Report

Requirement of Yeast SGS1 and SRS2 Genes for Replication and Transcription

See allHide authors and affiliations

Science  17 Dec 1999:
Vol. 286, Issue 5448, pp. 2339-2342
DOI: 10.1126/science.286.5448.2339

Abstract

The SGS1 gene of the yeastSaccharomyces cerevisiae encodes a DNA helicase with homology to the human Bloom's syndrome gene BLM and the Werner's syndrome gene WRN. The SRS2 gene of yeast also encodes a DNA helicase. Simultaneous deletion ofSGS1 and SRS2 is lethal in yeast. Here, using a conditional mutation of SGS1, it is shown that DNA replication and RNA polymerase I transcription are drastically inhibited in the srs2Δ sgs1-ts strain at the restrictive temperature. Thus, SGS1 and SRS2function in DNA replication and RNA polymerase I transcription. These functions may contribute to the various defects observed in Werner's and Bloom's syndromes.

Patients with Bloom's syndrome (BS) and Werner's syndrome (WS) suffer from growth retardation and a high incidence of cancers. In addition, individuals with WS age prematurely (1). Both syndromes display hyper-recombination and an increased incidence of karyotypic abnormalities (1). The SGS1 gene ofSaccharomyces cerevisiae is a member of the RecQ family of DNA helicases (2), and it shares extensive homology with the human Bloom's syndrome gene BLM and the Werner's syndrome gene WRN (3). The helicases encoded bySGS1, BLM, and WRN function in the 3′ → 5′ direction (4). A 3′ → 5′ DNA exonuclease activity is also present in the WRN protein (5). Sgs1 protein interacts physically with DNA topoisomerases II and III, and the deletion of SGS1 results in a reduction in growth rate, an elevation in the rate of mitotic recombination, and a decrease in the fidelity of chromosome segregation (2). As in WS, thesgs1 deletion (sgs1Δ) decreases the average life-span of cells and accelerates aging (6). The WRN and Sgs1 proteins are concentrated in the nucleolus (6,7), and sgs1Δ cells display premature fragmentation of their nucleoli (6). In addition,sgs1Δ cells accumulate large numbers of extrachromosomal ribosomal DNA (rDNA) circles (ERCs), and ERC accumulation in yeast leads to senescence (8). Here we elucidate the cellular function of SGS1.

Like SGS1, the S. cerevisiae SRS2 gene encodes a DNA helicase with 3′ → 5′ polarity, and mutations inSRS2 result in hyper-recombination (9). To examine whether the Sgs1 and Srs2 proteins function redundantly in a biological process such as DNA replication, we tried to generate thesgs1Δ srs2Δ double mutant strain by deleting the SRS2 gene from the sgs1Δ strain or by deleting the SGS1 gene from the srs2Δ strain (10). All our attempts, however, were unsuccessful, which suggested that the sgs1Δ srs2Δ combination was lethal. To verify this, we crossed a MATα yeast strain, which carried the sgs1Δ mutation marked with theLEU2 gene, to a MAT a strain, which carried the srs2Δ mutation marked with URA3(10). From analysis of 100 tetrads, we obtained 239 spores, none of which carried both the URA3 andLEU2 markers. Thus, deletion of both SGS1 andSRS2 is lethal.

Next, we isolated a recessive mutation of SGS1 that confers a temperature-sensitive (ts) conditional lethal phenotype in thesrs2Δ sgs1Δ mutant strain (11). The srs2Δ sgs1Δ strain carrying this sgs1-ts mutation in plasmid pPM980 (srs2Δ sgs1-ts) grew at the permissive temperature (25°C) but did not grow at the nonpermissive temperature (39°C), whereas the srs2Δ sgs1Δ strain carrying the wild-type SGS1 gene in plasmid pPM914 (srs2Δ SGS1) grew at 39°C (Fig. 1A). Sequence analysis has revealed four mutational alterations in the sgs1-ts mutant gene that consist of a GAA (Glu) to AAA (Lys) change in codon 171, two consecutive AGA (Arg) to AAA (Lys) changes in codons 1048 and 1049, and a CGC (Arg) to TGC (Cys) change in codon 1267. The Arg 1048 and 1049 codons are invariant among the yeast SGS1 and humanBLM and WRN genes.

Figure 1

DNA and RNA synthesis in the srs2ΔSGS1 and srs2Δ sgs1-ts mutant strains. (A) Growth of srs2Δ SGS1 andsrs2Δ sgs1-ts strains. Strains were grown at 25°C in YPD medium. At time 0, the cultures were divided. Half of the culture was left at 25°C, the other half was shifted to 39°C, and the cell density was determined (OD600) at the indicated time points. (B and C) DNA and RNA synthesis in the mutant strains. [3H]uracil was added to cultures grown at 25°C, and cultures were incubated for another 20 min at 25°C to allow the nucleotide pools to equilibrate. Cultures were then split (time 0). Half of the culture was incubated at 25°C and the other half at 39°C. Radioactivity at each time point was determined by liquid scintillation counting. cpm, counts per minute. ○,srs2Δ SGS1 strain at 25°C and •, 39°C ; ▵, srs2Δ sgs1-ts strain at 25°C and ▴, 39°C.

