Structural Basis for the Recruitment of Ctf18-RFC to the Replisome

Abstract

Ctf18-RFC is an alternative PCNA loader which plays important but poorly understood roles in multiple DNA replication-associated processes. To fulfill its specialist roles, the Ctf18-RFC clamp loader contains a unique module in which the Dcc1-Ctf8 complex is bound to the C terminus of Ctf18 (the Ctf18-1-8 module). Here, we report the structural and functional characterization of the heterotetrameric complex formed between Ctf18-1-8 and a 63 kDa fragment of DNA polymerase ɛ. Our data reveal that Ctf18-1-8 binds stably to the polymerase and far from its other functional sites, suggesting that Ctf18-RFC could be associated with Pol ɛ throughout normal replication as the leading strand clamp loader. We also show that Pol ɛ and double-stranded DNA compete to bind the same winged-helix domain on Dcc1, with Pol ɛ being the preferred binding partner, thus suggesting that there are two alternative pathways to recruit Ctf18-RFC to sites of replication.

A stable leading strand polymerase/clamp loader complex is required for normal and perturbed eukaryotic DNA replication

“…This showed that Ctf18-1-8 forms a tight complex with Pol2 CAT with a K D of 40 nM (Fig. 1D), over 30-fold stronger than the 1.3 µM K D measured for the truncated Pol2(1-528) construct (Grabarczyk, Silkenat and Kisker, 2018).…”

“…Previously, we showed that yeast Ctf18-1-8 forms a relatively transient complex with a truncated construct of the Pol ε catalytic domain, Pol2(1-528) (Fig, 1A), through the WH3 domain of Dcc1 (Dcc1 WH3 ) and exonuclease domain of Pol2 (Pol2 EXO ) (Grabarczyk, Silkenat and Kisker, 2018). Although this minimal complex could be disrupted in vitro by a structure-guided triple mutation, Dcc1 R367A/R376A/R380A (henceforth Dcc1-3A) (Grabarczyk, Silkenat and Kisker, 2018), a new crystal form of this complex revealed that this interface is highly flexible (Figure S1A). We therefore used cryo-EM to solve the structure of Ctf18-1-8 in complex with the complete catalytic domain of Pol2 (residues 1-1192, henceforth Pol2 CAT ) to 4.2 Å resolution.…”

“…We show this is achieved through an unusual protein-protein mechanism that utilizes electrostatic slippery interfaces to enable high binding plasticity. These low-specificity electrostatic patches explain why it was previously thought that the Ctf18-1-8 complex binds to DNA (Wade et al ., 2017; Grabarczyk, Silkenat and Kisker, 2018).…”

“…This module is separated from the RFC catalytic module by a large predicted unstructured linker (Fig 1A). The Ctf18-1-8 module comprises a triple-barrel domain (TBD) formed from all three proteins and a winged helix hook (WHH), composed of a series of winged-helix domains in the C-terminus of Dcc1 resembling a hook (Grabarczyk, Silkenat and Kisker, 2018; Wade et al ., 2017). Mutations in Ctf18 that disrupt formation of this Ctf18-1-8 module result in similar but not identical phenotypes as entire deletions of Ctf18 (Okimoto et al ., 2016; García-Rodríguez et al ., 2015), suggesting that this module is critical for at least some of the specialized functions of Ctf18-RFC.…”

“…The Pol2 catalytic domain (Pol2 CAT ) is connected by an unstructured linker to the remaining parts of Pol ε, which form an integral part of the CMGE (CMG-Pol ε) core replisome complex (Fig 1A) (Goswami et al ., 2018; Zhou et al ., 2017). It was recently shown that the last winged helix domain, WH3, of Dcc1 assumes a dual function: it can transiently interact with either DNA or a minimal construct of Pol2 (Grabarczyk, Silkenat and Kisker, 2018). However, despite a triple amino acid mutation within the WH3 domain that completely abolishes both interactions in vitro , surprisingly, deletion of WH3 in vivo only causes a small loss of Ctf18-RFC from replicating forks, while a cohesion defect is only seen when both WH2 and WH3 are deleted (Wade et al ., 2017).…”