Fold-switching proteins

Abstract

Globular proteins are expected to assume folds with fixed secondary structures, α-helices and β-sheets. Fold-switching proteins challenge this expectation by remodeling their secondary and/or tertiary structures in response to cellular stimuli. Though these shapeshifting proteins were once thought to be haphazard evolutionary byproducts with little intrinsic biological relevance, recent work has shown that evolution has selected for their dual-folding behavior, which plays critical roles in biological processes across all kingdoms of life. The widening scope of fold switching draws attention to the ways it challenges conventional wisdom, raising fundamental unanswered questions about protein structure, biophysics, and evolution. Here we discuss the progress being made to answer these questions and suggest future directions for the field.

Figure 1.. Three well-characterized fold-switching proteins.

(a) The C-terminal domain (CTD, teal) of E. coli RfaH reversibly switches from an α-helical to β-sheet fold upon binding RNA polymerase and operon polarity suppressor DNA. Adapted from (82). (b) Under physiological conditions, the human chemokine XCL1 (mustard) reversibly interconverts between a chemokine fold involved in signaling and a dimeric β-sheet fold that binds bacterial and fungal pathogens. (c) The C-terminal subdomain of S. elongatus KaiB (red) switches secondary structures when its ground state tetrameric form dissociates into a monomer and binds the circadian clock protein KaiC. Single-folding regions of all proteins are colored gray. Secondary structures diagrams were generated using SSDraw (21) with corresponding residue numbers above and below; all ribbon diagrams in this paper were generated with PyMOL (97).

Figure 2.. Schematic representation of energy landscapes modeled by structure-based models (SBMs).

(a) In single-basin SBM, the energy landscape is shaped as a funnel biased toward a single native conformation, typically derived from the contact map of the folded structure; Ubiquitin is used as an example here (PDB ID: 3EHV). This enables efficient sampling of folding pathways and thermodynamic properties. (b) For metamorphic proteins such as KaiB (PDB IDs: 2QKE and 5JYT), which adopt multiple native states, the SBM must be generalized to a double-well model. This is achieved by incorporating separate energy terms, each favoring a different native topology, thereby allowing exploration of fold-switching dynamics and conformational equilibria. Energy terms are derived from reference structures. These structures can be obtained through experiments or simulation modeling.

Figure 3.. Evolution of fold switching in protein families

(a) Evolved fold switching is illustrated by helix-turn-helix (HTH) response regulators evolving to winged-helix (wH) folds. FixJ (black) is an example of a HTH response regulator; KdpE (yellow) exemplifies a wH fold (left). Maximum-likelihood phylogenetic trees suggest an evolutionary path between response regulators with HTH and wH folds. Sequences with C-terminal domains annotated as HTH/wH from NCBI protein records are gray/yellow. The clade with 12 sequences bridging the two folds is highlighted in pink. AlphaFold models suggest that stepwise mutations through the bridge sequence caused HTH proteins to switch to wH. Distance units in both trees are arbitrary, though sequences further in space have more distant evolutionary relationships. (B) Adaptive fold switching can be either monophyletic (XCL1) or possibly polyphyletic (RfaH). Monophyletic Fold-Switching: fold switchers all came from a single common ancestor. This simplified phylogenetic tree traces XCL1’s evolutionary path from its last common ancestor (Anc.0), which possessed the canonical chemokine fold, to two distinct subfamilies highlighted in red and yellow. The red subfamily retains only the chemokine fold, while the yellow subfamily can adopt both the chemokine fold and an alternative fold. Dishman et al. propose that fold-switching in XCL1 allows the protein to adapt its function to physiological needs: at sites of infection, the dimeric alternative fold helps fight bacteria, while the monomeric chemokine fold activates leukocytes via the XCR1 receptor, thus providing two functions without requiring the synthesis of a new protein or fold. The emergence of fold-switching (indicated by a yellow bar on the tree) was detected by analyzing ancestral sequences (Anc.0 to Anc.4) that were resurrected using ancestral sequence reconstruction (ASR). Access to a second native-state structure arose after the loss of a conserved disulfide bond (Anc. 2) and the accumulation of mutations that either destabilized the chemokine fold or favored the alternative fold (yellow). (C) Polyphyletic Fold-Switching: In contrast, polyphyletic fold-switching refers to multiple, independent occurrences of fold-switching in the RfaH/NusG family, which contains highly diverse sequences. In this tree, NusG branches that are non-switching folds are highlighted in blue, while the RfaH fold-switching branch is shown in red. The phylogeny was constructed using Maximum Likelihood analysis of ~6,000 unique sequences from the NusG/RfaH family, with branch support validated by bootstrapping. The large red square marks Q57818 in the clade containing all archaeal sequences (shown in black) and represents an Spt5 protein used as a reference for profile realignment in the multiple sequence alignment. The tree is rooted between the predominantly non-fold-switching NusG subfamily (blue) and the fold-switching clades.

Figure 4.. Predicting E. coli proteins that may switch folds.

(A). 2126 E. coli and phage proteins were run through CF-random to test whether they switch folds (59). Seashell-like image represents these proteins by length; the inner circle represents 1111 candidates for which sufficiently deep MSAs could not be generated, and outer, the 2126 proteins that were then run through CF-random. If two or more distinct conformations were identified, such as in the case of the successfully identified fold-switching E. coli protein, RfaH, we tested for dual-fold coevolution using ACE (95). If coevolutionary evidence for both folds was found, the protein was considered a putative fold switcher. Light gray/black contacts on upper/lower diagonals are unique to CF-random predicted dominant/alternative conformations. Teal contacts are from ACE. Medium gray contacts are common to both folds. (B). Putative fold-switching proteins are involved in diverse functions. (C). Examples of putative hits. CF-random correctly identified the fold-switching protein MinE from its thousands of candidates, indicated by green check. NinH is transcription factor protein that may undergo an α-helix-to-β-sheet transition, and YmcE is a bacterial antitoxin predicted to assume two different folds with lower confidence; both YmcE and NinH are putative switchers indicated by gray question marks.