Augmented cellular trafficking and endosomal escape of porous silicon nanoparticles via zwitterionic bilayer polymer surface engineering.

Biomaterials xxx (2014) 1e13

Contents lists available at ScienceDirect

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

Augmented cellular trafficking and endosomal escape of porous silicon nanoparticles via zwitterionic bilayer polymer surface engineering €kila € a, b, Mohammad-Ali Shahbazi a, *, Patrick V. Almeida a, Ermei M. Ma b b a lder A. Santos a, * Martti H. Kaasalainen , Jarno J. Salonen , Jouni T. Hirvonen , He a b

Division of Pharmaceutical Chemistry and Technology, Faculty of Pharmacy, University of Helsinki, FI-00014 Helsinki, Finland Laboratory of Industrial Physics, Department of Physics and Astronomy, University of Turku, FI-20014, Finland

a r t i c l e i n f o

a b s t r a c t

Article history: Received 3 May 2014 Accepted 10 May 2014 Available online xxx

The development of a stable vehicle with low toxicity, high cellular internalization, efficient endosomal escape, and optimal drug release profile is a key bottleneck in nanomedicine. To overcome all these problems, we have developed a successful layer-by-layer method to covalently conjugate polyethyleneimine (PEI) and poly(methyl vinyl ether-co-maleic acid) (PMVE-MA) copolymer on the surface of undecylenic acid functionalized thermally hydrocarbonized porous silicon nanoparticles (UnTHCPSi NPs), forming a bilayer zwitterionic nanocomposite containing free positive charge groups of hyperbranched PEI disguised by the PMVE-MA polymer. The surface smoothness, charge and hydrophilicity of the developed NPs considerably improved the colloidal and plasma stabilities via enhanced suspensibility and charge repulsion. Furthermore, despite the surface negative charge of the bilayer polymerconjugated NPs, the cellular trafficking and endosomal escape were significantly increased in both MDA-MB-231 and MCF-7 breast cancer cells. Remarkably, we also showed that the conjugation of surface free amine groups of the highly toxic UnTHCPSi-PEI (Un-P) NPs to the carboxylic groups of PMVE-MA renders acceptable safety features to the system and preserves the endosomal escape properties via proton sponge mechanism of the free available amine groups located inside the hyper-branched PEI layer. Moreover, the double layer protection not only controlled the aggregation of the NPs and reduced the toxicity, but also sustained the drug release of an anticancer drug, methotrexate, with further improved cytotoxicity profile of the drug-loaded particles. These results provide a proof-of-concept evidence that such zwitterionic polymer-based PSi nanocomposites can be extensively used as a promising candidate for cytosolic drug delivery. © 2014 Elsevier Ltd. All rights reserved.

Keywords: Porous silicon nanoparticles Surface functionalization Cellular uptake Endosomal escape Polyethyleneimine Poly(methyl vinyl ether-co-maleic acid)

1. Introduction The success in cancer nano-therapy is extremely dependent on the development of carriers that are able to efficiently deliver therapeutic agents to the cytosolic compartment of target cells with minimal toxicity [1]. Among many different types of nanoparticles (NPs) applied for this aim, honeycomb-like porous silicon (PSi) NPs show remarkable advantages, including high surface area, stable nanostructure, tunable pore diameter, two functional surfaces (external particle surface and internal pore surface), modifiable

* Corresponding authors. Tel.: þ358 2941 59661, þ358 2941 59673; fax: þ358 2941 59138. E-mail addresses: m.a.shahbazi@helsinki.fi (M.-A. Shahbazi), helder.santos@ helsinki.fi (H.A. Santos).

shape and size, effective protection of the therapeutic cargos from undesirable degradation, as well as superior safety at concentrations adequate for pharmacological applications [2e5]. Currently, despite the above mentioned advantages, there are still concerns regarding the potential of PSi NPs at the cellular level due to the low cellular interaction and entry into the cells [6,7]. In addition, the internalized PSi NPs usually become entrapped inside endosomes and ultimately end up in the lysosome, a biological compartment in charge for enzymatic degradation and inactivation of different compounds [8]. Therefore, these nanostructures will possess a high potential for cancer therapy, if rendering the ability to breach cellular membrane and reach the cytoplasm or nucleus of the cell, a difficult task to achieve due to the complexity of the biological barriers [9,10]. Since many types of developed NPs with favorable physicochemical properties in vitro cannot be successfully

http://dx.doi.org/10.1016/j.biomaterials.2014.05.020 0142-9612/© 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Shahbazi M-A, et al., Augmented cellular trafficking and endosomal escape of porous silicon nanoparticles via zwitterionic bilayer polymer surface engineering, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.05.020

2

M.-A. Shahbazi et al. / Biomaterials xxx (2014) 1e13

applied in vivo owing to the limited intracellular functionality, numerous attempts have been made to find new approaches to attenuate this drawback by enhancing their cellular uptake and endosomal escape [11,12]. For example, cell penetrating peptides have been widely used for cellular uptake enhancement despite several disadvantages, such as low metabolic stability, possible immunogenicity, and dependency of their membrane translocation ability on the amino acids arrangement and site of conjugation with the NPs [13e15]. To overcome these problems, new alternative strategies are essential to achieve favorable therapeutic effect for the nanomedicines. Surface polymeric functionalization is one of the desirable methods that can be applied not only to affect the NP's properties by manipulation of the intrinsic size, shape, charge, smoothness, hydrophilicity/hydrophobicity, homogeneity, and stability, but also to act as a driving force for improving cellular internalization, endosomal escape, and drug release profile with the final aim of achieving a subtle therapeutic effect [16e18]. Although many reports in the literature have demonstrated the ability of some specific types of positively charged polymers, such as polyethyleneimine (PEI), for successful improvement of cellular internalization and endosomal escape [19,20], these materials have shown drawbacks in terms of making overt pores in the lipid bilayers, which can eventually lead to cellular toxicity by disturbing the concentration balance of ions and proteins that are essential to maintain the normal function of the cells [21,22], Thus, the application of specific polymer-conjugated nanostructures that are capable to avoid cellular toxicity while increasing cellular trafficking and release their payloads in the cytoplasm is essential for designing superior nanomedicines. Despite rapidly increasing progress in the polymer and copolymer based drug delivery systems with different compositions, shapes, and properties [23e25], there is still lack of deep understanding about the fate of surface polymeric functionalized NPs at the cellular level. Thus, clarification of the benefits associated to the surface functionalization of NPs with different polymers may bring new advantages to this burgeoning area of research. Despite current attempts to reduce cellular toxicity and improve cellular interaction using functionalization of the PSi NPs with bioadhesive negatively charged polymers [26], combining these properties with endosomal escape behavior is crucial for effective therapeutic effect. Accordingly, this study reports the preparation, characterization and in vitro fate of a new class of bilayer polymer-conjugated PSi NPs developed using the surface conjugation of PEI and poly(methyl vinyl ether-co-maleic acid) (PMVE-MA). The covalent attachment of the copolymers to the PSi's surface was made via the 1-ethyl-3-[3-dimethylaminopropyl] carbodiimide hydrochloride/N-hydroxysuccinimide (EDC/NHS) mediated covalent binding. The developed bilayer nanocomposite, namely UnTHCPSi-PEI-PMVE-MA NPs (Un-P-P), was studied in terms of cytotoxicity effect on cancer cells, stability in the aqueous buffer and in human plasma, cellular internalization, drug release, and antiproliferation effect on cancer cells.

The hydrodynamic diameter (Z-average), PDI and surface zeta-potential of the NPs were measured using a Zetasizer Nano ZS instrument (Malvern Instruments Ltd, UK). Typically, polymer-conjugated and unconjugated PSi NPs were centrifuged and redispersed in MilliQ-water with a final concentration of 30 mg/mL prior to the measurements. All measurements were repeated at least three times. The chemical composition and interaction of the UnTHPSi NPs and the polymers were studied by ATReFTIR. The ATReFTIR spectra of all samples were obtained using a Bruker VERTEX 70 series FTIR spectrometer (Bruker Optics, Germany) with a horizontal ATR sampling accessory (MIRacle, Pike Technology, Inc.). The ATReFTIR spectra were recorded in the wavenumber region of 4000650 cm1 with a resolution of 4 cm1 at RT using OPUS 5.5 software. Prior to each measurement, all the NPs were left to dry at RT for 48 h. The morphology of the NPs was studied using TEM (Jeol JEM-1400, Jeol Ltd., Japan). Samples were prepared in ethanol and dropped on a carbon coated copper TEM grid before air-drying at RT overnight. The colloidal stability of the NPs was also screened by leaving the samples at RT for 4 h after preparation.

