sodium caseinate by spray drying

Journal of Food Engineering 225 (2018) 34e41

Contents lists available at ScienceDirect

Journal of Food Engineering journal homepage: www.elsevier.com/locate/jfoodeng

In-vitro digestion of refined kenaf seed oil microencapsulated in b-cyclodextrin/gum arabic/sodium caseinate by spray drying Sook Chin Chew a, Chin Ping Tan b, Kar Lin Nyam a, * a b

Department of Food Science and Nutrition, Faculty of Applied Sciences, UCSI University, 56000 Kuala Lumpur, Malaysia Department of Food Technology, Faculty of Food Science and Technology, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia

a r t i c l e i n f o

a b s t r a c t

Article history: Available online 30 January 2018

Refined kenaf seed oil was microencapsulated by spray drying using the wall materials of b-cyclodextrin (b-CD), gum arabic (GA), and sodium caseinate (SC) to produce three different models (SC:b-CD, GA:b-CD, GA:SC:b-CD) of microencapsulated refined kenaf seed oil (MRKSO). An in-vitro digestion was used to simulate the human gastrointestinal digestion to examine the oil release behavior of MRKSO, changes in antioxidant activity and bioactive compounds of undigested oil, digested oil, and digested MRKSO samples. The results showed that three models of the MRKSO offered good protection by a lower percentage of oil released (1.43e6.44%) in the simulated gastric fluid and a high percentage of oil released (81.10e91.19%) after simulated gastric and intestinal phases digestion. The degree of lipolysis was in the order of SC:b-CD > GA:SC:b-CD > GA:b-CD > un-encapsulated oil. Among three models of MRKSO, GA:SC:b-CD offered better bioaccessibility by showing an increase in DPPH (20.0% increase) and ABTS (5.0% increase) values, phenolic content (130.4% increase), tocopherol and tocotrienol contents (147.7% increase), as well as slower degradation of phytosterol contents (59.4% decrease) after in-vitro digestion, compared to the undigested kenaf seed oil. © 2018 Elsevier Ltd. All rights reserved.

Keywords: Kenaf seed oil Spray drying Oil release Antioxidant activity Tocopherol Phytosterol

1. Introduction Currently, the interest in functional and nutritious oil has increased due to its abundance of unsaturated fatty acids and bioactive compounds (Chew et al., 2015). Kenaf (Hibiscus cannabinus L.) is a valuable fiber plant native to India and Africa from the Malvaceae family. Kenaf seeds have been widely studied to yield kenaf seed oil that is high in monounsaturated and polyunsaturated fatty acids. Moreover, abundance of lipophilic bioactive compounds is presented in the kenaf seed oil, such as tocopherols and phytosterols, which can offer antioxidant activity (Chew et al., 2016). Thus, refined kenaf seed oil has been suggested to be used as edible and functional oil due to its unique composition (Chew et al., 2016, 2017a). However, the incorporation of these functional ingredients into foods, maintaining their stability and functionality to the gastrointestinal (GI) tract for absorption by the human body is a major challenge faced by the food industry. Presence of high amount of oxygen in the GI tract can favor the oxidation of unsaturated fatty acids (Chew et al., 2015).

* Corresponding author. E-mail address: [email protected] (K.L. Nyam). https://doi.org/10.1016/j.jfoodeng.2018.01.018 0260-8774/© 2018 Elsevier Ltd. All rights reserved.

Microencapsulation has been focused recently to provide a physical barrier to protect the functional oils against degradation of unsaturated fatty acids and bioactive compounds (Chew and Nyam, 2016). Microencapsulation offers controlled release property by delivering the functional ingredients to the small intestine where they are absorbed into the bloodstream (Chew et al., 2015; Timilsena et al., 2017). Spray drying is the most common and economically feasible technology by using the hot gas stream to transform the fluid product into dry powder form to encapsulate functional ingredients in powder form for food applications (Daza et al., 2016; Edris et al., 2016; Goyal et al., 2015). Spray drying has been used to encapsulate kenaf seed oil in the previous study by using maltodextrin, sodium caseinate, and soy lecithin as the wall materials (Ng et al., 2013). However, high glycemic index (>130) and potentially unhealthy effects of maltodextrin have discouraged its application as wall material (SunWaterhouse and Waterhouse, 2015). Dietary fiber has been encouraged to be used as wall material in the encapsulation process due to their health benefits and protective effect on the core substances (Chew et al., 2015). Cyclodextrins (CDs) are dietary fibers that have low glycemic index and can act as prebiotics in the GI tract. CDs are beneficial in controlling human body weight and

S.C. Chew et al. / Journal of Food Engineering 225 (2018) 34e41

blood lipid profile (Fenyvesi et al., 2016). b-CD is the most commonly used among the CDs in the encapsulation process due to their ability to bind the hydrophobic oil in the interior cavity and the hydrophilic exterior surface can bind with water or hydrophilic head of other wall materials to form inclusion complex (Cheong and Nyam, 2016; Hundre et al., 2015). Cheong et al. (2016) reported that the kenaf seed oil encapsulated in nanoemulsion by bCD, sodium caseinate and tween 20 offered a good lipid digestion and good bioaccessibility of antioxidants, as well as low degradation rate of phytosterols during the simulated digestion. It is of interest to study the b-CD to encapsulate kenaf seed oil in a powder form in order to develop the functional product. In this context, b-CD will be blended with other wall materials, such as gum arabic and sodium caseinate to encapsulate refined kenaf seed oil. Gum arabic (GA) is a dietary fiber that can ferment in the colon to produce short-chain fatty acids that offer prebiotic effect and also possesses high solubility, low viscosity and good emulsifying properties (Babiker et al., 2012; Fernandes et al., 2013). Sodium caseinate (SC) is a water-soluble milk protein that could offer good emulsifying and encapsulation properties (Wang et al., 2016). To the best of our knowledge, no information has been reported on the microencapsulation by using the formulation of bCD/GA/SC by spray drying. Hence, in order to ensure the delivery of the refined kenaf seed oil to the targeted sites of the human GI tract, it is important to test the stability and the release behavior of the oil during in-vitro digestion using simulated digestive fluids. Previous study reported the formation of a secondary emulsion by addition of b-CD to the primary emulsion had remarkably increased its stability by binding of the hydroxyl group of b-CD with the hydrophilic head group of other wall materials via hydrogen bonding in the emulsion (Cheong and Nyam, 2016). Thus, b-CD would add as a secondary layer into the primary emulsion which contained the required wall materials and oil as the secondary emulsion in this study. In this study, microencapsulated refined kenaf seed oil (MRKSO) was subject to dissolution tests in gastric fluid and intestinal fluid to understand the release behavior of refined kenaf seed oil from MRKSO during their passage through the human GI digestion system. The bioaccessibility of the released refined kenaf seed oil from MRKSO was evaluated by compared its antioxidant activity and bioactive compounds after the simulated digestion with undigested and digested un-encapsulated kenaf seed oils.

