Biological membranes are compositionally asymmetric, with distinct lipid mixtures in each leaflet, yet how this asymmetry influences lateral membrane organization remains poorly understood. Here, we use calcium-induced hemifusion to generate asymmetric giant unilamellar vesicles (aGUVs) and investigate how lipid composition modulates interleaflet coupling of liquid-liquid phase separation. Symmetric GUVs composed of cholesterol, the high-melting lipid DPPC, and a low-melting phosphatidylcholine (either 14:1-PC or 16:1-PC) were prepared at compositions exhibiting coexisting liquid-ordered (Lo) and liquid-disordered (Ld) phases. Hemifusion with a uniformly mixed supported lipid bilayer selectively altered the outer leaflet composition, producing aGUVs with controlled but variable asymmetry. Fluorescence measurements of outer leaflet exchange revealed substantial vesicle-to-vesicle variability, resulting in overlapping populations of phase-separated and uniformly mixed aGUVs. To account for this variability, we developed a statistical framework that jointly models the distribution of exchange fractions and the location of a phase boundary in asymmetric composition space, allowing all observed vesicles to contribute to the analysis. We find that aGUVs containing 14:1-PC require significantly greater outer leaflet exchange to abolish phase separation than those containing 16:1-PC. Only in the 14:1-PC system do we observe vesicles exhibiting coexistence of distinct anti-registered phases, a theoretically predicted but rarely observed regime consistent with large hydrophobic mismatch. By expressing symmetric and asymmetric miscibility boundaries in a common compositional framework, we introduce a phenomenological parameter, Δ∗, that quantifies the direction and strength of interleaflet coupling. These results demonstrate that modest changes in lipid chain length can markedly alter asymmetric phase boundaries, and provide a quantitative link between experimental observations, leaflet dominance concepts, and coupled-leaflet theories of membrane organization.

General anaesthetics are known to modulate pentameric ligand-gated ion channels (pLGICs), including GABAA, glycine, serotonin and glutamate receptors, among others. The bacterial (Gloeobacter) ion channel GLIC captures anaesthetic modulation while providing high-resolution structures in each of its distinct functional conformations, allowing for molecular definition of state-dependent binding and quantification of allosteric mechanisms. This study uses molecular dynamics, metadynamics and free energy simulations to calculate the free energy surfaces of propofol binding to open and closed channels. We observe distinct binding sites in the pore, within and between transmembrane domain subunits, as well as in the ECD, that differ in location and free energies between open and closed states. Each transmembrane site was seen to bifurcate into two adjacent sites, with overlapping residues that agree with X-ray crystallography, but with other residues identified as strongly interacting that provide predictions for experimental verification. The investigation revealed distinct pathways between intra-subunit, inter-subunit and pore sites in the open and closed states. Multiple propofol binding to neighbouring subunits was seen to modify the binding surface and lead to a third inter-subunit site within the closed state. Calculated free energy changes are consistent with previous suggestions of bimodal modulation via competing sites that act to inhibit or potentiate the channel. Inhibition via pore binding involves pore occlusion by propofol, with minor changes to pore shape and size, notably tending to close the channel when propofol is bound. Binding to intra-subunit sites was shown to thermodynamically favour the open state, in agreement with experimental evidence for potentiation. Calculations of binding to inter-subunit sites were found to involve opposing effects in its constituent sites, leading to no clear net shift in equilibrium. These observations offer molecular explanations for the effects of propofol that are likely to apply to other pLGICs, assisting in the development of improved anaesthetics with better safety margins in future.

Many proteins contain intrinsically disordered regions (IDRs) that lack stable 3-dimensional structure. IDR behavior is poorly understood, leading to challenges for biochemical and computational analysis of IDR-containing proteins. Formins are a diverse set of homodimers containing an IDR - the FH1 domain - that facilitates polymerization of the cytoskeletal protein actin by increasing the local concentration of actin monomers at the actin assembly site. A commonly accepted model of formin-based actin polymerization involves a capture-and-deliver process: one or more binding sites (proline-rich motifs, PRMs) "capture" actin monomers and then "deliver" actin to the actin assembly site. There is evidence that formin FH1 domains are dimerized on both ends, but much research has been performed with formin constructs lacking the N-terminal dimerization site. Here, we ask: What happens when N-terminal dimerization is added to the standard model of formin-mediated actin assembly? We extend the kinetic model of FH1-mediated actin polymerization by incorporating a coarse-grain polymer model of FH1 domain dynamics, modeling the FH1 domain as a freely-jointed chain. We find that N-terminal dimerization can impact polymerization rates by modifying binding site accessibility and/or local concentration of binding sites (PRMs) at the actin assembly site. Which effect dominates depends on kinetic parameters and formin properties such as FH1 domain length and binding site location. Additionally, we demonstrate that our model can be fit to experimental data and used to make predictions for the effects of N-terminal dimerization on a variety of formin family members.

Dopamine is a clinically essential pharmaceutical agent extensively used in cardiology and emergency medicine. During the drug treatment, interactions between dopamine and lipid membranes are unavoidable. However, their molecular-level interaction remains insufficiently understood. In this research, Sum frequency generation (SFG) spectroscopy was utilized to investigate the molecular-level interaction of dopamine with the negatively charged phosphatidylglycerol (PG) lipid bilayer, in real time and in situ. Our results reveal that both the outer and inner PG leaflets remain unaffected at a low dopamine concentration of 1.0 mM. In contrast, at 10 mM dopamine, sight perturbations to the outer PG leaflet were observed due to the hydrogen bonds and electrostatic interactions formed between them. The average tilt angles of the CD3 groups were calculated to be 33° and 38° before and after the addition of 10 mM dopamine, respectively. To the best of our understanding, this work experimentally quantified the dopamine-induced orientation variations of lipid molecules in a bilayer for the first time. These findings provide molecular-level insights into dopamine-membrane interactions, offering potential guidance for the development of new cardiovascular therapeutics involving dopamine-related membrane processes, such as treatments for hypotension, shock, and heart failure.

