DOI: 10.1002/chem.201304070
Communication
& Magnetic Resonance Imaging
Diblock-Copolymer-Mediated Self-Assembly of Protein-Stabilized Iron Oxide Nanoparticle Clusters for Magnetic Resonance Imaging Sari Thk,[b] Ari Laiho,[c] and Mauri A. Kostiainen*[a] netic nanoparticles and enzymes.[5–7] The biggest advantage of protein cages from the materials point of view is their well-defined size and shape, hollow inner cavity, water solubility, and reversible assembly/disassembly process from small protein subunits into the fully assembled cage.[8, 9] Furthermore, the outer and inner surfaces of protein cages can easily be functionalized with genetic or chemical methods, which allow the particles to be decorated with functional moieties in a controlled manner.[10–12] Ferritin protein cages are found in almost all living organisms, ranging from mammals to Archaea and bacteria.[13] In nature, ferritins are intracellular proteins that play a key role in the isolation and storage of iron.[14] In this work, we have utilized recombinant ferritin particles from the hyperthermophile Pyrococcus furiosus.[15, 16] This protein cage is composed of 24 structurally similar protein subunits that assemble into a hollow, roughly spherical shell with an octahedral (432) point-group symmetry (Figure 1a). One subunit is approximate-
Abstract: Superparamagnetic iron oxide nanoparticles (SPIONs) can be used as efficient transverse relaxivity (T2) contrast agents in magnetic resonance imaging (MRI). Organizing small (D < 10 nm) SPIONs into large assemblies can considerably enhance their relaxivity. However, this assembly process is difficult to control and can easily result in unwanted aggregation and precipitation, which might further lead to lower contrast agent performance. Herein, we present highly stable protein–polymer double-stabilized SPIONs for improving contrast in MRI. We used a cationic–neutral double hydrophilic poly(N-methyl-2-vinyl pyridinium iodide-block-poly(ethylene oxide) diblock copolymer (P2QVP-b-PEO) to mediate the self-assembly of protein-cage-encapsulated iron oxide (g-Fe2O3) nanoparticles (magnetoferritin) into stable PEO-coated clusters. This approach relies on electrostatic interactions between the cationic N-methyl-2-vinylpyridinium iodide block and magnetoferritin protein cage surface (pI 4.5) to form a dense core, whereas the neutral ethylene oxide block provides a stabilizing biocompatible shell. Formation of the complexes was studied in aqueous solvent medium with dynamic light scattering (DLS) and cryogenic transmission electron microcopy (cryo-TEM). DLS results indicated that the hydrodynamic diameter (Dh) of the clusters is approximately 200 nm, and cryo-TEM showed that the clusters have an anisotropic stringlike morphology. MRI studies showed that in the clusters the longitudinal relaxivity (r1) is decreased and the transverse relaxivity (r2) is increased relative to free magnetoferritin (MF), thus indicating that clusters can provide considerable contrast enhancement.
Ferritins and virus capsids are nanoscale protein cage particles, which have evolved in nature to function in many biological transport and storage processes, such as iron deposit and viral infection.[1–4] Recent developments in nanotechnology have utilized these particles as nanocontainers for the encapsulation of various synthetic and biological materials, for example, mag[a] Prof. M. A. Kostiainen Biohybrid Materials Department of Biotechnology and Chemical Technology Aalto University, 00076 Aalto (Finland) E-mail: [email protected]
Figure 1. a) Structure of the ferritin cage from Pyrococcus furiosus. Top: Crystal structure viewed along the three-fold symmetry axis. Bottom: Cage with omitted protein subunits shows the hollow inner cavity of FT. b) Quaternization of the poly-2-vinylpyridine block with methyl iodide and electrostatic interactions with the surface of the ferritin cage. Circled: Calculated[51] electrostatic potential projected onto the solvent-accessible surface of three protein subunits surrounding the pore at the threefold axis. Values range from 0 kBTe 1 (blue) to 9 kBTe 1 (red), for which kB = Boltzmann constant, T = absolute temperature, and e = elementary charge.
[c] Dr. A. Laiho AMI Centre, Aalto NeuroImaging, School of Science Aalto University, 00076 Aalto (Finland) Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201304070.
[b] S. Thk Molecular Materials, Department of Applied Physics Aalto University, 00076 Aalto (Finland) Chem. Eur. J. 2014, 20, 2718 – 2722
2718
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication ly 20 kDa in size, and the outer and inner diameters of the cage are approximately 12 and 8 nm, respectively. The surface of the ferritin cage is negatively charged (pI 4.5) in a neutral environment, thus allowing high-affinity electrostatic interactions with polycationic compounds and particles.[8, 17–19] The negative charge density is heterogeneously distributed along the outer cage surface and is primarily located around the eight pores at the threefold symmetry axes (Figure 1b). Empty ferritin cages (apoferritin, FT) can be filled with iron oxide Fe3O4-g-Fe2O3 nanoparticles through synthetic methods to yield superparamagnetic magnetoferritin (MF) particles.[5] Magnetoferritin exhibits remarkable magnetic properties relative to simple ferritin proteins[18, 20] and has previously been utilized, for example, as a magnetic resonance imaging (MRI) contrast agent[21] with suggested suitability for in vivo imaging of the liver and spleen.[22] MF also allows inherent targeting and visualization of tumor tissues without the use of any additional chemical components, thus making it an ideal building block for biomedical applications.[23] MRI is an important noninvasive method for medical imaging. Contrast in MRI originates from different relaxation rates of water protons obtained from different tissue types, and it can be enhanced in two ways by using contrast agents. Longitudinal relaxivity (T1) contrast agents, such as chelated gadolinium, can interact with local water protons and shorten their spin–lattice relaxation time, which causes signal brightening. However, transverse relaxivity (T2) contrast agents induce decoherence in proton spin phasing and therefore enhance spin– spin relaxation to give darker (negative) contrast in MR images. In general, superparamagnetic iron oxide nanoparticles (SPIONs) shorten the relaxation time of water protons and can therefore function as T2 signal darkening contrast agents for MRI.[24, 25] T2 relaxivity and consequently also the contrast enhancement increase with increasing particle size.[26] Alternatively, small SPIONs can be organized into larger assemblies to increase MRI contrast.[27–29] However, severe aggregation and assemblies that are too large to exhibit suitable pharmacokinetic profiles should be avoided.[30, 31] Several different approaches have been used to address this problem and prevent aggregation.[32] For example, polymersomes derived from amphiphilic block copolymers have been used to direct the packing of iron oxide nanoparticles into dense self-assembled structures with excellent contrast properties.[33] Double hydrophilic block copolymers have also been utilized to create core–shell structures with dense, electrostatically cross-linked, iron oxide nanoparticle–polyelectrolyte cores and neutral polymer coronas.[34, 35] In this paper, we show that a cationic–neutral double hydrophilic poly(N-methyl-2-vinyl pyridinium iodide-block-poly(ethylene oxide) diblock copolymer (P2QVP-b-PEO) can self-assemble with ferritin and magnetoferritin particles into clusters with well-defined size and morphology. The assembly process is driven by the electrostatic attraction between the cationic P2QVP block and anionic ferritin surface (Figure 1b). The PEO block is organized on the surface of the clusters and limits the cluster growth and provides a neutral biocompatible corona with excellent solution stability. MRI imaging shows that MF Chem. Eur. J. 2014, 20, 2718 – 2722
www.chemeurj.org
