Cellular and Molecular Processes Are Differently Influenced in Primary Neural Cells by Slight Changes in the Physicochemical Properties of Multicore Magnetic Nanoparticles

Herein, we use two exemplary superparamagnetic iron oxide multicore nanoparticles (SPIONs) to illustrate the significant influence of slightly different physicochemical properties on the cellular and molecular processes that define SPION interplay with primary neural cells. Particularly, we have designed two different SPION structures, NFA (i.e., a denser multicore structure accompanied by a slightly less negative surface charge and a higher magnetic response) and NFD (i.e., a larger surface area and more negatively charged), and identified specific biological responses dependent on SPION type, concentration, exposure time, and magnetic actuation. Interestingly, NFA SPIONs display a higher cell uptake, likely driven by their less negative surface and smaller protein corona, more significantly impacting cell viability and complexity. The tight contact of both SPIONs with neural cell membranes results in the significant augmentation of phosphatidylcholine, phosphatidylserine, and sphingomyelin and the reduction of free fatty acids and triacylglycerides for both SPIONs. Nonetheless, NFD induces greater effects on lipids, especially under magnetic actuation, likely indicating a preferential membranal location and/or a tighter interaction with membrane lipids than NFA, in agreement with their lower cell uptake. From a functional perspective, these lipid changes correlate with an increase in plasma membrane fluidity, again larger for more negatively charged nanoparticles (NFD). Finally, the mRNA expression of iron-related genes such as Ireb-2 and Fth-1 remains unaltered, while TfR-1 is only detected in SPION-treated cells. Taken together, these results demonstrate the substantial impact that minor physicochemical differences of nanomaterials may exert in the specific targeting of cellular and molecular processes. A denser multicore structure generated by autoclave-based production is accompanied by a slight difference in surface charge and magnetic properties that become decisive for the biological impact of these SPIONs. Their capacity to markedly modify the lipidic cell content makes them attractive as lipid-targetable nanomedicines.


