Accepted Manuscript Use of Alginate, Chitosan and Cellulose Nanocrystals as Emulsion Stabilizers in the Synthesis of Biodegradable Polymeric Nanoparticles Nicoletta Rescignano, Elena Fortunati, Ilaria Armentano, Rebeca Hernandez, Carmen Mijangos, Rossana Pasquino, José Maria Kenny PII: DOI: Reference:

S0021-9797(14)00985-0 http://dx.doi.org/10.1016/j.jcis.2014.12.032 YJCIS 20075

To appear in:

Journal of Colloid and Interface Science

Received Date: Accepted Date:

9 October 2014 12 December 2014

Please cite this article as: N. Rescignano, E. Fortunati, I. Armentano, R. Hernandez, C. Mijangos, R. Pasquino, J.M. Kenny, Use of Alginate, Chitosan and Cellulose Nanocrystals as Emulsion Stabilizers in the Synthesis of Biodegradable Polymeric Nanoparticles, Journal of Colloid and Interface Science (2014), doi: http://dx.doi.org/ 10.1016/j.jcis.2014.12.032

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Use of Alginate, Chitosan and Cellulose Nanocrystals as Emulsion Stabilizers in the Synthesis of Biodegradable Polymeric Nanoparticles

Nicoletta Rescignano*1, Elena Fortunati2, Ilaria Armentano2, Rebeca Hernandez1, Carmen Mijangos1, Rossana Pasquino3 and José Maria Kenny1,2. 1

Institute of Polymer Science and Technology, CSIC, Juan de la Cierva 3, Madrid 28006, Spain. 2 Materials Engineering Centre, UdR INSTM, University of Perugia, Strada di Pentima , 4, 05100, Terni, Italy. 3 Department of Chemical, Materials and Industrial Engineering, University of Naples Federico II, P.le Tecchio 80, 80125 Napoli, Italy

Abstract Biopolymeric nanoparticles (NPs) based on a biodegradable poly (DL-Lactide-co-Glycolide) PLGA copolymer matrix combined with alginate, chitosan and nanostructured cellulose crystals as three different natural emulsion stabilizers, were synthesized by a double emulsion (water/oil/water) method with subsequent solvent evaporation. The morphological, thermal, chemical and rheological properties of the novel designed NPs and the effect of the different emulsion stabilizers used during the synthesis were deeply investigated in order to optimize the synthesis procedure and the development of biodegradable nanoparticles based on natural polymers coated with natural polymers. The morphological analysis of the produced nanoparticles showed that all the different formulations presented a spherical shape with smooth surface. Infrared investigations showed that the PLGA copolymer maintained its backbone structure and confirmed the presence of chitosan, alginate and cellulose nanocrystals (CNC) on the nanoparticles surface. The results

obtained suggest that PLGA nanoparticles with CNC as emulsion stabilizer might represent promising formulations opening new perspective in the field of self-assembly of biodegradable nanomaterials for medical and pharmaceutical applications. Keywords:

poly

(DL-Lactide-co-Glycolide),

biodegradable

nanoparticles,

cellulose

nanocrystals, chitosan, sodium alginate, emulsion stabilizer.

*Corresponding author: [email protected] C/ Juan de la Cierva 3, 28006 Madrid, España. Tel: (+34) 915 622 900; Fax: (+34) 915 644 853.

2

1. Introduction Nanoparticles (NPs) are submicron-sized polymeric colloidal particles studied extensively as particulate carriers in several pharmaceutical and medical applications [1, 2]. The sub-micron size of NPs offers a number of distinct advantages over microparticles [3]. Probably the main advantage is the general relatively higher intracellular uptake of nanoparticles compared to microparticles [4]. A number of different polymers, both synthetic and natural, have been utilized in formulating biodegradable NPs [5]. Nanoparticles based on biodegradable polyester (such as polylactic acid (PLA), polylactic-coglycolic acid (PLGA) and polycaprolactone (PCL)) represent one of the most promising candidates for in-vivo diagnosis and treatment of cancer, from preclinical development to clinical translation [1]. The use of these amphiphilic polymers results in the formation of nanoparticles with a hydrophobic core and a hydrophilic shell. The core–shell structure allows them to encapsulate and carry poorly water-soluble drugs [4] and to release these drugs at a sustained rate in the optimal range of drug concentration [6]. PLGA nanoparticles have been mostly prepared by emulsification–diffusion [7], solvent emulsion–evaporation [8], interfacial deposition [9] and nanoprecipitation method [10]. The acidic nature of PLGA monomers is not suitable for drugs or bioactive molecules [11, 12]. However, few approaches to overcome these problems have been developed. For this, PLGA nanomedicine

formulations

are

blended

with

alginate,

chitosan,

pectin

[13],

poly(propylenefumarate) [14] polyvinyalcohol (PVA) [15], poly(orthoester), etc. [16]. Emulsion solvent evaporation techniques are the frequently used methods to produce NPs, wherein a significant amount of PVA is generally employed as stabilizing agent [17, 18]; however, PVA is difficult to remove from the surface of the produced particles. Research efforts on the development of sustainable, biodegradable, and nontoxic solid particle emulsion stabilizers have been carried out since several decades due to the importance of emulsion systems and derived particles in food, cosmetics and pharmaceutical industries. 3

