Biomaterials 35 (2014) 1257e1266

Contents lists available at ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

Copper-free azideealkyne cycloaddition of targeting peptides to porous silicon nanoparticles for intracellular drug uptake Chang-Fang Wang a, Ermei M. Mäkilä a, b, Martti H. Kaasalainen b, Dongfei Liu a, Mirkka P. Sarparanta c, Anu J. Airaksinen c, Jarno J. Salonen b, Jouni T. Hirvonen a, Hélder A. Santos a, * a b c

Division of Pharmaceutical Technology, Faculty of Pharmacy, University of Helsinki, Viikinkaari 5 E (PO. Box 56), FI-00014 Helsinki, Finland Laboratory of Industrial Physics, Department of Physics and Astronomy, University of Turku, FI-20014 Turku, Finland Laboratory of Radiochemistry, Department of Chemistry, University of Helsinki, FI-00014 Helsinki, Finland

a r t i c l e i n f o

a b s t r a c t

Article history: Received 13 August 2013 Accepted 22 October 2013 Available online 7 November 2013

Porous silicon (PSi) has been demonstrated as a promising drug delivery vector for poorly water-soluble drugs. Here, a simple and efficient method based on copper-free click chemistry was used to introduce targeting moieties to PSi nanoparticles in order to enhance the intracellular uptake and tumor specific targeting hydrophobic drug delivery. Two RGD derivatives (RGDS and iRGD) with azide-terminated groups were conjugated to bicyclononyne-functionalized PSi nanoparticles via copper-free azideealkyne cycloaddition. The surface functionalization was performed in aqueous solution at 37
C for 30 min resulting in conjugation efficiencies of 15.2 and 3.4% (molar ratios) and the nanoparticle size increased from 165.6 nm to 179.6 and 188.8 nm for RGDS and iRGD, respectively. The peptides modification enhanced the cell uptake efficiency of PSi nanoparticles in EA.hy926 cells. PSi-RGDS and PSi-iRGD nanoparticles loaded with sorafenib showed a similar trend for the in vitro antiproliferation activity compared to sorafenib dissolved in dimethyl sulfoxide. Furthermore, sorafenib-loaded PSi-RGDS deliver the drug intracellulary efficiently due to the higher surface conjugation ratio, resulting in enhanced in vitro antiproliferation effect. Our results highlight the surface functionalization methodology for PSi nanoparticles applied here as a universal method to introduce functional moieties onto the surface of PSi nanoparticles and demonstrate their potential active targeting properties for anticancer drug delivery. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Porous silicon nanoparticle Copper-free click chemistry Surface modification Drug delivery Intracellular uptake Cancer therapy

1. Introduction Cancer is one of the most threatening diseases for humans [1]. Sustaining chronic proliferation is the most fundamental characteristic in cancer cells. Normal tissues carefully produce growthpromoting signals and regulate the cell formation, whereas cancer cells irregularly release the growth-promoting signals resulting in uncontrollable proliferation and development [2]. Chemotherapy combined with local therapies such as surgery or radiotherapy, has been applied as one of the strategies to treat cancer. However, non-specifically delivered and potent chemotherapeutic agents distributed in the whole body harm also the normal tissues. Thus, actively targeting delivery of the chemotherapeutic agents to the tumor tissue can increase the local drug

* Corresponding author. Tel.: þ358 9 191 59661; fax: þ358 9 191 59144. E-mail address: helder.santos@helsinki.fi (H.A. Santos). 0142-9612/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biomaterials.2013.10.065

concentration in the tumor and lower the systemic exposure, hence reducing the side effects of the drugs [3,4]. Neovascularization plays a crucial role for the development and formation of cancer, because cancer tissue, similarly to normal tissues, requires sustaining nutrients and oxygen as well as eliminating metabolic wastes and carbon dioxide. Tumor angiogenic vessels express biomarkers that are not present in the established blood vessels of normal tissues [3]. A number of cell-specific epitopes and biomarkers have been explored to show specific binding to certain antibodies, peptides or small molecules [4]. These differentially expressed biomarkers can be used as docking sites accumulating drug molecules and/or drug carriers at the tumor tissue e targeting the drug delivery. Integrins are the key type of regulators of angiogenesis [5]. anb3 integrin is the most abundantly expressed integrin by the neovascular endothelial cells during angiogenesis and tumor progression, but not in normal quiescent endothelial cells [6]. Besides integrin anb3, integrin anb5 interacts with vascular endothelial growth factor (VEGF) or transforming

1258

C.-F. Wang et al. / Biomaterials 35 (2014) 1257e1266

growth factor a to induce angiogenesis [5]. The three-amino acid peptide arginine-glycine-aspartic acid (RGD) has been identified as a ligand for integrins such as the anb3, anb5, and a5b1 [7,8]. RGDfunctionalized nanoparticles superiorly accumulated within tumorassociated blood vessels, but showed little binding to other vascular beds [9,10]. iRGD is a disulfide-based 9-amino acid cyclized peptide, identified by phage display as a tumor targeting and tissue penetrating peptide [11]. It can first associate with tumor cells by the specific affinity to anb3/5 integrins on tumor endothelium. Following association on the surface of the tumor cells, iRGD is cleaved between lysine (K)/arginine (R) and glycine (G) by proteolysis to produce a C-terminal motif, which can form neuropilin-1 mediated cell internalization [11]. Most of the chemotherapeutic agents are poorly water-soluble drugs which limits their application. Porous silicon (PSi) has a number of unique properties that render it as a potential drug delivery vehicle in biomedical application [12,13], such as increasing the solubility of poorly water-soluble drugs [14], high drug loading capacity [15] and tunable surface structure for controlled release properties by different surface chemistries [16]. For example, carboxylic acid- and amine-modified [17,18] surfaces of PSi can be used for further chemical functionalization [16]. Targeting moieties of E-selectin and folic acid have been incorporated into PSi microparticles for tissue targeting drug delivery based on ethyl(dimethylaminopropyl) carbodiimide/N-hydroxysuccinimide (EDC/NHS) coupling chemistry [19,20]. However, EDC/NHS coupling method is not very selective as it involves the reaction between an amine and carboxylic acid groups, which widely exist in peptides. This approach can thus distort the biologically active motifs presented in peptides during the conjugation to the nanovectors. The azideealkyne Huisgen cycloaddition click reaction has been reported as a simple method to couple organic molecules between the azide and alkyne chemical activities in high yields under mild conditions, with high selectivity in the presence of a diverse range of other functional groups [21]. Strain-promoted azideealkyne cycloaddition (SPAAC) click reaction, also called copper-free click reaction, relies on the release of ring-strain during the transition from a bent triple bond (bicyclononyne) to a double bond (triazole) [22]. SPAAC avoids using copper ion as a catalyst, which can reduce cell and further in vivo biological toxicity. SPAAC is therefore an important tool for surface modification of nanomaterials [23,24]. The antiangiogenic drug sorafenib has been employed in solid tumor malignancies [25] and in acute myelogenous leukemia [3,26]. It inhibits tumor cell proliferation and tumor angiogenesis by targeting RAf kinase, platelet-derived growth factor, VEGF receptor 2 & 3 kinases and the c-kit receptor [27]. However, the low solubility and the first-pass effect of sorafenib limit its therapeutic use in treatment [28]. Poorly dissolved sorafenib can also cause large adverse reactions [29,30]. To avoid the severe side effects, sorafenib encapsulated by nanoparticles such as solid lipid or amphiphilic polymer based nanoparticles have been investigated in preclinical studies for intravenous administration and passive targeting drug delivery [31,32]. Here, PSi nanoparticles were peptide-functionalized for actively targeting sorafenib to the tumor tissue and thus enhance the cellular uptake and drug delivery efficiency. In this study, RGD targeting derivative peptides, RGDS and iRGD were conjugated to 3-aminopropyltriethoxysilane-modified thermally carbonized PSi (APS-TCPSi) nanoparticles via SPAAC click reaction for targeted drug delivery to tumor neovasculature. SPAAC can be used for a highly selective and effective chemical conjugation to couple the targeting peptides to the PSi nanoparticles. Sorafenib was used to test the drug delivery efficiency of the PSi nanoparticles after peptide-functionalization. Endothelial EA.hy926 cells, a hybride

