fish oil
Content
JFS: Food Engineering and Physical Properties
Encapsulation of Fish Oil by an Enzymatic Gelation Process Using Transglutaminase Cross-linked Proteins
Y.-H. CHO, H.K. SHIM, AND J. PARK ABSTRACT: To improve the storage stability and achieve controlled release, fish oil containing docosahexaenoic acid was encapsulated using double emulsification and subsequent enzymatic gelation method, using microbial transglutaminase cross-linked proteins. Isolated soy protein was selected as a wall material because it showed better emulsion stability and higher reactivity with MTGase than other proteins. Microcapsules prepared by this method showed a high stability against oxygen and a low water solubility, which subsequently resulted in sustained release of fish oil. Results indicate that this microencapsulation process is suitable for preparing protein-based microcapsules containing sensitive ingredients for controlled release and stability improvement. Keywords: fish oil, microencapsulation, microbial transglutaminase, isolated soy protein, double emulsification
Introduction
here is increasing information about the nutritional and health benefits of long chain polyunsaturated fatty acids (LC-PUFA) for humans (Kitessa and others 2001). LC-PUFAs, such as arachidonic acid (AA) and docosahexaenoic acid (DHA), are structural components of cell membrane phospholipids and precursors of eicosanoids and play important roles in the development of the central nervous system, including the retina (Smit and others 2000). Fish oil is an important source of DHA and many studies have shown that fish-oil supplementation increases the DHA content of blood components (Markrides and others 1995). However, the application of fish oils in the food system has a limitation because oils are unstable against oxygen during storage. Also, the unpleasant taste and odor of fish oil can cause deterioration in the quality of food products. Research has been performed to overcome the limitations of fish oils. Ahn and others (1991) enhanced oxidative stability of EPA and DHA by adding lecithin. However, the sensory evaluation of fish oil remained a problem. Several methods have been studied for encapsulation of fish oil, and a variety of materials have been used as a wall material, including cellulose (Kantor and others 1990), gum Arabic and gelatin (Maruyama and Yamamoto 1988), agar and waxy corn starch (Chang and Ha 2000), and milk proteins (Keogh and others 2001). However, carbohydratebased microcapsules or protein-based microcapsules prepared by spray-drying are water soluble and thus not suitable for controlled-release applications. Protein films are, in general, excellent oxygen and aroma barriers and can be used in developing microcapsules for controlled core release in food applications (Lee and Rosenberg 2000a). Microcapsules using proteins as wall materials have been prepared by simple/complex coacervation (Arshady 1990) and techniques consistMS 20030315 Submitted 6/6/03, Revised 10/1/00, Accepted 10/11/00. Authors Cho, Shim, and Park are with Dept. of Biotechnology, Yonsei Univ., Seoul 120-749, South Korea. Direct inquiries to author Park (foodpro@ yonsei.ac.kr).
T
ing of double emulsification and subsequent cross-linking with glutaraldehyde or heat-induced gelation (Lee and Rosenberg 2000a). Both approaches provide a means for achieving high core retention and render the resulting microcapsule water insoluble (Lee and Rosenberg 2000b). However, the use of protein-based wall materials in the food industry for sensitive ingredients is limited because proteins are generally unstable with heating and damaged by organic solvents, and the cross-linking agent is usually harmful. Therefore, if a novel, mild cross-linking agent is developed, and proteins are converted into stable forms, the application of proteins would be greatly increased (Babiker 2000). Microbial transglutaminase (MTGase) is an enzyme excreted from Streptoverticillium mobaraense that catalyzes acyl-transfer reactions, resulting in formation of -( -glutaminyl) lysine intraand inter-molecular cross-links in proteins (Nielsen 1995). Several studies have been conducted on cross-linking various proteins using this enzyme to form biopolymer films (Nio and others 1986; Motoki and others 1987; Yildirim and Hettiarachchy 1998; Lim and others 1999; Liu and others 1999; Babiker 2000; Babin and Dickinson 2001). These attempts focused on the preparation of biodegradable films, while we have attempted to use MTGase crosslinked proteins as a wall material in the encapsulation process. We investigated a number of proteins for use as base materials for preparing biodegradable films using MTGase and developed a double emulsification and subsequent enzymatic gelation method using isolated soy protein as a wall material. Soy protein is an abundant byproduct of the soybean-oil industry. Soy proteins have important applications in many food products because of their high nutritional value and their contribution to food texture and emulsifying properties (Roesch and Corredig 2002). Microcapusules containing fish oil were prepared using oil-inwater-in-oil (O/W/O) double emulsification and 3 different gelation methods, including enzymatic gelation using MTGase cross-linked protein, heat gelation, and a combination of the enzyme and heat gelation methods. The physicochemical properties, microstructure, and controlled release of 3 types of fish-oil microcapsules were investigated and compared.
Vol. 68, Nr. 9, 2003—JOURNAL OF FOOD SCIENCE 2717
© 2003 Institute of Food Technologists
Further reproduction prohibited without permission
Food Engineering and Physical Properties
Encapsulation by TGase cross-linked protein . . .
Materials and Methods
Materials
Fish oils were supplied by Loders Croklaan (Marinol D-40, Wormerveer, The Netherlands) and stored at –18 °C. Soluble wheat protein (SWP; Sam-Mi Industrial Co., Ansan, South Korea), whey protein isolate (WPI; Sam-Mi Industrial Co.), Na-caseinate (Sam-Mi Industrial Co.), and isolated soy protein (ISP; Sam-Asia, Seoul, South Korea) were selected as wall materials. Proteins were crosslinked using MTGase (Ajinomoto, Tokyo, Japan) that is derived from S. mobaraense. Before use, MTGase powder (4 mg) was suspended in 0.5 M Tris-acetate buffer (1 mL, pH 6.0). Span 80 (Duksan Chemical Co., Seoul, South Korea) was selected as an emulsifier for enhancement of the emulsion stability. Corn oil was supplied by the CJ Co. (Seoul, South Korea). water, and then MTGase (0.025%) was added into protein wall solutions. A CIWE (O/W primary emulsion) was prepared by blending a core (fish oil)/wall (1:2) mixture at 9500 rpm for 10 min using Ultra-Turrax T25TM. Corn oil (400 g) containing 3% emulsifier (Span 80) was preheated at 50 °C in an incubator to dissolve emulsifiers. The CIWE (100 g) was slowly added to the corn oil under the above conditions and then incubated at 37 °C for 4 h. The protein matrices of the dispersed phase of the double emulsion were gelled by reaction with MTGase. Microcapsules were separated from the corn oil by filtration through Whatman nr 4 filter paper (Whatman Intl. Ltd., Maidstone, U.K.), washed with ethanol 3 times, and freeze dried at 10 °C for 24 h using a freeze dryer (FD-558, Heto, Llerod, Denmark). (B) Combination of enzymatic and heat gelation (MTGase/ heat-gelled capsule). The encapsulation process was the same as (A) except for the gelation method. After preparing an O/W/O double emulsion, the temperature of the emulsion was adjusted to 85 °C, and then gelation was performed for 4 h. (C) Double emulsification and heat gelation (heat-gelled capsule). The encapsulation process was the same as (A) except for preparation of the wall solution and the gelation method. A wall solution containing 10% ISP was prepared in distilled water without a reaction with MTGase. An O/W/O emulsion was prepared and heated at 85 °C, then gelation was performed for 40 min.
