Impaginato 71 Adv. Hort. Sci., 2020 34(1): 71­79 DOI: 10.13128/ahsc­8403 Varietal differences in sweetness value and flesh juiciness among persimmon cultivars M. Shiraishi (*), H. Asakuma Fukuoka Agricultural and Forestry Research Center, Chikushino 818‐8549, Japan. Key words: breeding stock, Diospyros kaki Thunb., flesh juiciness, sweetness value. Abstract: To promote persimmon breeding project, we analyzed the sugar com­ position (a ratio of sucrose to hexose sugars, SH ratio) and flesh juiciness of 43 persimmon cultivars (Diospyros kaki Thunb.) consisting of 24 pollination­con­ stant non­astringent (PCNA)­types and 19 non­PCNA­types, together with other fruit quality traits. The cultivar collection includes newly­released cultivars after 1990 and commercially­produced local cultivars in Japan. These cultivars were broadly classified into three types: sucrose accumulators, intermediate accu­ mulators, hexose accumulators. Analysis of variance showed that the genotypic effect on the SH ratio and flesh juiciness is high with negligibly small environ­ mental variance, indicating that SH ratio and flesh juiciness can be determined by a one­year trial without tree replication. Highly varietal diversity in the SH ratio and flesh juiciness was observed within and between persimmon cultivar types. Sweetness value (SSC × SH ratio) of the cultivars/selections seems to be a useful predictor of fruit sweetness. In terms of palatability, however, persim­ mon cultivar’s improvement should be performed on the sweetness value in association with flesh juiciness. 1. Introduction Persimmon (Diospyros kaki Thunb.) is believed to have originated in Eastern Asia, is produced worldwide including in Azerbaijan, Brazil, China, Iran, Israel, Italy, Japan, Korea, New Zealand, and Spain (FAOSTAT, 2017). A number of local varieties has been developed in China, Korea, and Japan during a long history of domestication (Parfitt et al., 2015; Sato and Yamada, 2016; Yesiloglu et al., 2018). Persimmon cultivars can be classi­ fied into four types: pollination­constant astringent (PCA); pollination­ variant astringent (PVA); pollination­variant non­astringent (PVNA); polli­ nation­constant non­astringent (PCNA), based on seed formation, change in flesh color, and nature of astringency loss (Hume, 1914; Ikeda et al., 1985; Yonemori et al., 2000). Among these types, fruits of PCA­ and PVA­ type cultivars are always astringent without postharvest treatment such as application of carbon dioxide gas or ethanol vapor. PVNA­type cultivars require pollination and seed formation to lose astringency as well as flesh (*) Corresponding author: mikioshi@farc.pref.fukuoka.jp Citation: SHIRAISHI M., ASAKUMA H., 2020 ­ Varietal diffe‐ rences in sweetness value and flesh juiciness among persimmon. ­ Adv. Hort. Sci., 34(1): 71­79. Copyright: © 2020 Shiraishi M., Asakuma H. This is an open access, peer reviewed article published by Firenze University Press (http://www.fupress.net/index.php/ahs/) and distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Competing Interests: The authors declare no competing interests. Received for publication 19 July 2019 Accepted for publication 12 November 2019 AHS Advances in Horticultural Science http://creativecommons.org/licenses/by/4.0/ http://creativecommons.org/licenses/by/4.0/ http://creativecommons.org/licenses/by/4.0/ Adv. Hort. Sci., 2020 34(1): 71­79 72 browning. Fruits of PCNA­type cultivars naturally lose their astringency because they terminate accumulat­ ing tannins at early fruit development stage. In par­ ticular, PCNA­type persimmon cultivars are highly desired for fresh consumption on a worldwide level. Regarding eating quality of PCNA­type cultivar, the great stress is laid upon the evaluation of soluble solids content (SSC, °Brix) using refractometer because sugars generally contribute a very large pro­ portion to the SSC in the ripening stage (reviewed by Giordani et al., 2011). However, Ban et al. (2010) and Mitani et al. (2015) postulated that flesh juiciness is a decisieve parameter to determine the texture of per­ simmon fruit, where fruits with higher firmness had a tendency to be less juicy as observed in apple (Harker et al., 2003). On the other hand, the percentage sucrose in flesh significantly affects the sweetness of fruit, such as East Asian pear (Kajiura et al., 1979), oriental melon (Zhang and Li, 2005), peach (Suzuki et al., 1990; Cirilli et al., 2016) and strawberry (Sone et al., 2000). Thus, understanding how varietal differ­ ences influence sugar composition will provide valu­ able information for genetic improvement of the palatability of persimmon fruit in future breeding programs. Giordani et al. (2011) classified worldwide persimmon cultivars such as ‘Atago’, ‘Fuyu’, ‘Rojo Brillante’ and ‘Triumph’ into three groups based on the cluster analysis of sugar composition. To date, however, there has been no comprehensive research on the sugar composition of the recent Japanese PCNA­ and non­PCNA­type cultivars. Since 1990, new PCNA­ and non­PCNA­type persimmon cultivar have been released in Japan by the National Institute of Fruit Tree Science (Yakushiji and Nakatsuka, 2007; Yamada et al., 2012 a, b). These new cultivars have large sized fruit, brilliant skin color, and are highly palatable, but their sugar composition remains large­ ly unknown. Asakuma and Shiraishi (2017) showed that geno­ typic effect of sugar composition (SH ratio) and flesh juiciness is significantly high, whereas the year, geno­ type × year interaction, among trees within geno­ type, and tree × year interaction is small or negligi­ ble. In experimental field of the current study, we have preserved a Japanese persimmon cultivar col­ lection of both PCNA­ and non­PCNA­type cultivars as breeding stocks since 1980. Although the total num­ ber of preserved cultivars is less than 50, the cultivar collection includes newly­released cultivars after 1990 and commercially­produced local cultivars in Japan. In the present study, we (1) re­confirmed small year variability in the sugar composition and flesh juiciness by the analysis of variance, and (2) dis­ cussed varietal differences in the sugar composition and flesh juiciness among persimmon cultivars in association with palatability. 2. Materials and Methods Plant materials for yearly variation of sugar composi‐ tion Four PCNA­type cultivars of Japanese persimmon (‘Akiou’, ‘Fuyu’, ‘Matsumotowase­Fuyu’, ‘Taishu’) were used. These cultivars are grown for commercial­ ly marketable fruit production by normal cultural practices, including pruning, flower and fruit thin­ ning, irrigation, soil and pest management (Yamada, 2006) in an open­field of the Fukuoka Agricultural and Forestry Research Center, Fukuoka, Japan (33°50ʹ N and 130°57ʹ E). According to the reference value derived from our annual survey from 2008 to 2015 (Table S1), mature representative eight fruits per tree were sampled from three trees per geno­ type for late­October to late­November in 2014 to 2016 seasons depending on each cultivar’s optimal ripening time and skin color. The statistical fixed­effect model (Table S2) that we adopted to express the phenotypic value (Asakuma and Shiraishi, 2017) is: Pijkc = m + Gi + Yk + (GYik) + Tij + (TYijk) +Eijkc where Pijkc is the phenotypic value of the cth fruit of the jth tree of the ith genotype in the kth year; m is the overall mean; Gi is an effect contributed by the ith genotype; Yk is an effect of the kth year; GYik is the interaction between the ith genotype and the kth year; Tij is an effect of the jth tree of the ith geno­ type; TYijk is the interaction between the jth tree of the ith genotype and the kth year; and Eijkc is an effect of the cth fruit of the jth tree of the ith genotype in the kth year. ANOVA provided the variance associat­ ed with genotype (σg 2), among years (σy 2), genotype × year interaction (σgy 2), among trees within genotypes (σt 2), tree × year interaction (σty 2) and among fruits within tree (σ2). Plant materials for varietal difference in sugar com‐ position A total of 43 of Japanese persimmon cultivars con­ sisting of 23 PCNA­types (Fig. S1), 9 PVNA­types (Fig. S2), 6 PVA­types (Fig. S3), and 5 PCA­types (Fig. S4) was analyzed in 2016. Cultural practices were con­ Shiraishi and Asakuma ‐ Sweetness value and flesh juiciness of persimmon fruit 73 formed to Plant materials for yearly variation of sugar composition. Depending on the optimal ripen­ ing time of the cultivar, five to eight mature fruits per one­tree of each cultivar were harvested from late­ September to early­December in 2016 season according to the reference values based on above­ described cultivation records (Table S1). Classification of fruit ripening time was performed in accordance with the definitions