35

2.2. Refining of kenaf seed oil Crude kenaf seed oil was extracted using Soxhlet extractor at 60  C for 3 h from the kenaf seeds according to the previously reported method (Chew et al., 2017a). The remaining solvent was evaporated off using Buchi Multivapor P-6 (Büchi Labortechnik AG, Switzerland) to recover the crude kenaf seed oil. A chemical refining process, which includes degumming, neutralization, bleaching, and deodorization was carried out to produce refined kenaf seed oil (Chew et al., 2017a). The crude oil was pre-treated with the 0.09% of phosphoric acid (85% concentration) for 10 min, followed by treated with 22.4% of Milli-Q water for 30 min at 40  C with 175 rpm. Then, the degummed oil was neutralized by adding 3.75% w/w of an excess of the stoichiometric ratio of NaOH solution (16 ºBe) at 40  C for 20 min. The neutralized oil was reacted with 1.5% w/w of acid-activated bleaching earth at 70  C for 40 min. Finally, the bleached oil was deodorized with a lab-scale glass deodorizer at 220  C with a reduced pressure of 9e12 mbar for 90 min. 2.3. Preparation of emulsions

2. Materials and methods

Three models of MRKSO were employed in this study, which were SC:b-CD (ratio 2:1), GA:b-CD (ratio 2:1) and GA:SC:b-CD (ratio 4:1:1). The wall material concentration was 20% w/w for SC:b-CD model, and 30% w/w for GA:b-CD and GA:SC:b-CD models of MRKSO. Tween 20 (1% w/w) was added into each formulation. Required amounts of wall materials (gum Arabic, sodium caseinate and Tween 20) were dissolved in distilled water at 50  C using T25 digital Ultra-Turrax homogenizer (IKA, Germany) and stored overnight at shaker (SK-300, Lab Companion, Korea) to ensure full hydration. After that, required amount of refined kenaf seed oil was added drop-wise into the solution under magnetic stirring at 1000 rpm and continued stirred for 10 min after all the oil was incorporated into the aqueous phase to form the primary emulsion. The oil-to-wall ratio was kept constant at 1:4 in each model in order to evaluate the effect of different wall constituents. Then, required amount of b-CD was dissolved into the primary emulsion using T25 digital Ultra-Turrax homogenizer (IKA, Germany) at 9000 rpm for 5 min to form a secondary emulsion. The emulsion was then homogenized using a Labsonic®P ultrasonic homogenizer (Sartorius AG, Germany) at the amplitude of 100% for 5 min to get the final emulsion. The selection of the wall materials and the ratio was based on the initial screening carried out in the laboratory (data not shown).

2.1. Materials

2.4. Microencapsulation by spray drying

Kenaf seeds (variety: V36) were purchased from the National Kenaf and Tobacco Board (Kelantan, Malaysia). b-CD was purchased from Zibo Qianhui Fine Chemical Co., Ltd. (Shandong, China). Gum arabic and sodium caseinate were purchased from VIS Food Tech Ingredient Supplies (Kuala Lumpur, Malaysia). Bile salts, 2,2'-azinobis(3-ethylbenzothia zoline-6-sulfonic acid) diammonium salt (ABTS), 2,2'-diphenyl-1-picrylhydrazyl radical (DPPH
, 95%), 5acholestane, 6-hydroxy-2,5,7,8-tetram-ethylchroman-2-carboxylic acid [Trolox (TE), 97%], N,O-bis (trimethylsilyl) trifluoroacetamide (BSTFA) with 1 mL/100 mL trimethylchlorosilane (TMCS), phytosterols standard (b-sitosterol, campesterol, and stigmasterol), pepsin from porcine gastric mucosa (P7125), and pancreatin from porcine pancreas (P3292) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Tocopherols and tocotrienols standards (a-, b-, g-, and d) were the products of ChromaDex, Inc. (California, USA). Tween 20 and other chemicals used were purchased from Merck Co., Ltd. (Darmstadt, Germany).

The emulsion was spray dried using Mini Spray Dryer B-290 (Büchi Labortechnik AG, Switzerland) with 0.45 m in height and 0.14 m in diameter of the glass dryer chamber. The spray dryer consisted of a two-fluid nozzle composed of an internal tip with an opening of 0.7 mm in diameter and an external ring with an opening of 1.5 mm in diameter. The inlet temperature was 160  C and the emulsion was fed into the main chamber through a peristaltic pump at a pump setting of 20% with a feed rate of 8 ± 2 g/ min. The compressor air pressure was 600 kPa and the atomiser pressure was 450 ± 10 kPa. The resultant powder was collected from both the drying chamber wall and from the cyclone. The powder was stored in freezer at 20  C for further analysis. 2.5. In-vitro release study 2.5.1. Preparation of simulated digestive fluids Simulated gastric fluid (SGF) was prepared by following the

