INTRODUCTION

Oriental persimmon (Diospyros kaki Thunb.), which is believed to have originate in China, is a popular fruit, particularly in East Asian countries such as China, Korea, and Japan (Yonemori et al., 2000). The astringent taste of persimmon is due to the presence of tannins that accumulate in the vacuoles of tannin cells, which are idioblasts of the parenchyma. Depending on the effect of seed formation on the loss of natural astringency in the flesh of the fruits at the harvest, persimmon cultivars are classified into four types: Pollination-constant non-astringent (PCNA), pollination variant non-astringent (PVNA), pollination variant astringent (PVA), and pollination constant astringent (PCA) (Kajiura, 1946). PCNA cultivars show certain differences to the other three cultivar types with respect to tannin accumulation in the flesh, and are notably characterized by smaller tannin cells and lower tannin contents (Yonemori & Matsushima, 1985). Furthermore, whereas tannin accumulation typically ceases early in the flesh development of PCNA cultivars, the other three types show continual tannin accumulation up to maturation. The early termination of tannin accumulation in PCNA cultivars is accompanied by the oxidative deactivation of tannins at maturation, resulting in a loss of astringency (Yonemori & Matsushima, 1985).

The genes involved in tannin synthesis in PCNA cultivars are homozygous recessive (Ikeda et al., 1985), and thus PCNA-type F₁ offspring can only be obtained by a reciprocal cross between PCNA cultivars. However, excluding bud sports and similar phenomena, there are only 18 known PCNA cultivars grown in the central part of Japan (Yamada et al., 2012), and as such, there exists a lack of genetic variation. Thus, reciprocal crosses between PCNA cultivars can typically lead to inbreeding depression (Yamada, 1993;Yamada et al., 1994). Accordingly, favorable traits from non-PCNA cultivars need to be introduced in order to overcome inbreeding depression in crosses between PCNA cultivars. However, since crosses between PCNA and non-PCNA types do not produce PCNA-type F₁ offspring, it is necessary to perform back crosses between the PCNA-type and non- PCNA-type F₁ individuals, which produces about 15% ratio of PCNA-type offspring (Ikeda et al., 1985). The oriental persimmon is a hexaploid (Namikawa and Higashi, 1928), but its genomic composition has not been elucidated.

In this latter regard, Kanzaki et al. previously used an amplified fragment length polymorphism (AFLP) method (Vos et al., 1995;Kanzaki et al., 2001) for early screening of PCNA-type F₁, and developed the non-PCNA-specific molecular marker EACC/MCTA-400. This marker is associated with the ASTRINGENCY (AST) locus, which regulates flesh astringency in persimmon. However, it is unable to effectively differentiate PCNA and non-PCNA types in the offspring of crosses using ‘Kurokuma’.

Kanzaki et al. (2010) also used restriction fragment length polymorphism (RFLP) analysis to detect three indel mutations between the ast and AST-linked regions, and developed the following primers: AST-F, PCNA-F (2 forward primers), and 5R3R (1 reverse primer). Use of these markers facilitated the effective screening of PCNA-type F₁ offspring in crosses using ‘Kurokuma’, ‘Nishimurawase’ and ‘Aizumishirazu’. However, although these molecular markers can accurately differentiate between PCNA-type and non-PCNA-types, marker analysis requires initial quantification, and even if the markers are quantified, there can be differences in the size of PCR bands that make differentiation difficult. In addition, given that the experimental process is complex, it can be difficult to efficiently analyze large groups. Therefore, the objectives of this study were to develop new high-resolution melting (HRM)-based molecular markers that can be simply, rapidly, and accurately analyzed.

MATERIALS AND METHODS

Plant materials

Among the genetic resources planted at the Pear Research Institute of the National Institute of Horticultural and Herbal Science, we selected three PCNA and three non-PCNA cultivars with discernible phenotypes. From each sample, we extracted total DNA from 1g of fresh young leaves using a DNeasy® Plant Maxi Kit (QIAGEN).

