A residue-specific shift in stability and amyloidogenicity of antibody variable domains

Abstract

Variable (V) domains of antibodies are essential for antigen recognition by our adaptive immune system. However, some variants of the light chain V domains (VL) form pathogenic amyloid fibrils in patients. It is so far unclear which residues play a key role in governing these processes. Here, we show that the conserved residue 2 of VL domains is crucial for controlling its thermodynamic stability and fibril formation. Hydrophobic side chains at position 2 stabilize the domain, whereas charged residues destabilize and lead to amyloid fibril formation. NMR experiments identified several segments within the core of the VL domain to be affected by changes in residue 2. Furthermore, molecular dynamic simulations showed that hydrophobic side chains at position 2 remain buried in a hydrophobic pocket, and charged side chains show a high flexibility. This results in a predicted difference in the dissociation free energy of ∼10 kJ mol(-1), which is in excellent agreement with our experimental values. Interestingly, this switch point is found only in VL domains of the κ family and not in VLλ or in VH domains, despite a highly similar domain architecture. Our results reveal novel insight into the architecture of variable domains and the prerequisites for formation of amyloid fibrils. This might also contribute to the rational design of stable variable antibody domains.

Keywords: Amyloid; Antibody; Domain Architecture; Molecular Dynamics; Nuclear Magnetic Resonance (NMR); Protein Stability; Variable Domains.

FIGURE 1.

Sequence analysis of V_Lκ domains, spectroscopic characterization of MAK33 V_Lκ and 1OPG V_Lκ. a, multiple sequence alignment of V_Lκ sequences. Five representative sequences are shown. MAK33 V_Lκ differs from 1VGE V_Lκ at 40 positions and from 1OPG V_Lκ at five positions as marked by red rectangles. Identical CDRs between MAK33 V_Lκ and 1OPG V_Lκ domains are highlighted in different colors. b, far-UV CD spectra of native (continuous line) and temperature-denatured (dotted line; 60 °C) MAK33 V_Lκ and 1OPG V_Lκ in PBS buffer. c, near-UV CD spectra of native MAK33 V_Lκ (green line) and 1OPG V_Lκ (blue line). Intrinsic tryptophan fluorescence spectra of native (continuous line) and 3 m GdmCl-denatured (dotted lines) MAK33 V_Lκ (d) and 1OPG V_Lκ (e) in PBS buffer. f, ANS fluorescence spectra of native MAK33 V_Lκ (green line) and 1OPG V_Lκ (blue line). The gray line is PBS/ANS without protein. g, thermally induced aggregation of MAK33 V_Lκ and 1OPG V_Lκ domains monitored by Rayleigh (elastic) light scattering at 440 nm.

FIGURE 2.

Stability and amyloidogenic propensity of MAK33 V_Lκ and 1OPG V_Lκ domains. a, thermal unfolding transitions of MAK33 V_Lκ (green symbols) and 1OPG V_Lκ (blue symbols). The solid lines indicate the theoretical curves derived by fitting the data to a Boltzmann function for MAK33 VLκ (green) and 1OPG V_Lκ (blue) to obtain transition midpoints (T_melt). b, GdmCl-induced unfolding transitions of MAK33 VLκ (green) and 1OPG V_Lκ (blue). The reversibility of the GdmCl-induced unfolding process is shown by the overlay of unfolding symbols (open circles) and refolding (closed circles) experiments. The solid lines show the fit to a two-state mechanism for both MAK33 V_Lκ (green line) and 1OPG V_Lκ (blue line) to obtain thermodynamic stability values (ΔG_U) and the cooperativity parameters (m values). c, transmission electron microscopy micrographs of MAK33 VLκ and 1OPG V_Lκ from amyloid induction experiments at neutral pH, 37 °C, and 1 week incubation with gentle agitation. Scale bars, 200 nm.

FIGURE 3.

