Large-Scale Analysis of Hydrogen Bond Interaction Patterns in Protein–Ligand Interfaces

Nittinger, Eva; Inhester, Therese; Bietz, Stefan; Meyder, Agnes; Schomburg, Karen T.; Lange, Gudrun; Klein, Robert J.; Rarey, Matthias

doi:10.1021/acs.jmedchem.7b00101

Cited by 78 publications

(72 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In general, if the protein has already been targeted by covalent inhibitors, the library can be compiled by collecting commercially available ligands and/or by enumerating synthetically accessible compounds bearing the same warhead type as the crystallized inhibitor.F or JAK3, the vast majority of known covalent inhibitors bind through an acrylamide warhead, while for KRas G12C ,m ost of the known potent covalent inhibitors bind through either an acrylamide or av inylsulfonamide warhead. [37,38] In a DUck simulation, the ligands are pulled from 2.5 to 5.0 relative to the defined H-bond interaction point in the protein, during a user-defined number of MD and SMD replicas. [16] Prior to docking simulations, LigPrep by Schrçdinger was used to prepare 3D conformations from SMILES codes and to generate tautomeric and ionization states at pH 6-8 while retaining specified chiralities.…”

Section: Methodsmentioning

confidence: 99%

“…For DUck, only H-bonds are assessed, as they are known to be key contributors to affinity in many targets. [37,38] In a DUck simulation, the ligands are pulled from 2.5 to 5.0 relative to the defined H-bond interaction point in the protein, during a user-defined number of MD and SMD replicas. The force necessary to pull out the ligand is then used to calculate aw ork value (W QB ), which corresponds to the strength of the H-bond.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

DUckCov: a Dynamic Undocking‐Based Virtual Screening Protocol for Covalent Binders

et al. 2019

View full text Add to dashboard Cite

Thanks to recent guidelines, the design of safe and effective covalent drugs has gained significant interest. Other than targeting non‐conserved nucleophilic residues, optimizing the noncovalent binding framework is important to improve potency and selectivity of covalent binders toward the desired target. Significant efforts have been made in extending the computational toolkits to include a covalent mechanism of protein targeting, like in the development of covalent docking methods for binding mode prediction. To highlight the value of the noncovalent complex in the covalent binding process, here we describe a new protocol using tethered and constrained docking in combination with Dynamic Undocking (DUck) as a tool to privilege strong protein binders for the identification of novel covalent inhibitors. At the end of the protocol, dedicated covalent docking methods were used to rank and select the virtual hits based on the predicted binding mode. By validating the method on JAK3 and KRas, we demonstrate how this fast iterative protocol can be applied to explore a wide chemical space and identify potent targeted covalent inhibitors.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

DUckCov: a Dynamic Undocking‐Based Virtual Screening Protocol for Covalent Binders

et al. 2019

View full text Add to dashboard Cite

show abstract

“…To assess the quality of directionality profiles calculated with B3LYP/cc-pVDZ we used MP2/aug-cc-pVTZ to compute domes for donor and acceptor sites of water (15,75), imidazole (25,92), methyl-acetamide (18,90), methyl-thioacetamide (not numbered as donor; compound 62 as acceptor), phenol (26,55), and TMAO (97). Results are summarized in Supplementary Table S2…”

Section: Supplementary Informationmentioning

confidence: 99%

“…These analyses are based on the assumption that HB strength and directionality can be inferred from structures because "interaction geometries are conserved with fidelities that correspond to their strengths" 9,12 . Statistical distributions of such observed geometries can be then translated into chemical potentials 13,14 Over the past decades, several chemical groups were studied using statistical approaches, including halogens [15][16][17] , sulfur and oxygen [18][19][20] , weak and strong donors 8 , among others 17,[21][22][23][24][25] . Despite providing invaluable information, retrospective structural analysis has several issues that limit the coverage of HB properties.…”

Section: Introductionmentioning

confidence: 99%

“…2). Such overall increased tolerance is likely responsible of the high propensity of proteins backbone to form stable HBs that are co-linear with the C=O bond such as in α-helices 5,25 . When oxygen is bound directly to sulfur, as in DMSO 94, the partial triple bond characteristics 40 increases acceptor strength to about 140% that of water, and further reduce directionality (D [HF] = 0.19, Fig.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Charting Hydrogen Bond Anisotropy

