Studies of membranotropic and fusogenic activity of two putative HCV fusion peptides

Over the past decades, membranotropic peptides such as positively charged cell-penetrating peptides (CPPs) or amphipathic antimicrobial peptides (AMPs) have received increasing interest in order to improve therapeutic agent cellular uptake.As far as we are concerned, we were interested in studying HCV fusion peptides as putative anchors. Two peptides, HCV6 and HCV7, were identified and conjugated to a fluorescent tag NBD and tested for their interaction with liposomes as model membranes. DSC and spectrofluorescence analyses demonstrate HCV7 propensity to insert or internalize in vesicles containing anionic lipids DMPG whereas no activity was observed with zwitterionic DMPC. This behavior could be explained by the peptide sequence containing a cationic arginine residue. On the contrary, HCV6 did not exhibit any membranotropic activity but was the only sequence able to induce liposomes' fusion or aggregation monitored by spectrofluorescence and DLS. This two peptides mild activity was related to their inefficient structuration in contact with membrane mimetics, which was demonstrated by CD and NMR experiments.Altogether, our data allowed us to identify two promising membrane-active peptides from E1 and E2 HCV viral proteins, one fusogenic (HCV6) and the other membranotropic (HCV7). The latter was also confirmed by fluorescence microscopy with CHO cells, indicating that HCV7 could cross the plasma membrane via an endocytosis process. Therefore, this study provides new evidences supporting the identification of HCV6 as the HCV fusion peptide as well as insights on a novel membranotropic peptide from the HCV-E2 viral protein.     

Exploring peptide membrane interaction using surface plasmon resonance: differentiation between pore formation versus membrane disruption by lytic peptides

Lytic peptides comprise a large group of membrane-active peptides used in the defensive and offensive systems of all organisms. Differentiating between their modes of interaction with membranes is crucial for understanding how these peptides select their target cells. Here we utilized SPR to study the interaction between lytic peptides and lipid bilayers (L1 sensor chip). Using studies also on hybrid monolayers (HPA sensor chip) revealed that SPR is a powerful tool for obtaining a real-time monitoring of the steps involved in the mode of action of membrane-active peptides, some of which previously could not be detected directly by other techniques and reported here for the first time. We investigated the mode of action of peptides that represent two major families: (i) the bee venom, melittin, as a model of a non-cell-selective peptide that forms transmembrane pores and (ii) magainin and a diastereomer of melittin (four amino acids were replaced by their D enantiomers), as models of bacteria-selective non-pore-forming peptides. Fitting the SPR data to different interaction models allows differentiating between two major steps: membrane binding and membrane insertion. Melittin binds to PC/cholesterol approximately 450-fold better than its diastereomer and magainin, mainly because it is inserted into the inner leaflet (2/3 of the binding energy), whereas the other two are not. In contrast, there is only a slight difference in the binding of all the peptides to negatively charged PE/PG mono- and bilayer membranes (in the first and second steps), indicating that the inner leaflet contributes only slightly to their binding to PE/PG bilayers. Furthermore, the 100-fold stronger binding of the cell-selective peptides to PE/PG as compared with PC/cholesterol resulted only from electrostatic attraction to the negatively charged headgroups of the outer leaflet. These results clearly differentiate between the two general mechanisms: pore formation by melittin only in zwitterionic membranes and a detergent-like effect (carpet mechanism) for all the peptides in negatively charged membranes, in agreement with their biological function.     