can provide efficient T2 contrast enhancement, which can be even further increased by cluster formation with P2QVP-b-PEO. P2VP29-b-PEO227 (Mw/Mn : 1.13) was quaternized with methyl iodide in dimethylformamide.[36] The first indirect evidence of a successful reaction was that the P2QVP-b-PEO product became highly water soluble owing to the charged N-methyl2-vinylpyridinium group, whereas the starting material was only sparingly soluble in neutral water. The structure of P2QVP-b-PEO was characterized by 1H NMR and FTIR spectroscopy. 1H NMR spectroscopy in D2O showed a broad N-methyl signal at d 3.5–4.5 ppm (partially overlapping with PEO CH2 signals), whereas the aromatic region showed the expected signals at d = 7.0–8.5 ppm. FTIR spectra also supported the fact that the reaction had succeeded by showing new absorption peaks at 1627 and 1513 cm 1 from the skeletal vibrations of quaternized pyridine rings. Furthermore, the signal from the unmodified pyridine rings at 1590 cm 1 was decreased or vanished from the spectrum of the quaternized product.[37] On the basis of the 1H NMR and FTIR spectroscopic characterization data, the extent of quaternization was determined to be 60– 80 % depending on the reaction (see the Supporting Information). Dynamic light scattering (DLS) was used to study the electrostatic interaction between the protein cages and the quaternized diblock copolymer and homopolymer. Initially, we studied the interaction between FT and P2QVP-b-PEO. FT and MF cages are known to behave and bind cationic compounds in a similar manner, which is to be expected since the encapsulated materials inside the cage do not change its outside properties.[8] A solution that contained only free FT (60 mg L 1) showed the expected peak at a hydrodynamic diameter (Dh) of approximately 12 nm, which is in excellent agreement with diameter of the ferritin cage on the basis of the crystal structure. When the FT solution was titrated with P2QVP-b-PEO polymer solution (0.1–10 mg mL 1), the free ferritin particles were efficiently clustered by the polymer into assemblies with a diameter of approximately 200 nm. In the DLS size-distribution profiles, this could be observed as a decrease in the peak that corresponded to the free protein cage as the polymer concentration increased. At the same time, another larger peak appeared that corresponded to the clusters of P2QVP-b-PEO and FT (Figure 2a). This data can be plotted to show the decrease of the free FT volume fraction (vol %, primary y axis) and the increase in the size of the secondary assembly clusters (Dh, secondary y axis) (Figure 2b). When the polymer concentration reached 6 mg L 1, free FT particles were no longer observed. The cluster reached its maximum size (Dh 200 nm) at a polymer concentration of 10 mg L 1 (Figure 2b), after which the cluster size remained constant. Figure 2c shows the measured electrophoretic mobility for free FT (0.2 mg mL 1), free P2QVP-b-PEO (0.19 mg mL 1), and their complex. It can be clearly observed that FT and P2QVP-bPEO are oppositely charged as their measured electrophoretic values are 1.3 and 1.1 mm cm V 1 s 1, respectively. The mobility of the complex was measured to be small (0.3 mm cm V 1 s 1) and between the free components, thus indicating that a complex with a neutral net charge had been formed. As a control
2719
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication
Figure 3. Cryo-TEM images of a) diblock copolymer (P2QVP-b-PEO) and b) homopolymer–MF clusters. With P2QVP-b-PEO, 100–200 nm stringlike clusters are observed, unlike the clusters formed with P2QVP homopolymer, which are large amorphous aggregates. Dimensions for the small insets below the larger images are 100 100 nm. Figure 2. a) DLS size-distribution profiles of FT (60 mg mL 1) with different P2QVP-b-PEO concentrations. Inset: Corresponding second-order autocorrelation curves. b) DLS titration data plotted to show the decrease in free FT and increase in the size of the forming FT–polymer cluster (secondary assembly). c) Electrophoretic mobility for the free diblock copolymer and FT as well as their complex. d) DLS data for the titration of FT with P2QVP homopolymer.
experiment, a titration of FT with quaternized P2QVP50 homopolymer was also carried out (Figure 2d). The binding efficiency of the homopolymer was found to be on the same order as the diblock copolymer and no free FT was observed at homopolymer concentrations above 5 mg L 1. However, without the stabilizing effect of the PEO block, the P2QVP-FT clusters had a much larger size (> 1000 nm), which indicated aggregation. The binding and cluster formation observed here with the two polymers and FT are in good agreement with our previously published results on similar multivalent cationic polymers and negatively charged protein cages.[38–41] The clusters formed with P2QVP-b-PEO diblock copolymer, PQ2VP homopolymer, and MF were visualized with cryogenic transmission electron microscopy (cryo-TEM). MF encapsulates an iron oxide nanoparticle, which creates a dense higheratomic-number region inside the protein cage and makes the imaging of MF particles particularly easy with electron microscopy. The morphology and size of the diblock copolymer–MF complexes is related to the chemical block composition of the copolymer and to the relative volume ratios of the two blocks. Many morphological studies of different block copolymer and nanoparticle complexes have been reported, and in many cases the weight fraction of the hydrophilic noninteracting block has been observed to control the polymer–nanoparticle complex morphology.[42, 43] With our materials, when P2QVP-b-PEO forms a complex with MF, the volume fraction of the core (which consists of cationic P2QVP bound with MF) increases and becomes less hydrophilic owing to charge neutralization. Because of the relatively short PEO chain and the consequent small corona crowding, elongated stringlike structures are observed (FigChem. Eur. J. 2014, 20, 2718 – 2722
www.chemeurj.org
ure 3a).[44] The observed MF chains are typically 100–200 nm in length and linear (single MF wide). It should be noted that even though larger magnetic nanoparticles can, as such, form elongated stringlike structures owing to strong dipole–dipole interactions that overcome Brownian motion, MF particles are not magnetic enough to form linear structures in the absence of a polymer. Previous studies on the assembly of multiblock copolymers that contain a PEO block have shown that the shorter PEO block (lower weight fraction of PEO, wPEO = 0.39– 0.32) dilates the morphology to wormlike structures.[44–46] Similar elongated or cylindrical structures from block copolymers complexed with superparamagnetic nanoparticles or horse spleen ferritin were reported by Berret et al.[34] and Shin et al.[43] As with the DLS experiments, The P2QVP homopolymer was again used as a reference material. In accordance with previous results, homopolymer–MF arranged into large loosely bound aggregates, which directly supports the stabilizing role of PEO in our system. To evaluate the ability of small PEO-stabilized MF clusters to enhance the contrast in MRI, agarose gel phantom samples that contained different amounts of free MF or clusters formed with P2QVP-b-PEO were studied. T1 and T2 relaxation times were measured from 1 % phantoms (200 mL) with a 3-T Magnetom Skyra whole-body scanner using a standard 20-channel head/neck coil. Mean signal intensities from each sample were determined and plotted against the echo time (TE) and inversion time (TI) and fitted to an exponential function to extract spin–lattice (T1) and spin–spin (T2) relaxation times, respectively. Figure 4a and b show the respective relaxation rates (1/T1 and 1/T2) plotted against the iron atom concentration and linear fits to obtain the longitudinal (r1) and transverse (r2) proton relaxation rates in response to iron concentration (relaxivities). The measured longitudinal relaxivity for free MF with the chosen imaging parameters is r1 = 1.46 s 1 mm 1, which corresponds well to the previously determined values for MF with similar iron oxide cores.[21, 47] The longitudinal relax-
2720
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication sus can be self-assembled into stringlike clusters with a double hydrophilic and cationic diblock copolymer (P2QVP-b-PEO). These clusters are effectively double-stabilized: first, by the block copolymer and second, by the protein cage. Double stabilization also allows the system to disassemble without severe aggregation taking place as the free particles are also effectively stabilized. The PEO chain is crucial for the formation and stabilization of small clusters (Dh 200 nm) because large amorphous aggregates are formed when using only P2QVP homopolymer. MRI experiments showed that even such small clusters can modulate spin–lattice (T1) or spin–spin (T2) relaxation of the surrounding protons by enhancing their realignment or spin dephasing, respectively. Consequently, such clusters can provide efficient contrast enhancement for MRI and pave the way toward highly stable double-stabilized SPIONs and their in vivo application.
