the assessment of in vivo toxicity is the survival rate of Zebrafish embryos. 9 In the case of iron oxide NPs (Fe1 and Fe2), the survival rate was almost 100% in both 10 cases, in agreement with other studies in which different formulations of water soluble 11 iron oxide NPs were also tested in Zebrafish embryos.[18, 30] However, the survival rate 12 for manganese based NPs (MnFe1 and MnFe2) was 100 % only for low doses, showing 13 a high percentage of mortality for a concentration of 100 g/mL of MNPs after 6 days 14 post fertilization. (Figure 2b and S11). This toxic effect could be due to a slow release of 15 the Mn cations from the inorganic core, which would lead to neurotoxicity by 16 mitochondrial dysfunction.[31, 32] Similar results have been described for cobalt ferrite 17 NPs, which were proved very lethal even at low concentration.[33] Therefore, mortality 18 of the Zebrafish embryos was found to be nanoparticle type-, dose- and time-dependent. 19 The discordant results obtained between cell cultures and Zebrafish embryos in the case 20 of manganese ferrite NPs demonstrate the importance of including animal models in the 21 toxicity studies of NPs designed for in vivo use. 22 In summary, whereas the in vitro cytotoxicity assays in cell culture showed low toxicity 23 for all MNPs, the in vivo assays in Zebrafish embryos revealed that only iron oxide 24 nanoparticles are suitable for in vivo use. Specifically, only Fe1 and Fe2 were selected as 25 8 potential contrast agents for in vivo imaging based on their low toxicity. Also, these iron 1 based NPs presented a shape-dependent behaviour in terms of relaxivity. Cubic MNPs 2 showed mainly T2 contrast at low and high magnetic fields, whereas spherical MNPs 3 showed dual T1 and T2 contrast at low magnetic field and only T2 contrast at 9.4 T. Based 4 on these results, the engineered cubic Fe2 was selected among all NPs as the most 5 promising contrast agent for MRI, being then characterized in vivo in Balb/c mice. Thus, 6 Fe2 (5 mg of Fe per kg of body weight) was intravenously injected in the tail vein of 7 Balb/c mice, and followed by MRI.[34] Short term pharmacokinetics (up to 30 min) 8 showed a rapid liver uptake, with a maximum relative enhancement (RE, Figure 3) at one 9 minute after injection. The kidneys also showed a significant RE, which remained almost 10 constant during the 30 min of the dynamic MRI experiment, involving that a significant 11 amount of Fe2 was circulating during this time period, as discussed elsewhere.[24, 35] 12 Moreover, long term pharmacokinetics based on quantitative T2 mapping showed a clear 13 T2 decrease in liver and kidneys at 1 h after injection, with ΔT2 of - 5.5 and -15.8 in liver 14 and kidneys, respectively. These quantitative data further support that, in spite of being 15 partially taken up by the kupffer cells of the liver, a significant amount of Fe2 is still 16 present in the bloodstream after one hour, which means that these MNPs show good 17 bioavailability, as previously described by our group for similar PEGylated magnetic 18 NPs.