Understanding the relationship between telomeres, telomerase, and DNA G-quadruplexes

Abstract

Cancer cells elongate their telomeres - G-rich repetitive sequences found at the end of linear chromosomes, allowing limitless replicative potential in these cells. Approximately 85% of cancers use telomerase to extend telomeres, making it an attractive potential anti-cancer target. The G-rich nature of telomeres allows the formation of DNA G-quadruplex secondary structures. Previous data had demonstrated that telomeric G-quadruplex substrates could not be extended by ciliate telomerase (Zahler et al., 1991). However, while the above observation is true for anti-parallel G-quadruplexes, parallel G-quadruplexes were shown to be substrates for ciliate telomerase (Oganesian et al., 2006). Whether human telomerase could extend parallel G-quadruplexes was unknown. In this thesis, I confirmed that human telomerase, like ciliate telomerase, can extend parallel, intermolecular G-quadruplexes in vitro. The ability of telomerase to extend G-quadruplexes is also true for parallel, intramolecular G-quadruplexes, indicating that the parallel nature of the structure allows telomerase extension. Extension of parallel G-quadruplexes using both biochemical and single-molecule FRET microscopy revealed that parallel G-quadruplexes are bound by telomerase as a distinct substrate and partially unfolded, allowing hybridisation of the RNA template. This partially unwound G-quadruplex is extended by human telomerase to the hTR template boundary, followed by translocation and complete G-quadruplex unfolding. Stabilisation of the parallel G-quadruplex using a parallel-G-quadruplex-specific ligand NMM did not inhibit telomerase activity demonstrating that chemically-stabilised parallel G-quadruplexes can be extended by human telomerase. Using a G-quadruplex specific antibody I showed that G-quadruplexes at telomeres increased after NMM treatment, indicating that parallel G-quadruplexes exist at human telomeres in vivo, and that telomeres with G-quadruplexes are a site of localisation for human telomerase. A potential protective effect of Gquadruplexes at uncapped telomeres was also investigated. In Saccharomyces cerevisiae lacking cdc13, equivalent in function to mammalian POT1, the DNA damage response could be suppressed by stabilising Gquadruplexes, showing that G-quadruplexes can have a protective effect at uncapped telomeres, but whether this is true at mammalian telomeres was unknown. In chapter 3 of this thesis I demonstrated that the DNA damage response at uncapped telomeres was suppressed by G-quadruplex stabilising ligands in G1 cells. I showed that G-quadruplex-telomere colocalisation increase in the absence of POT1, consistent with in vitro FRET experiments (Hwang et al., 2012). Treatment of POT1-deficient telomeres in G1 with G-quadruplex stabilising ligands reduced G-quadruplex-telomeres colocalisation. I provide preliminary data indicating that the nucleotide excision repair pathway is responsible for this phenotype, and that loss of stabilised telomeric G-quadruplexes is linked to the DNA damage response suppression phenotype. This thesis provides a body of work that improves our understanding of the role of G-quadruplexes at telomeres.