Info

silenced region within and ▼

downstream from the array Weakest

Silencing

Figure 17.2. Silenced loci in yeast. The 4 major silent loci are shown, which fall into 3 classes: The silent mating type cassettes HMRa and HMLa, telomeres and the tandem array of ribosomal RNA genes called the rDNA array or RDN1 (the genetic designation of this locus). In yeast, mating type is determined by the sets of genes expressed at the MAT locus, either a and a. Yeast has 3 copies of this locus, one which is expressed (MAT) and two which are strongly silenced (HMR and HML). These silent mating type cassettes contain flanking DNA elements, called E and I, which silence the MAT genes between them. The E and I sites in turn contain unidirectional binding sites for known DNA binding factors, RAP1 protein (which binds to the Rapl site), ABF1 protein (which binds to the Abfl site), and Origin Recognition Complex (which binds to the ACS element) (Laurenson and Rine 1992). The proteins encoded by SIR1, SIR2, SIR3 and SIR4 are required for silencing at the silent mating type cassettes. The TGi_3 repeats in telomeres contain arrays of RAP1 protein sites, which tether SIR2, SIR3 and SIR4 proteins to telomeres and silence nearby genes (Gottschling et al., 1990). The rDNA array binds a different set of proteins that include SIR2 protein and silence RNA polymerase II transcribed genes placed within these repeats (Straight et al., 1999).

C1-3A

Under a variety of mutant conditions that alter telomere chromatin structure, SIR proteins appeared to be released from the telomere and to relocalize at the silent mating type cassettes (Buck and Shore, 1995). Subsequently, SIR3 protein was shown to move from telomeres to newly made double-strand breaks (Martin et al., 1999; Mills et al., 1999). Finally, this relocalization of SIR proteins from telomeres plays an important role in the replicative aging of yeast, illustrating how release of proteins from telomeres can have an impact on lifespan (see Telomeres and Yeast Replicative Aging, below).

In addition to SIR proteins, RIF1 and RIF2 proteins also interact with the C-terminus of RAP1 protein bound at telomeres (Hardy et al., 1992; Wotton and Shore, 1997). One clear function of the RIF proteins is in telomere length control: cells that lack the C-terminal 160 amino acids of RAP1 protein have telomeres that grow from -330 bp of TG1-3 to 2-4 kb of TG1-3, and cells deleted for both RIF1 and RIF2 also have 2-4 kb TG1-3 telomere tracts (Kyrion et al., 1992; Wotton and Shore, 1997). Recent evidence also suggests that RIF proteins may interact with proteins or DNA independent of RAP1 protein while regulating telomere length (Levy and Blackburn, 2004).

Orthologs for RAP1 and RIF1 proteins have been identified in S. pombe and in humans, and show differences from budding yeasts. While RAP1 protein binds directly to DNA in S. cerevisiae, S. pombe and human cells have different Myb-related double-stranded telomere DNA binding proteins called taz1+ (in fission yeast) and TRF1 and TRF2 (in human cells). The RAP1 of fission yeast and humans lacks the DNA binding domain, and instead is tethered to telomeres by interactions with taz1 protein or TRF2, respectively. As in S. cerevisiae, the RIF1 protein ortholog also interacts with the C-terminus of the Rap1 ortholog in S. pombe and humans. While fission yeast rap1 and rif1 proteins and human Rap1 protein behave as negative regulators of telomere length (reviewed in detail by Smogorzewska and De Lange (2004)), human Rif1 protein apparently leaves telomeres to play a role in the ATM-dependent DNA damage response (Silverman et al., 2004). Recruitment of the human telomere binding protein TRF2 to DSBs has also been recently reported (Bradshaw et al., 2005). These properties of human Rif1 and TRF2 suggest that human telomeres may also serve as a reservoir of factors for other cellular processes such as the response to DSBs. The release of such factors could potentially be regulated by stochastic processes, e.g., a DSB activating a protein kinase cascade, or could be gradual as telomeres shorten over repeated cell divisions. A DNA damage function for the yeast Rif1 proteins has not been tested yet.