cell line resulting from the fusion of primary human umbilical vein cells (HUVEC) with adenocarcinomic human alveolar basal epithelial cells (A549) [33], was used as an in vitro cell model to test the drug delivery systems. Overall, we have developed targeting peptides functionalized PSi nanoparticles via the SPAAC click reaction for targeted drug delivery. 2. Materials and methods 2.1. Materials and cell culturing (1R,8S,9s)-bicyclo[6.1.0]non-4-yn-9-ylmethyl succinimidyl carbonate (BCNe NHS) was purchased from Synaffix (Nijmegen, The Netherlands). N-terminal azidoalanine-functionalized peptides RGDS and iRGD were customized from GenicBio (Shanghai, China). Sorafenib was obtained from LC Laboratories (Woburn, MA, USA). Dulbecco’s phosphate buffer saline (10  PBS) and Hank’s balanced salt solution (10  HBSS), hypoxanthineeaminopterinethymidine (HAT, 50) were purchased from GibcoÒ (Carlsbad, CA, USA). Dulbecco’s Modified Eagle’s Medium (DMEM), fetal bovine serum (FBS), trypsin (2.5%), sodium pyruvate, nonessential amino acids (NEAA, 100), L-glutamine (100), penicillin-streptomycin (100) were purchased from HyClone (Waltham, USA). CellTiter-GloÒ luminescent cell viability assay kit was purchased from Promega (Madison, USA). Endothelial EA.hy926 (ATCC) cells were incubated in DMEM contained supplements of 10% FBS, 4.5 g/L glucose, 1% sodium pyruvate, 1% NEAA, 1% L-glutamine 1% penicillin-streptomycin (100 IU/mL) and 2% of HAT in 75 cm2 flasks at 37
C with humidified atmosphere containing 5% CO2. 2.2. Preparation of TCPSi nanoparticles Multilayer PSi films were produced by electrochemically etching monocrystalline, pþ-type Si h100i wafers with a 0.01e0.02 U cm resistivity in a 1:1 (v/v) aqueous hydrofluoric acid (HF, 38%)eethanol electrolyte as described previously [18]. The free standing films were thermally carbonized with acetylene [14] after which the obtained TCPSi was immersed into HF to generate silanol termination for aminosilane (3-aminopropyl ethoxysilane) attachment following the previously described process using a 10 v-% APS-toluene solution [18]. The size reduction of the APS-TCPSi multilayer films to nanoparticles was done by wet milling using a 5 v-% APTES-toluene solution as the milling liquid [34]. After milling, the excess silane was removed by replacing the liquid and redispersing the nanoparticles to fresh toluene and ethanol at least three times using centrifugation. 2.3. Peptide-modified TCPSi nanoparticles via SPAAC Bicyclononyne-functionalized TCPSi (TCPSi-BCN) nanoparticles were prepared by dissolving 3 mg of BCNeNHS in 400 mL of dimethylformamide (DMF) with 2 mg of APS-TCPSi nanoparticles suspended in 200 mL 4-(2-hydroxyethyl)-1piperazineethanesulfonic acid (HEPES) buffer (0.1 M, pH 7.8) under vigorous mixing at room temperature. After 45 min, the TCPSi nanoparticles were collected from the reaction mixture by centrifugation (Sorvall RC 5B plus, Thermo Fisher Scientific, USA) at 10,000 relative centrifugal force (rcf) for 3 min and washed with 1 mL of DMF/water (60/40%, v/v), ethanol and water three times to obtain TCPSi-BCN resuspended in Milli-Q water. The conjugation of RGDS and iRGD to TCPSi-BCN nanoparticles via SPAAC was achieved by mixing azide-functionalized RGDS (0.5 mg) and iRGD (1 mg), respectively, to 1 mg of TCPSi-BCN nanoparticles suspended in 500 mL Milli-Q water. The reaction solution was mixed at 37
C for 30 min. The TCPSi nanoparticles were then harvested from the reaction mixture by centrifugation and washed with 1 mL ethanol and water three times to remove unreacted peptides. For confocal fluorescence imaging and flow cytometry analysis, the APS-TCPSi nanoparticles were first covalently labeled with fluorescein isothiocyanate isomer I (FITC). Briefly, 2 mg of APS-TCPSi nanoparticles were mixed with 0.1 mg of FITC in 400 mL ethanol and 100 mL 0.1 M HEPES (pH 7.8). After 30 min, the TCPSi nanoparticles were isolated from the reaction mixture and washed three times with ethanol to remove the unreacted FITC. The FITC-labeled TCPSi nanoparticles were continually reacted with BCNeNHS and the peptides as described above to obtain peptide-functionalized FITC-labeled TCPSi nanoparticles. 2.4. Physicochemical characterization of TCPSi nanoparticles The qualitative surface chemical characterization of peptide-functionalized TCPSi nanoparticles were performed by Fourier transform Infrared Spectroscopy (FTIR) with Vertex 70 FTIR spectrometer (Bruker Optics, USA) using a horizontal attenuated total reflectance (ATR) accessary (MIRacle, PIKE Technologies, USA). The transmittance spectra were recorded between 4000 and 650 cm1 with a 4 cm1 resolution, and the data collected using an OPUS 5.5 software (Bruker Optics Inc, MA, USA). The physical parameters of the nanoparticles were determined by nitrogen sorption at 196
C using TriStar 3000 (Micromeritics Inc., USA). The specific surface area of the APS-TCPSi nanoparticles was calculated using the Brunauer-EmmettTeller theory. The total pore volume was obtained as the total adsorbed amount at

