Nose-to-brain delivery of BACE1 siRNA loaded in solid lipid nanoparticles for Alzheimer’s therapy

Colloids and Surfaces B: Biointerfaces 152 (2017) 296–301

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Nose-to-brain delivery of BACE1 siRNA loaded in solid lipid nanoparticles for Alzheimer’s therapy Giovanna Rassu a , Elena Soddu a , Anna Maria Posadino b , Gianfranco Pintus c , Bruno Sarmento d,e , Paolo Giunchedi a , Elisabetta Gavini a,∗ a

Department of Chemistry and Pharmacy, University of Sassari, via Muroni 23a, 07100, Sassari, Italy Department of Biomedical Sciences, University of Sassari, viale San Pietro 43b, 07100, Sassari, Italy c Department of Biomedical Sciences, College of Health Science, Qatar University, P.O. Box 2713, Doha, Qatar d CESPU, Instituto de Investigac¸ão e Formac¸ão Avanc¸ada em Ciências e Tecnologias da Saúde, Rua Central de Gandra 1317, 4585-116, Gandra-PRD, Portugal e INEB, Instituto de Engenharia Biomédica, Rua do Campo Alegre, 823, 4150-180, Porto, Portugal b

a r t i c l e

i n f o

Article history: Received 1 September 2016 Received in revised form 16 January 2017 Accepted 17 January 2017 Available online 19 January 2017 Keywords: Nose-to-brain transport siRNA Solid lipid nanoparticle Chitosan RVG-9R cell-penetrating peptide

a b s t r a c t We designed a delivery system to obtain an efficient and optimal nose-to-brain transport of BACE1 siRNA, potentially useful in the treatment of Alzheimer’s disease. We selected a cell-penetrating peptide, the short peptide derived from rabies virus glycoprotein known as RVG-9R, to increase the transcellular pathway in neuronal cells. The optimal molar ratio between RVG-9R and BACE1 siRNA was elucidated. The complex between the two was then encapsulated. We propose chitosan-coated and uncoated solid lipid nanoparticles (SLNs) as a nasal delivery system capable of exploiting both olfactory and trigeminal nerve pathways. The coating process had an effect on the zeta potential, obtaining positively-charged nanoparticles, and on siRNA protection. The positive charge of the coating formulation ensured mucoadhesiveness to the particles and also prolonged residence time in the nasal cavity. We studied the cellular transport of siRNA released from the SLNs using Caco-2 as a model of epithelial-like phenotypes. We found that siRNA permeates the monolayer to a greater extent when released from any of the studied formulations than from bare siRNA, and primarily from chitosan-coated SLNs. © 2017 Elsevier B.V. All rights reserved.

1. Introduction Direct nose-to-brain transport of drugs or biologics (proteins, oligonucleotides, or viral vectors) is feasible via the olfactory or trigeminal nerve system [1–5]. Olfactory nerve axons originating in the olfactory bulb (OB) penetrate the cribriform plate and terminate at the apical surface of the olfactory neuroepithelium (1–2 cm2 surface area) [1,3,4]; it is located at the roof of the nasal cavity [6]. Moreover, filaments of the olfactory nerves are present both in the anterior and posterior parts at the middle turbinate. In addition, the respiratory mucosa (150 cm2 surface area) is densely inner-

Abbreviations: BACE1, ␤-secretase responsible for amyloid-␤ (A␤) generation in the brain; BBB, blood−brain barrier; Caco-2, human epithelial colorectal adenocarcinoma cells; C, chitosan coated nanoparticles; CNS, central nervous system; CPPs, cell-penetrating peptides; CS, low molecular weight chitosan; CSF, cerebrospinal fluid; 6-FAM, 6-carboxyfluorescein; NR, uncoated nanoparticles; OB, olfactory bulb; PVA, poly(vinyl alcohol); RVG-9R, Chimeric Rabies Virus Glycoprotein fragment; SLN, solid lipid nanoparticles. ∗ Corresponding author. E-mail address: [email protected] (E. Gavini). http://dx.doi.org/10.1016/j.colsurfb.2017.01.031 0927-7765/© 2017 Elsevier B.V. All rights reserved.

vated by sensory and parasympathetic trigeminal nerves and is even more extensive than the olfactory nerve. Sensory maxillary branches innervate the deepest nasal segments, including the olfactory cleft [4]. Unlike olfactory sensory neurons, the trigeminal nerve endings do not penetrate the mucosal surface. Access of molecules to the dense network of trigeminal nerve endings is limited by their ability to cross the mucosal layer [2,4]. Transfer of substances into the brain occurs by slow intra-axonal transport or by faster transfer along the perineural space surrounding the nerve cells into the cerebrospinal fluid (CSF) and/or into the interstitial fluid of the brain [4]. In fact, some substances may be transported by an intracellular pathway within neurons following adsorptive, receptor-mediated, or non-specific fluid phase endocytosis. If they cross the olfactory epithelium, they can reach the OB by intraneuronal transport; alternatively, intracellular pathways across the respiratory epithelium potentially include endocytosis into peripheral trigeminal nerve processes resulting in intracellular transport to the brainstem [2]. Other substances may cross either the olfactory or respiratory epithelia by paracellular or transcellular transport, reaching the lamina propria and then the cranial compartment or into general circulation [2]. The presence of tight

