Stacking Interactions of Druglike Heterocycles with Nucleobases

Area of Science:

• Computational chemistry
• Drug discovery
• RNA-targeting therapeutics

Background:

• Stacking interactions between small molecules and RNA are crucial for developing RNA-targeting inhibitors.
• Understanding these interactions is vital for designing effective therapeutic agents.

Purpose of the Study:

• To computationally analyze stacking interactions between druglike heterocycles and natural nucleobases.
• To develop a predictive model for heterocycle-nucleobase stacking interaction strength.
• To improve the accuracy of computational methods for predicting these interactions.

Main Methods:

• Computational analysis of 54 druglike heterocycles and natural nucleobases.
• Symmetry-adapted perturbation theory to study interaction strengths.
• Development of a multivariate predictive model using electrostatic potential descriptors.
• Modification of molecular mechanics force fields for improved accuracy.

Main Results:

• Heterocycle choice significantly tunes stacking interaction strength and geometry.
• Electrostatic and dispersion effects are primary modulators of interaction strength.
• A predictive model accurately estimates maximum stacking interaction strengths for 1854 heterocycles.
• Modified force fields reduce prediction errors in stacking interaction energies.

Conclusions:

• Computational modeling provides a powerful approach to understanding and predicting heterocycle-nucleobase stacking interactions.
• The developed predictive model and improved force fields can accelerate the design of novel RNA-targeting drugs.
• Optimized stacking interactions were observed in a case study with ribocil and a bacterial riboswitch.