In a groundbreaking advance in protein biophysics, researchers have unveiled detailed insights into the energy landscapes of proteins, challenging long-standing assumptions about protein stability and cooperativity. By harnessing site-resolved hydrogen-deuterium exchange (HDX) nuclear magnetic resonance (NMR) and innovative protein design strategies, the study reveals that low cooperativity in protein folding is typically associated with clusters of unstable residues rather than uniformly distributed instability. This finding, published in Nature, opens up new avenues for understanding protein dynamics, stability, and functionality on an unprecedented scale.
Proteins, the molecular workhorses of the cell, achieve their function through precise three-dimensional conformations stabilized by intricate energetic interactions. Traditionally, it was believed that proteins fold cooperatively, meaning that their structural components stabilize or destabilize together as a single unit. However, the current study disrupts this notion by demonstrating that while highly cooperative domains show uniform energy profiles across their secondary structures, low-cooperativity domains exhibit heterogeneous stability landscapes. This phenomenon manifests as specific folded elements with markedly lower stability than the rest of the protein.
By applying sophisticated HDX NMR experiments on a set of eight representative protein domains, including five low-cooperativity and three high-cooperativity examples, the researchers quantified site-specific opening energies. These data allowed the visualization of spatial distributions of stability within protein folds. For instance, in the de novo designed HHH_rd4_0557 protein, a subtle gradation of stability was observed across its three helices, whereas its counterpart HHH_rd4_0518 demonstrated a stark contrast, with one helix showing substantially reduced opening energy. Importantly, despite this discrepancy in stability, solution NMR structures confirmed that even the less stable helices adopt their native folds in solution, reflecting dynamic equilibria between native and excited states.
Further validating the structural integrity of these low-stability but correctly folded elements, the team solved solution NMR structures of designed proteins like HHH_rd4_0518 and EEHEE_rd4_0871. These structures closely matched both the computational design models and predictive AlphaFold 2 representations, indicating that native-like folding is maintained even amid regions of fluctuating local stability. This reconciliation between dynamic stability and structural fidelity challenges the simplistic binary view of protein segments as either stable or unfolded and suggests a more nuanced energetic mosaic in folded proteins.
The distinctive clustering of low-stability residues in certain domains sets them apart from typical highly cooperative proteins, which exhibit more uniform energy distribution. For example, in the LysM family, low-cooperativity variants showed instability concentrated in alpha-helical and beta-strand segments, while highly cooperative LysM domains demonstrated more consistent stability across their secondary structures. Notably, one outlier, LysM_1380, deviated from this pattern, displaying extensive heterogeneity in stability without clear spatial clustering, signaling that while clustering of unstable elements is common in low-cooperativity proteins, it is not an absolute requirement.
This heterogeneity in stability has profound implications for protein function. Unstable elements within an otherwise stable domain could act as dynamic switches or facilitate conformational flexibility essential for activity. The study’s revelation that globally stable proteins can harbor relatively unstable regions expands our understanding of the balance between rigidity and plasticity required for biological function. Such knowledge is crucial for the rational design of proteins with tailored stability profiles for therapeutic or industrial applications.
Equally revolutionary is the demonstration that protein folding energetics can be dissected with unprecedented resolution by combining HDX NMR under varying pH and temperature conditions. This method elucidates the energy barriers to local unfolding events, allowing researchers to deconvolute the cooperative nature of folding in atomic detail. By characterizing these opening energies across multiple de novo designed and natural domains, the study sets a new standard for mapping energy landscapes comprehensively.
Moreover, the findings highlight that cooperative folding is not an all-or-none phenomenon. Instead, residues within the same structural element may unfold simultaneously or through multiple partially open excited states with similar energies. This complexity underscores the intricate choreography governing protein folding thermodynamics and kinetics and suggests that future models must accommodate multiple isoenergetic pathways.
In light of these insights, the work also raises fascinating possibilities for protein engineering. Harnessing knowledge of spatially clustered instabilities could enable the design of proteins with engineered dynamic regions, tailored allosteric properties, or controlled folding pathways. The confirmation that designed proteins faithfully replicate intended folds, yet retain variable internal stability landscapes, signals the maturity of computational design tools integrated with experimental validation.
This comprehensive approach bridges the gap between purely computational predictions and experimental validation, leveraging the predictive power of AlphaFold alongside high-resolution NMR characterization. The resulting synergy provides a reliable pipeline for de novo protein design that accommodates the energetic complexity and dynamics innate to natural proteins.
Ultimately, this study represents a significant leap toward decoding the fundamental principles of protein energy landscapes. Its revelations about the spatial distribution of cooperative and non-cooperative regions offer new perspectives on how proteins negotiate the balance between structural stability and functional flexibility. As such, the work is poised to impact diverse fields from molecular biology to drug design and synthetic biology, illuminating the energetic underpinnings of life’s molecular machines.
As the authors conclude, low cooperativity often arises from distinct unstable elements rather than dispersed instability, with ensemble dynamics that maintain fold integrity despite local fluctuations. This nuanced understanding overturns simple dogma and sets the stage for future explorations into the energetic architectures of proteins, both natural and engineered. This landmark study heralds a new era of protein science, where large-scale, detailed energy landscape mapping becomes routine, accelerating discovery and innovation in the molecular life sciences.
Subject of Research: Protein energy landscapes, folding cooperativity, and spatial stability distribution in natural and de novo designed proteins.
Article Title: Large-scale discovery, analysis and design of protein energy landscapes.
Article References:
Ferrari, Á.J.R., Dixit, S.M., Thibeault, J. et al. Large-scale discovery, analysis and design of protein energy landscapes. Nature (2026). https://doi.org/10.1038/s41586-026-10465-z
DOI: https://doi.org/10.1038/s41586-026-10465-z
Keywords: Protein folding, cooperativity, energy landscape, HDX NMR, de novo protein design, structural dynamics, AlphaFold, stability distribution, protein engineering.
