Preparation of blueberry anthocyanin liposomes and changes of vesicle properties, physicochemical properties, in vitro release, and antioxidant activity before and after chitosan modification

Abstract The preparation of blueberry anthocyanin liposomes (BAL) was optimized by response surface methodology. Then, chitosan was used to modify BAL and the environmental stability, in vitro release, and antioxidant activity studies of anthocyanin liposome (An‐Lip), and chitosan‐modified anthocyanin liposome (CS‐An‐Lip) was studied. The results showed that the particle size, zeta potential, and entrapment efficiency of BAL were 210.7 ± 1.8 nm, ‐ 20.0 ± 1.0 mV, and 82.13%, respectively. After chitosan modification, the encapsulation efficiency and zeta potential of anthocyanin liposomes were improved. The results of environmental stability analysis showed that under certain conditions, the addition of chitosan could stabilize the color characteristics of anthocyanins and the loading amount of anthocyanins (LC%). In vitro release and simulated gastrointestinal digestion experiments showed that the addition of chitosan not only prolonged the sustained‐release time of anthocyanins, but also prolonged the residence time of anthocyanins in vivo, giving full play to the drug effect. In addition, the antioxidant activity test results showed that CS‐An‐Lip increased the antioxidant activity of anthocyanins.

anti-obesity, anti-diabetic, and vision protection activities (Amini et al., 2017;Kamenickova et al., 2013). However, the biological application of anthocyanins is limited due to their extreme susceptibility to temperature, light, pH, oxygen, ascorbic acid, enzymes, and metal ions (Vemana et al., 2020). In recent years, how to improve the stability and bioavailability of anthocyanins has become a new research hotspot. Encapsulation is an important way to protect environmentally sensitive bioactive molecules (Shishir et al., 2018). So far, many carrier systems have been developed and used to encapsulate anthocyanins to enhance storage and gastrointestinal stability and to improve the bioavailability and delivery of anthocyanins at target points, such as micelles, dendrimers, complex condensates, cyclodextrins, solid lipid nanoparticles, cellulose nanocrystal, polyelectrolyte complex, nano liposome, and different polymer formulations (Cruz et al., 2019;Shaddel et al., 2018;Flores et al., 2015;Lin et al., 2019). In this study, we focused on the application of lipid-based anthocyanin release system.
Liposomes are lipid-based colloidal delivery system, which consist of one or more phospholipid bilayer membranes surrounding an aqueous core. Liposomes can be employed for the delivery of both hydrophobic and hydrophilic compounds even for dual drug delivery (Gomez and Fernandze, 2006). Liposomes are considered a promising and safe carrier because of their biocompatibility, amphiphilic, nontoxic, and non-immunogenicity. Lipid molecules are easily biodegradable. It can enhance the transcellular transport by transient disruption of cellular lipophilic bilayers and could improve the para-cellular drug transport through altering the tight junctions (Dumont et al., 2018). In recent years, liposome has a wide range of potential applications in drug delivery, cosmetics, and food formulations. But during storage, liposomes easily aggregate. The fusion of liposomal membranes can cause burst release of payload due to oxidation during storage and hydrolysis of phospholipid under low pH and enzymatic conditions (Shishir, Karim, Gowd, Xie, et al., 2019;. So, how to effectively improve the function of liposome in complex environments became the focus of research. Some research suggests presence of glycolipids, glycoproteins, and proteins in the cell membrane helps to enhance its stability and function in the fluid-mosaic membrane model (Nicolson, 2014). These coating materials can be attached to the surface of the liposome by covalent bonds, hydrogen bonds, van der Waals forces, hydrophobic interactions, and/or electrostatic interactions. Alternatively, the surfactant can be incorporated into the lipid bilayer itself to modify its surface properties, such as phospholipids, glycolipids, sterols, and surfactants (Seong et al., 2018).
Chitosan is a nontoxic, biodegradable polysaccharide.
Modification of chitosan not only can slow the degradation of liposomes, but also prevents interplay among components during the process of storage and transportation. Due to the protonated amine groups on the repeated glycoside residues, chitosan has a cationic charge in an acidic medium and an increase in water solubility. The cationic nature of chitosan has attracted much attention as a polyelectrolyte excipient for medical use. In particular, the potential of chitosan as a carrier for anionic macromolecules (such as siRNA) has been highlighted by its effective intracellular and local delivery (Alavi et al., 2016;Zhang et al., 2018). In addition to electrostatic interactions with counterion molecules, chitosan has been reported to open tight junctions, alter the secondary structure of the stratum corneum keratin, and enhance cell membrane fluidity, thereby increasing skin permeability.
In this study, anthocyanin liposomes were prepared by thinfilm ultrasonic dispersion method, optimized by response surface method, established the preparation process, and verified the preparation process of anthocyanin liposomes by quadratic polynomial regression model. An-Lip was characterized by microscope and particle size analyzer. The formation of An-Lip was verified by infrared spectroscopy and then modified with chitosan. An-Lip was used as control. The physicochemical properties of CS-An-Lip were determined by mean particle size, polydispersity index (PDI), zeta potential, and morphology. The environmental stability of liposomes under the conditions of temperature and time was investigated. The in vitro release, bioavailability, and antioxidant activity of the liposomes were further investigated. It provides a basis for the rational design of new liposome drug delivery system.

