Hybrid RSM-GA approach to optimize extraction conditions for blueberry anthocyanins with high antioxidant activity

Abstract

BACKGROUND:

Blueberry contains an abundance of anthocyanins, which are a bioactive component of this fruit. Anthocyanins can be extracted via various methods, and each has pros and cons.

OBJECTIVE:

This current study reported the optimal conditions for the ultrasonic-assisted enzymatic extraction of blueberry anthocyanins simulated using response surface methodology (RSM) coupled with a genetic algorithm (GA).

METHODS:

The Box–Behnken design (BBD) was used for the RSM, and the extraction conditions were as follows: temperature, 42°C; ultrasonic power, 310 W; enzyme volume, 0.25%; and extraction time, 42 min.

RESULTS:

The maximum predicted extraction yield was 6.67 mg/g. The antioxidant activity of anthocyanins extracted via RSM and GA was based on the hydroxyl free radical activity and supersonic anion free radical activity of 230.50±12.76μg/ml and 4.41±0.36μg/ml, respectively. Anthocyanins exracted by the proposed method has stronger free radical removal capacity than that of Vc.

CONCLUSIONS:

This study shows that the combination of RSM with GA represents an optimized method for extracting blueberry anthocyanins for use in the food industry. This method can maintain high antioxidant potential and can be used as an alternative strategy for high-value products.

Keywords

Blueberry anthocyanins response surface methodology genetic algorithm antioxidant

1 Introduction

Blueberry (Ericaceae: Vaccinium) is a popular berry that includes a number of nutrients and antioxidants. Blueberries are rich in anthocyanins, vitamin C, superoxide dismutase (SOD), arbutin, and other active ingredients [1, 2]. Several research studies have shown that blueberry fruit has unique functions in preventing brain nerve aging, enhancing heart function, maintaining gut health, and improving eyesight and anti-cancer activities [3 –5]. Therefore, blueberry is a multi-nutritive, multi-functional fruit with high medicinal value. Blueberry has been recognized as one of the five main healthy foods by the United Nations Food and Agriculture Organization [6]. Due to its rich nutritional ingredients, blueberry has been used for the development of different functional foods, including fruit juice, yogurt, fruit wine,anthocyanin capsules and dried products [7 –10].

In recent years, extraction and separation of functional ingredients such as anthocyanins from plants have become research hotspot [11]. Anthocyanins are extracted from plants using hot water, organic solvent, ultrasonic, microwave, and supercritical fluid extraction. Each method has advantages and disadvantages [12 –14]. In ultrasonic extraction, the unique physical characteristics of ultrasonic waves can promote the deformation or breaking of plant cell walls, so that the effective components of plants can be fully extracted. The extraction rate is significantly higher than that of traditional pure organic solvent extraction; thus, ultrasonic extraction has been widely used to extract anthocyanins from natural products, such as blueberry pomace and purple sweet potatoes [15, 16].

Enzymatic treatments can destroy cell membranes and other components, leading to unbound from proteins and polysaccharides, and dissolve anthocyanin components more easily. This method has the advantages of mild conditions, high recovery rates, low energy consumption, and no destruction of anthocyanins caused by high temperatures [17].

Response surface methodology (RSM) is an experimental method commonly used to study the extraction process of functional ingredient. It considers the random error of the experiment and fits the complex unknown function relationship in a small area with a simple mathematical model. By optimizing the experimental conditions, RSM can continuously analyze each experiment level [18].

Genetic algorithm (GA) is a simulated evolutionary algorithm based on Darwin’s evolutionary theory and Mendel’s population genetic theory; it simulates the evolutionary process of natural organisms to develop a random global search for the optimal solution. It has the advantages of fast search speed and good stability and is not constrained by the continuity of the optimization function. This approach can efficiently determine the optimal global solution by overcoming the shortcomings of the RSM, such as the deviation of the design level from the optimal conditions due to the lack of experience and continuity of the optimization function [19].

This study selected extraction temperature, ultrasonic power, enzyme volume, and extraction time as the factors for investigation. The extraction rate of blueberry anthocyanins was used as an evaluation index. Based on the Box–Behnken response surface design, the objective function of blueberry anthocyanin extraction rate was established. Furthermore, GA was used to optimize the response surface design, and determine the optimal extraction conditions for blueberry anthocyanins. We then conducted a preliminary study on antioxidant activity in vitro. This approach lays a theoretical foundation for studying the preservation of blueberry anthocyanins, and it warrants large-scale validation.

2 Materials and methods

2.1 Raw material

Fresh blueberries were provided by Zhejiang Xinchang Zhaofeng Ecological Garden. These were sliced, dried in a freeze dryer (Free Zone 2.5 Plus, Labconco, Kansas City, MO, USA) to achieve a constant weight, crushed, sieved through 60 mesh, and stored in a –20°C refrigerator.

2.2 Chemicals

Cellulase (enzyme activity≥50 U/mg) was purchased from Yuanye Biotechnology Co., Ltd (Shanghai, China). 2,2-diphenyl-1-picrylhydrazyl (DPPH; 97% purity), cyanidin 3-glucoside (>97%, HPLC grade), nitroblue tetrazolium (NBT; molecular biology grade, 98%), beta-nicotinamide adenine dinucleotide disodium salt (NADH-2Na; 96%, HPLC grade), o-phenanthroline, and FeSO₄.7H₂O were obtained from Shanghai Dingguo Biotechnology Co., Ltd (Shanghai, China). The water was super pure grade, and the other reagents were analytically pure.