2. Material and methods

2.8. Cell viability studies

2.1. Fabrication of the UnTHCPSi NPs

To assess the biocompatibility of the NPs, their toxicity towards MCF-7 and MDA-MB-231 cells was evaluated by measuring the ATP activity as described elsewhere [27,28], and are explained in more detailed in Supporting Information. Therapeutic efficiency of the developed nanocarriers was investigated by the cell proliferation evaluation of the cells after a 6 h exposure to free methotrexate (MTX) and drug-loaded NPs using the same protocol applied for cell viability determination.

The preparation and characterization of PSi NPs have been previously described in detail elsewhere (Scheme 1) [2,7,27]. Details can be found in Supporting Information. 2.2. Preparation of UnTHCPSi-PEI (Un-P) NPs The carboxyl groups of UnTHCPSi NPs were covalently conjugated to the amine groups of branched PEI (average Mw ~ 25,000), as shown in Scheme 1. To successfully accomplish the covalent conjugation, 1.5 mg of UnTHCPSi NPs was dispersed in 4 mL of 10 mM 2-(N-morpholino)ethanesulfonic acid (MES) saline buffer at pH 5.2. Following dispersion, 8 mL and 6 mg of EDC and NHS, respectively, were added and mixed for 2 h to activate the carboxyl groups of the PSi NPs. Finally, the surface activated PSi NPs were exposed to an excess of hyper-branched PEI with a ratio of 1:10

(NPs:polymer) and vigorously stirred (800 rpm) at room temperature (RT) overnight. Since the amount of PEI used to covalently cross-link to the NPs was very high, we assume that all carboxyl groups of the PSi NPs were reacted with the polymer, leaving no free carboxyl group for further conjugation. To remove the excess of unconjugated polymer, the surface modified UnTHCPSi NPs were extensively rinsed with MiliQwater, and re-suspended in Hank's balanced salt solution (HBSS)(4-(2hydroxyethyl)-1-piperazineethanesulfonic acid) (HEPES) buffer (pH 7.4). 2.3. Formation of UnTHCPSi-PEI-PMVE-MA (Un-P-P) nanocomposites The preparation of Un-P-P nanocomposites was achieved by the addition of PMVE-MA copolymer onto the PEI-functionalized UnTHCPSi NPs, as shown in Scheme 1. Briefly, PMVE-MA copolymer was first obtained from poly(methyl vinyl ether-co-maleic anhydrate) (PMVE-MAh, average Mw ~ 216,000) by dissolving the later one in HBSSeHEPES buffer (pH 5.2) at 70 C for 3 h. Next, PMVE-MA polymer was activated for 2 h by addition of EDC/NHS to the solution, and subsequently, added with a ratio of 1:1 to Un-P NPs dispersed in the same buffer. The obtained polymer-conjugated PSi NPs were washed twice with MilliQ-water by repeated centrifugation at 15,000 rpm for 5 min and redispersed by pipetting in order to ensure that no ungrafted polymer or free reagents were present in the final obtained product. The NPs were then redispersed in HBSSeHEPES buffer (pH 7.4). 2.4. Characterization of the NPs

2.5. Stability in human plasma and aqueous solution To evaluate the impact of the PSi NPsʼ stability, 300 mg of the bare and polymerconjugated PSi NPs were dispersed in 200 mL of PBS (pH 7.4). The NPs were then mixed with 1500 mL of human plasma and kept at 37 C for 2 h under stirring at 800 rpm. Sampling was performed at different pre-determined time intervals (15, 30, 60, 90 and 120 min) to measure the particle size and PDI using Zetasizer Nano ZS instrument. Anonymous donor human plasma was obtained from the Finnish Red Cross Blood Service. For colloidal stability investigation in aqueous solution, all types of the NPs were dispersed in PBS (pH 7.4) and the particle size and PDI were measured in the above mentioned time points. 2.6. Fluorescent labeling of the NPs Fluorescein isothiocyanate (FITC) was conjugated to the NPs for imaging purposes. FITC was added to a mixture of HEPES (0.1 M; pH 7.5) and ethanol (1:4, v/v). Then, FITC-conjugated NPs were obtained by addition of the NPs to the FITC solution in a 10:1 ratio. 2.7. Cell lines and culture conditions For the in vitro studies, MDA-MB-231 and MCF-7 breast cancer cells were cultured according to the protocols described in detail in the Supporting Information.

2.9. Cellular uptake imaging TEM imaging was used to evaluate the cellular uptake potential of the NPs after exposure to both MCF-7 and MDA-MB-231 breast cancer cells. In this experiment, 13 mm round shape coverslips were placed at the bottom of 24-well plates (Corning Inc. Life Sciences, USA). Next, 105 cells were seeded in Dulbecco's modified Eagle's and Roswell Park Memorial Institute 1640 media for MCF-7 and MDA-MB-231 cells,



3

Scheme 1. Schematic representation of the production of UnTHCPSi NPs from boron doped pþ Si wafer and typical illustration of the conjugation of UnTHCPSi NPs with PEI and PMVE-MA polymers via EDC/NHS mediated covalent binding. The presence of carboxyl groups on the surface of the UnTHCPSi NPs render them the potential for further surface functionalization with different types of polymers. respectively, and allowed to attach overnight. After removing the cell culture media, 500 mL of the NPs suspensions (50 mg/mL) were added to each well prior incubation at 37 C for 6 h. The particle suspension was then carefully aspirated and the coverslips were washed twice with HBSSeHEPES. The cells were fixed with 2.5% glutaraldehyde in 0.1 m PBS solution (pH 7.4) for 1 h at RT. After fixation, the wells were washed twice with HBSSeHEPES (pH 7.4) and sodium cacodylate buffer (NaCac) for 3 min prior to post-fixation with 1% osmium tetroxide in 0.1 m NaCac buffer (pH 7.4). The cells were then dehydrated with 30e100% ethanol for 10 min each and embedded in epoxy resin. Ultrathin sections (60 nm) were cut parallel to the coverslip, post-stained with uranyl acetate and lead citrate, and observed by TEM. 2.10. Cellular association analyses by flow cytometry Cells were seeded in 6-well plates at a density of 7 105 cells/well and incubated at 37 C overnight. The cells were then washed with HBSSeHEPES (pH 7.4) and incubated with FITC-labeled NPs (50 mg/mL) for 6 h. For flow cytometry analyses, cells incubated with the NPs were detached by incubating with 300 mL of trypsin-PBSEDTA mixture. NPs not associated with the cells were then removed by washing three times with HBSSeHEPES and subsequent centrifugation at 500 rpm for 3 min. Thereafter, fixation of the cells was performed with 2.5% glutaraldehyde in PBS for 30 min and samples were re-suspended in 700 mL of the HBSSeHEPES (pH 7.4) prior to measurements using a LSR II flow cytometer (BD Biosciences, USA) with a laser excitation wavelength of 488 nm. To quench the fluorescence of nanoparticles conjugated on the surface of the cell membrane, trypan blue (TB, 0.005% v/v) was used before cell fixation, followed by three time washing. About 10,000 events were exactly obtained for each sample. Data were analyzed and plotted using Weasel software. 2.11. Intracellular distribution: endosomal escape and cytosolic NP delivery The intracellular distribution of the NPs in MCF-7 and MDA-MB-231 breast cancer cells were examined by an inverted confocal fluorescence microscopy (Leica

SP5 II HCS A, Germany). For this purpose, the cells were seeded into Lab-Tek 8Chamber Slides (Thermo Fisher Scientific, USA) at a density of 5 104 cells per well. After 24 h incubation, the cell medium was removed and 200 mL of the FITClabeled NPs (50 mg/mL) was added to each chamber. After 6 h incubation at 37 C, the NPs were removed and washed three times with HBSSeHEPES (pH 7.4). Thereafter, staining of the acidic organelles of the cells was done by adding 200 mL of the LysoTracker® Blue DND-22 (50 nM; Invitrogen, USA) and incubation for 30 min at 37 C. Washing was then performed twice with HBSSeHEPES (pH 7.4) to remove any free tracking agent. Next, the cells' plasma membrane was stained by adding 200 mL of the CellMask (3 mg/mL; Invitrogen, USA) and incubation for 3 min at 37 C. Finally, the stained solution was rinsed twice with HBSSeHEPES buffer, the cells fixed by 2.5% glutaraldehyde for 20 min and the intracellular localization was observed by confocal microscope.

