- Email: [email protected]
Acidity and taste in kiwifruit
Postharvest Biology and Technology 32 (2004) 159–168
Acidity and taste in kiwifruit K. Marsh∗ , S. Attanayake, S. Walker, A. Gunson, H. Boldingh, E. MacRae The Horticulture and Food Research Institute of New Zealand Ltd., Private Bag 92169, Auckland, New Zealand Received 26 March 2003; accepted 8 November 2003
Abstract Although total titratable acidity levels in ‘Hayward’ kiwifruit appear quite stable during storage at 0 ◦ C under New Zealand conditions it is known that citric acid levels decline but malic acid levels are maintained. By contrast, malic acid levels tend to increase with storage at 4 ◦ C. These observations formed the basis of a sensory comparison of fruit stored at 0, 4, and 10 ◦ C which were also analysed for acidity, sweetness, and the individual acid concentrations during 6 weeks of storage. The fruit were ripened to equivalent firmness, and presented to sensory panellists. Acid perception increased at 4 ◦ C, which correlated with an increase in malic acid concentration, but also with a decline in sweetness perception, soluble solids content, and dry matter in fruit stored at 4 and 10 ◦ C compared to fruit stored at 0 ◦ C. Although fruit were presented to panellists at equivalent fruit firmness (4.4–7.3N), the results were further confounded by changes in perceived texture in fruit held at the warmer temperatures. Ratings for core firmness, flesh firmness, and fibrous flesh texture increased in the fruit at 4 and 10 ◦ C compared to fruit at 0 ◦ C. Storage at 4 and 10 ◦ C also led to an increase in stalky, woody flavours compared to storage at 0 ◦ C. Findings for an increase in the perception of acidity from fruit stored at 4 ◦ C could not be unequivocably attributed to the changes in acidity occurring during storage. However, they were consistent with findings from pulp experiments, where malic and citric acid had been added to ‘Hayward’ kiwifruit pulp. The results had implications for the storage of ‘Hayward’ kiwifruit at less than optimum fruit temperatures and raised interesting questions about the changes in metabolism occurring in fruit at the three temperatures studied. © 2003 Elsevier B.V. All rights reserved. Keywords: Sugar; Acid; Actinidia; Storage; Malate; Citrate; Titratable acidity; Sensory assessment
1. Introduction Consumer preference for kiwifruit (Actinidia deliciosa (A. Chev.) C.F. Liang and A.R. Ferguson var. deliciosa ‘Hayward’) is determined primarily by the sugar–acid balance with fruit firmness and fruit volatile content causing a large moderating effect (Jaeger et al., 2003). A rapid loss of starch and in∗ Corresponding author. Tel.: +64-9-815-4200x7178; fax: +64-9-815-4202. E-mail address: [email protected] (K. Marsh).
crease in soluble solids content provides one of the main criteria for harvest of fruit in New Zealand, but fruit also accumulate acids during growth to produce the characteristic acidity in ‘Hayward’ fruit (Okuse and Ryugo, 1981; Walton and De Jong, 1990). At harvest, kiwifruit contain 0.9–2.5% total acidity, with 40–50% as citrate 40–50% as quinate, and 10% as malate. There are at least three distinct tissue zones within a kiwifruit and the balance of the different acids changes within these zones, so sampling is an important consideration for fruit acidity measurements (MacRae et al., 1989a). The proportion
0925-5214/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.postharvbio.2003.11.001
160
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168
of citrate is highest in the inner cortex, quinate is highest in the outer cortex, whereas the core has the lowest total acid content (about half the other zones), predominantly citrate. During storage and ripening the titratable acidity in ‘Hayward’ kiwifruit changes little (Matsumoto et al., 1983; MacRae et al., 1989b) or declines (Ben-Arie et al., 1982; Crisosto and Crisosto, 2001; Tombesi et al., 1993) depending on growing location. In Israel, California and Italy, high acidity was recorded at harvest (2–2.5%), which dropped sharply (0.5–1.5%) after 1–2 months of storage. In New Zealand, however, acidity was lower at harvest (∼1.4%) and remained stable during storage and subsequent ripening (MacRae et al., 1989b). In particular, MacRae et al. (1989b) found the proportion of citrate declined steadily, whilst malate levels tended to be quite stable during storage at 0 ◦ C. By contrast, storage at 4 ◦ C led to increased malate concentrations. These fruit, when presented to a consumer panel (MacRae et al., 1990), were perceived as less sweet than similar fruit stored at 0 ◦ C, although SSC was similar for both sets of fruit. When a standardised kiwifruit pulp was used (Rossiter et al., 2000; Marsh et al., 2003), consumers could detect additions of sugars and acids at levels found normally within the range of fruit available for purchase. In particular, while addition of all acids increased perception of acidity, addition of quinate also affected perception of ‘Hayward’ flavour and banana-like flavours. While this is one of the first reports of the taste characteristics of quinic acid, several workers have shown that citric acid is less sour than malic acid at similar molarity (mM H+ ) (Noble et al., 1986; Hartwig and McDaniel, 1995). If pulps are to be used to guide selection of new cultivars, and to predict sugar, acid, and fruit volatile compositions which might be found desirable by consumers, it would be useful to test the validity of at least one result in intact fruit. The results obtained in the above experiments (MacRae et al., 1989b, 1990) suggest that it is possible to manipulate acid composition in kiwifruit that are otherwise similar by storing the fruit at 4 ◦ C. We report here, the outcome of such an attempt and show that although acidity was affected, the results for fruit taste were confounded by alterations in other metabolite and texture characteristics with storage temperature.
2. Methods and materials 2.1. Fruit material Fruit from three orchards (designated orchard 1, 2, and 3) in the Bay of Plenty, New Zealand were harvested and commercially packed into 13 single layer trays (with poly-liners) for each orchard. The trays (36 fruit per tray) were transported immediately to HortResearch, Auckland. A ten fruit subsample of each orchard was taken for at-harvest assessment of fruit firmness (Lallu et al., 1989), juice soluble solids (%SSC) as an average of samples from stem and distal ends using a refractometer (Model ATC-1E, AtagoTM , Japan), juice pH (an equal mix from stem and distal ends using an ISFET electrode) (KS701, ShindingenTM ), dry matter (%DM; MacRae et al., 1989a) and the titratable acidity of fruit flesh (%TA; MacRae et al., 1989b). Samples (5 g) were also taken for acid analysis (see below). The 12 remaining trays of fruit for each orchard were divided into 3 different storage temperatures (0, 4, or 10 ◦ C) and stored for a period of 7 weeks. Acidity (pH, %TA), SSC and fruit firmness were assessed on a sample of one fruit per tray at 2-weekly intervals. After 7 weeks cool storage trays of fruit were held at 20 ◦ C for a further 2 weeks and fruit firmness monitored. As the fruit in each tray approached a mean firmness of 8N, the fruit were placed in store at 0 ◦ C and held for presentation to a trained panel of assessors (1–14 days). Although fruit from all orchards softened most quickly at 10 ◦ C, less quickly at 4 ◦ C and least quickly at 0 ◦ C, we were able to obtain 10 replicate sets of fruit from each storage temperature which were matched for fruit firmness (4.4–7.3N) from orchard 1 (designated line 1) on each of the two assessment days. Line 2, was a combination of five matched sets of fruit from orchard 2, and five matched sets of fruit from orchard 3. On the second tasting day (4 days later) similar sets of fruit were obtained and each panellist assessed fruit from the same orchard × temperature combinations as on tasting day 1. 2.2. Sensory panels Panellists experienced in descriptive analysis of kiwifruit participated in eight 1 h training sessions concentrating on descriptive analysis of ‘Hayward’ fruit,
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168
on days prior to testing. Over the course of the training, definitions of the odour, flavour and textural attributes found were generated and appropriate reference standards were established. These included: grassy door (cis-3-hexenel 40 l l−1 ), sulphur odour (cooked egg white), vomit odour (butyric acid 50 g l−1 ), fruit candy odour (ethyl-hexanoate 700 l l−1 ), acid taste (1 g l−1 malic acid), sweet taste (20 g l−1 sucrose), and fibrous flesh texture (canned pineapple). All attributes were assessed using 150 mm unstructured line scales anchored at 0, absent and 150, extreme. Fruit from each storage temperature, from both lines, were presented to 10 panellists. The firmness of all the fruit (4.4–7.3N) was recorded and each panellist was presented with a set of three fruit from a particular orchard which were accurately matched for firmness (e.g. 4.4–5.2N). Testing took place over 2 days with each panellist assessing 6 fruit on each test day (three fruit from each line)—using a design to reduce order and carry-over effects but which presented the fruit from the same line together. Ten-minute breaks were enforced after the first three samples to reduce panellist fatigue. Samples were presented to panellists monadically and were placed stem end up in individual three-digit labelled plastic cups. Assessors were directed to cut fresh slices from the top of the fruit for odour assessment, the middle section of the fruit for flavour assessment and the remaining third of the fruit for texture assessment. Testing was carried out in sensory booths which were maintained at 20 ◦ C with a positive airflow to remove odours from the testing area. MicroleneTM filtered water and plain water crackers were provided as palate cleansers. Prior to presentation, the ends of the fruit were sampled for SSC and juice pH, and two thirds (longitudinal) of the remainder presented to assessors. The final third of the fruit (excluding the penetrometer puncture) was sampled for DM (equatorial region) then the remainder was snap frozen in liquid N2 for subsequent analysis of TA and organic acid content. 2.3. Acid analysis TA was measured on a 5 g sample of frozen tissue, which had been macerated in 25 ml of distilled water using a polytron (KinematicaTM , Luzern, Switzerland) and by titration to pH 8.2 with 0.1N NaOH using an automatic titrator (716 DMS Titrino, Metrohm, Herisau, Switzerland). The same fruit tissue was sub-
161
sampled for organic acid composition (1 g of frozen tissue in 10 ml of 80% methanol), homogenised with a polytron (KinematicaTM ), and centrifuged (5000 × g) to separate tissue from solution. The supernatant was then used for measurement of acid content. Citrate and malate were measured enzymatically by adapting methods of Velterop and Vos (2001) for use with a 96-well microplate spectrophotometer (SpectraMax Plus 384, Molecular Devices). Reagents were provided by Boehringer–Mannheim GmbH. In addition, the same extracts were analysed by gas liquid chromatography (GLC) using a DB1701 column (J&W Scientific). Samples were prepared using a modification (micro-column) of the protocol described by Redgwell (1980) with tartaric acid used as an internal standard for quantification. 2.4. Data analysis Data were subjected to a repeated measures analysis of variance (ANOVA) to determine if differences between treatments existed. Tukey’s least significant differences (LSDs) were calculated to compare differences between means following a significant ANOVA effect. All effects and interactions were evaluated at the 5% level.
