Development of blueberry (Vaccinium corymbosum L.) waste powder as a potential food ingredient with functional properties

Abstract

Salta province, northwestern Argentina, produces blueberries for export and discards fruits with a potential quantity of bioactive compounds. These bioactive compounds have health-promoting properties that prevent or delay the appearance of chronic diseases. This study aimed to formulate blueberry microcapsules using discarded fruit, to determine and evaluate the effect of spray-drying and lyophilization on the bioactive compounds and their physical properties. Fourteen capsule prototypes were obtained by applying a randomized full factorial design with two factors: type of drying and type of wall material. The former factor had two levels (spray-drying and lyophilization) and the latter had three levels, each with defined quantities to be used, namely maltodextrin (0%, 10%, 15%, and 30%), gum Arabic (0%, 10%, 15%, and 30%), and modified starch (0%, 10%, 15%, and 30%). Spray-drying, lyophilization, total polyphenols, anthocyanins, proanthocyanidins, antioxidant activity, and the physical properties of the microcapsules were analyzed using ANOVA, PCA, and cluster analysis. Results showed significant differences between the two processes (P < 0,05), with lyophilization being better at preserving bioactive compounds. The PCA test also showed a positive association between lyophilization and bioactive compounds, while spray-drying powders were related to negative characteristics, like moisture and water activity.

Introduction

There is a growing trend in consumer demand for fruits, due to their good taste and health-promoting properties, which prevent or delay the appearance of chronic diseases like different types of cancer, diabetes, stroke, and obesity. This function is attributed to the bioactive compounds present in fruits. These substances efficiently interact with proteins, DNA, and other biological molecules, causing a protective effect on cells and can, therefore, be used for designing natural therapeutic agents (Schmidhuber 2019).

According to data from FAO and SAVE FOOD organizations, 20% of dairy products, 30% of cereals, 35% of fish, 20% of meat, and 20% of fruit, and 45% of vegetables are lost. Food waste refers to good-quality food that is fit for human consumption but is not consumed because it is discarded at the end of the distribution and consumption chain (Lipinski et al, 2013).

Salta Province, Argentina, produces blueberries for export and discards a large quantity of the fruit mainly because of poor size and shape, although berries have adequate organoleptic (color, aroma, texture) and microbiological attributes. This underutilization can be attributed to poor consumption and the lack of knowledge about the benefits of berries among the population in the region. Blueberries are small fruits with a high content of polyphenols, like anthocyanins, pigments from the flavonoid group with antioxidant power responsible for their purple and blue color. However, their industrial valorization is limited due to the instability of environmental factors, such as pH, temperature, light, metallic ions, enzymatic action, oxygen, ascorbic acid, sugar and their degradation products, proteins, and sulfur dioxide (Wilkowska et al. 2016). In order to increase the stability of these compounds, the most common current technique is microencapsulation by spray- and freeze-drying. Both dehydration processes are based on the entrapment of an ingredient in a coating material like polysaccharides, gums, gelatins, and proteins (Lee and Wong 2014). It is a technological strategy used for trans.

forming discarded fruits by dehydration methods and for developing by-products with healthy properties and added value. The aims of this study were to formulate blueberry microcapsules using discarded fruit, to evaluate the effect of spray-drying and lyophilization, and to determine and analyze the bioactive compounds and their physical properties.

Materials and methods

Discarded Blueberries, Emerald variety, in perfect conditions of quality from Extraberries S.A, Metán, Salta, Argentina, were stored in a freezer at −18 ± 2 °C in airtight bags of 250 ± 5 g capacity.

Blueberry juice

Berries were thawed at room temperature and washed by immersion with drinking water. Then, they were processed with a Braun Minipimer for 15 ± 1 min, and the pomace was filtered through a metal mesh (60). The juice obtained was mixed with seven wall materials at a maximum concentration of 30%. The formulations are indicated in Table 1.

• Encapsulation Procedure: Fourteen capsule prototypes were obtained by applying a randomized full factorial design with two factors: type of drying and type of wall material. The former factor had two levels (atomization and lyophilization) and the latter had three levels, each with defined quantities to be used (maltodextrin (0%, 10%, 15%, and 30%), gum Arabic (0%, 10%, 15%, and 30%), and modified starch (0%, 10%, 15%, and 30%).

