Modulation Of Surface Charge, Particle Size And Morphological Properties

Colloids and Surfaces B: Biointerfaces 44 (2005) 65–73

Modulation of surface charge, particle size and morphological properties of chitosan–TPP nanoparticles intended for gene delivery Quan Gan a,∗ , Tao Wang a , Colette Cochrane a , Paul McCarron b a

School of Chemical Engineering, Queen’s University Belfast, Belfast BT9 5AG, UK b School of Pharmacy, Queen’s University Belfast, Belfast BT9 7BL, UK

Received 14 March 2005; received in revised form 6 June 2005; accepted 8 June 2005

Abstract This work investigates the polyanion initiated gelation process in fabricating chitosan–TPP (tripolyphosphate) nanoparticles in the size range of 100–250 nm intended to be used as carriers for the delivery of gene or protein macromolecules. It demonstrates that ionic gelation of cationic chitosan molecules offers a flexible and easily controllable process for systematically and predictably manipulating particle size and surface charge which are important properties in determining gene transfection efficacy if the nanoparticles are used as non-viral vectors for gene delivery, or as delivery carriers for protein molecules. Variations in chitosan molecular weight, chitosan concentration, chitosan to TPP weight ratio and solution pH value were examined systematically for their effects on nanoparticle size, intensity of surface charge, and tendency of particle aggregation so as to enable speedy fabrication of chitosan nanoparticles with predetermined properties. The chitosan–TPP nanoparticles exhibited a high positive surface charge across a wide pH range, and the isoelectric point (IEP) of the nanoparticles was found to be at pH 9.0. Detailed imaging analysis of the particle morphology revealed that the nanoparticles possess typical shapes of polyhedrons (e.g., pentagon and hexagon), indicating a similar crystallisation mechanism during the particle formation and growth process. This study demonstrates that systematic design and modulation of the surface charge and particle size of chitosan–TPP nanoparticles can be readily achieved with the right control of critical processing parameters, especially the chitosan to TPP weight ratio. © 2005 Elsevier B.V. All rights reserved. Keywords: Chitosan nanoparticles; Nanoparticle surface charge; Nanoparticle morphology; Ionic gelation; Gene delivery

1. Introduction Chitosan is a non-toxic biodegradable polycationic polymer with low immunogenicity. It has been extensively investigated for formulating carrier and delivery systems for therapeutic macrosolutes, particularly genes and protein molecules primarily because positively charged chitosan can be easily complexed with negatively charged DNAs and proteins [1,2]. Chitosan can effectively bind DNA and protect it from nuclease degradation [3,4]. It has advantages of not necessitating sonication and organic solvents for its preparation, therefore minimizing possible damage to DNA during the complexation process. ∗

Corresponding author. Tel.: +44 2890 274463; fax: +44 2890 381753. E-mail address: [email protected] (Q. Gan).

0927-7765/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2005.06.001

Furthermore, there is evidence demonstrating that cationic polymers play an important role in both membrane adhesion [5] and lysosomal escape [6] of the encapsulated DNA, providing a potential explanation for the superiority of polymermediated gene transfer relative to naked DNA administration in many applications. These hybrid DNA–chitosan systems can be classified into two categories which differ in their mechanism of formation and morphology: complexes and nanospheres. Gentle mixing, followed by incubation, of chitosan and DNA solutions generated ‘broad’ distributions of chitosan–DNA particulate complexes with mean sizes between 100 and 600 nm, depending on the molecular weight of the chitosan. Since particle formation was elicited solely by the tropism of the two oppositely charged macromolecules for one another, these particles were termed

