Use of whey proteins for encapsulation and controlled delivery applications

Journal of Food Engineering 83 (2007) 31–40 www.elsevier.com/locate/jfoodeng

Use of whey proteins for encapsulation and controlled delivery applications Sundaram Gunasekaran *, Sanghoon Ko, Lan Xiao University of Wisconsin-Madison, Department of Biological Systems Engineering, 460 Henry Mall, Madison, WI 53706, USA Received 11 August 2006; received in revised form 1 November 2006; accepted 2 November 2006 Available online 14 December 2006

Abstract Whey proteins can be used as hydrogels and nanoparticle systems for encapsulation and controlled delivery of bioactive compounds. Whey protein concentrate (WPC) hydrogels exhibited pH-sensitive swelling behavior with minimum swelling ratio near the isoelectric point (pI) of whey proteins (5.1). The controlled drug release behavior of WPC hydrogels was studied using caffeine as a model drug. Consistent with the swelling behavior, the release of encapsulated model drug from the hydrogels was slower when the pH was below pI than it was at pH above pI. The swelling and release behavior of the WPC hydrogels can be changed easily with different layers of alginate coating. Nanoparticles of about 60 nm average particle size were prepared with beta-lactoglobulin (BLG) using a desolvation method by preheating the BLG solution to 60 °C. The stability of the particles was investigated by degradation experiments at neutral and acidic conditions with and without proteolytic enzyme. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Bioactive carrier; Beta-lactoglobulin; Controlled release; Encapsulation; Hydrogel; Nanoparticle; Swelling ratio; Whey protein

1. Introduction Whey proteins are valuable by-products from the cheese industry. They are used widely in a variety of foods primarily for their superior gelling and emulsification properties. b-Lactoglobulin (BLG) is the main whey protein component and its principal gelling agent. The physicochemical properties of the whey proteins suggest that they may be suitable for other novel food and nonfood applications. For example, whey protein gels may be used as pH-sensitive hydrogels for the controlled delivery of biologically-active substances (Gunasekaran, Xiao, & Ould Eleya, 2006). A hydrogel can be defined as a three-dimensional network that exhibits the ability to swell in water and retains a significant fraction of water within its structure. There is a wide variety of hydrogels

*

Corresponding author. Tel.: +1 608 262 1019; fax: +1 608 262 1228. E-mail address: [email protected] (S. Gunasekaran).

0260-8774/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.jfoodeng.2006.11.001

made from natural and synthetic polymers. Their ability to absorb water is due to the presence of hydrophilic groups such as –OH, –CONH–, –CONH2, –COOH, and –SO3H (Flory, 1953). Hydrogels can be neutral or ionic in nature. Because of their potential applications in controlled and site-specific drug release, the swelling behavior of hydrogels has been extensively studied. The driving force for swelling arises from the water–polymer thermodynamic mixing energy contribution to the overall free energy, which is coupled with an elastic polymer contribution (Kudela, 1985). For ionic hydrogels, the ionic interaction between charged polymer and free ions also contributes to swelling (Katchalsky, Lifson, & Eisenberg, 1951). Various hydrogels have been developed as controlled drug release carriers using water-soluble, biodegradable polymeric materials (Gudeman & Peppas, 1995; Park, Choi, Kim, & Kim, 1998; Park, Song, Kim, & Kim, 1998; Wang, Turhan, & Gunasekaran, 2004; Wen & Stevenson, 1993) including synthetic or natural polymers.

32

S. Gunasekaran et al. / Journal of Food Engineering 83 (2007) 31–40

Among the natural polymers used to develop pH-sensitive hydrogels are alginates (Park, Choi, et al., 1998; Park, Song, et al., 1998) and chitosan (Deyao, Tao, Goosen, Min, & He, 1993; Wang et al., 2004). The latter is usually cross-linked with other polymers such as poly(vinyl alcohol) (Wang et al., 2004) or poly(ether) (Deyao et al., 1993) using glutaraldehyde to produce semi-interpenetrating networks. Park, Choi, et al. (1998) reported that pH-sensitive hydrogels can be prepared from egg albumin simply through heat-induced gelation. They investigated the effect of gel preparation conditions, particularly the initial pH of the protein solution, on the swelling of dried albumin gel in phosphate buffer solutions. The albumin hydrogels exhibited pH-sensitive swelling behavior; the degree of swelling was low around the protein isoelectric point (pI) (pH 4) and increased with pH. Strong or weak heat-induced gels, with high or low water-holding capacity may be prepared from whey protein solutions simply by adjusting several of the gelation variables: concentration, pH, ionic strength, etc. Thus, it is possible to design heat-induced whey protein gels with good pH-sensitivity, tailored permeability, and mechanical properties that can be used as bioactive carriers. The advantages of using whey protein-based gels as potential devices for controlled release of bioactives is that they are entirely biodegradable and there is no need for any chemical crosslinking agents in their preparation. These are two of the major requirements for wide use of hydrogels not only in the pharmaceutical area but also in many food and bioprocessing applications. Furthermore, whey proteins may also be formed into nanoparticles. Nanoparticles are matrix systems of a dense polymeric network in which an active molecule may be dispersed throughout the matrix (Nakache, Poulain, Candau, Orecchioni, & Irache, 2000). Since nanoparticles are submicron and sub-cellular in size, they have versatile advantages for targeted, site-specific delivery purposes (Vinagradov, Bronich, & Kabanov, 2002) as they can penetrate circulating systems and target sites. The nanoparticles offer the feasibility to entrap drugs or bioactive compounds within but not chemically bound to them. Various biocompatible and biodegradable biopolymers have been used in the formation of nanoparticles to maximize delivery efficiency and increase the desirable benefits (Coester, Kreuter, Briesen, & Langer, 2000; Kreuter, 1994; Rhaese, Briesen, Rubsamen-Waigmann, Kreuter, & Langer, 2003). Albumin nanoparticles have been extensively investigated with respect to their preparation methods and release properties (Langer et al., 2003; Loo et al., 2004; Vural, Kas, Hincal, & Cave, 1990). Human serum albumin (HSA) and bovine serum albumin (BSA) have been used as natural matrix materials for delivery devices (Brannon-Peppas, 1995). The objectives of this research were to investigate the use of whey proteins as (1) pH-sensitive hydrogels and (2) nanoparticle systems.

