Correlation in Domain Fluctuations Navigates Target Search of a Viral Peptide along RNA

Abstract

Biological macromolecules often exhibit correlations in fluctuations involving distinct domains. This study decodes their functional implications in RNA–protein recognition and target-specific binding. The target search of a peptide along RNA in a viral TAR–Tat complex is closely monitored using atomistic simulations, steered molecular dynamics simulations, free energy calculations, and a machine-learning-based clustering technique. An anticorrelated domain fluctuation is identified between the tetraloop and the bulge region in the apo form of TAR RNA that sets a hierarchy in the domain-specific fluctuations at each binding event and that directs the succeeding binding footsteps. Thus, at each binding footstep, the dynamic partner selects an RNA location for binding where it senses a higher fluctuation, which is conventionally reduced upon binding. This event stimulates an alternate domain fluctuation, which then dictates sequential binding footstep/s and thus the search progresses. Our cross-correlation maps show that the fluctuations relay from one domain to another specific domain until the anticorrelation between those interdomain fluctuations sustains. Artificial attenuation of that hierarchical domain fluctuation inhibits specific RNA binding. The binding is completed with the arrival of a few long-lived water molecules that mediate slightly distant RNA–protein sites and finally stabilize the overall complex. The study underscores the functional importance of naturally designed fluctuating RNA motifs (bulge, tetraloop) and their interplay in dictating the directionality of the search in a highly dynamic environment.

Influence of ion and hydration atmospheres on RNA structure and dynamics: insights from advanced theoretical and computational methods

“…We have investigated various systems ranging from simple duplex (BIV TAR RNA), triplex (BWYV RNA PK) to numerous complex folded RNA structures (flavivirus sfRNA, SARS-CoV-2 RNA PK, SAM-I riboswitch aptamer, to name a few). 12,19,51,73,139 In all these systems we have primarily observed three distinguishable modes of interaction, representing their characteristic peaks in the RDF plot as shown in Fig. 1A.…”

The Three-Dimensional Reference Interaction Site Model Approach as a Promising Tool for Studying Hydrated Viruses and Their Complexes with Ligands

“…Like any other study, computational studies are aimed at understanding the mechanism of viral infection as it relates to the development of new small-molecule drugs that inhibit virus-host cell binding. A wide range of approaches are actively involved in non-empirical studies of viruses-from a bioinformatics analysis (see, for instance, [4,5]) to coarse-grained computational modeling [6] and long-timescale allatom MD simulations (see, for instance, [7][8][9][10]) as well as a fashionable machine-learningbased clustering technique [11] including deep learning [12,13]. A detailed review of various computational methods and their combinations can be found in [3], where such aspects as molecular biophysics, bioinformatics, cheminformatics, machine learning, and mathematics are discussed.…”

RNA’s Dynamic Conformational Selection and Entropic Allosteric Mechanism in Controlling Cascade Protein Binding Events

“…In fact, using atomistic simulations, recently, we identified intricate allosteric cross-talk between the distant bulge and loop in BIV TAR RNA. 30 There, we have shown how the hierarchical fluctuations of the loop and bulge guide a protein to find the target binding site along RNA.…”