Acknowledgements The authors thank J. Seitsonen for the cryo-TEM analysis and R. Sepponen for the support with MRI. J. McKee is thanked for fruitful discussions. Financial support from the Academy of Finland (grant nos. 137582, 267497, and 263504), the Emil Aaltonen Foundation, and Aalto University is gratefully acknowledged. This work made use of the Aalto University Nanomicroscopy Centre (Aalto NMC).
Figure 4. Magnetic resonance imaging of phantom samples. Comparison of a) spin–lattice (1/T1) and b) spin–spin (1/T2) relaxation rates between MF– polymer clusters and free MF. c) T1-weighted and d) T2-weighted 2D MRI images of 200 mL 1 % agarose gel phantoms with different MF concentrations (normalized to the amount of iron). e) 2D images of gel phantoms, in which the MF–polymer clusters (cFe = 3.6 mm) have been injected into the center of two samples, clearly show the location of the contrast agent.
ivity of MF can be decreased significantly (89 %) by forming small clusters with P2QVP-b-PEO for which r1 = 0.77 s 1 mm 1. The transverse relaxivity is changed to the opposite direction owing to the formation of stringlike clusters. Figure 4b shows clearly that the longitudinal relaxivity for the clusters (r2 = 36.7 s 1 mm 1) is higher than that for the free MF (r2 = 28.4 s 1 mm 1). A high r2/r1 ratio is particularly important for T2weighted images, in which a higher ratio improves the contrast. With P2QVP-b-PEO clusters the ratio is 47.7, which is enough to achieve significant contrast improvement at moderate iron concentrations. Importantly, the r2/r1 ratio is improved in the stringlike clusters while avoiding the formation of large aggregates. The observed changes in the longitudinal and transverse relaxivities can be explained by a clustering factor and lower diffusivity of water[48] near a well-hydrated core, which is facilitated by the hydrophilic PEO chains, as suggested by Gillis and co-workers[49] and Qin et al.,[50] respectively. Visual inspection of the imaged phantom samples shows clearly the effect of increasing iron concentration on the MRI contrast (Figure 4c–e). T1-weighted images show a signal brightening effect, whereas contrast in T2-weighted images is darkened. Figure 4c shows that at concentrations as low as cFe = 0.45 mm the sample is considerably brighter than the blank phantom without contrast agent (bottom two images). Furthermore, clusters that have been injected into the center of the gel phantom can be easily visualized spatially as a dark spot in the center of T2-weighted 2D images (Figure 4e). In conclusion, we have shown that empty (FT) or iron oxide nanoparticle filled (MF) ferritin particles from Pyrococcus furioChem. Eur. J. 2014, 20, 2718 – 2722
www.chemeurj.org
Keywords: block copolymers · iron · magnetic resonance imaging · nanoparticles · self-assembly [1] M. Young, D. Willits, M. Uchida, T. Douglas, Annu. Rev. Phytopathol. 2008, 46, 361 – 384. [2] M. Uchida, S. Kang, C. Reichhardt, K. Harlen, T. Douglas, BBA-Gen. Subj. 2010, 1800, 834 – 845. [3] M. Uchida, M. T. Klem, M. Allen, P. Suci, M. Flenniken, E. Gillitzer, Z. Varpness, L. O. Liepold, M. Young, T. Douglas, Adv. Mater. 2007, 19, 1025 – 1042. [4] T. Douglas, M. Young, Science 2006, 312, 873 – 875. [5] F. C. Meldrum, B. R. Heywood, S. Mann, Science 1992, 257, 522 – 523. [6] M. Comellas-Aragones, H. Engelkamp, V. I. Claessen, N. A. J. M. Sommerdijk, A. E. Rowan, P. C. M. Christianen, J. C. Maan, B. J. M. Verduin, J. J. L. M. Cornelissen, R. J. M. Nolte, Nat. Nanotechnol. 2007, 2, 635 – 639. [7] A. de La Escosura, R. J. M. Nolte, J. J. L. M. Cornelissen, J. Mater. Chem. 2009, 19, 2274 – 2278. [8] M. A. Kostiainen, P. Hiekkataipale, A. Laiho, V. Lemieux, J. Seitsonen, J. Ruokolainen, P. Ceci, Nat. Nanotechnol. 2013, 8, 52 – 56. [9] T. Douglas, M. Young, Nature 1998, 393, 152 – 155. [10] S. L. Kuan, Y. Wu, T. Weil, Macromol. Rapid Commun. 2013, 34, 380 – 392. [11] K. S. Raja, Q. Wang, M. G. Finn, ChemBioChem 2003, 4, 1348 – 1351. [12] E. Strable, J. E. Johnson, M. G. Finn, Nano Lett. 2004, 4, 1385 – 1389. [13] E. C. Theil, Annu. Rev. Biochem. 1987, 56, 289 – 315. [14] W. H. Massover, Micron 1993, 24, 389 – 437. [15] J. Tatur, P.-L. Hagedoorn, M. Overeijnder, W. Hagen, Extremophiles 2006, 10, 139 – 148. [16] J. Tatur, W. R. Hagen, P. M. Matias, J. Biol. Inorg. Chem. 2007, 12, 615 – 630. [17] S. Srivastava, B. Samanta, B. J. Jordan, R. Hong, Q. Xiao, M. T. Tuominen, V. M. Rotello, J. Am. Chem. Soc. 2007, 129, 11776 – 11780. [18] M. A. Kostiainen, P. Ceci, M. Fornara, P. Hiekkataipale, O. Kasyutich, R. J. M. Nolte, J. J. L. M. Cornelissen, R. D. Desautels, J. van Lierop, ACS Nano 2011, 5, 6394 – 6402.