3. Results 3.1. Fruit composition at harvest and during storage Fruit from the three orchards had similar compositions at harvest (Table 1), except orchard 3 appeared to have slightly higher TA and DM levels. All orchards had similar proportions of the three principle acids as analysed by gas liquid chromotography (Table 2), but Table 1 Harvest characteristics of ‘Hayward’ kiwifruit from three orchards (n = 10) Orchard
Fruit firmness
SSC (%)
Juice pH
TA (%)
DM (%)
1 2 3
71.7N aa 71.1N a 74.0N a
7.31 a 6.82 a 6.58 a
3.24 a 3.19 a 3.21 a
1.25 a 1.33 ab 1.48 b
16.2 ab 15.8 b 16.8 a
a Means in the same column, followed by the same letter are not significantly different at the 5% level (Tukey’s LSD).
Orchard Malate Malatea 1 2 3
1.4 ab 0.8 b 1.7 a
1.7 a 1.0 b 1.7 a
Quinate Citrate 8.2 a 7.8 a 7.4 a
Citratea
9.9 b 10.1 b 11.3 ab 12.6 a 12.6 a 11.3 ab
Total (GC) 19.5 a 19.8 a 21.6 a
a Individual acid measured by the enzymatic method shown in this column. b Means in the same column, followed by the same letter are not significantly different at the 5% level (Tukey’s LSD).
citrate was lower in fruit from orchard 1, and malate was low in fruit from orchard 2. Subsequent sampling of the fruit trays showed orchard 2 to have similar malate levels to the other two orchards (Fig. 3). Therefore, this appears to be due to the low sample numbers and chance selection of low malate containing fruit. Two methods were used to measure acid concentrations and there was a good correlation between malate concentrations (r 2 = 0.63, see mean values in Table 2), which was not quite so good for citrate (r2 = 0.40). The fruit stored at different temperatures softened as expected, with fruit held at 10 and 4 ◦ C softening faster than fruit held at 0 ◦ C (Fig. 1). There was no change in TA during the storage period at any of the three temperatures (Fig. 2) However, malate remained at similar levels (orchard 1) or increased (orchards 2 and 3) in fruit held at 4 ◦ C (Fig. 3). By contrast, citrate decreased during storage at 0 and 4 ◦ C, and remained at similar levels or decreased slightly at 10 ◦ C (Fig. 4). 3.2. Fruit composition when consumed SSC and DM decreased in fruit that had been stored at 4 and 10 ◦ C compared to fruit stored at 0 ◦ C for both fruit lines tasted (Table 3). In contrast, TA did not vary between treatments except for fruit from line 1 which had slightly reduced levels of TA at 0 ◦ C compared to fruit from other temperatures (Table 3). The acid concentrations varied between treatments (Table 4). Malate concentrations were significantly higher in fruit which had been stored at 4 ◦ C compared to 0 ◦ C for both fruit lines (Table 4), with fruit stored at 10 ◦ C having somewhat intermediate malate
80
60
40
20
(a)
0 0
2
4
6
0
2
4
6
0
2
4
6
80
Fruit firmness (N)
Table 2 Mean acid levels (g kg−1 ) measured by gas chromatography in fresh fruit from three orchards (n = 10)
Fruit firmness (N)
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168
(b)
60
40
20
0
80
Fruit firmness (N)
162
60
40
20
0
(c)
Time in storage (weeks) o
0C
o
4C
o
10 C
Fig. 1. Changes in fruit firmness for ‘Hayward’ kiwifruit sampled from trays of fruit stored at three different temperatures for (a) orchard 1, (b) orchard 2 and (c) orchard 3.
levels. There were no differences in the citrate levels between fruit stored at 0 or 4 ◦ C. Fruit stored at 10 ◦ C had higher levels of citrate than fruit held at 0 or 4 ◦ C for line 1, but not for line 2. The citrate content made the highest contribution to the total acids, so the total
3.0
-1
1.4
1.2
2.5 2.0 1.5 1.0
1.0
(a)
163
3.5
1.6
Malate (g kg )
Titratable acidity (%)
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168
0.0 0
2
4
6
(a)
0.0 0
2
4
6
3.5 3.0
1.4 -1
Malate (g kg )
Titratable acidity (%)
1.6
1.2
1.0 0
2
4
1.5
6
1.6
(b)
0.0 0
2
4
6
0
2
4
6
3.5 1.4
3.0 -1
1.2
1.0 0.0 0
(c)
2.0
1.0
0.0
Malate (g kg )
Titratable acidity (%)
(b)
2.5
2
4
6
2.5 2.0 1.5 1.0
Time in storage (weeks) 0.0 o
0C
o
4C
o
10 C
Fig. 2. Changes in total titratable acidity (TA) for ‘Hayward’ kiwifruit sampled from trays of fruit stored at three different temperatures for (a) orchard 1, (b) orchard 2 and (c) orchard 3.
acid content was also significantly higher in the fruit stored at 10 ◦ C in line 1. There were no differences in the quinate concentrations observed in fruit stored at different temperatures and ripened to equivalent firmness (Table 4).
(c)
Time in storage (weeks)
0C
4C
10C
Fig. 3. Changes in malate concentration (determined by enzymatic method) for ‘Hayward’ kiwifruit sampled from trays of fruit stored at three different temperatures for (a) orchard 1, (b) orchard 2 and (c) orchard 3.
164
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168 Table 3 Key characteristics of ‘Hayward’ kiwifruit presented to sensory panellists (each fruit line is averaged over both days; n = 20)
-1
Citrate (g kg )
14
12
10
Fruit
Storage temperature (◦ C)
DM (%)
SSC (%)
TA (%)
Line 1
0 4 10
16.6 aa 15.8 b 15.5 b
13.9 a 13.1 b 13.0 b
1.27 b 1.34 a 1.31 ab
Line 2
0 4 10
16.5 a 15.3 b 14.6 b
14.0 a 12.5 b 11.7 b
1.37 a 1.33 a 1.33 a
8
6
(a)
a
0 0
2
4
6
-1
Citrate (g kg )
14
seemed to have been balanced by a decrease in other acid components of the fruit. The decline in citrate content in fruit stored at 4 ◦ C is larger on a molar basis than the gain in malate, so it would be expected to have the largest effect on titratable acidity. Differences in acid concentration can also occur with no change in titratable acidity if the acid is in the basic form, e.g. potassium citrate.
12
10
8
6
(b)
Means within each fruit line followed by the same letter are not significantly different at the 5% level (Tukey’s LSD).
0 0
2
4
6
3.3. Sensory perception of the fruit
-1
Citrate (g kg )
14
For both fruit lines and on two separate testing days, the trained sensory panel gave higher values for acid taste, for grassy odour, and lower values for sweetness in the fruit stored at 4 ◦ C compared to those stored at 0 ◦ C (Tables 5 and 6). There were some small differences between testing days, but these affected the significance of the effects, and not the overall trends. Data for the mean effects over 2 days for each of the
12
10
8
6 0 0
(c)
2
4
6
Time in storage (weeks)
0C
4C
10C
Fig. 4. Changes in citrate concentration (determined by enzymatic method) for ‘Hayward’ kiwifruit sampled from trays of fruit stored at three different temperatures for (a) orchard 1, (b) orchard 2 and (c) orchard 3.
