The Saccharomyces cerevisiae Srb8-Srb11 complex functions with the SAGA complex during Gal4-activated transcription

Abstract

The Saccharomyces cerevisiae SAGA (Spt-Ada-Gcn5-acetyltransferase) complex functions as a coactivator during Gal4-activated transcription. A functional interaction between the SAGA component Spt3 and TATA-binding protein (TBP) is important for TBP binding at Gal4-activated promoters. To better understand the role of SAGA and other factors in Gal4-activated transcription, we selected for suppressors that bypass the requirement for SAGA. We obtained eight complementation groups and identified the genes corresponding to three of the groups as NHP10, HDA1, and SRB9. In contrast to the srb9 suppressor mutation that we identified, an srb9Delta mutation causes a strong defect in Gal4-activated transcription. Our studies have focused on this requirement for Srb9. Srb9 is part of the Srb8-Srb11 complex, associated with the Mediator coactivator. Srb8-Srb11 contains the Srb10 kinase, whose activity is important for GAL1 transcription. Our data suggest that Srb8-Srb11, including Srb10 kinase activity, is directly involved in Gal4 activation. By chromatin immunoprecipitation studies, Srb9 is present at the GAL1 promoter upon induction and facilitates the recruitment or stable association of TBP. Furthermore, the association of Srb9 with the GAL1 upstream activation sequence requires SAGA and specifically Spt3. Finally, Srb9 association also requires TBP. These results suggest that Srb8-Srb11 associates with the GAL1 promoter subsequent to SAGA binding, and that the binding of TBP and Srb8-Srb11 is interdependent.

FIG. 1.

(A) Phenotypes of suppressors of the spt20Δ Gal⁻ phenotype. Ten-fold serial dilutions of the indicated yeast strains were spotted on medium containing glucose (YPD), galactose (YPgal), caffeine (YPcaf), and lacking inositol (SD-Ino). The most concentrated spot contains cells from a culture at a concentration of 1 × 10⁷ cells/ml. The srb9Δ mutants cause a weak Ino⁻ phenotype. (B) The srb9^sup allele does not suppress the Gal⁻ phenotypes of the gal11Δ and snf2Δ mutants. Ten-fold serial dilutions of the indicated yeast strains were spotted on medium containing glucose or galactose as described for panel A. (C) The srb9^sup allele partially bypasses the requirement for Spt20 in GAL1 transcription. Yeast strains were grown in YPraf medium and induced with galactose for 20, 40, 60, and 120 min. Samples were removed for RNA extraction and Northern analysis at each time point. SNR190 served as the loading control.

FIG. 2.

(A) The srb8Δ-srb11Δ mutants and the srb10-3 mutant all have Gal⁻ phenotypes. Indicated strains were patched onto YPD medium and replica plated to YPD and YPgal plates. Photographs were taken on days 2 and 3 after replica plating to show that the srb8Δ-srb11Δ mutants and the srb10-3 mutants have Gal⁻ phenotypes (day 2), but these phenotypes are weaker than that observed with an spt20Δ strain (day 3). (B) Srb9 protein and Srb10 kinase activity are important for transcription of the GAL1 gene. Yeast strains were grown in YPraf medium and induced with galactose for 20, 40, 60, and 120 min. Samples were removed for RNA extraction and Northern blotting at each time point. SNR190 served as the loading control. (C) The Srb9 protein and Srb10 kinase activity are important for TBP binding to the GAL1 TATA box. A ChIP assay was conducted with strains grown in medium containing raffinose, followed by the addition of galactose for 20 min (top) or 120 min (bottom) prior to cross-linking with formaldehyde. All ChIP experiments have been quantified as the ratio of the percentage of immunoprecipitation DNA of the sample primer set (e.g., GAL1 TATA) compared to the percentage of immunoprecipitation of a control primer set that amplifies a region on chromosome V. Each data point represents the mean and standard error of at least three independent experiments.

FIG. 3.

