Molecular simulations integrated with experiments for probing the interaction dynamics and binding mechanisms of intrinsically disordered proteins.

Intrinsically disordered proteins (IDPs) exploit their plasticity to deploy a rich panoply of soft interactions and binding phenomena. Advances in tailoring molecular simulations for IDPs combined with experimental cross-validation offer an atomistic view of the mechanisms that control IDP binding, function, and dysfunction. The emerging theme is that unbound IDPs autonomously form transient local structures and self-interactions that determine their binding behavior. Recent results have shed light on whether and how IDPs fold, stay disordered or drive condensation upon binding; how they achieve binding specificity and select among competing partners. The disorder-binding paradigm is now being proactively used by researchers to target IDPs for rational drug design and engineer molecular responsive elements for biosensing applications.

Curious Case of MAD2 Protein: Diverse Folding Intermediates Leading to Alternate Native States.

Anfinsen's dogma postulates that for one sequence there will be only one unique structure that is necessary for the functioning of the protein. However, over the years there have been a number of departures from this postulate. As far as function is considered, there are growing examples of proteins that "moonlight", perform multiple unrelated functions. With the discovery of intrinsically disordered proteins, morpheeins, chameleonic sequences, and metamorphic proteins that can switch folds, we have acquired a more nuanced understanding of protein folding and dynamics. Appearing to apparently contradict the classical folding paradigm, metamorphic proteins are considered exotic species. In this work, we have explored the free energy landscape and folding pathways of the metamorphic protein MAD2 which is an important component of the spindle checkpoint. It coexists in two alternate states: the inactive open state and the active closed state. Using a dual-basin structure-based model approach we have shown that a variety of intermediates and multiple pathways are available to MAD2 to fold into its alternate forms. This approach involves performing molecular dynamics simulations of coarse-grained models of MAD2 where the structural information regarding both of its native conformations is explicitly included in terms of their native contacts in the force field used. Detailed analyses have indicated that some of the contacts within the protein play a key role in determining which folding pathway will be selected and point to a probable long-range communication between the N and the C termini of the protein that seems to control its folding. Finally, our work also provides a rationale for the experimentally observed preference of the ΔC10 variant of MAD2 to exist in the open state.

Role of Calcium in Modulating the Conformational Landscape and Peptide Binding Induced Closing of Calmodulin.

Metal ions play an essential role in several cellular functions. Calcium is a ubiquitous regulator and is involved in numerous physiological processes. A class of proteins have evolved that sense calcium levels inside cells and act as effector molecules. Calmodulin is one such protein that gets activated after binding to calcium and thereafter interacts with its many targets. Calmodulin comprises two homologous domains that are connected by a flexible linker. The calcium-dependent flexibility of the linker results in numerous conformations of calmodulin. In this work using microsecond long MD simulations and well-tempered metadynamics, we explore how the calcium induced conformation dynamics of calmodulin is different from the inherent fluctuations of apocalmodulin and whether it has any role in preparing calmodulin for its interaction with its target-smooth muscle myosin light chain kinase (smMLCK). We have observed that calcium bound calmodulin could explore states that are predisposed toward peptide binding. We also found that though the binding of calmodulin to smMLCK peptide is calcium-independent, calcium regulates the domain to which the peptide will be bound. On the basis of our findings, we have proposed two alternate pathways for smMLCK peptide binding to calmodulin as directed by the ambient calcium concentrations. Our work proposes how calmodulin functions under physiologically dynamic calcium concentrations.

Intersubunit Assisted Folding of DNA Binding Domains in Dimeric Catabolite Activator Protein.

The abundance of protein dimers and multidomain proteins is a testimony to their importance in various cellular functions. Several mechanisms exist, explaining how they assemble. The energy landscape theory has shown that, irrespective of the mechanism followed, folding and binding of dimers and multidomain proteins are funneled processes. Using a structure based model, we have characterized the folding landscape and dimerization mechanism of the DNA binding domain (DBD) in a complex multidomain, homodimeric transcription factor, catabolite activator protein (CAP). The DBD is tethered to the nucleotide binding domain (NBD) of CAP. Our investigation revealed that, as the tethered DBD of CAP transitions from an unfolded to the folded state, complementary folding and backtracking occur between the individual subunits within the DBD. This redistributes the entropies of the DBDs in both the subunits and might play a role in consequently modulating the free energy surface to reduce the entropic folding barrier. This redistribution of entropies forms the basis of an unusual intersubunit assisted folding mechanism whereby each subunit acts as a chaperone for the other. We have also investigated the effect of tethering on the folding landscape of DBD and found that the folding landscape can change depending on the tethering conditions.

Enzyme activity of α-chymotrypsin: Deactivation by gold nano-cluster and reactivation by glutathione.