66

Q. Gan et al. / Colloids and Surfaces B: Biointerfaces 44 (2005) 65–73

‘complexes’. The simplicity of chitosan–DNA complexes is both an advantage and a drawback. Though such complexes are extremely easy to synthesize, the fact remains that their transfection efficacy is significantly below that of cationic liposomes in vitro and viral vectors in vivo. Aside from N/P ratio and chitosan molecular weight, there remain few parameters in the synthesis protocol which can be modulated in an effort to augment transfection. To address this issue, investigators sought to develop more sophisticated DNA-loaded chitosan nanoparticles. The application of DNA–chitosan nanoparticles has advanced in vitro DNA transfection research, and data have been accumulating that shows their usefulness for gene delivery [7,8]. The therapeutic efficacy of the nanoparticles could be due to their ability to protect the therapeutic agent from degradation due to lysosomal enzymes. Due to their sub-cellular and sub-micron size, chitosan–TPP nanoparticles can penetrate deep into tissues through fine capillaries, cross the fenestration present in the epithelial lining (e.g., liver) [9]. This allows efficient delivery of therapeutic agents to target sites in the body. Also, by modulating nanoparticles characteristics, such as enzymatic degradation rate, size and surface charge density, one can control the release of a therapeutic agent from nanoparticles to achieve desired therapeutic level in target tissue for required duration for optimal therapeutic efficacy [6]. The major drawback associated with using chitosan as non-viral gene delivery system is the relatively low transfection rate in comparison to viral vectors, even though the later has its own limitations in patient safety, difficulty in scaleup production, and possible toxicity, immune responses, and inflammatory responses. It is understood that transfection efficacy of cationic polymers depends primarily on: (i) particle size, which determines their intracellular uptake, different pathways of their uptake, intracellular trafficking and sorting into different intracellular compartments, and (ii) the intensity of particle surface charge which influence their ability to efficiently condensate DNA and interact with cell. Stable and reproducible chitosan nanoparticles were in early days formulated via chemical cross-linking in waterin-oil emulsion system for entraping and delivering drugs [10]. However, the negative effects of cross-linking agents, e.g., glutaraldehyde, on cell viability and the integrity of macromolecular drugs led to the development of preparation method under mild conditions. Preparation methods by ionically cross-linking cationic chitosan with specific polyanions were particularly successful as, aside from its complexation with negatively charged polymers, chitosan has the ability to gel spontaneously on contact with multivalent polyanions due to the formation of inter- and intramolecular cross-linkage mediated by these polyanions. Among some polyanions investigated, tripolyphosphate (TPP) is the most popular because of its non-toxic property and quick gelling ability. The chitosan–TPP nano system exhibits some attractive features which render them promising carriers for the delivery of macromolecules. These features include forma-

tion under mild conditions; homogeneous and adjustable size and a positive surface charge that can be easily modulated and a great capacity for the association of peptides, proteins, oligonucleotides, and plasmids [11]. Therefore, ability to control and modulate the properties of chitosan–TPP nanoparticles, most importantly particle size and density of surface charge, is central in determining gene transfection efficiency. It is important that these characteristic properties be predictably produced and easily modulated in a flexible and reliable nano fabrication process with high yield and particle stability. It is therefore the focus of this paper to report on how systematically manipulating processing parameters in the TPP initiated chitosan gelation to obtain predictable and optimal nanoparticle properties for desired applications in relation to gene/protein delivery. 2. Materials and experimental methods 2.1. Materials Three different molecular weight chitosan, derived from crab shell, in the form of fibrils flakes were obtained from Sigma–Aldrich [Catalogue No. LMW 448869, MMW 448877, HMW 419419]. The degree of deacetylation for the low molecular weight chitosan (LMW Chitosan), medium molecular weight chitosan (MWM chitosan) and high molecular weight chitosan (HMW chitosan) is 86.6%, 84.7%, and 82.5%, respectively. Sodium Tripolyphosphate was purchased from Sigma–Aldrich Chemical Co. Ltd. All other reagents used were of analytical grade. 2.2. Purification of chitosan Since medical applications of animal derived biomaterials entail an inherent risk of protein contamination which has in recent years aroused great awareness and anxiety among the public, drug companies, and the industry regulators, it is of utmost importance to ensure that chitosan intended for medical applications is of the highest purity and free of protein contamination. It is therefore decided to further purify the purchased chitosan materials and examine whether there are changes in chemical as well as physical properties before and after purification. The origin and purity of purchased chitosan material depends on its source, season, and conditions of the chemical deacetylation process, which may vary across different suppliers. Further purification process is crucial to ensure that the starting chitosan material for nanoparticle fabrication possesses the highest purity and integrity. In this work, purchased chitosan materials were subjected to a vigorous purification process which involved mixing the solid chitosan flakes in 1 M NaOH solution, allowing 1 g of chitosan for 10 ml NaOH solution. This solid–liquid mixture was heated and continuously stirred for 2 h at 70 ◦ C, and then filtered using a Buchner funnel. Chitosan was insoluble in the caustic solution, and the recovered flakes were washed thoroughly and dried at 40 ◦ C for 12 h.

