Stability of Purple Corn Anthocyanin Encapsulated by Maltodextrin, and Its Combinations with Gum Arabic and Whey Protein Isolate

Purple corn anthocyanins are important natural colourants with cheap prices and rich bioactivities. However, their stability is limited. Microencapsulation is an effective way to improve anthocyanin stability and the influence of the type of wall material on the stability of encapsulated anthocyanin is very important. In this study, maltodextrin (MD) and its combination with whey protein isolate (WPI) or gum arabic (GA) were utilised as wall materials to obtain encapsulated purple corn anthocyanins (PCAs) (MD–PCA, MD–WPI–PCA, MD–GA–PCA) using spray drying. The effect of the amount of the wall material was determined by encapsulation efficiency, anthocyanin content, and colour. On this basis, the effects of the types of wall materials on the physicochemical characteristics, storage, and digestion stabilities of encapsulated PCA, as well as their stabilities in chewing tablets, were investigated. The highest encapsulation efficiency, suitable colour, and anthocyanin content were obtained with the mass ratios 1:1 PCA to MD, 2:3 PCA to MD–GA, and 1:1 PCA to MD–WPI. Microencapsulation increased PCA storage and digestion stabilities. All three types of PCA microcapsules had low water content and hygroscopicity and good water solubility. MD–PCA had the strongest stability when stored at 25 °C; MD–GA–PCA—when stored at 40 °C, or in the presence of 5000 Lux light illumination; MD–WPI–PCA—when stored in 75% relative humidity or during gastric–intestinal digestion, but its resistance to 40 °C temperature and light illumination was lower than those for the two others. When used in chewing tablets, MD encapsulation was most stable in the presence of Ca2+, VC, or Fe2+ and improved PCA digestion stability. In conclusion, MD is a good choice for PCA encapsulation in regular conditions. MD–GA and MD–WPI can be used when considering high storage temperature (or light illumination) and high humidity (or for high digestion stability), respectively. The results of this study provide a reference for the storage and application of PCA.


Introduction
Anthocyanins are pigments found in purple, black, and red vegetables, fruits, and grains [1]. Anthocyanins exhibit antioxidant, anticancer, and antidiabetic effects [2]. However, anthocyanins are extremely sensitive to pH, temperature, light, metal ions, and enzymes, which leads to their degradation and low utilisation during storage, processing, and digestion [3]. Encapsulation technologies, such as freeze drying, spray drying, liposome entrapment, and nanoencapsulation, have emerged as effective approaches that entrap anthocyanins inside the wall materials to slow their degradation under various environmental factors during processing and storage and control their release during di-

Encapsulation Efficiency
The method described by Robert et al. [16] was used, with modifications. Microcapsules (0.2 g) were dissolved in 10 mL of citric acid-sodium citrate buffer solution (0.1 M, pH 3.0) for determining total PCA content in microcapsules, which included anthocyanin content on microcapsule surface and inside the microcapsule, while 10 mL of absolute ethanol was used to wash the surface of microcapsules to determine PCA on the surface of microcapsules.

Colour
Following the method of Bhagya Raj and Dash [17], the colour of the microcapsules was measured using a colourimeter (NH310, Shenzhen Sanenshi Technology Company, Shenzhen, China). The L*, a*, and b* parameters were recorded. L* (brightness) ranged from 0 (black) to 100 (white); +a* (−a*) and +b* (−b*) represent the red (green) and yellow (blue) value, respectively. The total difference (∆E) was calculated as follows: where L 0 * and L t * represent the colour parameters of the microcapsules at the beginning and after t days of storage, respectively, and a 0 * and a t *, b 0 *, and b t * are expressed in the same manner. According to the method of AOAC [18], microcapsules (1 g) were weighed and dried in an oven (DHG-9070A, Shanghai Jinghong Experimental Equipment, Shanghai, China) at 105 • C (air velocity, 1400 rpm) to constant weight. The moisture content was expressed as the mass percentage of the evaporated water in the microcapsules before drying.

Solubility
Following the description of Shittu and Lawal [19] with slight modifications, microcapsules (0.1 g) were put into distilled water (10 mL, 30 • C) and magnetically agitated (100 rpm) for 30 min, then centrifuged (8000× g, 10 min). The resulting supernatant was dried at 105 • C to constant weight. The solubility represents the mass percentage of the dried solid in the microcapsules before drying.