G. Rassu et al. / Colloids and Surfaces B: Biointerfaces 152 (2017) 296–301

junctions at the epithelial barrier as well as a mucus layer covering the respiratory and olfactory area of the nasal cavity limit paracellular permeability by molecules [3]. The ability to deliver therapeutically-relevant amounts of drugs directly from the nasal cavity to the central nervous system (CNS) to treat neurological diseases is dependent on the availability of efficient drug delivery systems [7]. Generally, nanoparticles increase the drug concentration of the encapsulated drug in the CSF or brain tissue after intranasal administration as compared to the same drug administered as a simple solution [7]. A penetration enhancer and a mucoadhesive polymer are often included in the formulation to increase drug permeability and decrease mucociliary clearance [5,8–14]. Recently, intranasal delivery has been proposed as a means to increase the delivery of small interfering RNA (siRNA) to the CNS [15–19]. These entities are potential therapeutic agents for many genetically-influenced diseases (Alzheimer’s disease, Parkinson’s disease and brain tumors) [20], but the clinical utilization has been limited because they have short half-lives as a result of poor in vivo stability and poor cell membrane permeance/permeability [21]. Moreover, because naked siRNA molecules are water-soluble and carry a net negative charge, they are subject to excretion from the mucosa following administration [17]. A variety of delivery strategies have been developed to overcome these limitations of the clinical utilization of siRNA: cell-penetrating peptides (CPPs) [22,23], cationic nanoemulsion of omega-3 fatty acids [24], methoxy poly(ethylene glycol)/polycaprolactone copolymers conjugated with a cellpenetrating peptide [17], and gelatin nanoparticles [19]. Of these, CPPs have already shown a great ability to improve therapeutic molecule delivery to treat CNS diseases [25]. Neurotoxins are often used as drug carriers to specifically target the CNS [26]. In particular, the 29-amino-acid peptide derived from the rabies virus glycoprotein (RVG) binds to acetylcholine receptors, which are highly expressed in the neuronal cells and the CNS. This peptide is able to conjugate siRNAs instead of using electrostatic complexation between the oligonucleotide and the 9 arginine residues of the RVG peptide (RVG-9R siRNA) [26]. Nicotinic acetylcholine receptor is expressed in the nasal cavity, presumably on intraepithelial trigeminal nerve endings [27] and in the OB [28]. The aim of this work was to propose a nasal delivery system capable of exploiting both olfactory and trigeminal nerve pathways to promote siRNA delivery/transport to the CNS. Specifically:

(1) the complex between RVG-9R and BACE1 siRNA was prepared to protect the oligonucleotide and enhance the intracellular pathway by receptor-mediated endocytosis within the neurons; (2) the complexes in solid lipid nanoparticles (SLNs) were encapsulated to favor their transport across the olfactory epithelium and to increase penetration across the mucosal surface of the respiratory epithelium so that the trigeminal nerve endings could be reached; (3) chitosan was used to modify nanoparticle surfaces to increase paracellular transport by opening the tight junctions and at the same time increase mucoadhesiveness to the system.

The major ␤-secretase responsible for amyloid-␤ (A␤) generation in the brain is BACE1. BACE1 siRNA specifically influence the ␤-cleavage of amyloid precursor proteins and may be a potential therapeutic approach for treating Alzheimer’s disease [29]. Lipid-based materials are the most widely-used biomaterials for nanoparticulate siRNA delivery [28].