Q. Gan et al. / Colloids and Surfaces B: Biointerfaces 44 (2005) 65–73

The NaOH treated chitosan flakes were dissolved in 0.1 M acetic acid solution which was filtered using a filter paper to remove residues of insoluble particles. One molar NaOH solution was used to adjust pH value of the filtrate to pH 8.0, resulting in purified chitosan in the form of white precipitates. The precipitated chitosan was washed thoroughly using deionized water, and the product was vacuum-dried at room temperature for 24 h. The dried samples were used for FT-IR analysis and preparation of the chitosan–TPP nanoparticles. 2.3. Preparation of chitosan–TPP nanoparticles Chitosan solutions of different concentration and molecular weight were prepared by dissolving purified chitosan with sonication in 1% (w/v) acetic acid solution until the solution was transparent. Once dissolved, the chitosan solution was diluted with deionized water to produce chitosan solutions of different concentrations at 0.05%, 0.10%, 0.15%, 0.20%, 0.25%, and 0.30% (weight/volume). Tripolyphosphate was dissolved in deionized water at the concentration 0.7 mg/ml. The chitosan solution was flush mixed with an equal volume of TPP solution and the formation of chitosan–TPP nanoparticles started spontaneously via the TPP initiated ionic gelation mechanism. The nanoparticles were formed at selected chitosan to TPP weight ratios of 3:1, 4:1, 5:1, 6:1 and 7:1. The nanoparticle suspensions were gently stirred for 60 min at room temperature before being subjected to further analysis and applications. 2.4. FT-IR In FT-IR analysis of both purified and raw chitosan samples, transmittance spectra were obtained using a PerkinElmer FT-IR spectrometer (SPECTRUM 1000) fitted with an attenuated total reflectance mode (ATR) cell. The equipment was positioned in a laboratory maintained at 25 ± 1 ◦ C. A small chitosan sample (7.0–9.0 mg) was placed on a NaCl plate and subjected to light within the infrared spectrum. The instrument operated with a resolution of 4 cm−1 and 128 scans were collected for each sample. The IR absorbency scans were analysed between 700 and 4000 cm−1 for changes in the intensity of the sample peaks.

67

Netherlands). One drop of dilute chitosan–TPP nanoparticles solution was syringe placed on a carbon film 300 mesh copper grid, allowing to sit until air-dried. The sample was stained with 1 M uranyl acetate solution for 5 s at 7 ◦ C before viewing on the TEM.

3. Results and discussion 3.1. Purification and characterisation of supplied chitosan material FT-IR was used to identify if there were variations in chemical functional groups present at the surface of chitosan samples of different molecular weight, and to determine variations among purified and unpurified chitosan samples. By comparing the characteristic transmittance spectrum of different chitosan samples, it is possible to ascertain changes in the constituent surface functional groups (e.g., NH2 , CH2 NH) during the purification process, an indication of removal of impurities from the purchased chitosan material. Fig. 1 compares the transmittance spectrum of purified HMW chitosan with the original supplied materials. The contrasting difference in the spectrum evidently demonstrates the changes in surface chemistry of the original supplied chitosan material after purification, indicating possible removal of impurities, such as protein molecules and pigments. No attempts were made in this paper to identify what particular chemical bonds are associated with the spectrum peaks which require additional verification beyond the scope of this study. The variation in the transmittance spectrum for LMW and MMW chitosan samples was much less pronounced than the HMW chitosan before and after purification. Fig. 2 compares the FT-IR transmittance spectrum of purified chitosan samples of different molecular weight. The figure reveals variations in corresponding peak values of the same functional groups among different molecular weight samples, but not significant variations in presence of the functional groups themselves.

2.5. Measurement of size and zeta potential of chitosan–TPP nanoparticles Measurement of physical size, zeta potential and polydispersity (size distribution) of the chitosan–TPP nanoparticles were performed using a 3000HSA Zetasizer (Malvern Instruments, England). 2.6. Morphology observation The morphological characteristics of the nanoparticles were examined using a high resolution TEM (Transmission Electron Microscope) machine (Tecnai F-20, Phillips Co.,

Fig. 1. Representative FT-IR transmittance spectra of purified and unpurified HMW chitosan samples.

68

Q. Gan et al. / Colloids and Surfaces B: Biointerfaces 44 (2005) 65–73

Fig. 2. Representative FT-IR transmittance spectra of purified HMW, MMW and LMW chitosan samples.