Premature ventricular contractions (PVCs) are abnormal heartbeats that can trigger life-threatening arrhythmias. One mechanism responsible for PVCs is triggered activity. In this process, spontaneous calcium (Ca2+) release from the sarcoplasmic reticulum and subsequent Ca2+ waves at the subcellular level promote transient inward currents that lead to delayed afterdepolarizations at the cellular level. In uniform, healthy tissue, such events are often benign because the surrounding healthy cells act as an electrical sink, suppressing the abnormal signals from a few pathological source cells. This phenomenon is known as "source-sink mismatch." In this study, using computational modeling, we show how structural and functional tissue heterogeneities can overcome this protective mechanism. Regions of weak electrical coupling or specific cellular arrangements can electrically isolate small clusters of pathological cells. Within such isolated regions, the number of abnormal cells can more easily reach the "critical mass" required to overcome the source-sink mismatch and initiate a PVC. This principle, where the spatial arrangement of a minority group enables it to determine a local outcome of the whole group, is analogous to the political concept of gerrymandering. Our findings provide a novel mechanistic framework for understanding arrhythmia initiation and highlight the critical role of tissue heterogeneity.

DNA packaging imposes fundamental physical constraints on genomes across the tree of life. However, most of our mechanistic understanding of these processes comes from the eukaryotic nucleosome, where highly basic histones, along with their flexible tails, coordinate DNA compaction and gene accessibility. Large DNA viruses challenge this paradigm by assembling nucleosome-like particles with a divergent histone architecture. These viral assemblies lack full canonical histone tails, contain covalently fused domains linked by structured connectors, and exhibit altered surface electrostatics, features that collectively impose distinct biophysical properties on viral chromatin. Here, we use multi-microsecond all-atom molecular dynamics simulations to dissect how histone fusion, tail loss, and connector architecture reshape the structural and energetic behavior of the Melbournevirus nucleosome, a model system for studying viral chromatin organization. We find that viral systems exhibit elevated DNA unwrapping, weaker and more transient histone-DNA contacts, and localized flexibility at connector regions. Conformational adaptation at histone junctions partially offsets these effects, with structural shifts tuned to local DNA geometry during wrapping transitions. By capturing how nucleosome dynamics shift across time and sequence, our study provides a detailed view of how chromatin architecture can be reconfigured in viral nucleosome-like systems.

The pancreatic β cell contains several functional subpopulations of insulin secretory granules (ISGs). These subpopulations vary in maturity, age, and secretory capacity. Differences in protein and lipid composition of ISGs are correlated with disease but require further study to understand how ISG remodeling regulates normal biology. Due to limitations in traditional separation methods, the extent of these subpopulations, any overlap between them, and how they are affected by insulinotropic signals have not been determined. In this work, we adapted direct current insulator-based dielectrophoresis (DC-iDEP) to separate ISGs isolated from INS-1E cells, an immortalized rat insulinoma cell line model, according to their electrokinetic mobility ratio (EKMr). We were able to separate ISG subpopulations from unstimulated cells to determine a baseline distribution and identify characteristic profiles for immature, young, and old ISGs. We then analyzed the distributions of subpopulations in cells stimulated with insulin secretion signals known to induce biophysical remodeling and maturation. We found significant changes in each subpopulation studied in response to stimulation, consistent with the increases in maturation and crystallization and changes in size reported in the literature. This work provides new insights into how the cell controls ISG remodeling and may drive future development of more effective therapies.

The proline-rich domain (PRD) of the tumor suppressor p53 plays a central role in modulating conformational dynamics and molecular interactions, yet its intrinsic structural behavior remains incompletely understood. Here, we combine extensive all-atom molecular dynamics simulations with biophysical validation to characterize the conformational ensemble of the p53 PRD. The domain behaves as an intrinsically disordered region, sampling a highly heterogeneous ensemble with average end-to-end distance and radius of gyration of 52.5 Å and 21.8 Å, respectively. Despite this disorder, transient local structure is prominent: unordered conformations dominate, followed by substantial polyproline II (PPII) content, with β-bends and turns linking conserved PXXP motifs. Circular dichroism and small-angle X-ray scattering experiments corroborate the largely disordered yet partially structured nature of the PRD. Ramachandran and contact analyses reveal that consecutive prolines, particularly Pro71-Pro72, impose steric constraints that stabilize locally extended conformations and restrict backbone collapse. To approximate the PRD within full-length p53, additional simulations were performed with restrained terminal distances, yielding reduced conformational variability and improved agreement with small-angle X-ray scattering data while preserving secondary-structure propensities. PPII helices emerge as particularly robust features, acting as stiff spacers linking the transactivation domain to downstream regions. Finally, simulations of clinically relevant variants reveal mutation-specific local perturbations: P72R disrupts consecutive proline rigidity and increases flexibility, whereas P82L abolishes a PXXP motif and its associated PPII helix. These results identify proline-mediated rigidity and transient PPII structure as key determinants of the dynamic conformational landscape of the p53 PRD.