2.12. Drug loading and release Loading of the model anticancer drug, MTX, into the NPs was performed by the immersion method using concentrated aqueous solutions of the drug. First, 10 mg of MTX was dissolved in 1 mL of PBS (pH 8). Then, MTX was added to the NPs at a weight ratio of 20:1 w/w (drug:NPs). After 90 min stirring (400 rpm) at RT. MTXloaded NPs were separated from the supernatant by centrifugation at 15,000 rpm for 7 min. To remove drug molecules loosely adsorbed on the surface of the NPs, the NPs were gently washed twice with MiliQ-water. Prior to HPLC analyses, the MTX-loaded NPs were re-suspended in 10 mL of PBS at pH 8, tip-sonicated for 30 s and stirred (800 rpm) at RT for 60 min. The suspensions were then centrifuged at 15,000 rpm for 7 min and the supernatant was used for the detection of MTX using a reverse-phase HPLC analysis. The amount of loaded MTX was determined by HPLC (Agilent 1260, Agilent Technologies, USA), using a C18 column (4.6 100 mm, 3.5 mm, Zorbax Eclipse plus C18, USA). The mobile phase consisted of 2:1 ratio of 0.2 M Na2HPO4 and 0.1 M citric acid (pH 6) mixed with


4


the acetonitrile (90:10, v/v). The flow rate was 1.0 mL/min and the injected volume of the sample was 20 mL. The column temperature and detection wavelength were set at 30 C and 240 nm, respectively. To determine the in vitro drug release of MTX from the NPs, 250 mg of the MTXloaded NPs were redispersed in 20 mL of PBS (pH 7.4) at 37 C and at a stirring speed of 100 rpm. At pre-determined time points, 200 mL of the PBS solution were withdrawn and replaced with equal volume of the corresponding fresh pre-warmed medium to retain a constant volume of the release medium. After sampling, the aliquots were centrifuged for 2 min at 15,000 rpm and the amount of MTX released from the particles in the supernatant was analyzed by HPLC as described above. All measurements were repeated at least three times. 2.13. Statistical analyses Measured values were expressed as mean standard deviation (s.d.) of at least three independent experiments. Results in the abovementioned assays were evaluated by means of one-way analysis of variance (ANOVA) with the level of significance set at probabilities of *p < 0.05, **p < 0.01 or ***p < 0.001.

3. Results and discussion 3.1. Fabrication and characterization of the bare and polymerconjugated PSi NPs Major challenges for the preparation of NPs are to design and prepare desired structures with low toxicity, high stability, favorable drug release profiles and acceptable cellular uptake. Although the potential of bare UnTHCPSi NPs has already been shown in the field of drug delivery [27], we have functionalized their surface to further improve the biological and physicochemical properties of the NPs for efficient intracellular drug delivery. The UnTHCPSi NPs were prepared according to the procedure depicted in Scheme 1 [29,30]. For polymeric functionalization, previously reported approaches have shown some limitations for techniques such as surface polymeric absorption, non-covalent assembly, and surface-initiated polymerizations, because of the susceptibility for rapid displacement of the polymer by species that are prevalent in cell medium or biological fluids [31e34]. This deficiency can lead to misinterpretation of the results owing to a remarkable change in the chemical nature, surface charge, as well as colloidal and biological stabilities of the NPs [35,36].

In order to overcome some of these problems, we have conjugated the PEI and PMVE-MA copolymers onto the surface of UnTHCPSi NPs using EDC/NHS chemistry in aqueous solution. The carboxyl groups of UnTHCPSi NPs were used as the precursor for polymeric functionalization, as depicted in Scheme 1. First, the carboxyl groups of the UnTHCPSi NPs were covalently conjugated to the amine groups of PEI polymer to form UnTHCPSi-PEI NPs. The free amines presented on the surface of UnTHCPSi-PEI NPs were used as anchors for attaching the PMVE-MA polymer. During the preparation, we sought to conjugate the carboxyl groups of PMVEMA with just the external amine groups on the surface of the PEI, leaving free amine groups in the interior part of the hyperbranched PEI network available for further biological functions like endosomal escape despite the negative surface charge of the particles. Indeed, the hyper-branched structure of the PEI hinders the carboxyl groups of PMVE-MA to penetrate inside the PEI chain, and thus, PMVE-MA is only conjugated with the external free amine groups of PEI, an achievement confirmed by attenuated total reflectance Fourier transform infrared (ATReFTIR) spectroscopy experiments. The chemical structure and composition of the polymers and the prepared NPs are shown in Figure S1 in the Supporting Information. The mean particle size (Z-average) of the UnTHCPSi, Un-P, and UnTHCPSi-PEI-PMVE-MA (Un-P-P) NPs, determined by dynamic light scattering (DLS) measurements, was 149.1 ± 0.8, 211.3 ± 12.4 and 320.7 ± 15.1 nm, respectively (Fig. 1A). As expected, the zetapotential of negatively charged UnTHCPSi NPs changed from 21.8 ± 0.6 mV to þ34.0 ± 0.9 mV and 31.7 ± 1.7 mV, respectively, after PEI and PMVE-MA conjugation (Fig. 1B). This confirmed the successful polymer conjugation onto the surface of UnTHCPSi NPs and formation of bilayer polymer-conjugated PSi NPs. The polydispersity index (PDI), which is an indicative of particle size distribution, was found to be less than 0.2 in all three samples, showing the low polydispersity of the prepared NPs (Fig. 1C). One of the unavoidable problems associated with PSi NPs is their intrinsic instability in aqueous solutions due to their tendency

Fig. 1. Size (A), zeta-potential (B) and PDI (C) of the PSi NPs after preparation, and colloidal stability (D) after incubation of the PSi NPs at RT for 4 h in static conditions.



to aggregate, leading to significant decrease in their interfacial area, dispersibility, and cellular association [3]. Therefore, the stability of the bare and polymer-functionalized PSi NPs in phosphate buffer saline (PBS, pH 7.4) was evaluated by comparing the visual appearance of the NPs after storing at RT for 4 h. It was observed that the native UnTHCPSi NPs aggregated and precipitated over time due to the hydrophobic properties, whereas the polymerfunctionalized NPs were quite stable and very well dispersed (Fig. 1D). The presence of both PEI and PMVE-MA on the surface of the PSi NPs not only improved the hydrophilicity of the particles, but also played a crucial role in augmenting the aqueous dispersibility of the NPs as a result of the electrostatic or steric repulsion forces [36,37], and thus, preventing the PSi NP's aggregation. Furthermore, the increase in the zeta-potential also indicated a further improvement of the stability by enhancing the repulsion force between the particles [24,25]. To scrutinize the morphological changes of the UnTHCPSi NPs after polymeric conjugation, all the prepared NPs were also characterized by transmission electron microscopy (TEM). As illustrated in Fig. 2, the PSi NPs show an irregular shape. Besides, although the pores in the bare UnTHCPSi NPs were visible (Fig. 2A), the PEIconjugated UnTHCPSi NPs exhibited a coverage onto the NPs' surface (Fig. 2B), suggesting that the hyper-branched PEI can form a network on the surface of the NPs by conjugating to the free carboxylic groups of the UnTHCPSi NPs. After the second polymeric conjugation with PMVE-MA and formation of the Un-P-P NPs, the pores were not observable anymore and, instead, a polymeric layer was visible onto the NP's surface (Fig. 2C), indicating structural changes of the UnTHCPSi NPs after polymer conjugation. The chemical conjugation of the polymers onto the PSi NP's surface was also analyzed by ATReFTIR spectroscopy. Fig. 3 shows that a gradual change in the surface chemistry of the UnTHCPSi NPs took place during the polymer conjugation. The UnTHCPSi NPs displayed the bands corresponding to eCH stretching, carboxyl C¼O stretching, and eCH2 deformation at 2922, 1709, and 1469 cm1, respectively [30]. In the ATReFTIR spectrum of PEI polymer, two small bands in the ranges of 3400e3300 and 3330e3250 cm1, and a bend vibration absorption of NeH bond at 1604 cm1 were appeared due to the presence of primary amine groups. In addition, the bands at 1471, 1291, 1119, and 913 cm1 were representatives of the CH2 bending, CeN stretching, secondary amines, and NeH bending vibration, respectively [38e40]. The