The purification study demonstrates convincingly that further purification of supplied chitosan materials is essential especially with high molecular weight chitosans. 3.2. Modulation of particle size Particle size is one of the most significant determinant in mucosal and epithelial tissue uptake of nanoparticles and in the intracellular trafficking of the particles [6]. Smaller size nanoparticles (∼100 nm) demonstrated more than 3fold greater arterial uptake compared to larger nanoparticles (∼275 nm) in an ex vivo canine carotid artery model [12,13] as the smaller nanoparticles were able to penetrate throughout the sub-mucosal layers while the larger size micron-particles were predominantly localized in the epithelial lining. Prabha et al. [14] investigated the gene transfection levels of different size fractions of PLGA nanoparticles and found that the lower size nanoparticle fraction produced a 27-fold higher transfection in COS-7 cells and 4-fold higher transfection in HEK 293 cells for the same dose of nanoparticles. These studies also suggested that uniform particle size distribution are important to enhance the nanoparticle-mediated gene expression. Chitosan’s ability of quick gelling on contact with polyanions relies on the formation of inter- and intramolecular cross-linkages mediated by these polyanions. Nanoparticles are formed immediately upon mixing of TPP and chitosan solutions as molecular linkages were formed between TPP phosphates and chitosan amino groups. Size and size distribution of the chitosan–TPP nanoparticles depend largely on concentration, molecular weight, and conditions of mixing, i.e., stirring or sonication. Fig. 3 shows the effect of chitosan concentration on particles size at three different molecular weight. It demonstrates that the size of HMW chitosan nanoparticles was mostly affected by the increased chitosan solution concentration, and the increase in size with concentration showed a linear relationship within the tested range.

Fig. 3. Effect of chitosan concentration from 0.05% to 0.30% (w/v) on particles size with three different chitosan molecular weight. Chitosan to TPP mass ratio = 5:1, T = 20 ± 1 ◦ C, pH 5.0.

The effect of chitosan to TPP weight ratio on particle size was also very prominent (Fig. 4), showing a linear increase of size with increasing chitosan to TPP weight ratio within the tested chitosan to TPP ratio range. These linear relationships provide a simple processing window for manipulating and optimising the nano size for intended applications. 3.3. Modulation of particle surface charge Chitosan has a rigid crystalline structure through inter- and intra- molecular hydrogen bonding. Chitosan molecules in aqueous solutions adopt extended conformation with a more flexible chain because of the electrostatic charge repulsion between the chains. When chitosan and TPP were mixed with each other in dilute acetic acid, they spontaneously formed compact nano complexes with an overall positive surface charge, and the density of the surface charge is reflected by measured zeta potential values.

Fig. 4. Effect of chitosan to TPP mass ratio from 3:1 to 7:1 on particles size with three different chitosan molecular weight. c = 0.50% (w/v), T = 20 ±1 ◦ C, pH 5.0.

Q. Gan et al. / Colloids and Surfaces B: Biointerfaces 44 (2005) 65–73

Fig. 5. Effect of chitosan to TPP mass ratio on particle zeta potential with three different chitosan molecular weight. c = 0.50% (w/v), T = 20 ± 1 ◦ C, pH 5.0.

It was reported that the ability of nanoparticles to escape the endo-lysosomes was dependent on the surface charge of the nanoparticles [15,16]. Nanoparticles which show transition in their surface charge from anionic at pH 7 to cationic in the acidic endosomal pH (pH 4–5) were found to escape the endosomal compartment whereas the nanoparticles which remain negatively charged at pH 4–5 were retained mostly in the endosomal compartment. Thus, by varying the surface charge, one could potentially be able to direct the nanoparticles either to lysosomes or to cytoplasm [6]. The efficacy of nanoparticles as drug carriers is also closely related to their interaction, predominantly influenced by surface charge, with proteins and enzymes in different body fluids. Calvo et al. [17] analysed the interaction phenomenon between lysozyme, a positively charged enzyme that is highly concentrated in mucosas, and poly-␧-caprolactone coated nanoparticles, and found that the interaction of lysozyme with the nanoparticles and their consequent degradation was highly dependent on their surface charge. The zeta potential of chitosan–TPP nanoparticles increased linearly with increasing chitosan to TPP weight ratio from 3:1 to 7:1 (Fig. 5). Again, this simple linear relationship could be easily explored for modulating the particle surface charge density to facilitate the adhesion properties and transport properties of the nanoparticles. The effect of chitosan concentration on zeta potential was also investigated at a fixed chitosan to TPP weight ratio of 5:1. The results in Fig. 6 show that, unlike the trend of particle size increase with increasing chitosan concentration, the zeta potential decreased with increasing chitosan concentration. 3.4. The effect of solution pH Chitosan is a weak base polysaccharide, having an average amino group density of 0.837 per disaccharide unit [18], and insoluble at neutral and alkaline pH values. In an acidic

