The deacetylase Sir2 from the yeast Clavispora lusitaniae lacks the evolutionarily conserved capacity to generate subtelomeric heterochromatin

Abstract

Deacetylases of the Sir2 or sirtuin family are thought to regulate life cycle progression and life span in response to nutrient availability. This family has undergone successive rounds of duplication and diversification, enabling the enzymes to perform a wide variety of biological functions. Two evolutionarily conserved functions of yeast Sir2 proteins are the generation of repressive chromatin in subtelomeric domains and the suppression of unbalanced recombination within the tandem rDNA array. Here, we describe the function of the Sir2 ortholog ClHst1 in the yeast Clavispora lusitaniae, an occasional opportunistic pathogen. ClHst1 was localized to the non-transcribed spacer regions of the rDNA repeats and deacetylated histones at these loci, indicating that, like other Sir2 proteins, ClHst1 modulates chromatin structure at the rDNA repeats. However, we found no evidence that ClHst1 associates with subtelomeric regions or impacts gene expression directly. This surprising observation highlights the plasticity of sirtuin function. Related yeast species, including Candida albicans, possess an additional Sir2 family member. Thus, it is likely that the ancestral Candida SIR2/HST1 gene was duplicated and subfunctionalized, such that HST1 retained the capacity to regulate rDNA whereas SIR2 had other functions, perhaps including the generation of subtelomeric chromatin. After subsequent species diversification, the SIR2 paralog was apparently lost in the C. lusitaniae lineage. Thus, C. lusitaniae presents an opportunity to discover how subtelomeric chromatin can be reconfigured.

Figure 1. ClHst1 associates with the rDNA repeats.

(A) Features of the three rDNA loci in C. lusitaniae are illustrated. The Pol I-transcribed 35S gene is processed to generate the 18S, 5.8S, and 25S rRNAs (arrows). The Pol III-transcribed 5S gene divides the non-transcribed spacer. All three loci occur near the ends of supercontigs, but only the right end of supercontig 3 has a telomere repeat (gray box), indicating the end of the chromosome. Vertical bars indicate the beginning (3L) or end (8R) of the supercontig sequence. (B) The association of ClHst1 across an rDNA repeat was examined by chromatin IP. Anti-myc antibody was used to immunoprecipitate proteins from ClHST1 (LRY2826) or ClHST1-MYC (LRY2858) yeast. The relative enrichment of each probe compared to a control locus, ClPRI2, is indicated.

Figure 2. ClHst1 causes deacetylation at the rDNA repeats.

(A) The relative abundance of acetylated histones was assessed by immunoblotting. Whole-cell lysates were prepared from replicate samples of ClHST1 yeast (LRY2544) untreated or treated with 25 mM nicotinamide (NAM) or 10 µM trichostatin A (TSA) and from Clhst1Δ yeast (LRY2671). Proteins were electrophoretically separated, transferred to membranes, and probed with antibodies against tetra-acetylated H4, H3K9Ac, or H3K56Ac. (B) The acetylation of histones H3 and H4 across the rDNA repeat was examined by chromatin IP. Total H3, tetra-acetylated H4, H4-K16Ac, H3-K9Ac, or H3-K56Ac was immunoprecipitated from ClHST1 (LRY2544) yeast. For each immunoprecipitation, the recovery of probes was normalized to the control locus, ClPRI2. (C–E) The change in acetylation in deacetylase-deficient yeast relative to wild-type untreated yeast was determined. H3-K9Ac, tetra-acetylated H4, H4-K16Ac, or H3-K56Ac was immunoprecipitated from yeast. For each probe, the enrichment was determined relative to that of total H3. This value was then compared to wild-type untreated cells. Yeast strains used were (C) Clhst1Δ (LRY2671), (D) ClHST1 (LRY2544) treated with 25 mM nicotinamide (NAM), or (E) ClHST1 (LRY2544) treated with 10 µM trichostatin A (TSA).

Figure 3. ClHst1 co-localizes with ClNet1 at the rDNA repeats.

The association of ClNet1 and ClHst1 across an rDNA repeat was examined by chromatin IP. Anti-HA and anti-myc antibodies were used to immunoprecipitate proteins from ClHST1 ClNET1 (LRY2826) or ClHST1-MYC ClNET1-HA (LRY2917) yeast. The relative enrichment of each probe compared to the control locus, ClPRI2, is indicated.

