Two‐Dimensional Correlation and Two‐Dimensional Co‐Distribution Spectroscopies of Proteins

Abstract

Given the complexity of proteins, the broad nature of infrared bands, and the overlap of the HOH bending mode of water within the 1700–1600 cm −1 spectral region, it was often considered too difficult to include side‐chain vibrational modes in a description of a protein's dynamic behavior in solution. Yet, it is the roles of side‐chains that allow a description of a protein's dynamics to become a site‐specific understanding of stability and provide a basis for the development of protein drugs. This article describes how, in certain cases, site‐specific analysis of protein infrared spectra can be performed using two‐dimensional infrared correlation spectroscopy ( 2D‐COS ) developed by Dr. Isao Noda. The 2D‐COS algorithm spreads overlapped peaks within a spectral region of interest into a second dimension that allows for a complete evaluation of all of the peaks that change during a perturbation. 2D‐COS analysis has been implemented successfully for the study of proteins under different perturbations such as pressure, chemical, pH, concentration, and most commonly temperature. Initially, the interest was in determining conformational changes during thermal unfolding, which then led to evaluating solvent accessibility through hydrogen/deuterium exchange, leading to the use of isotope editing to probe local structural changes and interactions. Herein, we will explore the extent that both 2D‐COS and two‐dimensional co‐distribution spectroscopy ( 2D‐CDS ) have been implemented in the study of proteins. However, we were also keen to include recent advances to site‐specifically monitor protein deamidation in aqueous solution. Protein deamidation, like oxidation and glycosylation, is one of the processes that can initiate protein degradation pathways which could lead to decreased target binding, decreased stability, and possibly aggregation. These concerns have become part of the critical quality attributes ( CQAs ) evaluated during therapeutic protein development because they can potentially impact drug safety and efficacy. To date, tandem liquid chromatography–mass spectrometry ( LC‐MS ) techniques have been the primary method used to ascertain the presence and sites of deamidation in protein samples. However, the LC‐MS process requires multiple steps, including enzymatic digestion and peptide mapping. 2D‐COS and 2D‐CDS analyses provide an alternative that allows the determination of the deamidation sites and the evaluation of the stability of the protein populations within a sample without any sample preparation steps. The 2D‐COS method utilizes an innovative platform comprised of a quantum cascade laser microscope ( QCLM ), slide cell array, and software. The QCLM provides an enhanced signal/noise ratio and real‐time acquisition of hyperspectral images of an array of proteins in solution under accurate thermal control. The method monitors the deamidation process during thermal stress. Spectral data is then subject to 2D‐COS analysis to determine the extent of and site at which both asparagine and glutamine deamidation occur. In addition, 2D‐CDS analysis allows assessment of changes in stability in the distributions of the protein populations during the perturbation. The results presented here for the deamidation of the NISTmAb have been validated by LC‐MS to evaluate this new analytical tool. We foresee this approach as a required step in a therapeutic protein's developability assessment and release testing.