Residue frequency distribution analysis at position 2 of variable domains and of the different positions between MAK33 V_Lκ and 1OPG V_Lκ domains. Relative distribution of amino acid at various positions of light (a–e) and heavy chains (f). Green arrowheads indicate residue in MAK33 V_Lκ, and the corresponding positions in 1OPG V_Lκ are shown by blue arrowheads. With the exception of positions 100 and 106, the residues found in MAK33 V_Lκ at positions 2, 3, and 102 are more frequent than those at corresponding positions in 1OPG V_Lκ. The database could not separate λ LCs from the κ LCs during the distribution analysis. But an observation of most of the sequences show that Ser-2 is almost entirely contributed by the λ LCs (orange arrowhead in a). f, red arrowheads indicate the most frequent residue at position 2 of heavy chains. Sequences from all organisms were considered for the analysis. Analysis was based on the Kabat numbering scheme accessed through the antibody (abYsis database) (Dr. Andrew C. R. Martin's Group, University College London).

FIGURE 4.

Stability of the different V_Lκ variants. Thermal unfolding transitions of 1OPG V_Lκ variants (a), MAK33 V_Lκ variants (b), and 1VGE V_Lκ variants (d). The solid lines indicate the theoretical curves derived by fitting the data to a Boltzmann function to obtain transition midpoints (T_melt). c, increase or decrease in stability caused by the different substitutions in 1OPG V_Lκ and MAK33 V_Lκ, obtained by subtracting the T_melt and ΔG_U values of the wild type V_L from those of their corresponding mutants. e, GdmCl-induced unfolding transitions of 1VGE V_Lκ variants. The reversibility of the unfolding process is shown by the overlay of unfolding symbols (open circles) and refolding (closed circles) experiments. The solid lines show the fit to a two-state mechanism for V_Lκ variants to obtain thermodynamic stability values (ΔG_U) and the cooperativity parameters (m values).

FIGURE 5.

Amyloidogenicity of V_Lκ variants. 30 μm of each V_Lκ variant in a PBS buffer at pH 7.4 in the presence of ThT was subjected to ultrasonic pulses at 37 °C (n = 3). ThT fluorescence was monitored over time for 1OPG V_Lκ variants (a) and MAK33 V_Lκ variants (b), and TEM micrographs (c) were acquired at the end of the assay to detect the presence of fibrils. Variants with higher thermodynamic stabilities (>10 kJ mol⁻¹), OP-E2I/S102T and OP-IAT in a and MK-V_Lκ-WT and MK-T102S in b, did not bind ThT, and as a result the curves of both proteins are superimposed.

FIGURE 6.

Stability of the different MAK33 V_Lκ N-terminal variants and the different position mutants. a, thermal unfolding transitions of MAK33 V_Lκ N-terminal variants. The solid lines indicate the theoretical curves derived by fitting the data to a Boltzmann function to obtain transition midpoints (T_melt). b, GdmCl-induced unfolding transitions of MAK33 V_Lκ N-terminal variants. The reversibility of the unfolding process is shown by the overlay of unfolding symbols (open circles) and refolding (closed circles) experiments. The solid lines show the fit to a two-state mechanism for V_Lκ variants to obtain thermodynamic stability values (ΔG_U) and the cooperativity parameters (m values), for a qualitative comparison of the data. c, increase or decrease in stability caused by different amino acid substitutions at position 2 of 1OPG V_Lκ and MAK33 V_Lκ, obtained by subtracting the T_melt value of the wild type V_L from those of their corresponding mutants.

FIGURE 7.

Multiple sequence alignment of V_Lλ (a) and V_H (b) domains is irrespective of organism. Five representative sequences are shown. Only the first 60 residues are presented. All sequences have a less diverse framework within each family. The N terminus of V_Lλ is mostly a QAV, ESV, or PSV motif with Ser the most frequent at position 2. In V_H, the N terminus is mostly an EVQ, AVQ, or QVK motif with Val the most frequent residue at position 2. Stability of 1AQK V_Lλ and 1VGE V_H variants; thermal unfolding transitions of the different 1AQK V_Lλ (c) and 1VGE V_H variants (e). The solid lines indicate the theoretical curves derived by fitting the data to a Boltzmann function to obtain transition midpoints (T_melt). d, GdmCl-induced unfolding transitions of 1AQK V_Lλ variants monitored by tryptophan fluorescence. The reversibility of the unfolding process is shown by the overlay of unfolding symbols (open circles) and refolding (closed circles) experiments. The solid lines show the fit to a two-state mechanism for all 1AQK V_Lλ variants to obtain thermodynamic stability values (ΔG_U) and the cooperativity parameters (m values). GdmCl-induced unfolding transitions of 1VGE V_H variants could not be performed due to less protein amounts.

