Abstract
Blueberries are one of the most popular and widely consumed berries, they are usually consumed fresh, dried or processed into juice. During the production of juices large quantities of berry press residues are generated, which are an industrial by-product and they accumulate as food waste. Optimal management of this industrial fruit by-product could help in utilising berry press residues with the purpose to valorise and add economic value. Blueberry press residues consist of skins and seeds which hold valuable compounds with antioxidant activity –polyphenolics. To facilitate the objective of bioactive compound release, the cell walls must rupture. Pectolytic enzymes could be used to degrade the structural polysaccharides of the berry skin cell wall, thus releasing the contents of the cells into the extraction medium. The physical characteristics of berry skins allows the use of pectolytic enzymes for the disruption of cells; however, the effects of enzymatic hydrolysis could be further improved using ultrasound. In this study the combined effects of enzyme and ultrasonic treatments were evaluated and optimised using the Response Surface Methodology approach in order to increase the release of polyphenolic compounds (especially anthocyanins). The optimised method was further tested to evaluate the application potential of enzyme and ultrasound treatment to prepare blueberry or lingonberry juice with increased antioxidant activity and more vibrant colour. The obtained results provide an optional method of berry press residue valorisation to produce higher quality juice or extract bioactive compounds from this type of food-waste.
Introduction
Consumption of berries, considering health benefits and excellent taste properties, is becoming increasingly popular worldwide. Growing popularity increases production and consumption of blueberries (Vaccinium corymbosum L.). Blueberries have excellent taste properties, attractive appearance, ease of transportation and good storage stability, balanced sweet-sour taste and high nutritional value [1]. Global production of blueberries has reached 682,790 tons (80.2 % in Americas, 18.2 % in Europe) [2]. Widely recognised are the various health-promoting properties of blueberries as their consumption can reduce risk of cardiovascular disease, cancer [3, 4]. Blueberry fruits have been used fresh, dried and as juice for treatment for gastrointestinal tract and diabetes, but herbal supplements containing these berries are available in the market as aids to improve vision, to treat diarrhoea, considering their antimicrobial, anti-inflammatory and antioxidant properties [5–7]. Biological and pharmacological activity of blueberries and bilberries and other Vaccinium species berries is associated with high concentrations of polyphenolics, especially anthocyanins and their capacity to scavenge oxygen and other radical species and thus reduce the oxidative stress [8, 9].
Significant amounts of fresh berries are processed into juices; however, this process leaves a waste product, berry press residues, which has limited application potential, despite being rich in phytochemicals. Press residues contain high concentration of polyphenolics, especially anthocyanins, which are the molecules that give berries their specific colour [10]. Blueberries and lingonberries (Vaccinium vitis-idaea L.) are berries with colourless flesh; however, the berry skin is brightly pigmented –anthocyanins are located in the vacuoles of the berry skin cells [11]. Conventional juice preparation methods are inefficient in terms of complete berry material utilisation; therefore, a large portion of the valuable polyphenolics remain in the press residues, which is considered a waste.
The structural polysaccharides that make up the plant cells can be degraded using pectolytic enzymes to aid the release of the cell contents into the surrounding medium [12]. Enzyme treatment has been used to increase anthocyanin and proanthocyanidin contents in wine with various degrees of success as well as other types of biomass [13–15]. In addition to the increased polyphenolic content of the juice, the addition of enzymes during the juice preparation steps help in the clarification and handling of the finished product [16]. The most commonly used enzymes are pectinases, which are used to increase juice yields, clarity, colour and aroma of apple [17], orange [18], pear [19] and grape juices [20].
Ultrasound assisted extraction is a process where the phenomenon of cavitation is employed in order to cause the rupture of plant cells, thus releasing its contents [21]. Ultrasound treatment has been used to extract a variety of plant materials –most applications emphasise the increase of phytochemical contents in the final product (extract) [22, 23].
Both described treatments have been used successfully for various purposes, however, only a few studies have evaluated the combination of the two treatments. Effects of ultrasound assisted enzymatic extraction were evaluated for preparation of mulberry (Morus nigra) juice, where the optimised extraction conditions (by Response Surface Methodology - RSM) provided significantly higher contents of phytochemicals than untreated juice [24]. The combination of enzyme and ultrasound treatments was also tested to prepare wine must. Commercially available enzymes were tested, and the results showed that anthocyanin contents, proanthocyanidins and total soluble solids increased significantly [25] improving the appearance (colour intensity increased by 18%) [11] of the must.