by Yamada et al. (1995). Fruit shape index, FSI (longitudinal diameter/transverse diameter) was defined according to Maeda et al. (2018) with slight modifications (Fig. S5). Cracking of apex and calyx­end of fruit was examined by UPOV guideline (UPOV, 2004). Astringency removal of PCA­ and PVA­type cultivars was performed by treatment with either ethanol vapor or carbon dioxide gas depending on the cultivar (Yamada, 2006). The sam­ pled fruit was weighed, and each fruit was horizon­ tally cut to measure fruit skin color, flesh firmness, flesh juiciness, soluble solids content (SSC, °Brix), and sugar composition. Analysis of fruit quality traits Fruit skin color, flesh firmness, flesh juiciness, sol­ uble solids content, and sugar composition was mea­ sured according to Asakuma and Shiraishi (2017). Fruit skin color around fruit apex was measured using a chromameter (CR­300, Minoruta, Tokyo, Japan), and expressed as the value of color chart following formula: Color chart (CC) = − 9.485 Ln (hue angle) + 44.503, R2 = 0.9278. Flesh firmness (kg) was determined by a handheld universal pressure tester with a 5.0­mm­diameter ×10.0­mm­height columnar plunger (KM­5, FUJI­ WARA SCIENTIFIC, Tokyo, Japan). Fifteen to 20 g of peeled flesh was weighed and wrapped in one layer of medical gauze. After hand­pressing (only one press) for 15 s, the squeezed juice (flesh juiciness) was measured using a 25­mL mess cylinder and expressed as mL g­1 FW. Soluble solids content (SSC) of the resulting juice samples was determined as the °Brix value using a portable calibrated electronic refractometer (PAL­1, Atago, Tokyo, Japan). For the analysis of sugar composition, 8 to 10 g of peeled flesh was weighed and transferred to a 50­mL heat­tolerant tube and partially screw­capped. The flesh sample was immediately microwave­irradiated at 730 W for 60 s before extracting the sugars. The irradiated sample was ground in a laboratory blender with ~40 mL deionized water. The puree was cen­ trifuged at 5000 × g at 25°C for 10 min. The resulting supernatant was brought to 50 mL with deionized water and filtered through a 0.45­µm filter. Sugar composition was analyzed using a HPLC (LC­10A, Shimadzu, Kyoto, Japan) consisting of a SCL­10A sys­ tem controller, LC­10AD pumps, a CTO­10A column oven, and a RID­10A refractive index detector. The column (SCR­101N, 7.9 × 300 mm, Shimadzu, Kyoto, Japan) was operated at 60°C with 0.8 mL min­1 of water. The injection volume was 10 to 20 µL. 3. Results and Discussion Variation in sugar composition HPLC profiles of sugar composition were obtained from persimmon cultivars examined with three major peaks assigned as sucrose, glucose, and fructose, respectively, thereby expressing as the SH ratio, a ratio of sucrose to hexose sugars (Hirano et al., 1995). As shown in figure 1, we classified persimmon cultivars into three types according to Zheng and Sugiura (1990) with slight modification; sucrose accu­ mulators (percentage sucrose ≥55.1%, SH ratio≥ 1.23), intermediate accumulators (percentage sucrose in 45.0 to 55.0%, SH ratio in 0.82 to 1.22), hexose accumulators (percentage sucrose ≤44.9%, SH ratio ≤0.81). Previous studies have shown that the Fig. 1 ­ Sugar composition of persimmon cultivars from the HPLC analysis. A: Sucrose accumulators [‘Fuyu (PCNA­type)’, ‘Akagaki (PVNA­type)’, ‘Atago (PCA­type)’]; B: Intermediate accumulator [‘Maekawajiro (PCNA­type)’, ‘Saefuji (PVNA­type)’, ‘Aizumishirazu (PVA­type)’]; C: Hexose accumulators [‘Soshu (PCNA­type)’, ‘Nishimurawase (PVNA­type)’, ‘Hiratanenashi (PVA­ type)’] PCA: pollination­constant astringent PVA: pollina­ tion­variant astringent PVNA: pollination­variant non­ astringent PCNA: pollination­constant non­astringent. Adv. Hort. Sci., 2020 34(1): 71­79 74 sucrose percentage in persimmon cultivar ranged between 12 and 70%, resulting in varietal difference in the sugar composition or SH ratio (Tsuji and Komiyama, 1987; Zheng and Sugiura, 1990; Hirano et al., 1995; Hirai et al., 2004; Suzuki et al., 2010; Asakuma and Shiraishi, 2017). Sugar profiles of PCNA­ and non­PCNA­types from well­known culti­ vars such as ‘Atago’ (sucrose accumulator), ‘Hiratanenashi’ (hexose accumulator), ‘Soshu’ (hex­ ose accumulator) and ‘Fuyu’ (sucrose accumulator) in previous reports are in agreement with those in the present study. Table 1 shows highly varietal difference in the SH ratio among the PCNA­type