Envelope glycoproteins from different virus families appear utterly dissimilar in primary amino acid sequence and domain organization, structure, and even function. Receptor binding and fusion of some virus particles are mediated by the same protein. For others, these functions are segregated into two distinct proteins. Some viral proteins can mediate fusion at the cell surface, while others require activation by acidic pH in endocytic vesicles. Nevertheless, despite these differences, the fusion mechanism is remarkably similar between all fusion proteins from different virus families and relies on conformational changes in the viral protein or subunit that mediates fusion. Based on protein structure, viral fusion proteins can be assigned to one of three classes. Class I includes the most extensively studied examples of fusion proteins, exemplified by influenza virus HA. The study of this protein elucidated the mechanism of fusion with general features that are common to all fusion proteins regardless of class.
Class I Fusion Proteins
In addition to influenza virus HA, this class includes the human immunodeficiency virus type 1 envelope glycoprotein and paramyxovirus fusion proteins. These proteins are initially synthesized as a polyprotein precursor that is thermo-dynamically stable until subsequently cleaved. Proteolytic cleavage is a determinant of tropism: for example, inefficient cleavage of some avian influenza HA protein precursors in mammalian cells limits their zoonotic potential. This tropism restriction occurs because cleavage is essential for the release of the fusion peptide, a highly hydrophobic sequence that can insert into lipid membranes and that lies at the cleaved, extreme N terminus of the transmembrane subunit. Following cleavage, the fusion peptide has to be sequestered until the virus particle leaves the producing cell and reaches the target cellular membranes; otherwise it can insert into membranes prematurely.
The process is best illustrated for influenza virus HA. In the envelope protein precursor, HA0, the fusion peptide sequence is positioned at the junction of the two subunits HA1 and HA2, forming a loop where proteolytic cleavage occurs (Fig. 5.12). After cleavage, the fusion peptide translocates to a cavity formed by both HA1 and HA2 residues so that its hydrophobic residues are buried. In order for fusion to occur, the fusion peptide must be exposed. Therefore, protein regions that keep the fusion peptide shielded have to rearrange. This rearrangement is dependent on a signal indicating that the virus particle has reached the appropriate target membrane and is known as the fusion trigger. Acid pH is a common fusion trigger for viral fusion proteins, including those from class II and III, but fusion triggers can differ among viruses. In all cases, however, following exposure, the fusion peptide must reach the target membrane in order to penetrate it. This movement is achieved by extensive conformational changes in the transmembrane subunits of fusion proteins (Fig. 5.12). Upon fusion activation, protein segments that previously held the fusion peptide close to the transmembrane subunit core structure fold to project the fusion peptide toward the target membrane.
Figure 5.12 Conformational changes of class I proteins during fusion. The envelope glycoprotein of influenza viruses is synthesized as a precursor, HA0, that is cleaved at a specific sequence that forms a loop (green and red). For simplicity, the diagram of only one of three members of a trimer is shown, with HA1 in gray and HA2 in various colors to highlight conformational changes (PDB ID: 1HA0, 4FNK, 1HTM). The viral membrane would be located at the bottom of the image and the target membrane at the top. Following cleavage, HA1 and HA2 remain attached by disulfide bonds and other interactions. The fusion peptide (red) is now buried in the stem of the trimer structure, but the overall structure of the protein is not altered. HA1 mediates binding to the receptor and virus particles are endocytosed. Import of protons into the endosome triggers conformational rearrangements. The HA1 subunits tilt away from the core to expose the fusion peptide (not shown). A segment of HA2 (black) assumes a helical conformation that now extends N-terminal sequences, including the fusion peptide, toward the target membrane. At the C terminus of the protein, segments (blue and purple) fold against the central helix (cyan). This fold, known as the hairpin, brings the viral membrane at the C terminus of HA2 close to the target cell membrane so that fusion can occur.
Insertion of the fusion peptide into the target membrane results in the adoption of an extended intermediate structure. This structure bridges the two membranes but does not bring them close enough for fusion to occur (Fig. 5.13). Consequently, additional changes in protein conformation are required. Amino acid sequence comparison reveals a second region, in addition to the fusion peptide, that is similar in several types of viral fusion proteins. This region is a heptad repeat (a repeated sequence of seven amino acids) consisting of leucine or isoleucine residues that folds into an α-helical coiled coil known as a leucine zipper. Leucine zippers mediate protein-protein interactions and were therefore originally thought to be responsible for envelope protein oligomerization. However, mutational analysis of the leucine zippers from several viral transmembrane proteins, including measles virus F and human immunodeficiency virus type 1 TM, demonstrated that this motif was not important for synthesis, transport (and hence oligomerization), and incorporation into the virus particles, but was absolutely required for fusion.
Figure 5.13 Influenza virus entry. The globular heads of HA trimers mediate binding of the virus to sialic acid-containing cell receptors. The virus-receptor complex is endocytosed, and import of H+ ions into the endosome acidifies the interior. Upon acidification, HA undergoes conformational rearrangements that produce a fusogenic protein. The globular heads (gray) are pulled to the side, exposing the fusion peptides (red), and the loop region of native HA (black) becomes a coiled coil, moving the fusion peptides to the top of the molecule near the cell membrane. This structure is referred to as an extended hairpin. At the viral membrane, an α-helix (purple) packs against the trimer core. The coiled-coil bundles, or hairpins, bring the fusion peptides and the transmembrane domains together, moving the cell and viral membranes close so that fusion can occur. To allow release of vRNP into the cytoplasm, the H+ ions in the acidic endosome are pumped into the particle interior by the M2 ion channel. As a result, vRNP is primed to dissociate from M1 after fusion of the viral and endosomal membranes. The released vRNPs are imported into the nucleus through the nuclear pore complex via a nuclear localization signal-dependent mechanism (see “Import of Influenza Virus Ribonucleoprotein” below). Adapted from Carr CM, Kim PS. 1994. Science 266:234–236, with permission.
The leucine zippers of the three fusion protein monomers form a core three-stranded coiled coil. Following the insertion of the fusion peptide into the target membrane, the C-terminal part of each fusion protein monomer folds around this central coiled coil in what is described as a “hairpin” (Fig. 5.12 and 5.13). In some fusion proteins, like the human immunodeficiency virus type 1 TM, these C-terminal regions also assume an α-helical conformation that “zip up” around the central coiled coil, resulting in a six-helical bundle. The overall structure of the hairpin is strikingly similar in different fusion proteins (Fig. 5.14).
Folding of the fusion protein into this hairpin decreases the distance between the viral and cell membranes, thereby permitting