Assay of microbial transglutaminase activity
MTGase activity was measured using the colorimetric hydroxaminate assay method. This assay was performed in 0.5 mL of an enzyme-substrate solution containing 0.1 mL of MTGase solution, 0.1 mL of 0.5 M Tris-acetate buffer (pH 6.0), 0.2 mL of 0.3 M CBZ-Lglutaminylglycine, 0.05 mL of 0.02 M EDTA, and 0.05 mL of 4 M hydroxylamine. After 10 min of incubation at 37 °C, 0.5 mL of ammonium sulfate was added to stop the reaction. The resulting red color was measured at 525 nm. One unit of MTGase activity was defined as the amount of enzyme needed to produce 1 micromole of hydroxamic acid per minute.
Determination of emulsion stability
A sample of liquid emulsion was transferred to a 10-mL mass cylinder, which was then capped and stored for 24 h. The volume of oil separated from emulsion in the mass cylinder was read. Results are reported as an emulsion stability index (ESI) with a range of possible results from 0 to 1.
Food Engineering and Physical Properties
Preparation of O/W emulsions for selection of the wall material
Protein solutions of 5% and 10% were prepared with and without MTGase (0.025%). An O/W emulsion was prepared by homomixing fish oil in each protein solution with Ultra-Turrax T25TM homogenizer (tool S25N-25G, IKA Labortechnik, Staufen, Germany) at 9500 rpm for 10 min.
Rheological measurement
The viscoelastic properties of MTGase-treated protein solutions were measured using a 40-mm-dia rheometer (AR1000, TA Instrument, New Castle, Del., U.S.A.) with cone steel geometry. Protein solutions were carefully poured into the rheometer cup (40-mm dia) and covered with a thin layer of vacuum oil to prevent evaporation. The temperature was maintained at 37 °C for 60 min. The storage (G ) and loss (G ) moduli were recorded as a function of time at a frequency of 1 Hz and at a maximum strain of 0.05 in the linear region. Each sample was assayed in triplicate.
A value of 0 represents poor emulsion stability while a value of 1 represents high emulsion stability (Cho 2000). Each sample was assayed in triplicate.
Particle-size distribution
The microcapsules were dispersed in distilled water. The particle-size distribution, the mean particle size, and the specific surface area of microcapsules were determined using a particle-size analyzer (Model 1064, Cilas, Orleans, France).
Scanning electron microscopy (SEM) Electrophoretic analysis of polymerization
SDS-PAGE was performed according to the method of Laemmli (1970). Samples were run on 12.5% gels. Proteins were dissolved in a sample buffer in the presence of -mercaptoethanol, heated for 10 min at 100 °C, and loaded onto the gel at 20- L/well. Gels were stained with Coomassie brilliant blue R-250. The overall morphology and the outer topography of dry microcapsules were observed by scanning electron microscopy (JSM5410L, Jeol Co., Peabody, Mass., U.S.A.) at an accelerating voltage of 10 kV. Microcapsules were mounted on metal stubs using double-sided adhesive tape and coated with Pb-Pt in an ion sputter ( 1030, Hitachi, Tokyo, Japan) under an argon atmosphere.
Preparation of emulsion and microcapsules
(A) Double emulsification and enzymatic gelation (MTGasegelled capsule). Microcapsules were prepared using a process consisting of double emulsification and subsequent enzymatic gelation. In general, a core-in-wall emulsion (CIWE) was prepared and dispersed in corn oil to form an O/W/O double emulsion. The protein solution of the dispersed phase of the double emulsion turned into protein gel during incubation at 37 °C. Wall solutions containing 10% ISP were prepared in distilled
2718
Water solubility
The water solubility of protein capsules was determined according to the method of Lee and Rosenberg (2000a). Microcapsule samples (1.0 g) were placed in a glass vial containing 25 mL of water and incubated at 4 °C, 25 °C, and 37 °C for 12 h. The concentration of soluble protein in the aqueous phase was determined using a “Sigma diagnostic” protein assay kit (Sigma Diagnostics, St. Louis, Mo., U.S.A.). Bovine serum albumin (BSA) was used as standard protein. Each sample was assayed in triplicate.
URLs and E-mail addresses are active links at www.ift.org
JOURNAL OF FOOD SCIENCE—Vol. 68, Nr. 9, 2003
Encapsulation by TGase cross-linked protein . . .
Controlled release and core retention
To 10.0 g of microcapsule, 20 mL of a pepsin solution (2 mg/mL citric acid, pH 2.0) was added and incubated at 37 °C for 4 h. Samples were taken at intervals of 1 h until 4 h storage time, and the total oil content was measured. The controlled release of fish oil from the encapsulated powder was expressed as a proportion (%) of fish oil released from the microcapsules to fish oil retained in the dry microcapsules. To determine the total oil retention, samples were treated in a pepsin solution for 5 h. The total oil retention in the encapsulated powder was expressed as a proportion (%) of fish oil retained in the dry microcapsules to fish oil initially included in the primary emulsion. that is, oxidative stability, water solubility, release rate, and particle size.
Results and Discussion
Selection of a protein-based wall material
Stability of the primary emulsion is necessary for successful stabilization of a multiple emulsion. Proteins can stabilize the emulsion by adsorbing to an O/W interface. For this reason, different proteins show different emulsion stabilizing activities. To select a proper protein wall material, O/W emulsions at 5% and 10% protein concentrations with and without MTGase treatment were prepared and their emulsion stabilities were measured (Figure 1). SWP and WPI showed poor emulsifying properties without MTGase and increased emulsion stabilities after MTGase treatment. The emulsion stabilities of ISP and Na-caseinate increased with an increasing protein concentration. The emulsion stabilizing activity of ISP usually owes to the complexity of the tertiary and quaternary structures of the 2 components, 7S and 11S globulins (Liu and others 1999). Soluble wheat protein (SWP), whey protein isolate (WPI), Nacaseinate, and isolated soy protein (ISP) were selected as MTGase substrates and their reactivities were compared. In most cases, the viscosity and elasticity of proteins cross-linked by MTGase are increased. However, the viscoelastic properties of WPI and SWP solutions were not changed after MTGase treatment (Figure 2). The storage modulus G of the ISP solution increased faster with time (G > 50 Pa at 10 min) than did the modulus for Na-caseinate (G > 50 Pa at 30 min). The final G value of the ISP solution after MTGase treatment was 520 Pa, whereas the G value of the Nacaseinate solution was 60 Pa. The gelation properties of ISP and Nacaseinate were improved by MTGase treatment and ISP had superior gelling capacity than did the other proteins. ISP was selected as a wall material for microcapsule preparation.
Oxidative stability
The peroxide value of microencapsuled/nonencapsuled fish oil was estimated using a p-anisidine value (p-A.V.). Samples were stored in an incubator at 50 °C and withdrawn every 2 d. One gram of sample was placed in a 25-mL volumetric flask and dissolved with n-hexane. The absorbance of this solution was measured at 350 nm. Five milliliters of the fat solution and the solvent was transferred to a test tube and 1 mL of p-anisidine reagent was added. After 10 min, the absorbance of the fat solution in the test tube was measured as a blank in the reference cell. The p-A.V. is given by the formula:
where: As = absorbance of the fat solution after reaction with the p-anisidine reagent; Ab = absorbance of the fat solution; and m = mass (in g) of the test portion.
Statistical analysis
A Pearson analysis of variance (ANOVA) correlation analysis using SAS program (SAS 1993) was performed with the 4 variables,
Optimization of emulsification and the encapsulation process
Stability of the primary emulsion is necessary for successful sta-
Figure 1—Effects of different protein-based wall materials and their concentrations on emulsion stability. = 5% protein concentration without MTGase treatment; = 10% protein concentration without MTGase treatment; = 10% protein concentration with MTGase treatment.
