Highlights from the Faraday Discussion on Structural and Functional Asymmetry of Plasma Membranes, London, UK, April 2025

Abstract

The Faraday Discussion on Structural and functional asymmetry of plasma membranes was held from April 23rd to 25th 2025 at Burlington House in London, the home of the Royal Society of Chemistry. Established scientists, early-career researchers and graduate students gathered to discuss recent developments in the analysis, modelling and engineering of plasma membranes and their mimics. As a Faraday Discussion, the focus was on facilitating intellectual debate, with papers circulated in advance and concise summaries presented before opening the floor to discussion. Lively debate was encouraged and attendees participated in an open and engaging dialogue of pertinent questions in the field. In this report we summarise the presented results, ensuing discussions and key areas of future exploration in membrane asymmetry.

Introduction & Spiers memorial lecture

Gerald Feigenson (Cornell Univeristy, USA) kicked off proceedings at Burlington House (Fig. 1) with the Spiers memorial lecture with an overview of key historical findings on membrane asymmetry, including the work of Dennis Chapman. He observed the timeliness of holding this meeting in the 100th anniversary year of Gorter and Grendel's seminal paper in which it was first proposed that biological membranes are lipid bilayers composed of a membrane only two lipid molecules thick.¹ Gerald brought enthusiasm, stating his privilege to be working in the field of membrane asymmetry and contributing to the fundamental theory of this new state of matter. He noted the challenges in establishing the asymmetry of plasma membranes, let alone in investigating the underlying principles of this asymmetry. He attributed this to the “unordinary chemistry” of the plasma membrane, a term he coined regarding the experimental challenges of membranes in which products can escape the bilayer – in “ordinary chemistry” you would never expect your reagents to escape through the reaction vessel!

Fig. 1 The courtyard of Burlington House and the entrance to the Royal Society of Chemistry.

The historic role of phospholipases and SMases in elucidating plasma membrane composition was introduced, in particular in the work of Laurens Van Deenen, a leading figure in biomembrane research. In true Faraday Discussion style, Gerald challenged established and contemporary proposals, particularly regarding the impact of phospholipase activity on membrane pressure. This initiated a lively discussion on the number of phospholipids in the cytoplasmic vs. outer leaflet of plasma membranes that was to be revisited many times over the course of the Faraday Discussion (Fig. 2).

Fig. 2 Meeting attendees in the courtyard outside Burlington House.

Key aims for the field were also proposed, including the development of methods to reliably extract the plasma membrane from biological cells (without extracting additional cytoplasmic lipids), and to facilitate spatiotemporal resolution of membrane asymmetry, for example the development of increasingly powerful imaging methods.

Session 1: Plasma membrane asymmetry and lipid homeostasis

The first discussion session explored the role of lipid bilayer asymmetry under physiological conditions, and how active cortical forces translate those chemical imbalances into lateral structure.

Erwin London (Stony Brook, USA) set the scene by revisiting classic questions about lipid membrane domain formation through the lens of leaflet asymmetry. He discussed the use of lipid scramblases on living cells to interrogate how leaflet asymmetry disruption impacts membrane organisation, whilst highlighting factors that complicate the interpretation of FRET and membrane order experiments (https://doi.org/10.1039/d4fd00211c).

Kandice R. Levental (University of Virginia, USA) presented work exploring the membrane changes observed during receptor-mediated activation of mast and T cells (https://doi.org/10.1039/d4fd00205a). They found activated immune cells to exhibit plasma membrane scrambling, characterised by foci of externalised PS and PE existing for surprisingly long time periods (1–2 h), before resolving back into an asymmetric state, suggesting a distinct (yet-to-be determined) function of the asymmetry loss. With their work, the Levental group demonstrated the essence of scientific exploration, asking questions on their observations: why don’t the foci drift laterally? Are other lipids scrambled, too? How is the scrambling reversed? Are the scrambled regions physiologically isolated? And what are the functional consequences of local asymmetry loss?