2721
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication [19] M. A. Kostiainen, O. Kasyutich, J. J. L. M. Cornelissen, R. J. M. Nolte, Nat. Chem. 2010, 2, 394 – 399. [20] S. Gider, D. D. Awschalom, T. Douglas, S. Mann, M. Chaparala, Science 1995, 268, 77 – 80. [21] J. W. Bulte, T. Douglas, S. Mann, R. B. Frankel, B. M. Moskowitz, R. A. Brooks, C. D. Baumgarner, J. Vymazal, M. P. Strub, J. A. Frank, J. Magn. Reson. Imaging 1994, 4, 497 – 505. [22] J. W. M. Bulte, T. Douglas, S. Mann, J. Vymazal, P. G. Laughlin, J. A. Frank, Acad. Radiol. 1995, 2, 871 – 878. [23] K. Fan, C. Cao, Y. Pan, D. Lu, D. Yang, J. Feng, L. Song, M. Liang, X. Yan, Nat. Nanotechnol. 2012, 7, 459 – 464. [24] A. K. Gupta, M. Gupta, Biomaterials 2005, 26, 3995 – 4021. [25] G. Mikhaylov, U. Mikac, A. A. Magaeva, V. I. Itin, E. P. Naiden, I. Psakhye, L. Babes, T. Reinheckel, C. Peters, R. Zeiser, Nat. Nanotechnol. 2011, 6, 594 – 602. [26] U. I. Tromsdorf, N. C. Bigall, M. G. Kaul, O. T. Bruns, M. S. Nikolic, B. Mollwitz, R. Sperling, R. Reimer, H. Hohenberg, W. J. Parak, Nano Lett. 2007, 7, 2422 – 2427. [27] J. M. Perez, L. Josephson, T. O’Loughlin, D. Hogemann, R. Weissleder, Nat. Biotechnol. 2002, 20, 816 – 820. [28] L. Josephson, J. M. Perez, R. Weissleder, Angew. Chem. 2001, 113, 3304 – 3306; Angew. Chem. Int. Ed. 2001, 40, 3204 – 3206. [29] J. M. Perez, F. J. Simeone, Y. Saeki, L. Josephson, R. Weissleder, J. Am. Chem. Soc. 2003, 125, 10192 – 10193. [30] T. K. Jain, M. K. Reddy, M. A. Morales, D. L. Leslie-Pelecky, V. Labhasetwar, Mol. Pharm. 2008, 5, 316 – 327. [31] C. Chouly, D. Pouliquen, I. Lucet, J. J. Jeune, P. Jallet, J. Microencapsulation 1996, 13, 245 – 255. [32] S. M. Ryan, G. Mantovani, X. Wang, D. M. Haddleton, D. J. Brayden, Expert Opin. Drug Delivery 2008, 5, 371 – 383. [33] R. J. Hickey, A. S. Haynes, J. M. Kikkawa, S.-J. Park, J. Am. Chem. Soc. 2011, 133, 1517 – 1525. [34] J.-F. Berret, N. Schonbeck, F. Gazeau, D. El Kharrat, O. Sandre, A. Vacher, M. Airiau, J. Am. Chem. Soc. 2006, 128, 1755 – 1761. [35] M. Dan, D. F. Scott, P. Hardy, R. J. Wydra, J. Z. Hilt, R. Yokel, Y. Bae, Pharm. Res. 2013, 30, 552 – 561. [36] I. K. Voets, A. de Keizer, P. de Waard, P. M. Frederik, P. H. H. Bomans, H. Schmalz, A. Walther, S. M. King, F. A. M. Leermakers, M. A. Cohen Stuart,
Chem. Eur. J. 2014, 20, 2718 – 2722
www.chemeurj.org
[37] [38] [39] [40] [41] [42]
[43] [44] [45] [46] [47]
[48] [49] [50] [51]
Angew. Chem. 2006, 118, 6825 – 6828; Angew. Chem. Int. Ed. 2006, 45, 6673 – 6676. I. Tokarev, M. Orlov, S. Minko, Adv. Mater. 2006, 18, 2458 – 2460. G. Doni, M. A. Kostiainen, A. Danani, G. M. Pavan, Nano Lett. 2011, 11, 723 – 728. M. A. Kostiainen, P. Hiekkataipale, J. A. de La Torre, R. J. M. Nolte, J. J. L. M. Cornelissen, J. Mater. Chem. 2011, 21, 2112 – 2117. M. A. Kostiainen, C. Pietsch, R. Hoogenboom, R. J. M. Nolte, J. J. L. M. Cornelissen, Adv. Funct. Mater. 2011, 21, 2012 – 2019. J. Mikkil, H. Rosilo, S. Nummelin, J. Seitsonen, J. Ruokolainen, M. A. Kostiainen, ACS Macro Lett. 2013, 2, 720 – 724. B. L. Sanchez-Gaytan, W. Cui, Y. Kim, M. A. Mendez-Polanco, T. V Duncan, M. Fryd, B. B. Wayland, S.-J. Park, Angew. Chem. 2007, 119, 9395 – 9398; Angew. Chem. Int. Ed. 2007, 46, 9235 – 9238. Y. Lin, A. Bçker, J. He, K. Sill, H. Xiang, C. Abetz, X. Li, J. Wang, T. Emrick, S. Long, Nature 2005, 434, 55 – 59. Z. Li, E. Kesselman, Y. Talmon, M. A. Hillmyer, T. P. Lodge, Science 2004, 306, 98 – 101. F. S. Bates, G. H. Fredrickson, Annu. Rev. Phys. Chem. 1990, 41, 525 – 557. Y. Geng, F. Ahmed, N. Bhasin, D. E. Discher, J. Phys. Chem. B 2005, 109, 3772 – 3779. M. Uchida, M. Terashima, C. H. Cunningham, Y. Suzuki, D. A. Willits, A. F. Willis, P. C. Yang, P. S. Tsao, M. V. McConnell, M. J. Young, Magn. Reson. Med. 2008, 60, 1073 – 1081. C. Paquet, H. W. de Haan, D. M. Leek, H.-Y. Lin, B. Xiang, G. Tian, A. Kell, B. Simard, ACS Nano 2011, 5, 3104 – 3112. U. Hfeli, M. Zborowski, A. Roch, Y. Gossuin, R. N. Muller, P. Gillis, J. Magn. Magn. Mater. 2005, 293, 532 – 539. J. Qin, S. Laurent, Y. S. Jo, A. Roch, M. Mikhaylova, Z. M. Bhujwalla, R. N. Muller, M. Muhammed, Adv. Mater. 2007, 19, 1874 – 1878. T. J. Dolinsky, J. E. Nielsen, J. A. McCammon, N. A. Baker, Nucl. Acids Res. 2004, 32, W665 – W667.
