The temperature dependence of binding entropy is a selective pressure in protein evolution
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (PREreview)
Abstract
Proteins operate through ligand and solvent interactions governed by thermodynamics, yet the enthalpy-entropy trade-offs that guide their functional evolution remain poorly understood. The LacI/GalR family (LGF) of transcription factors provides a system for examining how these trade-offs evolve over billions of years but has seldom been studied from a full thermodynamic perspective. While the evolution of ligand specificity has been well-studied, the thermodynamic determinants underlying the changes in specificity is not well understood – especially how proteins alter thermodynamic strategies to optimize affinity. By reconstructing LGF ancestors, we reveal a shift from entropy-driven binding in the most distant ancestor, to enthalpy-driven binding in the most recent ancestor and extant LacI. The most distant ancestor is characterized by the ability to bind its ligand in an open and dynamic conformation, and we propose that entropically-driven binding is driven by the presence of entropic reservoirs. This thermodynamic binding trade-off between the most distant and most recent ancestor is in accordance with the concept of ancient life that existed in a hot Earth environment, where higher temperatures enhanced entropically-driven binding. This suggests that molecular binding mechanisms evolved not just for ligand specificity, but to adapt to environmental pressures such as cooling Earth temperatures where enthalpic binding modes are favored.
Article activity feed
-
This Zenodo record is a permanently preserved version of a PREreview. You can view the complete PREreview at https://prereview.org/reviews/15742764.
Summary
Ligand binding is driven by the combination of enthalpic and entropic thermodynamic terms, however, how evolution traverses the energy landscape to produce different specificities for ligands is not fully understood. In this work, the authors used an ancestral reconstruction of the LGF transcription factor family to identify the possible identities of the major branching transcription factors to study how enthalpic and entropic ligand binding modes may have evolved over time. They show using both DSF and ITC how the thermal stability of ancestral reconstructed TF's is lost while the thermodynamic binding components gradually switch for different carbon substrates from an entropic to …
This Zenodo record is a permanently preserved version of a PREreview. You can view the complete PREreview at https://prereview.org/reviews/15742764.
Summary
Ligand binding is driven by the combination of enthalpic and entropic thermodynamic terms, however, how evolution traverses the energy landscape to produce different specificities for ligands is not fully understood. In this work, the authors used an ancestral reconstruction of the LGF transcription factor family to identify the possible identities of the major branching transcription factors to study how enthalpic and entropic ligand binding modes may have evolved over time. They show using both DSF and ITC how the thermal stability of ancestral reconstructed TF's is lost while the thermodynamic binding components gradually switch for different carbon substrates from an entropic to enthalpic binding mode. The authors follow up their thermodynamic experiments with structural studies of the crystallized ancestral TF's with their respective substrates bound to provide a structural basis for their thermodynamic observations. Their structural analysis suggests two major sources for the thermodynamic binding modes: first, the substrate binding site evolved away from predominantly bulky hydrophobic residues in the most distant ancestor to smaller polar residues which resulted in a change in ligand binding towards forming specific hydrogen bonds both with the TF directly but also through an extensive water network (enthalpic component). Second, the authors compared the most distantly related TF's to illustrate the evolution of greater protein stability as illustrated by the greater order exhibited by Anc4, particularly in a loop region that is distal to the binding site. Further, they also show that ligand binding to Anc1 does not induce a greater degree of order compared to the apo protein which they propose represents a redistribution of entropy away from the ligand binding site.
Major points
Point 1. The authors propose that Anc1 has a spatial redistribution of entropy away from the ligand binding site to the distal loops to compensate for the loss of conformational entropy upon binding. Can they test this hypothesis by truncating or stabilizing (by point mutation) the loops? Despite a cooler earth at present, there are still organisms that live at hot temperatures. Do the extent orthologues in these organisms show entropically driven binding? Do the ancestors reported in this function as transcription factors at higher temperatures? Can the authors propose an experiment to test this? It's interesting that Anc1 is the most thermally stable of the TF's (based on the hypothesized relationship between earth temperature and protein thermal stability) yet the structure suggests it's the most disordered compared to Anc4. Can the authors comment on how this fits within their proposed model?
