SEARCH

SEARCH BY CITATION

Keywords:

  • chitosan;
  • electrostatic interaction;
  • modified lecithin;
  • nanoparticles;
  • self-assembly

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Conclusions
  7. Acknowledgments
  8. References

ABSTRACT:  The formation of biocompatible nanoparticles via the self-assembly of chitosan (CHI) and modified lecithin (ML) was studied. Stable nanoparticles in the size range of 123 to 350 nm were formed at over a wide molar mixing ratios of CHI/ML solutions (amino group/phosphate group) (NH3+/PO3) and total polyelectrolyte (PE) concentrations (0.1 to 1 wt%) except at intermediate molar ratios when the surface charge was close to neutrality. Zeta-potentials of the nanoparticles were found to be independent of the total PE concentrations. Nanoparticles exhibited excellent stability at over an extended pH (pHs 3 to 6) and ionic strength range (≤ 500 mM NaCl concentration). The particle size and zeta-potential of the nanoparticles increased with the molecular weight of CHI. Transmission electron microscopy suggested that nanoparticles were generally spherical in shape with CHI constituting the exterior of its surface at high molar mixing ratios. Dextran-fluorescein isothiocyanate, bovine serum albumin, and Coomassie brilliant blue as models of nonionic, positively and negatively charged compounds were encapsulated within the nanoparticles at between 8.7% and 62.7% efficiency. The ability of the nanoparticle suspensions to be converted to lyophilized powder or concentrated suspension was also demonstrated.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Conclusions
  7. Acknowledgments
  8. References

The utilization of nanoparticles as delivery system for bioactive food components has gained wide attention in recent years due to its ability to enhance bioavailability of active ingredients, revolutionize controlled release, confer protection to the bioactive compounds against environmental stress, and improve sensory aspects to name a few (Sanguansri and Augustin 2006). The distinct functional properties of the nanoparticles as excellent delivery vehicles may be related to its abilities in conferring high stability, high carrier capacity, feasibility in entrapping both hydrophilic and hydrophobic compounds, and its multiple routes of administration capability (Gelperina and others 2005). However, the vast majority of the present nanoparticle researches focus mainly in fabricating nanoparticle-based drug delivery systems. Furthermore, most of the processes used to produce these nanoparticles often involved the utilization of chemicals deemed to be nonbiocompatible and are therefore not suitable for applications in food that usually requires compounds to be generally recognized as safe (GRAS) (Chen and others 2006).

To overcome the problem associated with toxicity issues of using nonbiocompatible chemicals, complexation of polyelectrolytes (PEs) through the “bottom-up” approach is perhaps one of the simplest methods to create nanoparticles without the need to use any organic solvents or cross-linking agents. This method allows nanostructures to be generated from individual atoms or molecules that are capable of self-assembly (Moraru and others 2003; Chen and others 2006). The driving force for the formation of these self-organized nanoparticles usually involves electrostatic interactions between PEs carrying opposite charges or noncovalent intermolelcular forces such as hydrogen bonding, hydrophobic interactions, van der Waals forces, or dipole-charge transfer (Schatz and others 2004b; Graveland-Bikker and de Kruif 2006). Besides being an energy-saving process, this approach is also the more favorable method as compared to the conventional “top-down” approach as it only involves spontaneous formation of nanoparticles under mild conditions, thus ensuring that the bioactive compounds encapsulated within the nanoparticles are protected from inactivation due to physical or chemical stresses imposed on them during its fabrication (Ichikawa and others 2005).

Various biocompatible and degradable natural polymers can be employed in the formation of nanoparticles via this self-assembly method. In the present study, chitosan (CHI) and modified lecithin (ML) have been chosen as the polycationic and anionic compounds, respectively. CHI is a derivative of chitin, a biopolymer found in abundance in nature. CHI has attracted much attention due to its biocompatibility, biodegradability, nontoxic nature, low immunogenicity, antibacterial properties, mucoadhesivity, and its ability to act as an absorption enhancer (van der Lubben and others 2001; Vila and others 2002). With an estimated pKa value of 6.2 to 7, CHI is highly positively charged in acidic medium due to the protonation of its amino groups (NH3+). The positive NH3+ on the CHI polymer chain can bind to negatively charged surfaces via hydrogen or electrostatic bonding (Shahidi and Abuzaytoun 2005). This distinct property of CHI has been exploited for complexation with various polyanions to form nanoparticles such as β-lactoglobulin (Chen and Subirade 2005), carboxymethyl cellulose (Ichikawa and others 2005), poly(L-glutamic acid) (Dai and others 2007), lecithin (Sonvico and others 2006), dextran sulfate (Schatz and others 2004a, 2004b), and carboxymethyl konjac glucomannan (Du and others 2005).

Modified lecithin (ML) on the other hand is produced through the partial hydrolysis of soy lecithin by phospholipase A2. Phospholipase A2 acts on the sn-1–OH side chain of lecithin to produce lysophosphatidylcholine while keeping the other nonreacted phospholipids compounds intact, depending on the degree of hydrolysis (Sono 2005). We selected this biomaterial because it is easily soluble and dispersible in aqueous solution in contrast to the unmodified lecithin which is insoluble in aqueous solution, thus accelerates the formation of the desired nanoparticles. Moreover, lysophosphatidylcholine had been credited to markedly enhance the uptake of carotenoids solubilized in mixed micelles by Caco-2 human intestinal cells and in vivo studies of rats (Sugawara and others 2001; Lakshminarayana and others 2006).