The aim of this study was to test and optimise enzyme and ultrasound assisted extraction variables from blueberry press residues and provide a novel applicable approach for the juice preparation in order to increase radical scavenging potential, total polyphenolic and anthocyanin contents in berry juice. Enzyme treatment effect on stability of individual anthocyanins and their concentration in juice preparations was evaluated. Blueberry, and other berry juices, is an increasingly popular drink in Northern Europe and as the production increases also the produced food-waste increases. The berry press residues have high-added value in terms of phytochemical contents; therefore, it is important to maximise the utilisation of this juice production side-product, preferably eliminating the need of additional processing steps.
Materials and methods
Plant material
Blueberries (Vaccinium corymbosum L.) and lingonberries (Vaccinium vitis-idaea L.) used in this study were harvested in the summer of 2020. Blueberries (cv. ‘Blue Crop’) were harvested in a commercial farm (Z/S “Strēlnieki”) located on the outskirts of Jūrmala City, Latvia. Lingonberries were harvested in the forests surrounding town of Engure, Latvia. Berries were frozen and kept at -20°C until further use.
Blueberry press residues were prepared by thawing the frozen blueberries at room temperature. Blueberries were homogenised using porcelain mortar and placed into a hydraulic juice extractor (BioWin, Łódź, Poland) where pressure was applied to the berry biomass until all the juice was drained. Resulting blueberry press residues were immediately frozen in air-tight zip-lock bags at –20°C, a sample was taken to determine the moisture of prepared press residues (moisture content of blueberry press residues 73.1%). Enzyme and ultrasound treatments were performed using wet press residues to mimic industrial processes, the results were expressed per 100 g of dry press residues, which have been calculated using the previously determined moisture content.
Blueberries and lingonberries used for experiments for juice yield determination were slightly thawed at room temperature and homogenised using blade homogenizer (IKA, Staufen, Germany). Homogenised berries were used for the experiments immediately after the homogenisation process.
Enzyme and ultrasound treatment
Berry biomass treatment was performed using berry press residues or fresh, homogenised berries. All of the performed extractions were done using 5.0 g of berry biomass weighed out in 100 mL glass extraction bottles. The concentration of enzyme used was 0.5 mL, which was kept constant for all of the experiments. The volume of demineralised water (50 mL) used for the enzyme treatment was kept constant. Ultrasound treatment of samples was done using a 300 W (100 kHz) water-cooled ultrasound bath (Cole-Parmer, Chicago, IL, USA).
Single-factor experiments
Single-factor experiments were performed in order to identify the variables affecting the extraction yields. The variables were tested at two levels –pH 2 and 5; no ultrasound and 15-minute ultrasound treatment; addition of 25% and 50% ethanol; incubation of enzymes at room temperature (21°C) and at 50°C. Each of the experiments were conducted separately by changing only the specific variable, while keeping the rest of the extraction parameters constant.
RSM optimisation experiments
Exact conditions of samples prepared for the RSM optimisation can be found below (Experimental design and statistical analysis) and the experimental design can be found in in Supplementary Table 1. RSM optimisation of extraction variables was done by 1. adjusting the pH of extraction medium; 2. addition of enzyme (0.5 mL) to the sample; 3. sample incubation with the enzyme; 4. addition of ethanol; 5. ultrasound treatment.
Application experiments
Optimised extraction conditions were tested using 50 g whole blueberries and lingonberries (thawed and crushed in a mortar). The homogenised berries were transferred to a 100 mL extraction bottle and the optimised treatment conditions to retrieve polyphenolics (Table 6) were applied to measure TPC contents and DPPH activity and the optimised anthocyanin (Table 5) extraction was used to prepare samples for UPLC-PDA analysis. The extraction parameters were applied in the same order as in the RSM optimisation experiments. Samples were prepared in triplicate.
Optimised extraction conditions of polyphenolics from blueberry press residue using enzyme and ultrasound treatment
Optimised extraction conditions of polyphenolics from blueberry press residue using enzyme and ultrasound treatment
Optimised extraction conditions of anthocyanins from blueberry press residue using enzyme and ultrasound treatment
Prepared samples were filtered through a Whatman (Maidstone, United Kingdom) qualitative filter paper and stored at 4°C until further analysis, which were conducted no later than within 48 h.
