Changes in fresh-market and sensory attributes of blackberry genotypes after postharvest storage

2.1 Chemicals

The chemicals used in this study were purchased from commercial sources. Sodium hydroxide (0.1 N) for titratable acidity analysis was purchased from VWR (West Chester, PA). The solvents and standards used for analysis included HPLC-grade methanol, acetone, formic acid and acetic acid purchased from EMD Millipore (Darmstatd, Germany). For the quantification of ellagitannins and flavonols, ellagic acid and rutin standards were purchased from Sigma Aldrich (St. Louis, MO). For the quantification of anthocyanins peaks, a cyanidin 3-glucoside standard was purchased from Polyphenols-AS (Sandnes, Norway). Gallic acid from Sigma Aldrich (St. Louis, MO) was used to quantify total phenolics.

2.2 Blackberries

In 2015, blackberries were harvested at the shiny-black stage of maturity at the University of Arkansas Fruit Research Station, Clarksville, AR. Blackberries were harvested directly into 240-g, low-profile, vented clamshells, placed in chilled coolers and transported to the Department of Food Science, Fayetteville, AR. Approximately 4 kg of blackberries were harvested from each of the 11 genotypes, five cultivars (Natchez, Osage, Ouachita, Prime-Ark^® 45 and Prime-Ark^® Traveler) and six selections (A-2418, A-2434, A-2450, A-2453, A-2491 and APF-268). Six genotypes were harvested one week and the remaining five genotypes were harvested the next week.

Upon arrival at the Department of Food Science, the blackberries were randomized and separated in triplicates (three clamshells by genotype and storage) and evaluated at harvest (day 0) and after storage (day 7). Harvest (day 0) is defined in this study as the day after actual harvest, since the blackberries were harvested and transported then stored overnight at 2.0°C ± 1.0°C and 90.0% ± 1.0% RH. Postharvest evaluations and sensory analysis were performed at harvest and 7 days after storage (2.0°C ± 1.0°C and 90.0% ± 1.0% RH). For chemical composition analysis, 150 g of blackberries were placed in plastic freezer bags in triplicate and frozen in a commercial freezer (–20°C ± 2°C) at harvest and after 7 days of storage. The clamshells for this study were filled with blackberries that were commercially marketable (whole, intact berries without damage or flaws) to simulate commercial fresh-market standards.

2.3 Fruit quality

The fruit quality attributes of the blackberries were evaluated in triplicate clamshells per genotype at harvest (day 0) and after storage (day 7). The clamshells were filled with blackberries to commercial capacity (30–48 berries per genotype). The fruit quality attributes were evaluated on the fresh blackberries and included berry firmness, incidence of red drupelets, weight loss and unmarketable fruit.

2.4 Chemical composition

The chemical composition of the blackberries was evaluated in triplicate per genotype at harvest (day 0) and after storage (day 7). Chemical composition included analysis of basic blackberry composition and phytochemical compounds.

2.5 Sensory analysis

Descriptive sensory analyses of fresh blackberries were performed at the Sensory and Consumer Research Center at the University of Arkansas, Fayetteville, following the methodology established by Threlfall et al. 2016 [7]. The eleven genotypes were harvested with six genotypes harvested in the first week and five genotypes harvested the following week. The sensory panel had long breaks between replications to avoid panel fatigue. Descriptive sensory analyses were performed by panelists at harvest (day 0) and after storage (7 days). Six genotypes were evaluated on 17th of June (day 0) and then on the 22nd of June (day 7), then the other five genotypes were evaluated on 22nd of June (day 0) and 29th of June (day 7). Each panelist evaluated four berries of each genotype in duplicate. The blackberry samples were served monadically at room temperature (24°C) with random three-digit codes in a randomized complete block design. Sample serving order was randomized across replication to prevent presentation order bias. Panelists cleansed their palates with unsalted crackers and water between samples. The panelists used a modified Sensory Spectrum^® method, an objective method for describing the intensity of attributes in products using standards for the attributes. The descriptive panelists (n = 8-9) used a lexicon of sensory terms for fresh-market blackberries (Table 1) [7] and assessed each attribute per genotype and replication at a particular intensity according to the reference and using the universal aromatic scale. The descriptive panel identified and evaluated fresh blackberry attributes for appearance (color, uniformity of color and glossiness), basic tastes (sweet, sour and bitter), feeling factors (astringent and metallic), overall aromatic impact, and texture (hardness and popping), using a 15-point scale (0 = less of the attribute and 15 = more of the attribute in terms of intensity). The color of blackberries was scored based on the Royal Horticulture Society Chart (5th edition, London) Black Group 203 A = 15.0 score for intensity [7].

After harvest, the fruit from each of the 11 genotypes was completely randomized. Three replications were used to evaluate fruit quality and chemical composition analyses, and two replications were used for the descriptive sensory analyses.

