Catherine Breton

The use of molecular markers for germplasm management in a French olive collection.

Abstract.With more than 100 accessions, the CBNMP olive collection includes a major part of the French germplasm. We used molecular markers to characterise all accessions and to study genetic relationships between cultivars. Firstly, 497 olive trees were genotyped using 32 RAPD markers. We identified 114 RAPD profiles and detected several cases of mislabelling, synonymy and homonymy. Secondly, for each RAPD profile, one tree was analysed using mtDNA RFLPs to determine the cytoplasmic lineage of each cultivar and using five nuclear SSR loci. French germplasm displayed ME1, MOM and MCK mitotypes with ME1 prevailing (84%). Based on SSR markers, we revealed a slight differentiation between French cultivars growing in the West and the East side of the Rhône Valley. This study allowed us to construct a molecular data-base for the reference collection and to analyse genetic diversity for further prospecting, and for introducing new olive accessions.

Oleaster (var. sylvestris) and subsp. cuspidata are suitable genetic resources for improvement of the olive (Olea europaea subsp. europaea var. europaea)

The olive belongs to Olea, a complex genus and to a species with six subspecies. Subsp. europaea includes both the cultivated olive and the oleaster, the wild ancestor of the olive. Little is known on the phenotype of fruits from subsp. cuspidata. We aimed to compare europaea and cuspidata trees for oil content and composition and to verify whether natural hybrids may exist between the two subspecies. Specimens were from Kenya and putative natural hybrids between cuspidata and europaea were from Stellenbosch (South Africa). Cultivar and oleaster trees were sampled in France (continental and Corsica), Italy (Continental and Sardinia) and Tunisia. We have examined the fruit (drupe) structure between subsp. cuspidata and europaea and extracted oil from the pulp, seed and total drupe. Comparison for oil content was made between the cultivated olive and the oleaster and some cuspidata trees from Kenya. A few of cuspidata individuals from Botanical gardens do not enable phenotyping for oil content and composition. Oil composition analyses were performed for the main fatty acids to compare the three taxa. We used microsatellite markers at 11 loci to compute genetic distances between cuspidata, oleaster and cultivar trees and to reveal eventual hybrids. The SSR polymorphisms were huge between the two subspecies and they revealed that putative hybrids were true hybrids sampled around olive orchards. The whole comparison of oil content shows that olive cultivars display higher oil content than the subsp. cuspidata and that the oleaster trees are intermediate, whereas for oil composition of the drupe, cuspidata from Kenya shows less oleic acid than europaea. However, the cuspidata trees show seed oil composition similarities with the other two taxa. The discussion deals with possible reasons to explain the differences and of the possible uses of these taxa for breeding both the olive and brown olive.

New hypothesis elucidates self-incompatibility in the olive tree regarding S-alleles dominance relationships as in the sporophytic model

Most olive varieties are not strictly self-incompatible, nevertheless, they request foreign pollen to enhance fruit yield, and consequently orchards should contain pollinisers to ensure fruit set of the main variety. The best way to choose pollinisers is to experiment numerous crosses in a diallel design. Here, the genetic mode of inheritance of SI in the olive is deciphered and it does not correspond to the GSI type, but to the SSI type. It leaves S-allele dominance relationship expression in the male (pollen and pollen tube), but not in the female (stigma and style). Thus, a pair-wise combination of varieties may be inter-compatible in one direction (male to female, or female to male) and inter-incompatible in the other direction. Dominance relationships also explain different levels of self-pollination observed in varieties. Little efficient pollinisers were found and predicted in varieties; nevertheless, some new efficient pair-wise allele combinations are predicted and could be created. This model enables one to forecast compatibility without waiting for several years of yield records and to choose pollinisers in silico.

A model based on S-allele dominance relationships to explain pseudo self-fertility of varieties in the olive tree

Self-fertility is largely decreased and even prevented by various mechanisms because, broadly, it causes inbreeding depression, although some species have retained self-reproduction regimes. Species of plants that display the self-incompatible sporophytic type of self-incompatibility may rarely self-pollinate. It is only possible in the absence of foreign compatible pollen. In the olive tree with a sporophytic mechanism, we will show that three co-dominant S-alleles R1, R3 and R5 do not lead to the same level of self-fertility. All varieties that carry R1 are less self-fertile than those that carry R5, whatever the other S-alleles, while those carrying R3 are intermediate. S-allele pair-wise combinations that differ by two or three levels of dominance, and not the other combinations allow self-fertility, and moreover each S-allele R1, R3 and R5 decreases, maintains and enhances the self-fertility rate, respectively.

Are olive cultivars distinguishable from oleaster trees based on morphology of drupes and pits, oil composition and microsatellite polymorphisms?

Abstract The study aims to determine which traits may help to differentiate cultivars and oleaster trees based on drupe and pit morphological traits, oil content and composition and nuclear DNA markers at seven SSR loci. We determined for one cultivar (Gerboui) the steady drupe, pit morphological and oil composition variation ranges in six different contrasted agro-systems. We then investigated 33 cultivars and 40 oleaster trees. PCA analysis on morphological traits was examined. Gas chromatography was used to determine fatty acid composition of 20 cultivars and 13 oleaster trees, and the Soxhlet method was used to determine fruit oil content. SSR genotyping was performed in poly-acry- lamide gels. Five clusters based on UPGMA mixed olive cultivars and oleaster trees suggesting kinship relationships between some cultivars and some oleasters; and thus cultivars could derive from domestication of oleaster trees.

Potential of combining morphometry and ancient DNA information to investigate grapevine domestication

The goal of this work was to explore the possibility (1) of carrying out both morphogeometric and archaeological DNA analyses on the same grape pips and (2) of comparing different molecular markers to reveal DNA variation, namely Simple Sequence Repeats (SSRs) and Single Nucleotide Polymorphisms (SNPs). We focused on waterlogged seeds originating from three Roman and one medieval archaeological sites in France. Our first results indicate that taking photographs of pips is not detrimental to the preservation of DNA, provided a specific protocol is respected. Regarding the genetic markers, obtaining reliable information in sufficient quantity proved very difficult using SSRs. SNPs have a much more interesting potential, providing greater success rates and reliability. Here in four archaeological pips we studied 842 SNPs, derived from known polymorphisms in several genes, including one gene related to sex. Phylogenies built using these genetic markers indicate that three pips from the Roman site of Gasquinoy are close to modern wild grapevines and/or the female group, while the only medieval pip from Colletière is hermaphrodite and close to the modern cultivated group. Morphogeometrical results are in agreement with these findings. We conclude that the combined use of SNP markers and morphogeometry is promising for deciphering the intricate history of grapevine domestication.