  Two processes were carried out to obtain microcapsules. Spray-drying was performed in Buchhi Spray at an inlet temperature of 140 ± 2 °C and an outlet temperature of 65 ± 2 °C. Lyophilization was conducted in a Rificor lyophilizer. Each sample was previously frozen in an ultrafreezer (Righi) at −80 ± 2 °C for 30 min and then desiccated under vacuum for 16 h until room temperature. Finally, the samples were milled in a grinder. Both powders were stored in amber-colored containers until analysis.

• Bioactive compounds (BAC) for BAC analysis: 1 ± 0.01 g of powder was weighed in an Erlenmeyer flask with 10 mL of absolute ethanol and kept for 1 h at 30 ± 1 ºC in a thermostatic bath with constant agitation. Then the BAC extract was filtered and leveled to 25 mL with distilled water and stored at 2–8 ºC. The BAC were measured in a UV VIS Genesis spectrophotometer.

  • Total Anthocyanins (TAC) were determined using the pH-differential method by Giusty Wrolstald (2000). An aliquot was diluted (1:2 v = v) with two buffers, pH 1.0 (0.025 M potassium chloride) and pH 4.5 (0.4 M sodium acetate). After 15 ± 1 min of incubation at room temperature, the absorbance was measured at 510 and 700 nm. The TAC were expressed as cyanidin-3-glucoside/100 g, according to Eqs. 1 and 2:

    $$\text{Absorbance}\left(\text{A}\right)={\left({\text{A}}_{510\text{nm}}-{\text{A}}_{700\text{nm}}\right)}_{\text{pH1}.0}-{\left({\text{A}}_{510\text{nm}}-{\text{A}}_{700\text{nm}}\right)}_{\text{pH4}.5}$$ 1

    $$\text{Total Anthocyanin}\left(\text{mg}/100\text{g}\right)=\text{A x Mw x DF x 1}00/\left(\epsilon \text{x 1}\right)$$ 2

    where A is (A_{510 nm}–A_{700 nm}) _pH1.0–(A_{510 nm}–A_{700 nm}), _pH4.5; Mw is the molecular weight (449.20 g/mol); DF is the dilution factor; _ε is the molar extinction coefficient (26,900 L/M/cm for cyanidin-3-glucoside); and l is the pathlength (1cm).
  • Proanthocyanidins (PA) were evaluated with the vanillin-HCL method described by Price et al. (1978) as follows: 20 μL of the extract was mixed with 180 μL of methanol and stirred. Then 1.2 mL of vanillin (at 4% w/v in methanol) was incorporated. Finally, 600 μL of concentrated HCl was added and protected from light for 30 min. The samples were measured at a wavelength of 500 nm. The values obtained were calculated in a calibration curve made with catechin, and the content was expressed in mg CE/100 g (mg catechin equivalent/100 g).

• Total Polyphenols (TPF) were quantified according to Singleton and Rossi (1965) with some modifications. The extracts (50 ul) were oxidized by the Folin-Ciocalteu reagent (100 μl) and mixed with distilled water (1500 ul). Finally, they were neutralized with sodium carbonate concentrate to 20% (30 μl). The absorbance was measured at 765 nm, and the results were expressed on a calibration curve as mgGAE/100 g (mg gallic acid/100 g).

• Antioxidant Activity (AA) was measured using the bleaching method of 2.20-azinobis-[3-ethylbenzothiazoline-6 + -sulfonic acid] radical cation (ABTS) described by Re et al. (1999). ABTS was dissolved in distilled water to yield a 7 mM solution. Radical cation was prepared by incubating the ABTS solution with a 2.45 mM potassium persulfate solution for 16 h in darkness at room temperature and subsequently diluted with methanol to a final absorbance of 0.7 at 734 nm. For AA determination, 30 ul of the sample was placed in a cuvette containing 3 mL of the ABTS solution. Absorbance was measured at 734 nm, and the results were indicated as inhibition percent (%Inhibition).

• Microencapsulation Efficiency (ME) following a methodology modified from Idham et al. (2010). Total anthocyanin was determined following a modifed (TA) and surface anthocyanin (SA) contents were measured. For TA, 100 ± 2 mg of microcapsules were weighed, 1 mL distilled water was added, and samples were ground using pestle and mortar to destroy the microcapsule membrane. Then, 10 mL of ethanol was added for 5 min and then filtered.

Table 1.

