Urea-induced denaturation of apolipoprotein serum amyloid A reveals marginal stability of hexamer

Abstract

Serum Amyloid A (SAA) is an acute phase reactant protein that is predominantly found bound to high-density lipoprotein in plasma. Upon inflammation, the plasma concentration of SAA can increase dramatically, occasionally leading to the development of amyloid A (AA) amyloidosis, which involves the deposition of SAA amyloid fibrils in major organs. We previously found that the murine isoform SAA2.2 exists in aqueous solution as a hexamer containing a central channel. Here we show using various biophysical and biochemical techniques that the SAA2.2 hexamer can be totally dissociated into monomer by approximately 2 M urea, with the concerted loss of its alpha-helical structure. However, limited trypsin proteolysis experiments in urea showed a conserved digestion profile, suggesting the preservation of major backbone topological features in the urea-denatured state of SAA2.2. The marginal stability of hexameric SAA2.2 and the presence of residual structure in the denatured monomeric protein suggest that both forms may interconvert in vivo to exert different functions to meet the various needs during normal physiological conditions and in response to inflammatory stimuli.

Figure 1.

SAA2.2 urea-induced hexamer to monomer dissociation monitored by glutaraldehyde cross-linking and SDS-PAGE. SAA2.2 samples contained 50 μg/mL of protein in MOPS buffer at pH 7.4 and were preincubated at 20°C for 10 min before cross-linking with 0.7% (v/v) glutaraldehyde.

Figure 2.

SAA2.2 urea-induced quaternary structural changes monitored by size exclusion chromatography. (A) Hexamer to monomer transition is completed by 2 M urea. (B) At urea concentrations higher than 2 M, there is a decrease in the monomer peak retention time. SAA2.2 samples contained 0.1 mg/mL protein in 20 mM Tris buffer with 0.4M NaCl. Separation was achieved using a Superdex 75 PC 3.2/30 column at a flow rate of 0.1 mL/min.

Figure 3.

SAA2.2 urea-induced denaturation monitored by fluorescence polarization. (A) Fluorescence polarization plotted as a function of urea concentration. SAA2.2 samples contained 50 μg/mL of protein in MOPS buffer at pH 7.4 and were incubated at 20°C for 10 min. The symbols represent three independent experiments. (B) Fluorescence emission spectra of SAA2.2 samples containing the indicated urea concentrations.

Figure 4.

SAA2.2 urea denaturation monitored by far UV CD. SAA2.2 samples contained 50 μg/mL of protein in MOPS buffer at pH 7.4 and were incubated at 20°C for 10 min. (A) Wavelength scans taken at some urea concentrations of the sample. (B) Molar residue ellipticity [θ] at 222 nm plotted against urea concentration. Symbols represent two independent experiments.

Figure 5.

Limited trypsin digestion and reverse-phase HPLC analysis of SAA2.2 samples incubated in different concentrations of urea. SAA2.2 samples contained 0.1 mg/mL protein in MOPS buffer at pH 7.4 and were preincubated at 20°C for 30 min before limited trypsin digestion at an SAA:trypsin ratio of 120:1. The reaction was quenched with trifluoroacetic acid and analyzed by reverse-phase HPLC as described in Materials and Methods.

Figure 6.

Comparison of the urea denaturation curves obtained from different methods. The GCL/SDS–polyacrylamide gel was analyzed, and the monomer percentage was plotted against urea concentration. For the SEC curve, the fractional area of the slow-moving SAA2.2 peak in SEC was used. The polarization and CD data was internally normalized using the values at 0 (0%) and 5.8M (100%) urea.