Slow Transition Path Times Reveal a Complex Folding Barrier in a Designed Protein

Abstract

De-novo designed proteins have received wide interest as potential platforms for nano-engineering and biomedicine. While much work is being done in the design of thermodynamically stable proteins, the folding process of artificially designed proteins is not well-studied. Here we used single-molecule force spectroscopy by optical tweezers to study the folding of ROSS, a de-novo designed 2x2 Rossmann fold. We measured a barrier crossing time in the millisecond range, much slower than what has been reported for other systems. While long transition times can be explained by barrier roughness or slow diffusion, we show that isotropic roughness cannot explain the measured transition path time distribution. Instead, this study shows that the slow barrier crossing of ROSS is caused by the population of three short-lived high-energy intermediates. In addition, we identify incomplete and off-pathway folding events with different barrier crossing dynamics. Our results hint at the presence of a complex transition barrier that may be a common feature of many artificially designed proteins.

What can we learn from transition path time distributions for protein folding and unfolding?

“…Previous experiments have indicated the existence of one or multiple wells along the reaction coordinate. 36,37,42 What is the physics and chemistry of such wells? This is difficult to answer and calls for much more extensive modelling of the unfolded to folded transition free energy surface.…”

Barrier Crossing Dynamics from Single-Molecule Measurements

“…(C) Importantly, roughness of the free energy landscape alone does not necessarily lead to a broad distribution p TP ( t ): The coefficient of variation C is always less than 1 for diffusive dynamics in any one-dimensional potential, no matter how rugged it is. For a recent discussion of the effect of energetic roughness of the transition path times in 1D potentials, see refs and .…”

The loader complex Scc2/4 forms co-condensates with DNA as loading sites for cohesin

“…For 30 nM and 50 nM Scc2/4, 10 and 13 tethers were analyzed, respectively. Using an inversion of the eWLC model (Wang et al, 1997), 𝜉(𝐹) can then be converted into a contour length 𝐿 𝑐 (Mehlich et al, 2020).…”