Suppressor mutations that make the essential transcription factor Spn1/Iws1 dispensable in Saccharomyces cerevisiae

Abstract

Spn1/Iws1 is an essential eukaryotic transcription elongation factor that is conserved from yeast to humans as an integral member of the RNA polymerase II elongation complex. Several studies have shown that Spn1 functions as a histone chaperone to control transcription, RNA splicing, genome stability, and histone modifications. However, the precise role of Spn1 is not understood, and there is little understanding of why it is essential for viability. To address these issues, we have isolated 8 suppressor mutations that bypass the essential requirement for Spn1 in Saccharomyces cerevisiae. Unexpectedly, the suppressors identify several functionally distinct complexes and activities, including the histone chaperone FACT, the histone methyltransferase Set2, the Rpd3S histone deacetylase complex, the histone acetyltransferase Rtt109, the nucleosome remodeler Chd1, and a member of the SAGA coactivator complex, Sgf73. The identification of these distinct groups suggests that there are multiple ways in which Spn1 bypass can occur, including changes in histone acetylation and alterations in other histone chaperones. Thus, Spn1 may function to overcome repressive chromatin by multiple mechanisms during transcription. Our results suggest that bypassing a subset of these functions allows viability in the absence of Spn1.

Keywords: Iws1; Spn1; chromatin; histone chaperone; transcription.

Fig. 1.

The isolation and analysis of Spn1 bypass suppressors. a) A schematic showing the selection and subsequent screen used to isolate Spn1 bypass suppressors. Briefly, spn1Δ yeast strains that also have deletions of HIS3 and URA3 were grown. These strains are viable because they contain a wild-type copy of the SPN1 gene in a plasmid, which also has wild-type copies of URA3 and HIS3. Suppressors were selected on medium containing 5-FOA, which counterselects against cells that have URA3. To screen for the suppressor strains that lost the plasmid, rather than those that acquired a ura3 mutation, the 5-FOA-resistant candidates were screened for those that were also His⁻. We isolated 105 independent spn1Δ suppressors. Whole-genome sequencing of the 30 strongest mutants, followed by gene replacements, identified 8 genes with spn1Δ suppressors. b) Dilution spot tests of spn1Δ suppressors. Shown are 5-FOA (suppression) and YPD (permissive) plates after 7 days at 30°C. c) Suppression of spn1Δ by set2Δ and by H3K36A. Plates are shown after 5 days at 30°C. d) The suppressors of spn1Δ also suppress Spn1 depletion mediated by SPN1-AID. Plates are shown after 3 days at 30°C. Plates labeled NAA contain 500 µM NAA to enable Spn1 depletion. Here we tested suppression of pob3-E154K, while in (a) we tested pob3-E171K. Both mutations target the dimerization domain of Pob3 as previously described (Viktorovskaya et al. 2021). e) Analysis of suppression by double mutants, shown after 3 days of incubation at 30°C.

Fig. 2.

ChIP-seq analysis of histone H4 acetylation. a) Comparison of histone H4 acetylation levels in strains with and without Spn1 depletion. Top: metagene plots showing the average level of histone H4 acetylation normalized to total levels of histone H4 for 3,087 non-overlapping coding genes in Spn1 non-depleted and depleted conditions. The solid line and the shading are the median and interquartile range of each sample. Bottom: a heatmap showing the same set of genes aligned by the transcription start site (TSS) and arranged by transcript length. The white dotted line on the right represents the position of the cleavage and polyadenylation site. b) As in (a) except comparing Spn1 non-depleted to set2Δ Spn1 non-depleted. c) As in (a) except comparing Spn1 non-depleted to set2Δ Spn1-depleted. d) As in A except comparing set2Δ Spn1 non-depleted to set2Δ Spn1-depleted. e) Single gene examples of ChIP-seq analysis. The H4 acetylation levels normalized to total H4 levels are shown for 3 genes, MYO5, EGT2, and NHA1. These 3 genes showed notable differences in H4ac between the Spn1 depleted and non-depleted conditions. They do not have significant changes in mRNA levels after Spn1 depletion (Reim et al. 2020).

Fig. 3.

Spn1 bypass by set2Δ requires histone acetylation. a) Dilution spot tests to examine whether Gcn5 is required for Spn1 bypass by set2Δ. Plates are shown after 3 days at 30°C. b) Dilution spot tests to examine whether Esa1 is required for Spn1 bypass by set2Δ. Plates are shown after 2 days of incubation at 34°C.

Fig. 4.

Epistasis tests between Spn1 bypass suppressors and gcn5Δ. The plates shown were incubated at 30°C for 3 days.

Fig. 5.

Genetic analysis to understand rtt109Δ bypass of Spn1. a) Dilution spot tests to examine the loss of 3 different histone acetyltransferases for Spn1 bypass. Plates are shown after 3 days at 30°C. b) The top panel shows a schematic with the spn1Δ rtt109Δ strain that contains 2 plasmids, a URA3-labeled plasmid with SPN1 and a TRP1-labeled plasmid with one of 3 options for rtt109: RTT109, rtt109-D89A, and no additional gene or empty plasmid (vector). These strains were tested for growth on 5-FOA to assay the phenotype of cells that have lost the URA3 SPN1 plasmid. The bottom panel shows the dilution spot tests of an rtt109 catalytically dead mutant. Plates are shown after 3 days at 30°C. c) Dilution spot tests to analyze mutants that impair different aspects of Rtt109 activity. Plates are shown after 3 days at 30°C.