Extended Data Figure 7. Further evidence for two distinct catalytic metal sites during branching.

a–d, Branching requires two catalytic divalent metals (a). The 5′ splice site pro-R_p oxygen and the U80 pro-S_p oxygen interact with a metal distinct from the metal that interacts with the 5′ splice site pro-R_p oxygen and the U80 pro-R_p and G78 pro-S_p oxygens (b–d). See Supplementary Note 7 for discussion. Spliceosomes were assayed for as in Fig. 3e; where indicated, MgCl₂ or MnCl₂ were present at 1mM. Curves represent Hill fits to the data. The data in a are reproduced here from Fig. 3e to aid comparison. In panels a–d, values are averages; error bars, s.d. (n=3). Cd²⁺ was limiting in b to sensitize the assay to binding of a second metal, and Cd²⁺ was saturating in c to show that G78-PS(S_p) and U80-PS(S_p) have the potential to be rescued at levels comparable to U6 WT. Panel d shows that U80-PS(S_p) eliminates the affect of Mn²⁺ on the titration curves. e–j, Metal binding during branching for different combinations of sulfur substitutions in the substrate and U6 in the presence of the indicated metals, as reflected by the data in panels a–d. Panels e and f reflect data in a,b,c; g and h, data in b,c; i and j, data in b,c,d. Relevant U6 ligands are colored red in each panel; the nucleophile is colored orange. Metals are colored magenta (Cd2+) and blue (Mn2+), and their interactions with specific U6 ligands are depicted as dashed lines, with differential shading intensity meant to illustrate differences in the expected strength of interaction with oxygen vs. sulfur, as inferred from studies with model compounds38,61. Shading of metals bound at M1 and M2 is further adjusted to reflect experimental observations. Panels for G78-PS(Sp) would look the same as those for U80-PS(Rp) (g and h).