Structural basis for membrane fusion by enveloped viruses.

Enveloped viruses such as HIV-1, influenza virus, and Ebola virus express a surface glycoprotein that mediates both cell attachment and fusion of viral and cellular membranes. The membrane fusion process leads to the release of viral proteins and the RNA genome into the host cell, initiating an infectious cycle. This review focuses on the HIV-1 gp41 membrane fusion protein and discusses the structural similarities of viral membrane fusion proteins from diverse families such as Retroviridae (HIV-1), Orthomyxoviridae (influenza virus), and Filoviridae (Ebola virus). Their structural organization suggests that they have all evolved to use a similar strategy to promote fusion of viral and cellular membranes. This observation led to the proposal of a general model for viral membrane fusion, which will be discussed in detail.     

Cholesterol alters the inhibitory efficiency of peptide-based membrane fusion inhibitor

The membrane composition modulates membrane fusion by altering membrane physical properties and the structure, organization and dynamics of fusion proteins and peptides. The journey of developing peptide-based viral fusion inhibitors is often stalled by the change in lipid composition of viral and target membranes. This makes it important to study the role of membrane composition on the organization, dynamics and fusion inhibiting abilities of the peptide-based fusion inhibitors. Cholesterol, an important constituent of mammalian cell membrane, modulates bilayer properties in multiple ways and impart its effect on the membrane fusion. We have previously shown that TG-23 peptide derived from phagosomal coat protein, coronin 1, shows significant inhibition of fusion between membranes without cholesterol. In this work, we have studied the effect of the TG-23 peptide on the polyethylene glycol-mediated membrane fusion in presence of different concentrations of membrane cholesterol. Our results show that the inhibitory effect of TG-23 is being completely reversed in cholesterol containing membranes. We have evaluated the structure, organization, dynamics and depth of penetration of TG-23 in membranes having different lipid compositions and its effect on membrane properties. Our results demonstrate that cholesterol does not affect the secondary structure of the peptide, however, alters the depth of penetration of the peptide and modifies peptide organization and dynamics. The cholesterol dependent change in organization and dynamics of the peptide influences its efficacy in membrane fusion. Therefore, we envisage that the study of peptide organization and dynamics is extremely important to determine the effect of peptide on the membrane fusion.     

Fatty acids can substitute the HIV fusion peptide in lipid merging and fusion: an analogy between viral and palmitoylated eukaryotic fusion proteins

Various fusion proteins from eukaryotes and viruses share structural similarities such as a coiled coil motif. However, compared with eukaryotic proteins, a viral fusion protein contains a fusion peptide (FP), which is an N-terminal hydrophobic fragment that is primarily involved in directing fusion via anchoring the protein to the target cell membrane. In various eukaryotic fusion proteins the membrane targeting domain is cysteine-rich and must undergo palmitoylation prior to the fusion process. Here we examined whether fatty acids can replace the FP of human immunodeficiency virus type 1 (HIV-1), thereby discerning between the contributions of the sequence versus hydrophobicity of the FP in the lipid-merging process. For that purpose, we structurally and functionally characterized peptides derived from the N terminus of HIV fusion protein - gp41 in which the FP is lacking or replaced by fatty acids. We found that fatty acid conjugation dramatically enhanced the capability of the peptides to induce lipid mixing and aggregation of zwitterionic phospholipids composing the outer leaflet of eukaryotic cell membranes. The enhanced effect of the acylated peptides on membranes was further supported by real-time atomic force microscopy (AFM) showing nanoscale holes in zwitterionic membranes. Membrane-binding experiments revealed that fatty acid conjugation did not increase the affinity of the peptides to the membrane significantly. Furthermore, all free and acylated peptides exhibited similar α-helical structures in solution and in zwitterionic membranes. Interestingly, the fusogenic active conformation of N36 in negatively charged membranes composing the inner leaflet of eukaryotic cells is β-sheet. Apparently, N-terminal heptad repeat (NHR) can change its conformation as a response to a change in the charge of the membrane head group. Overall, the data suggest an analogy between the eukaryotic cysteine-rich domains and the viral fusion peptide, and mark the hydrophobic nature of FP as an important characteristic for its role in lipid merging.     