Methods involving surface modifications or the addition of co-surfactant compounds [19, 20] have been developed to produce symmetrical or asymmetrical particles, but such methods increase the use of chemicals and make the process complicated and expensive for large-scale production. Therefore, research efforts are being currently focused on the development of environmentally friendly, bio-based nanocomposites, but few studies, up to now, have described the use of chemicals derived from renewable resources for NP stabilization [21-23]. Cellulose is one of the most widespread biopolymers and is a good candidate due to its sustainability, biodegradability, and nontoxicity. In particular, cellulose nanocrystals (CNC), extracted by hydrolysis of different natural sources, are typically rigid rod-shaped monocrystalline cellulose domains, 1–100 nm in diameter and from tens to hundreds of nanometers in length [24]. They are involved in different applications, as reinforcements in polymer nanocomposites, in the biomedical fields and in optoelectronic devices applications, [25, 26] , and, more recently, their potential applications as stabilizers in chemical reactions, were investigated [27]. Kalashnikova et al. 2011 [28] showed the possibility to use bacterial cellulose nanocrystals as stabilizers at the hexadecane–water interface, promoting monodispersed oil in water droplets around 4 mm in diameter stable for several months. The authors proved the possibility to design and produce nanosized cellulosic particles, which were directly able to stabilize irreversibly an oil-in-water emulsion suggesting also that the nanocrystals can be used to prepare a monodispersed emulsion without any further modification or wrapping process. Furthermore, other polysaccharides have been considered as stabilizer agents in different reactions. Wang et al. prepared PLGA based nanoparticles using chitosan and/or alginate as emulsion stabilizers [29]. The hydrophilic properties of these molecules and their abundant functional groups for further modification make these natural polymers excellent candidates for biomedical applications.

4

Alginate and chitosan are two major naturally occurring polysaccharides which have been widely used as main components for drug delivery systems and tissue scaffolds. Alginate, commercially available as alginate sodium salt and commonly called sodium alginate (Alg), is a linear polysaccharide normally isolated from many strains of marine brown seaweed and algae. Alginate has the ability to bind multivalent cations, leading to the formation of covalent bonds and insoluble hydrogels. This anionic polysaccharide forms strong gels with calcium cations leading to microspheres with good strength and flexibility. Such crosslinking leads to a stiffer polymer system, reducing swelling in water and organic solvents. Alginate has carboxyl groups which may introduce negative charge to the polymer at appropriate pH [30]. Chitosan is a biodegradable polysaccharide extracted from crustacean shells. It has been shown to be non-toxic in a range of toxicity tests, both in experimental animals [31] and humans [32]. Chitosan is a natural cationic polymer obtained by deacetylating chitin comprising copolymers of (1,4)-glucosamine and N-acetyl-D-glucosamine. Furthermore, according to the therapeutic requirements, the NP composition (i.e., use of Alg or Chitosan) can also be considered as a useful mean to tune the drug release rate and the NP interactions with living tissues [33]. NPs are usually dispersed in water for drug delivery and the properties of their suspensions under flow are important to determine their stability and efficiency. Water has unique properties of stabilizing particle-particle interaction, thanks to ion-related electrostatic stabilization. However, other parameters, such as pH, aeration, depletion agents etc. can affect the stability of the suspensions. Rheology is a powerful method to measure the ability of the suspension to flow under external forces [34, 35]. Experimental and theoretical studies have shown that morphological properties of the particles (shape, size, distribution) as well as fluid/particle and particle/particle interactions can strongly influence the fluid rheology [36, 37].

5

In the present work, poly (DL-Lactide-co-Glycolide) copolymer based biopolymeric nanoparticles were synthesized by a double emulsion (water/oil/water) method comparing the effect of alginate, chitosan and nanostructured cellulose crystals as natural emulsion stabilizers on the morphological (in terms of sizes and shapes), thermal, surface, chemical and rheological properties of the produced nanoparticles. The idea is to optimize the procedure of PLGA based NP synthesis using two natural polymers, alginate and chitosan, and a nanostructured natural system, cellulose nanocrystals, as emulsion stabilizers during the NP production. The advantages and drawbacks of the different formulations are deeply investigated and discussed.

Experimental part 2.1 Materials A poly (DL-Lactide-co-Glycolide) copolymer, (PLGA) (I.V. 0.95-1.20dL g-1, Mw 91,600– 120,000 g/mol) ether terminated, with a 50/50 ratio (PLA/PGA) supplied by Absorbables Polymers/Lactel (Durect Corporation), was used as biodegradable polymer for the nanoparticle (NP) preparation. Microcrystalline cellulose (MCC, dimensions of 10-15 μm, bulk density 0.5 g mL-1 at 25 °C), supplied by Sigma Aldrich, was used as start material for cellulose nanocrystal (CNC) synthesis. Sodium alginate (Alg) and chitosan (Chit) were purchased from Sigma Aldrich and used as received. According to the manufacturer, the molecular weight of chitosan is approximately 50000-190000 Da. The molecular weight of sodium alginate was obtained through viscosity measurements by applying the Mark-Houwink equation,

, where

is the

intrinsic viscosity and k and a, are constants for a solvent-polymer pair at 25 °C. For alginate, = 2·10-5 dL g-1 and

=1.0 [39]. The molecular weight thus obtained was 165500 Da for

sodium alginate. 6

2.2 Synthesis of cellulose nanocrystals Cellulose nanocrystals, were obtained by a hydrolysis process starting from the commercial microcrystalline cellulose (MCC) treated by sulphuric acid hydrolysis (64 %wt/wt) at 45 °C for 30 min as previously reported [40]. After acid removal, dialysis and ultrasonic treatment, the resultant cellulose nanocrystal aqueous suspension was approximately 0.5% wt/wt and the hydrolysis procedure yield was ca. 20%. Mixed bed ion exchange resin (Dowex Marathon MR-3 hydrogen and hydroxide form) was added to the CNC water solution and it was thermally stabilized by addition of 1.0 %v/v of 0.25 mol L-1 NaOH. The obtained CNC showed the typical rod-like structure with dimensions ranging from 100 to 200 nm in length and 5-10 nm in width, as previously reported [41]. The synthesized CNC were then used as one of the three emulsion stabilizer for the polymeric nanoparticle preparation.

