Estimation of anisotropic bending rigidities and spontaneous curvatures of crescent curvature-inducing proteins from tethered-vesicle experimental data

TL;DR

This study estimates the anisotropic bending rigidities and spontaneous curvatures of crescent-shaped proteins from tethered-vesicle data, revealing differences between I-BAR and N-BAR domains and suggesting cluster formation in N-BAR proteins.

Contribution

The paper introduces a mean-field theoretical approach to quantify anisotropic bending properties of crescent curvature-inducing proteins from experimental data.

Findings

I-BAR domain fits well with a single anisotropic parameter set

Isotropic model fits poorly compared to anisotropic model

N-BAR domain shows deviations suggesting cluster formation

Abstract

The Bin/amphiphysin/Rvs (BAR) superfamily proteins have a crescent binding domain and bend biomembranes along the domain axis. However, their anisotropic bending rigidities and spontaneous curvatures have not been experimentally determined. Here, we estimated these values from the bound protein densities on tethered vesicles using a mean-field theory of anisotropic bending energy and orientation-dependent excluded volume. The dependence curves of the protein density on the membrane curvature are fitted to the experimental data for the I-BAR and N-BAR domains reported by C. Prevost et al. Nat. Commun. 6, 8529 (2015) and F.-C. Tsai et al. Soft Matter 17, 4254 (2021), respectively. For the I-BAR domain, all three density curves of different chemical potentials exhibit excellent fits with a single parameter set of anisotropic bending energy. When the classical isotropic bending energy is used instead, one of the curves can be fitted well, but the others exhibit large deviations. In contrast, for the N-BAR domain, two curves are not well-fitted simultaneously using the anisotropic model, although it is significantly improved compared to the isotropic model. This deviation likely suggests a cluster formation of the N-BAR domains.