From the End to the Middle

de Amitabha Gupta

Accurate DNA replication is essential for proper cellular growth and requires an adequate and balanced supply of dNTPs. In Saccharomyces cerevisiae, de novo dNTP synthesis through nucleotide reduction by the Ribonucleotide reductase (RNR) enzyme is the sole method of production. Hence, RNR activitity is highly regulated via allosteric control, transcriptional control, differential localization of subunits, and direct inhibition of the large subunit, Rnr1, by Sml1. Loss of RNR regulation results in increased mutation due to an imbalance or an absolute change in the dNTP levels in cells. In this study, I describe how mutants in dNTP regulation, including sml1∆, play a role in telomere length homeostasis. Reduction in total dNTP concentration results in a modest decrease in telomere length, while altering the ratios between the four dNTPs has a much more pronounced effect. The altered telomere lengths correlate with the relative amount of dGTP and are dependent on telomerase. At reduced levels of relative dGTP, telomerase repeatedly stalls and dissociates from telomeres, thereby leading to shorter telomeres. Conversely, with elevated relative dGTP levels, telomerase is able to processively add nucleotides and even shows low levels of repeat addition processivity. The correlation between telomerase activity and dGTP is conserved in human telomerase, which shows increased repeat addition processivity at increased dGTP concentrations. Thus, telomere length homeostasis is also sensitive to dNTP regulation in the cell via a conserved dependence on dGTP. RNR regulation is, however, relaxed in the cell following DNA damage to allow for an increase in dNTP levels to repair the damage. In response to various forms of damage, Rad53 and Dun1 are activated and then phosphorylate numerous downstream targets, including the Rnr1 inhibitor Sml1. In this study, it was shown that the phosphorylation of Sml1 triggers its ubiquitylation by the Rad6-Ubr2-Mub1 ubiquitin ligase complex and subsequent degradation by the 26S proteasome. Furthermore, I was able to identify novel genes involved in the degradation of Sml1. Of the genes identified, many are involved in the spindle assembly checkpoint (SAC), cohesin establishment, and kinetochore integrity. The loss of SML1 in mutants of these genes resulted in synthetic growth defects that were not due to the loss of dNTP regulation, indicating a second dNTP-independent function for Sml1. Analysis of the double mutants revealed elevated chromosome loss and aberrant spindle dynamics, pointing to a role for Sml1 in the spindle/kinetochore. Through analysis of kinetochore assembly kinetics, Sml1 was found to be the functional human Mis18α ortholog involved in timely establishment of the kinetochore. Thus, Sml1 has a novel structural function at the kinetochore in addition to its role in dNTP regulation.