The kinase activity of DNA-PK is required to protect mammalian telomeres

Abstract

The kinase activity of DNA-dependent protein kinase (DNA-PK) is required for efficient repair of DNA double-strand breaks (DSB) by non-homologous end joining (NHEJ). DNA-PK also participates in protection of mammalian telomeres, the natural ends of chromosomes. Here we investigate whether the kinase activity of DNA-PK is similarly required for effective telomere protection. DNA-PK proficient mouse cells were exposed to a highly specific inhibitor of DNA-PK phosphorylation designated IC86621. Chromosomal end-to-end fusions were induced in a concentration-dependent manner, demonstrating that the telomere end-protection role of DNA-PK requires its kinase activity. These fusions were uniformly chromatid-type, consistent with a role for DNA-PK in capping telomeres after DNA replication. Additionally, fusions involved exclusively telomeres produced via leading-strand DNA synthesis. Unexpectedly, the rate of telomeric fusions induced by IC86621 exceeded that which occurs spontaneously in DNA-dependent protein kinase catalytic subunit (DNA-PKcs) mutant cells by up to 110-fold. One explanation, that IC86621 might inhibit other, as yet unknown proteins, was ruled out when the drug failed to induce fusions in DNA-PKcs knock-out mouse cells. IC86621 did not induce fusions in Ku70 knock-out cells suggesting the drug requires the holoenzyme to be effective. ATM also is required for effective chromosome end protection. IC86621 increased fusions in ATM knock-out cells suggesting DNA-PK and ATM act in different telomere pathways. These results indicate that the kinase activity of DNA-PK is crucial to reestablishing a protective terminal structure, specifically on telomeres replicated by leading-strand DNA synthesis.

Introduction

In order to protect the integrity of the genome, it is essential that a cell be able to distinguish natural chromosome ends (telomeres) from unnatural ends created by DNA double-strand breaks (DSBs). Mammalian cells possess at least two mechanisms that repair DSBs; non-homologous end joining (NHEJ) and homologous recombination (HR). Mutations that compromise either DSB repair mechanism result in genomic instability [1], [2], [3]. Equally important to maintenance of genomic stability are functional telomeres, highly specialized complexes of repetitive DNA and their associated binding proteins [4]. By protecting natural chromosome ends, telomere binding proteins have a function precisely opposite to that of DSB repair. The importance of functional telomeres in preserving genomic stability is often under-appreciated because, under most circumstances, they protect chromosome ends extremely well. However when telomere dysfunction does occur, either through replication dependent erosion of telomeric sequence or loss of structural integrity of the terminal cap, the consequences are severe. They include triggering cellular senescence and induction of chromosomal rearrangements that may contribute to carcinogenesis [5], [6].

The molecular basis of telomere function includes a paradoxical requirement for several proteins better known in relation to DNA DSB repair. Three such proteins—Ku70, Ku80, and the catalytic subunit of DNA-dependent protein kinase (DNA-PKcs)—promote DSB repair through NHEJ [7], [8], [9]. Knock-out mutations in any of these three genes, or a hypomorphic severe combined immuno-deficiency (scid) allele of DNA-PKcs, cause spontaneous end-to-end fusions of chromosomes that retain large blocks of telomere sequence at the point of fusion and contribute significantly to chromosomal instability [10], [11]. Telomeric fusions occurred in these cells at a rate of approximately 1 per 60 cells per cell cycle. This is a sharp increase over the rate of spontaneous fusion in DNA-PK proficient cells, in which they are essentially undetectable. Interestingly, a rate of fusion two orders of magnitude greater has been reported when a dominant-negative allele of another telomere capping protein, telomere repeat binding factor 2 (TRF2), is expressed in human cells [12], [13]. Ku70, Ku80 and DNA-PKcs have each been located at mammalian telomeres [1], [14]. However, it has yet to be shown that they assemble into a functional, catalytically active holoenzyme, or that the kinase activity of DNA-PKcs is required for its role in telomere protection.

To investigate further the role of DNA-PKcs in telomere end-capping function, we exposed mouse and human cells to a novel chemical inhibitor of its kinase activity. This experimental drug (1-(2-hydroxy-4-morpholin-4-yl-phenyl)-ethanone), designated IC86621 (ICOS Corporation, Bothell, WA), has been shown to be highly specific for DNA-PKcs in comparison to other DNA-PK inhibitors such as wortmannin. In particular, IC86621 has little effect on related members of the phosphotidyl inositol kinase family such as ATM and ATR. Biochemical and genetic studies of IC86621 demonstrating its function as a specific inhibitor of DNA-PK kinase activity have been reported [15], [16]. Treatment of wild-type cells with IC86621 for one cell cycle induced telomere dysfunction (end-capping failure) that resulted in a much higher frequency of telomeric fusion than that seen in DNA-PKcs mutant mouse cells. Importantly, the telomeric fusions formed after replication and exclusively involved telomeres produced via leading-strand DNA synthesis. The studies presented here demonstrate that the kinase activity of DNA-PKcs is essential to effectively protect the natural ends of mammalian chromosomes and further support the view that this activity is necessary for the proper processing of the leading-strand telomere immediately following replication.