C.-F. Wang et al. / Biomaterials 35 (2014) 1257e1266 a relative pressure p/p0 ¼ 0.97. The average pore diameter was calculated from the obtained surface area and pore volume by assuming the pores as cylindrical. The amount of peptides covalently conjugated to TCPSi nanoparticles were determined by elemental analysis using a Vario Micro cube CHN analyzer (Elementar Analysensysteme, GmbH, Germany). The percentage of carbon (C), hydrogen (H) and nitrogen (N) were recorded. The conjugation efficiency was calculated based on the percentage of each element and the chemical structure of each molecule. The size and zeta (z)-potential measurements of TCPSi nanoparticles were carried out by dynamic light scattering (DLS) using a Zetasizer Nano ZS (Malvern Instruments, UK) equipped with a HeeNe laser beam (633 nm, fixed scattering angle of 173
) at 25
C. 2.5. Cytotoxicity studies of TCPSi nanoparticles The cytotoxicity of the peptide-functionalized TCPSi nanoparticles was evaluated by CellTiter-GloÒ luminescent cell viability assay kit. EA.hy926 endothelial cells were cultured with cell medium without nanoparticles and with 1% of Triton X-100 as positive and negative controls, respectively. About 2  104 cells/well in 100 mL cell medium was seeded in 96-well plates (Corning Life Sciences, USA). After overnight incubation, cells were exposed to the nanoparticles with different concentrations (25, 50, 100, and 250 mg/mL). After 6 h incubation, the nanoparticles suspension was removed and the cells were washed once with 1  HBSS (pH 7.4). The number of viable cells was determined by the CellTiter-GloÒ assay modified from the supplier’s protocol. The luminescence was measured with a Varioskan Flash fluorometer (Thermo Fisher Scientific, USA). 2.6. Nanoparticles plasma association For intravenous drug administration, plasma protein can interact with the nanoparticles and influence the biofate of the nanoparticles. The different particle composition and surface chemistry will determine the nanoparticle interactions with plasma components [35]. The size and z-potential change of the TCPSi nanoparticles during exposure to plasma media were studied by incubation of 100 mg/mL of the TCPSi nanoparticles with human plasma at 37  1
C during 2 h. Anonymous donor plasma was supplied by the Finnish Red Cross Blood Service under the permission from the respective institutional ethical committee. Samples were withdrawn at different time points and diluted 10 times for DLS analysis. All the experiments were performed at least in triplicates. 2.7. Cell uptake of nanoparticles RGD-functionalized TCPSi nanoparticles can specifically interact with integrin anb3/5 on the endothelial cell membrane and enhance the cell uptake. First, the subcellular localization of non-fluorescently labeled TCPSi nanoparticles was evaluated by transmission electron microscopy (TEM). About 2 mL/well of 5  105 cells/ well of EA.hy926 cells were seeded in 6-well plates containing 18  18 mm cover slip (Menzel-Gläser, Braunschweig, Germany) in each well. After reaching 80% confluency, 100 mg/mL of each nanoparticle in media replaced the culture medium. After 3 h incubation, the nanoparticle suspensions were removed and the cells were washed three times with HBSS (pH 7.4). 1 mL/well of 2.5% glutaraldehyde was added to the cells and incubated at 37
C for 20 min for cell fixation, followed by washing with HBSS for three times. Ultrathin sections of both control and exposed cells to TCPSi nanoparticles were prepared as described elsewhere [36]. The images were obtained with a Jeol TEM (Jeol, JEM-1400, Tokyo, Japan) with voltage 80 kV and magnification between 250 and 10,000. Confocal fluorescence imaging and flow cytometry were used to qualitatively and quantitatively, respectively, evaluate the cell uptake of fluorescently labeled TCPSi nanoparticles. In the case of confocal fluorescence, 200 mL of 8  104 cells/well were seeded in Lab-TekÒ chamber slides (Thermo Fisher Scientific, USA). After reaching 80% confluency, the cell culture medium was replaced by 200 mL of 100 and 250 mg/mL of each nanoparticle in media. Cells cultured with media without nanoparticles were used as controls. After 3 h incubation, the nanoparticles suspension was removed and the cells were rinsed three times with HBSS (pH 7.4). Cell membrane was stained with CellMaskÔ Orange (Invitrogen, USA) according the manufacturer protocol, followed by incubation with 2.5% glutaraldehyde for 20 min at 37
C for cell fixation. Confocal images were taken with a Leica SP5 inverted confocal microscope (Leica Microsystems, Germany), equipped with argon (488 nm) and DPSS (561 nm) lasers and HCX Plan Apochromate 63/1.2-0.6 oil immersion objective. For flow cytometry, 1 mL of 2  105 cells/well was seeded in 12-well plates. After reaching 80% confluency, the culture medium was replaced with 50, 100 and 250 mg/ mL of each nanoparticle and incubated with the cells for 3 h. After removing the nanoparticle solutions and washing the cells three times with HBSS (pH 7.4) to remove non-adherent nanoparticles, the cells were harvested by trypsinethylenediaminetetraacetic acid and treated with trypan blue (0.04% v/v) to quench the fluorescence of possible surface adherent nanoparticles. Flow cytometry was performed with an LSR II flow cytometer (BD Biosciences, USA) with laser excitation wavelength of 488 nm using a FACSDiva software. 10,000 events were collected for each sample and the data were analyzed by Flowjo 7.6 software (Tree Star, Ashland, OR, USA).

1259

2.8. Drug loading and release The poorly water-soluble antiangiogenic drug sorafenib was used to test the drug loading and delivery efficiency of the TCPSi nanoparticles before and after peptide-functionalization. The TCPSi nanoparticles were immersed in 15 mg/mL of sorafenib dissolved in acetone solution with a ratio of 1 mg of TCPSi nanoparticle to 1 mL of drug solution, and stirred for 2 h at room temperature. The drug-loaded TCPSi nanoparticles were collected and washed once with Milli-Q water. The loading degree was determined by immersing 200 mg of the drug-loaded TCPSi nanoparticles to 1 mL acetonitrile/water mixture (42:58%, v/v) under vigorous stirring for 1 h. The amount of sorafenib released from the nanoparticles was determined by HPLC (Agilent 1260, Agilent Technologies, USA). The chromatographic separation was achieved using a Zorbax C18 (4.6  100 mm, 5 mm) column. The mobile phase was composed of 0.2% trifluoroacetic acid (pH 2) and acetonitrile at a ratio of 42:58% (v/v) with 1.0 mL/min flow rate and UV detector set at wavelength of 254 nm. The dissolution tests were performed at 37  1
C in sink conditions by immersing 200 mg of each TCPSi nanoparticles in 50 mL of buffer dissolution media using a shaking method and a shaking speed rate of 150 rpm. The release media used were pH 5.5 (2-(N-morpholino)ethanesulfonic acid; MES), pH 7.4 (HEPES), DMEM with 10% FBS, DMEM without FBS, pH 5.5 (MES) with 10% FBS and pH 7.4 (HEPES) with 10% FBS. The dissolution of sorafenib alone was done by adding about 200 mg of sorafenib to 150 mL of DMEM with or without 10% FBS. 200 mL of samples were withdrawn from each dissolution test at different time points. The collected samples were centrifuged at 16,100 rcf for 3 min and analyzed by HPLC using the method described above. 2.9. Antiproliferation activity The in vitro antiproliferation effect of sorafenib-loaded into peptidefunctionalized TCPSi nanoparticles were evaluated by cell proliferation experiments. Briefly, EA.hy926 endothelial cells were seeded in 96-well plates at the density of 1  104 cells/well and allowed to attach overnight. Then, the cell culture medium was replaced by 100 mL of media containing different concentrations of sorafenib (1, 2, 5, 7.5, 10. 15 and 20 mM), TCPSi nanoparticles (10, 20, 50, 100 and 150 mg/mL), drug-loaded nanoparticles (with nanoparticles concentration of 10, 20, 50, 100 and 150 mg/mL) for 24 h. The cell amount was determined by the CellTiterGloÒ luminescence cell viability assay kit. Each experiment was performed at least in triplicate. 2.10. Statistical analyses Results of the assays are expressed as mean  s.d. of at least three independent experiments. Student’s t-test was used to evaluate the significant differences with probabilities set of *p < 0.05, **p < 0.01 and ***p < 0.005 using Origin 8.6 (OriginLab Corp., USA).

