DNET: A Graph-Based Tool and Workflow for Dynamic Hydrogen-Bond Networks and Applications for Visual Rhodopsins

Area of Science:

• Biochemistry and structural biology
• Computational biology and bioinformatics
• Molecular biophysics

Background:

• G Protein-Coupled Receptors (GPCRs) are crucial for cellular signaling and are key drug targets.
• GPCR activation involves structural changes transmitted through transmembrane domains, influenced by dynamic hydrogen-bond networks.
• Understanding these networks is vital for elucidating GPCR function and drug design.

Purpose of the Study:

• Introduce DNET, a novel graph-based computational tool for analyzing dynamic protein-water hydrogen-bond networks.
• Characterize the H-bond network dynamics and pKa fluctuations in jumping spider rhodopsin 1 (JSR-1).
• Investigate the impact of mutations on the JSR-1 H-bond network and its functional implications.

Main Methods:

• Developed DNET, a portable Python tool for processing simulation trajectories and computing dynamic H-bond networks.
• Integrated DNET with PROPKA to analyze pKa fluctuations within H-bond networks.
• Applied DNET to study wild-type and mutated JSR-1 proteins, analyzing UV-vis spectroscopy and H-bond dynamics.

Main Results:

• DNET efficiently computes dynamic protein-water H-bond networks and provides detailed residue-level analyses.
• JSR-1 exhibits complex retinal H-bond network dynamics with single and multiple conformational modes.
• Mutations in JSR-1 alter the electrostatic environment of the retinal Schiff base and its H-bond network.

Conclusions:

• DNET is a valuable tool for dissecting complex H-bond dynamics in GPCRs.
• The study reveals intricate H-bond network dynamics and associated pKa fluctuations in a visual rhodopsin.
• Mutations significantly impact the H-bond network, offering insights into rhodopsin function and potential therapeutic strategies.