By Els Parton, Randy De Palma, Gustaff Borghs; IMEC, Leuven, Belgium
Magnetic nanoparticles (MNPs) can be used in a wide variety of biomedical applications [1, 2], ranging from contrast agents for magnetic resonance imaging to the deterioration of cancer cells via hyperthermia treatment (see figure). Most of these promising applications require well-defined and controllable interactions between the MNPs and living cells. A novel and straight forward coating method was recently developed, enabling stable particle functionalization. Furthermore, this functionalization method allows the introduction of a tunable coating that can be specifically adapted according to the application envisioned, thereby ensuring more controllable cellular interactions, and paving the way to a bright future in medicine.
Biomedical applications
MRI. A well-known application in the field of diagnosis is the use of MNPs as contrast agents for magnetic resonance imaging (MRI), which is used to better differentiate healthy and pathological tissues and to visualize various biological events inside the body. Due to their low toxicity, iron oxide MNPs have received US Food and Drug Administration approval to be used as MRI signal enhancers.
 Figure 1. Biomedical applications based on the controlled interactions between living cells and biologically activated magnetic nanoparticles. |
Magnetic labeling. Another diagnostic application is the magnetic labeling of all kinds of biological entities, such as cells, DNA, and proteins. An interesting application is the labeling of stem cells to noninvasively monitor the distribution and fate of transplanted stem cells in the human body. Furthermore, MNPs show great promise as magnetic labels in biosensing with many advantages compared to conventional labels such as enzymes, fluorescent dyes, chemiluminescent molecules, and radioisotopes. For example, magnetically labeled cancer cells can be purified, transported, and detected on a single chip surface, enabling simple and cost-effective cancer screening in a lab-on-a-chip approach.
Controlled drug release. Aside from their small size and low toxicity to humans, MNPs can be transported through an external magnetic field gradient, penetrating deep into the human tissue. In this way, controlled transport of drugs to target sites can be achieved. The latter usage is realized by attaching a drug to a biocompatible MNP carrier, injecting the ferrofluid into the bloodstream, and applying an external magnetic field to concentrate the drug/carrier complexes at the target site. As one example, this principle is used with cytotoxic drugs in cancer treatments.
Hyperthermia. Another interesting therapy is based on the ability of MNPs to be heated when a time-varying magnetic field is applied. This characteristic is used to burn away cancer cells (hyperthermia), often in combination with chemotherapy. It is in fact known that cancer cells are more sensitive to temperatures in excess of 41°C than their normal counterparts. Both applications present a bright future for targeted therapy, which can specifically destroy a desired target without deteriorating healthy surrounding tissue.
Cell isolation. Finally, the attraction between an external magnet and the MNPs enables separation of a wide variety of biological entities. Examples are the isolation of cancer cells in blood samples or stem cells in bone marrow to allow for improved diagnosis and the removal of toxins from the human body. Furthermore, MNPs can be biologically activated to allow the uptake of cells via endocytotic pathways, thereby allowing certain cellular compartments to be specifically addressed. Once taken up, the desired cellular compartments can be magnetically isolated and accurately studied using proteomic analysis.
Challenges
There are two main challenges to make all the above-discussed biomedical applications come true: 1) a good synthesis route for manufacturing monodisperse MNPs with diameters <20nm; and 2) a good method to functionalize the surface of the nanoparticles [2]. The latter determines the ability of the MNPs to interact in a well-defined and controllable manner with living cells. Such an interaction is mainly achieved by coating the nanoparticles with biological ligands specific for certain receptors on the cell surface (i.e., receptor-mediated interaction). However, in some cases, a chemical functionality can also be "attractive" for a cell surface (i.e., nonspecific interaction). Once bound to the cell surface, the nanoparticle can stay there or a mechanism of cellular uptake can be triggered by which the nanoparticle is engulfed through the cellular membrane and brought into the cell body ((see figure).
MNP production
 Figure 2. Possible interactions between (bio-active) magnetic nanoparticles and living cells. |
Magnetic nanoparticles for biomedical applications have to be uniform in size and monodisperse so that each nanoparticle has nearly identical physical and chemical properties [2]. Most synthesis routes for monodisperse MNPs are based on the general principle of a short nucleation step, followed by a slower growth process on the existing nuclei. A major difficulty encountered during the synthesis of these MNPs is to keep them stable in solution without showing any sign of nanoparticle aggregation. As for all nanoparticles, aggregation is also commonly observed for MNPs due to their extremely large surface-to-volume ratio and the large surface energy they express. Furthermore, they could magnetically interact with each other when not properly stabilized.