297

2. Materials and methods 2.1. Materials Fluorescence-labeled siRNA, targeting the BACE1 gene, was purchased from Sigma-Aldrich (Milan, Italy). The siRNA sequence was as follows: 5 -CUGUUAUCAUGGAGGGCUU-3 (sense). The siRNA was labeled by 6-carboxyfluorescein (6-FAM) at the 3 end of the sense strand (6-FAM−BACE1 siRNA). The chimeric RVG fragment (RVG-9R; sequence: YTIWMPENPRPGTPCDIFTNSRGKRASNGGGGRRRRRRRRR) was purchased from AnaSpec (Seraing, Belgium). Low-molecular-weight chitosan (CS), 82.6% deacetylated, viscosity 62 cP (1 wt% in 1% acetic acid at 25 ◦ C), and polyvinyl alcohol (PVA), 87–90% hydrolyzed, MW 53,820, viscosity 6 cP (4% in H2O at 20 ◦ C), were obtained from Sigma-Aldrich (Milan, Italy), as were DNA-free water and cell culture products. Witepsol E 85 solid triglycerides were kindle-provided by Cremer Oleo (Hamburg, Germany). 2.2. Electrophoretic mobility shift assay 6-FAM−BACE1 siRNA, 100 pmol, was incubated with the RVG9R peptide at 1:0.1; 1:0.5; 1:1; 1:5, and 1:10 molar ratios (siRNA:peptide) for 15 min in dark conditions. The RNA-protein complexes were separated by electrophoresis using a 2% agarose gel in 0.5X TRIS-acetate/EDTA for 30 min at 80 V. Controls were siRNA sans peptide. Qualitative analysis of the oligonucleotide was evaluated through the fluorescence of the 6-FAM−BACE1 siRNA. To identify the peptide, the gel was colored with a Comassie solution (40% methanol, 10% acetic acid, and 0.25% Coomassie R Brilliant Blue; Sigma Aldrich, Milan, Italy) for 3 h under mild oscillation and then bleached (20% methanol and 10% acetic acid) overnight at 4 ◦ C. 2.3. Preparation of nanoparticles SLN were prepared using a modified solvent emulsificationevaporation method based on a w/o/w double-emulsion technique [30–32]. Uncoated nanoparticles (NR) as well as those coated with CS (C) were prepared. Briefly, 200 mg lipid was dissolved in 2 mL dichloromethane, then 0.2 mL RVG-9R/BACE1 siRNA complex solution at a 1:10 molecular ratio was added to the fatty mixture to form the primary emulsion. The emulsion was homogenized using a Bioblock Vibracell sonicator (Fisher Bioblock Scientific, Illkirch, France) for 30 s at 70% amplitude. This primary emulsion was poured into 10 mL PVA (2% w/v) and homogenized again under the same conditions. The resulting double emulsion was placed under magnetic stirring at room temperature to remove the organic solvent by evaporation. A CS solution (1% w/v) was prepared by dissolving CS in water containing 1% (v/v) acetic acid (pH 4.6) and PVA (2% w/v). Nanoparticles were added at 1:1 w/w and the mixture was magnetically stirred overnight, allowing surface layer deposition of the polymer on the particles [33]. The same procedure was followed to prepare unloaded SLNs (NRb and Cb). 2.4. Nanoparticles characterization After production, the nanoparticles were characterized for their average particle size (dm), polydispersity index (PI), and average zeta potential (ZP). All samples were diluted with Milli-Q water to a suitable concentration for both particle size and ZP analyses, which were performed using a 90Plus/90Plus/BI-MAS ZetaPlus (Brookhaven Instruments Corporation, NY, USA). All measurements were performed in triplicate. Morphological characterization of SLNs was obtained using scanning electron microscopy. Samples were mounted onto metal

298

G. Rassu et al. / Colloids and Surfaces B: Biointerfaces 152 (2017) 296–301

stubs and were then vacuum-coated with a layer of gold/palladium before observation in a FEI Quanta 400 FEG SEM microscope (FEI, Hillsboro, OR, USA). 2.5. Evaluation of nanoparticle fluorescence Fluorescence of all prepared formulations was measured on 200-␮L samples using a Tecan GENios Plus micro-plate reader (Tecan Schweiz AG, Männedorf, Switzerland) in a light-protected condition. Excitation and emission wavelengths used for fluorescence quantification were 492 nm and 520 nm, respectively. Calibration curves were created using 6-FAM−BACE1 siRNA and the complex in concentrations ranging from 25 to 200 nM. 2.6. Cell culture and permeability studies Using T25 flasks, Caco-2 cells were grown in DMEM supplemented with 10% (v/v) foetal bovine serum, 1% (v/v) nonessential amino acids, 1% (v/v) L-glutamine, and 1% (v/v) penicillin and streptomycin at 37 ◦ C, under a 5% CO2 water-saturated atmosphere. Upon confluence, the cells were harvested using trypsin–EDTA 1X and used for the experiments as following described. For the permeability study, cells were seeded on the apical chamber of 6 multi-wells TranswellTM plates (1 × 105 cells/ml). Plated cells were maintained under identical conditions, medium was changed every 2 days, and cells used when confluence was observed under an inverted optical microscope. The day of the experiment, the culture medium was removed, and the cell monolayers were washed with 1.5 mL and 2.5 mL of Hank’s Balanced Salt Solution (HBBS) at 37 ◦ C on the apical and basolateral sides, respectively. All permeability experiments were initiated (time zero) by adding both the non-formulated (L) and formulated complexes with SLNs (NR and C) to the apical compartment at the final 6-FAM-BACE1 siRNA concentration of 1.39 ␮g/well in HBBS. At various time intervals (15, 30, 60, and 180 min), 200 ␮L sample was collected from the basolateral compartment and replaced by 200 ␮L of a prewarmed HBSS buffer in order to maintain the total volume. Sample fluorescence was measured using the GENios Plus micro-plate reader on black plates (96 wells, Greiner) at the same wavelengths used for the nanoparticles (␭ex: 492 nm; ␭em: and 520 nm). The amount of 6-FAM-BACE1 siRNA transferred to the basal chamber, at the above-mentioned times points, was expressed as Relative Fluorescence Unit (RFU) and calculated with the following formula: CF =

