Regulation of HLTF-mediated PCNA polyubiquitination by RFC and PCNA monoubiquitination levels determines choice of damage tolerance pathway

Abstract

DNA-damage tolerance protects cells via at least two sub-pathways regulated by proliferating cell nuclear antigen (PCNA) ubiquitination in eukaryotes: translesion DNA synthesis (TLS) and template switching (TS), which are stimulated by mono- and polyubiquitination, respectively. However, how cells choose between the two pathways remains unclear. The regulation of ubiquitin ligases catalyzing polyubiquitination, such as helicase-like transcription factor (HLTF), could play a role in the choice of pathway. Here, we demonstrate that the ligase activity of HLTF is stimulated by double-stranded DNA via HIRAN domain-dependent recruitment to stalled primer ends. Replication factor C (RFC) and PCNA located at primer ends, however, suppress en bloc polyubiquitination in the complex, redirecting toward sequential chain elongation. When PCNA in the complex is monoubiquitinated by RAD6-RAD18, the resulting ubiquitin moiety is immediately polyubiquitinated by coexisting HLTF, indicating a coupling reaction between mono- and polyubiquitination. By contrast, when PCNA was monoubiquitinated in the absence of HLTF, it was not polyubiquitinated by subsequently recruited HLTF unless all three-subunits of PCNA were monoubiquitinated, indicating that the uncoupling reaction specifically occurs on three-subunit-monoubiquitinated PCNA. We discuss the physiological relevance of the different modes of the polyubiquitination to the choice of cells between TLS and TS under different conditions.

Spatiotemporal regulation of PCNA ubiquitination in damage tolerance pathways

“…The chain-formation activity of HLTF is absolutely DNA-dependent (Masuda, Suzuki, Kawai, Hishiki, et al 2012;Masuda et al 2018). In particular, structures with 3 0 -OH primer ends have the strongest stimulatory effect, even if the primer end has a one base flap (Masuda et al 2018). This property is consistent with the idea that the key structure for the stimulation of the chain-formation activity is the stalled primer end, as stalled primer ends do not always result in stable base pairing.…”

“…Mutations in the SWI2/SNF2 helicase domain compromising the ATPase activity reduce the chain-formation activity. Importantly, the levels of ATPase activity in each mutant correspond to the levels of chain-formation activity, indicating a functional coupling between the RING and SWI2/SNF2 helicase domains (Masuda et al 2018).…”

“…Biochemical studies of PCNA polyubiquitination K63-linked PCNA polyubiquitination by RAD5 in S. cerevisiae, or in humans by HLTF or SHPRH with the RAD6-RAD18 and MMS2-UBC13 complexes, has been characterized in vitro (Unk et al 2006(Unk et al , 2008Carlile et al 2009;Parker and Ulrich 2009). The reaction mechanisms of human HLTF have been intensively analyzed (Masuda, Suzuki, Kawai, Hishiki, et al 2012;Masuda et al 2018).…”

“…The chain-formation activity of HLTF is absolutely DNA-dependent (Masuda, Suzuki, Kawai, Hishiki, et al 2012;Masuda et al 2018). In particular, structures with 3 0 -OH primer ends have the strongest stimulatory effect, even if the primer end has a one base flap (Masuda et al 2018).…”

“…This property is consistent with the idea that the key structure for the stimulation of the chain-formation activity is the stalled primer end, as stalled primer ends do not always result in stable base pairing. The N-terminal HIRAN (HIP116 Rad5p N-terminal) domain, a 3 0 -OH-binding module (Achar et al 2015;Hishiki et al 2015;Kile et al 2015), is required for primer end-mediated stimulation (Masuda et al 2018); however, it is dispensable for the E3 activity (Achar et al 2015;Masuda et al 2018). Mutations in the SWI2/SNF2 helicase domain compromising the ATPase activity reduce the chain-formation activity.…”