36

S.C. Chew et al. / Journal of Food Engineering 225 (2018) 34e41

method outlined in US Pharmacopeia (United States Pharmacopeia, 2007), whereby dissolving 2.0 g of NaCl in 900 mL of distilled water and followed by the addition of 7.0 mL of HCl (36%). After that, 3.2 g of pepsin derived from porcine mucosa was added into the solution. The pH of the mixture was adjusted to pH 1.5 with either 1 M NaOH or 1 M HCl. Distilled water was added to make up the final volume into 1000 mL. SIF was prepared by following to the method reported by Chung et al. (2011) and Timilsena et al. (2017). To prepare SIF, 6.8 g of KH2PO4, 8.76 g of NaCl, 5 g of bile salts, and 6.4 g of pancreatin were dissolved in 900 mL of distilled water. The pH of the mixture was adjusted to pH 6.8 with either 1 M NaOH or 1 M HCl. Distilled water was added to make up the final volume into 1000 mL. 2.5.2. In-vitro digestion of MRKSO MRKSO (2.0 g) was incubated in 20 mL of SGF in an agitated water bath with 100 rpm at 37 ± 0.5  C for 2 h. After gastric digestion, the digested mixture was immediately adjusted to pH 6.8 to inactivate pepsin. After that, 20 mL of SIF was added into the digested MRKSO and then incubated in an agitated water bath with 100 rpm at 37 ± 0.5  C for 2 h. The oil release behavior of MRKSO in simulated GI digestion was determined by dispersion of MRKSO in SGF only and dispersion of MRKSO in SGF and followed by SIF, respectively. 2.5.3. Determination of oil released from MRKSO Hexane (25 mL) was added into the digested fluid to extract the released oil from MRKSO. The mixture was then centrifuged at 8000 rpm for 5 min to separate the hexane fraction from the digested fluid. The hexane fraction was collected and the extraction process was repeated three times. The obtained hexane fractions were combined and evaporated using a Buchi Multivapor P-6 at 40  C under a reduced pressure of 125 mbar. After that, residual solvent was fully removed by flushing with 99.9% nitrogen. The total oil of MRKSO was determined according to the previously described method (Li et al., 2015) with some modifications. Distilled water (20 mL) and 40 mL of ethanol:hexane (1:1, v/v) was added into a Schott bottle with 2 g of MRKSO. After that, the mixture was homogenized using a Labsonic®P ultrasonic homogenizer (Sartorius AG, Germany) at the amplitude of 100% for 2 min. The resultant mixture was centrifuged at 8000 rpm for 5 min to separate the hexane fraction from the digested fluid. The hexane layer was collected and the extraction process was repeated by adding 20 mL of hexane and followed by the homogenization and centrifugation steps for three times. The obtained hexane fractions were evaporated using a Buchi Multivapor P-6 at 40  C and 125 mbar. The residual solvent was completely removed by purging with 99.9% nitrogen to recover the total oil. The released oil was calculated by Eq. (1) and expressed in percentage (%):

Released oilð%Þ ¼

Released oil ðgÞ  100 Total oil ðgÞ

(1)

2.5.4. Free fatty acid (FFA) release The amount of FFA released from digested microcapsules was used to indicate the lipase activity on the extent of lipolysis. The digestion patterns of the un-encapsulated kenaf seed oil and three models of the MRKSO were evaluated using previously described methods with some modifications (Sarkar et al., 2010; Timilsena et al., 2017). Un-encapsulated refined kenaf seed oil (1 g) and 4 g of MRKSO (equivalent to 1 g oil) were used in the digestion models. The amount of FFA released was tested on the digestion mixtures obtained at the end of gastric digestion and intestinal digestion at

30, 60, 90, and 120 min. NaOH (0.1 N) was used to titrate the digestion mixture with phenolphthalein as an indicator. The degree of lipolysis was expressed as milligram of FFA (as oleic acid) per gram of oil by using Eq. (2). Blank was prepared by replacing the sample with Milli-Q water and the amount of FFA released was used as the control for the calculation of the degree of lipid hydrolysis in each stage of digestion.

FFA ðmg=g oilÞ ¼

282:47  V  N W

(2)

Where V is the volume of NaOH titrated (mL), N is the strength of NaOH (N), and W is the mass of oil used in the digestion model (g). 2.6. Assessment of bioaccessibility of released oil after in-vitro digestion 2.6.1. Radical scavenging activity assays 2,2-Diphenyl-1-picrylhydrazyl (DPPH
) radical scavenging activity assay was carried out using previously described methods by Chew et al. (2016). 2,20 -Azino-bis(3-ethylbenzothiazoline-6-sulphonic acid) (ABTS
þ) radical scavenging activity assay was carried out according to Chew et al. (2016) with slight modification. Oil sample (20 mL) was added to 1980 mL of ABTS radical solution and allowed to react for 3 min. The absorbance was measured at 734 nm against ethanol (blank). Radical scavenging activity of DPPH and ABTS was calculated using Eq. (2), where As and Ac are the absorbance of the sample and control, respectively. Measurements were calibrated to a standard curve of Trolox at concentrations of 0.02e0.1 mg/mL with calibration equation of y ¼ 146.3 x e 0.154 (R2 ¼ 0.996) for DPPH and y ¼ 512.9 x e 0.290 (R2 ¼ 0.997) for ABTS. The values were expressed as mg Trolox equivalents (mg Teq/100 g oil).

  As  100% Radical scavenging activity ð%Þ ¼ 1  Ac

(3)

2.6.2. Total phenolic content (TPC) TPC analysis was carried out according to the previously described method (Chew et al., 2016). The TPC of the samples was expressed in mg gallic acid equivalents (GAE)/100 g oil via a calibration curve of gallic acid at concentrations of 0.02e0.1 mg/mL with calibration equation of y ¼ 2.262 x þ 0.001 (R2 ¼ 0.997). 2.6.3. Determination of tocopherols and tocotrienols contents The tocopherols and tocotrienols contents were analyzed using a HPLC (Agilent Technologies 1200 Series, USA), equipped with a UVevis detector and a reversed phase column (Cosmosil 5PFP packed column, 5 mm  250 mm x 4.6 mm) (Nacalai Tesque, Inc., Japan). Isopropanol (1 mL) was added into 0.5 g of oil sample and vortexed for 2 min. The mixture was filtered with syringe filter (0.45 mm). Mobile phase (Methanol/water, 90:10, v/v) was used to separate the tocopherols and tocotrienols by isocratic elution with a flow rate of 1 mL/min for 30 min. The peaks were detected at 295 nm. The concentration of tocopherols and tocotrienols was expressed in mg/100 g oil through the calibration curve of a series of isopropanol solutions of a-tocotrienol, g-tocotrienol, d-tocotrienol, a-tocopherol, b-tocopherol, g-tocopherol, d-tocopherol at a concentration of 0.2e1 g/L (Chew et al., 2017a). 2.6.4. Determination of phytosterols content The phytosterols content in the oil samples were determined according to the previously published method (Chew et al., 2017b).