The enzymes used for the treatments were pectinase or cellulase enzyme mixtures produced and kindly donated by Biocatalysts Ltd. (Cardiff, United Kingdom) Enzymes used were chosen based on manufacturer suggestions and their designed uses for industrial production of apple, vegetable and grape juice, which are similar types of biomass as blueberries. Optimisation of enzyme and ultrasound treatment was done using Enzyme 1 (Table 1).
Summary of used enzymes for optimisation of combined treatment variables and application tests
ND –not defined by the manufacturer.
Determination of total anthocyanins in the prepared extracts was done according to a previously modified pH –differential method [8]. 0.025 M potassium chloride buffer solution (pH 1.0) and 0.4M sodium acetate buffer solution (pH 4.5) were prepared adjusting the pH with concentrated HCl. 0.5 mL sample or its dilution was added to 5 mL of both buffers and after 30 min incubation in the dark, the solutions were measured at 520 and 700 nm against a buffer blank using a UV-VIS spectrophotometer (Shimadzu, Kyoto, Japan). The total anthocyanin concentration was calculated using (Equation 1):
where A = (A520 nm –A700 nm) pH 1.0 –(A520 nm –A700 nm) pH 4.5; MW (molecular weight) = 449.2 g mol-1 for cyanidin-3-glucoside; DF = dilution factor; l = cuvette path length in cm (1 cm); ɛ= 26 900 molar absorptivity, in L mol –1 cm–1 for cyanidin-3-glucoside.
Quantification of total polyphenolic was performed using a modified Folin –Ciocalteu method [8]. 5 mL of 0.2 M Folin –Ciocalteu (FC) reagent (Sigma Aldrich, Darmstadt, Germany) were transferred to the reaction tube, 1 mL of sample or its dilution was added to the FC, mixed and incubated for 5 min. After the incubation 4 mL of NaCO3 solution in water (7.5%, w v-1 %) were added and mixed, incubated for 30 minutes in the dark. The resulting sample was measured in a UV-VIS spectrophotometer at 715 nm. Samples measured against reagent blank. Gallic acid standard solutions were prepared in the concentration range 0 –0.5 mg mL-1 and measured in the same manner as the samples. Concentration of total polyphenolics was calculated using linear regression equation and expressed as total polyphenolics in gallic acid equivalent (GE) per 100 g of dry berries, if not stated otherwise.
Determination of antiradical scavenging activity
Determination of antiradical scavenging activity was done using a modified 2,2-diphenyl-1-picrylhydrazyl (DPPH, Sigma Aldrich) free radical determination method [26]. 10 mg of DPPH were weighed out and dissolved in 250 mL 80% methanol. The absorption of the prepared solution was 0.95–1.10 AU at 517 nm. 5 mL of the prepared DPPH solution were transferred to the reaction tube and 200μL of the sample or its dilution was added. The reaction tube was vortexed and placed in the dark for 30 minutes or until a stable absorbance measurement could be taken. Samples were measured against a 80% methanol blank using a UV-VIS spectrophotometer. A Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid) standard curve was prepared in the concentration range 0.1 –0.6 mg mL-1. DPPH antiradical scavenging potential was expressed as mg Trolox equivalents (TE) per 100 g of dry berries if not stated otherwise.
UPLC-PDA analysis of anthocyanins and anthocyanidins
Ultra–performance liquid chromatography (UPLC) quantitative and qualitative analyses of anthocyanins and anthocyanidins were carried out using a Waters (Milford, MA, USA) ACQUITY UPLC system equipped with a Quaternary Solvent Manager (QSM), a Sample Manager –Flow-through Needle (cooled to 4°C) (SM–FTN), a column heater (CH–A) and a photo–diode array (PDA) λ detector. Data was collected using Waters Empower software. Analyses were carried out at 35°C using a C18 column (Acquity UPLC BEH C18 2.1×150 mm i.d., 1.7μm) with a column pre–filter (Frit and Nut 0.2μm, 2.1 mm). The mobile phase consisted of aqueous 5.0% formic acid (A) and methanol/1.0% formic acid in water (70:30 v v-1) (B). The flow rate was 0.250 mL min-1, and the gradient elution was from 80% to 75% of solvent A in 15 minutes, from 75% to 60% in 7 minutes, and from 60% to 0% in 18 minutes, followed by 10 min. of stabilisation at 80%. The total sample run time was 40 minutes, injection volume - 2.0μL. Quantification of anthocyanins was done using external calibration curves prepared from cyanidin 3-O-arabinoside chloride and cyanidin chloride (Extrasynthese, Genay, France) standard mixtures (3–100 ppm; R2 >0.9990). The limit of detection (LOD) and limit of quantification (LOQ), respectively defined as a 3:1 and 10:1 peak-to-noise ratio, were 0.75μg mL-1 and 2.20μg mL-1 respectively.