2. Materials and methods 2.1. A whey protein hydrogels 2.1.1. Gel preparation Whey protein concentrate (WPC) powder obtained from a commercial source (Davisco Foods International Inc., Eden Prairie, MN, USA) was used. This powder contained 82.5% protein, 6.8% fat, 3.4% ash, and 6.5% lactose (data supplied by the manufacturer). Two sets of WPC gels were prepared: Set 1: WPC concentration = 15% (w/v) and pH = 5.1, 5.7, 6.2, 6.8, 7.2, and 10.0; Set 2: pH = 10.0 and WPC concentration = 12%, 15%, and 18%. Gels were prepared in cylindrical (10-mm inner diameter, 40-mm long) stainless steel tubes by filling them with protein solutions. The tubes were closed with rubber stoppers, sealed with vinyl electrical tape, and placed vertically in a water bath and heated isothermally at 80 ± 0.5 °C for 1 h. The gels were cooled to room temperature and left at 4 °C overnight before being removed from the tubes and cut into 10-mm diameter, 2-mm thick tablets and dried in an enclosed desiccator until the tablets reached a constant mass (i.e., within ±0.001 g). The WPI gels were also prepared with an encapsulated model drug, caffeine (Aldrich Chemical Company, Inc., F.W. 194.19) which was chosen because several desirable properties such as ease of detection; thermal stability at 80 °C, the gelling temperature of WPC; water solubility; and it does not interact with WPC. Caffeine (0.15 g) was dissolved in 15% WPC solution (3.0 g WPC in 17.0 g pH 10.0 buffer solution) to get 1:20 drug/WPC mass ratio. The drug-loaded WPC gel tablets (10-mm diameter, 2mm thickness) were prepared and dried as described above. A portion of the WPC gels (prepared at 15% concentration and pH 10) and caffeine-encapsulated WPC gels were alginate coated by placing the tablets in 1% sodium alginate solution for 2 min (Kikuchi et al., 1999). The gels were then cured in a 0.1 M CaCl2 solution for 15 min and for 30 min to gel alginate on the surface. Additional layers of alginate coating were applied by repeating the procedure twice (for two alginate layers) or four times (for four alginate layers), as needed. The thickness of each alginate layer was measured using a micrometer to be about 37.5 ± 1.0 lm. The alginate-coated gels were washed twice using deionized water and dried in a desiccator. The alginate coating is desirable in case of protein gels to prevent gel hydrolysis by proteolytic enzymes in the stomach (e.g., pepsin). 2.1.2. Swelling experiment Triplicates of dried gel tablets were weighed and placed in 20 mL of pH 1.8, 2.9, 5.8, 6.6, 7.6, 10.1, 11.4 phosphate buffer solutions (ionic strength = 0.2 M). Temperature was maintained at 37.5 ± 0.5 °C in an incubator (Model 2005, VWR scientific Inc. West Chester, PA). Gels were removed from the buffer solution periodically, blotted dry with tissue paper and weighed. Swelling ratio (SR) was calculated

S. Gunasekaran et al. / Journal of Food Engineering 83 (2007) 31–40

by measuring the mass of wet gel (mw) and dry gel (md) as follows: SR ¼

mw  md md

ð1Þ

2.1.3. In vitro drug release experiment The in vitro drug release tests were carried out using the USP 23n.1 dissolution test apparatus (Model 2000, Distek Inc., North Brunswick, NJ, USA) fixed with six rotating baskets. The dissolution medium was USP phosphate buffer, pH 7.5 (300 mL, 37.5 ± 0.5 °C), and the speed of rotation was 100 ± 1 rpm. Three dried tablets were placed into each of three baskets, and 3-mL samples were collected from the release medium at regular intervals. After each sample collection, the same amount of fresh release medium at the same temperature was added back. The amount of drug released was monitored by a UV–VIS spectrophotometer (DMS 300, Varian Inc., Palo Alto, CA) at 272 nm. The UV standard absorbance curve for caffeine was established. In the concentration range (2  105–6  105 M) investigated, the UV absorbance obeyed the Beer’s law (Xiao, 2003). 2.2. Whey protein nanoparticles 2.2.1. Preparation of BLG nanoparticles BLG nanoparticles were prepared by a desolvation method (Langer et al., 2003; Loo et al., 2004; Marty, Oppenheimer, & Speiser, 1978; Weber, Coester, Kreuter, & Langer, 2000). Two percent (w/v) solution of BLG in 10 mM NaCl at pH 9.0 was stirred on a 500 rpm magnetic stirrer at room temperature and acetone, a desolvating agent, was added at 1 mL/min rate until the solution became just turbid. The rate of acetone addition was controlled carefully since it also influences the resulting particle size (Langer et al., 2003). The amount of acetone addition for BLG formation was 22.5 mL. At the end of acetone addition the solution pH was 8.1. After the desolvation process, 0.01 mL of a 4% glutaraldehyde–ethanol solution was added to induce particle cross-linking and stirred continuously at room temperature for 3 h. The nanoparticles formed were purified by five cycles of centrifugation and dispersion. For each centrifugation step, BLG solution was centrifuged at 25,000g (Optima LE-80K, Beckman Coulter Inc., Fullerton, CA, USA) for 30 min. After centrifugation BLG pellets were redispersed to the original volume of acetone solution at pH 9.0 to prevent particle aggregation among the particles. Each redispersion step was performed in an ultrasonication bath (Branson, Danbury, CT, USA). The excess cross-linking agent were removed from the particles by purification steps. The resulting nanoparticles were stored in absolute ethanol at 4 °C. In order to decrease the size of the nanoparticles, the BLG solution prepared as above was heated in a water bath at 60 °C for 30 min before the desolvation process.