Synthesis and characterization of iron oxide nanoparticles
The synthesis of multicore SPIONs was based on previous works. 41,42 Two heating systems were used here, one based on a heating mantle and glass under reflux and mechanical stirring conditions (NFD), and other based on a teflon lined stainless steel autoclave (NFA), which is a closed system that allows to minimize handling. In both cases, 3.2 mmol of FeCl 3 •6H 2 O and 1.6 mmol of FeCl 2 •4H 2 O were dissolved in 64 g of a mixture of DEG and NMDEA at 50 %. Separately, NaOH was dissolved in 32 g of the same DEG and NMDEA mixture under ultrasounds for 30 min. The two separated solutions were mixed for 5 min, transferred to the glass beaker or to the autoclave and heated at 220 ºC during 16 h. The precipitate was recovered using a magnet and washed with a mixture of ethanol and ethyl acetate four times. The nanoparticles were subjected to an acid treatment S3 following a protocol previously reported. 43 The precipitate was dispersed in nitric acid (10 %). Then, Fe(NO 3 ) 3 •9H 2 O was added and the mixture was heated to 80 ºC for 45 min. After cooling, the nanoparticles were dispersed with nitric acid (10 %) again. Finally, the sample was washed by magnetic decantation with ethanol and acetone, and finally dispersed in water. Acetone residues were eliminated using a rotatory evaporator.
All synthesized magnetic nanoparticles were coated with citric acid. For this purpose, the pH of the nanoparticle suspension previously prepared containing 20 mg of Fe was adjusted to pH 2 and 13 mL of 0.1 M citric acid were added, heated at 80 °C for 30 min and then, centrifuged, washed with distilled water and dispersed, first at pH 11, and then, at pH 7. Colloidal properties were analyzed by Dynamic Light Scattering (DLS) in a NanoZsizer apparatus from Malvern to determine the hydrodynamic size (D hydro ) and surface charge. The core size and morphology of the nanoparticles were analyzed by transmission electron microscopy (TEM, JEOL JEM 1010). Size distribution was determined by measuring ~150 particles with the ImageJ digital software. Data were fitted to a log-normal curve and the mean value (D TEM ) obtained. Surface and internal composition was analyzed by energy-dispersive X-ray spectroscopy (EDX) and the spectra were taken on a field emission scanning electron microscope (SEM, FEI Verios 460) at 2 kV accelerating voltage and a probe current of 13 pA. In addition, Fourier-transform infrared spectroscopy (FTIR) was carried out at in a spectrophotometer Bruker Vertex 70V with 2 cm -1 resolutions in KBr pellets and the FTIR absorbance spectra were collected over the 4000-500 cm -1 range.
Magnetic characterization was carried out using Vibrating Sample Magnetometry (VSM, Oxford instrument) and SQUID magnetometers (Quantum Design). Hysteresis S4 loops were recorded at room temperature after applying a magnetic field of ± 5 T and saturation magnetization by extrapolation to infinity field (Ms) and coercivity (Hc) were obtained in the VSM. The sample consisted in 100 µL of the colloidal suspension dried on a piece of cotton. Zero Field Cooling and Field Cooling (ZFC/FC) magnetization curves were recorded between 300 and 5 K at 100 Oe in the SQUID. Fe concentration in the colloidal and cell suspensions was determined by inductively coupled plasma optical emission spectroscopy (ICP-OES) in a PERKIN ELMER OPTIMA 2100 DV apparatus after digestion with nitric acid and aqua regia at 90 ºC (1.2 x 10 6 cells in 10 mL). Samples were digested in aqua regia and diluted to a known volume. The heating efficiency was evaluated in a Five Celes apparatus with an Osensa temperature probe, measuring the temperature change under the application of an alternating magnetic field (280 kHz-20 mT and 90 kHz-60 mT) in 1 mL of sample at 20-25 mM Fe concentration. The crystal structure of the sample was identified by X-ray diffraction using a Bruker D8 ADVANCE diffractometer with Cu Kα radiation, between 10º and 90º in 2. Crystal size was calculated from the (311) peak broadening.
Dynamic light scattering was used to assess the protein corona formation when NFA and NFD nanoparticles were incubated with fresh (and complete) Neurobasal culture media at different time points (0, 15, 30, 60 and 120 min and 24 h) at 37 ºC in a 5% CO 2 atmosphere (inside a sterile cell incubator).

Primary neural cells isolation, culture and SPIONs exposure
Embryonic neural progenitor cells (ENPCs) were obtained from cerebral cortices of Wistar rat embryos as previously described. 44 Adult female Wistar rats were provided by S5 the animal facilities of the National Hospital for Paraplegics and sacrificed when gestation reached 16-17 days (E16-E17). All the experimental protocols for cell collection adhered to the regulations of the European Commission (directives 2010/63/EU and 86/609/EEC) and the Spanish Government (RD53/2013 and ECC/566/2015) for the protection of animals used for scientific purposes. A total of 15 independent cell cultures (N ≥ 3 per cell assay and triplicates per culture condition) were carried out, with cell viability being always above 85 %. Prior to cell culture, plates were coated with an aqueous solution of poly( Llysine) (PLL) (45 µg mL -1 ) for 1 h at room temperature, later washed with sterile water, and finally conditioned for 1-2 h in complete culture medium in a sterile incubator at 37 ºC under a 5% CO 2 atmosphere. Cell seeding density was 25 x 10 3 cells cm -2 for all experiments except for qRT-PCR studies in which seeding density was doubled to increase the quantity of mRNA per condition. For flow cytometry studies, ENPC suspensions right after isolation were exposed to SPIONs at a density of 100,000 cells mL -1 (a total suspension of 5 mL per treatment condition). Cells were maintained for different time points (days-in-vitro, DIV) in complete Neurobasal TM culture medium containing B-27 supplement (2 %), streptomycin (100 UI mL -1 ), penicillin (100 UI ml -1 ), and GlutaMAX (1 %), which was half-replaced every 3-4 days. At different culture times as specifically indicated for each type of assay, cells were exposed to either NFA or NFD SPIONs at the different concentrations of selection in a sterile incubator at 37 ºC under a 5% CO 2 atmosphere. After 24h of exposure, SPION-exposed cells were gently washed with phosphate buffer saline (PBS) twice and maintained in fresh culture media for another 24 h.

