Studies on breeding for shattering habit in paddy rice cultivar 'Asahi'

'Asahi' is a recommended paddy rice (Oryza sativa L. ssp. japonica) cultivar of Okayama Prefecture, Japan. This cultivar was developed by pure-line selection from a landrace in Japan. 'Asahi' has very easily-shattering habit. To improve the shattering habit of 'Asahi', I examined the evaluation method for shattering habit of japonica rice by grasping the panicles. Then I searched the morphological marker trait to evaluate the shattering habit of japonica rice. I also developed pairs of near-isogenic lines for shattering habit of japonica rice by recurrent selfing method. The effects of major factor which controls shattering habit and different genetic backgrounds on grain-shattering degrees were analyzed using the pairs of near-isogenic lines. Finally, I advocated breeding program of the commercial cultivar of near isogenic line for shattering resistant of 'Asahi'. Outline of the results is as follows. 1. Origin and history of 'Asahi' 'Asahi' was registered as a recommended paddy rice cultivar in Okayama Prefecture in 1925. However, the detailed breeding procedure of this cultivar has not been published yet. Available materials relating to breeding procedure of 'Asahi' were perused and arranged. Firstly, the original 'Asahi' was selected from a native landrace 'Hinode' in 1909 by Mr. Shinjiro Yamamoto. The original 'Asahi' was widely grown in Western Japan in the 1920s. Okayama Prefectural Agricultural Experiment Station received 'Asahi' from Kyoto Prefecture in 1917. 'Asahi No.47' was developed through a pure-line selection by Okayama Prefectural Agricultural Experiment Station from 1925 to 1931. Thereafter, 'Asahi No.47' has been grown as 'Asahi' in Okayama Prefecture. 2. Evaluation method for shattering habit of rice by panicle grasping The index for evaluation of shattering habit and the necessary conditions for the method to evaluate the shattering habit by panicle grasping were clarified. Each of the four human evaluators tested 50 plants and grasped 1, 3 or 5 panicles each time on seven cultivars, comprised of three hardly shattering, one moderately, one slightly easily, one easily, and one very easily shattering habits. The data were analyzed as follows. (1) The index for evaluation of shattering habit The rate of shedding per panicle and the number of shedding grains per panicle were contrasted. Three-way (cultivars, evaluators and number of grasped panicles each time) analysis of variance (ANOVA) showed three main effects were statistically significant for the number of shedding grains, whereas for the rate of shedding, two main effects of cultivars and number of grasped panicles each time were statistically significant but the effects of evaluators was not significant. The rate of shedding per panicle was not influenced by evaluators, which were different in hand sizes or in grasped numbers of grains. Therefore, I selected the rate of shedding as an index for evaluation of shattering habit. (2) The number of panicles grasped each time The reproducibility used for investigation on the heritability of the shedding rate, which was estimated from the variance components of the two-way (cultivars and evaluators) ANOVA, was analyzed. The heritability was higher when the evaluator grasped three panicles than one or five panicles. Therefore, it was decided that the number of panicles grasped each time per plant was three. (3) The number of plants necessary to evaluate the shattering habit The effects of reduction in the number of plants used for evaluation of the shattering habit were analyzed by random sampling test. The rate of shedding varied widely with decreasing the number of plants for investigation on moderately shattering, slightly easily shattering, easily shattering and very easily shattering cultivars by 20 times of random sampling test. Therefore, an individual test precludes us from distinguishing the moderately, slightly easily, easily, and very easily shattering cultivars. Use of at least 40 plants was recommended to evaluate shattering habit accurately by the panicle grasping test. (4) The practicality of the individual test for the shattering habit of rice by the panicle grasping method The shedding rate was investigated by grasping three panicles in one hand on each plant. Each of the four evaluators tested 50 plants on each cultivar. From the frequency distribution of shedding rate, a confidence interval was decided by the Clopper-Pearson method. These data suggest that an individual test for the shattering habit by the panicle grasping method can classify hybrid individuals at the segregating generation on the rice breeding into hardly-shattering type and easily-shattering type (moderately, slightly easily, easily, and very easily shattering grades) at 96% confidence level, although the grade of shattering habit could not be classified. Thus, this method may be useful only for selecting the plants with a hardly-shattering habit. 3. Search for the morphological marker trait of shattering habit in japonica rice To facilitate the selection of moderately shattering habit in japonica rice breeding, the marker trait of shattering habit was searched. The grain-shattering pattern and microscopic morphology of the separation zone on pedicels of five japonica rice cultivars with different shattering habits were compared. Almost all grains were separated between the sterile lemma and rudimentary glume in cultivars with moderately shattering, slightly easily shattering, easily shattering and very easily shattering habits. On the other hand, approximately 50% of the grains of the cultivar with hardly shattering habit were torn off at the bent portion of pedicel. A separation pileus was formed on the terminal of the pedicel when the grain was separated between the sterile lemma and rudimentary glume. The pileuses were most prominent in cultivars with easily shattering and very easily shattering habits, while cultivars with a moderately shattering habit had more flat pileuses than prominent pileuses. Additionally, in the cultivar with a hardly shattering habit all pileuses were flat. Development of fibrous cell walls on the separation pileus varied with degree of shattering habit. The easier the grain shattered, the poorer was the development of fibrous cell walls. These results strongly suggest that the shape of separation pileus and the development of fibrous cell walls are related to the shattering habit in japonica rice. Additionally, it is highly probable that the moderately shattering habit is characterized by the grain-shattering pattern that separates between sterile lemma and rudimentary glume with the flat shape of pileus. 4. Effects of major gene and different genetic backgrounds on the degree of shattering habit Eleven pairs of near-isogenic lines (NILs) for shattering habit were developed by recurrent selfing method from a cross between japonica rice cultivars 'Asahi' and 'Setokogane'. 'Asahi' has very easily shattering habit, while 'Setokogane' has hardly shattering habit. In the developing process of NILs, phenotype of Fsub(1) plants and the segregation ratio of shattering type ('hard to shatter' or 'easy to shatter') in Fsub(2) and single seed descent Fsub(3) populations were investigated by the individual test of panicle grasping method. These data suggest that one dominant major gene controls hardly shattering phenotype in this cross combination. I investigated the grain-shattering degrees of NILs by panicle grasping method using 42 plants of each NIL with 3 replications. Unexpectedly, grain-shattering degrees in some lines with easily shattering habit of NILs were significantly different. Accordingly, the effect of genetic background on the grain-shattering degree was also analyzed using 11 pairs of NILs. Not only major gene which controls either type of easily shattering or hardly shattering but also genetic background affected the grain-shattering degree in the lines with easily shattering type. These data suggest that additional loci with small effects on grain-shattering degree exist in the cross between 'Asahi' and 'Setokogane'. Additionally, a dominant major gene derived from 'Setokogane', which involves hardly shattering habit, may behave epistatic to other loci. These pairs of NILs will be useful in studies of shattering effects on the harvesting losses and genes controlling the trait. 5. Breeding program of the commercial cultivar of NIL of 'Asahi' for resistance to shattering From the results of this study, it is likely that the hardly shattering trait of 'Setokogane' can be introduced into 'Asahi' by recurrent backcrossing method. This breeding program is currently in progress at Research Institute for Agriculture, Okayama Prefectural Technology Center for Agriculture, Forestry and Fisheries. The developing procedure of the commercial near-isogenic cultivar of 'Asahi' for resistance to shattering was shown in this section. The BCsub(4)Fsub(4) lines were bred in 2015.