Received: October 17, 2013 Published online on February 12, 2014
2722
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication
& Magnetic Resonance Imaging
Diblock-Copolymer-Mediated Self-Assembly of Protein-Stabilized Iron Oxide Nanoparticle Clusters for Magnetic Resonance Imaging Sari Thk,[b] Ari Laiho,[c] and Mauri A. Kostiainen*[a] netic nanoparticles and enzymes.[5–7] The biggest advantage of protein cages from the materials point of view is their well-defined size and shape, hollow inner cavity, water solubility, and reversible assembly/disassembly process from small protein subunits into the fully assembled cage.[8, 9] Furthermore, the outer and inner surfaces of protein cages can easily be functionalized with genetic or chemical methods, which allow the particles to be decorated with functional moieties in a controlled manner.[10–12] Ferritin protein cages are found in almost all living organisms, ranging from mammals to Archaea and bacteria.[13] In nature, ferritins are intracellular proteins that play a key role in the isolation and storage of iron.[14] In this work, we have utilized recombinant ferritin particles from the hyperthermophile Pyrococcus furiosus.[15, 16] This protein cage is composed of 24 structurally similar protein subunits that assemble into a hollow, roughly spherical shell with an octahedral (432) point-group symmetry (Figure 1a). One subunit is approximate-
Abstract: Superparamagnetic iron oxide nanoparticles (SPIONs) can be used as efficient transverse relaxivity (T2) contrast agents in magnetic resonance imaging (MRI). Organizing small (D < 10 nm) SPIONs into large assemblies can considerably enhance their relaxivity. However, this assembly process is difficult to control and can easily result in unwanted aggregation and precipitation, which might further lead to lower contrast agent performance. Herein, we present highly stable protein–polymer double-stabilized SPIONs for improving contrast in MRI. We used a cationic–neutral double hydrophilic poly(N-methyl-2-vinyl pyridinium iodide-block-poly(ethylene oxide) diblock copolymer (P2QVP-b-PEO) to mediate the self-assembly of protein-cage-encapsulated iron oxide (g-Fe2O3) nanoparticles (magnetoferritin) into stable PEO-coated clusters. This approach relies on electrostatic interactions between the cationic N-methyl-2-vinylpyridinium iodide block and magnetoferritin protein cage surface (pI 4.5) to form a dense core, whereas the neutral ethylene oxide block provides a stabilizing biocompatible shell. Formation of the complexes was studied in aqueous solvent medium with dynamic light scattering (DLS) and cryogenic transmission electron microcopy (cryo-TEM). DLS results indicated that the hydrodynamic diameter (Dh) of the clusters is approximately 200 nm, and cryo-TEM showed that the clusters have an anisotropic stringlike morphology. MRI studies showed that in the clusters the longitudinal relaxivity (r1) is decreased and the transverse relaxivity (r2) is increased relative to free magnetoferritin (MF), thus indicating that clusters can provide considerable contrast enhancement.
Ferritins and virus capsids are nanoscale protein cage particles, which have evolved in nature to function in many biological transport and storage processes, such as iron deposit and viral infection.[1–4] Recent developments in nanotechnology have utilized these particles as nanocontainers for the encapsulation of various synthetic and biological materials, for example, mag[a] Prof. M. A. Kostiainen Biohybrid Materials Department of Biotechnology and Chemical Technology Aalto University, 00076 Aalto (Finland) E-mail: [email protected]
Figure 1. a) Structure of the ferritin cage from Pyrococcus furiosus. Top: Crystal structure viewed along the three-fold symmetry axis. Bottom: Cage with omitted protein subunits shows the hollow inner cavity of FT. b) Quaternization of the poly-2-vinylpyridine block with methyl iodide and electrostatic interactions with the surface of the ferritin cage. Circled: Calculated[51] electrostatic potential projected onto the solvent-accessible surface of three protein subunits surrounding the pore at the threefold axis. Values range from 0 kBTe 1 (blue) to 9 kBTe 1 (red), for which kB = Boltzmann constant, T = absolute temperature, and e = elementary charge.
[c] Dr. A. Laiho AMI Centre, Aalto NeuroImaging, School of Science Aalto University, 00076 Aalto (Finland) Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201304070.
[b] S. Thk Molecular Materials, Department of Applied Physics Aalto University, 00076 Aalto (Finland) Chem. Eur. J. 2014, 20, 2718 – 2722
2718
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication ly 20 kDa in size, and the outer and inner diameters of the cage are approximately 12 and 8 nm, respectively. The surface of the ferritin cage is negatively charged (pI 4.5) in a neutral environment, thus allowing high-affinity electrostatic interactions with polycationic compounds and particles.[8, 17–19] The negative charge density is heterogeneously distributed along the outer cage surface and is primarily located around the eight pores at the threefold symmetry axes (Figure 1b). Empty ferritin cages (apoferritin, FT) can be filled with iron oxide Fe3O4-g-Fe2O3 nanoparticles through synthetic methods to yield superparamagnetic magnetoferritin (MF) particles.[5] Magnetoferritin exhibits remarkable magnetic properties relative to simple ferritin proteins[18, 20] and has previously been utilized, for example, as a magnetic resonance imaging (MRI) contrast agent[21] with suggested suitability for in vivo imaging of the liver and spleen.[22] MF also allows inherent targeting and visualization of tumor tissues without the use of any additional chemical components, thus making it an ideal building block for biomedical applications.[23] MRI is an important noninvasive method for medical imaging. Contrast in MRI originates from different relaxation rates of water protons obtained from different tissue types, and it can be enhanced in two ways by using contrast agents. Longitudinal relaxivity (T1) contrast agents, such as chelated gadolinium, can interact with local water protons and shorten their spin–lattice relaxation time, which causes signal brightening. However, transverse relaxivity (T2) contrast agents induce decoherence in proton spin phasing and therefore enhance spin– spin relaxation to give darker (negative) contrast in MR images. In general, superparamagnetic iron oxide nanoparticles (SPIONs) shorten the relaxation time of water protons and can therefore function as T2 signal darkening contrast agents for MRI.[24, 25] T2 relaxivity and consequently also the contrast enhancement increase with increasing particle size.[26] Alternatively, small SPIONs can be organized into larger assemblies to increase MRI contrast.[27–29] However, severe aggregation and assemblies that are too large to exhibit suitable pharmacokinetic profiles should be avoided.[30, 31] Several different approaches have been used to address this problem and prevent aggregation.[32] For example, polymersomes derived from amphiphilic block copolymers have been used to direct the packing of iron oxide nanoparticles into dense self-assembled structures with excellent contrast properties.[33] Double hydrophilic block copolymers have also been utilized to create core–shell structures with dense, electrostatically cross-linked, iron oxide nanoparticle–polyelectrolyte cores and neutral polymer coronas.[34, 35] In this paper, we show that a cationic–neutral double hydrophilic poly(N-methyl-2-vinyl pyridinium iodide-block-poly(ethylene oxide) diblock copolymer (P2QVP-b-PEO) can self-assemble with ferritin and magnetoferritin particles into clusters with well-defined size and morphology. The assembly process is driven by the electrostatic attraction between the cationic P2QVP block and anionic ferritin surface (Figure 1b). The PEO block is organized on the surface of the clusters and limits the cluster growth and provides a neutral biocompatible corona with excellent solution stability. MRI imaging shows that MF Chem. Eur. J. 2014, 20, 2718 – 2722
www.chemeurj.org
