Proteins must fold into precise protein structure to function. The folding process is governed by the thermodynamic principle that the native state represents the global free energy minimum under physiological conditions. Understanding protein folding thermodynamics is fundamental to predicting structure, designing proteins, and treating diseases caused by misfolding.
The Thermodynamic Framework
Protein folding is described by the Gibbs free energy equation: ΔG = ΔH − TΔS. For folding to occur spontaneously, ΔG must be negative. The folded state is stabilized by favorable enthalpy contributions from hydrogen bonds, van der Waals interactions, and electrostatic interactions. The unfolded state has higher conformational entropy because the polypeptide chain is flexible. However, the burial of hydrophobic side chains in the protein interior releases ordered water molecules into bulk solvent, producing a large favorable entropy gain that drives folding.
The free energy difference between the folded and unfolded states is remarkably small for a stable protein, typically 5–15 kcal/mol, equivalent to only a few hydrogen bonds. This marginal stability means that single point mutations can significantly alter folding thermodynamics and lead to disease.
The Hydrophobic Effect
The hydrophobic effect is the primary driving force for protein folding. Nonpolar side chains are sequestered in the protein core to minimize contact with water. When hydrophobic groups are exposed to water, adjacent water molecules form ordered clathrate-like cages around them, decreasing entropy. Burial of these groups releases the ordered water molecules, increasing the entropy of the solvent and providing a large favorable contribution to ΔG.
The hydrophobic effect is temperature-dependent. It is strongest at room temperature and weakens at low temperatures, explaining cold denaturation. At high temperatures, the entropic penalty of ordering water around exposed hydrophobic groups eventually favors unfolding.
Hydrogen Bonding and Electrostatic Interactions
Hydrogen bonds within the protein backbone (between amide NH and carbonyl CO groups) and between side chains stabilize secondary structures like α-helices and β-sheets. Although water can compete for hydrogen bonds in the unfolded state, the low dielectric environment of the protein interior strengthens these interactions.
Electrostatic interactions include salt bridges between oppositely charged side chains and dipole interactions. In the folded state, charged groups are often paired or exposed on the surface. Buried charges are rare and usually functionally important, such as in enzyme active sites.
van der Waals Interactions
Close packing of atoms in the protein core generates favorable van der Waals interactions. These weak, short-range forces depend on the complementarity of interacting surfaces. The tight packing in native proteins, comparable to that in organic crystals, indicates that van der Waals interactions contribute significantly to folding stability.
Conformational Entropy
The unfolded polypeptide chain has substantial conformational entropy because each backbone dihedral angle can sample multiple rotameric states. Folding restricts these degrees of freedom, creating an unfavorable entropy change of 1–2 kcal/mol per residue. This entropic penalty must be overcome by the favorable interactions described above.
Experimental Measurement
Differential scanning calorimetry (DSC) directly measures the heat capacity change upon unfolding, providing ΔH, ΔCp, and Tm. Circular dichroism (CD) spectroscopy monitors secondary structure content as a function of temperature. Fluorescence spectroscopy follows tryptophan burial. NMR spectroscopy provides residue-level resolution of folding pathways. Isothermal titration calorimetry (ITC) measures binding thermodynamics that couple to folding.
Two-State vs. Multi-State Folding
Small proteins often fold via a two-state mechanism in which only the fully folded and fully unfolded states are significantly populated. The folding transition is cooperative, meaning that partial unfolding is thermodynamically unfavorable. Larger proteins and those with multiple domains may exhibit multi-state folding with stable intermediates.
Protein Misfolding and Disease
Failure to fold correctly or maintain the native state leads to aggregation and disease. Amyloidoses such as Alzheimer’s, Parkinson’s, and prion diseases involve the formation of cross-β-sheet aggregates. Mutations that destabilize the native state promote misfolding. Cellular quality control systems including chaperones and the ubiquitin-proteasome pathway normally manage misfolded proteins, but their capacity can be exceeded in aging and disease.
Computational Approaches
Molecular dynamics simulations, often combined with enhanced sampling techniques, model the folding landscape at atomic resolution. Coarse-grained models reduce computational cost and can capture folding pathways for larger proteins. Machine learning methods such as AlphaFold predict native structures from sequence but do not directly provide thermodynamic information.
