Spt10 and Spt21 Are Required for Transcriptional Silencing in Saccharomyces cerevisiae

SPT10 and SPT21 are required for telomere position effect. All strains were grown in glucose, which represses expression of the GAL1pr-SPT10 allele. (A) Fivefold dilution spot assays with strains carrying the telV-R::URA3 reporter. Cultures were grown in YPD overnight to saturation, subjected to 5-fold serial dilutions, and spotted onto the indicated media. All strains have the telV-R::URA3 reporter, except for the wild-type URA3 strain, which has the wild-type URA3 gene at the endogenous locus, where it is not silenced. The mild growth defect of GAL1pr-SPT10 and spt21Δ mutants can be seen on the complete plate. (B) Real-time reverse transcription-PCR (RT-PCR) analysis of silencing of YFR057W. The measurement of YFR057W transcript levels was normalized to ACT1. Values are expressed relative to the wild-type level, which was assigned a value of 1. Shown are the means ± SEM for at least three independent experiments. The y axis was truncated to facilitate comparison; the value for sir2Δ is indicated above the corresponding bar.

GAL1pr-SPT10 and spt21Δ show mating type silencing defects in a MATa background. All strains were grown in glucose, which represses expression of the GAL1pr-SPT10 allele. (A) Quantitative mating assays with MATa and MATα haploid strains. Cells of each query strain were incubated for 5 h with a wild-type lawn of the opposite mating type, and the percentage of mated query cells was calculated (see Materials and Methods). Shown are the means ± standard deviations for at least three independent experiments. (B) Northern blot analysis of silencing at HMLα. The strains in lanes 1 to 11 are MATa, and the strain in lane 13 is MATα. The α1 and α2 transcripts are silenced in wild-type MATa haploids and expressed in wild-type MATα haploids.

Levels of rDNA silencing are increased in GAL1pr-SPT10 and spt21Δ mutants. All strains were grown in glucose, which represses expression of the GAL1pr-SPT10 allele. Fivefold dilution spot tests were done on the indicated strains, all of which carry an mURA3 reporter in a single copy within the rDNA. Growth on medium without uracil (−Ura) assesses the degree of reporter silencing, and complete medium controls for growth. Papillation reflects the stochastic nature of silencing changes.

(A) A high-copy-number plasmid encoding all four histones can partially suppress spt21Δ and GAL1pr-SPT10 silencing phenotypes. All strains were grown in glucose, which represses expression of the GAL1pr-SPT10 allele. The indicated strains carry either an empty vector (2μ LEU2) or a high-copy-number plasmid encoding the first copy of each histone protein (2μ LEU2 HTA1-HTB1 HHT1-HHF1). Fivefold dilution spot tests were grown on either medium without leucine (−Leu), which selects for plasmid maintenance, or on −Leu plus 5-FOA. Cells are able to grow in the presence of 5-FOA only when the telomere position effect is intact, and they are able to silence URA3 expression. (B) A high-copy-number plasmid with SIR3 can partially suppress the silencing defects of GAL1pr-SPT10 and spt21Δ. Fivefold dilution spot tests of telomere position effect were performed on the indicated strains carrying either an empty vector (2μ-LEU2) or a high-copy-number plasmid with SIR3 (2μ-LEU2-SIR3). All strains have the telV-R::URA3 reporter.

Measurement of Sir protein levels and Sir protein association with silenced chromatin in GAL1pr-SPT10 and spt21Δ mutants. All strains were grown in glucose, which represses expression of the GAL1pr-SPT10 allele. (A) Measurement of Sir protein levels. Total protein was isolated from the indicated strains and analyzed by Western analysis. Pgk1 is a glycolytic enzyme and serves as a loading control. (B) Chromatin immunoprecipitation (ChIP) of Sir2 on chromosome VI. Antibodies specific to Sir2 were used to immunoprecipitate the proteins in sonicated chromatin extracts from wild-type, GAL1pr-SPT10, and spt21Δ strains. Following reversal of cross-linking, the amount of immunoprecipitated DNA was quantitated using real-time PCR with primers specific to regions 0.6 kb and 2.8 kb from the right telomere of chromosome VI. Values were calculated relative to binding at the ACT1 gene, which is not subjected to silencing, and normalized to input DNA. Shown is the relative enrichment of Sir2 binding at the two subtelomeric locations. Plotted are the means ± SEM for four independent experiments. (C) ChIP of Sir3 was performed as described for Sir2, using an anti-Sir3 antiserum. (D) ChIP of histone H3 was conducted as described for Sir2, using an antiserum specific to histone H3.

Measurement of histone H3 K79 dimethylation and histone H4 acetylation levels at telomere VI-R in GAL1pr-SPT10 and spt21Δ mutants. All strains were grown in glucose, which represses expression of the GAL1pr-SPT10 allele. (A) ChIP of histone H3 K79 dimethylation was performed using the same strains, primers, and calculations as for Fig. 6, using antibodies specific to dimethyl histone H3 K79. Values are shown relative to total histone H3 ChIP. (B) ChIP of acetylated histone H4. (C) Western blot to measure total cellular levels of dimethylated histone H3 K79, acetylated histone H4, total histone H3, and Pgk1 as a loading control.