pH effects on the stability and dimerization of procaspase-3

Abstract

pH-dependent conformational changes are known to occur in dimeric procaspase-3, and they have been shown to affect the rate of automaturation. We studied the equilibrium unfolding of procaspase-3(C163S) as a function of pH (between pH 8.5 and pH 4) in order to examine these changes in the context of folding and stability. The data show that the procaspase dimer undergoes a pH-dependent dissociation below pH 5, so that the protein is mostly monomeric at pH 4. Consistent with this, the dimer unfolds via a four-state process between pH 8.5 and pH 4.75, in which the native dimer isomerizes to a dimeric intermediate, and the dimeric intermediate dissociates to a monomer, which then unfolds. In contrast, a small protein concentration dependence was observed by circular dichroism, but not by fluorescence emission, at pH 4.5 and pH 4.2. There was no protein-concentration dependence to the data collected at pH 4. Overall, the results are consistent with the redistribution of the population of native dimer (N(2)) to dimeric intermediate (I(2)) to monomeric intermediate (I), as the pH is lowered so that at pH 4, the "native" ensemble resembles the monomeric intermediate (I) observed during unfolding at higher pH. An emerging picture of the monomeric procaspase is discussed. Procaspase-3 is most stable at pH approximately 7 (24-26 kcal/mol), and while the stability decreased with pH, it was observed that dimerization contributes the majority (>70%) of the conformational free energy.

Figure 1.

(A) Structure of mature caspase-3. The positions of helices 1, 4, and 5, and of the active site tryptophanyl residues (W206 and W214) are indicated. Active-site loops L1, L2, L2′, and L4 are indicated. For clarity, only one active site is labeled. (B) Residues involved in charge–charge interactions across the dimer interface. The prime (′) indicates residues from the second heterodimer. The structures were generated using PDB entry 1CP3 and PyMOL (Delano Scientific). (C) Average emission wavelength versus pH for procaspase-3(C163S). The solid line represents a fit of the data as described (Bose and Clark 2001). The pK_a values obtained from the fits are described in the text. (D) Size exclusion chromatography of procaspase-3(C163S) at pH 5 (solid line) or pH 4 (dashed line). The BSA monomer (66 kDa) eluted at 13.4 mL.

Figure 2.

Equilibrium unfolding of procaspase-3(C163S) at pH 7.2. Unfolding was measured by circular dichroism at 228 nm (▵) and by fluorescence average emission wavelength with excitation at 280 nm (○) or at 295 nm (□). For all experiments, the protein concentration was 1 μM. The data were fit simultaneously (solid lines) as described in Materials and Methods using Igor Pro (WaveMetrics, Inc.), and the conformational free energies and m-values obtained from the fits are shown in Figure 6 ▶.

Figure 3.

Equilibrium unfolding vs. urea of procaspase-3(C163S) from pH 8.5 to pH 4.75. The pH is indicated at the top of each column. The protein concentrations are as follows: 0.25 μM (⋄), 0.5 μM (▵), 1 μM (□) and 2 μM (○). The solid triangles (▴) represent refolding data for 0.5 μM protein concentration to show reversibility. (A,E,I) Unfolding was monitored by fluorescence emission with excitation at 280 nm; (B,F,J) unfolding was monitored by fluorescence emission with excitation at 295 nm; (C,G,K) unfolding was monitored by circular dichroism at 228 nm. (D,H,L) An overlay of data for 1 μM protein concentration and fluorescence emission with excitation at 280 nm (○) or 295 nm (□) and CD at 228 nm (▵). The solid lines represent global fits of the data as described in Materials and Methods.

Figure 4.

Equilibrium unfolding versus urea of procaspase-3(C163S) from pH 4.5 to pH 4.0. The pH is indicated at the top of each column. The protein concentrations are as follows: 0.25 μM (⋄), 0.5 μM (▵), 1 μM (□) and 2 μM (○). (▴) Refolding data for 0.5 μM protein concentration to show reversibility. (A,E,I) Unfolding was monitored by fluorescence emission with excitation at 280 nm; (B,F,J) unfolding was monitored by fluorescence emission with excitation at 295 nm; (C,G,K) unfolding was monitored by circular dichroism at 228 nm. (D,H,L) An overlay of data for 1 μM protein concentration and fluorescence emission with excitation at 280 nm (○) or 295 nm (□) and CD at 228 nm (▵). The solid lines represent global fits of the data as described in Materials and Methods.

Figure 5.

Fraction of species as a function of urea concentration at pH 8.5 (A), pH 6.5 (B), pH 4.75 (C), pH 4.5 (D), pH 4.2 (E), and pH 4.0 (F). The fractions of native, dimeric intermediate, monomeric intermediate, and the unfolded protein were calculated as a function of urea concentration from fits of the data shown in Figures 3 ▶ and 4 ▶, and the parameters shown in Figure 6 ▶ for protein concentrations of 0.25, 0.5, 1, or 2 μM. N₂ and N refer to the dimeric and monomeric native procaspase-3(C163S), I₂ and I are the dimeric and monomeric intermediates, and U refers to the unfolded species. (○) fraction of I₂; (□) fraction of I. The data for 0.25 μM are represented by the dashed lines, and the data for 2 μM are represented by the solid symbols.

Figure 6.

(A) Plot of total free energy (°G_total) versus pH. (B) Plot of total cooperativity index (m_total) versus pH. The error bars show the standard error obtained from the global fits of 11 unfolding curves at each pH (four each from two sets of fluorescence data and three from CD data). (C) Mid-point of the unfolding transition (urea_1/2) for the N₂ ⇆I₂ transition (•) and the I ⇆U transition (▪).