33

During acetone addition, solution pH lowered but was subsequently readjusted to 9.0. 2.2.2. Determination of particle size and zeta-potential Average size, distribution, and zeta-potential of BLG nanoparticles were determined by photon correlation spectroscopy (PCS) using a commercial particle size analyzer (Zetasizer 3000HSA, Malvern Instruments Ltd., Malvern, UK) which employs a 10-mW Helium–Neon laser operating at 633 nm. For size measurement, the nanoparticles prepared were dispersed in 10 mL of distilled water with 1:400 (w/v) ratio at pH 9.0. The pH of the sample was adjusted with HCl or NaOH. The dispersions were stirred continuously at room temperature for 30 min, and then measured at 25 °C with a scattering angle of 90°. Each sample was measured 10 times to obtain average particle size and its distribution using Zetasizer operation software (PCS: zeta mode v 1.51, Malvern Instruments Ltd., Malvern, UK). For zeta-potential measurement, 10 mL of nanoparticle dispersion with 1:400 (w/v) ratio was controlled at 10 predefined pH values between 2 and 11. Both measurements were repeated three times for all samples. 2.2.3. In vitro degradation of BLG nanoparticles To determine the degradation stability of BLG nanoparticles in vitro degradation was performed at 37 °C in pH 7.4 phosphate-buffered saline (PBS) according to published procedures (Gopferich, 1996) under acidic and neutral conditions. For the acidic condition, 30 mL of 0.1 M PBS solution adjusted to pH 2.0 was used with and without 0.6 mL of a 0.1% pepsin solution. For the neutral condition, 30 mL of PBS at pH 7.4 was used with or without 1000 BSEE unit/mL of a trypsin. Prior to mixing the particles, the solutions were pre-incubated at 37 °C for 1 h. After mixing, the nanoparticle suspensions were incubated at 37 °C in a horizontal shaker water bath with 10 shakings per minute. Aliquots of 1 mL of the solution were sampled at periodic intervals for 98 h. Each sample was centrifuged at 10,000 rpm for 5 min. The centrifugation precipitates regular particles but still disperses the degraded debris. The supernatant was removed and the pellet was redispersed by 1 mL of 20% trichloroacetic acid (TCA) solution. The redispersion step was performed in an ultrasonication bath (Branson, Danbury, CT, USA) to disperse the particles completely. The absorbance of the particles in TCA solution was measured at 800 nm and the absorbance was used to represent the uncorrupted BLG nanoparticles in acidic and neutral environments with and without a proteolytic enzyme. 3. Results and discussion 3.1. pH-sensitive whey protein hydrogel 3.1.1. Effect of swelling medium pH on the kinetic and equilibrium swelling The swelling kinetics of 15% WPC hydrogel denatured at pH 10.0 kept in different swelling media pHs are shown

S. Gunasekaran et al. / Journal of Food Engineering 83 (2007) 31–40 6.00

SR (g H2O / gdry gel)

5.00 4.00 3.00 2.00

pH1.8 pH7.6

1.00

pH10.0

0.00 0

50

100

150

200

250

Time (min)

1

SRt/SR∝

0.8

0.6 pH1.8 pH7.6

0.4

pH10.0

0.2 0

50

100 150 Time (min)

200

250

Fig. 1. Swelling kinetics of WPC hydrogels at different pHs. Swelling ratio (SR) vs. time (top). Normalized swelling ratio (SRt/SR1) vs. time (bottom).

in Fig. 1. The SR of WPC hydrogels is sensitive to the swelling medium pH; the higher the swelling medium pH, the faster the swelling. At pH 10.0, the gels reached the equilibrium SR value in about 50 min while at pH 1.8, it took almost twice as long. The kinetics of swelling may be understood by considering several simultaneous effects. The contours of time vs. penetrant uptake curve deviate more often from the classical Fickian model. In these cases, the sorption process is not a passive diffusion of the solvent molecules into the void spaces of the network but includes a concomitant relaxation of the network segments resulting from the advancing solvent front, which leads to plasticization of the material and a large increase in volume. The generalized semi-empirical equation used to describe the swelling kinetics is (Harogoppad & Aminabhavi, 1991; Rathna, Li, & Gunasekaran, 2004; Valencia & Pierola, 2002) SRt ¼ Ktn SR1