S.C. Chew et al. / Journal of Food Engineering 225 (2018) 34e41

Gas chromatography (GC) (Agilent Technologies 7890A, USA) equipped with a flame ionization detector (FID) and a HP-5 column (30 m  0.32 mm i.d., 0.25 mm film thickness) was used to analyze the phytosterols content. The peaks were quantified by comparing the retention time to the standards of campesterol, stigmasterol, and b-sitosterol. 2.7. Statistical analysis Data were collected in duplicate samples and measurements were replicated two times. The results were reported as mean ± standard deviation and analyzed using statistical software of MINITAB 16 (Minitab Inc, Pennsylvania, USA). One-way analysis of variance (ANOVA) followed by Tukey's post hoc test was conducted to determine the significant difference among the samples based on a 95% confidence level (p < 0.05). 3. Results and discussion 3.1. In-vitro digestibility of MRKSO In order to examine the potential of MRKSO for oral delivery via a food or as a supplement for targeted delivery of unsaturated fatty acids and lipophilic bioactive to specific site of the GI, it is important to evaluate the release behavior of MRKSO during GI transit. Due to the smaller particle size of MRKSO, the MRKSO would not bite in the mouth phase and straight swallowed into the gastric phase. Thus, the MRKSO is first digested in the stomach at low pH ~1.5, and followed by digestion in the intestine at pH ~6.8 during GI digestion. The digestion duration for each individual may vary quite considerably as it depends on each individual characteristics and metabolism rate, as well as the food properties. A digestion time of 2 h was usually used in the gastric and intestinal stage of the digestion model that showed by the previous study (Hur et al., 2011). Therefore, the release behavior of MRKSO for sequential exposure of SGF for 2 h and SIF for 2 h was evaluated based on the percentage of oil released from different models of MRKSO to simulate the GI digestion conditions. Table 1 shows that the percentage of oil released from MRKSO was the lowest for the model of SC:b-CD (1.43%), followed by GA:b-CD (3.43%) and GA:SC:b-CD (6.44%) after exposure to the gastric phase. The oil released from the model of SC:b-CD MRKSO was significantly lower (p < 0.05) than other models of MRKSO. This might be due to the heatinduced protein-protein interaction and aggregation of proteins at low pH condition in the gastric environment, as explained in a previous study (Kosaraju et al., 2009). The MRKSO model of SC:bCD contains higher proportion of SC (13.33% w/w) due to the ratio of 2:1 for SC:b-CD. Thus, aggregation of the microcapsules (SC:bCD) was observed during the simulated gastric digestion, so this prevents the access of proteolytic enzymes into the MRKSO (Kosaraju et al., 2009). In addition, the low pH condition in the gastric environment will cause the entrapment of oil inside the randomly aggregated protein mass (Binsi et al., 2016). However, for the MRKSO model of GA:b-CD, there was only a little protein

37

content presented in the shell wall matrix that can be digested by the proteolytic enzymes in the gastric environment as GA contains two different polymers (glycoprotein and polysaccharide). On the other hand, protein amount in the MRKSO model of GA:SC:b-CD was lower (5% w/w), thus GA minimised the extent of protein aggregation, leading to higher exposure of the shell wall matrix to the proteolytic enzymes (Binsi et al., 2016). This resulted in the higher enzymatic degradation of the shell wall matrix of GA:SC:b-CD model compared to other models of MRKSO in the gastric phase. However, the oil released after exposure to the gastric phase digestion from the three models of MRKSO in this study was considered low compared to the previous studies. A very high amount of oil released was found in the flaxseed oil powder with chickpea, lentil protein isolate and maltodextrin (36.6e43.4%) (Karaca et al., 2013), 54.09e64.62% in spray dried fish roe powder with GA (Binsi et al., 2016) and 37.9e79.1% in chia seed oil complex coacervates with chia seed protein isolate and chia seed gum (Timilsena et al., 2017) after exposure to the gastric phase digestion. This large difference in oil released may be due to the difference in the technique and wall materials used in the microencapsulation process. This study showed that the three shell wall matrices of bCD combined with the SC and/or GA by spray drying were resistant to enzymatic and acidic degradation in the adverse gastric environment. After an in-vitro digestion in SGF and SIF, the oil released from MRKSO model of GA:SC:b-CD was significantly higher (p < 0.05) than the model of SC:b-CD. There was no significant difference (p > 0.05) in the oil released between the models of GA:SC:b-CD and GA:b-CD. Pancreatin presented in the SIF contains the enzymes of amylase, trypsin, and lipase, which can digest starch, proteins, and fats. Breaking down of both protein and polysaccharide in the shell wall matrices was hydrolyzed by the pancreatin presented in the SIF. This would have facilitated the emission of oil from the degraded polymer matrix, which resulted in increasing the amount of released oil in SIF. The lower amount of oil released from the MRKSO model of SC:b-CD might be due to the heat treatment of SC during the spray drying resulted in the lower digestibility by trypsin and chemotrypsin as a result of protein aggregation and conformational changes in the protein structure during SGF digestion. On the other hand, the protein and polysaccharide composition in the models of GA:SC:b-CD and GA:b-CD could be well digested by the enzymes as their lesser extent of protein aggregation in the SGF. Ahmad et al. (2017) showed that microencapsulation of folic acid by b-CD enables slow and target release of folic acid in the intestine. b-CD is able to protect the bioactive compound in the acidic condition and offer target release property in the intestine as b-CD is a slow digestible starch also. GA makes the MRKSO specific due to GA is not digested in the intestine by the pancreatic enzymes but it can ferment in the colon to release shortchain fatty acids, that offering various potential health benefits. However, there was high percentage of oil released from the MRKSO models of GA:SC:b-CD and GA:b-CD due to the alkaline pH of intestinal fluid making it to exclude the entrapped oil from the cleavage site of the partially digested matrix (Binsi et al., 2016). 3.2. Extent of lipolysis

Table 1 Percentage of oil released from the models of MRKSO after in-vitro digestion in simulated gastric and intestine phases. Sample

Gastric phase

Gastric phase þ intestinal phase

SC:b-CD GA:b-CD GA:SC:b-CD

1.43 ± 0.26c 3.43 ± 0.68b 6.44 ± 0.69a

81.10 ± 4.35b 87.40 ± 2.18ab 91.19 ± 0.74a

a,b,c Values with different superscript letters in the same column differ significantly (p < 0.05).