Bearing in mind the potential benefits of utilizing these 2 materials, the objective of our study was to use the self-assembly approach to produce biocompatible nanoparticles suitable for applications in food and to characterize its properties as well as the encapsulation capabilities of some hydrophilic compounds within these nanoparticles.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Conclusions
  7. Acknowledgments
  8. References

Materials

Modified lecithin (ML) (SLP WhiteLyso®) (lot nr 18.03.22 and 17.08.24) produced through the enzymatic hydrolysis of soy lecithin by phospholipase A2 was kindly donated by Tsuji Oil Mills Co Ltd. (Matsuzaka, Japan). As stated by the supplier, ML consisted of lysophosphatidylcholine (18% to 30%), phosphatidylinositol (10% to 20%), phosphatidylcholine (2% to 8%), phosphatidylethanolamine (1% to 7%) and phosphatidic acid (0% to 5%). “Chitosan HD®” from Yaegaki Bio-industry Co Ltd. (Himeji, Japan) was used as low molecular weight (LMW) chitosan (CHI) (MW = 58000 Da; degree of deacetylation [DD]= 98%). “100D(VL)®” (MW = 120000 Da; DD = 95%) and “100D®” (MW = 200000 Da; DD = 95%) obtained from Dainichi Seika Color & Chemicals Mfg. Co. Ltd. (Tokyo, Japan) were used as medium molecular weight (MMW) and high molecular weight (HMW) CHI, respectively. Both ML and CHI were used as received. Bovine serum albumin (BSA) (MW = 68000 Da) and dextran-fluorescein isothiocyanate (DX-FITC) (MW = 4000 Da) were obtained from Sigma Aldrich (St. Louis, Mo., U.S.A.). Coomassie brilliant blue R250 (CBB) (MW = 825 Da) was purchased from Fluka A.G. (Buchs, Switzerland). All the other reagents were of analytical grade and were purchased from Wako Pure Chemical Industries Ltd. (Osaka, Japan). Milli-Q water was used for the preparation of all solutions.

Determination of phospholipids content in ML

The phospholipids content in ML was determined quantitatively according to the method of Rouser and others (1970). Briefly, ML powder sample was weighed and transferred into a clean glass tube. Subsequently, 0.65 mL of perchloric acid was added into the tube. The tube was placed in a heating block (Dry-Block Multi Stirrer Series MU 925S-11, Orix, Scinics, Japan) set at 180 °C for about 30 min until the disappearance of the yellow color. After cooling, 3.3 mL of Milli-Q water, 0.5 mL of 2.5% ammonium molybdate solution followed by 10% ascorbic acid solution were added into the tube with agitation by vortex after each addition of the solution. The tube was placed in a boiling water bath for 5 min and then was let to cool before measurements. Potassium dihydrogen phosphate (KH2PO4) was used as the standard. Absorbance of the cool sample and standard were read at 800 nm using a UV/VIS spectrophotometer (V-570, Jasco, Tokyo, Japan). The amount of phospholipids was calculated directly on a molar basis from the amount of phosphorus (P) and on a weight basis after multiplying the amount of P by 25.

Preparation of nanoparticles

One wt% of CHI solution (LMW CHI was used throughout the study unless otherwise stated) and 1 wt% ML solution were prepared by dissolving CHI and ML powder separately in 50 mM acetate buffer solution containing 0.02 wt% sodium azide (as microbial agent) overnight. Both the solutions were adjusted to pH 3 with HCl and filtered with 0.45 μm cellulose acetate hydrophilic filter (Advantec, Japan) before used. The CHI solution was mixed to ML solution at varying volume ratios of both solutions (1: 1 to 1: 80 which corresponds to NH3+/PO3 molar mixing ratios of 4.9: 1 to 0.06: 1) to form total PE concentration of 1 wt% with the assumption that all the amino and phosphate groups were ionized at pH 3. The mixed solutions were then mixed vigorously for 30 s and subsequent left overnight at room temperature prior to analysis. The same procedure was repeated with total PE concentrations of 0.5 wt% and 0.1 wt%. For all the subsequent experiments, the preparation of nanoparticle suspension was fixed at NH3+/PO3 molar mixing ratio of 4.9: 1, total PE concentration of 0.5 wt% and pH 3 unless stated otherwise. Stability of the nanoparticle suspensions at various pH (pHs 3 to 6) and ionic strength (0 to 1000 mM NaCl) was also investigated by adjusting the suspensions with either NaOH or NaCl solutions. The influence of the MW of CHI on the properties of nanoparticles (total PE concentration of 0.1 wt%) formed was also investigated using CHI with different MW (that is, LMW, MMW, and HMW).

