Search

Trait variation within species is shaped by the genotype and the environment an individual is exposed to. Genomic information is inherited from the parents and forms the basis of the phenotype of an organism. The genetic variation between parents becomes differently distributed between their offspring, leading to trait variation in the offspring. Each trait can be affected by many genes, therefore the genetic architecture can be complex. In complex traits, multiple loci contribute to the ultimate trait value. However, complex traits are shaped not only by genetic variation but also by the environment and the interaction between genotype and environment. The interplay between genetic and environmental variation can affect the fitness of an organism. Chapter 2 discusses how genotype and environment have shaped the phenotype of the nematode Caenorhabditis elegans, the model species used in this thesis, resulting in a laboratory adapted domesticized strain known as Bristol N2. Bristol N2 has been cultivated in the laboratory for over a decade, leading to the fixation of novel mutations in several genes that strongly affect its phenotype. Genotypic variation arisen by novel mutations in the genes npr-1, glb-5, and nath-10 was fixed in N2 due to the laboratory environment. The allelic variation in npr-1 affects the behaviour of this animal in an environment dependent manner, showcasing the interplay between genotype and environment. However, the altered behaviour warrants caution for interpretation of results obtained in the N2 strain. The genotypic effects on trait variation can be large, and one of the more powerful population designs to study these effects are introgression lines. In chapter 3 the construction of a genome-wide introgression line (IL) panel between the N2 and the CB4856 strain is described. This panel contains loci of N2 introgressed in a homogeneous CB4856 background. It is demonstrated that together with CB4856-in-N2 ILs this new genome-wide introgression line library strongly facilitates the dissection of genetic interactions. Chapter 4 and 5 investigate natural variation in infection with Orsay virus, a natural pathogen of the nematode C. elegans. In chapter 4 an assay is developed and tested on two wild-type strains (N2 and JU1580) and two mutant strains with mutations in the RNAi pathway. The development of the virus infection in the separate strains can be traced and the influence of genotype and age on the progression of the infection can be quantified. Furthermore, it is demonstrated that heritable RNAi plays a role in the viral load upon Orsay virus infection, an example of an epigenetically inherited environmental influence. In chapter 5 the assay is applied on an N2xCB4856 recombinant inbred line (RIL) population, after observing a lower viral load in CB4856 compared to N2. The RIL analysis resulted in the identification of two QTL on chromosome IV. These quantitative trait loci (QTL) were verified by CB4856-in-N2 ILs, but the IL analysis also indicated that there could be genetic interactions affecting the QTL. By a transcriptome analysis and a candidate gene search, the gene cul-6 was identified as a candidate underlying the allelic variation between the N2 and CB4856 strain. Chapters 6 and 7 investigate the influence of genetic interactions and the environment on the genetic architecture of gene expression. In chapter 6 a N2xCB4856 RIL population was exposed to heat stress, leading to the identification of a trans-band on the top of chromosome IV. By analysis of candidate genes, cmk-1, egl-4, and eor-1 were implicated as contributing to the heat-stress induced transcriptional response affected by natural variation between N2 and CB4856. Furthermore, the genes with an expression-QTL on the trans-band were indicative of a stress response phenotype. By analysis of CB4856-in-N2 ILs, it was found that this locus leads to increased recovery from stress. In chapter 7 two-loci genetic interactions were mapped for gene expression in a N2xCB4856 RIL panel. These epistatic interactions were confirmed by measuring gene expression in a novel population of inbred line containing the full set of loci combinations. It was found that genetic interactions in gene expression can be identified by clustering and are pervasive. These genetically interacting loci affect evolutionary conserved genes. In conclusion, this thesis unveils the mechanisms underlying the genetic architecture of complex traits in C. elegans resulting from genotype and interactions between genotype and environment. It provides tools to unravel these interactions in C. elegans, by providing the community with new resources such as the N2-in-CB4856 introgression lines. Although C. elegans has been a very powerful platform for quantitative trait dissection, we need to expand our mechanistic understanding of polygenic traits.

Tomato yellow leaf curl disease, a devastating disease of tomato, is caused by a complex of begomoviruses generally referred to as Tomato yellow leaf curl virus (TYLCV). Almost all breeding for TYLCV resistance has been based on the introgression of the Ty-1 and Ty-3 resistance loci derived from Solanum chilense LA1969 and LA1932/LA2779 respectively. The aim of this thesis was to fine map, clone and characterize these two TYLCV resistance genes. The Ty-1 gene has been used in tomato breeding already for almost 20 years. Its exact genetic location was however unknown which made precise marker assisted breeding difficult. Here we have analysed the recombination behavior of the chromosomal region where Ty-1 is introgressed by applying newly developed molecular markers in two F2 populations obtained from two commercial Ty-1 carrying hybrids. A big S. chilense introgression was detected in both populations that coveralmost the whole short arm and a part of the long arm of chromosome 6. In this introgression recombination suppression was detected and Fluorescence in situ Hybridization (FISH) analysis revealed two chromosomal rearrangements between S. lycopersicum and S. chilense LA1969. These rearrangements are most likely the cause of the observed recombination suppression. Using disease tests on progeny of informative recombinants Ty-1 was mapped to a region of approximately 600 kb which partly overlapped with the mapped region for Ty-3, which led to the indication that Ty-1 and Ty-3 could be allelic. Altogether these results nicely demonstrate the usefulness of FISH as a powerful tool to aid in the accurate mapping of genes that are introgressed from wild species into cultivated tomato (Chapter 2). To further fine map and ultimately clone Ty-1 and Ty-3 more plants were screened for recombination events and consequently recombinant inbred lines were generated. By developing new markers in combination with disease tests both genes were fine mapped to a very small, almost similar genomic region (approximately 70 kb). Using a Tobacco Rattle Virus-Virus Induced Gene Silencing approach, the resistance genes were finally identified. It was shown that Ty-1 and Ty-3 are allelic and that they code for a RNA-dependent RNA polymerase (RDR) belonging to the RDRɣ type which has an atypical DFDGD motif in the catalytic domain. In contrast to the RDRαtype, characterized by a catalytic DLDGD motif, no clear function has yet been described for the RDRɣ type. With the identification of Ty-1/Ty-3, a completely new class of resistance genes was unveiled (Chapter 3). The Ty-1/Ty-3 allele is characterized by a 4 amino acid insertion at the 5-prime part of the protein and by a catalytic DFDGD motif. The allelic variation of this gene was examined using cDNA from five S. chilense derived lines and using draft assemblies of whole genome sequences from more than 50 tomato cultivars, landraces and related wild species. Tobacco Rattle Virus induced gene silencing was used to silence Ty-1/Ty-3 and altogether showed that resistance was compromised in three out of five S. chilense derived lines tested. One line with resistance derived from S. chilense LA1971 remained resistance after silencing of Ty-1/Ty-3. For another line (8783, derived from LA1932) only 4 out of 13 plants showed symptoms after silencing,but silencing in this line was inefficient because only one out of three PDS controls showed photobleaching. Comparison of the two typical features of the Ty-1/Ty-3 gene showed no sequence variation amongst S. chilense derived lines. The catalytic domain was found to be conserved among all tomato lines and species analysed, while the characteristic 4 amino acid insertion was also observed in three species closely related to S. chilense, e.g.Solanum corneliomulleri, Solanum peruvianum and Solanum huaylasense. This indicated that most S. chilense accessions most likely carry a functional TYLCV resistance locus on chromosome 6, allelic to Ty-1/Ty-3, and Solanum species related to S. chilense could possibly be useful for future TYLCV resistance breeding (Chapter 4). The Ty-1 gene encoded an RDR and for this reason most likely conferred resistance involving amplification of the siRNA signal. In the last experimental chapter (Chapter 5) this hypothesis was tested. It was shown that upon TYLCV challenging of resistant Ty-1 and Ty-3 lines low virus titers were detected concomitant with the production of relatively high levels of siRNAs. In contrast to the situation in susceptible tomato Moneymaker where high virus titers were observed, but the amount of siRNAs produced lower compared to those in Ty-1 and Ty-3. Analysis of the spatial genomic siRNA distribution showed a consistent and subtle enrichment for siRNAs derived from the CP (V1) and C3 gene in Ty-1 and Ty-3 lines compared with Moneymaker. In tomato plants containing the Ty-2 resistance gene, included as a control and not an RDR, the virus was hardly detectable but the siRNA profile similar to the one observed in TYLCV-challenged susceptible tomato Moneymaker. Furthermore, genome methylation analysis revealed a relative hypermethylation of the TYLCV CP (V1) promoter region in genomic DNA collected from Ty-1 in comparison to susceptible tomato Moneymaker. Altogether this thesis describes the mapping, cloning and characterization of the TYLCV resistance genes Ty-1 and Ty-3. Future breeding efforts for TYLCV resistance can now exploit in-gene markers and the insights obtained can possibly direct future research and breeding efforts on plant virus resistance (Chapter 6).