Madan Rao (National Centre for Biological Sciences (TIFR), India) next addressed the audience, highlighting the role of nonequilibrium forces in cell membrane organisation in a move from an experimental to theoretical perspective. He unveiled an “active Flory–Huggins” framework accounting for active contractile stresses, arising from PS–actomyosin interactions in the inner leaflet, predicting active emulsions of lo domains above the equilibrium phase transition temperature (https://doi.org/10.1039/d4fd00207e).

Subsequent to the brief presentations, the speakers lined up on stage to be exposed to a vigorous discussion, ranging from probe artefacts to theoretical limits of asymmetric stability. This crucial part of the Faraday Discussion allowed both panellists and audience members to debate where the field is heading and how to best tackle emerging questions. Participants denoted a pressing need to develop strategies for controlling asymmetric membrane systems in the absence of Ca²⁺. Discussing the presented work in a shared effort, it was established that in some cases lipid bilayer asymmetry suppresses phase separation, whereas in others, it induces phase separation.

Closing remarks of the session were devoted to a long-standing hurdle: designing a “truly ideal” probe. The panel agreed that the next generation of lifetime reporters must be (i) chemically compact, (ii) leaflet-locked and (iii) biased towards liquid-ordered domains.

Following the introductory lecture given by Gerald Feigenson, Ilya Levental, drafter of the contemporary proposal Gerald challenged, gave an impromptu lecture defending his laboratory's work on the numerical asymmetry of phospholipids in plasma membranes. He elaborated on their careful considerations of the choice of enzymes to digest all lipid targets. He also discussed the necessity of SMase, addressing the slow-down of PLA2 due to a potential build-up of monolayer pressure in the outer leaflet after an accumulation of PLA2 digestion products. He highlighted methodological difficulties in measuring this pressure, and shared contrary evidence in mono- and bilayer systems from literature. He ended his talk by stating the primacy of evidence over historical convention or fame.

Session 2: Engineering plasma membrane mimics

This session focussed on the development and functionalities of synthetic membranes with leaflet asymmetry (Fig. 3).

Fig. 3 Session two underway at the library at Burlington House, which served as the lecture hall for the Faraday discussion.

Heiko Heerklotz (University of Toronto, Canada; University of Freiburg, BIOSS Center for Biological Signaling Studies, Germany) began the session introducing his work that demonstrated that asymmetric large unilamellar vesicles (aLUVs) can be prepared in the gel phase using lipid exchange with cyclodextrin. The temperature at which the aLUVs were prepared was seen to influence bilayer coupling, where aLUVs prepared with both leaflets in the fluid state showed stronger coupling than aLUVs prepared with either leaflet in the gel state (https://doi.org/10.1039/d5fd00003c).

Neha Kamat (Northwestern University, USA) next presented work showing that the efficiency of T-cell activation by antibody-coated lipid nanocarriers was correlated to the proximity of the melting transition temperature of the lipid nanocarrier to the temperature at which the study was conducted. Closely matching the LUV melting transition (T_m) to the temperature of the activation studies resulted in the highest percentage of activated Jurkat cells. She proposed that close to the T_m, membrane density fluctuations should promote local accumulations of the antibody in the LUVs, and increased membrane undulations should facilitate antigen–receptor binding (https://doi.org/10.1039/d5fd00002e).

Arne Gericke (Worcester Polytechnic Institute, USA) next joined us online, and presented useful insights into the fabrication of asymmetric Giant Unilamellar Vesicles (aGUVs) with phosphoinositide lipids sequestered in only one leaflet. The hemifusion method between a GUV and a supported lipid bilayer using 1 mM Mg²⁺ was shown to successfully produce aGUVs with phosphoinositide lipids in the outer leaflet (https://doi.org/10.1039/d4fd00191e).

Francisco Barrera (University of Tennessee, USA) rounded up the short talks describing the development of a simple and robust method for polysaccharide functionalization of artificial membranes, in particular the conjugation of chondroitin sulphate to LUVs, supported lipid bilayers (SLBs) and giant unilamellar vesicles (GUVs). This will be useful in future work interrogating the role of the carbohydrate coating of plasma membranes known as the glycocalyx (https://doi.org/10.1039/d4fd00195h).