Experimental design and statistical analysis
Variables to be optimised were firstly identified by performing experiments where a single factor was changed at a time. Preliminary range of extraction variables were identified –incubation temperature, length of the ultrasound treatment, amount of added ethanol before ultrasonication, pH, incubation time. The identified extraction parameters were further optimised using Response Surface Methodology (RSM).
A five factor and three level central composite design (CCD) was performed (with three central points), allowing to determine the combined effects of the independent variables on the response. Possible effects of unexplained variability due to extrinsic factors in the two observed responses (total polyphenolics, total anthocyanins) were minimised by randomising the run (experiment) order. The range and the levels of the variables chosen based on the preliminary results are summarised in Table 2.
Independent variables and their levels used in the response surface design
Independent variables and their levels used in the response surface design
The experimental design of CCD consisted of 56 experimental points with 4 central points (28 experiments with 2 repetitions, Supplementary Table 1). The data obtained for RSM was fitted to a second-order polynomial model with a generalised second order (quadratic) model (Equation 2):
where β0 is the regression coefficient for intercept, βj is the regression coefficient for linear terms, βjj is the regression coefficient for quadratic terms, βij is the regression coefficient for interaction terms, and the independent variables are shown as x i and x j . Experimental data was first fitted to test the obtained model and the resulting prediction values were further used to generate a tri-dimensional regression models. The RSM design and optimal extraction conditions were determined by maximising desirability using a response profiler. All the statistical analysis and experimental design were created and tested using SAS JMP® software (Cary, NC, USA). The obtained fit statistics of the generated quadratic regression models were validated based on the determination coefficients by ANOVA and resulting P-values, summary of fit statistics can be found in Table 3. ANOVA with post-hoc Tukey’s HSD was used to compare means between different treatments. Paired t-test was used to determine statistical differences before and after the sample treatment.
Analysis of variance (ANOVA) for the response surface model on the total polyphenolic and anthocyanin yield from blueberry press residues
Preliminary single-factor experiments
Identification of the most important variables that have the most influence on the final yield of either total polyphenolics or anthocyanins was done by altering single factor in the enzyme treated samples. Sample amount (5.0 g), amount of enzyme (0.5 mL) and extraction solvent (50 mL water) were kept constant, as these variables are stated as optimal by the producer of the enzyme. Influence of pH, amount of added ethanol after incubation with enzyme, duration of ultrasound treatment and incubation temperature were tested with low and high variable values.
The tested treatment conditions revealed that pH of the incubation medium is an important factor –lower pH values (pH 2) produces significantly higher TPC, ACN values and thus the DPPH free radical scavenging activity also increase (Fig. 1A). A relation between the TPC and DPPH increase was observed, indicating that polyphenolics are likely the substances responsible for this activity. Similar relations have been observed for polyphenolics from different sources [27], for example, grapes [28], green tea [29], elderberries [30]. Ultrasound treatment significantly increased the TPC and ACN yield, although DPPH activity did not increase, this indicates that ultrasound helps in release of polyphenolics that have low antiradical scavenging activity (Fig. 1B). Addition of ethanol during the ultrasonication step after the enzyme treatment significantly increased the TPC yield (from 1.48 to 1.86 g 100 g-1, 20% increase) (Fig. 1C). Increasing incubation temperature provided positive effects for the extraction of TPC while ACN concentration did not change as compared to the control sample (Fig. 1D). The DPPH value increased without the increase of ACN concentration, which indicates that other groups of polyphenolics have higher antiradical activity than anthocyanins.