To determine whether the srs2Δ sgs1-ts mutant was defective in DNA or RNA synthesis, we examined the incorporation of radioactive label into these macromolecules (12,13). DNA and RNA synthesis were severely inhibited in thesrs2Δ sgs1-ts cells incubated at 39°C (Fig. 1, B and C) but were not affected in the srs2Δsgs1-ts strain at 25°C or in the srs2Δ SGS1strain at 25° and 39°C (Fig. 1, B and C). Growth and DNA and RNA synthesis were not affected in the sgs1Δ or thesgs1-ts mutant strains at 39°C (14).

To verify the defect in RNA synthesis in thesrs2Δ sgs1-ts mutant, we measured the rate of total RNA synthesis by pulse-labeling cells with [3H]uracil after a shift to the restrictive temperature (12, 13). Both the srs2Δsgs1-ts and srs2Δ SGS1 strains showed a rapid drop in RNA synthesis after a shift to 39°C (Fig. 2A). This inhibition was transient and was caused by heat shock (15). The srs2ΔSGS1 strain recovered from heat shock after about 1 hour and resumed RNA synthesis. The srs2Δ sgs1-tsmutant, however, displayed a reduced rate of RNA synthesis (Fig. 2A).

Figure 2

RNA polymerase I, II, and III transcription in the srs2Δ SGS1 and srs2Δsgs1-ts mutant strains. (A) Total RNA synthesis in thesrs2Δ SGS1 and srs2Δ sgs1-tsstrains. Strains were grown at 25°C in YPD medium. Cultures were divided into several aliquots, [3H]uracil was added to one aliquot, and cells were pulse-labeled at 25°C for 10 min (sample at time 0). The other aliquots were incubated at 39°C for the indicated times, and were then pulse-labeled for 10 min at 39°C. Total cellular RNA was isolated from each pulse-labeled sample. Radioactivity in the same amount of total RNA of each sample was determined by liquid scintillation counting, and incorporation of [3H]uracil (counts per minute per microgram of total RNA) was normalized by the counts present in thesrs2Δ SGS1 sample that was pulse-labeled at 25°C (time 0). •, srs2Δ SGS1; ▴,srs2Δ sgs1-ts. (B) Amounts of rRNAs transcribed by RNA polymerase I in the srs2ΔSGS1 and srs2Δ sgs1-ts strains after shift to 39°C. Total RNA obtained from cells pulse-labeled with [3H]uracil as in (A) was subjected to electrophoresis and fluorography. (C) Amounts of HIS3,MET19, ACT1, and POL1 mRNAs in the srs2Δ SGS1 andsrs2Δ sgs1-ts strains after shift to 39°C. (D) Heat shock-inducible synthesis of HSP26 mRNA and galactose-inducible synthesis of GAL10 andGAL7 mRNAs in the srs2Δ SGS1 andsrs2Δ sgs1-ts strains. (E) Amounts of intron-containing precursors of tRNAIle(tRNAI) and tRNATrp (tRNAW)in thesrs2Δ SGS1 and srs2Δsgs1-ts strains after shift to 39°C. For (C), (D), and (E), cultures were transferred from 25° to 39°C, and mRNA amounts (C and D) and amounts of intron-containing precursors of tRNAs (E) were examined at the indicated times by Northern hybridization.

Results of RNA analysis suggested that the synthesis of large ribosomal RNAs (rRNAs) was affected in the srs2Δ sgs1-tsstrain (16). The synthesis of RNA polymerase I–dependent rRNA was specifically analyzed by pulse-labeling RNA with [3H]uracil and subjecting the RNA to gel electrophoresis and fluorography (12, 13). Upon transfer to 39°C, the srs2Δ SGS1 strain showed a transient decrease in the synthesis of all large rRNA species (27S, 25S, 20S, and 18S) and was followed by nearly full recovery after 2 hours (Fig. 2B). In contrast, the srs2Δ sgs1-ts strain showed a dramatic decrease in all large rRNA species upon shift to 39°C, with no recovery to normal amounts even after 5 hours at 39°C (Fig. 2B). Thus, inactivation of both SRS2 and SGS1 leads to a defect in RNA polymerase I transcription.

