A Genetic Screen for Ribosomal DNA Silencing Defects Identifies Multiple DNA Replication and Chromatin-Modulating Factors

Abstract

Transcriptional silencing in Saccharomyces cerevisiaeoccurs at several genetic loci, including the ribosomal DNA (rDNA). Silencing at telomeres (telomere position effect [TPE]) and the cryptic mating-type loci (HML and HMR) depends on the silent information regulator genes, SIR1,SIR2, SIR3, and SIR4. However, silencing of polymerase II-transcribed reporter genes integrated within the rDNA locus (rDNA silencing) requires only SIR2. The mechanism of rDNA silencing is therefore distinct from TPE and HM silencing. Few genes other than SIR2 have so far been linked to the rDNA silencing process. To identify additional non-Sir factors that affect rDNA silencing, we performed a genetic screen designed to isolate mutations which alter the expression of reporter genes integrated within the rDNA. We isolated two classes of mutants: those with a loss of rDNA silencing (lrs) phenotype and those with an increased rDNA silencing (irs) phenotype. Using transposon mutagenesis,lrs mutants were found in 11 different genes, and irs mutants were found in 22 different genes. Surprisingly, we did not isolate any genes involved in rRNA transcription. Instead, multiple genes associated with DNA replication and modulation of chromatin structure were isolated. We describe these two gene classes, and two previously uncharacterized genes, LRS4 and IRS4. Further characterization of the lrs and irs mutants revealed that many had alterations in rDNA chromatin structure. Several lrs mutants, including those in the cdc17 and rfc1 genes, caused lengthened telomeres, consistent with the hypothesis that telomere length modulates rDNA silencing. Mutations in the HDB (RPD3) histone deacetylase complex paradoxically increased rDNA silencing by a SIR2-dependent, SIR3-independent mechanism. Mutations in rpd3 also restored mating competence selectively to sir3Δ MATα strains, suggesting restoration of silencing at HMR in a sir3mutant background.

ACKNOWLEDGMENTS

We thank Michael Hampsey, Danny Reinberg, David Shore, and Susan Gasser for communicating results prior to publication, Paul Kaufman, Jasper Rine, and David Shore for helpful discussions, and Greg Cost and Nurjana Bachman for comments on the manuscript. We also thank Forrest Spencer for providing access to a photography-equipped stereoscopic microscope, Virginia Zakian and Lorraine Pillus for the telomere probe plasmid, and Mike Snyder and Susan Michaelis for the transposon insertion library.

J.S.S. was supported by a postdoctoral fellowship from the Leukemia Society of America and an NIH postdoctoral training grant. This work was supported in part by National Institutes of Health grants CA16519 and GM36481 to J.D.B.