2013 Volume 82 Issue 3 Pages 222-226
The self-incompatible (SI) Japanese pear cultivar ‘Kosui’ was pollinated with pollen collected from a chronically gamma-irradiated ‘Kosui’ tree, and a progeny was obtained. This progeny resulted in the identification of self-compatible (SC) breeding selection, designated 415-1, which showed 74.4% fruit set in a self-pollination test. PCR-based genetic analysis revealed that 415-1 has S-RNase genotype S4S5, which is the same as that of the parent, ‘Kosui’. Pollination trials were used to investigate whether 415-1 harbors a stylar-part mutation or a pollen-part mutation in its SI locus. When 415-1 was pollinated with pollen from cultivars of the same genotype (‘Syuugyoku’ and ‘Oushuu’), no seed-containing fruit were set, indicating that 415-1 contains functional S4- and S5-RNase alleles. On the other hand, when ‘Syuugyoku’ and ‘Oushuu’ were pollinated with pollen from 415-1, fruit-bearing seeds were produced; therefore, we conclude that 415-1 carries a pollen-part mutation at the SI locus. This new self-compatible breeding selection will be useful for the development of new Japanese pear cultivars with SC.
Japanese pear (Pyrus pyrifolia Nakai) is one of the most important fruit in Japan, and breeders have worked for many years to develop disease-resistant cultivars with high fruit quality. For example, the National Agriculture and Food Research Organization Institute of Fruit Tree Science (NIFTS) has been carrying out Japanese pear-breeding projects since 1935. Initially, the goal of these projects was to produce cultivars with high fruit quality and resistance to black spot disease caused by Alternaria alternata (Kajiura and Sato, 1990). Many excellent cultivars have been produced, including ‘Kosui’ (syn. ‘Kousui’), ‘Hosui’ (syn. ‘Housui’), and ‘Akizuki’. All of these cultivars have high fruit quality, high levels of resistance to black spot disease, appropriate harvesting times, and excellent fruit taste. ‘Kosui’, ‘Hosui’, and ‘Akizuki’ are now the leading Japanese pear cultivars.
Most Japanese pear cultivars are SI, and pollinizers must be inter-planted in commercial orchards. Twenty percent of the total annual labor time in Japanese pear production is spent on artificial pollination and fruit thinning; therefore, SC has become an important objective in Japanese pear-breeding programs, since this would allow for stable fruit set without artificial pollination. Our current goals are to breed SC cultivars with high fruit quality and resistance to black spot disease.
SI systems are genetic mechanisms that prevent self-fertilization and promote out-crossing, and are wide-spread in flowering plants (De Nettancourt, 2001). Japanese pear exhibits gametophytic SI, controlled by a single S-locus with multiple alleles (Kikuchi, 1929). The SI locus contains at least two protein-coding genes: one expressed in the haploid pollen, and the other encoding an S-ribonuclease (S-RNase), expressed in the diploid tissues of the style. When the S-haplotype of the pollen matches one of the two S-haplotypes of the pistil, pollen tube growth is arrested in the style, and no fertilization occurs. Because of this SI system, Japanese pear orchards require pollinizers, and artificial pollination is essential to guarantee optimal fruit production.
The SC cultivar ‘Osa Nijisseiki’ was derived from a natural mutation of ‘Nijisseiki’, and was found in Tottori prefecture in western Japan (Furuta, 1980). Pollination tests and morphological analyses revealed that ‘Osa Nijisseiki’ has the same fruit quality and tree characteristics as ‘Nijisseiki’. Sato (1993) reported that ‘Osa Nijisseiki’ is a stylar-part mutant that lacks pistil SI function and retains pollen SI function. However, the use of ‘Osa Nijisseiki’ has not become widespread in Japanese pear orchards because this cultivar shows susceptibility to black spot disease and does not have higher fruit quality than the more popular cultivars. Beginning in 1988 at Tottori University, ‘Osa Nijisseiki’ was used as a parent in breeding other SC cultivars. This resulted in the commercial SC cultivars ‘Akibae’ and ‘Zuisyu’ (Tamura, 2006).