2.3 Biopolymeric nanoparticle preparation Biodegradable nanoparticles were prepared by double emulsion (water/oil/water) method with subsequent solvent evaporation [42]. Briefly, 0.25 g of biodegradable polymer were dissolved in 5 mL of chloroform, with 2 h magnetic stirring. This solution was emulsified with 2 ml of bi-distilled water using a tip sonicator (VIBRA CELL Sonics mod. VC 750) for 15 minutes. The resulting emulsion was mixed with 2 %wt/v of emulsion stabilizers (Alg, CNC and Chit) in 20 mL of aqueous solution (or acetic acid for Chit), for 30 min with sonication, in order to form the second emulsion. For the solvent evaporation, the second emulsion was transferred in 200 mL of 0.2 %wt/v emulsion stabilizer aqueous solution (or acid acetic solution in the case of Chit) and was magnetically stirred over night at room temperature. The obtained nanospheres were collected by centrifugation (centrifuge Sigma 2-16P) at 11000 rpm for 20 minutes and washed out four times with distilled water in the case of alginate and CNC while,

7

when Chit was employed as emulsion stabilizer, the washings were performed with acetic acid solution (1 %v/v in water). PLGA particles prepared with sodium alginate, cellulose nanocrystals and chitosan as emulsion stabilizer were designated as Alg-PLGANPs, CNC-PLGANPs and Chit-PLGANPs, respectively.

2.4 Characterization methods 2.4.1 Morphological characterization The morphological characterization of the Alg-PLGANPs, CNC-PLGANPs and ChitPLGANPs was performed by a Field Emission Scanning Electron Microscope (FESEM Supra 25, Zeiss). Samples were deposited onto fluorinated tin oxide (FTO) substrates using a dropcasting method, allowing them to dry at room temperature for 24 h and gold-coated by an Agar automatic sputter coater. NP dimensions were calculated by the analysis of the corresponding FESEM images using Image J software. The mean diameter and the particle size distribution were obtained from the analysis of at least 250 individual nanoparticles. Dynamic Light Scattering (DLS) was used for the determination of NPs average diameter and zeta potential employing a Malvern Nano ZS instrument with a 633 nm laser diode and a backscattering detection angle of 173º. Samples were housed in quartz cuvettes of 1 cm optical path length. The electrophoretic mobility was transformed into Zeta potential (ξ) using the Smoluchowski equation. All measurements were repeated three times and the average of three runs was computed.

2.4.2 Thermal and chemical analysis Thermal properties of the Alg-PLGANPs, CNC-PLGANPs and Chit-PLGANPs systems were assessed by differential scanning calorimetry (DSC 822/e Mettler-Toledo, USA). Ten milligrams of dried samples were placed in covered aluminum pans and then placed in the 8

DSC sample holder. For all samples, the measurements were performed in the temperature range from −25 °C to 200 °C, at 10 °C min-1, performing two heating and one cooling scans. The same temperature range was selected for all the proposed formulations in order to have information about the thermal phenomena avoiding, at the same time, any degradation mechanism. The values of the glass transition temperature (Tg) were obtained from the dynamic thermograms, using the midpoint between the intersections of the two parallel baselines, before and after the Tg. Thermogravimetric analysis (TGA) was performed using a quartz rod microbalance (Seiko Exstar 6300, Japon) in the following conditions: sample weight 10 mg, nitrogen flow (250 mL min-1), temperature range 30–600 °C, heating rate 10 °C min-1. Fourier transform infrared (FT-IR) spectroscopy of the Alg-PLGANPs, CNC-PLGANPs and Chit-PLGANPs was carried out with a FT-IR spectrophotometer (Jasco FT-IR 615, Japan). The spectra were performed on KBr disks in the range from 4000 to 400 cm-1 with a 4 cm-1 resolution.

2.4.3 Rheological analysis Nanoparticle dispersions in bidistilled water have been prepared at two different solid contents, 0.5 %w/v and 1 %w/v. The solutions were stirred overnight to ensure homogeneous solutions. Rheological measurements were performed at room temperature using a stresscontrolled rheometer (Bohlin CVO 120, Malvern instruments) equipped with a 60 mm smooth stainless steel disks with cone-plate geometry (cone angle 0.0175 rad), in order to maximize the torque value and allow good quality measurements. A steel cylindrical cage enveloping the cone-plate system has been used to prevent sample drying. Steady tests have been performed to measure the zero shear viscosity of the solutions with different nanoparticles content and type. In all cases, the time scale associated with sedimentation was larger than the time scale associated to the experiment [29]. 9

2.4.4. Bulk Settling experiments Several milligrams of three different lioiphyilised samples were dispersed in distilled water during 24 h with magnetic agitation in order to obtain the uniform dispersion of NPs in the solvent. Observations were made after 30 min and 24 hours con concentration (C=10mg in 5 ml de H2O). The colloidal stability of Alg-PLGANPs, Chit-PLGANPs dispersion was determined by measuring the variation of average diameter with the time, at 25ºC. Seven consecutive measurements were taken and then averaged and plotted in figure 6.

3. Results and discussion 3.1 Determination of particle size and zeta potential of the synthesized nanoparticles The combination of natural polymers such as alginate, chitosan or nanostructured cellulose nanocrystals with biodegradable PLGA for the formation of biopolymeric nanoparticles offer some interesting advantages: first of all the hydrophilic and the hydrophobic behaviors of the polymers (emulsion stabilizers and PLGA, respectively) provide peculiar properties that facilitate the interaction with the biological environment and hydrophobic drugs. Particle size and their surface properties play a key role in determining the drug release and cellular uptake as well as their in vivo pharmacokinetics and biodistribution. Furthermore, the characterization of the particles size facilitates understanding of the dispersion and aggregation processes. FESEM images corresponding to the three samples under study (Figure 1), Alg-PLGANPs, Chit-PLGANPs and CNC-PLGANPs, demonstrate the successful preparation of PLGA particles by the double emulsion technique. In all cases, PLGA particles are characterized by a smooth surface and a spherical shape. In addition, it is evident that the particle size is related to the kind of emulsion stabilizer employed. The histograms corresponding to each of the prepared samples are also shown in Figure 1. Alg-PLGANPs show a broad particle size 10

