Strukturelle Biologie der Titin/Myosin Interaktion
Final Report Abstract
The stiff entropic spring of the muscle cell, titin, extends from the Z-disc to the M-line and can also be regarded as a cytoskeletal scaffold due to its modular architecture of ~ 300 Ig and fn3 domains. Along the zone of the myosin filaments fn3 and Ig domains arrange into super-repeats of eleven domains that match the periodicity of the helical head domain of the myosin filaments. The aim of the current project was to determine the atomic details of the interaction between the titin derived two and three fn3 domain constructs and myosin. To this end the structure of the previously studied titin constructs (A77-78, A80-82 and A84-860) should be solved to high resolution. NMR experiments complemented by mutational studies should identify the interaction surface on the known constructs. This was expected to yield the basis for developing a model for the interaction to myosin. When we compared the interaction data obtained through different techniques we found discrepancies between the low-resolution methods like co-sedimentation and solid phase binding assays and the high-resolution method NMR. The latter only confirmed the interaction between the titin fragments and myosin but not the one with actin. Moreover, more elaborate measurements revealed that also the titin myosin interaction is very weak on a two to three domain level so that a detailed study of the interaction surface did not appear to be feasible. It is conceivable that the cooperative effect of multiple domains together is needed to form a strong binding interface but even larger fragments are not suitable for high-resolution studies. However, we used the insights gained from this study to develop a method for affinity constants determination of medium and weak interactions that is applicable to a large range of protein sizes and uses the high reliability of NMR spectroscopy to unambiguously determine whether an interaction is present in solution. This saturation transfer difference experiment requires only minute amounts of the unlabelled ”receptor” protein and is also applicable for high-molecular weight complexes unlike the commonly used chemical shift perturbation experiment. The affinity of the ligand is determined through competition with a known isotope-labelled ligand. The receptor protein and the ligands of unknown affinity are used without employing a specific labelling scheme. The well-known ligand is the only component in system that requires labelling. For the second part of the project, determination of the three-dimensional structure of the fn3 fragments, we tried to crystallize the fragments in collaboration with the group of Prof. Mayans in Basel. The smallest, the two domain construct A7778, diffracted well and the structure could be solved. In parallel, we assigned its backbone resonances and could confirm that the solution structure determined by NMR is comparable. Through SAXS measurements we obtained models of the domain arrangement of the two larger fragments, A80-82 and A84-86. Conserved surfaces in adjacent titin fn3 domains appear to be arranged in a consecutive way and it is likely that this represents the scaffold for the interaction to myosin.
Publications
- (2007) Domain arrangement in double and triple domain constructs from the skeletal muscle scaffold protein titin. EMBL Hamburg Annual report 2006
Muhle-Goll, C., Fehér, K., Konarev, P. and Svergun, D.