Part of the following discussion centred on how to prepare simple synthetic systems with rationally designed membrane asymmetry. This is commonly achieved by engineering lipid flip-flop in a symmetric system, by using exchange proteins, lipases such as PLA2, or methylcyclodextrins. It was noted that lipid flip-flop may impact membrane tension, and that it is important to consider the mechanical stress induced in asymmetric synthetic membranes, and whether it is possible to relax this with mechanical methods such as the use of a micropipette or osmotic stress. This led to many questions, including the fundamental question of how we define asymmetry. Besides lipid composition, factors such as the internal and external buffers must be considered, whilst even synthetic systems are never perfectly symmetric. Finally, the importance of avoiding imaging artefacts was discussed. It was raised that asymmetry cannot be reliably characterised using fluorescence intensity readouts alone.

Session 3: Structure and dynamics of asymmetric membranes

This session brought together experimental, theoretical, and computational studies that explore how compositional and structural asymmetries influence membrane mechanics, remodelling, and functional organization.

Markus Deserno (Carnegie Mellon University, USA) opened the third session by introducing a thermodynamic framework for leaflet coexistence in asymmetric ternary lipid systems via coarse-grained simulations. He concluded that the differential stress not only enables compositional asymmetry via cholesterol transition but also contributes to the total torque that induces membrane curvature (https://doi.org/10.1039/d4fd00196f).

Reinhard Lipowsky (Max Planck Institute of Colloids and Interfaces, Germany) presented quantitative analysis on membrane tension in flat bilayers, unilamellar nanovesicles and GUVs. The study distinguished the two leaflet tensions from the bilayer tension for planar and nanovesicle bilayers, and the contributions from the mechanical membrane tension and the curvature-elastic tension to the total membrane tension (https://doi.org/10.1039/d4fd00184b).

Michael Kozlov (Tel Aviv University, Israel) rounded up the first section of Session 3 with a theoretical model on the differential intra-membrane contact energies between a pure lipid domain and the monolayer beneath, and the resultant lipid matrix deformations. He demonstrated the universally attractive interactions for systems with a domain to monolayer rigidity ratio smaller than a critical value. For highly rigid domains, the interaction becomes repulsive at short distances and attractive at long distances (https://doi.org/10.1039/d4fd00186a).

The resulting discussion explored how asymmetry influences phase behaviour and curvature dynamics in membranes, with particular interest in the emergence of coexisting phases, the mechanical properties of asymmetric membranes, such as tension and torque, and the challenges of experimentally measuring and validating these parameters.

The second section of Session 3 began with a presentation from Daniel Huster (University of Leipzig, Germany), who investigated the structural and dynamic consequences of lipid asymmetry in multilamellar bilayers. They used POPE/POPG in the inner leaflet and either POPC or POPS in the outer leaflet, employing ²H NMR to compare the chain order and relaxation dynamics with symmetric membranes. They found that leaflet asymmetry induces a redistribution of acyl chain order from the lower to the upper part of the chains, and increased membrane stiffness in the headgroup instead of the chain segments, resulting from the lateral stress and lateral pressure profile differences (https://doi.org/10.1039/d4fd00192c).

Next, Milka Doktorova (Stockholm University, Sweden) presented atomistic molecular dynamics simulations of asymmetric lipid bilayers to generate synthetic experimental observables. Their work showed that lipid packing decreases with increasing cholesterol concentration for asymmetric bilayers under differential stresses. The differential leaflet packing densities can be revealed by small-angle scattering experimentally, and high sensitivity to phospholipid imbalances was shown in computationally generated cryo-EM intensity profiles (https://doi.org/10.1039/d4fd00200h).

The final talk of this section was delivered by Oded Farago (Ben Gurion University of the Negev, Israel). They performed Monte Carlo simulations of ultra coarse-grained asymmetric lipid bilayers and introduced a novel technique to quantify elastic properties, spontaneous curvature and non-local bending modulus. Phase separation into regions with different densities was observed for increased degrees of asymmetry, and larger condensed lipid density than liquid to gel density, which led to local membrane bending and buckling due to the coupling between these regions (https://doi.org/10.1039/d4fd00182f).