S6
For magnetic stimulation, cell culture Petri dishes of 3.5 cm in diameter were used.
Prior to ENPC culture, Petri dishes were coated with PLL as described above. Cells were then seeded and maintained in culture for different time points depending on the tests to be carried out. A total of six different conditions were evaluated: (1) cells without SPIONs nor magnetic field (Control), (2) cells only exposed to an alternating magnetic field (AMF), (3) cells only exposed to NFA SPIONs (NFA), (4) cells only exposed to NFD SPIONs (NFD), (5) cells exposed to both NFA SPIONs and a magnetic field (NFA + AMF) and (6) cells exposed to both NFD SPIONs and a magnetic field (NFD + AMF). ENPC cultures were exposed to the selected concentration/s of either NFA or NFD SPIONs 48 h before the application of the magnetic field. The hyperthermia treatment of selection was an alternating magnetic field of 20 mT for 60 min, with a frequency of 280 kHz, applied by using a Five Celes equipment.

Morphological studies by electron microscopies
Cell culture morphology after the exposure to SPIONs was first studied by fieldemission scanning electron microscopy (FESEM). Samples were fixed with glutaraldehyde (2.5 % in PBS) for 45 min. After gentle washing in distilled water, dehydration was performed by using series of ethanol solutions for 15 min (2 washes) and a final dehydration in absolute ethanol for 30 min. Samples were then dried at room temperature for at least 48 h. After mounting in stubs and coating with a nanometer-thick Chromium layer under vacuum, the morphology of the samples was immediately after characterized by using a field-emission Philips XL30 S-FEG microscope. For the visualization of SPION uptake by cells in suspension, ENPCs right after isolation were exposed to either NFA or NFD at different concentrations for 4 h at 37 °C.
Cell suspensions were then centrifuged at 300 g for 4 min. The obtained pellet of cells was fixed at room temperature with 4 % PFA and 2 % glutaraldehyde in 0.1 M phosphate buffer at pH 7.4 for 2 h, embedded in a gelatin matrix (10 % in bi-distilled water) and kept on ice until gelatin solidified. Then, it was cut in small cubes to proceed with embedding in epoxy resin as for a tissue block. Post-fixation was carried out with 1 % OsO 4 and 0.8 % K 3 Fe(CN) 6 in water at 4 °C for 1 h. Samples were dehydrated with ethanol and embedded in epoxy TAAB 812 resin (Laboratories, Berkshire, England) according to standard S8 procedures. After polymerization, 80-nm-thick (ultrathin) sections were obtained and stained with 2 % uranyl acetate solution in water and Reynolds lead citrate and examined at 100 kV in a Jeol JEM 1400 Flash (Tokyo, Japan) microscope. Pictures were taken with an OneView (Gatan) digital CMOS camera (4K x 4K). Images were acquired in both bright and dark field modes to clearly identify SPIONs inside the cells.

Viability studies by confocal fluorescence microscopy
Cell viability in culture was analyzed using a Live/Dead® viability kit according to manufacturer's instructions (Life Technologies). Briefly, the live and dead cells were stained with calcein and ethidium homodimer-1 (EthD-1), respectively. After staining, samples were visualized by using a Leica SP5 confocal laser scanning microscope. The fluorescence of both probes was excited using an argon laser tuned to 488 nm. After excitation, emitted fluorescence was separated by using a triple dichroic filter 488/561/633 and measured at 505-570 nm for green fluorescence (calcein) and 630-750 nm for red fluorescence (EthD-1). Collected images (n ≥ 5 per condition) were analyzed using the ImageJ software to quantify the area occupied by those positively stained for each marker with respect to the total image area.