Formulations according to wall material and proportion used

Formulation	Maltodextrin (MD)	Gum Arabic (AG)	Modified Starch (MS)
1	30%	–	–
2	15%	15%	–
3	–	30%	–
4	10%	10%	10%
5	–	15%	15%
6	–	–	30%
7	15%	–	15%

SA extraction was performed by quickly washing with 10 mL ethanol in a vortex for 10 ± 2 s, followed by centrifugation at 3.000 xg for 3 ± 1 min at 20 ± 2 °C. After phase separation, the supernatant was collected and filtered. Quantification was performed using the pH-differential method described by Giusty Wrolstald (2000).

TA was calculated as cyanindin-3-glucoside with the following equation:

$$\text{TA}\left(\text{mg}/\text{L}\right)=\mathrm{\Delta }\text{A}\epsilon \times 1\times \text{Mw}\times 103\times \text{DF}$$

where ΔA is (A_510nm–A_700nm) _pH1.0–(A_510nm–A7_00nm) _{pH4.5; ε} is the molar extinction coefficient (26.900 L/M/cm for cyanidin-3-glucoside); 1 is pathlength (1 cm); Mw is the molecular weight (449.2 g/M for cyanidin-3-glucoside); 103 is the Factor Conversion from g to mg; and DF is the Dilution factor.

The microencapsulation efficiency was expressed as percentages and calculated with TA and SA results using the equation below:

$$\%\text{ME}=\left(\text{TA}-\text{SA}\right)/\text{TA x 1}00$$

Physicochemical properties

Percent polymeric color (PPC)

Percent Polymeric Color (PPC) was determined according to the Giusty Wrolstald (2000) technique. The percent polymeric color indicates high discoloration and losses of monomeric anthocyanins in samples treated with sodium bisulfite (Polymeric Color) and distilled water (Color Density). First, to obtain the dilution factor, 1 g of sample was mixed in 5 mL of distilled water and diluted in potassium chloride buffer 0.025 M (pH 1.0). For analysis, 1 mL potassium metabisulfite was added to 5 mL of diluted sample (bisulfite bleached sample), and 1 mL of water was added to 5 mL of the diluted sample (nonbleached, control sample). After equilibrating for 15 min, the samples were evaluated at λ = 700, 510, and 420 nm.

The PPC was obtained from the following equation:

$$\text{Color Density}-\text{Polymeric Color}=({\text{A}}_{\text{42 nm}}-{\text{A}}_{700\text{nm}})-\left({\text{A}}_{510\text{nm}}-{\text{A}}_{700\text{nm}}\right)\text{x FD}$$

Finally, to obtain the percent of polymeric color, the following equation was used:

$$\text{Percent Polymeric Color}=\text{Color Density}/\text{Polymeric Color x 1}00$$

Water activity (aw) was measured using an electronic dewpoint water activity meter Aqualab Series 3TE (Decagon Devices, Pullman, Washington, USA). The samples were then stabilized at 25 ± 2 °C for 15 ± 1 min.

Solubility was determined according to Cano-Chauca et al. (2005) with some modifications. 1 ± 0,01 g of microcapsules were hydrated in 20 mL of distilled water and centrifuged at 3000 xg for 5 min. The supernatant (5 mL) was transferred to a porcelain capsule and oven-dried at 105 ± 1 °C until constant weight. Solubility was calculated by weight difference.

Moisture and soluble solids were measured according to A.O.A.C (2000) methods (925.09/ 983.17).

Statistical analysis

An analysis of variance (one-way ANOVA) was performed to evaluate any significant differences between the two drying techniques and the different wall materials used. Multiple mean comparisons were evaluated by Tukey’s post-hoc test. Principal Component Analysis (PCA) was conducted to correlate the treatments with bioactive compounds and physicochemical characteristics, where a correlation matrix was used and the minimum Eigenvalue was set at 1. Also, a hierarchical cluster analysis was carried out using Ward´s method and Euclidean distance. All calculations were carried out using the software INFOSTAT Student Version 2018.

Results and discussion

Bioactive compounds

The results obtained for the atomization process are shown in Table 2. ANOVA analysis showed significant differences in TAC, PA, and TPF. ME had similarities between some formulations, like AG30% and AG15% + MS15%. Besides, MS30% and MD10% + AG10% + MS10%, MD30% and MD15% + MS15% were also similar, which could be attributed to the wall material and emulsion viscosity (Norkaew et al. 2019).

