Splicing factor Spf30 assists exosome-mediated gene silencing in fission yeast

Abstract

Heterochromatin assembly in fission yeast relies on the processing of cognate noncoding RNAs by both the RNA interference and the exosome degradation pathways. Recent evidence indicates that splicing factors facilitate the cotranscriptional processing of centromeric transcripts into small interfering RNAs (siRNAs). In contrast, how the exosome contributes to heterochromatin assembly and whether it also relies upon splicing factors were unknown. We provide here evidence that fission yeast Spf30 is a splicing factor involved in the exosome pathway of heterochromatin silencing. Spf30 and Dis3, the main exosome RNase, colocalize at centromeric heterochromatin and euchromatic genes. At the centromeres, Dis3 helps recruiting Spf30, whose deficiency phenocopies the dis3-54 mutant: heterochromatin is impaired, as evidenced by reduced silencing and the accumulation of polyadenylated centromeric transcripts, but the production of siRNAs appears to be unaffected. Consistent with a direct role, Spf30 binds centromeric transcripts and locates at the centromeres in an RNA-dependent manner. We propose that Spf30, bound to nascent centromeric transcripts, perhaps with other splicing factors, assists their processing by the exosome. Splicing factor intercession may thus be a common feature of gene silencing pathways.

FIG. 1.

The ssl38 gene, synthetically lethal with swi6 when mutated, encodes a protein structurally similar to the human splicing factor SPF30. (A to C) The ssl38-ts mutation is synthetically lethal with the swi6Δ mutation, thermosensitive for growth, and hypersensitive to thiabendazole (TBZ). (A) Cells cultured in the absence of thiamine were serially diluted and spotted on indicated synthetic media. Thiamine represses nmt-swi6 transcription. (B and C) Cells were serially diluted, spotted onto YES+A (B) or YES+A plus TBZ (C). (D) Ssl38 is a nuclear protein. Exponentially growing ssl38-GFP cells were treated with Hoechst 33342 to stain DNA and then directly observed under a fluorescence microscope. Bar, 5 μm. (E) Ssl38 is similar to the Smndc1/Spf30 splicing factor. Region of sequence identity between Ssl38 and Spf30 orthologs, determined by using BLASTP, are indicated as shaded areas with the percentages underneath.

FIG. 2.

Ssl38 is required for efficient splicing. (A) Ssl38 coimmunoprecipitates with Prp1. Soluble protein extracts were subjected to immunoprecipitation with anti-GFP or anti-myc antibodies. Inputs (lanes I), immunoprecipitated (lanes IP) fractions, and proteins washed away (lanes FT) were analyzed by Western blotting with anti-myc or anti-GFP antibodies. (B) Unspliced transcripts accumulate in ssl38-ts. The scheme of the RT-PCR assay used to assess splicing efficiency is shown. Total RNA extracted from indicated cells was reverse transcribed in the presence (+RT) or absence (−RT) of reverse transcriptase. Complementary DNAs (cDNAs) were PCR amplified and in-gel quantified. Unspliced, upper band; spliced, lower band. (C) The steady-state level of act1 transcripts or 25S RNA is not overtly affected by ssl38-ts. Same procedure as in panel B, except that cDNAs were quantified by real-time PCR. The indicated values correspond to averages and mean deviations (m.d.) calculated from three (25S RNA) or two (act1) independent samples.

FIG. 3.

Chromosome segregation is defective in spf30-38 cells. (A) Spf30 deficiency affects chromosome segregation during mitosis. Cells fixed at 25, 32, or 37°C were processed for immunofluorescence against α-tubulin (Tubulin). DNA was stained with DAPI (4′,6′-diamidino-2-phenylindole), and chromosome segregation was examined in late-anaphase cells (mitotic spindle length, >5 μm; n > 100). Mutant cells exhibited three distinct phenotypes. (a) Lagging chromosomes: DAPI signals away from one spindle pole. (b) Uneven chromosome distribution: unbalanced DAPI signals at the poles. (c) Absence of or sustained delay in chromosome movement: DAPI signals stuck around the spindle center. Bar, 5 μm. (B) spf30-38 cells exhibit a Mad2-dependent delay in early mitosis. Fixed cells were processed as in panel A and prophase/prometaphase figures (mitotic spindle length, <2 μm) scored (n = 400 to 1,000). **, the difference with the wild-type (wt) control is statistically highly significant (P < 0.01, as determined by the chi-square test); *, borderline of statistical significance (P < 0.05); •, statistically not significant. (C) The absence of Mad2 increases the rate of missegregation events in spf30-38 cells. The same cells as in panel B were examined for chromosome segregation defects in late anaphase (n > 100). (D) spf30-38 alleviates silencing at the central domain of centromere 1. Centromere 1 with the central domain (cnt1) surrounded by outer repeats left and right (otr1), made of dg and dh elements, is represented, and the location of the ura4⁺ marker gene is indicated. Strains whose sole ura4⁺ gene was inserted within cnt1 (cnt1::ura4⁺), were grown at 25 or 32°C or shifted to 37°C for 6 h. A total of 500 ng of total RNA was reverse transcribed by using oligo(dT) primers, in the presence (+RT) or absence (−RT) of reverse transcriptase. ura4⁺ and 25S cDNAs were quantified by real-time PCR, and their ratios were determined. The indicated values correspond to averages and m.d. calculated based on two PCRs.