The differences in malate concentration are in the order of 1 g kg−1 which should result in 0.1% increase in titratable acidity and this occurred in line 1, but not in line 2. For line 2, changes in titratable acidity
Table 4 Extractable acid content (g kg−1 ) of ‘Hayward’ kiwifruit presented to sensory panellists sampled and analysed by gas chromatography (each fruit line is averaged over both days; n = 20) Fruit
Storage temperature (◦ C)
Malate
Line 1
0 4 10
1.50 ba 2.53 a 2.42 a
8.3 b 8.0 b 9.6 a
7.2 a 6.8 a 7.7 a
17.0 b 17.4 ab 19.7 a
Line 2
0 4 10
1.45 c 2.43 a 1.84 b
10.3 a 8.9 a 10.0 a
8.5 a 6.9 a 7.6 a
20.2 a 18.3 a 19.5 a
Citrate Quinate Total
a Means within each fruit line followed by the same letter are not significantly different at the 5% level (Tukey’s LSD).
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168 Table 5 The mean scores for trained panellists assessing the sensory attributes (0: absent, 150: extreme) of ‘Hayward’ kiwifruit from line 1 over two testing days after storage at different temperatures Attribute
Storage temperature (◦ C) 0
Fruit candy door Sulphur door Grassy door Vomit door Flesh firmness Core firmness Sweetness Acidity Juiciness Metallic flavour Characteristic Hayward flavour Woody/stalky flavour Metallic aftertaste Astringent aftertaste Ease of breakdown Fibrous flesh texture Gelatinous texture
4
10
Attribute
Fruit candy door Sulphur door Grassy door Vomit door Flesh firmness Core firmness Sweetness Acidity Juiciness Metallic flavour Characteristic Hayward flavour Woody/stalky flavour Metallic aftertaste Astringent aftertaste Ease of breakdown Fibrous flesh texture Gelatinous texture
aa b b a b b a b a a a
8.9 b 47.9 a 46.8 a 24.3 b 56.8 a 52.7 a 70.4 b 83.4 a 58 b 39.5 a 59 b
17.2 44.4 41.3 19.7 58.8 59.6 74.1 70.7 58.2 44.7 63.6
b a ab b a a b ab b a b
23.3 20.0 15.9 23.3 11.8 16.9 15 15 12.7 12.3 13.6
10.8 28.5 10.9 50.9 15.7 43 a
b a a b b
17.3 26.4 10.2 76 a 23.9 18.6
19.7 32.2 14.7 76 a 32.3 20.4
a a a
8.6 12.8 12.8 14.1 14.3 13.5
ab a a ab b
a b
Table 6 The mean scores for trained panellists assessing the sensory attributes (0: absent, 150: extreme) of kiwifruit from line 2 over two testing days after storage at different temperatures
LSD (Tukey)
82.9 21.3 29.3 51.5 27.4 23.2 89.1 60.9 78.8 41.7 79.6
165
Storage temperature (◦ C)
LSD (Tukey)
0
4
10
40.8 aa 28 b 31.8 b 43.6 a 39 b 25.2 b 78.8 a 74 b 70.9 a 46 a 66.6 a
9.2 b 43.5 a 46.3 ab 19.2 b 64 a 63.2 a 64 b 89.7 a 58.7 a 42.4 a 58.3 a
16.5 b 37.1 ab 50.8 a 37 ab 49.4 ab 49.3 a 64.1 b 77.4 ab 60.9 a 46.5 a 58.7 a
14.3 15.3 16.0 24.2 14.9 22.9 13.1 14.8 13.3 14.1 12.3
13.7 29.2 13.5 55.4 14.6 31.6
21.7 28.4 17.4 82.2 34 a 11.8
19.2 a 29.2 a 16 a 70.6 a 27.6 ab 17 b
10.8 10.1 10.1 13.6 17.2 10.1
a a a b b a
a a a a b
Fruit were assessed within the firmness range 4.4–7.3N. a Means in the same row followed by the same letter are not significantly different at the 5% level (Tukey’s LSD).
Fruit were assessed within the firmness range 4.4–7.3N. a Means in the same row, followed by the same letter are not significantly different at the 5% level (Tukey’s LSD).
fruit lines is reported (Tables 5 and 6) (Fig. 5) and there is a consistency in the results for the two fruit lines. For acidity, the fruit stored at 10 ◦ C were intermediate between the 4 and 0 ◦ C fruit and not significantly different from either group, but for many other parameters (including sweetness) the fruit at 10 ◦ C gave results similar to those at 4 ◦ C. Parameters such as DM and SSC, which are generally very highly correlated with sweetness character of the fruit, decreased with fruit temperature in measurements made on the tasted fruit (Table 3). The principal components analysis featured in Fig. 5 highlights the dependence of fruit taste on several aspects of the fruit. The fruit stored at 0 ◦ C cluster in an area associated with high sweetness, and typical Hayward flavour. By contrast, fruit stored at 4 ◦ C cluster in a region of higher acidity and less sweetness. The first principal component correlates with sweetness/DM and the second with the perception of acidity and individual acid values. Although the design of the sensory panels ensured that fruit were balanced for firmness (penetrometer measurement) the fruit from different temperatures
were perceived by sensory panellists as being very different in texture by the sensory panellists. Ratings for core firmness, flesh firmness, and fibrous flesh texture increased in the fruit at 4 and 10 ◦ C compared to fruit at 0 ◦ C, which correlates with our empirical observations of rubbery textures in fruit stored at temperatures greater than 0 ◦ C. Greater rubberiness has been reported previously in fruit stored at 4 ◦ C when compared to fruit stored at 0 ◦ C (MacRae et al., 1990). Storage at 4 and 10 ◦ C also tended to lead to an increase in stalky, woody flavours compared to storage at 0 ◦ C (Table 3).
4. Discussion We have clearly demonstrated that kiwifruit stored at temperatures above 0 ◦ C undergo different metabolism compared to those stored at 0 ◦ C. This resulted in fruit with quite different sensory characteristics. However, we were unable to manipulate fruit to show altered acid metabolism alone. Fruit stored
166
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168
line 1 malate
2
TA citrate quinate Acidity DM
0
Stalky/woody flavour
Brix
pH
-2
Sweetness Hayward flavour Brix/Acid ratio
-4
Second principal component 21.2 %
temp 0 temp 4 temp 10
-2
0
2
4
First principal component 24.7 %
malate citrate
2
quinate Acidity
TA
Stalky/woody flavour
0
Hayward flavour Sweetness pH
DM Brix
-2
Second principal component 21.2 %
4
line 2 temp 0 temp 4 temp 10
Brix/Acid ratio
-4
-2
0
2
First principal component 24.7 %
Fig. 5. A principal components comparison of sensory characteristics and analytical measures for fruit from two orchard lines stored at three different temperatures.
