Dear Editor,
Heterochromatin, the highly-condensed nucleosome arrays in chromosomes, plays an important role in regulating multiple cellular events including transcriptional silencing and chromosome segregation (
Allis and Jenuwein, 2016). Aberrant heterochromatin formation is linked to tumor progression, as well as other severe physiological disorders (
Hahn et al., 2010). However, the mechanism by which factors contribute to heterochromatin formation remains elusive.
In
Saccharomyces cerevisiae, heterochromatin forms at two hidden mating-type loci (
HM loci, namely
HMR and
HML), telomeres and rRNA-encoding DNA (rDNA) region (
Grunstein and Gasser, 2013). Besides the four silent information regulators, Sir1p–Sir4p, it was reported that histone chaperones are also required for heterochromatin formation. For instance, H3–H4 chaperones chromatin assembly factor 1 (CAF-1), anti-silencing function protein 1 (Asf1p), histone transcription regulator 1 (Hir1p), and regulator of Ty1 transposition protein 106 (Rtt106p) are important for heterochromatin silencing (
Enomoto and Berman, 1998;
Huang et al., 2005;
Huang et al., 2007;
Sharp et al., 2001). Genetic studies revealed two parallel pathways among these histone chaperones: Asf1p, Hir1p, and Rtt106p seem to mediate a genetically distinguishable silencing pathway from CAF-1. Interestingly, cells deficient in both pathways are only partially defective in heterochromatin silencing, which suggests other factors might also be involved (
Huang et al., 2007).
Histone chaperone complex FACT (facilitates chromatin transcription/transaction), consisting of two essential subunits, Spt16p and Pob3p (SSRP1 in humans), plays a critical role in transcription, replication, and other chromatin based processes in eukaryotes (
Formosa, 2012). While both Spt16p and Pob3p are essential in
S.
cerevisiae, Pob3p is dispensable in
Schizosaccharomyces pombe, and contributes to heterochromatin silencing (
Lejeune et al., 2007). However, whether FACT also has a role in heterochromatin silencing in
S.
cerevisiae remains to be determined. Here, we report that Spt16p functions redundantly with CAF-1 and Rtt106p in heterochromatin silencing and is required for heterochromatin formation at both mating-type loci and telomeres in budding yeast.
To identify other histone chaperones that might function in heterochromatin silencing, we performed a small-scale candidate gene screen using a GFP-based reporter gene silencing assay (Fig. 1A). The expression of GFP gene inserted at the silent
HMR locus (
hmr::
GFP) was used to monitor the degree of silencing by quantifying the percentage of yeast cells expressing GFP (Fig. 1A). Yeast cells harboring double deletion of
CAC1 (the large subunit of histone chaperone CAF-1) and
RTT106 was made as a query strain (Fig. 1B). Consistent with prior studies, 47.5% of
cac1Δ
rtt106Δ cells lost GFP silencing while only 0.4% wild-type cells expressed GFP, likely from the background. As a positive control,
sir3Δ almost completely abolishes cells’ ability in heterochromatin silencing (99.8%). Interestingly,
spt16-m, a newly identified Spt16p mutant allele in our group, dramatically exacerbates the phenotype of
cac1Δ
rtt106Δ (77.3%) (Fig. 1C). The
spt16-m allele contains two amino acids substitution (
K692AR693A) in its tandem pleckstrin homology (PH) domain and displays a subtle defect in transcription initiation but a substantial defect in DNA replication-coupled nucleosome assembly (
Yang et al., 2016). Thus, this result indicates that Spt16p may also be involved in heterochromatin silencing.
In yeast, Sir1p is involved in the establishment of heterochromatin silencing at the
HM loci (
Rusche et al., 2003). To dissect the role of Spt16p in heterochromatin silencing in yeast, we first examined the role of Spt16p at the
HMR locus in combination with
sir1Δ (Fig. 1D and 1E). Consistent with the published results,
sir1Δ causes a mild silencing defect (3.47%). GFP silencing appears to be normal in
spt16-m cells (0.4%). However,
spt16-m strongly exacerbates the silencing defect of
sir1Δ (40.8%) (Fig. 1D). Similar synergistic behavior between
SPT16 and
SIR1 was also observed in the background of
cac1Δ
rtt106Δ (Fig. S1). These results suggest that Spt16p cooperates with Sir1p, CAF-1, and Rtt106 in heterochromatin silencing at the
HMR locus.
To quantify the transcriptional activity at the
HMR locus, we used reverse transcription quantitative PCR (RT-qPCR) to analyze the expression level of
A1 gene that resides within
HMR region (Fig. 1F). Consistent with the previous observations,
A1 gene is silenced in wild-type cells, but is highly transcribed in
sir3Δ strain. The expression level of
A1 in
spt16-m sir1Δ double mutant cells is dramatically higher than that in either
spt16-m or sir1Δ single mutant cells. Moreover, we also analyzed the expression level of a gene
YFR057W near the telomere of chromosome VI (Fig. 1G). Although no aberrant expression was detected in
spt16-m strain, the transcription level has been apparently elevated in
spt16-m sir1Δ strain, indicating a potential role of Spt16p and Sir1p in telomere silencing. While Sir1p is dispensable for telomeric silencing at the modified telomere (
Aparicio et al., 1991), it has also been suggested that
SIR1 contributes to silencing at natural telomeres which contains intact subtelomeric repeat sequences (
Pryde and Louis, 1999). Together, our data suggested that the cooperation between Spt16p and Sir1p might be required for efficient silencing at both
HM loci and telomeres.
