Fkh1 and Fkh2 associate with Sir2 to control CLB2 transcription under normal and oxidative stress conditions

Abstract

The Forkhead (Fkh) box family of transcription factors is evolutionary conserved from yeast to higher eukaryotes and its members are involved in many physiological processes including metabolism, DNA repair, cell cycle, stress resistance, apoptosis, and aging. In budding yeast, four Fkh transcription factors were identified, namely Fkh1, Fkh2, Fhl1, and Hcm1, which are implicated in chromatin silencing, cell cycle regulation, and stress response. These factors impinge transcriptional regulation during cell cycle progression, and histone deacetylases (HDACs) play an essential role in this process, e.g., the nuclear localization of Hcm1 depends on Sir2 activity, whereas Sin3/Rpd3 silence cell cycle specific gene transcription in G2/M phase. However, a direct involvement of Sir2 in Fkh1/Fkh2-dependent regulation of target genes is at present unknown. Here, we show that Fkh1 and Fkh2 associate with Sir2 in G1 and M phase, and that Fkh1/Fkh2-mediated activation of reporter genes is antagonized by Sir2. We further report that Sir2 overexpression strongly affects cell growth in an Fkh1/Fkh2-dependent manner. In addition, Sir2 regulates the expression of the mitotic cyclin Clb2 through Fkh1/Fkh2-mediated binding to the CLB2 promoter in G1 and M phase. We finally demonstrate that Sir2 is also enriched at the CLB2 promoter under stress conditions, and that the nuclear localization of Sir2 is dependent on Fkh1 and Fkh2. Taken together, our results show a functional interplay between Fkh1/Fkh2 and Sir2 suggesting a novel mechanism of cell cycle repression. Thus, in budding yeast, not only the regulation of G2/M gene expression but also the protective response against stress could be directly coordinated by Fkh1 and Fkh2.

Keywords: Fkh1; Fkh2; Sir2; budding yeast; cell cycle; silencing; stress.

Figure 1

Fkh1 and Fkh2 associate with Sir2. (A) Pull-down assay. GST and GST-Sir2 proteins expressed in E. coli were immobilized on glutathione sepharose beads and incubated with lysates derived from yeast strains carrying Myc-tagged FKH1, FKH2, and HCM1. Immunodetection of co-precipitated Fkh1-Myc, Fkh2-Myc and Hcm1-Myc was performed with a mouse α-Myc antibody. (B) BiFC analysis. Yeast cells expressing the fusion protein Sir2-VC were transformed with plasmids encoding the fusion proteins VN-Fkh1, VN-Fkh2, and VN-Hcm1 under the control of the constitutive GPD promoter. Venus signals were analyzed.

Figure 2

Sir2 represses gene transcription via Fkh transcription factors. (A) Reporter gene activity assay. Yeast cultures were spotted in 1:5 serial dilutions onto SDII and SDIV media, and cell growth was analyzed after 5 days. (B) Liquid β-galactosidase assay. Protein lysates were prepared from exponentially growing yeast cells expressing LexA-Fkh1, LexA-Fkh2, or LexA-Hcm1 fusion proteins. Each bar represents the mean average obtained from three independent experiments. (C) Genetic interaction studies. Wild type strain BY4741 and deletion strains fkh1Δ, fkh2Δ, sir2Δ, fkh1Δsir2Δ, and fkh2Δsir2Δ were transformed with p423GAL -Sir2 or vector p423GAL as control. Subsequently, yeast cells were grown to mid-exponential phase, spotted in 1:5 serial dilutions on glucose or on galactose plates, and growth of yeast cells was analyzed after 3 days. The assay was performed three times, and one representative experiment is shown.

Figure 3

Sir2 regulates the CLB2 transcript level and binds to the CLB2 promoter in a cell cycle-dependent manner. (A,B) Quantitative Real-Time PCR. Total RNA was isolated from (A) hydroxyurea- or (B) nocodazole- arrested wild type, fkh1Δ, fkh2Δ, fkh1Δfkh2Δ, sir2Δ, fkh1Δsir2Δ, and fkh2Δsir2Δ cells. The ACT1 gene was used as control. Each bar represents the mean average obtained from three independent experiments. (C) ChIP assay. Protein/DNA complexes were precipitated from cells grown in exponential phase or synchronized with α-factor (G1 phase), hydroxyurea (S phase) or nocodazole (M phase) using an anti-Myc antibody. Each bar represents the mean average obtained from three independent experiments. (D) BiFC analysis. Haploid cells expressing the fusion protein Sir2-VC were transformed with plasmid p426GPD-VN-Fkh1, and selected transformants were synchronized in exponential growth (OD₆₀₀ ~0.6) with α-factor. Arrested cells were released into fresh media and samples were collected every 10 min for the detection of BiFC signals by fluorescence microscopy. DNA content of samples was determined by propidium iodide staining and FACS analysis.