69

Fig. 6. Effect of chitosan concentration on particle zeta potential at three different chitosan molecular weight. Chitosan to TPP mass ratio = 5:1, T = 20 ± 1 ◦ C, pH 5.0.

medium, the amine groups will be positively charged, conferring to the polysaccharide a high charge density [19]. Therefore, the surface charge density of chitosan molecules is strongly dependent on solution pH [20,21], and the ionic cross-linking process for the formation of chitosan–TPP nanoparticles is pH-responsive, providing opportunities to modulate the formulation and properties of the chitosan–TPP nanoparticles. To study the effects of changing environmental pH values on nanoparticle size and zeta potential, chitosan–TPP nanoparticles formulated with MMW chitosan at fixed concentration of 0.15% (w/v) and different chitosan to TPP mass ratio between 4:1 and 7:1 were examined at varying chitosan solution pH values. The variations in particle size and zeta potential with chitosan solution pH are shown separately in Fig. 7A and B. The nanoparticles formed at solution pH 5.5 had a smaller size but a higher particle zeta potential. Fig. 7B also demonstrated an interesting trough in zeta potential values at pH 5.0. The surface charge reversal of nanoparticles in the acidic solution of endo-lysosomes is proposed as the mechanism responsible for the endo-lysosomal escape of the nanoparticles into cytoplasmic compartment for effective release and gene expression [22]. Surface charge reversal occurs when protons or hydronium ions from bulk solution are transferred to nanoparticle surface under acidic conditions, resulting an increased surface charge density and zeta potential, which would allow stronger electrostatic interactions between the nanoparticles and tissue cells, leading to localized destabilisation of the cell membrane and escape of nanoparticles into cytoplasmic compartment. Low molecular weight chitosan–TPP nanoparticles produced at solution pH 5.5 were tested for their size and zeta potential response to changing pH values of the residing solution medium by simply adjusting the solution pH to a predetermined value by titration with either 1 M NaOH or 1 M HCl solutions. Fig. 8A and B showed that both measured par-

70

Q. Gan et al. / Colloids and Surfaces B: Biointerfaces 44 (2005) 65–73

Fig. 7. (A) Effect of solution pH on MMW chitosan–TPP particle size at different chitosan to TPP mass ratio. c = 0.15% (w/v), T = 20 ± 1 ◦ C. (B) Effect of solution pH value on MMW chitosan–TPP nanoparticles zeta potential. c = 0.15% (w/v), T = 20 ± 1 ◦ C.

ticle size and zeta potential are very sensitive to the changing pH values of the residing aqueous environment, indicating the surface density of protonised amino groups and the degree of protonisation are reversibly responsive to changing solution pH values. The increase in measured average particle size could be caused mainly by particle aggregation when solution pH value increased, rather than by further growth of the individual particle size after initial formation. The sharp increase in size at pH > 6.0 suggests that the degree of protonisation at surface of the particles were reduced, decreasing electrostatic repulsion between the particles thereby increasing the probability of particle aggregation. The idea of deprotonisation of the particle surface was supported by results presented in Fig. 8B which shows a continual decrease in the positive zeta potential well before the pH value reached pH 6.0. Fig. 8B also shows that an isoelectric point for the chitosan–TPP nanoparticles is located at around pH 9.0. Gelling of the nanoparticle colloidal system could easily occur when the overall particle surface charge is neutral at the isoelectric point. The gelling mechanism in relation to swinging pH values has been investigated for producing smart responsive nanoparticle systems for targeted drug delivery [20,21].

Fig. 8. (A) Responsive particle size change in relation to changing residing solution pH values from 3.2 to 12.2. LMW chitosan–TPP nanoparticles produced at conditions c = 0.50% (w/v), chitosan to TPP mass ratio = 5:1, T = 20 ± 1 ◦ C, pH 5.5. (B) Responsive particle zeta potential change in relation to changing residing solution pH values from 3.2 to 12.2. LMW chitosan–TPP nanoparticles produced at conditions c = 0.50% (w/v), chitosan to TPP mass ratio = 5:1, T = 20 ± 1 ◦ C, pH 5.5.

