Contribution of Trf4/5 and the nuclear exosome to genome stability through regulation of histone mRNA levels in Saccharomyces cerevisiae

Abstract

Balanced levels of histones are crucial for chromosome stability, and one major component of this control regulates histone mRNA amounts. The Saccharomyces cerevisiae poly(A) polymerases Trf4 and Trf5 are involved in a quality control mechanism that mediates polyadenylation and consequent degradation of various RNA species by the nuclear exosome. None of the known RNA targets, however, explains the fact that trf mutants have specific cell cycle defects consistent with a role in maintaining genome stability. Here, we investigate the role of Trf4/5 in regulation of histone mRNA levels. We show that loss of Trf4 and Trf5, or of Rrp6, a component of the nuclear exosome, results in elevated levels of transcripts encoding DNA replication-dependent histones. Suggesting that increased histone levels account for the phenotypes of trf mutants, we find that TRF4 shows synthetic genetic interactions with genes that negatively regulate histone levels, including RAD53. Moreover, synthetic lethality of trf4Delta rad53Delta is rescued by reducing histone levels whereas overproduction of histones is deleterious to trf's and rrp6Delta mutants. These results identify TRF4, TRF5, and RRP6 as new players in the regulation of histone mRNA levels in yeast. To our knowledge, the histone transcripts are the first mRNAs that are upregulated in Trf mutants.

Figure 1.—

Characterization of the S-phase progression in trf4/5 mutants and determination of endogenous levels of expression of Trf4 and Trf5. (A) Cells from wild type, trf4Δ (AC1946), and trf5Δ (AC1947) were released from α-factor arrest at 30°. Samples were collected each 15 min and analyzed by flow cytometry. (B) Cells from wild type and trf4-ts trf5Δ (AC1968) were arrested with α-factor at 37° as described in materials and methods. G1-arrested cells were released into the cell cycle at 37° into YPD media (columns 1 and 2) or into YPD containing 200 mm hydroxyurea (HU). Samples were collected at the indicated time points and processed for flow cytometry. Note the slow S-phase progression of trf4-ts trf5Δ (compare columns 1 and 2). (C) Cells from HA-tagged versions of Trf4 and Trf5 were arrested in G1 with α-factor, in S phase with hydroxyurea (HU), and at G2/M with nocodazole (Noc). Equivalent numbers of cells were lysed and whole-cell extracts (WCE) were analyzed on 8% SDS–PAGE gels as described in materials and methods. The levels of HA-Trf4 and HA-Trf5 were detected by anti-HA Western blot. Anti-PSTAIRE was used as loading control. The indicated ratio is calculated as intensity of anti-HA/anti-PSTAIRE signals as determined by densitometry and normalized to 1 for G1-arrested cells. (D) The steady-state levels of Trf4 and Trf5 were determined by analysis of WCE of asynchronous exponential growing cells as in C. The indicated ratio is calculated as intensity of anti-HA/anti-PSTAIRE and normalized to 1 for HA-Trf4.

Figure 2.—

Mutations in TRF4 and TRF5 do not lead to defects in Rad53 phosphorylation. (A) Cultures from the indicated genotypes were released from α-factor arrest into YPD containing 0.02% methyl methanesulfonate (MMS) at 30°. Samples were collected before release into MMS or 30 min later and analyzed by anti-Rad53 Western blot. The asterisk indicates a cross-reacting band that was used as loading control. mec1Δ (U953-61A) and rad53Δ (U960-5C) strains were used as control for Rad53 antibody. (B) Wild-type and trf4-ts trf5Δ (AC1968) cells were released from α-factor arrest at the restrictive temperature into YPD media containing 0.02% MMS where indicated and analyzed by anti-Rad53 Western blot at 30 min and 90 min after release from G1 into MMS.

