Molecular analysis and visualisation of adventitious root formation in rose (Rosa L.)

authored by

David Wamhoff

supervised by

Traud Winkelmann

Abstract

Roses (Rosa L.) are among the most important ornamental crops and of great importance in the market segments of cut flowers, garden and pot plants. Due to their highly heterozygous nature, roses must be propagated vegeta-tively to achieve true-to-type multiplication. Most cut and garden rose cultivars are still propagated by xenovegeta-tive propagation, which is labour-intensive and costly. Therefore, propagation by cuttings is of increasing interest to rose breeders. However, prerequisite for this is adventitious root (AR) formation, a complex multiphasic process that is under highly quantitative genetic control. In roses, huge genotypic differences between individual cultivars can be observed and some genotypes even do not form roots at all. Factors that limit AR formation are difficult to identify due to the complexity of the overall process, particularly in the early stages. Therefore, a better under-standing of the process of AR formation in roses on the molecular level and knowledge about the anatomical and biochemical characteristics in the different developmental stages is desirable. The aims of this study were to unravel possible constraints of AR formation in different developmental phases and to achieve a better understanding of the genetic control of this quantitative trait in rose. To characterise the genetic regulation of AR formation, existing genome-wide association studies (GWAS) in a population of 96 garden roses were followed up by analysing a larger number of genotypes and of cuttings per genotype to increase the power of the GWAS. Genotypic single nucleotide polymorphism (SNP) information generated with the 68k WagRhSNP Axiom® SNP array was available for the population of the 96 garden roses and a population of 96 cut roses. First, the popu-lations were characterised for their AR formation in vivo. The experiments were performed with stem cuttings in a hydroponic rooting system under greenhouse conditions for 6 weeks. Analyses of rooting percentage after 6 weeks, root number and root fresh mass revealed no significant associations, but several SNPs with strong allele dosage effects for these traits were identified, that were located in distinct peak regions. Of these, ten SNPs were located in the coding region of the putative phosphoinositide phosphatase SUPRESSOR OF ACTIN 9 (SAC9). This is the first di-rect evidence that the phosphoinositide (PI) signalling pathway may also be involved in the regulation of AR for-mation. For one of these SNPs - namely RhK5_69_1627P - an allele-specific PCR marker was derived and verified in an independent group of rose genotypes homozygous for this SNP. Thus, this allele-specific PCR marker can be used in rose breeding programs to improve AR formation by marker-assisted selection. In further GWAS analyses the early events in AR formation in vitro were investigated for 106 rose genotypes. The formation of root primordia (RP), AR formation and anatomical characteristics of the shoot base were correlated. Significant correlations were found for several rooting-associated traits with vasculature dimensions after 1 week of culture. In addition, it was observed that some genotypes showed RP in high numbers after 1 week, but low AR formation percentages after 3 weeks. GWAS analyses for traits at different stages of AR development showed a higher contribution of distinct genomic regions to phenotypic variation for traits at earlier developmental stages and identified nine SNPs to be significantly associated with different traits. The combination of early RP and later AR formation data as an index enabled to identify genomic regions and factors that potentially prevent RP from emer-gence. In addition, two SNPs located in homologues of genes that are involved in the PI signalling pathway showed strong allele dosage effects for RP/AR formation traits. The visualisation of the early stages in AR formation is of particular interest for genotypes with low rooting capacity, but challenging as most methods are destructive and do not allow observations over time. Magnetic resonance imaging was used to non-destructively visualise microstructures in AR formation already 3 days post excision. Both, the difficult-to-root genotype ᶦMariatheresiaᶦ (MT) and the easy-to-root genotype ᶦHerzogin Friederikeᶦ (HF) showed a pronounced formation of RP. Conspicuous differences in the MRI signal in the region of the shoot cortex in front of the RP referred to further investigation of the biochemical composition in this region of the shoot base. Spatially-resolved Fourier-transformed infrared spectroscopy revealed distinct differences in cell wall composition. In particular, absorbance for lignin and hemicelluloses were significantly higher for MT compared to HF. Thus, not the RP development limited the AR formation capacity in MT, but the emergence of RP was hindered by the altered composition and mechanical strength of the cell walls in regions of the cortex. In conclusion, by genetically dissecting AR formation using GWAS, this study contributed to a better understanding and potential improvement of AR formation in rose in breeding programmes and for the first time provided evi-dence of a possible role of the PI signalling pathway in AR formation. The combination of imaging techniques made it possible to identify anatomical and biochemical factors that limit AR formation, particularly in the early events of AR formation.

Organisation(s)

Section Reproduction and Development

Type

Doctoral thesis

No. of pages

168

Publication date

10.09.2024

Publication status

Published

Electronic version(s)

https://doi.org/10.15488/17972 (Access: Open)