| Materials and chemicals
Blueberry anthocyanin was prepared in the laboratory, and its purity was 98% by HPLC. Soybean lecithin (SP) was purchased from

| Preparation of An-Lip and CS-An-Lip
All the liposomes were prepared by film ultrasonic dispersion method with some minor modifications (Sylvester et al., 2017).
Firstly, SP and cholesterol were dissolved in chloroform (4 mL) and ether (8 mL) at the suitable ratio, and BA was dissolved in a phosphate buffer solution (PBS, 4 mL, pH7.3). The organic phrase was mixed completely by ultrasound in ice water to form stable water in oil type (W/O) emulsion. Secondly, the emulsion was placed in a round-bottom flask and then evaporated to form a gel by using a rotary evaporator, and residual organic solvent was removed by pumping in a vacuum for 1-2 h. After that, BA solution was injected into round-bottom flask, and then the resulting mixture was mixed uniform by using a rotary evaporator for 1 h; the mixture was homogenized using an ultrasonic cell disintegrator (JY92-II DN, Xinzhi Bio-technology and Science Inc.) for 20 min.
Ultimately, PBS was injected into the mixture and rotary evaporation for 30 min, homogenized using an ultrasonic cell disintegrator for 40 min, the liposome was successively filtered using 0.45 µm and 0.22 µm millipore membrane for sterilization, and stored at 4°C for further experiment.
CS-An-Lip is prepared by electrostatically depositing a cationic chitosan layer onto the surface of an anionic liposome loaded with anthocyanins, in short, by dissolving chitosan at pH 5.5. A chitosan solution was prepared in a 0.1% (v/v) acetic acid solution, and 0.45 μm was filtered to remove insoluble impurities. The An-Lip dispersion was slowly added to an equal volume of chitosan solution and further mixed for 30 min with magnetic stirring at 25℃. Leave the resulting mixture still overnight at 4°C. The excess acetic acid of the chitosan-coated liposomes was removed by centrifugation at 3,000 rpm for 3 min; then, the precipitated liposome pellet was resuspended in a 5% dextrose solution.

| Experimental design
RSM was developed to acquire the optimal preparation conditions by describing the relationships between the variables and the response (Luo et al., 2014). Three levels (determined in preliminary experiments) were studied for each of the three independent variables: ratio of SP to Cholesterol (A, w/w), ratio of SP to BA (B, w/w), and rotary evaporation temperature (C) ( Table 1).
Details of the 17 design points are shown in Table 2. All the experiments were developed in triplicate. The experimental data were subjected to multiple regression analysis to obtain the adjusted polynomial equation. ANOVA was done to identify the significant coefficients. The three-dimensional and contour graphs were built to identify the optimal combination of independent variables. For the analysis of the results, the Design-Expert V 8.6.0 software was used.

