Use of population structure and parentage analyses to elucidate the spread of native cultivars of Japanese chestnut

Abstract

The Japanese chestnut (Castanea crenata Siebold et Zucc.) is naturally distributed throughout Japan and is cultivated for its fruit (nuts) throughout the country. Many native cultivars have cultivation records going back more than 100 years. Researchers have speculated that native cultivars that originated in the Tanba region, the most important region of cultivation, were spread throughout Japan. To clarify the breeding history and spreading pattern of Japanese chestnut cultivars, we estimated the population structure of a set of 60 native chestnut cultivars via hierarchical clustering and Bayesian model-based clustering. Both analyses gave similar results. The cultivars were divided into two main clusters: one with cultivars from the Tanba region, the other with cultivars from other areas of Japan. However, there were some exceptions to this pattern, suggesting that propagation of clones and seeds by humans was a part of the spreading process. Additionally, parent–offspring relationships were estimated from the data obtained for 175 simple sequence repeat markers. Out of the 60 genotypes, nine putative parent–offspring pairs and eight putative parent–offspring trios were identified. These results suggest that native cultivars are likely to have been selected from crosses of older native cultivars. In particular, some native cultivars from outside the Tanba region had parent–offspring relationships with cultivars from the Tanba region. This result suggests that cultivars from outside the Tanba region had been crossed with cultivars from the Tanba region and then selected as a means of introducing favorable traits from the Tanba cultivars.