Multifunctional Cytotoxic Stealth Nanoparticles. A Model Approach

NANO LETTERS

Multifunctional Cytotoxic Stealth Nanoparticles. A Model Approach with Potential for Cancer Therapy

2009 Vol. 9, No. 2 636-642

Gre´gory F. Schneider,†,‡,§ Vladimir Subr,‡ Karel Ulbrich,‡ and Gero Decher*,† Centre National de la Recherche Scientifique (CNRS UPR022), Institut Charles Sadron, 23 rue du Loess, F-67034 Strasbourg-Cedex, France, UniVersite´ Louis Pasteur (ULP), 1 rue Blaise Pascal, F-67008 Strasbourg-Cedex, France, and Institute of Macromolecular Chemistry, V.V.i., AS CR, HeyroVsky Sq. 2, 162 06 Prague 6, Czech Republic Received October 2, 2008; Revised Manuscript Received December 4, 2008

ABSTRACT Here we report on a highly versatile nanoparticle-based core/shell drug delivery system consisting of cytotoxic stealth carrier particles. Their multifunctional shells, mandatory for addressing different diagnostic/treatment requirements, are constructed using a single assembly process in which various different functionalities are incorporated in a modular fashion. More specifically, we discuss a robust electrostatic and covalent layer-by-layer (LBL) assembly strategy as engineering approach toward nanoparticles with multilayer shells that combine all of the following properties: (i) a small size distribution of the nanoparticle carrier, (ii) a high stability in physiological media, (iii) attachment of a pro-drug in covalent form and thus a low toxicity of the carrier system, (iv) the triggered release and activation of the drug only after endocytosis and enzymatic cleavage, and (v) “stealthiness” and thus protection against uptake by macrophages. The fact that we employ small nanoparticles as carriers is predicted to enhance the accumulation of active drug in the tumor tissue (i.e., enhanced permeability and retention of tumor tissues, EPR). To establish this system as a proof of concept, we use the smallest nanoparticles within the interesting size range of about 25-100 nm for EPR targeting since these are the most difficult to functionalize and because they possess the highest surface area. On the basis of gold nanoparticle cores, our system allows for precise control of particle size and size distribution and also for easy monitoring of the dispersion stability by the naked eye.

The therapeutic width of cytotoxic agents is often small because such drugs tend to not be selective for cancer cells. There is hope that new systemic therapies will provide methods that can overcome nonspecific uptake by mononuclear phagocytic cells and by other nontargeted cells. Therefore the search for more selective treatments has become an issue of considerable research over the past years, and in particular, nanosized carrier systems are emerging, and are considered to possess high potential for improvement over existing cancer therapies.1,2 Many different systems have been attracting attention, originally starting with polymeric drugs, continuing later with stealth liposomes, while the focus today is shifting toward polymeric or inorganic nanoparticles,2,3 and even nanoemulsions or nanobubbles.4 While the requirements for (cancer) treatment are clear and the available nanofabrication methods are improving substantially, there is still no general and robust approach available for rationally * Corresponding author, [email protected]. † Centre National de la Recherche Scientifique (CNRS UPR022), Institut Charles Sadron, and Universite´ Louis Pasteur (ULP). ‡ Institute of Macromolecular Chemistry, v.v.i., AS CR. § Present address: Department of Chemistry and Chemical Biology, Harvard University, 12 Oxford St, Cambridge, MA 02138. 10.1021/nl802990w CCC: $40.75 Published on Web 01/26/2009

 2009 American Chemical Society

designing and realizing well-defined nanosized carrier systems with respect to the clinical needs. Several nanoparticles (NPs), including gold NPs,5 quantum dots,6-9 and magnetic nanoparticles10,11 have already been considered for therapeutic or diagnostic purposes,12 despite the fact that such applications are still seriously limited by several factors such as the potential aggregation of the nanoparticles in physiological media, their short circulation time in vivo due to elimination from the blood stream by macrophages of the mononuclear phagocytic system (MPS),13 or significant uptake by the liver before reaching any target. Other important drawbacks for potential applications include a low loading capacity, a lack of site-specific targeting, or limited release of the carried biologically active compound at the site of interest. Scheme 1 introduces the ideal design for a highly versatile nanoparticle-based core/shell system in which the shell is constructed in a modular fashion allowing the incorporation of various functionalities that are mandatory for addressing different diagnostic/treatment requirements. Layer-by-layer deposition technology14-16 allows such multifunctional systems to be realized in a built-to-order fashion in which a