[36] Finally, the possible side effects produced by the intravenous injection of MNPs 19 were also evaluated by histological analyses. Thus, hematoxylin and eosin stain was 20 performed on tissue samples of liver and kidneys, showing no significant alterations after 21 exposure to Fe2 compared to controls. Liver sections showed normal appearance, without 22 the vacuolated swelling of the cytoplasm of hepatocytes that is typically indicative of 23 acute and subacute liver injury. Similarly, kidneys presented normal tubular brushborders 24 and intact glomeruli surrounding Bowman’s capsule, evidential of no kidney injury 25 9 (Figure 4).[36, 37] Moreover, the mouse body weight was also followed during 28 days 1 after the administration of Fe2, showing the same pattern as the control group, in 2 agreement with the absence of toxic effect in vivo (Figure 4). 3 3. Conclusion 4 In conclusion, an exhaustive toxicity screening, based on cell cultures, Zebrafish embryos 5 and mice, is proposed to properly evaluate the toxicity of magnetic nanoparticles 6 developed as potential MRI contrast agents. Our results demonstrate that toxicity tests 7 based on cell cultures alone are not enough to evaluate the toxicity of nanomaterials that 8 are intended for in vivo use. Unfortunately, this is a common practice in many of the 9 studies in the field, but in our experience, the use animal models should always be 10 included as part of the protocol to evaluate the toxicity of new nanomaterials for in vivo 11 applications. The Zebrafish model is an excellent option as an intermediate screening step 12 to properly evaluate toxicity avoiding the excessive use of rodents. In this work, two 13 different magnetic cores, based on ferrite and manganese ferrite oxides, with different 14 sizes, from 3 to 20 nm, were studied. A detailed characterization on the PEGylated MNPs 15 showed high long term stability, no in vitro toxicity and good T2 and dual T2/T1 MRI 16 contrast at high and low magnetic fields, respectively. An intermediate in vivo toxicity 17 screening was performed on Zebrafish embryos to select the best candidate for the MRI 18 experiments in mice. No mortality and normal hatching rate were obtained on the 19 Zebrafish embryos exposed to iron oxide NPs. However, in contrast to in vitro results, 20 significant toxicity was obtained for magnetic NPs based on manganese ferrite. A low 21 percentage of survival was found on the embryos exposed to the highest concentration of 22 manganese ferrite NPs after 6 days post fertilization. From all the tested MNPs, Fe1 and 23 Fe2 showed the lowest toxicity and between these two, the Fe2 (20 nm cubic shape) 24 presented the best characteristics as a potential MRI contrast agent. Therefore, this MNP 25 16 Cytotoxicity assays. Briefly, the N-13 cells were plated at a density of 1 x 104 cells/well 1 in a 96-well plate at 37 ºC in 5 % CO2 atmosphere (200 L per well, number of repetitions 2 = 5). After 24 h of culture, the medium in the wells was replaced with fresh medium 3 containing gallol derived magnetic nanoparticles in varying concentrations from 0.1 4 g/mL to 100 g/mL. After 24 h, the supernatant of each well was replaced by 200 L of 5 fresh medium with 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide 6 (MTT) (0.5 mg/mL). After 2 h of incubation at 37 ºC and 5 % CO2 the medium was 7 removed and the formazan crystals were solubilized with 200 L of DMSO, and the 8 solution was vigorously mixed to dissolve the reacted dye. Two controls were performed 9 to evaluate the cytotoxicity: as negative control, cells unexposed to MNPs were used, and 10 as positive control, the cells were exposed to Triton X-100 (1 % v/v) for 15 min prior to 11 the MTT procedure. The absorbance of each well was read on a microplate reader 12 (Dynatech MR7000 instruments) at 550 nm. The relative cell viability (%) and its error 13 related to control wells containing cell culture medium without nanoparticles were 14 calculated by the equations: 15 𝑹𝒆𝒍𝒂𝒕𝒊𝒗𝒆 𝑪𝒆𝒍𝒍 𝑽𝒊𝒂𝒃𝒊𝒍𝒊𝒕𝒚 (𝑹𝑪𝑽) (%)=[𝑨𝒃𝒔]𝒕𝒆𝒔𝒕− [𝑨𝒃𝒔] 𝑷𝒐𝒔.