Sunflower improvement by conventional breeding is severely restricted by the availability of a rather limited gene pool owing to natural incompatibilities, even between related species, and by the time scale of most breeding programs. Therefore, much attention has been directed recently to the newly emerging and novel technologies of plant cell and molecular biology that provide a powerful means to supplement and complement the traditional methods of plant improvement. The concept of DNA-based markers has revolutionized our ability to directly access any part of the plant genome, and has led to new opportunities such as map-based cloning and directed plant breeding. Efficient regeneration of fertile plants from cultured cells and protoplats, combined with novel methods of DNA delivery and selection of transformed cells, has resulted in the production of transgenic sunflower. Agronomically useful genes, which confer resistance to insect pests and pathogens, have been introduced. However, further molecular improvement of sunflower could be limited most by the lack of our knowledge about, and access to, important and useful genes (e.g., those controlling multigenic traits like yield, and resistance to biotic and abi otic stresses). Therefore, high priority should be given to the development of ultra high density (UHD) linkage maps and the development of new tools for high-throughput genome and expression analyses.

Molecular characterization of 28 CMS sources and the fertile cytoplasm of H. annuus in sunflower demonstrated that cytoplasms different by their origin show a considerable similarity that cannot be expected from their pedigree. Cluster analysis using the UPGMA method allowed differentiating 10 mitochondrial (mt) types based on RFLP data. Characteristic proteins specific for one or more CMS sources could be identified which correspond to the mt types. In addition, the data obtained on fertility restoration for a subset of nine new CMS sources support the classification of the cytoplasms made on mtDNA level as demonstrated in the UPGMA dendrogram. A map-based cloning approach is followed for the isolation of the restorer gene Rf1, which restores pollen fertility in hybrids based on PET1. Markers closely linked to the restorer gene could be identified using AFLP and RAPD technologies. A bacterial artificial chromosome (BAC) library was constructed using the sunflower restorer line RHA325 and pBeloBAC11 as vector. The first sunflower BAC library corresponds to 1.7 haploid genome equivalents (104,736 clones, average insert size approx. 50 kb). Use of STS markers, which had been developed from RAPD markers, allowed to identify three BAC clones by colony hybridization against high density filters. DNA fingerprinting using HindIII allowed to build a contig for the restorer locus Rf1.

In sunflower, commercial hybrid breeding is based on a single CMS inducing cytoplasm, the so-called PET1 cytoplasm. The introduction of one dominant, nuclear-encoded restorer gene (Rf1) is in most cases sufficient for fertility restoration. Little has been learned so far about the mode of action of the restorer gene Rf1. For map-based cloning of the restorer gene Rf1, an F2 population of the cross RHA325 (cms) x HA342 has been used. The χ²-test confirmed segregation for one dominant gene which corresponds to Rf1. For the AFLP analyses 256 EcoRI/MseI primer combinations have been used so far. In addition, RAPD analyses were performed using 1,200 decamer primers. Twenty-three primers had polymorphic amplification products, differentiating the bulks, and could therefore be mapped. The hybridization of the marker HP4 against a BAC library resulted in three positive clones. The overlapping end of the smallest clone was used to get a new hybridization against the BAC library.

Sunflower is a plant of the American continent, but now many wild species are spread in different part of the world. Along the coast of the Inhambane Bay, Mozambique, two sunflower species, H. argophyllus and H. debilis, both of Texas origin, grow far apart. However, in two sites on sandy soil they grow together and many plants have been recognized as hybrids, sharing morphological traits typical of the two species. Some putative hybrid plants produced seed from which a new generation was raised at University of Udine. We present data coming from morphological and AFLP analyses of plants grown in the spring 2000. With morphological traits typical of the two species we have build up a classical hybrid index by adding the scores of each species. Clear evidence of introgressed plants results also by looking at the unequality of the AFLP bands present in H. debilis and H. argophyllus and shared in the introgressed material. A partial pollen grain staining confirms the presence of reduced fertility of some morphologically introgressed plants. From a breeding point of view, these wild species and the introgressed material seems to have great potential value for their specificity. The AFLP band pattern from H. annuus material appears to be quite unique and a cluster analysis shows a clear diversity of the three species. Deeper investigation is required in order to assess the extent of these data and to establish whether plants with intermedi ate traits can represent a new taxonomic unit.