3. Results and discussion 3.1. Characterization of the modified TCPSi nanoparticles The size of APS-TCPSi nanoparticles prepared for the studies was 165.6  1.1 nm. Nitrogen sorption results indicated that the nanoparticles retained a porous structure according to the shape of the isotherm. The calculated specific surface area was 245  13 m2/g with a pore volume of 0.70  0.01 cm3/g and an average pore diameter of 11.5  0.5 nm. The zeta (z)-potential of the APS-TCPSi obtained from electrophoretic mobility measurements was 35.1  0.8 mV. The conjugation of the azide-terminated peptides onto the surface of the APS-TCPSi nanoparticles was achieved by two steps (Scheme 1). Firstly, bicyclononyne (BCN) was covalently attached onto the APS-TCPSi nanoparticles by the reaction between BCNe NHS and the surface amine groups of the APS-TCPSi nanoparticles. Secondly, the nanoparticles were functionalized with RGD peptides in aqueous solution at 37
C by SPAAC click reaction of the TCPSiBCN and the azide-terminated RGD derivative peptides. Fig. 1 shows the attenuated total reflectance (ATR)-FTIR spectra of APS-TCPSi, TCPSi-BCN, and peptide-functionalized TCPSi nanoparticles (TCPSi-RGDS and TCPSi-iRGD). The ATR-FTIR spectra of TCPSi-BCN exhibited two characteristic bands corresponding to n(NC ¼ O) (amide I) and d(CNeH) (amide II) vibrations at 1693 and 1525 cm1, respectively, indicating the formation of the carbamate bond between the APS-TCPSi nanoparticles and the BCN moiety.

1260

C.-F. Wang et al. / Biomaterials 35 (2014) 1257e1266

Scheme 1. Peptides (RGDS and iRGD) conjugated to APS-TCPSi nanoparticles via the SPAAC reaction. In the first step BCNeNHS reacts with the amine group of APS-TCPSi nanoparticles at pH 7.8 and at room temperature (RT), and in the second step the azide-terminated peptides are added to the nanoparticle solution and allowed to react at 37
C for 30 min to form the peptide-functionalized TCPSi nanoparticles. The molecular structure of sorafenib used in this study is also presented.

The ATR-FTIR spectra of the TCPSi-RGDS and TCPSi-iRGD nanoparticles showed bands at 1665 cm1 (amide I, NC ¼ O) and 1529 cm1 (amide II, N(H)eC(O)), shifting to 1693 and 1525 cm1, respectively, compared to the ATR-FTIR spectra of initial TCPSi-BCN. This is due to the contribution of the peptides structure. Control experiments were conducted using azide-functionalized peptides physically mixed with APS-TCPSi nanoparticles (without bicyclononyne moiety) in similar conditions as the reaction of peptides conjugated to the PSi nanoparticles. The ATR-FTIR spectrum of this sample showed the same profile as the initial APS-TCPSi (Fig. S1). These confirmed the success of the covalently conjugation of the peptides onto the APS-TCPSi nanoparticles surface via SPAAC. Elemental analysis was also employed to confirm the successful surface peptide-modification onto the APS-TCPSi nanoparticles (Table S1). The quantitative ratio of the peptides conjugation efficiency to the surface of the APS-TCPSi nanoparticles was calculated based on the percentage of each element in the samples. The results showed that on average, ca. 15.2% and 3.4% (molar ratios) of the amine functional groups of the APS-TCPSi nanoparticles were conjugated to the peptides RGDS and iRGD, respectively. RGDS is a linear 4-amino acid peptide, whereas iRGD is a cyclized 9-amino acid peptide. We hypothesize that the structural ring molecules of iRGD possibly covered a larger surface area of the TCPSi

nanoparticles after a certain amount of iRGD was attached to the surface of the nanoparticles, and thus, hampered subsequent attachment of peptides to the close vicinity of the previously attached ones due to steric hindrance. It has been reported that for polymersome post-route surface modification ca. 5% of the outside surface functional groups were accessible to be functionalized with PEGylated molecules using the SPAAC click reaction due to surface steric hindrance [37]. Remarkably, in our study ca. 15.4% of the amine groups of the TCPSi nanoparticles were conjugated with RGDS, demonstrating the successfully conjugation of the peptides onto the TCPSi nanoparticles surface. In order to further use the PSi nanoparticles for imaging experiments, APS-TCPSi nanoparticles were also covalently labeled with FITC, followed by the conjugation of the peptides (RGDS and iRGD) to the FITC-labeled TCPSi nanoparticles via SPAAC. ATR-FTIR results showed that RGDS and iRGD were also successfully conjugated to the FITC-labeled APS-TCPSi nanoparticles (Fig. S2). Table 1 presents the size and the z-potential of the TCPSi nanoparticles before and after peptide-functionalization determined by DLS. The average size of the PSi nanoparticles was increased from 166.6 nm (APS-TCPSi) to 179.1 nm (TCPSi-RGDS) and 188.8 nm (TCPSi-iRGD) after peptide-functionalization. The polydispersity index (PDI) of the APS-TCPSi (0.081) and peptidefunctionalized TCPSi nanoparticles (0.082 and 0.120 for TCPSiRGDS and TCPSi-iRGD, respectively) indicated very monodisperse TCPSi nanoparticle dispersions. The z-potential of the PSi nanoparticles changed from þ35.1 mV (APS-TCPSi) to þ11.3 mV (TCPSiRGDS) and þ11.2 mV (TCPSi-iRGD). The reduction in the z-potential can be attributed to the neutral charge of the peptides, which corroborates with the success of the peptides conjugation onto the surface of the TCPSi nanoparticles. Despite of the changed z-potential values, no evidence of agglomeration was observed.

Table 1 Physiochemical characterization of the size, z-potential and drug loading degree of APS-TCPSi, TCPSi-RGDS and TCPSi-iRGD. Data represent the mean  s.d. of at least three independent measurements.

