Kartographierung des Schwingungsenergietransfers in Proteinen mit Hilfe nichtkanonischer Aminosäuren und Untersuchung seiner Rolle in der allosterischen Signaltransduktion
Physikalische Chemie von Molekülen, Flüssigkeiten und Grenzflächen, Biophysikalische Chemie
Zusammenfassung der Projektergebnisse
Molecules store and transfer energy via the vibrations of their atoms. It is assumed that vibrational energy and its transfer (VET - vibrational energy transfer) play an important role in the function of proteins. This concerns, for example, the dissipation of excess energy during chemical reactions in enzymes or after optical excitation, for example in photoreceptors. The coupling of protein vibrations to the motion along the reaction coordinate in chemical reactions in proteins that act as catalysts (enzymes), as well as the communication between different parts of a protein (allostery), are further processes in the context of which the role of VET is discussed. Many studies on VET are based on computer simulations. In simulations, it is comparatively easy to inject vibrational energy into a protein at a precisely defined point and to follow its transport through the protein on the picosecond time scale. In our project, we have now developed an approach that allows us to do exactly the same in an experiment - with the aim of mapping the transport through the protein, investigating its mechanisms and understanding its significance for protein function. To inject vibrational energy into a protein at a precisely defined location and to track it´s transfer at the level of individual amino acids (AS) through the protein, we have developed a pair of non-natural amino acids. The first amino acid (azulenylalanine, AzAla) serves as an injector of vibrational energy. In the experiment, it converts light energy from a short laser pulse (about 100 fs) in the visible spectral range into vibrational energy within less than 1 ps and injects it into the protein. The second amino acid (azidohomoalanine, Aha) serves as a sensor for incoming vibrational energy, which can be read out in the infrared spectral range with a second laser pulse. The next important step was to make it possible to incorporate these amino acids into the protein precisely at the points between which the energy transfer is to be investigated. To this end, we expanded the natural genetic code that describes the structure of the proteins as well as the mechanisms for amino acid incorporation in a way that the nonnatural amino acids can be incorporated at the desired sites. The next major challenge proved to be the detection of VET signals in the infrared spectrum. These are not only very small, but also superimposed by other signals and measurement artifacts. Various technical developments were necessary in order to be able to record signals down to the range of a few µOD (among the smallest that have been measured to date with femtosecond infrared spectroscopy). We have also started to apply the methods developed as part of the project. The shortest path between two sites in the protein is usually not along chemical bonds. VET could take place along the bonds, but also via other contacts between amino acids that could shorten the path (so-called non-covalent contacts such as hydrogen bonds and van der Waals contacts). How efficient is VET via such contacts compared to transfer via bonds along the protein backbone? Our measurements on a hairpin-shaped peptide have shown (in cooperation with theoretical work, Gerhard Stock, Freiburg) that even a hydrogen bond, that shortcuts a distance of 3 amino acids along the backbone, provides more efficient transfer than going along the backbone. The next largest system in which we started mapping VET was the protein domain PDZ3, a model system for communication between different protein parts (allostery). The mapping has not yet been completed, but it is already clear that VET is indeed anisotropic, i.e. it does not occur uniformly in all directions: The vibrational energy reaches some protein parts twice as fast as others, despite being at the same distance. The third system that we have started to investigate is the enzyme format dehydrogenase (FDH), for whose catalytic activity a role of protein low frequency vibrations is being discussed. Here we have succeeded in detecting VET from the periphery of the enzyme to the active center where the reaction takes place. We now have a versatile experimental methodology at our disposal to investigate VET in a wide variety of proteins and in the context of different protein functions.