2.3 Methods

2.3.1 Extraction of anthocyanins from blueberry

Freeze-dried blueberry powder and distilled water were mixed in a 1:1 ratio, and the pH of the mixed solution was adjusted after magnetic stirring for 30 min. 0.2% cellulase (w/w) was added for enzymolysis. Ultrasonic-assisted extraction (KS-600DE ultrasonic cleaner; Kunshan Ultrasonic Instruments Co., Ltd., Kunshan City, China) was performed to obtain blueberry anthocyanin extract. After extraction and filtration, the extract was evaporated using a rotary evaporator (RE-52A; Yarong Biochemical Instrument Factory, Shanghai, China) to obtain an anthocyanin concentrate. The concentrated solution was stored at 4°C until testing.

2.3.2 Measurement of anthocyanins

Total anthocyanins were measured using the pH differential method described by Flores-Sosa et al. (2021) [20]. We added 1 mL of a concentrated solution of blueberry anthocyanins to each of the two 10 mL brown volumetric flasks. Then, a buffer solution at a pH of 1.0 and a buffer solution at a pH of 4.5 were added to a volume of 1 mL. After the solution was shaken and stored in the dark at 4°C for 2 h, the absorbance at 520 nm and 700 nm was measured using a spectrophotometer (UV-9000; Metash, Shanghai, China).

Anthocyanin content was estimated as follows: $C = \frac{D \times MW \times DF \times 1000}{ɛ \times l}$ (1)

where sample absorbance D is equal to (A₅₂₀ –A₇₀₀) at pH 1.0 and (A₅₂₀–A₇₀₀) at pH 4.5, MW corresponds to the molecular weight of cyanidin 3-glucoside (449.2 g.mol^-1), ɛ is the molar absorptivity coefficient of cyanidin 3-glucoside (26,900 L·cm^-1·mol^-1), DF is the dilution factor used, and l is the path length (1 cm). The results were expressed as equivalents of cyanidin 3-glucoside in mg/g.

2.3.3 Single-factor experiments

Four extraction factors, namely, temperature (10–50°C), ultrasonic power (100–500 W), enzyme volume (0.1–0.5%), and extraction time (20–60 min), were analyzed in single-factor experiments. Only one factor was changed in each experiment to determine the preliminary range of extraction factors, and the final values were as follows: temperature, 30°C; ultrasonic power, 300 W; enzyme volume, 0.2%; and extraction time, 40 min. Each experiment was performed in triplicate, and the anthocyanin content was determined as described in Section 2.3.2.

2.3.4 RSM design

The extraction was optimized using the Box–Behnken design (BBD) in the RSM based on the single-factor experiments. The BBD with four variables [X1: temperature (°C); X2: ultrasonic power (W); X3: enzyme volume; and X4 extraction time (min)] at three levels (30, 40, and 50°C; 200, 300, and 400 W; 0.1, 0.2, and 0.3%; and 30, 40, and 50 min, respectively) was used to determine the optimum anthocyanin extraction parameters (Table 1). Mathematical modeling and regression analysis of the experimental data were performed using Design-Expert 12.0.3.0 64-bit statistical software (Minneapolis, MN, USA).

Table 1
Optimizing anthocyanin extraction parameters: factors and levels

Code Temperature A (°C) Ultrasonic Power B (W) Enzyme volume C (mg/mL) Time D (min)

–1 30 200 0.1 30

0 40 300 0.2 40

1 50 400 0.3 50

Code	Temperature A (°C)	Ultrasonic Power B (W)	Enzyme volume C (mg/mL)	Time D (min)
–1	30	200	0.1	30
0	40	300	0.2	40
1	50	400	0.3	50

2.3.5 Optimization using GA

GA is a stochastic nonlinear optimization algorithm with infinite iteration characteristics, making its prediction and optimization values infinitely close to the actual value [21]. The RSM model was used as the fitness function of GA to optimize the extraction of blueberry anthocyanins. In the optimization process, random uniform and dispersion cross functions were selected, and MATLAB R2014b was used to optimize the ultrasonic-assisted enzymatic extraction of blueberry anthocyanins. The objective function obtained was as follows: $Maximize y = f (x); X^{L} i \leq X_{i} \leq X^{U} i, i = 1, 2, \dots, n$ (2)

where f(x) is the objective function obtained from the RSM model, X is the input vector, Y is the yield of anthocyanins from blueberries (mg/g), and X ^L _i and X ^U _i are the lower and upper boundaries of the input vector, respectively.

The upper and lower limits of each factor level were selected as the constraint conditions for the GA optimization. The constraint conditions for the optimal anthocyanin yield are shown in the following formula: $30 \leq x 1 \leq 50; 200 \leq x 2 \leq 400; 0 . 1 \leq x 3 \leq 0 . 3; 30 \leq x 4 \leq 50$ (3)

2.3.6 Determination of the antioxidant activity of blueberry anthocyanins

We added 10 mL of blueberry anthocyanin extract to 30 mL of n-hexane liquid to extract and remove lipid substances. The sublayer extract was further extracted with ethyl acetate (extract: ethyl acetate = 1:4; volume ratio) to remove fat-soluble flavonoids. Subsequently, the extract was purified using AB-8 macroporous resin, eluted using an eluting reagent at pH 3.0 (pH adjusted using hydrochloric acid), and 60% ethanol to wash away sugars, and the anthocyanins were eluted with the extract. The eluent was concentrated via suspension evaporation and freeze-dried in a vacuum to obtain freeze-dried blueberry anthocyanin powder. An appropriate amount of freeze-dried blueberry anthocyanin powder and positive control ascorbic acid (Vc) were prepared in a series of solutions in pure water at different concentrations to determine antioxidant activity.