A total of 230 hybrids have been introduced in a sunflower breeding program as initial material and submitted to selection for the following traits: earliness, size of head, position of head related to the stem, stay green, tolerance to Sclerotinia sclerotiorum, tolerance to Phomopsis sp., CMS-Rf ability. Some interspecific hybrids were promoted through advanced generations of selfing, such as H. annuus x STR, ARG x H. annuus, RIG x H. annuus, (ANN x DEB) x H. annuus, the others were tested for CMS and restoration ability by crossing and offspring evaluation. The new CMS source ANT-1 is the most promising source for a new CMS-pollen fertility restoration system in sunflower hybrid production. Interspecific hybrids are important as donors and crosses for introgressing new favorable alleles to parent inbred lines.

We started to collect wild sunflower (Helianthus L.) species in 1998. The main task of this collection is to promote the development of new breeding material. In particular, it was planned to use the wild species as resistance sources to the major diseases of sunflower and as Rf genes sources to the new CMS types. The collection consists presently of 140 accessions of annual wild species and 110 samples of 27 perennial ones. The assessment of the wild sun flowers collection for resistance to Sclerotinia (Sclerotinia sclerotiorum), Phomopsis (Phomopsis helianthi) and broomrape (Orobanche cumana) was carried out on artificial infectious backgrounds. Resistance to other diseases was estimated simultaneously under natural infection. The immunologic potential of the collection was assessed three times during the vegetative period: at early stages of ontogeny (first week of June), at full flowering (second week of August) and at the end of the vegetation period (fourth week of September). A group of 101 samples showed various types of resistance to a complex of diseases. F1 and F2 generations were produced from more than 100 cross combinations of sunflower inbred lines and wild annual species. The tolerance to Phomopsis in field conditions was recorded in the F1 and F2 interspecific hybrids. Also, several sources of fertility restoration for new CMS types (ANN1, ARG1, PEF1, GIG1 and RIG1) have been found among the accessions of wild annual sunflower species. Thus, certain samples from the collection may be useful for breeding to disease resistance. The development of new initial mate rial for sunflower breeding has already started.

The trend in healthier human diets is to decrease the consumption of the saturated fatty acids. Sunflower oil, which is fourth in production among edible vegetable oils in the world, contains 65 g kg-1 palmitic and 45 g kg-1 stearic acids, both saturated fatty acids. These levels are high compared with rapeseed (Brassica napus L.) oil with 40 g kg-1 palmitic and 20 g kg-1 stearic acids. A reduction of saturated fats in traditional sunflower oil would lead to a healthier edible oil. The objective of this preliminary study was to search the vast genetic diversity available from wild ancestors of cultivated sunflower for potential sources of reduced palmitic and stearic fatty acids. Achene oil of one popula tion of wild H. giganteus L. (GIG-102) from INRA, Montpellier, France, had 47 g kg-1 palmitic acid and 18 g kg-1 stearic acid. The combined 65 g kg-1 palmitic and stearic acids is 40% lower than the present level of these fats in sunflower oil. The level of saturated fatty acids observed in the population remained low when plants were grown in the greenhouse under uniform conditions. In the greenhouse, palmitic acid averaged 48 g kg-1, while stearic acid averaged 16 g kg-1. This would indicate that palmitic and stearic acid concentrations are under genetic control with potential for incorporation into cultivated sunflower. Crossing this population with an inbred cultivated line produced F1 plants with an achene oil that averaged 39 g kg-1 palmitic and 26 g kg-1 stearic acid. The inbred cultivated parent averaged 55 g kg-1 palmitic and 51 g kg-1 stearic acid. F2 plants produced an achene oil that averaged 47 g kg-1 palmitic and 29 g kg-1 stearic acid, for a total of 76 g kg-1. When F1 plants were backcrossed to the cultivated inbred, BC1F1 plants produced an achene oil that averaged 47 g kg-1 palmitic and 28 g kg-1 stearic acid for a total of 75 g kg-1. The inbred cultivated parent averaged 54 g kg-1 palmitic and 56 g kg-1 stearic acid, for a total of 110 g kg-1. Preliminary information indicate that palmitic and stearic fatty acids in sunflower oil can be reduced by introducing genes from a wild peren nial progenitor into cultivated sunflower. Further research will be needed to determine the inheritance of these fatty acids. Other agronomic traits will also have to be monitored during the introgression of genes for reduced saturated fatty acids into cultivated sunflower.