𝑪𝒐𝒏𝒕𝒓𝒐𝒍 [𝑨𝒃𝒔] 𝑵𝒆𝒈.𝑪𝒐𝒏𝒕𝒓𝒐𝒍−[𝑨𝒃𝒔] 𝑷𝒐𝒔.𝑪𝒐𝒏𝒕𝒓𝒐𝒍×𝟏𝟎𝟎 16 17 𝑬𝒓𝒓𝒐𝒓 (%)=𝐑𝐂𝐕𝐭𝐞𝐬𝐭× √([𝛔]𝐭𝐞𝐬𝐭 [𝐀𝐛𝐬]𝐭𝐞𝐬𝐭) 𝟐 +( [𝛔]𝐜𝐨𝐧𝐭𝐫𝐨𝐥 [𝐀𝐛𝐬]𝐜𝐨𝐧𝐭𝐫𝐨𝐥) 𝟐 18 where  is the standard deviation. 19 Non parametric tests were used for statistical analysis using IBM SPSS package v22. 20 Teratogenicity assay. Wild type zebrafish were outcrossed at day 0. Embryos well 21 collected and incubated at 28 ºC in E3 medium for 4 h. Not fertilized eggs were then 22 removed. Properly developing embryos were then incubated with different concentrations 23 of the test products dissolved in E3 medium. 20-30 eggs were placed in 8-well square 24 17 petri-dishes in a volume of 4 mL. Survival, hatching and malformations were observed 1 in 24, 48, 72 and 144 hpf when experiments were finished 2 In vivo Magnetic Resonance Imaging. In vivo mice experiments were performed in 3 accordance with the ethical guidelines of Andalusian government. Male Balb/c mice (n = 4 3) with ca. 22 g in weight, provided by Janvier Labs were used. Animals were 5 anesthetized with 1 % isoflurane, the tail vein was cannulated and then the animals were 6 placed in the magnet, where respiration and body temperature were monitored throughout 7 the entire MRI experiment. The magnetic nanoparticle was administered intravenously 8 via tail vein at a concentration of 5 mg of Fe per kg of body weight. 9 All the MRI experiments were carried out on a 9.4 T Bruker Biospec system equipped 10 with a 400 mT/m gradients and a 40 mm quadrature bird-cage resonator. High resolution 11 T2-weighted images were acquired using a turbo-RARE sequence with respiratory gating 12 (TE = 16 ms, TR = 1000 ms, 4 averages, 156 m in-plane resolution and 1 mm slice 13 thickness). Quantitative T2 measurements were also performed using a multi-echo spin 14 echo sequence (TEs ranging from 7 ms to 448 ms, TR = 3500 ms, FOV = 4 cm, matrix 15 size = 128x128, slice thickness = 1 mm). The time-courses were followed by using a 16 turbo-RARE sequence with the same parameters indicated above, but only 1 average to 17 improve temporal resolution (1 image every 30 seconds). The acquisition scheme was as 18 follows: T2-weighted, quantitative T2, intravenous injection of the gallol derived magnetic 19 nanoparticles, time-course for 35 min, quantitative T2 and T2 weighted. The first 35 min 20 time courses were analyzed semi-quantitatively using the following expression: 21 𝑹𝑬=|𝑰 𝒕 −𝑰 𝟎 𝑰 𝟎 ×𝟏𝟎𝟎 | 22 23 18 where RE is the modulus of relative signal enhancement, It is the signal intensity at any 1 given time after the nanoparticles injection, and I0 is the signal intensity before the 2 injection. 3 Long-term pharmacokinetics were measured by quantitative T2 mapping at 0 h and 1 h. 4 Pharmacokinetics were obtained by calculating the average values within different 5 regions of interest (ROIs) placed on the following tissues: liver and kidneys, spleen and 6 muscle. 