AF-M-informed model of how the POLE1 catalytic domain is positioned near CMGs exit channel by the interaction of CTF18 (red) with the winged-helix domain of MCM7 (cyan) and the POLE1 catalytic domain. The model was assembled as follows: An AF-M prediction of POLE1 (C-terminal non-catalytic domain shown as gray ribbon, N-terminal catalytic domain shown as light blue ribbon) was aligned on the C-terminal, non-catalytic lobe of POLE1 in the cryo-EM structure of the human replisome (PDB:7PLO,69of which only MCM2-7, CDC45, and GINS are shown, POLE1 hidden). To model the primer template, the structure of yeast POLE1 catalytic domain with a primer template (PDB:4M8O) was aligned on the POLE1-NTD shown (yeast POLE1 hidden; primer template shown). We also generated an AF-M prediction of a complex of CTF18, CTF8, DCC1, and the NTD of POLE1 (which matches key features of an analogous experimental structure, PDB:6S2E70) and aligned it on the POLE1-NTD shown. This revealed that the CTF18 (yellow)-DCC1 (orange)-CTF8 (orange) complex binds the distal side of POLE1-NTD. Separately, MCM3, MCM7, and CTF18 were folded together and aligned on MCM7 from 7PLO. This shows that a movement of 30 (yellow arrow) would superimpose the CTF18 aligned on MCM7 (red) and the CTF18 aligned on POLE1-NTD (yellow). Given the reported flexibility between the NTD and CTD of POLE174 and some predicted flexibility between the NTD and CTD of CTF18 (https://alphafold.ebi.ac.uk/entry/Q8WVB6), CTF18 should be able to bind MCM7 and POLE1-NTD simultaneously. In this way, CTF18 would tether the POLE1-NTD near the rear exit channel of CMG, with the leading strand template (dark blue strand) being fed into the active site.
