Abstract

Blueberries are rich in bioactive anthocyanins, which are associated with health benefits contributing to reducing the risk of diabetes and cardiovascular diseases. The objective of this study was to improve the yield of anthocyanins extracted from rabbiteye blueberry (Vaccinium ashei) fruits cultivated in Nanjing using ultrasound-assisted extraction and to identify the individual anthocyanins present in the extract. The extraction conditions of blueberry anthocyanins were optimized using response surface methodology. The Box–Behnken test was designed to investigate the effect of extraction using different ethanol concentrations, extraction time, and liquid-to-solid ratios. The optimum conditions of the extraction derived from the model were as follows: extraction time 24 h at 30°C using 72.50% ethanol which contains 0.02% v/v hydrochloric acid as a solvent and liquid-to-solid ratio 20 : 1 v/w. The extraction yield was 16.21 ± 0.44 mg/g under these optimum conditions. The 13 peaks of the anthocyanin extract from rabbiteye blueberry fruits in Nanjing were tentatively identified using high-performance liquid chromatography (HPLC) and high-performance liquid chromatography-electrospray ionization interface-mass spectrometer (HPLC-ESI-MS), which are the derivatives of delphinidin, cyanidin, petunidin, peonidin, and malvidin that were glycosylated by glucose, galactose, or arabinose. This research provides a reliable scientific basis for efficacious extraction and identification of anthocyanins from blueberry fruits, which would be helpful for further investigation of the function and application of blueberry anthocyanins extract to human health.

1. Introduction

Blueberries (Vaccinium spp.) have a lot of health benefits, such as a super antioxidant function, with high levels of anthocyanins [1]. Blueberries are mainly distributed in the northern hemisphere with more than 450 varieties, and almost all of the commercial blueberries are harvested from three species, rabbiteye blueberry (V. ashei Reade; syn. V. virgatum Ait.), highbush blueberry (V. corymbosum L.), and lowbush blueberry (V. angustifolium Aiton and V. myrtilloides) [2]. Rabbiteye blueberries have strong adaptability to be grown under warmer climate conditions [3], like in the hilly area of Southern China, and there are large-scale rabbiteye blueberry planting bases in Nanjing, China.

Ultrasound-assisted extraction (UAE) has been used in the food and pharmaceutical sectors to extract the bioactive compounds, such as anthocyanins [4]. The eminence of the UAE is affordable equipment cost, easy operation, less solvent usage, repeatable various samples, environmental-friendly, short extraction time with great capacity, and applicable for large industrial scale [5]. Moreover, UAE could diminish the damage of anthocyanins [6].

The extraction techniques may be improved by optimization of various parameters in order to obtain high extraction efficiency and yield in an economically advantageous process [7]. Optimization of UAE conditions using response surface methodology (RSM) has been applied to enhance the efficiency of extraction of flavonoids [6]; however, few reports were about the optimization of anthocyanins extraction from blueberries using RSM. Hence, the objectives of the present study were to optimize the UAE parameters (ethanol concentration, extraction time, and liquid-to-solid ratio) of anthocyanins from rabbiteye blueberry (V. ashei) fruits cultivated in Nanjing using RSM. In addition, high-performance liquid chromatography-electrospray ionization interface-mass spectrometer (HPLC-ESI-MS) was used in this research to identify the individual anthocyanins in rabbiteye blueberries.

2. Materials and Methods

2.1. Plant Materials and Chemicals

Rabbiteye (V. ashei) blueberry fruits were harvested in Fujiabian Orchard Picking (Nanjing, China) in July 2017 and stored at −20°C avoiding light. Hydrochloric acid (36–38%), potassium chloride, sodium acetate trihydrate, glacial acetic acid, ethanol, and all the other reagents were brought from Sinopharm Group Chemical Reagent Co., Ltd (Shanghai, China). All chemical reagents used in this study were of pure analytical grade.

