Designing cooperativity into the designed protein Top7

Abstract

The topology of the designed protein Top7 is not found in natural proteins. Top7 shows signatures of non-cooperative folding in both experimental studies and computer simulations. In particular, molecular dynamics of coarse-grained structure-based models of Top7 show a well-populated C-terminal folding-intermediate. Since most similarly sized globular proteins are cooperative folders, the non-natural topology of Top7 has been suggested as a reason for its non-cooperative folding. Here, we computationally examine the folding of Top7 with the intent of making it cooperative. We find that its folding cooperativity can be increased in two ways: (a) Optimization of packing interactions in the N-terminal half of the protein enables further folding of the C-terminal intermediate. (b) Reduction in the packing density of the C-terminal region destabilizes the intermediate. In practice, these strategies are implemented in our Top7 model through modifications to the contact-map. These modifications do not alter the topology of Top7 but result in cooperative folding. Amino-acid mutations that mimic these modifications also lead to a significant increase in folding cooperativity. Finally, we devise a method to randomize the sizes of amino-acids within the same topology, and confirm that the structure of Top7 makes its folding sensitive to packing interactions. In contrast, the ribosomal protein S6, which has secondary structure similar to Top7, has folding which is much less sensitive to packing perturbations. Thus, it should be possible to make a sequence fold cooperatively to the structure of Top7, but to do so its side-chain packing needs to be carefully designed.

Modeling Non‐Native Interactions in Designed Proteins

“…However, in the crystal structure (1QYS.pdb), the C-terminal half is more ordered than the N-terminal half. [25,42,48] These MD simulations correctly show that the C-terminal half of the protein folds early and forms the intermediate ensemble (Figure 3D). [46] Several variant SBMs of Top7 (which include only native interactions) fold similarly through a well-populated intermediate ( Figure 3C,D).…”

“…[44] In addition to the target structure, the design algorithm usually involves a scoring function that ranks the sequences according to how well they fit the structure and a method to span sequence space, which may depend on the scores of the sequences. [48,55] (b) How well does the designed sequence fit the structure? [48,55] (b) How well does the designed sequence fit the structure?…”

“…[45] Folding cooperativity involves the all-or-nothing folding of a protein, with the population of very few intermediates (usually none). However, since then, SBMs [48] and eSBMs [58] of Top7 have been used to predict specific mutations that may make its folding more cooperative. Although folding cooperativity is a useful feature to have, it is difficult to encode in protein design algorithms.…”

“…[43] Given a structure and a designed sequence, one can ask two questions: (a) How easy is it to design a sequence for the particular structure? [48,55] (b) How well does the designed sequence fit the structure? Here we concentrate on the second question.…”

“…[44] We choose Top7 because it has been extensively studied, both experimentally [45][46][47] and through structure-based simulations. [25,42,48,49]…”