NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Coffin JM, Hughes SH, Varmus HE, editors. Retroviruses. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory Press; 1997.

Cover of Retroviruses


Show details

Membrane Fusion and Viral Entry

It is clear from the discussion of viral tropism that the recognition and binding of a high-affinity receptor by a retroviral glycoprotein are just the initial steps in a complex series of events that result in fusion of the viral membrane with that of the cell. As with most viruses, there is strong evidence that the machinery for fusion and entry lies within the retroviral glycoprotein, with the receptor providing little more than a specific point of recognition and attachment. The details of this process remain unresolved, although recent studies are beginning to elucidate steps in this pathway and point to common mechanisms shared between retroviruses and other enveloped viruses.

Entry Point into the Cell

There are two different pathways by which the interaction between virus and receptor leads to membrane fusion. Several enveloped viruses including influenza virus, VSV, and the alphaviruses require the acidic environment of the endosome to activate the fusogenic potential of their envelope glycoprotein complexes. Thus, after binding a receptor molecule on the cell surface, they are taken up into coated pits, undergo endocytosis, and then release their replicative machinery into the cell following fusion with the endosomal membrane. The entry of such viruses can be abrogated by the addition of lysosomotropic compounds such as chloroquine, ammonium chloride, and amantidine, which raise the pH of the endosome above that required for activation of the fusion process. In addition, transient exposure to low pH following receptor binding can trigger fusion at the cell surface by these normally endosome-dependent viral glycoproteins.

In contrast, the entry of most retroviruses is not dependent on exposure to an acidic environment. The efficiency of infection and membrane fusion by HIV-1 (Stein et al. 1987; McClure et al. 1988) and RSV-A (Gilbert et al. 1990) are unaffected by the presence of lysosomotropic agents, and fusion of HIV-1 with cell surface membranes occurs with equal efficiency over a pH range of 5.0–8.0 (Sinangil et al. 1988). Furthermore, among mammalian retroviruses, entry of only ecotropic MLV (or pseudotypes) is affected by raising endosomal pH (Andersen and Nexo 1983; McClure et al. 1990). Interestingly, dependence on exposure to low pH does not appear to be linked specifically to an interaction with the ecotropic receptor. Through the construction of chimeric amphotropic-ecotropic envelope glycoproteins, it has been possible to dissect the region defining low pH dependence from that defining receptor recognition (Nussbaum et al. 1993).

The requirement of MLV-E for an acidic point in the entry pathway is complex, however, since in the fusogenic rat XC cell line, ammonium chloride does not inhibit entry, and exposure of MLV-E to low pH at the cell surface neither enhances cell-cell fusion in mouse fibroblasts nor inactivates free virus (McClure et al. 1990; Nussbaum et al. 1993). Activation may involve low pH only indirectly: Cleavage of the MLV gp70 on viral entry can be observed. In contrast to low pH, exposure to proteinase enhances cell-cell fusion by MLV; conversely, the proteinase inhibitor, leupeptin, inhibits MLV infection (Andersen 1985, 1987; Andersen and Skov 1989). Thus, infection by MLV-E might be dependent on cleavage of the glycoprotein by a low-pH-dependent enzyme that is located in the endosome (McClure et al. 1990). The only retroviral glycoprotein that mimics those of the endosomally dependent viruses is that of MMTV. Exposure of cells expressing this glycoprotein to low-pH medium results in cell-cell fusion, similar to that seen with the influenza virus hemagglutinin (Redmond et al. 1984). At the present time, however, the pH dependence of MMTV entry has not been established.

In summary, the majority of retroviruses do not require exposure to low pH to activate the fusion process following receptor binding. Entry for the majority of these viruses is likely to occur through fusion with the plasma membrane as is observed with members of the paramyxovirus family.

Kinetics of Fusion and Entry

After binding to the high-affinity receptor on the cell surface, a finite period of time is required before viral entry is resistant to treatment of the cells with anti-receptor or neutralizing antibodies. During this time, temperature-dependent processes occur that include membrane fusion and internalization of the viral replication machinery. Early experiments with ASLV, in which virus was allowed to bind at 4oC, showed that viral internalization initiates rapidly (within 5 min) following a shift to 37oC (Steck and Rubin 1966). More recent experiments with this virus have shown that whereas binding of virus to cells occurred linearly with time at 4oC, no membrane fusion occurs at temperatures below 20oC and that fusion increases as temperatures are raised to 38oC. Fusion initiates rapidly (within 5 min) upon shift to physiological temperatures and is essentially complete by 3 hours (Gilbert et al. 1990).