Fig. 1. ATR-FTIR of APS-TCPSi (a), TCPSi-BCN (b), TCPSi-RGDS (c) and TCPSi-iRGD (d). Arrows indicate the bands corresponding to the amide I and amide II.

Nanoparticle

Size (nm)

PDI

z-potential (mV)

Loading degree (w-%) of sorafenib

APS-TCPSi TCPSi-RGDS TCPSi-iRGD

165.6  1.1 179.6  1.2 188.8  6.9

0.08  0.02 0.08  0.01 0.13  0.04

35.1  0.8 11.3  0.4 11.2  0.8

6.06  0.65 6.12  0.87 5.64  0.56

C.-F. Wang et al. / Biomaterials 35 (2014) 1257e1266

3.2. In vitro cell viability The charge, size, and surface chemical composition of the PSi nanoparticles have critical influence on the cytotoxicity of the nanoparticles [38,39]. In order to confirm the potential application of the aforementioned TCPSi nanoparticles as drug delivery vehicles, the cytotoxicity of the TCPSi nanoparticles before and after peptide-modification was tested in the EA.hy926 endothelial cell line. The results showed that none of the three nanoparticles (APSTCPSi, TCPSi-RGDS and TCPSi-iRGD) caused significant cytotoxicity at particle concentrations up to 250 mg/mL (Fig. 2). 3.3. Human plasma association studies The first effect of biological defense barrier that a drug delivery nanovector encounters after intravenous administration is the plasma proteins, which by association with the nanoparticles will determine their biofate. Nanoparticle properties, such as high surface area and porous structure, surface charge, surface composition, size and shape all play important roles in the plasma protein interactions [40]. Thus, the surface charge of the vector and opsonization of the nanoparticles will affect the drug delivery efficiency. Herein, the nanoparticleeplasma interactions were tested by incubating the PSi nanoparticles with human plasma up to 2 h at 37
C (Fig. 3a). The size for the APS-TCPSi nanoparticles was increased from ca. 170 to ca. 320 nm during the first 10 min incubation with human plasma, indicating certain degree of protein adsorption onto the nanoparticles in human plasma and consequent nanoparticle aggregation. After that, the size of the nanoparticles was decreased continuously from ca. 320 to ca. 260 nm at 2 h. This decrease in the particle size could be explained by the initial burst in the protein adsorption onto the nanoparticle’s surface, followed by a slow detachment of the outermost layer(s) of the adsorbed proteins weakly interacting with the hard protein corona layer [40], and thus, slightly inducing nanoparticle disaggregation over the time.

1261

Similarly, the size of TCPSi-RGDS and TCPSi-iRGD were both increased to ca. 340 and 370 nm, respectively, during the first 10 min incubation with human plasma. Furthermore, a similar decrease as for the APS-TCPSi nanoparticles was observed for the peptide-functionalized TCPSi nanoparticles. The PDI values of the nanoparticles after incubating with human plasma showed an increase due to the protein adsorption onto the nanoparticles’ surface. No significant differences were observed between the APSTCPSi and the peptide-functionalized TCPSi nanoparticles. The surface z-potential of all the three TCPSi nanoparticles was decreased after the protein association from positive values (Table 1) to ca. 14 mV and kept constant during the incubation with human plasma (Fig. 3b). Similar results have also been reported elsewhere with PSi-based nanoparticles with other different surface properties [41,42]. The nanoparticle surface chemistry drives selective adsorption of plasma components, which impacts the cellular interactions and nanoparticle’s biodistribution. It has been demonstrated that cationic, but not anionic or neutral liposomes, as well as PSi-based nanoparticles associated with the luminal surface of tumor endothelial cells, are subsequently internalized [35,43]. Our results showed that the peptide-functionalization to the cationic APS-TCPSi nanoparticles did not significantly influence the non-specific plasma opsonization. 3.4. Intracellular TCPSi nanoparticle distribution To determine the intracellular distribution of APS-TCPSi, TCPSiRGDS and TCPSi-iRGD nanoparticles, they were incubated with the EA.hy926 endothelial cells for 3 h at 37
C. Fig. 4 shows the TEM images of the morphology of the cells and the intracellular distribution of both the APS-TCPSi and the peptide-functionalized TCPSi nanoparticles. In general, the TCPSi nanoparticles were uptake and distributed inside the cells, particularly in the cytosol rather than in the nucleus. It can be seen that the TCPSi-RGDS and TCPSi-iRGD accumulated more extensively inside the cells than the APS-TCPSi nanoparticles. As a result of the peptide-functionalization, an

Fig. 2. In vitro cytotoxicity of APS-TCPSi, TCPSi-RGDS and TCPSi-iRGD nanoparticles after incubation with EA.hy926 endothelial cells for 6 h at 37
C. The viability was determined by an ATP-based CellTiter-GloÒ luminescence assay. Error bars represent mean  s.d. (n  3).

1262

C.-F. Wang et al. / Biomaterials 35 (2014) 1257e1266

Fig. 3. Effect on the average size (a, black line indicates the size and the grey line indicates the PDI) and z-potential (b) of the TCPSi nanoparticles after incubation with human plasma up to 120 min at 37
C. Error bars represent mean  s.d. (n  3).

enhanced intracellular uptake was observed for the TCPSi nanoparticles. 3.5. TCPSi nanoparticle intracellular uptake In order to accurately determine the improvement of cell uptake efficiency by the targeting peptide-functionalization of TCPSi nanoparticles, the TCPSi nanoparticles were first covalently labeled with FITC and then conjugated with targeting peptides (RGDS and iRGD) via SPAAC. The cellular internalization of TCPSi nanoparticles was examined by confocal fluorescence microscopy and flow cytometry. Fig. 5a shows the z-stack confocal fluorescence microscopy scanning of the EA.hy926 cells incubated with FITC-labeled TCPSi nanoparticles. At higher particle concentrations all the three studied TCPSi nanoparticles presented strong fluorescent signals in the cells. For each concentration, both peptidefunctionalized TCPSi nanoparticles were uptake more efficiently than the APS-TCPSi nanoparticles. Furthermore, cells treated with TCPSi-RGDS had clearly more nanoparticles located inside the cells than the cells treated with the TCPSi-iRGD nanoparticles at concentrations of 100 and 250 mg/mL. The ability of the peptide-functionalized PSi nanoparticles to enhance the cell internalization was also quantified by flow cytometry (Fig. 5b and c). Similarly to the results of the confocal

fluorescence microscopy, all the three PSi nanoparticles showed concentration dependent cell uptake. The peptides functionalization of the TCPSi nanoparticles was found to significantly enhance the cellular uptake of the APS-TCPSi nanoparticles. TCPSi-RGDS nanoparticles had significantly highest nanoparticle internalization (ca. 73% of treated cells associated with the nanoparticles), followed by the TCPSi-iRGD with ca. 62% of the treated cells associated with the PSi nanoparticles. At the concentration of 250 mg/ mL, APS-TCPSi associated the least (ca. 51%) to the cells. A similar trend was also observed in the cells treated with 100 and 50 mg/mL of APS-TCPSi nanoparticles. The aforementioned results demonstrated that both the peptide-functionalized TCPSi nanoparticles (TCPSi-RGDS and TCPSi-iRGD) were superior for endothelial cellular internalization compared to the APS-TCPSi nanoparticles, despite of the lower positive surface charge (z-potential) of the TCPSi-RGDS and TCPSi-iRGD nanoparticles. Considering that the nanoparticle’s positive surface charge is beneficial for the non-specific cell internalization of nanoparticles [43,44], the peptide-modified PSi nanoparticles with lower positive surface charge showed better cellular uptake, which implies that the properties of the peptides enhanced the specific cellular internalization further of the APS-TCPSi nanoparticles. Interestingly, TCPSi-RGDS nanoparticles had higher cellular uptake efficiency than TCPSi-iRGD, even though iRGD has been reported to