Machine learning-enabled discovery and design of membrane-active peptides

Antimicrobial peptides are a class of membrane-active peptides that form a critical component of innate host immunity and possess a diversity of sequence and structure. Machine learning approaches have been profitably employed to efficiently screen sequence space and guide experiment towards promising candidates with high putative activity. In this mini-review, we provide an introduction to antimicrobial peptides and summarize recent advances in machine learning-enabled antimicrobial peptide discovery and design with a focus on a recent work Lee et al. Proc. Natl. Acad. Sci. USA 2016;113(48):13588–13593. This study reports the development of a support vector machine classifier to aid in the design of membrane active peptides. We use this model to discover membrane activity as a multiplexed function in diverse peptide families and provide interpretable understanding of the physicochemical properties and mechanisms governing membrane activity. Experimental validation of the classifier reveals it to have learned membrane activity as a unifying signature of antimicrobial peptides with diverse modes of action. Some of the discriminating rules by which it performs classification are in line with existing “human learned” understanding, but it also unveils new previously unknown determinants and multidimensional couplings governing membrane activity. Integrating machine learning with targeted experimentation can guide both antimicrobial peptide discovery and design and new understanding of the properties and mechanisms underpinning their modes of action.     

Biophysical and functional assays for viral membrane fusion peptides

Membrane fusion is a protein catalyzed biophysical reaction that involves the simultaneous intermixing of two phospholipid bilayers and of the aqueous compartments bound by their respective bilayers. In the case of enveloped virus fusogens, short hydrophobic or amphipathic fusion peptides that are components of the larger fusion complex are essential for the membrane merger event. The process of cell–cell membrane fusion and syncytium formation induced by the nonenveloped fusogenic orthoreoviruses is driven by the Fusion-Associated Small Transmembrane (FAST) proteins, which are similarly dependent on the action of fusion peptides. In this article, we describe some simple methods for the biophysical characterization of viral membrane fusion peptides. Liposomes serve as an ideal model system for characterizing peptide–membrane interactions because their size, shape and composition can be readily manipulated. We present details of fluorescence assays used to elucidate the kinetics of membrane fusion as well as complimentary assays used to characterize peptide-induced liposome binding and aggregation.     

Antimicrobial and cell-penetrating peptides induce lipid vesicle fusion by folding and aggregation

According to their distinct biological functions, membrane-active peptides are generally classified as antimicrobial (AMP), cell-penetrating (CPP), or fusion peptides (FP). The former two classes are known to have some structural and physicochemical similarities, but fusogenic peptides tend to have rather different features and sequences. Nevertheless, we found that many CPPs and some AMPs exhibit a pronounced fusogenic activity, as measured by a lipid mixing assay with vesicles composed of typical eukaryotic lipids. Compared to the HIV fusion peptide (FP23) as a representative standard, all designer-made peptides showed much higher lipid-mixing activities (MSI-103, MAP, transportan, penetratin, Pep1). Native sequences, on the other hand, were less fusogenic (magainin 2, PGLa, gramicidin S), and pre-aggregated ones were inactive (alamethicin, SAP). The peptide structures were characterized by circular dichroism before and after interacting with the lipid vesicles. A striking correlation between the extent of conformational change and the respective fusion activities was found for the series of peptides investigated here. At the same time, the CD data show that lipid mixing can be triggered by any type of conformation acquired upon binding, whether α-helical, β-stranded, or other. These observations suggest that lipid vesicle fusion can simply be driven by the energy released upon membrane binding, peptide folding, and possibly further aggregation. This comparative study of AMPs, CPPs, and FPs emphasizes the multifunctional aspects of membrane-active peptides, and it suggests that the origin of a peptide (native sequence or designer-made) may be more relevant to define its functional range than any given name.     

Membrane interactions of antimicrobial peptides from Australian tree frogs

The skin secretions of amphibians are rich in host defence peptides. The membrane interactions of the antimicrobial peptides, aurein 1.2, citropin 1.1 and maculatin 1.1, isolated from Australian tree frogs, are reviewed. Although all three peptides are amphipathic α-helices, the mode of action of these membrane-active peptides is not defined. The peptides have a net positive charge and range in length from 13 to 21 residues, with the longest, maculatin 1.1, having a proline at position 15. Interestingly, alanine substitution at Pro-15 leads to loss of activity. The effects of these peptides on phospholipid bilayers indicate different mechanisms for pore formation and lysis of model membranes, with the shorter peptides exhibiting a carpet-like mechanism and the longest peptide forming pores in phospholipid bilayer membranes.     

