Mediator and SAGA have distinct roles in Pol II preinitiation complex assembly and function

Abstract

A key feature of RNA polymerase II (Pol II) preinitiation complexes (PICs) is their ability to coordinate transcription initiation with chromatin modification and remodeling. To understand how this coordination is achieved, we employed extensive proteomic and mechanistic analyses to study the composition and assembly of PICs in HeLa cell and mouse embryonic stem cell (ESC) nuclear extracts. Strikingly, most of the machinery that is necessary for transcription initiation on chromatin is part of the PIC. The PIC is nearly identical between ESCs and HeLa cells and contains two major coactivator complexes: Mediator and SAGA. Genome-wide analysis of Mediator reveals that it has a close correlation with Pol II, TATA-binding protein, and messenger RNA levels and thus may play a major role in PIC assembly. Moreover, Mediator coordinates assembly of the Pol II initiation factors and chromatin machinery into a PIC in vitro, whereas SAGA acts after PIC assembly to allow transcription on chromatin.

Figure 1. PIC Analysis in HeLa and Mouse ES Cell Nuclear Extracts

(A) In vitro transcription from HeLa extract and mouse ES cell nuclear extract on naked DNA and chromatin templates. (B) MuDPIT Analysis of PICs formed on naked DNA and chromatin templates in HeLa and ES cell extracts in the presence and absence of GAL4-VP16. The primary data were analyzed by MS Sort (See Supplemental Table 1) and the resulting activator-enriched (1.5-fold or greater; above dotted line) protein complexes were scored by average NSAF abundance. Average unique NSAFs (Uq-) for some complexes are also shown as described in the text (see supplementary information for detailed description of analysis). The fold enrichment and average NSAF values are color-coded according to the percentile of proteins shown within each experimental condition (e.g. both fold enrichment and NSAF values for Mediator are above the 90th percentile of proteins shown in the chart for each experimental condition and are therefore colored red (see key)). (C) Immunoblots representing known PIC components assembled during a 54-min time course on naked DNA templates using either HeLa (left) or mouse ES cell (right) nuclear extract.

Figure 2. Mediator Composition and Function in PIC Assembly

(A) Genomewide analysis of Mediator, TBP and Pol II distribution in ES cells. Genes were ranked by average binding of Pol II (-log (P-val)) within a 10-kb window surrounding the transcription start site of mouse promoters with a significant enrichment for any of the three factors (p-val <10⁻⁵). Expression data (RPKM) were plotted consequently for the same ranking order. (B) Chart depicting average NSAFs of complexes detected in MuDPIT analysis of purified FLAG-Med29 from sonicated HeLa and ES cell nuclei at 100, 200, 300 and 500 mM KCl (indicated as a gradient). ES cell Mediator was isolated from nuclei at 420 mM KCl. After MuDPIT and MS Sort, average NSAFs for proteins were ranked by abundance relative to the HeLa 100 mM dataset. Values are color-coded according to percentile rank within all detected complexes. (C) Pearson correlation comparison of Mediator-associated factors in unmodified chromatin and PICs assembled in vitro. (D) Immunoblots of PIC assembly comparing the recruitment of Mediator-associated protein complexes identified by MuDPIT in Mock- or Mediator-depleted HeLa nuclear extracts. Pure Mediator was added back to depleted extracts to rescue binding.

Figure 3. SAGA Composition and Role in PIC Assembly and Function

(A) Chart depicting average NSAFs of complexes detected in MuDPIT analysis of FLAG-Spt3 from sonicated HeLa nuclei at 150 and 300 mM KCl and from soluble HeLa nuclear extract at 450 mM. Values are color-coded according to percentile rank within detected complexes. (B) Immunoblots of PIC assembly comparing the recruitment of complexes identified by MuDPIT in Mock- or SAGA-depleted HeLa nuclear extracts plus or minus pure SAGA. Different modules of SAGA were detected using antibodies against the TRRAP, HAT/core (GCN5), SPT (SPT3), Deubiquitination (ATXN7), and TAF (TAF5L) modules. (C) In vitro transcription in SAGA-depleted extracts on naked DNA and unmodified chromatin templates plus or minus pure SAGA. A 2-fold longer exposure of transcription on chromatin is shown. (D) Immunoblots of immobilized template recruitment experiments with GAL4-VP16 and pure Mediator and/or SAGA. (E) Model showing that activators recruit Mediator and SAGA independently. Mediator recruits the indicated factors. SAGA functions on chromatin.