Stability of Haskap Berry (Lonicera Caerulea L.) Anthocyanins at Different Storage and Processing Conditions

The effect of freezing, frozen storage (–18 C for 6 months), thawing, juice extraction, and hot-air drying on the anthocyanin profile of haskap berry (Lonicera caerulea L.) was investigated using RP-HPLC. Five anthocyanins (ANCs) were quantified: cyanidin 3,5-di-glucoside (4.27 % of the total ANCs), cyanidin 3-glucoside (89.39 %), cyanidin 3-rutinoside (2.07 %), pelargonidin 3-glucoside (0.83 %), and peonidin 3-O-glucoside (3.44 %). Freezing did not significantly affect the content of individual ANCs, while frozen storage resulted in significant reductions (16.00-24.50 %). Thawing the frozen berries in the microwave oven retained the highest content of different ANCs. The highest degradation, however, occurred while thawing at room temperature. Extracting juice from the berries significantly reduced the content of individual ANCs. Drying the berries to 25 % moisture content at 60, 100, and 140 oC reduced the individual ANCs by 73.85-76.19, 78.46-80.95 and 90.77-95.40 %, respectively. The overall stability of the five ANCs during storage and processing is summarized by the following trend (from most to least stable): peonidin 3-O-glucoside  pelargonidin 3-glucoside  cyanidin 3,5-diglucoside  cyanidin 3-rutinoside  cyanidin 3-glucoside.


Introduction
The haskap berry (Lonicera caerulea L.) has been recently introduced as a commercial crop to the North American market.Some varieties and cultivars are currently available in Canada and USA (Bors et al., 2012).These berries are either consumed fresh or processed into juice, pastries, jams, ice cream, and dried berries (Celli et al., 2014).Haskap berries have attracted attention for their distinct profile of phenolic phytochemicals (Jurikova et al., 2012).They are particularly rich in anthocyanins (ANCs) with varied health benefits (Paredes-Lopez et al., 2010).The total anthocyanin content (TAC) of haskap berry was found to be up to 13.00 mg cyanidin 3-glucoside (C-3-G) equivalents per g fresh weight (FW) (Bakowska et al., 2007;Fan et al., 2011;Rupasinghe et al., 2012).This berry has much higher antioxidant potential than that reported for blueberry, blackberry, raspberry, bilberry, strawberry, sea buckthorn and black currant (Rop et al., 2011;Raudsepp et al., 2013;Celli et al., 2014).
ANCs are the most important water-soluble phytochemicals in nature (Harborne, 1998).They are responsible for the distinguished colors of several fruits and vegetables.They are distinctive from the other flavonoids by their ability to form flavylium cations (Fig. 1) (Mazza, 2007).ANCs consist of an aglycon base or flavylium ring (anthocyanidin), sugars, and may contain acylating groups (Bueno et al., 2012).From the several anthocyanidins found in nature, only cyanidin, delphinidin, petunidin, peonidin, pelargonidin, and malvidin (Fig. 1) are of importance in human nutrition (Harborne, 1998;Jaganath & Crozier, 2010;Bueno et al., 2012).Food ANCs play important roles in preventing various diseases including cancer, diabetes, cardiovascular diseases, and obesity.They are also associated with improving immunity and night vision, retarding aging and reducing the risk of degenerative disorders (Jing, 2006;Nikkhah et al., 2008).These beneficial effects of ANCs are attributed to their antioxidant, detoxification, anti-proliferation, anti-angiogenic, and anti-inflammatory activities (Miguel, 2011).juice extraction and subsequently sold as a dried berry product.Despite the varied functions and health benefits of ANCs, they are very labile and undergo significant breakage and structural changes during storage and processing (Ochoa et al., 1999;Lohachoompol et al., 2004;Sadilova et al., 2006).Due to their highly reactive nature, ANCs readily degrade to colorless or brown compounds.Loss of ANCs is also accelerated by the presence of oxygen and enzymes, and during the high temperature processing (Jackman et al., 1987).
Figure 1.Structures of major ANCs (adapted from Jing, 2006) The effect of storage, thawing, extraction and drying conditions on the retention of haskap berry ANCs has been investigated in our laboratory.The total anthocyanin content (TAC) was reduced in different varieties by 39.31-59.24and 36.87-56.57% upon frozen storage for 6 months at-18 ºC and-32 º C, respectively (Khattab et al., 2015a).The reduction in TAC was 32.14-53.25, 28.55-51.45 and 18.92-47.22% when the frozen fruits were thawed at room temperature (25±2 º C), fridge (4 º C) and in the microwave oven, respectively (Khattab et al., 2015b).In a recent study (Khattab et al., 2016a), we have investigated the effect of juice extraction and drying conditions on the stability of haskap berry ANCs.The juice extraction process significantly reduced the TAC by 48.18 % in the pressed residues as compared to that of the whole berries.Furthermore, the TAC decreased significantly at different drying temperatures with a strong positive correlation between the drying temperature and the degradation rate.
Our previous studies on haskap berries did not report on the levels of individual ANCs during storage and processing.It is important to determine the anthocyanin profiles for different haskap berry products and the effect of storage and processing conditions, as this will help develop strategies to improve the nutritional content of these products.Therefore, this study examines the effect of the processing chain including freezing, frozen storage, thawing, juice extraction, and drying conditions on the stability and retention of specific individual ANCs contained in haskap berry fractions.

