Summary Mutations in the tumor suppressor SPOP (Speckle-type POZ protein) cause prostate, breast and other solid tumors. SPOP is a substrate adaptor of the cullin3-RING ubiquitin ligase and localizes to nuclear speckles. Although cancer-associated mutations in SPOP interfere with substrate recruitment to the ligase, mechanisms underlying assembly of SPOP with its substrates in liquid nuclear bodies, and effects of SPOP mutations on assembly are poorly understood. Here we show that substrates trigger phase separation of SPOP in vitro and co-localization in membraneless organelles in cells. Enzymatic activity correlates with cellular co-localization and in vitro mesoscale assembly formation. Diseaseassociated SPOP mutations that lead to the accumulation of proto-oncogenic proteins interfere with phase separation and co-localization in membraneless organelles, suggesting that substrate-directed phase separation of this E3 ligase underlies the regulation of ubiquitin-dependent proteostasis.

The free energy landscape theory has been very successful in rationalizing the folding behaviour of globular proteins, as this representation provides intuitive information on the number of states involved in the folding process, their populations and pathways of interconversion. We extend here this formalism to the case of the Aβ40 peptide, a 40-residue intrinsically disordered protein fragment associated with Alzheimer’s disease. By using an advanced sampling technique that enables free energy calculations to reach convergence also in the case of highly disordered states of proteins, we provide a precise structural characterization of the free energy landscape of this peptide. We find that such landscape has inverted features with respect to those typical of folded proteins. While the global free energy minimum consists of highly disordered structures, higher free energy regions correspond to a large variety of transiently structured conformations with secondary structure elements arranged in several different manners, and are not separated from each other by sizeable free energy barriers. From this peculiar structure of the free energy landscape we predict that this peptide should become more structured and not only more compact, with increasing temperatures, and we show that this is the case through a series of biophysical measurements.

The use of free-energy landscapes rationalizes a wide range of aspects of protein behavior by providing a clear illustration of the different states accessible to these molecules, as well as of their populations and pathways of interconversion. The determination of the free-energy landscapes of proteins by computational methods is, however, very challenging as it requires an extensive sampling of their conformational spaces. We describe here a technique to achieve this goal with relatively limited computational resources by incorporating nuclear magnetic resonance (NMR) chemical shifts as collective variables in metadynamics simulations. As in this approach the chemical shifts are not used as structural restraints, the resulting free-energy landscapes correspond to the force fields used in the simulations. We illustrate this approach in the case of the third Ig-binding domain of protein G from streptococcal bacteria (GB3). Our calculations reveal the existence of a folding intermediate of GB3 with nonnative structural elements. Furthermore, the availability of the free-energy landscape enables the folding mechanism of GB3 to be elucidated by analyzing the conformational ensembles corresponding to the native, intermediate, and unfolded states, as well as the transition states between them. Taken together, these results show that, by incorporating experimental data as collective variables in metadynamics simulations, it is possible to enhance the sampling efficiency by two or more orders of magnitude with respect to standard molecular dynamics simulations, and thus to estimate free-energy differences among the different states of a protein with a k B T accuracy by generating trajectories of just a few microseconds.NMR spectroscopy | protein folding | protein structure determination | bias-exchange metadynamics | enhanced sampling I n the past two decades, a series of experimental and theoretical advances has made it possible to obtain a detailed understanding of the molecular mechanisms underlying the folding process (1-6). With the increasing power of computers (7), as well as the improvements in force fields (8, 9), atomistic simulations are also becoming increasingly important because they can generate highly detailed descriptions of the motions of proteins (10-12). A supercomputer specifically designed to integrate Newton's equations of motion of proteins (7) recently broke the millisecond time barrier. This achievement has allowed the direct calculation of repeated folding events for several fastfolding proteins (13) and the characterization of molecular mechanisms underlying protein dynamics and function (14). Reliable descriptions of the folding process have also been obtained by exploiting enhanced sampling techniques (15, 16), including replica-exchange molecular dynamics (17), metadynamics (18, 19), and distributed computing (20).It has also been realized that by bringing together experimental measurements and computational methods, it is possible to expand the range of problems that may be addressed (4,...

Intrinsically disordered proteins (IDPs), which are expected to be largely unstructured under physiological conditions, make up a large fraction of eukaryotic proteins. Molecular dynamics simulations have been utilized to probe structural characteristics of these proteins, which are not always easily accessible to experiments. However, exploration of the conformational space by brute force molecular dynamics simulations is often limited by short time scales. Present literature provides a number of enhanced sampling methods to explore protein conformational space in molecular simulations more efficiently. In this work, we present a comparison of two enhanced sampling methods: temperature replica exchange molecular dynamics and bias exchange metadynamics. By investigating both the free energy landscape as a function of pertinent order parameters and the per-residue secondary structures of an IDP, namely, human islet amyloid polypeptide, we found that the two methods yield similar results as expected. We also highlight the practical difference between the two methods by describing the path that we followed to obtain both sets of data.