at 4 ◦ C were 1–2% lower in SSC than those at 0 ◦ C and there was a slight increase in TA measured in fruit line 1 stored at 4 ◦ C but not in fruit line 2. The malate content showed a consistent increase at 4 ◦ C compared to 0 ◦ C in both fruit lines. Trained assessors were able to distinguish between fruit held at different
storage temperatures, with the most acid, least sweet, fruit being defined as those held at 4 ◦ C, and the least acid, most sweet, fruit being those held at 0 ◦ C (and 10 ◦ C). These results could equally be attributed to a decrease in SSC, or to an increase in malate content. Interestingly, the SSC for fruit stored at 4 and 10 ◦ C is similar but some differences in acid content occurred between fruit stored at these two temperatures. Based on previous research with whole fruit (MacRae et al., 1989a,1990; Paterson et al., 1991; Crisosto and Crisosto, 2001) an increase in SSC of fruit due to later harvests from the same location and growing condition is normally detected by assessors as having increased acceptability, and flavour intensity. Fruit from the same source in our experiments had higher SSC at ripeness when stored at 0 ◦ C compared to 4 ◦ C. This could explain the perceptions of greater sweetness and reduced acidity in the fruit stored at 0 ◦ C compared to the other storage regimes. In experiments using fruit pulps adulterated with sugars, higher SSC (with similar TA and acid concentrations) reduced perception of acidity and increased flavour acceptability (Rossiter et al., 2000). An alternative interpretation of our results, is that the greatest increase in perception of acidity occurred with fruit stored at 4 ◦ C. This fruit consistently had 50–100% more malate than fruit stored at 0 ◦ C. Although it is difficult to separate this increase in malate from confounding factors such as the decline in fruit dry matter, or changes in fruit texture, we had very consistent responses from trained panellists. The rise in off-flavours, and acid taste were consistent with experiments using ‘Hayward’ pulp with malate added (Marsh et al., 2003) and a rise in off flavours is also consistent with earlier experiments on fruit stored at 4 ◦ C (MacRae et al., 1990). Studies with fruit pre-sorted for dry matter content indicate that low dry matter fruit have lower acidity, yet the fruit are perceived to be less sweet, more acid (unpublished data). Decreased sweetness may well be driving the increase in acid taste observed in both examples. Alternatively, when malic acid and citric acid were added to fruit pulps at similar moles H+ , malate was significantly more sour tasting than citric acid (Marsh et al., 2003), as has been previously found in model solutions (Noble et al., 1986; Hartwig and McDaniel, 1995). Similar experiments using additions of citric acid to mango homogenates have demonstrated that acid con-
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168
centration affects ratings for sweet and sour and ratings for some flavour characters (Malundo et al., 2001). As with kiwifruit (Paterson et al., 1991) and tomato (Baldwin et al., 1998), TA correlated with and was a useful predictor of sour taste in mango. The combined results of experiments with fruit sorted for dry matter, pulp experiments (Marsh et al., 2003) and the storage experiment reported here show that a strategy of improving flavour in kiwifruit should not include raising malate concentrations. The marked decrease in acidity during storage and/or ripening reported by workers in warmer countries (Ben-Arie et al., 1982; Crisosto et al., 1984; Crisosto and Crisosto, 2001) deserves some attention. Our experience is for a slight initial decline then maintenance of levels of titratable acidity of around 0.8–1.0% during storage/ripening (MacRae et al., 1989b) (Fig. 2, Table 3), but the results described here confirm that some metabolism of acid occurs during this time and that storage temperature can change the balance of the three major acids in the fruit (Table 4). Research on fruit from California has shown that there are generally higher respiration rates recorded in the field for fruit growing at higher temperatures (Walton and De Jong, 1990). Such fruit appear to rapidly metabolise a particular acid pool when fruit are first placed in storage (Crisosto et al., 1984; Crisosto and Crisosto, 2001). Marsh et al. (2000a,b) recently described the significance of tonoplast gradients for acid accumulation. These gradients provide a store of energy which can be utilised via a pyrophosphatase shunt to produce ATP under anaerobic conditions or in the absence of other energy sources. As proton gradients were utilised the pH of the tonoplast and the squeezed juice would increase, and a sharp rise in juice pH during fruit ripening has been recorded in some instances (Matsumoto et al., 1983; Ben-Arie et al., 1982). A hypothesis to explain the different behaviour of fruit harvested from different environments is that the cooler conditions present in New Zealand allow induction of enzymes (e.g. V-PPase) to utilise tonoplast gradients and reduce titratable acidity before harvest. This is in contrast to fruit from warmer climates in which the gradients are maintained until dissipation is triggered by ripening or cool storage (Matsumoto et al., 1983; Crisosto and Crisosto, 2001). There are reports of low input technologies for storing kiwifruit in warmer climates (Sawada et al., 1992;
167
Nanos et al., 1997). Storage at temperatures other than 0 ◦ C will affect fruit flavour in New Zealand fruit, and an increase of 3–4 ◦ C for an extended period will result in decreased sweetness and/or increased perception of acid taste as well as off-flavours. There does, however, seem to be a difference in the ripening behaviour of fruit from warmer climates with respect to acidity such that the alternative storage strategies may be more appropriate for fruit grown in these climates.
Acknowledgements This research was funded by the New Zealand Foundation for Research, Science and Technology Contract No. C06X0006 and C06X0213. Panellist results were collected using Compusense five V.4.2.
References Baldwin, E.A., Scott, J.W., Einstein, M.A., Malundo, T.M., Carr, B.T., Shewfelt, R.L., Tandon, K.S., 1998. Relationship between sensory and instrumental analysis for tomato flavour. J. Am. Soc. Hortic. Sci. 123, 906–915. Ben-Arie, R., Gross, J., Sonego, L., 1982. Changes in ripening parameters and pigments of the Chinese gooseberry (kiwi) during ripening and storage. Sci. Hortic. 18, 65–70. Crisosto, C.H., Crisosto, G., 2001. Understanding consumer acceptance of early harvested ‘Hayward’ kiwifruit. Postharvest Biol. Technol. 22, 205–213. Crisosto, G., Mitchell, G., Arpaia, M.L., Mayer, G., 1984. The effect of growing location and harvest maturity on the storage performance and quality of ‘Hayward’ kiwifruit. J. Am. Soc. Hortic. Sci. 109, 584–587. Hartwig, P., McDaniel, M., 1995. Flavour characteristics of lactic, malic, citric, and acetic acids at various pH levels. J. Food Sci. 60, 384–388. Jaeger, S.R., Rossiter, K.L., Wismer, W.V., Harker, F.R., 2003. Consumer-driven product development in the kiwifruit industry. Food Qual. Pref. 14, 187–198. Lallu, N., Searle, A.N., MacRae, E.A., 1989. An investigation of ripening and handling strategies for early season kiwifruit (Actinidia deliciosa cv. Hayward). J. Sci. Food Agric. 47, 387– 400. MacRae, E.A., Bowen, J.H., Stec, M.H., 1989a. Maturation of kiwifruit (Actinidia deliciosa cv. Hayward) from two orchards: differences in composition of the tissue zones. J. Sci. Food Agric. 47, 401–416. MacRae, E.A., Lallu, N., Searle, A., Bowen, J., 1989b. Changes in the softening and composition of kiwifruit (Actinidia deliciosa) affected by maturity at harvest and postharvest treatments. J. Sci. Food Agric. 49, 413–430.
168
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168
MacRae, E.A., Stec, M.G., Triggs, C.M., 1990. Effects of post-harvest treatment on the sensory qualities of kiwfruit harvested at different maturities. J. Sci. Food Agric. 50, 533– 546. Malundo, T.M.M., Shewfelt, R.L., Ware, G.O., Baldwin, E.A., 2001. Sugars and acids influence flavour properties of mango (Mangifera indica). J. Am. Soc. Hortic. Sci. 126, 115–121. Marsh, K.B., Gonzalez, P.G., Echeverria, E., 2000a. PPi formation by reversal of the tonoplast bound V-PPiase from valencia orange juice cells. J. Am. Soc. Hortic. Sci. 125, 420–424. Marsh, K.B., Gonzalez, P.G., Echeverria, E., 2000b. Partial characterization of H+ -translocating inorganic pyrophosphatase from three citrus varieties differing in vacuolar pH. Physiol. Plant. 111, 519–526. Marsh, K.B., Rossiter, K., Lau, K., Walker, S., Gunson, A., MacRae, E., 2003. Using fruit pulps to explore flavour in kiwifruit. Acta Hortic. 610, 229–238. Matsumoto, S., Obara, T., Luh, B.S., 1983. Changes in chemical constituents of kiwifruit during postharvest ripening. J. Food Sci. 48, 607–611. Nanos, G.D., Kittas, C., Sfakiotakis, E., 1997. In: Porlingis, J. (Ed.), An alternative low-cost storage for kiwifruit. Acta Hortic. 444, 625–628. Noble, A.C., Philbrick, K.C., Boulton, R.B., 1986. Comparison of sourness of organic acid anions at equal pH and titratable acidity. J. Sens. Stud. 1, 1–8.
Okuse, I., Ryugo, K., 1981. Compositional changes in the developing ‘Hayward’ kiwifruit in California. J. Am. Soc. Hortic. Sci. 106, 73–76. Paterson, V.J., MacRae, E.A., Young, H., 1991. Relationships between sensory properties and chemical composition of kiwifruit (Actinidia deliciosa). J. Sci. Food Agric. 57, 235– 251. Redgwell, R.J., 1980. Fractionation of plant extracts using ion-exchange. Sephadex Anal. Biochem. 107, 44–50. Rossiter, K.L., Young, H., Walker, S.B., Miller, M., Dawson, D.M., 2000. The effects of sugars and acids on consumer acceptability of kiwifruit. J. Sens. Stud. 15, 241–250. Sawada, T., Morishima, H., Imou, K., Kawagoe, Y., 1992. Studies on the storage and ripening of kiwifruit (part 1). J. Jpn. Soc. Agric. Mach. 54, 61–67. Tombesi, A., Antognozzi, E., Palliotti, A., 1993. Influence of light exposure on characteristics and storage life of kiwifruit. N.Z. J. Crop. Hortic. Sci. 21, 85–90. Velterop, J.S., Vos, F., 2001. A rapid and inexpensive microplate assay for the enzymatic determination of glucose, fructose, sucrose, l-malate and citrate in tomato (Lycopersicon esculentum) extracts and in orange juice. Phytochem. Anal. 12, 299–304. Walton, E.F., De Jong, T.M., 1990. Growth and compositional changes in kiwifruit berries from three Californian locations. Ann. Bot. 66, 285–298.