cultivars, especially for Mid­ to Late­Oct. ripening ones. In cultivars as sucrose accumulators, large amounts of sucrose recorded in ‘Sodawase (SH ratio = 3.29)’ followed by ‘Okugosho (SH ratio = 2.35)’, ‘Hanagosho (SH ratio = 2.17)’, ‘Suruga (SH ratio = 2.02)’, ‘Fuyu (SH ratio = 1.93)’, and ‘Shinshu (SH ratio = 1.89)’. In the hexose accumulator, large amounts of hexose were present in ‘Kishu (SH ratio = 0.21)’, ‘Soshu (SH ratio = 0.23)’, ‘Izu (SH ratio = 0.35)’, followed by ‘Tenjingosho (SH ratio = 0.61)’, ‘Taishu (SH ratio = 0.64)’ and ‘Taiga (SH ratio = 0.67)’. Among the remainders, sucrose and hexose seemed to be accumulated approximately equal amounts with 1.10 in SH ratio (‘Maekawajiro’), 1.03 in SH ratio (‘Misatogosho’), and 0.96 in SH ratio (‘Reigyoku’). Table 1 ­ Sugar composition of pollination­constant non­astringent (PCNA) type of Japanese persimmon cultivars in 2016 Fruit ripening time (Late‐Sep., Early‐ to Late‐Oct., Early‐ to Late‐Nov., Early‐Dec.) was classified according to Yamada et al. (1995). SH ratio= sucrose [g 100 g­1 FW] / hexoses (glucose + fructose) [g 100 g­1 FW]. Sweetness value= soluble solids content (reference value in Table S1) x SH ratio. Fruit ripening time/ cultivar or selection Sugar composition (g 100 g­1 FW) Sugar composition (%) Type of sugar accumulation SH ratio Sweetness value Flesh juiciness (mL g­1 FW)Sucrose Glucose Fructose Total Sucrose Glucose Fructose Early‐Oct. Soushu 1.98 4.96 4.60 11.5 17.2 43.0 39.9 Hexose 0.21 3.1 0.27 Mid‐Oct. Izu 3.03 4.53 4.22 11,8 25.7 38.5 35.8 Hexose 0.35 5.3 0.29 Shinshuu 9.32 2.62 2.32 14.3 65.4 18.4 16.3 Sucrose 1.89 32.3 0.22 Late‐Oct. Sodawase 11.66 1.90 1.64 15.2 76.7 12.5 10.8 Sucrose 3.29 58.9 0.27 Reigyoku 6.44 3.59 3.12 13.2 49.0 27.3 23.7 Intermediate 0.96 16.1 0.36 Taiga 5.25 4.21 3.68 13.1 40.0 32.0 28.0 Hexose 0.67 11.1 0.32 Kanshu 5.93 3.87 3.48 13.3 44.7 29.1 26.2 Hexose 0.81 13.2 0.17 Kishu 2.66 5.90 5.56 14.1 18.8 41.8 39.4 Hexose 0.23 3.4 0.24 Taishu 5.30 4.52 3.76 13.6 39.0 33.3 27.7 Hexose 0.64 10.1 0.35 Maekawajiro 6.83 3.48 2.71 13.0 52.5 26.7 20.8 Intermediate 1.10 18.2 0.22 Tenjingosho 6.00 5.33 4.46 15.8 38.0 33.8 28.2 Hexose 0.61 10,5 0.17 Early‐Nov. Akiou 9.14 3.40 2.68 15.2 60.1 22.3 17.6 Sucrose 1.57 27.6 0.38 Misatogosho 7.37 4.18 2.96 14.5 50.8 28.8 20.4 Intermediate 1.03 18.6 0.23 Uenishiwase 7.63 3.35 2.48 13.5 56.7 24.9 18.4 Sucrose 1.31 20.8 0.13 Matsumotowase­Fuyu 9.16 2.85 2.41 14.4 63.5 19.8 16.7 Sucrose 1.74 26.8 0.27 Mid‐Nov. Mushirodagosho 5.53 3.77 3.57 12.9 43.0 29.3 27.7 Hexose 0.75 10.8 0.33 Youhou 5.99 4.17 4.11 14.3 42.0 29.2 28.8 Hexose 0.72 12.2 0.18 Late‐Nov. Fuyu 9.79 2.75 2.32 14.9 65.9 18.5 15.6 Sucrose 1.93 31.7 0.27 Okitsu­20 (Ro­19) 8.64 3.61 3.06 15.3 56.4 23.6 20.0 Sucrose 1.30 25.5 0.31 Taiho 7.92 3.50 2.88 14.3 55.4 24.5 20.1 Sucrose 1.24 21.2 0.36 Early‐Dec. Okugosho 9.43 2.14 1.88 13.5 70.1 15.9 14.0 Sucrose 2.35 43.0 0.24 Suruga 8.86 2.26 2.13 13.3 66.9 17.1 16.1 Sucrose 2.02 35.4 0.26 Hanagosho 9.03 2.20 1.96 13.2 68.5 16.7 14.9 Sucrose 2.17 37.1 0.25 Shiraishi and Asakuma ‐ Sweetness value and flesh juiciness of persimmon fruit 75 (Table 3) were all sucrose accumulators, except for ‘Kawazokogaki’ in SH ratio of 1.12 (intermediate accumulator). Conversely, most PVA­type cultivars (Table 3) could be classified as hexose accumulators, with SH ratios ranging from 0.39 (‘Tonewase’) to 0.67 (‘Hiratanenashi’) with the exception of ‘Koshu­ In the PVNA­type cultivars (Table 2), ‘Nishi­ murawase’ (SH = 0.47) was hexose accumulator, whereas the sucrose accumulators had SH ratios, var­ ing from 1.83 (‘Akagaki’) to 2.77 (‘Rendaiji’). The SH ratio of intermediate accumulators ranged from 0.86 (‘Fudegaki’) to 1.22 (‘Zenjimaru’). PCA­type cultivars Table 2 ­ Sugar composition and other fruit traits of pollination­variant non­astringent (PVNA) type of Japanese persimmon cultivars in 2016 Fruit ripening time (Late‐Sep., Early‐ to Late‐Oct., Early‐ to Late‐Nov., Early‐Dec.) was classified according to Yamada et al. (1995). SH ratio= sucrose [g 100 g­1 FW] / hexoses (glucose + fructose) [g 100 g­1 FW]. Sweetness value= soluble solids content (reference value in Table 1S) x SH ratio. Table 3 ­ Sugar composition of pollination­constant astringent (PCA) and pollination­variant astringent (PVA) type of Japanese Fruit ripening time (Late‐Sep., Early‐ to Late‐Oct., Early‐ to Late‐Nov., Early‐Dec.) was