Single factor experiments where (A) influence of pH (B) duration of ultrasound treatment (US) (C) addition of ethanol (EtOH) and (D) incubation temperature was tested to increase DPPH antiradical scavenging activity, total anthocyanin yield (ACN) and total polyphenolics yield (TPC). DPPH values represent g TE 100 g-1 dry press residues. Error bars represent measurement standard deviation (n = 3). Paired t-test was applied with p-values: ns - not significant; *0.05; **0.01.
Performed experiments revealed effects of various parameters on the ACN and TPC yields. The identified parameters were chosen for further optimisation using RSM, the value range of the variables can be found in Table 1. In addition to the above variables, also the incubation time was optimised as it has been previously recorded that prolonged exposure to oxygen and temperature can degrade several groups of polyphenolics [31].
Model fitting
In order to examine the combined effects of the previously chosen extraction variables using RSM, a CCD of 56 runs with 4 centre points was performed randomly. A summary of ANOVA of the used design can be found in Table 3 –analysis of variance was performed in order to determine if the obtained quadratic model was significant. The obtained F- and P- values for the model were significant. The obtained F- and P- values for the lack-of-fit show significant fit which indicates that the model prediction equation is adequate for predicting the total polyphenolic and anthocyanin yields under any combination of variable values (Table 3). The adjusted R2 and total correlation R2, 0.9607, 0.9225 and 0.9970, 0.9994 for the models of total polyphenolics and anthocyanins, respectively, indicate adequate accuracy and general availability of the designed polynomial models (Table 3).
Variables and their interactions were investigated for their significance on the total polyphenolic and anthocyanin yields (Table 4). The linear coefficients X2, X3, X4 were significant (P < 0.05) in both models, however, the X1 was significant only in the total polyphenolics model, while X5 was significant in the total anthocyanin model (Table 4). Significance of the variable interaction pairs is summarised in Table 4.
The effects tests of the predicted quadratic polynomial models for the two investigated responses –total polyphenolics and total anthocyanins
The effects tests of the predicted quadratic polynomial models for the two investigated responses –total polyphenolics and total anthocyanins
∗Significant at α= 0.05.
Response surface methodology was used to examine the effects of the used variables on the yield of total extractable anthocyanins. Anthocyanins have differing colour depending on the pH of the environment, they can degrade when exposed to elevated temperatures and oxidise when exposed to oxygen for prolonged periods of time [32]. The extraction of anthocyanins is an intricate process, where a variety of factors must be accounted for, therefore it is crucial to evaluate the optimal conditions of anthocyanin release from food-waste biomass, such as blueberry press residues.
Extraction variables (Table 5) and their combinations (Table 4) were optimised to maximise the desirability of anthocyanin yield. The obtained optimal conditions provide 98.9% desirability (Table 5). The minimal values obtained provided 166 mg anthocyanins 100 g-1, while the conditions of maximised desirability provided 1103 mg 100 g-1 blueberry press residues. Anthocyanin yield at minimal conditions is only 15% of what could be obtained at the optimised conditions. The duration of ultrasound treatment after the incubation with enzymes increased the extraction yield of anthocyanins –enzyme treatment liquifies the berry skins and the cavitation caused by the ultrasound waves further breaks down the cells, thus releasing the cell contents into the extraction medium (Fig. 2A). Stability of anthocyanins depends on the pH of the medium, therefore the lower the pH used for the extraction, the higher the anthocyanin yield. The performed optimisation considered the lowest pH value of the extraction medium to be pH 2, however, according to previous research by other authors, the true optimum of maximum anthocyanin stability can be reached in the range 0–1.5 pH [33]. The anthocyanin flavylium salts are more stable in highly acidic environment, by increasing the pH the anthocyanin molecule loses a proton turning into quinoidal base, which is able to bond with water forming a colourless compound chromenol. The pH optimum for anthocyanin stability is outside the range of the prepared RSM model, however, in this specific scenario, where enzymes are used to aid the extraction, also the enzyme stability must be considered, as low pH can degrade the used enzyme [34]. Instability of anthocyanin molecules can be observed as the increase of incubation time and temperature significantly decreases the anthocyanin yield (Fig. 2B, 2C). The optimal temperature for the enzyme activity, as stated by the manufacturer, is 50–60°C, however, for the optimal extraction of valuable anthocyanins, it is important to keep the temperature as low as possible in order to avoid degradation of various groups of phytochemicals [31]. Ethanol was added to the samples before the sonification in order to increase the solubility of anthocyanins, the identified optimum in this study was 50%, with indications that the true optimum lays outside the tested range, however, adding more ethanol to the sample (water phase) will dilute the sample and significantly increase the volume, requiring further concentration steps for sample analysis. In a previous study the optimal ethanol concentration for anthocyanin extraction was identified to be 40%, however, the pH of the used extraction solvent was lower, since trifluoroacetic acid (TFA) was used [8]. It has also been previously identified that the enzyme treatment temperature has a prominent role on the release of anthocyanins –higher temperatures render the breakdown of the cell wall matrix, making the anthocyanins more accessible for the extraction [35]. However, by further increase of temperature the anthocyanins are a subject to thermal degradation, leading to lower extraction yields [36].