Haskap Berries
Haskap berries (Lonicera caerulea L.); variety Indigo Gem (26 kilograms) were obtained from LaHave Natural Farms, Blockhouse, Nova Scotia, Canada.Upon receiving the fruits, they were analyzed for their anthocyanin profile.The fruits were then frozen and stored at -18 º C. Half the berries were thawed and analyzed the next day to study the effect of the freezing process.The other half was kept frozen for 6 months to study the effect of frozen storage, thawing methods and drying conditions.

Chemicals and Phenolic Standards
All chemicals used for this research were of analytical and HPLC grades and were procured from Sigma Aldrich (Oakville, Ontario, Canada) and Fisher Scientific (Ottawa, Ontario, Canada).Authentic anthocyanin standards were obtained from Sigma Aldrich, Canada.

Fruit Pressing and Juice Extraction
The juice extraction was carried out as previously described (Khattab et al., 2016b), using a lab-scale manual multi-fruit juice extractor (F.Dick 9060600, 6L; Friedrich DICK, Deizisau, Germany).In this study (Fig. 2) the frozen berries were allowed to thaw at different conditions including room temperature (252 C for 12 h), and in the refrigerator (4 C for 22 h) and microwave oven (1000 Watts for 0.29 h) according to Khattab et al. (2015b).The berries thawed at room temperature were loaded to the juice extractor (in 3 kg batches) and manually pressed.The pressing was done until obtaining 70 % of the original fruit weight as juice.The pressed berries were osmotically treated by mixing with sucrose (20 % of their weight), and left to infuse for 24 h.The mixture was then gently pressed using the juice extractor to drain the liquid part (syrup) without drastically affecting or rupturing the berries.The leftover berries were analyzed and dried.

Drying Process
The drying process was carried out according to Khattab et al. (2016a) using a lab-scale hot-air drying oven (Isotemp  Oven, Model 630F, Fisher Scientific, USA) at 60, 100, and 140 °C.The drying continued up to 48 h and the time needed to reach 25 % moisture content was recorded.

Moisture Content
Moisture content of the haskap berry samples was determined using a hot-air drying oven (Isotemp  630F, Fisher Scientific, USA) at 103± 2º C and atmospheric pressure until constant mass was reached (ISO, 2009).