Effect of gold nanoclusters (Au-NCs) on the circular dichroism (CD) spectra and enzymatic activity of α-chymotrypsin (ChT) (towards hydrolysis of a substrate, N-succinyl-l-phenylalanine p-nitroanilide) are studied. The CD spectra indicate that on binding to Au-NC, ChT is completely unfolded, resulting in nearly zero ellipticity. α-chymotrypsin (ChT) coated gold nano-clusters exhibit almost no enzymatic activity. Addition of glutathione (GSH) or oxidized glutathione (GSSG) restore the enzyme activity of α-chymotrypsin by 30-45%. ChT coated Au-NC exhibits two emission maxima-one at 480nm (corresponding to Au10) and one at 640nm (Au25). On addition of glutathione (GSH) or oxidized glutathione (GSSG) the emission peak at 640nm vanishes and only one peak at 480nm (Au10) remains. MALDI mass spectrometry studies suggest addition of glutathione (GSH) to α-chymotrypsin capped Au-NCs results in the formation of glutathione-capped Au-NCs and α-chymotrypsin is released from Au-NCs. CD spectroscopy indicates that the conformation of the released α-chymotrypsin is different from that of the native α-chymotrypsin.

Effect of alcohol on the structure of cytochrome C: FCS and molecular dynamics simulations.

Effect of ethanol on the size and structure of a protein cytochrome C (Cyt C) is investigated using fluorescence correlation spectroscopy (FCS) and molecular dynamics (MD) simulations. For FCS studies, Cyt C is covalently labeled with a fluorescent probe, alexa 488. FCS studies indicate that on addition of ethanol, the size of the protein varies non-monotonically. The size of Cyt C increases (i.e., the protein unfolds) on addition of alcohol (ethanol) up to a mole fraction of 0.2 (44.75% v/v) and decreases at higher alcohol concentration. In order to provide a molecular origin of this structural transition, we explore the conformational free energy landscape of Cyt C as a function of radius of gyration (Rg) at different compositions of water-ethanol binary mixture using MD simulations. Cyt C exhibits a minimum at Rg ∼ 13 Å in bulk water (0% alcohol). Upon increasing ethanol concentration, a second minimum appears in the free energy surface with gradually larger Rg up to χEtOH ∼ 0.2 (44.75% v/v). This suggests gradual unfolding of the protein. At a higher concentration of alcohol (χEtOH > 0.2), the minimum at large Rg vanishes, indicating compaction. Analysis of the contact map and the solvent organization around protein indicates a preferential solvation of the hydrophobic residues by ethanol up to χEtOH = 0.2 (44.75% v/v) and this causes the gradual unfolding of the protein. At high concentration (χEtOH = 0.3 (58% v/v)), due to structural organization in bulk water-ethanol binary mixture, the extent of preferential solvation by ethanol decreases. This causes a structural transition of Cyt C towards a more compact state.

Fluorescence Dynamics in the Endoplasmic Reticulum of a Live Cell: Time-Resolved Confocal Microscopy.

Fluorescence dynamics in the endoplasmic reticulum (ER) of a live non-cancer lung cell (WI38) and a lung cancer cell (A549) are studied by using time-resolved confocal microscopy. To selectively study the organelle, ER, we have used an ER-Tracker dye. From the emission maximum (λmaxem) of the ER-Tracker dye, polarity (i.e. dielectric constant, ϵ) in the ER region of the cells (≈500 nm in WI38 and ≈510 nm in A549) is estimated to be similar to that of chloroform (λmaxem =506 nm, ϵ≈5). The red shift by 10 nm in λmaxem in the cancer cell (A549) suggests a slightly higher polarity compared to the non-cancer cell (WI38). The fluorescence intensity of the ER-Tracker dye exhibits prolonged intermittent oscillations on a timescale of 2-6 seconds for the cancer cell (A549). For the non-cancer cell (WI38), such fluorescence oscillations are much less prominent. The marked fluorescence intensity oscillations in the cancer cell are attributed to enhanced calcium oscillations. The average solvent relaxation time (<τs >) of the ER region in the lung cancer cell (A549, 250±50 ps) is about four times faster than that in the non-cancer cell (WI38, 1000±50 ps).

Unfolding and refolding of a protein by cholesterol and cyclodextrin: a single molecule study.

Unfolding/refolding of a plasma protein, human serum albumin (HSA), is studied using fluorescence correlation spectroscopy (FCS) and single molecule fluorescence resonance energy transfer (sm-FRET). Addition of cholesterol causes unfolding of HSA resulting in an increase in the hydrodynamic diameter (dH = 2rH) from 76 Å in the native state to 120 Å upon addition of 1 mM cholesterol. Addition of ß-cyclodextrin to HSA (unfolded by cholesterol) restores the hydrodynamic diameter back to 78 Å. The cholesterol induced unfolding and ß-cyclodextrin induced refolding are also monitored by measuring the distance between a FRET donor (CPM dye, D) and a FRET acceptor (Alexa 488, A) covalently attached to the protein (HSA). It is observed that the average D-A distance increases from 45 ± 1 Å at 0 mM cholesterol to 51 ± 1 Å at 1 mM cholesterol. Upon addition of ß-cyclodextrin, the D-A distance is restored to 45 ± 1 Å. The binding study indicates that nearly 94% of HSA molecules remain bound to cholesterol in the absence of ß-cyclodextrin and only 5% binds to cholesterol in the presence of ß-cyclodextrin. As much as 57% of the HSA and 99% of the cholesterol molecules bind to ß-cyclodextrin. Thus ß-cyclodextrin removes cholesterol from HSA by hydrophobic binding to cholesterol ("strip off") and also, itself binds to HSA. The conformational dynamics results suggest that addition of ß-cyclodextrin restores native like binding free energy and folding dynamics.