2.2. UAE and Single-Factor Analysis

Frozen fruits of blueberries were thawed and homogenized. A certain amount of blueberry samples were extracted by ethanol containing some hydrochloric acid (HCl). The single factors for blueberry anthocyanins extraction procedures included HCl dosage (0.02, 0.05, 0.10, 0.30, and 0.50% v/v), ethanol concentration (50, 60, 70, 80, and 95%), extraction time (0.5, 1, 3, 6, and 24 h), extraction temperature (25, 30, 40, 50, and 60°C), and liquid-to-solid ratio (3 : 1, 5 : 1, 10 : 1, 15 : 1, and 20 : 1 v/w). UAE was carried out by initially sonicating the samples in an RH7200DB CNC ultrasonic device (Kunshan Ultrasonic Instrument Co. Ltd., Suzhou, China) at 100 W and 40°C for 20 minutes.

2.3. Box–Behnken Response Surface Optimization

Extraction time (A), ethanol concentration (B), and liquid-to-solid ratio (C) were chosen as independent variables (Table 1). RSM was used to specify the supreme conditions for the extraction of blueberry anthocyanins. Design Expert version 7.0 (StatEase Corp., Minneapolis, MN, USA) was used to encode and integrate each level of each factor, and the 17 sets of experimental protocols were designed. A Box–Behnken Design (BBD) matrix was performed in the RSM experimental design. Blueberry anthocyanins were extracted according to these protocols, and the extraction yield was obtained. Experimental runs were randomized to lessen the consequence of unforeseen variability in the noticed responses. The general second-order polynomial model used for the response surface analysis was as follows:where the response function (Y) was partitioned into linear, quadratic, and interactive components; b0 is defined as the constant; bi is defined as a linear coefficient; bii is defined as a quadratic coefficient; bim is defined as a cross-product coefficient; and Xi and Xm are the levels of the independent variables. Analysis of variance (ANOVA) tables were yielded, and the impact and regression coefficients of the individual linear, quadratic, and interaction terms were determined.

2.4. Determination of the Yield of Total Anthocyanin

The spectrophotometric pH differential method was used to determine the total anthocyanin content in the extracts [8]. Anthocyanin crude extract was taken and, respectively, added to potassium chloride buffer (pH 1.0) and sodium acetate buffer (pH 4.5). The distilled water was used as a blank, and cyanidin-3-glucoside was used as the standard. The absorbance of each dilution was measured at 520 nm and 700 nm using a UV-6300 visible spectrophotometer (Shanghai Media Instrument Co., Ltd., Shanghai, China). Absorbance (A) was calculated as follows:

The extraction yield of the total anthocyanin content expressed as mg cyanidin-3-glucoside equivalent per gram of blueberry sample was computed by the method proposed by Chen et al. [9]:where MW is the molecular weight of cyanidin-3-glucoside (449.2 g/mol), DF is the dilution factor of extract, V is the volume of extract (L), 1000 is the factor for conversion from g to mg, ε is the molar extinction coefficient of cyanidin-3-glucoside (26900 L/mol·cm), L is the cuvette length (cm), and wt is the weight of the sample (g). The extraction rate (%) was the percent of extracted anthocyanin yield divided by the total anthocyanin.

2.5. HPLC Analysis

After purification by AB-8 macroporous adsorption resin, the sample was filtered through a 0.22 μm polyvinylidene fluoride (PVDF) membrane before being subjected to HPLC. HPLC analysis was conducted in an Agilent-1200 (Agilent Technologies, USA) which was equipped with a G1311A binary pump and a G1315D diode array detector (DAD). Chromatographic analysis was performed on a 250 mm × 4.6 mm, 5 μm particle size, end-capped reverse-phase Eclipse XDB-C18 column (Agilent Technologies, USA). Mobile phase A was 1% phosphoric acid dissolved in ultrapure water, whereas mobile phase B was 100% acetonitrile. The elution gradient was as follows: 5% B (from 0 to 5 min), 5% to 10% B (from 5 to 15 min), 10% B (from 15 to 25 min), 10% to 12% B (from 25 to 35 min), 12% to 15% B (from 35 to 50 min), 15% to 18% B (from 50 to 60 min), 18% to 25% B (from 60 to 80 min), and 25% to 30% B (from 80 to 90 min). The running temperature was 25°C, and the injection volume was 10 µL. The detection was conducted at 520 nm at a flow rate of 0.6 mL/min.