This PhD project started with the composition of a diverse panel of genotypes that represented, (i) roughly the flavor variation in the commercial Capsicum annuum breeding program of Rijk Zwaan, (ii) parents of available mapping populations and (iii) some genotypes that were expected to have extraordinary flavors. The complete set consisted of 35 genotypes of which 24 genotypes were non-pungent. Volatile and non-volatile compounds as well as some breeding parameters were measured in mature fruits of all genotypes throughout the growing season. In addition, from three harvests the non-pungent genotypes were evaluated for taste by a trained descriptive sensory panel. The biochemical profiling with use of SPME-GC-MS allowed visualization of between- and within-species volatile compound variation. Principal components analysis (PCA) on the intensity patterns of 391 putative volatile compounds revealed individual grouping of C. chinense, C. baccatum var. pendulum and C. annuum, indicating potentially interesting volatile variation present in the former two groups. A large group of saturated and unsaturated esters were mainly responsible for the individual grouping of the C. chinense accessions. Due to the elevated acid concentrations and aberrant volatile profiles of the C. baccatum var. pendulum accessions PEN45 and PEN79, the two BIL populations derived from these accessions were identified as interesting candidates for further study. Compared to e.g. Mazurka the citrate concentration of the C. baccatum accessions was 2.5-3 times higher and the malate concentrations were even up to 12 times higher (Chapter 2). Based on the non-pungent genotypes, we found highly correlated clusters of volatiles and non-volatiles, which could be related to metabolic pathways and common biochemical precursors (Chapter 3). Contrasts between genotypes were caused by both qualitative and quantitative differences in these metabolic clusters, with the phenolic derivatives, higher alkanes, sesquiterpenes and lipid derived volatiles forming the major determinants. For the description of the non-pungent genotypes the panelists used fourteen attributes to describe the flavor sensation in the mouth/throat, which were the texture attributes crunchiness, stickiness of the skin, toughness and juiciness, the basic taste attributes sweetness and sourness and the retronasal flavor attributes aroma intensity, grassiness, green bean, carrot, fruity/apple, perfume, petrochemical and musty. The variation in flavor could be reduced into two major sensory contrasts, which were a texture related contrast and the basic sweet-sour contrast. The structure of the PCA plots resulting from the analysis with one harvest (Chapter 3) and the analysis with the combined three harvests (Chapter 4) remained almost identical, indicating the stability of these contrasts. To relate the sensory attributes to the metabolite data and to determine the importance of the individual compounds we used Random Forest regression on the individual harvests and on the three harvests together. Several predictors for the attributes aroma, fruity/apple, sourness and sweetness were found in common between harvests, which we proposed as key-metabolites involved in flavor determination of sweet pepper (Chapter 4). This list contains compounds with known relations to attributes, like sweetness and sugars, but also several compounds with new relations. In this respect we have demonstrated for the first time, that the metabolites p-menth-1-en-9-al, (E)-β-ocimene, (Z)-2-penten-1-ol, and 1-methyl-1,4-cyclohexadiene are related to fruity/apple taste and/or sweetness of pepper. For sourness the only compound with a consistent significant contribution was an unknown C6H8O2 compound. We postulated therefore the hypothesis that in pepper the effect of sourness related metabolites is masked by other volatile and non-volatile compounds or texture differences (Chapter 3). Subsequently in Chapter 4 we described a clear sweetness-sourness interaction and demonstrated that the masking effect of fructose and other sugars explained why we did not find organic acids contributing to the prediction of sourness. The major sensory attributes were also predicted between harvests. The Random Forest predictions of the texture related attributes (juiciness, toughness, crunchiness and stickiness of the skin) and sweetness were very good. The predictions of the attributes aroma intensity, sourness and fruity/apple were somewhat lower and more variable between harvests, especially in the second harvest. In general, we concluded that prediction of attributes with higher heritabilities works better and is more consistent over harvests (Chapter 4). Based on the results of the initial experiments (Chapter 2) the species C. baccatum was chosen for further study. To exploit the potential flavor wealth of C. baccatum PEN45 we combined interspecific crossing with embryo rescue, resulting in a multi-parent BC2S1 population, that was characterized for sensory and biochemical variation (Chapter 5). We developed a population specific genetic linkage map for QTL mapping of characterized traits. Because of the complex structure of our BC2S1 mapping population we encountered several limitations, such as accidental co-segregation, underrepresentation of color linked markers and pre-selection leading to skewness, which might have resulted in false positive or missed QTLs. Despite these limitations, we were still fairly well able to map several biochemical, physical and sensory traits, as demonstrated at first for the (monogenic) control traits red color and pungency in the BC2S1 mapping population and in second instance by validation of genetic effects via an experiment with near-isogenic lines (NILs).This two-step approach turned out to be very powerful, since it led to the identification of the main results from this thesis: (i) Asmall C. baccatum LG3 introgression causing an extraordinary effect on flavor, which resulted in significantly higher scores for the attributes aroma, flowers, spices, celery and chives. In an attempt to identify the responsible biochemical compounds few consistently up- and down-regulated metabolites were detected, including the well-known pepper compound 2-isobutyl-3-methoxypyrazine (down) and 6-methyl-4-oxo-5-heptenal (up); (ii) Two introgressions (LG10.1 and LG1) had major effects on terpenoid content of mature fruits, affecting at least fifteen different monoterpenes; (iii) A second LG3 fragment resulted in a strong increase in Brix (total soluble solids) without negative effects on fruit size (Chapter 5). In Chapter 6 some extra sensory results of the pungent genotypes are given and a comparison between the two C. baccatum pendulum BILs (PEN45 and PEN79 derived) is made in light of the overall results. Finally the perspectives for breeding are discussed and presented in the form of a flowchart for flavor improvement.

Tomato yellow leaf curl disease, a devastating disease of tomato, is caused by a complex of begomoviruses generally referred to as Tomato yellow leaf curl virus (TYLCV). Almost all breeding for TYLCV resistance has been based on the introgression of the Ty-1 and Ty-3 resistance loci derived from Solanum chilense LA1969 and LA1932/LA2779 respectively. The aim of this thesis was to fine map, clone and characterize these two TYLCV resistance genes. The Ty-1 gene has been used in tomato breeding already for almost 20 years. Its exact genetic location was however unknown which made precise marker assisted breeding difficult. Here we have analysed the recombination behavior of the chromosomal region where Ty-1 is introgressed by applying newly developed molecular markers in two F2 populations obtained from two commercial Ty-1 carrying hybrids. A big S. chilense introgression was detected in both populations that coveralmost the whole short arm and a part of the long arm of chromosome 6. In this introgression recombination suppression was detected and Fluorescence in situ Hybridization (FISH) analysis revealed two chromosomal rearrangements between S. lycopersicum and S. chilense LA1969. These rearrangements are most likely the cause of the observed recombination suppression. Using disease tests on progeny of informative recombinants Ty-1 was mapped to a region of approximately 600 kb which partly overlapped with the mapped region for Ty-3, which led to the indication that Ty-1 and Ty-3 could be allelic. Altogether these results nicely demonstrate the usefulness of FISH as a powerful tool to aid in the accurate mapping of genes that are introgressed from wild species into cultivated tomato (Chapter 2). To further fine map and ultimately clone Ty-1 and Ty-3 more plants were screened for recombination events and consequently recombinant inbred lines were generated. By developing new markers in combination with disease tests both genes were fine mapped to a very small, almost similar genomic region (approximately 70 kb). Using a Tobacco Rattle Virus-Virus Induced Gene Silencing approach, the resistance genes were finally identified. It was shown that Ty-1 and Ty-3 are allelic and that they code for a RNA-dependent RNA polymerase (RDR) belonging to the RDRɣ type which has an atypical DFDGD motif in the catalytic domain. In contrast to the RDRαtype, characterized by a catalytic DLDGD motif, no clear function has yet been described for the RDRɣ type. With the identification of Ty-1/Ty-3, a completely new class of resistance genes was unveiled (Chapter 3). The Ty-1/Ty-3 allele is characterized by a 4 amino acid insertion at the 5-prime part of the protein and by a catalytic DFDGD motif. The allelic variation of this gene was examined using cDNA from five S. chilense derived lines and using draft assemblies of whole genome sequences from more than 50 tomato cultivars, landraces and related wild species. Tobacco Rattle Virus induced gene silencing was used to silence Ty-1/Ty-3 and altogether showed that resistance was compromised in three out of five S. chilense derived lines tested. One line with resistance derived from S. chilense LA1971 remained resistance after silencing of Ty-1/Ty-3. For another line (8783, derived from LA1932) only 4 out of 13 plants showed symptoms after silencing,but silencing in this line was inefficient because only one out of three PDS controls showed photobleaching. Comparison of the two typical features of the Ty-1/Ty-3 gene showed no sequence variation amongst S. chilense derived lines. The catalytic domain was found to be conserved among all tomato lines and species analysed, while the characteristic 4 amino acid insertion was also observed in three species closely related to S. chilense, e.g.Solanum corneliomulleri, Solanum peruvianum and Solanum huaylasense. This indicated that most S. chilense accessions most likely carry a functional TYLCV resistance locus on chromosome 6, allelic to Ty-1/Ty-3, and Solanum species related to S. chilense could possibly be useful for future TYLCV resistance breeding (Chapter 4). The Ty-1 gene encoded an RDR and for this reason most likely conferred resistance involving amplification of the siRNA signal. In the last experimental chapter (Chapter 5) this hypothesis was tested. It was shown that upon TYLCV challenging of resistant Ty-1 and Ty-3 lines low virus titers were detected concomitant with the production of relatively high levels of siRNAs. In contrast to the situation in susceptible tomato Moneymaker where high virus titers were observed, but the amount of siRNAs produced lower compared to those in Ty-1 and Ty-3. Analysis of the spatial genomic siRNA distribution showed a consistent and subtle enrichment for siRNAs derived from the CP (V1) and C3 gene in Ty-1 and Ty-3 lines compared with Moneymaker. In tomato plants containing the Ty-2 resistance gene, included as a control and not an RDR, the virus was hardly detectable but the siRNA profile similar to the one observed in TYLCV-challenged susceptible tomato Moneymaker. Furthermore, genome methylation analysis revealed a relative hypermethylation of the TYLCV CP (V1) promoter region in genomic DNA collected from Ty-1 in comparison to susceptible tomato Moneymaker. Altogether this thesis describes the mapping, cloning and characterization of the TYLCV resistance genes Ty-1 and Ty-3. Future breeding efforts for TYLCV resistance can now exploit in-gene markers and the insights obtained can possibly direct future research and breeding efforts on plant virus resistance (Chapter 6).

Many plant species can hybridise and produce fertile offspring. Hybridization between cultivated species and their wild relatives has raised concerns with regard to GM crops, as it constitutes a possible route along which the transgene could disperse from crops into related wild species, establish itself in the natural population, and persist under natural conditions. This may cause unintended ecological consequences such as the formation of more invasive weeds and genetic erosion. After crop-wild hybridization, the persistence of the hybrids and of the crop genes (including the transgenes) in later generations depends on their genetic make-up, which consists of specific combinations of wild and crop genomic segments, and on the environmental conditions. Therefore, knowledge on the dynamics of crop-wild hybridization and introgression using conventional crop varieties is needed as it constitutes the baseline for putting into perspective the effects of transgene introgression under natural conditions. This study focused on understanding the genetic process of hybridization and introgression from cultivated to wild relative species using cultivated lettuce (Lactuca sativa L.) and its closest wild relative, prickly lettuce (L. serriola L.) as a crop-wild complex model. Natural populations of prickly lettuce in Europe were tested for the occurrence of crop-wild hybrids using Bayesian-based programmes (Structure, InStruct and NewHybrids) which returned a crop-wild hybrid occurrence of 7%. However, the geographical location of the hybrids led to the conclusion that crop-wild hybridization is not the cause of the recent northward spread of L. serriola in Europe. To test for the importance of the crop genomic segments to the performance of lettuce crop-wild hybrids, three hybrid generations were created (F2 as a selfing generation, and BC1 and BC2, backcrossed to the wild genotype to mimick the introgression process) by crossing L. sativa (cv. Dynamite) with L. serriola collected from Eys (the Netherlands). The three populations were genotyped with Single Nucleotide Polymorphism markers for genetic analysis. Because of the importance of the abiotic stresses as selection factors under natural conditions and the prospective of generating GM crop varieties with enhanced abiotic stress tolerance, the three populations were evaluated for vigour at the rosette stage under greenhouse conditions of non-stress, drought, salinity and nutrient deficiency. The BC1 population was also evaluated under field conditions at two locations (Wageningen and Sijbekarspel, the Netherlands) for hybrid germination, vigour, survival and reproduction. Based on the location and allelic effect of the QTLs for germination, vigour, survival and reproductive traits in the current lettuce crop-wild cross, genomic regions were suggested where transgenes could be or could not be located in order to mitigate their persistence in crop-wild hybrids through genetic hitchhiking and background selection.