2.3.6.1. DPPH radical-scavenging ability The DPPH assay was performed according to the method by Liu et al. method (2020) with slight modifications [22]. We mixed 4 mL of DPPH solution (0.5 mmol/L) with 4 mL of sample solutions of different concentrations (0–60μg/mL). After reacting in the dark at 37°C for 30 min, the absorbance of the mixture at 517 nm was measured using a UV-Vis spectrophotometer (UV-9000; Metash, Shanghai, China). The absorbance was recorded as A₁. Using absolute ethanol as the control, the measured absorbance was recorded as A₀. The DPPH solution was then replaced with anhydrous ethanol, and the absorbance was recorded as A₂.

Equation (4) was employed to calculate the DPPH radical-scavenging rate. $DPPH • radical scavenging rate (%) = (1 - \frac{A_{1} - A_{2}}{A_{0}}) \times 100$ (4)

2.3.6.2. OH radical-scavenging ability The ·OH radical scavenging ability was assessed using Quan’s method with some modifications [23]. In the experimental group, the reaction mixture consisted of the sample to be tested: 1.864 mmol/L o-phenanthroline solution, phosphate-buffered solution (PBS; pH 7.4), 1.864 mmol/L FeSO₄·7H₂O solution, and 0.03% H₂O₂ at a ratio of 1:1:2:1:1. The mixture was incubated at 37°C for 60 min. The absorbance of the mixture was recorded at 536 nm (A_s). The absorbance of water was recorded (A_n) as a reference, and the absorbance of water instead of the sample and H₂O₂ was recorded (A_b). $OH radical - scavenging (%) = (\frac{A_{s} - A_{n}}{A_{b} - A_{n}}) \times 100$ (5)

2.3.6.3. O₂^- radical-scavenging ability O₂^- radical-scavenging ability was determined using the assay of Goto et al. (2019) [24]. We dissolved NBT, NADH-2Na, and PMS in 0.1 mmol/L PBS and prepared 120μMol/L PMS solution, 300μMol/L NBT solution, 936μMol/L NADH-2Na solution, respectively.

The sample was mixed with the above solution according to a volume ratio of 1:1:2:2 and reacted at 25°C for 5 min. Then, we measured the absorbance at 560 nm with a spectrophotometer to obtain A₁. The absorbance measured with ethanol instead of the sample was A₀, and the absorbance measured with PBS instead of the added reagent was A₂. The formula for calculating the superoxide anion radical scavenging rate was as follows: $O_{2}^{-} radical - scavenging ability (%) = (1 - \frac{A_{1} - A_{2}}{A_{0}}) \times 100$ (6)

2.3.7 Statistical analysis

All data were presented as the mean±standard deviation (n = 3). Statistical analyses were performed using SAS version 6.12 (SAS Inc., Cary, NC, USA). Duncan’s multiple range test was used to analyze the differences between multiple groups of samples. Data were plotted using GraphPad Prism 9, and responses to the surface drawing were visualized using Design Expert 12.0.3.0 64-bit. MATLAB 2014b software was used for the GA analysis.

3 Results and discussion

3.1 Effect of single factors on anthocyanin extraction

Generally, high temperatures can accelerate mass transfer in a medium, reduce viscosity and surface tension of a solvent, strengthen cavitation effects, and improve extraction rates [25]. The extraction rate of blueberry anthocyanins initially increased and then decreased as the extraction temperature increased (Fig. 1A). When the temperature reached 40°C, the extraction rate of blueberry anthocyanins reached its maximum and then decreased with further increases in temperature. This observation is due to the poor stability of anthocyanins and also the increased temperatures break down the hydroxyl groups on the benzene ring of anthocyanins and degrade anthocyanins, decreasing the extraction rate [26, 27]. Moreover, high temperatures partially inactivate the enzyme, reducing the catalytic performance and extraction rate of anthocyanins [28]. Therefore, this result suggests that 40°C is considered a more appropriate extraction temperature.

Fig. 1

Effects of temperature (A), ultrasonic power (B), enzyme volume (C), and extraction time (D) on the yield of anthocyanins.

Fig. 2

Response surface (3D) showing the interactive effects of the extraction temperature (X1), ultrasonic power (X2), enzyme volume (X3), and extraction time (X4) on the yield of anthocyanins.

As for the influence of ultrasonic power on anthocyanins, the extraction rate of blueberry anthocyanins increased first and then decreased with increasing ultrasonic power (Fig. 1B). When the ultrasonic power was 300 W, the extraction rate of blueberry anthocyanins reached maximum. Ultrasound-assisted extraction caused the entire system to form numerous bubbles due to vibration collisions. These bubbles collide with other bubbles and burst, resulting in the rupture of the blueberry cell wall, thereby increasing the extraction rate of blueberry anthocyanins [29, 30]. However, when the ultrasonic power exceeded 300 W, the bubbles in the entire ultrasonic system saturated, resulting in a smaller impact force of the burst and a weaker cavitation effect [31], which slowed down the dissolution of anthocyanins. Higher ultrasonic power also produces excessive heat, resulting in the degradation of blueberry anthocyanins and reducing the extraction rate [12, 32].