HIV demonstrates a similar temperature dependence of fusion, most of which occurs rapidly following incubation at 37oC (Sinangil et al. 1988; Fu et al. 1993; Frey et al. 1995). In contrast, influenza virus shows equivalent levels of virus-induced fusion at both 20oC and 37oC (Frey et al. 1995). Using fluorescence dequenching, Sinangil et al. (1988) showed that membrane mixing occurs within 5 minutes of shifting to 37oC, and similarly rapid kinetics can be observed in time interval series of electron micrographs of HIV infection of T cells (Grewe et al. 1990). Somewhat slower and more diverse kinetics are observed when viral entry is measured by the time required for 50% of the infectious units to escape being blocked by the competitive anti-CD4 antibody Leu3A (Srivastava et al. 1991). In highly permissive cells, 50% entry occurs within 30 minutes compared to an average of 4 hours in less permissive cell lines. The rate of entry correlates with the efficiency of entry, such that viral stocks exhibit the highest titers on those cells in which entry is most rapid (Srivastava et al. 1991).

Studies of HIV-1 Env-induced cell-cell fusion, using a microscopic assay to follow redistribution of fluorescent dyes, have shown that this process occurs much more slowly than virus-cell fusion (Dimitrov et al. 1991; Dimitrov and Blumenthal 1994). Although the presence of CD4 on the target cell is essential for fusion to occur, there is no correlation between the surface concentration of the receptor and the extent of HIV Env-mediated fusion (Dimitrov et al. 1991). Fusion is dependent on the presence of calcium ions (Dimitrov et al. 1993), and since calcium is not required for gp120/ CD4 binding, it must have a role in a presently undefined part of the fusion process. Using imageenhanced microscopy, it is possible to show that cells make contact within minutes, in many cases by using microspikes to touch and adhere to adjacent cells, but that fusion only occurs after relatively long lag times (15 min to hours). The process of syncytium formation, which is much slower, involves subsequent fusion to other cells, rather than simultaneous fusion of many cells (Dimitrov and Blumenthal 1994).

Conformational Changes

It is known, for the hemagglutinin of influenza virus and the glycoprotein complex of Semliki Forest virus, that the acid environment of the endosomal compartment induces a conformational change in the proteins that is necessary for activation of the fusion process (for review, see White et al. 1983; White 1992). In the case of HIV, SIV, and ASLV-A, analogous conformational changes have been observed following binding of their respective receptor molecules (Sattentau and Moore 1991; Sattentau et al. 1993; Gilbert et al. 1995), and recent experiments suggest that these may parallel in molecular terms those of the influenza virus hemagglutinin.

Soluble-receptor-induced HIV-1 Env Dissociation

Infection of cells by laboratory strains of HIV-1 can be efficiently blocked by pretreatment of the virus with sCD4 (Smith et al. 1987; Deen et al. 1988; Fisher et al. 1988; Hussey et al. 1988; Traunecker et al. 1988). Experiments aimed at defining the mechanism of this neutralization demonstrated that this treatment induces rapid shedding of gp120 from virions and loss of their characteristic surface spikes (Kirsh et al. 1990; Moore et al. 1990, 1991b; Hart et al. 1991). Dissociation can be mediated by molecules corresponding to domain 1 (106 amino acids) of CD4 alone and depends on a functional binding site, since mutations in CD4 that disrupt gp120 binding also block dissociation. It appears to require binding of at least two CD4 molecules by each glycoprotein oligomer (Hart et al. 1991; Moore et al. 1991b; Sattentau and Moore 1991; Moore and Klasse 1992; Fu et al. 1993).

sCD4-induced Epitope Exposure with HIV-1 Env

As discussed previously (see section above Retroviral env Gene Products: Receptor Binding and Host-range Determinants, Structural Analysis of Env Proteins), brief treatment of influenza virus with low pH activates fusion and results in structural rearrangement of hemagglutinin as detected by the exposure of antigenic epitopes and protease cleavage sites normally sequestered within the molecule (Skehel et al. 1982; Daniels et al. 1983; White et al. 1983). With other viruses, including retroviruses, that do not require low pH for entry, similar changes can be induced directly by interaction with the receptor. Thus, dissociation of gp120 by CD4 is accompanied by exposure of epitopes on gp41 not reactive on the native molecule (Hart et al. 1991; Sattentau and Moore 1991). Epitope exposure does not depend on loss of gp120 since similar results can be obtained following CD4 binding at 4oC where dissociation does not occur (Sattentau and Moore 1991; Sattentau et al. 1995). Moreover, CD4 binding at low temperature leads to increased exposure of V2 and V3 loop epitopes on the gp120 molecule itself and increased susceptibility to thrombin cleavage within the V3 loop. These alterations are consistent with conformational changes that expose regions of V3 which are normally sequestered (Sattentau and Moore 1991; Sattentau et al. 1993). Similarly, treatment of HIV-1 with nondissociating concentrations of sCD4 results in the exposure of discontinuous epitopes on the gp120 molecule, apparently by inducing the movement of the V1/V2 loops on the surface of the molecule (Wyatt et al. 1995). These effects are not limited to sCD4 treatment of virions or infected cells since a similar exposure of gp41 epitopes can be seen at the interface of fusing cells (Sattentau and Moore 1993). Thus, cell surface CD4, like sCD4, can induce Env rearrangement following gp120 binding.