Scheme 1. Schematic Depiction of Nanoparticles Coated with Multilayer Shells as New Drug Delivery Systema

a The multifunctionality arises from the stepwise construction of the shell that is assembled by the LBL method. The internal layers are arbitrarily split in two compartments (yellow (1) and red (2)) in order to indicate that different functionalities can be integrated in a modular way in different layers. Theoretically there is no limit with respect to the number of different internal compartments. The yellow compartment (1) serves primarily to compatibilize between the core and the external layers enabling the use of different core materials while maintaining the same functionalization process. However, the yellow compartment (1) as well as the red one (2) may also serve to incorporate additional functional entities such as drugs, catalysts for biochemical reactions, radionuclids for radiotherapy, proteins/nucleotides for bioactivity, or contrast agents for detection. In the present study this was not realized, but such options are well established from LBL films on flat substrates. The external layers carry functionalities such as enzymatically cleavable drugs or ligands for receptor mediated targeting, both of which must be accessible on the outside. Only the functionalization with an attached drug was chosen in the present study. The attachment of additional ligands, however, follows the same principle which has already been described for a variety of different functional groups in the past.

single deposition process can be used for the fabrication of complex and chemically highly different layer architectures of the functional NP shell. The idea to employ a sole coating technology for the deposition of all individual layers constitutes an enormous advantage with respect to an implementation of any nanoparticle fabrication process in any future industrial production. In the present report we present a model system composed of a very small core (∼13 nm in diameter; to demonstrate the power of the LBL-coating technology even for smallsbut larger than 10 nmsnanoparticles, the most difficult to functionalize with polymeric shells, while limiting aggregation and bridging of nanoparticles), a few compatibilization layers (that allow adapting to different core materials), and a single multifunctional layer providing cytotoxicity as well as stealthiness (to demonstrate a simple case of multifunctionality). Our present system can potentially be extended to include the other functionalities as indicated in Scheme 1. Please note that complex shells with several functions (e.g., for multidrug cocktails) cannot simply be replaced by a mixture of different nanoparticles each of which is equipped with a different functional shell since each component of such a mixture may have a body distribution of its own. Therefore each NP component would end up in different parts of the body and the synergistic effect of a multiple drug treatment would be lost. Please note as well that our present example, the covalent attachment of the anticancer drug to the polymer chain, is not a strict necessity and thus not a principal limitation for the delivery of therapeutic agents. Drugs other than doxorubicin have already been incorporated in and released from planar multilayer films.17 More specifically, we report on a robust electrostatic and covalent layer-by-layer (LBL) assembly strategy for the synthesis of tunable cytotoxic stealth nanoparticles possessing Nano Lett., Vol. 9, No. 2, 2009

all of the following advantages: (i) a well adjustable size in combination with a small polydispersity of the nanoparticle carrier; (ii) a high stability in physiological media; (iii) an attachment of a pro-drug in covalent form and thus a low toxicity of the carrier system (with options to load drugs also noncovalently); (iv) a triggered release and such activation of the drug only after endocytosis and enzymatic cleavage; (v) “stealthiness” and thus protection against uptake by macrophages; (vi) already a competitive drug loading capacity in comparison with other systems (with options to increase the loading); (vii) an expected low acute toxicity for human cells due to the use of gold cores ensheathed by a “stealth” corona layer (please note that the acute toxicity of gold nanoparticles is a controversial issue); (viii) the possibility to mildly dissolve the nanoparticle cores prior to an eventual treatment thus reducing the potential risk of longterm toxicity induced by accumulation of the gold cores. Integrating these eight important properties in a single drug delivery system would be highly advantageous for the design of application-relevant pro-drug systems, but an approach combining more than half of theses aspects has, to our knowledge, not been reported in the past. Size control (point i in the list of desirable properties above) in combination with the absence of aggregation in biological media (point ii) is especially important with respect to passive tumor targeting. Particles must have highly controlled dimensions, small enough to extravasate through the endothelium of the vessels of the tumor tissue but big enough to be retained in the tumor itself by the EPR effect.18 Since the optimum particle size for the treatment of specific cancers is not yet well-known, we have decided to work with very small nanoparticles (∼10 nm) since they represent the most challenging system with respect to dispersion stability and functionalization. Methods developed for the function637