Telomere proteins that act at double-strand breaks Telomeres can be viewed as specialized double-strand breaks (DSBs). While telomeres have specific sequences and associated proteins, many of the proteins that interact with and process telomeres also play important roles in the recognition and processing of DSBs at internal chromosomal loci. These proteins include the S. cerevisiae DNA damage checkpoint kinases Tel1p and Mec1p (ATM and ATR in humans), the multifunctional MRX complex (MRN in humans), and the Yku70p-Yku80p DNA-end binding heterodimer (Ku70-Ku86 in humans) (Maser and DePinho, 2004). Yeast cells lacking these proteins maintain telomeres with fewer simple sequence repeats, indicating their role in telomere replication, and have defects in responding to and processing DSBs (Lustig and Petes, 1986; Morrow et al., 1995; Porter et al., 1996; Ritchie et al., 1999; Ritchie and Petes, 2000). The simple sequence repeats that make up telomere DNA and the combination of proteins that associate with these sequences prevent them from being processed as DSBs by the cellular machinery.

The first DSB-associated protein found to directly associate with telomeres was the Yku70/Yku80 hetero-dimer. This DNA end-binding protein, first found in mammalian cells, binds to DNA breaks and activates DNA-activated protein kinase catalytic subunit (DNA-PKcs), a member of the ATM family (Anderson and Lees 1992). The yeast Ku proteins were found to associate with the telomere and play a role in telomere replication by recruiting telomerase (Fisher et al., 2004; Gravel et al., 1998; Peterson et al., 2001). Since then, Ku heterodimer association with the telomere in fission yeast and humans has also been demonstrated (Bailey et al., 1999; Baumann and Cech, 2000).

Upon induction of a DSB, several proteins recognize and process this DNA lesion and signal the cell cycle machinery to pause. An elegant study investigating the choreography of these events in yeast found that the first proteins to bind are the MRX complex (composed of proteins encoded by MRE11, RAD50 and XRS2) and TEL1 protein, followed by recruitment of the RPA complex (Lisby et al., 2004). RPA then recruits MEC1 protein and MRX and TEL1 protein leave the DSB. Importantly, phenotypes for mutants in each of these genes include changes in the length of the TG1-3 repeats in budding yeast (Ritchie et al., 1999; Schramke et al., 2004; Smith et al., 2000; Tsukamoto et al., 2001). So what are these proteins? TEL1 and MEC1 are DNA damage protein kinases of the ATM family, whose orthologs are, respectively, tel1+ and rad3+ in S. pombe and ATM and ATR in humans. MRX is a multifunctional complex in vivo that has helicase and nuclease activity in vitro, whose orthologs in fission yeast and humans are called MRN complex (for Mre11-Rad50-Nbs1) (reviewed by Smogorzewska and De Lange (2004)). RPA is a conserved complex that functions in DNA synthesis and repair (reviewed by Schramke et al., (2004) and Smith et al., (2000)). Interestingly, cells lacking TEL1 and MEC1 in budding yeast, or tel1+ and rad3+ in fission yeast, have telomeres that gradually shorten as telomerase cannot access the telomere (Chan et al., 2001; Naito et al., 1998;

Ritchie et al., 1999). Experiments in budding yeast have shown that TEL1 protein associates with telomeres when they reach ~1/3 their normal length, and TEL1 association is a step in the pathway that allows telomere elongation by telomerase. In contrast, MEC1 protein telomere association is observed when survivor/ALT cells arise in the culture, suggesting a role for MEC1 protein when the telomere repeats become short enough for the telomere to lose its characteristic protection of the chromosome end (Hector et al., submitted). These data indicate different roles for these checkpoint kinases in maintaining and monitoring telomere length. As all of these proteins have a role in DNA damage signaling, they most likely play some yet-to-be defined role in how the cell senses a short telomere to pause the cell cycle. This function appears to be conserved in humans, as senescing cells show increased association of ATM with telomeres.