| Encapsulation efficiency (EE%) and loading capacity (LC%) of An-Lip and CS-An-Lip
The encapsulation efficiency (EE) is defined as the ratio of An-Lip and CS-An-Lip entrapped in liposomes to that in the delivery system, which is an important parameter for nanoliposomes when defined as delivery systems (Peng et al., 2010). It was calculated to determine the concentration of entrapped An-Lip and CS-An-Lip in nanoliposomes and unentrapped An-Lip and CS-An-Lip in the aqueous phase. The An-Lip and CS-An-Lip were separated from the aqueous phase using a freezing centrifuge (5810R, Eppendorf). A 0.5 mL nanoliposome suspension was taken and spun at 10,000 rpm for 30 min at 4°C. The same suspension was ruptured using a certain volume of ethanol, and the total amount of An was determined spectrophotometrically. The percentage of encapsulation efficiency (EE) was calculated according to Equation (Zhao et al., 2016): where W en is the amount of free An, and W total is the total amount of An present in 0.5 mL of nanoliposomes (W total and W en were measured by spectrophotometer and then calculated).
2.4.2 | Determination of particle size, polydispersity index, and ζ-potential The particle size, polydispersity index, and ζ-potential of An-Lip were determined using a NanoBrook 90Plus Zeta at 25°C (Tang et al., 2000). About 200 µL of each sample was taken and dispersed in deionized water to a final volume of 3 mL for the determination process. All data were calculated as the average of at least triplicate measurements.

| Surface morphology observation
An-Lip prepared at the optimum conditions was studied using a field-emission scanning electron microscope (SEM, Model H-7650, Hitachi, High-Technologies Co, Ltd.) operating at an accelerating voltage of 5 kV using a secondary electron detector 14. Take a small amount of sample powder deposited onto silicon wafer mounted on aluminum specimen stubs, and sputter coating (5 nm) with platinum to observe.

| Fourier transform infrared (FTIR) analysis
Fourier transform infrared (Perkin Elmer) was used to confirm the binding of anthocyanins to liposomes. The samples were mixed with 10% mannitol (used as drying protectant) to protect the structural integrity of liposomes (Guldiken et al., 2017), Then, the mixture was freeze-dried and lay the sample flat in the sample tank for analysis.
The scanning range was set at 4,000 cm −1 to 400 cm −1 . Each spectrum was obtained by 64 scans, and the resolution was set at 4 cm −1 . (Katouzian and Taheri, 2021).

| Storage environment stability
The stabilization effect of An-Lip and CS-An-Lip was assessed by measuring the change of A 520 and release rate of anthocyanin with time at room temperature and 4°C.

| Color changes of anthocyanin liposome
The tristimulus color coordinates of the anthocyanin liposomes were measured using an instrument colorimeter (Chroma Meter CR-400, Konica Minolta). The instrument is calibrated with a standard black and white plate. The measured parameters are L*(white/black), a* (red/green), and b* (yellow/blue).

| In vitro release of anthocyanins in liposomes
The in vitro release kinetics of anthocyanins from liposomes was performed by the previously described dialysis bag method  with minor modifications. Briefly, 2 mL liposome samples were transferred to a dialysis bag (Mw cutoff 8-14 kDa, Solarbio).
The dialysis bag was then incubated in 50 mL phosphate buffer (pH 7, 5 mM) release medium at 10,956 g in a shaker for 72 h at 37°C.
At specific time intervals, remove 2 mL of the release medium and replace with fresh medium. The release medium was diluted with MeOH to calculate the total cumulative amount of anthocyanins released from the liposomes. Anthocyanin content was measured at various sampling times at 520 nm using a spectrophotometer.

| Simulation of in vitro digestion for infusions
A simulated gastrointestinal (GIT) model was used to assess the bioavailability of anthocyanins in liposomes. As described elsewhere (Li et al., 2018) and with some modifications, a simplified model of simulated GIT was established, including the stomach and small intestine. The entire digestion process was carried out in a 37°C water bath shaker (DHSZ, Jiangsu Taicang Experimental Equipment Co.).
All solutions were preheated at 37°C prior to mixing.