Viability and internalization studies by flow cytometry
First, flow cytometry studies were carried out to analyze cell viability and the effect of SPIONs internalization in cell size and complexity in ENPC suspensions. Conditions investigated included: (1) cells without SPIONs (control), (2) cells exposed to NFA SPIONs at 4 different concentrations (0.001, 0.01, 0.025 and 0.05 mg Fe mL -1 ), and (3) cells exposed to NFD SPIONs at 4 different concentrations (0.001, 0.01, 0.025 and 0.05 mg S9 Fe mL -1 ). Exposure times were 1, 2, 4 and 24 h in a sterile incubator at 37 ºC under a CO 2 atmosphere (5 %). Briefly, ENPC suspensions right after isolation were exposed to the different treatments in 15 mL Falcon® tubes (5 mL). After the corresponding incubation times, cells were centrifuged at 300 g for 4 min, the supernatants were discarded, and the pellets were suspended in fresh culture media (1 mL). Subsequently, samples were filtered using 12 x 75 mm tubes with a 35 µm nylon cell strainer cap (Falcon) in order to remove cell aggregates and centrifuged again at 300 g at 4 ºC for 4 min. Supernatants were discarded and cellular pellets suspended in 100 µL of ice-cold Annexin V Binding Buffer 1x for staining with Annexin V-FITC and 7-amino-actinomycin D (7-AAD) following the manufacturer instructions (Beckman Coulter Life Sciences). Samples were analyzed on a FACS Canto II cytometer (BD Biosciences) within 30 min after staining and recorded for 2 min with at least 10,000 events recorded in the FSC gate. Flow cytometry data analysis was carried out with the FlowJo 10.7 software (BD Biosciences) following the gating strategy depicted in Figure S1. In detail, after cellular aggregates exclusion (A), FSC gate was plotted on an FSC vs SSC dot-plot to remove cellular debris (B). Then, FSC-gated events were analyzed for Annexin V and 7-AAD staining (C). Three subpopulations were identified: (1) Early apoptosis, with events Annexin V + /7-AAD -; (2) Late apoptosis, with events Annexin V + /7-AAD + ; and (3) Live cells which were double negative cells (Annexin V -/7-AAD -). The subpopulation characterized by low FSC and high SSC scattering was further gated into FSC gate as FSC low region (D). Subsequently, Annexin V/7-AAD staining was analyzed in the FSC low subpopulation in the same way as the FSC-gated population (E). Finally, changes induced by SPIONs treatment in the SSC intensity of FSC low -gated events were monitored by plotting them in a histogram (F). The SSC high interval was defined by using the corresponding control samples without nanoparticles.

S10
In order to unravel the cellular route for SPION uptake in this cell type, ENPC suspensions were pre-incubated for 2h in a sterile incubator at 37 ºC under a CO 2 atmosphere (5 %) with the following inhibitors covering the most common entry pathways: (i) chlorpromazine to affect clathrin-dependent endocytosis (5 µg mL -1 ), (ii) amiloride to block macropinocytosis, specifically affecting the sodium-proton exchange (0.5 µg mL -1 ), (iii) cytochalasin D to influence macropinocytosis by acting on actin polymerization (2 µM), (iv) genistein to affect clathrin-independent endocytosis (5 µM), and (v) wortmannin to inhibit PIK3 and other kinases during phagocytosis (5 µg mL -1 ). After inhibitors preincubation, cell suspensions were then exposed to 0.05 mg mL -1 of either NFA or NFD SPIONs for 2 h. Cells without either blocking agents or SPIONs, cells incubated with the different inhibitors in the absence of SPIONs and cells only exposed to the two types of SPIONs (without inhibitors) served as control samples. After corresponding incubations, cell suspensions were prepared for flow cytometry analyses as described above. To corroborate that the incubation conditions were harmless for these primary neural cells, duplicate samples of control cells without either blocking agents or SPIONs and cells exposed to NFA and NFD SPIONs without inhibitors were seeded into PLL-coated petri dishes right after this 4h-treatment. Cell viability and neural differentiation was analyzed after 7 DIV by confocal fluorescence microscopy as described in other sections.

