Fig. 4 in A single residue determines substrate preference in benzylisoquinoline alkaloid N-methyltransferases

Fig. 4. Superimposition of residue 204 identity on a maximum-likelihood phylogenetic analysis of N-methyltransferase gene candidates. The cladogram was adapted from that deposited in DRYAD by Hagel et al. (2015b) (https://doi.org/10.5061/dryad.bh276/18) and the corresponding sequences are described in Supplementary Dataset 4. CjCNMT was not included in the original analysis but is most similar to TfCNMT (86% amino acid identity). Key nodes are labelled with support values representing the number of 1000 bootstrapped trees in which the associated taxa clustered together. Putative BIA NMT sequences were aligned by Clustal Omega and the identity of the residue corresponding to E204 in CjCNMT was mapped onto each taxa (glutamic acid, green; glycine, purple; alanine, cyan; glutamine, yellow; serine; orange). BIA NMTs for which in vitro functional characterization data was available are indicated with black arrows and those subjected to yeast-based in vivo characterization in this work are indicated with white arrows. Prior to this report, sequence-based prediction of BIA NMT sub-functionalization was reliable for TNMT-like enzymes only due to their clustering as a single well-defined clade. CNMT-, RNMT- and PavNMT-like sequences are distributed amongst several less well-defined clades.