alization of particles with a size of about 100 nm and higher require typically important modifications when applied to particles with half or one-third of this size. It is however, generally straightforward to functionalize bigger nanoparticles with methods developed for smaller ones. As a model pro-drug (points iii and iv) we have chosen a well-established and carefully optimized polymeric drug carrier system based on N-(2-hydroxypropyl)methacrylamide (HPMA) copolymers carrying oligopeptide spaced doxorubicin side groups.19-21 HPMA provides a hydrophilic backbone that limits phagocytosis and uptake by the liver (point v). Doxorubicin (Dox) is a very potent DNA intercalation agent whose cytotoxicity for healthy cells is effectively reduced here by a covalent attachment to the nanocarrier via an oligopeptide spacer. This spacer has been designed for specific cleavage by lysosomal enzymes, namely, cathepsins, for releasing the doxorubicin exclusively after endocytotic uptake by the tumor cells. For the purpose of the present paper, the functionality of this copolymer was extended by integrating an additional monomer to allow for covalent LBL deposition. As core materials we have selected gold nanoparticles for which we and others have already reported efficient functionalization protocols (e.g., no aggregation and high yields) based on LBL deposition22-27 since the first attempts by Caruso.28 The acute toxicity of gold nanoparticles toward human cells can be negligible, but this topic is controversial since toxicity is likely to depend on many parameters, namely, particle size, concentration, and surface functionalization to name a few.29 While stealth particles are screened from recognition and thus much less likely to show toxicity of the core than, e.g., bare particles, the gold cores are also easily dissolved by mild methods, which completely prevents any undesirable side effects arising from the gold nanoparticle templates (point viii). Last but not least, the intense and aggregation sensitive color of the gold nanoparticles facilitates the monitoring of the dispersion stability during all functionalization steps. While we start out with a simple example, it should be emphasized that each component of the multilayered architecture can be optimized for a specific purpose as outlined in Scheme 1, such as the size of the template (core), the number of internal/external layers, or the functionality of the internal layers (drugs, contrast agents, radionucleids, proteins, catalysts, etc.). The outer layer (or corona layer) provides for stealthiness but may also carry cancer-specific recognition units (e.g., such as folic acid,30 CREKA31 or SIKVAV peptides,32 or more generally multivalent ligands33), to further enhance targeting. These units can be attached to the HPMA backbone in a similar way as our model drug. In the initial experiments reported here, we designed a core/shell structure of the multifunctional nanoparticles (MFNP) as outlined in Scheme 1. First, the colloidal core composed of a gold nanoparticle (AuNP) was coated with five internal primer layers of poly(allylamine) (PAH) and poly(stryrenesulfonate) (PSS) to create a defined polyamine surface that is independent of the core material. We denote this five-layer architecture as (PAH/PSS)2/PAH (i.e., Au5+; 638

the number “5” refers to the total number of layers and the charge “+” refers to the charge of the outer layer).22 Then a functional terpolymer (F-HPMA) was attached by covalent LBL deposition34 to Au5+ yielding the final functional core/ shell MFNP nanocarrier (Figure 1A).35 The terpolymer (structure presented in Figure 1B; Mw ) 56400 g/mol, Mw/Mn ) 1.9, see Supporting Information for synthetic and characterization procedures) bears three different monomer repeat units, the majority (∼90%) being N-(2-hydroxypropyl)methacarylamide providing for a highly water solvated corona layer (or external layer according to Scheme 1), a prerequisite for a strong steric stabilization and a limited opsonization of MFNP by serum proteins. The second monomer unit is N-methacryloyl-glycyl-glycyl thiazolidine-2-thione (Ma-X-TT, X ) GG; ∼8%) to enable covalent attachment of the terpolymer to Au5+ through the aminolysis of thiazolidine-2-thione reactive groups (TT) by the primary amino groups present on the surface of the PAH corona layer. The third comonomer is N-methacryloyl-glycylDL-phenylalanyl-leucyl-glycyl doxorubicin (Ma-Y-Dox; ∼2%) which is well-known for incorporating a nontoxic pro-drug form of doxorubicin into copolymers and optimized for providing release of the active drug after endocytosis through the enzymatic degradation of the oligopeptide spacer (Y).36 The stability of (PSS/PAH)n-coated NP dispersions is further enhanced by the attachment of the F-HPMA layer. While (PSS/PAH)n-coated particles are stable over years in pure water, such dispersions were shown to aggregate when the ionic strength approaches physiological concentrations.27 The attachment of the F-HPMA corona layer proceeds without any perturbation of the dispersion stability as evidenced by transmission electron microscopy (TEM) micrographs (Figure 1C) and a small displacement of the surface plasmon resonance band of the gold cores (e.g., ∆λmax < 10 nm; Figure 1D). In contrast, the attachment of the terpolymer further stabilizes the particles and prevents their aggregation in simple phosphate-buffered saline (PBS) at pH 7.4 or even in PBS containing human serum albumin (HSA, 0.5 mg/mL).37 In order to quantify this stabilizing effect in the different media, we performed a statistical evaluation of the NP aggregation (the number of Au cores present in the aggregates observed) for a total of over 2000 Au cores (Supplementary Figure S1). This analysis revealed the need for a minimum number of five internal LBL layers (n ) 2) prior to the attachment of F-HPMA for keeping aggregation below 20%, for keeping the maximum number of cores per aggregate below three, and for keeping the aggregate size still below 50 nm, even if PBS is used as the dispersion medium. A further advantage of AuNP cores is that they can easily be dissolved by KCN which causes the surface plasmon band to completely vanish from the spectrum (Figure 1E), resulting in the formation of nonaggregated functional nanocapsules (Figure 1F).22 The absence of the plasmon peak in the nanocapsule suspension allows the direct determination of the doxorubicin concentration in the sample by UV-vis spectroscopy. After calibration with free and polymer-bound Nano Lett., Vol. 9, No. 2, 2009