distribution centered at ~1 µm and with particle sizes ranging from ~250 nm to ~ 2 µm. Some particles with diameters higher than ~ 2 µm are also observed. In contrast, a different behavior was detected in the case of Chit-PLGANPs for which the particle size distribution is much narrower with an average particle size of ~400 nm. From these results, it is clear that the employment of chitosan as emulsion stabilizer in the preparation of the PLGA based nanoparticles by double emulsion constitutes a more efficient strategy to achieve particles in the nanometer range. The results could be explained in terms of the difference in the molecular weight of the two polymeric emulsion stabilizers as the chitosan employed in this work has a molecular weight which is ~6 times lower than the corresponding to sodium alginate as determined through viscosity measurements. As previously reported [43], a emulsion stabilizer with a high molecular weight may increase the viscosity of the emulsion resulting in an increased difficulty in breaking up the emulsion into smaller droplets and yielding bigger particles in agreement with our results. Insert Fig.1: Figure1: FESEM images a) Alg-PLGANPs, b) Chit-PLGANPs, and c) CNC-PLGANPs. The corresponding histograms are also shown

Interestingly, the employment of a polymeric nanostructured emulsion stabilizer such as cellulose nanocrystals also yields PLGA particles with a narrow particle size distribution centered at ~560 nm. As previously reported, the rod-like shape of cellulose nanocrystals gives rise to a percolating network which yields a high stabilization at the oil–water interface occurring in emulsion processes [44]. Therefore we can assume that CNC, once confined at the interface of the oil-in-water emulsion for the preparation of PLGA particles, might associate among them according to an aggregative process. This might increase the cohesion and stability of the PLGA droplets formed in the process of double emulsion thus increasing the efficiency of CNCs as emulsion stabilizer. 11

The Zeta (ξ) potential, which is the electrostatic potential that exists at the shear plane of a particle, is related to both surface charge and the local environment of the particle. Zeta potential is commonly used as an important parameter in colloid science to understand the colloid electrostatic interactions [45, 46] . In fact, zeta potential can greatly influence particle stability in suspension through the electrostatic repulsion between particles. The surface charge values may be positive or negative depending upon the nature of the polymer or the material used for surface modification. The zeta potential is an important parameter in the examination of the stability of nanoparticle dispersions in aqueous media. The zeta potential values of the particles are: -31.4±0.5 mV for Alg-PLGANPs, 42.9±2 mV for Chit-PLGANPs and -41±0.5mV for CNC-PLGANPs. The positive zeta potential found on the surface of the Chit-PLGANPs is due to the cationic characteristics of the chitosan chains and it is also affected by the chitosan concentrations used [47]. The negative charge in the presence of alginate on the particles surface was detected as expected from polyanions; the negative charge of the alginate polymer originates from the negative carboxyl (-CO2=) groups [48]. The cellulose nanocrystals confer a negative charge to the PLGA NPs. [49, 50], since CNC suspensions in neutral water present a negative zeta potential, as previously reported [51]. The negative values were ascribed to the presence of a larger amount of negatively charged sulfate groups on the surface of the cellulose induced by the acid hydrolysis reaction [52-54] .

3.2 Thermal and chemical analysis Thermal and chemical analyses of PLGA based nanoparticles synthesized with the three different emulsion stabilizers were carried out in order to evaluate the effect of nanostructured cellulose crystals, and alginate or chitosan polymers on the final properties of the biopolymeric nanoparticles.

12

DSC analysis was used to investigate the glass transition and crystallization/melting phenomena of polymeric NPs in relation to the composition and preparation method. Table 1 summarizes the thermal properties during the two heating and the cooling scans for NPs compared with PLGA pellet. DSC analysis demonstrated the amorphous character of PLGA (Tg = 50 °C) maintained also in the case of all PLGA based NPs [42]. The glass transition temperatures of the produced PLGA nanoparticles detected during the first heating scan, show higher values for all the formulations with respect to the cooling and the second heating scans as consequence of the synthesis process that affects the thermal behavior of the NPs. Moreover, a more pronounced enthalpic relaxation peak in the glass transition region (data not shown), commonly ascribed to a physical aging in polymers, was detected at the first heating for all the NP formulations [42]. Moreover, no particular effects of the different emulsion stabilizers on the thermal behavior of PLGA nanoparticles were detected and no crystallization and/or melting phenomena were present in the considered temperature range as a consequence of cellulose, chitosan or alginate introduction. Insert Fig. 2: Figure 2: DTG curves of CNC-PLGANPs (a), Alg-PLGANPs (b) and Chit-PLGANPs (c) compared with PLGA pellet, CNC (a), alginate (b) and chitosan (c) powders.

Thermogravimetric analysis was also applied in order to have information about the thermal stability of the produced PLGA nanoparticles and the influence of different emulsion stabilizers on this property. Figure 3 reports the derivative weight loss curves for CNCPLGANPs, Alg-PLGANPs and Chit-PLGANPs compared with PLGA pellet and cellulose nanocrystals, alginate and chitosan powder thermograms while in Table 1 the maximum degradation (Tmax) and the Tonset temperatures for each material studied were reported. PLGA neat copolymer was characterized by a single step degradation behavior with a main peak centered at (345.7±1.0) °C. 13

Insert Table 1: Table 1: Thermal properties of biopolymeric nanoparticles.