Like Japanese pear, sweet cherry is a member of the family Rosaceae and is generally SI. With the aim of overcoming SI, Lewis of the John Innes Institute in England made nominally incompatible crosses with pollen that had been X-ray irradiated. From the cross of ‘Emperor Francis’ (S3S4) with irradiated pollen of ‘Napoleon’ (S3S4), several SC seedlings were obtained. These included John Innes Seedling 2420 (JI 2420) (Bošković et al., 2000; Lewis, 1949; Lewis and Crowe, 1954; Matthews and Lapins, 1967). The SC cultivar of sweet cherry, ‘Stella’, was bred by scientists at the Canadian Department of Agriculture and named in 1968. ‘Stella’ originated from the cross ‘Lambert’ × JI 2420, which was made in 1956 (Lapins, 1971); therefore, it has been demonstrated that irradiation can be used to produce SC mutants in the family Rosaceae.
The goal of this study was to develop a new SC Japanese pear-breeding selection using chronically gamma-irradiated ‘Kosui’. The irradiated pollen was used to pollinate non-irradiated ‘Kosui’ flowers, and an SC offspring tree was identified. The SC of this plant was investigated using pollination tests and DNA markers.
The 8 Japanese pear cultivars (with S genotypes in parentheses), ‘Gold Nijisseiki’ (unknown), ‘Hosui’ (S3S5), ‘Imamuraaki’ (S1S6), ‘Kosui’ (S4S5), ‘Nijisseiki’ (S2S4), ‘Oushuu’ (unknown), ‘Okusankichi’ (S5S7), and ‘Syuugyoku’ (S4S5) were used in this experiment. All plant materials were obtained from the collection of the NARO Institute of Fruit Tree Science (NIFTS). The S genotypes listed above have been determined by pollination test (Terami et al., 1946) and PCR-RFLP (Castillo et al., 2001; Ishimizu et al., 1999).
Gamma irradiation conditionsIn 1962, a grafted tree of ‘Kosui’ was planted in the gamma field at the Institute of Radiation Breeding, National Institute of Agrobiological Science (Hitachiomiya, Japan) at a distance 77 m from a 60Co source, and thereafter the tree has been irradiated chronically at about 0.024 Gy/8h.
Development of progenyAnthers of the gamma-irradiated ‘Kosui’ were collected from flowers of the tree in 1993, dried to collect the pollen, and the pollen was stored in a freezer until use. In vitro germination capacity of the pollen was tested in medium (1% agar and 10% sucrose) according to the general method (Moore and Janick, 1983). After 24 hr incubation, the number of grain producing pollen tubes was counted under a microscope. In the following year, 450 balloon-stage flowers of a non-irradiated ‘Kosui’ tree were pollinated with the pollen of chronically irradiated ‘Kosui’. The flowers were covered with paper bags to prevent out-crossing. Fruits that developed from the pollinated flowers were harvested and the seeds collected in August. The seeds were stored at 4°C and planted the following spring. A single developed seedling was grown in a nursery for one year and then transplanted to a breeding field.
Pollination test of compatibilityAnthers were gathered from flowers of the pollen parents and dried to collect the pollen. Flowers of the seed parents were emasculated and then pollinated with the collected pollen (Tables 2, 4, and 5). The flowers were covered with paper bags to prevent out-crossing. The numbers of pollinated flowers and harvested fruit with seeds were recorded about 70 days later. A plant was considered to be compatible if more than 30% of the pollinated flowers set seeded fruit.
| Selection/Cultivar | No. of pollinated flowers | No. of fruit with seeds | Rate of fruit set (%) | Result |
|---|---|---|---|---|
| ‘Kosui’ | 60 | 3 | 5.0 | incompatible |
| 415-1 | 90 | 67 | 74.4 | compatible |
| Seed parent (S-genotype) | Pollen parent (S-genotype) | No. of pollinated flowers | No. of fruit with seeds | Rate of fruit set (%) | Result |
|---|---|---|---|---|---|
| 415-1 (S4S5) | ‘Syuugyoku’ (S4S5) | 32 | 0 | 0.0 | incompatible |
| 415-1 (S4S5) | ‘Oushuu’ (S4S5) | 40 | 0 | 0.0 | incompatible |
| ‘Kosui’ (S4S5) | ‘Oushuu’ (S4S5) | 30 | 1 | 3.3 | incompatible |
| ‘Gold Nijisseiki’ (S2S4) | ‘Oushuu’ (S4S5) | 30 | 28 | 93.3 | compatible |
| ‘Hosui’ (S3S5) | ‘Oushuu’ (S4S5) | 30 | 22 | 73.3 | compatible |