Particle size analysis and zeta-potential measurements

Dynamic light scattering (DLS) and electrophoretic mobility measurements for determining nanoparticles size and zeta-potential values, respectively, were conducted using Zetasizer Nano ZS (Malvern Instruments Ltd., Worcestershire, U.K.). The mean particle diameter, z-average, was calculated by cumulants analysis based on the intensity of light scattered owing to the Brownian motion of the particles. Electrophoretic mobility measurements were based on the laser doppler technique, using the monomodal mode and the Smoluchowski approximation to convert electrophoretic mobility to zeta-potential. All the DLS and electrophoretic mobility measurements were carried out at 25 °C. Nanoparticle suspensions were diluted 10-fold with buffer solution at the appropriate pH or ionic strength conditions prior to analysis to avoid multiple scattering effects.

Transmission electron microscopy (TEM) analysis

Transmission electron micrographs of the nanoparticles were obtained on a JEOL-1010 (JEOL, Tokyo, Japan) using the negative staining method. Briefly, 1 drop of nanoparticle suspension, which was diluted 100-fold with buffer solution, was placed onto a copper grid that had been pre-layered with polyvinyl formal film. Negative staining involved the exposure of sample on the grid to 1 drop of 2% uranyl acetate solution for about 30 s before washing off with water. Excess liquid was blotted from the grid by lightly touching the edge or corner of the filter paper to the edge of the grid. Samples were then air dried before examination on the TEM.

Encapsulation of chemical compounds

DX-FITC, BSA, and CBB were used as models of nonionic, positively charged, and negatively charged compounds. All of these compounds were dissolved in buffer solution at pH 3 to a concentration of 10 mg/mL. Nonionic or positively charged compound (40 μL) was first added to the ML solution (4 mL) followed by CHI solution (4 mL) and mixing after every addition. As for the encapsulation of negatively charged compound, the procedure was similar except for the reverse mixing order of ML and CHI. These nanoparticle suspensions were left overnight prior to centrifugation at 93000 ×g for 1 h at 20 °C to separate the encapsulated and unencapsulated compounds. The amount of free BSA and CBB in the supernatant was measured by UV/VIS spectrophotometer (V-570, Jasco, Tokyo, Japan) at an absorbance of 595 and 570 nm, respectively. Measurement of BSA was based on the Bradford protein microassay. The amount of free DX-FITC in the supernatant was determined using a spectrofluorometer (FP-6500, Jasco, Tokyo, Japan) at an excitation wavelength (λex) of 460 nm and emission wavelength (λem) of 520 nm. All of the reported values were the means of at least 2 measurements from at least 2 experiment replications. The encapsulation efficiency (EE) for all the compounds was calculated based on the following equation:

  • image

where A is the total amount of compound in the nanoparticle suspension and B is the amount of unencapsulated compound remaining in the supernatant.

Freeze-drying studies

Sucrose, dextran, or sorbitol as cryoprotectant agents were incorporated into the nanoparticle suspensions at concentrations ranging from 0% to 20% followed by freezing in liquid nitrogen. The samples were then freeze-dried at 1 Pa (FD-81, Eyela, Tokyo, Japan) for 24 h. The lyophilized samples were reconstituted in buffer solution to its initial suspension volume prior to particle size analysis.

Evaporation studies

Nanoparticle suspensions were concentrated to the desired PE concentration by evaporation using a rotary evaporator (NE-1101, Eyela, Tokyo, Japan) under reduced pressure (20 hPa at 45 °C). The pressure was reduced manually to avoid foaming. The resulting concentrated suspensions were diluted with buffer solution to its initial concentration to evaluate any changes to its mean particle size after evaporation.

Statistical analysis

All of the reported values were the means of at least 2 measurements from at least 2 experiment replications. The mean particle size and zeta-potential of the encapsulated nanoparticles were analyzed by one-way analysis of variance (ANOVA) using SAS (version 9.1.3; SAS Inst. Inc., Cary, N.C., U.S.A.). Significant differences (P < 0.05) between means were determined by the least significant difference (LSD) test.

Results and Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Conclusions
  7. Acknowledgments
  8. References

Formation of the CHI/ML nanoparticles occurred when the high density NH3+ groups inherent in CHI solution at pH 3 formed electrostatic linkage with the phosphate groups (PO3) of ML, mainly through an entropy driven process due to the liberation of low molecular counterions (Dautzenberg 2001), which can lead to either the formation of water-soluble complexes or precipitates. Based on the assumption that 1 mole of ML contains 1 mole of phosphate group (PO3), the phospholipids purity obtained with the ML sample was determined to be approximately 96%. The complexation process between both the PEs was spontaneous because as soon as the 2 solutions were mixed, an increase in the turbidity of the solution was observed immediately. This is an indication that the interaction between the 2 oppositely charged PEs was mainly governed by kinetics as a result of the vast difference in the pKa of both the PEs (ΔpKa approximately 4.7) (pKa of phosphate group is 1.5). Throughout this study, the pH was controlled at 3 so as to ensure the full protonation of CHI. At this pH, ML was also fully dissociated to form an anionic. By using PE solutions having the same pH, factors that might have contributed to the neutralization reaction could be eliminated. In the formation of PE complexation as in the case with our nanoparticles, a number of parameters have been identified as influencing the electrostatic interaction in the formation process such as the physicochemical properties of each PE (relative molar masses and charge density), PE concentrations and mixing ratios in the reaction medium as well as some external parameters such as pH, ionic strength, mixing order of reactants, and rate of mixing (Schmitt and others 1998; Schatz and others 2004b).