Figure 1. (A) From left to right: 13 nm sized gold nanoparticles (AuNPs) as obtained after synthesis, stabilized by adsorbed sodium citrate; AuNPs coated with five primer layers of PAH and PSS (i.e., Au5+:Au/(PAH/PSS)2/PAH); Au5+ further coated with a external layer of F-HPMA (yielding MFNP). The red circles represent doxorubicin moieties (Dox) and are at scale with respect to the size of Dox molecules and the density of Dox moieties on the nanoparticle surface. (B) Chemical structures of the terpolymer and the polyelectrolytes used for the layer-by-layer deposition. The terpolymer possesses three repeating units: (i) about 90 mol % of the hydrophilic backbone component N-(2-hydroxypropyl)methacrylamide (HPMA), (ii) about 8 mol % of N-methacryloyl-glycyl-glycyl thiazolidine-2-thione (Ma-Gly-Gly-TT) for covalent LBL attachment, and (iii) about 2 mol % of N-methacryloyl-glycyl-DL-phenylalanyl-leucyl-glycyl doxorubicin (Ma-Gly-PheLeu-Gly-Dox) as active load in the form of a pro-drug. The internal compatibilization layers, assembled by LBL deposition, are composed of poly(allylamine hydrochloride) (PAH) and poly(styrenesulfonate) (PSS). (C) Transmission electron micrographs of Au/(PAH/PSS)2/ PAH (Au5+) dispersed in pure water, (ii) Au5+/F-HPMA (MFNP) dispersed in pure water, (iii) MFNP dispersed in phosphate-buffered saline pH 7.4 (PBS). The bottom right inset represents Au5+ and MFNP, respectively, dispersed in PBS in the presence of 0.5 mg/mL human serum albumin (HSA). (D) UV-vis spectra of native AuNPs dispersed in pure water (O, dotted line), Au5+ dispersed in pure water (0), MFNP dispersed in pure water (b) and in PBS (9); (b) and (9) are undistinguishable. All spectra were normalized to yield the same absorbance value at the wavelength of 440 nm. (E) UV-vis spectra of MFNP before (b) and after (9) the dissolution of Au cores by an excess of KCN. Insets are TEM micrographs of MFNP before and after the dissolution of the cores. (F) Overview TEM micrographs of four MFNPs after the etching of the Au core (dark centers are an artifact from uranyl acetate staining; it accumulates at the inner interface of the nanocapsule).

doxorubicin, we calculate from the optical absorbance of the nanocapsule suspension that there is an average of about 50 F-HPMA chains adsorbed per nanoparticle. Since every F-HPMA chain carries an average of 6.5 doxorubicin molecules, this corresponds to about 325 doxorubicin moieties per particle, the UV-vis spectra are shown in Figure 1F. Nano Lett., Vol. 9, No. 2, 2009

The fact that doxorubicin would be released from the MFNPs surface by enzymatic clevage of the oligopeptide spacer is less than trivial considering the high density of F-HPMA chains in the corona layer. We have already demonstrated that the F-HPMA layer likely prevents the adsorption of the nanoparticles by human serum albumine, an effect that could also hinder lysosomal enzymes (here 639

Figure 2. (A) Cathepsin B induced release of doxorubicin from MFNPs by specific cleavage of the tetrapeptide (Y ) Gly-Phe-Leu-Gly) spacer between doxorubicin and the F-HPMA terpolymer backbone. The control with a Y ) Gly-Gly spacer is not cleaved. (B) Optical micrographs of TPA differentiated THP-1 leukemia monocytes after 72 h of incubation without nanoparticles (images on the left “THP-1” depicts the control), incubated with MFNPs (center images demonstrate the stealthiness), and incubated with Au5+ (images on the right show the strong uptake of particles without F-HPMA layer).

