The Sir2-Sum1 complex represses transcription using both promoter-specific and long-range mechanisms to regulate cell identity and sexual cycle in the yeast Kluyveromyces lactis

Abstract

Deacetylases of the Sir2 family regulate lifespan and response to stress. We have examined the evolutionary history of Sir2 and Hst1, which arose by gene duplication in budding yeast and which participate in distinct mechanisms of gene repression. In Saccharomyces cerevisiae, Sir2 interacts with the SIR complex to generate long-range silenced chromatin at the cryptic mating-type loci, HMLalpha and HMRa. Hst1 interacts with the SUM1 complex to repress sporulation genes through a promoter-specific mechanism. We examined the functions of the non-duplicated Sir2 and its partners, Sir4 and Sum1, in the yeast Kluyveromyces lactis, a species that diverged from Saccharomyces prior to the duplication of Sir2 and Hst1. KlSir2 interacts with both KlSir4 and KlSum1 and represses the same sets of target genes as ScSir2 and ScHst1, indicating that Sir2 and Hst1 subfunctionalized after duplication. However, the KlSir4-KlSir2 and KlSum1-KlSir2 complexes do not function as the analogous complexes do in S. cerevisiae. KlSir4 contributes to an extended repressive chromatin only at HMLalpha and not at HMRa. In contrast, the role of KlSum1 is broader. It employs both long-range and promoter-specific mechanisms to repress cryptic mating-type loci, cell-type-specific genes, and sporulation genes and represents an important regulator of cell identity and the sexual cycle. This study reveals that a single repressive complex can act through two distinct mechanisms to regulate gene expression and illustrates how mechanisms by which regulatory proteins act can change over evolutionary time.

Figure 1. KlSir2 co-precipitates with KlSir4 and KlSum1.

(A) KlSir2-HA, KlSir4-Flag or myc-KlSum1 was precipitated from a lysate prepared from wild-type (LRY2285) or rfm1Δ (LRY2528) strains, and the precipitated material was examined by immunoblotting with an antibody against the HA tag to detect KlSir2-HA. The input represents 33% of the IP. (B) KlSir2-HA or KlSir4-Flag was immunoprecipitated from a sum1Δ strain (LRY2158), and the precipitated material was examined with an antibody against the HA tag to detect KlSir2-HA or the Flag tag to detect KlSir4-Flag. (C) KlSir2-HA or myc-KlSum1 was immunoprecipitated from a sir4Δ strain (LRY2282), and the precipitated material was examined with an antibody against the HA tag to detect KlSir2-HA or the myc tag to detect myc-KlSum1.

Figure 2. KlSir2, KlSir4, and KlSum1 silence the cryptic mating-type locus HMLα.

Quantitative RT–PCR analysis of HMLα1, HMLα2 and HMLα3 mRNA in wild-type (CK213), sir2Δ (SAY569), sir4Δ (LRY2038), sum1Δ (LRY2035), rfm1Δ (LRY2528), and sir2Δ sum1Δ (LRY2533) strains. The amount of cDNA was first normalized to the control locus ACT1. The values shown here represent the relative amount of cDNA for each deletion strain compared to the wild-type strain. Error bars represent the SEM.

Figure 3. KlSir2, KlSir4, KlSum1, and KlRfma1 spread across HMLα.

(A) The association of KlSir2-HA, KlSir4-Flag, myc-KlSum1 (LRY2239) and KlRfm1-HA (LRY2327) with HMLα as assessed by chromatin IP followed by quantitative PCR. The y-axis represents the relative enrichment normalized to a control locus, RRP7, which is not detectably associated with KlSir2, KlSir4 or KlSum1. A diagram of the HMLα locus is shown under the x-axis. The aqua bar represents the characterized silencer and the orange and brown boxes represent sequences found at HMLα, MAT, and HMR a loci. Asterisks indicate the peaks of enrichment. (B) The association of KlSir2-HA, KlSir4-Flag and myc-KlSum1 with the MATα locus in a strain in which the α-cassette is only found at MAT (LRY2398). (C) The association of KlSir2-HA and myc-KlSum1 with HMLα in a sir4Δ strain (LRY2281). (D) The association of KlSir2-HA and KlSir4-Flag with HMLα in a sum1Δ strain (LRY2158). (E) The association of KlSir2-HA, KlSir4-Flag and myc-KlSum1 with HMLα in a rfm1Δ strain (LRY2528). (F) The association of KlSir4-Flag and myc-KlSum1 with HMLα in a sir2Δ strain (LRY2388). All y-axes are set to the same scale to facilitate the comparison of protein associations in different experiments. Error bars represent the SEM.