Fig. 4. Intracellular uptake and distribution of APS-TCPSi, TCPSi-RGDS and TCPSi-iRGD nanoparticles. TEM images of ultrathin sections of EA.hy926 endothelial cells as control or incubated with APS-TCPSi, TCPSi-RGDS, and TCPSi-iRGD nanoparticles at 37
C (scale bars are 5 and 2 mm, and 200 nm, respectively, from top to down).

C.-F. Wang et al. / Biomaterials 35 (2014) 1257e1266

1263

Fig. 5. (a) Confocal fluorescence microscopy images of EA.hy926 endothelial cells incubated for 3 h with APS-TCPSi, TCPSi-RGDS and TCPSi-iRGD nanoparticles, showing the maximum intensity projections of z-stacks of TCPSi nanoparticles’ internalization. The APS-TCPSi nanoparticles were covalently labeled with FITC (green color) and the cell membrane was stained with CellMaskÔ (orange color). Scale bars are 10 mm. (b) Flow cytometry histogram graphics of the PSi nanoparticles at concentrations of 250 mg/mL. (c) Flow cytometry and quantitative determination for the internalized APS-TCPSi, TCPSi-RGDS and TCPSi-iRGD nanoparticles. Cells were treated with trypan blue (0.04% v/v) to quench their surface fluorescence and discriminate between celleparticle association and nanoparticle internalization. 10,000 events were evaluated for each measurement. Error bars represent s.d. (n  2). The level of the significant differences of the treated PSi nanoparticles was set at probabilities of *p < 0.05, **p < 0.01 and ***p < 0.005 (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.).

improve cell targeting and tissue penetrating properties [11]. The reason might be due to the differences in the efficiency of the surface modification for the two peptides in our case. The linear 4amino acid peptide RGDS had much higher conjugation efficiency (15.4%) than the cyclized 9-amino acid peptide iRGD (3.4%), which is also possibly reflected in the lower cellular internalization of the latter peptide-modified TCPSi nanoparticles. 3.6. Drug loading and release Sorafenib (Scheme 1) is an antiangiogenic drug currently investigated for treatment of renal and lung cancers. However, it is a poor water-soluble drug that causes severe side effects which limits its clinical application [28e30]. In order to achieve the tissue specific delivery of sorafenib and increase the drug delivery efficacy, targeting peptide-modified TCPSi nanoparticles loaded with sorafenib were investigated. The drug loading efficacy in APS-TCPSi, TCPSi-RGDS and TCPSi-iRGD nanoparticles is listed in Table 1. Functionalization of the TCPSi nanoparticles with RGDS and iRGD via SPAAC did not significantly affect the drug loading into the TCPSi nanoparticles.

Next, we tested the drug release in different media. We first observed that the drug was not released either from the APS-TCPSi nanoparticles or from the TCPSi-RGDS or TCPSi-iRGD nanoparticles (Fig. S3) in aqueous buffers at pH-values of 5.5 and 7.4. Sorafenib has very poor water solubility. Thus, after loading the drug into the TCPSi nanoparticles, the dissolution was performed in DMEM containing 10% of FBS. The drug release results showed a fast drug dissolution from all the three TCPSi nanoparticles tested (Fig. 6a). About 69, 70 and 79% of the drug was released from APS-TCPSi, TCPSi-RGDS and TCPSi-iRGD nanoparticles, respectively, in the first 5 min, followed by a sustained release. Similarly, sorafenib was also released from the TCPSi nanoparticles in 10% FBS at pH-values of 7.4 and 5.5 (Fig. 6b and c), but sorafenib was not released from the TCPSi nanoparticles in DMEM without FBS (Fig. 6d). In order to evaluate the effect of the FBS on the release of sorafenib from TCPSi nanoparticles, 10% FBS was added to the dissolution media after 2 h incubation of the sorafenib-loaded TCPSi nanoparticles in DMEM without FBS. When 10% of FBS was added to the dissolution media, 64, 67 and 56% of sorafenib was dissolved in 5 min from the APS-TCPSi, TCPSi-RGDS and TCPSi-iRGD nanoparticles, respectively, followed by a continuous drug release (Fig. 6e). As a

1264

C.-F. Wang et al. / Biomaterials 35 (2014) 1257e1266

Fig. 6. Dissolution profiles in aqueous solutions of sorafenib-loaded in APS-TCPSi, TCPSi-RGDS and TCPSi-iRGD. (a) DMEM containing 10% of FBS, (b) buffer solution at pH 7.4 with 10% FBS, (c) buffer solution at pH 5.5 with 10% FBS, (d) DMEM without FBS, (e) DMEM without FBS and with 10% FBS at 120 min, and (f) bulk sorafenib in DMEM with and without 10% FBS. SF denotes for sorafenib.

control, the bulk sorafenib dissolution was also performed in DMEM with and without FBS (Fig. 6f). The bulk sorafenib showed extremely poor solubility in DMEM without FBS up to 7 days. Slow, and continuous dissolution profile up to 7 days with 14% of total amount of sorafenib was released in DMEM containing 10% FBS. Overall, when loaded into all the three PSi nanoparticles, sorafenib showed enhanced dissolution behavior in aqueous solutions containing 10% FBS compared to the bulk drug. Surface functionalization

of the TCPSi nanoparticles with either of the peptides did not significantly affect the drug loading degree and release profiles. Interestingly, the sorafenib-loaded PSi nanoparticles kept the drug inside the pores of the particles without leakage into the buffer in the absence of FBS. Thus, we envisaged that the TCPSi nanoparticles loaded with sorafenib can be stored for long periods of times in aqueous serum-free solutions without compromising the stability of the formulation.