classified according to Yamada et al. (1995). SH ratio= sucrose [g 100 g­1 FW] / hexoses (glucose + fructose) [g 100 g­1 FW]. Sweetness value= soluble solids content (reference value in Table 1S) x SH ratio. Fruit ripening time/ cultivar or selection Sugar composition (g 100 g­1 FW) Sugar composition (%) Type of sugar accumulation SH ratio Sweetness value Flesh juiciness (mL g­1 FW)Sucrose Glucose Fructose Total Sucrose Glucose Fructose Late‐Sept. Nishimurawase 3.47 3.22 4.16 10.9 32.0 29.7 38.3 Hexose 0.47 7.1 0.16 Early‐Oct. Akagaki 8.13 2.24 2.2 12.6 64.7 17.8 17.5 Sucrose 1.83 29.3 0.20 Mid‐Oct. Fudegaki 5.45 3.29 3.06 11.8 46.2 27.9 25.9 Intermediate 0.86 14.0 0.21 Ganzan 8.56 2.38 2.11 13.1 65.6 18.2 16.2 Sucrose 1.91 31.1 0.18 Oomiyawase 10.17 1.90 1.83 13.9 73.2 13.7 13.2 Sucrose 2.73 46.1 0.20 Saefuji 7.12 3.93 3.55 14.6 48.8 26.9 24.3 Intermediate 0.95 16.4 0.28 Late‐Oct. Rendaiji 8.45 1.58 1.47 11.5 73.5 13.7 12.8 Sucrose 2.77 41.0 0.20 Early‐Nov. Zenjimaru 8.05 3.41 3.20 14.7 54.9 23.3 21.8 Intermediate 1.22 21.5 0.25 Early‐Dec. Shogatsu 9.84 1.90 1.80 13.5 72.7 14.0 13.3 Sucrose 2.66 46.3 0.19 Fruit ripening time /cultivar or selection Type of astrin­ gency Sugar composition (g 100 g­1 FW) Sugar composition (%) Type of sugar accumula­ tion SH ratio Sweetness value Flesh juiciness (mL g­1 FW)Sucrose Glucose Fructose Total Sucrose Glucose Fructose Mid‐Oct. Ichidagaki PCA 10.43 3.59 3.13 17.1 60.8 20.9 18.2 Sucrose 1.55 30.5 0.35 Tonewase PVA 3.77 5.13 4.48 13.4 28.2 38.3 33.5 Hexose 0.39 5.9 0.25 Late‐Oct. Saijo PCA 8.69 2.12 1.97 12.8 68.0 16.6 15.4 Sucrose 2.12 34.8 0.28 Kawazokogaki PCA 7.80 3.80 3.15 14.8 52.9 25.8 21.4 Intermediate 1.12 18.1 0.31 Early‐Nov. Hiratanenashi PVA 5.54 4.50 3.81 13.8 40.0 32.5 27.5 Hexose 0.67 11.3 0.22 Koshuhyakume PVA 9.62 2.18 1.88 13.7 70.3 15.9 13.7 Sucrose 2.37 41.0 0.31 Hagakushi PCA 8.56 2.32 2.34 13.2 64.8 17.5 17.7 Sucrose 1.84 32.4 0.26 Taigetsu PVA 4.34 4.74 4.25 13.3 32.6 35.6 31.9 Hexose 0.48 7.5 0.43 Taiten PVA 5.93 3.80 2.61 12.3 48.1 30.8 21.2 Intermediate 0.93 15.7 0.46 Mid‐Nov. Aizumishirazu PVA 5.85 3.52 3.40 12.8 45.8 27.6 26.6 Intermediate 0.85 12.7 0.21 Atago PCA 8.15 2.02 1.78 12.0 68.2 16.9 14.9 Sucrose 2.14 33.8 0.23 76 Adv. Hort. Sci., 2020 34(1): 71­79 hyakume’ (SH = 2.37), which was classified as a sucrose accumulator. ‘Taiten’ was an intermediate accumulator in SH = 0.93. Despite the small number of cultivars examined in the present study, there are no general relationships between astringency type (PCNA, PVNA, PVA, and PCA) and sugar accumulation type. In terms of bio­ chemical consideration, varietal difference in the SH ratio can be explained by the degree of sucrose cleavage due to the activity of vacuolar acid invertase (Hirai et al. , 1986; Zheng and Sugiura, 1990). Furthermore, recent transcriptional studies on the sugar accumulation­related key genes postulated that varietal differences in the SH ratio may be result­ ed from the balance between sucrose synthase and vacuolar acid invertase activities in persimmon fruit (Suzuki et al., 2010; Shiraishi and Asakuma, 2019). To date, the genetic mechanisms controlling sucrose accumulation in persimmon fruit remain unclear. However, we hypothesize that dominance of sucrose accumulation over hexose accumulation in persim­ mon fruit on the basis of our ongoing breeding pro­ gram (unpublished data). Environmental variance in sugar composition, flesh juiciness and SSC Table 4 shows the contribution of variance from each trait to the total variance. The variance of geno­ type (σg 2) was high for SH ratio in 66.9% and flesh juiciness in 61.7%. As a whole, the variance of year (σy 2), among trees within genotype (σt 2), genotype × year (σgy 2) and tree × year (σty 2) interactions were small or negligible, varying from 0.0 to 6.6% of the total variance. Similar to previous report (Asakuma and Shiraishi, 2017), the present results indicated that adding year or tree replications will not be effi­ cient in reducing the environmental variance for SH ratio and flesh juiciness. Mitani et al. (2015) also showed that genotypic effect of flesh juiciness is sig­ nificantly high. It is thus considered that the genotyp­ ic effect on the SH ratio and flesh juiciness is high with negligibly small environmental variance, and that these traits can be determined by a one­year trial without tree replication. In contrast, σg 2 of SSC