Back

Atomic Conformational Dynamics and Actin-Crosslinking Function of Alpha-Actinin Revealed by SimHS-AFMfit

Ngo, K. X.; Sumikama, T.; Vuillemot, R.; Nguyen, H. G.; Uyeda, T.; Le, N. T. P.; Grudinin, S.

2025-04-10 biophysics
10.1101/2025.04.06.647477 bioRxiv
Show abstract

Many molecular systems, such as intrinsically disordered proteins and flexible multi-domain complexes, are highly dynamic and often inaccessible to conventional X-ray crystallography or cryo-EM due to their conformational heterogeneity and flexibility. As a result, resolving their atomic-level dynamics remains a significant challenge. In this study, we present SimHS-AFMfit-MD, an integrative framework that combines high-speed atomic force microscopy (HS-AFM), molecular dynamics (MD) simulations, and AFMfit-based structural modeling to reconstruct dynamic protein conformations at atomic resolution. Using alpha-actinin, an actin crosslinking protein, as a challenging test system, we show that AFMfit guided by nonlinear normal mode analysis (AFMfit-NMA) enables accurate structural fitting, while guiding AFMfit with MD trajectories (AFMfit-MD) further enhances the flexible fitting performance, achieving closer agreement with unbiased all-atom MD simulation results. This strategy allows us to convert thousands of three-dimensional HS-AFM images into atomic-scale conformational ensembles, revealing the twisting and bending transitions underlying Ca{superscript 2}-bound and Ca{superscript 2}-unbound alpha-actinin. Together, our results establish a hybrid computational-experimental approach that bridges the spatial and, to some extent, temporal resolution gaps between simulation and imaging, paving the way for real-time visualization of protein conformational dynamics at the atomic scale. Graphic Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=77 SRC="FIGDIR/small/647477v3_ufig1.gif" ALT="Figure 1"> View larger version (25K): org.highwire.dtl.DTLVardef@147c1aforg.highwire.dtl.DTLVardef@1fd0036org.highwire.dtl.DTLVardef@118da3forg.highwire.dtl.DTLVardef@a062c3_HPS_FORMAT_FIGEXP M_FIG C_FIG

Published in Nano Letters (predicted rank #17) · training set

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.