| Seed parent (S-genotype) | Pollen parent (S-genotype) | No. of pollinated flowers | No. of fruit with seeds | Rate of fruit set (%) | Result |
|---|---|---|---|---|---|
| ‘Syuugyoku’ (S4S5) | 415-1 (S4S5) | 105 | 60 | 57.1 | compatible |
| ‘Oushuu’ (S4S5) | 415-1 (S4S5) | 42 | 32 | 76.2 | compatible |
| ‘Oushuu’ (S4S5) | ‘Kosui’ (S4S5) | 80 | 2 | 2.5 | incompatible |
The S-RNase genotypes of Japanese pear cultivars were determined by PCR-RFLP according to the protocol of Ishimizu et al. (1999) and Takasaki et al. (2004). Briefly, genomic DNA was extracted from fresh young leaves using a modified CTAB protocol (Yamamoto et al. 2000). The forward and reverse primers for amplification of the S1- to S7-RNase alleles were: “FTQQYQ” (5′-TTTACGCAGCAATATCAG-3′), labeled with the fluorescent chemical, and unlabeled “anti-IIWPNV” (5′-AC(A/G)TTCGGCCAAATAATT-3′). Amplifications were performed using 10 cycles of 94°C for 15 s, 48°C for 30 s, and 70°C for 2 min, followed by 20 cycles of 94°C for 15 s, 48°C for 30 s, and 70°C for 2.5 min, with a final extension at 70°C for 7 min. The PCR products were digested with 7 restriction endonucleases: SfcI, AflII, PpuMI, NdeI, AlwNI, HincII, and AccII. The S1 and S3–S7 PCR-RFLP products were separated and detected using a PRISM 377 DNA sequencer (Applied Biosystems Japan Ltd., Tokyo, Japan). The sizes of the digested fragments were calculated using internal standard DNA markers (GeneScan-500TAMRA; Applied Biosystems Japan Ltd.) with the GeneScan software (Applied Biosystems Japan Ltd.). The digested fragments from the S2-RNase allele were separated by 1.0% agarose gel electrophoresis and visualized by ethidium bromide staining. S1- to S7-RNase genotypes were identified by comparing with S1- to S7-haplotypes of ‘Hosui’ (S3S5), ‘Imamuraaki’ (S1S6), ‘Nijisseiki’ (S2S4), ‘Okusankichi’ (S5S7).
Emasculated ‘Kosui’ flowers were pollinated using pollen from a chronically gamma-irradiated ‘Kosui’ tree (Table 1). The viability of the stored pollen was about 55.3%. Of the 450 pollinated flowers, only two produced fruits with seeds, and each of these fruits had one seed. The 2 seeds were planted in the nursery, but only one seedling survived. This selection was named 415-1 and used in this study.
| Seed parent | Pollen parent | No. of pollinated flowers | No. of fruit with seeds | No. of obtained seeds | No. of viable seedlings |
|---|---|---|---|---|---|
| ‘Kosui’ | Gamma-irradiated ‘Kosui’ | 450 | 2 | 2 | 1 |
Even though the chronically gamma-irradiated ‘Kosui’ pollen was viable, only 2 fruit with seeds were obtained. This result indicated that the original ‘Kosui’ cultivar was strictly SI and SI mutation of the irradiated pollen was very rare.
Self-fertilization of 415-1Beginning 8 years after planting, the SC of 415-1 was investigated using self-pollination tests. Over a period of 3 years, 415-1 flowers were self-pollinated, and the rate of fruit set was 74.4% (Table 2). In comparison, control self-pollination tests with the original ‘Kosui’ cultivar resulted in only 5.0% fruit set, indicating that ‘Kosui’ is SI.
Generally, compatible crosses in Japanese pear give fruit set percentages ranging from 31% to 100% (Sato, 1993). Therefore, the fruit set percentage derived from the selfing of 415-1 (74.4%) was in the range for compatible crosses, and was enough for 415-1 to be characterized as SC.
S-RNase genotyping of 415-1 by PCR-RFLPThe S-RNase genotypes of the breeding selection 415-1 and ‘Oushuu’ were determined by PCR-RFLP. DNA samples were subjected to PCR using a set of S-RNase gene-specific primers. Two PCR products of about 368 bp and 376 bp were detected in the 415-1 sample (Table 3). After digestion with the S4-specific endonuclease NdeI, the 368 bp product was cut into 2 fragments. After digestion with the S3- and S5-specific endonuclease Ppu MI or the S5-specific endonuclease AlwNI, the 376 bp product was cut into 2 fragments, respectively. Neither of the PCR products was digested by any other endonuclease used in this experiment. These results indicated that the S-RNase genotype of 415-1 was S4S5. This S-RNase genotype is the same as those of ‘Kosui’, ‘Syuugyoku’, and ‘Oushuu’ (Table 3).