Direct Sequencing

The 5R region, which is a fosmid-end sequence isolated from a fosmid library of Diospyros. lotus is linked to AST alleles A1, A2, and A3 and also includes indel-3 of the ast allele (Akagi et al., 2006;Kanzaki et al., 2009;Kanzaki et al., 2010). Previously, several primers have been designed from the sequences flanking Indel-3 (Kanzaki et al., 2010), and we used these primers for sequencing the indel-3 region. Sequencing reactions were performed in a DNA Engine Tetrad 2 Peltier Thermal Cycler (BIO-RAD) using ABI BigDye(R) Terminator v3.1 Cycle Sequencing Kits (Applied Biosystems), following the protocols supplied by the manufacturer. Single-pass sequencing was performed on each template using the AST-F and AST-R primer pair. Fluorescent-labeled fragments were purified according to the manufacturer’s (Applied Biosystems) recommended protocol, as this removes the unincorporated terminators and dNTPs. The purified samples thus obtained were analyzed electrophoretically using an ABI 3730XL DNA Analyzer (Applied Biosystems).

Sequence alignment and HRM analysis

Multiple alignment of the obtained sequences was performed using BioEdit v 7.0 software (Tom Hall, North Carolina State University, USA). Differences between PCNA and non-PCNA sequences at the AST locus were analyzed by pairwise sequence alignment, and the results were used in primer design for HRM analysis. PCR was performed using several combinations of primers (Table 1) in a total volume of 25 μL, containing 0.2 mM of each dNTP, 1 U of EX Taq (Takara Bio), 0.2 pmole of each primer, 0.8× reaction buffer, 2 μM fluorescent nucleic acid (Thermo Fisher Scientific), and 50 ng of total DNA. The PCR conditions were as follows: initial denaturation at 95°C for 30 s, followed by 45 cycles of 95°C for 10 s, 60°C for 5 s, 72°C for 5 s. The high-resolution melting conditions were 95°C for 60 s, 40°C for 60 s, 65°C for 1 s and 97°C for 1 s, with subsequent cooling to 37°C for 30 s.

RESULT AND DISCUSSION

In South Korea, breeding of new persimmon cultivars started in earnest in 2007 at the Pear Research Institute of the National Institute of Horticultural and Herbal Science, Naju, Korea. As of 2018, nine PCNA, two PVNA, and three astringent (PCA/PVA) cultivars had been bred. However, these cultivars are variously characterized by adverse traits attributable to inbreeding depression, notably cracking of the fruit apex, micro-cracking and calyx-end cracking, caused by repeated crossing within the small PCNA gene pool. In attempts to prevent inbreeding depression, we have introduced non-PCNA cultivars that are free of such physiological disorders into the breeding program. Annually, we have obtained a large number of F₁ and, BC₁ offspring through crossbreeding PCNA and non-PCNA cultivars, and we are currently using markers to select PCNA offsprings at an early stage.

In order to avoid unnecessary cultivation of non-PCNA offspring in the selection field, we applied DNA markerassisted selection (MAS) to select the PCNA offspring at nursery stage. Previously, Kanzaki et al. (2010) identified the sequence of an AST- and ast-linked region by screening a genomic library with a 5R probe, and identified three types of dominant AST-alleles, A1, A2, and A3, respectively, in ‘Kurokuma’. Accordingly, in the present study, we sought to screen sequences of near the 5R region using five PCNA cultivars and five non-PCNA cultivars to develop a new DNA maker for High-resolution melting (HRM) analysis.

We initially performed PCR using AST-F/AST-R primers in order to re-sequence the 5R region, and then performed multiplex PCR using the five PCNA and non- PCNA cultivars in order to verify the presence or absence of indel-3 in the ast-linked region of the 5R region, using the PCR conditions described by Kanzaki et al. (2010). After PCR, resequencing was performed using the products, and sequence alignment using BioEdit was conducted on sequences from the PCNA and non-PCNA cultivars in order detect indel-3 in the ast-linked region. We accordingly succeeded in identifying a sequence that differed between the PCNA and non-PCNA cultivars, and confirmed that this sequence was that of indel-3 in the ast-linked region (Fig. 1).