 SV  TV

× F(t−1) + Ft

where CF = cumulative fluorescence; SV = Volume of sample withdrawn (ml); TV = Total volume in the well (ml); Ft = Fluorescence reading at the time “t”; Ft–1 = Fluorescence reading previous to the time ‘t’. 2.7. Statistical analysis Data were analyzed using the nonparametric Kruskal–Wallis test; individual differences were evaluated using a post hoc Dunn’s multiple comparison test (GraphPad Prism, version 6.02; GraphPad Software Incorporated). When suitable, analysis of variance (ANOVA) followed by the Tukey test was performed. 3. Results 3.1. Electrophoretic mobility shift assay The ideal ratio between the peptide and the oligonucleotide for complex formation was at first investigated. Based on the fluorescence emitted from 6-FAM-BACE1 siRNA, the oligonucleotide could

Fig. 1. (A) Binding capacity based on the fluorescence of 6-FAM-BACE1 siRNA toward the peptide at various molar ratios followed by electrophoretic mobility shift assay. (B) Gel details highlighting the free 6FAM-BACE1 siRNA at those molar ratios too low for binding. (C) Identification of the free RVG-9R and the complex ® with the oligonucleotide after Coomassie Brilliant Blue staining. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.).

Table 1 Average particle size (dm), polydispersity index (PI), and zeta–potential (ZP) of prepared SLNs, given as mean ± standard deviation (SD). Formulation

dm (nm) ± SD

PI ± SD

ZP (mV) ± SD

NR C NRb Cb

335.76 ± 34.81a 358.44 ± 25.89b 419.47 ± 24.36a 469.71 ± 49.07b

0.013 ± 0.00a 0.028 ± 0.02b 0.26 ± 0.09a 0.30 ± 0.04b

−17.31 ± 0.68a,c +10.54 ± 0.75b,c −12.52 ± 0.99a,d +14.47 ± 0.19b,d

P < 0.05: a NR versus (vs) NRb; b C vs Cb; c NR vs C; d NRb vs Cb.

be identified. Qualitative analysis showed that the molar ratios of 1:5 and 1:10 (siRNA:RVG) provided the complexes where all the oligonucleotide employed interacted with the peptide (Fig. 1). At lower ratios (1:0.1, 1:0.5, and 1:1), the siRNA had the same electrophoretic mobility as the control (siRNA-free), indicating that at any of these ratios, the electrostatic interactions with the peptide were insufficient (Fig. 1). Presence of the peptide in the gel was highlighted by ® Coomassie Brilliant Blue staining, allowing the estimation that only when the molar ratio was 1:10 the peptide able to complex with the oligonucleotide, surrounding it completely. Only in this case, in fact, was the peptide linked to the dye (Fig. 1C). For this reason, siRNA seems completely surrounded, and thus protected, by RVG-9R if the molar quantity is at least 10 times that of the oligonucleotide, justifying use of a 1:10 ratio in later stages of this work. 3.2. Nanoparticle characterization Characterization of SLNs is shown in Table 1. The preparative method, based on a w/o/w double-emulsion technique, was found to be simple and fast, allowing SLN suspensions characterized by low PI values and suggesting the monodispersity of the systems. All unloaded nanoparticles were of average size, and PI values were higher than those of loaded particles (P < 0.05). Coating with CS did not significantly modify the mean diameter nor the PI of the nanoparticles (P > 0.05), but it did appear to affect the surface charge. All uncoated nanoparticles were negatively charged, but the presence of CS seemed to impart a positive charge to the particles’ surfaces (Table 1), demonstrating that coating occurred.

G. Rassu et al. / Colloids and Surfaces B: Biointerfaces 152 (2017) 296–301

299

Fig. 2. Scanning electron microscopy shows the morphology of prepared solid lipid nanoparticles: (A) Unloaded and uncoated, NRb, (B) uncoated, NR, (C) unloaded and with chitosan, Cb, and (D) chitosan-coated, C. In all panels, magnification is 75,000 × .

Fig. 3. Fluorescence signals of solid lipid nanoparticle dispersions. All data are presented as the means of triplicate measures ± standard deviations. P < 0.05. a) Unloaded chitosan nanoparticles (Cb) vs. chitosan-coated nanoparticles (C); b) Unloaded and uncoated nanoparticles (NRb) vs. uncoated nanoparticles (NR); c) chitosan-coated (C) vs. uncoated nanoparticles (NR). RFU, relative fluorescence units.