HPLC Analysis of Anthocyanins from Haskap Berries
Samples were extracted with 80 % acidified methanol and prepared for HPLC analysis according to Khattab et al. (2015a).The anthocyanin profiles of fresh, frozen, pressed, and dried haskap berries were analyzed according to Khattab et al. (2015c) using the reversed-phase DAD-HPLC (Agilent 1100 Series, Agilent Technologies, Hewlett-Packard, Waldbronn, Germany).Chromatograms were acquired at 520 nm and data were analyzed using the Agilent ChemStation software (version A10.02).

Statistical Analysis
All experiments were done in triplicates and data were analyzed using a one factor analysis of variance (ANOVA).Tukey-Kramer mean separation tests were done for multiple comparisons with SigmaStat software (version 3.5).The significance was accepted at p≤ 0.05.

Effect of Freezing and Frozen Storage on the Anthocyanin Profile of Haskap Berries
The results of the present study (Table 1) indicate that the freezing process had no significant effect on the content of individual ANCs of haskap berries.The total content of all ANCs was reduced by only 1.79 % as compared to the fresh fruit.In the food industry, a storage temperature of -18 ºC effectively reduces the chemical and biological spoilage of foods and extends their shelf life.However, freezing causes cell rupture and division allowing reactions between enzymes and their substrates (Tomá s-Barberá n & Espí n, 2001).
Therefore, ANCs may be degraded during freezing and more extensively during thawing due to their interaction with oxidative enzymes.The effects of freezing and frozen storage on anthocyanin content and phenolic profile of different kinds of berries have been investigated by other researchers (Bushway et al., 1992;de Ancos et al., 2000;Hä kkinen et al., 2000;Mullen et al., 2002).According to Selman (1992), the process of freezing itself does not alter the nutritive value of the product being frozen.Upon freezing raspberries at -30 ºC within 3 hours of picking, Mullen et al. (2002) found no significant differences either in the levels of the individual ANCs or in the TAC of the fresh and frozen raspberries.This might be because ANCs in frozen fruits become more easily extractable due to degradation of cell structures during frozen storage over time.The enhanced extractability might have surmounted any ANCs degradation that might have occurred during the freezing process.In some cases, ANCs have been even reported to increase during freezing (de Ancos et al., 2000).
The effect of frozen storage at -18 ºC for 6 months on the individual ANCs from haskap berries is shown in Table 1.The content of the five ANCs significantly decreased upon storage.The highest reduction (24.50 %) was recorded for C-3-G (the most abundant ANC), while the smallest decrease was that of peonidin 3-O-glucoside (16.00 %).This agrees with de Ancos et al. (2000) who found that C-3-G demonstrated a more significant degradation during frozen storage compared to the other ANCs found in raspberries.Structural analysis showed that less free hydroxyl groups and more methoxy groups in the B-ring of the aglycon improve anthocyanin stability (Liu et al., 2014).

Effect of Thawing Conditions on the Anthocyanin Profile of Haskap Berries
Freezing techniques affect how the food thaws and its subsequent structural and compositional changes.Quick freezing retains cell integrity than slower freezing due to the smaller intracellular ice crystals formed.Cell integrity is further influenced by thawing regimes as quick thawing better retains fruit quality (Delgado & Rubiolo, 2005).The freezing and thawing chain ruptures the cells allowing reactions between enzymes and their substrates.Anthocyanins, therefore, may degrade during thawing due to their interaction with oxidative enzymes like polyphenol oxidases and peroxidases that have been reported to be active even at lower temperatures (Chisari et al., 2007).
In the present study, we compared the effect of thawing methods (room temperature, refrigerator, and microwave oven) on the anthocyanin profile of frozen-stored haskap berries.The content of individual ANCs of haskap berries as affected by thawing methods is shown in Table 2.All thawing methods reduced the ANCs of haskap berries with significant differences among them.In agreement with the present study, the quality of frozen food has been reported as being more affected by the thawing process than by the freezing itself (Kim et al., 2011).Thawing has a major impact on the food quality, as the compounds normally kept apart in the intact cell can mix and react with each other (Kmiecik et al. 1995).

