Kinetic stability and sequence/structure studies of urine-derived Bence-Jones proteins from multiple myeloma and light chain amyloidosis patients

Abstract

It is now accepted that the ability of a protein to form amyloid fibrils could be associated both kinetic and thermodynamic protein folding parameters. A recent study from our laboratory using recombinant full-length (encompassing the variable and constant domain) immunoglobulin light chains found a strong kinetic control of the protein unfolding for these proteins. In this study, we are extending our analysis by using urine-derived Bence Jones proteins (BJPs) from five patients with light chain (AL) amyloidosis and four patients with multiple myeloma (MM). We observed lower stability in κ proteins compared to λ proteins (for both MM and AL proteins) in agreement with previous studies. The kinetic component of protein stability is not a universal feature of BJPs and the hysteresis observed during refolding reactions could be attributed to the inability of the protein to refold all domains. The most stable proteins exhibited 3-state unfolding transitions. While these proteins do not refold reversibly, partial refolding shows 2-state partial refolding transitions, suggesting that one of the domains (possibly the variable domain) does not refold completely. Sequences were aligned with their respective germlines and the location and nature of the mutations were analyzed. The location of the mutations were analyzed and compared with the stability and amyloidogenic properties for the proteins in this study, increasing our understanding of light chain unfolding and amyloidogenic potential.

Keywords: Bence-Jones; Circular dichroism; Kinetic stability; Light chain amyloidosis; Protein aggregation; Thermal unfolding.

Figure 1

Far-UV CD spectra of (A) MM1λ (green) and MM2λ (brown), (B) MM1κ (purple) and MM1κ(orange), (C) AL2λ (blue), (D) AL1κ(red) and AL2κ (black); in PBS (pH 7.4) at 4 °C, before (filled symbols) and after irreversible thermal unfolding (open symbols). In all cases, samples were prepared at 2.0 μM and measured using a 1 cm cuvette. Both heating and cooling were carried out at 0.5 °C min⁻¹.

Figure 2

Thermal denaturation curves of (A) MM1λ and (β) MM2λ, following changes by circular dichroism as a function of thermal scan rate. In all cases, melting curves were measured continuously from 4 °C to 90 °C every 0.5 °C. Heating scan-rates employed 0.5 °C min⁻¹ (black), 1.0 °C min⁻¹(blue), 1.5 °C min⁻¹ (green), 2.0 °C min⁻¹ (orange) and 2.5 °C min⁻¹ (red). Samples were prepared at 2.0 μM and measured using a 1 cm cuvette with continuous stirring. Error bars correspond to the standard error obtained from 3 independent experiments.

Figure 3

Thermal denaturation curves of (A) MM1κ and MM2κ, and (B) AL2κ and AL1κ following changes by circular dichroism as a function of thermal scan rate. In all cases, melting curves were measured continuously from 4 °C to 90 °C every 0.5 °C. Heating scan-rates employed 0.5 °C min⁻¹ (black), 1.0 °C min⁻¹(blue), 1.5 °C min⁻¹ (green), 2.0 °C min⁻¹ (orange) and 2.5 °C min⁻¹ (red). Samples were prepared at 2.0 μM and measured using a 1 cm cuvette with continuous stirring. Error bars correspond to the standard error obtained from 3 independent experiments.

Figure 4

Changes in the enthalpy reference independent of the applied scan rate ΔH^‡, as a function of the scan rate calculated for (A) MM1λ (green) and MM2λ (brown), (B) MM1κ (purple) and MM1κ(orange), (C) AL2λ (blue), (D) AL1κ(red) and AL2κ (black). Continuous represent the averaged ΔH^‡ value. Data are the average of triplicates obtained from fitting each thermal unfolding transition using Eq. (4). Error bars correspond to the standard error obtained from 3 independent experiments.

Figure 5

Comparison of the stability parameters (T* values) obtained from fitting the thermal transitions in figures 2 and 3, for: (A) MM1λ (green) and MM2λ (brown), (B) MM1κ (purple) and MM1κ(orange), (C) AL2λ (blue), (D) AL1κ(red) and AL2κ (black). Proteins were grouped as function of the disease and isotype to facilitate comparisons. Continuous line represent averaged T* values, and error bars represent the standard error obtained from 3 independent experiments.

Figure 6

Comparison of the scan rate dependence of the apparent Tm values (Tmapp), obtained from fitting the thermal transitions in figures 2 and 3, for: (A) MM1λ (green) and MM2λ (brown), (B) MM1κ (purple) and MM1κ(orange), (C) AL2λ (blue), (D) AL1κ(red) and AL2κ (black). Proteins were grouped as function of the disease and isotype to facilitate comparisons. Dashed lines represent the fitting to a linear function. Error bars correspond to the standard error obtained from 3 independent experiments.

Figure 7

Location of the mutations found in the most stable proteins and the only ones with 3-state unfolding behavior (top); proteins with small scan rate dependency (middle), and proteins with the largest scan rate dependency (bottom). Red highlights and text denotes non-conservative mutations.1b6d.pdb was used for the κ proteins, 1lil.pdb was used for the λ proteins.

Figure 8

Comparison of the relevant structural elements of the κ and λ BJPs. The variable (V_L) and constant (C_L) domains are connected by the join region (~10 aa), and each of the domains has one buried disulfide bond that stabilizes the beta-sandwich fold. Proline residues within V_L and C_L fragment are shown in a stick representation and highlighted in blue. cis- prolines are indicated by the red circles, and the rest are trans- prolines.