cathepsin B) from reaching their oligopeptide target. The accessibility of the glycine-phenylalanine-leucine-glycine (GFLG) spacer and its cleavability by cathepsin B was studied by high-performance liquid chromatography (HPLC) using fluorescence detection. The results of the doxorubicin release from different MFNPs (e.g., Y ) GFLG vs Y ) GG) are reported in Figure 2A. We observed that about 50% of bound doxorubicin was released from MFNPs after 36 h of incubation with cathepsin B. In these “static” enzymatic experiments, the enzyme is only added at time t ) 0 and the loss of the proteolytic activity of cathepsin B as a function of time was not compensated. Such a reduction of proteolytic activity never occurs after the endocytosis of nanoparticles by cells in vitro or in vivo and the values reported here can therefore be taken as minimum release values. The fact that the release of doxorubicin is entirely due to the specific cleavage by cathepsin B and not to unspecific hydrolysis was corroborated by comparison with nanoparticles coated with a corona layer in which the doxorubicin was attached to the terpolymer backbone via a noncleaVable Gly-Gly spacer (Y ) GG). In this case the doxorubicin release with time was below detection (Figure 2A). An extremely critical issue for engineering potentially long circulating drug carriers in vivo depends on whether the F-HPMA corona layer would render the nanoparticles “stealthy”. In this study we used TPA-differentiated THP-1 monocytes as model macrophages for testing stealthiness. Au5+ and MFNPs were incubated in vitro for 72 h in the presence of THP-1 monocytes (Figure 2B). Again we took benefit from the strong extinction value of the plasmon band of the gold cores to monitor the phagocytosis of the two different nanoparticles by the THP-1 cells. After the full incubation time (e.g., 72 h), macrophages incubated with MFNPs exhibited the same pale color as macrophages that were kept in the absence of nanoparticles (Figure 2B, middle). In contrast, macrophages incubated with Au5+ turned dark pink/purple after less than 6 h of exposure 640

to nanoparticles (Figure 2B, right). While this coloration in the presence of Au5+ particles does not permit distinguishing if the particles are merely adsorbed to the surface of the macrophages or if they are truly internalized, the absence of this coloration in the cases of MFNPs (independently of n) is a clear proof that these nanoparticles are indeed stealthy: they are neither adsorbed to the macrophage surface nor internalized. In addition, it still needs to be ruled out that the presence of highly cytotoxic MFNPs nanoparticles (as they can potentially deliver the doxorubicin) harms or kills the THP-1 cells which may also prevent them from becoming colored. This proof was simply established by adding Au5+ particles to the macrophages after a previous incubation with MFNPs nanoparticles. The observation of a strong coloration of the THP-1 cells after addition of the Au5+ particles clearly demonstrates that they were not significantly affected by the cytotoxic particles even after incubation times longer than 3 days. (e.g., MFNPs were not internalized by macrophages because of their stealthiness). Over the last years, layer-by-layer assembly has established itself as a versatile tool for the fabrication of multifunctional nanosized coatings. For films on flat surfaces, we see a bright future with respect to the “wet bench” fabrication of thin film devices such as touch sensors38 or in the field of biomaterials39 even with respect to programmability of architectures for responsive designs.40 Mo¨hwald and colleagues have pushed the use of the LBL concept toward the coating of micrometer-sized colloids;41 we have now fully unleashed the LBL toolbox for use with nanoparticles. In the present paper we show that several important demands for nanocarrier systems, which are widely thought to constitute the next generation of drugs, can be engineered in a straightforward way. The example we provide here is intended for a potential application in cancer treatment, but the modular concept behind LBL deposition allows to easily extend our approach to other therapies. Most importantly, Nano Lett., Vol. 9, No. 2, 2009

we show that the stealthing of nano-objects in a layer-bylayer fashion is achieved with a copolymer of HPMA, which opens an entirely new possibility to engineer different multifunctional nanoparticles for obtaining highly stable dispersions with an “a` la carte” functionality. This concept makes our new nanocarrier system a likely candidate for testing EPR targeting which requires very good control of particle sizes. We are aware that such experiments with animals require substantial quantities of nanoparticles; however, recent progress in nanoparticle separation technologies42 have increased the expectation for being able to scaleup our process. Experimental Section. Reagents, instrumentation, and synthetic and characterization procedures are available in the Supporting Information. Synthesis of Au/(PAH/PSS)2/PAH (e.g., Au5+) and MFNPs. The procedure yielding Au5+ was inspired from our previous protocol22 with slight modification in order to reach a maximal amount of isolated functional nanoparticles (e.g., one Au core per colloidal aggregate).27 Due to strong adsorption of PAH-coated AuNPs (e.g., Au5+) on glass during the covalent attachment of F-HPMA, reaction flasks composed of standard glass were preliminary coated with poly(ethyleneimine) (PEI, BASF-Lupasol). For that, glass flasks were filled with a mixture HCl/MeOH (1:1) (30 min) and replaced by pure sulfuric acid (30 min). They were then abundantly rinsed with ultrapure water and filled with an aqueous solution of poly(ethyleneimine) (2.5 mg/mL). PEI adsorption was carried for 30 min. The PEI solution was then removed, and flasks were intensively rinsed with ultrapure water and were dried at room temperature. A fresh aqueous stock solution of F-HPMA at 1.0 mg/mL was prepared 10 min before use. A 992 µL portion of this solution was mixed with ultrapure water in order to reach a final volume of 18 mL. Au5+ was injected in order to reach a final absorbance in the 18 mL mixture of 0.43 (this corresponds to approximately 500 copolymer chains of F-HPMA per particles, e.g., ∼15000 reactive TT groups per particles). Precisely 10 s after the injection of Au5+ into F-HPMA, 600 µL of H3BO3 (46.52 mg/mL) and 600 µL of Na2B4O7·10H2O (69.44 mg/mL) were simultaneously added. The F-HPMA adsorption reaction was carried out for 6 h at room temperature, in the dark. After 6 h, 6.6 µL of an aqueous solution of 1-aminopropan-2-ol (1 vol %) is added to aminolyze unreacted TT groups. The mixture was allowed to react for an additional hour and was then distributed in some 1.5 mL Eppendorf Safe Lock Tubes and centrifuged for 2 h at 16000g and 25 °C. The supernatant was removed and replaced by ultrapure water. The centrifugation/redispersion cycle was reiterated once or twice (twice for analytical determination of the number of copolymer chains and doxorubicin per particles). After centrifugation, liquid pellets of MFNPs were collected into a glass vial (e.g., MFNPs stock solution), and the concentration in MFNPs was calculated by the measurement of the absorbance at 440 nm. This revealed a concentration in particles in the MFNP stock solution of 51.8 nmol/L. Nano Lett., Vol. 9, No. 2, 2009

