Comparison of fruit morphology and nutrition metabolism in different cultivars of kiwifruit across developmental stages

Experimental materials and field experiments

The kiwifruit cultivars used in this experiment were ‘Kuilv’, ‘Hongyang’ and ‘Hayward’. Field experiments were carried out in Mianyang of Sichuan Province (31°13′N, 104°16′E) in 2018. The vines were trained to a Pergola system with row-to-row distance of 4 m and plant-to-plant distance of 3 m. Ten kiwi fruit vines of the same age (5-years-old), showing good growth trends and orientation and the same fruit load, were selected for each cultivar. All field experiments were performed as typical for commercial kiwifruit orchards in the area, with water supplied by drip irrigation. At 15 days after anthesis (DAA), fruits of each cultivar were selected randomly from different parts of the vine, marked for subsequent experiments and harvested when appropriate. They were first picked at 30 DAA, followed by harvests at 15-day intervals (10 fruits per vine). The samples were placed in separate paper storage bags and marked. Fruit samples were collected in four biological replicates, snap-frozen in liquid nitrogen, and then stored at −20 °C for subsequent analysis. When measuring physiological indexes, the samples were pressed into a pulp using a juicer at low temperature, in order to reduce any unnecessary chemical reactions.

Evaluation of fruit weight and shape

At regular intervals (15 days apart), four fruits were picked from 10 labelled kiwifruit vines for each kiwifruit cultivar. Fruit fresh weight was measured according to the method reported by Nardozza et al. (2010). Shape factors such as length and the minimum and maximum diameters were measured using a Vernier caliper.

Measurement of total soluble solids (TSS), dry matter (DM), starch and total soluble sugars and carbohydrates

To determine the TSS, juice was obtained from both ends of the fruit and its content of TSS was determined using a refractometer (WYT-4, China). The DM content was measured after drying a 2-mm-thick cross-sectional slice per fruit sample for 24 h at 65 °C, according to the method reported by Nardozza et al. (2010).

Starch assay was performed by colorimetry after samples were digested enzymatically, as described by Qiu et al. (2020) with the following modifications. After centrifugation with 70% ethanol, the supernatant was used to determine soluble sugars and the pellet was used to assay starch accumulation. Anthranone sulfate solution was heated at 100 °C for 11 min and cooled on ice. Absorbance was read at 630 nm. The pellet was dissolved using 1.1% HCl and boiled for 30 min. After cooling and centrifugation, the supernatant was collected. The subsequent procedure was the same as for the measurement of soluble sugars.

Carbohydrate accumulation was measured and calculated using the method described by Barboni & Chiaramonti (2006), using an Agilent 1200 Series instrument with a refractive index detector (Agilent, USA). After that, the extract was separated using an amino column (Inertsil NH₂ 250 mm × 4.6 mm, 5 µm), and then eluted isocratically for 20 min at 85 °C.

Titratable acid (TA)

TA was determined by the traditional titratable method described by Suh, Jung & Lee (2018). Briefly, 20 mL of the juice was titrated with 0.1 M NaOH. Organic acids were assessed according to the method of Zheng et al. (2009). Briefly, 10 g of kiwifruit was transferred into a 50 mL centrifuge tube containing 12.5 mL NH₄H₂PO₄ (40 mmol/L, pH 2.5) and centrifuged at 12,000× g at 4 °C for 15 min. The supernatants were collected after being filtered (0.45 µm filter, Millipore, Bedford, USA) and stored at 4 °C for further analysis. HPLC analysis was performed using an Agilent 1200 Series device with a C18 column (4.6 mm ×250 mm, 5 µm; Waters, UK) at a wavelength of 210 nm. The mobile phase was NH₄H₂PO₄ (40 mmol/L, pH 2.5).

Evaluation of VC content

VC content was measured using the 2, 6-dichlorophenol titration method (Alhassan et al., 2019). Briefly, 20 mL of 1% oxalic acid solution was added to 5 g of thawed kiwi fruit pulp. The solution was shaken, and the volume of the solution was adjusted to 50 mL. Samples were placed in the dark for 2 h, centrifuged at 8,000× g at 4 °C for 10 min, and 10 mL of the supernatant was used to determine the VC content. The extract was titrated with 2,6-dichlorophenol indophenol salt solution (1 mL ≈ 0.02 mg VC) until the solution color became light red and remained unchanged for 15 s. The titration was repeated three times, and the volume of 2,6-dichlorophenol indophenol salt solution used for titration was calculated.

Superoxide dismutase (SOD), peroxidase (POD) and catalase (CAT) activity assays

SOD activity was determined using the method of Xu (2019) (Yang et al., 2015). Briefly, 0.05 mL of enzyme extract was added to sodium phosphate buffer (0.1 M, pH 7.4), followed by addition of 1.35 mL detection reagent consisting of methionine, Ethylene Diamine Tetraacetic Acid (EDTA), Nitrotetrazolium Blue chloride (NBT) and riboflavin. The reaction was maintained in a 37 °C water bath for 30 min. Absorbance at 550 nm was measured using a spectrophotometer (TU-1901, Unico, China), and sodium phosphate buffer (0.1 M, pH 7.4) was used for comparison. One unit of enzyme activity was defined as the amount of enzyme that could inhibit 50% of the reduction of NBT.