Hygroscopicity
Following the description of Kurek et al. [20], with small modifications, microcapsules (1 g) were stored in a container with saturated sodium chloride solution at 25 • C for one week. Hygroscopicity was the mass of absorbed water per 100 g of microcapsules.

Fluidity
Fluidity was expressed by the angle of repose (θ) measured as described by Huang et al. [21], with small modifications. Microcapsules (10 g) were poured on the horizontal plane through a funnel to make them naturally stacked, then the height (h) of powder pile and the radius (r) of the co-powder pile were measured, and θ was calculated according to the following formula:

Morphology of PCA Microcapsules
After gold spraying, the morphology of the microcapsules was observed using a scanning electron microscope (SEM) HITACHI Regulus8100 (Tokyo, Japan). Gold layer was 10 nm. Multiple magnifications were 1000 and 2000. Accelerating voltage was 3.0 kV. Objective aperture was 50 µm. Working distance was 9.1 mm.

Storage of PCA Microcapsules
PCA microcapsules and freeze-dried unencapsulated PCA (freeze-dried PCA extract solution described in Section 2.1 was used as a control) were placed in conical bottles with stoppers and stored at different temperatures (25 • C, 40 • C) in the dark, 25 • C in the presence of 5000 Lux illumination intensity (Illuminometer, LS-4000uv1000-7000LX, Beijing Star Technology Co., Ltd., Beijing, China), and 25 • C with relative humidity 75% (prepared using a saturated NaCl solution) in the dark for 30 days. Microcapsules (0.1 g) removed after a period of time were used for the determination of colour difference (Section 2.4) or dissolved in 10 mL of citric acid-sodium citrate buffer solution (0.1 M, pH 3.0) with an ultrasonic cleaner (25 • C) for determining total PCA content (pH differential method) [11] and 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging ability.

DPPH Free Radical Scavenging Ability
Following the slightly modified method of Chen et al. [22], 0.2 mL sample solution (Section 2.7) was mixed and reacted with 0.2 mL of DPPH ethanol working solution for 30 min at room temperature in the dark, and the absorbance was measured at 517 nm.

Degradation Kinetics of PCA Microcapsules during Storage
First-order kinetics were applied to evaluate anthocyanin degradation under different storage conditions according to Akhavan Mahdavi et al. [23] and Nazareth et al. [24]. The first-order kinetics are expressed by Equation (1), and the half-time (t 1/2 ) was calculated using Equation (2): where C 0 and C t are the anthocyanin content at the start and after t days of storage (mg/g), respectively, t is the storage day (d), and k is kinetic constant (d −1 ).

Preparation and Storage of Chewing Tablets
Formulation of chewing tablets: The weight per tablet was 0.5 g. The amounts for V C , ferrous fumarate, and calcium carbonate were 9%, 4.5%, and 37.5% per tablet, respectively, when added alone (not combined) to the tablet. The amounts of microcapsules, magnesium stearate, citric acid, and stevioside were 50%, 1%, 2%, and 0.1%, respectively, while the rest was supplemented by microcrystalline cellulose.
The preparation of the chewing tablets was based on the description of Wu et al. [25] with some modifications. The materials were weighed, screened through a 100-mesh sieve, and then mixed. A small amount of water was added to the mixture to create a damp mass, which was screened into granules using an 18-mesh sieve. The granules were dried in an oven (DHG-9070A, Shanghai Jinghong Experimental Equipment, Shanghai, China) at 50 • C for two hours, passed through a 20-mesh sieve, and compressed into tablets using a tablet machine (DF-4, Tianjing Port East Technology Development Company, Tianjin, China).

In Vitro Digestion
In vitro digestion was conducted according to the description of Zhang et al. [26], with some modifications. PCA microcapsules and chewing tablets (0.2 g) were dissolved in 50 mL 0.9% NaCl and adjusted to pH 2 with 1 M HCl solution, followed by adding 0.16 g pepsin (activity, 1:3000), then shaken (120 rpm/min) at 37 • C for two hours to stimulate gastric digestion. Thereafter, the pH of the digested solution was adjusted to pH 7.5 with 0.5 M NaHCO 3 , 12 mL of 2 mg/mL trypsin solution and 6 mL of 12 mg/mL bovine bile salt were added, and the mixture was shaken (120 rpm/min) at 37 • C for two hours to stimulate intestinal digestion. During gastric and internal digestion, 1 mL of the digestion solution was collected every 30 min and heated in a boiling water bath for one minute to kill the enzyme, then cooled and centrifuged for 15 min (6000× g), and the anthocyanin content in the supernatant was determined.

