To investigate the role of the yeast telomere-, silencing-, and UAS-binding protein RAP1 in telomere position effects, we have characterized two sets of mutant cells: (1) a set of rap1 alleles (termed the rap1 t alleles) that produce truncated RAP1 proteins missing the carboxy-terminal 144-165 amino acids; and (2) null mutants of the RIF1 gene, encoding a protein capable of interaction with the carboxyl terminus of RAP1. The data presented here indicate that loss of the carboxyl terminus of RAP1 abolishes position effects at yeast telomeres and diminishes silencing at the HML locus. Elimination of position effects in these cells is associated with increased accessibility to the Escherichia coli dam methylase in vivo. Thus, the carboxy-terminal domain of RAP1 is required for telomere position effects. In contrast, rifl deletion alleles increase the frequency of repressed cells. Using the rap1 t alleles to generate wild-type cells differing only in telomere tract lengths, we also show that telomere position effects are highly sensitive to changes in the size (or structure) of the telomeric tract. Longer poly(GI_3T) tracts can increase the frequency of transcriptional repression at the telomere, suggesting that telomeric poly(Gl_3 T) tracts play an active role in the formation or stability of subtelomeric transcriptional states.

The yeast protein RAP1, initially described as a transcriptional regulator, binds in vitro to sequences found in a number of seemingly unrelated genomic loci. These include the silencers at the transcriptionally repressed mating-type genes, the promoters of many genes important for cell growth, and the poly[(cytosine)1-3 adenine] [poly(C1-3A)] repeats of telomeres. Because RAP1 binds in vitro to the poly(C1-3A) repeats of telomeres, it has been suggested that RAP1 may be involved in telomere function in vivo. In order to test this hypothesis, the telomere tract lengths of yeast strains that contained conditionally lethal (ts) rap1 mutations were analyzed. Several rap1ts alleles reduced telomere length in a temperature-dependent manner. In addition, plasmids that contain small, synthetic telomeres with intact or mutant RAP1 binding sites were tested for their ability to function as substrates for poly(C1-3A) addition in vivo. Mutations in the RAP1 binding sites reduced the efficiency of the addition reaction.

The Saccharomyces cerevisiae DNA-binding protein RAP1 is capable of binding in vitro to sequences from a wide variety of genomic loci, including upstream activating sequence elements, the HML and HMR silencer regions, and the poly(Gl13T) tracts of telomeres. Recent biochemical and genetic studies have suggested that RAP1 physically and functionally interacts with the yeast telomere. To further investigate the role of RAP1 at the telomere, we have identified and characterized three intragenic suppressors of a temperature-sensitive allele ofRAP], rapl-S. These telomere deficiency (rapl') alleles confer several novel phenotypes. First, telomere tract size elongates to up to 4 kb greater than sizes of wild-type or rapl-5 telomeres. Second, telomeres are highly unstable and are subject to rapid, but reversible, deletion of part or all of the increase in telomeric tract length. Telomeric deletion does not require the R4D52 or RADI gene product. Third, chromosome loss and nondisjunction rates are eleva ted 15-to 30-fold above wild-type levels. Sequencing analysis has shown that each rapl' allele contains a nonsense mutation within a discrete region between amino acids 663 and 684. Mobility shift and Western immunoblot analyses indicate that each allele produces a truncated RAP1 protein, lacking the C-terminal 144 to 165 amino acids but capable of efficient DNA binding. These data suggest that RAP1 is a central regulator of both telomere and chromosome stability and define a C-terminal domain that, while dispensable for viability, is required for these telomeric functions.Telomeres, the structures present at the ends of linear eukaryotic chromosomes, are essential for the stability and complete replication of the chromosome. Telomeric DNA, at its extreme terminus, is composed of a variable number of simple sequence repeats. These repeats usually contain a G-rich strand, oriented in a 5'-to-3' direction toward the terminus. In contrast to other genomic sequences, telomeres are replicated inexactly. The size and, in some organisms, the sequence of an individual telomere vary among different cells of a population (60). The average size of telomeric DNA appears to be maintained through a regulated equilibrium between the loss and gain of telomeric sequences (36,45,46,60). One component involved in this process is the ribonucleoprotein complex telomerase (16-18, 37, 48, 49, 59). This enzyme, identified in both ciliate and mammalian cells, is capable of catalyzing the addition of the G-rich simple sequences onto the 3' end of single-stranded substrates, utilizing a sequence within its RNA component as a template. A similar activity is likely to explain the properties of telomere addition in yeasts and other organisms in which telomerase has not yet been identified (60). The essentiality of maintaining telomeric sequences is underscored by the recombinational instability, chromosome loss, and lethality caused by loss of part or all of telomeric DNA (34, 60).Telomeres are packaged in vivo into unique nonnucleosomal complexes, f...

The Ku heterodimer, conserved in a wide range of eukaryotes, plays a multiplicity of roles in yeast. First, binding of Ku, which is composed of a 70 kDa (Hdf1p) and an 80 kDa (Hdf2p) subunit [1-3], to double-strand breaks promotes non-homologous end-to-end joining of DNA [3]. Second, Ku appears to participate in DNA replication, regulating both the number of rounds of replication permissible within the cell cycle and the structure of the initiation complex [3,4]. Furthermore, mutations in HDF1 or HDF2 rapidly reduce telomeric poly (TG1-3) tract size [1-3], hinting also at a possible telomeric function of Ku. We show here that the two subunits of the Ku heterodimer play a key role in maintaining the integrity of telomere structure. Mutations in either Ku subunit increased the single-strandedness of the telomere in a cell-cycle-independent fashion, unlike wild-type cells which form 3' poly(TG1-3) overhangs exclusively in late S phase [5]. In addition, mutations enhanced the instability of elongated telomeres to degradation and recombination. Both Ku subunits genetically interacted with the putative single-stranded telomere-binding protein Cdc13p. We propose that Ku protects the telomere against nucleases and recombinases.