Quantification of Doxorubicin-Loading per Nanoparticle. Two hundred microliters of concentrated stock solution of MFNP (centrifuged/redispersed three times) was mixed under vigorous stirring with 100 µL of KCN (10.0 mg/mL) and 100 µL of ultrapure water for 2 h to completely dissolve the gold nanocores. UV-vis spectroscopy revealed 50.5 FHPMA chains per MFNP after a calibration with a stock solution of F-HPMA (Figure 1E). Enzymatic Doxorubicin-Release Profiles. The enzymatic release of doxorubicin from Gly-DL-PheLeuGly-Dox side chains on MFNPs was assayed with cathepsin B. Five hundred milliliters of a buffer solution pH ) 6.0 containing 250 mL of 0.1 N KH2PO4 (3.2 g of KH2PO4), 14.1 mL of 0.1 M NaOH, 146 mg of EDTA, and 235.9 mL of ultrapure water (solution A) was prepared. Thirty milligrams of reduced glutathione was disolved in 10 mL of solution A (that is the solution B). In a quartz cuvette with a path length of 10 mm, 415 µL of solution A was mixed with 500 µL of the solution B. Then 60 µL of a solution of cathepsin B (2 mg in 500 µL) was added. The cuvette was incubated at 37 °C for 5 min. At this step, reduced glutathione activate the enzyme. Twenty-five microliters of a solution containing the substrate (13.93 mg of NR-benzoyl-L-arginine 4-nitroanilide dissolved in 346.4 mg of DMSO) is added in the cuvette, and the mixture is rapidly mixed and the absorbance at 410 nm was measured (A410nm ) 0.104). After 10 min of incubation at 37 °C, the absorbance at 410 nm was measured again (A410nm ) 1.122). A ∆A410nm of 0.330 for a cuvette of 10 mm path length over a time course of 10 min corresponds to a concentration of 2 × 10-7 mol/L in active enzyme. A 415 µL portion of solution A, 500 µL of solution B, and 20 µL of Cathepsin B (2 × 10-7 mol/L) were incubated for 5 min at 37 °C. Then 100 µL of the MFNPs stock solution was added. The enzymatic degradation was carried out on MFNP, and the released doxorubicin was analyzed by HPLC equipped with TSK SGX C18 column and fluorescence detector FluoroMonitor 4100 at an excitation wavelength of 480 nm and an emission wavelength of 560 nm after 12, 26, and 36 h. Before the analysis, the enzymatic reaction was stopped by freezing the mixture and colloids were removed by centrifugation. Only the supernatant was injected in the HPLC. In Vitro Incubation of Au5+ and MFNPs with Macrophages. Human monocytic THP-1 cells (leukemia) were cultured into culture flask of 25 cm2 (Greiner, ref 690 195) in RPMI 1640 (Rosewell Parc Memorial Institute) containing 10 mM HEPES, 14 mM glucose, 1 mM sodium pyruvate, 10% fetal calf serum, 40 µg/mL gentamicine (stock solution 200 mg/mL), and 50 µM 2-mercaptoethanol (stock solution at 100 mM). Every 2 days, cells were counted using a numeration Bu¨rker cell and were diluted in supplemented RMPI in order to reach 200000 cells/mL. After 2 days of culture, cell concentration reaches generally 700000 cells/ mL. For their differentiation in macrophages by TPA (e.g., phorbol ester 12-O tetra decanoyl phorbol 13 acetate), 1 mL of RPMI (300000 cells) supplemented with 5 nM TPA was distributed into a 24-well plate (Corning Inc. 3526, Corning, NY) in order to obtain differentiated THP-1 cells (typically 1.5 µL of a TPA solution at 1.0 mg/mL in PBS is injected 641