| Selection/Cultivar | S-allele-specific restriction endonucleases
|
S-RNase genotype | ||||||
|---|---|---|---|---|---|---|---|---|
| Sfc I (S1 specific) | Afl II (S2 specific) | Ppu MI (S3,S5 specific) | Nde I (S4 specific) | Alw NI (S5 specific) | Hinc II (S6 specific) | Acc II (S6,S7 specific) | ||
| 415-1 | − | − | + | + | + | − | − | S4S5 |
| ‘Kosui’ | − | − | + | + | + | − | − | S4S5 |
| ‘Syuugyoku’ | − | − | + | + | + | − | − | S4S5 |
| ‘Oushuu’ | − | − | + | + | + | − | − | S4S5 |
| ‘Gold Nijisseiki’ | − | + | − | + | − | − | − | S2S4 |
| ‘Imamuraaki’ | + | − | − | − | − | + | − | S1S6 |
| ‘Nijisseiki’ | − | + | − | + | − | − | − | S2S4 |
| ‘Hosui’ | − | − | ++ | − | + | − | − | S3S5 |
| ‘Okusankichi’ | − | − | + | − | + | − | + | S5S7 |
++: Two S-RNase fragments digested with restriction endonucleases.
+: One of two S-RNase fragments digested with restriction endonucleases.
−: S-RNase fragments undigested with restriction endonucleases.
Emasculated flowers of 415-1 (S4S5) were pollinated with pollen from ‘Syuugyoku’ (S4S5) and ‘Oushuu’ (S4S5). The following control crosses were also performed: ‘Kosui’ (S4S5) × ‘Oushuu’, ‘Gold Nijisseiki’ (S2S4) × ‘Oushuu’, and ‘Hosui’ (S3S5) × ‘Oushuu’ (30 crosses each). No seeded fruit were obtained from either the 415-1 × ‘Syuugyoku’ or the 415-1 × ‘Oushuu’ crosses, and only one seeded fruit was obtained from the ‘Kosui’ × ‘Oushuu’ crosses (Table 4). On the other hand, high rates of seeded fruit set were obtained from the ‘Gold Nijisseiki’ × ‘Oushuu’ and the ‘Hosui’ × ‘Oushuu’ crosses, indicating that the pollen of ‘Oushuu’ was viable. These results indicated that 415-1 contained active S4- and S5-RNases. Therefore, the styles of 415- 1 retain the natural SI phenotype and can prevent the growth of pollen from plants with the same S-RNase genotype.
Pollen SI phenotype of 415-1Pollen from 415-1 was used to pollinate 105 ‘Syuugyoku’ flowers and 42 ‘Oushuu’ flowers. After 70 days, these pollinations resulted in 60 fruit with seeds from ‘Syuugyoku’ and 32 fruit with seeds from ‘Oushuu’ (57.1% and 76.2% seeded fruit set, respectively) (Table 5). The control cross (‘Oushuu’ × ‘Kosui’; 80 flowers) resulted in only 2 fruit with seeds, indicating that the styles of ‘Oushuu’ contained functional S-RNases. These results indicated that the pollen of 415-1 could be fertilized without block in the styles of plants with the S4S5 genotype. Thus, only the pollen SI function of 415-1 was affected by gamma-induced mutagenesis. In contrast, pollination tests revealed that the SC cultivar ‘Osa Nijisseiki’ lacks pistil function but retains pollen function, and is therefore called a stylar-part mutant of ‘Nijisseiki’ (Sato, 1993). The selection 415-1 obtained in this study is the first pollen-part mutant of Japanese pear to be reported.
Since Japanese pear is generally SI, pollinizers and artificial pollination are required for commercial Japanese pear production. Therefore, it is desirable to produce SC Japanese pear cultivars with good fruit quality, comparable with the quality of the leading cultivars ‘Kosui’ and ‘Hosui’. To this end, we attempted to develop an SC breeding selection using gamma-induced mutagenesis. This resulted in the identification of a new breeding selection named 415-1, which carries a pollen-part mutation in its SI system. The SC selection 415-1 can be used as a parent for SC cultivar breeding, and it will also provide useful material for research on SI in Japanese pear. We expect that future studies will shed light on the inheritance of the pollen SC mutation, and help to explain the molecular nature of this mutation.