The in-vitro digestion model was consisted simulated gastric and intestinal phases, but the actual lipid digestion was happened in the intestinal phase due to the presence of pancreatic lipase enzyme. Thus, the extent of lipolysis of the samples was measured in an interval of 30 min in the intestinal digestion in this study. Fig. 1 shows that the FFA released in the samples of unencapsulated oil and MRKSO in the range of 17.04e376.72 mg/g oil at the end of gastric digestion. There was no gastric lipase

38

S.C. Chew et al. / Journal of Food Engineering 225 (2018) 34e41

Fig. 1. Amount of FFA released (mg FFA/g oil) from the un-encapsulated oil and three different models of MRKSO during in-vitro digestion.

presented in the SGF, thus there was no enzymatic hydrolysis of lipids. The increase of FFA might be due to the hydrolytic breakdown of triglycerides by the heat (37  C), acid, and moisture existed in the gastric phase (Timilsena et al., 2017). The FFA content of all the samples was increased considerably when the samples entered the intestinal digestion phase. The FFA released was in the order of SC:b-CD > GA:SC:b-CD > GA:b-CD > un-encapsulated oil. The result showed that spray drying technique helped to improve the digestibility of kenaf seed oil. This is due to the larger droplet size of the oil (Cheong et al., 2016) and microencapsulation helped to reduce the oil droplet size by produced in micro-powder form by spray drying. Thus, the total surface area of MRKSO available for the action of pancreatic lipase was increased when the oil droplets are smaller resulted in an increased rate of lipid digestion by lipase. Different models of MRKSO showed different digestion patterns as the different components of wall materials. The wall materials might affect the diffusion and adsorption of lipase on the oil droplets (Timilsena et al., 2017). Although the digested SC:b-CD model of MRKSO showed 81.10% of oil released after simulated digestion, which was lower than the oil released of digested GA:bCD and GA:SC:b-CD models of MRKSO, the FFA released in the digested SC:b-CD MRKSO model was the highest. This might be due to the presence of undigestible fiber of GA in the MRKSO models of GA:b-CD and GA:SC:b-CD, that interfered with the lipase activity that decreased the rate of lipolysis. The previous study reported that presence of protein, fibers, and polysaccharides on the complex coacervates microcapsules was reduced the digestibility of the oil compared to the unencapsulated oil (Timilsena et al., 2017). Thus, this study showed that the spray dried microcapsules have improved the lipolysis of the kenaf seed oil. 3.3. Assessment of antioxidant activity after in-vitro digestion Un-encapsulated refined kenaf seed oil was also digested in order to assess the effect of the digestion in the oil composition. Thus, refined kenaf seed oil, digested refined kenaf seed oil and three different models of MRKSO were examined in this study.

Table 2 shows the results of radical scavenging activities of DPPH and ABTS, as well as the TPC of the studied samples. The DPPH radical scavenging activity was significantly increased (p < 0.05) for the digested refined oil and GA:SC:b-CD model of MRKSO after simulated GI digestion. Pazinatto et al. (2013) reported digestive enzymes were able to change the chemical structure of soluble and insoluble components leading to the increase of its availability on the matrix's surface, thus giving their reducing properties to DPPH radical. Some compounds with increased antioxidant activity would be formed during the GI digestion (You et al., 2010). There was no significant difference (p > 0.05) in the DPPH radical scavenging activity between the refined oil and GA:b-CD model of MRKSO, while there was a significantly decreased (p < 0.05) in the SC:b-CD model of MRKSO. The decrease of DPPH radical scavenging activity in digested SC:b-CD model of MRKSO was due to the polyphenols-protein interaction (Ozdal et al., 2013). The ABTS radical scavenging activity was slightly increased for the digested refined oil and digested MRKSO models of GA:b-CD and GA:SC:bCD, but there was no significant difference when compared to the refined oil. The ABTS radical scavenging activity of the MRKSO model of SC:b-CD was significantly decreased (p < 0.05) after simulated digestion. You et al. (2010) suggested that the increase of hydrophilic characteristic of the GI digests after GI digestion favors the increase of the ABTS values. In addition, the increase of radical scavenging activity in the MRKSO samples might be due to the antioxidant activity attributed by GA and b-CD due to the presence of their hydroxyl groups, which able to donate a hydrogen atom to scavenge the free radical (Ashwar et al., 2016). The TPC of digested refined oil and digested MRKSO models of GA:b-CD and GA:SC:b-CD was significantly increased (p < 0.05) after the simulated digestion, while there was no significant difference (p > 0.05) in the TPC between the refined oil and digested SC:b-CD model of MRKSO. Overall, the MRKSO models of GA:b-CD and GA:SC:b-CD showed an increase of bioaccessibility in terms of TPC after in-vitro digestion. Cheong et al. (2016) reported that CDs could form complex with phenolic compounds to increase the total phenolic contents in aqueous solution. Hydroxypropyl-b-CD can

S.C. Chew et al. / Journal of Food Engineering 225 (2018) 34e41

39

Table 2 Changes in antioxidant activities and phenolic content of undigested oil, un-encapsulated digested oil and oil released after in-vitro digestion of three different models of MRKSO. Sample

DPPH (mg Trolox equiv/100 g oil)

ABTS (mg Trolox equiv/100 g oil)

TPC (mg/100 g oil)