Many plant species can hybridise and produce fertile offspring. Hybridization between cultivated species and their wild relatives has raised concerns with regard to GM crops, as it constitutes a possible route along which the transgene could disperse from crops into related wild species, establish itself in the natural population, and persist under natural conditions. This may cause unintended ecological consequences such as the formation of more invasive weeds and genetic erosion. After crop-wild hybridization, the persistence of the hybrids and of the crop genes (including the transgenes) in later generations depends on their genetic make-up, which consists of specific combinations of wild and crop genomic segments, and on the environmental conditions. Therefore, knowledge on the dynamics of crop-wild hybridization and introgression using conventional crop varieties is needed as it constitutes the baseline for putting into perspective the effects of transgene introgression under natural conditions. This study focused on understanding the genetic process of hybridization and introgression from cultivated to wild relative species using cultivated lettuce (Lactuca sativa L.) and its closest wild relative, prickly lettuce (L. serriola L.) as a crop-wild complex model. Natural populations of prickly lettuce in Europe were tested for the occurrence of crop-wild hybrids using Bayesian-based programmes (Structure, InStruct and NewHybrids) which returned a crop-wild hybrid occurrence of 7%. However, the geographical location of the hybrids led to the conclusion that crop-wild hybridization is not the cause of the recent northward spread of L. serriola in Europe. To test for the importance of the crop genomic segments to the performance of lettuce crop-wild hybrids, three hybrid generations were created (F2 as a selfing generation, and BC1 and BC2, backcrossed to the wild genotype to mimick the introgression process) by crossing L. sativa (cv. Dynamite) with L. serriola collected from Eys (the Netherlands). The three populations were genotyped with Single Nucleotide Polymorphism markers for genetic analysis. Because of the importance of the abiotic stresses as selection factors under natural conditions and the prospective of generating GM crop varieties with enhanced abiotic stress tolerance, the three populations were evaluated for vigour at the rosette stage under greenhouse conditions of non-stress, drought, salinity and nutrient deficiency. The BC1 population was also evaluated under field conditions at two locations (Wageningen and Sijbekarspel, the Netherlands) for hybrid germination, vigour, survival and reproduction. Based on the location and allelic effect of the QTLs for germination, vigour, survival and reproductive traits in the current lettuce crop-wild cross, genomic regions were suggested where transgenes could be or could not be located in order to mitigate their persistence in crop-wild hybrids through genetic hitchhiking and background selection.

Among the different abiotic and biotic stresses, Botrytis cinerea, Phytophthora infestans and high salt concentrations are world-wide the most destructive. Several wild relatives of tomato were identified as source for tolerance to these stresses. Three introgression line (IL) populations derived from S. habrochaites LA1777, S. pennellii LA716 and S. lycopersicoides LA2951 were employed to identify quantitative trait loci (QTL). For B. cinerea resistance twenty four QTLs were identified in S. habrochaites LA1777 and S. lycopersicoides LA2951. These QTLs resulted in reduced lesion size (LS) and disease incidence (DI) in leaves, stem or fruits. Five QTLs were found in S. habrochaites LA1777 for reduced LS in the interaction between tomato and P. infestans. For salt tolerance in the seedling stage ten QTLs were identified in S. pennellii LA716 and five in S. lycopersicoides LA2951. Some QTLs were semi-dominant with a non-additive or even epistatic effect. Many QTLs co-localized indicating that cross talk between coordinating pathways for abiotic and biotic stress might exist. The results provide the basis to combine QTLs with tolerance to abiotic and biotic stresses and for further narrowing down the size of the introgressions. The introgressions from these wild relatives which are involved in tolerance to multiple stresses are of interest for tomato breeders.

Among the different abiotic and biotic stresses, Botrytis cinerea, Phytophthora infestans and high salt concentrations are world-wide the most destructive. Several wild relatives of tomato were identified as source for tolerance to these stresses. Three introgression line (IL) populations derived from S. habrochaites LA1777, S. pennellii LA716 and S. lycopersicoides LA2951 were employed to identify quantitative trait loci (QTL). For B. cinerea resistance twenty four QTLs were identified in S. habrochaites LA1777 and S. lycopersicoides LA2951. These QTLs resulted in reduced lesion size (LS) and disease incidence (DI) in leaves, stem or fruits. Five QTLs were found in S. habrochaites LA1777 for reduced LS in the interaction between tomato and P. infestans. For salt tolerance in the seedling stage ten QTLs were identified in S. pennellii LA716 and five in S. lycopersicoides LA2951. Some QTLs were semi-dominant with a non-additive or even epistatic effect. Many QTLs co-localized indicating that cross talk between coordinating pathways for abiotic and biotic stress might exist. The results provide the basis to combine QTLs with tolerance to abiotic and biotic stresses and for further narrowing down the size of the introgressions. The introgressions from these wild relatives which are involved in tolerance to multiple stresses are of interest for tomato breeders.

Lilies (Lilium L.) are one of the most important ornamental bulbous crops for cut flower industry grown extensively in The Netherlands for last few decades. The genus Lilium consists of seven different sections with about 80 species. The species within genus Lilium (2n = 2x = 24) comprise a range of desirable and complementary characters. Besides being an important horticultural crop, lily (Lilium) also serves as an interesting model plant for molecular cytogenetic research and introgression breeding for several reasons like, i). Lily is a model crop for interspecific hybridization and it includes plants of different taxonomic species each of which possess valuable horticultural traits that need to be combined in the new cultivars. ii) Through careful selection n and 2n gametes can be obtained in interspecific hybrids. iii) The genomes of different species are so well differentiated genetically that the parental chromosomes can be clearly identified in the F1 hybrids as well as in the progenies through DNA in situ hybridization techniques. iv) The chromosomes are large enough and the number and position of homoeologous recombination sites can be clearly detected. v) The large and easily identified chromosomes in different lily species could be a potential source for the cytological mapping of the Lilium genomes. Taking advantage of these favourable attributes of lily, a molecular cytogenetic investigation was conducted to evaluate the amount of recombination and introgression of characters between Longiflorum - Asiatic (LA) and Oriental - Asiatic (OA) hybrids through the use of n and 2n gametes. For this purpose different F1 Longiflorum × Asiatic (LA) and Oriental × Asiatic (OA) hybrids were backcrossed with different Asiatic cultivars. Ovule and embryo rescue techniques were employed to get backcross (BC) progenies. Most of the F1 LA appeared to be sterile but some hybrids were able to produce only 2n gametes in considerable frequencies. However, in rare occasions it was also found that normal meiosis took place in few of the LA hybrids which resulted into the formation of normal n gametes. Ploidy level and intergenomic recombination was studied in LA interspecific hybrids in order to assess the possibility of functional n gametes and their potential use in introgression at diploid level in lily. A total of 104 BC1 LA interspecific lily hybrids were obtained, 27 diploids (2n = 2x = 24), 73 triploids (2n = 2x = 36) and 4 aneuploids (2x – 1, 2x + 2 or 2x + 3). Similarly, triploid BC1 (LAA) plants were backcrossed to diploid Asiatic parents. As a result 14 diploid BC2 progenies were produced. The intergenomic recombination and amount of introgression of respective genome (L and A) was assessed in these diploid genotypes through GISH (Genomic in situ Hybridization). Extensive intergenomic recombination was found among the chromosomes in LA hybrids. A large of amount of L- genome was transmitted from F1 LA hybrids to their subsequent BC1 progenies. However, very few segments of L- genome were introgressed from the BC1 diploid and triploid (LAA) plants to the BC2 progenies (Chapter 2). GISH identifies a considerable amount of recombination events amongst different interspecific lily hybrids (LA and OA) obtained from functional 2n gametes. Based on recombination sites on different chromosomes identified by GISH, cytological maps of three genomes of Lilium were constructed. For this purpose, BC progenies of two diploid interspecific hybrids of lily, viz., Longiflorum × Asiatic (LA) and Oriental × Asiatic (OA) were used. The BC progenies of LA hybrids consisted of both triploid (2n = 3x = 36) and diploid (2n = 2x = 24) with some aneuploid genotypes and those of OA hybrids mostly consisted of triploid (2n = 3x = 36) and some aneuploid genotypes. In LA hybrids 248 recombination sites were cytologically localized on 12 different chromosomes of each genomes (i.e., L and A). Similarly, 116 recombinant sites were marked on 12 chromosomes each from the BC progenies of OA hybrids (O and A genomes). The distances of the recombination sites from the centromeres are measured (in micrometres). Based on these recombination sites four cytological maps were constructed. Since an Asiatic parent was involved in both hybrids, viz., LA and OA, two maps were constructed for A genome which were indicated as Asiatic (L) and Asiatic (O) and one each for Longiflorum (A) and Oriental (A) genomes (Chapter 3). With a view to generate genetic variation via homoeologous recombination in BC progenies of LA and OA hybrids the most logical approach was the use of 2n gametes. 63 BC1 LA (LA × AA or AA × LA) and 53 OA (AA × OA) progeny plants were obtained after unilateral sexual polyploidization. 16 genotypes from F2 LA populations were obtained after bilateral sexual polyploidization through sib-mating of F1 LA hybrids. GISH was employed for the identification of the parental genomes, mode of origin of these progenies and measurement of the introgression in different interspecific lily hybrids. Most of the BC1 progeny plants (LA and OA) had originated through 2n gametes by First Division Restitution (FDR) mechanism. However, there were 12 genotypes in LA hybrids and four genotypes in OA hybrids that originated through 2n gametes formation as the result of Indeterminate Meiotic Restitution (IMR). A higher amount of recombination was found in LA hybrids as compared to OA hybrids. Intergenomic recombination was also determined in the sib-mated F2 LA population. In this case both parents had contributed gametes with the somatic number of chromosomes (i.e., 2n-2n) thus confirming the event of bilateral sexual polyploidization in interspecific LA hybrids. Based on these results, the relevance of interspecific lily hybrids obtained from uni- and bilateral sexual polyploidization leading to allotriploid and allotetraploid formation in interspecific lily hybrids is discussed in the context of introgression and mapping (Chapter 4). Molecular markers are an important tool for the construction of genetic linkage maps, as the first step in the genetic dissection of the required traits leading to crop improvement followed by the marker assisted breeding in different plants. Lilium has one of the largest genome in plant kingdom and genetic mapping in lilies is constrained by its large genome. DArT (Diversity Array Technology), a molecular marker technique can detect and type DNA variation at several hundred genomic loci in parallel without relying on genome sequence information. The DArT technique was developed for Longiflorum × Asiatic (LA) lily hybrids to enable an efficient and effective genetic mapping with the production of a large numbers of markers in microarrays-based assay. The restriction enzyme PstI + TaqI combination generated the largest frequency of polymorphic genomic representations for a genotyping array. Genomic representations from 88 F1 LA plants were used to assemble a DArT genotyping microarray. A total of 687 DArT markers were developed and 382 polymorphic markers were mapped on 14 main linkage groups which is two more then the haploid chromosome number (i.e. n = 12). The resulting linkage map with 382 DArT markers spanned 1329 cM (3.5 cM/marker on average). The results highlighted the potential of DArT as a genetic technique for genome profiling in the context of molecular breeding and genomics, especially in crops with large genome sizes where other techniques proved to be less useful (Chapter 5). The results of the present investigation are of practical implication. These results show the advantages of the n gametes and their subsequent progenies which opened a new approach of lily breeding ‘the analytic breeding’ in the allopolyploids. It also shows the possibility of using certain triploid hybrids for further breeding. A comparison has been made between different types of interspecific crosses, the amount of intergenomic recombination and introgressions of chromosomal segments obtained after unilateral sexual polyploidization. Furthermore, bilateral sexually polyploidization via sib-mated F1 hybrids producing 2n gametes has been proven. The use of allotetraploids obtained from bilateral sexual polyploidization could be a novel approach in the breeding of LA-hybrids. These allotetraploid with recombinant chromosomal segment may be a potential source to generate genetic variation in subsequent progenies. The molecular cytogenetic GISH and FISH techniques proved to be a powerful tool useful for the construction of cytogenetic maps in interspecific crosses in crops with large genomes sizes like lily. These techniques are also used for the identification and integration of genetic maps with chromosome maps. FISH also helps to monitor the introgressed chromosome segment or marker of interest in the subsequent progenies. Application of the DArT technique proved to be an effective method to construct genetic linkage maps especially crops (like Lilium) with large genome sizes where other techniques might be less useful.