Enzyme-assisted extraction is a green, efficient, and economical method for extracting anthocyanins and phenols from plants [33]. The function of the enzyme, cellulase, is to destroy the cell wall of plants, thus improving the permeability of the cell membrane and facilitating the extraction of anthocyanins. The extraction rate of blueberry anthocyanins was first increased and then decreased with increasing cellulase concentration (Fig. 1 C). The extraction rate of blueberry anthocyanins reached maximum when the cellulase concentration was 0.2%. Afterwards, the blueberry anthocyanin extraction rate no longer increased with increasing cellulase concentration. Considering the economic cost, 0.20% was selected as the optimal enzyme dosage for this experiment.

The extraction rate first increased rapidly, peaked at 40 min, and did not increase with the increasing time (Fig. 1D). Extending the extraction time can make blueberry fully in contact with cellulase, thus improving the extraction rate. However, excessive duration of extraction may lead to enzymatic degradation of anthocyanins [34, 35] and increase the contact between anthocyanins and oxygen, thus leading to the degradation of anthocyanins and reduction in the extraction rate [36].

3.2 RSM analysis

3.2.1 Model equation establishment and variance analysis

The Box–Behnken response surface design was performed using Design-Expert 12 software. The experimental scheme and results are presented in Table 2. The blueberry anthocyanin extraction rate contains four factors: extraction temperature, ultrasonic power, enzyme volume, and extraction time. The regression fitting equation is obtained by evaluating the model’s determination coefficient (R²) and variance analysis and by comparing the fitting degree of the fitting equation. Anthocyanin yield from blueberries as a response is described by the following polynomial equation:

Table 2
Results of the Box–Behnken response surface design using Design-Expert 12 software

Number A Temperature (°C) B Ultrasonic Power (W) C Enzyme Volume (%) D Time (min) Y yield of anthocyanins (mg/g)

1 –1 –1 0 0 2.74

2 1 –1 0 0 2.90

3 –1 1 0 0 2.15

4 1 1 0 0 4.70

5 0 0 –1 –1 3.93

6 0 0 1 –1 3.58

7 0 0 –1 1 3.12

8 0 0 1 1 5.94

9 –1 0 0 –1 2.07

10 1 0 0 –1 3.45

11 –1 0 0 1 2.25

12 1 0 0 1 4.37

13 0 –1 –1 0 3.12

14 0 1 –1 0 5.37

15 0 –1 1 0 4.73

16 0 1 1 0 4.77

17 –1 0 –1 0 2.27

18 1 0 –1 0 4.36

19 –1 0 1 0 4.19

20 1 0 1 0 5.07

21 0 –1 0 –1 2.48

22 0 1 0 –1 3.45

23 0 –1 0 1 2.72

24 0 1 0 1 3.97

25 0 0 0 0 6.56

26 0 0 0 0 6.49

27 0 0 0 0 6.27

Number	A Temperature (°C)	B Ultrasonic Power (W)	C Enzyme Volume (%)	D Time (min)	Y yield of anthocyanins (mg/g)
1	–1	–1	0	0	2.74
2	1	–1	0	0	2.90
3	–1	1	0	0	2.15
4	1	1	0	0	4.70
5	0	0	–1	–1	3.93
6	0	0	1	–1	3.58
7	0	0	–1	1	3.12
8	0	0	1	1	5.94
9	–1	0	0	–1	2.07
10	1	0	0	–1	3.45
11	–1	0	0	1	2.25
12	1	0	0	1	4.37
13	0	–1	–1	0	3.12
14	0	1	–1	0	5.37
15	0	–1	1	0	4.73
16	0	1	1	0	4.77
17	–1	0	–1	0	2.27
18	1	0	–1	0	4.36
19	–1	0	1	0	4.19
20	1	0	1	0	5.07
21	0	–1	0	–1	2.48
22	0	1	0	–1	3.45
23	0	–1	0	1	2.72
24	0	1	0	1	3.97
25	0	0	0	0	6.56
26	0	0	0	0	6.49
27	0	0	0	0	6.27

$\begin{matrix} R_{1} = 6.44 + 0.765 A + 0 . 476667 B + 0 . 509167 C + 0 . 284167 D + 0 . 5975 AB - \\ 0 . 3025 AC + 0 . 185 AD - 0 . 5525 BC + 0 . 07 BD + 0 . 7925 CD - 1 {. 80917 A}^{2} - 1.48667 B^{2} - \\ 0.567917 C^{2} - 1.70792 D^{2} \end{matrix}$ (7) where A, B,C, and D represent temperature, ultrasonic power, enzyme volume, and time, respectively

The experimental results were analyzed using ANOVA of the binomial model based on Design Expert 12 software (Table 3). A value of P < 0.001 indicated that the regression equation was statistically significant. P > 0.05 indicates a lack of fit and suggests that there was no mismatch factor in the model and that the experimental results fit well with the model. The correlation coefficient R² = 0.956 demonstrates that the method is reliable and that the equation has practical significance in simulating an accurate analysis. The quadratic terms A², B², C², and D² and the factors AB, BC, and CD of the simulation equation reached a significant level (P < 0.05). According to the F value, the four factors (extraction temperature, ultrasonic power, enzyme volume, and extraction time) affected the extraction rate of blueberry anthocyanins in the following order: extraction temperature > enzyme volume > ultrasonic power > extraction time.