The discussion explored how lipid composition, temperature, and cholesterol concentration influence membrane structure, order, and phase behaviour in asymmetric systems. Broader discussions were raised about the interpretation of simulation outputs, such as cryo-EM profiles and area per lipid, and the experimental challenges of constructing and characterizing asymmetric membranes.

The third part of the third session began with the contribution of Sylvie Roke (École Polytechniqué Fédérale de Lausanne (EPFL), Switzerland). They used second harmonic water imaging to examine a variety of membrane mimetic systems, in addition to neuroblastoma plasma membranes. Their study provided insight into the different modes of hydration of lipid bilayers, as well as the differences in the passive ion transport of GUVs and free-standing lipid membranes (https://doi.org/10.1039/d4fd00197d).

Next, Alexander Fellows (Fritz-Haber-Institut der Max-Planck-Gesellschaft, Germany) presented his work on the impact of the lipid headgroup on the formation of heterogeneous phases in lipid membranes. In particular, they used phase-resolved azimuthal-scanned sum-frequency generation microscopy to spot differences in phase formations upon exposure of PS-lipids (https://doi.org/10.1039/d4fd00187g).

The last talk was given by John Conboy (University of Utah, USA). He dealt with the affinity of Ca²⁺ ions to PS-lipids and the role of the cation in the maintenance of PS-asymmetry. Sum-frequency vibrational spectroscopy revealed a high specificity of Ca²⁺ to PS-headgroups, vastly outperforming Mg²⁺ ions in preventing PS flip-flop in an SLB (https://doi.org/10.1039/d4fd00206g).

The discussion after the three presentations focused on the non-linear methods used in the presented work. The limitations and strengths of the presented methods were discussed in terms of the complexity or simplicity of examined systems. Whatever is added to the measured system can change results in unexpected ways, so a key challenge prevails in moving from the relatively simple presented systems to more complex systems in a quantitative way.

Gonen Golani (University of Haifa, Israel) started off the last section of Session 3 by presenting simulations using a continuum elastic model and supporting experiments to investigate how lipid composition asymmetry can affect the energy barriers of membrane fusion, specifically hemifusion-stalk formation and fusion-pore expansion. Their work showed that introducing different compositions in the monolayers can modulate both energy barriers, resulting in a preferred direction of fusion (https://doi.org/10.1039/d4fd00189c).

Giacomo Fiorin (National Institutes of Health, USA) delivered a talk on the change in bending energetics of asymmetric POPC/DOPC lipid bilayers with differential leaflet density imbalance, using all-atom molecular dynamics simulations. Small changes in leaflet density asymmetry can lead to increased membrane stiffness, highlighting a mechanism for the bending and remodelling of compositionally asymmetric membranes (https://doi.org/10.1039/d5fd00006h).

Axel Voigt (TU Dresden, Germany) closed the session by presenting a mesoscopic membrane model that extends the classical Canham–Helfrich framework by incorporating higher-order geometric bending terms and treating the membrane as a fluid deformable surface. The resulting equations were solved using surface finite element methods. Differential impacts on membrane shape evolution and stability were found from Gaussian curvature squared and quartic mean curvature deviations, and when combined with surface viscosity, were seen to significantly accelerate shape changes (https://doi.org/10.1039/d4fd00202d).

The ensuing discussion focused on the role of proteins in membrane fusion and curvature generation, including how protein-mediated effects influence energy barriers and the relevance of these barriers in biological contexts. Arguments were proposed on the origins of differential stress between leaflets, changes in bending rigidity, the rationale for including higher-order geometric terms in membrane models, and how such terms affect tubular shape stability, along with technical considerations around model equations and membrane friction parameters.

Session 4: Proteins in asymmetric membranes

The session on membrane–protein interactions began with Edward Lyman's (University of Delaware, USA) remarks on the reaction of G protein-coupled receptors (GPCRs) to the loss of asymmetry. In their simulations, the binding affinities of GPCRs before and after loss of lipid asymmetry were explored, where neither PS nor cholesterol abandoned their interaction sites. For his final remarks, he highlighted the need for the elucidation of protein structures in native lipid environments. He also cited the need for biologically relevant models of asymmetric systems other than red blood cell membranes and stressed the importance of a rigorous analysis of membrane–protein thermodynamics (https://doi.org/10.1039/d4fd00210e).