FIG. 4.

Spf30 preserves the integrity centromeric heterochromatin. (A) spf30-38 alleviates silencing of the otr1::ura4⁺ marker gene. The location of the ura4⁺ marker gene within centromere 1 is indicated. Total RNA was reverse transcribed in the presence (+RT) or absence (−RT) of reverse transcriptase, and ura4⁺ and 25S cDNAs were quantified by real-time PCR. Indicated values correspond to averages and m.d. calculated from duplicated samples. (B and C) Endogenous centromeric transcripts accumulate in spf30-38 cells. (B) Total RNA was reverse transcribed using a primer complementary to the forward (cen-for) or the reverse (cen-rev) strand of dh repeat, in the presence (+RT) or absence (−RT) of reverse transcriptase. act1 transcripts were used as control. The indicated values are averages and m.d. calculated from two independent experiments. (C) Centromeric transcripts were detected by Northern blotting, and blots were reprobed with act1 for a loading control. (D and E) Decreased amounts of Swi6 and H3K9-me at the centromeric outer repeats in spf30-38 cells. The Swi6 and H3K9-me levels at the outer repeats (cen-dg and cen-dh) were assessed by ChIP. The indicated values are averages and m.d. calculated from two qPCRs.

FIG. 5.

Spf30 functions in the exosome-mediated silencing pathway. (A) Spf30 deficiency alleviates silencing of a subtelomeric tel1L::ura4⁺ marker gene. Cells of indicated genotypes were serially diluted and spotted on nonselective (N/S), selective (−Ura), and counterselective (FOA) synthetic medium. 5-Fluoroorotic acid (5-FOA) is toxic to Ura⁺ cells. Plates were incubated at 32°C. (B) spf30-38 affects silencing at the mating-type locus. A colony color assay for ade6⁺ silencing when it is inserted within the mating-type locus (mat3::ade6⁺) was performed. The indicated strains were grown at 32°C on YES containing a low amount of adenine. (C) Spf30 deficiency does not overtly affect the production of siRNAs. Centromeric siRNAs were detected by Northern blotting, and blots were reprobed with snR58 for a loading control. (D) Polyadenylated cen-dh transcripts accumulate when Spf30 is deficient. The scheme of the polyadenylation assay is shown. Total RNA extracted from indicated strains grown at 30.5°C was reverse transcribed with the poly(dT) anchor primer and PCR amplified by using cen-dh strand-specific and anchor primers. (E) Dis3 overexpression partly suppresses spf30-38 silencing defect at 32°C. Strains transformed with the pCD174 episomal multicopy plasmid (pDis3) (26) or with an empty vector (pLEU2) were serially diluted and spotted onto selective media. (F) The steady-state amount of Dis3 protein is not altered in the spf30-38 mutant. Soluble protein extracts prepared from wild-type (wt) and spf30-38 cells expressing Dis3-HA and grown at 25 or 32°C or shifted at 37°C for 6 h were subjected to Western blotting with monoclonal 12CA5 anti-HA antibodies. Equal loading was verified by Ponceau staining. (G) Cryptic unstable transcripts accumulate in spf30-38 cells. Total RNA from indicated strains was reverse transcribed by using random hexamers, and cDNAs were quantified by real-time PCR.

FIG. 6.

Spf30, recruited to centromeric heterochromatin in a Dis3-dependent manner, binds nascent centromeric transcripts. (A) The chromosomal association of Dis3 is unaffected in spf30-38 cells. Dis3 levels at indicated chromosomal sites were assessed by ChIP. The indicated values correspond to averages and m.d. calculated from four (wild type) and six (spf30-38) experiments. (B) Spf30 colocalizes with Dis3 at centromeres and euchromatic genes and is reduced at the pericentric heterochromatin in dis3-54 cells. Spf30 binding was assessed by ChIP. The indicated values correspond to averages and m.d. calculated from four experiments. (C) Spf30 binds RNA species, among which were centromeric transcripts. The indicated strains were processed for RNA-IP by using polyclonal anti-GFP antibodies. RNA recovered in the input and immunoprecipitated fractions was reverse transcribed in the presence (+) or absence (−) of reverse transcriptase. cDNAs were quantified by real-time PCR. The indicated values correspond to averages and m.d. calculated from two independent experiments with two reverse transcriptions per experiment. (D) Spf30 associates with chromatin in an RNA-dependent manner. Chromatin extracts prepared for ChIP were treated with RNase or RNase plus inhibitor before immunoprecipitation. Enrichments in Spf30-GFP immunoprecipitated fractions were calculated as a ratio, i.e., the quantity in the immunoprecipitate/the quantity in the input, normalized to the ratio of the corresponding untagged control. The indicated values correspond to averages and m.d. calculated from two experiments.