Table 2.

Bioactive compounds in microcapsules obtained by atomization

Microcapsules	TAC (1) (mg cyanidin 3glucoside/100 g)	PA (2) (mg CE/100 g)	TPF (3) (mg GAE/100 g)	AA (4) (% Inhibition)	ME(5) (%)
MD 30%	2509.00 ± 20a	1810.42 ± 23a	8575.95 ± 12a	93.1 ± 0.05a	85.45 ± 0.11d
MD 15% + AG 15%	1020.93 ± 13b	976.00 ± 12b	8046.37 ± 10b	91.7 ± 0.07a	67.1 ± 0.14a
AG 30%	789.00 ± 15c	598.21 ± 10c	6874.09 ± 12c	93.8 ± 0.10a	74,65 ± 0.12b
MD 10% + GA10% + MS10%	1907.5 ± 20d	621.43 ± 13d	7985.45 ± 14d	90.78 ± 0.02a	82.35 ± 0.10c
AG15% + MS15%	3414.00 ± 15e	1879.35 ± 11e	4774.91 ± 10e	93.6 ± 0.11a	74.5 ± 0.12b
MS 30%	1586.00 ± 16f	1169.64 ± 10f	8208.75 ± 13f	93.46a ± 0.12a	81.42 ± 0.09c
MD 15% + MS 15%	2091.67 ± 12 g	1860.20 ± 10 g	8552.45 ± 12 g	90.89 ± 0.08a	86.27 ± 0.12d

1: Total Anthocyanins, 2: Proanthocyanidins, 3: Polyphenols, 4: Antioxidant Activity, 5: Microencapsulation efficiency

The values are mean ± standard deviations and those in the same column not sharing the same superscript letter are significantly different from each other (P < 0.05)

TAC ranged from 1020.93 to 3414.00 mg cyanidin-3-glucoside/100 g. The differences between microcapsules could be attributed to the wall materials used and emulsion stability. Similar results were described by Murali et al. (2014) in the study of microencapsulation of black carrot juice using spray-drying at 150 °C. The maximum retention of TAC was observed in maltodextrin (1461.23 mg cyanidin-3-glucoside /100 g) followed by gum Arabic (1381.47 mg cyanidin-3-glucoside/100 g), and tapioca starch (1085.77 mg cyanidin-3-glucoside/100 g).

PA content ranged from 598.20 to 1879.30 mg CE/100 g. The variations between the powders can be associated in the same way with the wall materials used. The proportion of proanthocyanidins analyzed ranged between 32 and 40% with respect to the total anthocyanins. These percentages were similar to those reported by Vázquez-Castilla et al. (2012), who found a 40% proportion of PA in blueberry.

The TPF on the formulated microcapsules was higher than the TAC, the values obtained varied between 4774.91 and 8575.97 mgGAE/100 g. Differences between microcapsules may be due to the presence of other phenolic compounds and flavonols that could form complexes with polysaccharides depending on the water solubility, molecular size, conformational mobility, and shape of polyphenols (Naczk and Shahidi 2004).

Antioxidant Activity in the formulated microcapsules exhibited between 90 and 93% inhibition. Arteaga and Arteaga (2016) obtained values ranging from 41.39 to52.22 in the study of optimization of the antioxidant capacity, anthocyanins, and rehydration in cranberry powder (Vaccinium corymbosum) microencapsulated with mixtures of hydrocolloids (Maltodextrin, gum Arabic, and modified starch) at 120 °C. The differences could be associated with the fruit growing conditions (light, soil, water).

Microencapsulation efficiency was greater than 60% in all formulations, with the microcapsule with the highest ME being the mixture of maltodextrin and modified starch (86.5%). The results were close to those reported by Akhavan et al., (2016) for the microencapsulation optimization of natural anthocyanins from barberry fruit using maltodextrin, gum Arabic, and gelatin. They reported percentages of ME of 90–96% using 25–35% of wall material at 150 ºC inlet temperature. The values also indicated that the proportion of wall materials and temperatures used were adequate (Akhavan et al. 2016).

The results of the microcapsules obtained by lyophilization are presented in Table 3. The ANOVA analysis showed significant differences related to the dehydration method applied (P < 0.05). Freeze-drying retained bioactive compounds better than spray-drying, which could be attributed to the heat lability of anthocyanins and the wall materials used. In this sense, Madene et al. (2006) reported that high temperatures can collapse the maltodextrin matrix, accelerating the degradation of microcapsules.