7 Statistical analysis. The statistical analysis was performed using the SPSS package (SPSS 8 Inc., Chicago, Illinois). Cell viability, hatching, survival rates of embryos, in vivo T2 9 values and weight variation are shown as mean ± standard deviation (SD). Student’s t- 10 test or two-way analysis of variance were used to determine significant differences 11 between different MNPs or different experimental conditions. The level of significance 12 was set at p<0.05. 13 Supporting Information 14 DLS/zeta potential, FTIR, In vitro relaxivities, Stability studies, Cytotoxicity assays and 15 Teratogenicity assays. 16 17 Acknowledgements 18 Financial support was provided by the Andalusian Ministry of Health (PI2013-0559 to 19 MPL), the Spanish Ministry of Economy and Competitiveness (CTQ2017-86655-R to 20 MPL and MLGM). MPL thanks to the V Plan Propio of the University of Seville for the 21 Postdoctoral Fellowship. The authors thank Juan F. López and M. Carmen Muñoz- 22 Hernández for assistance with TEM and MRI experiments. 23 24 19 1 Figure 1. TEM images of the four different nanoparticles (left): Fe1 (a), MnFe1 (b), Fe2 2 (c) and MnFe2 (d). The scale bar is equivalent to 50 nm in all the images. The size 3 distribution histograms are shown in each case. Diameters are expressed as the mean ± 4 SD, by measuring at least 100 particles. 5 6 7 Figure 2. A) Hatching rate of zebrafish embryos exposed to MNP at different 8 concentration 48 hpf B). Dose-dependent survival rate of the zebrafish embryos exposed 9 to MNP at 6 dpf. All data are expressed as the mean ± SD by analyzing 20-30 eggs per 10 NP concentration. 11 20 1 Figure 3. Left) In vivo time courses of Fe2 after intravenously injection in balb/c mice: 2 top) Liver; bottom) Kidneys. Right). Representative T2-weighted MR images at different 3 experimental times after the intravenous injection of Fe2. top) T2-weighted images of 4 liver at 0 and 1 h after injection. bottom) T2-weighted images of kidneys and muscle at 0 5 and 1 h respectively after injection. Table: T2 quantitative averages and ΔT2 values of 6 liver, kidneys and muscle at different times after intravenously injection of Fe2. The 7 average values were obtained by performing three experiments. 8 9 21 1 Figure 4. Representative histological sections of: liver, control (a) and Fe2 injected (b); 2 and kidney, control (c) and Fe2 injected (d). (Right) Weight profile of intravenously 3 injected mice and control mice in relative values. Data are expressed as the mean ± SD 4 (n=3). 5 6 References 7 [1] Bogart LK, Pourroy G, Murphy CJ, Puntes V, Pellegrino T, Rosenblum D, et al. 8 Nanoparticles for Imaging, Sensing, and Therapeutic Intervention. ACS Nano 9 2014;8:3107-22. 10 [2] Kreyling WG, Abdelmonem AM, Ali Z, Alves F, Geiser M, Haberl N, et al. In vivo 11 integrity of polymer-coated gold nanoparticles. Nature Nanotechnology 2015;10:619. 12 [3] Caro C, Dalmases M, Figuerola A, García-Martín ML, Leal MP. Highly water-stable 13 rare ternary Ag–Au–Se nanocomposites as long blood circulation time X-ray computed 14 tomography contrast agents. Nanoscale 2017;9:7242-51. 15 [4] Caro C, Pozo D. Polysaccharide colloids as smart vehicles in cancer therapy. Current 16 Pharmaceutical Design 2015;21:4822-36. 17 22 [5] Chen L, Watson C, Morsch M, Cole NJ, Chung RS, Saunders DN, et al. Improving 1 the Delivery of SOD1 Antisense Oligonucleotides to Motor Neurons Using Calcium 2 Phosphate-Lipid Nanoparticles. Frontiers in Neuroscience 2017;11. 