ð2Þ

where K is a characteristic constant of the system, which is a function of the geometry of the hydrogel tablet and the diffusion constant. Eq. (2) is valid when (SRt/SR1) < 0.6. Based on the value of the exponent n, this equation has been used to distinguish three types of sorption behavior:

case I, case II, and anomalous (Lucht & Peppas, 1987). Case I sorption is typified by n  0.5. It represents a perfect Fickian process, during which the rate of solvent penetration is slower, and hence being the rate-determining step, than the chain relaxation rate. For case II sorption n = 1.0, i.e., the mobility of the penetrant is substantially faster than the chain relaxation rate and the solvent uptake is directly proportional to time. Anomalous sorption occurs when 0.5 < n < 1.0. In this case, the rate of penetrant mobility and segmental relaxation are comparable. Thus, the relative importance of solvent diffusion and polymer matrix relaxation effects can be analyzed by examining the exponent n of the power law. As shown in Fig. 2, the equation fit well all cases and the model constants are listed in Table 1. At pH 10.0, n is 0.51 and the process may be considered diffusion-controlled case I sorption, while at pH 1.8 and pH 7.6, n is 0.56 and 0.65, respectively, and it may be the case of anomalous sorption. This kind of kinetic behavior can be understood considering the network structure of the WPC hydrogel. At pH 10, the polymer chain relaxation reduces greatly because of strong electrostatic repulsion among negative charges at the surface of the gel microstructure, so that water diffusion is faster than relaxation of polymer chain and swelling turns out to be diffusion-controlled. On the other hand, when the swelling medium pH is 7.6 most of the net negative charges were neutralized by the positive charges from the swelling medium, so smaller amount of net charges existed in the hydrogel. As a result, the electrostatic repulsion strongly

0.0 -0.1

Log (SRt/SR∝)

34

-0.2

y = 0.51x - 0.8 R2 = 0.9895

y = 0.65x - 1.05 R2 = 0.9916

-0.3

y = 0.56x - 1.12 R2 = 0.9997

-0.4 pH 1.8 pH 7.6 pH 10.0

-0.5 -0.6 0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

Log (time, min)

Fig. 2. Fitting of swelling kinetics of 15% whey protein prepared at pH 10.0 at different swelling media pHs by the power law, SR/SR1 = Ktn.

Table 1 Model (Eq. (2)) constants K and n for 15% WPC hydrogels prepared at pH 10 at different swelling medium pHs pH

K

n

R2

1.8 7.6 10.0

0.1585 0.0891 0.0759

0.56 0.65 0.56

0.9985 0.9916 0.9997

S. Gunasekaran et al. / Journal of Food Engineering 83 (2007) 31–40

decreases and the polymer chain relaxation increases to be comparable with water diffusion, resulting in an anomalous sorption mechanism. Hydrogel swelling is also governed by ionization of negatively-charged groups. When the swelling medium pH = 10, the number of negatively charged groups is the most, so the equilibrium SR (=5.5) is the highest because of the strong electrostatic repulsion. When the swelling medium pH = 7.6, the protons from the swelling medium neutralizes most of the negatively-charged groups, so the equilibrium SR (=3.8) is lower due to the reduced electrostatic repulsion. When the swelling medium pH = 1.8, all negatively-charged groups are neutralized; instead, there would be some positive amine groups. Because the amine groups in the hydrogel are fewer than the carboxyl groups, the net charges in this case are few, so that the equilibrium SR (=2.2) is very low. 3.1.2. Effect of concentration and pH on equilibrium swelling The equilibrium swelling ratios of hydrogels at different WPC concentrations (12%, 15%, and 18%) vs. pH values of swelling media are shown in Fig. 3A. At all swelling media pH values, 12% WPC hydrogel imbibed the most water while 18% WPC gels imbibed the least. This can be explained by the Flory’s swelling theory (Flory, 1953). At higher concentration the density of protein network is high and because of this, the equilibrium SR should decrease with increasing protein concentration. The equilibrium swelling ratios of 15% WPC hydrogels prepared at different pH values vs. pH of swelling medium are shown in Fig. 3B. At all swelling medium pHs, there is a general trend that the higher the gelation pH, the higher the equilibrium SR. Because the structure of thermally-denatured protein gels depends on pH of the protein solution. When pH < pI, the extent of increase in SR at acidic swelling medium was very small, because there are very few amine groups exist at protein chains so that the positive charges are very limited. SR reached the minimum when swelling medium pH was close to the pI of the whey protein (=5.4). This

A

12.0

is because the net charge of whey protein molecules is close to zero at pI, which means almost no electrostatic repulsion between chains in thermally denatured whey protein and minimum SR exists. On the other hand, when pH > pI, there are a lot of negatively-charged groups in the protein chains, so the gels would contain a lot of net charges when the swelling medium of high pH value is used, which results in increased equilibrium SR. The higher the pH, the more surface charges, the higher electrostatic repulsive force, and higher equilibrium SR. The linear regressions of equilibrium SR of WPC denatured at various pHs (5.1, 5.7, 6.2, 6.8, 7.2, and 10.0) are presented in Fig. 4A. The slope of these lines represents the pH-sensitivity which is plotted against pH of 15% WPC solution used for gel preparation also presented in Fig. 4B. The WPC hydrogels denatured at higher pH showed higher pH-sensitivity. As mentioned before, the gels denatured at higher pH value have higher surface net charges or negative charges, so that electrostatic repulsion between the charges led to the higher equilibrium SR and higher pH-sensitivity. 3.1.3. Effect of alginate coating on swelling Alginate can form very stable gel in the presence of Ca2+ and it is widely used for coating of polymer matrices used in controlled drug delivery system (Kikuchi et al., 1999). It is well known that alginate coating lowers the diffusion of solvent and encapsulated drug release. Fig. 5 shows such an effect of alginate coating on swelling of 15% WPC gel prepared at pH 10.0. The equilibrium SR and the rate of swelling decreased dramatically after coating whey protein gel with alginate compared with that of the gel without coating. It is well known that alginate gel formed through calcium ion bridges is very rigid and does not swell readily (Papageorgiou, Kasapis, & Gothard, 1994). Furthermore, n value determined (per Eq. (2)) for the alginate-coated gel is 0.44. Thus, we could say that alginate coating not only lowers the diffusion rate but also alters the swelling kinetics from anomalous sorption (n = 0.65 before alginate coating) to diffusion-controlled sorption.