Peptide mimics of SNARE transmembrane segments drive membrane fusion depending on their conformational plasticity

SNARE proteins are essential for different types of intracellular membrane fusion. Whereas interaction between their cytoplasmic domains is held responsible for establishing membrane proximity, the role of the transmembrane segments in the fusion process is currently not clear. Here, we used an in vitro approach based on lipid mixing and electron microscopy to examine a potential fusogenic activity of the transmembrane segments. We show that the presence of synthetic peptides representing the transmembrane segments of the presynaptic soluble N-ethylmaleimide-sensitive factor attachment protein receptors (SNAREs) synaptobrevin II (also referred to as VAMP II) or syntaxin 1A, but not of an unrelated control peptide, in liposomal membranes drives their fusion. Liposome aggregation by millimolar Ca(2+) concentrations strongly potentiated the effect of the peptides; this indicates that juxtaposition of the bilayers favours their fusion in the absence of the cytoplasmic SNARE domains. Peptide-driven fusion is reminiscent of natural membrane fusion, since it was suppressed by lysolipid and involved both bilayer leaflets. This suggests transient presence of a hemifusion intermediate followed by complete membrane merger. Structural studies of the peptides in lipid bilayers performed by Fourier transform infrared spectroscopy indicated mixtures of alpha-helical and beta-sheet conformations. In isotropic solution, circular dichroism spectroscopy showed the peptides to exist in a concentration-dependent equilibrium of alpha-helical and beta-sheet structures. Interestingly, the fusogenic activity decreased with increasing stability of the alpha-helical solution structure for a panel of variant peptides. Thus, structural plasticity of transmembrane segments may be important for SNARE protein function at a late step in membrane fusion.     

Conformation and interaction of the cyclic cationic antimicrobial peptides in lipid bilayers.

To investigate the role of peptide-membrane interactions in the biological activity of cyclic cationic peptides, the conformations and interactions of four membrane-active antimicrobial peptides [based on Gramicidin S (GS)] were examined in neutral and negatively charged micelles and phospholipid vesicles, using CD and fluorescence spectroscopy and ultracentrifugation techniques. Moreover, the effects of these peptides on the release of entrapped fluorescent dye from unilamellar vesicles of phosphatidylcholine (PC) and phosphatidylethanolamine/phosphatidylglycerol (PE/PG) were studied. The cyclic peptides include GS10 [Cyclo(VKLdYP)2], GS12 [Cyclo(VKLKdYPKVKLdYP)], GS14 [Cyclo(VKLKVdYPLKVKLdYP)] and [d-Lys]4GS14 [Cyclo(VKLdKVdYPLKVKLdYP)] (underlined residues are d-amino acids), were different in their ring size, structure and amphipathicity, and covered a broad spectrum of hemolytic and antimicrobial activities. Interaction of the peptides with the zwitterionic PC and negatively charged PE/PG vesicles were distinct from each other. The hydrophobic interaction seems to be the dominant factor in the hemolytic activity of the peptides, as well as their interaction with the PC vesicles. A combination of electrostatic and hydrophobic interactions of the peptides induces aggregation and fusion in PE/PG vesicles with different propensities in the order: [d-Lys]4GS14 > GS14 > GS12 > GS10. GS10 and GS14 are apparently located in the deeper levels of the membrane interfaces and closer to the hydrophobic core of the bilayers, whereas GS12 and [d-Lys]4GS14 reside closer to the outer boundary of the interface. Because of differing modes of interaction of the cyclic cationic peptides with lipid bilayers, the mechanism of their biological activity (and its relation to peptide-lipid interaction) proved to be versatile and complex, and dependent on the biophysical properties of both the peptides and membranes.     