Pervenets is a sunflower mutant having a seed oil oleic acid content greater than 65%. All lines derived from the mutant Pervenets population display seed oil with a high oleic acid content [HOAC], over 80%. Linoleic (LO) and [HOAC] genotypes carry specific RFLP markers, oleLOR and oleHOS, revealed by oleate-desaturase cDNA used as a probe. Linkage disequilibria between the OleHOS allele and Pervenets mutation, and oleLOR and [LO] were without any exception. We studied the inheritance of [HOAC] in F2 progenies from a cross [LO] x [HOAC]. This study has shown one dominant allele for [HOAC] that cosegregated with the oleate-desaturase oleHOS, [HOAC]-specific RFLP. Variance analyses revealed almost complete dominance and that oleHOS marker explained 86% of the OAC variation. F6 recombinant inbred lines segregating for [HOAC] / [LO] showed that all [HOAC] RI lines carried oleHOS, but half of the RI lines carrying oleHOS were [LO]. Conversely, the absence of [HOAC] RI lines carrying oleLOR is evidence against recombination having taken place between the oleHOS locus and the Pervenets mutation locus. The [HOAC] trait is therefore due to 2 independent loci: the locus carrying the oleHOS allele and another locus. One allele, supole, at this second locus may suppress the effect of the oleHOS allele on the [HOAC] trait. Preliminary results with SSR (170 microsatellites, 500 cM) indicated that the oleHL and the supole loci are independent and explained 47.14 and 32.2% of the OAC variation, respectively. The existence of the Pervenets mutation, supole and epistatic combinations for modifier alleles leading to [HOAC] trait has never been suggested. It may explain the discordant results found in our segregating populations and literature.

This work aims to clone the genomic region of the sunflower genome that may carry the Pervenets mutation causing high oleic content of seed oil. We previously reported that the oleHOS allele at the oleate-desaturase locus (oleHL) carries or is genetically closely linked to Pervenets mutation. We pointed out that the fragments that border the 5.8 kb EcoRI fragments carry the Pervenets mutation. The oleHOS reference physical map was constructed. It displays a common part made of the 5.8 kb EcoRI fragment, also present in oleLOR allele, and a specific region with the 7.9 kb EcoRI fragment, also carrying oleate-desaturase sequences. A genomic library was constructed in λfixII with an average insert size of 15 kb. Two millions clones were screened enabling to isolate clones from Group I, which were entirely sequenced and revealed carrying a gene for an oleate probably located in the RE. This corresponds to the invariant part of the oleHL locus. Two clones of Group II (11.1, 27.1) are overlapping but 11.1 is characterized by new specific restriction fragments and instability leading to a smear when probed with the oleate cDNA. Consequently, 11.1 sequence probably carries oleate repeated sequences that cause instability of the clone. The clone 11.1 is a good candidate to carry a part of the specific oleHOS allele, but due to its organization in repeats it is not yet sequenced. This organization leads to speculate on the mechanisms that could disturb oleate function in Pervenets [HOAC] genotypes.

A diallel crossing design was carried out during 1999 and 2000 seasons near Montpellier under field nursery conditions. Intercrosses of seven sunflower lines were performed between genetic (ms2) or cytoplasmic male sterile (PET1) and standard versions using bagged heads. These lines contrasted for oleic acid content in oil. Four lines with high oleic acid content originated from Pervenets. Three linoleic inbred lines were added to the crossing design. Male fertile lines were selfed in the same conditions. F1 seeds and those from selfed inbred lines were analyzed for fatty acid composition using gas chromatography. Year effect, reciprocal or maternal effects, general combining ability and specific combining ability effects were found to be significant. The level of dom inance of high oleic acid content appeared related to the genetic background. We found previously that 83HR4, a linoleic line of the diallel, was bearing a suppressor allele canceling Pervenets mutation effect. In reciprocal crosses with high oleic lines, this line had a strong maternal negative effect on oleic acid content.