C.-F. Wang et al. / Biomaterials 35 (2014) 1257e1266

1265

3.7. Antiproliferation activity The antiproliferation effect of sorafenib-loaded in APS-TCPSi, TCPSi-RGDS, and TCPSi-iRGD was also tested in the endothelial cell line EA.hy926 using an ATP-based luminescence cell proliferation assay. Sorafenib alone was first dissolved in dimethyl sulfoxide (DMSO) as a stock solution with a concentration of 2000 mM. Sorafenib-loaded in the three TCPSi nanoparticles showed a similar dose-dependent antiproliferative effect in the tumor cells, whereas TCPSi nanoparticles alone did not significantly inhibit the cell proliferation for the corresponding concentrations of the sorafenibloaded TCPSi nanoparticles (Fig. 7 and Fig S4). The antiproliferation effect of sorafenib-loaded TCPSi nanoparticles in vitro was similar to the sorafenib dissolved in DMSO. Although the antiproliferation effect of sorafenib-loaded in the peptide-functionalized TCPSi nanoparticles was similar to the APSTCPSi, the former nanoparticles were more specific as they were more efficiently internalized into the cells, thus enhancing the intracellular drug delivery efficiency and reducing possible delivery of the drug in the extracellular side, which can reduce extensive side effects to the healthy cells. Sorafenib-loaded in TCPSi nanoparticles were resuspended in cell media without FBS and added to the cells and incubated for 24 h. The results showed that sorafenibloaded in TCPSi-RGDS had higher cytotoxicity than the APS-TCPSi or TCPSi-iRGD nanoparticles (Fig. 8). These results highlight the fact that RGDS has a higher coupling efficacy onto the TCPSi nanoparticles surface, and thus, a higher enhanced cell uptake efficiency and superior in vitro targeting drug delivery capacity. 4. Conclusion An ideal drug delivery system that would efficiently deliver the drug to the targeting site with as little as possible exposure drug to healthy cells or tissues is still under extensive research. In this study, we highlight the successful conjugation of targeting peptides to the surfaces of the APS-TCPSi nanoparticles via copper-free azideealkyne cycloaddition, as well as the intracellular delivery of sorafenib-loaded peptide-functionalized TCPSi nanoparticles in EA.hy926 endothelial cells. These peptide-functionalized nanoparticles enhanced the endothelial cell internalization compared to APS-TCPSi nanoparticles alone, but did not influence the drug

Fig. 8. Growth inhibition of EA.hy926 endothelial cells treated with sorafenib-loaded TCPSi nanoparticles. TCPSi nanoparticles were added to cells in media without FBS. The level of the significant difference compared to the samples treated with TCPSi nanoparticles was set at probabilities of *p < 0.05 and **p < 0.01.

loading and dissolution profiles of sorafenib-loaded TCPSi nanoparticles. In general, the sorafenib-loaded TCPSi-RGDS and TCPSiiRGD nanoparticles showed a similar trend for the in vitro antiproliferation effect against EA.hy926 endothelial cells compared to sorafenib dissolved in DMSO. However, sorafenib-loaded TCPSiRGDS deliver the drug intracellulary more efficiently than sorafenib-loaded APS-TCPSi and TCPSi-iRGD nanoparticles, due to the higher surface conjugation ratio, resulting in enhanced in vitro antiproliferation activity. The surface peptide-functionalization described in this work is a universal method to introduce functional moieties onto the surface of TCPSi nanoparticles. Therefore, these peptide-functionalized TCPSi nanoparticles are potential platforms as drug delivery systems and should be further developed for targeted cancer therapy.

Conflict of interest The authors declare no competing financial interest.

Acknowledgments C.-F. Wang acknowledges PhD scholarship support from the Chinese Scholarship Council (grant no. 2009627022). H.A. Santos acknowledges the University of Helsinki, the Academy of Finland (decision no. 252215 and 256394) and the European Research Council under the European Union’s Seventh Framework Programme (FP/2007e2013) grant no. 310892 for financial support. Mr. H. Haddad (Department of Chemistry, University of Helsinki) is thanked for his help with the elemental analysis. Dr. Y. Lou (Division of Biopharmaceutics and Pharmacokinetics, University of Helsinki, Finland) is acknowledged for generously providing the EA.hy926 cells.

Fig. 7. Growth inhibition of EA.hy926 endothelial cells treated with sorafenib and sorafenib-loaded TCPSi nanoparticles. The nanoparticles were suspended in full cell media. Cells were exposed to pure sorafenib, and sorafenib-loaded TCPSi nanoparticles for 24 h at 37
C. Pure sorafenib was dissolved to 2 mM in DMSO as stock solution. The cell viability of 1% of DMSO was 82  7%.

Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.biomaterials.2013.10.065.

1266

C.-F. Wang et al. / Biomaterials 35 (2014) 1257e1266

References [1] Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100:57e70. [2] Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 2011;144:646e74. [3] Ruoslahti E, Bhatia SN, Sailor MJ. Targeting of drugs and nanoparticles to tumors. J Cell Biol 2010;188:759e68. [4] Ruoslahti E. Specialization of tumour vasculature. Nat Rev Cancer 2002;2:83e90. [5] Weis SM, Cheresh DA. Alpha V integrins in angiogenesis and cancer. Cold Spring Harb Perspect Med 2011;1:a006478. [6] Brooks PC, Clark RA, Cheresh DA. Requirement of vascular integrin alpha V beta 3 for angiogenesis. Science 1994;264:569e71. [7] Pierschbacher MD, Ruoslahti E. Cell attachment activity of fibronectin can be duplicated by small synthetic fragments of the molecule. Nature 1984;309: 30e3. [8] Pierschbacher MD, Ruoslahti E. Variants of the cell recognition site of fibronectin that retain attachment-promoting activity. Proc Natl Acad Sci USA 1984;81:5985e8. [9] Sugahara KN, Teesalu T, Karmali PP, Kotamraju VR, Agemy L, Greenwald DR, et al. Coadministration of a tumor-penetrating peptide enhances the efficacy of cancer drugs. Science 2010;328:1031e5. [10] Chen CW, Lu DW, Yeh MK, Shiau CY, Chiang CH. Novel RGD-lipid conjugatemodified liposomes for enhancing siRNA delivery in human retinal pigment epithelial cells. Int J Nanomed 2011;6:2567e80. [11] Sugahara KN, Teesalu T, Karmali PP, Kotamraju VR, Agemy L, Girard OM, et al. Tissue-penetrating delivery of compounds and nanoparticles into tumors. Cancer Cell 2009;16:510e20. [12] Shahbazi MA, Hamidi M, Makila EM, Zhang H, Almeida PV, Kaasalainen M, et al. The mechanisms of surface chemistry effects of mesoporous silicon nanoparticles on immunotoxicity and biocompatibility. Biomaterials 2013;34:7776e89. [13] Bimbo LM, Denisova OV, Makila E, Kaasalainen M, De Brabander JK, Hirvonen J, et al. Inhibition of influenza a virus infection in vitro by saliphenylhalamide-loaded porous silicon nanoparticles. ACS Nano 2013;7: 6884e93. [14] Salonen J, Laitinen L, Kaukonen AM, Tuura J, Bjorkqvist M, Heikkila T, et al. Mesoporous silicon microparticles for oral drug delivery: loading and release of five model drugs. J Control Release 2005;108:362e74. [15] Santos HA, Bimbo LM, Lehto VP, Airaksinen AJ, Salonen J, Hirvonen J. Multifunctional porous silicon for therapeutic drug delivery and imaging. Curr Drug Discov Technol 2011;8:228e49. [16] Jarvis KL, Barnes TJ, Prestidge CA. Surface chemical modification to control molecular interactions with porous silicon. J Colloid Interface Sci 2011;363: 327e33. [17] Wu EC, Andrew JS, Cheng L, Freeman WR, Pearson L, Sailor MJ. Real-time monitoring of sustained drug release using the optical properties of porous silicon photonic crystal particles. Biomaterials 2011;32:1957e66. [18] Makila E, Bimbo LM, Kaasalainen M, Herranz B, Airaksinen AJ, Heinonen M, et al. Amine modification of thermally carbonized porous silicon with silane coupling chemistry. Langmuir 2012;28:14045e54. [19] Mann AP, Tanaka T, Somasunderam A, Liu X, Gorenstein DG, Ferrari M. E-selectin-targeted porous silicon particle for nanoparticle delivery to the bone marrow. Adv Mater 2011;23:H278e82. [20] Tabasi O, Falamaki C, Khalaj Z. Functionalized mesoporous silicon for targeted-drug-delivery. Colloid Surf B 2012;98:18e25. [21] Kolb HC, Finn MG, Sharpless KB. Click chemistry: diverse chemical function from a few good reactions. Angew Chem Int Ed Engl 2001;40:2004e21. [22] Jewett JC, Bertozzi CR. Cu-free click cycloaddition reactions in chemical biology. Chem Soc Rev 2010;39:1272e9. [23] Sawoo S, Dutta P, Chakraborty A, Mukhopadhyay R, Bouloussa O, Sarkar A. A new bio-active surface for protein immobilisation via copper-free ‘click’ between azido SAM and alkynyl Fischer carbene complex. Chem Commun 2008;7:5957e9. [24] Lee DE, Na JH, Lee S, Kang CM, Kim HN, Han SJ, et al. Facile method to radiolabel glycol chitosan nanoparticles with 64Cu via copper-free click chemistry for microPET imaging. Mol Pharmacol 2013;10:2190e8.