Point 2. The possibility that ancestral reconstruction artificially stabilizes proteins has been acknowledged in the literature (e.g. PMID 27413048). Are the authors concerned that the changes in stability observed in their work might be due to the stabilizing effect of consensus mutations?
Point 3. The authors focus on the LBD of the LGF family for structural studies and point out that Anc1 (the most distant ancestor) exhibits a greater level of disorder compared to the most recent ancestor Anc4. Is this level of disorder also expected to occur in the DNA binding domain or is it disorder unique to the LBD? In other words, does evolution only act on one domain of this family or are there correlated changes to the DBD as well (allosteric mechanism)?
Point 4. It's interesting that D-fucose binding was largely lost by Anc2 (or not tested?), can the authors provide a structural reason for that similar to their analysis with Anc4? Further, with respect to Figure 4 can authors show (perhaps just an AlphaFold prediction) what the composition of the substrate binding site looks like between each ancestor? Was there a sudden change between Anc1 and Anc2 in composition or was it more gradual (also given the D-fucose binding is almost lost between Anc1 and Anc2 - again was that actually tested)?
Point 5. "It should be noted that he apo and ᴅ-fucose-bound ΔAnc1 structures were obtain from crystals from same crystal screening drop i.e., the observed differences are not due to differences in crystallization conditions". Was this a co-crystallization experiment where two crystals were looped from a single drop - one crystal led to a structure with fucose bound and the other was apo? Crystals with different symmetry (and different crystal packing) can grow in the same drop from identical conditions. The listed space groups in Supplementary Table 2 indicated that the space group was different for the apo and fucose-bound Anc1. Is there concern that the conformational change observed between holo and apo-protein is influenced by the differences in crystal packing? The cell dimensions are similar, can the authors check that the data indexing is consistent?
Point 6. The authors point out that LacI is a functional homodimer in Figure 1 but do not distinguish whether they are investigating the homodimer or monomeric form in subsequent experiments. It would be helpful to clarify which oligomeric state they are investigating in their experiments (DSF, ITC, etc.). See minor point 3.
Point 7. D-fucose is smaller and more hydrophobic than BMDG/lactose. It follows that a protein's binding pocket that is smaller and more hydrophobic (e.g. better packed) will favor D-fucose binding. Given that core packing is a well-established mechanism of protein stabilization (e.g. PMID 27425410), how do the authors think about whether this reflects well established principles in molecular recognition and protein stability vs novel mechanistic insight specific to sugar recognition evolution?
Minor points
Point 1. Check the Figure 4 legend matches the subpanel letters. E.g. panel "a" shows BMDG not D-fucose.
Point 2. IPTG is a synthetic analogue of allolactose and is unlikely to be encountered by evolution in the context of this work. Was this included because it was in the initial carbon source panel?
Point 3. Supplementary Figure 6. Only Anc3 and Anc4 appear to have a well defined transition in the CD melt curves. Are the fits to a sigmoidal curve meaningful for the other curves? How were the uncertainties calculated for these fits? Perhaps quote confidence intervals instead of SEM?
Point 4. "ᴅ-fucose retains degrees of freedom in the Anc1 binding pocket, contrary to the idea that ligands lose their conformational entropy on binding"
How was "degrees of freedom" assessed in this case? Were multiple conformations observed in the electron density maps?
Point 5. Ensemble refinement (PMID 23251785) was used to assess protein disorder, however, it is not mentioned in the results text. The Rfree values for the input models to Supplementary Table 5 to help comparison. The Rfree values were up to 5% worse for the ensembles compared to refinement of a single structure (e.g. 21.21 vs 26.21 for the Anc1 glycerol structure). This suggests that the ensemble is worse than a single model. The authors should justify the inclusion of these results.
Point 6. The lac operon regulates genes associated with the metabolism of lactose. What did the fuc operon regulate? (Perhaps the genes are hinted at in gray text in Figure 1b?)
Point 7. "These findings are suggestive of an evolutionary transition from binding of lactose/BMDG to ᴅ-fucose." The reverse, right? D-fucose in ancestor, Lactose in extant?
Point 8. Figure 2d could be improved by adding the results from all sugars tested with each ancestor.
Competing interests
The authors declare that they have no competing interests.
-
-