TEL2 protein may be another yeast telomere protein that plays a role in the cellular response to DSBs. The original TEL2 mutation, tel2-1, has telomeres one-half the length of wild-type cells and appears to control telomere length through the TEL1 pathway (Lustig and Petes, 1986). TEL2 protein binds to double and single-stranded telomere repeats in vitro, suggesting a direct role at the telomere similar to TEL1 protein (Kota and Runge 1998; Kota and Runge, 1999), and preliminary data indicate that TEL2 protein associates with telomeres in vivo (R. Kota and K.W. Runge, in preparation). However, deletion of the TEL2 gene is lethal, while deletion of the TEL1 gene is not, suggesting that TEL2 protein may act in another pathway as well (Runge and Zakian, 1996). Work in Caenorhabditis elegans suggests a role for TEL2 protein in aging and the response to DSBs. The C. elegans TEL2 ortholog was identified twice by mutation: once as clk-2, a mutation that slows the rate of aging, and once as rad-5, a DNA damage checkpoint mutation (Ahmed et al., 2001; Benard et al., 2001). The human ortholog of TEL2 has also been cloned and overexpressed in tissue culture cells, where it localizes in the cytoplasm (Jiang et al., 2003), but it is unclear if this overexpression reveals all of the normal locations for human Tel2 protein activity. Thus, while the in vivo functions of TEL2 are currently obscure, the efforts in many model organisms should soon reveal its mode of action at telomeres and in aging.

Chromatin remodeling complexes and telomere length While yeast telomeres form a nonnucleosomal complex based on micrococcal nuclease mapping (Wright et al., 1992), mutations in proteins known to modify nucleo-somes can alter telomere length. Every open reading frame in yeast has been deleted in a systematic fashion, and this library of strains is publicly available. A screen of each of these strains for changes in telomere length has revealed a number of unexpected gene deletions that alter telomere length. The genes identified include RSC2 (a member of the RSC remodeling complex), FMP26

(which interacts with the SAGA complex) and nine members of various histone deacetylase, histone methy-lase and histone ubiquitination complexes (Askree et al., 2004). Their effect on telomere length raises the possibility that many of these complexes may also modify telomere proteins to alter their functions or control the expression of telomere proteins. A third possibility was recently shown in Tetrahymena, where the placement of ordered nucleosome arrays adjacent to telomere repeats altered the length of the telomeric tract (Jacob et al., 2004), raising the possibility that chromatin-modifying proteins may alter telomere length in a similar way.

Unknown players in telomere metabolism The more telomeres are investigated, the more new processes are uncovered that require explanation. As mentioned previously, yeast telomeres have a 3' overhang of 12-14 bp during most of the cell cycle, but in late S-phase this end is resected to an overhang of >50 nt in a manner that does not require telomerase activity (Wellinger et al., 1996; Wellinger et al., 1993) (Figure 17.1). These data suggest that a nuclease degrades the 5' strand as part of normal telomere processing each cell cycle. The identity of this nuclease (or nucleases) and how it links telomeres to DNA replication will be of great interest. The process by which the 5' strand is resected also has important implications for cell senescence and aging. When telo-meres are short, degradation of the 5' strand would further reduce the number of double-stranded telomere repeats, and could convert short telomeres into structures that activate the short telomere checkpoint. Identification of the 5' telomere strand degrading nuclease may allow a direct test of this hypothesis.

Blood Pressure Health

Blood Pressure Health

Your heart pumps blood throughout your body using a network of tubing called arteries and capillaries which return the blood back to your heart via your veins. Blood pressure is the force of the blood pushing against the walls of your arteries as your heart beats.Learn more...

Get My Free Ebook


Post a comment