Fig. 5: LytMCD and LssCD interact with the PG sacculus.

a Samples containing LssCD and LytMCD bound to the intact, polymerised sacculus visualised by SDS-PAGE electrophoresis. Lytic activity of enzymes was evaluated in turbidity reduction assay against S. aureus TF5303 wild-type and TF5311 mutant during 1 h. The experiment was repeated three independent times and molecular weight marker is provided in fig. S10. b Schematic representation of the PG sacculus used for the protein pull-down assays. c Comparison of ¹H, ¹⁵N-correlation spectra of uniformly ²H, ¹³C, ¹⁵N-labelled LssCD and LytMCD free in solution (grey) and in the presence of intact, polymerised PG (crimson) from S. aureus TF5311. d CSP mapping on protein structure of Lss (upper panel) and LytM (lower panel) catalytic domains. CSPs superior to 1 standard deviation (SD > 0.23 ppm for LssCD; >0.16 ppm for LytMCD coloured in salmon) or 2 standard deviations (>0.35 ppm for LssCD; >0.23 ppm for LytMCD coloured in crimson) are displayed on the surface representation of respective proteins. Zn(II) is depicted as blue-grey sphere. e Combined H and N chemical shift perturbations calculated for each residue from the data displayed in panel c as the weighted-average distance between the resonance position of the free form of LytMCD or LssCD and its equivalent position in the form bound to intact sacculus. Bars are colour-coded with CSPs amplitudes as in panel d. The loop regions are highlighted in grey. Source data are provided as a Source Data file. f LytMCD or LssCD:hexadimuropeptide best energy model from the best HADDOCK cluster of solutions. Residues predicted to form hydrogen bonds with the ligand identified in at least 50% of the structures from the cluster of the best energy score are presented as sticks. g Comparison of HADDOCK models for the residues interacting with D-Ala-D-Ala (green colour) and glycan chain (grey colour) regions. For clarity, the glycan chain region was presented for LytMCD models only. Abbreviations: L loop.