Membrane cell fusion activity of the vaccinia virus A17-A27 protein complex

Vaccinia virus enters cells by endocytosis and via a membrane fusion mechanism mediated by viral envelope protein complexes. While several proteins have been implicated in the entry/fusion event, there is no direct proof for fusogenic activity of any viral protein in heterologous systems. Transient coexpression of A17 and A27 in mammalian cells led to syncytia formation in a pH-dependent manner, as ascertained by confocal fluorescent immunomicroscopy. The pH-dependent fusion activity was identified to reside in A17 amino-terminal ectodomain after overexpression in insect cells using recombinant baculoviruses. Through the use of A17 ectodomain deletion mutants, it was found that the domain important for fusion spanned between residues 18 and 34. To further characterize A17–A27 fusion activity in mammalian cells, 293T cell lines stably expressing A17, A27 or coexpressing both proteins were generated using lentivectors. A27 was exposed on the cell surface only when A17 was coexpressed. In addition, pH-dependent fusion activity was functionally demonstrated in mammalian cells by cytoplasmic transfer of fluorescent proteins, only when A17 and A27 were coexpressed. Bioinformatic tools were used to compare the putative A17–A27 protein complex with well-characterized fusion proteins. Finally, all experimental evidence was integrated into a working model for A17–A27-induced pH-dependent cell-to-cell fusion.     

Insights into a Structure-Based Mechanism of Viral Membrane Fusion

A number of different viral spike proteins, responsible for membrane fusion, show striking similarities in their core structures. The prospect of developing a general structure-based mechanism seems plausible in light of these newly determined structures. Influenza hemagglutinin (HA) is the best-studied fusion machine, whose action has previously been described by a hypothetical spring-loaded model. This model has recently been extended to explain the mechanism of other systems, such as HIV gp120–gp41. However, evidence supporting this idea is insufficient, requiring re-examination of the mechanism of HA-induced membrane fusion. Recent experiments with a shortened construct of HA, which is able to induce lipid mixing, have provided evidence for an alternative scenario for HA-induced membrane fusion and perhaps that of other viral systems.     

Effects of monovalent cations on Semliki Forest virus entry into BHK-21 cells

Infection of mammalian cells with Semliki Forest virus requires the endocytosis of the virus, its delivery to prelysosomal endosomes, and fusion of the viral envelope with the endosome membrane. Previous studies have indicated that the low endosomal pH triggers a conformational change in the viral spike glycoproteins rendering them fusogenic. In this paper, we demonstrate an additional factor(s) which regulates virus fusion in endosomes. We found that Semliki Forest virus is unable to penetrate or infect baby hamster kidney (BHK-21) cells grown in medium containing reduced Na+ concentrations. Virus endocytosis and degradation are nearly normal, the virus is transported to endosomes where a characteristic low pH-induced loss of trypsin-sensitivity of the E1 spike glycoprotein occurs. Nevertheless, the viral envelope fails to fuse with the endosomal membrane and the viral RNA is not released into the cytosol. As judged by the uptake of the voltage-sensitive probe [3H]triphenylmethyl phosphonium we observed a close correlation between conditions which inhibit virus infection and which cause depolarization of the cells. We propose that in intact cells, the fusion of Semliki Forest virus with the endosome membrane depends not only on acidic endosomal pH, but also on the maintenance of the potential.     

Membrane-containing viruses with icosahedrally symmetric capsids

Viruses with an icosahedrally symmetric protein capsid and a membrane infect hosts from all three domains of life. Similar architectural principles are shared by different viral families, as exemplified by double-stranded DNA viruses such as PRD1 and STIV. During virus assembly, the membrane lipids are selectively acquired from the host cell. The X-ray structure of bacteriophage PRD1 revealed that the lipids are asymmetrically distributed between the two leaflets and facet length is controlled by a tape-measure protein. In most membrane-containing viruses, viral and host membranes fuse during viral entry. In the best-understood systems of the alphaviruses, flaviviruses and herpes viruses, fusion is mediated by viral glycoproteins. Recent structural advances reveal how very different protein architectures can be used to form trimeric extensions that extend into the target cell membrane and then fold back to mediate fusion of the target and viral membranes.     