in 15 mL of RPMI). The cells were incubated at 37 °C under a 5% CO2-humidified atmosphere for 72 h. Then the medium containing TPA was replaced by 500 µL of fresh RPMI medium. In 700 µL of RPMI, 14 µL of MFNPs or Au5+ stock solution (both ∼50 nM in particles) was added and mixed. This 700 µL sample was then filtered under a sterile 0.22 µm Millex low binding protein filter (Millipore). Five hundred microliters of this medium was then injected in each well containing macrophages. Every 10 min, over a time laps of 72 h, optical photographs were taken from the wells with an inverted microscope equipped with a digital color camera (Princeton, CoolSnap FX). The microscope used was a LEICA DMIRE2 used in the bright field image mode, equipped with a halogen light (12 V, 100 W), a shutter, an objective LEICA N PLAN 10×/0.25, a motored plate Merzhauser equipped with a temperature/CO2-regulated box (CTI, ref IS-M-582) controlled by electronic devices (CTIController 3700 Digital for CO2, TempControl 37-2 Digital for the temperature). After 72 h, the medium was replaced by 1 mL of fresh RPMI and pictures were taken (Figure 2). In wells, where NPs were not internalized by macrophages (e.g., MFNPs), 500 µL of RPMI was replaced by 500 µL of RPMI containing 10 µL of Au5+ (∼50 nM). Time laps imaging was then continued for 72 h revealing the further phagocytotic activity of the macrophages toward Au5+. Acknowledgment. The following support is gratefully acknowledged: Marie Curie Training Site Fellowship, Contract No. HPMT-CT-2001-00396 (G.F.S.), the C.N.R.S., the Ministe`re de l’Education Nationale de la Recherche et de la Technologie, and a grant from the Academy of Sciences of the Czech Republic, No. KAN 200200651. We are grateful to the interdisciplinary collaboration and the invaluable support of Betty Heller from the Institut de Biologie Mole´culaire et Cellulaire, supervisor of the Cell Culture Department (IGBMC, Illkirch, France) and for the technical support from Jean Luc Vonesch, supervisor of the imaging department from the IGBMC, Didier Hentsch and Marcel Boeglin for time-laps imaging. Annette Thierry and Marc Schmutz are acknowledged for their invaluable support with transmission electron microscopy. The authors thank Dr. Aurore Schneider for helpful discussions. Supporting Information Available: Description of materials and methods, figures showing UV-vis spectra, plasmon bands, and TEM micrographs. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Ferrari, M. Nat. ReV. Cancer 2005, 5, 161–171. (2) Peer, D.; Karp, J. M.; Hong, S.; FaroKhzad, O. C.; Margalit, R.; Langer, R. Nat. Nanotechnol. 2007, 2, 751–760. (3) Torchilin, V. P. AdV. Drug DeliVery ReV. 2006, 58, 1532–1555. (4) Rapoport, N.; Gao, Z. G.; Kennedy, A. J. Natl. Cancer Inst. 2007, 99, 1095–1106. (5) Huang, X. H.; Jain, P. K.; El-Sayed, I. H.; El-Sayed, M. A. Nanomedicine 2007, 2, 681–693. (6) Gao, X.; Cui, Y.; Levenson, R. M.; Chung, L. W. K.; Nie, S. Nat. Biotechnol. 2004, 22, 969–976.