Neural differentiation by confocal fluorescence microscopy
An immune-labeling procedure was used to investigate the impact of SPIONs exposure on neural cell differentiation at different time points in culture. Briefly, cells were fixed with 4 % PFA in PBS for 10 min at room temperature, rinsed with PBS, and then permeabilized with saponin (0.25 % in PBS + 10 % of fetal bovine serum) for 10 min. The S11 following primary antibodies were selected: MAP-2 and β-III tubulin for labeling neurons, vimentin for labeling non-neuronal cells including glia and glial fibrillary acidic protein (GFAP) for specific targeting of astrocytes. Additionally, synaptophysin, the most abundant protein in the membrane of synaptic vesicles, was used to visualize synapses in neurons.

Lipidome studies
ENPC cultures on petri dishes (3.5 cm in diameter) were exposed to the different conditions and then trypsinized. Cell suspensions were centrifuged at 19,900 rpm at 4 ºC S12 for 15 min. NaCl (0.88 %, 100μL, at 4 ºC), EquiSPLASH internal standard (10 μL) and chloroform/methanol (2:1; 500 μL) were added to the pellets. Suspensions were then incubated at -40 ºC for 15 min and sonicated 30 s at 80 % with a UP50H homogenizer (5 times) a -40ºC. These last two steps were repeated three times. Samples were centrifuged at 14,000 rpm for 15 min at 4 ºC. The top phase (aqueous mixture and methanol) was MSMSALL acquisition mode was controlled by the Analyst software (Sciex). All precursors were selected in Q1, with an isolation window of 1Da in the MS interval of 200 S13 to 1200 umas, where the precursor ions were equally distributed in the isolation windows.
Colision energy for lipids in positive mode was CE 25 ± 15 V and CE -40 ± 15 V for those in negative mode. Data analysis was carried out by using the LipidView v1.3 software.
Lipid classes and species were identified based on m/z exacta and fragmentation patterns.