Structure and function of a paramyxovirus fusion protein

Paramyxoviruses initiate infection by attaching to cell surface receptors and fusing viral and cell membranes. Viral attachment proteins, hemagglutinin-neuraminidase (HN), hemagglutinin (HA), or glycoprotein (G), bind receptors while fusion (F) proteins direct membrane fusion. Because paramyxovirus fusion is pH independent, virus entry occurs at host cell plasma membranes. Paramyxovirus fusion also usually requires co-expression of both the attachment protein and the fusion (F) protein. Newcastle disease virus (NDV) has assumed increased importance as a prototype paramyxovirus because crystal structures of both the NDV F protein and the attachment protein (HN) have been determined. Furthermore, analysis of structure and function of both viral glycoproteins by mutation, reactivity of antibody, and peptides have defined domains of the NDV F protein important for virus fusion. These domains include the fusion peptide, the cytoplasmic domain, as well as heptad repeat (HR) domains. Peptides with sequences from HR domains inhibit fusion, and characterization of the mechanism of this inhibition provides evidence for conformational changes in the F protein upon activation of fusion. Both proteolytic cleavage of the F protein and interactions with the attachment protein are required for fusion activation in most systems. Subsequent steps in membrane merger directed by F protein are poorly understood.     

Structure and function of a paramyxovirus fusion protein

Paramyxoviruses initiate infection by attaching to cell surface receptors and fusing viral and cell membranes. Viral attachment proteins, hemagglutinin-neuraminidase (HN), hemagglutinin (HA), or glycoprotein (G), bind receptors while fusion (F) proteins direct membrane fusion. Because paramyxovirus fusion is pH independent, virus entry occurs at host cell plasma membranes. Paramyxovirus fusion also usually requires co-expression of both the attachment protein and the fusion (F) protein. Newcastle disease virus (NDV) has assumed increased importance as a prototype paramyxovirus because crystal structures of both the NDV F protein and the attachment protein (HN) have been determined. Furthermore, analysis of structure and function of both viral glycoproteins by mutation, reactivity of antibody, and peptides have defined domains of the NDV F protein important for virus fusion. These domains include the fusion peptide, the cytoplasmic domain, as well as heptad repeat (HR) domains. Peptides with sequences from HR domains inhibit fusion, and characterization of the mechanism of this inhibition provides evidence for conformational changes in the F protein upon activation of fusion. Both proteolytic cleavage of the F protein and interactions with the attachment protein are required for fusion activation in most systems. Subsequent steps in membrane merger directed by F protein are poorly understood.     

Influenza Hemifusion Phenotype Depends on Membrane Context: Differences in Cell–Cell and Virus–Cell Fusion

Influenza viral entry into the host cell cytoplasm is accomplished by a process of membrane fusion mediated by the viral hemagglutinin protein. Hemagglutinin acts in a pH-triggered fashion, inserting a short fusion peptide into the host membrane followed by refolding of a coiled-coil structure to draw the viral envelope and host membranes together. Mutations to this fusion peptide provide an important window into viral fusion mechanisms and protein–membrane interactions. Here, we show that a well-described fusion peptide mutant, G1S, has a phenotype that depends strongly on the viral membrane context. The G1S mutant is well known to cause a “hemifusion” phenotype based on experiments in transfected cells, where cells expressing G1S hemagglutinin can undergo lipid mixing in a pH-triggered fashion similar to virus but will not support fusion pores. We compare fusion by the G1S hemagglutinin mutant expressed either in cells or in influenza virions and show that this hemifusion phenotype occurs in transfected cells but that native virions are able to support full fusion, albeit at a slower rate and 10–100 × reduced infectious titer. We explain this with a quantitative model where the G1S mutant, instead of causing an absolute block of fusion, alters the protein stoichiometry required for fusion. This change slightly slows fusion at high hemagglutinin density, as on the viral surface, but at lower hemagglutinin density produces a hemifusion phenotype. The quantitative model thus reproduces the observed virus–cell and cell–cell fusion phenotypes, yielding a unified explanation where membrane context can control the observed viral fusion phenotype.     

