Microarray designed to find PIN polarity regulators downstream of TIR1/AFB

Roots of 5-day-old 35S::WRKY23-GR were treated for 6 hours with 10 μM DEX or DMSO, respectively. Wt Col-0 and 35S::WRKY23-SRDX plants were treated with 10 μM NAA or DMSO for 6 hours. Roots were subsequently collected for RNA isolation. All replicates were sampled in three independent experiments. Total RNA (200 μg per array) was hybridized to ATH1 Affymetrix Arabidopsis arrays in accordance with the standard procedures at the VIB Nucleomics Core. Data files containing probe level intensities (.cel files) were used for background correction and normalization was done in R (http://www.r-project.org) using the Bioconductor package affylmGUI (http://bioinf.wehi.edu.au/affylmGUI/) using the log2 scale RMA procedure. Genes with the same or contrasting WRKY23 expression profiles were selected by Pavlidis template matching in TMeV 4.0 (TIGR) (28). Finally, genes with a significant p value (< 0.001), denoting expression above background with minimum 2-fold change compared to the respective control, were retained for further analysis. First, by comparing 35S::WRKY23-GR roots with and without induction by dexamethasone, we obtained a set of 110 genes as potential WRKY23 targets, which were up-regulated in dexamethasone-treated seedlings. Next, we identified 950 genes, which were auxin-inducible in Col-0 wild type (Wt), but lost this auxin responsiveness in dominant-negative 35S::WRKY23-SRDX roots. The overlap between the two datasets yielded a list of 61 genes (Fig. S1B). Because WRKY23 acts downstream of the SLR/IAA14 transcriptional repressor, this genelist was compared with previously published microarray data on auxin-treated solitary root1 (slr-1) mutant seedlings