Magnetically triggered clustering of biotinylated iron oxide nanoparticles in the presence of streptavidinylated enzymes.

This work deals with the production and characterization of water-compatible, iron oxide based nanoparticles covered with functional poly(ethylene glycol) (PEG)-biotin surface groups (SPIO-PEG-biotin). Synthesis of the functionalized colloids occurred by incubating the oleate coated particles used as precursor magnetic fluid with anionic liposomes containing 14 mol% of a phospholipid-PEG-biotin conjugate. The latter was prepared by coupling dimyristoylphosphatidylethanolamine (DC(14:0)PE) to activated α-biotinylamido-ω -N-hydroxy-succinimidcarbonyl-PEG (NHS-PEG-biotin). Physical characterization of the oleate and PEG-biotin iron oxide nanocolloids revealed that they appear as colloidal stable clusters with a hydrodynamic diameter of 160 nm and zeta potentials of - 39 mV (oleate coated particles) and - 14 mV (PEG-biotin covered particles), respectively, as measured by light scattering techniques. Superconducting quantum interference device (SQUID) measurements revealed specific saturation magnetizations of 62-73 emu g(-1) Fe(3)O(4) and no hysteresis was observed at 300 K. MR relaxometry at 3 T revealed very high r(2) relaxivities and moderately high r(1) values. Thus, both nanocolloids can be classified as small, superparamagnetic, negative MR contrast agents. The capacity to functionalize the particles was illustrated by binding streptavidin alkaline phosphatase (SAP). It was found, however, that these complexes become highly aggregated after capturing them on the magnetic filter device during high-gradient magnetophoresis, thereby reducing the accessibility of the SAP.

Intracellular nanoparticle coating stability determines nanoparticle diagnostics efficacy and cell functionality.

Iron oxide nanoparticles (NPs) are frequently employed in biomedical research as magnetic resonance (MR) contrast agents where high intracellular levels are required to clearly depict signal alterations. To date, the toxicity and applicability of these particles have not been completely unraveled. Here, we show that endosomal localization of different iron oxide particles results in their degradation and in reduced MR contrast, the rate of which is governed mainly by the stability of the coating. The release of ferric iron generates reactive species, which greatly affect cell functionality. Lipid-coated NPs display the highest stability and furthermore exhibit intracellular clustering, which significantly enhances their MR properties and intracellular persistence. These findings are of considerable importance because, depending on the nature of the coating, particles can be rapidly degraded, thus completely annihilating their MR contrast to levels not detectable when compared to controls and greatly impeding cell functionality, thereby hindering their application in functional in vivo studies.

High intracellular iron oxide nanoparticle concentrations affect cellular cytoskeleton and focal adhesion kinase-mediated signaling.

Iron oxide nanoparticle internalization exerts detrimental effects on cell physiology for a variety of particles, but little is known about the mechanism involved. The effects of high intracellular levels of four types of iron oxide particles (Resovist, Endorem, very small organic particles, and magnetoliposomes (MLs)) on the viability and physiology of murine C17.2 neural progenitor cells and human blood outgrowth endothelial cells are reported. The particles diminish cellular proliferation and affect the actin cytoskeleton and microtubule network architectures as well as focal adhesion formation and maturation. The extent of the effects correlates with the intracellular concentration (= iron mass) of the particles, with the biggest effects for Resovist and MLs at the highest concentration (1000 microg Fe mL(-1)). Similarly, the expression of focal adhesion kinase (FAK) and the amount of activated kinase (pY397-FAK) are affected. The data suggest that high levels of perinuclear localized iron oxide nanoparticles diminish the efficiency of protein expression and sterically hinder the mature actin fibers, and could have detrimental effects on cell migration and differentiation.

Stable long-term intracellular labelling with fluorescently tagged cationic magnetoliposomes.