5

ATReFTIR spectrum of the pure PMVE-MAh showed the typical stretching bands of the anhydride groups (C¼O: 1855 and 1772 cm1; CeOeC: 918 cm1) and those corresponded to the methylether group (eOCH3: 1220 and 1088 cm1) [41]. Compared to the ATReFTIR spectrum of PMVE-MAh, it was found out that the carbonyl band is shifted to the lower wavenumbers (1692 cm1) in the ATReFTIR spectrum of PMVE-MA, confirming that the ring opening reaction of PMVE-MAh has been occurred and the hydroxyl groups are present (OeH bending). For the Un-P NPs, the most notable change compared to the bare UnTHCPSi NPs is the emergence of a conspicuous band at 1641 cm1, which is attributed to the primary amine groups overlapped with the weak C¼O stretching mode of the UnTHCPSi NPs. The visible band at 1471 cm1 is also assigned for the eCH2 stretching. Besides, the band at around 3400 is representative of NeH stretching, which further confirms the presence of PEI on the surface of the UnTHCPSi NPs. For the Un-P-P, compared to the PMVE-MA, the ATReFTIR bands assigned for the carbonyl group were shifted to a lower wavenumber, an observation that clearly indicates an amide formation in the bilayer polymer-conjugated NPs. Interestingly, despite the disappearance of the two small bands of the primary amines (3400e3250 cm1) due to the overshadowing with OeH stretching of the carboxylic groups, the band at 1578 cm1 for the NeH bending vibration could still be observed, indicating the presence of primary amines in the interior chains of the PEI layer, even after the PMVE-MA conjugation [42]. 3.2. Stability investigation After the physicochemical characterization of the polymerconjugated PSi NPs, it is crucial to understand the interactions between the NPs and the plasma proteins [43]. Ideally, nanocarriers injected into the bloodstream should show minimal interactions with plasma proteins in order to avoid agglomeration and rapid clearance by the macrophages [44]. To circumvent this problem, NPs can be conjugated, for example, with some kinds of polymers that resist the protein adsorption [45]. Our results showed that the polymer-conjugated NPs exhibited a significantly lower variation in size and PDI compared to the pure UnTHCPSi NPs, indicating less interactions with human plasma proteins (Fig. 4). Although it is known that the surface functionalization of NPs with PEI can increase the stability of the NPs by

Fig. 2. TEM images and magnifications of the UnTHCPSi (A), Un-P (B) and Un-P-P (C) NPs. The localization of the polymer layer onto the surface of the NPs is clear from left to right.


6


Fig. 3. FTIR spectra of UnTHCPSi, PEI, PMVE-MAh, PMVE-MA, Un-P, and Un-P-P are shown from number 1 to 6, respectively.

Fig. 4. The impact of the human plasma proteins on the size (A), and PDI (B) of UnTHCPSi, Un-P and Un-P-P NPs after 2 h incubation. The size (C) and PDI (D) of the NPs after dispersing in aqueous medium for 2 h, demonstrating high stability of the polymer-conjugated NPs over time. The results were calculated from the DLS measurements data as a function of time at 37 C. Values denote the mean ± s.d. (n 3).



reducing nonspecific protein adsorption [46,47], our results showed that PMVE-MA may also lead to a similar effect and decrease NP interaction with plasma protein [36,48,49]. Stability assessment in aqueous medium showed that while the hydrodynamic diameter and PDI of the UnTHCPSi NPs increased to over 1 mm and 0.4, respectively, in less than 90 min as a result of NP aggregation, polymer-functionalized particles showed no substantial change in both particles size and PDI. Further stability assessment of the NPs for 2 days revealed no change in the size of pure and polymer-conjugated NPs from 2 h to 48 h (data not shown). 3.3. Cellular toxicity In order to assess the in vitro safety and cytocompatibility of the bare and polymer-conjugated PSi nanocomposites, we analyzed the ATP activity [27,28] in both MDA-MB-231 and MCF-7 cells, following incubation with different concentrations of the NPs for 6 and 24 h. As shown in Fig. 5, unlike the satisfactory non-toxic effect of UnTHCPSi NPs, PEI-conjugated NPs showed high cytotoxicity in both cell lines, particularly at the highest NP concentration (100 mg/ mL). The very low cytotoxicity of UnTHCPSi NPs is hypothetically attributed to the surface negative charge and low aqueous stability of the NPs, which result in low cellular interaction and reduced cytotoxicity. The cytotoxicity of Un-P NPs is also mainly because of the positive charge of the free amines on the PEI polymer structure [50]. By disguising the free positive charges of PEI using PMVE-MA polymer, the cell viability was significantly improved after exposure to the Un-P-P nanocomposites. Comparing to the HBSS buffer control, although no significant toxic effect of Un-P-P nanocomposites was observed after 6 h exposure to the cells, the cell viability was significantly decreased to 64% for all concentrations in MDA-MB-231 and to 83% in MCF-7 cells for 50 and 100 mg/mL of NPs after 24 h. These results indicate that the Un-P-P

7

nanocomposites are safe for drug delivery applications for short incubation times and at concentrations lower than 100 mg/mL. 3.4. Cellular association and uptake One of the main concerns regarding the development of nanomedicines is to ensure efficient delivery and facile NP internalization into cancer cells. While the size of PSi NPs contribute significantly to the NPs' interaction with cells, the surface chemical composition seems to be one of the primary dictators of various imperative properties, such as dispersibility, biocompatibility, hydrophilicity, and cellular interactions [2,7,27,28]. Despite great efforts for applying PSi NPs in biomedical applications [51], there is still a need to improve their intracellular uptake and clarify their biofate. Thus, in order to understand the behavior of the PSi NPs at the cellular level, TEM and flow cytometry analyses were performed. TEM pictures showed that the association of UnTHCPSi NPs with both MDA-MB-231 and MCF-7 cell lines was negligible (Fig. 6), mainly due to the surface negative charge, lack of bioadhesivity and low aqueous stability, resulting in a remarkable reduction of cellular internalization [36]. Unlike the UnTHCPSi NPs, the Un-P and Un-P-P NPs were extensively taken up by both cell lines and appeared mainly located in the acidic compartments and cytosol. The results obtained for UnP NPs were in agreement with other reports in the literature showing that NPs with positive surface charge can interact with the negative charge moieties of the cell membrane, and subsequently become internalized, escaping from the endosome via proton sponge mechanism and eventually enter into the cytoplasm [52]. The strong cellular association of the Un-P-P NPs can be attributed to the high dispersibility of these NPs as well as bioadhesive properties of the PMVE-MA polymer [26,53]. In line with these results, there have been evidences of high uptake of

Fig. 5. Cell viability response of MDA-MB-231 (A and B) and MCF-7 (C and D) breast cancer cells after exposure to UnTHCPSi, Un-P and Un-P-P NPs. The ATP content was measured after 6 h (A and C) and 24 h (B and D) incubation with different concentrations of PSi NPs. Statistical analysis was made by ANOVA. All data sets were compared to the negative control HBSS (pH 7.4). The level of significance was set at probabilities of *p < 0.05, **p < 0.01, and ***p < 0.001. Error bars represent s.d. (n 3).