The exposure of epitopes in both TM and SU, coupled with the dissociation of gp120 observed with laboratory-adapted isolates of HIV-1, represents the most dramatic effect of sCD4 on Env conformation and may reflect the relative instability of the glycoprotein complexes of these isolates. The susceptibility of primary isolates of HIV-1 to sCD4 neutralization is significantly lower than that of cell-line-adapted isolates (Daar et al. 1990), and sCD4 is very inefficient at inducing gp120 dissociation in the nonadapted strains (Moore et al. 1992), perhaps reflecting the lower affinity for CD4 exhibited by the oligomeric form of Env from natural isolates (Moore et al. 1992). Indeed, the adaptation of viruses to growth in cell lines is accompanied by a higher affinity for CD4 and more readily dissociable gp120 (Moore et al. 1993b; Willey et al. 1994a), as well as increased sensitivity to antibody neutralization (Moore and Ho 1995). The differences between laboratory-adapted viral strains and those that exist in the infected individual probably reflect a relaxation of the constraints imposed on Env by the in vivo environment and the host immune response. The relative instability of the glycoprotein complex of adapted strains may allow more efficient infection of cultured cells, but the relative ease with which these viruses are neutralized would be expected to allow rapid elimination in an immune competent host.

Receptor-induced Conformational Changes in SIV and HIV-2 Env

The glycoproteins of HIV-2 and SIV also have a lower affinity for CD4, and these viruses are resistant to neutralization by sCD4. In contrast to the laboratory-adapted strains of HIV-1, both SIV and HIV-2 infection and syncytium formation can be enhanced by subinhibitory concentrations of the soluble receptor. Incubation of sCD4 with SIVAGM viruses not only accelerates the rate of infection, but also leads to a 100-fold increase in viral titer (Allan et al. 1990; Sekigawa et al. 1990; Werner et al. 1990; Clapham et al. 1991). A similar enhancement of infectivity can also be observed with HIV-1 when poorly neutralizable primary isolates are treated with subinhibitory concentrations of sCD4 (Sullivan et al. 1995). Treatment of SIV- or HIV-2-infected cells with the soluble receptor does not result in dissociation of gp120 or increased exposure of TM epitopes, indicative of stronger interactions between the Env subunits in these viruses (Allan et al. 1992; Sattentau et al. 1993). Nevertheless, this treatment does result in some conformational changes, since an increase in the accessibility of the V2 and V3 loops to antibody can be observed after sCD4 treatment (Sattentau et al. 1993), and antibodies from SIVAGM-infected monkeys that fail to neutralize virusin vitro can prevent the sCD4-induced enhancement of SIVAGM infection (Allan et al. 1990). These results are consistent with the suggestion that sCD4 enhancement of SIVAGM (and presumably of HIV-1 primary isolates and HIV-2) is a form of receptormediated activation (Allan et al. 1990), analogous to the low pH activation observed with influenza virus or Semliki Forest virus (Marsh and Helenius 1989).

Molecular Rearrangements following Receptor Binding: Formation of Coiled-Coils

As described above (see Retroviral env Gene Products), both crystallographic and peptide studies with influenza virus have shown that the low-pH-mediated conformational changes involve the transition of a region of extended polypeptide chain within the HA2 (TM) protein into an elongated helical structure (Carr and Kim 1993; Bullough et al. 1994). The newly formed helices of the hemagglutinin trimer form a triple-helix coiled-coil structure that is postulated to position the fusion peptide, located at the amino terminus of HA2, for insertion into the target cell membrane (see Fig. 8C).

