PMC

RTS1 is a high-copy suppressor of gcn5∆. (A) A screen for high-copy suppressors of gcn5∆ sas3-C357Y, P375A (LPY13321) lethality identified multiple independent isolates of RTS1. Suppression by IOC2 has been characterized (). (B) Several of the independently isolated transformants identified in the dosage suppression screen that rescue gcn5∆ sas3-C357Y, P375A lethality at high temperature are shown. (C) Unlike IOC2, RTS1 suppression of gcn5∆ (LPY10182) temperature sensitivity is independent of SAS3. Strains were grown on Leu⁻ media for 3 days at indicated temperatures prior to imaging.

HTB1-T91 dynamic modification is essential for growth. (A) Phosphomimetic mutation of T91 is lethal in gcn5∆ and causes slow growth in GCN5 cells. Site-directed mutagenesis was used to mutate htb1-T91A to D/E to generate pLP2689 and 2770 and transformed into LPY16434 and LPY14461 histone shuffle strains. (B) RTS1 overexpression does not rescue slow growth and lethality caused by htb1-T91D/E phosphomimetic mutations. Transformants were plated onto His⁻ Ura⁻ or His⁻ Leu⁻ Ura⁻ (growth) and 5-FOA to select against the WT histone plasmid as shown and grown for 3 days at 30°.

Directed “histome” screen identifies two H2B residues required for RTS1 suppression in gcn5∆. (A) Phosphorylatable S and T residues (blue), known phosphorylated residues (yellow), and select Y residues (underlined) were tested for function in RTS1 suppression in the gcn5∆ histone shuffle strains (LPY16290 hht1- hhf1∆:kanMX hht2-hhf2∆::kanMX hta2-htb2∆::HPH gcn5∆::kanMX, LPY16434 hht1-hhf1∆::kanMX hta1-htb1∆::natMX hta2-htb2∆::HPH gcn5∆::kanMX). (B) Nucleosome structure with H2B chains in blue. The Y40 and T91 residues of interest are highlighted in yellow and indicated with yellow arrows. (C) H2B (encoded by HTB1) residues Y40 and T91 (underlined in A) are required for RTS1 rescue of gcn5∆ temperature sensitivity. Mutant htb1 plasmids (pLP2482, htb1-T91A and pLP3250, htb1-Y40F) were transformed into LPY16434 and LPY14461 (hht1-hhf1∆::kanMX hta1-htb1∆::natMX hta2-htb2∆::HPH) containing RTS1 or vector control high-copy plasmids. 5-FOA was used to select against the WT histone plasmid from these strains before plating onto His⁻ Leu⁻ media as shown and grown for 3 days at the indicated temperature. (D) H2B-Y40 and T91 are required for RTS1 rescue of DNA damage sensitivity (0.05 M HU, 0.02% MMS). H2B-T91 is also required for RTS1 rescue of sensitivity to microtubule destabilization by NOC (2 µg/ml). Fresh histone mutant transformants were generated as above before plating and grown for 3 days at 30°.

RTS1 overexpression suppresses diverse gcn5∆ phenotypes. (A) RTS1 overexpression rescues gcn5∆ growth on Ura⁻ plates containing DNA damage inducers HU (0.1 M) and MMS (0.03%). (B) RTS1 overexpression partially restores growth of gcn5∆ in the presence of microtubule destabilizing nocodazole (NOC) (2 µg/ml). (C) Growth on nonfermentable carbon sources glycerol and ethanol improves in gcn5∆ with RTS1 overexpression. Cells were grown at 30° for 3 days (A and B) or 2 days (C). LPY5 (WT) and LPY10182 (gcn5∆::kanMX) transformants are shown. (D) RTS1 overexpression suppresses gcn5∆ slow progression into S phase. The bar1∆ (LPY11975) and gcn5∆ bar1∆ (LPY21272) transformants were arrested in G1 with pheromone α-factor at 30° for 90 and 120 min, respectively, and released into fresh medium for immediate collection (T₀) or to grow at 30° or 37° for 30 min prior to collection for flow cytometry. All data shown are representative of at least three independent experiments using independent transformants. See Figure S1 for additional analysis and summary.

