Short hydrogen bonds (SHBs), which have the donor-acceptor separations below 2.7 Å, exhibit prominent covalent characters. Although SHBs are prevalent in biological systems and are associated with various biological processes, ranging from enzymatic catalysis to bacterial photosensory, quantum effects are largely neglected in the discussion of their biological functions. In addition, exact determination of SHBs in proteins requires the biomolecular structures at atomic resolution, creating a challenge to the structural determination techniques such as X-ray crystallography, neutron scattering and NMR spectroscopy. To address this issue, we have conducted a statistical analysis of the top 1% highest-quality biomolecules in the Protein Data Bank to uncover the structural and chemical preferences of SHBs. Based on these features, we have further developed a novel machine learning model for the effective prediction of SHBs in proteins, which will facilitate the experimental and computational refinement of protein structures with moderate or low resolution. By performing <i>ab initio</i> calculations on 3665 SHBs that formed from the side chains of amino acids, we have unraveled the quantum nature of the SHBs and demonstrated how the equilibrium positions of the proton and the energy barriers for sharing the proton in the hydrogen bond depend on the heteroatom separations. Combining <i>ab initio</i> calculations and energy decomposition analysis, we have further elucidated the key role of electronic quantum effects, such as charge transfer and dispersion, in stabilizing the SHBs and making the Tyr-Asp side chain-side chain hydrogen bonds highly probable to form SHBs. Our calculations will contribute to the development of new force fields of proteins to include the quantum mechanical description of SHBs and promote the design of novel proteins and biomaterials with enhanced physical properties and biological functions.

Structures, machine learning predictions and quantum effects of short hydrogen bonds

ACS Fall 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.