There is suggestive evidence for a related mechanism in retroviral entry. First, a heptad repeat (leucine-zipper-like) region capable of forming a coiled-coil structure similar to that of hemagglutinin is found within the amino-terminal 100 amino acids of the TM protein of a majority of retroviruses. Second, peptides corresponding to this domain form stable coiled-coil structures (Wild et al. 1992). Third, the heptad repeat alone is capable of forming oligomers when fused to a foreign protein, such as to the carboxyl terminus of a monomeric form of bacterial protein A (Bernstein et al. 1995). Thus, this region has the ability to both oligomerize and mediate oligomerization of protein molecules. The domain is highly conserved among HIV-1 isolates and was originally postulated to have a role in oligomerization of the glycoprotein complex during assembly (Gallaher et al. 1989; Delwart and Mosialos 1990). Mutational analyses of the region argue against this role but show that even minor amino acid substitutions have dramatic effects on the biological activity of the HIV-1 glycoprotein, blocking cell-cell fusion and viral infectivity (Dubay et al. 1992; Chen et al. 1993; Chen 1994).

A role for interactions involving the heptad repeat region in entry is supported by the observation that peptides corresponding to this region are potent inhibitors of HIV-1-induced cell fusion as well as viral infection. This property correlates directly with the ability of the peptides to form stable coiled-coil structures, since the same mutations that block the biological function of Env reduce the melting temperature of the peptide's coiled-coil structure and also abrogate their inhibitory potential (Wild et al. 1994). It has been postulated therefore that binding of CD4 to gp120 might induce conformational changes equivalent to that seen during low pH treatment of the influenza virus hemagglutinin, with the resulting formation of an extended coiled-coil that can present the amino-terminal fusion peptide to the target cell membrane (see Fig. 9) (Wild et al. 1994; Bernstein et al. 1995). Such a mechanism would explain why mutations that block cleavage of the Env precursor effectively block fusion; in the absence of a free amino terminus, the fusion peptide would be unable to undergo such a dramatic displacement.

Latter Stages of the Fusion Process

Even if a parallel exists between myxoviruses and retroviruses, in the formation of a coiled-coil that presents the fusion peptide to the target cell membrane, many aspects of the fusion reaction remain to be elucidated. How, for example, does the amino-terminal fusion peptide of gp41 interact with the membrane and what molecular interactions must occur following insertion? Mutational analyses of the fusion peptides of HIV-1, HIV-2, and SIV, which show similarity to the fusion peptides of the paramyxoviruses (Gallaher 1987), have demonstrated the importance of hydrophobic residues within this region. Substitution of hydrophobic residues with hydrophilic residues reduces syncytium formation and viral infectivity, whereas enhancement of hydrophobicity increases both (Bosch et al. 1989; Freed et al. 1990; Freed and Myers 1992; Steffy et al. 1992). Theoretical predictions and empirical observations suggest that the fusion peptide penetrates the lipid bilayer in an oblique orientation and that mutations which prevent this penetration interfere with the peptide's ability to promote fusion (Martin et al. 1994). Nevertheless, it is not clear whether during viral entry the fusion peptide interacts directly with lipid or with membrane-associated proteins in a manner analogous to that of the signal peptides of secreted and membrane-spanning proteins.

Other domains of the TM protein also appear to have critical roles in a productive fusion reaction. The membrane-spanning anchor domain of the TM protein is an important structural feature, as might be expected if the initial result of fusion peptide insertion is an oligomeric structure that acts as a bridge between the viral and cell membranes. Because this domain of the HIV-1 glycoprotein can be replaced by the membrane anchor of CD4 and the chimeric protein retains fusogenicity (Vincent et al. 1993), there appears to be no requirement for a specific amino acid sequence within the transmembrane domain for the fusion reaction to occur. In contrast, substitution of a glycosylphosphatidylinositol (GPI) anchor for the HIV-1 protein anchor abrogates fusion and infectivity, even though this PI-linked protein is efficiently transported to the plasma membrane and incorporated into virions (Salzwedel et al. 1993; Weiss and White 1993). A similar result has been reported for a GPI-linked form of the influenza virus hemagglutinin protein, although in this case, lipid mixing of the outer leaflet of the bilayer (hemifusion) was seen in the absence of complete fusion and mixing of cell contents (Kemble et al. 1994). Thus, a protein component that spans both leaflets of the lipid bilayer appears to be critical for complete mixing of the viral and target membranes.

One event that must occur for fusion to proceed is the merging of apposing membranes and the formation of a fusion pore. Little is known about these steps in any of the viral systems studied to date. Models have been proposed for hemagglutinin-mediated fusion, in which the fusion pore is lined with several conformationally altered hemagglutinin trimers (Stegmann et al. 1989; White 1992), but direct evidence for such structures is lacking. Since this is a critical step in membrane fusion and a potential point at which retroviral entry could be blocked, it is likely to be an area of active future investigation.

Copyright © 1997, Cold Spring Harbor Laboratory Press.
Bookshelf ID: NBK19401


  • PubReader
  • Print View
  • Cite this Page

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...