can provide efficient T2 contrast enhancement, which can be even further increased by cluster formation with P2QVP-b-PEO. P2VP29-b-PEO227 (Mw/Mn : 1.13) was quaternized with methyl iodide in dimethylformamide.[36] The first indirect evidence of a successful reaction was that the P2QVP-b-PEO product became highly water soluble owing to the charged N-methyl2-vinylpyridinium group, whereas the starting material was only sparingly soluble in neutral water. The structure of P2QVP-b-PEO was characterized by 1H NMR and FTIR spectroscopy. 1H NMR spectroscopy in D2O showed a broad N-methyl signal at d 3.5–4.5 ppm (partially overlapping with PEO CH2 signals), whereas the aromatic region showed the expected signals at d = 7.0–8.5 ppm. FTIR spectra also supported the fact that the reaction had succeeded by showing new absorption peaks at 1627 and 1513 cm 1 from the skeletal vibrations of quaternized pyridine rings. Furthermore, the signal from the unmodified pyridine rings at 1590 cm 1 was decreased or vanished from the spectrum of the quaternized product.[37] On the basis of the 1H NMR and FTIR spectroscopic characterization data, the extent of quaternization was determined to be 60– 80 % depending on the reaction (see the Supporting Information). Dynamic light scattering (DLS) was used to study the electrostatic interaction between the protein cages and the quaternized diblock copolymer and homopolymer. Initially, we studied the interaction between FT and P2QVP-b-PEO. FT and MF cages are known to behave and bind cationic compounds in a similar manner, which is to be expected since the encapsulated materials inside the cage do not change its outside properties.[8] A solution that contained only free FT (60 mg L 1) showed the expected peak at a hydrodynamic diameter (Dh) of approximately 12 nm, which is in excellent agreement with diameter of the ferritin cage on the basis of the crystal structure. When the FT solution was titrated with P2QVP-b-PEO polymer solution (0.1–10 mg mL 1), the free ferritin particles were efficiently clustered by the polymer into assemblies with a diameter of approximately 200 nm. In the DLS size-distribution profiles, this could be observed as a decrease in the peak that corresponded to the free protein cage as the polymer concentration increased. At the same time, another larger peak appeared that corresponded to the clusters of P2QVP-b-PEO and FT (Figure 2a). This data can be plotted to show the decrease of the free FT volume fraction (vol %, primary y axis) and the increase in the size of the secondary assembly clusters (Dh, secondary y axis) (Figure 2b). When the polymer concentration reached 6 mg L 1, free FT particles were no longer observed. The cluster reached its maximum size (Dh 200 nm) at a polymer concentration of 10 mg L 1 (Figure 2b), after which the cluster size remained constant. Figure 2c shows the measured electrophoretic mobility for free FT (0.2 mg mL 1), free P2QVP-b-PEO (0.19 mg mL 1), and their complex. It can be clearly observed that FT and P2QVP-bPEO are oppositely charged as their measured electrophoretic values are 1.3 and 1.1 mm cm V 1 s 1, respectively. The mobility of the complex was measured to be small (0.3 mm cm V 1 s 1) and between the free components, thus indicating that a complex with a neutral net charge had been formed. As a control
2719
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication
Figure 3. Cryo-TEM images of a) diblock copolymer (P2QVP-b-PEO) and b) homopolymer–MF clusters. With P2QVP-b-PEO, 100–200 nm stringlike clusters are observed, unlike the clusters formed with P2QVP homopolymer, which are large amorphous aggregates. Dimensions for the small insets below the larger images are 100 100 nm. Figure 2. a) DLS size-distribution profiles of FT (60 mg mL 1) with different P2QVP-b-PEO concentrations. Inset: Corresponding second-order autocorrelation curves. b) DLS titration data plotted to show the decrease in free FT and increase in the size of the forming FT–polymer cluster (secondary assembly). c) Electrophoretic mobility for the free diblock copolymer and FT as well as their complex. d) DLS data for the titration of FT with P2QVP homopolymer.
experiment, a titration of FT with quaternized P2QVP50 homopolymer was also carried out (Figure 2d). The binding efficiency of the homopolymer was found to be on the same order as the diblock copolymer and no free FT was observed at homopolymer concentrations above 5 mg L 1. However, without the stabilizing effect of the PEO block, the P2QVP-FT clusters had a much larger size (> 1000 nm), which indicated aggregation. The binding and cluster formation observed here with the two polymers and FT are in good agreement with our previously published results on similar multivalent cationic polymers and negatively charged protein cages.[38–41] The clusters formed with P2QVP-b-PEO diblock copolymer, PQ2VP homopolymer, and MF were visualized with cryogenic transmission electron microscopy (cryo-TEM). MF encapsulates an iron oxide nanoparticle, which creates a dense higheratomic-number region inside the protein cage and makes the imaging of MF particles particularly easy with electron microscopy. The morphology and size of the diblock copolymer–MF complexes is related to the chemical block composition of the copolymer and to the relative volume ratios of the two blocks. Many morphological studies of different block copolymer and nanoparticle complexes have been reported, and in many cases the weight fraction of the hydrophilic noninteracting block has been observed to control the polymer–nanoparticle complex morphology.[42, 43] With our materials, when P2QVP-b-PEO forms a complex with MF, the volume fraction of the core (which consists of cationic P2QVP bound with MF) increases and becomes less hydrophilic owing to charge neutralization. Because of the relatively short PEO chain and the consequent small corona crowding, elongated stringlike structures are observed (FigChem. Eur. J. 2014, 20, 2718 – 2722
www.chemeurj.org
ure 3a).[44] The observed MF chains are typically 100–200 nm in length and linear (single MF wide). It should be noted that even though larger magnetic nanoparticles can, as such, form elongated stringlike structures owing to strong dipole–dipole interactions that overcome Brownian motion, MF particles are not magnetic enough to form linear structures in the absence of a polymer. Previous studies on the assembly of multiblock copolymers that contain a PEO block have shown that the shorter PEO block (lower weight fraction of PEO, wPEO = 0.39– 0.32) dilates the morphology to wormlike structures.[44–46] Similar elongated or cylindrical structures from block copolymers complexed with superparamagnetic nanoparticles or horse spleen ferritin were reported by Berret et al.[34] and Shin et al.[43] As with the DLS experiments, The P2QVP homopolymer was again used as a reference material. In accordance with previous results, homopolymer–MF arranged into large loosely bound aggregates, which directly supports the stabilizing role of PEO in our system. To evaluate the ability of small PEO-stabilized MF clusters to enhance the contrast in MRI, agarose gel phantom samples that contained different amounts of free MF or clusters formed with P2QVP-b-PEO were studied. T1 and T2 relaxation times were measured from 1 % phantoms (200 mL) with a 3-T Magnetom Skyra whole-body scanner using a standard 20-channel head/neck coil. Mean signal intensities from each sample were determined and plotted against the echo time (TE) and inversion time (TI) and fitted to an exponential function to extract spin–lattice (T1) and spin–spin (T2) relaxation times, respectively. Figure 4a and b show the respective relaxation rates (1/T1 and 1/T2) plotted against the iron atom concentration and linear fits to obtain the longitudinal (r1) and transverse (r2) proton relaxation rates in response to iron concentration (relaxivities). The measured longitudinal relaxivity for free MF with the chosen imaging parameters is r1 = 1.46 s 1 mm 1, which corresponds well to the previously determined values for MF with similar iron oxide cores.[21, 47] The longitudinal relax-
2720
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication sus can be self-assembled into stringlike clusters with a double hydrophilic and cationic diblock copolymer (P2QVP-b-PEO). These clusters are effectively double-stabilized: first, by the block copolymer and second, by the protein cage. Double stabilization also allows the system to disassemble without severe aggregation taking place as the free particles are also effectively stabilized. The PEO chain is crucial for the formation and stabilization of small clusters (Dh 200 nm) because large amorphous aggregates are formed when using only P2QVP homopolymer. MRI experiments showed that even such small clusters can modulate spin–lattice (T1) or spin–spin (T2) relaxation of the surrounding protons by enhancing their realignment or spin dephasing, respectively. Consequently, such clusters can provide efficient contrast enhancement for MRI and pave the way toward highly stable double-stabilized SPIONs and their in vivo application.