B

8.0

18%

6.0 4.0 2.0 0.0 0.0

12.0 pH10.0 pH7.2 pH6.8 pH6.2 pH5.1 pH5.7

10.0

SR (g H2O / gdry gel)

SR (g

H2O

/g dry gel)

10.0 12% 15%

35

8.0 6.0 4.0 2.0

2.0

4.0

6.0

8.0

pH of Swelling Medium

10.0

12.0

0.0 1.0

3.0

5.0

7.0

9.0

11.0

13.0

pH of Swelling Medium

Fig. 3. Equilibrium swelling ratio (SR) of WPC gels (A): of different concentrations (12%, 15%, 18%) prepared at pH 10.0 and (B): of protein preparation pH (5.1, 5.7, 6.2, 6.8, 7.2, and 10.0) at 15% WPC concentration.

36

S. Gunasekaran et al. / Journal of Food Engineering 83 (2007) 31–40

B pH7.2 pH6.8 pH6.2

8.0

pH5.7 pH5.1

6.0

pH10.0

2

SR (g H O / gdry gel )

10.

4.0 2.0 0.0 5.0

6.0

1.4

pH-sensitivity ((g H2O /g dry gel)/pH)

A 12.

7.0

8.0

9.0

10.0

11.0

1.2 1.0 0.8 0.6 0.4 0.2 0.0 5.0

12.0

6.0

7.0

8.0

9.0

10.

pH of denaturation solution

pH of swelling medium

Fig. 4. Equilibrium swelling ratio (SR) of 15% WPC gel (A): prepared at pH 10.0 with different swelling medium pHs and (B): pH-sensitivity of 15% whey protein hydrogel vs. the preparation pH. The pH-sensitivity was defined as the slope of the first order regression (A).

SR(gH2O/gdry gel)

4

stantially faster than at pH 1.8. The slower release at pH 1.8 is due to fewer net charges and electrostatic repulsion. This is consistent with pH-sensitive swelling behavior. The n values determined (per Eq. (2)) were, 0.5 at pH 7.6 and 0.47 at pH 1.8. These n values suggest that the release at both pHs is diffusion-controlled.

un-coatedWPC

3

2

3.1.5. Effect of alginate coating on drug release The caffeine release profile from 15% WPC hydrogel prepared at pH 10.0 with alginate coating is shown in Fig. 6B. The release profile from whey protein gel without alginate coating is also shown for comparison. It is obvious that the caffeine release rate is reduced significantly by alginate coating, which is consistent with the results of the swelling study. The Ca2+ induced alginate gel is very strong, rigid and hard to swell, so the diffusion through this coating is the rate limiting step for swelling and drug release. The release was prolonged by additional alginate layers on the hydrogel surface. The release profile of the sample with four alginate layer coating is interesting in that not only the release rate was significantly lower, but also the release kinetics changed to zero-order. Similar results

alginate-coated WPC

1

0 0

150

300

450

600

750

900

Time (min) Fig. 5. Comparison of swelling ratio (SR) of WPC gel (15%, gel preparation pH 10.0, swelling medium pH 7.5) with and without alginate coating.

1.0

0.8

B 1.0

0.9

pH 7.5

0.7

Fraction released

A Fraction released

3.1.4. Effect of release medium pH value on drug release Caffeine release profiles from the 15% WPC hydrogel prepared at pH 10.0 pH in release media pH of 7.6 and 1.8 are shown in Fig. 6A. At pH 7.6 caffeine release is sub-

pH 1.8

0.6 0.5 0.4 0.3

0.8

0.6

0.4 0 layer 1 layer, 15 min 1 layer, 30 min 2 layers, 30 min 2 layers, 15 min 4 layers, 30 min

0.2

0.2 0.1 0.0

0 0

40

80

120

160 200 240 280

Time (min)

320 360

0

50

100

150

200

250

300

350

Time (min)

Fig. 6. Effect of release medium pH value (on left) and different layers of alginate coating at pH 7.5 (on right) on the in vitro release profile of caffeine from WPC gel tablet.

S. Gunasekaran et al. / Journal of Food Engineering 83 (2007) 31–40

37

30

25

% Nanoparticles

20

15

10

36 nm

5

0 0

50

100

150

200

250

300

Diameter (nm) Fig. 7. Particle size distribution of BLG (average size, 131 ± 8 nm; size at peak, 127 ± 4 nm; half-bandwidth of 80% distribution, 36 ± 10 nm) nanoparticles prepared without preheating and pH adjustment.