Iron oxide nanocrystals that are dextran coated are widely exploited biomedically for magnetic resonance imaging (MRI), hyperthermia cancer treatment and drug or gene delivery. In this study, the use of an alternative coating consisting of a phospholipid bilayer directly attached to the magnetite core is described. The flexible nature of the magnetoliposome (ML) coat, together with the simple production procedure, allows rapid and easy modification of the coating, offering many exciting possibilities for the use of these particles in biomedical applications. Upon incubation of neutral MLs with an equimolar amount of cationic 1,2-distearoyl-3-trimethylammoniumpropane (DSTAP)-bearing vesicles, approximately one third of the cationic lipids are incorporated into the ML coat. This is in line with a theoretical model predicting transferability of only the outer leaflet phospholipids of bilayer structures. Most interestingly, the use of MLs containing 3.33 % DSTAP with a positive zeta-potential of (31.3+/-7.3) mV (mean +/-SD) at neutral pH, results in very heavy labelling of a variety of biological cells (up to (70.39+/-4.52) pg of Fe per cell, depending on the cell type) without cytotoxic effects. The results suggest the general applicability of these bionanocolloids for cell labelling. Mechanistically, the nanoparticles are primarily taken up by clathrin-mediated endocytosis and follow the endosomal pathway. The fate of the ML coat after internalisation has been studied with different fluorescent lipid conjugates, which because of the unique features of the ML coat can be differentially incorporated in either the inner or the outer layer of the ML bilayer. It is shown that, ultimately, iron oxide cores surrounded by an intact lipid bilayer appear in endosomal structures. Once internalised, MLs are not actively exocytosed and remain within the cell. The lack of exocytosis and the very high initial loading of the cells by MLs result in a highly persistent label, which can be detected, even in highly proliferative 3T3 fibroblasts, for up to at least one month (equivalent to approximately 30 cell doublings), which by far exceeds any values reported in the literature.

Targeting tumor cells and neovascularization using RGD-functionalized magnetoliposomes.

PURPOSE: Magnetoliposomes (MLs) have shown great potential as magnetic resonance imaging contrast agents and as delivery vehicles for cancer therapy. Targeting the MLs towards the tumor cells or neovascularization could ensure delivery of drugs at the tumor site. In this study, we evaluated the potential of MLs targeting the αvß3 integrin overexpressed on tumor neovascularization and different tumor cell types, including glioma and ovarian cancer. METHODS: MLs functionalized with a Texas Red fluorophore (anionic MLs), and with the fluorophore and the cyclic Arginine-Glycine-Aspartate (cRGD; cRGD-MLs) targeting the αvß3 integrin, were produced in-house. Swiss nude mice were subcutaneously injected with 107 human ovarian cancer SKOV-3 cells. Tumors were allowed to grow for 3 weeks before injection of anionic or cRGD-MLs. Biodistribution of MLs was followed up with a 7T preclinical magnetic resonance imaging (MRI) scanner and fluorescence imaging (FLI) right after injection, 2h, 4h, 24h and 48h post injection. Ex vivo intratumoral ML uptake was confirmed using FLI, electron paramagnetic resonance spectroscopy (EPR) and histology at different time points post injection. RESULTS: In vivo, we visualized a higher uptake of cRGD-MLs in SKOV-3 xenografts compared to control, anionic MLs with both MRI and FLI. Highest ML uptake was seen after 4h using MRI, but only after 24h using FLI indicating the lower sensitivity of this technique. Furthermore, ex vivo EPR and FLI confirmed the highest tumoral ML uptake at 4 h. Last, a Perl's stain supported the presence of our iron-based particles in SKOV-3 xenografts. CONCLUSION: Uptake of cRGD-MLs can be visualized using both MRI and FLI, even though the latter was less sensitive due to lower depth penetration. Furthermore, our results indicate that cRGD-MLs can be used to target SKOV-3 xenograft in Swiss nude mice. Therefore, the further development of this particles into theranostics would be of interest.

Synthesis, physicochemical characterization and MR relaxometry of aqueous ferrofluids.