Acknowledgements The authors thank J. Seitsonen for the cryo-TEM analysis and R. Sepponen for the support with MRI. J. McKee is thanked for fruitful discussions. Financial support from the Academy of Finland (grant nos. 137582, 267497, and 263504), the Emil Aaltonen Foundation, and Aalto University is gratefully acknowledged. This work made use of the Aalto University Nanomicroscopy Centre (Aalto NMC).
Figure 4. Magnetic resonance imaging of phantom samples. Comparison of a) spin–lattice (1/T1) and b) spin–spin (1/T2) relaxation rates between MF– polymer clusters and free MF. c) T1-weighted and d) T2-weighted 2D MRI images of 200 mL 1 % agarose gel phantoms with different MF concentrations (normalized to the amount of iron). e) 2D images of gel phantoms, in which the MF–polymer clusters (cFe = 3.6 mm) have been injected into the center of two samples, clearly show the location of the contrast agent.
ivity of MF can be decreased significantly (89 %) by forming small clusters with P2QVP-b-PEO for which r1 = 0.77 s 1 mm 1. The transverse relaxivity is changed to the opposite direction owing to the formation of stringlike clusters. Figure 4b shows clearly that the longitudinal relaxivity for the clusters (r2 = 36.7 s 1 mm 1) is higher than that for the free MF (r2 = 28.4 s 1 mm 1). A high r2/r1 ratio is particularly important for T2weighted images, in which a higher ratio improves the contrast. With P2QVP-b-PEO clusters the ratio is 47.7, which is enough to achieve significant contrast improvement at moderate iron concentrations. Importantly, the r2/r1 ratio is improved in the stringlike clusters while avoiding the formation of large aggregates. The observed changes in the longitudinal and transverse relaxivities can be explained by a clustering factor and lower diffusivity of water[48] near a well-hydrated core, which is facilitated by the hydrophilic PEO chains, as suggested by Gillis and co-workers[49] and Qin et al.,[50] respectively. Visual inspection of the imaged phantom samples shows clearly the effect of increasing iron concentration on the MRI contrast (Figure 4c–e). T1-weighted images show a signal brightening effect, whereas contrast in T2-weighted images is darkened. Figure 4c shows that at concentrations as low as cFe = 0.45 mm the sample is considerably brighter than the blank phantom without contrast agent (bottom two images). Furthermore, clusters that have been injected into the center of the gel phantom can be easily visualized spatially as a dark spot in the center of T2-weighted 2D images (Figure 4e). In conclusion, we have shown that empty (FT) or iron oxide nanoparticle filled (MF) ferritin particles from Pyrococcus furioChem. Eur. J. 2014, 20, 2718 – 2722
www.chemeurj.org
Keywords: block copolymers · iron · magnetic resonance imaging · nanoparticles · self-assembly [1] M. Young, D. Willits, M. Uchida, T. Douglas, Annu. Rev. Phytopathol. 2008, 46, 361 – 384. [2] M. Uchida, S. Kang, C. Reichhardt, K. Harlen, T. Douglas, BBA-Gen. Subj. 2010, 1800, 834 – 845. [3] M. Uchida, M. T. Klem, M. Allen, P. Suci, M. Flenniken, E. Gillitzer, Z. Varpness, L. O. Liepold, M. Young, T. Douglas, Adv. Mater. 2007, 19, 1025 – 1042. [4] T. Douglas, M. Young, Science 2006, 312, 873 – 875. [5] F. C. Meldrum, B. R. Heywood, S. Mann, Science 1992, 257, 522 – 523. [6] M. Comellas-Aragones, H. Engelkamp, V. I. Claessen, N. A. J. M. Sommerdijk, A. E. Rowan, P. C. M. Christianen, J. C. Maan, B. J. M. Verduin, J. J. L. M. Cornelissen, R. J. M. Nolte, Nat. Nanotechnol. 2007, 2, 635 – 639. [7] A. de La Escosura, R. J. M. Nolte, J. J. L. M. Cornelissen, J. Mater. Chem. 2009, 19, 2274 – 2278. [8] M. A. Kostiainen, P. Hiekkataipale, A. Laiho, V. Lemieux, J. Seitsonen, J. Ruokolainen, P. Ceci, Nat. Nanotechnol. 2013, 8, 52 – 56. [9] T. Douglas, M. Young, Nature 1998, 393, 152 – 155. [10] S. L. Kuan, Y. Wu, T. Weil, Macromol. Rapid Commun. 2013, 34, 380 – 392. [11] K. S. Raja, Q. Wang, M. G. Finn, ChemBioChem 2003, 4, 1348 – 1351. [12] E. Strable, J. E. Johnson, M. G. Finn, Nano Lett. 2004, 4, 1385 – 1389. [13] E. C. Theil, Annu. Rev. Biochem. 1987, 56, 289 – 315. [14] W. H. Massover, Micron 1993, 24, 389 – 437. [15] J. Tatur, P.-L. Hagedoorn, M. Overeijnder, W. Hagen, Extremophiles 2006, 10, 139 – 148. [16] J. Tatur, W. R. Hagen, P. M. Matias, J. Biol. Inorg. Chem. 2007, 12, 615 – 630. [17] S. Srivastava, B. Samanta, B. J. Jordan, R. Hong, Q. Xiao, M. T. Tuominen, V. M. Rotello, J. Am. Chem. Soc. 2007, 129, 11776 – 11780. [18] M. A. Kostiainen, P. Ceci, M. Fornara, P. Hiekkataipale, O. Kasyutich, R. J. M. Nolte, J. J. L. M. Cornelissen, R. D. Desautels, J. van Lierop, ACS Nano 2011, 5, 6394 – 6402.