3.5. Stability of the chitosan–TPP nanoparticle system The chitosan–TPP nanoparticle colloidal system is thermodynamically unstable, especially at unfavourable solution pH conditions and at high particle concentrations, because of high surface energy associated with the nano scale dimensions. Fig. 9 shows size growth kinetics of MMW chitosan–TPP particles at a dilute chitosan concentration 0.15% (w/v). Ionic gelation and growth of the chitosan–TPP nanoparticles were completed within the first 60 min with subsequently slight increase in particle size over the next 24 h. No apparent aggregation of particles was observed during this period at constant temperature and solution pH. However when the initial chitosan concentration was increased over and above a critical concentration, large aggregates formed instantaneously. The large aggregates were observed using the TEM imaging technique (Section 3.6),

Q. Gan et al. / Colloids and Surfaces B: Biointerfaces 44 (2005) 65–73

71

and the critical aggregation concentration was studied using Zetasizer measurement which showed a drastic size increase accompanied by a sudden large reduction in zeta potential at the critical concentration. The critical chitosan concentration for the spontaneous formation of aggregates depends on solution pH and chitosan molecular weight. Tables 1 and 2 show that the critical chitosan concentration for LMW, MMW and HMW chitosan is 0.65%, 0.25%, 0.15% (w/v) at pH 4.0, and 1.00%, 0.85%, 0.75% (w/v) at pH 5.0, respectively. 3.6. Morphological characteristics of chitosan–TPP nanoparticles Fig. 9. Kinetics of chitosan–TPP nanoparticle size growth. c = 0.15% (w/v), chitosan to TPP mass ratio = 5:1, T = 20 ± 1 ◦ C, pH 5.0.

The morphological characteristics of the LMW chitosan– TPP nanoparticles were examined using the TEM technique.

Table 1 Measured particles size and zeta potential at different chitosan molecular weight and concentration LMW chitosan (%) (w/v)

Size (nm)

Zeta (mv)

MMW chitosan (%) (w/v)

Size (nm)

Zeta (mv)

HMW chitosan (%) (w/v)

Size (nm)

Zeta (mv)

0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65

136.2 142.3 152.9 171.2 190.3 203.1 312.7 429.6 578.3 712.4 888.6 997.0 1628.6

48.3 44.2 41.0 39.7 37.3 35.6 33.3 31.6 30.4 28.2 26.9 25.2 14.3

0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65

145.3 150.5 165.2 182.3 2175.4 10016.2 – – – – – – –

43.9 40.3 39.1 37.2 16.2 15.2 – – – – – – –

0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65

155.0 188.9 3884.2 5964.3 6136.2 7854.2 – – – – – – –

37.7 34.8 17.4 17.2 16.7 16.9 – – – – – – –

Chitosan to TPP mass ratio = 4:1, T = 20 ± 1 ◦ C, pH 4.0. Table 2 Measured particles size and zeta potential with different chitosan molecular weight and concentration LMW chitosan (%) (w/v)

Size (nm)

Zeta (mv)

MMW chitosan (%) (w/v)

Size (nm)

Zeta (mv)

HMW chitosan (%) (w/v)

Size (nm)

Zeta (mv)

0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 1.10 1.20

143.2 152.1 159.2 172.8 181.9 189.6 273.2 387.2 519.6 604.3 692.1 726.6 740.6 846.7 858.9 893.6 908.6 938.7 998.6 1819.3 2059.0 2851.6

49.2 46.8 45.6 44.3 42.7 40.7 39.4 37.1 36.8 34.2 33.3 32.7 32.1 30.8 30.0 28.5 26.7 26.2 24.3 13.5 11.2 7.6

0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 1.10 1.20

156.1 163.7 170.7 181.5 192.2 209.8 310.2 427.5 546.3 654.3 721.5 783.2 848.3 881.0 936.8 988.6 2371.0 2742.7 4276.9 – – –

46.8 44.4 40.3 39.8 37.8 35.3 33.8 32.7 32.1 30.4 29.0 28.6 27.5 25.3 25.1 24.8 15.2 12.2 7.6 – – –

0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 1.10 1.20

162.7 176.1 208.9 216.8 234.2 257.0 362.3 478.0 566.6 697.3 748.5 828.0 898.5 976.8 1654.9 2400.5 5116.8 – – – – –

41.3 37.4 36.9 35.1 34.1 33.2 32.6 31.8 31.0 29.5 27.8 26.3 25.9 23.2 15.9 10.6 8.0 – – – – –

Chitosan to TPP mass ratio = 4:1, T = 20 ± 1 ◦ C, pH = 5.0.