URLs and E-mail addresses are active links at www.ift.org
Figure 2—Changes in storage modulus values of different protein solutions during MTGase treatment at 37 °C. = whey protein isolate (WPI); = soluble wheat protein (SWP); = Na-caseinate; = isolated soy protein (ISP). Vol. 68, Nr. 9, 2003—JOURNAL OF FOOD SCIENCE 2719
Food Engineering and Physical Properties
Encapsulation by TGase cross-linked protein . . .
bilization of a multiple emulsion. Emulsifying agents lower the surface tension of droplets and form a barrier to help prevent droplet coalescence (Risch and Reineccius 1988). Changes in the viscoelastic property of the ISP solution during MTGase treatment are shown in Figure 3. Both the storage modulus G and stability of the emulsion were increased with an increasing ISP concentration. A 10% ISP solution had a stable CIWE with a high viscoelastic value. Both the increased viscosity and the protein concentration favor the stability of the emulsion, in agreement with Hwang and others (1992). The concentration of MTGase influenced the gel formation of ISP (data not shown). ISP was used as an emulsifying agent for maintaining primary O/W emulsion (CIWE) stability. Other types of emulsifying agents are also required for preparation of an O/W/O double emulsion. Considering double emulsion stability and capsule productivity, Span 80 was the best hydrophobic surfactant for microcapsule preparation. SDS-PAGE analysis was performed to determine the effect of reaction time on the extent of MTGase polymerization of ISP (data not shown). Molecular weights for native ISP were between 20 and 220 kDa. The intensity of protein bands decreased with the reaction time up to 3 h and disappeared at 4 h, which indicated ISP treated with MTGase resulted in formation of high molecular-weight polymer aggregates, which did not enter the stacking gel. This was accompanied by a decrease in band intensities for the unpolymerized protein fractions. In contrast, no polymerization products were observed for the control. These results indicate formation of intermolecular cross-linked polymers catalyzed by MTGase. capsules prepared from enzymatic gelation exhibited some surface pores. These pores were probably the footprints of core droplets that were originally present at the surface and were lost during the microencapsulation process (Lee and Rosenberg 2000a). Heatgelled capsules were roughly spherical in shape and had very porous and wrinkled surfaces (Figure 5c). According to Lee and Rosenberg (2000a), these structural features suggest aggregation of protein matrices rather than formation of a continuous film. Water-solubility of microcapsules for controlled and sustained core release in an aqueous environment is of great importance to the functionality of microcapsules. Microcapsule design requires limited or delayed water solubility (Lee and Rosenberg 2000b). The water solubility of microcapsules is shown in Figure 6. MTGasegelled capsules had remarkably lower water solubility. MTGasegelled capsules were not influenced by the incubation temperature, whereas the solubility of heat-induced capsules was influenced by the incubation temperature. The solubility of the capsules increased with the temperature, influencing the microstructural properties. The difference in water solubility is probably related to differences in the cross-linking efficiency of proteinbased wall matrices obtained by enzyme-induced gelation and heat-induced gelation (Lee and Rosenberg 2000a).
Controlled release and core retention
The release mechanism of fish oil from microcapsules in a pepsin solution was investigated because MTGase-gelled capsules were insoluble in the water. The initial release of heat-gelled capsules was followed by 1st-order kinetics involving a core material burst out from the capsules (Figure 7). MTGase-gelled capsules and MTGase/heat-gelled capsules exhibited a sustained release of fish oil. These capsules showed a half-order release profile up to 120 min of incubation time. The release rate shifted toward 1st-order kinetics as time progressed. The controlled-release mechanism of these capsules was probably related to the surface characteristics of the microcapsules. The initial “burst effect” of heat-gelled capsules can be interpreted as a fast release of fish oil from the surface pores of the microcapsules (Cho 2000). At the end of the incubation time the fish-oil contents released from the microcapsules were 78%, 81%, and 98% for MTGase-
Food Engineering and Physical Properties
Physical properties of microcapsules
Microcapsules had a narrow particle-size range (30 to 60 m) with a relatively uniform distribution (Figure 4). There were no significant differences (P > 0.05) in the mean diameters of MTGasegelled, MTGase/heat-gelled, and heat-gelled capsules. Smaller particles of approximately 2 m were present that were probably fragments of denatured proteins. The morphology of the microcapsules containing fish oil was observed under a scanning electron microscope (Figure 5). MTGase-gelled capsules and MTGase/heat-gelled capsules had a spherical shape and exhibited a smooth, dent-free surface. Micro-
Figure 3—Effects of the isolated soy protein (ISP) concentration on the emulsion stability and storage modulus of emulsions. - = storage modulus; = emulsion stability index 2720
Figure 4—Particle size distribution of isolated soy protein (ISP) microcapsules containing fish oil. — — = MTGase— gelled capsule; . . . . . . = MTGase/heat-gelled capsule; ——— = heat-gelled capsule.
URLs and E-mail addresses are active links at www.ift.org
JOURNAL OF FOOD SCIENCE—Vol. 68, Nr. 9, 2003
Encapsulation by TGase cross-linked protein . . .
gelled capsule, MTGase/heat-gelled capsule, and heat-gelled capsule respectively, which was maintained over a 5-h incubation time. These values indicated the total fish-oil retention for each microcapsule type. Heat-induced gelation had the highest efficiency. In most cases, core retention for microcapsules prepared by the double emulsion/subsequent gelation method is lower than for spraydried microcapsules because core losses during the process can occur during formation of the O/W/O double emulsion during the cross-linking and washing stages (Lee and Rosenberg 2000b). The low core retention in microcapsules prepared by enzyme-induced gelation is probably because of the extended gelation time (4 h) compared with heat-induced gelation and chemical cross-linking (20 to 40 min). Some fish oil may be flushed from wall materials during the cross-linking stage. The high core retention of heatgelled capsules may be because of the shorter gelation time and the presence of effective emulsifier in the external phase of the double emulsion, thus assuring the stability of the double emulsion.
Figure 6—Changes in solubility of isolated soy protein (ISP) microcapsules at 3 different incubation temperatures ( = at 37 °C; = at 25 °C; = at 4 °C). (a) MTGase-gelled capsule, (b) MTGase/heat-gelled capsule, and (c) heatgelled capsule.
Figure 5—SEM micrographs of isolated soy protein (ISP) microcapsules containing fish oil. (a) MTGase-gelled capsule at 2000×, (b) MTGase/heat-gelled capsule at 2000×, and (c) heat-gelled capsule at 2000×. — = 15 m.
URLs and E-mail addresses are active links at www.ift.org
Figure 7—Cumulative release of fish oil from isolated soy protein (ISP) microcapsules during incubation at 37 °C. = MTGase-gelled capsule; = MTGase/heat-gelled capsule; = heat-gelled capsule. Vol. 68, Nr. 9, 2003—JOURNAL OF FOOD SCIENCE 2721
Food Engineering and Physical Properties
Encapsulation by TGase cross-linked protein . . .
Oxidative stability
The p-anisidine reagent reacts with oxidation products, such as aldehydes (principally 2-alkenals and 2, 4-dienals), producing a yellowish product. Hence, an increased p-anisidine value (p-A.V.) indicates an increase in the amount of the oxidation product. The rate of fat oxidation is a function of several factors, including water activity, availability of oxygen, and the presence of antioxidants (Kim and Morr 1996). Figure 8 shows changes of the p-A.V. value for nonencapsulated and encapsulated fish oil during storage. The increasing rate of the p-A.V. value for nonencapsulated fish oil was much faster than for encapsulated fish oil. ISP film acted as a good barrier, reducing the rate of fish-oil oxidation. Moreau and Rosenberg (1998) showed that the porosity of the matrix is affected by the fat core and encapsulant levels used. However, oxidation of the fat core material did not occur over a 12-mo period with protein wall matrices. Pearson correlation analysis was performed with the 3 variables of mean particle size of microcapsules, the release rate of fish oil, and water solubility of the microcapsules at 37 °C at incubation temperatures (Table 1). The stability study by p-anisidine production from encapsulated fish oil during storage was excluded from the variables for Pearson correlation analysis because no significant differences were observed among the 3 encapsulated samples. A strong correlation between the release rate of fish oil from microcapsules and the water solubility of microcapsules incubated at 37 °C was detected. Microcapsules having higher water solubility exhibited a faster release rate of fish oil. No linear correlation was observed between the release rate or the water solubility and the mean microcapsule particle size.