Lilies (Lilium L.) are one of the most important ornamental bulbous crops for cut flower industry grown extensively in The Netherlands for last few decades. The genus Lilium consists of seven different sections with about 80 species. The species within genus Lilium (2n = 2x = 24) comprise a range of desirable and complementary characters. Besides being an important horticultural crop, lily (Lilium) also serves as an interesting model plant for molecular cytogenetic research and introgression breeding for several reasons like, i). Lily is a model crop for interspecific hybridization and it includes plants of different taxonomic species each of which possess valuable horticultural traits that need to be combined in the new cultivars. ii) Through careful selection n and 2n gametes can be obtained in interspecific hybrids. iii) The genomes of different species are so well differentiated genetically that the parental chromosomes can be clearly identified in the F1 hybrids as well as in the progenies through DNA in situ hybridization techniques. iv) The chromosomes are large enough and the number and position of homoeologous recombination sites can be clearly detected. v) The large and easily identified chromosomes in different lily species could be a potential source for the cytological mapping of the Lilium genomes. Taking advantage of these favourable attributes of lily, a molecular cytogenetic investigation was conducted to evaluate the amount of recombination and introgression of characters between Longiflorum - Asiatic (LA) and Oriental - Asiatic (OA) hybrids through the use of n and 2n gametes. For this purpose different F1 Longiflorum × Asiatic (LA) and Oriental × Asiatic (OA) hybrids were backcrossed with different Asiatic cultivars. Ovule and embryo rescue techniques were employed to get backcross (BC) progenies. Most of the F1 LA appeared to be sterile but some hybrids were able to produce only 2n gametes in considerable frequencies. However, in rare occasions it was also found that normal meiosis took place in few of the LA hybrids which resulted into the formation of normal n gametes. Ploidy level and intergenomic recombination was studied in LA interspecific hybrids in order to assess the possibility of functional n gametes and their potential use in introgression at diploid level in lily. A total of 104 BC1 LA interspecific lily hybrids were obtained, 27 diploids (2n = 2x = 24), 73 triploids (2n = 2x = 36) and 4 aneuploids (2x – 1, 2x + 2 or 2x + 3). Similarly, triploid BC1 (LAA) plants were backcrossed to diploid Asiatic parents. As a result 14 diploid BC2 progenies were produced. The intergenomic recombination and amount of introgression of respective genome (L and A) was assessed in these diploid genotypes through GISH (Genomic in situ Hybridization). Extensive intergenomic recombination was found among the chromosomes in LA hybrids. A large of amount of L- genome was transmitted from F1 LA hybrids to their subsequent BC1 progenies. However, very few segments of L- genome were introgressed from the BC1 diploid and triploid (LAA) plants to the BC2 progenies (Chapter 2). GISH identifies a considerable amount of recombination events amongst different interspecific lily hybrids (LA and OA) obtained from functional 2n gametes. Based on recombination sites on different chromosomes identified by GISH, cytological maps of three genomes of Lilium were constructed. For this purpose, BC progenies of two diploid interspecific hybrids of lily, viz., Longiflorum × Asiatic (LA) and Oriental × Asiatic (OA) were used. The BC progenies of LA hybrids consisted of both triploid (2n = 3x = 36) and diploid (2n = 2x = 24) with some aneuploid genotypes and those of OA hybrids mostly consisted of triploid (2n = 3x = 36) and some aneuploid genotypes. In LA hybrids 248 recombination sites were cytologically localized on 12 different chromosomes of each genomes (i.e., L and A). Similarly, 116 recombinant sites were marked on 12 chromosomes each from the BC progenies of OA hybrids (O and A genomes). The distances of the recombination sites from the centromeres are measured (in micrometres). Based on these recombination sites four cytological maps were constructed. Since an Asiatic parent was involved in both hybrids, viz., LA and OA, two maps were constructed for A genome which were indicated as Asiatic (L) and Asiatic (O) and one each for Longiflorum (A) and Oriental (A) genomes (Chapter 3). With a view to generate genetic variation via homoeologous recombination in BC progenies of LA and OA hybrids the most logical approach was the use of 2n gametes. 63 BC1 LA (LA × AA or AA × LA) and 53 OA (AA × OA) progeny plants were obtained after unilateral sexual polyploidization. 16 genotypes from F2 LA populations were obtained after bilateral sexual polyploidization through sib-mating of F1 LA hybrids. GISH was employed for the identification of the parental genomes, mode of origin of these progenies and measurement of the introgression in different interspecific lily hybrids. Most of the BC1 progeny plants (LA and OA) had originated through 2n gametes by First Division Restitution (FDR) mechanism. However, there were 12 genotypes in LA hybrids and four genotypes in OA hybrids that originated through 2n gametes formation as the result of Indeterminate Meiotic Restitution (IMR). A higher amount of recombination was found in LA hybrids as compared to OA hybrids. Intergenomic recombination was also determined in the sib-mated F2 LA population. In this case both parents had contributed gametes with the somatic number of chromosomes (i.e., 2n-2n) thus confirming the event of bilateral sexual polyploidization in interspecific LA hybrids. Based on these results, the relevance of interspecific lily hybrids obtained from uni- and bilateral sexual polyploidization leading to allotriploid and allotetraploid formation in interspecific lily hybrids is discussed in the context of introgression and mapping (Chapter 4). Molecular markers are an important tool for the construction of genetic linkage maps, as the first step in the genetic dissection of the required traits leading to crop improvement followed by the marker assisted breeding in different plants. Lilium has one of the largest genome in plant kingdom and genetic mapping in lilies is constrained by its large genome. DArT (Diversity Array Technology), a molecular marker technique can detect and type DNA variation at several hundred genomic loci in parallel without relying on genome sequence information. The DArT technique was developed for Longiflorum × Asiatic (LA) lily hybrids to enable an efficient and effective genetic mapping with the production of a large numbers of markers in microarrays-based assay. The restriction enzyme PstI + TaqI combination generated the largest frequency of polymorphic genomic representations for a genotyping array. Genomic representations from 88 F1 LA plants were used to assemble a DArT genotyping microarray. A total of 687 DArT markers were developed and 382 polymorphic markers were mapped on 14 main linkage groups which is two more then the haploid chromosome number (i.e. n = 12). The resulting linkage map with 382 DArT markers spanned 1329 cM (3.5 cM/marker on average). The results highlighted the potential of DArT as a genetic technique for genome profiling in the context of molecular breeding and genomics, especially in crops with large genome sizes where other techniques proved to be less useful (Chapter 5). The results of the present investigation are of practical implication. These results show the advantages of the n gametes and their subsequent progenies which opened a new approach of lily breeding ‘the analytic breeding’ in the allopolyploids. It also shows the possibility of using certain triploid hybrids for further breeding. A comparison has been made between different types of interspecific crosses, the amount of intergenomic recombination and introgressions of chromosomal segments obtained after unilateral sexual polyploidization. Furthermore, bilateral sexually polyploidization via sib-mated F1 hybrids producing 2n gametes has been proven. The use of allotetraploids obtained from bilateral sexual polyploidization could be a novel approach in the breeding of LA-hybrids. These allotetraploid with recombinant chromosomal segment may be a potential source to generate genetic variation in subsequent progenies. The molecular cytogenetic GISH and FISH techniques proved to be a powerful tool useful for the construction of cytogenetic maps in interspecific crosses in crops with large genomes sizes like lily. These techniques are also used for the identification and integration of genetic maps with chromosome maps. FISH also helps to monitor the introgressed chromosome segment or marker of interest in the subsequent progenies. Application of the DArT technique proved to be an effective method to construct genetic linkage maps especially crops (like Lilium) with large genome sizes where other techniques might be less useful.

Lily, one of the economically most important ornamental crops, belongs to the genus Lilium of the family Liliaceae. There are about 80 species in Lilium which are categorized into seven sections, i.e., Lilium, Martagon, Pseudolirium, Archelirion, Sinomartagon, Leucolirion and Oxypetala. Usually, it is not so difficult to cross between the species within each section and the hybrids are fertile. However, it is very difficult to cross the species belonging to different sections. With cut style pollination followed by embryo rescue techniques, such distant interspecific crosses can be possible, but the hybrids are highly sterile. All modern lily cultivars have originated from hybridization among wild lily species. The three main lily cultivar groups, viz., Longiflorum, Asiatic and Oriental, originated from hybridization within one section, i.e., Leucolirion, Sinomartagon and Archelirion respectively. Up to now, about 150 Longiflorum, 4000 Asiatic cultivars, and 2000 Oriental cultivars have been registered. The genomes of Longiflorum (Leucolirion), Asiatic (Sinomartagon) and Oriental (Archelirion) are represented as L (Longiflorum), A (Asiatic) and O (Oriental) genome respectively. They possess quite different valuable traits, and one of the main goals of modern lily breedingareto combine the three distinctive groups in new cultivars.In this thesis, crosses among diploid Asiatic and Longiflorum cultivars, diploid F1 LA hybrids, triploid BC1 cultivars, allotetraploidand allopentaploid lilies were made. 11 diploid F1 LA hybrids and 19 triploids BC1 cultivars which were supplied by the Dutch lily breeding companies, and 23 new BC1, five BC2 and seven BC3 progenies were analyzed with conventional cytological methods, flow cytometry and genomic in situ hybridization.The configurations of metaphase I during meioses of the F1 LA hybrids are quantitatively variable, ranging from no bivalent to 12 bivalents in different pollen mother cells. This implies that LA hybrids have abnormal meiosis and normal meiosis, and indicates that LA hybrids have possibilities to produce aneuploid gametes, 2n gametes and n gametes (Chapter 2). Because the bivalents disjoin and the univalents divide simultaneously at anaphase I of the observed pollen mother cells, it is concluded that F1 LA hybrids have more potential to produce IMR 2n gametes than FDR 2n gametes I (Chapter 2). However, most of the BC1 progenies result from FDR 2n gametes and less from IMR 2n gametes. Probably, FDR 2n gametes have better viability than IMR 2n gametes because of chromosome and gene imbalance in the latter (Chapters 3 & 4).Besides the mode of 2n gamete formation, some crossover events, e.g., single, threestranddouble, four strand double, four strand triple crossover, etc, are clearly elucidated based on the GISH results from anaphase I of F1 LA hybrids (Chapter 2). The intergenomic recombinant chromosomes of triploid BC1 progenies mainly originate from single crossover (Chapters 3 & 4). The intergenomic recombinant chromosomes caused by other crossover events are confirmed in diploid BC1 progenies (Chapter 4).Based on GISH analysis of 19 BC1 cultivars from the Dutch lily breeding companies, 17 of the BC1 cultivars are eutriploid and two hypertriploid (Chapter 3). Nevertheless, among 45 new BC1 progenies, 10 of them are diploid, while the others are triploid (Chapter 4). This is the first reported finding that LA hybrid can produce functional haploid gametes. This finding might be valuable for lily introgression breeding. Only limited BC2 progenies, which originated from crosses between triploid BC1 cultivars and diploid Asiatic cultivars, were analyzed with GISH. They were predominantly diploid and they contained very few Longiflorum chromosomes or segments (Chapter 5).Allopentaploid lilies have relatively good male fertility as determined from their pollen germination. They were successfully crossed with Asiatic cultivars and Longiflorum cultivars. Most of the BC3 progenies were pseudoeuploids that possessed euploid chromosome numbers (2n=3x=36) but the parental genomes were aneuploid as a result of chromosome substitutions. Apart from this, one aneuploid (2n=3x+1=37) was also present. Both the pseudoeuploids and the aneuploids might contribute to genetic variation and are potentially useful for selection.Based on the results of lily interploid crosses (2x-3x, 2x-4x, 2x-5x and their reciprocals), diploid, triploid, tetraploid, and pentaploid lilies could be used as male parents when they had some degree of male fertility. Most diploid and triploid lilies could be used as female parents regardless of their male fertility. On the contrary, allotetraploid and allopentaploid lilies could hardly be used as female parents even though they had good fertility as estimated from pollen germination tests. The success and failure of interploid crosses in lilies depends on the viability of the gametes and the ploidy level of the secondary nucleus. From the process of tetrasporic eight-nucleate embryo sac formation in Lilium , it is derived that diploid, triploid, tetraploid, and pentaploid lilies produce tetraploid, hexaploid, octaploid and decaploid secondary nuclei in their embryo sacs respectively. Thus, It is suggested that tetraploid secondary nucleus might be ideal for lily endosperm development; hexaploid secondary nucleus are acceptable; but octaploid or higher secondary nuclei are not ideal (Chapter 5).The difference between 2x-4x and 4x-2x or 2x-5x and 5x-2x is obvious, because one was successful and the other not. A clear difference was also observed between 2x-3x and 3x-2x and between AA x LA and LA x AA. The difference is possibly caused by the frequency of functional gametes, ploidy level of endosperm and the interaction between embryo development and endosperm development (Chapter 5).2n gametes and n gametes might play different roles in lily breeding. The former is more useful for polyploid breeding; the latter may be valuable for introgression breeding. We could not over- or underestimate either of them, because the LA hybrids which could produce functional 2n gametes or n gametes are very limited. Especially, it is more difficult to find LA hybrids which produce functional n gametes. The advantages of 2n gametes over mitotic doubling in lily breeding have been well confirmed (Chapters 3 and 4). One of the further tasks is how to use the LA hybrids which produce viable n gametes in lily breeding.