Figure 4. Chemical structures of haskap berry ANCs identified in this study
The reductions in Table 2 may be attributed to the hydrolytic reactions that convert anthocyanin glycosides to chalcones, which intuitively degrade into aldehydes and phenolic acids (Kamiloglu et al., 2015).It is also possible that enzymes might have played a role in the reduction of ANCs (Howard et al., 2010).The highest reduction occurred when thawing at room temperature followed by that in the refrigerator, while microwave thawing caused the least reduction.The higher ANCs retention during the microwave thawing might be attributed to the shorter time taken (17 min) as compared to 12 and 22 h in the room temperature and refrigerator thawing, respectively.Using microwave, thawing time was reduced by seven times compared to convective thawing at atmospheric temperature when appropriate conditions were used (Tong et al., 1993).Thawing at lower temperature (refrigerator), despite taking significantly longer time, retained more ANCs compared to room temperature thawing.
These results agree with Oszmiański et al. (2009) where considerable ANC losses were reported after thawing strawberries stored frozen for several months.Moreover, the ANC contents of frozen fruits were found to depend on their thawing methods.The differences of ANC contents between bilberries thawed at 2-4°C and fruit thawed at room temperature (18-20°C) were approximately 10% (Kmiecik et al., 1995).Our results showed that the reductions in individual ANCs were 19.05-26.94,0.00-16.73and 0.00-12.04% after thawing at room, and using the refrigerator and microwave oven, respectively.A similar reduction trend was seen as that observed for berries in frozen storage where C-3-G showed the highest reduction while the lowest reductions were noticed in peonidin 3-O-glucoside and pelargonidin 3-glucoside.This is supported by other studies that report that C-3-G is one of the most reactive ANCs during processing (Rommel et al., 1990;Boyles et al., 1993;Garcia-Viguera et al., 1998).Furthermore, Fleschhut et al. (2006) reported that an increase in hydroxyl groups in the B-ring of the anthocyanin nucleus results in reduced stability.They found that cyanins seemed to be less stable than petunins and peonidins indicating that methylation of hydroxyl groups in B-ring increases the stability of ANCs.

Effect of Juice Extraction on the Anthocyanin Profile of Haskap Berries
The results of the present study (Table 3) indicate that both juice and syrup showed significantly lower ANC content (1.61 and 1.00 mg/g FW, respectively) than that of the thawed berries (4.07 mg/g FW).The syrup showed significantly lower content than that found for the juice fraction, which might be attributed to high extractability of these water soluble compounds.These results are not surprising as it is known that ANCs suffer significant degradation and structural changes during processing (Sadilova et al., 2006).Extraction of fruit juice causes major ANC losses yielding significantly lower ANC contents in the obtained juice as compared to the corresponding fruit used.1.43±0.20 b 0.89±0.08c 3.97±0.08a 13.49±0.18ANC 3 0.09±0.01 a 0.04±0.00b 0.02±0.00b 0.12±0.00 a 0.37±0.00ANC 4 0.04±0.01 a ND ND ND ND ANC 5 0.17±0.01 a 0.07±0.00b 0.05±0.01b 0.19±0.01 a 0.65±0.01ANC 1 : Cyanidin 3,5-di-glucoside; ANC 2 : Cyanidin 3-glucoside; ANC 3 : Cyanidin 3-rutinoside; ANC 4 : Pelargonidin 3-glucoside; ANC 5 : Peonidin 3-O-glucoside; FW: fresh weight; DW: dry weight.Values are means of duplicate analysis ± standard deviation (SD).* Thawed at room temperature.** The content of the five ANCs of the thawed berries were 1.21, 25.91, 0.71, 0.28 and 1.08, respectively based on the DW base.Values in the same row with similar superscript letters are not significantly different (0.05).
The content of ANCs in the juices affects their storage stability and shelf life, for example, only 11-15% of the original ANCs were detected in the commercial juices at their expiry date, after storage for 35-49 weeks at room temperature (Hellstrom et al., 2013).It is known that manufacturing processes lead to anthocyanin degradation and color alteration in berries (Hager et al., 2008).Hellstrom et al. (2013) attributed the lower anthocyanin content in commercial berry juices to the severe production processes applied industrially.Furthermore, processing blueberries into purees caused 43 % loss in total ANCs, compared to the levels in fresh fruit (Brownmiller et al., 2008).
In other studies, it was found that the stability of individual ANCs in food systems depended greatly on their chemical structure (Jackman et al., 1987) and different ANCs had different degradation kinetics in juices (Hellstrom et al., 2013).Hydroxyl, methoxyl, sugar, and acylated sugar substituent groups have pronounced effects on the stability of the ACNs.Diglycosidic substitution gives more stability than monoglycosidic substitution (Mazza & Miniati 1993).Moreover, acylation of the ANC molecule improves its stability by preventing it from hydration (Brouillard, 1981).Pelargonidin 3-glucoside totally degraded and was not detected in the juice, syrup or pressed berries.This is might be due to its marginal content in the initial thawed berries.For the other ANCs, the same reduction trend was observed as seen during the storage and thawing where C-3-G suffered the highest reduction (47.94 %) followed by cyanidin 3-rutinoside (47.89 %) and cyanidin 3,5-di-glucoside (45.45 %).The least reduction, however, was observed for peonidin 3-O-glucoside (39.81 %).