The synthesis and characterization of ferrofluid based MR contrast agents, which offer R2* versatility beyond that of ferucarbotran, is described. Ferrofluids were formed after stabilizing magnetite cores with dodecanoic acid (a), oleic acid (b), dodecylamine (c), citric acid (d) or tartaric acid (e). Core sizes were deduced from TEM micrographs. Magnetic properties were determined by SQUID magnetometry. Hydrodynamic particle diameters were determined by dynamic light scattering measurements. Zeta potentials were measured by combining laser Doppler velocimetry and phase analysis light scattering. Iron contents were evaluated colorimetrically. MR relaxometry including R1 and R2* was conducted in vitro using homogeneous ferrofluid samples. The average core diameters of ferrofluids a, b and c equaled 9.4 +/- 2.8 nm and approximately 2 nm for ferrofluids d and e. Magnetization measurements at 300 K revealed superparamagnetic behaviour for the dried 9 nm diameter cores and paramagnetic-like behaviour for the dried cores of ferrofluids d and e. Iron contents were between 32-75 mg Fe/mL, reflecting the ferrofluids' high particle concentrations. Hydrodynamic particle diameters equaled 100-120 nm (a, b and c). For the ferrofluids a, b, d and e coated with anions, strong negative zeta potential values between -27.5 mV and -54.0 mV were determined and a positive zeta potential value of +33.5 mV was found for ferrofluid c, covered with cationic dodecylammonium ions. MR relaxometry yielded R1-values of 1.9 +/- 0.3 (a), 4.0 +/- 0.8 (b), 5.2 +/- 1.0 (c), 0.124 +/- 0.002 (d) and 0.092 +/- 0.005 s(-1) mM(-1) (e), and R2*-values of 856 +/- 24 (a), 729 +/- 16 (b), 922 +/- 29 (c), 1.7 +/- 0.05 (d) and 0.49 +/- 0.05 s(-1) mM(-1) (e). Thus, the synthesized ferrofluids reveal a broad spectrum of R2* relaxivities. As a result, the various MR contrast agents have a great potential to be used in studies dealing with malignant tissue targeting or molecular imaging.

Adsorption of antiphospholipid antibodies on affinity magnetoliposomes.

Phosphatidylcholine-based magnetoliposomes containing specific ligands for biological molecules, so-called affinity magnetoliposomes (AML), may prove to be useful as adsorbents in applications such as diagnosis or anchoring and delivery of drugs at specific sites in the human body. In the present study, the performance of affinity magnetoliposomes to adsorb anticardiolipin antibodies (aCL) from a previously characterized pool of patients with autoimmune diseases is described. The magnetic vesicles were prepared by enrobing nanometer-sized colloidal magnetite particles with a phospholipid bilayer composed of dimyristoylphosphatidylcholine (DMPC) and the affinity lipid ligand cardiolipin (CL). Adsorption of antibodies onto the affinity magnetoliposomes assayed using a high-gradient magnetophoresis (HGM) system, in which the magnetoliposomes were first magnetically captured on stainless steel fibers, and which were subsequently overflowed either with a pool of sera from autoimmune patients or sera of healthy individuals as a control. The spectrophotometric assay showed stronger changes in absorbance spectra when the affinity magnetoliposomes containing cardiolipin were added to sera of autoimmune patients than when they were added to sera of healthy individuals. The breakthrough curves obtained from a frontal analyses of the adsorption in the magnetophoresis system showed a 10% difference for total adsorbed IgG when sera of autoimmune and healthy individuals were assayed on magnetoliposomes containing cardiolipin.

Improved Labeling of Pancreatic Islets Using Cationic Magnetoliposomes.

Pancreatic islets (PIs) transplantation is an alternative approach for the treatment of severe forms of type 1 diabetes (T1D). To monitor the success of transplantation, it is desirable to follow the location of engrafted PIs non-invasively. In vivo magnetic resonance imaging (MRI) of transplanted PIs is a feasible cell tracking method; however, this requires labeling with a suitable contrast agent prior to transplantation. We have tested the feasibility of cationic magnetoliposomes (MLs), compared to commercial contrast agents (Endorem and Resovist), by labeling insulinoma cells and freshly isolated rat PIs. It was possible to incorporate Magnetic Ressonance (MR)-detectable amounts of MLs in a shorter time (4 h) when compared to Endorem and Resovist. MLs did not show negative effects on the PIs' viability and functional parameters in vitro. Labeled islets were transplanted in the renal sub-capsular region of healthy mice. Hypointense contrast in MR images due to the labeled PIs was detected in vivo upon transplantation, while MR detection of PIs labeled with Endorem and Resovist was only possible after the addition of transfection agents. These findings indicate that MLs are suitable to image PIs, without affecting their function, which is promising for future longitudinal pre-clinical and clinical studies involving the assessment of PI transplantation.

