The development of meaningful molecular descriptions of proteins exhibiting high levels of flexibility is a key challenge for contemporary structural biology. Multi-domain proteins comprise 80% of eukaryotic proteomes, where inter-domain dynamics, mediated by intrinsically disordered linker domains, play key roles in a multitude of molecular recognition and signaling processes. These complex dynamic modes cannot be understood from static structures of either the entire protein or individual domains. Indeed intrinsically disordered linkers connecting folded subunits often encode the degrees of conformational flexibility essential to protein function. While crystallography describes individual points on the free-energy landscape, nuclear magnetic resonance (NMR), and small-angle X-ray scattering (SAXS) report on averages over ensembles of interchanging conformers present in solution, and as such offer the possibility to study the conformational behavior of proteins exhibiting high levels of domain dynamics.
Recently, we have developed a generally applicable approach for mapping the conformational energy landscape of multi-domain proteins that uses rational sampling of conformational space, via a tested statistical coil model (Flexible-Meccano), and a genetic algorithm (ASTEROIDS) to optimize members of an ensemble whose size is determined by reproduction of independent experimental data.
We have applied this approach to study the flexible, multi-domain protein U2AF65 in collaboration with Prof. Michael Sattler (Technical University Munich). In order to map the conformational-energy surface of the two domains RRM1 and RRM2, transient contact mapping from paramagnetic relaxation enhancements (PREs), orientational averaging from residual dipolar couplings (RDCs) and distance distribution functions from SAXS are simultaneously integrated into the analysis.
The spatial distribution of U2AF65 conformations is found to be highly anisotropic, comprising significantly populated inter-domain contacts that appear to be electrostatic in origin (Figure 1). This hypothesis is supported by the reduction of signature PREs reporting on expected interfaces with increasing salt concentration. The described spatial distribution reveals the complete spectrum of the unbound forms of U2AF65 that co-exist with the small percentage of a pre-formed RNA-bound domain arrangement required for RNA polypyrimidine-tract recognition by conformational selection. The conformational space sampled by the protein is thus dominated by transient electrostatic interactions that allow the RNA binding conformation to remain populated in solution, facilitating binding kinetics.
Figure 1. Representation of the conformational space sampled by the two domain protein RRM1/RRM2 of U2AF65 using the combination of RDCs, PREs and SAXS. The position of RRM2 with respect to RRM1 is shown as a probability distribution in terms of a 3D density maps (5% population contour). The conformational space is non-uniformly sampled, with an encounter complex due to electrostatic contacts between the two domains.
More generally, the approach to describing conformational equilibria of multi-domain proteins can be further combined with other experimental data that are sensitive to domain dynamics.
We are now applying the developed approaches to study other multi-domain proteins such as the C-terminal region, 627-NLS, of influenza PB2 polymerase, where domain dynamics appears to play a role in temperature adaptation to different hosts.
Recent related publications :
J. Huang, L. Warner, C. Sanchez, F. Gabel, T. Madl, C. Mackereth, M. Sattler* and M. Blackledge*.
Transient electrostatic interactions dominate the conformational equilibrium sampled by multidomain splicing factor U2AF65 : A combined NMR and SAXS study.
J. Am. Chem. Soc. 136, 7068-7076 (2014).