Three-dimensional response surface graphs showing the effects of the duration of ultrasound treatment (X1), ethanol addition (X2), incubation temperature (X3), pH (X4) and incubation time (X5) on the total anthocyanin extraction yield from blueberry press residues. (A) effects of X1X3, (B) X4X5, (C) X3X5 and (D) X2X5.
In the juice production process, it is important to increase the anthocyanin concentration in juice, as consumers are drawn towards more brightly coloured, clear, and natural juices. Addition of ethanol increases the solubility of anthocyanins (Fig. 2D) –such approach should be used when investigating anthocyanins in various berries or different types of biomass, as complete extraction of the analysed molecules allows more repeatable and precise results. Anthocyanins are found in the juice as well as the skins of berries, therefore it is crucial to ensure extraction of both types of tissue, as this can have major impact on the analysis of anthocyanin profiles and possibly provide false information on the quality of material [37, 38].
Blueberry press residues contain various groups of polyphenolics, for example, flavonols, flavanols, flavones, flavanones and others. While anthocyanins provide colour to the juice, which is an important factor for consumer attraction, they also have antiradical activity. Other groups of polyphenolics have even greater ability of free-radical scavenging, therefore optimisation of blueberry press residue extraction using enzyme and ultrasound treatment could provide products that have not only better visual appearance, but also increased health benefits. Polyphenolics are found in the vacuoles of the plant cells and during maceration and cold/hot-pressing they can be released only partly with the majority of these bioactive compounds remaining in the berry skins. To facilitate the objective of bioactive compound release, the cell walls must rupture. Pectolytic enzymes could be used to degrade the structural polysaccharides of the berry skin cell wall, thus releasing the contents of the cells into the extraction medium. The physical characteristics of berry skins allows the use of pectolytic enzymes for the disruption of cells; however, the effects of enzymatic hydrolysis could be further improved by the use of ultrasound. Ultrasound causes physical and chemical changes due to the phenomenon of cavitation; it is the formation and collapse of bubbles produced by the ultrasonic waves. The formed bubbles implode near the cell walls and causes the disruption of the cells, which improves release of the phenolic compounds. Combination of the two treatments significantly increases the value of juice in the terms of polyphenolic contents (Fig. 3).
Three-dimensional response surface graphs showing the effects of the duration of ultrasound treatment (X1), ethanol addition (X2), incubation temperature (X3), pH (X4) and incubation time (X5) on the total polyphenolic extraction yield from blueberry press residues. (A) effects of X4X5, (B) X3X5, (C) X1X2 and (D) X2X3.