3.2. Whey protein nanoparticles 3.2.1. Particle size and zeta-potential Fig. 7 shows the distribution of BLG nanoparticles formed without preheating and pH adjustment. For the BLG nanoparticles, the peak of the distribution was at 127 ± 4 nm. The number average diameter of BLG was 131 ± 8 nm, respectively. From the particles size distribution data the size range around the peak which contains 80% of the particles was calculated. One half of this 80% particle bandwidth was used as a measure of particle size dispersion. The half width of 80% particle bandwidth for BLG nanoparticles was 36 ± 10 nm. Fig. 8 shows the zeta-potential of BLG nanoparticles decreased with increasing pH. The surface of BLG is charged positively at acidic condition and negatively at neutral and basic conditions with the transition occurs at its pI. The particles size of BLG can be explained by its surface charge and surface hydrophobicity. BLG is characterized by a high content of charged amino acids (Brown, 1975; Papiz et al., 1986). At basic pH, the size of the protein aggregates as well as the void spaces within a particle generally decreases (Schmidt, 1981). In addition, proteins are generally more unfolded at basic pH which exposes more reactive sites for cross-linking (Kinsella & Whitehead, 1989). The unfolding of the BLG molecule at basic pH increases thiol–disulfide interchange reaction, which may

30

Zeta-potential (mV)

have been reported by others (Giunchedi, Gavini, Moretti, & Pirisino, 2000; Ritger & Peppas, 1987). The curing time of alginate gel seems to have no significant effect on the drug release behavior indicating that alginate coating completely cured within 15 min.

20 10 0 -10 -20 -30

pI of BLG: ~ 5.1

-40 -50 3

4

5

6

7

8

9

10

11

pH Fig. 8. Zeta-potential of BLG nanoparticles.

enhance particle formation but inhibit the formation of large aggregates. Our BLG nanoparticles were manufactured at pH 9.0 whose molecules would have negative charge. This condition resulted in small BLG particles charged negatively on their surface since coacervate precipitation was suppressed at pH 9.0. Another critical factor, the surface hydrophobicity dictates the propensity to bind non-polar amino acid groups to a hydrophobic part of its surface. Hydrophobic interactions between hydrophobic regions of unfolded polypeptide chains lead to their aggregation resulting in size increment (Ismond, Murray, & Arntfield, 1988). At basic pH, protein unfolding results in the change of the protein secondary structure. As pH increases, b-sheet formation in a protein increases due to an increase in hydrogen bonding (Krimm & Bandekar, 1986). At basic pH, a thiol group or previously hidden hydrophobic groups in BLG becomes exposed and the thiol–disulfide interchange reaction is

38

S. Gunasekaran et al. / Journal of Food Engineering 83 (2007) 31–40

accelerated. The degree of unfolding is related to the amino acid composition (Birdi, 1976). The effective hydrophobicity of BLG is 12.2. Thus, small hydrophobic interactions of BLG suppressed the aggregation of the molecules and then resulted in smaller particles. 3.2.2. Effect of preheating and pH adjustment Fig. 9 shows the size distribution of BLG nanoparticles formed after preheating to 60 °C and the pH readjusted to 9.0. The average particle size was 59 ± 5 nm with majority of the particles of size 50 ± 4 nm. In addition, the halfbandwidth of 80% of particles was narrower (27 ± 5 nm) than what was obtained without preheating and pH readjustment. This is a substantial improvement in both lowering the average particle size and improving the particle size uniformity. Preheating makes protein molecules unfolded so that the hydrophobic interactions between them are suppressed, which reduces self-aggregation. Further, by maintaining pH 9.0 we have generated high repulsive forces between BLG molecules and increased their unfolding. Therefore, we think that optimizing preheating and pH adjustment may be the key in preparing uniform BLG nanoparticles of sub-100-nm size range. 3.2.3. Degradation of nanoparticles Degradation plots of the BLG nanoparticles at acidic and neutral conditions are shown in Fig. 10A and B, respectively. All the degradation curves exhibited a typical rapid initial decrease in absorbance followed by a fairly stable tail region. By linearizing the initial and final regions, the degradation time (Dt) was determined at the intersection of the two linear segments. The average Dt values are listed in Table 2. BLG particles were relatively stable in the acidic environment (pH 2.0) (Fig. 10A). At the acidic pH far from

pI BLG unfolds only partially (Kinsella & Whitehead, 1989). As degree of unfolding increases, it is easier for the degradative factors such as invasion of water and proteolytic enzymes to attack. The degradation rate increased substantially when pepsin was added. The Dt value of BLG nanoparticles was 7.3 h. At neutral pH of 7.4, BLG nanoparticles was highly stable (Fig. 10B). Only, less than 20% of initial amount was degraded over 4 d. Addition of trypsin accelerated the degradation. For BLG particles Dt was 15 h. At neutral pH, trypsin is expected to attack specific sites on the surface and in the interior of the protein particles. The number of susceptible peptide bonds is likely important in determining the rate and extent of degradation. For trypsin, it has been reported that the main site at which hydrolysis occurs is the carboxyl group of basic amino acids, such as lysine and arginine (Magee, Halbert, & Wilmott, 1995). BLG has 15 lysine and 3 arginine residues whereas BSA has 59 lysine and 23 residues (Brown, 1975; Carter & Ho, 1994; Papiz et al., 1986). The portion of lysine and arginine residues of BLG is 9.3% and 1.9%, respectively (Kinsella & Whitehead, 1989). At pH 7.4 PBS containing trypsin, the BLG particles were degraded slowly. The proteolytic enzyme activity on the surface of the BLG particles were less since the amount of basic amino acids was limited. Several factors such as particle preparation technique, degradation environments, enzyme activity, surface area, porosity, tortuosity, and size can affect the degradation on the matrix of protein nanoparticles. The development of a dense cross-linking matrix for nanoparticles offers resistance against the proteolytic degradation since it is difficult for the enzymes penetrate into the particles. For BLG nanoparticles manufactured under the similar processing condition, they showed better resistance against enzyme degradation under both neutral and acidic environments.