Undigested oil Un-encapsulated oil SC:b-CD GA:b-CD GA:SC:b-CD

29.22 ± 1.49b 36.58 ± 1.50a 16.70 ± 2.11c 26.22 ± 2.04b 35.05 ± 1.17a

42.90 ± 0.85a 43.86 ± 0.69a 29.45 ± 3.58b 45.45 ± 1.64a 45.06 ± 3.83a

3.85 ± 0.15b 8.81 ± 1.28a 4.40 ± 0.42b 10.99 ± 1.46a 8.87 ± 1.31a

a,b,c

Values with different superscript letters in the same column differ significantly (p < 0.05).

improve its solubility in aqueous solution and enhance its antioxidant activity by complexation with polyphenols. Therefore, complexation of polyphenols with b-CD in the MRKSO model would increase its solubility in aqueous solution as well as its TPC values after digestion. However, complexation of polyphenols with proteins would change the structural, functional and nutritional properties of both compounds. The polyphenols-proteins interaction might reduce the digestibility of some amino acids and protein also, and resulted in the decrease in the bioaccessibility and total antioxidant capacity after simulated digestion (Ozdal et al., 2013). 3.4. Tocopherols and tocotrienols contents Table 3 shows that the total tocopherols and tocotrienols contents were increased (increase 40.7e150.5%) after the simulated GI digestion, which showed in the digested refined oil and digested MRKSO samples. This result was in accordance with the previous studies that reported on the digested microcapsules of kenaf seed oil (Chew et al., 2015) and nanoemulsion of kenaf seed oil (Cheong et al., 2016). This indicates a good releasing and solubility of tocopherols and tocotrienols from the released oil into bile salt mixed micelles that helps in the transport and adsorption across the intestinal wall, thus increases the bioavailability of the tocopherols and tocotrienols after simulated digestion (Cheong et al., 2016). The chemical form of the tocopherols and the nature of the food matrix that it comprised in would affect the bioavailability of tocopherols in 1e100% (O’Callaghan and O’Brien, 2010; Reboul et al., 2006). Tocopherols and tocotrienols are highly lipophilic molecule with very low water solubility. Thus, microencapsulation can help to incorporate the tocopherols and tocotrienols in microcapsules form to improve its solubility and thus improve its bioavailability or adsorption during GI digestion. In addition, long chain fatty acids in the refined kenaf seed oil such as C16:0 and C18:0 are accounted for the improved solubilization capacity for tocopherols and tocotrienols. This is because long-chain fatty acids can form mixed micelles that have a larger solubilization capacity for non-polar lipophilic molecules compared to the medium chain fatty acids (Yang and McClements, 2013). All of the tocopherols and tocotrienols isomers were increased after the simulated digestion, except for the a-tocopherol (decrease 27.7e68.3%). This might be due to the high amount of oxygen

presented in the GI environment caused the a-tocopherol oxidized as the a-tocopherol was found to be the most sensitive against the oxidative deterioration (Chew et al., 2017b). Previous studies showed that the degree of degradation of tocopherols was shown in the following order: a->(gþb)>d-tocopherol in the storage of rice bran oil, which indicated that the a-tocopherol was the least stable (Bruscatto et al., 2009). b-Tocopherol was not detected in the digested MRKSO samples due to its closer retention time with gtocopherol. b-Tocopherol was combined together with the gtocopherol due to their increased amount after the simulated digestion. The total tocopherols and tocotrienols contents in both of the MRKSO models of GA:b-CD and GA:SC:b-CD were significantly higher (p < 0.05) than others samples. This indicated the shell wall had protected the tocopherols and tocotrienols contents in the refined kenaf seed oil against GI environment and increased its bioavailability during the GI digestion. Previous study showed that the protein-based (maltodextrin:SC) microencapsulated walnut oil showed a higher decrease of tocopherol contents after simulated digestion compared to the carbohydrate-based microcapsules. Besides that, carbohydrate-based microcapsules offered the major stability to the concentration of a-tocopherol isomer (Calvo et al., 2012). This might be due to the synergism of tocopherols and polyphenols due to the increased phenolic contents in the MRKSO models of GA:b-CD and GA:SC:b-CD. However, polyphenols-protein interaction reduces the increment of tocopherols and tocotrienols contents in the SC:b-CD model of MRKSO. Increase of phenolic content in the lipid system has been shown to be effective in the retention of the tocopherols and tocotrienols contents and release of more free form of tocopherols and tocotrienols from dimeric or other esterified compounds (Shahidi and de Camargo, 2016). 3.5. Phytosterol contents In this study, b-Sitosterol was the most abundant phytosterol in kenaf seed oil, followed by campesterol and stigmasterol, which agreed with the previous study (Chew et al., 2016). Table 4 shows the phytosterol content was significantly decreased (p < 0.05) after the simulated digestion, which showed in the digested oil and MRKSO. The degradation of phytosterol content after simulated digestion was in accordance with the previous studies (Chew et al.,

Table 3 Changes in tocopherols and tocotrienols contents of undigested oil, un-encapsulated digested oil and oil released after in-vitro digestion of three different models of MRKSO. Content (mg/100 g)

Undigested oil

Un-encapsulated oil

SC:b-CD

GA:b-CD

GA:SC:b-CD

d-Tocotrienol g-Tocotrienol a-Tocotrienol d-Tocopherol (bþg)-Tocopherol a-Tocopherol

4.19 ± 0.17d 1.84 ± 0.09c 0.91 ± 0.07c 1.89 ± 0.05d 24.90 ± 0.26c 13.53 ± 0.38a 48.03 ± 0.79d

5.09 ± 0.04c 2.06 ± 0.06c 1.02 ± 0.06c 10.34 ± 0.96c 39.30 ± 1.54b 9.78 ± 0.17b 67.60 ± 2.63c

5.47 ± 0.34bc 4.10 ± 0.37b 3.44 ± 0.18b 25.12 ± 3.67b 37.91 ± 3.30b 4.29 ± 0.41e 80.34 ± 7.26b

5.87 ± 0.28ab 6.16 ± 0.13a 5.08 ± 0.25a 38.70 ± 1.29a 57.84 ± 0.79a 6.69 ± 0.09c 120.33 ± 1.78a

5.98 ± 0.19a 6.07 ± 0.16a 5.29 ± 0.23a 38.32 ± 0.83a 57.66 ± 2.98a 5.67 ± 0.21d 118.99 ± 4.55a

Sum a,b,c,d

Values with different superscript letters in the same column differ significantly (p < 0.05).