Statistical analyses were conducted using JMP^® (version 12.0; SAS Institute, Cary, NC). A univariate analysis of variance (ANOVA) was used to determine the significance of main factors (genotype and storage) and interaction (genotype*storage). For the analysis of sensory data, panelists were treated as random effect, while genotypes and storage were treated as fixed effects. Tukey’s test Honestly Significant Difference (HSD) and Slice test were used to detect significant differences (p < 0.05) among means and verify interactions at 95% significance level, respectively. Pairwise correlations and Principal Component Analysis were used to verify the relationship between/within fruit quality, chemical composition and sensory attributes.

At harvest, the blackberries had a range of firmness (4.9–9.0 N) and red drupelets (0.3–4.4%), whereas after storage the range of firmness was 4.0–10.1 N and red drupelets was 0–6.4%. There was an interaction between genotype and storage for firmness (p = 0.0344) of blackberries, but not for the percent of red drupelets. For most genotypes, storage did not affect firmness, except A-2491 had decreased firmness from 6.7 N at harvest to 4.0 N after storage (Table 4). Storage did not affect the development of red drupelets. The blackberry genotypes in this study had low red drupelets (<6%). A-2453, the only genotype evaluated with the unique crispy texture, had the lowest percent of red drupelets (0.7%) (Table 2). Color reversion for red drupelets formation can occur due to rupture of cell membranes and walls, causing leakage from vacuole to cytoplasm that changes the pH and the stability of anthocyanins [12]. Crispy genotypes, like A-2453, maintain the stability of cell membranes during storage resulting in lower incidence of red drupelets when compared to non-crispygenotypes [12].

Weight loss and unmarketable fruit of blackberries stored in clamshells were evaluated after storage. Weight loss was very low (<3%) for all the genotypes. A-2491 had the highest weight loss (3.0%) and A-2453 the lowest (0.2%) (Fig. 1). Another study found that ‘Ouachita’, ‘Natchez’ and ‘Navaho’ had 2–4% weight loss when stored at 2°C for 15 days [15]. There were large differences in the genotypes in terms of unmarketable fruit. ‘Natchez’ had the lowest percent of unmarketable fruit (2.9%), demonstrating the potential for ‘Natchez’ as a fresh-market fruit, while A-2418 had the highest unmarketable fruit (53.6%) (Fig. 2). Weight loss was negatively correlated to firmness (r = –0.68), so the more firm the blackberry, then the more potential for postharvest storage. Unmarketable fruit was positively correlated to leakage (r = 0.91) and decay (r = 0.61).

OPEN IN VIEWER

Fig.1

Weight loss (%) after 7 days of storage at 2°C for blackberry genotypes in 240-g, low profile, vented clamshells, Clarksville, AR, 2015. Means with different letter(s) are significantly different (p < 0.05) using Tukey’s HSD test.

OPEN IN VIEWER

Fig.2

Unmarketable fruit (%) after 7 days of storage at 2°C for blackberry genotypes in 240-g, low profile, vented clamshells, Clarksville, AR, 2015. Means with different letter(s) are significantly different (p < 0.05) using Tukey’s HSD test.

At harvest, the blackberries had a range of soluble solids (6.6–10.9%), pH (3.1–3.9) titratable acidity (0.5–1.3%), and SS/TA ratio (4.7–19.5). Other research has shown similar data with soluble solids of blackberries from 6.2–11.1% and pH from 2.5–4.1 [25]. Although some research has shown that soluble solids levels decreased [26] and some showed levels that increased [13] during storage of blackberries. A slight increase in soluble solids was expected, since weight loss in blackberries concentrates cell sap [13]. In this study, the level of soluble solids was not affected by storage. A-2434 and A-2491 had the highest levels of soluble solids, 10.4% and 10.6%, respectively, while A-2418 had the lowest (6.9%). However, storage induced an increase in pH and a reduction of titratable acidity. At harvest, genotypes had a pH of 3.3 and after 7 days of storage had pH of 3.4. A-2453 had the highest pH (3.8), while the pH of other genotypes ranged from 3.1 to 3.4 (Table 2).

There was an interaction between genotype and storage for titratable acidity (p = 0.0287) and SS/TA ratio (p = 0.0157). After 7 days of storage, the titratable acidity decreased by 28% for A-2418 (from 1.5 to 1.1%) and A-2434 (from 1.5 to 1.0%). The titratable acidity decreased by 40% (from 1.2 to 0.7%) in ‘Ouachita’, but increased by 28.0% (from 0.7 to 1.1%) in ‘Osage’ (Table 4). Reduction of 10.0% to 30.0% in titratable acidity was observed in shiny-black blackberries during storage for ‘Shawnee’, ‘Cheyenne’ and ‘Choctaw’, and an increase of 30.0% in ‘Navaho’ [13]. Preferential use of organic acids for respiration may have decreased the titratable acidity [26]. Genotypes with higher respiration rates may have higher degradation of organic acids. SS/TA ratio decreased 30.0% for A-2453 (from 19.5 to 13.6) and 51.0% for ‘Osage’ (from 15.2 to 7.5), and increased 53.0% for ‘Ouachita’ (from 7.3 to 15.4) (Table 4).