Figure 4. Telomere repeats are located at ends of chromosomes.

(A) Features at the ends of five supercontigs are illustrated. Telomere repeats are represented by gray boxes. Restriction sites are marked with their distance from the telomere repeat, and the distance between sites is indicated in italics. Probes used for Southern analysis are shown as black rectangles. (B) Digested genomic DNA (LRY2826) was separated electrophoretically, transferred to membranes, and hybridized with the indicated probes located near telomere repeat sequences.

Figure 5. ClHst1 is not associated with telomeres.

(A) Diagrams represent the genetic features of 10 kbp at ends of contigs bearing telomere repeat sequences. The positions of genes (black arrows), telomere repeat sequences (gray boxes), and PCR products used for chromatin IP analysis (red boxes) are shown. Two dubious ORFs have been omitted. CLUG_04101 overlaps the telomere repeat sequence of contig 5L, and CLUG_04103 is antisense to CLUG_4102. (B) The association of ClHst1 with telomeres was examined by chromatin IP. Anti-myc antibody was used to immunoprecipitate proteins from ClHST1 (LRY2826) or ClHST1-MYC (LRY2858) yeast. The relative enrichment of each probe compared to the control locus, ClPRI2, is indicated. (C) The association of ClRap1 with telomeres was examined by chromatin IP. Anti-myc antibody was used to immunoprecipitate proteins from ClRAP1 (LRY2826) or ClRAP1-MYC (LRY2859, LRY2860) yeast. The relative enrichment of each probe compared to the control locus, ClPRI2, is indicated. (D–F) The change in acetylation of histones H3 and H4 at the telomeres was examined in deacetylase-deficient yeast relative to wild-type. Tetra-acetylated H4, H4-K16Ac, H3-K9Ac, or H3-K56Ac was immunoprecipitated from yeast. For each immunoprecipitation, the enrichment was determined relative to that of total H3, and this value was compared to the value for wild-type untreated cells. Yeast strains used were (D) Clhst1Δ (LRY2671), (E) ClHST1 (LRY2544) treated with 25 mM nicotinamide (NAM), or (F) ClHST1 (LRY2544) treated with 10 µM trichostatin A (TSA).

Figure 6. Expression of subtelomeric genes did not change in the absence of ClHst1.

Average normalized read counts are plotted for all genes within 40,000 base pairs of a telomere repeat sequence. These genes are located on contigs 2L, 4R, 5L, 7L, and 7R. For each gene, the distance of the start codon from the telomere repeat sequence is plotted on the x-axis. The dashed rectangle in the upper panel is expanded in the lower panel. Strains examined were wild-type (LRY2544; open boxes) and hst1Δ (LRY2623 and LRY2671; shaded diamonds).

Figure 7. ClHst1 has a limited and indirect impact on gene expression.

(A) Expression levels were assessed for genes identified by RNA-seq as induced in Clhst1Δ strains. Expression was assessed by quantitative RT-PCR in ClHST1 (LRY2544, white) and four independently constructed Clhst1Δ strains (LRY2623, light gray; LRY2671; medium gray; LRY2672, dark gray; LRY2673, black). Levels of mRNA for each gene were first normalized to ClPRI2 (CLUG_00368) and then expressed relative to the WT ClHST1 strain. (B) Expression of genes repressed in hst1Δ strains was measured as in part A. For fold repression, the inverse of the fold expression was calculated. (C) The association of ClHst1 with promoters of candidate genes was examined by chromatin IP. Anti-myc antibody was used to immunoprecipitate proteins from ClHST1 (LRY2826) or ClHST1-MYC (LRY2858) yeast.

Figure 8. The ancestral SIR2 gene may have duplicated and subfunctionalized prior to loss of one paralog.

(A) A model is shown depicting the hypothesized evolutionary history of the SIR2 gene within the CTG clade. An early duplication followed by subfunctionalization partitioned the ancestral functions. The product of one gene (HST1) acted at the rDNA locus, whereas the other (SIR2) possessed different functions, perhaps including a role at telomeres. In some lineages, including C. lusitaniae, only the HST1 gene was retained. (B) A species tree indicates with an asterisk (*) organisms whose genomes encode the additional SIR2 gene. The maximum likelihood tree is based on .