Influences of different molar mixing ratios and total PE concentrations

In this section, the influences of different molar mixing ratios of CHI and ML (NH3+/PO3) as well as the total PE concentrations on the formation and properties of nanoparticles are described. CHI and ML solutions were mixed on a volume ratio basis of 1: 80 to 1: 1, which corresponds to the NH3+/PO3 molar mixing ratios of 0.06 to 4.9 at total PE concentrations of 0.1, 0.5, and 1 wt%. Varying the molar mixing ratios as well as the total PE concentrations will likely have an influence on the mean particle size and surface charge of the formed particles.

Stable nanoparticles in the zeta-potential and size ranges of –29 to +59 mV and 123 to 350 nm, respectively, were formed at over a wide range of molar mixing ratios and total PE concentrations (Figure 1A and 1B). However, precipitation of the nanoparticle suspensions was observed to overlap at the intermediate molar mixing ratios of 0.24 to 0.54 for all the total PE concentrations studied. The fact that the magnitudes of the zeta-potential were close to neutrality at these molar ratios resulted in the nanoparticles being more prone to aggregation due to the low repulsive forces between the particles (Figure 1A). In addition, an increase in the PE concentration favored the formation of larger particles (Figure 1B and 1C). Nanoparticles formed at total PE concentration of 1 wt% with a size range of 323 to 350 nm were bigger than those formed at 0.5 wt% (166 to 263 nm) and 0.1 wt% (123 to 213 nm). As more molecules are involved at higher concentration, the frequencies of coalition are expected to increase and this could have contributed to the formation of larger and less stable particles. Also, at higher PE concentration, the range of stable nanoparticles that could be formed became more limited. At PE concentration of 1 wt%, precipitation of the suspensions was observed at below the molar ratios of 1. In contrast, precipitations occurred at a much narrower molar mixing ratio ranges of 0.08 to 0.6 when PE concentration was at 0.5 wt% and the range was further reduced at 0.1 wt% PE concentration (0.2 to 0.54 molar mixing ratios). Similar results had also been reported in a few studies using biopolymers and synthetic PEs (Dautzenberg 1997; Buchhammer and others 2000; Schatz and others 2004a, 2004b). Tsuchida and others (1972) and Kabanov and Zezin (1984) suggested the formation of water-soluble polyelectrolyte complexes between a high MW PE present in excess with a low MW PE provided that both the PEs bear weak ionic groups, in agreement with our observation.

image

Figure 1—. Influence of chitosan and modified lecithin molar mixing ratio (NH3+/PO3) and polyelectrolyte concentration on the (A) zeta-potential and (B) mean particle size of nanoparticles at pH 3. P is the molar mixing ratios at which precipitation occurred. (C) Influence of polyelectrolyte concentration on the particle size distribution of nanoparticles at chitosan and modified lecithin molar mixing ratio (NH3+/PO3) of 4.9 at pH 3.

Download figure to PowerPoint

Though the particle size varied at different PE concentrations, interestingly, surface charge of the nanoparticles appeared to be independent of the total PE concentrations as the zeta-potentials portrayed trends, which were similar regardless of the PE concentrations. This indicated that the same ratios of individual PE molecule units were involved in the complexation of the PEs at any given molar ratio. Generally, as demonstrated in Figure 1A and 1B, zeta-potentials and mean particle sizes could be divided into 3 distinct regions: (1) in the presence of an excess ML, particles were negatively charged and remained at a plateau of around –29 mV at molar ratios of 0.06 to 0.24; (2) as more CHI was added, a steep increase in the zeta-potential was observed, followed by charge neutralization and charge inversion on the surface of the nanoparticles from negative to positive at over a narrow intermediate molar ratios of 0.24 to 0.54. In this region, the magnitudes of the zeta-potential were close to zero indicating that the number of negatively charged groups of ML balances the number of positively charged groups of CHI, resulting in the precipitation of the nanoparticle suspension because the electrostatic repulsion between the particles is no longer sufficiently strong to overcome the attractive interactions (McClements 2005); (3) when CHI was present in excess at the molar ratios of around 0.54 to 4.9, a gradual increase in the zeta-potential values to approximately +60 mV was observed. Saturation of the excess ML and CHI could have contributed to the observed near static zeta-potential values at both low and high molar mixing ratios. At the zones where the net charges of the particles were high, particles were highly stable especially at low total PE concentrations. The magnitude of zeta-potential of particle is one of the factors determining the stability of emulsion and suspension. Suspensions with absolute zeta-potential value above 60 mV show excellent stability while those above 30 mV are moderately stable. Below 20 mV, suspensions are marginally stable with pronounced aggregation occurring at below 5 mV (Keck 2006).

In other words, formation of stable nanoparticles was more favorable at low PE concentrations and at high molar mixing ratios when CHI was present in excess. Aiming at obtaining nanoparticles in the stable domain, we focused on using the nonaggregating nanoparticles formed at 0.5 wt% PE concentration and molar mixing ratio of 4.9 as representative sample for the subsequent experiments (unless stated otherwise).