The phytochemical levels varied among the genotypes but all increased after 7 days of storage at 2°C (Table 2). Before storage, the level of total phenolics was 382.2 mg GAE/100 g, the level of anthocyanins was 206.0 mg C3GE/100 g, the level of flavonols was 7.7 mg RE/100 g, and the level of ellagitannins was 27.1 mg EAE/100 g. After storage, the level of total phenolics increased 17.0%, anthocyanins 9.0%, flavonols 53.0% and ellagitannins 13.0%. It is possible that, in this study, stress conditions such as low temperature or pathogen attack during the postharvest storage could have increased these compounds. Postharvest storage temperature can affect the level of anthocyanins in small fruits, such as raspberries, strawberries, cherries, sour cherries and red currants [27]. The storage of blackberries for 15 days at 2°C, increased the level of total anthocyanins, total flavonols and total phenolics by 3.2%, 10.4% and 8.7%, respectively in ‘Natchez’, 38.0%, 19.8% and 32.0%, respectively in ‘Ouachita’, and 48.0%, 33.6% and 59.8%, respectively in ‘Navaho’, suggesting that genetic background may play an important role in phytochemical compound changes during storage [15]. After 7 days of storage at 2°C, the content of total phenolics increased in ‘Arapaho’ (289 to 323 mg GAE/100 g) and ‘Shawnee’ (175 to 364.6 mg GAE/100 g) and remained the same for ‘Choctaw’ and ‘Apache’, while total anthocyanins decreased in ‘Arapaho’ (144 to 119 mg C3GE/100 g) and increased in ‘Navaho’ (136 to 158 mg C3GE/100 g) [28]. In oranges, storage at 4°C increased the level of anthocyanins eight times compared to fruit stored at 25°C. The low temperature led to activation of phenylalanine ammonia lyase (PAL), chalcone synthase (CHS), dihydroflavonol 4-reductase (DFR) and UPD-glucose flavonoid glucosyl transferase (UFGT) genes involved in the biosynthesis of anthocyanins and suggests that genes of the anthocyanin biosynthesis pathway in oranges can be considered cor (cold-regulated) genes [29].

Selection A-2450 had the highest level of total phenolics and total anthocyanins, 502.8 mg GAE/100 g and 271.8 mg C3GE/100 g, respectively, while ‘Prime-Ark^® Traveler’ had the lowest levels (336.3 mg GAE/100 g and 124.4 mg C3GE/100 g). ‘Prime-Ark^® 45’ had the highest level of total flavonols (14.8 mg RE/100 g) and ‘Ouachita’ the lowest (6.3 mg RE/100 g). ‘Natchez’ had the highest level of total ellagitannins (42.8 mg EAE/100 g) and A-2453 the lowest (18.2 mg EAE/100 g) (Table 2). Blackberries showed a wide variation in their biochemical composition due to genotype differences, total phenolics ranged from 682 to 1040 mg GAE/100 g for total phenolics, total anthocyanins from 131 to 256 mg C3GE/100 g, total flavonols from 4 to 12 mg RE/100 g and total ellagitannins from 8 to 27 mg EAE/100 g [30].

At harvest, total anthocyanins and total flavonols were positively correlated to total phenolics (r = 0.71 and r = 0.68, respectively). Interestingly, red drupelet incidence (%) was negatively correlated with total phenolics (r = –0.77) and total flavonols (r = –0.66). After storage, only total flavonols were correlated with total phenolics (r = 0.68).

3.3 Sensory analysis

The descriptive sensory panelists evaluated appearance, texture, basic tastes, feeling factors and overall aromatic impact of the blackberry genotypes at harvest and after storage. The panelists did not detect differences in the color and uniformity of color of the blackberries. On the 15-point scale, the blackberries were scored 13.2–13.7, very dark and uniform. The intense, almost black color of blackberries can be correlated to composition and high levels of various anthocyanins [31]. A-2453 was the glossiest genotype, and ‘Prime-Ark^® 45’ was the least glossy. Glossiness of the blackberries was not impacted by storage.

