PMC

Kinetic refolding species plots for U (solid line), M (dashed line) and N₂ (dotted lines) were derived from the parameters obtained in the global fit (). Protein concentrations are 0.1 µM (blue), 1 µM (green) and 10 µM (red).

Perturbations relative to WT SOD1 were calculated from the parameters obtained in the global kinetic analysis for each protein. The black bars indicate the total change in free energy (2UN₂); the white bars indicate the change in free energy for the monomer (2U2M), and the grey bars indicate the change in free energy for the dimer association reaction (2MN₂).

Data were measured by far-UV CD at 230 nm at 10 µM monomer concentration. (•) WT, (□) A4V, (▴) L38V, (▵) G93A, (▪) L106V, (○) S134N. Solid (for open symbols) and dashed (for closed symbols) lines represent global fits of two to three protein concentrations to a two-state model. The corresponding free energies are listed in .

The fractional populations of (A) the monomer and (B) the unfolded state at equilibrium and 0 M urea for 0.1 µM (black), 1 µM (white) and 10 µM (grey) protein concentration were calculated based on the parameters from the global kinetic analysis (). The populations of the unfolded states for WT SOD1 and S134N are multiplied by 10 in panel B to make them visible.

(left) The free cysteine residues at the sites of the C6A and C111S mutations and the C57–C146 disulfide bond are shown in red and purple ball-and-stick models, respectively. The Zn-binding loop (loop IV; residues 49–82) is shown in pale blue, and the electrostatic loop (loop VII; residues 121–142) is in pale green. The Zn²⁺ and Cu²⁺ ions are shown as blue and orange spheres, respectively. The sites of ALS mutations (A4V, L38V, G93A, L106V, and S134N) are highlighted in yellow ball-and-stick models. (right) One monomer of SOD1 turned 90° around a horizontal axis toward the viewer. The figure was generated with PyMOL using pdb code 2c9v .

Refolding (filled circles) and unfolding (open circles) for WT, A4V, L38V, G93A, L106V, and S134N SOD1 were monitored by manual-mixing CD at 230 nm. The symbols correspond to the different final protein concentrations; 1–2 µM (pink), 4 µM (blue), 10 µM (black), and 30 µM (red). Lines are the inverse of the microscopic rate constants obtained in the global fits for the monomer refolding (solid green), monomer unfolding (dashed green), dimer association (solid blue), and dimer dissociation reaction (dashed blue). For clarity, the dimer association reaction has been normalized to 10 µM protein concentration. The relaxation times for the corresponding stable monomeric versions of these proteins are shown in grey circles, and the grey lines represent fits of the data to a two-state kinetic model, UM.

The folding reaction coordinates at 10 µM protein under native conditions are calculated from the global fits based on the results of in vitro kinetic measurements. The Tanford β value is used as the order parameter and is calculated from the kinetic m-values in according to β = m_f/(m_f−m_u). This order parameter is expected to be proportional to solvent accessible surface area. The black line represents the folding reaction coordinate of the WT protein, which is nearly coincident with the reaction coordinate for the S134N variant (not shown). Other disease mutations result in selective destabilization of the dimer and/or significant destabilization of the monomer. The A4V variant (blue) predominately affects the dimer stability, whereas the G93A variant (red) has the largest effect on the monomer stability. The L38V (green) and L106V (not shown) variants affect both steps. Based on the enhanced populations of partially-folded forms observed for the ALS-causing variants, aggregation (red arrows) is hypothesized to proceed from the monomer (M) or unfolded (U) species. The wild-type like reaction coordinate observed for the metal-deficient S134N suggests that even WT SOD can aggregate in the absence of metals. Although the reaction coordinate shown is semi-quantitative, one caveat to including both monomers and dimers i a single reaction coordinate is that it is necessary to plot them with respect to different reference states. The part of the reaction coordinate corresponding to the monomer reaction is, therefore, scaled with respect to a monomer reference state and the bimolecular step is scaled with respect to a dimer reference state. A prefactor of 1×10⁹ s⁻¹ and 1×10⁹ M⁻¹s⁻¹ was used for the unimolecular and bimolecular steps, respectively, in relating the rate to an activation free-energy using Kramers theory.