Rationalizing Diverse Binding Mechanisms to the Same Protein Fold: Insights for Ligand Recognition and Biosensor Design

The engineering of novel protein–ligand binding
interactions,
particularly for complex drug-like molecules, is an unsolved problem,
which could enable many practical applications of protein biosensors.
In this work, we analyzed two engineered biosensors, derived from
the plant hormone sensor PYR1, to recognize either the agrochemical
mandipropamid or the synthetic cannabinoid WIN55,212-2. Using a combination
of quantitative deep mutational scanning experiments and molecular
dynamics simulations, we demonstrated that mutations at common positions
can promote protein–ligand shape complementarity and revealed
prominent differences in the electrostatic networks needed to complement
diverse ligands. MD simulations indicate that both PYR1 protein–ligand
complexes bind a single conformer of their target ligand that is close
to the lowest free-energy conformer. Computational design using a
fixed conformer and rigid body orientation led to new WIN55,212-2
sensors with nanomolar limits of detection. This work reveals mechanisms
by which the versatile PYR1 biosensor scaffold can bind diverse ligands.
This work also provides computational methods to sample realistic
ligand conformers and rigid body alignments that simplify the computational
design of biosensors for novel ligands of interest.