Statistical Analysis
All experiments were repeated three times, and the data were expressed as mean ± standard deviation. The statistical significance analysis of each factor in the singlefactor test was calculated using SPSS Statistics V22.0.

Results and Discussion
3.1. Influence of the Amount of Wall Material on the EE and Colour of PCA As shown in Table 1, with an increase in the white colour addition to the wall material, the L* and b* values of the microcapsules increased, whereas the a* values decreased; the colour of the microcapsules became increasingly light pink and was not close to that of the unencapsulated PCA. With an increase in the wall material, the anthocyanin content of the microcapsules decreased. Among them, the anthocyanin content in MD-PCA, MD-GA-PCA, and MD-WPI-PCA ranged from 76.01 to 143.39 mg/100 g, 69.61 to 83.25 mg/100 g, and 59.58 to 136.99 mg/100 g, respectively, higher than that (53.27 to 108.46 mg/100 g) of red cabbage anthocyanin microcapsules prepared by MD or MD-GA (combination ratio of 35:15 or 25:25) [27]. With an increase in the wall material, the EE of the PCA increased and then began to decline slightly, which may have been due to the decreased atomisation efficiency of spray drying caused by the high viscosity of the whole system [23]. When the core wall ratio was 1:1 for MD-PCA, 2:3 for MD-GA-PCA, and 1:1 for MD-WPI-PCA, the EE reached a maximum of 94.6%, 93.4%, and 94.4%, respectively, and the colour and anthocyanin content were suitable. Therefore, these three ratios were selected to prepare the microcapsules and used to determine their properties and stabilities. Akhavan Mahdavi et al. [23] found a higher EE of MD-GA (combination ratio of 3:1) on barberry anthocyanin than on MD alone. Norkaew et al. [10] found that the combination of MD with GA or WPI at ratios of 7:3, 1:1, and 3:7 had similar EEs (reached 100%) to that of MD alone on black rice anthocyanin, suggesting that the EE of anthocyanins was influenced by anthocyanin structures, the different amount of the wall materials, and combination ratios of MD to GA or WPI.

Physical Properties of the Three Kinds of Encapsulated PCA
Generally, the food industry requires the moisture content of food powder to not exceed 4% [28]. A low water content can maintain the microcapsules in a glassy state, which can restrict undesirable impacts such as molecular mobility and oxidation [29]. The water contents of the three types of microcapsules were all lower than 4% (Table 2), suggesting high spray-drying efficiency, good storage stability, and overall acceptability of the microcapsule powders [6]. Solubility reflects the dissolution rate of microcapsules when they are redissolved in water. The solubility of the three kinds of PCA microcapsules was 79.92% to 81.23%. The order was as follows: MD-PCA > MD-WPI-PCA > MD-GA-PCA (Table 2). Similar results were reported by Sarabandi et al. [30], who found that eggplant peel extract encapsulated by MD had higher solubility than that encapsulated by MD-GA. Pieczykolan and Kurek [7] indicated that capsules with small size and low water content had greater solubility.
Higher hygroscopicity causes powder agglomeration and affects its fluidity and nutritional value [28]. Ng and Sulaiman [31] believed that under 75% relative humidity, the microencapsulated powder with moisture absorption in the range of 15.1-20.0 g/100 g dry matter has strong hygroscopicity, indicating that the hygroscopicity of the three kinds of microencapsules in this study was weak (6.08~9.56 g/100 g). Moreover, MD-WPI-PCA had the lowest hygroscopicity, while MD-PCA had the highest hygroscopicity, suggesting that MD-WPI-PCA is more conducive to storage under an environment with high humidity. Rajabi et al. [32] indicated that microcapsules produced by wall materials with higher glass transition temperatures had lower hygroscopicity, such as the higher glass transition temperature of GA than that of MD.
The angle of repose, also known as the free-flow angle, was used to analyse the fluidity of finished products; the smaller the angle of repose, the better the fluidity [33]. A product with an angle of repose higher than 45 • had poor fluidity [34]. The fluidity ranking of the three types of microcapsules was as follows: MD-PCA > MD-GA-PCA > MD-WPI-PCA. This may be due to the lower moisture content and mutual adhesion forces between the microcapsules [35].