Santos-Martins¹,

Forli

2019

Preprint

View full text Add to dashboard Cite

Hydrogen bond (HB) is an essential interaction in countless phenomena, and regulates the chemistry of life. HBs are characterized by two main features, strength and directionality, with a high degree of heterogeneity across different chemical groups. These characteristics are dependent on the electronic configuration of the atoms involved in the interaction, which, in turn, is influenced strongly by the molecular environment where they are found. Studies based on the analysis of HB in solid phase, such as X-ray crystallography, suffer from significant biases due to the packing forces. These will tend to better describe strong HBs at the expenses of weak ones, which are either distorted or under-represented. Using quantum mechanics (QM), we calculated interaction energies for about a hundred acceptor and donors, in a rigorously defined set of geometries. We performed about 180,000 independent QM calculations, covering all relevant angular components, and mapping strength and directionality in a context free from external biases, with both single-site and cooperative HBs. We show that by quantifying directionality, there is not correlation with strength, and therefore these two components need to be addressed separately. Results demonstrate that there are very strong HB acceptors (e.g., DMSO) with nearly isotropic interactions, and weak ones (e.g., thioacetone) with a sharp directional profile. Similarly, groups can have comparable directional propensity, but be very distant in the strength spectrum (e.g., thioacetone and pyridine). These findings have implications for biophysics and molecular recognition, providing new insight for chemical biology, protein engineering, and drug design. The results require rethinking the way directionality is described, with implications for the thermodynamics of HB. hydrogen bond | drug design | force field | non-covalent interaction Correspondence: forli@scripps.edu ResultsThe dataset contains 34 donors ( Fig.S1) and 67 acceptors ( Fig.S2, Methods). For the entire dataset, calculations were performed using B3LYP density functional with the cc-pVDZ basis set and the counterpoise correction, while MP2/aug-cc-pVTZ with counterpoise correction was used for smaller subsets of complexes. The validation of these two methods, their differences and their effect on the results has been discussed in Supplementary Methods. Strength.We measured the strength of HBs on the 34 donors ( Fig.S1) and 67 acceptors, performing B3LYP/cc-pVDZ calculations of donor-acceptor dimers interacting in optimal ge-July 5, 2019 | 1-20

show abstract

Structure‐Based Drug Design

Ferreira

Andricopulo

2021

Burger's Medicinal Chemistry and Drug Discovery

View full text Add to dashboard Cite

Structure‐based drug design (SBDD) is the backbone of modern pharmaceutical research and development. Its origins date back to the 1950s and 1960s when the first X‐ray protein structures and pioneering studies correlating ligand–receptor binding and drug activity came out. With this foundation, seminal work on computer graphics enabled the first visualization and manipulation of virtual protein structures. These milestones opened the avenues for the growth of protein crystallography and molecular modeling, which thenceforth evolved side by side to culminate in modern SBDD. Today, SBDD can handle molecular targets previously regarded as intractable or undruggable, such as protein–protein interactions. Another advance coming from SBDD, fragment‐based drug discovery has excelled in optimizing weak ligands into drug‐like compounds. SBDD has provided groundbreaking drugs for many therapeutic areas, including highly complex conditions and diseases that were previously considered a death sentence. In fact, a substantial fraction of the drugs approved recently have been developed with the support of structural data. This article provides a perspective on the central aspects of this exciting field and explores the fundamentals of molecular modeling and its integration with X‐ray diffraction. The discussion of key concepts is followed by representative examples of the practice of SBDD.

show abstract

Large-Scale Analysis of Hydrogen Bond Interaction Patterns in Protein–Ligand Interfaces

Cited by 78 publications

References 47 publications

DUckCov: a Dynamic Undocking‐Based Virtual Screening Protocol for Covalent Binders

DUckCov: a Dynamic Undocking‐Based Virtual Screening Protocol for Covalent Binders

Charting Hydrogen Bond Anisotropy

Structure‐Based Drug Design

Contact Info

Product

Resources

About