Extended Data Fig. 8: Orientation of the allenylamine (10) in WT enzyme and F231Y variant.

a) MD simulations show that the presence of Tyr231 induces a reorientation of the allene moiety in the active site of F231Y variant. The ∠(C1 – C2 – N1 – C3) dihedral angle measured along independent MD trajectories (3 replicas of 500 ns for each for both the WT and F231Y variant) describes the reorientation of the allene moiety in BesB active site. In the right panel, the representation of the normalized kernel density plot of the monitored dihedral angle along the MD simulations for allenylamine (10) (see the left figure) for WT enzyme (depicted in gray) and F231Y variant (depicted in blue) are shown. b) Representative snapshot corresponding to the most populated clusters extracted from Molecular Dynamics (MD) simulations describing the interactions between allenylimine (10), Arg61, Asp233 and Phe/Tyr231. As shown in the figure, the interaction between Asp233, Arg61 and PLP-phosphate is maintained in both the WT enzyme and F231Y variant. However, repositioning of the allene moiety with respect to Arg61 in F231Y prevents the deprotonation by PLP-phosphate group. The introduction of Tyr231 affects the geometric stability of the different species in the BesB active site and the orientation of the alkene/allene moiety with respect to possible deprotonation groups. While along MD simulation of ketimine (8) Tyr231 shows a similar conformation that the one adopted for Phe231 in the WT enzyme, along the simulations with intermediate-9 and allenylamine (10), residue Tyr231 samples different orientations and interacts directly with the PLP-phosphate group or other residues found in the active site (Asp233 and Asp305). This interaction causes a rearrangement of the species in the active site, as well as affects the stability of the different species. For the F231Y variant, the analysis suggests that the two deprotonation sources -Asp233 and the PLP-phosphate group- (that showed low deprotonation barriers in DFT calculations, see Supplementary Fig. 29) are not close to the Hd of the intermediates for an efficient deprotonation (see Supplementary Figs. 45, 46). These results may explain why the tyrosine at position 231 contributes to trap the allenic intermediate. See additional details in Supplementary Fig. 48.