We used replica exchange molecular dynamics (REMD) simulations to evaluate four different AMBER force fields and three different implicit solvent models. Our aim was to determine if these physics-based models captured the correct secondary structures of two α-helical and two β-peptides: the 14-mer EK helix of Baldwin and co-workers, the C-terminal helix of ribonuclease, the 16-mer C-terminal hairpin of protein G, and the trpzip2 miniprotein. The different models gave different results, but generally we found that AMBER ff96 plus the implicit solvent model of Onufriev, Bashford, and Case gave reasonable structures, and is fairly well-balanced between helix and sheet. We also observed differences in the strength of ion pairing in the solvent models, we but found that the native secondary structures were retained even when salt bridges were prevented in the conformational sampling. Overall, this work indicates that some of these all-atom physics-based force fields may be good starting points for protein folding and protein structure prediction. I. Are Physics-Based Force Fields Accurate Enough for Protein Structure Predictions?All-atom force fields, such as CHARMM, 1 AMBER, 2,3 and OPLS, 4 are widely used for computer simulations of the properties of proteins and small peptides. However, such force fields are not yet commonly used to predict native protein structures, in part because of computational limitations in the conformational sampling of systems with many degrees of freedom, as in proteins, but also in part because of concerns that the force fields themselves might not be good enough. It has been difficult to distinguish whether physics-based protein structure prediction has been hindered by insufficient sampling or by limitations in the force fields.However, a few tests on protein structure prediction have been possible so far because of mechanistically focused sampling or large dedicated computer resources. There are some notable successes indicating that modern physics-based force fields might be accurate enough to predict the native structures of proteins. Using supercomputer resources, Duan and Kollman performed a milestone microsecond molecular dynamics simulation of the 36-residue villin *Corresponding author. E-mail: E-mail: shell@engineering.ucsb.edu. NIH Public Access NIH-PA Author ManuscriptNIH-PA Author Manuscript NIH-PA Author Manuscript headpiece in explicit solvent starting from an unfolded conformation. 5 Scheraga's group, starting from a random configuration, folded the 46-residue protein A to within 3.5 Å rootmean-square deviation (rmsd) using Monte Carlo dynamics with an implicit solvation model. 6 Another milestone was the folding of villin by the Pande group on Folding@Home, a distributed grid computing system running on the screen savers of volunteer computers worldwide. 7Recently, higher accuracies have also been achieved. High-resolution structures of villin have recently been reached by Pande et al. 8 and Duan et al. 9,10 In addition, three groups have folded the 20-residue...

Sensing and responding to signals is a fundamental ability of living systems, but despite substantial progress in the computational design of new protein structures, there is no general approach for engineering arbitrary new protein sensors. Here, we describe a generalizable computational strategy for designing sensor-actuator proteins by building binding sites de novo into heterodimeric protein-protein interfaces and coupling ligand sensing to modular actuation through split reporters. Using this approach, we designed protein sensors that respond to farnesyl pyrophosphate, a metabolic intermediate in the production of valuable compounds. The sensors are functional in vitro and in cells, and the crystal structure of the engineered binding site closely matches the design model. Our computational design strategy opens broad avenues to link biological outputs to new signals.

Sensing and responding to signals is a fundamental ability of living systems, but despite remarkable progress in computational design of new protein structures, there is no general approach for engineering arbitrary new protein sensors. Here we describe a generalizable computational strategy for designing sensor/actuator proteins by building binding sites de novo into heterodimeric protein-protein interfaces and coupling ligand sensing to modular actuation via split reporters. Using this approach, we designed protein sensors that respond to farnesyl pyrophosphate, a metabolic intermediate in the production of valuable compounds. The sensors are functional in vitro and in cells, and the crystal structure of the engineered binding site matches the design model with atomic accuracy. Our computational design strategy opens broad avenues to link biological outputs to new signals.One Sentence SummaryAn engineering strategy to design modular synthetic signaling systems that respond to new small molecule inputs.

There is a growing interest in engineering proteins whose function can be controlled with the spatial and temporal precision of light. Here, we present a novel example of a functional light-triggered switch in the Ca-dependent cell–cell adhesion protein E-cadherin, created using a mechanism-based design strategy. We report an 18-fold change in apparent Ca2+ binding affinity upon illumination. Our results include a detailed examination of functional switching via linked changes in Ca2+ binding and cadherin dimerization. This design opens avenues toward controllable tools that could be applied to many long-standing questions about cadherin’s biological function in cell–cell adhesion and downstream signaling.