Influence of pH

The effect of pH on the mean particle size and zeta-potential of nanoparticles formed at 0.5 wt% total PE concentration and 4.9 mixing ratio was studied. As shown in Figure 2A, particle size remained unchanged from pHs 3 to 5 but increased appreciably at pH 6. Further increase of the pH to around 6.5 caused the particles to aggregate since this pH was in the reported range of the pKa of CHI. This also evidenced the fact that the surface properties of the nanoparticles were predominantly governed by CHI. Simultaneously, the zeta-potentials of the nanoparticles was found to decrease from about +60 mV at pH 3 to +25 mV at pH 6. This result was expected because as the pH increased, the degree of protonation of the amino groups in CHI decreased.

image

Figure 2—. Influence of pH on the (A) mean particle size and zeta-potential and (B) mean particle size and stability of nanoparticles prepared at chitosan and modified lecithin at molar mixing ratio (NH3+/PO3) of 4.9 and polyelectrolyte concentration of 0.5 wt% at pH 3.

Download figure to PowerPoint

On the other hand, nanoparticles were found to be stable for a period of at least 2 wk with no discernible changes to the mean particle size upon storage at pH 3 through 6 at room temperature (Figure 2B). However, based on the zeta-potential values at different pH, nanoparticles at pH 6 would be expected to be the least stable since the magnitude of its surface charge was the lowest (+25 mV) and vice versa for nanoparticles suspended at pH 3 (+60 mV).

Influence of ionic strength

In this section, the formation of nanoparticles at molar ratio of 4.9 between CHI and ML studied at different ionic strengths (0–1000 mM) by adding various concentrations of sodium chloride solution into the nanoparticle suspensions to form a final PE concentration of 0.4 wt% while maintaining the pH at 3 is described (Figure 3A). When NaCl was initially introduced into the nanoparticle suspension, shrinkage of the nanoparticles by about 15% and a decrease in the polydispersity index (data not shown) was observed. The initial decrease of the mean particle size of the nanoparticles in the presence of small amount of salt was probably due to the screening of the excess NH3+ (since CHI is present in excess at the investigated molar mixing ratio) by the counterion. This could have resulted in a reduced intraparticle repulsion within the nanoparticles, which led to the shrinkage of the nanoparticles. Up to 500 mM NaCl, particles having similar size were formed and their particle size distributions were monomodal. However, as more NaCl was added, the screening of the electrical double layer of the nanoparticles progressively reduced the electrostatic repulsion between the nanoparticles, until it reached a critical salt level whereby it was no longer capable of preventing flocculation. This eventually led to a macroscopic phase separation characterized by a clear supernatant and a bulky precipitate when NaCl concentration was at 1000 mM. The particle size distributions of the nanoparticles at this stage became multimodal. Besides attributing the aggregation to electrostatic screening effect, ion binding could also account for the aggregation behavior observed in the particle size measurements. Since it has been known that the presence of free mobile ions can have an effect on electrostatic stabilization, this result again proved that the driving force between ML and CHI was electrostatic in nature.

image

Figure 3—. Influence of NaCl concentrations on the (A) mean particle size and (B) mean particle size and stability of nanoparticles prepared at chitosan and modified lecithin molar mixing ratio (NH3+/PO3) of 4.9 and polyelectrolyte concentration of 0.4 wt% at pH 3.

Download figure to PowerPoint

As demonstrated in Figure 3B, mean particle size of the nanoparticles remained constant for over 2 wk at all NaCl concentrations investigated except for nanoparticles suspended in 500 mM NaCl which showed a gradual increase in the particle size throughout the study period. Nanoparticles at this NaCl concentration were more susceptible to aggregation due to its much lower zeta-potential magnitude (data not shown).

Influence of molecular weight of CHI

In an attempt to study the effect of different MW of CHI on the formation of nanoparticles, CHI having MW of 58, 125, and 250 kDa with almost similar degree of deacetylation were chosen. In our preliminary study, total PE concentration of 0.5 wt% was used but it was found to be unsuitable for this purpose because a viscous gel was formed in the mixture containing CHI at 125 and 250 kDa instead of a stable nanoparticle suspension. Therefore, it was decided that total PE concentration of 0.1 wt% was more appropriate for this part of study.

As evident in Figure 4, the formation of larger structures occurred with increasing MW of CHI. This result was in correlation with previous studies (Schatz and others 2004a; Ichikawa and others 2005; Sonvico and others 2006). It has been suggested that the increase in the hydrodynamic size of the particles was attributed to the ability of the longer hydrophilic chain of HMW CHI to form thicker outer electrosteric/electrostatic stabilizing shell or may be due to the projection of the hydrophilic chain from the surface of the particles (Schatz and others 2004b; Sonvico and others 2006) while the smaller dimension particles were believed to be more compact and uniform (Schatz and others 2004b). On the other hand, the increase in the MW of CHI was also parallel to an increase in the zeta-potentials but our results contravened with the observation in an earlier study, which reported a constant zeta-potential value even as the MW of CHI was increased (Sonvico and others 2006) since the electrostatic interaction between the anionic and cationic PEs could be assumed to be fairly similar. We postulated that the nanoparticles formed from the HMW CHI were more loosely complexed in contrast to those formed from LMW CHI, hence allowed more CHI to be exposed. Another possible explanation was that the entropy of mixing that opposes adsorption could have been much greater for the LMW CHI than that at HMW CHI and therefore accounted for the higher zeta-potentials as the MW increased.

image

Figure 4—. Influence of chitosan molecular weight on the mean particle size and zeta-potential of nanoparticles prepared at chitosan and modified lecithin molar mixing ratio (NH3+/PO3) of 4.9 and polyelectrolyte concentration of 0.1 wt% at pH 3.