Table 1—Correlation analysis: Pearson correlation coefficients Particle size ( m) Particle size ( m) Release rate (%) Water solubility (%) 1.0000 0.3536 0.6166 Release rate (%) 0.3536 1.0000 0.9544 Water solubility (%) 0.6166 0.9544 1.0000
Conclusions
For gel formation of a double emulsion, a 4-h incubation time was required. There were no significant differences in the mean diameter of microcapsules prepared with different gelation methods. The size of capsules manufactured by the O/W/O double emulsification method was not influenced by the gelation method. However, the capsule morphology was highly influenced by the gelation method. Microcapsules prepared by enzyme (MTGase)-induced gelation were spherical in shape with a dent-free surface, while heat-gelled capsules were wrinkled with a rough surface. MTGase-gelled capsules exhibited lower water solubility and sustained release in a pepsin solution. However, the total oil retention for MTGase-gelled capsules was lower than for heat-gelled capsules. Regardless of the gelation method, encapsulated fish oil was more stable than nonencapsulated fish oil for each microcapsule type, indicating that the encapsulation method presented herein is effective for protection of the core material from oxidation. A microencapsulation process consisting of double emulsification and subsequent gelation is suitable for protecting sensitive ingredients and for controlled release in the food industry.
Food Engineering and Physical Properties
M
TGase modified intra- and inter-molecular interactions with protein-based wall materials increasing the emulsion stability and rheological properties of proteins. ISP exhibited the highest emulsion stability and gelling capacity. A CIWE (primary O/W emulsion) prepared with 10% ISP was stable without any emulsifier.
References
Ahn T, Kim J, Kim H, Park K, Choi C. 1991. Antioxidative effect of commercial lecithin on the oxidative stability of fish oil. Korean J Food Sci Technol 23(5):578– 81. Arshady R. 1990. Microspheres and microcapsules, a survey of manufacturing techniques. Part II: Coacervation. Polym Eng Sci 30(15):905–14. Babiker EE. 2000. Effect of transglutaminase treatment on the functional properties of native and chymotrypsin-digested soy protein. Food Chem 70:139– 45. Babin H, Dickinson E. 2001. Influence of transglutaminase treatment on the thermoreversible gelation of gelatin. Food Hydrocolloid 15:271–6. Chang PS, Ha JS. 2000. Optimization of fish oil microencapsulation by response surface methodology and its storage stability. Korean J Food Sci Technol 32(3):646–53. Cho YH. 2000. Development of emulsification and encapsulation process for the stability improvement and controlled release of fruit flavor compounds [DPhil thesis]. Seoul, Korea: Yonsei Univ. 47 p. Hwang J, Kim YS, Pyun YR. 1992. Effect of protein and oil concentration on the emulsion stability of soy protein isolate. J Korean Agric Chem Soc 35(6):457– 61. Kantor ML, Steiner SS, Pack HM, inventors; Clinical Technology Associate Inc, assignee. 1990 Apr 16. Microencapsulation of fish oil. Japan patent 02-103289. Keogh MK, O’Kennedy BT, Kelly J, Auty MA, Kelly PM, Fureby A, Haar AM. 2001. Stability of oxidation of spray-dried fish oil powder microencapsulated using milk ingredients. J Food Sci 66(2):217–24. Kim YD, Morr CV. 1996. Microencapsulation properties of gum Arabic and several food proteins: spray-dried orange oil emulsion particles. J Agric Food Chem 44(5):1314–20. Kitessa SM, Gulati SK, Ashes JR, Fleck E, Scott TW, Nichols PD. 2001. Utilisation of fish oil in ruminants I. Fish oil metabolisom in sheep. Animal Feed Sci Technol 89:189–99. Laemmli UK. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophase T4. Nature 222:680–5. Lee SJ, Rosenberg M. 2000a. Whey protein-based microcapsules prepared by double emulsification and heat gelation. Lebensm-Wiss Technol 33:80–8. Lee SJ, Rosenberg M. 2000b. Preparation and some properties of water-insoluble, whey protein based microcapsules. J Microencaps 17:29–44. Lim LT, Mine Y, Tung MA. 1999. Barriers and tensile properties of transglutaminase cross-linked gelatin films as affected by relative humidity, temperature, and glycerol content. J Food Sci 64(4):616–22. Liu M, Lee D, Damodaran S. 1999. Emulsifying properties of acidic sub-units of soy 11S globulin. J Agric Food Chem 47(12):4970–5. Makrides M, Neuman M, Simmer K, Pater J, Gibson R. 1995. Are long-chain polyunsaturated fatty acids essential nutrient in infancy? Lancet 345:1463–8. Maruyama T, Yamamoto Y, inventors; Meiji Milk Product Co., assignee. 1988 Feb
Figure 8—Changes in the p-anisidine concentration in fish oil in isolated soy protein (ISP) microcapsules during storage at 50 °C. = nonencapsulated fish oil; = MTGasegelled capsule; = MTGase/heat-gelled capsule; = heatgelled capsule. 2722
JOURNAL OF FOOD SCIENCE—Vol. 68, Nr. 9, 2003
URLs and E-mail addresses are active links at www.ift.org
Encapsulation by TGase cross-linked protein . . .
1. Preparation of microcapsule containing fats of highly unsaturated fatty acid. Japan patent 63-023736. Moreau DL, Rosenberg M. 1998. Porosity of whey protein-based microcapsules containing anhydrous milkfat measured by gas displacement phenometry. J Food Sci 63:819–23. Motoki M, Aso H, Seguro K, Nio N. 1987. s1-Casein film preparation using transglutaminase. Agric Biol Chem 51(4):993–6. Nielsen PM. 1995. Reactions and potential industrial applications of transglutaminase: review of literature and patents. Food Biotech 6(3):119–56. Nio N, Motoki M, Takinami K. 1986. Gelation of protein emulsion by transglutaminase. Agric Biol Chem 50(6):1409–12. Risch SJ, Reineccius GA. 1988. Spray dried orange oil-effect of emulsion size on flavor retention and shelf stability. In: Risch SJ, Reineccius GA, editors. Flavor encapsulation. ACS symposium series 370. Chicago, Ill.: American Chemical Society. p 67–77. Roesch RR, Corredig M. 2002. Characterization of oil-in-water emulsions prepared with commercial soy protein concentrate. J Food Sci 67(8):2837–42. SAS. 1993. SAS procedures guide, version 6. Cary, N.C.: SAS Institute, Inc. Smit EN, Oelen EA, Seerat E, Boersma ER, Muskiet FAJ. 2000. Fish oil supplementation improves docosahexaenoic acid status of malnourished infants. Arch Dis Child 82:366–9. Yildirim M, Hettiarachchy NS. 1998. Properties of films produced by cross-linking whey proteins and 11S globulin using transglutaminase. J Food. Sci 63:248–52.
