Comparability of protein therapeutics: quantitative comparison of second-derivative amide I infrared spectra.

Comparability determination for protein therapeutics requires an assessment of their higher order structure, usually by using spectroscopic methods. One of the most common techniques used to determine secondary structure composition of proteins is analysis of the second derivative of the amide I region of Fourier transform infrared (FTIR) spectra. A number of algorithms have been described for quantitative comparison of second-derivative amide I FTIR spectra, but no systematic evaluation has been conducted to assess these approaches. In this study, the two most common methods, spectral correlation coefficient and area of overlap (AO), are compared for their ability to determine spectral comparability of a protein as a function of changes in pH or temperature. Two other algorithms were considered as well. Recently, a QC compare similarity function found in OMNIC software has been reported as being useful in comparing amide I FTIR spectra. In addition, a new algorithm, termed modified AO, is described herein. These four methods were evaluated for their ability to determine comparability for second-derivative amide I FTIR spectra of four model proteins. The result is a framework for quantitative determination of whether any two spectra differ significantly.