Functional Analysis of the Cell Cycle Protein Dpb11 in Response to DNA Damage and Replicative Stress

DNA molecule is complex, fragile and can suffer different damages. Specific DNA repair mechanisms were evolved to respond to these challenges, and to allow a faithful transmission of genetic information throughout generations. If the damaging conditions are extensive, a mechanism called DNA damage checkpoint takes care of arresting the progression of the cell division cycle to allow the cell to repair the damage before proceeding further. Genes involved in the DNA damage checkpoint are conserved throughout evolution and mutations in the human genes are known to produce severe illnesses - like Ataxia Telangiectasia - and genomic instability, which is usually considered as the onset of cancer: indeed checkpoint genes, like BRCA1, were found to be mutated in different types of cancers. The yeast Saccharomyces cerevisiae has been widely used to study the DNA damage checkpoint because, despite its evolutionary distance, the easiness in generating knockout and mutant strains has facilitated the understanding of the underlying mechanisms. In this yeast, as in humans, the signal that activates the checkpoint is represented by the ssDNA covered by RPA, to which many different checkpoint and repair factors are recruited. ssDNA signals are responsible for the activation of Mec1 (hATR), the apical kinase of the checkpoint pathway, but in humans two other factors are required for this signalling to occur: a ring-like heterotrimer - the PCNA-like complex - which is loaded onto DNA in response to damage and which recruits the second factor, TopBP1. Once active, Mec1 kinase phosphorylates a series of substrates, among which there is the Ddc1 subunit of the PCNA-like complex, and the Rad9 protein; phosphorylated Rad9 allows the recruitment of Rad53, the central kinase of the checkpoint whose Mec1-dependent activation contributes to cell survival after DNA damage and replication stress. To be phosphorylated by DNA-bound Mec1, the Rad9 protein must be recruited to chromatin: this process involves the binding of a Rad9 domain - the Tudor domain - to a methylated lysine on histone H3. Indeed, cells mutated in the conserved H3 lysine, in the Tudor domain or in the histone methyltransferase Dot1 are defective in Rad9 and Rad53 phosphorylation when DNA is damaged in the G1 phase of the cell cycle. Surprisingly, when these mutants receive a DNA damage in mitosis, they are still able to phosphorylate Rad9 and Rad53, suggesting the presence of a second pathway that, in M phase, provides an alternative way for Rad9 to be phosphorylated. In this thesis evidences regarding this alternative pathway for Rad9 recruitment and phosphorylation are provided. This pathway depends upon the C-terminal tail of Dpb11, the yeast homologue of human TopBP1, and on the Mec1-dependent phosphorylation of threonine 602 of the Ddc1 subunit of the PCNA-like complex. We show that Dpb11 itself is phosphorylated after DNA damage and that this phosphorylation is reduced in the presence of a non-phosphorylatable 602-residue on Ddc1, suggesting that in these conditions Dpb11 cannot be functionally recruited. Supporting this idea the two-hybrid interaction between Ddc1 and Dpb11 requires the presence of a functional Mec1 kinase. Although being capable of in vitro stimulation of Mec1 kinase activity, after UV irradiation in M phase, Dpb11 is not required for Mec1 to phosphorylate its binding partner Ddc2. On the other hand, we provide evidences that Dpb11 performs its Mec1 activation task during the response to global replication stress; indeed Dpb11 and the PCNA-like complex are independently required to obtain a proper phosphorylation of histone H2A - here used as a marker of Mec1 kinase activity - and a full Rad53 activation. Consistent with this observation ddc1Δdpb11-1 mutants are extremely sensitive to chronic exposition to hydroxyurea, a commonly used chemotherapeutic drug that generates replication stress by reducing the concentration of dNTPs in the cell. We also provide evidence that this lethality is not due to classical checkpoint functions like the stabilisation of stalled replication forks or the ability to delay entrance in M phase. We suggest also that other proteins known to be involved in checkpoint activation after hydroxyurea treatment are working in the pathway in which Dpb11 is involved.