To circumvent the aggregation problem, a strong repulsive force must be created to counteract the magnetic and surface-related attractions. Such a repulsive force can be achieved by electrostatic or by steric repulsion [2]. The first method uses ionic compounds to coat the particles; the second approach—which offers a more efficient stabilization—coats the particles with large molecules such as polymers, or surfactants containing long-chain hydrocarbons.
There are many protocols for the synthesis of MNPs based on micro-emulsion, co-precipitation, and other water-based methods [1, 3]. The disadvantages of these water-based methods are that the size uniformity and crystallinity of the MNPs are rather poor, and nanoparticle aggregation is commonly observed. Recently, Sun et al. developed a new and simple synthetic procedure [2–4] that enables the synthesis of very monodisperse and highly crystalline MNPs with sizes between 3nm and 20nm without showing any sign of nanoparticle aggregation.
A typical process involves the high-temperature decomposition (>220°C) of an organic iron precursor in the presence of hydrophobic ligands such as oleic acid [2–3]. These hydrophobic ligands form a dense coating around the nanoparticles, thereby avoiding their aggregation. This method has been further adapted by other researchers to synthesize all types of MNPs containing different materials such as cobalt, manganese, nickel, platinum, etc. [3]. Although this thermal-decomposition method has the advantage of producing very monodisperse and highly crystalline particles, a major disadvantage is that the resulting nanoparticles are soluble only in nonpolar solvents due to their coating with hydrophobic ligands. Hence, to make MNPs suitable for biological applications, the hydrophobic ligand coating needs to be replaced by a hydrophilic, biocompatible, and functional coating that allows controlled interaction with different types of biological species, such as cells, proteins, or DNA.
Coating the MNPs
Ligand exchange can be used to replace the hydrophobic coating of the above-mentioned MNPs with hydrophilic molecules. The ligand exchange method involves adding an excess of ligand to the nanoparticle solution to displace the original ligand on the nanoparticles’ surface. In this way, research groups have realized ferrite MNPs covered with hydrophilic ligands containing carboxylate, phosphate, and alcohol end groups, making the particles water-soluble. However, the long-term stability of these water-soluble nanoparticles has not been unambiguously proven due to the weak (noncovalent) binding of the ligands to the MNPs.
Based on its expertise in silane self-assembled monolayers for biosensor applications [5], IMEC recently worked out a procedure to coat MNPs (via the thermal-decomposition method) with silane monolayers [6]. The advantage of this approach is that it enables stable, water-soluble monodisperse MNPs bearing a huge variety of functional end groups. This capability makes it possible to tune the surface functionality of the nanoparticles, optimizing it for every specific application.
A systematic study of silane ligand exchange, screening nine commercially available silane monolayers, showed that the ligand exchange is very effective [6]. The original hydrophobic ligand was fully replaced by the silane self-assembled monolayer, forming a dense organic layer with the functional end groups presented to the surrounding liquid.
Furthermore, it was shown that the presence of these end groups strongly determines the water dispersibility of the nanoparticles. Out of these nine different silanes, only the amino, carboxylic acid, and poly(ethylene glycol) end functions were found to render the nanoparticles perfectly soluble in aqueous solutions over a wide pH range (see figure).
 Figure 3. Conversion to water-soluble magnetic nanoparticles using silane ligand exchange with amino, carboxylic acid, and poly(ethylene glycol) end functions. |
Enhanced long-term stability and increased resistance against mild acid and alkaline environments were observed compared to those of other ligands commonly used to stabilize MNPs. These characteristics are due to the strong covalent linkage of the silane layers onto the nanoparticles’ surface. Combined with Sun’s method for synthesizing MNPs, the functionalization procedure described above results in monodisperse, water-soluble (thus biocompatible) MNPs with properties that can easily be tuned to the requirements of the specific application.
Applications for living cells
Biomedical applications such as magnetic resonance imaging (contrast enhancement) and hyperthermia of cancer cells require that the MNPs interact with the body’s cells. Since nanoparticles have a much higher surface area per unit volume than larger particles, they therefore possess a much greater interface with their surroundings, and the interactions with their surroundings very much depend on the coating of their surface. A test was done to see if nanoparticles, realized according to Sun’s method and coated using IMEC’s new procedure, were “attractive” to cells. This test was done in close collaboration with the U. Hospital of Leuven, Belgium.
