Naganathan Athi N.

Differential scanning calorimetry (DSC) is a very powerful tool for investigating protein folding and stability because its experimental output reflects the energetics of all conformations that become minimally populated during thermal unfolding. Accordingly, analysis of DSC experiments with simple thermodynamic models has been key for developing our understanding of protein stability during the past five decades. The discovery of ultrafast folding proteins, which have naturally broad conformational ensembles and minimally cooperative unfolding, opens the possibility of probing the complete folding free energy landscape, including those conformations at the top of the barrier to folding, via DSC. Exploiting this opportunity requires high-quality experiments and the implementation of novel analytical methods based on statistical mechanics. Here, we cover the recent exciting developments in this front, describing the new analytical procedures in detail as well as providing experimental guidelines for performing such analysis.

Mutational perturbations of protein structures, i.e., phi-value analysis, are commonly employed to probe the extent of involvement of a particular residue in the rate-determining step(s) of folding. This generally involves the measurement of folding thermodynamic parameters and kinetic rate constants for the wild-type and mutant proteins. While computational approaches have been reasonably successful in understanding and predicting the effect of mutations on folding thermodynamics, it has been challenging to explore the same on kinetics due to confounding structural, energetic, and dynamic factors. Accordingly, the frequent observation of fractional phi-values (mean of ~0.3) has resisted a precise and consistent interpretation. Here, we describe how to construct, parameterize, and employ a simple one-dimensional free energy surface model that is grounded in the basic tenets of the energy landscape theory to predict and simulate the effect of mutations on folding kinetics. As a proof of principle, we simulate one-dimensional free energy profiles of 806 mutations from 24 different proteins employing just the experimental destabilization as input, reproduce the relative unfolding activation free energies with a correlation of 0.91, and show that the mean phi-value of 0.3 essentially corresponds to the extent of stabilization energy gained at the barrier top while folding.

Deciphering the folding mechanism of small single-domain proteins has a long and well-chartered history that has been and still is aided by numerous experimental and computational approaches. The computational tools at the disposal of the folding community range from all-atom molecular simulations to structure-based models. In this review, we highlight one such structure-based statistical mechanical model termed the Wako-Saitô-Munõz-Eaton (WSME) model. We have, over the past few years, made the model physically more realistic by systematically introducing mean-field terms for solvation and electrostatics apart from conventional packing interactions. The WSME model can simply be calibrated with equilibrium unfolding curves and various features such as heat capacity thermograms, free-energy surfaces or profiles and hence the folding mechanism, changes in stability upon point mutations or certain post-translational modifications, thermodynamic vs. dynamic effects and possible connections with function fallout of the model without additional calibration. The model requires only a small set of tunable thermodynamic parameters (∼3-4) allowing for a tremendous scope in further improvement of its energy function. Most importantly, it can be employed as a rapid, physical and ensemble-based tool to directly characterize experimental equilibrium and kinetic rate and amplitude data (in real world units), that is not conventionally possible in other native-centric treatments. We believe that the WSME model is now poised to address numerous questions in the field of protein folding including pathway heterogeneity, structural-energetic relations, quantifying disorder and the effect of point mutations in disease.