Cellulose nanocrystal powder shows the typical multi-step behaviour with first thermal transitions at low temperature and the main degradation region between 250 and 500°C (initial degradation temperatures at (250.2±1.1) °C and maximum degradation temperature at (263.9±1.0) °C) (Figure 3a). The multi-step behavior was maintained also in the case of CNCPLGANPs with a low intensity peak around 270 °C due to the CNC first-step degradation and a main peak at 361 °C with an increase of Tmax of about 16 °C with respect to the PLGA pellet. The presence of cellulose nanocrystals during the synthesis is able to stabilize the PLGA thermal behavior and this confirms the effectiveness in the use of CNC as emulsion stabilizer as discussed in the previous section. Less stable conditions were detected in the case of Alg-PLGANPs and Chit-PLGANPs (Figure 3b and c) and this effect is more evident in the case of the chitosan based formulation that presents the lowest maximum degradation temperature (289 °C) and the lowest value of initial degradation (about 213 °C). Concerning the chitosan degradation behavior (Figure 3c and Table 1), the thermal degradation of pure Chit is characterized by two significant weight losses as previously reported: the first one corresponds to moisture vaporization, the loss of adsorbed and bound water and the residues of acetic acid [55-57]; the second and main weight loss, which was observed at 300–400 °C, can be attributed to the degradation of the saccharide structure of chitosan molecule, including the dehydration of saccharide rings and the polymerization and decomposition of the acetylated and deacetylated units of chitin [58]. After the combination with PLGA, a decrease of the main degradation peak of Chit-PLGANPs is detected comparing with both PLGA and chitosan pellets. This fact could be attributed to the process of lyophylization employed to dry the samples. In this process, stress induced by water crystallization could induce breakage of the polymer chains thus giving rise to a decrease of their thermal stability[59].

14

FT-IR technique was employed to obtain a better understanding of the chemistry of the synthesized PLGA nanoparticles developed with different natural emulsion stabilizers. As a comparison, the pure polymer materials were also analyzed. Figure 3 reports the infrared spectra of CNC-PLGANPs, Alg-PLGANPs and Chit-PLGANPs in the 4000-2500 cm-1 (a) and 2300-400 cm-1 (b) frequency regions. A comparison between the cellulose nanocrystals and CNC-PLGANPs in the same frequency region is also reported (c,d). All the synthesized nanoparticles show the main peaks attributed to the functional groups of PLGA chemical structure, such as –CH, –CH2, –CH3 (2850–3000 cm−1), carbonyl –C=O (1760 cm−1), C–O (1050–1250 cm−1), ethyl –CH2 (1405–1465 cm−1), methyl –CH3 (1230– 1350 cm−1) and –OH stretching vibrations (3450–3700 cm−1) (Figure 5a and b)[60], confirming that the PLGA polymer maintained its backbone structure during the production of nanoparticles. Insert Fig. 3: Figure 3: FTIR spectra of CNC-PLGANPs, Alg-PLGANPs and Chit-PLGANPs in the 40002500 cm-1 (a) and 2300-400 cm-1 (b) frequency regions. A comparison between the cellulose nanocristals and CNC-PLGANPs in the same frequency region is also reported (c,d).

In Figure 3a all spectra show the absorption peaks at 2999, 2952, 2883 and 1853 cm-1 due to aliphatic C-H stretching vibrations, a new peak is visible only in the case of CNC-PLGANPs due to the cellulose nanocrystal presence, as underlined in the spectra in Figure 5c. In the AlgPLGANPs and Chit-PLGANPs spectra, the bands around 1030 cm−1 (C-O-C stretching) and 950 cm−1 (C-O stretching) are evident and assigned to Alginate and Chitosan saccharide structures respectively. In the CNC-PLGANPs samples the peaks at 1030 cm−1 is shifted at higher wavenumber (1061 cm-1), as shown in the Figure 3d, the phenomena might be explained by a weak interaction between the cellulose nanocrystals with the polymeric chains which might also be responsible for the increased thermal stability of CNC-PLGANPs as 15

previously shown by TGA measurements. Another appreciable change observed in the FT-IR was in the peaks at 3400-3200 cm-1 for -OH and -NH stretching region, where new peaks at 3274 and 3342 cm-1 are detectable in the CNC-PLGANPs samples, due to the CNC introduction.

3.4 Viscosity and settling behavior of aqueous dispersions Viscosity as a function of stress has been measured for all the systems study and zero shear viscosity values were extracted. Figure 4 (a) shows the viscosities as a function of stress corresponding to the aqueous dispersions of Chit-PLGANPs and CNC-PLGANPs measured at two different solid contents. For sake of clarity, the other studied systems are not shown along. In any case, all the solutions analyzed were Newtonian, with an increasing viscosity value with increasing solid content. No shear thinning effect due to the nanoparticle presence has been detected.

Insert Fig. 4: Figure 4: a) Shear viscosity as function of shear rate for two systems (cellulose and chitosan nanoparticle water suspensions) and for two solid loadings (0.5 %w/v and 1 %w/v); b) Normalized viscosity of water suspensions as function of the solid content of nanoparticles.

The fluid viscosity of a suspension usually increases with increasing nanoparticle volume fraction. Even when this fraction is small, the viscosity of the suspension can differ significantly from the low concentration prediction of the Einstein equation [31], which claims that the viscosity of the suspension  depends on the volume concentration  in the following way:

   0 (1  2.5 )

[1]

16

where 0 is the viscosity of the pure suspending medium. The viscosity values of the suspensions ( in equation [1]) have been divided by the viscosity of the pure bi-distilled water ( in equation [1]) and the obtained values, called normalized viscosity, have been plotted as a function of the solid content, as shown in Figure 4 (b). The viscosity increases with the nanoparticles content is evident, whatever type of nanoparticles has been considered, suggesting the hint of a non-aggregating system. The increase of viscosity corresponding to the aqueous dispersions of Chit-PLGANPs and AlgPLGANPs is almost the same and it is in agreement with the Einstein prediction [31]. Aqueous dispersions of CNC-PLGANPs show, instead, a much stronger increase of the viscosity with the concentration of the nanoparticles. Any conclusive explanation for this behavior was already found, but we speculate that this could be dependent on the dispersion quality and stability over the time of the different nanoparticles in water. This is clearly observed in figure 5 where the results corresponding to bulk settling tests performed on aqueous dispersions of the three samples under study are shown. Settling tests provide qualitative information on aggregation rates or cluster properties that can be valuable for practical applications [61, 62]. When alginate or chitosan are used as emulsion stabilizers, aqueous dispersions are not stable and PLGA particles precipitate. In the case of AlgPLGANPs, one possible explanation is the broad particle size distribution obtained and the presence of particles in the micron size range as shown in figure 5. In addition, AlgPLGANPs and Chit-PLGANPs were lyophylized before being re-dispersed in water which might induce difficulties for the re-dispersion of the nanoparticles in aqueous media [63]. This can be explained because during the freezing process previous to lyophilization, small crystals of water are formed and disrupt the stabilizer shell around the particle resulting in clusters in the nanoparticle dispersion which might induce the precipitation of particles of large size [64].