3 [6] Becerro AI, Fuente JMdl, García-Martín ML, González-Mancebo D, Ocaña M, Rojas 4 TC. HoF3 and DyF3 Nanoparticles as Contrast Agents for High-Field Magnetic 5 Resonance Imaging. 2017. 6 [7] Lahooti A, Sarkar S, Laurent S, Shanehsazzadeh S. Dual nano-sized contrast agents 7 in PET/MRI: a systematic review. Contrast Media & Molecular Imaging 2016;11:428- 8 47. 9 [8] Materia ME, Pernia Leal M, Scotto M, Balakrishnan PB, Kumar Avugadda S, García- 10 Martín ML, et al. Multifunctional Magnetic and Upconverting Nanobeads as Dual Modal 11 Imaging Tools. Bioconjugate Chemistry 2017;28:2707-14. 12 [9] Toth GB, Varallyay CG, Horvath A, Bashir MR, Choyke PL, Daldrup-Link HE, et al. 13 Current and potential imaging applications of ferumoxytol for magnetic resonance 14 imaging. Kidney Int 2017;92:47-66. 15 [10] Ventola CL. Progress in Nanomedicine: Approved and Investigational Nanodrugs. 16 P T 2017;42:742-55. 17 [11] Eskandari N, Nejadi Babadaei MM, Nikpur S, Ghasrahmad G, Attar F, Heshmati M, 18 et al. Biophysical, docking, and cellular studies on the effects of cerium oxide 19 nanoparticles on blood components: in vitro. International journal of nanomedicine 20 2018;13:4575-89. 21 [12] Liegertová M, Wrobel D, Herma R, Müllerová M, Šťastná LČ, Cuřínová P, et al. 22 Evaluation of toxicological and teratogenic effects of carbosilane glucose 23 glycodendrimers in zebrafish embryos and model rodent cell lines. Nanotoxicology 2018. 24 23 [13] Huang H, Delikanli S, Zeng H, Ferkey DM, Pralle A. Remote control of ion channels 1 and neurons through magnetic-field heating of nanoparticles. Nature Nanotechnology 2 2010;5:602. 3 [14] Doubrovinski K, Swan M, Polyakov O, Wieschaus EF. Measurement of cortical 4 elasticity in <em>Drosophila melanogaster</em> embryos using ferrofluids. 5 Proceedings of the National Academy of Sciences 2017;114:1051. 6 [15] Schilling TF, Webb J. Considering the zebrafish in a comparative context. Journal 7 of Experimental Zoology Part B: Molecular and Developmental Evolution 8 2007;308B:515-22. 9 [16] Fleming A, Alderton WK. Zebrafish in pharmaceutical industry research: Finding 10 the best fit. Drug Discovery Today: Disease Models 2013;10:e43-e50. 11 [17] George S, Xia T, Rallo R, Zhao Y, Ji Z, Lin S, et al. Use of a high-throughput 12 screening approach coupled with in vivo zebrafish embryo screening to develop hazard 13 ranking for engineered nanomaterials. ACS Nano 2011;5:1805-17. 14 [18] Rizzo LY, Golombek SK, Mertens ME, Pan Y, Laaf D, Broda J, et al. In vivo 15 nanotoxicity testing using the zebrafish embryo assay. Journal of Materials Chemistry B 16 2013;1:3918-25. 17 [19] Zhu X, Tian S, Cai Z. Toxicity Assessment of Iron Oxide Nanoparticles in Zebrafish 18 (Danio rerio) Early Life Stages. PLOS ONE 2012;7:e46286. 19 [20] Felix LC, Ortega VA, Ede JD, Goss GG. Physicochemical Characteristics of 20 Polymer-Coated Metal-Oxide Nanoparticles and their Toxicological Effects on Zebrafish 21 (Danio rerio) Development. Environmental Science & Technology 2013;47:6589-96. 22 [21] Zhang Y, Zhu L, Zhou Y, Chen J. Accumulation and elimination of iron oxide 23 nanomaterials in zebrafish (Danio rerio) upon chronic aqueous exposure. Journal of 24 Environmental Sciences 2015;30:223-30. 25 24 [22] Lee J-H, Huh Y-M, Jun Y-w, Seo J-w, Jang J-t, Song H-T, et al. Artificially 1 engineered magnetic nanoparticles for ultra-sensitive molecular imaging. Nature 2 Medicine 2006;13:95. 