Loading siRNA resulted in a more negative or less positive ZP versus that of uncoated SLNs (P < 0.05 for both coated and uncoated particles). Analysis by scanning electron microscopy confirmed the dimensional properties of the nanoparticles, showing that they were nearly rounded with smooth surfaces (Fig. 2). 3.3. Evaluation of nanoparticle fluorescence As shown in Fig. 3, the fluorescence of uncoated nanoparticles (NR) was much higher than that of chitosan-coated nanoparticles

Fig. 4. Fluorescence signals obtained from basolateral medium at various time intervals. All data are presented as the means of triplicate measures ± standard deviations. P < 0.05. a) non-formulated complex (L) vs. uncoated nanoparticles (NR); b) non-formulated complex (L) vs. chitosan-coated nanoparticles (C); c) non-coated (NR) vs. chitosan-coated nanoparticles (C). RFU, relative fluorescence units.

(C), indicating that the coating screens the fluorescence emission of 6-FAM−BACE1 siRNA. It was absorbed on the particle surface. 3.4. Permeability studies The permeation capacity of 6-FAM−BACE1siRNA alone and released from SLN dispersion through Caco-2 cells was evaluated (Fig. 4). The standalone siRNA does not quickly permeated the cell monolayer. A lag time of 15 min appeared when the siRNA was employed as non-formulated complex (L) or released from uncoated particles (NR), which reached 30 min in the case of the chitosan-coated nanoparticles (C). All tested samples exhibited

300

G. Rassu et al. / Colloids and Surfaces B: Biointerfaces 152 (2017) 296–301

a sinusoidal pattern of release that was nearly superimposable between the uncoated (NR) and chitosan-coated nanoparticles (C), whereas it resulted shifted toward lowest time values in the case of the non-formulated complex. However, the transport, as indicted by the changes in the measured fluorescence signals, was different among the tested samples. In particular, formulations were able to increase significantly the permeation of siRNA through epithelial cells primary at 60 and 180 min. In this context, chitosan coating resulted to be the most efficient at 60 min. Membrane integrity of the cells during the experiments have been detected by running a Trypan Blue Exclusion Test [34].

4. Discussion Gene therapy using siRNA represents a powerful tool for the treatment of a number of brain diseases. In particular, the silencing of the ␤-secretase gene, BACE1, leads to an improvement in the pathogenesis of Alzheimer’s disease [29]. The intranasal route has been explored for successful delivery of various neuropeptides; it is a non-invasive method of administration that allows therapeutic substances direct access to the CNS without the need to cross the blood–brain barrier (BBB) [2]. Renner and co-workers found that after intranasal administration, siRNA reaches the olfactory nerve bundles [16]. Nevertheless, lower membrane permeability acts as a significant limitation to the nasal absorption of polar high-molecular-weight biologics such as siRNA [24]. In addition, excretion in the mucosa following administration can occur because the biologics carry a net negative charge [17]. To surmount these limitations, some authors have proposed various strategies, including CPPs. Although widely suggested to facilitate crossing the BBB, CPP–protein conjugation for targeting intranasal drug delivery to the brain is rarely reported. Recently, Lin and collaborators utilized a low-molecular-weight protamine-CPP to facilitate noseto-brain transport of a protein. This protamine enhances, in vivo, drug exposure to the target tissues, whereas native protein shows tissue diffusion ability. The transcellular route was the major pathway responsible for the CPP–drug diffusion [23]. In this work, our selected CPP was the short peptide derived from the RVG. The RVGspacer-d-arginine, RVG-9R, is positively charged and able to bind negatively charged nucleic acids via charge interaction. Because this interaction is dose-dependent, we studied the optimal molar ratio between RVG-9R and BACE1 siRNA. The result agrees with siRNA binding studies: only the 1:10 molar ratio (siRNA:peptide) seems to result in a completely surrounded BACE1 siRNA, thus providing protection by RVG-9R. This ratio was found optimal for maximal transduction [35]. Furthermore, we chose to use RVG9R so that the intracellular pathway of siRNA could be enhanced by receptor-mediated endocytosis within the trigeminal and olfactory neurons. In fact, this CPP can specifically bind to the nicotinic acetylcholine receptor on neuronal cells [35]. We subsequently encapsulated the RVG-9R and BACE1 siRNA complex in SLNs and then coated them with chitosan. Encapsulation is a technique by which a drug can be included inside polymers avoiding exposition to enzymatic degradation [19,24] and allows transport and delivery of the drug to the olfactory and trigeminal nerves. Because RNA degradation is one of the important barriers to siRNA intracellular delivery, the therapeutic effect is lost [24]. Degradation of the drug occurs in the nasal cavity by various enzymes such as cytochrome-P450, peptidases, and proteases [36]. Prepared SLNs are able to encapsulate BACE1 siRNA. The coating process has an effect both in the ZP, obtaining positively-charged nanoparticles, and in siRNA protection. In fact, fluorescence emission data of SLN dispersions show that a small fraction of siRNA is absorbed on the uncoated particle surface; thus, the data for