2721
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Communication [19] M. A. Kostiainen, O. Kasyutich, J. J. L. M. Cornelissen, R. J. M. Nolte, Nat. Chem. 2010, 2, 394 – 399. [20] S. Gider, D. D. Awschalom, T. Douglas, S. Mann, M. Chaparala, Science 1995, 268, 77 – 80. [21] J. W. Bulte, T. Douglas, S. Mann, R. B. Frankel, B. M. Moskowitz, R. A. Brooks, C. D. Baumgarner, J. Vymazal, M. P. Strub, J. A. Frank, J. Magn. Reson. Imaging 1994, 4, 497 – 505. [22] J. W. M. Bulte, T. Douglas, S. Mann, J. Vymazal, P. G. Laughlin, J. A. Frank, Acad. Radiol. 1995, 2, 871 – 878. [23] K. Fan, C. Cao, Y. Pan, D. Lu, D. Yang, J. Feng, L. Song, M. Liang, X. Yan, Nat. Nanotechnol. 2012, 7, 459 – 464. [24] A. K. Gupta, M. Gupta, Biomaterials 2005, 26, 3995 – 4021. [25] G. Mikhaylov, U. Mikac, A. A. Magaeva, V. I. Itin, E. P. Naiden, I. Psakhye, L. Babes, T. Reinheckel, C. Peters, R. Zeiser, Nat. Nanotechnol. 2011, 6, 594 – 602. [26] U. I. Tromsdorf, N. C. Bigall, M. G. Kaul, O. T. Bruns, M. S. Nikolic, B. Mollwitz, R. Sperling, R. Reimer, H. Hohenberg, W. J. Parak, Nano Lett. 2007, 7, 2422 – 2427. [27] J. M. Perez, L. Josephson, T. O’Loughlin, D. Hogemann, R. Weissleder, Nat. Biotechnol. 2002, 20, 816 – 820. [28] L. Josephson, J. M. Perez, R. Weissleder, Angew. Chem. 2001, 113, 3304 – 3306; Angew. Chem. Int. Ed. 2001, 40, 3204 – 3206. [29] J. M. Perez, F. J. Simeone, Y. Saeki, L. Josephson, R. Weissleder, J. Am. Chem. Soc. 2003, 125, 10192 – 10193. [30] T. K. Jain, M. K. Reddy, M. A. Morales, D. L. Leslie-Pelecky, V. Labhasetwar, Mol. Pharm. 2008, 5, 316 – 327. [31] C. Chouly, D. Pouliquen, I. Lucet, J. J. Jeune, P. Jallet, J. Microencapsulation 1996, 13, 245 – 255. [32] S. M. Ryan, G. Mantovani, X. Wang, D. M. Haddleton, D. J. Brayden, Expert Opin. Drug Delivery 2008, 5, 371 – 383. [33] R. J. Hickey, A. S. Haynes, J. M. Kikkawa, S.-J. Park, J. Am. Chem. Soc. 2011, 133, 1517 – 1525. [34] J.-F. Berret, N. Schonbeck, F. Gazeau, D. El Kharrat, O. Sandre, A. Vacher, M. Airiau, J. Am. Chem. Soc. 2006, 128, 1755 – 1761. [35] M. Dan, D. F. Scott, P. Hardy, R. J. Wydra, J. Z. Hilt, R. Yokel, Y. Bae, Pharm. Res. 2013, 30, 552 – 561. [36] I. K. Voets, A. de Keizer, P. de Waard, P. M. Frederik, P. H. H. Bomans, H. Schmalz, A. Walther, S. M. King, F. A. M. Leermakers, M. A. Cohen Stuart,
Chem. Eur. J. 2014, 20, 2718 – 2722
www.chemeurj.org
[37] [38] [39] [40] [41] [42]
[43] [44] [45] [46] [47]
[48] [49] [50] [51]
Angew. Chem. 2006, 118, 6825 – 6828; Angew. Chem. Int. Ed. 2006, 45, 6673 – 6676. I. Tokarev, M. Orlov, S. Minko, Adv. Mater. 2006, 18, 2458 – 2460. G. Doni, M. A. Kostiainen, A. Danani, G. M. Pavan, Nano Lett. 2011, 11, 723 – 728. M. A. Kostiainen, P. Hiekkataipale, J. A. de La Torre, R. J. M. Nolte, J. J. L. M. Cornelissen, J. Mater. Chem. 2011, 21, 2112 – 2117. M. A. Kostiainen, C. Pietsch, R. Hoogenboom, R. J. M. Nolte, J. J. L. M. Cornelissen, Adv. Funct. Mater. 2011, 21, 2012 – 2019. J. Mikkil, H. Rosilo, S. Nummelin, J. Seitsonen, J. Ruokolainen, M. A. Kostiainen, ACS Macro Lett. 2013, 2, 720 – 724. B. L. Sanchez-Gaytan, W. Cui, Y. Kim, M. A. Mendez-Polanco, T. V Duncan, M. Fryd, B. B. Wayland, S.-J. Park, Angew. Chem. 2007, 119, 9395 – 9398; Angew. Chem. Int. Ed. 2007, 46, 9235 – 9238. Y. Lin, A. Bçker, J. He, K. Sill, H. Xiang, C. Abetz, X. Li, J. Wang, T. Emrick, S. Long, Nature 2005, 434, 55 – 59. Z. Li, E. Kesselman, Y. Talmon, M. A. Hillmyer, T. P. Lodge, Science 2004, 306, 98 – 101. F. S. Bates, G. H. Fredrickson, Annu. Rev. Phys. Chem. 1990, 41, 525 – 557. Y. Geng, F. Ahmed, N. Bhasin, D. E. Discher, J. Phys. Chem. B 2005, 109, 3772 – 3779. M. Uchida, M. Terashima, C. H. Cunningham, Y. Suzuki, D. A. Willits, A. F. Willis, P. C. Yang, P. S. Tsao, M. V. McConnell, M. J. Young, Magn. Reson. Med. 2008, 60, 1073 – 1081. C. Paquet, H. W. de Haan, D. M. Leek, H.-Y. Lin, B. Xiang, G. Tian, A. Kell, B. Simard, ACS Nano 2011, 5, 3104 – 3112. U. Hfeli, M. Zborowski, A. Roch, Y. Gossuin, R. N. Muller, P. Gillis, J. Magn. Magn. Mater. 2005, 293, 532 – 539. J. Qin, S. Laurent, Y. S. Jo, A. Roch, M. Mikhaylova, Z. M. Bhujwalla, R. N. Muller, M. Muhammed, Adv. Mater. 2007, 19, 1874 – 1878. T. J. Dolinsky, J. E. Nielsen, J. A. McCammon, N. A. Baker, Nucl. Acids Res. 2004, 32, W665 – W667.
Received: October 17, 2013 Published online on February 12, 2014
2722
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