72

Q. Gan et al. / Colloids and Surfaces B: Biointerfaces 44 (2005) 65–73

Fig. 10. TEM image of a single LMW chitosan–TPP nanoparticle.

TEM image of single chitosan–TPP nanoparticles (Fig. 10) reveals that the nanoparticles have a size range between 140 and 250 nm which conforms with the size measurement by photon correlation spectroscopy using Zeasizer 3000HAS. Fig. 11 shows the image of an aggregate of four distinctive single particles with clear joining boundaries formed alongside the regular geometry of the proximate polyhedron (pentagon and hexagon) shaped particles. Different to past reported works [9,23], the evidence of the formation of polyhydrons, instead of spheres, at nano-metric scale suggests a nucleation through ionic gelation followed by semi-crystal formation and growth. As previously described in Section 3.5, the formation of large nanoparticle aggregates depends on chitosan concentration and solution pH. Fig. 12 shows the TEM image of a

Fig. 11. TEM image of an aggregate of four single LMW chitosan–TPP nanoparticles with distinctive polyhedron shapes.

Fig. 12. TEM image of a large aggregate of LMW chitosan–TPP nanoparticles.

large aggregate formed with many distinctive single nanoparticles, each still possessing a similar nano-metric dimension as being shown in Fig. 10. Fig. 13 shows a TEM image of a chitosan–TPP particle incorporating protein molecules of bovine serum albumin (BSA). 0.03 mg/ml BSA molecules were added to an equal volume of LMW chitosan solution (1.5%, w/v) in acidic acid and gently mixed for 1 h before TPP solution (0.5%, w/v) was added to make up chitosan to TPP mass ratio at 5:1. The BSA incorporated nanoparticles have a size range of 300–350 nm,

Fig. 13. TEM image of a LMW chitosan–TPP nanoparticles incorporating BSA molecules.

Q. Gan et al. / Colloids and Surfaces B: Biointerfaces 44 (2005) 65–73

doubling the size of chitosan–TPP particles. The BSA incorporated particles possessed a typical spherical shape and smooth surface, which confirms the findings by Xu and Du [23], and Janes et al. [1]. Future work will investigate the encapsulation mechanism and efficiency of protein molecules in the ionic initiated chitosan–TPP nanoparticle formation process, and the release kinetics of protein molecules from the particle. 4. Conclusion The formation of high yield chitosan–TPP nanoparticles with predetermined nano-metric size and surface charge density can be simply manipulated and controlled by varying the key processing conditions of chitosan concentration, chitosan to TPP weight ratio, and solution pH value. Within the tested range of conditions, the increase in particle size and particle zeta potential showed a simple linear relationship with increasing chitosan to TPP weight ratio, but the zeta potential at fixed chitosan to TPP ratio showed a linear decrease with increasing chitosan concentration. Solution pH value and chitosan concentration also had profound influence on the stability of the nanoparticle system, and the critical chitosan concentrations for spontaneous formation of large particle aggregates at pH 5 were found to be 0.65%, 0.25%, 0.15% (w/v) at pH 4.0, and 1.00%, 0.85%, 0.75% (w/v) at pH 5.0 for LMW, MMW and HMW chitosan, respectively. The isoelectric point of the chitosan–TPP nanoparticles was found at around pH 9.0. Morphological study of the nanoparticles formed under different conditions revealed the formation of regularly shaped polyhydron particles, an indication of semicrystallisation mechanism during the particle formation and growth, suggesting the particles were of compact structure with orderly molecular arrangement, the discovery of which bears important implications on gene/protein encapsulation and release mechanisms. Acknowledgement To the European Social Funding programme for providing a scholarship to Colette Cochrane for this project.

References [1] K.A. Janes, P. Calvo, M.J. Alonso, Polysaccharide colloidal particles as delivery systems for macromolecules, Adv. Drug Deliv. Rev. 47 (2001) 83–97. [2] S.C.W. Richardson, H.V.J. Kolbe, R. Duncan, Potential of low molecular mass chitosan as a DNA delivery system: biocompatibility, body distribution and ability to complex and protect DNA, Int. J. Pharm. 178 (1999) 231–243.