In the following talk, Fabio Lolicato (Heidelberg University Biochemistry Center, Germany; University of Helsinki, Finland) presented his joint work with Walter Nickel on the acceleration of FGF2 secretion via asymmetry of PIP2. In their research they moved from in vitro systems to live cells and were able to show the effects of PIP2 asymmetry loss (https://doi.org/10.1039/d4fd00208c).

The emanating discussion touched on specific protein interactions with cholesterol, but also on the effect of differential stress, while giving glimpses into the intricacies of molecular dynamic simulations. After that, the discussion switched to the dynamics of FGF2 across the membrane and how it might transit PIP2 to the extracellular side. Additional suggestions were discussed that may influence the dynamics of FGF2.

The second part of Session 4 and the last block of this Faraday Discussion started off with a presentation by Jonathan Machin (University of Leeds, UK) on joint work with Sheena Radford. In their work, they used mutations of an outer membrane protein (OmpA) to induce charge asymmetry in liposomes. The degree of this induced dipole was shown to influence the folding and stability of additional proteins dependent on charge distribution (https://doi.org/10.1039/d4fd00180j).

The last presentation of this session was given by Josef Lorent (UCLouvain, Belgium), in which he explored the adaptability of transmembrane protein domains to their asymmetric lipid environment. In a bioinformatic approach, they compared different protein properties over a variety of clades and correlated various amino acid properties to their respective local membrane properties (https://doi.org/10.1039/d4fd00199k).

In the following discussion, the focus lay on the folding kinetics of outer membrane proteins and the influence of charge distribution and lipid density. Transitioning to the work of Josef Lorent, a question about possible bias arose in connection to the positive-inside rule. The use of machine learning was discussed among other suggestions to provide further insight on the correlations between membrane asymmetry and transmembrane proteins.

Poster session and social events

A poster session was held on the evening of the first day, with prizes awarded to Manpreet Kaur (Heidelberg University, Germany) and Iwona Swiderska (EPFL, Switzerland) (Fig. 4 and 5).

Fig. 4 Poster Winner Manpreet Kaur with some of the organisers.

Fig. 5 The Physical Chemistry Chemical Physics Poster Prize winner Iwona Swiderska with some of the organisers.

We enjoyed a wonderful dinner in Burlington House, being treated to not one but two musical performances. The first performance included a selection of Franz Schubert songs performed by Félix Goñi (baritone) and Bendit Chan (piano), before Georg Pabst, John Seddon and Félix Goñi rounded up the musical entertainment with a series of original material (Fig. 6 and 7).

Fig. 6 Georg Pabst, John Seddon and Félix Goñi providing musical entertainment during the dinner service.

Fig. 7 Georg Pabst, John Seddon and Félix Goñi in scientific rather than musical discussion.

Closing remarks lecture (Goñi's Swansong)

For the concluding remarks, Félix Goñi entered the stage, proving to be full of surprises (Fig. 8). Well aware of how exhausted the participants had become through the past two and a half days, he resorted to somewhat extreme measures to keep the audience engaged, adding stuntman to his long list of talents. After a historical outline of the prevailing century of research into lipid asymmetry, he added his own thoughts to the ongoing discussion about phospholipid distribution in the plasma membrane. Announcing his retirement from active research, the overall talk felt like a final farewell to the membrane research community, but not without stirring up the pot a last time before closing shop for good. He provided a nod to T. S. Eliot in framing the joy of scientific exploration.

Fig. 8 Introductory lecture and closing remarks speakers: Gerald Feigenson and Félix Goñi.

We shall not cease from exploration And the end of all our exploring Will be to arrive where we started And know the place for the first time. – T. S. Eliot²

Acknowledgements

With thanks to the organising committee John Seddon, Georg Pabst, Rumiana Dimova, Syma Khalid, Karin A. Riske and Raya Sorkin.

Notes and references

1. E. Gorter and F. Grendel, J. Exp. Med., 1925, 41(4), 439–443.
2. T. S. Eliot, Little Gidding, Four Quartets, Harcourt Brace & Co., New York, 1971, p. 59.

---