2.6. HPLC-ESI-MS Analysis

HPLC-ESI-MS analysis was carried out using an Agilent-1100 HPLC system equipped with a UV detector and LCQ ion-trap mass spectrometer (MS) fitted with an electrospray ionization interface (ESI) (Agilent Technologies, USA). The analytical column was a Zorbax SB-C18 column (250 mm × 4.6 mm, 5 μm). Mobile phase A was 6% formic acid dissolved in ultrapure water, whereas mobile phase B was 6% formic acid dissolved in acetonitrile. The elution gradient was the same as described above. The ESI capillary voltage was 3.0 kV in positive-ion (NI) mode with the capillary temperature at 350°C. A nebulizing gas of 1.5 L/min and a drying gas of 10 L/min were applied for ionization using nitrogen. ESI was performed with the scan range between 100 and 1200 m/z.

2.7. Statistical Analysis

All data presented are the mean value ± standard deviation (SD) of three independent experiments. Figures were obtained using GraphPad Prism Version 7 (GraphPad Software, Inc., CA, USA). One-way ANOVA or Sidak’s multiple-comparison test was performed to determine the statistical differences between different groups. Differences were considered significant at .

3. Results and Discussion

3.1. Single-Factor Experimental Analysis of Blueberry Anthocyanin Extraction

Most of the anthocyanins are more stable in acidic conditions, and the derivation takes place under neutral or alkaline conditions [10]. However, adding an excessive amount of strong acid may lead to the partial hydrolysis of the glycosidic bonds and acyl groups of anthocyanins or cause the linkages with metals or copigments break during the extraction process, thereby affecting the extraction yield of anthocyanins [11]. Figure 1(a) presents the extraction yield of blueberry anthocyanin samples according to a different amount of HCl concentration used for the extraction process. All treatments yielded entirely no significantly different results (). The extraction yield of anthocyanins reached the highest at 5.187 ± 0.845 mg/g when the amount of HCl was 0.02% (v/v).

Ethanol is preferably used in food, cosmetic, and pharmaceutical industries after classified with the GRAS (Generally Recognized as Safe) status by the US FDA [12]. An increase of ethanol concentration decreased the polarity difference between anthocyanins and solvent and then induced the reduction of other water-soluble constituents, such as polysaccharide and pectin, which prevented the dissolution of anthocyanins [9]. The effect of ethanol concentration on the extraction yield of blueberry anthocyanins is shown in Figure 1(b). The blueberry anthocyanin extraction yield tended to increase gradually in tandem with increasing ethanol concentration. When the ethanol concentration was 95%, anthocyanin extraction yield significantly increased and reached the highest at 4.511 ± 0.052 mg/g ().

The number of cavitation microbubbles increased as the extension of extraction time [13]. Nevertheless, long-time extraction could cause the degradation of anthocyanins [14]. The effect of extraction time on the extraction yield of blueberry anthocyanins is shown in Figure 1(c). The extraction time was extended from 0.5 h to 24 h, and the extraction yield of blueberry anthocyanin was not significantly increased in tandem with increasing extraction time. When the extraction time was 24 h, anthocyanin extraction yield reached the highest at 7.210 ± 0.458 mg/g. However, extraction time of 0.5 h also obtained a good anthocyanin extraction yield at 6.707 ± 0.277 mg/g ( vs. 24 h group). Considering the saving of time and energy, extraction time of 0.5 h was chosen in the single-factor experiments.

You et al. [15] extracted blueberry fruits with the extraction temperature controlled at 30°C to prevent the destruction of glycosidic bonds by high temperature, since the degradation rate of blueberry anthocyanin with the extraction temperature at 80°C was 36 times that at 40°C [16]. However, Figure 1(d) shows that there was no significant difference in the results among all the treatments (); when the extraction temperature was 30°C, the extraction yield of anthocyanin reached the highest of 6.166 ± 0.113 mg/g.

A high liquid-to-solid ratio could dissolve the constituents (anthocyanins and polyphenols) more effectively and thus increase the mass transfer rates and the extraction yields [17]. In this study, the liquid-to-solid ratios were extended from 3 : 1 to 20 : 1 v/w. Figure 1(e) shows that the extraction yield of blueberry anthocyanin significantly increased with increasing liquid-to-solid ratio (). When the liquid-to-solid ratio was 20 : 1 v/w, anthocyanin extraction yield reached the highest of 9.818 ± 0.229 mg/g.

In accordance with the results of preliminary single-factor experiments, the range and center point values of them are presented in Table 1.