Storage did not impact the hardness of the blackberries, but hardness varied by genotype with ‘Natchez’, ‘Ouachita’, ‘Prime-Ark^® Traveler’ and A-2491 as the softest and A-2453 the firmest. There was an interaction between genotype and storage for the popping attribute (p = 0.0209). Panelists identified a decrease of the popping attribute for the genotypes A-2434, A-2450, A-2491, ‘Natchez’ and ‘Prime-Ark 45^®’ after 7 days of storage (Table 5).

In terms of basic tastes, there were interactions between genotype and storage for the sweet (p = 0.0279) and sour attributes (p < 0.0001), but the interactions were not significant for bitter basic taste, the feeling factors or the overall aromatic impact. Before storage, panelists identified A-2491 as the sweetest (soluble solids 10.9%) and A-2418 as the least sweet (soluble solids 6.6%) and found the sweetness of A-2491 and ‘Natchez’ decreased during storage. In terms of sourness after storage, A-2434, A-2450, A-2491, ‘Natchez’, ‘Prime-Ark^® 45’ and ‘Prime-Ark^® Traveler’ showed a reduction of sourness, while ‘Osage’ showed an increase. After storage, A-2418 was the sourest, while A-2491 the least sour. These findings correspond with our analytical results that showed variation in titratable acidity for the genotypes after 7 days of storage. Panelists did not find differences among genotypes for bitterness, but found that berries were bitterer before storage, but overall bitterness was low (<2.2 on the 15-point scale).

Astringent feeling factor varied among genotypes and increased after storage. It is interesting to note that panelists identified A-2453 as the least astringent and A-2450 as the most astringent, since A-2453 had the lowest content of ellagitannins and A-2450 the highest content of total phenolics and anthocyanins. Astringent feeling factor is related to phenolic compounds such as anthocyanins [32]. The metallic feeling factor (biting into tin foil as a reference) of the blackberries was very low. There was not a difference between genotypes found, but the metallic feeling increased after storage.

The intensity of overall aromatics ranged from 7.0 to 8.3 with A-2491 having the highest intensity, but the overall aromatics were impacted by storage with a decrease from 7.9 to 7.6 in the intensity of overall aromatics after 7 days of storage. Soluble solids were positively correlated with sweet basic taste and overall aromatic impact (r = 0.74 and 0.68, respectively).

Principal component analysis was used to segregate genotypes into different organoleptic groups (Table 6). The perception of sensory attributes and analytical attributes was reduced to three principal components, which explained 71.1% and 72.2% of data variance, at harvest and after storage, respectively.

At harvest, the positive loadings in component 1 were astringent feeling factor, metallic feeling factor and titratable acidity associated with APF-268 and A-2434, while the negative loadings were pH and SS/TA ratio associated with A-2453. The positive loadings in component 2 were sour taste, popping, total flavonols and total phenolics associated with A-2450 and ‘Prime-Ark^® 45’, while the negative loadings were red drupelets incidence and color associated with ‘Ouachita’. The positive loadings in component 3 were sweet taste, overall aromatic impact, uniformity of color and soluble solids associated with A-2491 and ‘Natchez’, while the negative loading was hardness associated with A-2418.

After storage, the positive loadings in component 1 were sour taste, astringent feeling factor, metallic feeling factor and titratable acidity associated with A-2418 and A-2450, while the negative loadings were sweet taste, soluble solids and SS/TA ratio associated with A-2491. The positive loadings in component 2 were glossiness, hardness and popping associated with A-2453, while the negative loadings were bitter taste, overall aromatic impact, total flavonols and total ellagitannins associated with ‘Prime-Ark^® 45’ and ‘Natchez’. The positive loadings in component 3 were color, uniformity of color, total phenolics and total anthocyanins associated with A-2453 and ‘Natchez’, while the negative loading was red drupelet incidence associated with ‘Ouachita’.

The shifts in the multidimensional sensory and analytical dimensions between 0 and 7 days of storage indicate the need for examining changes in genotypes over time. For example, at day 0, sweetness and soluble solids are more strongly associated with uniformity of color and overall aromatic impact; the negative correlation with sourness was less pronounced at day 0 than at day 7. In contrast, in the day 7 principal component analysis solution for component 1, sour taste, astringent feeling, metallic feeling and titratable acidity are negatively associated with sweet taste, soluble solids, and SS/TA ratio. Sourness and titratable acidity of many cultivars decreased over time, which partly drove the changes in the sensory and analytical space. Higher fruit quality is associated with higher sweetness, lower sourness and lower bitterness [23]. Since lower sourness and higher sweetness are associated with higher consumer liking, blackberries like A-2491, a key sample on the sweetness dimension, may be more desirable after 7 days of storage than at harvest. As another example, bitter taste is more strongly associated with total flavonols and total ellagitannins on component 2 at day 7. Since higher bitterness is associated with lower consumer liking [23], A-2453, a key sample, which is negatively associated with high flavonols and bitter taste may be liked more than other cultivars which are more bitter.