High-Resolution THz Spectroscopy of Crystalline Peptides: Exploring Hydrogen-Bonding Networks

摘要

The sensitivity of CW THz spectroscopy to hydrogen bonding networks of peptide crystals is explored through investigations of the lowest frequency vibrational modes including three forms of trialanine, and two isostructural dipeptides which form hydrophobic nanotubes, alanyl isoleucine and isoleucyl alanine. THz spectra were obtained in the range 0.6 cm~(-1) to 100 cm~(-1) at 4.2 K for the peptides under investigation. Three issues related to the THz absorption features have been addressed: ⅰ) the impact of the intermolecular hydrogen bonding network, ⅱ) effects arising from weak hydrophobic interactions with water and ⅲ) line broadening due to crystal defects. First, large variations in the THz spectra of antiparallel β-sheet trialanine are observed due to changes in the degree of hydration. Second, subtle frequency shifts of ～ 1 cm~(-1) due to the weak hydrophobic interactions of dipeptide molecules with intracrystalline water are reported for the hydrated dipeptide alanyl isoleucine. In contrast, little or no shift was observed for the unhydrated but isostructural retroanalogue, isoleucyl alanine. Finally, line-broadening effects which can be attributed to the relative abundance of crystal defects are reported for the parallel-β-sheet form of rntrialanine. The small crystalline peptides studied here provide benchmark systems for validating computational models at the level of semi-empirical force fields and, because of the small molecular size, at the ab initio levels of theory. The experimental results give insight into the nature and scope of computational models necessary to accurately capture the structure and spectra of these simple peptides. Studies of simple systems are a necessary first step towards delineating the model requirements imposed by the very low energy modes of this regime, where long range and many body interactions can have a significant impact. Comparison with computational results will aid in advancing the current state-of-the-art in computational software to accurately model more complex biological systems.