ABSTRACT: Transcriptional cofactors communicate regulatory cues from enhancers to promoters and are central effectors of transcription activation and gene expression, which is a hallmark of all multicellular organisms. However, the extent to which different cofactors display intrinsic specificity for distinct promoters is unclear. Testing intrinsic COF – CP preferences requires the systematic assessment of transcriptional activation for many CPs in the presence or absence of a given COF in an otherwise constant standardized reporter system. We therefore combined a plasmid-based high-throughput reporter assay, Self-Transcribing Active Core Promoter-sequencing (STAP-seq), with the specific recruitment of individual COFs to create a high-throughput activator bypass-like assay. Using this assay, we tested whether 23 different individually tethered Drosophila melanogaster COFs could activate transcription from any of 72,000 CP candidates in S2 cells. We selected COFs that represent different functional classes and enzymatic activities: histone-acetyl transferases (Nejire - the fly ortholog of CBP/P300, Mof, Gcn5 - ortholog of pCAF, Atac2, Tip60, Br140), three components of the COMPASS-like histone H3K4 methyltransferase complexes (Lpt, Trr and Trx), nucleosome remodeling ATP-ase (Brm), two Chromo or Chromo-shadow domain (Chro and Mof) and three Bromodomain (Brd8, Brd9 and fs(1)h - ortholog of Brd4) -containing COFs, three subunits of the Mediator Complex (MED15, MED24 and MED25), three subunits of the TFIID general transcription factor (Taf4, Tbp and Trf2) and three less well-characterized COFs (EMSY, Gfzf and Pzg). We used the strong transcriptional activator P65 as a positive and GFP as a negative control. In addition, we tested 6 COFs and the 2 controls in two other Drosophila cell lines: ovarian somatic cells (OSC) and Kc167 cells. We uncovered five classes of core promoters, distinguished by their differential response to specific cofactors and by the presence of known sequence elements, such as the TATA-box, Down-stream Promoter Element, or TCT motif. Moreover, the five promoter types show distinct chromatin properties at endogenous loci. The observed compatibilities between cofactors and promoters can explain how different enhancers specifically activate distinct sets of genes, alternative promoters within the same genes, or even distinct transcription starting sites within the same promoters. Thus, cofactor–promoter compatibilities may underlie distinct transcriptional programs.