25

% Nanoparticles

20

15

10 27 nm 5

0 0

50

100

150

200

250

300

Diameter (nm) Fig. 9. Particle size distribution of BLG (average size, 59 ± 5 nm; size at peak, 50 ± 4 nm; half-bandwidth of 80% distribution, 27 ± 5 nm) nanoparticles prepared with preheating at 60 °C and pH adjustment at 9.0.

S. Gunasekaran et al. / Journal of Food Engineering 83 (2007) 31–40

0.4

0.4

Absorbance

B 0.5

Absorbance

A 0.5

0.3 0.2 0.1

39

0.3 0.2 0.1

0.0

0.0 0

20

40

60

80

100

Time (h)

0

20

40

60

80

100

Time (h)

Fig. 10. Degradation of BLG nanoparticles with (solid triangles) or without (open triangles) pepsin at pH 2.0 (A) and with or without trypsin in pH 7.4 PBS (B). Table 2 Degradation time (Dt, h) of BLG nanoparticles Acidic condition (pH 2.0)

Neutral condition (pH 7.4)

No enzyme

Pepsin

No enzyme

Trypsin

22.1 ± 12.4

7.3 ± 0.3

–a

15.4 ± 1.4

a

Could not be determined.

Only 11.2% basic amino acid residues (lysine + arginine) retarded the hydrolysis of the BLG nanoparticles. Thus, we could attribute the resistance of the BLG nanoparticles against the enzyme attack to its dense structure and small portion of basic amino acid composition. 4. Conclusions Whey proteins can be used as hydrogels and/or nanoparticles systems for controlled release of bioactive compounds. As hydrogels, they exhibit pH-sensitive swelling ability especially at pH above their isoelectric point. The release kinetic of the hydrogels parallel that of their swelling ability. The release properties can be conveniently altered by appropriately coating with sodium alginate. Nanoparticles of sub-100-nm size can be prepared from beta-lactoglobulin (BLG). The average particle size can be lowered by preheating the BLG solution to 60 °C. Preheating also improves particle size uniformity. The BLG nanoparticles are more stable at neutral conditions than at acidic conditions with and without proteolytic enzymes. Acknowledgement We thank Davisco Foods International Inc. for providing the whey proteins for our study. References Birdi, K. S. (1976). Interaction of insulin with lipid monolayers. Journal of Colloid and Interface Science, 57, 228–232. Brannon-Peppas, L. (1995). Recent advances on the use of biodegradable microparticles and nanoparticles in controlled drug delivery. International Journal of Pharmaceutics, 116, 1–9.

Brown, J. R. (1975). Structure of bovine serum albumin. Federation Proceedings, 34, 591. Carter, D. C., & Ho, J. X. (1994). Structure of serum albumin. Advances in Protein Chemistry, 45, 153–203. Coester, C., Kreuter, J., Briesen, H. v., & Langer, K. (2000). Preparation of avidin-labelled gelatin nanoparticles as carriers for biotinylated peptide nucleic acid (PNA). International Journal of Pharmaceutics, 196, 147–149. Deyao, K., Tao, P., Goosen, M. F. A., Min, J. M., & He, Y. Y. (1993). pH-sensitivity of hydrogels based on complex-forming chitosanpolyether interpenetrating polymer network. Journal of Applied Polymer Science, 48, 343–354. Flory, P. J. (1953). Principles of polymer chemistry. Ithaca, NY: Cornell University Press. Giunchedi, P., Gavini, E., Moretti, M. D. L., & Pirisino, G. (2000). Evaluation of alginate compressed matrices a prolonged drug delivery system. AAPS PharmSciTech, 1, 19. Gopferich, A. (1996). Mechanisms of polymer degradation and erosion. Biomaterials, 17, 103–114. Gudeman, L. F., & Peppas, N. A. (1995). Preparation and characterization of pH-sensitive, interpenetrating networks of poly(vinyl alcohol) and poly(acrylic acid). Journal of Applied Polymer Science, 55, 919–928. Gunasekaran, S., Xiao, L., & Ould Eleya, M. M. (2006). Whey protein concentrate hydrogels as bioactive carriers. Journal of Applied Polymer Science, 99, 2470–2476. Harogoppad, S. B., & Aminabhavi, T. M. (1991). Diffusion and sorption of organic liquids through polymer membranes. 5. Neoprene, styrene butadiene rubber, ethylene propylene diene terpolymer, and naturalrubber versus hydrocarbons (C8-C16). Macromolecules, 24, 2598–2605. Ismond, M. A. H., Murray, E. D., & Arntfield, S. D. (1988). The role of non-covalent forces in micelle formation by vicilin from Vicia Faba. The effect of urea, guanidine hydrochloride and sucrose on protein interactions. Food Chemistry, 29, 189–198. Katchalsky, A., Lifson, S., & Eisenberg, H. (1951). Equation of swelling for polyelectrolyte gels. Journal of Polymer Science, 7, 571–574. Kikuchi, A., Kawabuchi, M., Watanabe, A., Sugihara, M., Sakurai, Y., & Okano, T. (1999). Effect of Ca2+-alginate gel dissolution on release of dextran with different molecular weights. Journal of Controlled Release, 58, 21–28. Kinsella, J. E., & Whitehead, D. M. (1989). Protein in whey: chemical, physical, and functional properties. Advances in Food and Nutrition Research, 33, 343–391. Kreuter, J. (1994). Nanoparticles. In J. Kreuter (Ed.), Colloidal drug delivery systems (pp. 219–342). New York: Marcel Dekker. Krimm, S., & Bandekar, J. (1986). Vibrational spectroscopy and conformation of peptides, polypeptides, and proteins. Advances in Protein Chemistry, 38, 181–364.