8


Fig. 6. TEM images of MDA-MB-231 and MCF-7 cells treated with 50 mg/mL of bare and polymer-conjugated PSi NPs for 6 h. A very small amount of the bare UnTHCPSi NPs were found onto the surface of the MDA-MB-231 cells and almost none were visible in the vicinity of the MCF-7 cells. The positively (Un-P) and negatively (Un-P-P) charged surface NPs appear in the close vicinity of the cell's membrane, as well as inside the vesicular endosomes and in the cytosol. The Un-P NPs were also observed in the nucleus of MDA-MB231 cells. The numbers represent the magnified parts of the cells. Scale bars are 7 mm.

negatively charged particles in different cell lines [54e57], despite the unfavorable interaction between them and the negatively charged cell membranes. The cellular internalization of such NPs usually occurs through nonspecific binding of the particles to the scarce cationic sites on the cell membrane surface and their subsequent endocytosis; however, other mechanisms, such as the cell surface targeting by the polymer, facilitation of cellular uptake mechanisms, and cell membrane penetrating functionality could also be involved here [58,59]. To confirm the TEM results, flow cytometry experiments were conducted in the presence and absence of TB (acts as a fluorescence quenching agent). The results were in very good agreement with the TEM imaging. In non-treated TB samples, while there was no change in the fluorescence peak of the bare UnTHCPSi NPs in both cell lines compared to the control (Fig. 7A and C), the fluorescence intensity of the cells incubated with Un-P and Un-P-P NPs was considerably shifted to the right, a clear evidence for the high cellular association of the polymer-conjugated NPs and the cells. Moreover, TB was not able to completely quench all the fluorescence of the cells treated with the polymer-conjugated NPs (Fig. 7B and D), revealing the cellular uptake of the particles in 6 h. Taken all together, the TEM and the flow cytometry studies allowed a reliable comparison between the surface adsorbed and the internalized PSi NPs as a function of the NP's surface functionalization by means of innovative bilayer assembly of polymer-conjugated PSi NPs that bear zwitterionic property owing to the presence of both free positive and negative charges in its structure. To the best of our knowledge, such a system has not been reported yet; however, a few studies are available representing the conjugation of some

widely used polymers on the surface of Si particles with the aim to prolong drug release time [18]. 3.5. Intracellular distribution and endosomal escape capability of polymer-conjugated PSi nanocomposites In order to demonstrate the potential of the prepared polymerconjugated PSi NPs for cancer therapy, the cellular trafficking and endosomal escape in both breast cancer cell lines were further explored by confocal fluorescence microscopy and simultaneous imaging of the PSi NPs, cell membrane, and the acidic compartments of the cells. Since NPs entrapped in endosomes and lysosomes can be easily degraded by specific enzymes, the endosomal escape of NPs is a critical prerequisite for effective cancer therapy [60]. Thus, we have imaged the endosomal escape capability of the UnTHCPSi NPs before and after polymer conjugation. While our results showed negligible cell uptake of UnTHCPSi NPs in both MDA-MB-231 and MCF-7 cells, an increase in the cellular internalization of Un-P and Un-P-P NPs was confirmed by observing turquoise color in the merged picture owing to the colocalization of green fluorescence of the PSi NPs and the blue color of MDA-MB-231 and MCF-7 cells (Fig. 8A and B). In addition, for both cell lines, only a few NPs were located inside the acidic compartments, as shown by the overlapped yellow color in the merged picture. In the merged picture, it is also clear that some of the particles are still on the surface of the cell membrane and present green color. These observations indicate that most of the NPs were able to interact with the cells, escape from the acidic compartments and localize in the cytosol. To more easily observe



9

Fig. 7. Flow cytometry analyses of MDA-MB-231 (A and B) and MCF-7 (C and D) cells. The cellular association experiments were performed in MDA-MB-231 and MCF-7 cells before (A and C) and after (B and D) extracellular fluorescence quenching by TB to evaluate both cellular uptake and extracellular binding of the NPs with the cell membrane. The cells were incubated with UnTHCPSi, Un-P, and Un-P-P NPs for 6 h at 37 C. The results are representative of at least two independent experiments.

the staining compartments, Lysotracker® and CellMask® were, respectively, pseudo-colored with red and blue using Leica Application Suite Advanced Fluorescence (LAS AF) software. According to our physicochemical and characterization results, the free available amine groups in the PEI layer of the Un-P-P NPs can act as endosomal escape inducing motifs via “proton-sponge” or “endosome buffering” effect [61]. Besides, this could be probably due to the fact that, although the maleic acid amide (MAA) constructed by the conjugation of carboxyl groups of maleic acid to the amine groups of PEI in Un-P-P is stable at extracellular neutral pH, it is capable to be rapidly hydrolyzed at the endosomal acidic pH [62,63], leading to a high impact on the proton sponge effect because of the presence of both previously available free hidden amine groups of the PEI and those that become free after PMVE-MA dissociation. Another possibility could be the ability of PMVE-MA layer to fuse into the lipid bilayer of the acidic compartments endosomes and disrupt them, similarly to what was observed in previous reports with other specific anionic polymers [50,64]. To show the potential of the developed nanosystems, high cellular interaction of the polymer-conjugated UnTHCPSi NPs with the cells after 3 h is shown in the Figure S2 of the Supporting Information. The TEM and confocal results showed that the Un-P NPs can cross the cell membrane of the MDA-MB-231 cells after 6 h incubation, confirming that the PEI conjugation successfully endow the UnTHCPSi NPs with cellular association function. In contrast, observed lower cellular uptake by MCF-7 cells can be ascribed to the higher resistant of the MCF-7 cells compared to the MDA-MB231 cells as well as its tendency to make condense clusters,

minimizing its interaction with the NPs. The pronounced differences in the cellular uptake and endosomal escape of the bare and polymer-conjugated NPs support the idea that surface chemical modification is among the most efficient strategies for the modulation of biological functions of NPs. However, the cellular uptake of NPs depends not only on the surface chemistry, but also on many other factors such as the size, shape, surface charge, purity of the NPs, aggregation/agglomeration, NP-cell incubation conditions, etc., making the actual uptake assessment of a given type of NP rather complex [48]. Furthermore, based on our results, it seems that the cell type can considerably affect the cellular trafficking profiles of the developed NPs. Therefore, cell-specific dependence of uptake, endosomal escape, and cytosolic delivery of these NPs should be further evaluated in the future. 3.6. Drug loading and release We next investigated the drug-loading of the NPs by comparing the total drug loading degree before and after polymer conjugation. MTX was chosen as a model anticancer drug due to its potency in breast cancer treatment [65], and the presence of both carboxyl and amine groups in its structure. The latter unique property of this drug can enhance the probability of its interaction with the amine and carboxyl groups of the polymers conjugated to the NPs and, consequently, increase its loading degree in the PSi NPs. The loading degree of MTX in the UnTHCPSi NPs was 6.4 ± 1.2%, whereas PEI and PMVE-MA conjugation improved the MTX loading degree to 12.6 ± 0.1 and 14.0 ± 0.5%, respectively. The low MTX-


10


Fig. 8. Confocal fluorescence microscopy analysis of the MDA-MB-231 (A) and MCF-7 (B) cells incubated with HBSS buffer (pH 7.4) as control (a), 50 mg/mL of UnTHCPSi (b), Un-P (c), and Un-P-P (d) at 37 C for 6 h. Lysotracker and CellMask for staining the lysosomes and the cell membranes are shown in red and blue pseudo-colors, respectively. Merged panel shows the internalized NPs outside of the acidic compartments in turquoise color and those co-localized in early endosomes and lysosomes in yellow color. The NPs that are located on the surface of the cell membrane are shown in green color in the merged panels. Scale bars are 8 mm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)



11

NPs, as well as hydrogen and electrostatic binding with the drug molecules [18]. A plausible explanation for the slower drug release rate from the Un-P-P NPs compared to the Un-P NPs is the presence of both carboxyl and amine groups in the MTX structure, which increases the possibility of interaction with the amine groups of PEI and the carboxyl groups of PMVE-MA in Un-P-P NPs. 3.7. Therapeutic efficiency