Compared to the optimal conditions of anthocyanin extraction, polyphenolic extraction requires longer ultrasound treatment and incubation time –this indicates that the polyphenolics are bound and incorporated into the cell membranes, as more cell degradation is necessary. The frequency of ultrasound used can also have significant influence on the enzyme activity –low sonication frequency can possibly increase activity of pectinase and other enzymes that can disintegrate cell membranes thus releasing polyphenolic compounds [39, 40]. Low frequencies of ultrasound (20–40 kHz) generate large cavitation bubbles which ensure more violent cellular degradation, thus increasing solvent penetration and extraction rate [41]. The optimal extraction length was identified to be at the maximum (30 min) of prepared model, indicating that the true optimum is outside of the optimisation range –to decrease the extraction time and increase extraction yield, the use of ultrasound at different frequencies should be tested. Optimised polyphenolic extraction provides 99.2% desirability (Table 6), when the extraction variables are maximised simultaneously. The minimal values obtained provided 0.45 g polyphenolics 100 g-1, while the conditions of maximised desirability provided 2.45 mg 100 g-1 blueberry press residues. Polyphenolic yield at minimal conditions is 19% of what could be obtained at the optimised conditions. Similarly, to anthocyanin extraction, also polyphenolics favour lower, more acidic extraction medium (Fig. 3A). Exposure to higher temperatures has effect on the extraction yield, however, it was not as pronounced as for the anthocyanins (Fig. 3B). The slightly less polar solvent, ethanol, when added to the extraction after the enzyme treatment significantly increases the extraction yield of polyphenolics (Fig. 3C, 3D). Both the pH and ethanol concentrations were at the maximum of the used RSM model, indicating possibility to increase the extraction yield. As previously reported the optimal ethanol concentration for polyphenolics extraction was identified to be 57% with addition of 0.9% TFA, however, the stability of used enzyme preparations must be considered as well as prolonged exposure of polyphenolics to avoid pH induced degradation [8, 42]. Approach where ethanol and enzyme treatments are combined could lead to creation of new products, as large molecular polyphenolics (for example procyanidins) could be released from the cell-wall membranes, as well as possibly new substances, previously not found in blueberries, could be identified.
Optimised treatment conditions were used to extract juice from blueberries and lingonberries in order to increase the juice yield as well as the total polyphenolic content and antiradical activity. Control sample 1 was a non-treated sample, while control sample 2 was a sample treated with ultrasound. Obtained results show that the juice yield of both berries increased when ultrasound treatment was used, however, it was not significant. When an enzyme was used in combination with the ultrasound treatment, significantly higher juice yields were obtained (Table 7). Also, the dry matter of the juice increased when a combination of treatments was used, indicating release of soluble compounds into the juice. Total polyphenolics and antiradical activity increased based on the used enzyme (Table 7). Larger increase, compared to control, was observed in lingonberry juice (up to 22%). This increase could be attributed to the lower pH of the juice for lingonberry (pH 3.03) than for the blueberry (pH 4.12), as it was previously concluded, that lower pH increases the activity of enzyme used as well as the stability of polyphenolic compounds. In a study where the activity of different enzymes was evaluated, it was concluded that lower pH (2-3) and lower incubation temperature significantly increases enzyme activity [43]. Enzyme 1 showed the best results for juice production from blueberries, while enzyme 5 was more effective for processing of lingonberries (Table 1, 7). Another aspect of juice production is the visual appearance of the juice –by using the combination of the two treatments it was possible to obtain clarified, vibrant, aromatic juice. Non-treated lingonberry juice was a viscous slurry, which is difficult to press and filter, however, enzyme treatment solves these issues. To possibly increase the juice and TPC yields from blueberries, more acidic juice (for example lingonberry) could be added to the enzyme-treated biomass, in order to lower the pH. The high concentration of benzoic and other organic acids [44] in lingonberry juice lowers the pH which in turn increasing the enzyme activity, simultaneously working as natural preservatives of the juice.
Application of combined enzyme and ultrasound treatment for juice production from lingonberries and blueberries
Application of combined enzyme and ultrasound treatment for juice production from lingonberries and blueberries
Connecting letters next to the TPC and DPPH values represent significant differences between the measurements (ANOVA, Tukeys HSD test, n = 3, α= 0.05).
Anthocyanin stability after the enzyme/ultrasound treatment was evaluated using UPLC-PDA analysis. In total 19 (Supplementary Table 2) and 8 (Supplementary Table 3) individual anthocyanins were identified in the juice samples in blueberry and lingonberry, respectively.