chitosan-coated particles is superimposable on those for unloaded formulations. This conclusion was confirmed by ZP. In fact, the ZP of uncoated nanoparticles is more negative than that for unloaded and uncoated nanoparticles. Despite the fact that PVA is classified as a nonionic polymer, its acetate groups gain negative charge in this case [37,38]. The positive charge obtained by the formulation for chitosancoated particles ensures mucoadhesiveness and a prolonged residence time in the nasal cavity. In fact, chitosan coating permits electrostatic interactions to form with the negatively-charged surfaces of epithelial cells to reduce mucociliary clearance [5]. Furthermore, chitosan is unable to induce a transient opening in the tight junctions of nasal epithelia that enhances the paracellular transport of drugs. Therefore, the siRNA that is released can pass in the cerebrospinal fluid through olfactory epithelial cells [39] or the trigeminal nerve endings in the respiratory epithelium [4]. Based on the evidence that CPPs increase the transcellular pathway in a neuronal cell and that chitosan can increase paracellular transport through epithelia, we studied the cell transport of siRNA released from SLNs using Caco-2 as a model for epithelial-like phenotypes [13]. The olfactory epithelium, in fact, is a modified form of respiratory epithelium, which has a pseudostratified epithelium that, apart from the supporting epithelial cells, with villi and basal cells, also contains olfactory receptor cells (the cribriform plate). The siRNA permeates the monolayer when released from the formulation much better than bare siRNA. The intracellular passage appears slow but with high intensity mainly from coated SLNs. Both in vitro and ex vivo, a lag time has previously been observed in the permeation profiles of drugs released from a chitosan formulation [13]. This experiment highlights the importance of an encapsulation approach, as stated by other authors [17,19,22–24]. SLN, after diffusion through the nasal mucus layer, can reach the cells where the encapsulated complex could be released next to the cells. The release kinetic of drug from SLN depends on the amount of the drug entrapped into the lipid matrix and that adsorbed onto the surfaces. During particle production, in fact, partition of hydrophilic molecules from the oil phase to the aqueous surfactant phase occurs. This partition generates systems with different percentages of initial burst release followed by prolonged release [40]. Drug diffusion through the lipid matrix of SLNs and/or drug release by erosion of the lipid matrix are the two possible release mechanisms. Of course both mechanisms can occur. The drug released can diffuse through the cells by (a) the transcellular pathway, receptor-mediated endocytosis (Nicotinic acetylcholine receptor) or passive diffusion, (b) the paracellular pathway through tight junctions between the cells. In addition, SLN, due to their small size, could themselves diffuse via transcellular or paracellular pathways and release the drug inside the cells and/or farther on them. Therefore the total transport through the mucosa would be the resultant of these different pathways. We analyzed only drugs transported through Caco-2 cells; we do not know if nanoparticles were involved in the intracellular transport. However, we can assume that it is, based on literature data. Gartziandia and co-workers suggest that lipid-based nanoparticles could be more appropriate than polymeric nanoparticles for intranasal administration when their transport is desirable [41]. Moreover, they found that nanostructured lipid carrier that were chitosan coated were transported across olfactory cell monolayers much more efficiently than those that were uncoated (22% vs 8%) [41]. Kim and collaborators report that gelatin nanoparticles loaded with siRNA after intranasal delivery to rats were found in extracellular and intracellular compartments of many brain regions, including the OB, cerebral cortex, and striatum. Moreover, the therapeutic potency of the formulation was significantly greater than

