Implementation of an FTIR calibration curve for fast and objective determination of changes in protein secondary structure during formulation development

Abstract

The aim of this study was to develop a quick and objective method for the determination of changes in protein secondary structure by Fourier transform infrared spectroscopy (FTIR). Structural shifts from native regions (α-helix, intramolecular β-sheet) to aggregated strands (intermolecular β-sheet) were used to evaluate protein damage. FTIR spectra of 16 different proteins were recorded and quantified by peak fitting of the non-deconvolved and baseline corrected amide I bands. The resulting percentile secondary structures were correlated with the shape and intensity of the area normalized amide I bands using an interval partial least squares algorithm (iPLS). Structural elements were focused on the following regions: α-helix 1660–1650 cm −1, intramolecular β-sheet 1695–1683 cm −1 and 1644–1620 cm −1, intermolecular β-sheet 1620–1595 cm −1. Three calibration curves were created from the data sets. Calculated α-helix content ranged from 0% to 79.59%, intramolecular β-sheet from 10.64% to 63.89% and intermolecular β-sheet from 0.23% to 9.70%. The linear relationship between actual values (as determined by peak fitting) and calculated values was evaluated by correlation coefficient and root mean square error of calibration while cross-validation was performed to detect possible outliers. Results were verified by including two proteins as validation standards and comparing the calculated values to peak fitting and X-ray data. Structural changes of human serum albumin (HSA) due to elevated temperatures and the fibrillation of glucagon were quantified by calibration curve analysis. Performance and reliability of the iPLS algorithm were evaluated by comparing calculated secondary structure elements with results from peak fitting and circular dichroism. Different methods for the determination of secondary structure gave slightly different results but overall tendencies concurred. Additionally, formation of HSA aggregates could be linked to increasing β-sheet content by comparing SEC–HPLC and turbidity analysis with results from the FTIR calibration curves. In summary, quantification of the α-helix to β-sheet transition by iPLS analysis proves to be a feasible and objective way for the determination of damage to protein secondary structure.