Fig. 9. Drug release profiles of MTX-loaded UnTHCPSi and polymer-conjugated PSi NPs in PBS (pH 7.4) at 37 C. Values represents the mean ± s.d. (n 3).

loading in the bare UnTHCPSi can be attributed to the low affinity of the drug to the pores of the bare NPs because of the intrinsic hydrophobic nature of the pores. Contrarily, polymer-conjugated NPs are prone to inter-particle hydrogen and electrostatic bonding with the drug, leading to improved drug loading. This suggests that the polymer conjugation increase the loading of the drug due to the more interactions of the drug's functional groups with the free amine and carboxyl groups of the polymer-conjugated PSi NPs. The drug release profiles of MTX-loaded PSi NPs were also evaluated. Fig. 9 shows the sustained drug release profiles after polymeric conjugation. The UnTHCPSi NPs released all the MTX in less than 5 min in PBS (pH 7.4) due to the rapid diffusion of the drug from the pores. Contrarily, the release rate from PEI- and PEI-PMVEMA conjugated PSi NPs were significantly slower than that of observed for bare PSi NPs, showing a constant drug release up to about 95 and 70% within the first 3 h, respectively. Afterwards, no significant change was observed in the cumulative drug release until 12 h. This suggests that the polymer layers provide an easy means to tune the drug release profile by capping on top of the PSi

In order to demonstrate that the Un-P and Un-P-P NPs can efficiently deliver the anticancer drug inside the cells, the cell proliferation in breast cancer cells exposed to the MTX-loaded PSi NPs was analyzed using an ATP-based activity assay. Since the Un-P NPs showed quite cytotoxic effect in the cell viability assessment (Fig. 5), only MTX-loaded Un-P-P particles were examined for anticancer therapeutic efficiency in a safe concentration range of the NPs and compared with equal molar ratio of the free drug and MTX-loaded UnTHCPSi NPs after 6 h incubation (Fig. 10). The results showed that the free drug and the MTX-loaded UnTHCPSi NPs have the same proliferation inhibition pattern, reducing the cell viability from 90% to less than 35% for MDA-MB-231 cells and from 95% to around 50% for MCF-7 cells. In contrast, MTX-loaded Un-P-P NPs exhibited more efficient therapeutic efficiency than the free and MTX-loaded UnTHCPSi NPs in both cell lines. In MDA-MB-231 and MCF-7 cells, the proliferation rate was decreased from 64% to less than 10% and from 77% to around 27%, respectively, after exposure to the Un-P-P NPs. All data sets were compared with a negative control of HBSS (pH 7.4), considered as 100% proliferation. These results clearly revealed more potent cytotoxic effect of the developed nanosystem in the breast cancer cells. The lower effect of the MTX-loaded UnTHCPSi NPs can be attributed to the low cellular uptake of the particles. In addition, the higher proliferation inhibition of the MTX-Un-P-P in the MDA-MB-231 cells compared to the MCF-7 is attributed to the higher cellular internalization of the NPs in the former one. Since the anticancer effect of the MTX is related to its potency to penetrate inside the cells and inhibit folic acid reductase and DNA synthesis during cellular replication [66,67], higher therapeutic efficiency of the Un-P-P NPs can be probably described by the ability of the nanocarriers to deliver higher amount of the MTX inside the cells, resulting in higher cellular toxicity effect. To further elucidate the high potential of the

Fig. 10. Proliferation of MDA-MB-231 and MCF-7 cells after exposure to free MTX, MTX-UnTHCPSi and MTX-Un-P-P for 6 h at 37 C. In MDA-MB-231 cells, free MTX and MTXUnTHCPSi showed a mild concentration-independent cell proliferation reduction and MTX-Un-P-P revealed potent concentration-dependent cell proliferation decrease. In MCF7 cells, while free drug and MTX-UnTHCPSi showed very high cell proliferation, MTX-Un-P-P induced a moderate cell proliferation reduction effect independent of the concentration tested. Data are expressed as the mean of three independent experiments ± s.d. (n 3).


12


developed nanosystems, the cell proliferation assay was also performed based on different concentrations of the drug loaded NPs or in the absence of MTX, shown in the Figure S3 of Supporting Information. These results represented the outstanding suppressor potential of the drug loaded polymer-conjugated NPs on cell proliferation. 4. Conclusion Surface chemistry modification of NPs is crucial for further surface functionalization, reactivity adjustments, toxicity reduction, stability enhancement, and cellular interactions regulation. Here, we have reported an easy and versatile method for preparing efficient zwitterionic PSi NPs by covalent conjugation of the PEI and PMVE-MA polymers onto the surface of UnTHCPSi NPs. The results revealed that the polymer-conjugated NPs not only show relatively small size, reduced PDI, and favorable colloidal and human plasma stability, but also possess a high potential for intracellular drug delivery in breast cancer cells, owing to the feasibility of the NPs to confine the anticancer drug in the pores of the particles and also between the free spaces of the polymeric network. Furthermore, enhanced bioadhesive functionality and cellular association, low toxicity, improved endosomal escape via the free positive charges of the PEI concealed under the PMVE-MA layer, as well as sustained drug release due to the new donated physicochemical properties of the NPs after polymer conjugation were among the most promising features of the developed nanocomposites. Overall, we demonstrated that the interaction of polymerconjugated PSi NPs with cells is not only dependent on the subtle structural changes on the surface of the PSi NPs using the polymers, but also on how the functional groups of polymers are spatially arranged on the surface and between different polymer layers. Overall, the results provide a proof-of-concept evidence that zwitterionic polymer-based PSi nanocomposites can be extensively used as a promising candidate for cytosolic drug delivery. Acknowledgments Dr. H. A. Santos acknowledges financial support from the Academy of Finland (decision nos. 252215 and 256394), the University of Helsinki Research Funds, the Biocentrum Helsinki and the European Research Council under the European Union's Seventh Framework Programme (FP/2007e2013) grant no. 310892. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.biomaterials.2014.05.020. References [1] Zhao HL, Xue C, Du JL, Ren M, Xia S, Cheng YG, et al. Sustained and cancer cell targeted cytosolic delivery of Onconase results in potent antitumor effects. J Control Release 2012;159:346e52. €kil€ [2] Ma a 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. €nen J, M€ € E, Laaksonen TJ, Laaksonen P, [3] Sarparanta M, Bimbo LM, Rytko akila et al. Intravenous delivery of hydrophobin-functionalized porous silicon nanoparticles: stability, plasma protein adsorption and biodistribution. Mol Pharmaceutics 2012;9:654e63. [4] 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. [5] Shahbazi MA, Herranz B, Santos HA. Nanostructured porous Si-based nanoparticles for targeted drug delivery. Biomatter 2012;2:296e312. € E, Laaksonen T, Lehto VP, Salonen J, Hirvonen J, et al. Drug [6] Bimbo LM, M€ akila permeation across intestinal epithelial cells using porous silicon nanoparticles. Biomaterials 2011;32:2625e33.