In recent years, consumers have become concerned with reducing the saturated fat content of their diet. Studies have indicated that high levels of saturated fat consumption are correlated with increased risk of coronary heart disease. The total saturated fat content of oil from current sunflower hybrids averages about 130 g kg-1. To identify sunflower germplasm with reduced saturated fatty acid composition, a total of 884 cultivated sunflower accessions from the USDA-ARS North Central Regional Plant Introduction Station, Ames, Iowa, were screened for fatty acid composition by gas chromatography. PI 250542, a cultivar collected in Egypt by Paul Knowles and deposited into the National Plant Germplasm System in 1958, was identified as an accession with reduced saturated fatty acid content. The fatty acid composition of 26 halfseeds of PI 250542 was determined, and the seeds with lowest saturated fatty acids were grown in the greenhouse. Pollen from a single plant was used to pol linate NMS HA 89, and the F1 seed was grown in the field and self-pollinated. After three generations of selection by half-seed analysis, two lines with a low saturated fatty acid trait were selected. Line RS1 has a striped black and dark gray achene, whereas line RS2 has a light gray achene which often bleaches to white when grown in the field. The total saturated fatty acid composition of RS1 including C16 to C24 fatty acids was 77 g kg-1, and for RS2 was 76 g kg-1 when grown at Fargo, North Dakota, in 2000. To determine inheritance of the reduced saturated fatty acid trait, RS1 and RS2 were pollinated by HA 821, a high saturated fatty acid line, and grown in the greenhouse. The resulting F1 seeds were slightly higher in saturated fatty acids than the RS1 or RS2 parents, but far lower than the HA 821 parent, suggesting that the reduced saturated fatty acid trait was partially dominant.

The 49 North American wild Helianthus species have long survived extreme environments and possess resistance or tolerance genes to salt, drought, insects, diseases, as well as cytoplasmic male-sterility and fertility restoration, and are valuable genetic resources for sunflower improvement. Gene transfer from wild species into cultivated background depends on the success of interspecific hybridization, F1 fertility, chromosome pairing for genetic recombination, efficient screening methods, and sufficient progenies for selection. Most wild annual species x sunflower crosses produce F1 seeds and can be backcrossed easily. For the more difficult perennial x sunflower crosses where hybrid seed set is rare, a two-step embryo culture technique has been established for rescuing immature interspecific embryos prior to abortion. The chromosome homology between genomes of wild species and cultivated sunflower is high, and the meiotic chromosome pairing of wild diploid x cultivated F1 is reasonably good, except for the multivalent formation, bridges, and fragments due to translocation and inversion differences. Chromosomal doubling by colchicine treatment substantially increases the F1 fertility, improves backcrossing success, and leads to the production of amphiploids. The amphiploids have restored fertility and can be maintained by sib-pollina tion, and will serve as a bridge for gene transfer. Using this approach, we suc cessfully transferred genes resistant to the new broomrape race F, which attacks all the cultivated sunflower in Spain. Inheritance studies suggest a sin gle dominant gene provides resistance.

Broomrape (Orobanche cumana Wallr.) populations belonging to the new race F in Spain have overcome all known resistance genes, Or1 to Or5 , in cultivated sunflower (Helianthus annuus L.) and are spreading rapidly. Resistance to race F of this parasitic weed has been found in wild perennial species of Helianthus and has been introgressed into cultivated sunflower. The objective of this study was to characterize the inheritance of resistance genes in cultivated sunflower derived from wild perennial species H. divaricatus and H. grosseserratus, respectively. Crosses between resistant cultivated lines and the susceptible line P21 were made, and the F1's were resistant when evaluated for broomrape resistance using a highly virulent population of race F, indicating dominance of resistance genes. Comparison of resistance of the segregating populations, F2 and BC1F1, to both parents confirmed the dominance observed in the F1 and indicated that resistance is under the control of a single dominant gene. This dominance of resistance genes will greatly simplify the breeding for resistance.

Three sunflower inbred lines differing in resistance to Sclerotinia were grown on nutritive medium supplemented with different concentrations of oxalic acid. After two weeks of culture, plant height, fresh and dry weight of above-ground part, length of first pair of leaves, root length, and fresh and dry weight of root were measured. Data obtained were analysed using ANOVA and LSD test. Out of the tested parameters, plant height and root length were found to be the most reliable indicators of plant resistance/susceptibility to Sclerotinia. Based on the data obtained, oxalic acid concentrations of 4 mM, 3 mM and 2 mM were chosen for further work in which test would be done on a larger number of genotypes.

