Optical Absorbance Sensitivity to Rugged Energy Landscape

Abstract

Previously we have shown that underdamped long-range intramolecular vibrations exist in proteins using anisotropic absorption in the terahertz frequency range [1,2]. Measurements on free chicken egg white lysozyme (CEWL) and CEWL bound to the inhibitor tri-acetylglucosamine (3NAG) found the anisotropic absorption dramatically changes with inhibitor binding. To associate the spectral features with specific motions that are enhanced or diminished by the binding, we compare the measured spectra to calculations using normal mode analysis (NMA). Starting structures were randomly chosen frames from a molecular dynamics trajectory of a fully hydrated CEWL molecule. For each starting structure, we minimize the energy with small structural perturbations, and then perform NMA using CHARMM. These calculations find that the vibrational density of states (VDOS) is largely unchanged for a given starting structure, whereas the optical absorbance is different for each minimized structure. This variation arises from the sensitivity of the optical absorption to the strength and direction of the transition dipole, which is strongly dependent on the actual displacements for a given mode energy. This suggests that to properly model optical absorbance, one must perform an ensemble average of the calculated spectra, whereas for comparisons with the VDOS measurements, a single minimized structure is sufficient. The VDOS does not appear to be strongly affected by the ruggedness of the energy landscape, whereas the optical absorbance is. Given that strong bands are measured experimentally, the results suggest that in spite of the variation in the optical spectra, there are vibrations that are more prevalent than others, and possibly this biasing towards these motions enhances function. [1] Niessen, K.A., Xu, M., Markelz, A.G. (2015). Biophysical Reviews 7(2): 201-216. http://dx.doi.org/10.1007/s12551-015-0168-4. [2] Acbas, G., Niessen, K.A., Snell, E.H., Markelz, A.G. (2014) Nature Communications 5, 3076 http://dx.doi.org/10.1038/ncomms4076.