POD activity was measured using the method described by Yang et al. (2015) using guaiacol as the donor and H₂O₂ as the substrate (Yang et al., 2015). One unit of POD activity was defined as an increase in absorbance of 0.01 per minute at 460 nm under the measurement conditions.

CAT activity was determined according to the method of Xu (2019) with modifications. The test mixture included 2.25 mL of H₂O₂ prepared using sodium phosphate buffer (0.1 M, pH 7.4) and 0.05 mL of enzyme solution (Xu et al., 2019). Increase in absorbance at 240 nm was recorded at 25 °C for 3 min. One unit of CAT activity was defined as the amount of enzyme extract that caused a 0.01 min⁻¹ change in absorbance.

Data analysis

Experimental data are presented as the means ± standard errors from three independent experimental replications. Data were analyzed using SPSS version 23.0 (SPSS, Chicago, IL, USA). A single factor ANOVA (ANOVA-LSD) was used to make comparisons among groups. A difference was considered statistically significant at p < 0.05.

Fruit growth and development of the three species

The flowering period and ripening time were dramatically different among the cultivars, although they were planted in the same resource orchard. The full-bloom period of ‘Kuilv’ and ‘Hongyang’ occurred in the middle of April, one month earlier than that of ‘Hayward’ (the middle of May). Meanwhile, the ripening time of ‘Kuilv’ was in late August, and the total fruit development period was about 120 days, which was shorter than those of ‘Hongyang’ (135 DAA) and ‘Hayward’ (165 DAA).

Although the fruit weight of the three cultivars increased throughout the growth period, there were still some differences. Throughout the growth period, the fruit weight of ‘Hayward’ and ‘Hongyang’ showed a single ‘S’ shape, which could be divided into three stages: rapid, slow and stagnant growth periods. At the end of the rapid growth period (75 DAA for ‘Hongyang’ and ‘Hayward’), the fresh weight of ‘Hongyang’ and ‘Hayward’ was up to 84.06% and 62.46% of that at fruit harvest. During the stagnant growth period, the fruit weight remained steady, and the final fruit weights of ‘Hongyang’ and ‘Hayward’ were about 66.18 g and 81.29 g, respectively. In comparison, the ‘Kuilv’ fruit weight showed a double-S-shaped growth curve throughout the growth period (Fig. 1A), suggesting that ‘Kuilv’ fruit had two rapid growth periods. In the first rapid growth period, the ‘Kuilv’ fruit weight reached 78.80% of the final fruit weight. In the second rapid growth period following the slow growth period, it reached 98.97% of the final fruit weight. The final harvest weight of ‘Kuilv’ was about 16.04 g, which was far less than those of ‘Hayward’ and ‘Hongyang’.

The vertical diameter, maximum diameter and minimum diameter of fruit are important characteristics indicative of fruit quality. Figure 1B shows the shape and size of fruits of ‘Hayward’, ‘Hongyang’ and ‘Kuilv’ at the time of picking. The ‘Hayward’ fruit had the largest size while the ‘Kuilv’ fruit had the smallest size. The vertical diameter, maximum diameter and minimum diameter were 32.99 mm, 25.81 mm and 22.44 mm for the ‘Kuilv’ fruit; 48.29 mm, 40.00 mm and 36.33 mm for the ‘Hongyang’ fruit, and 55.55 mm, 48.97 mm and 39.16 mm for the ‘Hayward’ fruit (Fig. 1B), respectively.

Figures 1C and 1D shows the size and vertical section of ‘Hayward’, ‘Hongyang’ and ‘Kuilv’ fruits when they were picked. The ‘Kuilv’ fruit had dark green flesh with no hair on the skin. The ‘Hongyang’ fruit had yellow flesh with a red center and a hairless skin. The ‘Hayward’ fruit had green flesh with dense hair on the skin.

Carbohydrate changes in the three species during fruit development

In the early stage of fruit development, there was little fluctuation in TSS content of all species; TSS then increased dramatically before the fruit matured (Fig. 2A). The fruit TSS content in ‘Hayward’ was lower (8.23%) at later stages than that in ‘Hongyang’ and ‘Kuilv’ (around 8.85% and 8.5%, respectively).

Fruit of the three species all showed a trend of increasing DM with developmental age, yet with different accumulation patterns (Fig. 2B). Specifically, the DM of ‘Hayward’ fruit increased linearly from 30 DAA to 120 DAA. This was followed by a second, slower phase of fruit growth from 120 DAA to 165 DAA, during which the DM remained constant (16.29%). In contrast, the DM of ‘Hongyang’ fruit increased throughout development, with a relatively higher accumulation rate at 45–60 DAA and 90–105 DAA. The DM of ‘Kuilv’ fruit increased in a linear pattern, ranging from 6.72% to 25.14%.