The nonselective cation channel TRPV1 is responsible for transducing noxious stimuli into action potentials propagating through peripheral nerves. It is activated by temperatures greater than 43 °C, while remaining completely nonconductive at temperatures lower than this threshold. The origin of this sharp response, which makes TRPV1 a biological temperature sensor, is not understood. Here we used molecular dynamics simulations and free energy calculations to characterize the molecular determinants of the transition between nonconductive and conductive states. We found that hydration of the pore and thus ion permeation depends critically on the polar character of its molecular surface: in this narrow hydrophobic enclosure, the motion of a polar side-chain is sufficient to stabilize either the dry or wet state. The conformation of this side-chain is in turn coupled to the hydration state of four peripheral cavities, which undergo a dewetting transition at the activation temperature.

Mitogen-activated protein kinases, which include p38, are essential for cell differentiation and autophagy. The current model for p38 activation involves activation-loop phosphorylation with subsequent substrate binding leading to substrate phosphorylation. Despite extensive efforts, the molecular mechanism of activation remains unclear. Here, using NMR spectroscopy, we show how the modulation of protein dynamics across timescales activates p38. We find that activation-loop phosphorylation does not change the average conformation of p38; rather it quenches the loop ps-ns dynamics. We then show that substrate binding to nonphosphorylated and phosphorylated p38 results in uniform µs-ms backbone dynamics at catalytically essential regions and across the entire molecule, respectively. Together, these results show that phosphorylation and substrate binding flatten the energy landscape of the protein, making essential elements of allostery and activation dynamically accessible. The high degree of structural conservation among ser/thr kinases suggests that elements of this mechanism may be conserved across the kinase family.

In protein structure prediction it is essential to score quickly and reliably large sets of models by selecting the ones that are closest to the native state. We here present a novel statistical potential constructed by Bayesian analysis measuring a few structural observables on a set of 500 experimental protein structures. Even though employing much less parameters than current state-of-the-art methods, our potential is capable of discriminating with an unprecedented reliability the native state in large sets of misfolded models of the same protein. We also introduce the new idea that thermal fluctuations cannot be neglected for scoring models that are very similar to each other. In these cases, the best structure can be recognized only by comparing the probability distributions of our potential over short finite temperature molecular dynamics simulations starting from the competing models. 1-4 are energy functions derived from databases of known protein conformations that empirically aim to capture the most relevant aspects of the physical chemistry of protein structure and function. They are derived by measuring the probability of an observable in an ensemble of experimental structures relative to a reference state 2,6,[16][17][18] . The conversion of the probability into an energy function is normally done employing Boltzmann's law 3 . Theory of conditional probabilities 5,6 , linear and quadratic programming 7 and information theory 8 have been invoked to justify the approach. The simplest observable 1,2 that one can use to characterize a structure is the presence of a contact between two specific residues. This procedure has been generalized to include more and more complex observables 3,6,7,[9][10][11][12][13][14][15] , making the KBPs more and more accurate. , a scoring function derived using an elegant Bayesian analysis, the composite scoring function QMEAN6 35,36 and the potential RF_CB_SRS_OD introduced by Rykunov and Fiser 18 are particularly successfull, even when tested on CASP targets 18,28 . However, even the best performing KBP is not capable of distinguishing the folded state in all the decoy sets, indicating that improvements are still possible. Moreover, the state-of-the-art KBPs exploit many complex observables, such as the distance between pair of residues or atoms, which significantly boosts the number of parameters. The weights of the various terms (which may include correlated observables, such as value of the torsions and the presence of secondary structure elements) are optimized on the decoy sets in order to obtain the best performance 10,35 . This may affect the robustness of the potentials and the possibility to use them for different purposes other than fold recognition.We here introduce a statistical potential that we call BACH (Bayesian Analysis Conformation Hunt). Its definition relies on a few binary structural observables, such as the presence of short range contacts between pair of residues, in the spirit of the original works 1,2 on KBPs. As a consequence, the energy functio...

Patterns of interacting amino acids are so preserved within protein families that the sole analysis of evolutionary comutations can identify pairs of contacting residues. It is also known that evolution conserves functional dynamics, i.e., the concerted motion or displacement of large protein regions or domains. Is it, therefore, possible to use a pure sequence-based analysis to identify these dynamical domains? To address this question, we introduce here a general coevolutionary coupling analysis strategy and apply it to a curated sequence database of hundreds of protein families. For most families, the sequence-based method partitions amino acids into a few clusters. When viewed in the context of the native structure, these clusters have the signature characteristics of viable protein domains: They are spatially separated but individually compact. They have a direct functional bearing too, as shown for various reference cases. We conclude that even large-scale structural and functionally related properties can be recovered from inference methods applied to evolutionary-related sequences. The method introduced here is available as a software package and web server (spectrus.sissa.it/spectrus-evo_webserver).