Figure 4. KlSir2 and KlSum1, but not KlSir4, silence and spread across HMRa.

(A) Quantitative RT-PCR analysis of HMR a 1 and HMR a 2 in wild-type (SAY538), sir2Δ (SAY544), sir4Δ (LRY1946), sum1Δ (LRY1947), and rfm1Δ (LRY2529) strains. The fold induction was determined as for Figure 2. (B) The association of KlSir2-HA, KlSir4-Flag, myc-KlSum1 (LRY2285) and KlRfm1-HA (LRY2328) with HMRa as assessed by chromatin IP followed by quantitative PCR. (C) The association of KlSir2-HA (LRY2528), myc-KlSum1 and KlSir4-Flag (LRY2529) with HMR a in a rfm1Δ strain. (D) The association of KlSir2-HA with HMR a in a sum1Δ strain (LRY2126) and the association of myc-KlSum1 with HMR a in a sir2Δ strain (LRY2390). All y-axes are set to the same scale to facilitate the comparison of protein associations in different experiments. Error bars represent the SEM.

Figure 5. KlSir2, KlSum1, and KlRfm1 repress sporulation genes in a promoter-specific manner.

(A) Quantitative RT-PCR analysis of CDA2 (KLLA0C17226g), SPS4 (KLLA0F08679g), SPR3 (KLLA0B08129g) and SPS2 (KLLA0C01001g) mRNA in wild-type (SAY538), sir2Δ (SAY544), sum1Δ (LRY1947), rfm1Δ (LRY2529) and sir4Δ (LRY1946) strains. The fold induction was determined as for Figure 2. (B) The association of KlSir2-HA, KlSir4-Flag, myc-KlSum1 (LRY2285) and KlRfm1-HA (LRY2328) with the promoters of CDA2, SPS4 and SPR3 was assessed by chromatin IP followed by quantitative PCR. The y-axis is a log-scale. (C) Distribution of KlSir2-HA, KlSir4-Flag, myc-KlSum1 (LRY2285) and KlRfm1-HA (LRY2328) across the CDA2 locus. The blue bar in the schematic represents the conserved MSE sequence. (D) The association of KlSir2-HA with CDA2 in a sum1Δ strain (LRY2126) and association of myc-KlSum1 with CDA2 in a sir2Δ strain (LRY2390). (E) The association of KlSir2-HA and myc-KlSum1 with CDA2 in a rfm1Δ strain (LRY2529). All y-axes are set to the same scale to compare changes in protein association across experiments. Error bars represent the SEM.

Figure 6. KlSir2 and KlSum1 repress cell-type–specific genes.

(A) Association of KlSir2-HA, KlSir4-Flag, myc-KlSum1 (LRY2285) and KlRfm1-HA (LRY2328) at the MFα1 (KLLA0E19173g), STE18 (KLLA0E06138g) and BAR1 (KLLA0D15917g) promoters in a MATα strain as assessed by chromatin IP followed by quantitative PCR. The y-axis is a log-scale. (B) Quantitative RT–PCR analysis of MFα1 mRNA in MATα wild-type (SAY538), sir2Δ (SAY544), sum1Δ (LRY1947), and sir4Δ (LRY1946) strains and MAT a wild-type (CK213), sir2Δ (SAY569), sum1Δ (LRY2035) and sir4Δ (LRY2038) strains. (C) Quantitative RT-PCR analysis of STE18 mRNA in the same strains analyzed in (B). (D) Quantitative RT-PCR analysis of BAR1 mRNA in the same strains analyzed in panel B. Error bars represent the SEM. (E) PCR amplification of MAT loci in strains analyzed in (B–D) using mating-type specific primers.

Figure 7. Mechanisms of repression mediated by KlSum1-KlSir2.

The KlSum1-KlSir2 complex participates in multiple mechanisms of repression: a promoter-specific mechanism that represses mid-sporulation and cell-type specific genes (top) as well as a long-range spreading mechanism that silences the cryptic mating-type loci either with Sir4 (HMLα; bottom) or without Sir4 (HMR a; middle). The darker shaded proteins represent stronger association than the lighter colored proteins.