3.2. Optimization of Blueberry Anthocyanin Extraction Yield

Table 2 shows the treatments with coded levels and the experimental results of blueberry anthocyanin extraction yield. The yield ranged from 1.668 to 16.210 mg/g. By applying the multiple regression equations to the experimental data, the response variable (Y) and the test variables are related to the following second-order polynomial equation:

Table 3 shows the ANOVA for the regression equation. The linear parameters (A and C) and the interaction parameter (AC) were significant (), indicating that the model used to adjust response variables was qualified to deputize the relationship between the response values and the independent variables. The lack of fit was used to verify the sufficiency of the model and was not significant (), indicating that the model could adequately fit the experimental data. The probability value of this model was highly significant (), indicating that the relationship between the blueberry anthocyanin extraction yield and the other independent variables was significant, the multiple regression equation was high, and the experimental error was small. The results revealed that UAE was an effective technique for blueberry anthocyanin extraction. A suitable ultrasonic time, the concentration of ethanol, and the liquid-to-solid ratio were important factors in the extraction of blueberry anthocyanins. RSM was successfully implemented to optimize the yield of anthocyanin extraction.

Figure 2 shows the response surface analysis of blueberry anthocyanin extraction. Figures 2(a) and 2(b) show the effect of the interaction between extraction time and ethanol concentration on the extraction yield of blueberry anthocyanins. An increase of extraction time with the same ethanol concentration has no significant effect on the extraction yield of anthocyanins, while an increase of ethanol concentration with the same extraction time also has no obvious effect on the extraction yield of anthocyanins. In a different study performed by Dibazar et al. [6], the interaction between extraction time (3–20 min) and ethanol concentration (20–100%) also had no significant effect on the yield of total anthocyanins content from Nova Scotia lowbush blueberry fruit (V. Angustifolium Aiton).

It can be seen from Figures 2(c) and 2(d) that the contours were elliptical and the slope of the response surface was steep, indicating that the extraction time has a significant interaction with the liquid-to-solid ratio on the extraction yield of blueberry anthocyanins. The response surface curve opened downwards and has the highest point. The maximum value corresponds to the respective maximum of the two factors. The model with the extraction time and liquid-to-solid ratio as independent variables was significant. Our experiment results showed that the single-factor extraction time and liquid-to-solid ratio and their interactions have significant influences on the extraction yields of blueberry anthocyanins. This result was slightly different from the experimental result performed by Chen et al. [9] in which the linear factors of extraction time (40–80 min) and liquid-to-solid ratio (15 : 1–25 : 1) also had a significant effect on the anthocyanin yield, but the interaction between extraction time and the liquid-to-solid ratio had no significant effect on the anthocyanin yield from chokeberry (Aronia melanocarpa) fruits.

Figures 2(e) and 2(f) show the effect of the interaction between ethanol concentration and liquid-to-solid ratio on the extraction yield of blueberry anthocyanins. An increase of ethanol concentration with the same liquid-to-solid ratio has no significant effect on the extraction yield of anthocyanins, while an increase of liquid-to-solid ratio with the same ethanol concentration has a slight effect on the extraction yield of anthocyanins and the change was not significant. This result was different from the experimental result performed by Zou et al. [18] in which the single-factor methanol concentration and the liquid-to-solid ratio had a highly significant effect on the extraction yields and the interaction between methanol concentration (30–70%) and liquid-to-solid ratio (15 : 1–25 : 1) had a significant effect on the yield of anthocyanins from mulberry fruit. However, in our single-factor experimental analysis, ethanol concentration also had a highly significant effect on the extraction yield.

The response surface optimization is a combination of statistical and mathematical analysis method based on multifactor and multilevel model design and experimental results to enhance the optimization of process conditions [19]. Through the analysis of the multivariate quadratic regression equation of the model, the best extraction conditions were as follows: the extraction time was 24 h, the ethanol concentration was 72.50% acidified with 0.02% (v/v) hydrochloric acid, the liquid-to-solid ratio was 20 : 1 ml/g, and the extraction temperature was 30°C. Under these optimum conditions, the yield of anthocyanins reached 16.21 ± 0.44 mg/g frozen-fresh weight, which was higher than the extraction rate under the previous nonsupreme extraction conditions. As comparison, the optimum conditions of anthocyanins extraction from Nova Scotia lowbush blueberry fruit (V. Angustifolium Aiton) including 60% methanol acidified with 1% (v/v) acetic acid, 50 ml/g liquid-to-solid ratio at 65°C for 11.5 min, yielded 13.22 mg/g frozen-dry weight as the maximum yield [6].