Simulation studies of the interaction of antimicrobial peptides and lipid bilayers

Experimental studies of a number of antimicrobial peptides are sufficiently detailed to allow computer simulations to make a significant contribution to understanding their mechanisms of action at an atomic level. In this review we focus on simulation studies of alamethicin, melittin, dermaseptin and related antimicrobial, membrane-active peptides. All of these peptides form amphipathic alpha-helices. Simulations allow us to explore the interactions of such peptides with lipid bilayers, and to understand the effects of such interactions on the conformational dynamics of the peptides. Mean field methods employ an empirical energy function, such as a simple hydrophobicity potential, to provide an approximation to the membrane. Mean field approaches allow us to predict the optimal orientation of a peptide helix relative to a bilayer. Molecular dynamics simulations that include an atomistic model of the bilayer and surrounding solvent provide a more detailed insight into peptide-bilayer interactions. In the case of alamethicin, all-atom simulations have allowed us to explore several steps along the route from binding to the membrane surface to formation of transbilayer ion channels. For those antimicrobial peptides such as dermaseptin which prefer to remain at the surface of a bilayer, molecular dynamics simulations allow us to explore the favourable interactions between the peptide helix sidechains and the phospholipid headgroups.     

Molecular dynamics simulation of the membrane binding and disruption mechanisms by antimicrobial scorpion venom-derived peptides

Pandinin 2 (Pin2) is an alpha-helical polycationic peptide, identified and characterized from venom of the African scorpion Pandinus imperator with high antimicrobial activity against Gram-positive bacteria and less active against Gram-negative bacteria, however it has demonstrated strong hemolytic activity against sheep red blood cells. In the chemically synthesized Pin2GVG analog, the GVG motif grants it low hemolytic activity while keeping its antimicrobial activity. In this work, we performed 12 μs all-atom molecular dynamics simulation of the antimicrobial peptides (AMPs) Pin2 and Pin2GVG to explore their adsorption mechanism and the role of their constituent amino acid residues when interacting with pure POPC and pure POPG membrane bilayers. Starting from an α-helical conformation, both AMPs are attracted at different rates to the POPC and POPG bilayer surfaces due to the electrostatic interaction between the positively charged amino acid residues and the charged moieties of the membranes. Since POPG is an anionic membrane, the PAMs adhesion is stronger to the POPG membrane than to the POPC membrane and they are stabilized more rapidly. This study reveals that, before the insertion begins, Pin2 and Pin2GVG remained partially folded in the POPC surface during the first 300 and 600 ns, respectively, while they are mostly unfolded in the POPG surface during most of the simulation time. The unfolded structures provide for a large number of intermolecular hydrogen bonds and stronger electrostatic interactions with the POPG surface. The results show that the aromatic residues at the N-terminus of Pin2 initiate the insertion process in both POPC and POPG bilayers. As for Pin2GVG in POPC the C-terminus residues seem to initiate the insertion process while in POPG this process seems to be slowed down due to a strong electrostatic attraction. The membrane conformational effects upon PAMs binding are measured in terms of the area per lipid and the contact surface area. Several replicas of the systems lead to the same observations.     

Real-time quantitative analysis of lipid disordering by aurein 1.2 during membrane adsorption, destabilisation and lysis

