Contributors: Chester, Tank, Alexander
Additional interacting residues
Investigated residues beyond the catalytic glutamates (reasoning: glutamates unlikely to act alone in binding/catalysis).
- GH1: All cellobiose-enzyme interactions were hydrogen bonds (interacting residues tabulated in source data).
- GH10: Mix of hydrophobic interactions and hydrogen bonds (tabulated).
PLIP flagged minor structural issues in both structures during analysis; confirmed via online sources that as long as the downloaded PDB and cellobiose positions are unchanged, this does not affect interpretation.
Notenboom et al. cross-reference
Cross-referenced docking results against Notenboom et al.’s crystallography/mutation paper on Cex cellulose/xylan specificity:
That paper identified GLN87 as a steric hindrance factor for Cex. Our docking prediction instead suggests GLN87 contributes a hydrogen bond to cellobiose binding.
Two hypotheses considered:
- Hypothesis 1 (likely): The paper’s mutation showed no change in K_d because steric hindrance removal and hydrogen bond removal cancel each other out. This explains the null result without requiring the structures to differ.
- Hypothesis 2 (rejected): AlphaFold3 structure differs meaningfully from the crystal structure. Rejected — RMSD between the two is low enough to rule this out (validated June 18).
PoseView cross-check
PoseView used as an alternative to LigPlot+ to independently verify interactions:
- GH1: Different interacting residues identified vs PLIP, but the key catalytic glutamates were still present in both analyses.
- GH10: PoseView showed only GLU233 interacting — GLU127 absent from the PoseView interaction diagram. This is a possible explanation for GH10’s lower beta-glucosidase activity relative to GH1: GLU127 may not be functioning as effectively as expected, hindering cellobiose binding or catalysis.
Flagged for further investigation: is GLU127 inhibited by steric hindrance, or by charges from non-interacting neighbouring residues?