Download figure to PowerPoint

Transmission electron microscopy (TEM) observations

TEM images of CHI-ML nanoparticles prepared at total PE concentration of 0.5 wt% and molar ratio of 4.9 at pH 3 using the negative staining method are shown in Figure 5. The hydrophilic portions of the particle are stained black while the hydrophobic components are unstained. The clearly visible color contrast of the interior of the nanoparticles showed the distribution of the nanoparticle's hydrophilic and hydrophobic components, which was a reflection of its internal structure. From the TEM images, nanoparticles appeared to be basically spherical in shape with a rather compact morphology. However, the particle size observed with TEM appeared to be smaller as compared to the result determined by light scattering. This discrepancy could be most likely explained to be the shrinking of the nanoparticles during the air-drying process prior to the TEM observation. Moreover, the projection of the hydrophilic chain of CHI from the surface of the nanoparticle toward the exterior as revealed in the TEM image could have accounted for the bigger particle size as measured by light scattering.

image

Figure 5—. Transmission electron micrograph (TEM) image of chitosan-modified lecithin nanoparticle complexes prepared at chitosan and modified lecithin molar mixing ratio (NH3+/PO3) of 4.9 and total polyelectrolyte concentration of 0.5 wt% at pH 3.

Download figure to PowerPoint

Possible mechanism of nanoparticle formation

Based on the TEM observation and quantitative results, we proposed the mechanism of nanoparticle formation as illustrated in Figure 6. To explain our point, we made the assumption based on the formation of positively charged nanoparticles at high molar mixing ratios and negatively charged nanoparticles at low molar mixing ratios. Positively charged nanoparticles were obtained by the addition of an excess amount of CHI to ML solution. When CHI first comes into contact with ML, complexation between the 2 oppositely charged PEs started to take place immediately by initially forming a randomly loose coil with the hydrophilic chain of the CHI extending to the exterior. As more CHI is available for interaction with ML, both the PEs rearrange into a more ordered configuration and tightly packed structure, as driven by the overall entropy gain as a result of counterions and water molecules releases. The uncomplexed segment of the positively charged CHI present is expected to constitute the exterior of the nanoparticles, hence giving the nanoparticles a highly positively charged property. Therefore, the higher the MW of CHI used, the longer the hydrophilic chain of CHI and in turn attribute to a larger hydrodynamic size of the nanoparticles. The presence of an excess amount of CHI also provides stability to the nanoparticles against flocculation due to its electrostatic repulsive force.

image

Figure 6—. Schematic representation of possible mechanism of nanoparticles formation based on the polyelectrolyte complexation of chitosan and modified lecithin (not drawn to scale).

Download figure to PowerPoint

In the case of negatively charged nanoparticles, CHI solution was added to an excess amount of ML solution. ML being more flexible than CHI would favor a complete ion-pairing with CHI leading closely to charge neutralization. In addition, the charges of the unpaired phosphate groups are not sufficient to ensure stabilization of the nanoparticle suspension, hence explain the susceptibility of particles to aggregation at low molar mixing ratios.

Encapsulation efficiency

The encapsulation efficiencies of various charged compounds into the nanoparticles were also investigated (Table 1). Nanoparticles were prepared at pH 3 at molar mixing ratio of 4.9, total PE concentration of 0.5 wt% with the hydrophilic compounds encapsulated to a final concentration of 50 μg/mL in the nanoparticle suspension. BSA and CBB were chosen as models of positively and negatively charged compounds, respectively. BSA is positively charged at pH below its pI value of 5.1 and vice versa when it is above the pI value. Results obtained revealed that CBB was entrapped in the nanoparticles at 62.7%, surpassing the 32.7% encapsulation efficiency of BSA at pH 3. Since CBB is a small molecule with high anionic densities, it is expected to bind strongly to the amino group of CHI during the electrostatic interaction between the 2 functional groups, hence explained the high encapsulation efficiency as well as the significantly (P < 0.05) smaller particles formed. On the other hand, entrapment of DX-FITC, which is a nonionic compound into the nanoparticles yielded only 8.7%. Moreover, no significant (P > 0.05) difference was observed in its mean particle size after encapsulation. These results could be because the electrostatic interaction that occurred between DX-FITC and CHI or ML might not be as strong as those involving ionic compounds having a much higher charge densities (that is, BSA and CBB), thus explained the lower encapsulation efficiency and the insignificant change in its mean particle size.

Table 1—.  Properties and encapsulation efficiency of model hydrophilic compounds into nanoparticles.A
Entrapped compoundsMw (Da)Mean particleBsize (nm)ζ-potentialB(mV)Encapsulation efficiency (%)
  1. APrepared at chitosan and modified lecithin molar mixing ratio (NH3+/PO3) of 4.9 and polyelectrolyte concentration of 0.5 wt% at pH 3.

  2. BData are expressed as means ± standard deviation of at least 2 measurements from at least duplicate experiments. Mean values with different letters within the same column are significantly different (P < 0.05).

