In vivo target of a transcriptional activator revealed by fluorescence resonance energy transfer

Abstract

Our understanding of eukaryotic transcriptional activation mechanisms has been hampered by an inability to identify the direct in vivo targets of activator proteins, primarily because of lack of appropriate experimental methods. To circumvent this problem, we have developed a fluorescence resonance energy transfer (FRET) assay to monitor interactions with transcriptional activation domains in living cells. We use this method to show that the Tra1 subunit of the SAGA (Spt/Ada/Gcn5/acetyltransferase) complex is the direct in vivo target of the yeast activator Gal4. Chromatin-immunoprecipitation experiments demonstrate that the Gal4-Tra1 interaction is required for recruitment of SAGA to the upstream activating sequence (UAS), and SAGA, in turn, recruits the Mediator complex to the UAS. The UAS-bound Mediator is required for recruitment of the general transcription factors to the core promoter. Thus, our results identify the in vivo target of an activator and show how the activator-target interaction leads to transcriptional stimulation. The FRET assay we describe is a general method that can be used to identify the in vivo targets of other activators.

Figure 1.

Development of a FRET assay for detecting interactions with Gal4 in vivo. (A) Experimental strategy. Direct interaction between Gal4-ECFP and an EYFP-tagged protein (black) results in a peak of fluorescence emission at 525 nm and a concomitant reduction in donor (ECFP) emission. (B) Fluorescence emission spectra in yeast strains coexpressing Gal4-ECFP and Gal80-EYFP, or expressing Gal4-ECFP alone, Gal80-EYFP alone, or Gal4-ECFP and unfused EYFP expressed from the GAL80 promoter. (C) Fluorescence emission spectra in Gal4-ECFP/Gal80-EYFP cells in which the EYFP acceptor fluorophore was photobleached at 514 nm prior to excitation of the ECFP donor fluorophore with the 405-nm laser line. (PB) Photobleaching.

Figure 2.

Identification of the SAGA subunit that is contacted by Gal4. (A) Fluorescence emission spectra in yeast strains coexpressing Gal4-ECFP and one of the 14 SAGA subunits fused to EYFP. (B) FRET efficiency in yeast strains coexpressing Gal4-ECFP and Gal80-EYFP or one of the 14 SAGA subunits fused to EYFP. FRET efficiency (E_FRET) was calculated using the formula E_FRET = (I_post - I_pre)/I_post, where I_pre and I_post are the fluorescence intensities of ECFP before and after photobleaching, respectively (see Materials and Methods for details). Spectra were recorded from three cells; the average FRET efficiency and standard deviation are shown. (C) Fluorescence emission spectra for all 14 Gal4-ECFP/SAGA subunit-EYFP strains following excitation of the EYFP acceptor fluorophore with the 488-nm laser line. (D) Growth of strains expressing SAGA subunit-EYFP fusions in dextrose (YPD) and galactose (YPG) media. (E) Analysis of PHO84 transcription in strains expressing SAGA-EYFP fusion proteins. Transcription was monitored by primer-extension analysis, quantitated, and presented as the percent PHO84 transcription relative to the untagged strain FY23.

Figure 3.

Confirmation of the Gal4-Tra1 interaction and analysis of other putative Gal4 targets. (A) Fluorescence emission spectra in Gal4-ECFP/Tra1-EYFP cells that have been photobleached as described in Fig. 1C. (PB) Photobleaching. (B) Fluorescence emission spectra of yeast strains expressing Gal4-EYFP and an N-terminal ECFP-tagged SAGA subunit. (C) FRET efficiency in yeast strains coexpressing Gal4-EYFP and ECFP-SAGA fusion proteins. FRET efficiency was calculated as described in Fig. 2B. (D) Fluorescence emission spectra in cells expressing Gal4-ECFP and one of four EYFP-tagged possible targets, TBP, TFIIB, Srb4, or Gal11.

Figure 4.

The Gal4-Tra1 interaction occurs only in galactose medium. (A) Fluorescence emission spectra in Gal4-ECFP/Tra1-EYFP cells at various time points following a shift from raffinose to galactose medium. (B) Kinetics of the Gal4-Tra1 interaction, monitored by FRET (data from A), and transcriptional induction of GAL1, monitored by primer-extension (Bhaumik and Green 2002; inset), following a shift from raffinose to galactose medium. (C) Fluorescence emission spectra in Gal4-ECFP/Gal80-EYFP cells in dextrose, raffinose, and galactose media.

Figure 5.

The Gal4-Tra1 interaction is required for SAGA recruitment and transcriptional activation. (A) Fluorescence emission spectra of yeast strains expressing Tra1-EYFP and ECFP fused to either Gal4 or derivatives lacking its AD, Gal4(ΔAD)-ECFP, or its DNA-binding domain, Gal4(ΔDBD)-ECFP. (B) Transcriptional analysis of GAL1 and SED1 by primer-extension (left) and ChIP analysis of Gal4 and SAGA recruitment to the GAL1 UAS (right) following inactivation of Tra1. Formaldehyde-based in vivo cross-linking and ChIP was performed as previously described (Bhaumik and Green 2001). Immunoprecipitated (IP) DNA was amplified by PCR using primer pairs corresponding to the GAL1 UAS, and IP DNA was quantitated and presented as the ratio of IP to input relative to wild type. (C) Fluorescence emission spectra in Gal4-ECFP/Tra1-EYFP cells in the presence and absence of the SAGA subunit Spt20. (Inset) Immunoblot showing Tra1-EYFP levels in wild-type and spt20Δ strains. (D) ChIP analysis of Tra1 recruitment to the GAL1 UAS in wild-type cells and in an spt20Δ deletion mutant.

Figure 6.

An ordered protein interaction network at the GAL1 UAS facilitates PIC assembly. (A) ChIP analysis of Gal4, SAGA, and Mediator binding at GAL1. IP DNA was amplified by PCR using primer pairs corresponding to the UAS or core promoter of GAL1 (Bhaumik and Green 2001). (B) SAGA is required for recruitment of Mediator to the GAL1 UAS. A ChIP assay was performed using primer pairs corresponding to the GAL1 UAS, and IP DNA was quantitated and presented as described in Fig. 5B. (C) ChIP analysis of Gal4, SAGA, and Mediator recruitment at the GAL1 UAS following inactivation of Srb4.

Figure 7.

Recruitment of SAGA and Mediator to the GAL1 UAS is required for PIC assembly at the core promoter. (A) PIC assembly at the GAL1 core promoter is dependent on Gal4, SAGA, and Mediator. (B) Analysis of Gal4, SAGA, and Mediator recruitment and PIC assembly following inactivation of TBP and TFIIB. (C) Schematic summary. (Top) At the GAL1 UAS, SAGA serves as an “adaptor” that recruits Mediator to the Gal4 AD, and ultimately results in PIC assembly at the core promoter. The interaction between Gal4 and SAGA occurs directly via Tra1. Other interactions, indicated by arrows, have not been demonstrated to be direct. (Bottom) At some promoters, the activator (Act) functions by direct contact with Mediator.