Figure 3.—

Genetic interactions between TRF4 and genes involved in the DNA damage checkpoint and replication stress pathways and genes required for transcriptional repression of histones. (A–E) trf4Δ is synthetically lethal with rad53Δ and synthetically sick with a rad53 kinase-dead mutant (rad53K227A) but not with other genes involved in the checkpoint activation pathway. Strains of the indicated genotypes were crossed and tetrads dissected. In each case, circles indicate expected and recovered double mutants. Squares indicate trf4Δ spores. All dissection plates were incubated at 30° for at least 3 days. Strains crossed were (A) AC2061 and AC2123, (B) AC2055 and AC2163, (C) U953-61A and AC2115, (D) SPY40 and AC1959, and (E) AC1957 and AC1959. (F) trf4Δ asf1Δ shows reduced fitness and is inviable at 37°. Strains AC1946 and AC2122 were used for this cross. (G) trf4Δ hir1Δ shows reduced fitness. Strains AC1946 and AC2225 were crossed as described for A–F.

Figure 4.—

Quantitative analysis of histone and other transcripts in the trf4-ts trf5Δ mutant. (A) Real-time PCR analysis of the levels of several transcripts in the trf4-ts trf5Δ mutant at the restrictive temperature. Cells were released from α-factor arrest. RNA extraction was performed on cells collected at the time points after G1 release indicated in B (45 min for wild type and 65 min for the trf4-ts trf5Δ mutant, circled sections) at 37°. RNA levels were determined as described in materials and methods. Reactions were run in triplicate, and results are plotted as fold increase in expression of the mutant over wild type after normalizing to ACT1 mRNA levels. (B) Flow cytometry used to monitor the stage of S-phase progression in trf4-ts trf5Δ (AC1968) in comparison to wild type. (C) Total RNA from exponentially growing trf4-ts trf5Δ cells after a 3-hr shift to 37° was separated on a 1.2% agarose–formaldehyde gel. Northern blot analysis was performed using the HHF2 probe to detect HHF2 transcripts, as described in materials and methods. TSA1 mRNA was detected in the same blot and used as loading control because it did not vary under these conditions. The numbers beneath the blots were derived through quantification of RNA levels and indicate the ratio of mutant to wild type. (D) The same total RNA samples used for cDNA preparation for real-time PCR (A and B) were used to prepare the poly(A)+-enriched fraction and analyzed by Northern blot as in C. The numbers beneath the blots were derived through quantification of RNA levels and indicate the ratio of mutant to wild type. (E) HHF2 is polyadenylated by Pap1. Exponentially growing cells of wild type or pap1-1 (AC2207) were shifted to 37° for 1 hr to inactivate Pap1. Total RNA was analyzed by PAGE–urea Northern blots. As a loading control, ADH1 mRNA was used because it is not affected by Pap1 inactivation. (F) Introduction of TRF4 in trf4-ts trf5Δ cells lowers the levels of HHF2 transcript. Total RNA from exponentially growing cells after shift to 37° for 3 hr was separated in 6.5% PAGE–urea gels. Wild-type cells carried the pRS316 vector; trf4-ts trf5Δ carried pRS316 or pRS316TRF4 plasmids, as indicated. TSA1 was detected in the same blot and used as loading control. Note the lower loading in the middle lane. (G) Real-time PCR analysis of the levels of several transcripts in the trf4Δ mutant in asynchronous cells. Reactions were run in triplicate, and results are plotted as fold of expression of trf4Δ over wild type after normalizing to ACT1 mRNA levels. (H) Analysis of HHF2 transcript in trf4Δ and trf5Δ single mutants. Total RNA from exponentially growing cells of wild type, AC1946, and AC1947 was analyzed by PAGE–urea Northern blot.