Effect of Drying Conditions on the Anthocyanin Profile of Haskap Berries
The effect of hot-air drying at different temperatures is shown in Table 4.The drying time taken to reach a moisture content of 25 % was 16.0, 5.6 and 2.5 h at 60, 100 and 140 C, respectively.Upon drying to this moisture content, the reductions in the individual ANCs of the dried berries were 73.85-79.99, 78.46-82.73 and 90.77-100.00% at the three drying temperatures, respectively.The HPLC chromatograms of pressed and dried haskap berries at different temperatures are illustrated in Fig. 5.The TAC decreased by 71.85 and 88.30 % after 8 h of drying at 60 and 100 C, respectively.Even at lower temperature (60 C), the degradation of ANCs continued with drying time to reach more than 95.00 % after 48 h where all ANCs were significantly degraded with pelargonidin 3-glucoside, cyanidin 3,5-di-glucoside and C-3-G being the most affected.However, ANCs were completely degraded after 32 h of drying at 100 C.Logarithmic anthocyanin degradation with an arithmetic increase in temperature has been frequently reported (Drdak & Daucik, 1999;Rhim, 2002).The high temperatures blanching (95 ºC for 3 min in combination with pasteurisation) involved in processing blueberries into purees resulted in 43 % loss in total ANCs (Brownmiller et al., 2008).In addition, ANCs were significantly decreased as a result of jam and marmalade processing of black carrots (Kamiloglu et al., 2015).After 20 weeks of jam and marmalade storage, ANCs were significantly higher at 4 °C that at 25 °C.
Drying at 140 C (Table 4), however, had a significant effect on the ANCs.Both pelargonidin 3-glucoside and cyanidin 3,5-diglucoside were completely degraded upon drying to 25% moisture content (2.5 h).The other ANCs were reduced by 90.77 to 95.40 % with C-3-G and peonidin 3-O-glucoside being the highest and least degraded ones, respectively.After 5 h of drying at this temperature, only C-3-G was detected in the samples but no other ANCs.These results agree with Drdak and Daucik (1999) and Rhim (2002).Garcia-Viguera et al. (1999) reported anthocyanin losses of 10 to 80 % during jam processing.Moreover, C-3-G and pelargonidin 3-glucoside in blackberry and strawberry puree were significantly reduced by thermal treatment at 70 C for 2 min (Patras et al., 2009).Furthermore, the content of total ANCs of dehydrated potato flakes decreased by 23-45 % during the dehydration process at 100-150 °C (Nayak, 2011).The reduction of ANCs upon drying might be attributed to the heat labile nature of ANCs.The exposure to oxygen at higher temperatures might have also contributed to their degradation (Welcha et al., 2008).Sadilova et al. (2006) observed that only 50 % of ANCs were retained after heating elderberry for 3 h at 95 C.Similar losses in raspberry purees were reported by Ochoa et al. (1999).This reduction might also be attributed to the osmotic treatment during which more ANCs might have leached out into the osmotic solution.Lohachoompol et al. (2004) found that the reduction in the TAC was 41 % in the dried blueberries and increased to 49 % when drying was preceded with osmotic treatment.The loss of ANCs was further attributed to several factors, including residual enzyme activity or condensation reactions of ANCs with other phenolics at higher temperatures (Jackman et al., 1987;Brownmiller et al., 2008).Fracassetti et al. (2013) found that storage of freeze-dried wild blueberry powder for 49 days at 25, 42, 60, and 80 °C reduced single and total ANCs at all temperatures.The reduction in ANCs depended on the temperature and occurred slowly up to 3% at day 14 at 25 and 42 °C, whereas it was faster, reaching 60 and 85 % after three days at 60 and 80 °C, respectively.
The stability of ANCs is influenced by the aglycon B-ring substituents and the presence of additional hydroxyl groups decreases the aglycon stability in neutral media.Furthermore, mono-and diglycosides derivatives are more stable than the non-glycosylated aglycons (Castañeda-Ovando et al., 2009).Our results showed that among the three cyanidins investigated in this study, cyanidin 3,5-diglucoside (with two sugar moieties) showed the highest stability followed by cyanidin 3-rutinoside and C-3-G.
The reductions in individual ANCs were 73.85-76.19, 78.46-80.95 and 90.77-95.40% upon drying to 25 % moisture content at 60, 100 and 140 º C, respectively.This excludes pelargonidin 3-glucoside (the least abundant anthocyanin in haskap berries) which was completely degraded at both 100 and 140 º C. Cyanidin 3,5-diglucoside also disappeared in the samples dried at 140 º C. Peonidin 3-O-glucoside, however, was the most stable anthocyanin at different temperatures.Drying haskap berries at 60 ºC is recommended for better retention of ANCs.Drying at this temperature was recommended by Garba and Kaur (2014) who investigated the influence of hot air drying (40-60 º C) on the TAC of black carrot and found that the optimum retention of ANCs was attained from drying at 60 º C. In the study by Zoric et al. (2014), the effect of heating temperature (80-120 °C) and processing time (5-50 min) on the stability of ANCs in freeze-dried sour cherry pastes was explored.They found that C-3-G was the most unstable among ANCs.