Wrapping things up about virus RNA replication

All single-stranded 'positive-sense' RNA viruses that infect mammalian, insect or plant cells rearrange internal cellular membranes to provide an environment facilitating virus replication. A striking feature of these unique membrane structures is the induction of 70-100 nm vesicles (either free within the cytoplasm, associated with other induced vesicles or bound within a surrounding membrane) harbouring the viral replication complex (RC). Although similar in appearance, the cellular composition of these vesicles appears to vary for different viruses, implying different organelle origins for the intracellular sites of viral RNA replication. Genetic analysis has revealed that induction of these membrane structures can be attributed to a particular viral gene product, usually a non-structural protein. This review will highlight our current knowledge of the formation and composition of virus RCs and describe some of the similarities and differences in RNA-membrane interactions observed between the virus families Flaviviridae and Picornaviridae.     

Involvement of lipids in different steps of the flavivirus fusion mechanism

Flavivirus membrane fusion is triggered by acidic pH and mediated by the major envelope protein E. A structurally very similar fusion protein is found in alphaviruses, and these molecules are designated class II viral fusion proteins. In contrast to that of flaviviruses, however, alphavirus fusion has been shown to be absolutely dependent on the presence of cholesterol and sphingomyelin in the target membrane, suggesting significant differences in the fusion protein-membrane interactions that lead to fusion. With the flavivirus tick-borne encephalitis virus (TBEV), we have therefore conducted a study on the lipid requirements of viral fusion with liposomes and on the processes preceding fusion, specifically, the membrane-binding step and the fusion-associated oligomeric switch from E protein dimers to trimers. As with alphaviruses, cholesterol had a strong promoting effect on membrane binding and trimerization of the fusion protein, and-as shown by the use of cholesterol analogs-the underlying interactions involve the 3beta-hydroxyl group at C-3 in both viral systems. In contrast to alphaviruses, however, these effects are much less pronounced with respect to the overall fusion of TBEV and can only be demonstrated when fusion is slowed down by lowering the temperature. The data presented thus suggest the existence of structurally related interactions of the flavivirus and alphavirus fusion proteins with cholesterol in the molecular processes required for fusion but, at the same time, point to significant differences between the class II fusion machineries of these viruses.     

Fusion of Semliki Forest virus with cholesterol-containing liposomes at low pH: a specific requirement for sphingolipids

Semliki Forest virus (SFV) utilizes a membrane fusion strategy to introduce its genome into the host cell. After binding to cell-surface receptors, virus particles are internalized through receptor-mediated endocytosis and directed to the endosomal cell compartment. Subsequently, triggered by the acid pH in the lumen of the endosomes, the viral envelope fuses with the endosomal membrane. As a result of this fusion reaction the viral RNA gains access to the cell cytosol. Low-pH-induced fusion of SFV, in model systems as well as in cells, has been demonstrated previously to be strictly dependent on the presence of cholesterol in the target membrane. In this paper, we show that fusion of SFV with cholesterol-containing liposomes depends on sphingomyelin (SM) or other sphingolipids in the target membrane, ceramide representing the sphingolipid minimally required for mediating the process. The action of the sphingolipid is confined to the actual fusion event, cholesterol being necessary and sufficient tor low-pH-dependent binding of the virus to target membranes. The 3-hydroxyl group on the sphingosine backbone plays a key role in the SFV fusion reaction, since 3-deoxy-sphingomyelin does not support the process. This, and the remarkably low levels of sphingolipid required for half-maximal fusion (1–2 mol%), suggest that the sphingolipid does not play a structural role in SFV fusion, but rather acts as a co-factor, possibly through activation of the viral fusion protein. Domain formation between cholesterol and sphingolipid, although it may facilitate SFV fusion, is unlikely to play a crucial role in the process.     