was small in 14.6%, followed by σy 2 in 12.5% and σgy 2 in 8.8%. Other variance components, σt 2 and σty 2 of SSC were small or negligible with 1.5 and 4.2%, respectively. Furthermore, the ratio of σt 2/(σy 2 + σgy 2) is calculated as 0.07. If the ratio exceeds 1.0, tree replications should be required. However, our results indicate that repeated yearly measurements are more efficient than replicated trees to estimate the genetic variance of SSC as observed in grape (Sato et al., 2000) and persimmon (Yamada et al., 1993) Using variance components in Table 4, the error variance (σE 2) of each trait can be obtained by the fol­ lowing equation (cf. Yamada et al., 1993): (σy 2/3) + (σgy 2/3) + (σt 2/3) + {σty 2/(3×3)} + {σ2/(3×3×8)}. The σE 2 of SH ratio, flesh juiciness, and SSC is 0.01366, 0.00014, and 0.16003, respectively. Broad­ sense heritability (σg 2/{σg 2+σE 2}) results in high for SH ratio in 0.95 and flesh juiciness in 0.97, whereas low for SSC in 0.63. In general, a high broad­sense heri­ tability means that most of the variation among genotypes is caused by genetic variation and not environmental variation. Knowing the heritability can be of value when the breeder will make an effective selection. In this study, the high heritability of SH ratio and flesh juiciness is useful to discriminate genetic sweetness and juiciness of persimmon fruit, respectively. Yamada et al. (1993) elucidated that an increase in yearly repetition instead of tree replica­ tions substantially reduced σE 2 in the measurements for SSC and fruit weight to clarify the genetic proper­ ties of genotypes. In the present study, SSC of each Table 4 ­ Estimates of variance component and their percentage to total variance obtained from the analysis of variance Negative value was assumed to be zero SH ratio= sucrose/hexoses (glucose + fructose) Variance components SH ratio Flesh juiciness Solube solids content (SSC) σg2 (genotype) 0.2636 (66.9%) 0.0052 (61.7%) 0.2725 (14.6%) σy2 (year) 0.0260 (6.6%) 0.0000 (0.0%) 0.2326 (12.5%) σgy2(genotype× year) 0.0000 (0.0%) 0.0003 (3.5%) 0.1642 (8.8%) σt2(among trees within genotype) 0.0062 (1.6%) 0.0000 (0.0%) 0.0285 (1.5%) σty2(tree × year) 0.0161 (4.1%) 0.0000 (0.0%) 0.0791 (4.2%) σ2(among fruit within tree) 0.0819 (20.8%) 0.0029 (34.8%) 1.0869 (58.3%) Shiraishi and Asakuma ‐ Sweetness value and flesh juiciness of persimmon fruit 77 cultivar/selection was evaluated as the reference value based on one tree with more than five years field trials (Table S1). Varietal difference in sweetness value and flesh juici‐ ness Sugars represent a crucial component of fruit edi­ ble quality, principally conferring sweetness, one of the main attributes influencing the degree of con­ sumer acceptance. The ratio of constitutive sugars determines the sweetness of fruits; the higher the sucrose percentage, the stronger the organoleptic perception of sweetness in Asian pear (Kajiura et al., 1979), oriental melon (Zhang and Li, 2005), peach (Cirilli et al., 2016) and strawberry (Sone et al., 2000). In this study, we proposed a new index entitled “sweetness value” evaluating fruit sweetness by the equation: SSC × SH ratio. In place of Table S3, SSC value in Table S1 was used for calculation of sweet­ ness value because of the above­described environ­ mental error. As shown in Tables 1­3, the sweetness value varied due to the propotional level of sucrose content, ranging from 3.1 to 58.9 in PCNA­, 7.1 to 46.3 in PVNA­, and 5.9 to 41.0 in PVA­ and PCA­type cultivars. Corresponding to sugar accumulation type, hexose accumulators exhibited lower sweetness value in 3.1 (‘Soushu’) to 11.1 (‘Taiga’), whereas sweetness values of sucrose accumulators were high­ er in 20.8 (‘Uenishiwase’) to 58.9 (‘Sodawase’). These results indicate that sweetness value seems to be a useful predictor of fruit sweetness in persimmon genotype. In our previous sensory tests (Asakuma and Shiraishi, 2017), the less­sweet genotypes exhib­ ited SH ratios below 0.3, while highly­sweet geno­ types had SH ratios exceeding 1.0. Given that the SSC of genotype is around 16 (average value in 16.6 of 43 persimmon cultivars/selections in Table S1), sweet­ ness value of highly­ and less­sweet genotype is expected as exceeding 16 and below 4.8, respective­ ly. For instance, its sensory sweetness of ‘Kishu’ has been evaluated to be lower than that of ‘Fuyu’ by several persimmon