Table 3

ANOVA Results of the binomial model using the Design Expert 12 software

Source	Sum of Squares	df	Mean Square	F-Value	P-Value
Model	47.90	14	3.42	42.95	<0.0001	significant
A-A	7.02	1	7.02	88.14	<0.0001	*
B-B	2.73	1	2.73	34.22	<0.0001	*
C-C	3.11	1	3.11	39.05	<0.0001	*
D-D	0.969	1	0.969	12.16	0.0045	*
AB	1.43	1	1.43	17.92	0.0012	*
AC	0.366	1	0.366	4.59	0.0533
AD	0.137	1	0.137	1.72	0.2145
BC	1.22	1	1.22	15.32	0.0021	*
BD	0.02	1	0.02	0.246	0.6289
CD	2.51	1	2.51	31.53	0.0001	*
A²	17.46	1	17.46	219.09	<0.0001	*
B²	11.79	1	11.79	147.94	<0.0001	*
C²	1.72	1	1.72	21.59	0.0006	*
D²	15.56	1	15.56	195.26	<0.0001
Residual	0.9561	12	0.08
Lack of Fit	0.9103	10	0.09	3.98	0.2176	not significant
Pure Error	0.0458	2	0.023
Cor Total	48.86	26

3.2.2 Interactive effects of extracting factors on anthocyanin extraction

The 3D response surface of the above experimental model was drawn using Design Expert 12 software. The 3D response surface analysis of the four factors revealed the interaction intensity between the extraction conditions of blueberry anthocyanins: enzyme volume and extraction time > extraction temperature and ultrasonic power > ultrasonic power and enzyme volume > extraction temperature and enzyme volume > extraction temperature and extraction time > ultrasonic power and extraction time. This interacting impact was consistent with the ANOVA results. The best extraction conditions of blueberry anthocyanins were as follows: temperature of 42°C, ultrasonic power of 310 W, enzyme volume of 0.25%, and extraction time of 42 min. The regression model predicted that the yield of anthocyanins would be 6.67 mg/g.

3.2.3 Optimization of extraction parameters of blueberry anthocyanins based on GA

The GA toolbox in MATLAB R2014b was used to optimize the extraction process. When the number of operation iterations was 60, the yield of blueberry anthocyanins reached a maximum (Fig. 3). Under the optimal conditions, the coding values of the experimental factors extraction temperature, ultrasonic power, enzyme volume, and extraction time were 0.2004, 0.1153, 0.4843, and 0.2085, respectively; and the corresponding factor levels were 41.99°C, 310.78 W, 0.249% and 42.01 min, respectively. Under these conditions, the theoretical yield of blueberry anthocyanins is 6.67 mg/g.

Fig. 3

Results optimization with genetic algorithm.

3.2.4 Validation experiment

To verify the method’s reliability, the process parameters were revised as follows: extraction temperature of 42°C, ultrasonic power of 310 W, enzyme volume of 0.25%, and extraction time of 42 min. Experiments were performed in triplicate using these parameters. The experimental extraction rate of anthocyanins was 6.53±0.14 mg/g, and the relative error between the experimental and theoretical values was 2.10%, indicating that it was feasible to optimize the extraction process of blueberry anthocyanins via GA.

3.2.5 Determination of the antioxidant activity of blueberry anthocyanins

The antioxidant ability of anthocyanins can be determined by measuring their DPPH free radical-scavenging rate. The IC₅₀ (the concentration value of the antioxidant when the clearance rate was 50%) was used as an indicator to evaluate the ability of anthocyanins to scavenge DPPH free radicals. The smaller the IC₅₀, the better the scavenging effect. To eliminate DPPH, it is necessary to pair a single electron in anthocyanin with a single electron in DPPH. The DPPH clearance rate of Vc was dependent on anthocyanin concentrations (Fig. 4A), which was consistent with the results of Wu [37]. The calculated IC₅₀ of blueberry anthocyanins and ascorbic acid (Vc) to scavenge 0.5 mmol/L DPPH free radicals was 27.80±1.25μg/mL and 17.67±0.43μg/mL, respectively. At concentrations from 10–60μg/mL, the DPPH radical scavenging rate of Vc was significantly higher than the blueberry anthocyanins (P < 0.05). When the concentration was > 40μg/mL, the scavenging rate of Vc for the DPPH radical reached saturation.

Fig. 4

Free radical scavenging ability (DPPH free radical, ·OH free radical, and O₂^- free radical) of blueberry anthocyanins.