Lily, one of the economically most important ornamental crops, belongs to the genus Lilium of the family Liliaceae. There are about 80 species in Lilium which are categorized into seven sections, i.e., Lilium, Martagon, Pseudolirium, Archelirion, Sinomartagon, Leucolirion and Oxypetala. Usually, it is not so difficult to cross between the species within each section and the hybrids are fertile. However, it is very difficult to cross the species belonging to different sections. With cut style pollination followed by embryo rescue techniques, such distant interspecific crosses can be possible, but the hybrids are highly sterile. All modern lily cultivars have originated from hybridization among wild lily species. The three main lily cultivar groups, viz., Longiflorum, Asiatic and Oriental, originated from hybridization within one section, i.e., Leucolirion, Sinomartagon and Archelirion respectively. Up to now, about 150 Longiflorum, 4000 Asiatic cultivars, and 2000 Oriental cultivars have been registered. The genomes of Longiflorum (Leucolirion), Asiatic (Sinomartagon) and Oriental (Archelirion) are represented as L (Longiflorum), A (Asiatic) and O (Oriental) genome respectively. They possess quite different valuable traits, and one of the main goals of modern lily breedingareto combine the three distinctive groups in new cultivars.In this thesis, crosses among diploid Asiatic and Longiflorum cultivars, diploid F1 LA hybrids, triploid BC1 cultivars, allotetraploidand allopentaploid lilies were made. 11 diploid F1 LA hybrids and 19 triploids BC1 cultivars which were supplied by the Dutch lily breeding companies, and 23 new BC1, five BC2 and seven BC3 progenies were analyzed with conventional cytological methods, flow cytometry and genomic in situ hybridization.The configurations of metaphase I during meioses of the F1 LA hybrids are quantitatively variable, ranging from no bivalent to 12 bivalents in different pollen mother cells. This implies that LA hybrids have abnormal meiosis and normal meiosis, and indicates that LA hybrids have possibilities to produce aneuploid gametes, 2n gametes and n gametes (Chapter 2). Because the bivalents disjoin and the univalents divide simultaneously at anaphase I of the observed pollen mother cells, it is concluded that F1 LA hybrids have more potential to produce IMR 2n gametes than FDR 2n gametes I (Chapter 2). However, most of the BC1 progenies result from FDR 2n gametes and less from IMR 2n gametes. Probably, FDR 2n gametes have better viability than IMR 2n gametes because of chromosome and gene imbalance in the latter (Chapters 3 & 4).Besides the mode of 2n gamete formation, some crossover events, e.g., single, threestranddouble, four strand double, four strand triple crossover, etc, are clearly elucidated based on the GISH results from anaphase I of F1 LA hybrids (Chapter 2). The intergenomic recombinant chromosomes of triploid BC1 progenies mainly originate from single crossover (Chapters 3 & 4). The intergenomic recombinant chromosomes caused by other crossover events are confirmed in diploid BC1 progenies (Chapter 4).Based on GISH analysis of 19 BC1 cultivars from the Dutch lily breeding companies, 17 of the BC1 cultivars are eutriploid and two hypertriploid (Chapter 3). Nevertheless, among 45 new BC1 progenies, 10 of them are diploid, while the others are triploid (Chapter 4). This is the first reported finding that LA hybrid can produce functional haploid gametes. This finding might be valuable for lily introgression breeding. Only limited BC2 progenies, which originated from crosses between triploid BC1 cultivars and diploid Asiatic cultivars, were analyzed with GISH. They were predominantly diploid and they contained very few Longiflorum chromosomes or segments (Chapter 5).Allopentaploid lilies have relatively good male fertility as determined from their pollen germination. They were successfully crossed with Asiatic cultivars and Longiflorum cultivars. Most of the BC3 progenies were pseudoeuploids that possessed euploid chromosome numbers (2n=3x=36) but the parental genomes were aneuploid as a result of chromosome substitutions. Apart from this, one aneuploid (2n=3x+1=37) was also present. Both the pseudoeuploids and the aneuploids might contribute to genetic variation and are potentially useful for selection.Based on the results of lily interploid crosses (2x-3x, 2x-4x, 2x-5x and their reciprocals), diploid, triploid, tetraploid, and pentaploid lilies could be used as male parents when they had some degree of male fertility. Most diploid and triploid lilies could be used as female parents regardless of their male fertility. On the contrary, allotetraploid and allopentaploid lilies could hardly be used as female parents even though they had good fertility as estimated from pollen germination tests. The success and failure of interploid crosses in lilies depends on the viability of the gametes and the ploidy level of the secondary nucleus. From the process of tetrasporic eight-nucleate embryo sac formation in Lilium , it is derived that diploid, triploid, tetraploid, and pentaploid lilies produce tetraploid, hexaploid, octaploid and decaploid secondary nuclei in their embryo sacs respectively. Thus, It is suggested that tetraploid secondary nucleus might be ideal for lily endosperm development; hexaploid secondary nucleus are acceptable; but octaploid or higher secondary nuclei are not ideal (Chapter 5).The difference between 2x-4x and 4x-2x or 2x-5x and 5x-2x is obvious, because one was successful and the other not. A clear difference was also observed between 2x-3x and 3x-2x and between AA x LA and LA x AA. The difference is possibly caused by the frequency of functional gametes, ploidy level of endosperm and the interaction between embryo development and endosperm development (Chapter 5).2n gametes and n gametes might play different roles in lily breeding. The former is more useful for polyploid breeding; the latter may be valuable for introgression breeding. We could not over- or underestimate either of them, because the LA hybrids which could produce functional 2n gametes or n gametes are very limited. Especially, it is more difficult to find LA hybrids which produce functional n gametes. The advantages of 2n gametes over mitotic doubling in lily breeding have been well confirmed (Chapters 3 and 4). One of the further tasks is how to use the LA hybrids which produce viable n gametes in lily breeding.

Besides being an important horticultural crop, lily ( Lilium ) also serves as an interesting model plant for molecularcytogeneticresearch for several reasons. a) The crop includes cultivars of different taxonomic species each of which possess valuable horticultural traits that need to be combined in the new cultivars. b) The genomes of different species are so well differentiated genetically that the individual chromosomes can be clearly identified in the F 1 hybrids as well as in the progenies through DNA in situ hybridization techniques. c) The chromosomes are so large that the number and position of homoeologous recombinant segments can be clearly detected. d) Through careful selection, or through nitrous oxide treatment, 2n gametes can be obtained in sterileinterspecifichybrids. Taking advantage of thesefavourableattributes of lily, a molecularcytogeneticinvestigation was conducted on the possibilities of introgression of characters between Oriental and Asiatic lilies through the use of 2n gametes.For this purpose, more than 700 genotypes of F 1interspecifichybrid between Oriental × Asiatic lilies were produced by using special techniques of ovule and/or embryo culture techniques. All the diploidinterspecificOA hybrids (2n=2x=24) were highly sterile but 12 genotypes were found to produce 2n gametes in notable frequencies. Six of these were analyzed through GISH forintergenomicrecombination as well as meiotic nuclear restitution. In all cases there was evidence for the occurrence of first division restitution (FDR) gametes through three different types of cytological abnormalities, viz., post metaphase I division (PMI); post metaphase II (PMII) and asymmetric cytokinesis (Chapter 2). The 12 genotypes of OA hybrid that produced 2n gametes were used as parents, both as male and female, in crosses with both parental species, i.e., AA and OO, as well as the allotetraploid4x-OA(derived from doubling the chromosome number of OA hybrids withoryzalin). From these crosses, 246 triploid and 14 tetraploid progenies were obtained. The chromosome constitution of some of theallotriploidBC1 progenies wasanalysedthrough GISH. This confirmed the presence of both O and A genome chromosomes with some of them possessing recombinant segments. Theserecombinationshad obviously occurred during the development of 2n gametes in the F 1 hybrid (Chapter 3). A selected population from the BC 1 progenies (38 plants) was analyzed by GISH and FISH techniques in order to identify chromosomes as well as to determine Intergenomic recombination. These analyses indicated that most of the progeny plants had originated through FDR. However, a small proportion of the progeny originated through IMR. Three kind of plants where identified when considering the presence of the number of homoeologous chromosomes and the recombinant segments: a) plants in which both reciprocal products from a crossover were present ( O / A , A / O , where O represents the centromere of the O genome and A, the recombinant segment of Asiatic chromosome and vice versa); (b) plants in which one normal chromosome of the O genome and a recombinant chromosome where present (O, A / O ) and (c) plants in which one normal chromosome of the A genome and a recombinant chromosome where present (A, O / A ) (Chapter 4). Furthermore, with the aim of utilizing allotriploids (2n=3x=36) for introgression breeding two kinds of crosses were made, a) diploid Asiatic cultivars were used as female parent and allotriploids AOA hybrids as male parents and vice versa (2x - 3x and 3x - 2x); b) allotriploids were crossed with allotetraploids (4x-OA) and other 2n gametes producers. In all these crosses two kinds of allotriploids were used, those that had originated through crosses of 4x-OA with diploid Asiatic cultivars and those that had originated from crosses using 2n gametes of OA hybrids with diploid Asiatic cultivars. Flowcytometry analyses were used for ploidy determination and these results were confirmed by chromosome counting in some cases showing concurrence. In the case of the 2x - 3x and 3x - 2x crosses, diploid and near-diploid progeny was obtained. In the 3x - 4x crosses near-tetraploid, tetraploid, near-pentaploid and pentaploid progenies were obtained. These results indicate that the allotriploids produced aneuploid as well as euploid (x, 2x,3x) pollen grains. Through GISH analysis the identification of the parental as well as recombinant chromosomes was possible. Those allotriploids which originated from 2n gametes transmitted their recombinant chromosomes to the progeny achieving introgression of the recombinant segments.(Chapter 5).Finally, through N2O treatments the formation of 2n gametes was induced in some of the genotypes of OA hybrids. Out of eight genotypes that were treated, six were totally sterile and two were known to produce 2n gametes in low frequencies. In the case of the totally sterile genotypes, three of them were able to produce progeny and in one of the known 2n gamete producers the production of 2n pollen was considerably increased from 0-30% to 55-80%. GISH analysis in the progeny revealedintergenomicrecombination between the parental genomes, indicating that FDR and IMR restitution mechanisms were responsible for the 2n pollen formation.(Chapter 6).These results show the advantages of the 2n gametes to generate genetic variation and the possibility to use them to produce progeny. It also shows the possibility of using triploid hybrids for further breeding and that introgression of chromosome segments can be achieved through chromosome assortment and recombination among the parental genomes. Furthermore, ploidy levels of the progeny can be manipulated depending upon the ploidy level of the parents. The induction of 2n gametes by N2O treatments in sterileinterspecificlily hybrids provides many possibilities to generate completely new hybrids. This investigation shows that by using sexual polyploidization and by monitoring the progenies trough FISH and GISH techniques, more accurate insights into the process of introgression can be obtained.