Insert Fig. 5: 17

Figure 5: Stability of Alg-PLGANPs , Chit-PLGANPs and CNC-PLGANPs dispersions at two different times.

A different behaviour was observed for CNC-PLGANPs. The use of cellulose nanocrystals as stabilizer agent induces a good stability of the formed nanoparticles in the aqueous medium and therefore the dispersion remains stable during the test, even for several days, highlighting the positive effect of cellulose nanocrystals as stabilizer for the nanoparticle dispersions. One major difference with the other two stabilizer employed for this study, alginate and chitosan is that CNC, once confined on the surface of the PLGA particles, might associate with each other according to an aggregative process which produces multiple cooperative contacts distributed all along the rod-like cellulose crystals [65]. Recently, in fact, Kalashnikova and co-workers [66] emphasized the ability to stabilize oil-in-water emulsions for cellulose nanocrystals varying in crystalline allomorph, morphology, and hydrolysis processes related to the amphiphilic character of non-hydrophobized cellulose nanocrystals. The authors proved that the charge adjustment made it possible to determine the conditions where a low surface charge density, remains compatible with emulsification, as well as when assisted by charge screening regardless of the source relating the CNC properties to the stability characteristic. This evidence justifies, in the present case, the higher colloidal stability encountered for the aqueous dispersions of CNC-PLGANPs respect to the others and might also explain the increase of the viscosity of the resulting aqueous dispersions with respect to the results found for the other two polymeric stabilizers.

Insert figure 6: Particles size as function of time of Alg-PLGANPs and Chit-PLGANPs

18

A further confirmation of the presence of big particles due to the formation of aggregates and their precipitation in aqueous dispersions of Alg-PLGANPs and Chit-PLGANPs were achieved by measuring the average diameter as a function of time by means of DLS and the results are shown in figure 6. The data obtained from these unstable suspensions show a decrease in particle size with time. The apparent decrease in particle size is due to the settling of the larger particles and aggregates out of the field-of-view of the instrument.

4. Conclusions A promising approach to produce small, spherical and uniform biodegradable polymeric nanoparticles using three different natural emulsion stabilizers like as chitosan, alginate and cellulose nanocrystals has been analyzed in this paper, and the results have been discussed in terms of process parameters. Chitosan and alginate were previously considered as surface modifiers to make oppositely-charged PLGA nanoparticles [26] while bacterial cellulosic nanocrystals were employed to stabilize irreversibly oil in water Pickering emulsion nanocrystals [28]. The novelty of this paper is to use cellulose nanocrystals, obtained by hydrolysis starting from the commercial MCC, for the synthesis of PLGA nanoparticle by using the double emulsion technique. We demonstrated that the particle size is related to the kind of emulsion stabilizer employed. Indeed Alg-PLGANPs show particle size distribution with a maximum at ~1 µm and with particle sizes ranging from ~250 nm to roughly ~ 2 µm. In contrast, in the case of Chit-PLGANPs, the particle size distribution has an average particle size of ~400 nm. The employment of a polymeric nanostructured emulsion stabilizer such as cellulose nanocrystals also yields PLGA particles with a narrow particle size distribution

19

centered at ~560 nm. Cellulose nanocrystals give rise to a percolating network which yields a high stabilization at the oil–water interface occurring in emulsion processes [44]. The viscosity increase with the nanoparticles content is evident, whatever type of nanoparticles has been considered. In particular the aqueous dispersions of CNC-PLGANPs show a much stronger increase of the viscosity [43]. The surface properties of each particle were studied and the results obtained suggest that the use of CNC as emulsion stabilizer for the nanoparticles formation ensures stabilization of nanoparticles in aqueous dissolution, but yet more investigations are needed to develop the newer materials in biomedical and food application. Acknowledgements Authors would like to thank M. Criado for the determination of the molecular weight of the the alginate employed in this study, R Hernández thanks MINECO for a Ramon y Cajal contract and MAT 2011-24794 for Financial support. N. Rescignano thanks the European Project Genis Lab for the financial support.

REFERENCES [1] [2] 113. [3] 135.

A. Kumari, S.K. Yadav, S.C. Yadav, Colloids and Surfaces B: Biointerfaces 75 (2010) 1. C.E. Mora-Huertas, H. Fessi, A. Elaissari, International Journal of Pharmaceutics 385 (2010) S.V. Vinogradov, T.K. Bronich, A.V. Kabanov, Advanced drug delivery reviews 54 (2002)