RTS1 overexpression restores histone gene expression in asynchronous gcn5∆ populations. (A) RT-qPCR analysis of histone mRNA reveals lower histone gene expression in gcn5∆ at 37° relative to the RNA Pol III transcript, SCR1, except for HTA1, which revealed no significant change. Gene expression increased with RTS1 overexpression, except for HHT2. Error bars indicate standard deviation from three independent experiments using fresh 2µ vector (pLP136) and RTS1 (pLP2462) transformants derived from LPY5 (WT) and LPY10182 (gcn5∆::kanMX). Asterisks indicate P-values <0.05 by paired Student’s t-test: a single asterisk indicates significant difference compared to WT + vector and a double asterisk, compared to gcn5∆ + vector. All normalized averages and significant P-values are listed in Table S4. Each histone gene pair locus is diagrammed below the corresponding histograms. Oligonucleotide sequences for qPCR amplification of all histone genes except for HHF2 were originally published in . (B) RTS1 overexpression restores HTB1 expression upon release from α-factor arrest. Early log-phase transformants of WT (LPY11975) and gcn5∆ (21272) were arrested with α-factor at 30° and samples were taken every 20 min upon release to grow at 37° for RT-qPCR analysis. HTA1 induction is similar, but HTB1 induction is reduced in between gcn5∆ and WT. This difference is suppressed when RTS1 is overexpressed.

There is a functional interaction between RTS1 and SAGA. (A) Molecular models of three Gcn5-containing complexes in budding yeast (). (B) Genetic analysis of rts1∆ (LPY14653) with SAGA, ADA, and SLIK/SALSA subunits highlighted by black outlines. The rts1∆ strain was transformed with RTS1-URA3 (pLP2462) and crossed to generate the double mutants shown. Strains were plated onto Ura⁻ (growth) and 5-FOA to select against the RTS1 plasmid and grown for 3 days at 30°. Synthetic lethality is observed with rts1∆ gcn5∆ (LPY15178) and rts1∆ spt20∆ (LPY17484) but not with loss of ADA subunits encoded by AHC1 (LPY18424), AHC2 (LPY18495), or the SLIK/SALSA-specific RTG2 (LPY20692). Deletion of the SAGA-specific SPT8 (LPY21898), encoding a TBP-binding subunit, and UBP8 (LPY21081), encoding the deubiquitinase subunit, are also viable in rts1∆. (C) RTS1 improves gcn5∆ (LPY10182), spt20∆ (LPY16914), and spt8∆ (LPY6487) growth at high temperature and in DNA damaging conditions in all strains shown, including ubp8∆ (LPY8240). Strains were transformed with vector or RTS1 and plated onto Ura⁻ plates with and without MMS and grown for 3 days at 30° or 37.5° as indicated. (D) RTS1 overexpression results in increases in H3-K9, K14ac. Log-phase cells grown at indicated temperature were collected for protein lysate preparation; shown is a representative immunoblot. (E) Quantification of relative H3-K9, K14ac signal from three biological replicates is shown. Error bars indicate standard deviation. Asterisks indicate P-values <0.05 by paired Student’s t-test: a single asterisk indicates significant difference compared to WT + vector and a double asterisk, compared to gcn5∆ + vector. All normalized averages and significant P-values are listed in Table S4.

Loss of PP2A^Rts1 function impairs growth and RTS1 rescue in gcn5∆. (A) Cartoon representations of two yeast PP2A complexes are adapted from crystal structures (). (B) Only RTS1 overexpression, not CDC55, can suppress gcn5∆ temperature sensitivity. LPY10182 (gcn5∆::kanMX) was transformed with 2µ GCN5 (pLP1524), empty vector (pLP135), RTS1 (pLP2197), and CDC55 (pLP2330). Fresh transformants were grown overnight and plated onto Leu⁻ medium prewarmed to the indicated temperature, and grown for 4 days. (C) Loss of PP2A^Rts1 is lethal in gcn5∆. WT (LPY5), gcn5∆ rts1∆ (LPY15178), gcn5∆ cdc55∆ (LPY15178), gcn5∆ pph21∆ (LPY15296), gcn5∆ pph22∆ (LPY14692), gcn5∆ pph21∆ pph22∆ (LPY20694), and gcn5∆ tpd3∆ (LPY15416) are all shown transformed with pLP1640 (GCN5). The gcn5∆ pph21∆ pph22∆ mutant (LPY20694) was transformed with pLP2997 (PPH22). Cells were plated onto Ura⁻ and 5-FOA to select against the GCN5 or PPH22 CEN plasmid and grown for 3 days at 30°. (D) Treatment with the phosphoprotein phosphatase inhibitor OKA reduces RTS1 suppression of gcn5∆ temperature sensitivity. Log-phase cells were treated with 10 µM OKA or DMSO (control) for 1 hr prior to plating onto Ura⁻ plates prewarmed to 30° or 36.5°. OKA treatment caused temperature sensitivity in WT cells when grown at 37° (not shown), so a slightly lower temperature was used for this assay. Cells were grown for 3 days prior to counting CFUs on each plate. Survival at high temperature was determined by calculating the ratio of CFUs at 36.5° to 30° followed by normalization to WT + vector controls, where n indicates total CFUs counted in three experiments for each transformant, and the error bars show standard deviation. See Table S4 for all normalized relative survival ratios.