Towards efficient improvement of greenhouse grown roses: genetic analysis of vigour and powdery mildew resistance

Rose (Rosa) is one of the important ornamental crops worldwide. Greenhouse production of cut rose and pot rose requires much fossil fuel for optimal plant growth and is adversely affected by infection with powdery mildew ( Podosphaera pannosa ). The heavy energy input and frequent application of fungicides to control the disease contributes considerably to the product cost and environmental pollution. Genetic improvement of cultivars will befacilitated by a better understanding of the genetic variation and the inheritance architecture of the traits determining vigour and mildew resistance.To this end procedures were developed to screen for plant vigour and mildew resistance. These methods were subsequently used with success to study the genetic variation for the two traits present in a diploid and a tetraploid population, respectively. The populations were also molecularly characterized to enable genetic dissection of the variation of the traits. The achievements of the studies are presented in different chapters of this thesis.A high-density genetic map with a number of anchor markers was created for vigour study, using the diploid rose population (Chapter 2). Linkage maps were constructed using a total of 520 molecular markers including AFLP, SSR, PK, RGA, RFLP, SCAR and morphological markers. Seven linkage groups, each putatively corresponding to one of the seven rose chromosomes, were identified for female and male linkage groups spanning 487 cM and 490 cM, respectively. The linkage groups likely cover more than 90 % of the rose genome. The corresponding female and male linkage groups were subsequently integrated. The present linkage maps with robust types of markers are currently the most advanced ones in rose with regard to marker density and genome coverage. The mapped SSR and RFLP markers provide good anchor points for future map alignment studies.Phenotypic evaluation of the diploid population for vigour and its related traits was separately conducted in greenhouses in Denmark (DK) and The Netherlands (NL) under suboptimal growth conditions.A screening procedure for large-scale evaluation of rose genotypes for vigour was developed(Chapter 3). The population showed a continuous quantitative variation as well as a considerable transgression for all the traits. Genetic differences among the tested entries were highly significant and tended to be large for most of the traits in comparison to the effects of genotype by environment interaction. The estimates for heritability were high (68 to 92 %) and the relationships among most of the traits were also high ( r = 0.65 to 0.95). Total shoot dry weight and leaf area are suggested to be good criteria for early selection of rose genotypes with vigorous growth.QTL analyses for each of ten vigour-related traits identified ten chromosomal regions, scattered over the seven linkage groups, containing QTLs for one or more traits (Chapter 4). Considering each trait separately, a total of 42 QTLs was found. Among these QTLs, 24 were common in both DK and NL experiments, whereas eight were only detected in the NL experiment and ten in the DK experiment. The number of QTLs for individual traits varied from three to five, each determining 12 % to 35 % phenotypic variation. QTLs for highly correlated traits were frequently colocalized, indicating a common genetic basis. Clustering of QTLs for different traits was noted in some chromosome regions, suggesting co-localization of several separate genes or/and pleiotropy.The tetraploid population and its parents were tested for powdery mildew resistance under greenhouse conditions with two well-defined monospore isolates (2 and F1) using artificial inoculation with spore suspensions (Chapter 5). Disease score at 11 days post inoculation, latency period and rate of symptom development were used to describe the variation in resistance within the population. The tests for both isolates exhibited a wide and significant variation among genotypes for resistance. The distribution of the genotypic means of the disease scores was continuous and showed a considerable transgression. Analysis of the data indicated that the two isolates differed in pathogenicity. It is the first time that in rose an inoculation assay with spore suspensions has been successfully used. It is a reliable and effective way to screen large numbers of genotypes under greenhouse conditions for mildew evaluation in rose genetic and breeding studies.The tetraploid population for powdery mildew resistance was further molecularly characterized to allow genetic map construction and marker-trait association analysis (Chapter 6). Uni-parental as well as bi-parental simplex AFLP and SSR markers were used to construct female and male maps. These analyses resulted in 23 and 17 separate linkage groups for the maps. The length of the respective maps was 695 and 697 cM. Multi-allelic SSR markers were successfully used to assign most of the linkage groups to one of the seven rose chromosomes. Marker-trait association analyses identified a number of marker loci presumably linked to genes for mildew resistance. These marker loci determined only a moderate part of the heritable variation. Some isolate-specific markers were found from both parents and distributed on different chromosomes, indicating polygenic resistance of the population.The findings of the present study provide knowledge of the inheritance of the target traits at molecular level, which paves the road towards marker-assisted selection for breeding of new rose cultivars with vigorous growth for more efficient energy use and durable resistance to powdery mildew. This will finally result in energy efficient in the cultivation of roses in greenhouse.