G. Rassu et al. / Colloids and Surfaces B: Biointerfaces 152 (2017) 296–301

that of bare siRNA [19]. We are aware that our claims would require confirmation after in vivo animal tests. 5. Conclusion Coated and un-coated SLNs used to deliver BACE1 siRNA intranasally were designed. We hypothesized that this delivery system would be able to overcome all the barriers that prevent efficient and optimal nose-to-brain transport of siRNA. In particular, BACE1 siRNA was complexed with RVG-9R to protect the oligonucleotide and enhance the intracellular nerve pathway when released from SLNs. The complex was loaded in SLNs, and total encapsulation was obtained via chitosan coating, which modified the ZP and the mucoadhesiveness of the system as well as the permeation ability of the siRNA through the epithelial cell, enhancing the intracellular transport. Declaration of interest The authors report no conflicts of interest. Acknowledgments Authors thank Prof Paola Rapelli for donating the Caco-2 cells. This research did not receive specific grants from funding agencies in the public, commercial, or not-for-profit sectors. References [1] A. Mistry, S. Stolnik, L. Illum, Nanoparticles for direct nose-to-brain delivery of drugs, Int. J. Pharm. 379 (2009) 146–157. [2] J.J. Lochhead, R.G. Thorne, Intranasal delivery of biologics to the central nervous system, Adv. Drug Deliv. Rev. 64 (2012) 614–628. [3] C.V. Pardeshi, V.S. Belgamwar, Direct nose to brain drug delivery via integrated nerve pathways bypassing the blood–brain barrier: an excellent platform for brain targeting, Expert Opin. Drug Deliv. 10 (2013) 957–972. [4] P.G. Djupesland, J.C. Messina, R.A. Mahmoud, The nasal approach to delivering treatment for brain diseases: an anatomic, physiologic, and delivery technology overview, Ther. Deliv. 5 (2014) 709–733. [5] G. Rassu, E. Soddu, M. Cossu, E. Gavini, P. Giunchedi, A. Dalpiaz, Particulate formulations based on chitosan for nose-to-brain delivery of drugs. A review, J. Drug Deliv. Sci. Technol. 32 (2016) 77–87. [6] L. Illum, Nasal drug delivery-possibilities, problems and solutions, J. Control. Release 87 (2003) 187–198. [7] A. Mistry, S. Stolnik, L. Illum, Nose-to-brain delivery: investigation of the transport of nanoparticles with different surface characteristics and sizes in excised porcine olfactory epithelium, Mol. Pharm. 12 (2015) 2755–2766. [8] E. Gavini, G. Rassu, T. Haukvik, C. Lanni, M. Racchi, P. Giunchedi, Mucoadhesive microspheres for nasal administration of cyclodextrins, J. Drug Target. 17 (2009) 168–179. [9] E. Gavini, G. Rassu, L. Ferraro, A. Generosi, J.V. Rau, A. Brunetti, P. Giunchedi, A. Dalpiaz, Influence of chitosan glutamate on the in vivo intranasal absorption of rokitamycin from microspheres, J. Pharm. Sci. 100 (2011) 1488–1502. [10] E. Gavini, G. Rassu, V. Ciarnelli, G. Spada, M. Cossu, P. Giunchedi, Mucoadhesive drug delivery systems for nose-to-brain targeting of dopamine, J. Nanoneurosci. 2 (2012) 47–55. [11] E. Gavini, G. Rassu, L. Ferraro, S. Beggiato, A. Alhalaweh, S. Velaga, N. Marchetti, P. Bandiera, P. Giunchedi, A. Dalpiaz, Influence of polymeric microcarriers on the in vivo intranasal uptake of an anti-migraine drug for brain targeting, Eur. J. Pharm. Biopharm. 83 (2013) 174–183. [12] A. Dalpiaz, M. Fogagnolo, L. Ferraro, A. Capuzzo, B. Pavan, G. Rassu, A. Salis, P. Giunchedi, E. Gavini, Nasal chitosan microparticles target a zidovudine prodrug to brain HIV sanctuaries, Antivir. Res. 123 (2015) 146–157. [13] G. Rassu, E. Soddu, M. Cossu, A. Brundu, G. Cerri, N. Marchetti, L. Ferraro, R.F. Regan, P. Giunchedi, E. Gavini, A. Dalpiaz, Solid microparticles based on chitosan or methyl-␤-cyclodextrin: a first formulative approach to increase the nose-to-brain transport of deferoxamine mesylate, J. Control. Release 201 (2015) 68–77. [14] A. Yalcin, E. Soddu, E. Turunc Bayrakdar, Y. Uyanikgil, L. Kanit, G. Armagan, G. Rassu, E. Gavini, P. Giunchedi, Neuroprotective effects of engineered polymeric nasal microspheres containing hydroxypropyl-␤-cyclodextrin on ␤-Amyloid (1–42)–induced toxicity, J. Pharm. Sci. 105 (2016) 2372–2380. [15] I.D. Kim, S.W. Kim, J.K. Lee, Gene knockdown in the olfactory bulb, amygdala, and hypothalamus by intranasal siRNA administration, Korean J. Anat. 42 (2009) 285–292.