Acidity and taste in kiwifruit K. Marsh∗ , S. Attanayake, S. Walker, A. Gunson, H. Boldingh, E. MacRae The Horticulture and Food Research Institute of New Zealand Ltd., Private Bag 92169, Auckland, New Zealand Received 26 March 2003; accepted 8 November 2003
Abstract Although total titratable acidity levels in ‘Hayward’ kiwifruit appear quite stable during storage at 0 ◦ C under New Zealand conditions it is known that citric acid levels decline but malic acid levels are maintained. By contrast, malic acid levels tend to increase with storage at 4 ◦ C. These observations formed the basis of a sensory comparison of fruit stored at 0, 4, and 10 ◦ C which were also analysed for acidity, sweetness, and the individual acid concentrations during 6 weeks of storage. The fruit were ripened to equivalent firmness, and presented to sensory panellists. Acid perception increased at 4 ◦ C, which correlated with an increase in malic acid concentration, but also with a decline in sweetness perception, soluble solids content, and dry matter in fruit stored at 4 and 10 ◦ C compared to fruit stored at 0 ◦ C. Although fruit were presented to panellists at equivalent fruit firmness (4.4–7.3N), the results were further confounded by changes in perceived texture in fruit held at the warmer temperatures. Ratings for core firmness, flesh firmness, and fibrous flesh texture increased in the fruit at 4 and 10 ◦ C compared to fruit at 0 ◦ C. Storage at 4 and 10 ◦ C also led to an increase in stalky, woody flavours compared to storage at 0 ◦ C. Findings for an increase in the perception of acidity from fruit stored at 4 ◦ C could not be unequivocably attributed to the changes in acidity occurring during storage. However, they were consistent with findings from pulp experiments, where malic and citric acid had been added to ‘Hayward’ kiwifruit pulp. The results had implications for the storage of ‘Hayward’ kiwifruit at less than optimum fruit temperatures and raised interesting questions about the changes in metabolism occurring in fruit at the three temperatures studied. © 2003 Elsevier B.V. All rights reserved. Keywords: Sugar; Acid; Actinidia; Storage; Malate; Citrate; Titratable acidity; Sensory assessment
1. Introduction Consumer preference for kiwifruit (Actinidia deliciosa (A. Chev.) C.F. Liang and A.R. Ferguson var. deliciosa ‘Hayward’) is determined primarily by the sugar–acid balance with fruit firmness and fruit volatile content causing a large moderating effect (Jaeger et al., 2003). A rapid loss of starch and in∗ Corresponding author. Tel.: +64-9-815-4200x7178; fax: +64-9-815-4202. E-mail address: [email protected] (K. Marsh).
crease in soluble solids content provides one of the main criteria for harvest of fruit in New Zealand, but fruit also accumulate acids during growth to produce the characteristic acidity in ‘Hayward’ fruit (Okuse and Ryugo, 1981; Walton and De Jong, 1990). At harvest, kiwifruit contain 0.9–2.5% total acidity, with 40–50% as citrate 40–50% as quinate, and 10% as malate. There are at least three distinct tissue zones within a kiwifruit and the balance of the different acids changes within these zones, so sampling is an important consideration for fruit acidity measurements (MacRae et al., 1989a). The proportion
0925-5214/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.postharvbio.2003.11.001
160
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168
of citrate is highest in the inner cortex, quinate is highest in the outer cortex, whereas the core has the lowest total acid content (about half the other zones), predominantly citrate. During storage and ripening the titratable acidity in ‘Hayward’ kiwifruit changes little (Matsumoto et al., 1983; MacRae et al., 1989b) or declines (Ben-Arie et al., 1982; Crisosto and Crisosto, 2001; Tombesi et al., 1993) depending on growing location. In Israel, California and Italy, high acidity was recorded at harvest (2–2.5%), which dropped sharply (0.5–1.5%) after 1–2 months of storage. In New Zealand, however, acidity was lower at harvest (∼1.4%) and remained stable during storage and subsequent ripening (MacRae et al., 1989b). In particular, MacRae et al. (1989b) found the proportion of citrate declined steadily, whilst malate levels tended to be quite stable during storage at 0 ◦ C. By contrast, storage at 4 ◦ C led to increased malate concentrations. These fruit, when presented to a consumer panel (MacRae et al., 1990), were perceived as less sweet than similar fruit stored at 0 ◦ C, although SSC was similar for both sets of fruit. When a standardised kiwifruit pulp was used (Rossiter et al., 2000; Marsh et al., 2003), consumers could detect additions of sugars and acids at levels found normally within the range of fruit available for purchase. In particular, while addition of all acids increased perception of acidity, addition of quinate also affected perception of ‘Hayward’ flavour and banana-like flavours. While this is one of the first reports of the taste characteristics of quinic acid, several workers have shown that citric acid is less sour than malic acid at similar molarity (mM H+ ) (Noble et al., 1986; Hartwig and McDaniel, 1995). If pulps are to be used to guide selection of new cultivars, and to predict sugar, acid, and fruit volatile compositions which might be found desirable by consumers, it would be useful to test the validity of at least one result in intact fruit. The results obtained in the above experiments (MacRae et al., 1989b, 1990) suggest that it is possible to manipulate acid composition in kiwifruit that are otherwise similar by storing the fruit at 4 ◦ C. We report here, the outcome of such an attempt and show that although acidity was affected, the results for fruit taste were confounded by alterations in other metabolite and texture characteristics with storage temperature.
2. Methods and materials 2.1. Fruit material Fruit from three orchards (designated orchard 1, 2, and 3) in the Bay of Plenty, New Zealand were harvested and commercially packed into 13 single layer trays (with poly-liners) for each orchard. The trays (36 fruit per tray) were transported immediately to HortResearch, Auckland. A ten fruit subsample of each orchard was taken for at-harvest assessment of fruit firmness (Lallu et al., 1989), juice soluble solids (%SSC) as an average of samples from stem and distal ends using a refractometer (Model ATC-1E, AtagoTM , Japan), juice pH (an equal mix from stem and distal ends using an ISFET electrode) (KS701, ShindingenTM ), dry matter (%DM; MacRae et al., 1989a) and the titratable acidity of fruit flesh (%TA; MacRae et al., 1989b). Samples (5 g) were also taken for acid analysis (see below). The 12 remaining trays of fruit for each orchard were divided into 3 different storage temperatures (0, 4, or 10 ◦ C) and stored for a period of 7 weeks. Acidity (pH, %TA), SSC and fruit firmness were assessed on a sample of one fruit per tray at 2-weekly intervals. After 7 weeks cool storage trays of fruit were held at 20 ◦ C for a further 2 weeks and fruit firmness monitored. As the fruit in each tray approached a mean firmness of 8N, the fruit were placed in store at 0 ◦ C and held for presentation to a trained panel of assessors (1–14 days). Although fruit from all orchards softened most quickly at 10 ◦ C, less quickly at 4 ◦ C and least quickly at 0 ◦ C, we were able to obtain 10 replicate sets of fruit from each storage temperature which were matched for fruit firmness (4.4–7.3N) from orchard 1 (designated line 1) on each of the two assessment days. Line 2, was a combination of five matched sets of fruit from orchard 2, and five matched sets of fruit from orchard 3. On the second tasting day (4 days later) similar sets of fruit were obtained and each panellist assessed fruit from the same orchard × temperature combinations as on tasting day 1. 2.2. Sensory panels Panellists experienced in descriptive analysis of kiwifruit participated in eight 1 h training sessions concentrating on descriptive analysis of ‘Hayward’ fruit,
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168
on days prior to testing. Over the course of the training, definitions of the odour, flavour and textural attributes found were generated and appropriate reference standards were established. These included: grassy door (cis-3-hexenel 40 l l−1 ), sulphur odour (cooked egg white), vomit odour (butyric acid 50 g l−1 ), fruit candy odour (ethyl-hexanoate 700 l l−1 ), acid taste (1 g l−1 malic acid), sweet taste (20 g l−1 sucrose), and fibrous flesh texture (canned pineapple). All attributes were assessed using 150 mm unstructured line scales anchored at 0, absent and 150, extreme. Fruit from each storage temperature, from both lines, were presented to 10 panellists. The firmness of all the fruit (4.4–7.3N) was recorded and each panellist was presented with a set of three fruit from a particular orchard which were accurately matched for firmness (e.g. 4.4–5.2N). Testing took place over 2 days with each panellist assessing 6 fruit on each test day (three fruit from each line)—using a design to reduce order and carry-over effects but which presented the fruit from the same line together. Ten-minute breaks were enforced after the first three samples to reduce panellist fatigue. Samples were presented to panellists monadically and were placed stem end up in individual three-digit labelled plastic cups. Assessors were directed to cut fresh slices from the top of the fruit for odour assessment, the middle section of the fruit for flavour assessment and the remaining third of the fruit for texture assessment. Testing was carried out in sensory booths which were maintained at 20 ◦ C with a positive airflow to remove odours from the testing area. MicroleneTM filtered water and plain water crackers were provided as palate cleansers. Prior to presentation, the ends of the fruit were sampled for SSC and juice pH, and two thirds (longitudinal) of the remainder presented to assessors. The final third of the fruit (excluding the penetrometer puncture) was sampled for DM (equatorial region) then the remainder was snap frozen in liquid N2 for subsequent analysis of TA and organic acid content. 2.3. Acid analysis TA was measured on a 5 g sample of frozen tissue, which had been macerated in 25 ml of distilled water using a polytron (KinematicaTM , Luzern, Switzerland) and by titration to pH 8.2 with 0.1N NaOH using an automatic titrator (716 DMS Titrino, Metrohm, Herisau, Switzerland). The same fruit tissue was sub-
161
sampled for organic acid composition (1 g of frozen tissue in 10 ml of 80% methanol), homogenised with a polytron (KinematicaTM ), and centrifuged (5000 × g) to separate tissue from solution. The supernatant was then used for measurement of acid content. Citrate and malate were measured enzymatically by adapting methods of Velterop and Vos (2001) for use with a 96-well microplate spectrophotometer (SpectraMax Plus 384, Molecular Devices). Reagents were provided by Boehringer–Mannheim GmbH. In addition, the same extracts were analysed by gas liquid chromatography (GLC) using a DB1701 column (J&W Scientific). Samples were prepared using a modification (micro-column) of the protocol described by Redgwell (1980) with tartaric acid used as an internal standard for quantification. 2.4. Data analysis Data were subjected to a repeated measures analysis of variance (ANOVA) to determine if differences between treatments existed. Tukey’s least significant differences (LSDs) were calculated to compare differences between means following a significant ANOVA effect. All effects and interactions were evaluated at the 5% level.
