We found that 490 genes of promoter regions bound by Arx in all three experiments contained one or several motifs with at least 75% similarity to Arx-binding motif. This result was significantly higher than the number of motifs obtained in a set of control sequences. To Lapatinib structure validate these results in a more physiological situation, we decided to do similar experiments from E15.5 mouse embryonic brains, which correspond to an important time for neuronal migration and differentiation and which shows a high expression of Arx. We thus performed 3 independent experiments, hybridizing Arx-associated chromatin fragments to the same promoter microarrays. In total, 369 genes were found consistently enriched in Arx-immunoprecipitated material . Our results revealed that out of these 369 genes, 290 were common to those identified in transfected N2a cells . Then, using EMBOSS Profit as previously, we inspected the sequences identified in embryonic brain for both Arx-binding motifs . Surprisingly, out of 369 genes, only 74 were found to have at least one Arx-binding site as previously defined . Among these 74 genes, 65 genes were identified in both Arx-transfected N2a cells and mouse embryonic brain . Although we tried to identify new or degenerated motifs in these negative sequences, we were not able to find a motif that was significantly more present in Arx-bound sequences by comparison to control sequences. To determine the validity of our ChIP-chip results, we randomly selected 21 candidate Arx-bound genes displaying representative degrees of enrichment based on P-values in the list of 1006 genes obtained in total in ChIP experiments . We performed quantitative QFM-PCR on Arximmunoprecipitated material from transfected N2a cells and embryonic brain and compared the enrichment of these genes with total input DNA. Binding was confirmed for 19/21 genes in both transfected N2a cells and embryonic brain . In contrast, there was no enrichment of any of these genes in control immunoprecipitates . Similarly, Arx did not bind to Vapb, a negative control . These results confirmed ChIP-chip findings for Pten, which was only identified in N2a cells on microarrays and was also found negative in embryonic brain by ChIP-PCR. Similarly, Jph4 which was hardly positive in N2a cells by ChIP-chip was only confirmed in brain by ChIP-PCR . However, we observed that we were able to validate in both Arx-transfected N2a cells and E15.5 embryonic brain some genes, such as Sh3tc2, Lmo3, Epha3, Cdh2, Calb2 that were negative in ChIP experiments performed from embryonic brains, suggesting a higher sensitivy of the quantitative ChIP-PCR method compared to ChIP-chip, at least in embryonic brains.