URLs and E-mail addresses are active links at www.ift.org
Vol. 68, Nr. 9, 2003—JOURNAL OF FOOD SCIENCE
2723
Food Engineering and Physical Properties
Encapsulation of Fish Oil by an Enzymatic Gelation Process Using Transglutaminase Cross-linked Proteins
Y.-H. CHO, H.K. SHIM, AND J. PARK ABSTRACT: To improve the storage stability and achieve controlled release, fish oil containing docosahexaenoic acid was encapsulated using double emulsification and subsequent enzymatic gelation method, using microbial transglutaminase cross-linked proteins. Isolated soy protein was selected as a wall material because it showed better emulsion stability and higher reactivity with MTGase than other proteins. Microcapsules prepared by this method showed a high stability against oxygen and a low water solubility, which subsequently resulted in sustained release of fish oil. Results indicate that this microencapsulation process is suitable for preparing protein-based microcapsules containing sensitive ingredients for controlled release and stability improvement. Keywords: fish oil, microencapsulation, microbial transglutaminase, isolated soy protein, double emulsification
Introduction
here is increasing information about the nutritional and health benefits of long chain polyunsaturated fatty acids (LC-PUFA) for humans (Kitessa and others 2001). LC-PUFAs, such as arachidonic acid (AA) and docosahexaenoic acid (DHA), are structural components of cell membrane phospholipids and precursors of eicosanoids and play important roles in the development of the central nervous system, including the retina (Smit and others 2000). Fish oil is an important source of DHA and many studies have shown that fish-oil supplementation increases the DHA content of blood components (Markrides and others 1995). However, the application of fish oils in the food system has a limitation because oils are unstable against oxygen during storage. Also, the unpleasant taste and odor of fish oil can cause deterioration in the quality of food products. Research has been performed to overcome the limitations of fish oils. Ahn and others (1991) enhanced oxidative stability of EPA and DHA by adding lecithin. However, the sensory evaluation of fish oil remained a problem. Several methods have been studied for encapsulation of fish oil, and a variety of materials have been used as a wall material, including cellulose (Kantor and others 1990), gum Arabic and gelatin (Maruyama and Yamamoto 1988), agar and waxy corn starch (Chang and Ha 2000), and milk proteins (Keogh and others 2001). However, carbohydratebased microcapsules or protein-based microcapsules prepared by spray-drying are water soluble and thus not suitable for controlled-release applications. Protein films are, in general, excellent oxygen and aroma barriers and can be used in developing microcapsules for controlled core release in food applications (Lee and Rosenberg 2000a). Microcapsules using proteins as wall materials have been prepared by simple/complex coacervation (Arshady 1990) and techniques consistMS 20030315 Submitted 6/6/03, Revised 10/1/00, Accepted 10/11/00. Authors Cho, Shim, and Park are with Dept. of Biotechnology, Yonsei Univ., Seoul 120-749, South Korea. Direct inquiries to author Park (foodpro@ yonsei.ac.kr).
T
ing of double emulsification and subsequent cross-linking with glutaraldehyde or heat-induced gelation (Lee and Rosenberg 2000a). Both approaches provide a means for achieving high core retention and render the resulting microcapsule water insoluble (Lee and Rosenberg 2000b). However, the use of protein-based wall materials in the food industry for sensitive ingredients is limited because proteins are generally unstable with heating and damaged by organic solvents, and the cross-linking agent is usually harmful. Therefore, if a novel, mild cross-linking agent is developed, and proteins are converted into stable forms, the application of proteins would be greatly increased (Babiker 2000). Microbial transglutaminase (MTGase) is an enzyme excreted from Streptoverticillium mobaraense that catalyzes acyl-transfer reactions, resulting in formation of -( -glutaminyl) lysine intraand inter-molecular cross-links in proteins (Nielsen 1995). Several studies have been conducted on cross-linking various proteins using this enzyme to form biopolymer films (Nio and others 1986; Motoki and others 1987; Yildirim and Hettiarachchy 1998; Lim and others 1999; Liu and others 1999; Babiker 2000; Babin and Dickinson 2001). These attempts focused on the preparation of biodegradable films, while we have attempted to use MTGase crosslinked proteins as a wall material in the encapsulation process. We investigated a number of proteins for use as base materials for preparing biodegradable films using MTGase and developed a double emulsification and subsequent enzymatic gelation method using isolated soy protein as a wall material. Soy protein is an abundant byproduct of the soybean-oil industry. Soy proteins have important applications in many food products because of their high nutritional value and their contribution to food texture and emulsifying properties (Roesch and Corredig 2002). Microcapusules containing fish oil were prepared using oil-inwater-in-oil (O/W/O) double emulsification and 3 different gelation methods, including enzymatic gelation using MTGase cross-linked protein, heat gelation, and a combination of the enzyme and heat gelation methods. The physicochemical properties, microstructure, and controlled release of 3 types of fish-oil microcapsules were investigated and compared.
Vol. 68, Nr. 9, 2003—JOURNAL OF FOOD SCIENCE 2717
© 2003 Institute of Food Technologists
Further reproduction prohibited without permission
Food Engineering and Physical Properties
Encapsulation by TGase cross-linked protein . . .
Materials and Methods
Materials
Fish oils were supplied by Loders Croklaan (Marinol D-40, Wormerveer, The Netherlands) and stored at –18 °C. Soluble wheat protein (SWP; Sam-Mi Industrial Co., Ansan, South Korea), whey protein isolate (WPI; Sam-Mi Industrial Co.), Na-caseinate (Sam-Mi Industrial Co.), and isolated soy protein (ISP; Sam-Asia, Seoul, South Korea) were selected as wall materials. Proteins were crosslinked using MTGase (Ajinomoto, Tokyo, Japan) that is derived from S. mobaraense. Before use, MTGase powder (4 mg) was suspended in 0.5 M Tris-acetate buffer (1 mL, pH 6.0). Span 80 (Duksan Chemical Co., Seoul, South Korea) was selected as an emulsifier for enhancement of the emulsion stability. Corn oil was supplied by the CJ Co. (Seoul, South Korea). water, and then MTGase (0.025%) was added into protein wall solutions. A CIWE (O/W primary emulsion) was prepared by blending a core (fish oil)/wall (1:2) mixture at 9500 rpm for 10 min using Ultra-Turrax T25TM. Corn oil (400 g) containing 3% emulsifier (Span 80) was preheated at 50 °C in an incubator to dissolve emulsifiers. The CIWE (100 g) was slowly added to the corn oil under the above conditions and then incubated at 37 °C for 4 h. The protein matrices of the dispersed phase of the double emulsion were gelled by reaction with MTGase. Microcapsules were separated from the corn oil by filtration through Whatman nr 4 filter paper (Whatman Intl. Ltd., Maidstone, U.K.), washed with ethanol 3 times, and freeze dried at 10 °C for 24 h using a freeze dryer (FD-558, Heto, Llerod, Denmark). (B) Combination of enzymatic and heat gelation (MTGase/ heat-gelled capsule). The encapsulation process was the same as (A) except for the gelation method. After preparing an O/W/O double emulsion, the temperature of the emulsion was adjusted to 85 °C, and then gelation was performed for 4 h. (C) Double emulsification and heat gelation (heat-gelled capsule). The encapsulation process was the same as (A) except for preparation of the wall solution and the gelation method. A wall solution containing 10% ISP was prepared in distilled water without a reaction with MTGase. An O/W/O emulsion was prepared and heated at 85 °C, then gelation was performed for 40 min.