20

[4] N. Rescignano, L. Tarpani, R. Tiribuzi, S. Montesano, S. Martino, L. Latterini, J.M. Kenny, I. Armentano, Macromolecular Bioscience 13 (2013) 1204. [5] K.S. Soppimath, T.M. Aminabhavi, A.R. Kulkarni, W.E. Rudzinski, Journal of Controlled Release 70 (2001) 1. [6] L. Zhang, J.M. Chan, F.X. Gu, J.-W. Rhee, A.Z. Wang, A.F. Radovic-Moreno, F. Alexis, R. Langer, O.C. Farokhzad, ACS Nano 2 (2008) 1696. [7] D.K. Sahana, G. Mittal, V. Bhardwaj, M.N.V.R. Kumar, Journal of Pharmaceutical Sciences 97 (2008) 1530. [8] M. Zambaux, F. Bonneaux, R. Gref, E. Dellacherie, C. Vigneron, Journal of Controlled Release 60 (1999) 179. [9] C. Pinto Reis, R.J. Neufeld, A.J. Ribeiro, F. Veiga, Nanomedicine: Nanotechnology, Biology and Medicine 2 (2006) 8. [10] J.M. Barichello, M. Morishita, K. Takayama, T. Nagai, Drug development and industrial pharmacy 25 (1999) 471. [11] K.G. Carrasquillo, A.M. Stanley, J.C. Aponte-Carro, P. De Jésus, H.R. Costantino, C.J. Bosques, K. Griebenow, Journal of Controlled Release 76 (2001) 199. [12] M. RaviKumar, S. Mohapatra, X. Kong, P. Jena, U. Bakowsky, C. Lehrd, Journal of nanoscience and nanotechnology 4 (2004) 990. [13] L. Liu, Y.J. Won, P.H. Cooke, D.R. Coffin, M.L. Fishman, K.B. Hicks, P.X. Ma, Biomaterials 25 (2004) 3201. [14] E.L. Hedberg, H.C. Kroese-Deutman, C.K. Shih, R.S. Crowther, D.H. Carney, A.G. Mikos, J.A. Jansen, Biomaterials 26 (2005) 4616. [15] S.D. Patil, F. Papadimitrakopoulos, D.J. Burgess, Diabetes technology & therapeutics 6 (2004) 887. [16] L. Wang, C.-S. Chaw, Y.-Y. Yang, S.M. Moochhala, B. Zhao, S. Ng, J. Heller, Biomaterials 25 (2004) 3275. [17] N. Ahmed, M. Michelin-Jamois, H. Fessi, A. Elaissari, Soft Matter 8 (2012) 2554. [18] D. Ibraheem, M. Iqbal, G. Agusti, H. Fessi, A. Elaissari, Colloids and Surfaces A: Physicochemical and Engineering Aspects 445 (2014) 79. [19] B.P. Binks, J.A. Rodrigues, W.J. Frith, Langmuir 23 (2007) 3626. [20] I. Grosse, K. Estel, Colloid Polym Sci 278 (2000) 1000. [21] H. Ougiya, K. Watanabe, Y. Morinaga, F. Yoshinaga, Biosciences Biotechnology and Biochemistry 61 (1997) 1541. [22] J.J. Blaker, K.-Y. Lee, X. Li, A. Menner, A. Bismarck, Green Chemistry 11 (2009) 1321. [23] A. Yusoff, B.S. Murray, Food Hydrocolloids 25 (2011) 42. [24] E. Fortunati, M. Peltzer, I. Armentano, L. Torre, A. Jiménez, J.M. Kenny, Carbohydrate polymers 90 (2012) 948. [25] E.M. Fernandes, R.A. Pires, J.F. Mano, R.L. Reis, Progress in Polymer Science 38 (2013) 1415. [26] L. Valentini, M. Cardinali, E. Fortunati, L. Torre, J.M. Kenny, Materials Letters 105 (2013) 4. [27] L. Brinchi, F. Cotana, E. Fortunati, J. Kenny, Carbohydrate polymers 94 (2013) 154. [28] I. Kalashnikova, H. Bizot, B. Cathala, I. Capron, Langmuir 27 (2011) 7471. [29] Q. Wang, S. Jamal, M.S. Detamore, C. Berkland, Journal of Biomedical Materials Research Part A 96A (2011) 520. [30] H.A. Patel, in: (Ed.)^(Eds.); Google Patents, 1995. [31] S.B. Rao, C.P. Sharma, Journal of biomedical materials research 34 (1997) 21. [32] T.J. Aspden, J.D. Mason, N.S. Jones, J. Lowe, Ø. Skaugrud, L. Illum, Journal of Pharmaceutical Sciences 86 (1997) 509. [33] F. Ungaro, I. d'Angelo, C. Coletta, R. d'Emmanuele di Villa Bianca, R. Sorrentino, B. Perfetto, M.A. Tufano, A. Miro, M.I. La Rotonda, F. Quaglia, Journal of Controlled Release 157 (2012) 149. [34] T.S. Chow, Physical Review E 48 (1993) 1977. [35] G. Balasubramanian, S. Sen, I.K. Puri, Physics Letters A 376 (2012) 860. [36] C.W. Macosko, Rheology: Principles, Measurements, and Applications. Wiley, 1994. [37] J. Chevalier, O. Tillement, F. Ayela, Physical Review E 80 (2009) 051403. [38] M. Rinaudo, M. Milas, P.L. Dung, International Journal of Biological Macromolecules 15 (1993) 281. 21