In yeast, Sir2p is a nicotinamide adenine dinucleotide (NAD) dependent histone deacetylase, which deacetylates nearby histones to facilitate binding of Sir3p and Sir4p. Subsequently, more Sir2p is recruited through direct interaction with Sir3p and Sir4p to spread Sir proteins (
Grunstein and Gasser, 2013). A reduction of Sir2p and Sir3p proteins over heterochromatin region was reported in
cac1Δ
rtt106Δ strain (
Huang et al., 2007). To examine whether Spt16 affects the binding of Sir2–4p to different silent regions, we performed chromatin immunoprecipitation (ChIP) assay using antibodies against Sir2–4p (Fig. 2A–D). As reported, we found that all the Sir proteins are highly enriched at heterochromatin region rather than euchromatin region in wild-type cells, while
sir3Δ completely abolishes the Sir proteins binding over the heterochromatin region. Moreover, while
spt16-m mutation led to a mildly reduction of Sir2–4p binding at heterochromatin region, this mutation in combination with
cac1Δ
rtt106Δ led to a dramatic reduction of Sir2–4p binding comparing with either single or double mutant cells (Fig. 2A). This data suggests that Spt16p might function in parallel with CAF-1 and Rtt106p in regulating Sir proteins’ occupancy over heterochromatin regions. Furthermore, we found that the levels of Sir proteins drop dramatically in
spt16-m sir1Δ strain at both
HMR locus and telomere region (Fig. 2B and 2C). This dramatic defect is not due to alteration in Spt16p expression or chromatin occupancy in the mutant cells (Fig. S2). To further characterize the defect of heterochromatin formation in
spt16-m mutant cells, we also analyzed the distribution of all three Sir proteins across
HMR locus (Fig. 2D) and the histone modifications at
HMR locus (Fig. 2E). Silencers, including
E and
I elements, are genomic regions where heterochromatin formation initiates (
Grunstein and Gasser, 2013). In agreement with the silencer function of
HMR-E and
HMR-I that recruits Sir proteins and promotes formation of heterochromatin inwards toward the mating-type genes, an “M” shape distribution of Sir proteins was detected in wild-type cells. Compared with wild-type cells, the amount of Sir proteins is comparable in
sir1Δ or
spt16-m cells and the similar “M” pattern was observed, indicating that the silencer function is largely maintained in either single mutant strains. By contrast, the M pattern is lost in
spt16-m sir1Δ strain (Fig. 2D). Acetylation of H4 lysine 16 (H4K16Ac) is low at heterochromatin region to facilitate the Sir proteins binding but is high at euchromatin regions adjacent to heterochromatin to restrict heterochromatin spreading into euchromatin (
Grunstein and Gasser, 2013). We found that the level of H4K16Ac at the heterochromatin region is more dramatically enhanced in
spt16-m sir1Δ double mutant cells compared to either single mutant alone (Fig. 2E, left panel). No obvious alterations was observed in H3K56Ac, an acetylation mark of newly synthesized H3 (
Li et al., 2008), which was analyzed by ChIP assays side by side (Fig. 2E, right panel). Together, these data further supports the idea that Spt16p may cooperate with Sir1p and function in regulating the binding of Sir2–4p binding to heterochromatic region.
As Sir1p is known to be required for establishment of heterochromatin (
Pillus and Rine, 1989), we speculate that Spt16p may also have a role in this stage. To test this, we used a Sir3p-induction system in
sir3Δ cells to monitor the establishment of heterochromatin at silenced mating-type region (Fig. 2F). Before induction, no Sir3p proteins are present and heterochromatin silencing is completely abolished. Therefore,
MATa cells will not respond to the mating pheromone α-factor due to lost silencing at yeast silent mating-type locus. Upon induction of Sir3p expression, heterochromatin is re-established in these cells and as a result, the cells will respond to α-factor and form shmoos. Deletion of
SIR1 leads to reduced number of shmoos after Sir3p induction. Notably, comparing with wild-type cells,
spt16-m also exhibits an apparent defect in shmoo formation after induction of Sir3p. Moreover,
spt16-m exacerbates the defect of shmoo formation of
sir1Δ (Fig. 2F). Thus, we conclude that Spt16p has a role in the establishment of heterochromatin silencing and it may function cooperatively with Sir1p during this process.
In summary, we show that histone chaperone FACT is required for the heterochromatin silencing in
S.
cerevisiae via the study of its subunit Spt16p. We find that the
spt16-m mutant aggravates the
HMR silencing defect in cells lacking both
CAC1 and
RTT106 genes. Moreover, Spt16p functions in parallel with CAF-1 and Rtt106p in regulating Sir proteins’ occupancy. Therefore, besides functioning with CAF-1 and Rtt106p during nucleosome assembly (
Yang et al., 2016), we speculate that Spt16p also cooperates with CAF-1 and Rtt106 in heterochromatin formation. Additionally, we find that Spt16p likely functions cooperatively with Sir1p in the establishment of heterochromatin silencing at both telomeres and cryptic mating-type loci. Pob3p has been previously reported to be important for heterochromatin silencing in
S.
pombe (
Lejeune et al., 2007). Our data suggests that Spt16p might also play a crucial role during heterochromatin silencing in
S.
cerevisiae, a highly divergent eukaryotic species from
S.
pombe. Therefore, it is likely that FACT’s role in heterochromatin silencing is conserved among eukaryotes.
PDF
(664KB)