73

[3] Z. Cui, R.J. Mumper, Chitosan-based nanoparticles for topical genetic immunization, J. Control. Release 75 (2001) 409–419. [4] L. Illum, I. Jabbal-Gill, M. Hinchcliffe, A.N. Fisher, S.S. Davis, Chitosan as a novel nasal delivery system for vaccines, Adv. Drug Deliv. Rev. 51 (2001) 81–96. [5] K.A. Mislick, J.D. Baldeschwieler, Evidence for the role of proteoglycans in cation-mediated gene transfer, Proc. Natl. Acad. Sci. USA 93 (1996) 12349–12354. [6] J. Panyam, V. Labhasetwar, Biodegradable nanoparticles for drug and gene delivery to cells and tissue, Adv. Drug Deliv. Rev. 55 (2003) 329–347. [7] K. Corsi, F. Chellat, L. Yahia, J.C. Fernandes, Mesenchymal stem cells, MG63 and HEK293 transfection using chitosan–DNA nanoparticles, Biomaterials 24 (2003) 1255–1264. [8] K. Romoren, B.J. Thu, O. Evensen, Immersion delivery of plasmid DNA. II. A study of the potentials of a chitosan based delivery system in rainbow trout (Oncorhynchus mykiss) fry, J. Control. Release 85 (2002) 215–225. [9] S.V. Vinagradov, T.K. Bronich, A.V. Kabanov, Nanosized cationic hydrogels for drug delivery: preparation, properties and interactions with cells, Adv. Drug Deliv. Rev. 54 (2002) 223–233. [10] Y. Ohya, M. Shiratani, H. Kobayashi, T. Ouchi, Release behavior of 5-fluorouracil from chitosan-gel nanospheres immobilizing 5-fluorouracil coated with polysaccharides and their cell specific cytotoxicity, Pure Appl. Chem. A 31 (1994) 629–642. [11] X.Z. Shu, K.J. Zhu, A new approach to prepare tripolyphosphate/chitosan complex beads for controlled drug delivery, Int. J. Pharm. 201 (2000) 51–58. [12] C. Song, V. Labhasetwar, X. Cui, T. Underwood, R.J. Levy, Arterial uptake of biodegradable nanoparticles for intravascular local drug delivery: results with an acute dog model, J. Control. Release 54 (1998) 201–211. [13] M.P. Desai, V. Labhasetwar, G.L. Amidon, R.J. Levy, Gastrointestinal uptake of biodegradable microparticles: effect of particle size, Pharm. Res. 13 (1996) 1838–1845. [14] S. Prabha, W.Z. Zhou, J. Panyam, V. Labhasetwar, Size dependency of nanoparticles-mediated gene transfection: Studies with fractionnated nanoparticles, Int. J. Pharm. 244 (2002) 105– 115. [15] W. Paul, C. Sharma, Chitosan a drug carrier for the 21st century S.T.P, Pharm. Sci. 10 (2000) 5–22. [16] V. Dodane, D. Vilivalam, Pharmaceutical applications of chitosan, Pharm. Sci. Technol. Today 16 (1998) 246–253. [17] P. Calvo, J.L. Vila-Jato, M.J. Alonso, Effect of lysozyme on the stability of polyester nanocapsules and nanoparticles: stabilization approaches, Biomaterials 18 (1997) 1305–1310. [18] F.-L. Mi, H.-W. Sung, S.-S. Shyu, Synthesis and characterization of biodegradable TPP/genipin co-crosslinked chitosan gel beads, Polymer 44 (2003) 6521–6530. [19] K.W. Leong, H.Q. Mao, V.L. Truong-Le, DNA-polycation nanospheres as non-viral gene delivery vehicles, J. Control. Release 53 (1998) 183–193. [20] J.A. Ko, H.J. Park, S.J. Hwang, Preparation and characterization of chitosan microparticles intended for controlled drug delivery, J. Pharm. 249 (2002) 165–174. [21] X.Z. Shu, K.J. Zhu, Novel PH–sensitive citrate cross-linked chitosan film for controlled drug release, Int. J. Pharm. 212 (2001) 19–28. [22] K. Makino, H. Ohshima, T. Kondo, Transfer of protons from bulk solution to the surface of poly(l-lactide) microcapsules, J. Microencapsul. 3 (1986) 195–202. [23] Y. Xu, Y. Du, Effect of molecular structure of chitosan on protein delivery properties of chitosan nanoparticles, Int. J. Pharm. 250 (2003) 215–226.