Effective antimicrobial peptides (AMPs) distinguish between the host and microbial cells, show selective antimicrobial activity and exhibit a fast killing mechanism. Although understanding the structure-function characteristics of AMPs is important, the impact of the peptides on the architecture of membranes with different lipid compositions is also critical in understanding the molecular mechanism and specificity of membrane destabilisation. In this study, the destabilisation of supported lipid bilayers (SLBs) by the AMP aurein 1.2 was quantitatively analysed by dual polarisation interferometry. The lipid bilayers were formed on a planar silicon oxynitride chip, and composed of mixed synthetic lipids, or Escherichiacoli lipid extract. The molecular events leading sequentially from peptide adsorption to membrane lysis were examined in real time by changes in bilayer birefringence (lipid molecular ordering) as a function of membrane-bound peptide mass. Aurein 1.2 bound weakly without any change in membrane ordering at low peptide concentration (5 μM), indicating a surface-associated state without significant perturbation in membrane structure. At 10 μM peptide, marked reversible changes in molecular ordering were observed for all membranes except DMPE/DMPG. However, at 20 μM aurein 1.2, removal of lipid molecules, as determined by mass loss with a concomitant decrease in birefringence during the association phase, was observed for DMPC and DMPC/DMPG SLBs, which indicates membrane lysis by aurein. The membrane destabilisation induced by aurein 1.2 showed cooperativity at a particular peptide/lipid ratio with a critical mass/molecular ordering value. Furthermore, the extent of membrane lysis for DMPC/DMPG was nearly double that for DMPC. However, no lysis was observed for DMPC/DMPG/cholesterol, DMPE/DMPG and E. coli SLBs. The extent of birefringence changes with peptide mass suggested that aurein 1.2 binds to the membrane without inserting through the bilayer and membrane lysis occurs through detergent-like micellisation above a critical P/L ratio. Real-time quantitative analysis of the structural properties of membrane organisation has allowed the membrane destabilisation process to be resolved into multiple steps and provides comprehensive information to determine the molecular mechanism of aurein 1.2 action.     

Magainin 2 and PGLa in bacterial membrane mimics IV: Membrane curvature and partitioning

We previously reported that the synergistically enhanced antimicrobial activity of magainin 2 (MG2a) and PGLa is related to membrane adhesion and fusion. Here, we demonstrate that equimolar mixtures of MG2a and L18W-PGLa induce positive monolayer curvature stress and sense, at the same time, positive mean and Gaussian bilayer curvatures already at low amounts of bound peptide. The combination of both abilities—membrane curvature sensing and inducing—is most likely the base for the synergistically enhanced peptide activity. In addition, our coarse-grained simulations suggest that fusion stalks are promoted by decreasing the free-energy barrier for their formation rather than by stabilizing their shape. We also interrogated peptide partitioning as a function of lipid and peptide concentration using tryptophan fluorescence spectroscopy and peptide-induced leakage of dyes from lipid vesicles. In agreement with a previous report, we find increased membrane partitioning of L18W-PGLa in the presence of MG2a. However, this effect does not prevail to lipid concentrations higher than 1 mM, above which all peptides associate with the lipid bilayers. This implies that synergistic effects of MG2a and L18W-PGLa in previously reported experiments with lipid concentrations >1 mM are due to peptide-induced membrane remodeling and not their specific membrane partitioning.     

Dual Action of BPC194: A Membrane Active Peptide Killing Bacterial Cells

Membrane active peptides can perturb the lipid bilayer in several ways, such as poration and fusion of the target cell membrane, and thereby efficiently kill bacterial cells. We probe here the mechanistic basis of membrane poration and fusion caused by membrane-active, antimicrobial peptides. We show that the cyclic antimicrobial peptide, BPC194, inhibits growth of Gram-negative bacteria and ruptures the outer and inner membrane at the onset of killing, suggesting that not just poration is taking place at the cell envelope. To simplify the system and to better understand the mechanism of action, we performed Förster resonance energy transfer and cryogenic transmission electron microscopy studies in model membranes and show that the BPC194 causes fusion of vesicles. The fusogenic action is accompanied by leakage as probed by dual-color fluorescence burst analysis at a single liposome level. Atomistic molecular dynamics simulations reveal how the peptides are able to simultaneously perturb the membrane towards porated and fused states. We show that the cyclic antimicrobial peptides trigger both fusion and pore formation and that such large membrane perturbations have a similar mechanistic basis.     