301

[16] D.B. Renner, W.H. Frey II, L.R. Hanson, Intranasal delivery of siRNA to the olfactory bulbs of mice via the olfactory nerve pathway, Neurosci. Lett. 513 (2012) 193–197. [17] T. Kanazawa, F. Akiyama, S. Kakizaki, Y. Takashima, Y. Seta, Delivery of siRNA to the brain using a combination of nose-to-brain delivery and cell-penetrating peptide-modified nano-micelles, Biomaterials 34 (2013) 9220–9226. [18] T. Kanazawa, K. Morisaki, S. Suzuki, Y. Takashima, Prolongation of life in rats with malignant glioma by intranasal siRNA/drug codelivery to the brain with cell-penetrating peptide-modified micelles, Mol. Pharm. 11 (2014) 1471–1478. [19] I.-D. Kim, E. Sawicki, H.-K. Lee, E.-H. Lee, H.J. Park, P.-L. Han, K. Kim, H. Choi, J.-K. Lee, Robust neuroprotective effects of intranasally delivered iNOS siRNA encapsulated in gelatin nanoparticles in the post ischemic brain, Nanomedicine 12 (2016) 1219–1229. [20] A. Orlacchio, G. Bernardi, A. Orlacchio, S. Martino, RNA interference as a tool for Alzheimer’s disease therapy, Mini Rev. Med. Chem. 7 (2007) 1166–1176. [21] M.A. Behlke, Progress towards in vivo use of siRNAs, Mol. Ther. 13 (2006) 644–670. [22] F. Said Hassane, A.F. Saleh, R. Abes, M.J. Gait, B. Lebleu, Cell penetrating peptides: overview and applications to the delivery of oligonucleotides, Cell. Mol. Life Sci. 67 (2010) 715–726. [23] T. Lin, E. Liu, H. He, M.C. Shin, C. Moon, V.C. Yang, Y. Huang, Nose-to-brain delivery of macromolecules mediated by cell-penetrating peptides, Acta Pharm. Sin. B 6 (2016) 352–358. [24] S. Yadav, S.K. Gandham, R. Panicucci, M.M. Amiji, Intranasal brain delivery of cationic nanoemulsion-encapsulated TNF␣ siRNA in prevention of experimental neuroinflammation, Nanomedicine 12 (2016) 987–1002. [25] L.-L. Zou, J.-L. Ma, T. Wang, T.-B. Yang, C.-B. Liu, Cell-penetrating peptide-mediated therapeutic molecule delivery into the central nervous system, Curr. Neuropharmacol. 11 (2013) 197–208. [26] E. Soddu, G. Rassu, P. Giunchedi, B. Sarmento, E. Gavini, From naturally-occurring neurotoxic agents to CNS shuttles for drug delivery, Eur. J. Pharm. Sci. 74 (2015) 63–76. [27] H. Alimohammadi, W.L. Silver, Evidence for nicotinic acetylcholine receptors on nasal trigeminal nerve endings of the rat, Chem. Senses 25 (2000) 61–66. [28] K.L. Kozielski, S.Y. Tzeng, J.J. Green, Bioengineered nanoparticles for siRNA delivery, Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. 5 (2013) 449–468. [29] J.H. Stockley, R. Ravid, C. O’Neill, Altered ␤-secretase enzyme kinetics and levels of both BACE1 and BACE2 in the Alzheimer’s disease brain, FEBS Lett. 580 (2006) 6550–6560. [30] M.G. Fuentes, D. Torres, M.J. Alonso, Design of lipid nanoparticles for the oral delivery of hydrophilic macromolecules, Colloids Surf. B: Biointerfaces 27 (2003) 159–168. [31] N. Zhang, Q. Ping, G. Huang, W. Xu, Y. Cheng, X. Han, Lectin-modified solid lipid nanoparticles as carriers for oral administration of insulin, Int. J. Pharm. 327 (2006) 153–159. [32] S. Soares, P. Fonte, A. Costa, J. Andrade, V. Seabra, D. Ferreira, S. Reis, B. Sarmento, Effect of freeze-drying, cryoprotectants and storage conditions on the stability of secondary structure of insulin-loaded solid lipid nanoparticles, Int. J. Pharm. 456 (2013) 370–381. [33] F. Araújo, N. Shrestha, M.-A. Shahbazi, P. Fonte, E.M. Mäkilä, J.J. Salonen, J.T. Hirvonen, P.L. Granja, H.A. Santos, B. Sarmento, The impact of nanoparticles on the mucosal translocation and transport of GLP-1 across the intestinal epithelium, Biomaterials 35 (2014) 9199–9207. [34] T. Mukherjee, E. Squillantea, M. Gillespieb, J. Shao, Transepithelial electrical resistance is not a reliable measurement of the Caco-2 monolayer integrity in Transwell, Drug Deliv. 11 (Jan–Feb (1)) (2004) 11–18. [35] P. Kumar, H. Wu, J.L. McBride, K.-E. Jung, M.H. Kim, B.L. Davidson, S.K. Lee, P. Shankar, N. Manjunath, Transvascular delivery of small interfering RNA to the central nervous system, Nature 448 (2007) 39–43. [36] L. Casettari, L. Illum, Chitosan in nasal delivery systems for therapeutic drugs, J. Control. Release 190 (2014) 189–200. [37] M. Wi´sniewska, The temperature effect on electro kinetic properties of the silica–polyvinyl alcohol (PVA) system, Colloid. Polym. Sci. 289 (2011) 341–344. [38] J. Emami, H. Mohiti, H. Hamishehkar, J. Varshosaz, Formulation and optimization of solid lipid nanoparticle formulation for pulmonary delivery of budesonide using Taguchi and Box-Behnken design, Res. Pharm. Sci. 10 (2015) 17–33. [39] T.E. Finger, V.L. St Jeor, J.C. Kinnamon, W.L. Silver, Ultrastructure of substance P-and CGRP immunoreactive nerve fibers in the nasal epithelium of rodents, J. Comp. Neurol. 294 (1990) 293–305. [40] R.H. Müller, K. Mäder, S. Gohla, Solid lipid nanoparticles (SLN) for controlled drug delivery−a review of the state of the art, Eur. J. Pharm. Biopharm. 50 (2000) 161–177. [41] O. Gartziandia, S.P. Egusquiaguirre, J. Bianco, J.L. Pedraz, M. Igartua, R.M. Hernandez, V. Préat, A. Beloqui, Nanoparticle transport across in vitro olfactory cell monolayers, Int. J. Pharm. 499 (2016) 81–89.