[25] Wilhelm S, Carter C, Lynch M, Lowinger T, Dumas J, Smith RA, et al. Discovery and development of sorafenib: a multikinase inhibitor for treating cancer. Nat Rev Drug Discov 2006;5:835e44. [26] Mori S, Cortes J, Kantarjian H, Zhang W, Andreefi M, Ravandi F. Potential role of sorafenib in the treatment of acute myeloid leukemia. Leuk Lymphoma 2008;49:2246e55. [27] Wilhelm SM, Carter C, Tang L, Wilkie D, McNabola A, Rong H, et al. BAY 439006 exhibits broad spectrum oral antitumor activity and targets the RAF/ MEK/ERK pathway and receptor tyrosine kinases involved in tumor progression and angiogenesis. Cancer Res 2004;64:7099e109. [28] Jain L, Woo S, Gardner ER, Dahut WL, Kohn EC, Kummar S, et al. Population pharmacokinetic analysis of sorafenib in patients with solid tumours. Br J Clin Pharmacol 2011;72:294e305. [29] Boudou-Rouquette P, Narjoz C, Golmard JL, Thomas-Schoemann A, Mir O, Taieb F, et al. Early sorafenib-induced toxicity is associated with drug exposure and UGTIA9 genetic polymorphism in patients with solid tumors: a preliminary study. PloS One 2012;7:e42875. [30] Ravandi F, Cortes JE, Jones D, Faderl S, Garcia-Manero G, Konopleva MY, et al. Phase I/II study of combination therapy with sorafenib, idarubicin, and cytarabine in younger patients with acute myeloid leukemia. J Clin Oncol 2010;28:1856e62. [31] Zhang H, Zhang FM, Yan SJ. Preparation, in vitro release, and pharmacokinetics in rabbits of lyophilized injection of sorafenib solid lipid nanoparticles. Int J Nanomed 2012;7:2901e10. [32] Kim do H, Kim MD, Choi CW, Chung CW, Ha SH, Kim CH, et al. Antitumor activity of sorafenib-incorporated nanoparticles of dextran/poly(dl-lactide-coglycolide) block copolymer. Nanoscale Res Lett 2012;7:91. [33] Baranska P, Jerczynska H, Pawlowska Z, Koziolkiewicz W, Cierniewski CS. Expression of integrins and adhesive properties of human endothelial cell line EA.hy926. Cancer Genomics Proteomics 2005;2:265e9. [34] Bimbo LM, Sarparanta M, Santos HA, Airaksinen AJ, Makila E, Laaksonen T, et al. Biocompatibility of thermally hydrocarbonized porous silicon nanoparticles and their biodistribution in rats. ACS Nano 2010;4:3023e32. [35] Serda RE, Blanco E, Mack A, Stafford SJ, Amra S, Li Q, et al. Proteomic analysis of serum opsonins impacting biodistribution and cellular association of porous silicon microparticles. Mol Imaging 2011;10:43e55. [36] Bimbo LM, Makila E, Laaksonen T, Lehto VP, Salonen J, Hirvonen J, et al. Drug permeation across intestinal epithelial cells using porous silicon nanoparticles. Biomaterials 2011;32:2625e33. [37] Meeuwissen SA, Debets MF, van Hest JCM. Copper-free click chemistry on polymersomes: pre-vs. post-self-assembly functionalisation. Polym Chem 2012;3:1783e95. [38] Santos HA, Riikonen J, Salonen J, Makila E, Heikkila T, Laaksonen T, et al. In vitro cytotoxicity of porous silicon microparticles: effect of the particle concentration, surface chemistry and size. Acta Biomater 2010;6:2721e31. [39] Bhattacharjee S, Rietjens IM, Singh MP, Atkins TM, Purkait TK, Xu Z, et al. Cytotoxicity of surface-functionalized silicon and germanium nanoparticles: the dominant role of surface charges. Nanoscale 2013;5:4870e83. [40] Albanese A, Tang PS, Chan WC. The effect of nanoparticle size, shape, and surface chemistry on biological systems. Annu Rev Biomed Eng 2012;14: 1e16. [41] Sarparanta M, Bimbo LM, Rytkonen J, Makila E, Laaksonen TJ, Laaksonen P, et al. Intravenous delivery of hydrophobin-functionalized porous silicon nanoparticles: stability, plasma protein adsorption and biodistribution. Mol Pharmacol 2012;9:654e63. [42] Liu D, Mäkilä E, Zhang H, Herranz B, Kaasalainen M, Kinnari P, et al. Nanostructured porous silicon-solid lipid nanocomposite: towards enhanced cytocompatibility and stability, reduced cellular association, and prolonged drug release. Adv Funct Mater 2013;23:1893e902. [43] Krasnici S, Werner A, Eichhorn ME, Schmitt-Sody M, Pahernik SA, Sauer B, et al. Effect of the surface charge of liposomes on their uptake by angiogenic tumor vessels. Int J Cancer 2003;105:561e7. [44] Gratton SE, Ropp PA, Pohlhaus PD, Luft JC, Madden VJ, Napier ME, et al. The effect of particle design on cellular internalization pathways. Proc Natl Acad Sci USA 2008;105:11613e8.