Analysis of individual anthocyanins revealed that some of the present anthocyanins degrade due to certain enzymes. Cyanidin galactoside in lingonberry was found in both control samples, ultrasound treatment did not influence the anthocyanin contents, however, treatment with Enzyme 1, 2 and 5 significantly reduced the concentration of this specific anthocyanin. Enzyme 3 and 4 showed similar anthocyanin profiles as the untreated and ultrasound treated control samples, which indicates that these enzymes do not degrade anthocyanins, also the total anthocyanin concentrations are comparable to the control samples (Fig. 4A). Blueberry anthocyanins, depending on the used enzyme, showed similar pattern of anthocyanin degradation –Enzymes 3 and 4 showed comparable profiles as those of control samples, while Enzymes 1,2 and 5 showed degradation of certain anthocyanins. Significant decrease of malvidin galactoside was observed after the enzyme treatment (Fig. 4B). In lingonberries, as one of the main anthocyanins petundin-3-O-glucoside was identified. Slight changes in the concentrations of this anthocyanin were observed depending on the used enzyme, however, in a study where enzymes have been used to extract anthocyanins (cyanidin-3-O-glucoside, petunidin-3-O-glucoside) from Prunus nepalensis L. no change in the concentration was observed due to the degradation of anthocyanins by the enzymes used [35].
Differences in the relative concentration of main anthocyanins in (A) lingonberry and (B) blueberry juice after the enzyme and ultrasound treatment using the optimised juice extraction parameters. Error bars represent standard deviation. Connecting letters next to the Total amount of anthocyanins (mg L-1) represent significant differences between the sample measurements (ANOVA, Tukeys HSD test, n = 3, α= 0.05).
A pattern was observed where certain enzymes degraded the anthocyanins present in the berry juice (Fig. 4). Relative amounts of certain anthocyanins significantly decreased (75% to 8% in lingonberry) as well as total amount of anthocyanins (non-treated juice 1353.2 mg L-1; treated juice 277.6 mg L-1). Based on the observed changes it was assumed that these changes depend on the type of enzyme used (Table 1). Various factors can have influence on the anthocyanin (and other polyphenolic) stability. For example, in agreement with the findings of this study, prolonged incubation time can degrade certain groups of polyphenolics [45] as well as increased concentration of the enzyme used, due to rutinase activity [46]. As stated by the manufacturer pectinase activity of Enzymes 1, 2 and 5 were the main activities, with defined side-activities. Enzyme 1 is a combination of enzymes with pectinase and cellulase activities. Enzymes 3 and 4 showed comparable results based on the juice and TPC yields, additionally, the anthocyanins were not degraded. Enzymes 1 and 5 showed higher extraction potential of total polyphenolics (Table 7), while enzymes 3 and 4 were more suitable for the extraction of anthocyanins (Fig. 4). With the available information it is not possible to conclude what type of enzyme activity leads to the degradation of anthocyanins in berry juice, as the commercially available enzymes are often mixtures of various enzymes that possess both activities and the exact composition of these products is not disclosed.
Results presented in this study showed an approach where a combination of enzyme treatment and subsequent ultrasound treatment was used to maximise the extraction of valuable antioxidants from berries in order to increase the value of juice for human consumption. Response surface methodology was used to successfully optimise five extraction variables simultaneously (incubation time, incubation temperature, pH, addition of ethanol, duration of ultrasound treatment). Optimised berry treatment conditions were used to demonstrate application potential for preparation of juice with higher total polyphenolics yield and thus having higher antiradical scavenging activity. The performed experiments revealed specific anthocyanin degradation by certain pectolytic enzymes. A combination of enzyme treatment and further ultrasound treatment provides a novel approach for more efficient extraction of valuable phytochemicals. While the optimised conditions were shown to provide high efficiency, it has to be noted, that the conditions must be optimised for specific type of biomass used and the effects of the used treatments (choice of enzyme) must be meticulously evaluated on various extraction responses to provide consumers with high quality functional products. Juice industry generates large amounts of valuable food wastes that contain various phytochemicals. The proposed juice processing technique could lead towards zero-waste production, where the biomass in question, using innovative approaches, would be fully utilised without the production of food-waste.
Footnotes
Acknowledgments
In accordance with the contract No. 1.2.1.1/18/A/002 between “Latvian Food Competence Center” Ltd. and the Central Finance and Contracting Agency, the study is conducted by “Pharmeko Lettland” Ltd. with support from the European Regional Development Fund (ERDF within the framework of the project “Latvian Food Industry Competence Center”. Work of LK was supported by the patron “Mikrotīkls” Ltd. administered by the Foundation of University of Latvia.
Conflict of interest
The authors have no conflict of interest to report.