Magnetoliposomes as Contrast Agents for Longitudinal in vivo Assessment of Transplanted Pancreatic Islets in a Diabetic Rat Model.

Magnetoliposomes (MLs) were synthesized and tested for longitudinal monitoring of transplanted pancreatic islets using magnetic resonance imaging (MRI) in rat models. The rat insulinoma cell line INS-1E and isolated pancreatic islets from outbred and inbred rats were used to optimize labeling conditions in vitro. Strong MRI contrast was generated by islets exposed to 50 µg Fe/ml for 24 hours without any increased cell death, loss of function or other signs of toxicity. In vivo experiments showed that pancreatic islets (50-1000 units) labeled with MLs were detectable for up to 6 weeks post-transplantation in the kidney subcapsular space. Islets were also monitored for two weeks following transplantation through the portal vein of the liver. Hereby, islets labeled with MLs and transplanted under the left kidney capsule were able to correct hyperglycemia and had stable MRI signals until nephrectomy. Interestingly, in vivo MRI of streptozotocin induced diabetic rats transplanted with allogeneic islets demonstrated loss of MRI contrast between 7-16 days, indicative of loss of islet structure. MLs used in this study were not only beneficial for monitoring the location of transplanted islets in vivo with high sensitivity but also reported on islet integrity and hereby indirectly on islet function and rejection.

Optimal conditions for labelling of 3T3 fibroblasts with magnetoliposomes without affecting cellular viability.

A comparative study that deals with the internalisation of different types of magnetoliposomes (MLs) by 3T3 fibroblasts revealed that cationic MLs proved to be superior to neutral and anionic ones. Internalisation was visualised both by optical light and transmission electron microscopy. The latter showed that the cationic MLs ultimately ended up in lysosomal structures. The effect of increasing 1,2-dioleoyl-3-trimethylammonium propane (DOTAP) concentrations in the cationic ML coat has been elucidated. High uptake efficiency was only achieved with MLs that carry a high DOTAP payload. However, these structures also demonstrated toxic effects. The use of the saturated distearoyl analogue (DSTAP) at identical concentrations led to improved uptake efficiency and lower toxicity. By using iron-oxide-free vesicles, it was shown that the toxicity was due to lipid bilayer constituents and not the iron oxide. In conclusion, the use of DMPC-DSTAP (96.67:3.33; molar ratio) MLs results in an extremely high labelling of 3T3 fibroblasts with iron oxides (47.66 pg Fe per cell) without evoking any influence on cell viability.

PEGylation of cholecystokinin prolongs its anorectic effect in rats.

The anorectic compound CCK-9 was coupled to polyethylene glycol 5 kDa, 10 kDa, 20 kDa and 30 kDa, under different reaction conditions. Conjugates were purified by HPLC and characterized by MALDI-TOF MS. A 96% PEGylation yield was obtained in buffer pH 7.5 after 6h reaction at 20 degrees C. The anorectic activity was tested in vivo in rats. A single bolus intra-peritoneal injection of non-modified CCK-9 resulted in a significant initial food intake reduction 30 min after food presentation (87% compared to paired control group). When PEG-CCK-9 conjugates modified with polymers of molecular weight up to 20 kDa were injected, lower but statistically significant initial food intake reductions were obtained (76% for PEG 10 kDa-CCK-9 conjugate compared to control group). The cumulative food intake reduction of non-modified CCK-9 is normalized within 1-2h, whereas the PEG-CCK-9 molecules showed a prolonged anorectic activity lasting for 6h for PEG 5 kDa-CCK-9; 23 h for PEG 10 kDa-CCK-9 and between 8h and 23 h for PEG 20 kDa-CCK-9. For PEG 30 kDa-CCK-9 conjugate, neither an initial nor a cumulative FI reduction was observed. PEG-CCK-9 conjugates show a significantly prolonged anorectic activity in comparison to the non-modified peptide. This effect is most evident for the PEG 10 kDa-CCK-9 conjugate.