€kila € E, Laaksonen T, [7] Bimbo LM, Sarparanta M, Santos HA, Airaksinen AJ, Ma et al. Biocompatibility of thermally hydrocarbonized porous silicon nanoparticles and their biodistribution in rats. ACS Nano 2010;4:3023e32. [8] Varkouhi AK, Scholte M, Storm G, Haisma HJ. Endosomal escape pathways for delivery of biologicals. J Control Release 2011;151:220e8. [9] Boeneman K, Delehanty JB, Blanco-Canosa JB, Susumu K, Stewart MH, Oh E, et al. Selecting improved peptidyl motifs for cytosolic delivery of disparate protein and nanoparticle materials. ACS Nano 2013;7:3778e96. [10] Febvay S, Marini DM, Belcher AM, Clapham DE. Targeted cytosolic delivery of cell-impermeable compounds by nanoparticle-mediated, light-triggered endosome disruption. Nano Lett 2010;10:2211e9. [11] Chan CL, Majzoub RN, Shirazi RS, Ewert KK, Chen YJ, Liang KS, et al. Endosomal escape and transfection efficiency of PEGylated cationic liposome-DNA complexes prepared with an acid-labile PEG-lipid. Biomaterials 2012;33:4928e35. [12] Ke JH, Young TH. Multilayered polyplexes with the endosomal buffering polycation in the core and the cell uptake-favorable polycation in the outer layer for enhanced gene delivery. Biomaterials 2010;31:9366e72. [13] Verma A, Stellacci F. Effect of surface properties on nanoparticle-cell interactions. Small 2010;6:12e21. [14] Rennert R, Wespe C, Beck-Sickinger AG, Neundorf I. Developing novel hCT derived cell-penetrating peptides with improved metabolic stability. Biochim Biophys Acta 2006;1758:347e54. [15] Haas AK, Maisel D, Adelmann J, von Schwerin C, Kahnt I, Brinkmann U. Human-protein-derived peptides for intracellular delivery of biomolecules. Biochem J 2012;442:583e93. [16] Huhn D, Kantner K, Geidel C, Brandholt S, De Cock I, Soenen SJ, et al. Polymercoated nanoparticles interacting with proteins and cells: focusing on the sign of the net charge. ACS Nano 2013;7:3253e63. [17] Bhattacharyya S, Wang H, Ducheyne P. Polymer-coated mesoporous silica nanoparticles for the controlled release of macromolecules. Acta Biomater 2012;8:3429e35. [18] McInnes SJ, Irani Y, Williams KA, Voelcker NH. Controlled drug delivery from composites of nanostructured porous silicon and poly(L-lactide). Nanomedicine (Lond) 2012;7:995e1016. [19] Appelhans D, Komber H, Quadir MA, Richter S, Schwarz S, van der Vlist J, et al. Hyperbranched PEI with various oligosaccharide architectures: synthesis, characterization, ATP complexation, and cellular uptake properties. Biomacromolecules 2009;10:1114e24. [20] Lee SY, Huh MS, Lee S, Lee SJ, Chung H, Park JH, et al. Stability and cellular uptake of polymerized siRNA (poly-siRNA)/polyethylenimine (PEI) complexes for efficient gene silencing. J Control Release 2010;141:339e46. [21] Aravindan L, Bicknell KA, Brooks G, Khutoryanskiy VV, Williams AC. Effect of acyl chain length on transfection efficiency and toxicity of polyethylenimine. Int J Pharm 2009;378:201e10. [22] Florea BI, Meaney C, Junginger HE, Borchard G. Transfection efficiency and toxicity of polyethylenimine in differentiated Calu-3 and nondifferentiated COS-1 cell cultures. AAPS PharmSci 2002;4:E12. [23] Hamidi M, Shahbazi MA, Rostamizadeh K. Copolymers: efficient carriers for intelligent nanoparticulate drug targeting and gene therapy. Macromol Biosci 2012;12:144e64. [24] Shahbazi MA, Hamidi M. The impact of preparation parameters on typical attributes of chitosan-heparin nanohydrogels: particle size, loading efficiency, and drug release. Drug Dev Ind Pharm 2012;39:1774e82. [25] Shahbazi MA, Hamidi M, Mohammadi-Samani S. Preparation, optimization, and in-vitro/in-vivo/ex-vivo characterization of chitosan-heparin nanoparticles: drug-induced gelation. J Pharm Pharmacol 2013;65:1118e33. €kila € E, Correia A, Ferreira MP, Kaasalainen M, [26] Shahbazi MA, Almeida PV, Ma et al. Poly(methyl vinyl ether-alt-maleic acid)-functionalized porous silicon nanoparticles for enhanced stability and cellular internalization. Macromol Rapid Commun 2014;35:624e9. €kila € EM, Zhang H, Almeida PV, Kaasalainen M, [27] Shahbazi MA, Hamidi M, Ma et al. The mechanisms of surface chemistry effects of mesoporous silicon nanoparticles on immunotoxicity and biocompatibility. Biomaterials 2013;34: 7776e89. € E, Heikkila € T, Laaksonen T, et al. [28] Santos HA, Riikonen J, Salonen J, M€ akila In vitro cytotoxicity of porous silicon microparticles: effect of the particle concentration, surface chemistry and size. Acta Biomater 2010;6:2721e31. €kil€ [29] Zhang H, Liu D, Shahbazi M-A, Ma a E, Herranz-Blanco B, Salonen S, et al. Fabrication of a multifunctional nano-in-micro drug delivery platform by microfluidic templated encapsulation of porous silicon in polymer matrix. Adv Mater; 2014. http://dx.doi.org/10.1002/adma.201400953. € E, Sakka T, Salonen J, Ogata YH. Thermally promoted [30] Jalkanen T, M€ akila addition of undecylenic acid on thermally hydrocarbonized porous silicon optical reflectors. Nanoscale Res Lett 2012;7:311. [31] Muhammad F, Guo M, Qi W, Sun F, Wang A, Guo Y, et al. pH-Triggered controlled drug release from mesoporous silica nanoparticles via intracelluar dissolution of ZnO nanolids. J Am Chem Soc 2011;133:8778e81. [32] Lu J, Liong M, Zink JI, Tamanoi F. Mesoporous silica nanoparticles as a delivery system for hydrophobic anticancer drugs. Small 2007;3:1341e6. [33] Popat A, Hartono SB, Stahr F, Liu J, Qiao SZ, Qing Max Lu G. Mesoporous silica nanoparticles for bioadsorption, enzyme immobilisation, and delivery carriers. Nanoscale 2011;3:2801e18. [34] Singh N, Karambelkar A, Gu L, Lin K, Miller JS, Chen CS, et al. Bioresponsive mesoporous silica nanoparticles for triggered drug release. J Am Chem Soc 2011;133:19582e5.


M.-A. Shahbazi et al. / Biomaterials xxx (2014) 1e13 [35] Miles WC, Goff JD, Huffstetler PP, Reinholz CM, Pothayee N, Caba BL, et al. Synthesis and colloidal properties of polyether-magnetite complexes in water and phosphate-buffered saline. Langmuir 2009;25:803e13. [36] Ayala V, Herrera AP, Latorre-Esteves M, Torres-Lugo M, Rinaldi C. Effect of surface charge on the colloidal stability and in vitro uptake of carboxymethyl dextran-coated iron oxide nanoparticles. J Nanopart Res 2013;15:1e14. [37] French RA, Jacobson AR, Kim B, Isley SL, Penn RL, Baveye PC. Influence of ionic strength, pH, and cation valence on aggregation kinetics of titanium dioxide nanoparticles. Environ Sci Technol 2009;43:1354e9. [38] Muller M, Kessler B, Houbenov N, Bohata K, Pientka Z, Brynda E. pH dependence and protein selectivity of poly(ethyleneimine)/poly(acrylic acid) multilayers studied by in situ ATR-FTIR spectroscopy. Biomacromolecules 2006;7: 1285e94. [39] Hu X, Ji J. Covalent layer-by-layer assembly of hyperbranched polyether and polyethyleneimine: multilayer films providing possibilities for surface functionalization and local drug delivery. Biomacromolecules 2011;12:4264e71. [40] Sarkar K, Debnath M, Kundu PP. Preparation of low toxic fluorescent chitosangraft-polyethyleneimine copolymer for gene carrier. Carbohydr Polym 2013;92:2048e57. [41] Elizondo E, Sala S, Imbuluzqueta E, Gonzalez D, Blanco-Prieto MJ, Gamazo C, et al. High loading of gentamicin in bioadhesive PVM/MA nanostructured microparticles using compressed carbon-dioxide. Pharm Res 2011;28: 309e21. [42] Stuart BH. Infrared spectroscopy: fundamentals and applications. John Wiley & Sons Ltd; 2004. pp. 80e2. € nen ML, Ma €kila € EM, Kaasalainen MH, Rivinoja A, Salonen JJ, [43] Kinnari PJ, Hyvo et al. Tumour homing peptide-functionalized porous silicon nanovectors for cancer therapy. Biomaterials 2013;34:9134e41. [44] Aggarwal P, Hall JB, McLeland CB, Dobrovolskaia MA, McNeil SE. Nanoparticle interaction with plasma proteins as it relates to particle biodistribution, biocompatibility and therapeutic efficacy. Adv Drug Deliv Rev 2009;61: 428e37. [45] Xie J, Xu C, Kohler N, Hou Y, Sun S. Controlled PEGylation of monodisperse Fe3O4 nanoparticles for reduced non-specific uptake by macrophage cells. Adv Mater 2007;19:3163e6. [46] Xia T, Kovochich M, Liong M, Meng H, Kabehie S, George S, et al. Polyethyleneimine coating enhances the cellular uptake of mesoporous silica nanoparticles and allows safe delivery of siRNA and DNA constructs. ACS Nano 2009;3:3273e86. [47] Kim J, Kim HS, Lee N, Kim T, Kim H, Yu T, et al. Multifunctional uniform nanoparticles composed of a magnetite nanocrystal core and a mesoporous silica shell for magnetic resonance and fluorescence imaging and for drug delivery. Angew Chem Int Ed Engl 2008;47:8438e41. [48] Zhao F, Zhao Y, Liu Y, Chang X, Chen C. Cellular uptake, intracellular trafficking, and cytotoxicity of nanomaterials. Small 2011;7:1322e37. [49] Walczyk D, Bombelli FB, Monopoli MP, Lynch I, Dawson KA. What the cell “sees” in bionanoscience. J Am Chem Soc 2010;132:5761e8. [50] Ahmed M, Narain R. Cell line dependent uptake and transfection efficiencies of PEI-anionic glycopolymer systems. Biomaterials 2013;34:4368e76. [51] Santos HA, Bimbo LM, Herranz B, Shahbazi MA, Hirvonen J, Salonen J. Nanostructured porous silicon in preclinical imaging: moving from bench to bed side. J Mater Res 2013;28:152e64.