The three species had similar patterns of starch accumulation during fruit development (Fig. 3A). In ‘Kuilv’, which had the shortest time of fruit growth, the starch concentrations peaked at 90 DAA, earlier than the other species (105 DAA for ‘Hongyang’; 135 DAA for ‘Hayward’), and were lowest at the time of harvest. As the fruit weight continued to increase, starch degradation began to occur, coinciding with a fast rate of sugar accumulation (glucose, fructose and sucrose) in all three species.

Interestingly, the glucose concentration in ‘Kuilv’ fruit did not show a rapid increase before harvest (Fig. 3B). From 75 DAA to 120 DAA, the glucose concentration steadily increased, and ‘Kuilv’ had the lowest glucose concentration among the three cultivars. In contrast, the amount of glucose increased greatly from 105 DAA to 135 DAA in ‘Hongyang’ and from 120 DAA and 150 DAA in ‘Hayward’ fruits. There was a decrease of glucose accumulation in ‘Hayward’ before harvest, and the concentration in ‘Hongyang’ (1.30 g 100 g⁻¹ FW) was slightly higher than that in ‘Hayward’ (1.24 g 100 g⁻¹ FW).

In addition, fruit of ‘Kuilv’ showed only a small change (from 0.03 g 100 g⁻¹ FW to 0.33 g 100 g⁻¹ FW) in fructose content, while the fructose concentrations in ‘Hayward’ and ‘Hongyang’ fruits increased rapidly before maturity (Fig. 3C).

The sucrose content increased smoothly within a small range in ‘Hayward’ fruit (Fig. 3D). ‘Hongyang’ showed a similar trend to ‘Hayward’ of stable sucrose content, but a higher rate of increase compared with ‘Hayward’. By contrast, ‘Kuilv’ fruit showed a relatively greater change in sucrose content during development, which increased continuously to 0.71 g 100 g⁻¹ FW, higher than those of the other two cultivars, which were 0.16 g 100 g⁻¹ FW and 0.49 g 100 g¹ FW, respectively, at harvest.

Changes in organic acid content

We found significant differences in fruit acidity among the three kiwifruit cultivars throughout the growth period (Fig. 4). There was a rapid increase in titratable acid content in ‘Hayward’ fruit from 30 DAA to 45 DAA, reaching a maximum of 1.36% (w/w), then a decline from 45 DAA to 90 DAA (Fig. 4A). The total titratable acidity of ‘Kuilv’ fruit increased consistently, then decreased from 90 DAA to 120 DAA. However, the titratable acidity of ‘Hongyang’ fruit increased steadily until fruit maturity.

The concentration of citric acid (Fig. 4B) in fruits increased over time in the three species, but was apparently higher in ‘Kuilv’ than in the other two cultivars. Moreover, it decreased from 105 DAA to 120 DAA after reaching a peak (1.069 g 100 g⁻¹ FW) at 105 DAA in ‘Kuilv’. Meanwhile, the citric acid continued to accumulate during maturation in fruits of the other two species.

Unlike citric acid, the level of quinic acid (Fig. 4C) eventually declined in all three cultivars, but with different patterns of decrease. The quinic acid levels increased again during fruit maturation in ‘Hayward’ and ‘Hongyang’; for ‘Hayward’, the pick-up time was 120–150 DAA, while that for ‘Hongyang’ was 75–90 DAA. However, ‘Kuilv’ did not show an obvious increase in quinic acid during maturation and had the lowest level of quinic acid (0.31 g 100 g⁻¹ FW).

Malic acid (Fig. 4D) also decreased during the fruit development process, but all three cultivars showed an initial increase at 30–45 DAA, followed by a drop until maturity. At harvest, the content of malic acid in ‘Kuilv’ fruit was lower (0.13 g 100 g⁻¹ FW) than those in ‘Hongyang’ (0.26 g 100 g⁻¹ FW) and ‘Hayward’ (0.20 g 100 g⁻¹ FW) fruits.

In addition, the content of VC in ‘Kuilv’ fruit peaked at 45 DAA and then gradually decreased until harvest. The content of VC in ‘Hongyang’ fruit increased to the highest level at 60 DAA, then declined in a wave form, reaching the lowest level of 83.02 mg 100 g⁻¹ FW at 135 DAA. The VC concentration of ‘Hayward’ also fluctuated, but within a smaller range. In addition to the differences in the trend of change with time, the VC content also differed among the three kiwi fruits. ‘Kuilv’ had a higher VC content than ‘Hongyang’ or ‘Hayward’ at any time.

SOD, POD and CAT antioxidant enzyme activities in the three species

After kiwi fruit ripening, the SOD activity of ‘Hongyang’ fruit was highest among the three cultivars, 40.57% and 114.90% greater than that of ‘Hayward’ and ‘Kuilv’ fruit, respectively (Fig. 5A). In contrast, the POD activity of ‘Kuilv’ fruit was 115.94 U g⁻¹ FW min⁻¹, significantly higher than those of ‘Hayward’ and ‘Kuilv’ fruits (Fig. 5B). The CAT activity of ‘Kuilv’, ‘Hongyang’ and ‘Hayward’ fruits were 6.84 U g⁻¹ FW min⁻¹, 10.28 U g⁻¹ FW min⁻¹ and 31.08 U g⁻¹ FW min⁻¹, respectively (Fig. 5C).