Structural and functional properties of an unusual internal fusion peptide in a nonenveloped virus membrane fusion protein

The avian and Nelson Bay reoviruses are two of only a limited number of nonenveloped viruses capable of inducing cell-cell membrane fusion. These viruses encode the smallest known membrane fusion proteins (p10). We now show that a region of moderate hydrophobicity we call the hydrophobic patch (HP), present in the small N-terminal ectodomain of p10, shares the following characteristics with the fusion peptides of enveloped virus fusion proteins: (i) an abundance of glycine and alanine residues, (ii) a potential amphipathic secondary structure, (iii) membrane-seeking characteristics that correspond to the degree of hydrophobicity, and (iv) the ability to induce lipid mixing in a liposome fusion assay. The p10 HP is therefore predicted to provide a function in the mechanism of membrane fusion similar to those of the fusion peptides of enveloped virus fusion peptides, namely, association with and destabilization of opposing lipid bilayers. Mutational and biophysical analysis suggested that the internal fusion peptide of p10 lacks alpha-helical content and exists as a disulfide-stabilized loop structure. Similar kinked structures have been reported in the fusion peptides of several enveloped virus fusion proteins. The preservation of a predicted loop structure in the fusion peptide of this unusual nonenveloped virus membrane fusion protein supports an imperative role for a kinked fusion peptide motif in biological membrane fusion.     

The interaction of cationic antimicrobial peptides with vesicles containing synthetic glycolipids as models of the outer membrane of gram-negative bacteria.

Two simple lipid A analogues methyl 2,3-di-O-tetradecanoyl-alpha-D-glucopyranoside (GL1) and methyl 2,3-di-O-tetradecanoyl-alpha-D-glucopyranoside 4-O-phosphate (GL2) were synthesized and used for preparing mixed phosphocholine vesicles as models of the outer membrane of gram-negative bacteria. The interaction of these model membranes with magainin 2, a representative of the alpha-helical membrane active peptides, and apidaecin Ib and drosocin, two insect Pro-rich peptides which do not act at the level of the cellular membrane, were studied by CD and dye-releasing experiments. The CD spectra of apidaecin Ib and drosocin in the presence of GL1- or GL2-containing vesicles were consistent with largely unordered structures, whereas, according to the CD spectra, magainin 2 adopted an amphipathic alpha-helical conformation, particularly in the presence of negatively charged bilayers. The ability of the peptides to fold into amphipathic conformations was strictly correlated to their ability to bind and to permeabilize phospholipid as well as glycolipid membranes. Apidaecin Ib and drosocin, which are unable to adopt an amphipathic structure, showed negligible dye-leakage activity even in the presence of GL2-containing vesicles. It is reasonable to suppose that, as for the killing mechanism, the two classes of antimicrobial peptides follow different patterns to cross the bacterial outer membrane.     

Antimicrobial peptides with cell-penetrating peptide properties and vice versa

Antimicrobial peptides (AMPs) are a group of peptides that are active against a diverse spectrum of microorganisms. Due to their mode of action, AMPs are a promising class of molecules that could overcome the problems of increasing resistance of bacteria to conventional antibiotics. Furthermore, AMPs are strongly membrane-active and some are able to translocate into cells without the necessity for permanent membrane permeabilization. This feature has brought them into focus for use as transport vectors in the context of drug delivery. Since the plasma membrane restricts transport of bioactive substances into cells, great research interest lies in the development of innovative ways to overcome this barrier and to increase bioavailability. In this context, peptide-based transport systems, such as cell-penetrating peptides (CPPs), have come into focus, and their efficiency has been demonstrated in many different applications. However, more recently, also some AMPs have been used as efficient vectors for intracellular translocation of various active molecules. This review summarizes recent efforts in this interesting field of drug delivery. Moreover, some examples of the application of CPPs as efficient antimicrobial substances will be discussed.     

Absorbance-based assay for membrane disruption by antimicrobial peptides and synthetic copolymers using pyrroloquinoline quinone-loaded liposomes