SEM Analysis
As shown in Figure 1, the shapes of the three types of microcapsules were spherical. No cracks were observed on the microcapsule surface. The sizes of the MD-PCA microcapsules were 8~80 µm, which was more uniform and relatively smaller than those of MD-GA-PCA (10~80 µm) and MD-WPI (10~100 µm), consistent with the conclusion of Sendri et al. [36] that the size of the microcapsules encapsulated by complex wall materials is larger than that for a single wall material. Most microcapsules had concavities and wrinkles. An increasing number of deeper concavities and wrinkles was found in the microcapsules of black rice anthocyanins encapsulated with MD and their combination with GA or WPI [10]. These were caused by the deformation of the wall materials, resulting from the rapid evaporation of the atomised droplets by the higher temperature of the inlet's hot air. In addition, lower emulsifying and film-forming abilities can also lead to surface problems in microcapsules [23]. dry matter has strong hygroscopicity, indicating that the hygroscopicity of the three kinds of microencapsules in this study was weak (6.08~9.56 g/100 g). Moreover, MD-WPI-PCA had the lowest hygroscopicity, while MD-PCA had the highest hygroscopicity, suggesting that MD-WPI-PCA is more conducive to storage under an environment with high humidity. Rajabi et al. [32] indicated that microcapsules produced by wall materials with higher glass transition temperatures had lower hygroscopicity, such as the higher glass transition temperature of GA than that of MD. The angle of repose, also known as the free-flow angle, was used to analyse the fluidity of finished products; the smaller the angle of repose, the better the fluidity [33]. A product with an angle of repose higher than 45° had poor fluidity [34]. The fluidity ranking of the three types of microcapsules was as follows: MD-PCA > MD-GA-PCA > MD-WPI-PCA. This may be due to the lower moisture content and mutual adhesion forces between the microcapsules [35].

SEM Analysis
As shown in Figure 1, the shapes of the three types of microcapsules were spherical. No cracks were observed on the microcapsule surface. The sizes of the MD-PCA microcapsules were 8~80 μm, which was more uniform and relatively smaller than those of MD-GA-PCA (10~80 μm) and MD-WPI (10~100 μm), consistent with the conclusion of Sendri et al. [36] that the size of the microcapsules encapsulated by complex wall materials is larger than that for a single wall material. Most microcapsules had concavities and wrinkles. An increasing number of deeper concavities and wrinkles was found in the microcapsules of black rice anthocyanins encapsulated with MD and their combination with GA or WPI [10]. These were caused by the deformation of the wall materials, resulting from the rapid evaporation of the atomised droplets by the higher temperature of the inlet's hot air. In addition, lower emulsifying and film-forming abilities can also lead to surface problems in microcapsules [23].

Storage Stability Analysis of the Three Kinds of Encapsulated PCA under Different Temperatures
As shown in Figure 2, with the increase in storage time, the PCA content of all three kinds of microcapsules decreased, and ∆E value became larger due to decreased PCA retention in the microcapsules. However, the DPPH free-radical-scavenging activity initially increased slightly and then decreased. The result was consistent with that of Wu et al. [37], who studied the change in antioxidant activities of blueberry anthocyanin microcapsules encapsulated by WPI-GA during storage, through the determination of ABTS radical scavenging ability and ferric iron reducing antioxidant power (FRAP). They indicated that the small increase may be caused by the phenolic acids formed by the degradation of anthocyanins in the microcapsules in the early stage of storage, which slightly increased the antioxidant capacity, followed by the decomposition of anthocyanins and phenolic acid in the microcapsules in the late stage of storage, when the antioxidant activity began to decline. The degradation of PCA was more obvious at 40 • C since high temperature causes the generation of unstable aglycones resulting from the loss of glycosides in anthocyanins [38].
As shown in Figure 2, compared with unencapsulated PCA (control), the change extent of anthocyanin retention rate, antioxidant activity, and ∆E value of the three kinds of microcapsules were all smaller. Moreover, as shown in Table 3, compared with unencapsulated PCA, the k value of the degradation kinetics of the microcapsules decreased and t 1/2 increased, suggesting that the wall material protected the PCA from heat and slowed its degradation rate. Among them, MD-PCA showed the best stability at 25 • C (t 1/2 = 173.29 days), consistent with the results of Kang et al. [39], who compared the stability of chlorophyll encapsulated with MD and its combination with GA. However, MD-GA-PCA had the best stability under 40 • C (t 1/2 = 69.31 d), suggesting that MD and GA had a good synergistic effect on protecting PCA from degradation under high temperature [6]. Machado et al. [27] indicated that higher thermal stability of the wall material resulted in a stronger stability of its core.     a 1 -a 3 ,b 1 -b 3 ,c 1 -c 3 ,d 1 -d 3 ) are the change at 25 • C, 40 • C in dark, 25 • C in the presence of 5000 Lux illumination intensity, and 25 • C with relative humidity 75% in the dark, for 30 days, respectively. PCA indicates purple corn anthocyanin; MD-PCA, MD-GA-PCA, and MD-WPI-PCA indicate purple corn anthocyanin encapsulated by maltodextrin, the combination of maltodextrin and gum arabic, and the combination of maltodextrin and whey protein isolate, respectively. Different lowercase letters in the same column indicate significant difference (p < 0.05).