To determine the effect of the srs2Δ sgs1-tsmutation on RNA polymerase II transcription, we used Northern (RNA) blot analysis to examine the amounts of HIS3,MET19, ACT1, POL1, and other mRNAs (12, 13) in the srs2Δsgs1-ts and srs2Δ SGS1 strains before (sample at 0 min) and after shifting the cultures to 39°C (Fig. 2C) (17). However, we observed no decrease in the amounts of the seven mRNAs examined in the srs2Δsgs1-ts strain over the 6-hour period at 39°C (Fig. 2C) (17). We next examined the effect of thesrs2Δ sgs1-ts mutation on the inducible synthesis ofHSP26, GAL7, and GAL10 mRNAs (12, 13). Transfer of srs2ΔSGS1 and srs2Δ sgs1-ts strains from 25° to 39°C greatly increased heat shock–inducible HSP26 mRNA in both strains (Fig. 2D). To examine the induction of GAL7and GAL10 mRNAs, galactose was added and cultures were transferred immediately to 39°C. The synthesis of GAL7 andGAL10 mRNAs also was not affected in thesrs2Δ sgs1-ts mutant strains (Fig. 2D). Thus, RNA polymerase II transcription was not affected in thesrs2Δ sgs1-ts strain at the restrictive temperature.

To determine the effect of the srs2Δ sgs1-tsmutation on RNA polymerase III transcription, we examined the synthesis of Ile tRNA (tRNAIle) and Trp tRNA (tRNATrp). We used hybridization probes that were complementary to the intron sequences present in these tRNA genes to avoid the problem of high stability of tRNAs. Because tRNA introns are processed rapidly, with a half-life of <3 min (18), the amount of precursor tRNA containing the intron sequences reflects the rate of transcription of these genes. The amounts of precursor tRNAIle or tRNATrp were not affected in the srs2Δsgs1-ts cells incubated at 39°C for up to 6 hours (Fig. 2E).

To verify the effect of the srs2Δ sgs1-tsmutations on DNA synthesis, the srs2Δ SGS1 andsrs2Δ sgs1-ts strains were synchronized in G1 with the use of the yeast mating pheromone α factor (19). Cells were shifted to either the permissive or restrictive temperature after removal of the α factor, and their DNA content was monitored by flow cytometry (19). The kinetics of DNA synthesis were similar in both strains at the permissive temperature (14). At the restrictive temperature, there was no DNA synthesis in the srs2Δ sgs1-tsstrain even after at least 8 hours (Fig. 3). In contrast, DNA synthesis continued in the srs2Δ SGS1 strain at the restrictive temperature (Fig. 3). These observations were further confirmed by releasing these strains from α-factor arrest into medium containing [3H]uracil at 25° or 39°C and by determining the amount of radioactivity incorporated into DNA (20).

Figure 3

Inhibition of DNA replication in thesrs2Δ sgs1-ts mutant strain. Strains grown at 25°C were synchronized in G1 with mating pheromone α factor, released from α-factor arrest, and then shifted to 39°C (time 0). Samples were taken at the indicated time points, and their DNA content was determined by flow cytometry.

Thus, yeast SGS1 and SRS2 genes function in RNA polymerase I transcription and in DNA replication. Sgs1 and Srs2 proteins must play redundant roles, because the absence of eitherSGS1 or SRS2 is not lethal, and the effects on rRNA transcription and DNA replication are not seen in thesgs1Δ or srs2Δ single mutants but are evident only in the srs2Δ sgs1-ts double mutant. In humans, the BLM and WRN genes and the putativeSRS2 counterpart may function redundantly in RNA polymerase I transcription and DNA replication. The effects of thesrs2Δ sgs1-ts mutations are similar to the inhibition of RNA polymerase I transcription and DNA replication in the yeast top1 top2-ts double mutant at the restrictive temperature (21). It has been proposed that topoisomerases I and II act as a swivel to relax the torsional stress that arises during DNA replication, and which may also result from the high rate of transcription of rDNA (21).

Sgs1 and Srs2 may function in the unwinding of double-stranded DNA during movement of the replication fork. A role for Sgs1 and Srs2 in the initiation of DNA replication is also possible and is suggested by the observation that the Sgs1-WRN counterpart foci-forming activity–1 (FFA-1) from Xenopus laevis is a component of replication foci, which are the initiation sites of DNA replication (22). SGS1, SRS2, and their human counterparts may also function in the unwinding of rDNA during transcription and be responsible for the high rate of rRNA synthesis.

The involvement of Sgs1 in DNA replication and in RNA polymerase I transcription may help to explain the various defects observed in Bloom's and Werner's syndromes. A subtle deficiency in DNA replication may cause increased recombination, chromosome loss, and other karyotypic abnormalities. These manifestations, in turn, could account for the increased frequency of cancers in BS and WS patients. Ribosomal RNA chain elongation may be slowed in WS cells, which may render RNA polymerase I more prone to pausing that could trigger the formation of double strand breaks in rDNA. Repair of such breaks by nonhomologous end-joining could result in the accumulation of deletions within the genomic rDNA array and contribute to premature aging in WS patients.

  • * To whom correspondence should be addressed. E-mail: sprakash{at}scms.utmb.edu

REFERENCES AND NOTES

View Abstract

Navigate This Article