A simple homogeneous assay for the detection of membrane permeabilization by antimicrobial peptides and synthetic copolymers is described. Liposomes encapsulating pyrroloquinoline quinone (PQQ), the prosthetic group of the apoenzyme glucose dehydrogenase (GDH), are used to detect membrane permeabilization by the antimicrobial peptides MSI-594 and MSI-78 as well as various synthetic antimicrobial copolymers in an optical microwell assay. PQQ-loaded liposomes and the peptide or copolymer are added to wells of a 96-well microtiter plate. If the integrity of the liposome is compromised, the PQQ encapsulated in the liposomes is released and available for activating the apoenzyme. The release of PQQ catalyzes a color change in the presence of apo-GDH, glucose, and the redox dye 1,6-dichlorophenol indophenol (DCPIP) that can be evaluated through a visual color change. For more quantitative measurements, the absorbance change over a 30 min period was measured. The absorbance change is related to the activity and concentration for a given antimicrobial agent. Furthermore, by varying liposome compositions to include cholesterol, the potential toxicity of the peptide or polymer toward mammalian cells can be readily evaluated. The assay is simple and sensitive and will be useful for analyzing the membrane permeation/disruption properties of a host of antimicrobial peptides and synthetic polymers.     

Alpha-helical cationic antimicrobial peptides: relationships of structure and function

Antimicrobial peptides (AMPs), with their extraordinary properties, such as broad-spectrum activity, rapid action and difficult development of resistance, have become promising molecules as new antibiotics. Despite their various mechanisms of action, the interaction of AMPs with the bacterial cell membrane is the key step for their mode of action. Moreover, it is generally accepted that the membrane is the primary target of most AMPs, and the interaction between AMPs and eukaryotic cell membranes (causing toxicity to host cells) limits their clinical application. Therefore, researchers are engaged in reforming or de novo designing AMPs as a ‘single-edged sword’ that contains high antimicrobial activity yet low cytotoxicity against eukaryotic cells. To improve the antimicrobial activity of AMPs, the relationship between the structure and function of AMPs has been rigorously pursued. In this review, we focus on the current knowledge of α-helical cationic antimicrobial peptides, one of the most common types of AMPs in nature.     

Adhesion and merging of lipid bilayers: a method for measuring the free energy of adhesion and hemifusion

Lipid bilayers can be induced to adhere to each other by molecular mediators, and, depending on the lipid composition, such adhesion can lead to merging of the contacting monolayers in a process known as hemifusion. Such bilayer-bilayer reactions have never been systematically studied. In the course of our studies of membrane-active molecules, we encountered such reactions. We believe that they need to be understood whenever bilayer-bilayer interactions take place, such as during membrane fusion. For illustration, we discuss three examples: spontaneous adhesion between phospholipid bilayers induced by low pH, polymer-induced osmotic depletion attraction between lipid bilayers, and anionic lipid bilayers cross-bridged by multicationic peptides. Our purpose here is to describe a general method for studying such interactions. We used giant unilamellar vesicles, each of which was aspirated in a micropipette so that we could monitor the tension of the membrane and the membrane area changes during the bilayer-bilayer interaction. We devised a general method for measuring the free energy of adhesion or hemifusion. The results show that the energies of adhesion or hemifusion of lipid bilayers could vary over 2 orders of magnitude from −1 to −50 × 10⁻⁵ J/m² in these examples alone. Our method can be used to measure the energy of transition in each step of lipid transformation during membrane fusion. This is relevant for current research on membrane fusion, which focuses on how fusion proteins induce lipid transformations.     

Modulation of membrane curvature by peptides

The fusion of two stable bilayers likely proceeds through intermediates in which the membrane acquires curvature. The insertion of peptides into the membrane will affect its curvature tendency. Studies with a number of small viral fusion peptides indicate that these peptides promote negative curvature at low concentration. This is in accord with the curvature requirements to initiate membrane fusion according to the stalk-pore model. Although a characteristic of fusion peptides, the promotion of negative curvature is only one of several mechanisms by which fusion proteins accelerate the rate of fusion. In addition, the fusion peptide itself, as well as other regions in the viral fusion protein, facilitates membrane fusion by mechanisms that are largely independent of curvature. Leakage of the internal aqueous contents of liposomes is another manifestation of the alteration of membrane properties. Peptides exhibit quite different relative potencies between fusion and leakage that is determined by the structure and mode of insertion of the peptide into the membrane.