• 2018-07
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • Here we investigate in more detail how


    Here we investigate in more detail how Dot1 function contributes to the regulation of DNA damage tolerance. We find that Dot1 modulates the response to the alkylating agent MMS through its catalytic activity on H3K79. In fact, progressively reduced levels of H3K79 tri-methylation result in gradually increased resistance to MMS. In addition, we examine the functional interaction between Dot1 and an HA-tagged version of the Rad53 checkpoint kinase that also promotes increased MMS resistance and mutagenesis [40]. Our results indicate that there is a window of opportunity for TLS to act in the face of MMS lesions that is delineated by threshold levels of Rad53 activity. Moreover, we present evidence indicating that the contribution of Dot1 to DNA damage tolerance is exerted via Rad53 and controls the levels of Rev1 protein associated with chromosomes.
    Materials and methods
    Discussion We have previously described a role for the histone H3K79 methyltransferase Dot1 in the tolerance to alkylating DNA damage [34]. Here, we have further characterized this function of Dot1 by first analyzing the impact of different methylation states of H3K79 in the response to continuous MMS exposure. We provide evidence indicating that the regulation of DNA damage tolerance by Dot1 depends on its catalytic activity on H3K79 and not on other possible unknown substrate(s) because, like dot1Δ, both a catalytically defective dot1-G401V allele, and a non-methylatable H3-K79A version of histone H3 confer MMS hyper-resistance. To investigate how different degrees of H3K79 methylation affect MMS resistance, we have engineered and analyzed a set of strains with progressively crippled Dot1 activity, ranging between wild-type DOT1 and dot1Δ as the maximal and minimal Dot1 catalytic activity, respectively. We find a striking correlation between the decline of Dot1 activity and the increase in MMS resistance. This correlation was particularly evident for the loss of H3K79-me3, suggesting that this methylation state is the most relevant for the MMS response. Similarly, different functional relevance for the different methylation states of H3K79 in the coordination of DNA repair and checkpoint activation in response to UV has been proposed [62]. In contrast, it has been clearly demonstrated that Thio-TEPA silencing relies on global levels of H3K79 methylation, and not on specific methylation states [42]. Importantly, the MMS resistance conferred by the absence of Dot1 is independent of the silencing SIR complex [34]. The role of Dot1 in multiple nuclear processes, such as transcriptional silencing, meiotic checkpoint, DNA damage checkpoint or DSB repair, relies on the regulated binding of various key factors to specific chromosomal regions [25], [26], [27], [33]. In principle, the impact of Dot1 (i.e. H3K79 methylation) in MMS resistance and the higher number of MMS-induced chromosome-associated Rev1 foci in the dot1Δ mutant could emanate from a direct effect of a peculiar chromatin structure dictated by the H3K79 methylation status modulating the recruitment of the TLS machinery. However, our results support an alternative possibility implying that the effect of Dot1 in DNA damage tolerance is exerted indirectly through the regulation of the Rad53 checkpoint kinase. The fact that a rad53-HA mutant, characterized by reduced levels of the kinase, substantially phenocopies dot1Δ in the response to chronic MMS exposure ([34], [40]; this work) suggested a possible relationship between Dot1 and Rad53 in the regulation of tolerance to alkylating damage. Supporting this possibility, we find that the MMS resistance of both dot1Δ and rad53-HA depends on ubiquitylation of PCNA at K164, which is a crucial regulator of the TLS mechanism of DNA damage tolerance. Moreover, we show here that the increased MMS resistance of the dot1Δ mutant depends on Rad53. Both dot1Δ and rad53-HA mutants display enhanced resistance to alkylating damage and increased TLS-dependent MMS-induced mutagenesis. Strinkingly, in both mutants, the levels of Rad53 activated by MMS treatment are reduced compared to the wild type, but for different reasons. In the case of dot1Δ, Rad53 is produced at normal levels (see Figs. 5A, lanes 1 and 2 and 6A, lanes 5 and 7), but the inability to properly recruit the Rad9 adaptor to DNA damage sites results in defective activation of Rad53 ([27], [32], [33]; Fig. 5). On the other hand, the rad53-HA allele gives rise to a functional protein, which can be fully activated; however, it is highly unstable, resulting in the production of low levels of MMS-induced active kinase ([40]; see also Fig. 5A, lane 7). In the dot1Δ rad53-HA double mutant, the activation of the low amounts of kinase produced by the HA-tagged allele is further hampered by the absence of DOT1 resulting in barely detectable levels of phosphorylated kinase. Remarkably, whereas the dot1Δ and rad53-HA single mutants show increased MMS resistance, the dot1Δ rad53-HA double mutant shows strong MMS sensitivity implying that threshold levels of Rad53 activity determine the outcome of the cellular response to alkylating damage. We favor the scenario presented in Fig. 8 to explain our findings. In the face of MMS challenge that prevents the advance of replication forks, the wild-type strain fully activates Rad53 and the subsequent checkpoint responses controlled by this effector kinase, including cell cycle arrest, stabilization of replication forks and induction of DNA repair mechanisms [5]. In addition, high levels of Rad53 activity would negatively regulate the TLS mechanism of damage tolerance to prevent excessive mutagenesis. We propose that the sub-optimal levels of activated Rad53 present in rad53-HA mutants or in mutants that cripple H3K79 methylation (dot1-G401A, dot1-G401V, dot1Δ), while still preventing replication fork collapse, allow cell cycle progression and TLS-dependent replication across damage. Consistent with this idea, analysis of MMS-treated cells lacking the Rad53 phosphatases Pph3 an Ptc2 suggested that a graded response to the level of Rad53 phosphorylation occurs controlling replication fork restart [63], [64]. These authors propose that a cycle of Rad53 activation and deactivation coordinates DNA repair with TLS-dependent replication fork progression through damaged DNA by a mechanism involving the Cdc7-Dbf4 kinase activity [64]. Indeed, the cooperation of a functional Rad53-dependent checkpoint response with multiple pathways involving base excision repair, recombination and DNA damage tolerance has been shown to be crucial for a proper cellular response to alkylated DNA [65], [66]. Our analysis of the dot1Δ rad53-HA double mutant suggests that when Rad53 activity drops below a critical threshold level, damaged replication forks would irreversibly collapse, like in rad53Δ [67], [68], resulting in cell death and pronounced MMS sensitivity (Fig. 8). We note, however, that in contrast with rad53Δ, the low amounts of the Rad53-HA protein in the dot1Δ rad53-HA mutant must be sufficient to support viability in undamaged cells.