Three different types of silane layers bearing different charges were used as a coating, i.e., NH2, COOH, and polyethylene glycol (PEG), being respectively positive, negative, and neutral. In all three cases, the MNPs showed an excellent solubility when used in a cell medium such as DMEM-F12, typically used to keep the cells in a healthy condition. The influence of fetal calf serum (FCS) in the cell medium on the cellular interactions was also tested.
The interaction of the MNPs with living cells was evaluated based on the success rate of magnetic labeling and magnet-based separation. The success rate was very high—between 70% and 80% of the cells in the medium were magnetically labeled. The figure shows the results in greater detail: 1) When FCS was present in the cell medium, only the NH2-coated nanoparticles showed strong cellular interactions. 2) When no FCS was present, only the COOH-and PEG-coated nanoparticles allowed for magnetic labeling. To explain these results, we first investigated the binding of FCS onto the nanoparticle surface. NH2 nanoparticles showed a strong binding with the serum proteins, while PEG and COOH were found to repel these serum proteins from their surface.
 Figure 4. Influence of the nanoparticles’ surface function and presence of fetal calf serum on the nonspecific interactions with living cells. |
This clearly shows that the cells don’t show a strong affinity toward the NH2 nanoparticles except when they are coated with FCS proteins. Conversely, the cells show a strong affinity toward the COOH and PEG nanoparticles. However, when FCS is present in solution, it will compete with the nanoparticles and inhibit the interactions with the COOH and PEG nanoparticles. In none of the cases did the nanoparticles seem to be toxic to the living cells, and cell viabilities >70% were observed.
The results described above clearly indicate that the interactions between living cells and nanoparticles can be easily controlled by varying the surface functions of the nanoparticle. Furthermore, the presence of serum proteins can trigger or inhibit the nonspecific interactions, depending on the nanoparticles’ surface functions.
Conclusion
The combination of Sun’s thermal-decomposition method to manufacture high-quality MNPs and IMEC’s procedure to coat the nanoparticles via silane ligand exchange has proven to be a success. The method represents a generic and versatile method to synthesize highly stable, water-soluble ferrite MNPs with a variable ligand periphery and tunable surface properties. Furthermore, we have shown that the nonspecific interactions between nanoparticles and living cells are indeed dominated by the functional end groups at the nanoparticles’ surface.
Using the experimental techniques described, we could not distinguish between binding to the cell surface or cellular uptake. Future investigations will examine the promotion of cellular uptake to allow for endosomal proteomics that can, for example, contribute to a more fundamental understanding of the biochemistry behind Alzheimer’s disease. Furthermore, we will investigate the covalent grafting of biological ligands onto the nanoparticles’ surface for the targeted diagnosis and therapy of cancers, based on MRI and hyperthermia, respectively. Since a wide variety of silane surface functionalities are cells was evaluated based on the success rate commercially available, one can easily screen for a specific application, ranging from cell isolation, to medical imaging, to hyperthermia for cancer treatment.
Acknowledgments
The authors would like to thank the other members of the NEXT-NS and NEXT-BE group for their invaluable scientific support.
References
1. Q.A. Pankhurst, J. Connolly, S.K. Jones, J. Dobson, “Applications of Magnetic Nanoparticles in Biomedicine,” Journal of Physics D: Applied Physics, 36, pp. R167–R181, 2003.
2. C. Xu, S. Sun, “Monodisperse Magnetic Nanoparticles for Biomedical Applications,” Polymer International, 56, pp. 821–826, 2007
3. T. Hyeon, “Chemical Synthesis of Magnetic Nanoparticles,” Chemical Communication, pp. 927–934, 2007.
4. S. Sun, H. Zeng, D.B. Robinson, S. Raoux, P.M. Rice, S.X. Wang, G. Li, “Monodisperse MFe2O4 (M= Fe, Co, Mn) Nanoparticles,” Journal of the American Chemical Society, 126, pp. 273–279, 2004.
5. R. De Palma, W. Laureyn, F. Frederix, K. Bonroy, J.-J. Pireaux, G. Borghs, G. Maes, “Formation of Dense Self-assembled Monolayers of (nDecyl) tricholorosilanes on Ta/Ta2O5,” Langmuir, 23, pp. 443–451, 2007.
6. R. De Palma, S. Peeters, M.J. Van Bael, H. Van den Rul, K. Bonroy, W. Laureyn, et al., “Silane Ligand Exchange to Make Hydrophobic Super-paramagnetic Nanoparticles Water-dispersible,” Chemistry of Materials, 19, pp. 1821–1831, 2007.
Els Parton is a scientific editor at IMEC. She can be reached at [email protected]. Randy De Palma is working toward a PhD at IMEC, and Gustaaf Borghs is an IMEC research fellow.