Role of complementarity-determining regions 1 and 3 in pathologic amyloid formation by human immunoglobulin κ1 light chains

Abstract

Background

Immunoglobulin light chain (LC) amyloidosis is a life-threatening disease complicated by vast numbers of patient-specific mutations. We explored 14 patient-derived and engineered proteins related to κ1-family germline genes IGKVLD-33*01 and IGKVLD-39*01.

Methods

Hydrogen-deuterium exchange mass spectrometry analysis of conformational dynamics in recombinant LCs and their fragments was integrated with studies of thermal stability, proteolytic susceptibility, amyloid formation and amyloidogenic sequence propensity. The results were mapped on the structures of native and fibrillary proteins.

Results

Proteins from two κ1 subfamilies showed unexpected differences. Compared to their germline counterparts, amyloid LC related to IGKVLD-33*01 was less stable and formed amyloid faster, whereas amyloid LC related to IGKVLD-39*01 had similar stability and formed amyloid slower, suggesting different major factors influencing amyloidogenesis. In 33*01-related amyloid LC, these factors involved destabilization of the native structure and probable stabilization of amyloid. The atypical behavior of 39*01-related amyloid LC stemmed from increased dynamics/exposure of amyloidogenic segments in βC′V and βEV that could initiate aggregation and decreased dynamics/exposure near the Cys23–Cys88 disulfide.

Conclusions

The results suggest distinct amyloidogenic pathways for closely related LCs and point to the complementarity-defining regions CDR1 and CDR3, linked via the conserved internal disulfide, as key factors in amyloid formation.

Keywords:

• Hydrogen-deuterium exchange mass spectrometry
• light chain amyloidosis
• mutational effects on native and misfolded states
• protein stability and dynamics
• Protein aggregation, proteolysis and misfolding

Acknowledgements

The authors thank Mari Nakamura for help with data collection using electron microscopy and Thomas Wales for technical assistance with HDX MS. We thank Dr. Tatiana Prokaeva for helpful comments on the manuscript, and acknowledge her and Brian Spencer for cloning and sequencing of patient-based LCs. We are grateful to Dr. Gareth Morgan for helpful comments.

Author contributions

ESK, DP, OG and JRE designed research; DP, ESK and EAB performed research; ESK, DP, JRE and OG analyzed the data and wrote the paper; all authors discussed the results and edited the paper; OG, LHC and JRE obtained funding.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The HDX MS data have been deposited to the ProteomeXchange Consortium via the PRIDE [32] partner repository with the dataset identifier PXD039682. The summary of HDX MS experimental parameters, proteolytic maps, details of replicates, and the numeric values used to create all HDX MS figures are provided in Supplemental Data file 1.

Additional information

Funding

This study was supported by NIH R01 Grants GM067260 and GM135158, and the Wildflower Foundation.