| Tulip (Tulipa L.) is one of the most important ornamental crops in the world. It is widelyused as cut flower, pot plant and garden flower. Bulb rot caused by Fusarium oxysporum andTulip fire caused by Botrytis tulipae are two most important fungal diseases. The two diseasessignificantly reduce the quality of cut flower and bulb production, causing huge economiclosses each year. Due to a lack of resistant cultivars on the market, chemical control of thesediseases is needed and this makes tulip into a high consumer crop of pesticides. In recent yers,legalisation of many countries with respect to the application of pesticides in thebulb-growing industry has been tightened considerably. Therefore, breeding new materialswhich resistant to these diseases has become a major breeding target of most tulip breedingcompanies. Tulip has a long juvenil stage (4-5years), which means selection of a new cultivarthrough conventiaonal breeding approaches is a time consuming process. Breeding effeciencyis very low. Thus, fast selection methods are needed. Construct genetic linkage map andmapping QTLs related to the disease resistance. Discover molecular markers that closelylinked to the resistant QTLs and use them in marker assisted breeding can speed up theselection process considerably.This study used F1progeny of an interspecific cross between ‘Kees Nelis’ and ‘Cantata’as mapping population. Based on SNP, SSR, AFLP and NBS markers, the first parental mapshave been constructed using the ‘two-way pseudo-testcross’ strategy. Resistance to Fusariumoxysporum and Botrytis tulipae have been evaluated and QTLs related to the resistance havebeen identified. The study also investigated genetic diversity and structure of a collection of72tulip accessions.1. Tulip F1progeny and parents were genotyped with316SNP markers (151KN_SNP,165CA_SNP) using KASPar genotyping technology. A total of275SNP markers werepolymorphic, in which122KN_SNP only polymorphic in mother ‘Kees Nelis’ while121CA_SNP only polymorphic in father ‘Cantata’. Only1KN_SNP and4CA_SNP werepolymorphic in both parents.2. A total of444and380molecular markers which were polymorphic in F1progeny were selected to construct maternal and paternal map, respectively. For maternal map,342molecular markers (10SNP,2SSR,238AFLP and5NBS) were mapped on27linkagegroups. The maternal map covered1707cM in the genome, with an average marker densityaround3.9cM. Length of linkage groups ranged between17.7cM-130.1cM. The paternalmap consisted21linkage groups on which300molecular markers (99SNP,3SSR,190AFLPand8NBS) were mapped. Paternal map covered1201cM in the genome, with an averagemarker density around3.1cM. Length of linkage groups ranged between6.7cM–122.8cM.3. Resistance to Fusarium oxysporum was evaluated through soil infectionn testscarried out in two consecutive years and a GFP-tagged F. oxysporum inoculation test. It wasfound that resistance level of two parents were significantly separated. F1progeny showed acontinuous distribution from resistant to susceptible phenotypes, indicating the resistance is aquantitative trait. The phenotype distribution of F1progeny fitted normal distribution andshowed clear transgressive segregation. Correltation of the two years soil infect test wasmoderate (r=0.48). In this study, it is the first time to test fusarium resistance using a GFP–tagged F. oxysporum isolate for inoculation. Resistance level was quantified accurately bythe fluorescent signal from green fluorescent protein. Based on the parental genetic linkagemaps and3sets of phenotype data,6QTLs related to fusarium resistance were identified byKruskal-Wallis, IM and MQM. Not all QTLs were found in different disease tests and QTLsshowed different significance level. Contribution of single QTL to phenotypic variationranged from12.5%to20.7%.4. It was the first time to apply spot innoculation method to evaluate resistance toBotrytis tulipae in tulip. The resistance level of F1progeny and parents were quantified byboth manual mesurement and chlorophyll fluorescence analysis. Results showed thatresistance level of two parents were significantly separated. F1progeny showed a continuousdistribution, fitted normal distribution and showed clear transgressive segregation. Takeadvantage of chlorophyll fluorescence to quantify resistance to Botrytis tulipae in tulip avoidinaccurate results due to subjective factors. Based on the infection size (IS) and chlorophyllfluorescence data (DB and TB) as phenotype data,11putative QTLs were discovered byKruskal-Wallis test. However, only6QTLs were confirmed in IM and MQM. Contribution ofsingle QTL to phenotypic variation ranged from7.1%-27.8%.5. A total of121SNP markers were selected to analyze genetic diversity andrelationships among72tulip accessions. The total observed heterozygosity (Ho) among the72accessions was0.35. Ho of cultivar groups defined according to flowering time andmorphology ranged from0.22(Fosteriana) to0.43(Darwin hybrids). In UPGMA, PCoA andSTRUCTURE analysis,72tulip accessions were separated into3clusters. The first cluster included49T. gesneriana cultivars, which belong to8cultivar groups (Triumph, Single early,Single late, Double early, Double late, Fringed, Lily-flowered, Parrot). However, there was nosubgroups within the major group and genetic distances among these cultivar groups wererather small. The second cluster consisted of4T. fosteriana cultivars. The third clusterincluded cultivar ‘Purissima’,‘Flair’ and Darwin hybrids. Cultivar ‘Purissima’ and ‘Flair’were inferred as hybrids as STRUCTURE analysis clearly showed that these two cultivarspossessed both T. gesneriana and T. fosteriana genetic backgrounds. By Analysis ofmolecular variance (AMOVA), significant difference was detected among three clusters (FST=0.208，P<0.0001). In addition, by classify cultivars according to flowering time, a significantdifference was detected between early and late flowering cultivars (FST=0.072, P=0.02). |
PDF Full Text Request