Intrinsic disorder-based design of stable globular proteins

Directed stabilization of globular proteins via substitution of a minimal number of amino acid residues is one of the most complicated experimental tasks. In this work, we have successfully used algorithms for the evaluation of intrinsic disorder predisposition (such as PONDR® FIT and IsUnstruct) as tools for searching for the weakened regions in structured globular proteins. We have shown that the weakened regions found by these programs as regions with highest levels of predicted intrinsic disorder predisposition are a suitable target for introduction of stabilizing mutations.

Intrinsic Disorder-Based Design of Stable Globular Proteins

Directed stabilization of globular proteins via substitution of a minimal number of amino acid residues is one of the most complicated experimental tasks. This work summarizes our research on the effect of amino acid substitutions on the protein stability utilizing the outputs of the analysis of intrinsic disorder predisposition of target proteins. This allowed us to formulate the basis of one of the possible approaches to the stabilization of globular proteins. The idea is quite simple. To stabilize a protein as a whole, one needs to find its "weakest spot" and stabilize it, but the question is how this weak spot can be found in a query protein. Our approach is based on the utilization of the computational tools for the per-residue evaluation of intrinsic disorder predisposition to search for the "weakest spot" of a query protein (i.e., the region(s) with the highest local predisposition for intrinsic disorder). When such "weakest spot" is found, it can be stabilized through a limited number of point mutations by introducing order-promoting residues at hot spots, thereby increasing structural stability of a protein as a whole. Using this approach, we were able to obtain stable mutant forms of several globular proteins, such as Gαo, GFP, ribosome protein L1, and circular permutant of apical domain of GroEL.

Protein-RNA affinity of ribosomal protein L1 mutants does not correlate with the number of intermolecular interactions

Ribosomal protein L1, as part of the L1 stalk of the 50S ribosomal subunit, is implicated in directing tRNA movement through the ribosome during translocation. High-resolution crystal structures of four mutants (T217V, T217A, M218L and G219V) of the ribosomal protein L1 from Thermus thermophilus (TthL1) in complex with a specific 80 nt fragment of 23S rRNA and the structures of two of these mutants (T217V and G219V) in the RNA-unbound form are reported in this work. All mutations are located in the highly conserved triad Thr-Met-Gly, which is responsible for about 17% of all protein-RNA hydrogen bonds and 50% of solvent-inaccessible intermolecular hydrogen bonds. In the mutated proteins without bound RNA the RNA-binding regions show substantial conformational changes. On the other hand, in the complexes with RNA the structures of the RNA-binding surfaces in all studied mutants are very similar to the structure of the wild-type protein in complex with RNA. This shows that formation of the RNA complexes restores the distorted surfaces of the mutant proteins to a conformation characteristic of the wild-type protein complex. Domain I of the mutated TthL1 and helix 77 of 23S rRNA form a rigid body identical to that found in the complex of wild-type TthL1 with RNA, suggesting that the observed relative orientation is conserved and is probably important for ribosome function. Analysis of the complex structures and the kinetic data show that the number of intermolecular contacts and hydrogen bonds in the RNA-protein contact area does not correlate with the affinity of the protein for RNA and cannot be used as a measure of affinity.

High-resolution crystal structure of the isolated ribosomal L1 stalk

The crystal structure of the isolated full-length ribosomal L1 stalk, consisting of Thermus thermophilus ribosomal protein L1 in complex with a specific 80-nucleotide fragment of 23S rRNA, has been solved for the first time at high resolution. The structure revealed details of protein-RNA interactions in the L1 stalk. Analysis of the crystal packing enabled the identification of sticky sites on the protein and the 23S rRNA which may be important for ribosome assembly and function. The structure was used to model different conformational states of the ribosome. This approach provides an insight into the roles of domain II of L1 and helix 78 of rRNA in ribosome function.