Incongruent end structures of leading and lagging telomeres dictate the nature of end replication problem

Abstract

The end replication problem refers to the incomplete replication of parental DNA at telomeres, a process whose molecular depiction is hampered by the complex nature of telomere ends. Here we recapitulate this process using a synthetic de novo telomere in yeast and delineate distinct molecular fates of telomere ends in vivo. We show that the lagging strand telomere carries a [~]10 nt 3 overhang, while the leading strand telomere has a Yku-protected blunt end, which is prevalent on native telomeres. In addition, RNase H2 but not RNase H1 is mainly responsible for the removal of the last RNA primer. Consistently, in the absence of RNase H2 activity, RNA primer is retained on the lagging strand telomere, attenuating telomere erosion and delaying senescence in telomerase-null cells. These findings highlight incongruent end structures on telomeres and clarify that the primary culprit behind end replication problem is the incompletely replicated lagging strand telomere.