(A) Srb9 associates with the GAL1 UAS and requires SAGA for its binding. A ChIP assay was conducted with strains grown to mid-log phase in medium containing raffinose, followed by the addition of galactose, for 20 min (middle) or 120 min (bottom) prior to cross-linking with formaldehyde. An anti-Myc antibody that recognizes the Myc epitope on Srb9-13xMyc was used for immunoprecipitation. As a control, a strain lacking the Srb9-13xMyc-tagged protein was used (no tag). Shown is an example of the radioactive PCRs for Srb9 ChIP. The mean and standard error of at least three independent experiments are shown in the bar graph. The triangles above the bands indicate twofold differences in the level of input chromatin in the reaction, to ensure that the PCR was in the linear range. (B) Spt20 and Spt3 can bind to the GAL1 UAS largely independently of Srb9. A ChIP assay was conducted with strains grown in YPraf medium, followed by the addition of galactose for 20 min prior to cross-linking with formaldehyde. An antibody that recognizes the hemagglutinin epitope on Spt20 and Spt3 used for immunoprecipitation. The mean and standard error of at least three experiments are shown.

FIG. 4.

Srb9 association with the GAL1 UAS requires TBP binding. A ChIP assay was conducted on strains grown in YPraf medium, followed by the addition of galactose for 20 min prior to cross-linking with formaldehyde. An antibody that recognizes the hemagglutinin epitope on Spt20 and Spt3 in respective strains was used for immunoprecipitation, and a Myc antibody was used to immunoprecipitate the Srb9-13xMyc protein. The data shown are the relative fold enrichments, where the value for the wild-type strain has been set to 1. The actual fold enrichment values for Srb9 ChIP are 20.5 ± 6.5 for the wild type and 3.8 ± 0.7 for spt15-21. For the Spt3 ChIP, the values are 2.7 ± 1.2 for the wild type and 2.6 ± 0.8 for spt15-21. For the Spt20 ChIP, the values are 7.6 ± 1.2 for the wild type and 4.6 ± 2.4 for spt15-21.

FIG. 5.

SAGA and Srb9 associate with Gal4-binding sites in the absence of RNA Pol II. A ChIP assay was conducted with strains grown in YPraf medium, followed by the addition of galactose for 20 min prior to cross-linking with formaldehyde. Antibodies that recognize the hemagglutinin epitope on Spt20-3xHA, the Myc epitope on Srb9-13xMyc, and the 8WG16 anti-Rbp1 antibody were used for immunoprecipitation. Binding of factors to the isolated Gal4-binding sites is shown in white bars, and binding to the endogenous GAL1 UAS is shown in gray bars. The mean and standard error are shown for at least three experiments.

FIG. 6.

The Gal11 core Mediator component is not required for the association of SAGA but is required for the association of Srb8-Srb11, the Rox3 core Mediator component, and TBP. A ChIP assay was conducted with strains grown in YPraf medium, followed by the addition of galactose for 20 min prior to cross-linking with formaldehyde. Antibodies that recognize the Ada1 SAGA component, the Myc epitope on Srb9-13xMyc, TBP, and Rox3 were used for ChIP. Data shown represent the mean and standard error of at least three experiments.

FIG. 7.

Rox3 binding to the GAL1 UAS is partially dependent on SAGA and Srb8-Srb11. A ChIP assay was conducted with strains grown in medium containing raffinose, followed by the addition of galactose for 20 min prior to cross-linking with formaldehyde. Antibodies that recognize the Rox3 protein were used for the ChIP assay. Data represent the mean and standard error of at least three experiments.

FIG. 8.

A dependency pathway model for assembly of factors required for activation at GAL1. (A) Prior to induction, only Gal4 is bound to the GAL1 promoter. (B) Upon induction by galactose, Gal4 recruits SAGA. The recruitment of Mediator is dependent upon both SAGA and Gal4. (C) Spt3 of SAGA recruits TBP, and SAGA and Mediator recruit Srb8-Srb11. In the assembled structure, Srb8-Srb11 and TBP are mutually required for their association at the GAL1 promoter.