Table 3.

Bioactive compounds in microcapsules obtained by lyophilization

Microcapsules	TAC1 (mg cyanidin 3glucoside/100 g)	PA2 (mg CE/100 g)	TPF3 (mg GAE/100 g)	AA4 (% inhibition)	ME5(%)
MD 30%	7714.00 ± 20a	2414.67 ± 26a	8669.88 ± 12a	92.83 ± 0.02a	88.14 ± 0.05b
MD 15% + AG 15%	3515.00 ± 10b	1546.78 ± 23b	8808.75 ± 10b	93.06 ± 0.02a	88.10 ± 0.04a
AG 30%	3323.00 ± 16c	686.66 ± 19c	8008.75 ± 14c	92.85 ± 0.05a	92.18 ± 0.10b
MD10% + GA10% + MS10%	7272.00 ± 13d	2194.07 ± 16d	9808.75 ± 10d	93.18 ± 0.10a	94.70 ± 0.08b
AG15% + MS15%	2204.00 ± 12e	986.7 ± 20e	5268.55 ± 11e	92.73 ± 0.08a	86.5 ± 0.03a
MS 30%	5752.00 ± 12f	1598.46 ± 18f	9058.75 ± 15f	93.69 ± 0.04a	87.13 ± 0.05a
MD 15% + MS 15%	4408.00 ± 17 g	19,401.10 ± 20 g	9059.57 ± 12 g	93.00 ± 0.04a	93.12 ± 0.07b

1: Total Anthocyanins, 2: Proanthocyanidins, 3: Polyphenols, 4: Antioxidant Activity, 5:Microencapsulation efficiency

The values are mean ± standard deviations and those in the same column not sharing the same superscript letter are significantly different from each other (P < 0.05)

We also observed the same differences in ANOVA results for lyophilized microcapsules, and therefore, anthocyanins, proanthocyanidins, and polyphenols were significantly different. This could be related to environmental stresses such as thermal processing, freezing, and dehydration, which may induce instability of the encapsulated products (Norkaew et al. 2019). AA was similar in all formulations, meaning that all formulations had chelating properties. Besides, there were variations in ME that could be associated with physicochemical interactions of wall material aggregation (Norkaew et al. 2019).

TAC in the microcapsules obtained by lyophilization were higher, compared with the spray-drying method. This can be explained by phenolic losses in the atomization process linked to the large surface area exposed to the air (Oberoi et al. 2015) and high temperatures, whereas the loss during freeze-drying is associated with material grinding after the lyophilization (Kuck et al. 2016). TAC ranged between 2204 and 7714 mg cyanidin-3-glucoside/100 g. Similar values were reported by Rigon and Zapata (2015) in the study of microencapsulation of blackberry (Rubus Fruticosus), where anthocyanin content was 2765.14 ± 5.4 mg cyanidin-3-glucosid/100 g.

Proanthocyanidins represent values between 20 and 44%, similar to those of the spray-dried microcapsules. These values are in agreement with de quantities reported by Vázquez-Castilla et al. (2012) for blueberries.

Polyphenols in lyophilized microcapsules were higher than those in the atomization process. The values were between 5268.55 and 9808.75 mgGAE/100 g, which may be associated with other phenolics present in berries. Also, the heat treatment at 130 °C showed that polyphenol degradation may be associated with the destruction of aromatic rings and Maillard reaction products that are lost in the thermal process (Bustos et al. 2018).

The antioxidant activity studied in the microcapsules ranged between 92.73 and 93.69% inhibition. Similar proportions were described by Brauch et al. (2015) with retention of antioxidant activity over 88% in maqui juice freeze-dried with maltodextrin as wall material.

The microencapsulation efficiency obtained was between 87 and 93%. As observed for spray-drying, the best wall material was the combination of MD15% and MS15%. ME results were similar to those of Stănciuc et al. (2018) in the investigation of bioactives from elderberry (Sambucus nigra L.) using whey proteins as a microencapsulation agent. These authors found 97.13 ± 1.47% of ME. Akhavan et al., (2016) encapsulated anthocyanins from barberry (Berberis vulgaris) with two combinations, namely maltodextrin-gum Arabic and maltodextrin-gelatin, and reported ME between 86 and 96%, respectively.