3. Results 3.1. Fruit composition at harvest and during storage Fruit from the three orchards had similar compositions at harvest (Table 1), except orchard 3 appeared to have slightly higher TA and DM levels. All orchards had similar proportions of the three principle acids as analysed by gas liquid chromotography (Table 2), but Table 1 Harvest characteristics of ‘Hayward’ kiwifruit from three orchards (n = 10) Orchard
Fruit firmness
SSC (%)
Juice pH
TA (%)
DM (%)
1 2 3
71.7N aa 71.1N a 74.0N a
7.31 a 6.82 a 6.58 a
3.24 a 3.19 a 3.21 a
1.25 a 1.33 ab 1.48 b
16.2 ab 15.8 b 16.8 a
a Means in the same column, followed by the same letter are not significantly different at the 5% level (Tukey’s LSD).
Orchard Malate Malatea 1 2 3
1.4 ab 0.8 b 1.7 a
1.7 a 1.0 b 1.7 a
Quinate Citrate 8.2 a 7.8 a 7.4 a
Citratea
9.9 b 10.1 b 11.3 ab 12.6 a 12.6 a 11.3 ab
Total (GC) 19.5 a 19.8 a 21.6 a
a Individual acid measured by the enzymatic method shown in this column. b Means in the same column, followed by the same letter are not significantly different at the 5% level (Tukey’s LSD).
citrate was lower in fruit from orchard 1, and malate was low in fruit from orchard 2. Subsequent sampling of the fruit trays showed orchard 2 to have similar malate levels to the other two orchards (Fig. 3). Therefore, this appears to be due to the low sample numbers and chance selection of low malate containing fruit. Two methods were used to measure acid concentrations and there was a good correlation between malate concentrations (r 2 = 0.63, see mean values in Table 2), which was not quite so good for citrate (r2 = 0.40). The fruit stored at different temperatures softened as expected, with fruit held at 10 and 4 ◦ C softening faster than fruit held at 0 ◦ C (Fig. 1). There was no change in TA during the storage period at any of the three temperatures (Fig. 2) However, malate remained at similar levels (orchard 1) or increased (orchards 2 and 3) in fruit held at 4 ◦ C (Fig. 3). By contrast, citrate decreased during storage at 0 and 4 ◦ C, and remained at similar levels or decreased slightly at 10 ◦ C (Fig. 4). 3.2. Fruit composition when consumed SSC and DM decreased in fruit that had been stored at 4 and 10 ◦ C compared to fruit stored at 0 ◦ C for both fruit lines tasted (Table 3). In contrast, TA did not vary between treatments except for fruit from line 1 which had slightly reduced levels of TA at 0 ◦ C compared to fruit from other temperatures (Table 3). The acid concentrations varied between treatments (Table 4). Malate concentrations were significantly higher in fruit which had been stored at 4 ◦ C compared to 0 ◦ C for both fruit lines (Table 4), with fruit stored at 10 ◦ C having somewhat intermediate malate
80
60
40
20
(a)
0 0
2
4
6
0
2
4
6
0
2
4
6
80
Fruit firmness (N)
Table 2 Mean acid levels (g kg−1 ) measured by gas chromatography in fresh fruit from three orchards (n = 10)
Fruit firmness (N)
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168
(b)
60
40
20
0
80
Fruit firmness (N)
162
60
40
20
0
(c)
Time in storage (weeks) o
0C
o
4C
o
10 C
Fig. 1. Changes in fruit firmness for ‘Hayward’ kiwifruit sampled from trays of fruit stored at three different temperatures for (a) orchard 1, (b) orchard 2 and (c) orchard 3.
levels. There were no differences in the citrate levels between fruit stored at 0 or 4 ◦ C. Fruit stored at 10 ◦ C had higher levels of citrate than fruit held at 0 or 4 ◦ C for line 1, but not for line 2. The citrate content made the highest contribution to the total acids, so the total
3.0
-1
1.4
1.2
2.5 2.0 1.5 1.0
1.0
(a)
163
3.5
1.6
Malate (g kg )
Titratable acidity (%)
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168
0.0 0
2
4
6
(a)
0.0 0
2
4
6
3.5 3.0
1.4 -1
Malate (g kg )
Titratable acidity (%)
1.6
1.2
1.0 0
2
4
1.5
6
1.6
(b)
0.0 0
2
4
6
0
2
4
6
3.5 1.4
3.0 -1
1.2
1.0 0.0 0
(c)
2.0
1.0
0.0
Malate (g kg )
Titratable acidity (%)
(b)
2.5
2
4
6
2.5 2.0 1.5 1.0
Time in storage (weeks) 0.0 o
0C
o
4C
o
10 C
Fig. 2. Changes in total titratable acidity (TA) for ‘Hayward’ kiwifruit sampled from trays of fruit stored at three different temperatures for (a) orchard 1, (b) orchard 2 and (c) orchard 3.
acid content was also significantly higher in the fruit stored at 10 ◦ C in line 1. There were no differences in the quinate concentrations observed in fruit stored at different temperatures and ripened to equivalent firmness (Table 4).
(c)
Time in storage (weeks)
0C
4C
10C
Fig. 3. Changes in malate concentration (determined by enzymatic method) for ‘Hayward’ kiwifruit sampled from trays of fruit stored at three different temperatures for (a) orchard 1, (b) orchard 2 and (c) orchard 3.
164
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168 Table 3 Key characteristics of ‘Hayward’ kiwifruit presented to sensory panellists (each fruit line is averaged over both days; n = 20)
-1
Citrate (g kg )
14
12
10
Fruit
Storage temperature (◦ C)
DM (%)
SSC (%)
TA (%)
Line 1
0 4 10
16.6 aa 15.8 b 15.5 b
13.9 a 13.1 b 13.0 b
1.27 b 1.34 a 1.31 ab
Line 2
0 4 10
16.5 a 15.3 b 14.6 b
14.0 a 12.5 b 11.7 b
1.37 a 1.33 a 1.33 a
8
6
(a)
a
0 0
2
4
6
-1
Citrate (g kg )
14
seemed to have been balanced by a decrease in other acid components of the fruit. The decline in citrate content in fruit stored at 4 ◦ C is larger on a molar basis than the gain in malate, so it would be expected to have the largest effect on titratable acidity. Differences in acid concentration can also occur with no change in titratable acidity if the acid is in the basic form, e.g. potassium citrate.
12
10
8
6
(b)
Means within each fruit line followed by the same letter are not significantly different at the 5% level (Tukey’s LSD).
0 0
2
4
6
3.3. Sensory perception of the fruit
-1
Citrate (g kg )
14
For both fruit lines and on two separate testing days, the trained sensory panel gave higher values for acid taste, for grassy odour, and lower values for sweetness in the fruit stored at 4 ◦ C compared to those stored at 0 ◦ C (Tables 5 and 6). There were some small differences between testing days, but these affected the significance of the effects, and not the overall trends. Data for the mean effects over 2 days for each of the
12
10
8
6 0 0
(c)
2
4
6
Time in storage (weeks)
0C
4C
10C
Fig. 4. Changes in citrate concentration (determined by enzymatic method) for ‘Hayward’ kiwifruit sampled from trays of fruit stored at three different temperatures for (a) orchard 1, (b) orchard 2 and (c) orchard 3.