Besides being an important horticultural crop, lily ( Lilium ) also serves as an interesting model plant for molecularcytogeneticresearch for several reasons. a) The crop includes cultivars of different taxonomic species each of which possess valuable horticultural traits that need to be combined in the new cultivars. b) The genomes of different species are so well differentiated genetically that the individual chromosomes can be clearly identified in the F 1 hybrids as well as in the progenies through DNA in situ hybridization techniques. c) The chromosomes are so large that the number and position of homoeologous recombinant segments can be clearly detected. d) Through careful selection, or through nitrous oxide treatment, 2n gametes can be obtained in sterileinterspecifichybrids. Taking advantage of thesefavourableattributes of lily, a molecularcytogeneticinvestigation was conducted on the possibilities of introgression of characters between Oriental and Asiatic lilies through the use of 2n gametes.For this purpose, more than 700 genotypes of F 1interspecifichybrid between Oriental × Asiatic lilies were produced by using special techniques of ovule and/or embryo culture techniques. All the diploidinterspecificOA hybrids (2n=2x=24) were highly sterile but 12 genotypes were found to produce 2n gametes in notable frequencies. Six of these were analyzed through GISH forintergenomicrecombination as well as meiotic nuclear restitution. In all cases there was evidence for the occurrence of first division restitution (FDR) gametes through three different types of cytological abnormalities, viz., post metaphase I division (PMI); post metaphase II (PMII) and asymmetric cytokinesis (Chapter 2). The 12 genotypes of OA hybrid that produced 2n gametes were used as parents, both as male and female, in crosses with both parental species, i.e., AA and OO, as well as the allotetraploid4x-OA(derived from doubling the chromosome number of OA hybrids withoryzalin). From these crosses, 246 triploid and 14 tetraploid progenies were obtained. The chromosome constitution of some of theallotriploidBC1 progenies wasanalysedthrough GISH. This confirmed the presence of both O and A genome chromosomes with some of them possessing recombinant segments. Theserecombinationshad obviously occurred during the development of 2n gametes in the F 1 hybrid (Chapter 3). A selected population from the BC 1 progenies (38 plants) was analyzed by GISH and FISH techniques in order to identify chromosomes as well as to determine Intergenomic recombination. These analyses indicated that most of the progeny plants had originated through FDR. However, a small proportion of the progeny originated through IMR. Three kind of plants where identified when considering the presence of the number of homoeologous chromosomes and the recombinant segments: a) plants in which both reciprocal products from a crossover were present ( O / A , A / O , where O represents the centromere of the O genome and A, the recombinant segment of Asiatic chromosome and vice versa); (b) plants in which one normal chromosome of the O genome and a recombinant chromosome where present (O, A / O ) and (c) plants in which one normal chromosome of the A genome and a recombinant chromosome where present (A, O / A ) (Chapter 4). Furthermore, with the aim of utilizing allotriploids (2n=3x=36) for introgression breeding two kinds of crosses were made, a) diploid Asiatic cultivars were used as female parent and allotriploids AOA hybrids as male parents and vice versa (2x - 3x and 3x - 2x); b) allotriploids were crossed with allotetraploids (4x-OA) and other 2n gametes producers. In all these crosses two kinds of allotriploids were used, those that had originated through crosses of 4x-OA with diploid Asiatic cultivars and those that had originated from crosses using 2n gametes of OA hybrids with diploid Asiatic cultivars. Flowcytometry analyses were used for ploidy determination and these results were confirmed by chromosome counting in some cases showing concurrence. In the case of the 2x - 3x and 3x - 2x crosses, diploid and near-diploid progeny was obtained. In the 3x - 4x crosses near-tetraploid, tetraploid, near-pentaploid and pentaploid progenies were obtained. These results indicate that the allotriploids produced aneuploid as well as euploid (x, 2x,3x) pollen grains. Through GISH analysis the identification of the parental as well as recombinant chromosomes was possible. Those allotriploids which originated from 2n gametes transmitted their recombinant chromosomes to the progeny achieving introgression of the recombinant segments.(Chapter 5).Finally, through N2O treatments the formation of 2n gametes was induced in some of the genotypes of OA hybrids. Out of eight genotypes that were treated, six were totally sterile and two were known to produce 2n gametes in low frequencies. In the case of the totally sterile genotypes, three of them were able to produce progeny and in one of the known 2n gamete producers the production of 2n pollen was considerably increased from 0-30% to 55-80%. GISH analysis in the progeny revealedintergenomicrecombination between the parental genomes, indicating that FDR and IMR restitution mechanisms were responsible for the 2n pollen formation.(Chapter 6).These results show the advantages of the 2n gametes to generate genetic variation and the possibility to use them to produce progeny. It also shows the possibility of using triploid hybrids for further breeding and that introgression of chromosome segments can be achieved through chromosome assortment and recombination among the parental genomes. Furthermore, ploidy levels of the progeny can be manipulated depending upon the ploidy level of the parents. The induction of 2n gametes by N2O treatments in sterileinterspecificlily hybrids provides many possibilities to generate completely new hybrids. This investigation shows that by using sexual polyploidization and by monitoring the progenies trough FISH and GISH techniques, more accurate insights into the process of introgression can be obtained.

This thesis focused on the possibilities to improve greenhouse tomato (Lycopersicon esculentum cv. Moneymaker) in terms of yield. The domestic tomato has a very narrow genetic base which makes breeding for better performance a difficult task. The wild, crossable relatives of tomato present the possibility to introduce more variation in the gene pool of the cultivated tomato. An example of such an accession is Lycopersicon pennellii LA716. We used a classical approach for crop improvement while exploiting modern technological tools. The approach consists of introducing genetic variation by making crosses using L. pennellii LA716 as a male parent followed by recurrent backcrossing to greenhouse tomato and selecting for interesting traits in the hybrid progenies. The genetic variation emerging in the progenies of the interspecific cross was immortalized through the creation of a backcross inbred line library (BIL-population). A BIL-library is a group of genotypes with single introgressions which gives together a total coverage of the L. pennellii LA716 genome.The focus was not only on economic yield which by definition is the biomass allocated to the harvestable organs (fruits). But yield was also examined in a broader biological sense as total above ground biomass produced though out the period of growth. Further, we seek an understanding of growth, development and physiological processes which underlie yield. Relative growth rate (RGR) and its major physiological components; net assimilation rate (NAR) and leaf area ratio (LAR) were analyzed. The results of a growth, development, and physiological analysis conducted on old cultivars representing genetic variation in domesticated tomato cultivars fromEuropeshowed that genetic differences and the pattern of change in RGR during plant development are genotype specific. Correlation studies showed that genetic differences in RGR are directly affected by variations in NAR which is a measure for the efficiency of assimilation of photosynthates to biomass.Growth analyses enabled the identification of three BILs giving better growth in greenhouse tomato. BIL1.2, BIL2.3 and BIL8.2 showed an increase in RGR in the range of 5-3%, 15-4%, and 10-3% respectively. The faster RGR's observed in BIL1.2, BIL2.3 and BIL8.2 effected end point biomass production with respective increases of 16%, 22% and 3%. Additionally the results of BIL1.2 and BIL8.2 showed that the pattern of change (delta RGR) during plant development plays also a strong role in crop performance.Yield analysis showed that BIL1.2 and BIL2.3 produced 17-20% respectively 3% more fresh fruits.BILs 1.2, 2.3 and 8.2 produced about 16%, 22% and 3% more biomass (dry matter). Also, interesting trends in biomass allocation were observed, such as BIL1.2 which produced 20% more fresh fruits but only 16% more biomass. This was clearly different in BIL2.3 which produced 3% more fresh fruits but an amazing 22% more biomass. Finally BIL8.2 with 0.5% more fresh fruits produced 3% more biomass than Moneymaker. This shows that different BILs have different biomass allocation patterns and prompted us to execute a thorough analysis of biomass allocations to the above plant organs such as fruits, stem and leaves. It was seen that the biomass partitioning patterns were genotype specific and were controlled by either the presence of homozygous introgressions of L. pennellii LA716 or the absence of the replaced Moneymaker segment. The results suggested that BIL5.3 and BIL12.2 have a more efficient biomass allocation to the fruits. A tempting calculation shows that BIL2.3 in combination with the introgression of BIL5.3 or BIL 12.2 can give a potential 20-41% yield advantage.