Nanoparticles (without compound) 305.2 ± 3.5a63.0 ± 2.1a 
Bovine serum albumin68000283.5 ± 5.9b63.0 ± 1.8a32.7 ± 4.7
Dextran FITC4000299.2 ± 1.6a65.5 ± 1.7b 8.7 ± 1.3
Coomassie brilliant blue825268.3 ± 3.3c62.1 ± 2.1a62.7 ± 4.7

Freeze-drying of nanoparticles

To further improve the physical and chemical stability of the nanoparticles, as well as to facilitate convenience, an attempt to freeze-dry the nanoparticle suspension was carried out. Freeze-drying or lyophilization is a process to remove water from some materials by sublimation and desorption under vacuum while leaving the basic structure and composition of the material intact. From our preliminary experiment, though it was possible to freeze-dry the nanoparticle suspension, precipitation of the nanoparticles was observed when we reconstituted the freeze-dried powder in buffer solution. According to Abdelwahed and others (2006), cryoprotectant agents are usually added to protect the nanoparticles from stresses caused by freezing and desiccation during freeze-drying process. Figure 7 shows changes in the particle size of nanoparticles after freeze-drying with the addition of different quantity of cryoprotectant agents before and after redispersion in buffer solution. Sucrose showed the best performance among the 3 cryoprotectants investigated. As little as 5 wt% of sucrose was sufficient to control the nanoparticles back to its original size upon reconstitution in buffer solution. Even though the use of dextran as a cryoprotectant appeared to be feasible, the particle size of the nanoparticle was somehow still larger than its initial size (about 36% bigger). In fact, the nanoparticle suspension became increasingly viscous as more dextran was incorporated even after reconstituting the resulting freeze-dried powder in buffer solution. While it was possible for sorbitol to act as a cryoprotectant, a much higher amount (approximately 20 wt %) was needed to draw on the same effect as sucrose.

image

Figure 7—. Influence of freeze-drying on the mean particle size of nanoparticles prepared at chitosan and modified lecithin molar mixing ratio (NH3+/PO3) of 4.9, polyelectrolyte concentration of 0.5 wt% at pH 3 with the addition of different types of cyroprotectants.

Download figure to PowerPoint

Concentration of nanoparticle by evaporation

We also attempted to concentrate the nanoparticle suspensions via evaporation. Upon evaporation to the desired concentration, the concentrated liquid was diluted in buffer solution to avoid multiple scattering effects when measuring particle size. As the nanoparticle suspension was evaporated, the suspension became increasingly concentrated and the nanoparticles gradually increased in size (Figure 8). Nevertheless, a marked increase in the particle size was only observed when the suspension was concentrated 5-folds. When the suspension becomes more concentrated, the number of particles present per unit volume is expected to increase and this will likely increase the frequency of collision between the particles that could have led to the enlargement of the nanoparticles.

image

Figure 8—. Influence of evaporation on the mean particle size of nanoparticles prepared at chitosan and modified lecithin molar mixing ratio (NH3+/PO3) of 4.9, polyelectrolyte concentration of 0.5 wt% at pH 3.

Download figure to PowerPoint

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Conclusions
  7. Acknowledgments
  8. References

Nanoparticles could be formed spontaneously via the self-assembly of CHI and ML as driven by the electrostatic interactions between the 2 oppositely charged PEs. The dimensions and surface charges of the CHI/ML nanoparticles were dependent on the NH3+/PO3 molar mixing ratios, total PE concentrations, pH, ionic strength conditions, as well as the MW of CHI. Formation of nanoparticles was more favorable at high molar mixing ratios or when the total PEs were present at low concentrations. However, precipitation of the nanoparticle suspensions occurred at intermediate mixing ratios regardless of the PE concentrations when the surface charge of the particles was close to zero. TEM observation of the nanoparticles formed at high mixing ratios revealed a morphology, which is generally spherical in shape with CHI constituting the exterior of the nanoparticle surface. Nanoparticles were stably suspended at over a wide pH range (pHs 3 to 6) and NaCl concentrations (≤ 500 mM) for at least 2 wk. In addition, model water-soluble compounds could be entrapped within the nanoparticles at fairly good efficiency. The possibilities to lyophilize and concentrate the nanoparticle suspensions offer yet another attractive alternative for its utilization in food and pharmaceuticals besides its highly desirable biocompatible and nontoxic nature.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Conclusions
  7. Acknowledgments
  8. References