However, hydroxyl radicals are harmful to organisms. When the production and elimination of hydroxyl radicals are unbalanced, they affect the normal physiological function of the organism. ·OH is produced by the Fenton system. When ·OH is removed, the absorbance increases with increasing Fe^2 + concentration in the system. When the concentration of blueberry anthocyanins was 20–60μg/mL, the ·OH clearance rate was significantly higher than that of Vc (P < 0.05) (Fig. 4B). The IC₅₀ of blueberry anthocyanins to scavenge ·OH was 230.50±12.76μg/mL. Further, the ·OH clearance rate of Vc was low and not dose-dependent, whereas the ·OH clearance rate of blueberry anthocyanins was high and dose-dependent. The ·OH clearance rate of anthocyanins was lower than 50%. This phenomenon may be because, during the Fenton reaction, Fe^2 + always exists in the system. Also, there are multiple phenolic hydroxyl groups in the anthocyanin structure, making it easy for anthocyanins to react with Fe^2 + to produce precipitation, thus making some anthocyanins unable to exert their antioxidant capacity, leading to a decline in antioxidant activity.

The superoxide anion is one of the precursors of singlet oxygen molecules and ·OH, which can indirectly produce lipid peroxides. Therefore, the superoxide anion free radical scavenging efficiency is significant for identifying antioxidant capacity. The superoxide anion radical was generated by the PMS-NADH system. In this study, we determined the scavenging capacity of samples for superoxide anion radicals by reducing NBT. The superoxide anion radical scavenging ability of blueberry anthocyanins and Vc was concentration dependent, and the IC₅₀ of Vc was 21.68±0.83μg/mL, higher than the IC₅₀ of blueberry anthocyanins (4.41±0.36μg/mL). The above results showed that blueberry anthocyanins had a strong superoxide anion radical scavenging potential, significantly better than that of Vc (P < 0.05). Zhou also found that the EC₅₀ of blueberry anthocyanins and Vc for scavenging superoxide anion radicals were 29.17μg/mL and 84.38μg/mL, respectively [39]. This may be due to the weaker dissociation energy of the O-H bond and the more vital ability of blueberry anthocyanins to provide hydrogen for superoxide anions [34].

4 Conclusions

The influence of extraction factors on extracting blueberry anthocyanins was ordered as follows: extraction temperature > enzyme volume > ultrasonic power > extraction time. The optimal extraction process of blueberry anthocyanins obtained by RSM and GA was as follows: extraction temperature, 42°C; ultrasonic power, 310 W; enzyme volume, 0.25%; and extraction time, 42 min. Furthermore, blueberry anthocyanins exert strong antioxidative potential to scavenge ·OH and O₂^-. Our findings provide a practical extraction approach that could enhance the yield of blueberry anthocyanins; however, our results should be validated in larger-scale settings.

Footnotes

Acknowledgments

The authors thank Dr. Li-Shu Wang for helpful comments and language revision.

Funding

This work was supported by the National Key R&D Program of China [Grant No.2021YFD2100502-1-2]; and Hangzhou Agricultural and Social Development Research Initiative Design Project (Grant No. 20190101A06).

Conflict of interest

The authors have no conflict of interest to report.

Author contributions

Xiangjun Fang: Investigation, Data curation, Writing – Original draft

Weijie Wu: Software, Visualization

Honglei Mu: Methodology, Validation

Hangjun Chen: Resources, Data Curation

Xiaolin Zheng: Supervision, Writing - Review & Editing

Haiyan Gao: Project administration, Funding acquisition.

References

Rossi

, Woods

, Leisner

. Quantification of total phenolic, anthocyanin, and flavonoid content in a diverse panel of blueberry cultivars and ecotypes. Hortscience. 2022;57(8):901–909. DOI: 10.21273/HORTSCI16647-22

, Hu

, Liao

, Jiang

, Xiu

, Gaowa

, et al. Effect of atmospheric cold plasma treatment on antioxidant activities and reactive oxygen species production in postharvest blueberries during storage. J Sci Food Agr. 2020;100:5586–95. DOI: 10.1002/jsfa.10611

Duan

, Tarafdar

, Chaurasia

, Singh

, Bhargava

, Yang

, et al. Blueberry fruit valorization and valuable constituents: A review. Int J Food Microbiol. 2022;381(16):109890. DOI: 10.1016/j.ijfoodmicro.2022.109890

Giampieri

, Cianciosi

, Alvarez-Suarez

, Quiles

, Forbes-Hernandez

, Navarro-Hortal

et al. Anthocyanins: what do we know until now? J Berry Res. 2023;13(1):1–6. DOI: 10.3233/JBR-220087

Iqbal

, Javaid

, Hameeda

. Synergic interactions between berry polyphenols and gut microbiota in cardiovascular diseases. Med J Nutr Metab. 2022;15(4):555–73. DOI: 10.3233/MNM-220071

Wang

, Gao

, Tao

, Wu

, Cui

. Influence of γ-irradiation on the reactive-oxygen metabolism of blueberry fruit during cold storage. Innov Food Sci Emerg. 2017;41:397–403. DOI: 10.1016/j.ifset.2017.04.007

Liao

, Shen

, Manickam

, Li

, Tao

, Li

, et al. Investigation of blueberry juice fermentation by mixed probiotic strains: Regression modeling, machine learning optimization and comparison with fermentation by single strain in the phenolic and volatile profiles. Food Chem. 2023;405(PartB):134982. DOI: 10.1016/j.foodchem.2022.134982

Zheng

, He

, Li

, Tao

, Wang

, et al. Influence of adding Perilla seed oil on potato blueberry yogurt quality during storage at 4°C. Lwt - Food Sci Technol. 2022;168(1):113921. DOI: 10.1016/j.lwt.2022.113921

Varo

, Martin-Gomez

, Serratosa

, Merida

. Effect of potassium metabisulphite and potassium bicarbonate on color, phenolic compounds, vitamin C and antioxidant activity of blueberry wine. Lwt - Food Sci Technol. 2022;163(15):113585. DOI: 10.1016/j.lwt.2022.113585

10.

