The overall structural make-up of a protein biomolecule is complex and governed by folding, intramolecular, and intermolecular interactions. Repetitive patterns allow the identification of structural populations and monitoring the presence and changes in these structural populations reveals important information relating to the stability and ultimately the safety and efficacy of a biotherapeutic. The most common structures that are important in the characterization of folded proteins include parallel and anti-parallel beta sheets, alpha-helices, turns, and unordered structures. By tracking relative amounts of these components and comparing them to known reference materials, it is possible to gain confidence in the integrity of a biotherapeutic as well as identify undesired changes in functionality that will be critical to control and remedy at all stages in the development and formulation workflows.
Analysis of secondary structure in proteins using Infrared spectroscopy has a long tradition and is a well-accepted industry practice. At first glance, a typical IR spectrum focusing on the Amide I region appears broad, so extraction of useful information requires spectral enhancement. One process of extracting secondary structure information from a spectrum typically involves a band narrowing and curve fitting approach. For instance, Fourier self-deconvolution curve fitting, and partial least squares analysis are two well-known approaches. However, second derivative analysis is the most prominent method used and is the method employed in the AQS³delta software package.
Since the various structural components of the Amide I spectrum absorb at a range of wavelengths (Table 1), the Amide I band tends to be broad. The goal of a band fitting approach is to deconvolute the Amide I band into the various component bands which can then be assigned to the different types of secondary structure. A band narrowing step is first necessary to assign the relevant bands and can be performed by either a second derivative analysis or by Fourier deconvolution.
Table 1: Deconvoluted Amide I band frequencies and assignments to secondary structure.
An example of Gaussian curve fitting can be seen in Figure 1 where the red line represents the inverted and baselined second derivative spectrum of an antibody, and the colored gaussian curves represent the deconvoluted contributions of each spectral feature as outlined in Table 1.
Figure 1: Example curve fitting process of an Amide 1 spectrum obtained from MMS using the AQS3pro.
The advantage of using Microfluidic Modulation Spectroscopy (MMS) powered by the AQS³delta software to elucidate the structural composition of a protein therapeutic is that it is intuitive and easy to specify the populations that should be monitored. Post-analysis processing gives clear and meaningful results to show which secondary structural motifs have been maintained and which may have been compromised due to changes in experimental or formulation conditions.