breeders and growers, although ‘Kishu’ fruit has normally around 16 in SSC, which is comparable to ‘Fuyu’ (Yamada et al., 2009). In fact, sweetness value of ‘Kishu’ was 3.4 in contrast to that of ‘Fuyu’ in 31.7 (Table 2), which is in agreement with above­mentioned sensory sweetness. However, there can be inconsistencies when eval­ uating the eating quality between sweetness scores and cultivars having different flesh juiciness, particu­ larly less juice content (data not shown). In the pre­ sent study, highly varietal difference in flesh juiciness was observed, ranging from 0.13 (‘Uenishiwase’) to 0.38 (‘Akiou’) mL g­1 FW of PCNA type (Table 2), 0.16 (‘Nishimurawase’) to 0.28 (‘Saefuji’) mL g­1 FW of PVNA type (Table 3), and 0.21 (‘Aizumishirazu’) to 0.46 (‘Taiten’) mL g­1 FW of PVA and PCA type (Table 4). In general, the harder the flesh of fruits, the more chewing is required to breakdown the tissue and the longer it takes to release the juice (Harker et al., 2003). In peach (Suzuki et al., 1990), sweet cherry (Dever et al., 1996), and carrot (Horie and Hiramoto, 2009), flesh sweetness is significantly promoted by high juiciness. Similarly, Ban et al. (2010) and Mitani et al. (2015) revealed that flesh juiciness is consid­ ered to be crucial mouth­feel attribute in the overall taste of persimmon fruit. Asakuma and Shiraishi (2017) proposed that the new descriptor of flesh juiciness of persimmon fruit as “very juicy” (≥0.30 mL g­1 FW), “juicy” (0.21­0.29 mL g­1 FW) and “slightly juicy” (≤0.20 mL g­1 FW) based on the sensory juici­ ness. From this perspective, three sucrose accumu­ lating PCNA­type cultivars/selections (‘Akiou’, ‘Okitsu 20’, and ‘Taiho’) are considered to be promising breeding stocks because of the high sweetness value (21.2 to 27.6 ) and high juiciness (0.31 to 0.38 mL g­1 FW) together with large fruit size and brilliant fruit color (Tables S1 and S3). Thus, in terms of palatabili­ ty, persimmon cultivar’s improvement will be effec­ tively performed using a combination of sweetness value and flesh juiciness. Acknowledgements We are grateful to Dr. Hiroyuki Chijiwa, Fukuoka Agricultural and Forestry Research Center, for reviewing the manuscript and providing suggestions. We also thank Mrs. Masako Hirashima for technical assistance with fruit sampling and sugar extraction. References ASAKUMA H., SHIRAISHI M., 2017 ­ Proposed descriptors for the evaluation of skin color, flesh firmness and juici‐ ness, and sugar composition in Japanese persimmon breeding. ­ Euphytica, 213: 69. BAN Y., KONO A., MITANI N., SATO A., 2010 ­ Estimation of the proportions of offspring having soft fruit and juicy fruit as evaluated by sensory tests in Japanese persim‐ mon. ­ Sci. Hortic., 126(2): 145­151. Adv. Hort. Sci., 2020 34(1): 71­79 78 CIRILLI M., BASSI D., CIACCIULLI A., 2016 ­ Sugars in peach fruit: a breeding perspective. ­ Hortic. Res., 3: 15067. DEVER M.C., MACDONALD R.A., CLIFF M.A., LANA W.D., 1996 ­ Sensory evaluation of sweet cherry cultivars. ­ HortSci., 31(1): 150­153. FAOSTAT, 2017 ­ Production yearbook . ­ Food and Agriculture Organization of the United Nations, Rome, Italy. GIORDANI E., DOUMETT S., NIN S., DEL BUBBA M., 2011 ­ Selected primary and secondary metabolites in fresh persimmon (Diospyros kaki Thunb.): A review of analyt‐ ical methods and current knowledge of fruit composi‐ tion and health benefits. ­ Food Res. Int., 44(7): 1752­ 1767. HARKER F.R., LAU K., GUNSON F.A., 2003 ­ Juiciness of fresh fruit: A time‐intensity study. ­ Postharvest Biol. Technol., 29(1): 55­60. HIRAI S., KONDO T., KINOSHITA H., 2004 ­ The changes of sugars, tannins and sarcocarp cellular structures with maturation and removal of astringency of the persim‐ mon fruits. ­ Bull. Iida Women’s Juni. Colle., 21: 73­81. HIRAI S., ROKUHARA S., SHIMIZU S., 1986 ­ Changes of invertase activity in Japanese persimmon fruits during maturation, storage and processing. ­ Nippon Shokuhin Kogyo Gakkaishi, 33(6): 369­374. HIRANO S., YONEMORI K., SUGIURA A., 1995 ­ Involvement of sugar metabolism in persimmon growth inhibition by calyx lobe removal. ­ J. Amer. Soc. Hort. Sci., 120(1): 75­ 77. HORIE H., HIRAMOTO R., 2009 ­ Enhanced sweetness of carrot by steaming. ­ J. Cook Sci. Jpn., 42(3): 194­197. HUME H.H., 1914 ­ A kaki classification. ­ J. Hered., 5(9): 400­406. IKEDA I., YAMADA M., KURIHARA A., NISHIDA T., 1985 ­ Inheritance of astringency in