Figure 4

Interplay between Fkh transcription factors and Sir2 under oxidative stress conditions. (A) Growth analysis. Wild type, fkh1Δ, fkh2Δ, fkh1Δfkh2Δ, sir2Δ, fkh1Δsir2Δ, and fkh2Δsir2Δ deletion strains were grown in YPD medium to saturation overnight and spotted in 1:5 serial dilutions on CSM medium containing 2 mM H₂O₂ or 40 μM MD. Cell growth was analyzed after 3 days. The assay was performed three times, and one representative experiment is shown. (B) BiFC analysis. Haploid cells expressing Sir2-VC or Ndd1-VC were transformed with plasmid p426GPD-VN-Fkh1 or p426GPD-VN-Fkh2, respectively. Subsequently, yeast cells were analyzed for BiFC signals in exponential phase (OD₆₀₀ ~0.6), in stationary phase (OD₆₀₀ ~1.6) and in the presence of 2 mM H₂O₂ or 40 μM MD.

Figure 5

Sir2 binds to the CLB2 promoter via Fkh1 and Fkh2 under stress conditions. (A,B) ChIP assay. (A) Haploid cells were grown to exponential phase (OD₆₀₀ ~0.6), stationary phase (OD₆₀₀ ~1.6) or treated with 2 mM H₂O₂. Protein extracts were prepared and ChIP experiments were carried out using an anti-Myc antibody. The ACT1 gene was used as control. Each bar represents the mean average obtained from three independent experiments. (B) Haploid wild type and fkh1Δ or fkh2Δ cells expressing a C-terminal tagged Sir2-Myc fusion protein were grown to stationary phase (OD₆₀₀ ~1.6), and ChIP experiments were performed using an anti-Myc antibody. The ACT1 gene was used as control. Each bar represents the mean average obtained from three independent experiments. (C) Localization studies. Haploid wild type, fkh1Δ and fkh2Δ cells expressing GFP-tagged Sir2 were grown overnight to saturation (OD₆₀₀ ~1.6), and Sir2-GFP signals were analyzed by fluorescence microscopy.

Figure 6

CLB2 transcript and protein levels are altered under stress conditions. (A,B) Quantitative Real-Time PCR. Total RNA was prepared from exponentially growing wild type and fkh1Δ, fkh2Δ, fkh1Δfkh2Δ, sir2Δ, fkh1Δsir2Δ, and fkh2Δsir2Δ cells (OD₆₀₀ ~0.6) that were treated with 2 mM H₂O₂ for 90 min (A) or grown to stationary phase (OD₆₀₀ ~1.6) (B). The ACT1 gene was used as control. Each bar represents the mean average obtained from three independent experiments. (C) Protein extracts were isolated from exponential growing cells (OD₆₀₀ ~0.6), cells incubated with 2 mM H₂O₂ for 90 min or cells grown to stationary phase. Clb2 levels were determined by Western Blot using an α-Clb2 specific antibody. Coomassie Brilliant Blue protein staining was used as a loading control.

Figure 7

High Clb2 levels influence survival of yeast cells under oxidative stress conditions. (A) Survival of wild type and fkh1Δfkh2Δ cells upon nutrient limitation. Yeast strains were grown to saturation (OD₆₀₀ ~1.6) and further incubated in H₂O. Samples of cultures were spotted in 1:5 serial dilutions on SC medium after 14 or 21 days. (B,C) Clb2 overexpression in exponentially grown (B) and stationary (C) cells in presence of H₂O₂. Yeast cells transformed with plasmids p423GAL-Clb2 or p423GAL were incubated in SC medium containing 2% galactose with or without 2 mM H₂O₂ for 12 h. Yeast cultures were then spotted in 1:5 serial dilutions on glucose- or galactose-containing plates, and growth was analysed after 3 days. One representative experiment is shown.

Figure 8

Regulatory model of Clb2 expression. The transcription factor Hcm1 is required for periodic expression of the G2/M phase-specific regulators Fkh1, Fkh2, and Ndd1 (Pramila et al., 2006). Control of Clb2 expression requires the binding of Fkh1, Fkh2, and the transcriptional coactivator Ndd1 (Darieva et al., ; Reynolds et al., ; Pic-Taylor et al., 2004). In late M phase and during G1 phase, Ndd1 depleted from CLB2 promoter allows for binding of the HDAC Sir2. Our findings suggest a repressive association between Fkh1/Fkh2 and Sir2, with a predominant role for Fkh1. This is also supported by findings of (Hollenhorst et al., 2001). In line with our results, nuclear localization of Hcm1 was shown to be Sir2-dependent (Rodriguez-Colman et al., 2010).