Receptor-mediated biological responses are prolonged using hydrophobized ligands.

Hormone-receptor interactions occur following three-dimensional diffusion of the ligand to the membrane-embedded receptor. However, prior hydrophobization of the ligand might restrict its movement to two dimensions along the membrane surface, and the biological response might therefore be modulated. This idea was tested using the C-terminal nonapeptide, CCK9, of the satiating hormone, cholecystokinin (CCK). The hormone was lipidated by linking it covalently to distearoylphosphatidylethanolamine via a poly(ethylene glycol) (PEG) spacer. The desired conjugate was isolated by thin-layer chromatography and incorporated into preformed small unilamellar dimyristoylphosphatidylcholine (DMPC) vesicles. The hormone-bearing vesicles were injected intraperitoneally into Wistar rats and food intake monitored. Compared to the biological effect elicited by the same amount of soluble non-derivatized CCK9, food intake reduction showed a delayed onset, but lasted for a significantly longer time. We believe this prolonged effect was due to the transfer of the derivatized CCK9 from the vesicles to the natural membrane containing the hormone receptor. Ultimately, this event may result in sustained receptor occupation and, thus, food intake reduction. The underlying mechanism for the physiological effects observed may be of relevance in interpreting results obtained using artificial measuring devices; for example, the signal produced by biosensors may be drastically affected by the hydrophobicity of the ligand.

Biotinylated Stealth magnetoliposomes.

Dimyristoylphosphatidylethanolamine (DC(14:0)PE) and the dioleoyl analogue (DC(18:1cis)PE) were mixed with alpha-biotinylamido-omega-N-succinimidoxycarbonyl-poly(ethylene glycol) (NHS-PEG-biotin) and quantitatively converted to alpha-biotinylamido-omega-(dimyristoylphosphatidylethanolamino-carbonyl)polyethylene glycol (DC(14:0)PE-PEG-biotin) and the dioleoyl analogue DC(18:1cis)PE-PEG-biotin, respectively. As shown by thin-layer chromatography and 1H NMR spectroscopy, PEGylation of both phosphatidylethanolamine types went to completion if the reaction was performed in organic solvent in the presence of triethylamine. The resulting derivatives were successfully incorporated into both classical phospholipid vesicles and a phospholipid bilayer surrounding nanometer-sized magnetite cores. In the latter case, the so-called activated Stealth(1) magnetoliposomes were produced which very efficiently immobilized streptavidinylated alkaline phosphatase.

Attachment of water-soluble proteins to the surface of (magnetizable) phospholipid colloids via NeutrAvidin-derivatized phospholipids.

The present work describes the incorporation of a functionalized phospholipid derivative into the phospholipid bilayer of both classical small unilamellar vesicles and recently developed magnetoliposomes, resulting in unique biocolloid structures onto which peripheral water-soluble enzymes can be immobilized on the surfaces. In the first part of this work, a synthesis protocol is outlined for a universal membrane anchor for water-soluble proteins. Dioleoylphosphatidylethanolamine-N-dodecanyl was used as the starting lipid molecule. After activation of the terminal -COOH group, alpha,omega-diamino-poly(ethylene glycol), used as a hydrophilic, flexible spacer arm, was coupled covalently. Subsequently, NeutrAvidin was bound, after blocking the free -NH(2) groups with citraconic anhydride. In the second part, the resulting lipid-NeutrAvidin derivative was incorporated into small unilamellar vesicles comprised of dimyristoylphosphatidylglycerol. FPLC with Superdex 200 as the column matrix clearly showed that biotinylated alkaline phosphatase, which served as a representative model of water-soluble proteins, was attached to the vesicles. Furthermore, magnetoliposomes, constructed of the same type of phospholipid molecules, were presented as interesting colloids to assess the degree of enzyme immobilization in a rapid and elegant manner. Potential applications that can emerge from this study are briefly discussed.