Figure 5.—

Deletion of HHT2 and HHF2 suppresses the inviability of trf4Δ rad53Δ and histone overexpression is toxic to trf mutants. (A) Crosses were done in sml1Δ background between strains AC2192 and AC2062. trf4Δ rad53Δ sml1Δ expected spores are indicated by circles. trf4Δ rad53Δ sml1Δ hht2-hhf2Δ spores are indicated by squares. (B) Overexpression of the four histones exacerbates the ts phenotype of trf4-ts trf5Δ cells. Wild-type and AC2193 trf4-ts trf5Δ strains were transformed with control vector pRS425 or with pPK128, a 2μ plasmid expressing the four core histones (HHT1-HHF1-HTA1-HTB1, Table 2). Serial dilutions were spotted on plates lacking leucine and allowed to grow for 4 days at the indicated temperatures. (C) Overexpression of the four core histones increases the MMS and HU sensitivities of trf4Δ (AC1946). Dilution assays were performed as in B on the indicated media and at indicated temperatures. (D) Overexpression of histones is toxic to mutants in subunits of NuA3 and NuA4 histone acetyltransferase complexes. Strains from the indicated genotypes were transformed with control empty vector or pPK128 carrying the four histone genes. Transformants were streaked in −LEU plates at 37° and allowed to grow for 4 days.

Figure 6.—

The RNA-binding proteins Air1 and Air2 do not play a role in histone mRNA level regulation. HHF2 mRNA levels were analyzed in total RNA samples from asynchronous cells of indicated genotypes (strains AC2164, AC2208, AC2231, and AC2232) by PAGE–urea Northern blots. HHF2 RNA levels were normalized with respect to loading controls and the ratio of mutant to wild-type levels is shown.

Figure 7.—

rrp6Δ cells are sensitive to histone overexpression and display abnormally high levels of HHF2 mRNA. (A) Histone overexpression exacerbates the temperature-sensitive phenotype of rrp6Δ. Serial dilutions were performed as in Figure 5B, using strain AC2161 and the pRS425/pPK128 plasmids. Plates were incubated at the indicated temperatures for 3 days. (B) Levels of transcripts encoding for HHF2 were analyzed in exponentially growing cells of rrp6Δ and isogenic wild-type cells at 30°. Total RNA samples were analyzed in PAGE–urea gels and Northern blotting was performed as in Figure 4, F and H. Numbers are RNA levels normalized to wild type. (C) Cell cycle profile of HHF2 mRNA levels in rrp6Δ or wild-type cells following G1 release at 30°. RNA was collected from samples at the indicated times after α-factor release and analyzed by PAGE–urea Northern blots. PhosphorImager analysis of the Northern blot is shown in E. (D) Cell cycle progression of the cells analyzed in C was monitored by flow cytometry. (E) HHF2 mRNA levels from C were quantified using a PhosphorImager and normalized to the ADH1 control. The maximum intensity of the HHF2 band in wild type was defined as 1.0. ♦,wild type; ▪, rrp6Δ. (F) Cell cycle profile of HHF2 levels in trf4-ts trf5Δ or wild-type cells following G1 release at 37°. Analysis was carried out as in C. (G) Cell cycle progression of the samples analyzed in F was monitored by flow cytometry. (H) HHF2 mRNA levels from F were quantified using a PhosphorImager and normalized to ADH1 mRNA levels. The maximum intensity of the HHF2 band in wild type was defined as 1.0. ♦, wild type; ▪, trf4-ts trf5Δ.

Figure 8.—

Analysis of poly(A) tail length in rrp6Δ and trf4-ts trf5Δ mutants. (A) Overexpression of TRF4 does not lead to a detectable increase in the poly(A) tail length of HHF2 in wild-type or rrp6Δ backgrounds. Strains of indicated genotypes were grown in −URA raffinose medium to exponential phase. Overexpression of TRF4 and trf4DADA alleles was induced by adding 2% galactose to the medium for 3 hr. Total RNA was digested with RNase H and indicated oligonucleotides. Digested RNAs were analyzed by PAGE–urea Northern blots. (B) Analysis of poly(A) tail length of HHF2 mRNA in wild-type and trf4-ts trf5Δ cells. RNase H digestion and Northern blot are as in A.

Figure 9.—

Summary of pathways regulating histone levels in yeast. See text for details.