This thesis focused on the possibilities to improve greenhouse tomato (Lycopersicon esculentum cv. Moneymaker) in terms of yield. The domestic tomato has a very narrow genetic base which makes breeding for better performance a difficult task. The wild, crossable relatives of tomato present the possibility to introduce more variation in the gene pool of the cultivated tomato. An example of such an accession is Lycopersicon pennellii LA716. We used a classical approach for crop improvement while exploiting modern technological tools. The approach consists of introducing genetic variation by making crosses using L. pennellii LA716 as a male parent followed by recurrent backcrossing to greenhouse tomato and selecting for interesting traits in the hybrid progenies. The genetic variation emerging in the progenies of the interspecific cross was immortalized through the creation of a backcross inbred line library (BIL-population). A BIL-library is a group of genotypes with single introgressions which gives together a total coverage of the L. pennellii LA716 genome.The focus was not only on economic yield which by definition is the biomass allocated to the harvestable organs (fruits). But yield was also examined in a broader biological sense as total above ground biomass produced though out the period of growth. Further, we seek an understanding of growth, development and physiological processes which underlie yield. Relative growth rate (RGR) and its major physiological components; net assimilation rate (NAR) and leaf area ratio (LAR) were analyzed. The results of a growth, development, and physiological analysis conducted on old cultivars representing genetic variation in domesticated tomato cultivars fromEuropeshowed that genetic differences and the pattern of change in RGR during plant development are genotype specific. Correlation studies showed that genetic differences in RGR are directly affected by variations in NAR which is a measure for the efficiency of assimilation of photosynthates to biomass.Growth analyses enabled the identification of three BILs giving better growth in greenhouse tomato. BIL1.2, BIL2.3 and BIL8.2 showed an increase in RGR in the range of 5-3%, 15-4%, and 10-3% respectively. The faster RGR's observed in BIL1.2, BIL2.3 and BIL8.2 effected end point biomass production with respective increases of 16%, 22% and 3%. Additionally the results of BIL1.2 and BIL8.2 showed that the pattern of change (delta RGR) during plant development plays also a strong role in crop performance.Yield analysis showed that BIL1.2 and BIL2.3 produced 17-20% respectively 3% more fresh fruits.BILs 1.2, 2.3 and 8.2 produced about 16%, 22% and 3% more biomass (dry matter). Also, interesting trends in biomass allocation were observed, such as BIL1.2 which produced 20% more fresh fruits but only 16% more biomass. This was clearly different in BIL2.3 which produced 3% more fresh fruits but an amazing 22% more biomass. Finally BIL8.2 with 0.5% more fresh fruits produced 3% more biomass than Moneymaker. This shows that different BILs have different biomass allocation patterns and prompted us to execute a thorough analysis of biomass allocations to the above plant organs such as fruits, stem and leaves. It was seen that the biomass partitioning patterns were genotype specific and were controlled by either the presence of homozygous introgressions of L. pennellii LA716 or the absence of the replaced Moneymaker segment. The results suggested that BIL5.3 and BIL12.2 have a more efficient biomass allocation to the fruits. A tempting calculation shows that BIL2.3 in combination with the introgression of BIL5.3 or BIL 12.2 can give a potential 20-41% yield advantage.

Keywords: Allium cepa , Allium roylei , Japanese bunching onion, introgression breeding, organic agriculture, linkage map, AFLP, QTL analysis, Arbuscular Mycorrhizal Fungi, in vitro multiplicationA massive amount of synthetic fertilizers is needed to grow onions ( Allium cepa L.) due to their meager and inefficient root system. While the sustainability of such high-input systems is being questioned, low-input systems, such as organic agriculture, are gaining ground. For organic agriculture, plants have to be good nutrient scavengers. Therefore, productivity and stability of onion production in organic systems can be problematic. Plant breeding can improve the efficiency of onion roots, but breeding relies on available variation and there was no information about that in onions or in Allium. The aim of this thesis was to search for variation in root morphology in onion and in its allied species A. fistulosum L., to understand the role this variation could play in organic agriculture and to perform a genetic analysis of root traits. The variation found in root traits in onion was limited, although old onion cultivars had a higher root length density than modern ones. Huge variation was observed betweenonionand A. fistulosum . A. fistulosum developed substantially more stem-borne and lateral roots and, consequently, a much denser root system. Experiments carried out in an organic farm revealed that onion explored a smaller volume of soil and had a lower root density than A. fistulosum . In addition, onion, contrary to A. fistulosum , showed a reduction in total and fine root density when cultivated in a soil with low nitrogen content. It was also demonstrated that A. fistulosum was very responsive to indigenous and inoculated AMF (50 to 60% increase in both shoot biomass and root length). A Quantitative Trait Locus (QTL) analysis was done on the genetic linkage map of the progeny from the cross A. cepa x ( A. roylei x A. fistulosum ) to locate some of the genes responsible for the better performance of the A. fistulosum root system. All traits were evaluated in a replicated trial using in vitro cloned plants. QTLs were found for number of bulbs (1) and stem-borne roots (2) and, more interesting for breeding, for the number of lateral roots (1) and for the relative root length of fine and thick roots (1). The results showed the feasibility of breeding for onions with improved root systems using the interspecific hybrid between A. roylei and A. fistulosum as a genetic source. Some perspectives on the use of cultivars carrying such roots traits in onion organic production are highlighted.

Keywords: Allium cepa , Allium roylei , Japanese bunching onion, introgression breeding, organic agriculture, linkage map, AFLP, QTL analysis, Arbuscular Mycorrhizal Fungi, in vitro multiplicationA massive amount of synthetic fertilizers is needed to grow onions ( Allium cepa L.) due to their meager and inefficient root system. While the sustainability of such high-input systems is being questioned, low-input systems, such as organic agriculture, are gaining ground. For organic agriculture, plants have to be good nutrient scavengers. Therefore, productivity and stability of onion production in organic systems can be problematic. Plant breeding can improve the efficiency of onion roots, but breeding relies on available variation and there was no information about that in onions or in Allium. The aim of this thesis was to search for variation in root morphology in onion and in its allied species A. fistulosum L., to understand the role this variation could play in organic agriculture and to perform a genetic analysis of root traits. The variation found in root traits in onion was limited, although old onion cultivars had a higher root length density than modern ones. Huge variation was observed betweenonionand A. fistulosum . A. fistulosum developed substantially more stem-borne and lateral roots and, consequently, a much denser root system. Experiments carried out in an organic farm revealed that onion explored a smaller volume of soil and had a lower root density than A. fistulosum . In addition, onion, contrary to A. fistulosum , showed a reduction in total and fine root density when cultivated in a soil with low nitrogen content. It was also demonstrated that A. fistulosum was very responsive to indigenous and inoculated AMF (50 to 60% increase in both shoot biomass and root length). A Quantitative Trait Locus (QTL) analysis was done on the genetic linkage map of the progeny from the cross A. cepa x ( A. roylei x A. fistulosum ) to locate some of the genes responsible for the better performance of the A. fistulosum root system. All traits were evaluated in a replicated trial using in vitro cloned plants. QTLs were found for number of bulbs (1) and stem-borne roots (2) and, more interesting for breeding, for the number of lateral roots (1) and for the relative root length of fine and thick roots (1). The results showed the feasibility of breeding for onions with improved root systems using the interspecific hybrid between A. roylei and A. fistulosum as a genetic source. Some perspectives on the use of cultivars carrying such roots traits in onion organic production are highlighted.

This report was assigned by COGEM. It is meant to uncover possible lacunas in scientific knowledge in the field of crop-wild relative hybridisation and consecutive introgression of crop genes into wild relatives, as well as its ecological and evolutionary consequences. The motivation for this enquiry comes from the impending release of GM crops; the assessment and the a posteriori monitoring of the possible effects of such releases should be based on the best available scientific knowledge. The report reviews the current data and identifies the main research priorities to be implemented for further progress in this field

This report was assigned by COGEM. It is meant to uncover possible lacunas in scientific knowledge in the field of crop-wild relative hybridisation and consecutive introgression of crop genes into wild relatives, as well as its ecological and evolutionary consequences. The motivation for this enquiry comes from the impending release of GM crops; the assessment and the a posteriori monitoring of the possible effects of such releases should be based on the best available scientific knowledge. The report reviews the current data and identifies the main research priorities to be implemented for further progress in this field

Molecular markers provide plant breeding with an important and valuable new source of information. Linkage between molecular markers can be translated to genetic linkage maps, which have become an important tool in plant and animal genetics. Linkage between (quantitative) trait-data and occurrences of marker alleles allow identification of important genetic factors, underlying observable traits. Knowledge that results from such analyses, i.e. the location on the genome of important genetic factors (quantitative trait loci or QTLs), can and should be applied when making selection and breeding decisions.Selection of parents is an important issue in plant breeding. Basing selection on QTL information, i.e. applying marker-assisted selection, can result in an increased selection efficiency. This is especially true for quantitative traits with a low heritability. For efficient application of marker-assisted selection reliable and fairly complete QTL-mapping results are required. When QTLs were mapped for several traits a multiple trait-selection can be devised, through the use of a suitable index. In this case an ideal target genotype, containing favourable alleles for QTLs that affect the traits of interest, can be constructed and crosses can be made between selected parents in such a way that the probability of obtaining the target genotype is maximised. Although this approach looks promising, and simulation results show an improved selection performance, several problems remain which are limiting application in practise. A more reliable and complete mapping of QTLs, including mapping of interaction between QTLs, mapping of QTLs with a higher reliability, for instance resulting from a combined mapping of several traits, and mapping of QTLs in more diverse non-mapping types of populations could greatly contribute to an increased application of marker-assisted selection, and hence a more efficient selection in plant breeding.Although it is common practise to resort to unadapted material when searching for new genetic variation, the undesired characteristics that accompany the genes coding for the target trait of interest, limit the applicability of introducing 'foreign' genes. With the help of marker and QTL-analysis the genome region that harbours genes which are responsible for the desired characteristics can be identified more precisely and thus the size of the fragment that needs to be introgressed can remain restricted. Marker-assisted backcrossing allows a much more controlled method of gene introgression, limiting the amount of 'linkage-drag' and requiring less generations of backcrossing than conventional backcrossing for yielding suitable genotypes.Developments that favour application of marker-assisted selection are still progressing at a high rate. New technical enhancements in the field of molecular biology, new protocols and methods for identification of genetic factors, new versatile software for data analysis and visualisation all contribute to new ways of selection and breeding that take advantage of this newly acquired knowledge and information. These novel methods should be used to continue to create genetic improvement, in a faster or more efficient way than before, and to introduce quality enhancing genetic factors into cultivated crops.

Molecular markers provide plant breeding with an important and valuable new source of information. Linkage between molecular markers can be translated to genetic linkage maps, which have become an important tool in plant and animal genetics. Linkage between (quantitative) trait-data and occurrences of marker alleles allow identification of important genetic factors, underlying observable traits. Knowledge that results from such analyses, i.e. the location on the genome of important genetic factors (quantitative trait loci or QTLs), can and should be applied when making selection and breeding decisions.Selection of parents is an important issue in plant breeding. Basing selection on QTL information, i.e. applying marker-assisted selection, can result in an increased selection efficiency. This is especially true for quantitative traits with a low heritability. For efficient application of marker-assisted selection reliable and fairly complete QTL-mapping results are required. When QTLs were mapped for several traits a multiple trait-selection can be devised, through the use of a suitable index. In this case an ideal target genotype, containing favourable alleles for QTLs that affect the traits of interest, can be constructed and crosses can be made between selected parents in such a way that the probability of obtaining the target genotype is maximised. Although this approach looks promising, and simulation results show an improved selection performance, several problems remain which are limiting application in practise. A more reliable and complete mapping of QTLs, including mapping of interaction between QTLs, mapping of QTLs with a higher reliability, for instance resulting from a combined mapping of several traits, and mapping of QTLs in more diverse non-mapping types of populations could greatly contribute to an increased application of marker-assisted selection, and hence a more efficient selection in plant breeding.Although it is common practise to resort to unadapted material when searching for new genetic variation, the undesired characteristics that accompany the genes coding for the target trait of interest, limit the applicability of introducing 'foreign' genes. With the help of marker and QTL-analysis the genome region that harbours genes which are responsible for the desired characteristics can be identified more precisely and thus the size of the fragment that needs to be introgressed can remain restricted. Marker-assisted backcrossing allows a much more controlled method of gene introgression, limiting the amount of 'linkage-drag' and requiring less generations of backcrossing than conventional backcrossing for yielding suitable genotypes.Developments that favour application of marker-assisted selection are still progressing at a high rate. New technical enhancements in the field of molecular biology, new protocols and methods for identification of genetic factors, new versatile software for data analysis and visualisation all contribute to new ways of selection and breeding that take advantage of this newly acquired knowledge and information. These novel methods should be used to continue to create genetic improvement, in a faster or more efficient way than before, and to introduce quality enhancing genetic factors into cultivated crops.

