Genotyping-by-Sequencing facilitates genetic mapping of the inflorescence type in Hydrangea

Abstract

Inflorescences of Hydrangea macrophylla are classified as lacecap or mophead, according to the position and number of decorative flowers in the inflorescence. Lacecap inflorescences consist of many small, fertile flowers, which are surrounded by big and sterile, decorative flowers. In contrast, mophead inflorescences contain more decorative flowers that are distributed across the whole inflorescence, which give a ball-like shape. Hydrangea plants with mophead inflorescences are more attractive and thus preferred by consumers. The inflorescence type is inherited in a monogenic, dominant-recessive manner, in which the mophead type is recessive and the lacecap type dominant (Meier, 1990; Uemachi and Okumura, 2012). If a mophead plant is crossed with a lacecap plant, then all progenies will develop the lacecap inflorescences. After backcross with another mophead plant, the offspring will segregate for lacecap and mophead inflorescences, which allows the selection of mophead plants. However, Hydrangea plants need about 13 months to develop inflorescences, which delays determination of the inflorescence type considerably. Analysis of molecular markers, which are linked with the inflorescence type, will enable to follow mophead alleles in complex breeding programs. Furthermore, it will allow markerassisted selection of mophead plants already in the seedling stage. To identify genes, which control the inflorescence type, we crossed a mophead plant with a F1 lacecap plant and produced a pseudo-backcross population (pBC1). This population contains 424 individuals and segregates for the lacecap and mophead inflorescence type in a ratio of 3:1 (Χ2 = 0.034, non-significant at α = 0.05). However, we expect a segregation ratio of 1:1 for monogenic, dominant-recessive inheritance. A 3:1 ratio suggests rather a digenic, dominant-recessive inheritance. Thus, we propose that two genes control the inflorescence type in our population. Currently, we perform a QTL analysis to map these genes. In order to produce a genetic map, we performed a genome-wide marker analysis through applying genotyping-by-sequencing for 381 pBC1 plants. Preliminary sequencing data and mapping results will be presented.