Membrane permeability studies by flow cytometry
Flow cytometry studies were also carried out to examine the effect of SPIONs on cell membrane permeability. For this purpose, we employed two fluorescent dyes: FM TM 1-43 and Laurdan. Conditions investigated included: (1) cells without SPIONs and without dyes (control), (2) cells exposed to NFA SPIONs without dye, (3) cells exposed to NFD SPIONs without dye, (4) cells loaded with FM1-43, (5) cells exposed to NFA SPIONs and loaded with FM1-43, (6) cells exposed to NFD SPIONs and loaded with FM1-43, (7) cells loaded with Laurdan probe, (8) cells exposed to NFA SPIONs and loaded with Laurdan and (9) cells exposed to NFD SPIONs and loaded with Laurdan. Briefly, right after isolation, ENPC suspensions were exposed to NFA and NFD SPIONs, as appropriate, at a concentration of 0.05 mg mL -1 for 2 h in a sterile incubator at 37 ºC under a CO 2 atmosphere (5%). After treatment, cells were centrifuged at 300 g for 4 min, the supernatants removed, and the pellets suspended in 1 mL of either Hank's without calcium and magnesium (FM1-43) or Dulbecco's Modified Eagle Medium (DMEM; Laurdan). For FM1-43 staining, cells were filtered using 12 x 75 mm tubes with a 35 µm nylon cell strainer Cap (Falcon) to remove cell aggregates and centrifuged at 300 g for 4 min at 4 ºC.
Next, supernatants were discarded, and pellets were suspended in 1 mL of Hank's without sodium and magnesium. Subsequently, 2.5 µl of FM1-43 (2.5 µg mL -1 ) were added to the samples and incubated for 1 min. Labelled samples were then analyzed by using flow S14 cytometry. FM1-43 fluorescence was excited with a 488 nm laser and the emission fluorescence was collected with a V585/42 detector in the 564-606 nm. For Laurdan staining, cell pellets were incubated with 5 µl of Laurdan (5 µM) for 1 h at 37 ºC. Then, cells were filtered using the same tubes as before and centrifuged at 300 g for 4 min at 4 ºC. The gating strategy followed for FM1-43 studies is illustrated in Figure S2. In detail, after cell aggregates exclusion, singlets gate was plotted on an FSC vs SSC dot-plot and FSC gate was created to exclude cellular debris. Next, FSC-gated events, plotted on a was measured for all three subpopulations and two different regions were defined: a FM1-S15 43 + region (i.e. higher levels of FM1-43 fluorescence) and a FM1-43 low region (i.e. events with a lower fluorescence for FM1-43).
The gating strategy followed for Laurdan studies is illustrated in Figure S3. In detail, after cell aggregates exclusion, singlet events were plotted on an FSC vs SSC dotplot and FSC gate was created to exclude cellular debris. As samples treated with SPIONs Thermal Cycler (Bio-Rad). We also included a Melting curve from 60 °C to 95 °C (0.5 °C s -1 ) at the end of the program to verify the specificity of the PCR. Fluorescence was acquired during both the 60 °C and Melting steps. The assay specificity for all tested genes was confirmed by their unique Melting peaks with indicated mastermix for each amplicon.
In order to discard a potential contamination of reagents and/or primer-dimer artifacts, a S17 non-template control (NTC) was carried out using all the reagents except the sample. For all tested genes, NTC amplifications were always negative or delayed more than 5 cycles with respect to the experimental samples, which allowed ruling out contamination and primer-dimer artifacts. Technical triplicates were performed in order to correct pipetting errors in plate loading. In general, accurate technical repeats were obtained. In order to test for the presence of gDNA in RNA samples, we performed the Valid Prime assay, which measures the gDNA contribution to the qPCR signal using an optimized gDNA-specific assay (VPA) and a gDNA reference sample in this case. The VPA, targeting a nontranscribed locus that is present in exactly one copy per haploid normal genome, is used to measure gDNA contents in RT samples and the gDNA reference is used to normalize for GOI-specific differences in gDNA sensitivity. These measures provide a reliable correction for gDNA background in qPCR. Correction is possible for any GOI assay that consistently amplifies gDNA, given that the DNA contribution does not exceed 60 % of the signal. In this work, no correction is necessary because VAP gene has no signal except for gDNA sample. Data processing was carried out using the software GenEx v. 5.4.4 (MultiD Analyses AB, Gothenburg, Sweden), performing the subsequent steps: (1) Efficiency correction; (2) Average technical qPCR replicates, (3) Normalization with selected reference gene, and (4) Relative quantification 2 -Δ(ΔCq) , 45 being ΔCq = Cq value of each individual sample against the Cq value of calibrator group (first biological replicate of the NFA SPION group). An appropriate normalization strategy is essential to correct the experimental variability (e.g. integrity differences, pipetting errors). In this study, the putative reference gene analyzed was 18S.

S18
Results were expressed in conventional bar graphs as the mean ± standard error of the mean (SEM), unless otherwise indicated, of at least three independent experiments for each assay (N ≥ 3). Statistical analysis was performed by using the IBM SPSS Statistics software (version 28.0.1.0). Comparisons among groups were done by one-way analysis of variance (ANOVA) followed by either post-hoc Scheffé, Tukey HSD or Games-Howell tests (homogeneous vs. heterogeneous variances as dictated by Levene's test). Comparisons between two groups, when needed, were carried out by T test. In all cases, the significance level was defined as p < 0.05. Figure S1. Scheme illustrating the gating and analysis strategies followed in flow cytometry studies. S20 Figure S2. Scheme illustrating the gating and analysis strategies followed in flow cytometry studies with the FM TM 1-43 probe.   Energy-dispersive X-ray (EDX) profiles confirming the elemental composition of both SPIONs, discarding the presence of eventual contaminants. Analyses were conducted at 2 kV. (C) Infrared spectra for NFA (red) and NFD (green) samples: O-H stretching vibrations due to water (3400 cm -1 ), C-H doublet symmetric and asymmetric stretching (2940 and 2882 cm -1 ) and carboxylic groups (1632 and 1384 cm -1 ) due to the citric acid coating and, finally, C-O-C vibrations (1112 and 1044 cm -1 ) and primary alcohols (860 and 920 cm -1 ) assigned to some polyol rests from the synthesis.         FSC histogram overlays for the SSC high cell population in ENPC suspensions exposed to different internalization inhibitors. Cells were exposed to SPIONs at 0.05 mg Fe mL -1 for 2h.