| Stomach phase
The simulated gastric fluid (SGF) was prepared by dissolving NaCl (2 g), concentrated HCl (7 mL), and pepsin (3.2 g) into distilled water (1 L). The liposomes were mixed with SGF in a volume ratio of 1:1.
The pH of the simulated gastric phase was adjusted to about 1.5, and this digestion step took 4 h.

| Small intestine phase
The simulated intestinal fluid (SIF) was prepared by dissolving K 2 HPO 4 (6.8 g), NaCl (8.775 g), and pancreatin (3.2 g) into distilled water (1 L). Before the addition of SIF, the pH of mixtures taken from simulated stomach phase must be adjusted to 6.8 approximately, because the extremely acid environment could inactivate the pancreatin. Mixtures taken from simulated stomach phase were also mixed with SIF at a volume ratio of 1:1. The pH was adjusted to 7.0, and this digestion period took 4 h.

| Antioxidant activity
The DPPH and ABTS assays of anthocyanins liposomes were measured in accordance with a previously described method with slight where A 517 is the absorbance measured at 517 nm.
Accurately, suck 1 mL of sample solutions of different concentrations, successively add 1 mL of 9 mmol L −1 FeSO 4 solution, 1 mL of 9 mmol L −1 salicylic acid solution and 1 mL of 8.8 mmol L −1 H 2 O 2 solution, shake well, react in a 37°C water bath for 30 min, take deionized water as the reference solution and VC as the control, set 3 parallel in each group, and measure the absorbance value at 734 nm.
ABTS radical scavenging activity was calculated by using the following equation: where A 734 indicates the absorbance measured at 734 nm.

| Statistical analysis
Data were analyzed by Design-Expert V 8.6.0 software where appropriated. Results were described as means ± SE. One-way analysis of variance (ANOVA) was used to determine statistically significant difference among groups, and means of every two different groups were detected with test.

| Experimental design and data analysis
The combined effects of three factors with three levels on EE% were listed in Table 2 and were evaluated by a 17-run BBD for parameter optimization.
The fitted equation for predicting the maximum EE% was given as follows: Y = +82.02−29A+94B+0.60C+0.46AB+0.033AC−0.083 To show the type of interactions between the independent variables and the relationship between responses and levels of each variable, the response surface plots for EE% are presented in Figure 1.

| Effect of the independent variables on the encapsulation efficiency
In this study, A, B, C, AB, A 2 , B 2 , and C 2 are significant model terms.
The Model F-value of 213.58 implies the model is significant, which confirmed that the model could correctly explain and predict the results in the experiments. The R 2 value and R 2 ad could indicate that the regression models were fit for explaining the variability of response.
These results show that the effects of above various factors on An-Lip preparation are not simple linear relationship. The adjusted determination coefficient (R 2 adj) is 0.9917, indicating that the model is suitable to represent the actual relationship between response and variables, that is, the model is in line with this study (Table 3).
The corresponding response surface figure and contour map were drawn with the binomial fitting model. The results showed that with the increase of SP, the ratio of SP to cholesterol, and the ratio of SP to BA, EE% first increased and then decreased.
This phenomenon may be related to the saturation degree of liposomes. Cholesterol acts as a "flow buffer" in liposome system.
Higher than the phase transition temperature, the lipid bilayer can be arranged closely, thus reducing the drug leakage and increasing the EE%, the asymmetry, permeability, and rigidity of lipid bilayer membrane increased with the increase of cholesterol dosage, which resulted in drug penetration and decreased EE% (Li et al., 2018). In addition, the steeper the surface was, the more significant the effect indicated, so ratio of SP to cholesterol and ratio of SP to BA were the two main effects on encapsulation efficiency in this experiment. Surface openings in three-dimensional plots were downward. Encapsulation efficiency was on the rise as the growth of each factor, and then gradually reduced when it reached the peak. The peak value was also the maximum encapsulation efficiency point in these three factors (Figure 1).