Rotation-Activated and Cooperative Zipping Characterize Class I Viral Fusion Protein Dynamics

Class I viral fusion proteins are α-helical proteins that facilitate membrane fusion between viral and host membranes through large conformational transitions. Although prefusion and postfusion crystal structures have been solved for many of these proteins, details about how they transition between these states have remained elusive. This work presents the first, to our knowledge, computational survey of transitions between pre- and postfusion configurations for several class I viral fusion proteins using structure-based models to analyze their dynamics. As suggested by their structural similarities, all proteins share common mechanistic features during their transitions that can be characterized by a diffusive rotational search followed by cooperative N- and C-terminal zipping. Instead of predicting a stable spring-loaded intermediate, our model suggests that helical bundle formation is mediated by N- and C-terminal interactions late in the transition. Shared transition features suggest a global mechanism in which fusion is activated by slow protein-core rotation.     

Spring-loaded heptad repeat residues regulate the expression and activation of paramyxovirus fusion protein

During viral entry, the paramyxovirus fusion (F) protein fuses the viral envelope to a cellular membrane. Similar to other class I viral fusion glycoproteins, the F protein has two heptad repeat regions (HRA and HRB) that are important in membrane fusion and can be targeted by antiviral inhibitors. Upon activation of the F protein, HRA refolds from a spring-loaded, crumpled structure into a coiled coil that inserts a hydrophobic fusion peptide into the target membrane and binds to the HRB helices to form a fusogenic hairpin. To investigate how F protein conformational changes are regulated, we mutated in the Sendai virus F protein a highly conserved 10-residue sequence in HRA that undergoes major structural changes during protein refolding. Nine of the 15 mutations studied caused significant defects in F protein expression, processing, and fusogenicity. Conversely, the remaining six mutations enhanced the fusogenicity of the F protein, most likely by helping spring the HRA coil. Two of the residues that were neither located at "a" or "d" positions in the heptad repeat nor conserved among the paramyxoviruses were key regulators of the folding and fusion activity of the F protein, showing that residues not expected to be important in coiled-coil formation may play important roles in regulating membrane fusion. Overall, the data support the hypothesis that regions in the F protein that undergo dramatic changes in secondary and tertiary structure between the prefusion and hairpin conformations regulate F protein expression and activation.     

Intracellular and viral membrane fusion: a uniting mechanism

Structural and functional analyses have revealed remarkable mechanistic similarities between viral and intracellular fusion. Both fusion processes are driven by an orchestrated cascade of protein binding and folding reactions. After an initial tethering step, activation of the fusion machinery links the opposing membranes and protein folding pulls the membranes in close proximity; fusion pores form, open and dilate, and the process culminates in the complete merging of the lipid bilayers. Viral fusion is mediated by a single fusion protein, whereas the intracellular fusion machinery is split into matching halves, the v- and t-SNAREs. SNAREs, together with synaptotagmins, emerge as the key machinery for regulated exocytosis.     

Protein-induced fusion can be modulated by target membrane lipids through a structural switch at the level of the fusion peptide

Regulatory features of protein-induced membrane fusion are largely unclear, particularly at the level of the fusion peptide. Fusion peptides being part of larger protein complexes, such investigations are met with technical limitations. Here, we show that the fusion activity of influenza virus or Golgi membranes is strongly inhibited by minor amounts of (lyso)lipids when present in the target membrane but not when inserted into the viral or Golgi membrane itself. To investigate the underlying mechanism, we employ a membrane-anchored peptide system and show that fusion is similarly regulated by these lipids when inserted into the target but not when present in the peptide-containing membrane. Peptide-induced fusion is regulated by a reversible switch of secondary structure from a fusion-permissive alpha-helix to a nonfusogenic beta-sheet. The "on/off" activation of this switch is governed by minor amounts of (lyso)-phospholipids in targets, causing a drop in alpha-helix and a dramatic increase in beta-sheet contents. Concomitantly, fusion is inhibited, due to impaired peptide insertion into the target membrane. Our observations in biological fusion systems together with the model studies suggest that distinct lipids in target membranes provide a means for regulating membrane fusion by causing a reversible secondary structure switch of the fusion peptides.     