642

(7) Alivisatos, A. P. Nat. Biotechnol. 2004, 22, 47–52. (8) Michalet, X.; Pinaud, F. F.; Bentolila, L. A.; Tsay, J. M.; Doose, S.; Li, J. J.; Sundaresan, G.; Wu, A. M.; Gambhir, S. S.; Weiss, S. Science 2005, 307, 538–544. (9) Medintz, I. L.; Uyeda, H. T.; Goldman, E. R.; Mattoussi, H. Nat. Mater. 2005, 4, 435–446. (10) Duguet, E.; Vasseur, S.; Mornet, S.; Devoisselle, J. M. Nanomedicine 2006, 1, 157–168. (11) Weissleder, R.; Kelly, K.; Sun, E. Y.; Shtatland, T.; Josephson, L. Nat. Biotechnol. 2005, 23, 1418–1423. (12) Gu, F. X.; Karnik, R.; Wang, A. Z.; Alexis, F.; Levy-Nissenbaum, E.; Hong, S.; Langer, R. S.; Farokhzad, O. C. Nano Today 2007, 2, 14–21. (13) Lasser, A. Hum. Pathol. 1983, 14, 108–126. (14) Decher, G. Science 1997, 277, 1232–1237. (15) Caruso, F.; Caruso, R. A.; Mohwald, H. Science 1998, 282, 1111– 1114. (16) Donath, E.; Sukhorukov, G.; Caruso, F.; Davis, D. P.; Mohwald, H. Angew. Chem., Int. Ed. 1998, 37, 2201–2205. (17) Jessel, N.; Lavalle, P.; Ball, V.; Ogier, J.; Senger, B.; Picart, C.; Schaaf, P.; Voegel, J.-C.; Decher, G. In Elements of Macromolecular Structural Control; Gnanou, Y., Leibler, L., Matyiaszewski, K., Eds.; WileyVCH: Weinheim, 2007; Vol. 2, pp 1249-1306. (18) Maeda, H.; Wu, J.; Sawa, T.; Matsumura, Y.; Hori, K. J. Controlled Release 2000, 65, 271–284. (19) Duncan, R. Nat. ReV. Cancer 2006, 6, 688–701. (20) Kopecek, J.; Kopeckova, P.; Minko, T.; Lu, Z. R.; Peterson, C. M. J. Controlled Release 2001, 74, 147–158. (21) Ulbrich, K.; Sˇubr, V.; Strohalm, J.; Plocova´, D.; Jelı´nkova´, M.; Rı´hova´, B. J. Controlled Release 2000, 64, 63–79. (22) Schneider, G.; Decher, D. Nano Lett. 2004, 4, 1833–1839. (23) Schneider, G.; Decher, G.; Nerambourg, N.; Praho, R.; Werts, M. H. V.; Blanchard-Desce, M. Nano Lett. 2006, 6, 530–536. (24) Chanana, M.; Gliozzi, A.; Diaspro, A.; Chodnevskaja, I.; Huewel, S.; Moskalenko, V.; Ulrichs, K.; Galla, H. J.; Krol, S. Nano Lett. 2005, 5, 2605–2612. (25) Kim, J. S.; Rieter, W. J.; Taylor, K. M. L.; An, H.; Lin, W. L.; Lin, W. B. J. Am. Chem. Soc. 2007, 129, 8962–8963. (26) Wang, S. H.; Shi, X. Y.; Van Antwerp, M.; Cao, Z. Y.; Swanson, S. D.; Bi, X. D.; Baker, J. R. AdV. Funct. Mater. 2007, 17, 3043– 3050. (27) Schneider, G.; Decher, G. Langmuir 2008, 24, 1778–1789. (28) Gittins, D. I.; Caruso, F. AdV. Mater. 2000, 12, 1947–1949. (29) Sanvicens, N.; Marco, M. P. Trends Biotechnol. 2008, 26, 425–433. (30) Pan, D.; Turner, J. L.; Wooley, K. L. Chem. Commun. 2003, 2400– 2401. (31) Simberg, D.; Duza, T.; Park, J. H.; Essler, M.; Pilch, J.; Zhang, L. L.; Derfus, A. M.; Yang, M.; Hoffman, R. M.; Bhatia, S.; Sailor, M. J.; Ruoslahti, E. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 932–936. (32) Stevenson, M.; Hale, A. B. H.; Hale, S. J.; Green, N. K.; Black, G.; Fisher, K. D.; Ulbrich, K.; Fabra, A.; Seymour, L. W. Cancer Gene Ther. 2007, 14, 335–345. (33) Mammen, M.; Choi, S. K.; Whitesides, G. M. Angew. Chem., Int. Ed. 1998, 37, 2755–2794. (34) Multilayer thin films: Sequential assembly of nanocomposite materials; Decher, G., Schlenoff, J. B., Eds.; Wiley-VCH: Weinheim, 2003. (35) The synthesis and characterization of these macromolecules are described in the Supporting Information. (36) The approximate shell thickness revealed by TEM for Au5+ is 1.5 ( 0.5 nm and ∼5.0 ( 1.0 nm for MFNPs. (37) HSA was used as a model protein to monitor the serum protein adsorption. (38) Maheshwari, V.; Saraf, R. F. Science 2006, 312, 1501–1504. (39) Tang, Z. Y.; Wang, Y.; Podsiadlo, P.; Kotov, N. A. AdV. Mater. 2006, 18, 3203–3224. (40) Jessel, N.; Oulad-Abdeighani, M.; Meyer, F.; Lavalle, P.; Haikel, Y.; Schaaf, P.; Voegel, J. C. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 8618–8621. (41) Tong, W. J.; Gao, C. Y. J. Mater. Chem. 2008, 18, 3799–3812. (42) Sweeney, S. F.; Woehrle, G. H.; Hutchison, J. E. J. Am. Chem. Soc. 2006, 128, 3190–3197.

NL802990W

Nano Lett., Vol. 9, No. 2, 2009