3 [23] Park J, An K, Hwang Y, Park JEG, Noh HJ, Kim JY, et al. Ultra-large-scale 4 syntheses of monodisperse nanocrystals. Nature materials 2004;3:891-5. 5 [24] Pernia Leal M, Rivera-Fernández S, Franco JM, Pozo D, De La Fuente JM, García- 6 Martín ML. Long-circulating PEGylated manganese ferrite nanoparticles for MRI-based 7 molecular imaging. Nanoscale 2015;7:2050-9. 8 [25] Materia ME, Guardia P, Sathya A, Pernia Leal M, Marotta R, Di Corato R, et al. 9 Mesoscale assemblies of iron oxide nanocubes as heat mediators and image contrast 10 agents. Langmuir 2015;31:808-16. 11 [26] Kakwere H, Leal MP, Materia ME, Curcio A, Guardia P, Niculaes D, et al. 12 Functionalization of Strongly Interacting Magnetic Nanocubes with (Thermo)Responsive 13 Coating and Their Application in Hyperthermia and Heat-Triggered Drug Delivery. ACS 14 Applied Materials & Interfaces 2015;7:10132-45. 15 [27] Yang L, Ma L, Xin J, Li A, Sun C, Wei R, et al. Composition Tunable Manganese 16 Ferrite Nanoparticles for Optimized T2 Contrast Ability. Chemistry of Materials 17 2017;29:3038-47. 18 [28] Guo Y, Chen L, Wu J, Hua J, Yang L, Wang Q, et al. Parental co-exposure to 19 bisphenol A and nano-TiO2 causes thyroid endocrine disruption and developmental 20 neurotoxicity in zebrafish offspring. Science of The Total Environment 2019;650:557- 21 65. 22 [29] Pham DH, De Roo B, Nguyen XB, Vervaele M, Kecskés A, Ny A, et al. Use of 23 Zebrafish Larvae as a Multi-Endpoint Platform to Characterize the Toxicity Profile of 24 Silica Nanoparticles. Scientific Reports 2016;6. 25 25 [30] Cáceres-Vélez PR, Fascineli ML, Grisolia CK, de Oliveira Lima EC, Sousa MH, de 1 Morais PC, et al. Genotoxic and histopathological biomarkers for assessing the effects of 2 magnetic exfoliated vermiculite and exfoliated vermiculite in Danio rerio. Sci Total 3 Environ 2016;551-552:228-37. 4 [31] Dobson AW, Erikson KM, Aschner M. Manganese Neurotoxicity. Annals of the 5 New York Academy of Sciences 2004;1012:115-28. 6 [32] Caro C, García-Martín ML, Pernia Leal M. Manganese-Based Nanogels as pH 7 Switches for Magnetic Resonance Imaging. Biomacromolecules 2017;18:1617-23. 8 [33] Ahmad F, Liu X, Zhou Y, Yao H. An in vivo evaluation of acute toxicity of cobalt 9 ferrite (CoFe2O4) nanoparticles in larval-embryo Zebrafish (Danio rerio). Aquatic 10 Toxicology 2015;166:21-8. 11 [34] Caro C, Carmen Muñoz-Hernández M, Leal MP, García-Martín ML. In vivo 12 pharmacokinetics of magnetic nanoparticles. Methods in Molecular Biology2018. p. 13 409-19. 14 [35] Leal MP, Muñoz-Hernández C, Berry CC, García-Martín ML. In vivo 15 pharmacokinetics of T2 contrast agents based on iron oxide nanoparticles: optimization 16 of blood circulation times. RSC Advances 2015;5:76883-91. 17 [36] Pernia Leal M, Caro C, García-Martín ML. Shedding light on zwitterionic magnetic 18 nanoparticles: Limitations for in vivo applications. Nanoscale 2017;9:8176-84. 19 [37] Peter AI, Naidu ECS, Akang E, Ogedengbe OO, Offor U, Rambharose S, et al. 20 Investigating Organ Toxicity Profile of Tenofovir and Tenofovir Nanoparticle on the 21 Liver and Kidney: Experimental Animal Study. Toxicological Research 2018;34:221-9. 22 [38] Riedinger A, Pernia Leal M, Deka SR, George C, Franchini IR, Falqui A, et al. 23 "nanohybrids" based on pH-responsive hydrogels and inorganic nanoparticles for drug 24 delivery and sensor applications. Nano Letters 2011;11:3136-41. 25