40

S.C. Chew et al. / Journal of Food Engineering 225 (2018) 34e41

Table 4 Changes in phytosterols content of undigested oil, un-encapsulated digested oil and oil released after in-vitro digestion of three different models of MRKSO. Content (mg/100 g)

Undigested oil

Un-encapsulated oil

SC:b-CD

GA:b-CD

GA:SC:b-CD

Campesterol Stigmasterol b-Sitosterol Sum

46.16 ± 1.29a 20.00 ± 2.27a 374.33 ± 21.08a 440.49 ± 21.11a

38.73 ± 4.30b 12.61 ± 1.10b 271.32 ± 32.58b 322.66 ± 35.68b

6.46 ± 0.34cd 2.83 ± 0.21d 78.08 ± 0.91d 87.37 ± 0.83d

5.49 ± 0.41d 4.24 ± 0.36d 54.25 ± 4.49d 63.98 ± 4.17d

10.85 ± 0.68c 7.28 ± 0.37c 160.81 ± 12.65c 178.94 ± 11.97c

a,b,c,d

Values with different superscript letters in the same column differ significantly (p < 0.05).

2015; Cheong et al., 2016). This might be due to the hydrophobicity characteristic of the phytosterol, which is less soluble in the simulated enzyme fluid and resulted in the decrease of the bioavailability of phytosterol. However, the phytosterol content in the digested refined kenaf seed oil was significantly higher (p < 0.05) than the three models of MRKSO. According to Cheong et al. (2016), solvent extraction of the digested bulk oil from the SGF and SIF might contain some undigested oils. This is because of the large droplet size of the bulk oil, which might decrease the solubility and digestibility of the oil. Thus, some of the phytosterols were not liberated during the digestion as it was still remained in the large oil droplets. The MRKSO model of GA:SC:b-CD was showed to be the most protective shell wall matrix against the degradation of phytosterol content. There is no single wall material possesses all the characteristics required for an ideal encapsulating wall material, so combination of wall materials might improve the strengthening ability of encapsulated powders. 4. Conclusions The present study has shown that microencapsulation by spray drying by using the combination of wall materials of b-CD, GA and/ or SC offered controlled release property by protecting the refined kenaf seed oil in the SGF. There was at least 80% of refined kenaf seed oil released from digested MRKSO after simulated digestion, especially the highest oil released (91.19%) was shown by the GA:SC:b-CD model of MRKSO. The degree of lipolysis was in the order of SC:b-CD > GA:SC:b-CD > GA:b-CD > un-encapsulated oil. The GA:SC:b-CD model of MRKSO offered better protection to the oil during simulated digestion by showing an increase in the radical scavenging activity of DPPH (20.0% increase) and ABTS (5.0% increase), total phenolic content (130.4% increase), tocopherol and tocotrienol contents (147.7% increase). There was a slower degradation of phytosterol content in the GA:SC:b-CD model of MRKSO (59.4% decrease) compared to the GA:b-CD (85.5% decrease) and SC:b-CD (80.2% decrease) models of MRKSO. This study provides useful insights for the future development application of MRKSO in the food industry. Acknowledgement Financial support of this work by internal funding from CERVIE UCSI University (Proj-In-FAS-053) is gratefully acknowledged. References Ahmad, M., Qureshi, S., Maqsood, S., Gani, A., Masoodi, F.A., 2017. Micro-encapsulation of folic acid using horse chestnut starch and b-cyclodextrin: microcapsule characterization, release behaviour & antioxidant potential during GI tract conditions. Food Hydrocoll. 66, 154e160. Ashwar, B.A., Gani, A., Wani, I.A., Shah, A., Masoodi, F.A., Saxena, D.C., 2016. Production of resistant starch from rice by dual autoclaving retrogradation treatment: invitro digestibility, thermal and structural characterization. Food Hydrocoll. 56, 108e117. Babiker, R., Merghani, T.H., Elmusharaf, K., Badi, R.M., Lang, F., Saeed, A.M., 2012. Effects of gum Arabic ingestion on body mass index and body fat percentage in healthy adult females: two-arm randomized, placebo controlled, double-blind trial. Nutr. J. 11, 1e7.