The authors would like to thank Mrs. Fumiko Yukuhiro for her technical assistance with TEM analysis. This study was supported by Food Nanotechnology Project, MAFF, Japan.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Conclusions
  7. Acknowledgments
  8. References
  • Abdelwahed W, Degobert G, Stainmesse S, Fessi H. 2006. Freeze-drying of nanoparticles: formulation, process and storage considerations. Adv Drug Deliv Rev 58:1688713.
  • Buchhammer HM, Petzold G, Lunkwitz K. 2000. Nanoparticles based on polyelectrolyte complexes: effect of structure and net charge on the sorption capability for solved organic molecules. Colloid Polym Sci 278:8417.
  • Chen L, Subirade M. 2005. Chitosan/β-lactoglobulin core-shell nanoparticles as nutraceutical carriers. Biomaterials 26:604153.
  • Chen L, Remondetto GE, Subirade M. 2006. Food protein-based materials as nutraceutical delivery systems. Trends Food Sci Technol 17:27283.
  • Dai Z, Yin J, Yan S, Cao T, Ma J, Chen X. 2007. Polyelectrolyte complexes based on chitosan and poly(L-glutamic acid). Polym Int 56:11227.
  • Dautzenberg H. 1997. Polyelectrolyte complex formation in highly aggregating systems. 1. Effect of salt: polyelectrolyte complex formation in the presence of NaCl. Macromolecules 30:78105.
  • Dautzenberg H. 2001. Polyelectrolyte complex formation in highly aggregating systems: methodical aspects and general tendencies. In: RadevaT, editor. Physical chemistry of polyelectrolytes. New York : Marcel Dekker Inc. p 74392.
  • Du J, Sun R, Zhang S, Zhang LF, Xiong CD, Peng YX. 2005. Novel polyelectrolyte carboxymethyl konjac glucomannan-chitosan nanoparticles for drug delivery. I. Physicochemical characterization of the carboxymethyl konjac glucomannan-chitosan nanoparticles. Biopolymers 78:18.
  • Gelperina S, Kisich K, Iseman MD, Heifets L. 2005. The potential advantages of nanoparticle drug delivery systems in chemotherapy of tuberculosis. Am J Respir Crit Care Med 172:148790.
  • Graveland-Bikker JF, De Kruif CG. 2006. Unique milk protein based nanotubes: food and nanotechnology meet. Trends Food Sci Technol 7:196203.
  • Ichikawa S, Iwamoto S, Watanabe J. 2005. Formation of biocompatible nanoparticles by self-assembly of enzymatic hydrolysates of chitosan and carboxymethyl cellulose. Biosci Biotechnol Biochem 69:163742.
  • Kabanov VA, Zezin AB. 1984. A new class of complex water-soluble polyelectrolytes. Makromol Chem 6:25976.
  • Keck CM. 2006. Cyclosporine nanosuspensions: optimized size characterization and oral formulation. [PhD thesis]. Berlin , Germany : Freie Univ. Berlin. p 22132. Available from: FU Berlin Digitale Dissertation (http://www.diss.fu-berlin.de/2006/512/indexe.html). Accessed Jul 9, 2008.
  • Lakshminarayana R, Raju M, Krishnakantha TP, Baskaran V. 2006. Enhanced lutein bioavailability by lyso-phosphatidylcholine in rats. Mol Cell Biochem 281:10310.
  • McClements DJ. 2005. Food emulsions: principles, practice and techniques. 2nd ed. Boca Raton , Fla. : CRC Press.
  • Moraru C, Panchapakesan CP, Huang Q, Takhistov P, Liu S, Kokini JL. 2003. Nanotechnology: a new frontier in food science. Food Technol 57:1229.
  • Rouser G, Fleischer S, Yamamoto A. 1970. Two-dimensional thin layer chromatographic separation of polar lipids and determination of phospholipids by phosphorus analysis of spots. Lipids 5:4946.
  • Sanguansri P, Augustin MA. 2006. Nanoscale materials development—a food industry perspective. Trends Food Sci Technol 17:54756.
  • Schatz C, Domard A, Viton C, Pichot C, Delair T. 2004a. Versatile and efficient formation of colloids of biopolymer-based polyelectrolyte complexes. Biomacromolecules 5:188292.
  • Schatz C, Lucas J, Viton C, Domard A, Pichot C, Delair T. 2004b. Formation and properties of positively charged colloids based on polyelectrolyte complexes of biopolymers. Langmuir 20:776678.
  • Schmitt C, Sanchez C, Desobry-Banon S, Hardy J. 1998. Structure and technofunctional properties of protein-polysaccharide complexes: a review. Crit Rev Food Sci Nutr 38:689753.
  • Shahidi F, Abuzaytoun R. 2005. Chitin, chitosan, and co-products: chemistry, production, applications, and health effects. Adv Food Nutr Res 49:93135.
  • Sono R. 2005. Development and production of functional materials. Shokuhin Kougyo 48:19 (in Japanese).
  • Sonvico F, Cagnani A, Rossi A, Motta S, Di Bari MT, Cavatorta F, Alonso MJ, Deriu A, Colombo P. 2006. Formation of self-organized nanoparticles by lecithin/chitosan ionic interaction. Int J Pharm 324:6773.
  • Sugawara T, Kushiro M, Zhang H, Nara E, Ono H, Nagao A. 2001. Lysophosphatidylcholine enhances carotenoid uptake from mixed micelles by Caco-2 human intestinal cells. J Nutr 131:29217.
  • Tsuchida E, Osada Y, Sanada KJ. 1972. Interaction of poly(styrene sulfonate) with polycations carrying charges in the chain backbone. J Polym Sci 10:3397404.
  • Van Der Lubben IM, Verhoef JC, Borchard G, Junginger HE. 2001. Chitosan for mucosal vaccination. Adv Drug Deliv Rev 52:13944.
  • Vila A, Sánchez A, Tobío M, Calvo P, Alonso MJ. 2002. Design of biodegradable particles for protein delivery. J Control Release 78:1524.