| Validation of the optimized model
Box-Behnken design has the characteristics of good predictability, intuitive results, and simple operation. It can optimize the process parameters faster and more efficiently with the least number of experiments. At present, it has been widely used in drug prescription screening. In this experiment, Box-Behnken combined with response surface method was used to optimize the formulation of lipid preparation, which greatly reduced the workload, successfully constructed the model of the mass ratio of phospholipid to cholesterol, the volume ratio of phospholipid to anthocyanin and the effect of temperature on the entrapment efficiency, intuitively predicted the optimal formula, prepared three batches of samples according to the optimized conditions, measured their entrapment efficiency respectively, and carried out validation experiments, The predicted value of the optimal formulation liposome entrapment efficiency predicted by the model is 82.21%, the average value of the measured entrapment efficiency is 82.13%, and the deviation is 0.08%, indicating that the prediction of this model is accurate and reliable, the reproducibility of the formulation process is good, the particle size of the prepared liposome meets the requirements, the entrapment efficiency is stable, and it is suitable for the preparation of An-Lip.

| DLS, Zeta, PDI, and TEM characterization
An-Lip presents a standardized particle size distribution curve, showing the variation between particle sizes. The main particle sizes of An-Lip range from 100 nm to 300 nm (Figure 2a). Zeta potential was used as an indicator of the physical stability of liposomes. The zeta of An-Lip prepared in this study was −31.05 ± 1.0 mV (Figure 2b), indicating that the liposomes were negatively charged and the system was stable. Figure 2c shows the relationship between the intensity of scattered light and time in previous studies (Tang and Pikal, 2004;Guldiken et al., 2018).
The PDI values of nanoliposomes were 0.112-0.255. The PDI value of An-Lip was 0.237. In general, PDI measures particle size distribution from homogeneous to heterogeneous (0-1.0), where homogeneous particles are considered to be in the range of 0 to 0.3 (Zhao and Temelli, 2017). Therefore, the nanoliposome carrier developed in this study has uniform dispersion.

| Characterization of An-Lip and CS-An-Lip
As shown in Table 4, the coating of chitosan significantly increased the particle size and polydispersity index of the liposome. A higher chitosan content results in a larger particle size increase, which also indicates the formation of an additional chitosan layer. In the anthocyanin liposomes, the particle size increase of chitosan coating was more pronounced than that of empty liposomes. This indicates that the thicker layer is composed of negatively charged anthocyanins and positively charged shells. Although chitosan forms an interaction between the layers, the thicker layer does not seem to indicate that more anthocyanins are incorporated into the chitin liposomes because the chitin coating does not increase the load of anthocyanins in the liposomes.

| FTIR analysis
In order to further confirm the successful embedding of anthocyanins, Fourier transform infrared spectroscopy (FTIR) was used to characterize. In the range of 4,000-2,500 cm −1 , as shown in Figure 3a, An has an absorption peak at 3818-3339 cm −1 , and the stretching vibration of hydroxyl group occurs, but the absorption peak is obviously weakened, which is due to the reduction of carbonyl group in blueberry anthocyanins structure after purification (Cai et al., 2019). The skeleton vibration of benzene ring was mainly concentrated around 1596.60 cm −1 with only one absorption peak, and the unsaturated C-H deformation vibration absorption peak appeared at 1176.96 cm −1 of fingerprint region. Figure 4b shows The peaks observed at 1300-1100 cm −1 refer to the PO2 symmetric and asymmetric stretching vibration of a phospholipid, respectively, The An-Lip showed almost similar spectra to An which could be due to the well-encapsulation of An into the aqueous core of Lip. It is worth noting that the FTIR spectrum of An-Lip is similar to that of blank Lip, and there is no characteristic absorption

TA B L E 3 Analysis of variance
peak of An, which indicates that the coating mode of liposomes is physical coating. In Figure 4d, the characteristic spectrum of CH shows that the typical polysaccharide has a wide and strong peak at 3500-3000 cm −1 . The typical peak appears at 3339.