Physical properties

The physical properties analyzed in the microcapsules obtained by atomization and lyophilization are shown in Table 4. After the ANOVA test, we observed that the microcapsules were similar in some characteristics like solubility, moisture, and soluble solids in both processes. The losses and differences between them could be associated with grinding, which increased the likelihood of contact with the air, causing an oxidation reaction, one of the main factors leading to the degradation of the active substance (Hussain et al. 2018). On the other hand, significant differences were found for PPC and water activity, which could be related to the wall material. This can be associated with the ability of each material used to absorb water during drying and storage. It can also be attributed to the protection capacity offered by each polymer used Wilkowska et al. 2016).

Table 4.

Phisical properties in Spray Drying and Lyophilization Microcapsules

Microcapsules	Treatment	Percent Polymeric Color (%)	Water activity (Aw)	Solubility (%)	Moisture (%)	Soluble solids (ºBx)
MD30%	Spray-Drying	2.75 ± 0.2ab	0.51 ± 0.01ab	84.5 ± 0.01ª	7.00 ± 0.03a	12.4 ± 0.02ª
MD30%	Lyophilization	12.6 ± 0.15ab	0.48 ± 0.02ab	92.9 ± 0.01ª	8.7 ± 0.02a	12.40.01 ± ª
MD15% + AG15%	Spray-Drying	2.3 ± 0.18ª	0.5 ± 0.01ab	87.2 ± 0.02ª	5.2 ± ª0.02a	12.3 ± 0.01ª
MD15% + AG15%	Lyophilization	11.5 ± 0.10ª	0.51 ± 0.03ab	95.7 ± 0.04ª	8.7 ± 0.04ª	12.2 ± 0.01ª
AG30%	Spray-Drying	2.6 ± 0.12ab	0.6 ± 0.15b	91.3 ± 0.05ª	5.4 ± 0.02ª	12.2 ± 0.01ª
AG30%	Lyophilization	12.8 ± 0.22ab	0.56 ± 0.01b	82.18 ± 0.01ª	9.75 ± 0.01ª	12.3 ± 0.02ª
MD10% + GA10% + MS10%	Spray-Drying	3.15 ± 0.34ª	0.45 ± 0.03a	86.30 ± 0.02ª	6.30 ± 0.03ª	12.3 ± 0.01ª
MD10% + GA10% + MS10%	Lyophilization	8.9 ± 0.12ª	0.49 ± 0.01a	89.53 ± 0.01ª	8.8 ± 0.02ª	12.3 ± 0.01ª
AG15% + MS15%	Spray-Drying	2.95 ± 0.11ª	0.67 ± 0.01b	95.00 ± 0.03ª	6.00 ± 0.01a	12.2 ± 0.01ª
AG15% + MS15%	Lyophilization	9.6 ± 0.10ª	0.49 ± 0.03b	95.2 ± 0.02ª	6.7 ± 0.03ª	12.4 ± 0.01ª
MS30%	Spray-Drying	4.5 ± 0.12b	0.55 ± 0.01ab	93.3 ± 0.01ª	5.8 ± 0.01ª	12.6 ± 0.01ª
MS30%	Lyophilization	13.8 ± 0.13b	0.45 ± 0.04ab	89.01 ± 0.01ª	6.52 ± 0.02a	12.5 ± 0.01ª
MD15% + MS15%	Spray-Drying	5.3 ± 0.01b	0.57 ± 0.03ab	85.5 ± 0.04ª	6.00 ± 0.04a	12.1 ± 0.01ª
MD15% + MS 15%	Lyophilization	12.9 ± 0.20b	0.47 ± 0.02ab	97.3 ± 0.05ª	8.4 ± 0.05a	12.1 ± 0.01ª

The values are mean ± standard deviations and those in the same column not sharing the same superscript letter are significantly different from each other (P < 0.05)

Percent Polymeric Color in both processes was around 2–13.8%. The values were similar to those reported by García (2014) in the study of blackberry juice (Rubus urticaefolius, poir R.) with values of percent of Polymeric Color of 13.42%. Besides, the results obtained for both processes were low, indicating the efficacy of the treatment.

Water activity in spray- and freeze-dried powders was between 0.5 and 0.6. These results are in agreement with Nóbrega et al. (2015), who analyzed oven-dried acerola (Malpighia emarginata) pomace (0.20–0.40).