In vivo hepatocyte MR imaging using lactose functionalized magnetoliposomes.

The aim of this study was to assess a novel lactose functionalized magnetoliposomes (MLs) as an MR contrast agent to target hepatocytes as well as to evaluate the targeting ability of MLs for in vivo applications. In the present work, 17 nm sized iron oxide cores functionalized with anionic MLs bearing lactose moieties were used for targeting the asialoglycoprotein receptor (ASGP-r), which is highly expressed in hepatocytes. Non-functionalized anionic MLs were tested as negative controls. The size distribution of lactose and anionic MLs was determined by transmission electron microscopy (TEM) and dynamic light scattering (DLS). After intravenous administration of both MLs, contrast enhancement in the liver was observed by magnetic resonance imaging (MRI). Label retention was monitored non-invasively by MRI and validated with Prussian blue staining and TEM for up to eight days post MLs administration. Although the MRI signal intensity did not show significant differences between functionalized and non-functionalized particles, iron-specific Prussian blue staining and TEM analysis confirmed the uptake of lactose MLs mainly in hepatocytes. In contrast, non-functionalized anionic MLs were mainly taken up by Kupffer and sinusoidal cells. Target specificity was further confirmed by high-resolution MR imaging of phantoms containing isolated hepatocytes, Kupffer cell (KCs) and hepatic stellate cells (HSCs) fractions. Hypointense signal was observed for hepatocytes isolated from animals which received lactose MLs but not from animals which received anionic MLs. These data demonstrate that galactose-functionalized MLs can be used as a hepatocyte targeting MR contrast agent to potentially aid in the diagnosis of hepatic diseases if the non-specific uptake by KCs is taken into account.

Variability in contrast agent uptake by different but similar stem cell types.

The need to track and evaluate the fate of transplanted cells is an important issue in regenerative medicine. In order to accomplish this, pre-labelling cells with magnetic resonance imaging (MRI) contrast agents is a well-established method. Uptake of MRI contrast agents by non-phagocytic stem cells, and factors such as cell homeostasis or the adverse effects of contrast agents on cell biology have been extensively studied, but in the context of nanoparticle (NP)-specific parameters. Here, we have studied three different types of NPs (Endorem®, magnetoliposomes [MLs], and citrate coated C-200) to label relatively larger, mesenchymal stem cells (MSCs) and, much smaller yet faster proliferating, multipotent adult progenitor cells (MAPCs). Both cell types are similar, as they are isolated from bone marrow and have substantial regenerative potential, which make them interesting candidates for comparative experiments. Using NPs with different surface coatings and sizes, we found that differences in the proliferative and morphological characteristics of the cells used in the study are mainly responsible for the fate of endocytosed iron, intracellular iron concentration, and cytotoxic responses. The quantitative analysis, using high-resolution electron microscopy images, demonstrated a strong relationship between cell volume/surface, uptake, and cytotoxicity. Interestingly, uptake and toxicity trends are reversed if intracellular concentrations, and not amounts, are considered. This indicates that more attention should be paid to cellular parameters such as cell size and proliferation rate in comparative cell-labeling studies.

Investigating the toxic effects of iron oxide nanoparticles.

The use of iron oxide nanoparticles (IONPs) in biomedical research is steadily increasing, leading to the rapid development of novel IONP types and an increased exposure of cultured cells to a wide variety of IONPs. Due to the large variation in incubation conditions, IONP characteristics, and cell types studied, it is still unclear whether IONPs are generally safe or should be used with caution. During the past years, several contradictory observations have been reported, which highlight the great need for a more thorough understanding of cell-IONP interactions. To improve our knowledge in this field, there is a great need for standardized protocols and toxicity assays, that would allow to directly compare the cytotoxic potential of any IONP type with previously screened particles. Here, several approaches are described that allow to rapidly but thoroughly address several parameters which are of great impact for IONP-induced toxicity. These assays focus on acute cytotoxicity, induction of reactive oxygen species, measuring the amount of cell-associated iron, assessing cell morphology, cell proliferation, cell functionality, and possible pH-induced or intracellular IONP degradation. Together, these assays may form the basis for any detailed study on IONP cytotoxicity.