| Storage environment stability
The two liposomes were accurately transferred and placed at room temperature for sampling at 0, 1, 2, and 3 months. The encapsulation efficiency, drug-loading rate, zeta potential, and particle size were used as evaluation indexes to investigate the shell aggregation. The results are shown in Table 5. As is clear from the Table 5, the liposome has a lower degree of change in the encapsulation efficiency and the drug-loading rate at 4°C; thereby, it can be seen that the prepared liposome is suitable for storage under 4°C environments. In addition, under the same environment, the encapsulation efficiency of chitosan-modified anthocyanin liposomes was higher than that of unmodified liposomes, and the drug-loading rate decreased slowly.
Due to the incorporation of positively charged chitosan, the potential of anthocyanin liposomes is negatively converted to a positive charge, which also increases the charge of the liposome.

| Color changes of anthocyanin liposome
The color changes shown by anthocyanins liposomes during storage at 20°C for 2 weeks were examined. The results are presented

| In vitro release study
In order to effectively target organs and tissues passively, liposomes loaded with nutraceuticals should retain nutrients for a long time during the cycle. Therefore, the anthocyanin release be- These results indicate that the presence of chitosan can inhibit the release of anthocyanins from liposomes, which may be beneficial for extended release applications. It is speculated that chitosan increases the affinity of anthocyanins for the hydrophobic domains within the liposomes, thereby reducing their release tendency (Christodouleas et al., 2014).

TA B L E 5 Physical stability of anthocyanin liposomes and chitosan-coated liposomes at 4°C and 25°C
T (℃)

| Simulation of in vitro digestion for infusions
It can be seen from Figure (Zhao et al., 2015).
It can be seen from Figure 6 that the leakage rate of anthocyanins is time-dependent. The anthocyanin leakage rate increases with time, and the leakage rate tends to be stable at 120 min. It was also found that the stability of the liposome sample in the stomach environment was significantly higher than in the intestinal environment. This indicates that since the pH of SGF is 2, anthocyanins are very stable under acidic conditions. After 2.5 h of digestion, the unmodified liposome anthocyanin leakage rate was 45.85%, and the chitosan-modified liposome leakage rate was 35.76%, which significantly improved its stability in the stomach. This indicates that the polysaccharide-modified liposome can effectively inhibit the release of anthocyanins in gastric juice and help to reduce its destruction by enzymes and acids in the stomach.

| Antioxidant activity
Several chemical analysis methods have been established to evaluate the oxidation/reduction potential of plant compounds (Liang et al., 2017;Narod and Nazarali, 2014).

| CON CLUS IONS
In this study, the formulation of anthocyanin liposomes was optimized by response surface methodology, and nanoliposomes with high entrapment efficiency were successfully prepared. On this basis, chitosan was added to improve its stability and functionality. It was characterized by a variety of analytical equipment. Through the experimental design, the best sample is selected based on the central composite design. The average particle size of the best nano-liposome was less than 300 nm, and the zeta potential showed that the system was stable. FTIR spectra showed that there were electrostatic interactions and hydrogen bonds between phospholipid polar groups, chitosan amine part, and main anthocyanin extract polyphenols. In addition, the results of environmental stability analysis showed that the addition of chitosan could stabilize the color characteristics and content of anthocyanins under certain conditions. In vitro release and simulated gastrointestinal digestion experiments showed that the addition of chitosan not only prolonged the release of anthocyanins, but also prolonged the retention time of anthocyanins in vivo. In addition, the antioxidant activity test showed that CS-An-Lip could improve the antioxidant activity of anthocyanins. Therefore, the combination of biopolymer and chitosan on the Lip surface is a promising method, which has great application potential in food, dietary supplement, and pharmaceutical industry. In further research, it is recommended to implement the developed loaded chitosan to strengthen/enrich food and test the acceptability of the product through sensory evaluation. The interesting results obtained in this study will pave the way for the application of these nanostructures in different food systems.

ACK N OWLED G EM ENTS
The study was supported by the Jilin Science and Technology Development Program of China (Grant no. 18SS017; 20200404023YY).

CO N FLI C T O F I NTE R E S T
The authors declare that they do not have any conflict of interest.

E TH I C A L A PPROVA L
This study does not involve any human or animal testing.

I N FO R M E D CO N S E NT
Written informed consent was obtained from all study participants.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the finding of this study are available from the corresponding author upon resonable request.