Saponins are a large group of plant secondary metabolites including triterpenoids and steroids glycosylated with one or more sugar chains (Hostettmann and Marston, 1995). In the Medicago genus saponins are a complex mixture of triterpenic glycosides showing a broad spectrum of biological (antifungal, insecticidal, phytotoxic, allelopathic and hemolytic) and pharmacological properties (anticholesterolemic, anti-cancer adjuvant) (Tava and Avato, 2006). In spite of their role in plant defence mechanisms, their importance as antimicrobial compounds and their possible benefits for human health, knowledge of the genetic control of saponin biosynthesis is still lagging behind. Triterpenic saponins are derived from 61538;-amyrin that is converted to various aglycones by a series of oxidative reactions catalyzed by cytochromes P450; finally glycosyltransferases (GTs) convert aglycones into saponins. In M. truncatula at least three genes encoding early enzymes of the 61538;-amyrin skeleton formation have been functionally characterized (Suzuki et al., 2002). However, in M. truncatula no gene involved in the oxidative reactions of 61538;-amyrin has yet been isolated. In our laboratory, we have used M. truncatula mutant collections to search mutations affecting saponin biosynthesis. In an activation-tagging mutant collection of M. truncatula (Porceddu et al., 2008), one plant lacking hemolytic saponins was evidenced (lha-1 mutant); the T-DNA tagged locus was identified by inverse-PCR and the putative gene implied in this mutation turned out to be a cytochrome P450 named CYP716A12. A second collection, obtained by EMS-mutagenesis (Porceddu et al., 2008), was screened by TILLING analysis and two mutants for the CYP716A12 gene were identified (lha-2 and lha-3). Both these mutants lacked hemolytic saponins, although they accumulated soyasaponins. In addition, all lha mutants lacked sapogenin precursors of hemolytic saponins suggesting that the biosynthetic pathway was blocked at an early and common step. In order to validate the correspondence between the CYP716A12 gene and the lha phenotype, a vector carrying a full-length CDS of CYP716A12 was stably transformed into lha-1 mutant plants and restoration of the biosynthetic pathway of hemolytic saponins was confirmed by GC-MS analyses. This result confirmed the knock-out of CYP716A12 was responsible for the block in hemolytic sapogenin pathway. Objectives of this Ph.D project were: i) investigating the role of cytochrome CYP716A12 in saponin biosynthesis by heterologous expression in yeast (Saccharomyces cerevisiae), ii) studying gene expression levels and sapogenin content in M. truncatula plants at different developmental stages and iii) identifying the orthologue of CYP716A12 in M. sativa. To elucidate the function of CYP716A12, the gene was expressed in yeast (S. cerevisiae); first, the CYP716A12 coding sequence was introduced in a commercial yeast strain. In vivo and in vitro microsome assays were performed using 61538;-amyrin and oleanolic acid as possible substrates of the cytochrome, but no modifications of the supplemented substrates were evidenced. These negative results may be explained by the low efficiency of yeast NADPH-P450 reductase (Pompon et al., 1996) that has been suggested to be important for efficient electron transfer to cytochromes P450. For this reason a new experiment was performed using yeast strains WR and WAT11, overexpressing S. cerevisiae and Arabidopsis thaliana P450 reductase respectively. An in vivo assay was performed, but GC analysis did not show any additional peak except those of the substrates. Enzymatic activities in microsomes (in vitro assay) were then tested by supplying 61538;-amyrin, and erythrodiol as substrates; GC-MS analysis revealed the substrate transformation to yield oleanolic acid. These results indicate that CYP716A12 mainly catalyzes the sequential three-step oxidation at the C-28 position necessary to transform 61538;-amyrin into oleanolic acid. To identify which substrate could restore the saponin pathway in lha-1 mutant, plant microsomal fractions were used. Using 61538;-amyrin and oleanolic acid as substrate, wild-type microsomes produced all sapogenins, while mutant microsomes yielded non-hemolytic sapogenins only. In contrast, when supplemented with hederagenin, neither wild-type nor lha mutant microsomes restored the saponin pathway. These contrasting results may reflect the difficulty in setting up an efficient and stable plant microsome preparation. To investigate the expression profiles of the CYP716A12 gene, quantitative PCR analyses were performed on a wild-type line. Leaf stem and root samples were collected at three biological stages (vegetative growth, early flowering, and early pod setting). Roots displayed highest and most stable expression across stages, while in leaves a significant increase in expression appeared in the reproductive stages with a maximum at flowering. The same plants were examined for sapogenin content by GC/FID analysis. Hemolytic sapogenin content was significantly influenced by both phenological stages and plant organs. Finally, in order to transfer the new findings in hemolytic saponin biosynthesis in crop species, the ORF of the orthologous gene in M. sativa was isolated using the same primers utilized in M. truncatula. In summary, results of this Ph.D work revealed that CYP716A12 is a multifunctional oxidase catalyzing the sequential three-step oxidation at C-28 position necessary to transform 61538;-amyrin into oleanolic acid. CYP716A12 can use different substrates and plays a key role in hemolytic saponin biosynthesis as its disruption prevents the formation of any aglycone with an oxygenated group in C-28 and consequently the presence of triterpenes glycosylated in this position. In contrast to other transcripts involved in triterpenic saponin pathway, generally restricted to root system, CYP716A12 is expressed in different plant organs and developmental stages. Building upon this work, we plan to use the WAT11 yeast expression system to investigate the role of CYP716A12 cytochrome from M. sativa, and perform TILLING analysis on other candidate P450 genes in order to identify and characterize new functions possibly involved in saponin biosynthesis.