How to assess cytotoxicity of (iron oxide-based) nanoparticles: a technical note using cationic magnetoliposomes.

The range of different types of nanoparticles and their biomedical applications is rapidly growing, creating a need to thoroughly examine the effects these particles have on biological entities. One of the most commonly used nanoparticle types is iron oxide nanoparticles, which can be used as MRI contrast agents. The main research topic is the in vitro labeling of cells with iron oxide nanoparticles to render the cells detectable for MRI upon in vivo transplantation. For the correct evaluation of cell function and behavior in vivo, any effects of the nanoparticles on the cells must be completely ruled out. The present work provides a technical note where a detailed overview is given of several assays that could be useful to determine nanoparticle toxicity. The assays described focus on (i) nanoparticle internalization, (ii) immediate cell toxicity, (iii) cell proliferation, (iv) cell morphology, (v) cell functionality and (vi) cell physiology. Potential pitfalls, appropriate controls and advantages/disadvantages of the different assays are given. The main focus of this work is to provide a detailed guide to help other researchers in the field interested in setting up nanoparticle-toxicity studies.

Cytotoxic effects of iron oxide nanoparticles and implications for safety in cell labelling.

The in vitro labelling of cultured cells with iron oxide nanoparticles (NPs) is a frequent practice in biomedical research. To date, the potential cytotoxicity of these particles remains an issue of debate. In the present study, 4 different NP types (dextran-coated Endorem, carboxydextran-coated Resovist, lipid-coated magnetoliposomes (MLs) and citrate-coated very small iron oxide particles (VSOP)) are tested on a variety of cell types, being C17.2 neural progenitor cells, PC12 rat pheochromocytoma cells and human blood outgrowth endothelial cells. Using different NP concentrations, the effect of the NPs on cell morphology, cytoskeleton, proliferation, reactive oxygen species, functionality, viability and cellular homeostasis is investigated. Through a systematic study, the safe concentrations for every particle type are determined, showing that MLs can lead up to 67.37 ± 5.98 pg Fe/cell whereas VSOP are the most toxic particles and only reach 18.65 ± 2.07 pg Fe/cell. Using these concentrations, it is shown that for MRI up to 500 cells/µl labelled with VSOP are required to efficiently visualize in an agar phantom in contrast to only 50 cells/µl for MLs and 200 cells/µl for Endorem and Resovist. These results highlight the importance of in-depth cytotoxic evaluation of cell labelling studies as at non-toxic concentrations, some particles appear to be less suitable for the MR visualization of labelled cells.

Background migration of USPIO/MLs is a major drawback for in situ labeling of endogenous neural progenitor cells.

MR-labeling of endogenous neural progenitor cells (NPCs) to follow up cellular migration with in vivo magnetic resonance imaging (MRI) is a very promising tool in the rapidly growing field of cellular imaging. To date, most of the in situ labeling work has been performed using micron-sized iron oxide particles. In this work magnetoliposomes (MLs), i.e. ultrasmall superparamagnetic iron oxide cores (USPIOs), each individually coated by a phospholipid bilayer, were used as the MR contrast agent. One of the main advantages of MLs is that the phospholipid bilayer allows easy modification of the surface, which creates the opportunity to construct a wide range of MLs optimized for specific biomedical applications. We have investigated the ability of MLs to label endogenous NPCs after direct injection into the adult mouse brain. Whereas MRI revealed contrast relocation towards the olfactory bulb, our data strongly imply that this relocation is independent of the migration of endogenous NPCs but represents background migration of MLs along a white matter tract. Our findings suggest that the small size of USPIOs/MLs intrinsically limits their potential for in situ labeling of NPCs.