Binsi, P.K., Natasha, N., Sarkar, P.C., Ashraf, P.M., George, N., Ravishankar, C.N., 2016. Structural, functional and in vitro digestion characteristics of spray dried fish roe powder stabilised with gum arabic. Food Chem. 221, 1968e1708. Bruscatto, M.H., Zambiazi, R.C., Sganzerla, M., Pestana, V.R., Otero, D., Lima, R., Paiva, F., 2009. Degradation of tocopherols in rice bran oil submitted to heating at different temperatures. J. Chromatogr. Sci. 47, 762e765.  mez, D., 2012. In-vitro Calvo, P., Lozano, M., Espinosa-Mansilla, A., Gonz alez-Go evaluation of the availability of 6-3 and 6-6 fatty acids and tocopherols from microencapsulated walnut oil. Food Res. Int. 48, 316e321. Cheong, A.M., Nyam, K.L., 2016. Improvement of physical stability of kenaf seed oilin-water nanoemulsions by addition of b-cyclodextrin to primary emulsion containing sodium caseinate and Tween 20. J. Food Eng. 183, 24e31. Cheong, A.M., Tan, C.P., Nyam, K.L., 2016. In vitro evaluation of the structural and bioaccessibility of kenaf seed oil nanoemulsions stabilised by binary emulsifiers and b-cyclodextrin complexes. J. Food Eng. 189, 90e98. Chew, S.C., Nyam, K.L., 2016. Microencapsulation of kenaf seed oil by co-extrusion technology. J. Food Eng. 175, 43e50. Chew, S.C., Tan, C.P., Long, K., Nyam, K.L., 2015. In-vitro evaluation of kenaf seed oil in chitosan coated-high methoxyl pectin-alginate microcapsules. Ind. Crops. Prod. 76, 230e236. Chew, S.C., Tan, C.P., Long, K., Nyam, K.L., 2016. Effect of chemical refining on the quality of kenaf (Hibiscus cannabinus) seed oil. Ind. Crops. Prod. 89, 59e65. Chew, S.C., Tan, C.P., Nyam, K.L., 2017a. Application of response surface methodology for optimizing the deodorization parameters in chemical refining of kenaf seed oil. Sep. Purif. Technol. 184, 144e151. Chew, S.C., Tan, C.P., Nyam, K.L., 2017b. Comparative study of crude and refined kenaf (Hibiscus cannabinus L.) seed oil during accelerated storage. Food Sci. Biotechnol. 26, 63e69. Chung, C., Sanguansri, L., Augustin, M.A., 2011. In vitro lipolysis of fish oil microcapsules containing protein and resistant starch. Food Chem. 124, 1480e1489. Daza, L.D., Fujita, A., F avaro-Trindade, C.S., Rodrigues-Ract, J.N., Granato, D., Genovese, M.I., 2016. Effect of spray drying conditions on the physical properties of Cagaita (Eugenia dysenterica DC.) fruit extracts. Food. Bioprod. Process 97, 20e29. Edris, A.E., Kalemba, D., Adamiec, J., Pia˛ tkowski, M., 2016. Microencapsulation of Nigella sativa oleoresin by spray drying for food and nutraceutical applications. Food Chem. 204, 326e333.  Vikmon, M., Szente, L., 2016. Cyclodextrins in food technology and Fenyvesi, E., human nutrition: benefits and limitations. Crit. Rev. Food Sci. Nutr. 56, 1981e2004. Fernandes, R.V.B., Borges, S.V., Botrel, D.A., 2013. Influence of spray drying operating conditions on microencapsulated rosemary essential oil properties. Food Sci. Technol. 33, 171e178. Goyal, A., Sharma, V., Sihag, M.K., Tomar, S.K., Arora, S., Sabikhi, L., Singh, A.K., 2015. Development and physic-chemical characterization of microencapsulated flaxseed oil powder: a functional ingredient for omega-3 fortification. Powder Technol. 286, 527e537. Hundre, S.Y., Karthik, P., Anandharamakrishnan, C., 2015. Effect of whey protein isolate and b-cyclodextrin wall systems on stability of microencapsulated vanillin by spray-freeze drying method. Food Chem. 174, 16e24. Hur, S.J., Lim, B.O., Decker, E.A., McClements, D.J., 2011. In vitro human digestion models for food applications. Food Chem. 125, 1e12. Karaca, A.C., Nickerson, M., Low, N.H., 2013. Microcapsule production employing chickpea or lentil protein isolates and maltodextrin: physicochemical properties and oxidative protection of encapsulated flaxseed oil. Food Chem. 139, 448e457. Kosaraju, S.L., Weerakkody, R., Augustin, M.A., 2009. In-vitro evaluation of hydrocolloid-based encapsulated fish oil. Food Hydrocoll. 23, 1413e1419. Li, J., Xiong, S., Wang, F., Regenstein, J.M., Liu, R., 2015. Optimization of microencapsulation of fish oil with gum arabic/casein/beta-cyclodextrin mixtures by spray drying. Food Chem. 80, C1445eC1452. Ng, S.K., Lau, J.L.Y., Tan, C.P., Long, K., Nyam, K.L., 2013. Effect of accelerated storage on microencapsulated kenaf seed oil. J. Am. Oil Chem. Soc. 90, 1023e1029. O’Callaghan, Y., O’Brien, N., 2010. Bioaccessibility, cellular uptake and transepithelial transport of alpha-tocopherol and retinol from a range of supplemented foodstuffs assessed using the caco-2 cell model. Int. J. Food Sci. Technol. 45, 1436e1442. Ozdal, T., Capanoglu, E., Altay, F., 2013. A review on protein-phenolic interactions and associated changes. Food Res. Int. 51, 954e970. Pazinatto, C., Malta, L.G., Pastore, G.M., Netto, F.M., 2013. Antioxidant capacity of amaranth products: effects of thermal and enzymatic treatments. Food Sci. Technol. 33, 485e493.

S.C. Chew et al. / Journal of Food Engineering 225 (2018) 34e41 Reboul, R., Richelle, M., Perrot, E., Desmoulins-Malezet, C., Pirisi, V., Borel, P., 2006. Bioaccessibility of carotenoids and vitamin E from their main dietary sources. J. Agric. Food Chem. 54, 8749e8755. Sarkar, A., Horne, D.S., Singh, H., 2010. Pancreatin-induced coalescence of oil-in water emulsions in an in vitro duodenal model. Int. Dairy J. 20, 589e59. Shahidi, F., de Camargo, A.C., 2016. Tocopherols and tocotrienols in common and emerging dietary sources: occurrence, applications, and health benefits. Int. J. Mol. Sci. 17, E1745. Sun-Waterhouse, D., Waterhouse, G.I.N., 2015. Spray-drying of green or gold kiwifruit juice-milk mixtures; novel formulations and processes to retain natural fruit colour and antioxidants. Food Bioprocess Technol. 8, 191e207. Timilsena, Y.P., Adhikari, R., Barrow, C.J., Adhikari, B., 2017. Digestion behaviour of chia seed oil encapsulated in chia seed protein-gum complex coacervates. Food

41

Hydrocoll. 66, 71e81. United States Pharmacopeia (USP. 30), 2007. United States Pharmacopeial Convention. Wang, T., Hu, Q., Zhou, M., Xia, Y., Nieh, M.-P., Luo, Y., 2016. Development of “all natural” layer-by-layer redispersible solid lipid nanoparticles by nano spray drying technology. Eur. J. Pharm. Biopharm. 107, 273e285. Yang, Y., McClements, D.J., 2013. Vitamin E bioaccessibility: influence of carrier oil type on digestion and release of emulsified a-tocopherol acetate. Food Chem. 141, 473e481. You, L.J., Zhao, M.M., Regenstein, J.M., Ren, J.Y., 2010. Changes in the antioxidant activity of loach (Misgurnus anguillicaudatus) protein hydrolysates during a simulated gastrointestinal digestion. Food Chem. 120, 810e816.