All the non-astringent persimmon cultivars currently being grown in South Korea were introduced from Japan, and the non-astringent persimmon from Japan are homozygous recessive compared with astringent persimmon. Since persimmons are hexaploid, the genotype of PCNA cultivars in the AST/ast-linked region is aaaaaa, whereas the genotype of non-PCNA cultivars is A-----. In order to develop a marker that can differentiate PCNA and non- PCNA types, the forward primer was designed to include the base that is indicated by the gray arrow in Fig. 1, as we considered this to be the site where indel-3 is inserted in the ast-linked region, and we designed several primers that can detect A alleles but not a allele (Table 1).

In order to verify whether these primers can be used in cross-breeding, we analyzed PCNA and non-PCNA cultivars and the progeny of crosses between PCNA and PCNA and PCNA and non-PCNA cultivars (Table 2). In line with expectations, we found that the PCNA cultivars showed no bands, and that bands at the predicted molecular weight were only observed in astringent persimmons (Fig. 2A). In order to confirm these results, we used AST-F/PCNA-F/ AST-R primers (Kanzaki et al., 2010) and verified that PCNA and non-PCNA types were accurately separated (Fig. 2B).

In addition, in order to provide convenient analysis for early screening of PCNA lines in seedlings using MAS, we assessed the potential of HRM analysis using our developed primers. The analysis conditions were as follows The analysis indicated that the most accurate results were obtained when using the Hrm1-F/Hrm1-R primers (Fig. 3). The result of HRM analysis revealed that the PCNA cultivars showed a straight line [Fig. 3A and Fig 3C (green color)], whereas the non-PCNA cultivars showed varioussized peaks within the temperature range of 68 to 72ºC [Fig. 3B and Fig 3C (pink colors)].

CONCLUSION

PCNA cultivars are genetically recessive, so they can only be obtained from the progeny of crosses between PCNA types. However, as there are only 18 known PCNA cultivars from Japan, the genetic resources are limited. Under these circumstances, continual breeding results in inbreeding depression in the progeny, causing physiological disorders and making it difficult to breed high-quality cultivars. In order to overcome these limitations, the Pear Research Institute of the RDA is using high-quality culti-vated astringent persimmon as a parent. The F₁ offspring from a PCNA × non-PCNA cross are 100% non-PCNA; however, when the F₁ generation is backcrossed with a PCNA cultivar, the ratio of PCNA offspring in the BC₁ generation is reported to be 15%. Kanzaki et al. (2010) developed a marker for early screening of PCNA using MAS, and this marker was used in the crosses we performed using astringent persimmon. However, as the analysis method in cumbersome and time-consuming, it’s a little hard to readily be applied to a large number of seedlings. Given that, the marker we developed in this study uses the HRM method, we expect this to significantly reduce the time, labor, and cost of analyses compared with previous markers.

적 요

최근에 감에서 떫은맛을 조절하는 AST에 연관된 지역에서 분자 표지들이 개발되었다. 이중에서 sequence characterized amplified region (SCAR) marker는 5R region에 인접한 지 역에서 개발되었다. 하지만 이 SCAR마커는 분석 방법이 다 소 복잡하고 해석이 어려워 많은 교배실생을 분석할 경우에는 적합하지 않다. 우리는 5R 지역의 sequences에 기반하여 high-resolution melting (HRM)-based 분자 표지를 개발하였다. 개발된 HRM preimer set을 8개 품종의 단감 및 떫은감에 대 해 적용한 결과 단감 품종에서는 직선을 나타낸 반면 떫은감 품종에서는 다양한 크기의 곡선으로 나타나서 차이를 확인할 수 있었다. 결과적으로 이번 연구에서 개발된 HRM primer set은 분자 표지를 활용한 감 품종 육성 연구에 매우 효율적으 로 활용될 수 있을 것으로 기대된다.