Conclusion
ANCs were significantly affected by frozen storage and subsequent thawing by different methods.Microwave thawing revealed the highest ANC retention.Drying significantly reduced ANCs and higher drying temperatures resulted in higher degradation.All ANCs were significantly reduced during the frozen storage, thawing, juice extraction and drying.Cyanidins were more degradable than pelargonidin than peonidin.Microwave thawing and lower-temperature storage and processing are recommended for better retention of haskap berry ANCs.
Understanding the structure-stability relations and behavior of ANCs during storage and processing will help haskap and other berry processors to design high-quality berry products with improved nutritional/functional properties.

Figure 2 .
Figure 2. Schematic flowchart of the experimental work.Method of thawing is represented by A: Room temperature; B: Fridge temperature, C: Microwave oven.# Pressing until 70 % juice yield; ‡ Pressing to drain the remaining liquid; * HPLC profiling was conducted for these samples

Table 2 .
Effect of thawing conditions on the content of individual anthocyanins (mg/g) in haskap berries Peonidin 3-O-glucoside.Values are means of duplicate analysis ± standard deviation (SD).Values in the same row with similar superscript letters are not significantly different (0.05).

Table 3 .
Effect of juice extraction (until 70 % juice yield) on the content of individual anthocyanins (mg/g) in haskap berry products