Storage Stability of the Three Kinds of Encapsulated PCA under 5000 Lux Light Intensity
As shown in Figure 2, with the increase in light duration at 25 • C, the PCA retention rate, DPPH free radical scavenging activity, and the ∆E value showed the exact change trends as under different temperatures in the dark. However, light illumination accelerated the degradation of anthocyanin at 25 • C. Table 3 also shows that the degradation rate (k) under light was larger compared with that at 25 • C in dark, indicating that light accelerates the degradation of anthocyanins; thus, the half-life (t 1/2 ) decreased. Nie et al. [40] indicated that ultra-violet (UV) light changed anthocyanins from the ground state to the excited state, which then converted into C4 adducts after hydrolysis at the C4 position of the aglycon.
Among the three kinds of microcapsules, MD-GA-PCA microcapsules had the highest anthocyanin retention and lowest decrease extent of antioxidant activity and increase extent of ∆E value (Figure 2), suggesting that the combination of MD and GA can protect PCA against light more effectively. Akhavan Mahdavi et al. [23] found that barberry anthocyanins encapsulated by MD-GA had higher retention rates than those encapsulated by MD alone. Moreover, Table 3 shows that MD-GA-PCA had the smallest degradation rate (k) and the largest half-life (t 1/2 = 115.52 d), followed by MD-PCA and MD-WPI-PCA (t 1/2 = 99.02 and 77.02 d, respectively), whereas the t 1/2 of unencapsulated PCA was 69.31 d, indicating that MD-GA-PCA had the best protective ability for PCA against light, while the effect of MD-WPI was relatively low, consistent with the results shown in Figure 2.

Storage Stability of the Three Kinds of Encapsulated PCA under 75% Relative Humidity
As shown in Figure 2, with the increase in duration under 75% relative humidity in the dark, the changing trend of PCA retention rate, DPPH free radical scavenging activity, and the ∆E value showed the same change trends as those under the other three conditions. The degradation rates k of the encapsulated and unencapsulated PCA were higher than those for conditions at 25 • C in the dark without higher humidity, indicating that the humidity can accelerate their degradation. High relative humidity increases the mobility of molecules, thus accelerating the degradation reaction rate [23]. All three types of microcapsules exhibited a lower degradation rate k and a prolonged half-life (t 1/2 ), suggesting that the wall materials protected the PCA against humidity damage.
As shown in Figure 2, among the three kinds of microcapsules, the change extents of PCA retention rate, DPPH free radical scavenging activity, and the ∆E value of MD-WPI-PCA microcapsules were all smaller than those of MD-PCA and MD-GA. It was observed that the microcapsules of MD-PCA and MD-GA-PCA became sticky and caked on the 6th day of storage, while MD-WPI-PCA became slightly sticky by the 15th day but were still in powder form and maintained good fluidity, indicating that the hygroscopicity of MD-WPI-PCA was weaker than that of MD-PCA and MD-GA-PCA. Moreover, Table 3 shows that MD-WPI-PCA had the lowest degradation reaction rate k and longest t 1/2 compared with the other two wall materials encapsulated in PCA, indicating that MD-WPI had the strongest resistance to humidity and protected PCA against degradation caused by humidity more effectively, which may be due to its lower hygroscopicity [23]. Comparing the degradation rates k and t 1/2 under 75% humidity with those under other conditions, it was clear that MD-WPI had a stronger tolerance to humidity, whereas MD and MD-GA resisted light better.

