BindDB: Oct4 Case Study
Description
Epigenomic Profiling of Oct4 Promoter Due to the flexibility of the BindDB platform, more general questions related to epigenetic regulation in the ESC context can be asked. For example, how are the alternative promoters of the stem cell specific transcription factor OCT4 regulated? In both mESCs and hESCs, two main isoforms of Oct4 exist (Oct4A, Oct4B), which differ by approximately 3kb in their 5' regions and have the potential to be regulated by two distinct promoter regions (Wang and Dai, 2010). Interestingly, Oct4A and Oct4B display different temporal and spatial expression patterns (Cauffman et al., 2006). It has been hypothesized that the Oct4A is the pluripotent promoter, whereas, Oct4B is related to stress responses (Wang et al., 2009). To study these promoters in greater depth, we queried ‘Pou5f1’ in BindDB (Figure 2A). The tool automatically discerns the presence of two alternative promoters while searching for the gene's genomic location and directs the user to a heatmap of per-promoter results. From the heatmap, it is apparent that both promoters share some of their epigenetic signature including H3K4me3, H3K27ac and H3K9ac active chromatin marks (Figure 2B). Also, the OCT4 protein itself binds both of its promoters, along with other pluripotent factors such as UTX and KLF4. Beyond this common epigenetic core, the two promoters differ widely: The Oct4A promoter of the longer isoform displays binding for a plethora of factors and evidence of RNAPII suggests that Oct4 is actively transcribed in ESCs. In general, this type of large-scale epigenetic profile is characteristic of active gene regulation and goes far beyond the simple histone modification code. On the other hand, the Oct4B promoter exhibits binding for an entirely different set of factors including SMC1, SMC3, SA1, SA2, RAD21, and CTCF, the main components of the structural Cohesin complex. It is also bound by ZC3H11A, which we found to cluster globally with the Cohesin complex (data not shown). Cohesin binding to the Oct4B promoter suggests that a specific enhancer may be driving the expression of this second isoform in ESCs. On the other hand, the mediator proteins MED1 and MED12 bind the first promoter and do not cluster with the Cohesin complex as they are proposed to do. These findings may be particularly insightful into discovering the molecular mechanisms governing alternative promoter regulation of Oct4 and genes in general.