[39] K. King, Food Hydrocolloids 8 (1994) 83. [40] E. Fortunati, I. Armentano, Q. Zhou, D. Puglia, A. Terenzi, L.A. Berglund, J. Kenny, Polymer degradation and stability 97 (2012) 2027. [41] E. Fortunati, I. Armentano, Q. Zhou, A. Iannoni, E. Saino, L. Visai, L.A. Berglund, J. Kenny, Carbohydrate polymers 87 (2012) 1596. [42] N. Rescignano, M. Amelia, A. Credi, J. Kenny, I. Armentano, European Polymer Journal 48 (2012) 1152. [43] Y.-Y. Yang, T.-S. Chung, N. Ping Ng, Biomaterials 22 (2001) 231. [44] I. Capron, B. Cathala, Biomacromolecules 14 (2013) 291. [45] R.J. Hunter, Zeta Potential in Colloid Science: Principles and Applications. Academic Press, 1981. [46] Y. Zhang, M. Yang, J.-H. Park, J. Singelyn, H. Ma, M.J. Sailor, E. Ruoslahti, M. Ozkan, C. Ozkan, Small 5 (2009) 1990. [47] M.V. Lorevice, M.R.d. Moura, F.A. Aouada, L.H.C. Mattoso, Journal of nanoscience and nanotechnology 12 (2012) 2711. [48] S.H. Ching, B. Bhandari, R. Webb, N. Bansal, Food Hydrocolloids 43 (2015) 165. [49] M. Yan, S. Li, M. Zhang, C. Li, F. Dong, W. Li, BioResources 8 (2013) 6330. [50] P.B. Filson, B.E. Dawson-Andoh, D. Schwegler-Berry, Green Chemistry 11 (2009) 1808. [51] H. Kargarzadeh, I. Ahmad, I. Abdullah, A. Dufresne, S. Zainudin, R. Sheltami, Cellulose 19 (2012) 855. [52] D. Bondeson, A. Mathew, K. Oksman, Cellulose 13 (2006) 171. [53] X. Dong, J.-F. Revol, D. Gray, Cellulose 5 (1998) 19. [54] M. Roman, W.T. Winter, Biomacromolecules 5 (2004) 1671. [55] S. Tripathi, G. Mehrotra, P. Dutta, International Journal of Biological Macromolecules 45 (2009) 372. [56] C.-H. Chen, F.-Y. Wang, C.-F. Mao, W.-T. Liao, C.-D. Hsieh, International Journal of Biological Macromolecules 43 (2008) 37. [57] K. Lewandowska, Thermochimica Acta 493 (2009) 42. [58] A.T. Paulino, J.I. Simionato, J.C. Garcia, J. Nozaki, Carbohydrate polymers 64 (2006) 98. [59] W. Abdelwahed, G. Degobert, H. Fessi, International Journal of Pharmaceutics 309 (2006) 178. [60] N. Jalali, F. Moztarzadeh, M. Mozafari, S. Asgari, M. Motevalian, S.N. Alhosseini, Colloids and Surfaces A: Physicochemical and Engineering Aspects 392 (2011) 335. [61] C.G. Serrano, J.J. McDermott, D. Velegol, Nat Mater 10 (2011) 716. [62] S. Witharana, C. Hodges, D. Xu, X. Lai, Y. Ding, Journal of Nanoparticle research 14 (2012) 851. [63] A. Gursoy, L. Eroglu, S. Ulutin, M. Tasyurek, H. Fessi, F. Puisieux, J.P. Devissaguet, International Journal of Pharmaceutics 52 (1989) 101. [64] Y.N. Konan, R. Gurny, E. Allémann, International Journal of Pharmaceutics 233 (2002) 239. [65] A.J. Svagan, A. Musyanovych, M. Kappl, M. Bernhardt, G. Glasser, C. Wohnhaas, L.A. Berglund, J. Risbo, K. Landfester, Biomacromolecules 15 (2014) 1852. [66] I. Kalashnikova, H. Bizot, B. Cathala, I. Capron, Biomacromolecules 13 (2011) 267.

22

Figure and Table Captions Figure 1: FESEM images a) Alg-PLGANPs, b) Chit-PLGANPs, and c) CNC-PLGANPs. The corresponding histograms are also shown. Figure 2: DTG curves of CNC-PLGANPs (a), Alg-PLGANPs (b) and Chit-PLGANPs (c) compared with PLGA pellet, CNC (a), alginate (b) and chitosan (c) powders. Figure 3: FTIR spectra of CNC-PLGANPs, Alg-PLGANPs and Chit-PLGANPs in the 40002500 cm-1 (a) and 2300-400 cm-1 (b) frequency regions. A comparison between the cellulose nanocrystals and CNC-PLGANPs in the same frequency region is also reported (c,d) Figure 4: a) Shear viscosity as function of shear rate for two systems (cellulose and chitosan nanoparticle water suspensions) and for two solid loadings (0.5 %w/v and 1 %w/v ); b) Normalized viscosity of water suspensions as function of the solid content of nanoparticles. Figure 5: Stability of Alg-PLGANPs , Chit-PLGANPs and CNC-PLGANPs dispersions at two different times. Figure 6: Particles size as a function of time of Alg-PLGANPs and Chit-PLGANPs Table 1: Thermal properties of biopolymeric nanoparticles.

23

Fig.1

a Frequency

60

40

20

2 µm

0 500

1000

b

1500

2000

2500

diameter [nm]

100

Frequency

80

60

40

20

2 µm

0 0

c

500

1000

1500

2000

2500

3000

2000

2500

3000

diameter [nm]

100

Frequency

80

60

40

20

2 µm 0 0

500

1000

1500

24

Fig. 2

a) 0,20

0,15

0,10

0,20

PLGA pellet Chit powder Chit-PLGANPs

DTG [g/g min]

PLGA pellet CNC powder CNC-PLGANPs DTG [g/g min]

DTG [g/g min]

c)

b)

0,20

0,15

0,10

0,15

0,10

0,05

0,05

0,05

0,00

0,00

0,00

100

200

300

400

Temperature [°C]

500

600

100

200

300

Temperature [°C]

400

500

600

PLGA pellet Alg powder Alg-PLGANPs

100

200

300

400

500

600

Temperature [°C]

25

Fig. 3

26

Fig. 4

1.10 0.5% cellulose 1% cellulose 0.5% chitosan 1% chitosan

viscosity [mPa.s]

1.3 1.2 1.1 1.0

a

0.9 0.8 200

400

600

800 -1

shear rate [s ]

1000

normalized viscosity [-]

1.4

1.08

CELLULOSE ALGINATE CHITOSAN

1.06 1.04 1.02

b

1.00 0.98 0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

w [% weight/volume]

27

Fig. 5

Alg

Chit CNC

Time 0

30 min

24 hours

28

Fig. 6

29

Table 1:

Materials

PLGA pellet CNC-PLGANPs Alg-PLGANPs Chit-PLGANPs

DSC analysis

TGA analysis

First Heating scan

Cooling scan

Second Heating scan

Tg (°C) 50 51.8±0.1 51.4±0.1 52.4±0.2

Tg (°C) 47 42.4±0.1 42.5±0.6 44.5±0.1

Tg (°C) 50 46.9±0.4 47.2±0.9 46.6±0.2

Tmax (°C) 345.7±1.0 361.0±1.0 317±2.0 289±2.0

30

Graphical abstract

31

Highlights: - Nanoparticles have been prepared using three different emulsion stabilizers - Use of biocompatible components for a potential use in biomedical field - Polymeric nanoparticles obtained from the emulsions method have a size around 250 nm - The properties of systems have studied and compared

32