Figure 2: Structural differences between PCGF5 and PCGF4 in complex with RING1B.

(a) Overlay of the two complexes in the asymmetric unit in the crystal structure of RING1B–PCGF5–UbcH5c (chains A, B and C compared with chains D, E and F) is shown at left (root mean squared deviation 0.75 Å over 339 residues, using Coot SSM superpose⁶⁷; the copy consisting of chains A, B and C is used in subsequent figures). The structure of the previously solved RING1B–PCGF4–UbcH5c (PDB code: 3RPG³⁷) is shown at right for comparison. The PCGF4 and PCGF5 complexes are highly similar with an overall r.m.s.d of 1.11 Å over 340 residues. Helices within the PCGF are numbered α1 through α4. The β-strand labelled β1 on the PCGF4 cartoon is absent in PCGF5: insets show close-up views of the hydrogen-bonding pattern between RING1B and the PCGF5 or PCGF4 N termini. (b) Close-up view of the overlay of the two-stranded antiparallel β-sheet from previously solved RING1B–PCGF4 structures^37,38,47. PDB codes are shown in parentheses. (c) Activity comparison of PCGF chimeric complexes exchanging strand β1 and/or helix α1 as monitored by western blot. All PCGFs are in complex with GST–RING1B. ‘P4 S’ (swap) denotes PCGF5_1–17PCGF4_18–109, ‘P4 S strand’ denotes PCGF5_1–8PCGF4_9–109 and ‘P4 S helix’ denotes PCGF4_1–8PCGF5_9–17PCGF4_18–109.