Stabilities of Encapsulated PCA in Chewing Tablets with Ca 2+ , V C , and Fe 2+
As shown in Figure 3, with an increase in storage time, PCA retention rate and the value of a* of the chewing tablets decreased gradually, whereas ∆E increased, suggesting that PCA was unstable in the presence of Ca 2+ , V C , and Fe 2+ , similar to the results of Shi and Lv [41], who found that the addition of Ca 2+ and Fe 3+ decreased the absorbance of the PCA pigment solution. However, the retention rates of the three kinds of encapsulated PCA were higher than those of unencapsulated PCA; among them, the retention rate of MD-PCA was the highest and that of ∆E was the smallest, suggesting that MD can protect PCA from degradation caused by Ca 2+ , V C , and Fe 2+ more effectively. In particular, the effects of preventing degradation from V C and Fe 2+ were more obvious (retention rates of PCA 92.20% and 87.75%, respectively) than those of the other two wall materials. In addition, MD-WPI had stronger effects on preventing PCA degradation from Ca 2+ and V C and a weaker effect on preventing PCA degradation from Fe 2+ than MD-GA. Ren and Giusti [15] indicated that V C decreased the colour intensity of anthocyanins because of its condensation at the C4 position of anthocyanin and its activation of the oxidative cleavage of the pyrylium ring, and increased addition of WP to the PCA solution protected the destruction caused by V C through the formation of a complex with PCA, which is beneficial for the generation of flavylium cations.
value of a* of the chewing tablets decreased gradually, whereas ΔE increased, suggesting that PCA was unstable in the presence of Ca 2+ , VC, and Fe 2+ , similar to the results of Shi and Lv [41], who found that the addition of Ca 2+ and Fe 3+ decreased the absorbance of the PCA pigment solution. However, the retention rates of the three kinds of encapsulated PCA were higher than those of unencapsulated PCA; among them, the retention rate of MD-PCA was the highest and that of ΔE was the smallest, suggesting that MD can protect PCA from degradation caused by Ca 2+ , VC, and Fe 2+ more effectively. In particular, the effects of preventing degradation from VC and Fe 2+ were more obvious (retention rates of PCA 92.20% and 87.75%, respectively) than those of the other two wall materials. In addition, MD-WPI had stronger effects on preventing PCA degradation from Ca 2+ and VC and a weaker effect on preventing PCA degradation from Fe 2+ than MD-GA. Ren and Giusti [15] indicated that VC decreased the colour intensity of anthocyanins because of its condensation at the C4 position of anthocyanin and its activation of the oxidative cleavage of the pyrylium ring, and increased addition of WP to the PCA solution protected the destruction caused by VC through the formation of a complex with PCA, which is beneficial for the generation of flavylium cations.  a 1 -a 3 ,b 1 -b 3 ,c 1 -c 3 ) are the change in the presence of Ca 2+ , V C , and Fe 2+ , respectively. Note: PCA indicates purple corn anthocyanin; MD-PCA, MD-GA-PCA, and MD-WPI-PCA indicate purple corn anthocyanin encapsulated by maltodextrin, the combination of maltodextrin and gum arabic, and the combination of maltodextrin and whey protein isolate, respectively. Different lowercase letters in the same line or uppercase letters in the same day indicate significant difference (p < 0.05).

Digestion Stabilities of PCA from Microcapsules and Chewing Tablets
As shown in Figure 4, at the initial time, the PCA released from the MD-encapsulated microcapsules to the gastric digestion solution reached a maximum, whereas the PCA released from MD-GA and MD-WPI reached a maximum in the digestion solution after gastric digestion (0.5 h), mainly due to the high solubility of MD. This phenomenon was also observed by Norkaew et al. [10], who studied the release of anthocyanins from black rice during digestion. The content of PCA released from the three types of microcapsules during gastric digestion was steady, whereas the extent of change was larger during intestinal digestion, consistent with the results of Righi da Rosa et al. [6], who studied the stimulated digestion of microencapsulated blueberry anthocyanins. Compared with the initial value of anthocyanins in the digestion solution, the retention of MD-PCA, MD-GA-PCA, and MD-WPI-PCA after two hours of gastric digestion was 89.40%, 85.73%, and 92.95%, respectively. After two hours of intestinal digestion, the retention of MD-PCA, MD-GA-PCA, and MD-WPI-PCA decreased to 35.36%, 30.33%, and 43.12%, respectively, suggesting that MD-WPI can protect PCA more effectively during gastric and intestinal digestion, followed by MD and MD-GA. For the anthocyanins from the chewing tablets, the release of encapsulated PCA from the chewing tablets was slower, and its retention was 95.27% and 23.50% after gastric and intestinal digestion, respectively, while the retention of unencapsulated PCA was 82.67% and 14.81%, respectively, suggesting that encapsulation can improve the digestion stability of PCA from the chewing tablets, which was mainly because the wall materials delayed the release of PCA and avoided the degradation caused by the digestion solution.
letters in the same line or uppercase letters in the same day indicate significant difference (p < 0.05).