40

S. Gunasekaran et al. / Journal of Food Engineering 83 (2007) 31–40

Kudela, V. (1985). Hydrogels. New York: John Wiley & Sons, 783–807 pp. Langer, K., Balthasar, S., Vogel, V., Dinauer, N., Briesen, H. v., & Schubert, D. (2003). Optimization of the preparation process for human serum albumin (HSA) nanoparticles. International Journal of Pharmaceutics, 257, 169–180. Loo, C., Lin, A., Hirsch, L., Lee, M. H., Barton, J., Halas, N., et al. (2004). Nanoshell-enabled photonics-based imaging and therapy of cancer. Technology in Cancer Research & Treatment, 3, 33–40. Lucht, L. M., & Peppas, N. A. (1987). Transport of penetrants in the macromolecular structure of coals. 5. Anomalous transport in pretreated coal particles. Journal of Applied Polymer Science, 33, 1557–1566. Magee, G. A., Halbert, G. W., & Wilmott, N. (1995). Effect of process variables on the in-vitro degradation of protein microspheres. Journal of Controlled Release, 37, 11–19. Marty, J. J., Oppenheimer, R. C., & Speiser, P. (1978). Nanoparticles—a new colloidal drug delivery system. Pharmaceutica Acta Helvetiae, 53, 17–23. Nakache, E., Poulain, N., Candau, F., Orecchioni, A. M., & Irache, J. M. (2000). Biopolymer and polymer nanoparticles and their biomedical applications. In H. S. Nalwa (Ed.), Handbook of nanostructured materials and nanotechnology: organics, polymers ,and biological materials (pp. 577–635). San Diego: Academic Press. Papageorgiou, H., Kasapis, S., & Gothard, M. G. (1994). Structural and textural properties of calcium-induced, hot-made alginate gels. Carbohydrate Polymers, 24, 199–207. Papiz, M. Z., Sawyer, L., Eliopoulos, E. E., North, A. C. T., Findlay, J. B. C., Sivaprasadarao, R., et al. (1986). The structure of beta-lactoglobulin and its similarity to plasma retinol-binding protein. Nature, 324, 383–385. Park, H. Y., Choi, C. R., Kim, J. H., & Kim, W. S. (1998). Effect of pH on drug release from polysaccharide tablets. Drug Delivery, 5, 13–18. Park, H. Y., Song, I. H., Kim, J. H., & Kim, W. S. (1998). Preparation of thermally denatured albumin gel and its pH-sensitive swelling. International Journal of Pharmaceutics, 175, 231–236.

Rathna, G. V. N., Li, J., & Gunasekaran, S. (2004). Functionally-modified egg white albumen hydrogels. Polymer International, 53, 1994–2000. Rhaese, S., Briesen, H. v., Rubsamen-Waigmann, H., Kreuter, J., & Langer, K. (2003). Human serum albumin-polyethylenimine nanoparticles for gene delivery. Journal of Controlled Release, 92, 199–208. Ritger, P. L., & Peppas, N. A. (1987). A simple equation for description of solute release. II. Fickian and anomalous release from swellable devices. Journal of Controlled Release, 5, 37–42. Schmidt, R. H. (1981). Gelation and coagulation. In J. P. Cherry (Ed.), Protein functionality in foods. Washington, DC: Am. Chem. Soc., p. 131. Valencia, J., & Pierola, I. F. (2002). Swelling kinetics of poly(Nvinylimidazole-co-sodium styrenesulfonate) hydrogels. Journal of Applied Polymer Science, 83, 191–200. Vinagradov, S. V., Bronich, T. K., & Kabanov, A. V. (2002). Nanosized cationic hydrogels for drug delivery: preparation, properties and interactions with cells. Advanced Drug Delivery Reviews, 54, 223–233. Vural, I., Kas, H. S., Hincal, A. A., & Cave, G. (1990). Cyclophosphamide loaded albumin microspheres: 2. Release characteristics. Journal of Microencapsulation, 7, 511–516. Wang, T., Turhan, M., & Gunasekaran, S. (2004). Selected properties of pH-sensitive, biodegradable chitosan-poly(vinyl alcohol) hydrogel. Polymer International, 53, 911–918. Weber, C., Coester, C., Kreuter, J., & Langer, K. (2000). Desolvation process and surface characteristics of protein nanoparticles. International Journal of Pharmaceutics, 194, 91–102. Wen, S., & Stevenson, W. T. K. (1993). Synthetic pH-sensitive polyampholite hydrogels: a preliminary study. Colloid and Polymer Science, 271, 38–49. Xiao, L., 2003. Whey protein-based hydrogels: gelation swelling and drug release biological systems engineering, University of Wisconsin-Madison, Madison.