Investigating native state fluorescence emission of Immunoglobulin G using polarized Excitation Emission Matrix (pEEM) spectroscopy and PARAFAC

Intrinsic fluorescence spectroscopy (IFS) measurements for protein structural analysis can be enhanced by the use of anisotropy resolved multidimensional emission spectroscopy (ARMES). ARMES attempts to overcome the tryptophan (Trp) and tyrosine (Tyr) spectral overlap problem and resolve emitting components by combining anisotropy measurements with chemometric analysis. Here we investigate for the first time the application of polarized excitation-emission matrix (pEEM) measurements and Parallel Factor (PARAFAC) analysis to study IFS from an Immunoglobulin G (IgG) type protein, rabbit IgG (rIgG), in its native state. Protein IFS is a non-trilinear system primarily because of Forster resonance energy transfer (FRET). Non-trilinearity is also caused by inner filter effects, and Rayleigh/Raman scattering, both of which can be corrected by data pre-processing. However, IFS FRET cannot be corrected for, and thus here we carefully evaluated the impact of various different data preprocessing methods on IFS data which used for PARAFAC. Care must be taken with data pre-processing and interpolation, as both had an impact on PARAFAC modelling and the recovered anisotropy values, with residual shot noise from the Rayleigh scatter which overlapped the emission blue edge being the root cause. pEEM spectra from thawed rIgG solutions (15-35 degrees C temperature range) were collected with an expectation being that this temperature range should cause sufficient emission variation to facilitate component resolution but without major structural changes. However, the only significant changes observed were of the overall intensity due to thermal motion induced quenching and this was confirmed by the PARAFAC scores. PARAFAC resolved one major component (>99%) for the emission data (polarized and unpolarized) which mostly represented the large Tyr-to-Trp hetero-FRET process, with a second, very weak component (<1%) apparently a contribution from directly excited Trp emission. PARAFAC scores recovered from normalized pEEM data showed minimal change which was further proof for negligible structural change. The results of this study serves as the starting point for the use of PARAFAC analysis of IFS from IgG type proteins and important processes such as denaturation and aggregation.