RNAi-Untersuchungen und Überexpression von Genen der Rosmarinsäurebiosynthese

Rosmarinic acid biosynthesis has been elucidated in cell cultures of Coleus blumei (Lamiaceae). The involved enzymes are known and have been characterised and some of the corresponding genes have been cloned. Among the enzymes that have been heterologously expressed are hydroxyphenylpyruvate reductase (HPPR) and rosmarinic acid synthase (RAS), which might be specific for rosmarinic acid biosynthesis. For the evaluation of those enzymes’ function and their possible regulatory mechanisms RNAi down-regulation experiments have been performed. The RNAi effect on the rosmarinic acid content has been measured. The effect of overexpression of both genes on rosmarinic acid accumulation was also evaluated. The pHANNIBAL vector in combination with the binary vector pART27 was used for the cloning of the RNAi and overexpression constructs. The RNAi and overexpression constructs were integrated into the genome of Coleus blumei with help of Agrobacterium rhizogenes resulting in hairy root liquid cultures. Since there were different Agrobacterium rhizogenes strains and different Coleus blumei sterile plant cultures available, first test transformations were performed to find the best Coleus blumei/Agrobacterium rhizogenes combination. For the test transformations agrobacteria without additional vector DNA and agrobacteria transformed with the HPPR RNAi construct were used. The best combination was the plant line ColV1 with the Agrobacterium rhizogenes strain LBA15834. They generated the most viable hairy roots and showed an acceptable rosmarinic acid content of 1.5 % in the dry weight. Additionally those hairy roots demonstrated the most obvious RNAi effects. This plant/agrobacterium combination has been used for further transformation of Coleus blumei with RNAi and overexpression constructs of HPPR and RAS and the controls. As controls there was plant material transformed with the acrobacteria carrying the empty binary vector pART27 and untransformed agrobacteria. The obtained hairy root liquid cultures were analysed by molecular biological methods for the absence of agrobacteria, the presence of the inserted construct, the presence of the agrobacterial genes rolA, rolB, rolC and the transcript levels of HPPR and RAS. Southern blots were executed to determine the number of T-DNA insertions of the constructs. The 35S promoter of the pHANNIBAL vector and the nptII gene of the pART27 vector were used as probes. Several transformation events were observed in all the hairy root lines. The hairy root cultures were characterised to determine the best harvest time for the measurements of enzyme activities and rosmarinic acid content. Day twelve of the cultivation period was the best option. The hairy roots were analysed with respect to their rosmarinic acid content, the enzyme activities of tyrosine aminotransferase (TAT), phenylalanine ammonia-lyase (PAL), HPPR and RAS. The RNAi and overexpression effects were most pronounced in the rosmarinic acid contents. The RNAi lines showed a reduced and the HPPR overexpression lines an elevated content. The enzyme activities in protein extracts of the RNAi lines showed no reduction. The HPPR overexpression lines had an increased HPPR activity. RAS overexpression lines exhibited co-suppression and showed similar results as the controls or RAS RNAi lines. The hairy root lines have been analysed for the accumulation of intermediates of rosmarinic acid biosynthesis. Only caffeic acid could be detected but showed no correlation to the reduced or elevated rosmarinic acid content. L-Tyrosine and L-phenylalanine are the precursors for rosmarinic acid biosynthesis. Control lines and overexpression lines of HPPR and RAS were supplemented with these two amino acids. There was no change in the rosmarinic acid content compared to non-supplemented controls. The amino acids are not limiting for rosmarinic acid accumulation and the biosynthesis is not induced by the oversupply of these aromatic amino acids. HPPR of Coleus blumei is closely related to the cytosolic hydroxypyruvate reductase (HPR2) of Arabidopsis thaliana and was available as heterologously enzyme. The enzyme was characterised and the obtained data was compared with the data of the heterologously expressed HPPR and the native HPPR of Coleus blumei. The most obvious difference was the Km value for the rosmarinic acid biosynthesis substrate 4-hydroxyphenylpyruvate. The native HPPR of Coleus blumei and the HPR2 of Arabidopsis thaliana displayed a higher affinity for this substrate than the heterologously expressed HPPR of Coleus blumei. A genomic sequence of HPPR from Coleus blumei could be amplified with full-length primers. The sequence and position of an intron could be identified. The number of genomic copies of the HPPR could not be determined exactly by Southern blotting but more than one copy were observed.