The differences in malate concentration are in the order of 1 g kg−1 which should result in 0.1% increase in titratable acidity and this occurred in line 1, but not in line 2. For line 2, changes in titratable acidity
Table 4 Extractable acid content (g kg−1 ) of ‘Hayward’ kiwifruit presented to sensory panellists sampled and analysed by gas chromatography (each fruit line is averaged over both days; n = 20) Fruit
Storage temperature (◦ C)
Malate
Line 1
0 4 10
1.50 ba 2.53 a 2.42 a
8.3 b 8.0 b 9.6 a
7.2 a 6.8 a 7.7 a
17.0 b 17.4 ab 19.7 a
Line 2
0 4 10
1.45 c 2.43 a 1.84 b
10.3 a 8.9 a 10.0 a
8.5 a 6.9 a 7.6 a
20.2 a 18.3 a 19.5 a
Citrate Quinate Total
a Means within each fruit line followed by the same letter are not significantly different at the 5% level (Tukey’s LSD).
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168 Table 5 The mean scores for trained panellists assessing the sensory attributes (0: absent, 150: extreme) of ‘Hayward’ kiwifruit from line 1 over two testing days after storage at different temperatures Attribute
Storage temperature (◦ C) 0
Fruit candy door Sulphur door Grassy door Vomit door Flesh firmness Core firmness Sweetness Acidity Juiciness Metallic flavour Characteristic Hayward flavour Woody/stalky flavour Metallic aftertaste Astringent aftertaste Ease of breakdown Fibrous flesh texture Gelatinous texture
4
10
Attribute
Fruit candy door Sulphur door Grassy door Vomit door Flesh firmness Core firmness Sweetness Acidity Juiciness Metallic flavour Characteristic Hayward flavour Woody/stalky flavour Metallic aftertaste Astringent aftertaste Ease of breakdown Fibrous flesh texture Gelatinous texture
aa b b a b b a b a a a
8.9 b 47.9 a 46.8 a 24.3 b 56.8 a 52.7 a 70.4 b 83.4 a 58 b 39.5 a 59 b
17.2 44.4 41.3 19.7 58.8 59.6 74.1 70.7 58.2 44.7 63.6
b a ab b a a b ab b a b
23.3 20.0 15.9 23.3 11.8 16.9 15 15 12.7 12.3 13.6
10.8 28.5 10.9 50.9 15.7 43 a
b a a b b
17.3 26.4 10.2 76 a 23.9 18.6
19.7 32.2 14.7 76 a 32.3 20.4
a a a
8.6 12.8 12.8 14.1 14.3 13.5
ab a a ab b
a b
Table 6 The mean scores for trained panellists assessing the sensory attributes (0: absent, 150: extreme) of kiwifruit from line 2 over two testing days after storage at different temperatures
LSD (Tukey)
82.9 21.3 29.3 51.5 27.4 23.2 89.1 60.9 78.8 41.7 79.6
165
Storage temperature (◦ C)
LSD (Tukey)
0
4
10
40.8 aa 28 b 31.8 b 43.6 a 39 b 25.2 b 78.8 a 74 b 70.9 a 46 a 66.6 a
9.2 b 43.5 a 46.3 ab 19.2 b 64 a 63.2 a 64 b 89.7 a 58.7 a 42.4 a 58.3 a
16.5 b 37.1 ab 50.8 a 37 ab 49.4 ab 49.3 a 64.1 b 77.4 ab 60.9 a 46.5 a 58.7 a
14.3 15.3 16.0 24.2 14.9 22.9 13.1 14.8 13.3 14.1 12.3
13.7 29.2 13.5 55.4 14.6 31.6
21.7 28.4 17.4 82.2 34 a 11.8
19.2 a 29.2 a 16 a 70.6 a 27.6 ab 17 b
10.8 10.1 10.1 13.6 17.2 10.1
a a a b b a
a a a a b
Fruit were assessed within the firmness range 4.4–7.3N. a Means in the same row followed by the same letter are not significantly different at the 5% level (Tukey’s LSD).
Fruit were assessed within the firmness range 4.4–7.3N. a Means in the same row, followed by the same letter are not significantly different at the 5% level (Tukey’s LSD).
fruit lines is reported (Tables 5 and 6) (Fig. 5) and there is a consistency in the results for the two fruit lines. For acidity, the fruit stored at 10 ◦ C were intermediate between the 4 and 0 ◦ C fruit and not significantly different from either group, but for many other parameters (including sweetness) the fruit at 10 ◦ C gave results similar to those at 4 ◦ C. Parameters such as DM and SSC, which are generally very highly correlated with sweetness character of the fruit, decreased with fruit temperature in measurements made on the tasted fruit (Table 3). The principal components analysis featured in Fig. 5 highlights the dependence of fruit taste on several aspects of the fruit. The fruit stored at 0 ◦ C cluster in an area associated with high sweetness, and typical Hayward flavour. By contrast, fruit stored at 4 ◦ C cluster in a region of higher acidity and less sweetness. The first principal component correlates with sweetness/DM and the second with the perception of acidity and individual acid values. Although the design of the sensory panels ensured that fruit were balanced for firmness (penetrometer measurement) the fruit from different temperatures
were perceived by sensory panellists as being very different in texture by the sensory panellists. Ratings for core firmness, flesh firmness, and fibrous flesh texture increased in the fruit at 4 and 10 ◦ C compared to fruit at 0 ◦ C, which correlates with our empirical observations of rubbery textures in fruit stored at temperatures greater than 0 ◦ C. Greater rubberiness has been reported previously in fruit stored at 4 ◦ C when compared to fruit stored at 0 ◦ C (MacRae et al., 1990). Storage at 4 and 10 ◦ C also tended to lead to an increase in stalky, woody flavours compared to storage at 0 ◦ C (Table 3).
4. Discussion We have clearly demonstrated that kiwifruit stored at temperatures above 0 ◦ C undergo different metabolism compared to those stored at 0 ◦ C. This resulted in fruit with quite different sensory characteristics. However, we were unable to manipulate fruit to show altered acid metabolism alone. Fruit stored
166
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168
line 1 malate
2
TA citrate quinate Acidity DM
0
Stalky/woody flavour
Brix
pH
-2
Sweetness Hayward flavour Brix/Acid ratio
-4
Second principal component 21.2 %
temp 0 temp 4 temp 10
-2
0
2
4
First principal component 24.7 %
malate citrate
2
quinate Acidity
TA
Stalky/woody flavour
0
Hayward flavour Sweetness pH
DM Brix
-2
Second principal component 21.2 %
4
line 2 temp 0 temp 4 temp 10
Brix/Acid ratio
-4
-2
0
2
First principal component 24.7 %
Fig. 5. A principal components comparison of sensory characteristics and analytical measures for fruit from two orchard lines stored at three different temperatures.