, Li

, Zhang

, Cheng

, Wang

, Yang

, et al. Effects of different drying temperatures on flavor related quality of blueberry. J Berry Res. 2023;13(1):7–20. DOI: 10.3233/JBR-220071. DOI: 10.3233/JBR-220071

11.

Liao

, Zhang

, Chen

, Battino

, Giampieri

, Bai

, et al. Recovery of value-added anthocyanins from mulberry by a cation exchange chromatography. Cur Res Food Sci. 2022;(5):1445–145. DOI: 10.1016/j.crfs.2022.08.022

12.

Ngoc Nhon

, Diem My

, Tuong Vi

, Kim Lien

, Thao Minh

, Doan Duy

, et al. Enhancement of extraction effectiveness and stability of anthocyanin from Hibiscus sabdariffa L. J Agr Food Res. 2022;10:100408. DOI: 10.1016/j.jafr.2022.100408

13.

Kuasnei

, Wojeicchowski

, Santos

, Pinto

, Ferreira

SRS

, Zielinski

AAF

. Modifiers based on deep eutectic mixtures: A case study for the extraction of anthocyanins from black bean hulls using high pressure fluid technology. J Supercrit Fluid. 2022;191:105761. DOI: 10.1016/j.supflu.2022.105761

14.

Zhang

, Wang

, Wu

, Battino

, Giampieri

, Bai

, et al. Recovering high value-added anthocyanins from blueberry pomace with ultrasound-assisted extraction. Food Chem: X. 2022;16(30):100476. DOI: 10.1016/j.fochx.2022.100476

15.

Yusoff

, Mat

Taher Z

, Rahmat

, Chua

. review of ultrasound-assisted extraction for plant bioactive compounds: Phenolics, flavonoids, thymols, saponins and proteins. Food Res Int. 2022;157:111268. DOI: 10.1016/j.foodres.2022.111268

16.

Zhang

, Wang

, Wu

, Battino

, Giampieri

, Bai

, et al. Recovering high value-added anthocyanins from blueberry pomace with ultrasound-assisted extraction. Food Chem: X. 2022;16:100476. DOI: 10.1016/j.fochx.2022.100476

17.

Granato

, Fidelis

, Haapakoski

, dos Santos Lima

, Viil

, Hellström

, et al. Enzyme-assisted extraction of anthocyanins and other phenolic compounds from blackcurrant (Ribes nigrum L.) press cake: From processing to bioactivities. Food Chem. 2022;391:133240. DOI: 10.1016/j.foodchem.2022.133240

18.

Zhu

, Yu

, Jiao

, Tong

, Zhang

, Chang

, et al. Optimization of quercetin extraction method in Dendrobium officinale by response surface methodology. Heliyon. 2019;5(9):e02374. DOI: 10.1016/j.heliyon.2019.e02374

19.

Sarkar

, Salauddin

, Mukherjee

, Shariati

, Rebezov

, Tretyak

, et al. Application of bio-inspired optimization algorithms in food processing. Cur Res Food Sci. 2022;5:432–50. DOI: 10.1016/j.crfs.2022.02.006

20.

Flores-Sosa

, Soto-Magaña

, Gonzalez-de la Vara

, Sanchez-Segura

, Bah

, Rivera-Pastrana

, et al. Red drupelet reversion in blackberries caused by mechanical damage is not linked to a reduction in anthocyanin content. Postharvest Biol Tec. 2021;180:111618. DOI: 10.1016/j.postharvbio.2021.111618

21.

Zahedi

, Elkamel

, Lohi

. Genetic algorithm optimization of supercritical fluid extraction of nimbin from neem seeds. J Food Eng. 2010;97(2):127–34. DOI: 10.1016/j.jfoodeng.2009.10.001

22.

Liu

, Guo

, Wu

, Wang

, Golly

, Haile Ma

. The necessity of walnut proteolysis based on evaluation after in vitro simulated digestion: ACE inhibition and DPPH radical-scavenging activities. Food Chem. 2020;311:125960. DOI: 10.1016/j.foodchem.2019.125960

23.

Quan

, Zhang

, Wang

, Zeng

, Qin

, He

, et al. Assessment antioxidant properties of Torreya grandis protein enzymatic hydrolysates: Utilization of industrial by-products. Food Biosci. 2021;43:101325. DOI: 10.1016/j.fbio.2021.101325

24.

Goto

, Kuda

, Shikano

, Charrouf

, Yamauchi

, Yokozawa

, et al. Induction of superoxide anion radical-scavenging capacity in an argan press cake-suspension by fermentation using Lactobacillus plantarum Argan-L1. Lwt - Food Sci Technol. 2019;100:56–61. DOI: 10.1016/j.lwt.2018.10.033

25.

Tao

, Sun

. Enhancement of food processes by ultrasound: A review. Crit Rev Food Sci. 2015;55(4):570–94. DOI: 10.1080/10408398.2012

26.