Solubility and moisture content in spray- and freeze-dried powders were 84–94% for solubility and 5.4–9.7% for moisture. Higher contents were found in the freeze-drying treatment and similar results were reported by Franceschinis et al. (2014), who observed 2.41–6.11% moisture content for blackberry (Rubus sp.) juice microencapsulated with MD (Dextrose Equivalent 12) and trehalose dihydrate by spray- and freeze-drying. The variations in moisture in spray- and freeze-drying are the result of rapid freezing (lower than −40 °C), and hence the pores in the outer layer are smaller, which may hinder mass transfer and act as a barrier against sublimation, resulting in increased moisture retention (Ezhilarasi et al. 2013). The solubility results may be related to the high solubility of the encapsulating agents used as well as the particle size obtained, which means that the smaller particles have a greater surface area available for hydration (Franceschinis et al. 2014).

The soluble solid results were similar in both processes applied and higher than the fruit (13.8 ºBx). This difference can be attributed to the wall material added for the development of microcapsules.

PCA analysis was applied to all data to determine the most important variables that explain the relationships among the fourteen microcapsules formulated by atomization and freeze-drying. The first two components explained 64.6% of the overall variance (42.7% and 21.7% for PC1 and PC2, respectively) (Fig. 1). According to the Cluster analysis, it was observed that at a Euclidean distance of 4.5, four groups had similar characteristics. These groups are presented in the dendrogram in Fig. 1 and 2.

Fig. 1.

Principal components plot for bioactive compounds and physical properties in the different microcapsules

Fig. 2.

Dendogram of the cluster analisys

It can be observed along the first principal component (PC1) from left to right that moisture content and water activity were associated with three groups of samples: clusters I (AG1% + MS15%spray and AG15% spray), III (MD30%spray), and IV (AG15% + MS15%:liop, M30%spray, MD15% + MS15%spray, MD10% + AG10% + MS10%spray, MD15% + AG15%spray and AG30%liop). Samples MS30%liop, MD15% + MS14%liop, MD15% + AG15%liop, MD10% + AG10% + MS10%liop and MD30%liop (cluster II) were related to bioactive compound (proanthocyanidins, anthocyanins, and polyphenols). Antioxidant activity, percent polymeric color, encapsulation efficiency, solubility and soluble solids, as opposed to moisture content and water activity.

The PCA diagram supports the difference between spray- and freeze-drying treatments, confirming that most of the lyophilized microcapsules (MD15% + AG15%liop; MS30%liop; MD10% + AG10% + MS10%liop; MD15% + MS15%liop; MD30%liop) are strongly associated with bioactive compounds. This means that these formulations retain more of these compounds. Also, the lyophilized treatment has greater ME and values of polymeric color, which are associated with the higher proportion of proanthocyanidins and more stability of compounds due to copigmentation reactions (Giusty-Wrolstald 2000). The PCA also shows that spray-dried powders have a greater association with the moisture and water activity variables, which would indicate that they are more unstable during storage and susceptible to chemical and enzymatic degradation reactions.

Conclusion

Through drying technology, the use of discarded fruits allows the development of byproducts like powders with bioactive compounds and physical properties suitable for consumption, providing an alternative for the development of healthy ingredients and a solution to environmental problems induced by waste. Besides, the methodologies applied enable the bioactive compounds present in berries to be conserved in large quantities. Lyophilization is a better treatment for functional fractions of blueberries, and spray-dried powders have better physical properties, like moisture and water activity. According to the wall material, we observed a synergic effect in AG15%-MS15%, as this formulation in both treatments was less affected by the processes applied.

Abbreviations

AG

Gum Arabic

AA

Antioxidant Activity

BAC

Bioactive Compounds

GAE

Gallic Acid equivalent

MD

Maltodextrin

ME

Microencapsulation Efficiency

MS

Modified Starch

PA

Proanthocyanidins

PPC

Percent Polymeric Color

PCA

Principal Component Analysis

TAC

Total Anthocyanins

TPF

Total Polyphenols

Author contributions

All authors have conceived and planned the experiments. Olivares La Madrid, A. and Villalva, F. carried out the experiments. Lotufo Haddad, A. and Alcocer, J. contributed to the interpretation of the results. Armada, M. and Cravero, P. helped supervise the project and took the lead in writing the manuscript. All authors provided critical feedback and helped shape the research, analysis and manuscript.

Declarations

Conflict of interest

Authors declare no conflict of interest.