Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

A Generic Program for Multistate Protein Design

Figure 6

Curious cases from negative design.

A) Placing both F52 and W63 on RalA (green) destabilizes its interaction with Sec5 (magenta). In the docked conformation, the F52 and W63 rotamers collide in the least-awful-rotamer placement available. In the unbound state (orange) these residues relax out of collision. W63 disrupts binding with Sec5 through F52, but neither residue disrupts binding on its own. Unfortunately, W63 is incompatible with the RalA backbones from the crystal structures, though it is compatible with the RalB backbone in the NMR structure. Here, a discrepancy between the backbone conformations of Ral in its various states lead to a questionable design. B) The Missing Rotamer Problem encountered while trying to redesign chain E of human uracil-DNA glycosylase bound to a protein inhibitor (PDB ID: 1UGH). The mutation F267 on chain E (green) is consistently chosen by multistate design when optimizing for binding energy, not because F267 forms favorable contacts with its (polar) neighbors on chain I (cyan), but because the rotamer it finds in the bound state is absent from the set of rotamers for the unbound state; the best available rotamer for the unbound state (orange) has a high-energy collision with the C atom on residue 279. The green rotamer collides with the chain E backbone (with an energy 5.1 REU) and, in the unbound, is pruned by Rosetta's bump check machinery (threshold of 5.0 REU); however, in the bound state, slightly favorable interactions with the chain I backbone rescue this rotamer by pushing its energy just beneath the bump-check threshold (4.9 REU). Placing phenylalanine at 267 and anything besides glycine at 279 produces a large energy difference in the bound and unbound states which masquerades as an excellent binding energy.

Figure 6

doi: https://doi.org/10.1371/journal.pone.0020937.g006