Spring-loaded model revisited: paramyxovirus fusion requires engagement of a receptor binding protein beyond initial triggering of the fusion protein

During paramyxovirus entry into a host cell, receptor engagement by a specialized binding protein triggers conformational changes in the adjacent fusion protein (F), leading to fusion between the viral and cell membranes. According to the existing paradigm of paramyxovirus membrane fusion, the initial activation of F by the receptor binding protein sets off a spring-loaded mechanism whereby the F protein progresses independently through the subsequent steps in the fusion process, ending in membrane merger. For human parainfluenza virus type 3 (HPIV3), the receptor binding protein (hemagglutinin-neuraminidase [HN]) has three functions: receptor binding, receptor cleaving, and activating F. We report that continuous receptor engagement by HN activates F to advance through the series of structural rearrangements required for fusion. In contrast to the prevailing model, the role of HN-receptor engagement in the fusion process is required beyond an initiating step, i.e., it is still required even after the insertion of the fusion peptide into the target cell membrane, enabling F to mediate membrane merger. We also report that for Nipah virus, whose receptor binding protein has no receptor-cleaving activity, the continuous stimulation of the F protein by a receptor-engaged binding protein is key for fusion. We suggest a general model for paramyxovirus fusion activation in which receptor engagement plays an active role in F activation, and the continued engagement of the receptor binding protein is essential to F protein function until the onset of membrane merger. This model has broad implications for the mechanism of paramyxovirus fusion and for strategies to prevent viral entry.     

Membrane fusion activity of purified SipB, a Salmonella surface protein essential for mammalian cell invasion

An early event in Salmonella infection is the invasion of non-phagocytic intestinal epithelial cells. The pathogen is taken up by macropinocytosis, induced by contact-dependent delivery of bacterial proteins that subvert signalling pathways and promote cytoskeletal rearrangement. SipB, a Salmonella protein required for delivery and invasion, was shown to localize to the cell surface of bacteria invading mammalian target cells and to fractionate with outer membrane proteins. To investigate the properties of SipB, we purified the native full-length protein following expression in recombinant Escherichia coli. Purified SipB assembled into hexamers via an N-terminal protease-resistant domain predicted to form a trimeric coiled coil, reminiscent of viral envelope proteins that direct homotypic membrane fusion. The SipB protein integrated into both mammalian cell membranes and phospholipid vesicles without disturbing bilayer integrity, and it induced liposomal fusion that was optimal at neutral pH and influenced by membrane lipid composition. SipB directed heterotypic fusion, allowing delivery of contents from E. coli-derived liposomes into the cytosol of living mammalian cells.     

Mechanism of fusion of Sendai virus: role of hydrophobic interactions and mobility constraints of viral membrane proteins. Effects of polyethylene glycol

The mechanism of Sendai virus fusion was investigated by studying the effect of the dehydrating agent polyethylene glycol (PEG) on the interaction of the virus with erythrocyte membranes. The initial rate of virus fusion, monitored continuously by a fluorescence membrane fusion assay, increases approximately 5-fold in the presence of small amounts (4%, w/v) of PEG. The polymer did not trigger a massive nonspecific fusion event, as the limited number of virus particles that fuse per erythrocyte ghost remains unaltered. A mass action kinetic analysis reveals that the binding rate constant increases approximately 1.5-fold; however, the fusion rate constant is enhanced by about an order of magnitude. The results demonstrate that hydrophobic interaction forces dominate the actual fusion step of the virus. Below about 22 degrees C, the viral membrane proteins appear to be clustered, as revealed by temperature-dependent fluorescence measurements of fluorescently tagged viral proteins. Clustering is not modulated by the presence of PEG, and fusion at those conditions is not observed. It is concluded that in addition to hydrophobic interactions, constraints in the mobility of the viral membrane proteins codetermine the fusogenic capacity of the virus. Such constraints have to be relieved in order to allow the occurrence of the hydrophobic interactions. PEG primarily affects the surface properties of the viral membrane, including the properties of the membrane glycoproteins. We hypothesize that during virus-target membrane interaction but prior to the actual fusion reaction, the fusion protein may undergo a conformational change, triggered by an enhancement in hydrophobic environment, which accounts for the need to establish close, i.e. fusion-susceptible intermembrane contact between virus and target membrane.