13

[52] Fuller JE, Zugates GT, Ferreira LS, Ow HS, Nguyen NN, Wiesner UB, et al. Intracellular delivery of core-shell fluorescent silica nanoparticles. Biomaterials 2008;29:1526e32. [53] Irache JM, Huici M, Konecny M, Espuelas S, Campanero MA, Arbos P. Bioadhesive properties of Gantrez nanoparticles. Molecules 2005;10:126e45. [54] Cho EC, Xie J, Wurm PA, Xia Y. Understanding the role of surface charges in cellular adsorption versus internalization by selectively removing gold nanoparticles on the cell surface with a I2/KI etchant. Nano Lett 2009;9: 1080e4. [55] Villanueva A, Canete M, Roca AG, Calero M, Veintemillas-Verdaguer S, Serna CJ, et al. The influence of surface functionalization on the enhanced internalization of magnetic nanoparticles in cancer cells. Nanotechnology 2009;20:115103. [56] Wilhelm C, Billotey C, Roger J, Pons JN, Bacri JC, Gazeau F. Intracellular uptake of anionic superparamagnetic nanoparticles as a function of their surface coating. Biomaterials 2003;24:1001e11. [57] Patil S, Sandberg A, Heckert E, Self W, Seal S. Protein adsorption and cellular uptake of cerium oxide nanoparticles as a function of zeta potential. Biomaterials 2007;28:4600e7. [58] Wang D, Huang J, Wang X, Yu Y, Zhang H, Chen Y, et al. The eradication of breast cancer cells and stem cells by 8-hydroxyquinoline-loaded hyaluronan modified mesoporous silica nanoparticle-supported lipid bilayers containing docetaxel. Biomaterials 2013;34:7662e73. [59] Bondar OV, Sagitova AV, Badeev YV, Shtyrlin YG, Abdullin TI. Conjugation of succinic acid to non-ionogenic amphiphilic polymers modulates their interaction with cell plasma membrane and reduces cytotoxic activity. Colloids Surf B Biointerfaces 2013;109:204e11. [60] Pittella F, Zhang M, Lee Y, Kim HJ, Tockary T, Osada K, et al. Enhanced endosomal escape of siRNA-incorporating hybrid nanoparticles from calcium phosphate and PEG-block charge-conversional polymer for efficient gene knockdown with negligible cytotoxicity. Biomaterials 2011;32:3106e14. [61] Tu J, Wang T, Shi W, Wu G, Tian X, Wang Y, et al. Multifunctional ZnPc-loaded mesoporous silica nanoparticles for enhancement of photodynamic therapy efficacy by endolysosomal escape. Biomaterials 2012;33:7903e14. [62] Takemoto H, Miyata K, Hattori S, Ishii T, Suma T, Uchida S, et al. Acidic pHresponsive siRNA conjugate for reversible carrier stability and accelerated endosomal escape with reduced IFNalpha-associated immune response. Angew Chem Int Ed Engl 2013;52:6218e21. [63] Rozema DB, Ekena K, Lewis DL, Loomis AG, Wolff JA. Endosomolysis by masking of a membrane-active agent (EMMA) for cytoplasmic release of macromolecules. Bioconjug Chem 2003;14:51e7. [64] Alexander C. Temperature- and pH-responsive smart polymers for gene delivery. Expert Opin Drug Deliv 2006;3:573e81. [65] Barros S, Mencia N, Rodriguez L, Oleaga C, Santos C, Noe V, et al. The redox state of cytochrome c modulates resistance to methotrexate in human MCF7 breast cancer cells. PLoS One 2013;8:e63276. [66] Genestier L, Paillot R, Quemeneur L, Izeradjene K, Revillard JP. Mechanisms of action of methotrexate. Immunopharmacology 2000;47:247e57. [67] Tian H, Cronstein BN. Understanding the mechanisms of action of methotrexate: implications for the treatment of rheumatoid arthritis. Bull NYU Hosp Jt Dis 2007;65:168e73.


Distinct effects of endosomal escape and inhibition of endosomal trafficking on gene delivery via electrotransfection.

Silicon Nanocrystals and Silicon-Polymer Hybrids: Synthesis, Surface Engineering, and Applications.

Targeting bacterial biofilms via surface engineering of gold nanoparticles.

Nitric oxide-releasing porous silicon nanoparticles.

Surface engineering of porous silicon microparticles for intravitreal sustained delivery of rapamycin.

Dual-drug delivery by porous silicon nanoparticles for improved cellular uptake, sustained release, and combination therapy.

Colloidal stability of zwitterionic polymer-grafted gold nanoparticles in water.

Augmented cellular uptake of nanoparticles using tea catechins: effect of surface modification on nanoparticle-cell interaction.

Manufacturing Techniques and Surface Engineering of Polymer Based Nanoparticles for Targeted Drug Delivery to Cancer.

Colloidal interactions of inorganic nanoparticles grafted with zwitterionic polymer brushes and gels by surface-mediated seeded polymerization.

Pretargeted PET Imaging of trans-Cyclooctene-Modified Porous Silicon Nanoparticles.

Agglomeration of Luminescent Porous Silicon Nanoparticles in Colloidal Solutions.

Synthesis of zwitterionic polymer brushes hybrid silica nanoparticles via controlled polymerization for highly efficient enrichment of glycopeptides.

Surface chemistry dependent immunostimulative potential of porous silicon nanoplatforms.

A Mechanochemical Approach to Porous Silicon Nanoparticles Fabrication.

Cancer-targeting siRNA delivery from porous silicon nanoparticles.

Peptide-modified zwitterionic porous hydrogels for endothelial cell and vascular engineering.

Photoluminescent silicon nanocrystal-polymer hybrid materials via surface initiated reversible addition-fragmentation chain transfer (RAFT) polymerization.

Surface patterning of nanoparticles with polymer patches.

Endocytic pathways and endosomal trafficking: a primer.

Surface-initiated hyperbranched polyglycerol as an ultralow-fouling coating on glass, silicon, and porous silicon substrates.

Poly(methyl vinyl ether-alt-maleic acid)-functionalized porous silicon nanoparticles for enhanced stability and cellular internalization.

Distinct Trafficking of Cell Surface and Endosomal TIM-1 to the Immune Synapse.

Photophysical properties of luminescent silicon nanoparticles surface-modified with organic molecules via hydrosilylation.