Development of a consistent protocol for qPCR aanlysis in azalea flowers

Azalea flower colour is mainly determined by differences in anthocyanins and flavonols. Genes coding for key enzymes in the biosynthesis of these pigments were isolated and the cDNA sequences were characterised. The expression of these genes in the petals of azalea was used for the development of a suitable real-time RT-PCR protocol for gene expression analysis in azalea. Optimization was needed at all crucial steps from RNA isolation up to the final quantification. Disruption of the petal tissue with a Retsch Tissuelyser (Qiagen) prior to RNA isolation proved to be of similar quality compared to the use of a mortar and a pestle. For azalea, being a not well characterised species, almost no sequence information is available in databases. For the studied flower colour genes, only cDNA fragments were on hand, so the development of intron-exon boundary spanning primers was not achievable. DNase treatment of mRNA samples appeared to be a good alternative for preventing the co-amplification of contaminating DNA, as shown by the low or non-existing amplification in the noRT-samples. There exist many different quantification strategies, but, although most labour-intensive, the use of standard curves remains the most reliable method. However, reproducibility and stability of these dilution series was a major problem. This problem was circumvented by linearization of the used plasmids and by diluting them in a yeast tRNA solution. The most critical step for the whole quantification process was the selection of the right housekeeping genes. In previous analysis, only GAPDH was used for this purpose, but the use of at least two housekeeping genes is recommended. Starting from the ‘Flamenco’ cDNA library, 200 cDNA fragments were randomly picked and sequenced. The putative functions of these fragments were determined by comparison of the sequences with EMBL accessions, 60 potential azalea genes could be identified. Twelve of these genes were good candidate housekeeping genes (coding for enzymes active in photosynthesis, cell cycle, …) and primers suited for real-time PCR were developed. One primer pair gave no amplification; another one resulted in aspecific signal. The remaining ten genes were tested together with GAPDH in real-time PCR and by using the geNorm software (Vandesompele et al., 2002) we could state that the use of three housekeeping genes was sufficient for expression analysis in the petals of azalea. Finally, the expression level of the flower colour genes was determined on an LC480 (Roche). No significant relation was found between the flower colour and the expression profiles of the genes. Nevertheless, our aim of developing a reproducible protocol for qPCR analysis in azalea was achieved, since the expression levels of the genes of interest were extremely comparable both in the technical and the biological replicates.