Trypanosomosis is the most widespread parasitic disease of great importance in Africa affecting human and animals. There are breeds of cattle that are trypanotolerant. Trypanotolerance is an ability of these breeds to withstand the effects of trypanosome infections. The disease impedes production in Sub-Saharan Africa. In this thesis, the main focus is on marker-assisted breeding to improve trypanotolerance and mouse is used as model organism for cattle. We have concentrated on the opportunities to exploit individual genes (QTL) that affect trypanotolerance. We have shown how the size of a breeding experiment increases with the number and the size of the QTL region to introgress. Biological factors to be considered when estimating the required size of an introgression experiment are also highlighted. Strategies to optimize designs for introgression show that selecting two lines each carrying two of the donor's QTL allele during the backcross (when introgressing three QTL) is more advantageous than monitoring simultaneously the three QTL in terms of number of animals to maintain, number of genotyping and costs involved. An experiment in mice shows that QTL introgression is feasible and successful, and that the background genotype is an important factor to be considered when analyzing the results of such an experiment. This experiment is unique. The general discussion focuses on the introgression of trypanotolerance genes in cattle with an emphasis on SubSaharan zones of Africa. The introduction of trypanotolerant cattle in the humid and subhumid zones of Africa, however, should be done with care to avoid damage to the environment and bio-diversity.

This thesis describes the results of a molecular cytogenetic investigation of the process of introgression in Alstroemeria . The aim of this study was to transfer chromosomes or genes from one Alstroemeria species into another. For this, two distantly related species, A. aurea and A. inodora , were hybridized and the resulting hybrids were further backcrossed with the species A. inodora . To monitor the process of introgression accurately it was necessary to identify the individual chromosomes of both parental species.In addition to classical karyotyping, reliable criteria for unequivocal identification of all individual somatic chromosomes of the species A. aurea and A. inodora have been developed by applying fluorescence in situ hybridization (FISH) with tandemly repeated DNA probes (Chapter 2). Four probes, two species specific, A001-I (for A. aurea ) and D32-13 (for A. inodora ), and two ribosomal rDNA repeats, pTa71 (18S-25S) and pTa794 (5S), revealed specific banding patterns per chromosome, enabling identification of all individual chromosomes. In contrast to most plant species, both Alstroemeria species contained a large number of rDNA sites (per haploid genome: 20 in A. aurea and 16 in A. inodora ). Remarkably, the physical location of the rDNA sites varied greatly between the two species; only a few were found at similar positions.Besides identification of somatic chromosomes, FISH with the species specific probes was used for the identification of meiotic chromosomes to accurately determine the extent of homoeologous association in the diploid parental hybrid. Multicolour FISH experiments revealed a large number of bivalents per Metaphase I cell (average 6.7 bivalents per MI) of the hybrid A. aurea x A. inodora , which indicated that homoeologous recombination occurs frequently in the hybrid.Successful backcrossing of the highly sterile distant hybrid A. aurea x A. inodora with its parent A. inodora resulted in a small number of first generation backcross plants (BC1). To establish whether and how frequently homoeologous recombination occurred during meiosis in the hybrid, the BC1 plants were analysed with genomic in situ hybridization (GISH). The presence of recombinant chromosomes in all plants provided evidence for homoeologous recombination. All plants were analysed using a combination of FISH with the repetitive probes and GISH that allowed identification of all chromosomes and the construction of detailed karyotypes. In addition, identification of the recombinant chromosomes revealed which of the chromosomes of A. aurea and A. inodora were actually homoeologous.The recombinant chromosomes showed a large number of recombination points, indicating that multiple crossovers had occurred. The distribution of the recombination points indicated that chiasmata are formed along the entire length of the chromosomes. Furthermore, the analysis of recombinant chromosomes revealed structural differences (one translocation and one inversion) between two sets of homoeologous chromosomes. These differences were confirmed by analysis of metaphase I configurations of these chromosomes in the hybrid and the BC1 plants (Chapters 4 and 5).The detailed analysis of the genome constitution of the plants revealed two different types of plants, near diploids (2n=2x+1=17) and triploids (2n=3x=24), indicating that two different types of gametes were functional in the hybrid, viz., n- and 2n-gametes. The GISH and FISH analysis provided evidence that 2n-gametes were functional both from male and female parents and had originated in all cases through a mechanism genetically leading to first division restitution (FDR) (Chapter 4).In order to investigate the fate of the introgressed A. aurea segments in A. inodora , meiosis was analysed using GISH and FISH in the triploid BC1 plants. These plants contained two more or less complete genomes of A. inodora and one of A. aurea as a result of the FDR mechanism (Chapter 4). Metaphase I configurations were studied per set of three homo(eo)logous chromosomes. This analysis revealed, as expected, preferential pairing of the A. inodora chromosomes, whereas the A. aurea chromosomes usually were unpaired. Only low frequencies of homoeologous chromosome pairing were observed, which is in contrast to what was found in the parental hybrid. Interestingly, the recombinant chromosomes generally formed trivalents with their two homo(eo)logues. These chromosomes were only associated at homologous (recombinant) chromosome segments. At later stages of meiosis, normal distribution of the A. inodora chromosomes and the recombinant chromosomes into the gametes was observed, whereas the non-recombinant A. aurea chromosomes were randomly distributed among the gametes, pointing towards the perspective for obtaining addition lines or introgression of alien chromosome segments in Alstroemeria .For breeding, the most important observation, which was made in this molecular cytogenetic study of Alstroemeria , was the high frequency of pairing and recombination between homoeologous chromosomes in the interspecific hybrid and the almost absence of this phenomenon in the further backcross generations. In these backcrosses, preferential pairing between homologous chromosomes was observed. For efficient introgression breeding more emphasis should be given to the production of sufficient BC1 plants and the detection of desired recombinants in this generation.

This thesis describes the results of a molecular cytogenetic investigation of the process of introgression in Alstroemeria . The aim of this study was to transfer chromosomes or genes from one Alstroemeria species into another. For this, two distantly related species, A. aurea and A. inodora , were hybridized and the resulting hybrids were further backcrossed with the species A. inodora . To monitor the process of introgression accurately it was necessary to identify the individual chromosomes of both parental species.In addition to classical karyotyping, reliable criteria for unequivocal identification of all individual somatic chromosomes of the species A. aurea and A. inodora have been developed by applying fluorescence in situ hybridization (FISH) with tandemly repeated DNA probes (Chapter 2). Four probes, two species specific, A001-I (for A. aurea ) and D32-13 (for A. inodora ), and two ribosomal rDNA repeats, pTa71 (18S-25S) and pTa794 (5S), revealed specific banding patterns per chromosome, enabling identification of all individual chromosomes. In contrast to most plant species, both Alstroemeria species contained a large number of rDNA sites (per haploid genome: 20 in A. aurea and 16 in A. inodora ). Remarkably, the physical location of the rDNA sites varied greatly between the two species; only a few were found at similar positions.Besides identification of somatic chromosomes, FISH with the species specific probes was used for the identification of meiotic chromosomes to accurately determine the extent of homoeologous association in the diploid parental hybrid. Multicolour FISH experiments revealed a large number of bivalents per Metaphase I cell (average 6.7 bivalents per MI) of the hybrid A. aurea x A. inodora , which indicated that homoeologous recombination occurs frequently in the hybrid.Successful backcrossing of the highly sterile distant hybrid A. aurea x A. inodora with its parent A. inodora resulted in a small number of first generation backcross plants (BC1). To establish whether and how frequently homoeologous recombination occurred during meiosis in the hybrid, the BC1 plants were analysed with genomic in situ hybridization (GISH). The presence of recombinant chromosomes in all plants provided evidence for homoeologous recombination. All plants were analysed using a combination of FISH with the repetitive probes and GISH that allowed identification of all chromosomes and the construction of detailed karyotypes. In addition, identification of the recombinant chromosomes revealed which of the chromosomes of A. aurea and A. inodora were actually homoeologous.The recombinant chromosomes showed a large number of recombination points, indicating that multiple crossovers had occurred. The distribution of the recombination points indicated that chiasmata are formed along the entire length of the chromosomes. Furthermore, the analysis of recombinant chromosomes revealed structural differences (one translocation and one inversion) between two sets of homoeologous chromosomes. These differences were confirmed by analysis of metaphase I configurations of these chromosomes in the hybrid and the BC1 plants (Chapters 4 and 5).The detailed analysis of the genome constitution of the plants revealed two different types of plants, near diploids (2n=2x+1=17) and triploids (2n=3x=24), indicating that two different types of gametes were functional in the hybrid, viz., n- and 2n-gametes. The GISH and FISH analysis provided evidence that 2n-gametes were functional both from male and female parents and had originated in all cases through a mechanism genetically leading to first division restitution (FDR) (Chapter 4).In order to investigate the fate of the introgressed A. aurea segments in A. inodora , meiosis was analysed using GISH and FISH in the triploid BC1 plants. These plants contained two more or less complete genomes of A. inodora and one of A. aurea as a result of the FDR mechanism (Chapter 4). Metaphase I configurations were studied per set of three homo(eo)logous chromosomes. This analysis revealed, as expected, preferential pairing of the A. inodora chromosomes, whereas the A. aurea chromosomes usually were unpaired. Only low frequencies of homoeologous chromosome pairing were observed, which is in contrast to what was found in the parental hybrid. Interestingly, the recombinant chromosomes generally formed trivalents with their two homo(eo)logues. These chromosomes were only associated at homologous (recombinant) chromosome segments. At later stages of meiosis, normal distribution of the A. inodora chromosomes and the recombinant chromosomes into the gametes was observed, whereas the non-recombinant A. aurea chromosomes were randomly distributed among the gametes, pointing towards the perspective for obtaining addition lines or introgression of alien chromosome segments in Alstroemeria .For breeding, the most important observation, which was made in this molecular cytogenetic study of Alstroemeria , was the high frequency of pairing and recombination between homoeologous chromosomes in the interspecific hybrid and the almost absence of this phenomenon in the further backcross generations. In these backcrosses, preferential pairing between homologous chromosomes was observed. For efficient introgression breeding more emphasis should be given to the production of sufficient BC1 plants and the detection of desired recombinants in this generation.