Assay of microbial transglutaminase activity
MTGase activity was measured using the colorimetric hydroxaminate assay method. This assay was performed in 0.5 mL of an enzyme-substrate solution containing 0.1 mL of MTGase solution, 0.1 mL of 0.5 M Tris-acetate buffer (pH 6.0), 0.2 mL of 0.3 M CBZ-Lglutaminylglycine, 0.05 mL of 0.02 M EDTA, and 0.05 mL of 4 M hydroxylamine. After 10 min of incubation at 37 °C, 0.5 mL of ammonium sulfate was added to stop the reaction. The resulting red color was measured at 525 nm. One unit of MTGase activity was defined as the amount of enzyme needed to produce 1 micromole of hydroxamic acid per minute.
Determination of emulsion stability
A sample of liquid emulsion was transferred to a 10-mL mass cylinder, which was then capped and stored for 24 h. The volume of oil separated from emulsion in the mass cylinder was read. Results are reported as an emulsion stability index (ESI) with a range of possible results from 0 to 1.
Food Engineering and Physical Properties
Preparation of O/W emulsions for selection of the wall material
Protein solutions of 5% and 10% were prepared with and without MTGase (0.025%). An O/W emulsion was prepared by homomixing fish oil in each protein solution with Ultra-Turrax T25TM homogenizer (tool S25N-25G, IKA Labortechnik, Staufen, Germany) at 9500 rpm for 10 min.
Rheological measurement
The viscoelastic properties of MTGase-treated protein solutions were measured using a 40-mm-dia rheometer (AR1000, TA Instrument, New Castle, Del., U.S.A.) with cone steel geometry. Protein solutions were carefully poured into the rheometer cup (40-mm dia) and covered with a thin layer of vacuum oil to prevent evaporation. The temperature was maintained at 37 °C for 60 min. The storage (G ) and loss (G ) moduli were recorded as a function of time at a frequency of 1 Hz and at a maximum strain of 0.05 in the linear region. Each sample was assayed in triplicate.
A value of 0 represents poor emulsion stability while a value of 1 represents high emulsion stability (Cho 2000). Each sample was assayed in triplicate.
Particle-size distribution
The microcapsules were dispersed in distilled water. The particle-size distribution, the mean particle size, and the specific surface area of microcapsules were determined using a particle-size analyzer (Model 1064, Cilas, Orleans, France).
Scanning electron microscopy (SEM) Electrophoretic analysis of polymerization
SDS-PAGE was performed according to the method of Laemmli (1970). Samples were run on 12.5% gels. Proteins were dissolved in a sample buffer in the presence of -mercaptoethanol, heated for 10 min at 100 °C, and loaded onto the gel at 20- L/well. Gels were stained with Coomassie brilliant blue R-250. The overall morphology and the outer topography of dry microcapsules were observed by scanning electron microscopy (JSM5410L, Jeol Co., Peabody, Mass., U.S.A.) at an accelerating voltage of 10 kV. Microcapsules were mounted on metal stubs using double-sided adhesive tape and coated with Pb-Pt in an ion sputter ( 1030, Hitachi, Tokyo, Japan) under an argon atmosphere.
Preparation of emulsion and microcapsules
(A) Double emulsification and enzymatic gelation (MTGasegelled capsule). Microcapsules were prepared using a process consisting of double emulsification and subsequent enzymatic gelation. In general, a core-in-wall emulsion (CIWE) was prepared and dispersed in corn oil to form an O/W/O double emulsion. The protein solution of the dispersed phase of the double emulsion turned into protein gel during incubation at 37 °C. Wall solutions containing 10% ISP were prepared in distilled
2718
Water solubility
The water solubility of protein capsules was determined according to the method of Lee and Rosenberg (2000a). Microcapsule samples (1.0 g) were placed in a glass vial containing 25 mL of water and incubated at 4 °C, 25 °C, and 37 °C for 12 h. The concentration of soluble protein in the aqueous phase was determined using a “Sigma diagnostic” protein assay kit (Sigma Diagnostics, St. Louis, Mo., U.S.A.). Bovine serum albumin (BSA) was used as standard protein. Each sample was assayed in triplicate.
URLs and E-mail addresses are active links at www.ift.org
JOURNAL OF FOOD SCIENCE—Vol. 68, Nr. 9, 2003
Encapsulation by TGase cross-linked protein . . .
Controlled release and core retention
To 10.0 g of microcapsule, 20 mL of a pepsin solution (2 mg/mL citric acid, pH 2.0) was added and incubated at 37 °C for 4 h. Samples were taken at intervals of 1 h until 4 h storage time, and the total oil content was measured. The controlled release of fish oil from the encapsulated powder was expressed as a proportion (%) of fish oil released from the microcapsules to fish oil retained in the dry microcapsules. To determine the total oil retention, samples were treated in a pepsin solution for 5 h. The total oil retention in the encapsulated powder was expressed as a proportion (%) of fish oil retained in the dry microcapsules to fish oil initially included in the primary emulsion. that is, oxidative stability, water solubility, release rate, and particle size.
Results and Discussion
Selection of a protein-based wall material
Stability of the primary emulsion is necessary for successful stabilization of a multiple emulsion. Proteins can stabilize the emulsion by adsorbing to an O/W interface. For this reason, different proteins show different emulsion stabilizing activities. To select a proper protein wall material, O/W emulsions at 5% and 10% protein concentrations with and without MTGase treatment were prepared and their emulsion stabilities were measured (Figure 1). SWP and WPI showed poor emulsifying properties without MTGase and increased emulsion stabilities after MTGase treatment. The emulsion stabilities of ISP and Na-caseinate increased with an increasing protein concentration. The emulsion stabilizing activity of ISP usually owes to the complexity of the tertiary and quaternary structures of the 2 components, 7S and 11S globulins (Liu and others 1999). Soluble wheat protein (SWP), whey protein isolate (WPI), Nacaseinate, and isolated soy protein (ISP) were selected as MTGase substrates and their reactivities were compared. In most cases, the viscosity and elasticity of proteins cross-linked by MTGase are increased. However, the viscoelastic properties of WPI and SWP solutions were not changed after MTGase treatment (Figure 2). The storage modulus G of the ISP solution increased faster with time (G > 50 Pa at 10 min) than did the modulus for Na-caseinate (G > 50 Pa at 30 min). The final G value of the ISP solution after MTGase treatment was 520 Pa, whereas the G value of the Nacaseinate solution was 60 Pa. The gelation properties of ISP and Nacaseinate were improved by MTGase treatment and ISP had superior gelling capacity than did the other proteins. ISP was selected as a wall material for microcapsule preparation.
Oxidative stability
The peroxide value of microencapsuled/nonencapsuled fish oil was estimated using a p-anisidine value (p-A.V.). Samples were stored in an incubator at 50 °C and withdrawn every 2 d. One gram of sample was placed in a 25-mL volumetric flask and dissolved with n-hexane. The absorbance of this solution was measured at 350 nm. Five milliliters of the fat solution and the solvent was transferred to a test tube and 1 mL of p-anisidine reagent was added. After 10 min, the absorbance of the fat solution in the test tube was measured as a blank in the reference cell. The p-A.V. is given by the formula:
where: As = absorbance of the fat solution after reaction with the p-anisidine reagent; Ab = absorbance of the fat solution; and m = mass (in g) of the test portion.
Statistical analysis
A Pearson analysis of variance (ANOVA) correlation analysis using SAS program (SAS 1993) was performed with the 4 variables,
Optimization of emulsification and the encapsulation process
Stability of the primary emulsion is necessary for successful sta-
Figure 1—Effects of different protein-based wall materials and their concentrations on emulsion stability. = 5% protein concentration without MTGase treatment; = 10% protein concentration without MTGase treatment; = 10% protein concentration with MTGase treatment.