3.3. Identification of Blueberry Anthocyanins by HPLC and HPLC-ESI-MS

High-performance liquid chromatography (HPLC) has been the most widely used tool for the identification of anthocyanins, in which the individual anthocyanins can be separated by their polarity [20]. HPLC is commonly coupled with mass spectrometry (MS) detectors [21], which measure the mass-to-charge ratio (m/z) of individual ions, to get structural characterization by comparison with available published information of molecular ion and fragment ions which usually form a unique pattern [22]. Electrospray ionization (ESI) is the most successful interface used in HPLC-MS configuration [23]. Therefore, HPLC-ESI-MS combination provides an efficacious technique to identify the unknown components with high selectivity and sensitivity [24]. Here, the individual anthocyanins were identified mainly according to their retention times and MS spectra data [25]. As shown in Figure 3, the 13 different anthocyanins were detected in the extracts from rabbiteye blueberries cultivated in Nanjing.

Blueberry fruits are rich in anthocyanins, including malvidin, delphinidin, petunidin, cyanidin, peonidin, and pelargonidin, which were glycosylated by the sugar moieties of hexose (glucose or galactose) or pentose (arabinose) [26]. The determination of the molecular weights by HPLC-ESI-MS in Table 4 shows only five of the six most widespread anthocyanins in rabbiteye blueberry cultivated in Nanjing, namely, delphinidin (m/z 303), cyanidin (m/z 287), petunidin (m/z 317), peonidin (m/z 301), and malvidin (m/z 331). The derivate of pelargonidin was not found. The MS data showed the anthocyanins composed with either a hexose (indicated with Δ162 a.m.u.) or a pentose (indicated with Δ132 a.m.u.) [25].

Peak 1 had a similar mass spectra pattern ([M – H]+ 465.1 m/z) with peak 2, with the molecular weight of 303 m/z plus 162 a.m.u., indicating that these forms possibly were delphinidin derivatives. Based on their different polarities, peak 1 was tentatively identified as delphinidin-3-galactoside and peak 2 was tentatively identified as delphinidin-3-glucoside. The mass spectra of peaks 3 and 5 were both [M – H]+ 449.2 m/z, indicating that this form possibly was cyanidin derivative with the molecular weight of 287 m/z plus 162 a.m.u. Therefore, peaks 3 and 5 were identified as cyanidin-3-galactoside and cyanidin-3-glucoside, respectively. The mass spectrum of peak 4 was [M – H]+ 435.1 m/z with the molecular weight of 303 m/z plus 132 a.m.u., tentatively identified as delphinidin-3-arabinose. The mass spectra of peaks 6 and 7, [M – H]+ 479.1 m/z with the molecular weight of 317 m/z plus 162 a.m.u., indicated that they were petunidin-3-galactoside and petunidin-3-glucoside, while peak 9 was [M – H]+ 449.2 m/z, with the molecular weight of 317 m/z plus 132 a.m.u., which was identified as petunidin-3-arabinose. Peaks 8 and 10 were identified as peonidin-3-galactoside and peonidin-3-glucoside since the mass spectrum was [M – H]+ 463.2 or 463.1 m/z, with the molecular weight of 301 m/z plus 162 a.m.u. In addition, peaks 11 and 12 had the mass spectra [M – H]+ 493.1 m/z with the molecular weight of 331 m/z plus 162 a.m.u., while peak 13 was [M – H]+ 463.2 m/z with the molecular weight of 331 m/z plus 132 a.m.u., indicating that these forms were malvidin derivatives, including malvidin-3-galactoside, malvidin-3-glucoside, and malvidin-3-arabinose.

