Determination of the secondary structure content of proteins in aqueous solutions from their amide I and amide II infrared bands. Comparison between classical and partial least-squares methods.

TLDR: The results show that the best agreement between the secondary structure determined by X-ray crystallography and that predicted by infrared spectroscopy is obtained when both the amide I and II bands are used to generate the calibration set.

Abstract: A method for estimating protein secondary structure from infrared spectra has been developed. The infrared spectra of H{sub 2}O solutions of 13 proteins of known crystal structure have been recorded and corrected for the spectral contribution of water in the amide I and II region by using the algorithm of Dousseau et al. This calibration set of proteins has been analyzed by using either a classical least-squares (CLS) method or the partial least-squares (PLS) method. The pure-structure spectra calculated by the classical least-squares method are in good agreement with spectra of poly(L-lysine) in the {alpha}-helix, {beta}-sheet, and undefined conformations. The results show that the best agreement between the secondary structure determined by X-ray crystal-lography and that predicted by infrared spectroscopy is obtained when both the amide I and II bands are used to generate the calibration set, when the PLS method is used, and when it is assumed that the secondary structure of proteins is composed of only four types of structure: ordered and disordered {alpha}-helices, {beta}-sheet, and undefined conformation. Attempts to include turns in the secondary structure estimation have led to a loss of accuracy. The spectra of the calibration proteins were also recorded in {sup 2}H{sub 2}O solution.more » After correction for the contribution of the combination band of {sup 2}H{sub 2}O in the amide I{prime} band region, the spectra were analyzed with PLS, but the results were not as good as for the spectra obtained in H{sub 2}O, especially for the {alpha}-helical conformation.« less