at 4 ◦ C were 1–2% lower in SSC than those at 0 ◦ C and there was a slight increase in TA measured in fruit line 1 stored at 4 ◦ C but not in fruit line 2. The malate content showed a consistent increase at 4 ◦ C compared to 0 ◦ C in both fruit lines. Trained assessors were able to distinguish between fruit held at different
storage temperatures, with the most acid, least sweet, fruit being defined as those held at 4 ◦ C, and the least acid, most sweet, fruit being those held at 0 ◦ C (and 10 ◦ C). These results could equally be attributed to a decrease in SSC, or to an increase in malate content. Interestingly, the SSC for fruit stored at 4 and 10 ◦ C is similar but some differences in acid content occurred between fruit stored at these two temperatures. Based on previous research with whole fruit (MacRae et al., 1989a,1990; Paterson et al., 1991; Crisosto and Crisosto, 2001) an increase in SSC of fruit due to later harvests from the same location and growing condition is normally detected by assessors as having increased acceptability, and flavour intensity. Fruit from the same source in our experiments had higher SSC at ripeness when stored at 0 ◦ C compared to 4 ◦ C. This could explain the perceptions of greater sweetness and reduced acidity in the fruit stored at 0 ◦ C compared to the other storage regimes. In experiments using fruit pulps adulterated with sugars, higher SSC (with similar TA and acid concentrations) reduced perception of acidity and increased flavour acceptability (Rossiter et al., 2000). An alternative interpretation of our results, is that the greatest increase in perception of acidity occurred with fruit stored at 4 ◦ C. This fruit consistently had 50–100% more malate than fruit stored at 0 ◦ C. Although it is difficult to separate this increase in malate from confounding factors such as the decline in fruit dry matter, or changes in fruit texture, we had very consistent responses from trained panellists. The rise in off-flavours, and acid taste were consistent with experiments using ‘Hayward’ pulp with malate added (Marsh et al., 2003) and a rise in off flavours is also consistent with earlier experiments on fruit stored at 4 ◦ C (MacRae et al., 1990). Studies with fruit pre-sorted for dry matter content indicate that low dry matter fruit have lower acidity, yet the fruit are perceived to be less sweet, more acid (unpublished data). Decreased sweetness may well be driving the increase in acid taste observed in both examples. Alternatively, when malic acid and citric acid were added to fruit pulps at similar moles H+ , malate was significantly more sour tasting than citric acid (Marsh et al., 2003), as has been previously found in model solutions (Noble et al., 1986; Hartwig and McDaniel, 1995). Similar experiments using additions of citric acid to mango homogenates have demonstrated that acid con-
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168
centration affects ratings for sweet and sour and ratings for some flavour characters (Malundo et al., 2001). As with kiwifruit (Paterson et al., 1991) and tomato (Baldwin et al., 1998), TA correlated with and was a useful predictor of sour taste in mango. The combined results of experiments with fruit sorted for dry matter, pulp experiments (Marsh et al., 2003) and the storage experiment reported here show that a strategy of improving flavour in kiwifruit should not include raising malate concentrations. The marked decrease in acidity during storage and/or ripening reported by workers in warmer countries (Ben-Arie et al., 1982; Crisosto et al., 1984; Crisosto and Crisosto, 2001) deserves some attention. Our experience is for a slight initial decline then maintenance of levels of titratable acidity of around 0.8–1.0% during storage/ripening (MacRae et al., 1989b) (Fig. 2, Table 3), but the results described here confirm that some metabolism of acid occurs during this time and that storage temperature can change the balance of the three major acids in the fruit (Table 4). Research on fruit from California has shown that there are generally higher respiration rates recorded in the field for fruit growing at higher temperatures (Walton and De Jong, 1990). Such fruit appear to rapidly metabolise a particular acid pool when fruit are first placed in storage (Crisosto et al., 1984; Crisosto and Crisosto, 2001). Marsh et al. (2000a,b) recently described the significance of tonoplast gradients for acid accumulation. These gradients provide a store of energy which can be utilised via a pyrophosphatase shunt to produce ATP under anaerobic conditions or in the absence of other energy sources. As proton gradients were utilised the pH of the tonoplast and the squeezed juice would increase, and a sharp rise in juice pH during fruit ripening has been recorded in some instances (Matsumoto et al., 1983; Ben-Arie et al., 1982). A hypothesis to explain the different behaviour of fruit harvested from different environments is that the cooler conditions present in New Zealand allow induction of enzymes (e.g. V-PPase) to utilise tonoplast gradients and reduce titratable acidity before harvest. This is in contrast to fruit from warmer climates in which the gradients are maintained until dissipation is triggered by ripening or cool storage (Matsumoto et al., 1983; Crisosto and Crisosto, 2001). There are reports of low input technologies for storing kiwifruit in warmer climates (Sawada et al., 1992;
167
Nanos et al., 1997). Storage at temperatures other than 0 ◦ C will affect fruit flavour in New Zealand fruit, and an increase of 3–4 ◦ C for an extended period will result in decreased sweetness and/or increased perception of acid taste as well as off-flavours. There does, however, seem to be a difference in the ripening behaviour of fruit from warmer climates with respect to acidity such that the alternative storage strategies may be more appropriate for fruit grown in these climates.
Acknowledgements This research was funded by the New Zealand Foundation for Research, Science and Technology Contract No. C06X0006 and C06X0213. Panellist results were collected using Compusense five V.4.2.
References Baldwin, E.A., Scott, J.W., Einstein, M.A., Malundo, T.M., Carr, B.T., Shewfelt, R.L., Tandon, K.S., 1998. Relationship between sensory and instrumental analysis for tomato flavour. J. Am. Soc. Hortic. Sci. 123, 906–915. Ben-Arie, R., Gross, J., Sonego, L., 1982. Changes in ripening parameters and pigments of the Chinese gooseberry (kiwi) during ripening and storage. Sci. Hortic. 18, 65–70. Crisosto, C.H., Crisosto, G., 2001. Understanding consumer acceptance of early harvested ‘Hayward’ kiwifruit. Postharvest Biol. Technol. 22, 205–213. Crisosto, G., Mitchell, G., Arpaia, M.L., Mayer, G., 1984. The effect of growing location and harvest maturity on the storage performance and quality of ‘Hayward’ kiwifruit. J. Am. Soc. Hortic. Sci. 109, 584–587. Hartwig, P., McDaniel, M., 1995. Flavour characteristics of lactic, malic, citric, and acetic acids at various pH levels. J. Food Sci. 60, 384–388. Jaeger, S.R., Rossiter, K.L., Wismer, W.V., Harker, F.R., 2003. Consumer-driven product development in the kiwifruit industry. Food Qual. Pref. 14, 187–198. Lallu, N., Searle, A.N., MacRae, E.A., 1989. An investigation of ripening and handling strategies for early season kiwifruit (Actinidia deliciosa cv. Hayward). J. Sci. Food Agric. 47, 387– 400. MacRae, E.A., Bowen, J.H., Stec, M.H., 1989a. Maturation of kiwifruit (Actinidia deliciosa cv. Hayward) from two orchards: differences in composition of the tissue zones. J. Sci. Food Agric. 47, 401–416. MacRae, E.A., Lallu, N., Searle, A., Bowen, J., 1989b. Changes in the softening and composition of kiwifruit (Actinidia deliciosa) affected by maturity at harvest and postharvest treatments. J. Sci. Food Agric. 49, 413–430.
168
K. Marsh et al. / Postharvest Biology and Technology 32 (2004) 159–168
MacRae, E.A., Stec, M.G., Triggs, C.M., 1990. Effects of post-harvest treatment on the sensory qualities of kiwfruit harvested at different maturities. J. Sci. Food Agric. 50, 533– 546. Malundo, T.M.M., Shewfelt, R.L., Ware, G.O., Baldwin, E.A., 2001. Sugars and acids influence flavour properties of mango (Mangifera indica). J. Am. Soc. Hortic. Sci. 126, 115–121. Marsh, K.B., Gonzalez, P.G., Echeverria, E., 2000a. PPi formation by reversal of the tonoplast bound V-PPiase from valencia orange juice cells. J. Am. Soc. Hortic. Sci. 125, 420–424. Marsh, K.B., Gonzalez, P.G., Echeverria, E., 2000b. Partial characterization of H+ -translocating inorganic pyrophosphatase from three citrus varieties differing in vacuolar pH. Physiol. Plant. 111, 519–526. Marsh, K.B., Rossiter, K., Lau, K., Walker, S., Gunson, A., MacRae, E., 2003. Using fruit pulps to explore flavour in kiwifruit. Acta Hortic. 610, 229–238. Matsumoto, S., Obara, T., Luh, B.S., 1983. Changes in chemical constituents of kiwifruit during postharvest ripening. J. Food Sci. 48, 607–611. Nanos, G.D., Kittas, C., Sfakiotakis, E., 1997. In: Porlingis, J. (Ed.), An alternative low-cost storage for kiwifruit. Acta Hortic. 444, 625–628. Noble, A.C., Philbrick, K.C., Boulton, R.B., 1986. Comparison of sourness of organic acid anions at equal pH and titratable acidity. J. Sens. Stud. 1, 1–8.
Okuse, I., Ryugo, K., 1981. Compositional changes in the developing ‘Hayward’ kiwifruit in California. J. Am. Soc. Hortic. Sci. 106, 73–76. Paterson, V.J., MacRae, E.A., Young, H., 1991. Relationships between sensory properties and chemical composition of kiwifruit (Actinidia deliciosa). J. Sci. Food Agric. 57, 235– 251. Redgwell, R.J., 1980. Fractionation of plant extracts using ion-exchange. Sephadex Anal. Biochem. 107, 44–50. Rossiter, K.L., Young, H., Walker, S.B., Miller, M., Dawson, D.M., 2000. The effects of sugars and acids on consumer acceptability of kiwifruit. J. Sens. Stud. 15, 241–250. Sawada, T., Morishima, H., Imou, K., Kawagoe, Y., 1992. Studies on the storage and ripening of kiwifruit (part 1). J. Jpn. Soc. Agric. Mach. 54, 61–67. Tombesi, A., Antognozzi, E., Palliotti, A., 1993. Influence of light exposure on characteristics and storage life of kiwifruit. N.Z. J. Crop. Hortic. Sci. 21, 85–90. Velterop, J.S., Vos, F., 2001. A rapid and inexpensive microplate assay for the enzymatic determination of glucose, fructose, sucrose, l-malate and citrate in tomato (Lycopersicon esculentum) extracts and in orange juice. Phytochem. Anal. 12, 299–304. Walton, E.F., De Jong, T.M., 1990. Growth and compositional changes in kiwifruit berries from three Californian locations. Ann. Bot. 66, 285–298.