Japanese persimmon (Diospyros kaki Thunb.). ­ J. Japan. Soc. Hort. Sci., 54(1): 39­45. KAJIURA I., YAMAKI S., OMURA M., AKIHAMA T., MACHIDA Y., 1979 ­ Improvement of sugar content and composi‐ tion in fruits, and classifications of East Asian pears by the principal component analysis of sugar compositions in fruits. ­ Japan. J. Breed., 29(1): 1­12. MAEDA H., AKAGI T., TAO R., 2018 ­ Quantitative charac‐ terization of fruit shape and its differentiation pattern in diverse persimmon (Diospyros kaki) cultivars. ­ Sci. Hortic., 228: 41­48. MITANI N., YAMADA M., SATO A., KONO A., BAN Y., UENO T., SHIRAISHI M., 2015 ­ Objective evaluation of persim‐ mon juiciness: Estimation of environmental variance and varietal differences. ­ Sci. Hortic., 193: 374­380. PARFITT D.E., YONEMORI K., HONSHO C., NOZAKA M., KANZAKI S., SATO A., YAMADA M., 2015 ­ Relationships among Asian persimmon cultivars, astringent and non‐ astringent types. ­ Tree Genet. Genomes, 11: 24. SATO A., YAMADA M., 2016 ­ Persimmon breeding in Japan for pollination‐constant non‐astringent (PCNA) type with marker‐assisted selection. ­ Breed. Sci., 66(1): 60­68. SATO A., YAMADA M., IWANAMI H., HIRAKAWA N., 2000 ­ Optimal spatial and temporal measurement repetition for reducing environmental variation of berry traits in grape breeding. ­ Sci. Hort., 85: 75­83. SHIRAISHI M., ASAKUMA H., 2019 ­ Sugar accumulation profile during the ripening process in the pollination‐ constant non‐astringent Japanese persimmon ‘Akiou’. ­ Hort. J., 88(2): 172­179. SONE K., MOCHIZUKI T., NOGUCHI Y., 2000 ­ Relationship between stability of eating quality of strawberry culti‐ vars and their sugar and organic acid contents. ­ J. Japan. Soc. Hort. Sci., 69(6): 736­743. SUZUKI A., MURAKAMI Y., MAOTANI T., 1990 ­ Characters related to palatability and sweetness of peach fruit. ­ Bull. Fruit Tree Res. Stn., 17: 91­98. SUZUKI T., NIIKAWA T., SHIRATAKE K., 2010 ­ Relationship between sugar composition and gene expression of sugar metabolism enzymes in Japanese persimmon cul‐ tivars. ­ Hort. Res. (Japan) 9 (Suppl 1): 313. TSUJI M., KOMIYAMA Y., 1987 ­ Relationship between invertase activity and sugar composition of persimmon fruit. ­ Nippon Shokuhin Kogyo Gakkaishi, 34(7): 425­ 431. UPOV, 2004 ­ Persimmon. ­ International union for the protection of new varieties of plants, Geneva, Switzerland, pp. 12­13. YAKUSHIJI H., NAKATSUKA A., 2007 ­ Recent persimmon research in Japan. ­ Jpn. J. Plant Sci., 1: 42­62. YAMADA M., 2006 ­ Persimmon, pp. 72­81. ­ In: JAPAN SOC. HORT. SCIENCE (eds.) Horticulture in Japan. Nakanishi Printing, Kyoto, Japan. YAMADA M., SATO A., YAMANE H., MITANI N., IWANAMI H., SHIRAISHI M., HIRAKAWA N., UENO T., KONO A., YOSHIOKA M., 2012 a ­ New persimmon cultivar ‘Taigetsu’. ­ Bull. NARO Inst. Fruit Tree Sci., 14: 25­38. YAMADA M., SATO A., YAMANE H., MITANI N., IWANAMI H., SHIRAISHI M., HIRAKAWA N., UENO T., KONO A.,YOSHIOKA M., 2012 b ­ New persimmon cultivar ‘Taiten’. ­ Bull. NARO Inst. Fruit Tree Sci., 14: 39­52. YAMADA M., YAMANE H., SATO A., YOSHINAGA K., HIRAKAWA N., IWANAMI H., KAKUTANI M., OZAWA T., HIRABAYASHI T., MITANI N., SHIRAISHI M., SUMI T., YOSHIOKA M., NAKAJIMA I., 2009 ­ New Japanese per‐ simmon cultivar ‘Kishu’ ­ Bull. Natl. Inst. Fruit Tree Sci., 8: 25­38. YAMADA M., YAMANE H., UKAI Y., 1995 ­ Genetic analysis of fruit ripening time in Japanese persimmon. ­ J. Amer. Soc. Hort. Sci., 120(6): 886­890. YAMADA M., YAMANE H., YOSHINAGA K., UKAI Y., 1993 ­ Optimal spatial and temporal measurement repetition for selection in Japanese persimmon breeding. ­ HortSci., 28(8): 838­841. YESILOGLU T., CIMEN B., INCESU M., YILMAZ B., 2018 ­ Genetic diversity and breeding of persimmon, breeding and health benefits of fruit and nut crops . ­ IntechOpen. https://doi:10.5772/intechopen.74977. YONEMORI K., SUGIURA A., YAMADA M., 2000 ­ Shiraishi and Asakuma ‐ Sweetness value and flesh juiciness of persimmon fruit 79 Persimmon genetics and breeding. ­ Plant Breed. Rev., 19: 191­225. ZHANG M.F., LI Z.H., 2005 ­ A comparison of sugar‐accu‐ mulating patterns and relative compositions in develop‐ ing fruits of two oriental melon varieties as determined by HPLC. ­ Food Chem., 90(4): 785­790. ZHENG G.H., SUGIURA A., 1990 ­ Changes in sugar compo‐ sition in relation to invertase activity in the growth and ripening of persimmon (Diospyros kaki) fruits. ­ J. Japan. Soc. Hort. Sci., 59(2): 281­287.