From Hofmeister’s Curiosity to an Interesting Mechanism
In 1888, Franz Hofmeister published a curious observation: salts affect protein solubility in water in systematic ways. This led to the famous “Hofmeister Series,” a ranking of ions based on their ability to precipitate or solubilize proteins. Over the next century, many studies expanded on these observations of salt-induced effects on protein folding, but a unifying theory explaining how ions influence protein structure remained elusive.
Our recent study originated from a practical challenge rather than a theoretical hypothesis. In our lab’s ongoing work to study the shape memory effect of regenerated keratin — a structural protein abundant in wool, hair, and feathers — we observed some puzzling behaviors. When keratin is extracted using concentrated lithium bromide (LiBr), it does not form a fully solubilized protein solution. Instead, we observed that the proteins spontaneously aggregate into a thick, cohesive gel that can be readily separated from the surrounding solution. More unexpectedly, this protein gel solidifies almost immediately upon rehydration, without the need for dialysis or removal of the denaturants. These phenomenon contrasted sharply with the behavior observed when using organic denaturants such as urea or guanidine hydrochloride.
Illustration by Michael Rosnach (Disease Biophysics Group, Harvard University)
None of these phenomenon matched existing explanations for how LiBr supposedly works. If LiBr denatures proteins by directly binding to them, why would the keratin spontaneously separate out of solution? Why would it renature so quickly just by being placed back in water? These observations hinted at a fundamentally different mechanism, potentially not driven by direct ion-protein interactions, but something more subtle and systemic. We suspected that the answer lay not in the proteins, but in the water environment instead.
To investigate this phenomenon, we collaborated with Prof. Eugene Shakhnovich in the Department of Chemistry and Chemical Biology at Harvard. His team, particularly Junlang Liu, who became an important contributor and co-author, provided key tools including molecular dynamics simulations and an entropy-based model to characterize the water network. This collaboration enabled an interdisciplinary study that offered new insights into protein denaturation.
Reducing the Hydrophobic Effect by “Diluting” Water
Protein folding is governed by two competing factors: the entropy gained when a protein unfolds, increasing the disorder of the polypeptide chain, and the stabilizing enthalpic and entropic contributions that favor the folded state. In aqueous environments, hydrophobic components have the tendency of being buried within the protein core, minimizing disruption of the surrounding water network and stabilizing the folded structure of proteins.
But what if the water network itself were altered? What if certain ions could reduce the entropy cost associated with disturbing that network, thereby shifting the thermodynamic balance toward unfolding—not by directly interacting with the protein, but by weakening the stabilizing role of water?
We found evidence supporting this mechanism through a combination of spectroscopic measurements, thermodynamic modeling, and atomistic simulations. Our results show that LiBr and related salts denature proteins not by binding directly to them, but by perturbing the water structure. Molecular dynamics simulations revealed that LiBr solutions contain a significantly higher proportion of ordered, ion-bound water molecules, reducing the size and connectivity of the free water network that is capable of forming hydrogen bonds. This contraction of the network increases the tendency of proteins to unfold, following a pathway distinct from that of conventional organic denaturants.
Comparison of enthalpy and entropy driven denaturation mechanisms
This effect is akin to “diluting” water, not in concentration but in its structural and functional capacity. As solutes sequester water molecules into tightly bound shells, fewer are available to support the hydrogen-bonded network that stabilizes folded proteins. This reduction in water network integrity lowers the entropic penalty for unfolding and increases the likelihood of protein denaturation.
Translating Mechanistic Insights to Applications
Understanding the mechanism of denaturation allowed us to revisit our strategy for keratin regeneration. Recognizing that LiBr acts indirectly by restructuring the solvent environment, we developed a more efficient and sustainable process for protein extraction and regeneration. This process avoids harsh solvents, simplifies extraction, and supports a range of fabrication methods, providing a practical way to repurpose keratin-rich waste into valuable materials:
- Closed-loop recycling: Since LiBr does not unfold proteins through direct interaction, the same solution can be reused across multiple cycles, reducing chemical waste and simplifying processing.
- Simple separation: Denatured keratin can spontaneously aggregate into a condensed gel state, enabling simple separation from the LiBr solution without dialysis.
- Facile manufacturing: Upon rehydration, the keratin gel rapidly solidifies within seconds, allowing or a variety of manufacturing strategies including molding, film casting, fiber spinning, and 3D printing.
- Shape-memory effect: Recovery of the alpha-helix secondary structures also provides the regenerated keratin material with a hydration-responsive shape-memory effect.
Regenerated keratin material
In hindsight, the pieces had been accumulating for decades. Many early studies revealed ion-specific effects, yet no consistent mechanistic explanation emerged. The role of water entropy in protein thermodynamics, though occasionally mentioned, had not taken center stage. By assembling these pieces and grounding them in both experimental results and physical modeling, we sought to contribute a clearer understanding of protein denaturation.
While our work focused on keratin, we believe the implications extend more broadly. The critical role of solutes to modulate solvent structure and entropy landscape may influence not only protein behavior, but also the dissolution and assembly of synthetic polymers and other macromolecules. This perspective may help guide future studies across fields where solvent-mediated processes are central but not fully understood.
What began as a puzzling observation in keratin processing led us to a broader question about the role of water in molecular systems. In tracing that path, we came to a better understanding of how water, ions, and macromolecules shape one another across both biology and materials. In the end, this work speaks not just to protein regeneration, but to larger questions of solvation, entropy, and how we design with the forces that govern material behaviors.