Danisman

, Arslan

, Toklucu

. Kinetic analysis of anthocyanin degradation and polymeric colour formation in grape juice during heating. Czech J Food Sci. 2015;33(2):103–8. DOI: 10.17221/446/2014-cjfs

27.

Ursu

MGS

, Milea

, Pacularu-Burada

, Dumitrascu

, Râpeanu

, Stanciu

, et al. Optimizing of the extraction conditions for anthocyanin’s from purple corn flour (Zea mays L): Evidences on selected properties of optimized extract. Food Chem: X. 2013;17:100521. DOI: 10.1016/j.fochx.2022.100521

28.

, Smith

, Hossain Md

. Extraction of phenolics from citrus peels II. Enzyme-assisted extraction method. Sep Purif Technol. 2006;48(2):182–8. DOI: 10.1016/j.seppur.2005.07.005

29.

Khadhraoui

, Ummat

, Tiwari

, Fabiano-Tixier

, Chemat

. Review of ultrasound combinations with hybrid and innovative techniques for extraction and processing of food and natural products. Ultrason Sonochem. 2021;76:105625. DOI: 10.1016/j.ultsonch.2021.105625

30.

Kumar

, Srivastav

, Sharanagat

. Ultrasound assisted extraction (UAE) of bioactive compounds from fruit and vegetable processing by-products: A review. Ultrason Sonochem. 2020;70:105325. DOI: 10.1016/j.ultsonch.2020.105325

31.

Belwal

, Huang

, Li

, Duan

, Zhang

, Aalim

, et al. Optimization model for ultrasonic-assisted and scale-up extraction of anthocyaninsfrom Pyrus communis ’Starkrimson’ fruit peel. Food Chem. 2019;297:124993. DOI: 10.1016/j.foodchem.2019.124993

32.

, Gong

, Ma

, Xie

, Wei

, Xu

. Ultrasound treatment degrades, changes the color, and improves the antioxidant activity of the anthocyanins in red radish. Lwt - Food Sci Technol. 2022;165:113761. DOI: 10.1016/j.lwt.2022.113761

33.

Kumar

, Dahuja

, Sachdev

, Tomar

, Lorenzo

, Dhumal

, et al. Optimization of the use of cellulolytic enzyme preparation for the extraction of health promoting anthocyanins from black carrot using response surface methodology. Lwt - Food Sci Technol. 2022;163:113528. DOI: 10.1016/j.lwt.2022.113528

34.

Landbo

, Meyer

. Enzyme-assisted extraction of antioxidative phenols from black currant juice press residues (Ribesnigrum). J Agr Food Chem. 2001;49(7):3169–77. DOI: 10.1021/jf001443p

35.

Landbo

, Meyer

. Effects of different enzymatic maceration treatments on enhancement of anthocyanins and other phenolics in black currant juice. Innov Food Sci Emerg. 2004;5(4):503–13. DOI: 10.1016/j.ifset.2004.08.003

36.

Maran

, Manikandan

, Thirugnanasambandham

, Nivetha

, Dinesh

. Box-Behnken design based statistical modeling for ultrasound- assisted extraction of corn silk polysaccharide. Carbohyd Polym. 2013;92(1):604–11. DOI: 10.1016/j.carbpol.20

37.

, Zhou

, Chen

, Li

, Wang

, Zhang

. Comparison and screening of bioactive phenolic compounds in different blueberry cultivars: Evaluation of anti-oxidation and α-glucosidase inhibition effect. Food Res Int. 2017;100 Part 1:312–24. DOI: 10.1016/j.foodres.2017.07.004

38.

Zhou

, Xie

, Yang

, Liu

. Antioxidant activity of high purity blueberry anthocyanins and the effects on human intestinal microbiota. Lwt - Food Sci Technol. 2020;117:108621. DOI: 10.1016/j.lwt.2019.108621

39.

Xie

, Zhang

, Wang

, Shen

, Nie

, Xie

. Preparation: Characterization and antioxidant activities of acetylated polysaccharides from Cyclocarya paliurus leaves. Carbohyd Polym. 2015;133:596–604. DOI: 10.1016/j.carbpol.2015.07.031

Hybrid RSM-GA approach to optimize extraction conditions for blueberry anthocyanins with high antioxidant activity

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1 Introduction

2 Materials and methods

2.1 Raw material

2.2 Chemicals

2.3 Methods

2.3.1 Extraction of anthocyanins from blueberry

2.3.2 Measurement of anthocyanins

2.3.4 RSM design

Table 1 Optimizing anthocyanin extraction parameters: factors and levels Code Temperature A (°C) Ultrasonic Power B (W) Enzyme volume C (mg/mL) Time D (min) –1 30 200 0.1 30 0 40 300 0.2 40 1 50 400 0.3 50

3 Results and discussion

3.1 Effect of single factors on anthocyanin extraction

3.2.1 Model equation establishment and variance analysis

3.2.3 Optimization of extraction parameters of blueberry anthocyanins based on GA

3.2.5 Determination of the antioxidant activity of blueberry anthocyanins

Footnotes

Acknowledgments

Funding

Conflict of interest

Author contributions

References

Table 1
Optimizing anthocyanin extraction parameters: factors and levels

Code Temperature A (°C) Ultrasonic Power B (W) Enzyme volume C (mg/mL) Time D (min)

–1 30 200 0.1 30

0 40 300 0.2 40

1 50 400 0.3 50