Which end: dissecting Ku's function at telomeres and double-strand breaks.

Abstract

The ends of chromosomes are active places. These natural DNA termini must be protected from degradation, recombination, and end-to-end fusions—events that would be ultimately fatal for the genome. In addition, the enzyme telomerase ensures that telomeres in proliferating cells, such as single-celled organisms like budding yeast or the mammalian germ line, are completely replicated. Over the past decade, a plethora of telomereassociated proteins have been identified, based on their ability to regulate telomerase access, provide protection from nucleases and recombinases, and control telomere length (for review, see McEachern et al. 2000; de Lange 2002). In the budding yeast, at least a dozen proteins interact with just the terminal single-stranded overhang, with an additional dozen or more proteins that are associated with the duplex region of the telomere. At mammalian telomeres, which are bound by additional factors that are apparently not present in yeast, the situation is even more complex. To make the task of understanding events that take place at telomeres even more challenging, many telomere-associated proteins perform more than one biochemical activity at chromosome ends. For example, in human cells, the telomere repeat binding factor TRF2 protects chromosomes from end-to-end fusions, and also acts as a negative regulator of telomere length (van Steensel et al. 1998; Loayza and De Lange 2003). The yeast telomere end-binding factor Cdc13 similarly performs several discrete tasks: Cdc13 both positively and negatively regulates telomere elongation by telomerase, as well as shields chromosome termini from unregulated degradation by nucleases (Garvik et al. 1995; Nugent et al. 1996; Booth et al. 2001; Chandra et al. 2001). Analyses of these Cdc13-mediated processes have been greatly aided by separation-of-function mutations that appear to selectively impair each activity. In an analogous manner, recent genetic studies of the catalytic subunit of telomerase have helped to identify discrete regions that mediate enzyme processivity, interaction with other telomerase subunits, or recruitment of the telomerase complex to chromosome termini (Peng et al. 2001; Armbruster et al. 2003; Friedman et al. 2003; Kim et al. 2003). One factor that is critical to several aspects of telomere biology, yet has lagged behind in this genetic dissection, is the Ku heterodimer. Composed of 70and 80-kD subunits, Ku was first appreciated for its role in double-strand break (DSB) repair (Taccioli et al. 1994). In both mammalian and yeast cells, DSBs that arise via developmentally regulated site-specific rearrangements [e.g., V(D)J recombination in vertebrate cells], or are induced experimentally or as a result of environmental assaults such as irradiation, must be readily repaired to maintain genome stability. Ku plays a central role in this process by binding to DNA double-strand ends and recruiting additional factors, which are required to complete repair via nonhomologous end-joining (for review, see Jackson 2002). Somewhat paradoxically, Ku also functions at telomeres, even though these DNA termini are not normally substrates for end-joining reactions. In fact, telomeres must be specifically protected from end-to-end fusions to maintain genomic stability. Furthermore, the role that the Ku heterodimer plays at telomeres is complex, as evidenced by the fact yku70or yku80budding yeast strains, although still viable, exhibit an impressively large number of phenotypes, indicative of severe telomere dysfunction. In these Ku-defective strains, the structure of the extreme single-stranded termini is altered by increased exposure to nucleolytic activity (Gravel et al. 1998; Polotnianka et al. 1998; Maringele and Lydall 2002). Yeast strains that lack Ku also have substantially shortened telomeres (Boulton and Jackson 1996; Porter et al. 1996), in part because of this enhanced nuclease action (A.A. Bertuch and V. Lundblad, in prep.), but primarily because of impaired regulation of telomerase (Peterson et al. 2001; Stellwagen et al. 2003). Additionally, loss of Ku function in yeast disrupts the transcriptional silencing of telomere-proximal genes, by altering the composition of factors that comprise telomeric heterochromatin (Laroche et al. 1998; Nugent et al. 1998; Mishra and Shore 1999). Analysis of yku70or yku80yeast null strains, therefore, has firmly established that the Ku heterodimer is a multifunctional player that contributes to telomere Corresponding author. E-MAIL abertuch@bcm.tmc.edu; FAX (713) 798-3457. Article and publication are at http://www.genesdev.org/cgi/doi/10.1101/ gad.1146603.