Phoma macdonaldii, the causal agent of the sunflower black stem disease, is responsible for qualitative and quantitative damage which can result in up to 60% yield losses in France and worldwide. An early inoculation method has been developed in the laboratory to test sunflower genotype tolerance to Phoma. It was thus important to check whether the method was relevant by assessing the possible incidence of the growth stage of the plant on its reaction to Phoma. Six sunflower genotypes, which were known to be more or less susceptible to Phoma, were inoculated with an aggressive isolate at four different growth stages: cotyledon, unfolded leaves 3-6, budding and early flowering stage. Data analysis showed there is no interaction between the growth stage and the genotype: the test developed discriminates between genotypes early and the grading is representative of what occurs at a more advanced stage as it remains unchanged. Taking into account these data and the large genetic variability of Phoma macdonaldii, the strain-genotype interaction was investigated on isolates originating from France and other countries, in the search for possible pathotypes. For this purpose, ten isolates of Phoma, which were consid ered to be representative of the fungus variability, with differing aggressiveness, geographical origins and types of symptoms, were tested on ten different sunflower lines. Preliminary results suggest the existence of five pathotypes. Plant breeders could thus use these pathotypes for testing rapidly and reliably the tolerance of their genetic material to Phoma.

A severe dwarf mutant affecting vegetative and reproductive growth arose spontaneously in our sunflower-breeding nursery in 1999. The segregation data did not deviate significantly from the expected ratio for a recessive monogenic character. Periodic treatment with gibberellic acid (GA3) was effective to revert to the wild type phenotype and internode elongation was directly related to the GA3 concentration. The results suggest that the mutant gene is involved in the enzymatic pathway of GA synthesis.

Sclerotinia sclerotiorum (Lib.) de Bary is one of the major diseases of sunflower (Helianthus annuus). The pathogen can attack all parts of the plant at every stage of plant growth, predominantly the capitulum, leaf and stem. In the present study, several perennial Helianthus species of diverse origin were evaluated for resistance level to mid-stem and leaf infection using an inoculation method. The evaluation revealed considerable and significant differences among the genotypes in all recorded resistance traits. Three genotypes showed enhanced resistance levels to mid-stem reaction. Two genotypes exhibit resistances to leaf infection. Since all evaluated resistant genotypes in these trials are sexually incompatible with the Helianthus annuus genome, newly developed biotechnological methods gain particular importance for the transfer of the discovered levels of resistance into the cultivated sunflower.

In the framework of a bilateral project Italy – Romania (University of Udine and IC), an experiment was established at the University of Udine Experiment Station in order to: i) evaluate the genetic variability of the tested genotypes for Sclerotinia basal stem and head infection; ii) to test two new alternative screening methods against white head rot; iii) to investigate possible relationships among the different tests; iiii) to find resistance-associated fragments by AFLP analysis. The following tests were used: 1) basal stem infection, 2) ascospore head infection, 3) mycelium culture filtrate injection and 4) oxalic acid injection into the back face of the head. The segregation of responses between resistant and susceptible controls displayed the suitability of mycelium and ascospore tests adopted and the independence of the two tolerance mechanisms. Conversely, tests 3 and 4 showed a poor discrimination of the controls and a complete lack of relationships with mycelium and ascospore infections. Sunflower reaction against S. sclerotiorum may not be restricted to resistance to oxalic acid, but may involve other physiological mechanisms. The inbred line 28R (coming from H. argophyllus wild species) was not only most tolerant against both, basal stem and white head rot infections, but also it gave the best performance in oxalate and culture filtrate tests highlight ing a specific resistance to oxalate.

Plant breeders are always interested in variability and new genetic sources. In the past, the sources have been limited to species able to be sexually crossed. Biotechnological methods applying cell biology, genetic engineering and tissue culture techniques now provide almost unlimited strategies to create additional breeding resources. In the past 10 years, we spent much effort using several biotechnological methods to achieve sunflower plants with a superior resistance to Sclerotinia sclerotiorum, which is the most important fungal pathogen of this crop worldwide. Symmetric and asymmetric somatic hybridizations were the most effective and viable alternatives to overcome the severe Sclerotinia infections in this crop. These results have been confirmed in preliminary field experiments of the hybrids.