Mechanical Deformation Accelerates Protein Ageing

Abstract

A hallmark of tissue ageing is the irreversible oxidative modification of its proteins. We show that single proteins, kept unfolded and extended by a mechanical force, undergo accelerated ageing in times scales of minutes to days. A protein forced to be continuously unfolded completely loses its ability to contract by folding, becoming a labile polymer. Ageing rates vary among different proteins, but in all cases they lose their mechanical integrity. Random oxidative modification of cryptic side chains exposed by mechanical unfolding can be slowed by the addition of antioxidants such as ascorbic acid, or accelerated by oxidants. By contrast, proteins kept in the folded state and probed over week-long experiments show greatly reduced rates of ageing. We demonstrate a novel approach whereby protein ageing can be greatly accelerated: the constant unfolding of a protein for hours to days is equivalent to decades of exposure to free radicals under physiological conditions.

Keywords: force spectroscopy; oxidative damage; protein folding; protein structure; single-molecule studies.

Figure 1. Accelerated protein ageing by mechanical unfolding

A) Scheme of a magnetic tweezers setup with a tandem modular protein anchored to a glass surface and a paramagnetic bead. B) MT trace of a protein L octamer. First the protein is completely unfolded at 45 pN (8 unfolding steps are marked by arrow heads), and then refolded at 6.8 pN. Refolding occurs in two phases: an elastic contraction followed by stepwise folding contractions (dashed lines mark each folding level). We use two kinds of force protocols to study protein ageing. The protocol in C) keeps the protein mostly in the folded state, interrupted by brief unfolding pulses to probe the integrity of the protein. The protocol in D) first fully unfolds and extends the protein, after which the force is lowered and the protein is kept unfolded at 14 pN. Afterwards, the protein is allowed to fold again at low force followed by an unfolding pulse to check the number of domains still present (marked with arrows). A dashed line is drawn indicating the total protein extension.

Figure 2. Time dependent loss of folding contraction

A) Protein L octamer held in the extended state for Δt_U >3 hrs shows a complete loss of its folding contraction, retaining only its elastic recoil. B) Protein L octamer is unfolded and held extended in a solution with 5 mM AA. After holding the protein unfolded for >12 hours, 7 of 8 domains are recovered. Arrows indicate unfolding steps. C) Number of folded domains that remain vs the time that a protein is exposed unfolded to solvent, Δt_U. The data are fit with a single exponential (solid line) from which the decay rate τ and the standard error of the mean (SEM) are obtained. Four proteins were studied: I91 (τ = 0.07 ± 0.02 h; pink), I10 (τ = 0.09 ± 0.02 h; dark orange), ubiquitin (right y-axis; τ = 0.12 ± 0.03 h, dark blue) and protein L (τ = 5.3 ± 1.4 h, green). D) Protein L was studied under four conditions: adding 0.6% hydrogen peroxide to the solution (τ = 1.96 ± 0.4 h; brown), 5 mM AA (τ = 66.8 ± 22.9 h; yellow), kept in the folded state at 4.3 pN (τ = 92.5 ± 14.3 days; blue), and without illumination (τ = 12.6 ± 2.2 h; red).

Figure 3. Dissecting protein folding under force into its components

A) A force quench results in a characteristic elastic contraction, followed by a force-dependent stepwise folding contraction. B) The average size of unfolding and folding steps as a function of force. The data are fit using the FJC model with l_k= 1.1 nm and ΔL_c = 16.3 nm. C) Folding probability as a function of refolding force. Data are fit using a sigmoidal function with m = 8 pN and r = 0.93 pN. D) Elastic contraction measured from 45 pN to a refolding force F (40 to 4 pN) for a naïve protein (blue) and a fully-oxidized protein (red). The fits correspond to the elastic recoil force dependency assuming the FJC model with l_k = 0.47 nm, and normalized by the total contour length of each molecule. Error bars are SEM.

Figure 4. Protein folding in elasticity and ageing

A) Total recoil (TR) of protein L from 20 pN to different release forces (15 – 4 pN). The experimental data (red symbols) is well reproduced by Eq.1 (red line) using the parameters of Figure 3. The elastic recoil is calculated for comparison (blue line). B) Model of TR of 15 tandem FNIII domains, when the force is quenched from 20 pN to a lower force F. The change in TR is calculated with Eq.1 for an undamaged protein with its folding contraction intact (red line), compared to that of a damaged protein where the folding contraction is lost (blue line). Predictions of the change in TR for different accumulated ageing times at 6 pN are shown (dashed lines), using two ageing time constants τ = 0.1 h and 70 h. The inset is an analogy of pinching the skin that would apply a stretching force.