Digestion Stabilities of PCA from Microcapsules and Chewing Tablets
As shown in Figure 4, at the initial time, the PCA released from the MD-encapsulated microcapsules to the gastric digestion solution reached a maximum, whereas the PCA released from MD-GA and MD-WPI reached a maximum in the digestion solution after gastric digestion (0.5 h), mainly due to the high solubility of MD. This phenomenon was also observed by Norkaew et al. [10], who studied the release of anthocyanins from black rice during digestion. The content of PCA released from the three types of microcapsules during gastric digestion was steady, whereas the extent of change was larger during intestinal digestion, consistent with the results of Righi da Rosa et al. [6], who studied the stimulated digestion of microencapsulated blueberry anthocyanins. Compared with the initial value of anthocyanins in the digestion solution, the retention of MD-PCA, MD-GA-PCA, and MD-WPI-PCA after two hours of gastric digestion was 89.40%, 85.73%, and 92.95%, respectively. After two hours of intestinal digestion, the retention of MD-PCA, MD-GA-PCA, and MD-WPI-PCA decreased to 35.36%, 30.33%, and 43.12%, respectively, suggesting that MD-WPI can protect PCA more effectively during gastric and intestinal digestion, followed by MD and MD-GA. For the anthocyanins from the chewing tablets, the release of encapsulated PCA from the chewing tablets was slower, and its retention was 95.27% and 23.50% after gastric and intestinal digestion, respectively, while the retention of unencapsulated PCA was 82.67% and 14.81%, respectively, suggesting that encapsulation can improve the digestion stability of PCA from the chewing tablets, which was mainly because the wall materials delayed the release of PCA and avoided the degradation caused by the digestion solution.

Conclusions
When the mass ratio was 1:1 for PCA to MD, 2:3 for PCA to MD-GA combination (mass ratio of 1:1), and 1:1 for PCA to MD-WPI combination (mass ratio of 3:7), MD-PCA, MD-GA-PCA, and MD-WPI-PCA had the highest EEs (94.6%, 93.4%, and 94.4%, respectively), favourable colour, and anthocyanin content in microcapsules (117.22, 84.08, and 110.81 mg/100 g, respectively). The three types of encapsulated PCA were all spherical. Their sizes ranged from 8 to 100 µm. The moisture content of the three kinds of microcapsules was lower than 4%, their solubility was good (79.92~81.23%), and the hygroscopicity was weak (6.08~9.56 g/100 g). The order of sol-ubility was MD-PCA > MD-WPI-PCA > MD-GA-PCA, whereas the hygroscopicity and fluidity order was MD-PCA > MD-GA-PCA > MD-WPI-PCA. All three types of wall materials improved the PCA stability, whether in storage or during simulated digestion. MD-GA-PCA had the strongest resistance to the storage degradation caused by 40 • C temperature or 5000 Lux light illumination. MD-WPI-PCA showed the strongest resistance to degradation at 75% relative humidity and the highest stability during gastric and intestinal digestion. MD-PCA can maintain the strongest stability at 25 • C or when used in chewing tablets in the presence of Ca 2+ , V C , or Fe 2+ , especially the strongest resistance to V C and Fe 2+ . MD-WPI-PCA had stronger resistance to Ca 2+ and V C but weaker resistance to Fe 2+ than MD-GA-PCA. The results of this study provide a reference for the application of PCA to storage, digestion, and processing. MD is a good wall material for PCA encapsulation in regular conditions. MD-GA can be used when high temperatures or light illumination are difficult to avoid, and MD-WPI can be considered under very high humidity or for higher digestion stability. In the future, encapsulated PCA will be added to different kinds of food products to investigate their stability and their influence on food qualities.