FIGURE 8.

Structural properties of V_L variants monitored by NMR spectroscopy. a, ¹H-¹⁵N-HSQC spectra of MAK33 V_Lκ-WT and MAK33 V_Lκ-I2E. Both spectra were acquired at a protein concentration of 50 μm in 20 mm phosphate, 50 mm NaCl (pH 6.5) at 298 K on a 600 MHz spectrometer equipped with a cryoprobe. b, chemical shift changes of MAK33 V_Lκ caused by the I2E substitution. ¹H and ¹⁵N backbone chemical shifts were determined at a protein concentration of 50 μm at 298 K in 20 mm phosphate, 50 mm NaCl (pH 6.5). c, effects of I2E substitution in MAK33 V_Lκ. Strongly affected residues are marked on the MAK33 VLκ-WT crystal structure (PDB code 1FH5). Ile-2 is shown in red, and residues with chemical shift changes >0.05 ppm are shown in dark purple, and unassigned residues are shown in gray.

FIGURE 9.

a, comparison of backbone root mean square deviations (RMSD) from the corresponding experimental start structure versus data gathering simulation time for MAK33 V_Lκ-WT (black curve) and MAK33 V_Lκ-I2E (red curve). b, same for 1OPG V_Lκ-WT (red curve) and 1OPG V_Lκ-E2I (black curve). c, comparison of root mean square fluctuations of amino acid residues along the domain chain for 200-ns simulations of MAK33 VLκ-WT (black curve) and MAK33 VLκ-I2E (red curve). d, same for 1OPG V_Lκ-WT (red curve) and 1OPG V_Lκ-E2I (black curve).

FIGURE 10.

Heavy atom r.m.s.d. of residue 2 versus simulation time after best superposition on the complete backbone of the start structure for 1OPG V_Lκ-WT (red, Glu-2) and 1OPG V_Lκ-E2I (black, Ile-2) (a) and MAK33 V_Lκ-WT (black, Ile-2) and MAK33 V_Lκ-I2E (red, Glu-2) (b). c, calculated potential-of-mean force for the dissociation of the Glu-2 in 1OPG V_Lκ-WT (black curve), Ile-2 in 1OPG V_Lκ-E2I (green), Ile-2 in MAK33 V_Lκ-WT (red curve), and Glu-2 in MAK33 V_Lκ-I2E (blue curve) from the hydrophobic cavity region. d, N-terminal cavity region of MAK33 V_Lκ with the Ile-2 side chain buried in the cavity (protein schematic with Ile-2 as sticks model and adjacent side chains as van der Waals spheres). e, example of a simulation snapshot with a fully solvent-exposed Glu-2 side chain and several water molecules at the rim of the N-terminal cavity region of 1OPG V_Lκ. f, solvent-accessible surface representation in yellow of the binding cavity region for residue 2 in the case of MAK33 V_Lκ-WT (stick model of Ile-2); g. same for 1VGE V_H-WT with Val-2 as stick representation, and h, same for 1AQK V_Lλ-WT with Asn-2.

FIGURE 11.

a, comparison of backbone r.m.s.d. from the corresponding experimental start structure versus data gathering simulation time for 1AQK V_Lλ-WT (black curve) and 1AQK V_Lλ-N2E (red curve). b, same for 1VGE V_H-WT (black curve) and 1VGE V_H-V2E (red curve). Heavy atom r.m.s.d. of residue 2 versus simulation time after best superposition on the complete backbone of the start structure for 1AQK V_Lλ-WT (black, Asn-2) and 1AQK V_Lλ-N2E (red, Glu-2) (c) and 1VGE V_H-WT (black, Val-2) and 1VGE V_H-V2E (red, Glu-2) (d).