PUB - Publikationen an der Universität Bielefeld

MicroRNAs (miRNAs) are small noncoding RNAs whose function as modulators of gene expression is crucial for many aspects of plant and animal development. A major challenge in understanding the regulatory role of miRNAs is to accurately predict regulated targets. In this work, 281 novel miRNA targets in Arabidopsis were predicted employing the program RNAhybrid with additional assumptions based on already validated miRNA:target interactions. Comparing gene ontology (GO) annotation of both previously predicted/validated targets and novel predicted targets found in this work with the GO categorization for the whole genome revealed that, contrary to previously predicted/validated miRNA targets, there is no over-represented protein class among the novel predicted targets. Some GO annotation classes that were over-represented (e.g. transcription factors) or under-represented, now show distributions close to their representation in the whole genome. Nine putative miRNA targets were subjected to experimental validation, five of them were validated, including MYB101, MYB125, MRG1 and ACS8, which are targets of miR159, and GAE1, which is a target of miR161. The validation of four candidate targets failed. Among the novel validated miRNA targets, two were further analyzed: MYB101 and MRG1. Overexpression of MYB101 containing silent mutations in the miR159 binding site (MYB101mutBS) resulted in accumulation of MYB101 in tissues where the transcript is normally absent. The overexpression of wild-type MYB101 did not show this effect. Adult plants overexpressing MYB101mutBS were smaller than wild-type, whereas MYB101 overexpressors showed no difference to wild-type plants. Contrasting with the MYB101 transcript levels that are highest in pollen, the expression pattern of MYB101 analyzed by promoter-GUS lines revealed that the MYB101 promoter is active in seedlings (cotyledons, leaves and roots) and flowers, again showing a strong signal in pollen. These findings confirm the regulatory role of miR159 for proper MYB101 expression. MRG1 is found only in Arabidopsis and contains no conserved protein motif. The expression pattern of MRG1 analyzed by promoter-GUS lines revealed that the MRG1 promoter is active in many different tissues whereas the MRG1 transcript can be detected at very low levels only. The overexpression of MRG1 was only effective when silent mutations in the miR159 binding site had been introduced. In MRG1mutBS overexpressing plants several defects in leaf morphology were observed and the number of leaves was altered drastically. Nevertheless, plants overexpressing wild-type MYB101 showed similar, but weaker phenotypes. MRG1 protein, expressed as fusion protein with GFP, was localized in the nucleus of BY-2 protoplasts. Therefore, MRG1 may represent a novel regulator that affects leaf development, and miR159 controls the precise expression of MRG1. The expression patterns of MIR159A, MIR159B and MIR161 were analyzed by promoter-GUS lines. Although the promoters of both MIR159 genes show an overlapping expression pattern, promoter-GUS lines confirmed previous indications that MIR159A is the gene responsible for the majority of mature miR159 accumulation. Deletion analysis of the MIR159A promoter identified regions that have regulatory properties. The promoter activity of MIR161 confirmed that miR161 is a broadly expressed miRNA. This conclusion is also supported by the analysis of serial deletions of the MIR161 promoter. Even the smallest promoter fragment conferred high activity of the reporter protein. In addition, two regulatory regions where found within the miR161 promoter. The regulatory regions found within these two MIRNA promoters can now be used to identify proteins that drive the expression of these genes.