URLs and E-mail addresses are active links at www.ift.org
Figure 2—Changes in storage modulus values of different protein solutions during MTGase treatment at 37 °C. = whey protein isolate (WPI); = soluble wheat protein (SWP); = Na-caseinate; = isolated soy protein (ISP). Vol. 68, Nr. 9, 2003—JOURNAL OF FOOD SCIENCE 2719
Food Engineering and Physical Properties
Encapsulation by TGase cross-linked protein . . .
bilization of a multiple emulsion. Emulsifying agents lower the surface tension of droplets and form a barrier to help prevent droplet coalescence (Risch and Reineccius 1988). Changes in the viscoelastic property of the ISP solution during MTGase treatment are shown in Figure 3. Both the storage modulus G and stability of the emulsion were increased with an increasing ISP concentration. A 10% ISP solution had a stable CIWE with a high viscoelastic value. Both the increased viscosity and the protein concentration favor the stability of the emulsion, in agreement with Hwang and others (1992). The concentration of MTGase influenced the gel formation of ISP (data not shown). ISP was used as an emulsifying agent for maintaining primary O/W emulsion (CIWE) stability. Other types of emulsifying agents are also required for preparation of an O/W/O double emulsion. Considering double emulsion stability and capsule productivity, Span 80 was the best hydrophobic surfactant for microcapsule preparation. SDS-PAGE analysis was performed to determine the effect of reaction time on the extent of MTGase polymerization of ISP (data not shown). Molecular weights for native ISP were between 20 and 220 kDa. The intensity of protein bands decreased with the reaction time up to 3 h and disappeared at 4 h, which indicated ISP treated with MTGase resulted in formation of high molecular-weight polymer aggregates, which did not enter the stacking gel. This was accompanied by a decrease in band intensities for the unpolymerized protein fractions. In contrast, no polymerization products were observed for the control. These results indicate formation of intermolecular cross-linked polymers catalyzed by MTGase. capsules prepared from enzymatic gelation exhibited some surface pores. These pores were probably the footprints of core droplets that were originally present at the surface and were lost during the microencapsulation process (Lee and Rosenberg 2000a). Heatgelled capsules were roughly spherical in shape and had very porous and wrinkled surfaces (Figure 5c). According to Lee and Rosenberg (2000a), these structural features suggest aggregation of protein matrices rather than formation of a continuous film. Water-solubility of microcapsules for controlled and sustained core release in an aqueous environment is of great importance to the functionality of microcapsules. Microcapsule design requires limited or delayed water solubility (Lee and Rosenberg 2000b). The water solubility of microcapsules is shown in Figure 6. MTGasegelled capsules had remarkably lower water solubility. MTGasegelled capsules were not influenced by the incubation temperature, whereas the solubility of heat-induced capsules was influenced by the incubation temperature. The solubility of the capsules increased with the temperature, influencing the microstructural properties. The difference in water solubility is probably related to differences in the cross-linking efficiency of proteinbased wall matrices obtained by enzyme-induced gelation and heat-induced gelation (Lee and Rosenberg 2000a).
Controlled release and core retention
The release mechanism of fish oil from microcapsules in a pepsin solution was investigated because MTGase-gelled capsules were insoluble in the water. The initial release of heat-gelled capsules was followed by 1st-order kinetics involving a core material burst out from the capsules (Figure 7). MTGase-gelled capsules and MTGase/heat-gelled capsules exhibited a sustained release of fish oil. These capsules showed a half-order release profile up to 120 min of incubation time. The release rate shifted toward 1st-order kinetics as time progressed. The controlled-release mechanism of these capsules was probably related to the surface characteristics of the microcapsules. The initial “burst effect” of heat-gelled capsules can be interpreted as a fast release of fish oil from the surface pores of the microcapsules (Cho 2000). At the end of the incubation time the fish-oil contents released from the microcapsules were 78%, 81%, and 98% for MTGase-
Food Engineering and Physical Properties
Physical properties of microcapsules
Microcapsules had a narrow particle-size range (30 to 60 m) with a relatively uniform distribution (Figure 4). There were no significant differences (P > 0.05) in the mean diameters of MTGasegelled, MTGase/heat-gelled, and heat-gelled capsules. Smaller particles of approximately 2 m were present that were probably fragments of denatured proteins. The morphology of the microcapsules containing fish oil was observed under a scanning electron microscope (Figure 5). MTGase-gelled capsules and MTGase/heat-gelled capsules had a spherical shape and exhibited a smooth, dent-free surface. Micro-
Figure 3—Effects of the isolated soy protein (ISP) concentration on the emulsion stability and storage modulus of emulsions. - = storage modulus; = emulsion stability index 2720
Figure 4—Particle size distribution of isolated soy protein (ISP) microcapsules containing fish oil. — — = MTGase— gelled capsule; . . . . . . = MTGase/heat-gelled capsule; ——— = heat-gelled capsule.
URLs and E-mail addresses are active links at www.ift.org
JOURNAL OF FOOD SCIENCE—Vol. 68, Nr. 9, 2003
Encapsulation by TGase cross-linked protein . . .
gelled capsule, MTGase/heat-gelled capsule, and heat-gelled capsule respectively, which was maintained over a 5-h incubation time. These values indicated the total fish-oil retention for each microcapsule type. Heat-induced gelation had the highest efficiency. In most cases, core retention for microcapsules prepared by the double emulsion/subsequent gelation method is lower than for spraydried microcapsules because core losses during the process can occur during formation of the O/W/O double emulsion during the cross-linking and washing stages (Lee and Rosenberg 2000b). The low core retention in microcapsules prepared by enzyme-induced gelation is probably because of the extended gelation time (4 h) compared with heat-induced gelation and chemical cross-linking (20 to 40 min). Some fish oil may be flushed from wall materials during the cross-linking stage. The high core retention of heatgelled capsules may be because of the shorter gelation time and the presence of effective emulsifier in the external phase of the double emulsion, thus assuring the stability of the double emulsion.
Figure 6—Changes in solubility of isolated soy protein (ISP) microcapsules at 3 different incubation temperatures ( = at 37 °C; = at 25 °C; = at 4 °C). (a) MTGase-gelled capsule, (b) MTGase/heat-gelled capsule, and (c) heatgelled capsule.
Figure 5—SEM micrographs of isolated soy protein (ISP) microcapsules containing fish oil. (a) MTGase-gelled capsule at 2000×, (b) MTGase/heat-gelled capsule at 2000×, and (c) heat-gelled capsule at 2000×. — = 15 m.
URLs and E-mail addresses are active links at www.ift.org
Figure 7—Cumulative release of fish oil from isolated soy protein (ISP) microcapsules during incubation at 37 °C. = MTGase-gelled capsule; = MTGase/heat-gelled capsule; = heat-gelled capsule. Vol. 68, Nr. 9, 2003—JOURNAL OF FOOD SCIENCE 2721
Food Engineering and Physical Properties
Encapsulation by TGase cross-linked protein . . .
Oxidative stability
The p-anisidine reagent reacts with oxidation products, such as aldehydes (principally 2-alkenals and 2, 4-dienals), producing a yellowish product. Hence, an increased p-anisidine value (p-A.V.) indicates an increase in the amount of the oxidation product. The rate of fat oxidation is a function of several factors, including water activity, availability of oxygen, and the presence of antioxidants (Kim and Morr 1996). Figure 8 shows changes of the p-A.V. value for nonencapsulated and encapsulated fish oil during storage. The increasing rate of the p-A.V. value for nonencapsulated fish oil was much faster than for encapsulated fish oil. ISP film acted as a good barrier, reducing the rate of fish-oil oxidation. Moreau and Rosenberg (1998) showed that the porosity of the matrix is affected by the fat core and encapsulant levels used. However, oxidation of the fat core material did not occur over a 12-mo period with protein wall matrices. Pearson correlation analysis was performed with the 3 variables of mean particle size of microcapsules, the release rate of fish oil, and water solubility of the microcapsules at 37 °C at incubation temperatures (Table 1). The stability study by p-anisidine production from encapsulated fish oil during storage was excluded from the variables for Pearson correlation analysis because no significant differences were observed among the 3 encapsulated samples. A strong correlation between the release rate of fish oil from microcapsules and the water solubility of microcapsules incubated at 37 °C was detected. Microcapsules having higher water solubility exhibited a faster release rate of fish oil. No linear correlation was observed between the release rate or the water solubility and the mean microcapsule particle size.