Malvidin derivatives had the highest level among all the anthocyanins in the tested blueberries in Nanjing, while the derivative of pelargonidin was not found, which has similarity with anthocyanins from the fruits of rabbiteye blueberry cultivar Tifblue cultivated in Japan [27] and juices made from Tifblue cultivated in USA [28]. However, this result was a little different from anthocyanins in the fruits of rabbiteye blueberry cultivar Delite cultivated in Brazil that consisted of glycosylates and aglycones of delphinidin, cyanidin, delphinidin, and pelargonidin, but the derivatives of petunidin and peonidin were not found in it [29]. These differences were probably because the composition of anthocyanins in blueberry fruits is dependent on their origin, location, and the preharvest environmental conditions.

Darrow’s species (V. darrowii) have the highest total anthocyanins and antioxidant capacity followed by rabbiteye (V. ashei), black highbush (V. fuscatum), and southern highbush (V. corymbosum) blueberry fruits [30]. This may be because Darrow’s blueberry fruits are bigger than rabbiteye and highbush blueberry fruits. The warm climate in Nanjing may also influence the anthocyanins content in rabbiteye blueberry fruits because anthocyanins as thermally sensitive compounds were easy to degrade while exposed to higher temperatures.

4. Conclusion

In this study, the ultrasound-assisted extraction method was developed for the extraction of anthocyanins from rabbiteye blueberry fruits cultivated in Nanjing. Response surface methodology was successfully employed to optimize the yield of anthocyanin extraction. Experimental results showed that extraction time, liquid-to-solid ratio, and their interaction have significant influences on the extraction yields of blueberry anthocyanins. The best extraction conditions were as follows: extraction time 24 h, 72.50% ethanol acidified with 0.02% (v/v) hydrochloric acid, the liquid-to-solid ratio 20 : 1 ml/g, and the extraction temperature 30°C. Under these optimum conditions, the yield of anthocyanins reached 16.21 ± 0.44 mg/g frozen-fresh weight. After purification, 13 anthocyanins were detected from the extract of rabbiteye blueberry fruits in Nanjing. The determination of the molecular weights by HPLC-ESI-MS showed that malvidin, delphinidin, petunidin, cyanidin, and peonidin, which were glycosylated by the sugar moieties of hexose (glucose or galactose) or pentose (arabinose) were tentatively identified in rabbiteye blueberries cultivated in Nanjing. The derivative of pelargonidin was not found. Among the anthocyanins, malvidin derivatives had the highest level. This result is slightly dissimilar with that of some other species of blueberries that were cultivated in other countries, which may differ in their origin, location, and the preharvest environmental conditions. This research provided a reliable scientific basis for further application of anthocyanins from blueberries as a functional food ingredient or nutraceutical.

Data Availability

The Microsoft Excel, PDF, and GraphPad Prism data used to support the findings of this study are included in the supplementary information files.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Authors’ Contributions

R. P. Hutabarat and Y. D. Xiao contributed equally to this work.

Acknowledgments

This research was supported by the Natural Science Foundation Program of Jiangsu Province (BK20161376), “333 Project” Training Funding of Jiangsu Province (BRA2018380), and the Special Scientific Research Fund of Jiangsu Academy of Agricultural Sciences (ZX(18)3008).

Supplementary Materials

“Figure 1(a)” contains the statistical calculation results of the anthocyanin content of HCl concentration experiment. “Figure 1(b)” contains the statistical calculation results of the anthocyanin content of ethanol concentration experiments. “Figure 1(c)” contains the statistical calculation results of the anthocyanin content of extraction time experiment. “Figure 1(d)” contains the statistical calculation results of the anthocyanin content of extraction temperature experiment. “Figure 1(e)” contains the statistical calculation results of anthocyanin content of liquid-to-solid ratio experiment. “Supplementary data_Excel” contains the calculation results of anthocyanin contents of single-factor experiments conducted on 5 factors, which are HCl concentrations, ethanol concentrations, extraction time, extraction temperatures, and liquid-to-solid ratio and also the calculation results of anthocyanin contents of 17 response surface optimization experiments; therefore, the maximum extraction result was obtained. “HPLC” contains the 4 pictures of the separated individual anthocyanin peaks from the blueberry anthocyanin extract performed by HPLC where they were detected at 280, 320, 360, and 520 nm. However, the data used to support our findings are those obtained at the detection at 520 nm. “HPLC-ESI-MS” contains the pictures of the separated individual anthocyanin peaks from blueberry anthocyanins extract and the mass-to-charge ratio (m/z) measurements of individual ions along with the tentative identification of each peak. (Supplementary Materials)