Table 1—Correlation analysis: Pearson correlation coefficients Particle size ( m) Particle size ( m) Release rate (%) Water solubility (%) 1.0000 0.3536 0.6166 Release rate (%) 0.3536 1.0000 0.9544 Water solubility (%) 0.6166 0.9544 1.0000
Conclusions
For gel formation of a double emulsion, a 4-h incubation time was required. There were no significant differences in the mean diameter of microcapsules prepared with different gelation methods. The size of capsules manufactured by the O/W/O double emulsification method was not influenced by the gelation method. However, the capsule morphology was highly influenced by the gelation method. Microcapsules prepared by enzyme (MTGase)-induced gelation were spherical in shape with a dent-free surface, while heat-gelled capsules were wrinkled with a rough surface. MTGase-gelled capsules exhibited lower water solubility and sustained release in a pepsin solution. However, the total oil retention for MTGase-gelled capsules was lower than for heat-gelled capsules. Regardless of the gelation method, encapsulated fish oil was more stable than nonencapsulated fish oil for each microcapsule type, indicating that the encapsulation method presented herein is effective for protection of the core material from oxidation. A microencapsulation process consisting of double emulsification and subsequent gelation is suitable for protecting sensitive ingredients and for controlled release in the food industry.
Food Engineering and Physical Properties
M
TGase modified intra- and inter-molecular interactions with protein-based wall materials increasing the emulsion stability and rheological properties of proteins. ISP exhibited the highest emulsion stability and gelling capacity. A CIWE (primary O/W emulsion) prepared with 10% ISP was stable without any emulsifier.
References
Ahn T, Kim J, Kim H, Park K, Choi C. 1991. Antioxidative effect of commercial lecithin on the oxidative stability of fish oil. Korean J Food Sci Technol 23(5):578– 81. Arshady R. 1990. Microspheres and microcapsules, a survey of manufacturing techniques. Part II: Coacervation. Polym Eng Sci 30(15):905–14. Babiker EE. 2000. Effect of transglutaminase treatment on the functional properties of native and chymotrypsin-digested soy protein. Food Chem 70:139– 45. Babin H, Dickinson E. 2001. Influence of transglutaminase treatment on the thermoreversible gelation of gelatin. Food Hydrocolloid 15:271–6. Chang PS, Ha JS. 2000. Optimization of fish oil microencapsulation by response surface methodology and its storage stability. Korean J Food Sci Technol 32(3):646–53. Cho YH. 2000. Development of emulsification and encapsulation process for the stability improvement and controlled release of fruit flavor compounds [DPhil thesis]. Seoul, Korea: Yonsei Univ. 47 p. Hwang J, Kim YS, Pyun YR. 1992. Effect of protein and oil concentration on the emulsion stability of soy protein isolate. J Korean Agric Chem Soc 35(6):457– 61. Kantor ML, Steiner SS, Pack HM, inventors; Clinical Technology Associate Inc, assignee. 1990 Apr 16. Microencapsulation of fish oil. Japan patent 02-103289. Keogh MK, O’Kennedy BT, Kelly J, Auty MA, Kelly PM, Fureby A, Haar AM. 2001. Stability of oxidation of spray-dried fish oil powder microencapsulated using milk ingredients. J Food Sci 66(2):217–24. Kim YD, Morr CV. 1996. Microencapsulation properties of gum Arabic and several food proteins: spray-dried orange oil emulsion particles. J Agric Food Chem 44(5):1314–20. Kitessa SM, Gulati SK, Ashes JR, Fleck E, Scott TW, Nichols PD. 2001. Utilisation of fish oil in ruminants I. Fish oil metabolisom in sheep. Animal Feed Sci Technol 89:189–99. Laemmli UK. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophase T4. Nature 222:680–5. Lee SJ, Rosenberg M. 2000a. Whey protein-based microcapsules prepared by double emulsification and heat gelation. Lebensm-Wiss Technol 33:80–8. Lee SJ, Rosenberg M. 2000b. Preparation and some properties of water-insoluble, whey protein based microcapsules. J Microencaps 17:29–44. Lim LT, Mine Y, Tung MA. 1999. Barriers and tensile properties of transglutaminase cross-linked gelatin films as affected by relative humidity, temperature, and glycerol content. J Food Sci 64(4):616–22. Liu M, Lee D, Damodaran S. 1999. Emulsifying properties of acidic sub-units of soy 11S globulin. J Agric Food Chem 47(12):4970–5. Makrides M, Neuman M, Simmer K, Pater J, Gibson R. 1995. Are long-chain polyunsaturated fatty acids essential nutrient in infancy? Lancet 345:1463–8. Maruyama T, Yamamoto Y, inventors; Meiji Milk Product Co., assignee. 1988 Feb
Figure 8—Changes in the p-anisidine concentration in fish oil in isolated soy protein (ISP) microcapsules during storage at 50 °C. = nonencapsulated fish oil; = MTGasegelled capsule; = MTGase/heat-gelled capsule; = heatgelled capsule. 2722
JOURNAL OF FOOD SCIENCE—Vol. 68, Nr. 9, 2003
URLs and E-mail addresses are active links at www.ift.org
Encapsulation by TGase cross-linked protein . . .
1. Preparation of microcapsule containing fats of highly unsaturated fatty acid. Japan patent 63-023736. Moreau DL, Rosenberg M. 1998. Porosity of whey protein-based microcapsules containing anhydrous milkfat measured by gas displacement phenometry. J Food Sci 63:819–23. Motoki M, Aso H, Seguro K, Nio N. 1987. s1-Casein film preparation using transglutaminase. Agric Biol Chem 51(4):993–6. Nielsen PM. 1995. Reactions and potential industrial applications of transglutaminase: review of literature and patents. Food Biotech 6(3):119–56. Nio N, Motoki M, Takinami K. 1986. Gelation of protein emulsion by transglutaminase. Agric Biol Chem 50(6):1409–12. Risch SJ, Reineccius GA. 1988. Spray dried orange oil-effect of emulsion size on flavor retention and shelf stability. In: Risch SJ, Reineccius GA, editors. Flavor encapsulation. ACS symposium series 370. Chicago, Ill.: American Chemical Society. p 67–77. Roesch RR, Corredig M. 2002. Characterization of oil-in-water emulsions prepared with commercial soy protein concentrate. J Food Sci 67(8):2837–42. SAS. 1993. SAS procedures guide, version 6. Cary, N.C.: SAS Institute, Inc. Smit EN, Oelen EA, Seerat E, Boersma ER, Muskiet FAJ. 2000. Fish oil supplementation improves docosahexaenoic acid status of malnourished infants. Arch Dis Child 82:366–9. Yildirim M, Hettiarachchy NS. 1998. Properties of films produced by cross-linking whey proteins and 11S globulin using transglutaminase. J Food. Sci 63:248–52.
URLs and E-mail addresses are active links at www.ift.org
Vol. 68, Nr. 9, 2003—JOURNAL OF FOOD SCIENCE
2723
Food Engineering and Physical Properties
